From 6f5a1e71fef4c0cbaa2c1585d7105221c4517cae Mon Sep 17 00:00:00 2001 From: njlyon0 Date: Wed, 28 Aug 2024 12:25:41 -0400 Subject: [PATCH] style: standardized citation format across modules --- .../mod_data-viz/execute-results/html.json | 4 +-- .../figure-html/multi-modal-1.png | Bin 202336 -> 200775 bytes .../mod_data-viz/figure-html/nms-ord-1.png | Bin 100040 -> 100104 bytes .../execute-results/html.json | 4 +-- _freeze/mod_reports/execute-results/html.json | 4 +-- _freeze/mod_spatial/execute-results/html.json | 4 +-- _freeze/mod_stats/execute-results/html.json | 4 +-- .../figure-html/mem-explore-graph-1.png | Bin 189229 -> 189011 bytes _freeze/mod_wrangle/execute-results/html.json | 4 +-- mod_credit.qmd | 6 ++-- mod_data-disc.qmd | 10 +++--- mod_data-viz.qmd | 8 ++--- mod_facilitation.qmd | 4 +-- mod_findings.qmd | 5 +-- mod_interactivity.qmd | 6 ++-- mod_next-steps.qmd | 4 +-- mod_project-mgmt.qmd | 8 ++--- mod_reports.qmd | 11 +++---- mod_reproducibility.qmd | 29 +++++++++--------- mod_spatial.qmd | 12 ++++---- mod_stats.qmd | 18 +++++------ mod_team-sci.qmd | 12 ++++---- mod_thinking.qmd | 4 +-- mod_version-control.qmd | 8 ++--- mod_wrangle.qmd | 19 ++++++------ 25 files changed, 94 insertions(+), 94 deletions(-) diff --git a/_freeze/mod_data-viz/execute-results/html.json b/_freeze/mod_data-viz/execute-results/html.json index 779593f..efcbc3d 100644 --- a/_freeze/mod_data-viz/execute-results/html.json +++ b/_freeze/mod_data-viz/execute-results/html.json @@ -1,8 +1,8 @@ { - "hash": "9edb204615b994ea90fb63cf2bff58d2", + "hash": "cfb7361be24c401ef4cc557bf0e80724", "result": { "engine": "knitr", - "markdown": "---\ntitle: \"Data Visualization & Exploration\"\ncode-annotations: hover\n---\n\n\n## Overview\n\nData visualization is a fundamental part of working with data. Visualization can be only used in the final stages of a project to make figures for publication but it can also be hugely valuable for quality control and hypothesis development processes. This module focuses on the fundamentals of graph creation in an effort to empower you to apply those methods in the various contexts where you might find visualization to be helpful.\n\n## Learning Objectives\n\nAfter completing this module you will be able to: \n\n- Define fundamental `ggplot2` vocabulary\n- Identify appropriate graph types for given data type/distribution\n- Discuss differences between presentation- and publication-quality graphs\n- Explain how your graphs can be made more accessible\n\n## Needed Packages\n\nIf you'd like to follow along with the code chunks included throughout this module, you'll need to install the following packages:\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Note that these lines only need to be run once per computer\n## So you can skip this step if you've installed these before\ninstall.packages(\"tidyverse\")\ninstall.packages(\"lterdatasampler\")\ninstall.packages(\"supportR\")\ninstall.packages(\"cowplot\")\ninstall.packages(\"vegan\")\ninstall.packages(\"ape\")\n```\n:::\n\n\nWe'll go ahead and load some of these libraries as well to be able to better demonstrate these concepts.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load needed libraries\nlibrary(tidyverse)\n```\n:::\n\n\n## Graphing with `ggplot2`\n\n### `ggplot2` Fundamentals\n\nYou may already be familiar with the `ggplot2` package in R but if you are not, it is a popular graphing library based on [The Grammar of Graphics](https://bookshop.org/p/books/the-grammar-of-graphics-leland-wilkinson/1518348?ean=9780387245447). Every ggplot is composed of four elements:\n\n1. A 'core' `ggplot` function call\n2. Aesthetics\n3. Geometries\n4. Theme\n\nNote that the theme component may be implicit in some graphs because there is a suite of default theme elements that applies unless otherwise specified. \n\nThis module will use example data to demonstrate these tools but as we work through these topics you should feel free to substitute a dataset of your choosing! If you don't have one in mind, you can use the example dataset shown in the code chunks throughout this module. This dataset comes from the [`lterdatasampler` R package](https://lter.github.io/lterdatasampler/) and the data are about fiddler crabs (_Minuca pugnax_) at the [Plum Island Ecosystems (PIE) LTER](https://pie-lter.ecosystems.mbl.edu/welcome-plum-island-ecosystems-lter) site.\n\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load the lterdatasampler package\nlibrary(lterdatasampler)\n\n# Load the fiddler crab dataset\ndata(pie_crab)\n\n# Check its structure\nstr(pie_crab)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\ntibble [392 × 9] (S3: tbl_df/tbl/data.frame)\n $ date : Date[1:392], format: \"2016-07-24\" \"2016-07-24\" ...\n $ latitude : num [1:392] 30 30 30 30 30 30 30 30 30 30 ...\n $ site : chr [1:392] \"GTM\" \"GTM\" \"GTM\" \"GTM\" ...\n $ size : num [1:392] 12.4 14.2 14.5 12.9 12.4 ...\n $ air_temp : num [1:392] 21.8 21.8 21.8 21.8 21.8 ...\n $ air_temp_sd : num [1:392] 6.39 6.39 6.39 6.39 6.39 ...\n $ water_temp : num [1:392] 24.5 24.5 24.5 24.5 24.5 ...\n $ water_temp_sd: num [1:392] 6.12 6.12 6.12 6.12 6.12 ...\n $ name : chr [1:392] \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" ...\n```\n\n\n:::\n:::\n\n\nWith a dataset in hand, let's make a scatterplot of crab size on the Y-axis with latitude on the X. We'll forgo doing anything to the theme elements at this point to focus on the other three elements.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\nggplot(data = pie_crab, mapping = aes(x = latitude, y = size, fill = site)) + # <1>\n geom_point(pch = 21, size = 2, alpha = 0.5) # <2>\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/gg-1-1.png){fig-align='center' width=864}\n:::\n:::\n\n1. We're defining both the data and the X/Y aesthetics in this top-level bit of the plot. Also, note that each line ends with a plus sign\n2. Because we defined the data and aesthetics in the `ggplot()` function call above, this geometry can assume those mappings without re-specificying\n\nWe can improve on this graph by tweaking theme elements to make it use fewer of the default settings.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\nggplot(data = pie_crab, mapping = aes(x = latitude, y = size, fill = site)) +\n geom_point(pch = 21, size = 2, alpha = 0.5) +\n theme(legend.title = element_blank(), # <1>\n panel.background = element_blank(),\n axis.line = element_line(color = \"black\"))\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/gg-2-1.png){fig-align='center' width=864}\n:::\n:::\n\n1. All theme elements require these `element_...` helper functions. `element_blank` removes theme elements but otherwise you'll need to use the helper function that corresponds to the type of theme element (e.g., `element_text` for theme elements affecting graph text)\n\n### Multiple Geometries\n\nWe can further modify `ggplot2` graphs by adding _multiple_ geometries if you find it valuable to do so. Note however that geometry order matters! Geometries added later will be \"in front of\" those added earlier. Also, adding too much data to a plot will begin to make it difficult for others to understand the central take-away of the graph so you may want to be careful about the level of information density in each graph. Let's add boxplots behind the points to characterize the distribution of points more quantitatively.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\nggplot(data = pie_crab, mapping = aes(x = latitude, y = size, fill = site)) +\n geom_boxplot(pch = 21) + # <1>\n geom_point(pch = 21, size = 2, alpha = 0.5) +\n theme(legend.title = element_blank(), \n panel.background = element_blank(),\n axis.line = element_line(color = \"black\"))\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/gg-3-1.png){fig-align='center' width=864}\n:::\n:::\n\n1. By putting the boxplot geometry first we ensure that it doesn't cover up the points that overlap with the 'box' part of each boxplot\n\n:::{.callout-note icon=\"false\"}\n#### Activity: Graph Creation (P1)\n\nIn a script, attempt the following with one of either yours or your group's datasets:\n\n- Make a graph using `ggplot2`\n - Include at least one geometry\n - Include at least one aesthetic (beyond X/Y axes)\n - Modify at least one theme element from the default\n\n:::\n\n### Multiple Data Objects\n\n`ggplot2` also supports adding more than one data object to the same graph! While this module doesn't cover map creation, maps are a common example of a graph with more than one data object. Another common use would be to include both the full dataset and some summarized facet of it in the same plot.\n\nLet's calculate some summary statistics of crab size to include that in our plot.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load the supportR library\nlibrary(supportR)\n```\n\n::: {.cell-output .cell-output-stderr}\n\n```\n\nAttaching package: 'supportR'\n```\n\n\n:::\n\n::: {.cell-output .cell-output-stderr}\n\n```\nThe following object is masked from 'package:dplyr':\n\n count\n```\n\n\n:::\n\n```{.r .cell-code}\n# Summarize crab size within latitude groups\ncrab_summary <- supportR::summary_table(data = pie_crab, groups = c(\"site\", \"latitude\"),\n response = \"size\", drop_na = TRUE)\n\n# Check the structure\nstr(crab_summary)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n'data.frame':\t13 obs. of 6 variables:\n $ site : chr \"BC\" \"CC\" \"CT\" \"DB\" ...\n $ latitude : num 42.2 41.9 41.3 39.1 30 39.6 41.6 33.3 42.7 34.7 ...\n $ mean : num 16.2 16.8 14.7 15.6 12.4 ...\n $ std_dev : num 4.81 2.05 2.36 2.12 1.8 2.72 2.29 2.42 2.3 2.34 ...\n $ sample_size: int 37 27 33 30 28 30 29 30 28 25 ...\n $ std_error : num 0.79 0.39 0.41 0.39 0.34 0.5 0.43 0.44 0.43 0.47 ...\n```\n\n\n:::\n:::\n\n\nWith this data object in-hand, we can make a graph that includes both this and the original, unsummarized crab data. To better focus on the 'multiple data objects' bit of this example we'll pare down on the actual graph code.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\nggplot() + # <1>\n geom_point(pie_crab, mapping = aes(x = latitude, y = size, fill = site),\n pch = 21, size = 2, alpha = 0.2) + \n geom_errorbar(crab_summary, mapping = aes(x = latitude, # <2>\n ymax = mean + std_error,\n ymin = mean - std_error),\n width = 0.2) +\n geom_point(crab_summary, mapping = aes(x = latitude, y = mean, fill = site),\n pch = 23, size = 3) + \n theme(legend.title = element_blank(),\n panel.background = element_blank(),\n axis.line = element_line(color = \"black\"))\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/gg-4-1.png){fig-align='center' width=864}\n:::\n:::\n\n1. If you want multiple data objects in the same `ggplot2` graph you need to leave this top level `ggplot()` call _empty!_ Otherwise you'll get weird errors with aesthetics later in the graph\n2. This geometry adds the error bars and it's important that we add it before the summarized data points themselves if we want the error bars to be 'behind' their respective points\n\n:::{.callout-note icon=\"false\"}\n#### Activity: Graph Creation (P2)\n\nIn a script, attempt the following:\n\n- Add a second data object to the graph you made in the preceding activity\n - _Hint:_ If your first graph is unsummarized, add a summarized version (or vice versa)\n\n:::\n\n## Streamlining Graph Aesthetics\n\nSynthesis projects often generate an entire network of inter-related papers. Ensuring that all graphs across papers from a given team have a similar \"feel\" is a nice way of implying a certain standard of robustness for all of your group's projects. However, copy/pasting the theme elements of your graphs can (A) be cumbersome to do even once and (B) needs to be re-done every time you make a change anywhere. Fortunately, there is a better way!\n\n`ggplot2` supports adding theme elements to an object that can then be reused as needed elsewhere. This is the same theory behind wrapping repeated operations into custom functions.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\n# Define core theme elements\ntheme_synthesis <- theme(legend.position = \"none\",\n panel.background = element_blank(),\n axis.line = element_line(color = \"black\"),\n axis.text = element_text(size = 13)) # <1>\n\n# Create a graph\nggplot(pie_crab, aes(y = water_temp, x = air_temp, color = size, size = size)) +\n geom_point() +\n theme_synthesis +\n theme(legend.position = \"right\") # <2>\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/std-theme-1.png){fig-align='center' width=864}\n:::\n:::\n\n1. This theme element controls the text on the tick marks. `axis.title` controls the text in the _labels_ of the axes\n2. As a bonus, subsequent uses of `theme()` will replace defaults defined in your earlier theme object. So, you can design a set of theme elements that are _usually_ appropriate and then easily change just some of them as needed\n\n:::{.callout-note icon=\"false\"}\n#### Activity: Graph Creation (P3)\n\nIn a script, attempt the following:\n\n- Remove all theme edits from the graph you made in the preceding activity and assign them to a separate object\n - Then add that object to your graph\n- Make a second (different) graph and add your consolidated theme object to that graph as well\n\n:::\n\n## Multi-Panel Graphs\n\nIt is sometimes the case that you want to make a single graph file that has multiple panels. For many of us, we might default to creating the separate graphs that we want, exporting them, and then using software like Microsoft PowerPoint to stitch those panels into the single image we had in mind from the start. However, as all of us who have used this method know, this is hugely cumbersome when your advisor/committee/reviewers ask for edits and you now have to redo all of the manual work behind your multi-panel graph. \n\nFortunately, there are two nice entirely scripted alternatives that you might consider: **Faceted graphs** and **Plot grids**. See below for more information on both.\n\n:::{.panel-tabset}\n### Facets\n\nIn a faceted graph, every panel of the graph has the same aesthetics. These are often used when you want to show the relationship between two (or more) variables but separated by some other variable. In synthesis work, you might show the relationship between your core response and explanatory variables but facet by the original study. This would leave you with one panel per study where each would show the relationship only at that particular study.\n\nLet's check out an example.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\nggplot(pie_crab, aes(x = date, y = size, color = site))+\n geom_point(size = 2) +\n facet_wrap(. ~ site) + # <1>\n theme_bw() +\n theme(legend.position = \"none\") # <2>\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/facet-1-1.png){fig-align='center' width=576}\n:::\n:::\n\n1. This is a `ggplot2` function that assumes you want panels laid out in a regular grid. There are other `facet_...` alternatives that let you specify row versus column arrangement. You could also facet by multiple variables by putting something to the left of the tilde\n2. We can remove the legend because the site names are in the facet titles in the gray boxes\n\n### Plot Grids\n\nIn a plot grid, each panel is completely independent of all others. These are often used in publications where you want to highlight several _different_ relationships that have some thematic connection. In synthesis work, your hypotheses may be more complicated than in primary research and such a plot grid would then be necessary to put all visual evidence for a hypothesis in the same location. On a practical note, plot grids are also a common way of circumventing figure number limits enforced by journals.\n\nLet's check out an example that relies on the `cowplot` library.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\n# Load a needed library\nlibrary(cowplot)\n\n# Create the first graph\ncrab_p1 <- ggplot(pie_crab, aes(x = site, y = size, fill = site)) + # <1>\n geom_violin() +\n coord_flip() + # <2>\n theme_bw() +\n theme(legend.position = \"none\")\n\n# Create the second\ncrab_p2 <- ggplot(pie_crab, aes(x = air_temp, y = water_temp)) +\n geom_errorbar(aes(ymax = water_temp + water_temp_sd, ymin = water_temp - water_temp_sd),\n width = 0.1) +\n geom_errorbarh(aes(xmax = air_temp + air_temp_sd, xmin = air_temp - air_temp_sd), # <3>\n width = 0.1) +\n geom_point(aes(fill = site), pch = 23, size = 3) +\n theme_bw()\n\n# Assemble into a plot grid\ncowplot::plot_grid(crab_p1, crab_p2, labels = \"AUTO\", nrow = 1) # <4>\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/grid-1-1.png){fig-align='center' width=864}\n:::\n:::\n\n1. Note that we're assigning these graphs to objects!\n2. This is a handy function for flipping X and Y axes without re-mapping the aesthetics\n3. This geometry is responsible for _horizontal_ error bars (note the \"h\" at the end of the function name)\n4. The `labels = \"AUTO\"` argument means that each panel of the plot grid gets the next sequential capital letter. You could also substitute that for a vector with labels of your choosing\n:::\n\n:::{.callout-note icon=\"false\"}\n#### Activity: Graph Creation (P4)\n\nIn a script, attempt the following:\n\n- Assemble the two graphs you made in the preceding two activities into the appropriate type of multi-panel graph\n\n:::\n\n## Accessibility Considerations\n\nAfter you've made the graphs you need, it is good practice to revisit them with to ensure that they are as accessible as possible. You can of course also do this during the graph construction process but it is sometimes less onerous to tackle as a penultimate step in the figure creation process. There are many facets to accessibility and we've tried to cover just a few of them below.\n\n### Color Choice\n\nOne of the more well-known facets of accessibility in data visualization is choosing colors that are \"colorblind safe\". Such palettes still create distinctive colors for those with various forms of color blindness (e.g., deuteranomoly, protanomaly, etc.). The classic red-green heatmap for instance is very colorblind unsafe in that people with some forms of colorblindness cannot distinguish between those colors (hence the rise of the yellow-blue heatmap in recent years). Unforunately, the `ggplot2` default rainbow palette--while nice for exploratory purposes--_is not_ colorlbind sfae.\n\nSome websites (such as [colorbewer2.org](https://colorbrewer2.org/#type=sequential&scheme=YlGnBu&n=9)) include a simple checkbox for colorblindness safety which automatically limits the listed options to those that are colorblind safe. Alternately, you could use a browser plug-in (such as [Let's get color blind](https://chromewebstore.google.com/detail/lets-get-color-blind/bkdgdianpkfahpkmphgehigalpighjck) on Google Chrome) to simulate colorblindness on a particular page.\n\nOne extreme approach you could take is to dodge this issue entirely and format your graphs such that color either isn't used at all or only conveys information that is also conveyed in another graph aesthetic. We don't necessarily recommend this as color--when the palette is chosen correctly--can be a really nice way of making information-dense graphs more informative and easily-navigable by viewers.\n\n### Multiple Modalities\n\nRelated to the color conversation is the value of mapping multiple aesthetics to the same variable. By presenting information in multiple ways--even if that seems redundant--you enable a wider audience to gain an intuitive sense of what you're trying to display.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\nggplot(data = pie_crab, mapping = aes(x = latitude, y = size, \n fill = site, shape = site)) + # <1>\n geom_jitter(size = 2, width = 0.1, alpha = 0.6) + \n scale_shape_manual(values = c(21:25, 21:25, 21:23)) + # <2>\n theme_bw() +\n theme(legend.title = element_blank())\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/multi-modal-1.png){fig-align='center' width=864}\n:::\n:::\n\n1. In this graph we're mapping both the fill and shape aesthetics to site\n2. This is a little cumbersome but there are only five 'fill-able' shapes in R so we need to reuse some of them to have a unique one for each site. Using fill-able shapes is nice because you get a crisp black border around each point. See `?pch` for all available shapes\n\nIn the above graph, even though the rainbow palette is not ideal for reasons mentioned earlier, it is now much easier to tell the difference between sites with similar colors. For instance, \"NB\", \"NIB\", and \"PIE\" are all shades of light blue/teal. Now that they have unique shapes it is dramatically easier to look at the graph and identify which points correspond to which site.\n\n\n:::{.callout-warning icon=\"false\"}\n#### Discussion: Graph Accessibility\n\nWith a group discuss (some of) the following questions:\n\n- What are other facets of accessibility that you think are important to consider when making data visualizations?\n- What changes do you make to your graphs to increase accessibility?\n - What changes _could_ you make going forward?\n\n:::\n\n\n### Presentation vs. Publication\n\nOne final element of accessibility to consider is the difference between a '_presentation_-quality' graph and a '_publication_-quality' one. While it may be tempting to create a single version of a given graph and use it in both contexts that is likely to be less effective in helping you to get your point across than making small tweaks to two separate versions of what is otherwise the same graph.\n\n:::{.panel-tabset}\n### Presentation-Focused\n\n**Do:**\n\n- Increase size of text/points **greatly**\n - If possible, sit in the back row of the room where you'll present and look at your graphs from there\n- _Consider_ adding graph elements that highlight certain graph regions\n- Present summarized data (increases focus on big-picture trends and avoids discussion of minutiae)\n- Map multiple aesthetics to the same variables\n\n**Don't:**\n\n- Use technical language / jargon\n- Include _unnecessary_ background elements\n- Use multi-panel graphs (either faceted or plot grid)\n - If you have multiple graph panels, put each on its own slide!\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\nggplot(crab_summary, aes(x = latitude, y = mean, \n shape = reorder(site, latitude), # <1>\n fill = reorder(site, latitude))) +\n geom_vline(xintercept = 36.5, color = \"black\", linetype = 1) +\n geom_vline(xintercept = 41.5, color = \"black\", linetype = 2) + # <2>\n geom_errorbar(mapping = aes(ymax = mean + std_error, ymin = mean - std_error),\n width = 0.2) +\n geom_point(size = 4) + \n scale_shape_manual(values = c(21:25, 21:25, 21:23)) +\n labs(x = \"Latitude\", y = \"Mean Crab Size (mm)\") + # <3>\n theme(legend.title = element_blank(),\n axis.line = element_line(color = \"black\"),\n panel.background = element_blank(),\n axis.title = element_text(size = 17),\n axis.text = element_text(size = 15))\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/talk-graph-1.png){fig-align='center' width=864}\n:::\n:::\n\n1. We can use the `reorder` function to make the order of sites in the legend (from top to bottom) match the order of sites in the graph (from left to right)\n2. Adding vertical lines at particular parts in the graph can make comparisons within the same graph easier\n3. `labs` lets us customize the title and label text of a graph\n\n### Publication-Focused\n\n**Do:**\n\n- Increase size of text/points **slightly**\n - You want to be legible but you can more safely assume that many readers will be able to increase the zoom of their browser window if needed\n- Present un-summarized data (with or without summarized points included)\n - Many reviewers will want to get a sense for the \"real\" data so you should include unsummarized values wherever possible\n- Use multi-panel graphs\n - If multiple graphs \"tell a story\" together, then they should be included in the same file!\n- Map multiple aesthetics to the same variables\n- If publishing in a journal available in print, check to make sure your graph still makes sense in grayscale\n - There are nice browser plug-ins (like [Grayscale the Web](https://chromewebstore.google.com/detail/grayscale-the-web-save-si/mblmpdpfppogibmoobibfannckeeleag) for Google Chrome) for this too\n\n**Don't:**\n\n- Include _unnecessary_ background elements\n- Add graph elements that highlight certain graph regions\n - You can--and should--lean more heavily on the text of your publication to discuss particular areas of a graph\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\nggplot() +\n geom_point(pie_crab, mapping = aes(x = latitude, y = size,\n color = reorder(site, latitude)),\n pch = 19, size = 1, alpha = 0.3) +\n geom_errorbar(crab_summary, mapping = aes(x = latitude, y = mean, \n ymax = mean + std_error, \n ymin = mean - std_error),\n width = 0.2) +\n geom_point(crab_summary, mapping = aes(x = latitude, y = mean, \n shape = reorder(site, latitude),\n fill = reorder(site, latitude)),\n size = 4) +\n scale_shape_manual(values = c(21:25, 21:25, 21:23)) +\n labs(x = \"Latitude\", y = \"Mean Crab Carapace Width (mm)\") + # <1>\n theme(legend.title = element_blank(),\n axis.line = element_line(color = \"black\"),\n panel.background = element_blank(),\n axis.title = element_text(size = 15),\n axis.text = element_text(size = 13))\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/pub-graph-1.png){fig-align='center' width=864}\n:::\n:::\n\n1. Here we are using a reasonable amount of technical language\n\n:::\n\n## Ordination\n\nIf you are working with multivariate data (i.e., data where multiple columns are all response variables collectively) you may find ordination helpful. Ordination is the general term for many types of multivariate visualization but typically is used to refer to visualizing a distance or dissimiliarity measure of the data. Such measures collapse all of those columns of response variables into fewer (typically two) index values that are easier to visualize.\n\nThis is a common approach particularly in answering questions in community ecology or considering a suite of traits (e.g., life history, landscape, etc.) together. While the math behind reducing the dimensionality of your data is interesting, this module is focused on only the visualization facet of ordination so we'll avoid deeper discussion of the internal mechanics that underpin ordination.\n\nIn order to demonstrate two types of ordination we'll use a lichen community composition dataset included in the `vegan` package. However, ordination approaches are most often used on data with multiple groups so we'll need to make a simulated grouping column to divide the lichen community data.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load library\nlibrary(vegan)\n\n# Grab data\nutils::data(\"varespec\", package = \"vegan\")\n\n# Create a faux group column\ntreatment <- c(rep.int(\"Treatment A\", nrow(varespec) / 2),\n rep.int(\"Treatment B\", nrow(varespec) / 2))\n\n# Combine into one dataframe\nlichen_df <- cbind(treatment, varespec)\n\n# Check structure of first few columns\nstr(lichen_df[1:5])\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n'data.frame':\t24 obs. of 5 variables:\n $ treatment: chr \"Treatment A\" \"Treatment A\" \"Treatment A\" \"Treatment A\" ...\n $ Callvulg : num 0.55 0.67 0.1 0 0 ...\n $ Empenigr : num 11.13 0.17 1.55 15.13 12.68 ...\n $ Rhodtome : num 0 0 0 2.42 0 0 1.55 0 0.35 0.07 ...\n $ Vaccmyrt : num 0 0.35 0 5.92 0 ...\n```\n\n\n:::\n:::\n\n\n:::{.panel-tabset}\n\n### Metric Ordination\n\nMetric ordinations are typically used when you are concerned with retaining quantitative differences among particular points, _even after you've collapsed many response variables into just one or two_. For example, this is a common approach if you have a table of traits and want to compare the whole set of traits among groups while still being able to interpret the effect of a particular effect on the whole.\n\nTwo of the more common methods for metric ordination are Principal Components Analysis (PCA), and Principal Coordinates Analysis (PCoA / \"metric multidimensional scaling\"). The primary difference is that PCA works on the data directly while PCoA works on a distance matrix of the data. We'll use PCoA in this example because it is closer analog to the non-metric ordination discussed in the other tab. **If the holistic difference among groups is of interest,** (rather than metric point-to-point comparisons), **consider a non-metric ordination approach.**\n\nIn order to perform a PCoA ordination we first need to get a distance matrix of our response variables and then we can actually do the PCoA step. The distance matrix can be calculated with the `vegdist` function from the `vegan` package and the `pcoa` function in the `ape` package can do the actual PCoA.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load needed libraries\nlibrary(vegan); library(ape)\n\n# Get distance matrix\nlichen_dist <- vegan::vegdist(x = lichen_df[-1], method = \"kulczynski\") # <1>\n\n# Do PCoA\npcoa_points <- ape::pcoa(D = lichen_dist)\n```\n:::\n\n1. The `method` argument requires a distance/dissimilarity measure. Note that **if you use a non-metric measure** (e.g., Bray Curtis, etc.) **you lose many of the advantages conferred by using a metric ordination approach**.\n\nWith that in hand, we can make our ordination! While you could make this step-by-step on your own, we'll use the `ordination` function from the `supportR` package for convenience. This function automatically uses colorblind safe colors for up to 10 groups and has some useful base plot defaults (as well as including ellipses around the standard deviation of the centorid of all groups).\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\n# Load the library\nlibrary(supportR)\n\n# Make the ordination\nsupportR::ordination(mod = pcoa_points, grps = lichen_df$treatment, \n x = \"topleft\", legend = c(\"A\", \"B\")) #<1>\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/pcoa-ord-1.png){fig-align='center' width=672}\n:::\n:::\n\n1. This function allows several base plot arguments to be supplied to alter non-critical plot elements (e.g., legend position, point size, etc.)\n\nThe percentages included in parentheses on either axis label are the percent of the total variation in the data explained by each axis on its own. Use this information in combination with what the graph looks like to determine how different the groups truly are.\n\n\n### Non-Metric Ordination\n\nNon-metric ordinations are typically used when you care more about the relative differences among groups rather than specific measurements between particular points. For instance, you may want to assess whether the composition of insect communities differs between two experimental treatments. In such a case, your hypothesis likely depends more on the holistic difference between the treatments rather than some quantitative difference on one of the axes.\n\nThe most common non-metric ordination type is called Nonmetric Multidimensional Scaling (NMS / NMDS). This approach prioritizes making groups that are \"more different\" further apart than those that are less different. However, NMS uses a dissimilarity matrix which means that the _distance_ between any two specific points cannot be interpreted meaningfully. It _is_ appropriate though to interpret which cloud of points is closer to/further from another in aggregate. **If specific distances among points are of interest, consider a metric ordination approach.**\n\nIn order to perform an NMS ordination we'll first need to calculate a dissimilarity matrix for our response data. The vegan function `metaMDS` is useful for this. This function has many arguments but the most fundamental are the following:\n\n- `comm` = the dataframe of response variables (minus any non-numeric / grouping columns)\n- `distance` = the distance/dissimilarity metric to use\n - Note that there is no benefit to using a metric distance because when we make the ordination it will become non-metric\n- `k` = number of axes to decompose to -- typically two so the graph can be simple\n- `try` = number of attempts at minimizing \"stress\"\n - Stress is how NMS evaluates how good of a job it did at representing the true differences among groups (lower stress is better)\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load needed libraries\nlibrary(vegan)\n\n# Get dissimilarity matrix\ndissim_mat <- vegan::metaMDS(comm = lichen_df[-1], distance = \"bray\", k = 2,\n autotransform = F, expand = F, try = 50)\n```\n:::\n\n\nWith that in hand, we can make our ordination! While you could make this step-by-step on your own, we'll use the `ordination` function from the `supportR` package for convenience. This function automatically uses colorblind safe colors for up to 10 groups and has some useful base plot defaults (as well as including ellipses around the standard deviation of the centorid of all groups).\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\n# Load the library\nlibrary(supportR)\n\n# Make the ordination\nsupportR::ordination(mod = dissim_mat, grps = lichen_df$treatment, \n x = \"bottomright\", legend = c(\"A\", \"B\")) #<1>\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/nms-ord-1.png){fig-align='center' width=672}\n:::\n:::\n\n1. This function allows several base plot arguments to be supplied to alter non-critical plot elements (e.g., legend position, point size, etc.)\n\nIf the stress is less than 0.15 it is generally considered a good representation of the data. We can see that the ellipses do not overlap which indicates that the community composition of our two groups does seem to differ. We'd need to do real multivariate analysis if we wanted a _p_-value or AIC score to support that but as a visual tool this is still useful.\n\n:::\n\n## Maps\n\nYou may find it valuable to create a map as an additional way of visualizing data. Many synthesis groups do this--particularly when there is a strong spatial component to the research questions and/or hypotheses.\n\nCheck out the [bonus spatial data module](https://lter.github.io/ssecr/mod_spatial.html) for more information on map-making if this is of interest!\n\n## Additional Resources\n\n### Papers & Documents\n\n- NCEAS [Colorblind Safe Color Schemes](https://www.nceas.ucsb.edu/sites/default/files/2022-06/Colorblind%20Safe%20Color%20Schemes.pdf) reference document\n\n### Workshops & Courses\n\n- NCEAS Scientific Computing team's Coding in the Tidyverse workshop [`ggplot2` module](https://nceas.github.io/scicomp-workshop-tidyverse/visualize.html)\n- The Carpentries' Data Analysis and Visualization in R for Ecologists [`ggplot2` episode](https://datacarpentry.org/R-ecology-lesson/04-visualization-ggplot2.html)\n\n\n### Websites\n\n- \n", + "markdown": "---\ntitle: \"Data Visualization & Exploration\"\ncode-annotations: hover\n---\n\n\n## Overview\n\nData visualization is a fundamental part of working with data. Visualization can be only used in the final stages of a project to make figures for publication but it can also be hugely valuable for quality control and hypothesis development processes. This module focuses on the fundamentals of graph creation in an effort to empower you to apply those methods in the various contexts where you might find visualization to be helpful.\n\n## Learning Objectives\n\nAfter completing this module you will be able to: \n\n- Define fundamental `ggplot2` vocabulary\n- Identify appropriate graph types for given data type/distribution\n- Discuss differences between presentation- and publication-quality graphs\n- Explain how your graphs can be made more accessible\n\n## Needed Packages\n\nIf you'd like to follow along with the code chunks included throughout this module, you'll need to install the following packages:\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Note that these lines only need to be run once per computer\n## So you can skip this step if you've installed these before\ninstall.packages(\"tidyverse\")\ninstall.packages(\"lterdatasampler\")\ninstall.packages(\"supportR\")\ninstall.packages(\"cowplot\")\ninstall.packages(\"vegan\")\ninstall.packages(\"ape\")\n```\n:::\n\n\nWe'll go ahead and load some of these libraries as well to be able to better demonstrate these concepts.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load needed libraries\nlibrary(tidyverse)\n```\n:::\n\n\n## Graphing with `ggplot2`\n\n### `ggplot2` Fundamentals\n\nYou may already be familiar with the `ggplot2` package in R but if you are not, it is a popular graphing library based on [The Grammar of Graphics](https://bookshop.org/p/books/the-grammar-of-graphics-leland-wilkinson/1518348?ean=9780387245447). Every ggplot is composed of four elements:\n\n1. A 'core' `ggplot` function call\n2. Aesthetics\n3. Geometries\n4. Theme\n\nNote that the theme component may be implicit in some graphs because there is a suite of default theme elements that applies unless otherwise specified. \n\nThis module will use example data to demonstrate these tools but as we work through these topics you should feel free to substitute a dataset of your choosing! If you don't have one in mind, you can use the example dataset shown in the code chunks throughout this module. This dataset comes from the [`lterdatasampler` R package](https://lter.github.io/lterdatasampler/) and the data are about fiddler crabs (_Minuca pugnax_) at the [Plum Island Ecosystems (PIE) LTER](https://pie-lter.ecosystems.mbl.edu/welcome-plum-island-ecosystems-lter) site.\n\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load the lterdatasampler package\nlibrary(lterdatasampler)\n\n# Load the fiddler crab dataset\ndata(pie_crab)\n\n# Check its structure\nstr(pie_crab)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\ntibble [392 × 9] (S3: tbl_df/tbl/data.frame)\n $ date : Date[1:392], format: \"2016-07-24\" \"2016-07-24\" ...\n $ latitude : num [1:392] 30 30 30 30 30 30 30 30 30 30 ...\n $ site : chr [1:392] \"GTM\" \"GTM\" \"GTM\" \"GTM\" ...\n $ size : num [1:392] 12.4 14.2 14.5 12.9 12.4 ...\n $ air_temp : num [1:392] 21.8 21.8 21.8 21.8 21.8 ...\n $ air_temp_sd : num [1:392] 6.39 6.39 6.39 6.39 6.39 ...\n $ water_temp : num [1:392] 24.5 24.5 24.5 24.5 24.5 ...\n $ water_temp_sd: num [1:392] 6.12 6.12 6.12 6.12 6.12 ...\n $ name : chr [1:392] \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" ...\n```\n\n\n:::\n:::\n\n\nWith a dataset in hand, let's make a scatterplot of crab size on the Y-axis with latitude on the X. We'll forgo doing anything to the theme elements at this point to focus on the other three elements.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\nggplot(data = pie_crab, mapping = aes(x = latitude, y = size, fill = site)) + # <1>\n geom_point(pch = 21, size = 2, alpha = 0.5) # <2>\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/gg-1-1.png){fig-align='center' width=864}\n:::\n:::\n\n1. We're defining both the data and the X/Y aesthetics in this top-level bit of the plot. Also, note that each line ends with a plus sign\n2. Because we defined the data and aesthetics in the `ggplot()` function call above, this geometry can assume those mappings without re-specificying\n\nWe can improve on this graph by tweaking theme elements to make it use fewer of the default settings.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\nggplot(data = pie_crab, mapping = aes(x = latitude, y = size, fill = site)) +\n geom_point(pch = 21, size = 2, alpha = 0.5) +\n theme(legend.title = element_blank(), # <1>\n panel.background = element_blank(),\n axis.line = element_line(color = \"black\"))\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/gg-2-1.png){fig-align='center' width=864}\n:::\n:::\n\n1. All theme elements require these `element_...` helper functions. `element_blank` removes theme elements but otherwise you'll need to use the helper function that corresponds to the type of theme element (e.g., `element_text` for theme elements affecting graph text)\n\n### Multiple Geometries\n\nWe can further modify `ggplot2` graphs by adding _multiple_ geometries if you find it valuable to do so. Note however that geometry order matters! Geometries added later will be \"in front of\" those added earlier. Also, adding too much data to a plot will begin to make it difficult for others to understand the central take-away of the graph so you may want to be careful about the level of information density in each graph. Let's add boxplots behind the points to characterize the distribution of points more quantitatively.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\nggplot(data = pie_crab, mapping = aes(x = latitude, y = size, fill = site)) +\n geom_boxplot(pch = 21) + # <1>\n geom_point(pch = 21, size = 2, alpha = 0.5) +\n theme(legend.title = element_blank(), \n panel.background = element_blank(),\n axis.line = element_line(color = \"black\"))\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/gg-3-1.png){fig-align='center' width=864}\n:::\n:::\n\n1. By putting the boxplot geometry first we ensure that it doesn't cover up the points that overlap with the 'box' part of each boxplot\n\n:::{.callout-note icon=\"false\"}\n#### Activity: Graph Creation (P1)\n\nIn a script, attempt the following with one of either yours or your group's datasets:\n\n- Make a graph using `ggplot2`\n - Include at least one geometry\n - Include at least one aesthetic (beyond X/Y axes)\n - Modify at least one theme element from the default\n\n:::\n\n### Multiple Data Objects\n\n`ggplot2` also supports adding more than one data object to the same graph! While this module doesn't cover map creation, maps are a common example of a graph with more than one data object. Another common use would be to include both the full dataset and some summarized facet of it in the same plot.\n\nLet's calculate some summary statistics of crab size to include that in our plot.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load the supportR library\nlibrary(supportR)\n```\n\n::: {.cell-output .cell-output-stderr}\n\n```\n\nAttaching package: 'supportR'\n```\n\n\n:::\n\n::: {.cell-output .cell-output-stderr}\n\n```\nThe following object is masked from 'package:dplyr':\n\n count\n```\n\n\n:::\n\n```{.r .cell-code}\n# Summarize crab size within latitude groups\ncrab_summary <- supportR::summary_table(data = pie_crab, groups = c(\"site\", \"latitude\"),\n response = \"size\", drop_na = TRUE)\n\n# Check the structure\nstr(crab_summary)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n'data.frame':\t13 obs. of 6 variables:\n $ site : chr \"BC\" \"CC\" \"CT\" \"DB\" ...\n $ latitude : num 42.2 41.9 41.3 39.1 30 39.6 41.6 33.3 42.7 34.7 ...\n $ mean : num 16.2 16.8 14.7 15.6 12.4 ...\n $ std_dev : num 4.81 2.05 2.36 2.12 1.8 2.72 2.29 2.42 2.3 2.34 ...\n $ sample_size: int 37 27 33 30 28 30 29 30 28 25 ...\n $ std_error : num 0.79 0.39 0.41 0.39 0.34 0.5 0.43 0.44 0.43 0.47 ...\n```\n\n\n:::\n:::\n\n\nWith this data object in-hand, we can make a graph that includes both this and the original, unsummarized crab data. To better focus on the 'multiple data objects' bit of this example we'll pare down on the actual graph code.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\nggplot() + # <1>\n geom_point(pie_crab, mapping = aes(x = latitude, y = size, fill = site),\n pch = 21, size = 2, alpha = 0.2) + \n geom_errorbar(crab_summary, mapping = aes(x = latitude, # <2>\n ymax = mean + std_error,\n ymin = mean - std_error),\n width = 0.2) +\n geom_point(crab_summary, mapping = aes(x = latitude, y = mean, fill = site),\n pch = 23, size = 3) + \n theme(legend.title = element_blank(),\n panel.background = element_blank(),\n axis.line = element_line(color = \"black\"))\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/gg-4-1.png){fig-align='center' width=864}\n:::\n:::\n\n1. If you want multiple data objects in the same `ggplot2` graph you need to leave this top level `ggplot()` call _empty!_ Otherwise you'll get weird errors with aesthetics later in the graph\n2. This geometry adds the error bars and it's important that we add it before the summarized data points themselves if we want the error bars to be 'behind' their respective points\n\n:::{.callout-note icon=\"false\"}\n#### Activity: Graph Creation (P2)\n\nIn a script, attempt the following:\n\n- Add a second data object to the graph you made in the preceding activity\n - _Hint:_ If your first graph is unsummarized, add a summarized version (or vice versa)\n\n:::\n\n## Streamlining Graph Aesthetics\n\nSynthesis projects often generate an entire network of inter-related papers. Ensuring that all graphs across papers from a given team have a similar \"feel\" is a nice way of implying a certain standard of robustness for all of your group's projects. However, copy/pasting the theme elements of your graphs can (A) be cumbersome to do even once and (B) needs to be re-done every time you make a change anywhere. Fortunately, there is a better way!\n\n`ggplot2` supports adding theme elements to an object that can then be reused as needed elsewhere. This is the same theory behind wrapping repeated operations into custom functions.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\n# Define core theme elements\ntheme_synthesis <- theme(legend.position = \"none\",\n panel.background = element_blank(),\n axis.line = element_line(color = \"black\"),\n axis.text = element_text(size = 13)) # <1>\n\n# Create a graph\nggplot(pie_crab, aes(y = water_temp, x = air_temp, color = size, size = size)) +\n geom_point() +\n theme_synthesis +\n theme(legend.position = \"right\") # <2>\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/std-theme-1.png){fig-align='center' width=864}\n:::\n:::\n\n1. This theme element controls the text on the tick marks. `axis.title` controls the text in the _labels_ of the axes\n2. As a bonus, subsequent uses of `theme()` will replace defaults defined in your earlier theme object. So, you can design a set of theme elements that are _usually_ appropriate and then easily change just some of them as needed\n\n:::{.callout-note icon=\"false\"}\n#### Activity: Graph Creation (P3)\n\nIn a script, attempt the following:\n\n- Remove all theme edits from the graph you made in the preceding activity and assign them to a separate object\n - Then add that object to your graph\n- Make a second (different) graph and add your consolidated theme object to that graph as well\n\n:::\n\n## Multi-Panel Graphs\n\nIt is sometimes the case that you want to make a single graph file that has multiple panels. For many of us, we might default to creating the separate graphs that we want, exporting them, and then using software like Microsoft PowerPoint to stitch those panels into the single image we had in mind from the start. However, as all of us who have used this method know, this is hugely cumbersome when your advisor/committee/reviewers ask for edits and you now have to redo all of the manual work behind your multi-panel graph. \n\nFortunately, there are two nice entirely scripted alternatives that you might consider: **Faceted graphs** and **Plot grids**. See below for more information on both.\n\n:::{.panel-tabset}\n### Facets\n\nIn a faceted graph, every panel of the graph has the same aesthetics. These are often used when you want to show the relationship between two (or more) variables but separated by some other variable. In synthesis work, you might show the relationship between your core response and explanatory variables but facet by the original study. This would leave you with one panel per study where each would show the relationship only at that particular study.\n\nLet's check out an example.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\nggplot(pie_crab, aes(x = date, y = size, color = site))+\n geom_point(size = 2) +\n facet_wrap(. ~ site) + # <1>\n theme_bw() +\n theme(legend.position = \"none\") # <2>\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/facet-1-1.png){fig-align='center' width=576}\n:::\n:::\n\n1. This is a `ggplot2` function that assumes you want panels laid out in a regular grid. There are other `facet_...` alternatives that let you specify row versus column arrangement. You could also facet by multiple variables by putting something to the left of the tilde\n2. We can remove the legend because the site names are in the facet titles in the gray boxes\n\n### Plot Grids\n\nIn a plot grid, each panel is completely independent of all others. These are often used in publications where you want to highlight several _different_ relationships that have some thematic connection. In synthesis work, your hypotheses may be more complicated than in primary research and such a plot grid would then be necessary to put all visual evidence for a hypothesis in the same location. On a practical note, plot grids are also a common way of circumventing figure number limits enforced by journals.\n\nLet's check out an example that relies on the `cowplot` library.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\n# Load a needed library\nlibrary(cowplot)\n\n# Create the first graph\ncrab_p1 <- ggplot(pie_crab, aes(x = site, y = size, fill = site)) + # <1>\n geom_violin() +\n coord_flip() + # <2>\n theme_bw() +\n theme(legend.position = \"none\")\n\n# Create the second\ncrab_p2 <- ggplot(pie_crab, aes(x = air_temp, y = water_temp)) +\n geom_errorbar(aes(ymax = water_temp + water_temp_sd, ymin = water_temp - water_temp_sd),\n width = 0.1) +\n geom_errorbarh(aes(xmax = air_temp + air_temp_sd, xmin = air_temp - air_temp_sd), # <3>\n width = 0.1) +\n geom_point(aes(fill = site), pch = 23, size = 3) +\n theme_bw()\n\n# Assemble into a plot grid\ncowplot::plot_grid(crab_p1, crab_p2, labels = \"AUTO\", nrow = 1) # <4>\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/grid-1-1.png){fig-align='center' width=864}\n:::\n:::\n\n1. Note that we're assigning these graphs to objects!\n2. This is a handy function for flipping X and Y axes without re-mapping the aesthetics\n3. This geometry is responsible for _horizontal_ error bars (note the \"h\" at the end of the function name)\n4. The `labels = \"AUTO\"` argument means that each panel of the plot grid gets the next sequential capital letter. You could also substitute that for a vector with labels of your choosing\n:::\n\n:::{.callout-note icon=\"false\"}\n#### Activity: Graph Creation (P4)\n\nIn a script, attempt the following:\n\n- Assemble the two graphs you made in the preceding two activities into the appropriate type of multi-panel graph\n\n:::\n\n## Accessibility Considerations\n\nAfter you've made the graphs you need, it is good practice to revisit them with to ensure that they are as accessible as possible. You can of course also do this during the graph construction process but it is sometimes less onerous to tackle as a penultimate step in the figure creation process. There are many facets to accessibility and we've tried to cover just a few of them below.\n\n### Color Choice\n\nOne of the more well-known facets of accessibility in data visualization is choosing colors that are \"colorblind safe\". Such palettes still create distinctive colors for those with various forms of color blindness (e.g., deuteranomoly, protanomaly, etc.). The classic red-green heatmap for instance is very colorblind unsafe in that people with some forms of colorblindness cannot distinguish between those colors (hence the rise of the yellow-blue heatmap in recent years). Unforunately, the `ggplot2` default rainbow palette--while nice for exploratory purposes--_is not_ colorlbind sfae.\n\nSome websites (such as [colorbewer2.org](https://colorbrewer2.org/#type=sequential&scheme=YlGnBu&n=9)) include a simple checkbox for colorblindness safety which automatically limits the listed options to those that are colorblind safe. Alternately, you could use a browser plug-in (such as [Let's get color blind](https://chromewebstore.google.com/detail/lets-get-color-blind/bkdgdianpkfahpkmphgehigalpighjck) on Google Chrome) to simulate colorblindness on a particular page.\n\nOne extreme approach you could take is to dodge this issue entirely and format your graphs such that color either isn't used at all or only conveys information that is also conveyed in another graph aesthetic. We don't necessarily recommend this as color--when the palette is chosen correctly--can be a really nice way of making information-dense graphs more informative and easily-navigable by viewers.\n\n### Multiple Modalities\n\nRelated to the color conversation is the value of mapping multiple aesthetics to the same variable. By presenting information in multiple ways--even if that seems redundant--you enable a wider audience to gain an intuitive sense of what you're trying to display.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\nggplot(data = pie_crab, mapping = aes(x = latitude, y = size, \n fill = site, shape = site)) + # <1>\n geom_jitter(size = 2, width = 0.1, alpha = 0.6) + \n scale_shape_manual(values = c(21:25, 21:25, 21:23)) + # <2>\n theme_bw() +\n theme(legend.title = element_blank())\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/multi-modal-1.png){fig-align='center' width=864}\n:::\n:::\n\n1. In this graph we're mapping both the fill and shape aesthetics to site\n2. This is a little cumbersome but there are only five 'fill-able' shapes in R so we need to reuse some of them to have a unique one for each site. Using fill-able shapes is nice because you get a crisp black border around each point. See `?pch` for all available shapes\n\nIn the above graph, even though the rainbow palette is not ideal for reasons mentioned earlier, it is now much easier to tell the difference between sites with similar colors. For instance, \"NB\", \"NIB\", and \"PIE\" are all shades of light blue/teal. Now that they have unique shapes it is dramatically easier to look at the graph and identify which points correspond to which site.\n\n\n:::{.callout-warning icon=\"false\"}\n#### Discussion: Graph Accessibility\n\nWith a group discuss (some of) the following questions:\n\n- What are other facets of accessibility that you think are important to consider when making data visualizations?\n- What changes do you make to your graphs to increase accessibility?\n - What changes _could_ you make going forward?\n\n:::\n\n\n### Presentation vs. Publication\n\nOne final element of accessibility to consider is the difference between a '_presentation_-quality' graph and a '_publication_-quality' one. While it may be tempting to create a single version of a given graph and use it in both contexts that is likely to be less effective in helping you to get your point across than making small tweaks to two separate versions of what is otherwise the same graph.\n\n:::{.panel-tabset}\n### Presentation-Focused\n\n**Do:**\n\n- Increase size of text/points **greatly**\n - If possible, sit in the back row of the room where you'll present and look at your graphs from there\n- _Consider_ adding graph elements that highlight certain graph regions\n- Present summarized data (increases focus on big-picture trends and avoids discussion of minutiae)\n- Map multiple aesthetics to the same variables\n\n**Don't:**\n\n- Use technical language / jargon\n- Include _unnecessary_ background elements\n- Use multi-panel graphs (either faceted or plot grid)\n - If you have multiple graph panels, put each on its own slide!\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\nggplot(crab_summary, aes(x = latitude, y = mean, \n shape = reorder(site, latitude), # <1>\n fill = reorder(site, latitude))) +\n geom_vline(xintercept = 36.5, color = \"black\", linetype = 1) +\n geom_vline(xintercept = 41.5, color = \"black\", linetype = 2) + # <2>\n geom_errorbar(mapping = aes(ymax = mean + std_error, ymin = mean - std_error),\n width = 0.2) +\n geom_point(size = 4) + \n scale_shape_manual(values = c(21:25, 21:25, 21:23)) +\n labs(x = \"Latitude\", y = \"Mean Crab Size (mm)\") + # <3>\n theme(legend.title = element_blank(),\n axis.line = element_line(color = \"black\"),\n panel.background = element_blank(),\n axis.title = element_text(size = 17),\n axis.text = element_text(size = 15))\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/talk-graph-1.png){fig-align='center' width=864}\n:::\n:::\n\n1. We can use the `reorder` function to make the order of sites in the legend (from top to bottom) match the order of sites in the graph (from left to right)\n2. Adding vertical lines at particular parts in the graph can make comparisons within the same graph easier\n3. `labs` lets us customize the title and label text of a graph\n\n### Publication-Focused\n\n**Do:**\n\n- Increase size of text/points **slightly**\n - You want to be legible but you can more safely assume that many readers will be able to increase the zoom of their browser window if needed\n- Present un-summarized data (with or without summarized points included)\n - Many reviewers will want to get a sense for the \"real\" data so you should include unsummarized values wherever possible\n- Use multi-panel graphs\n - If multiple graphs \"tell a story\" together, then they should be included in the same file!\n- Map multiple aesthetics to the same variables\n- If publishing in a journal available in print, check to make sure your graph still makes sense in grayscale\n - There are nice browser plug-ins (like [Grayscale the Web](https://chromewebstore.google.com/detail/grayscale-the-web-save-si/mblmpdpfppogibmoobibfannckeeleag) for Google Chrome) for this too\n\n**Don't:**\n\n- Include _unnecessary_ background elements\n- Add graph elements that highlight certain graph regions\n - You can--and should--lean more heavily on the text of your publication to discuss particular areas of a graph\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\nggplot() +\n geom_point(pie_crab, mapping = aes(x = latitude, y = size,\n color = reorder(site, latitude)),\n pch = 19, size = 1, alpha = 0.3) +\n geom_errorbar(crab_summary, mapping = aes(x = latitude, y = mean, \n ymax = mean + std_error, \n ymin = mean - std_error),\n width = 0.2) +\n geom_point(crab_summary, mapping = aes(x = latitude, y = mean, \n shape = reorder(site, latitude),\n fill = reorder(site, latitude)),\n size = 4) +\n scale_shape_manual(values = c(21:25, 21:25, 21:23)) +\n labs(x = \"Latitude\", y = \"Mean Crab Carapace Width (mm)\") + # <1>\n theme(legend.title = element_blank(),\n axis.line = element_line(color = \"black\"),\n panel.background = element_blank(),\n axis.title = element_text(size = 15),\n axis.text = element_text(size = 13))\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/pub-graph-1.png){fig-align='center' width=864}\n:::\n:::\n\n1. Here we are using a reasonable amount of technical language\n\n:::\n\n## Ordination\n\nIf you are working with multivariate data (i.e., data where multiple columns are all response variables collectively) you may find ordination helpful. Ordination is the general term for many types of multivariate visualization but typically is used to refer to visualizing a distance or dissimiliarity measure of the data. Such measures collapse all of those columns of response variables into fewer (typically two) index values that are easier to visualize.\n\nThis is a common approach particularly in answering questions in community ecology or considering a suite of traits (e.g., life history, landscape, etc.) together. While the math behind reducing the dimensionality of your data is interesting, this module is focused on only the visualization facet of ordination so we'll avoid deeper discussion of the internal mechanics that underpin ordination.\n\nIn order to demonstrate two types of ordination we'll use a lichen community composition dataset included in the `vegan` package. However, ordination approaches are most often used on data with multiple groups so we'll need to make a simulated grouping column to divide the lichen community data.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load library\nlibrary(vegan)\n\n# Grab data\nutils::data(\"varespec\", package = \"vegan\")\n\n# Create a faux group column\ntreatment <- c(rep.int(\"Treatment A\", nrow(varespec) / 2),\n rep.int(\"Treatment B\", nrow(varespec) / 2))\n\n# Combine into one dataframe\nlichen_df <- cbind(treatment, varespec)\n\n# Check structure of first few columns\nstr(lichen_df[1:5])\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n'data.frame':\t24 obs. of 5 variables:\n $ treatment: chr \"Treatment A\" \"Treatment A\" \"Treatment A\" \"Treatment A\" ...\n $ Callvulg : num 0.55 0.67 0.1 0 0 ...\n $ Empenigr : num 11.13 0.17 1.55 15.13 12.68 ...\n $ Rhodtome : num 0 0 0 2.42 0 0 1.55 0 0.35 0.07 ...\n $ Vaccmyrt : num 0 0.35 0 5.92 0 ...\n```\n\n\n:::\n:::\n\n\n:::{.panel-tabset}\n\n### Metric Ordination\n\nMetric ordinations are typically used when you are concerned with retaining quantitative differences among particular points, _even after you've collapsed many response variables into just one or two_. For example, this is a common approach if you have a table of traits and want to compare the whole set of traits among groups while still being able to interpret the effect of a particular effect on the whole.\n\nTwo of the more common methods for metric ordination are Principal Components Analysis (PCA), and Principal Coordinates Analysis (PCoA / \"metric multidimensional scaling\"). The primary difference is that PCA works on the data directly while PCoA works on a distance matrix of the data. We'll use PCoA in this example because it is closer analog to the non-metric ordination discussed in the other tab. **If the holistic difference among groups is of interest,** (rather than metric point-to-point comparisons), **consider a non-metric ordination approach.**\n\nIn order to perform a PCoA ordination we first need to get a distance matrix of our response variables and then we can actually do the PCoA step. The distance matrix can be calculated with the `vegdist` function from the `vegan` package and the `pcoa` function in the `ape` package can do the actual PCoA.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load needed libraries\nlibrary(vegan); library(ape)\n\n# Get distance matrix\nlichen_dist <- vegan::vegdist(x = lichen_df[-1], method = \"kulczynski\") # <1>\n\n# Do PCoA\npcoa_points <- ape::pcoa(D = lichen_dist)\n```\n:::\n\n1. The `method` argument requires a distance/dissimilarity measure. Note that **if you use a non-metric measure** (e.g., Bray Curtis, etc.) **you lose many of the advantages conferred by using a metric ordination approach**.\n\nWith that in hand, we can make our ordination! While you could make this step-by-step on your own, we'll use the `ordination` function from the `supportR` package for convenience. This function automatically uses colorblind safe colors for up to 10 groups and has some useful base plot defaults (as well as including ellipses around the standard deviation of the centorid of all groups).\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\n# Load the library\nlibrary(supportR)\n\n# Make the ordination\nsupportR::ordination(mod = pcoa_points, grps = lichen_df$treatment, \n x = \"topleft\", legend = c(\"A\", \"B\")) #<1>\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/pcoa-ord-1.png){fig-align='center' width=672}\n:::\n:::\n\n1. This function allows several base plot arguments to be supplied to alter non-critical plot elements (e.g., legend position, point size, etc.)\n\nThe percentages included in parentheses on either axis label are the percent of the total variation in the data explained by each axis on its own. Use this information in combination with what the graph looks like to determine how different the groups truly are.\n\n\n### Non-Metric Ordination\n\nNon-metric ordinations are typically used when you care more about the relative differences among groups rather than specific measurements between particular points. For instance, you may want to assess whether the composition of insect communities differs between two experimental treatments. In such a case, your hypothesis likely depends more on the holistic difference between the treatments rather than some quantitative difference on one of the axes.\n\nThe most common non-metric ordination type is called Nonmetric Multidimensional Scaling (NMS / NMDS). This approach prioritizes making groups that are \"more different\" further apart than those that are less different. However, NMS uses a dissimilarity matrix which means that the _distance_ between any two specific points cannot be interpreted meaningfully. It _is_ appropriate though to interpret which cloud of points is closer to/further from another in aggregate. **If specific distances among points are of interest, consider a metric ordination approach.**\n\nIn order to perform an NMS ordination we'll first need to calculate a dissimilarity matrix for our response data. The vegan function `metaMDS` is useful for this. This function has many arguments but the most fundamental are the following:\n\n- `comm` = the dataframe of response variables (minus any non-numeric / grouping columns)\n- `distance` = the distance/dissimilarity metric to use\n - Note that there is no benefit to using a metric distance because when we make the ordination it will become non-metric\n- `k` = number of axes to decompose to -- typically two so the graph can be simple\n- `try` = number of attempts at minimizing \"stress\"\n - Stress is how NMS evaluates how good of a job it did at representing the true differences among groups (lower stress is better)\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load needed libraries\nlibrary(vegan)\n\n# Get dissimilarity matrix\ndissim_mat <- vegan::metaMDS(comm = lichen_df[-1], distance = \"bray\", k = 2,\n autotransform = F, expand = F, try = 50)\n```\n:::\n\n\nWith that in hand, we can make our ordination! While you could make this step-by-step on your own, we'll use the `ordination` function from the `supportR` package for convenience. This function automatically uses colorblind safe colors for up to 10 groups and has some useful base plot defaults (as well as including ellipses around the standard deviation of the centorid of all groups).\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\n# Load the library\nlibrary(supportR)\n\n# Make the ordination\nsupportR::ordination(mod = dissim_mat, grps = lichen_df$treatment, \n x = \"bottomright\", legend = c(\"A\", \"B\")) #<1>\n```\n\n::: {.cell-output-display}\n![](mod_data-viz_files/figure-html/nms-ord-1.png){fig-align='center' width=672}\n:::\n:::\n\n1. This function allows several base plot arguments to be supplied to alter non-critical plot elements (e.g., legend position, point size, etc.)\n\nIf the stress is less than 0.15 it is generally considered a good representation of the data. We can see that the ellipses do not overlap which indicates that the community composition of our two groups does seem to differ. We'd need to do real multivariate analysis if we wanted a _p_-value or AIC score to support that but as a visual tool this is still useful.\n\n:::\n\n## Maps\n\nYou may find it valuable to create a map as an additional way of visualizing data. Many synthesis groups do this--particularly when there is a strong spatial component to the research questions and/or hypotheses.\n\nCheck out the [bonus spatial data module](https://lter.github.io/ssecr/mod_spatial.html) for more information on map-making if this is of interest!\n\n## Additional Resources\n\n### Papers & Documents\n\n- National Center for Ecological Analysis and Synthesis (NCEAS). [Colorblind Safe Color Schemes](https://www.nceas.ucsb.edu/sites/default/files/2022-06/Colorblind%20Safe%20Color%20Schemes.pdf). **2022**.\n\n### Workshops & Courses\n\n- The Carpentries. [Data Analysis and Visualization in R for Ecologists: Data Visualization with `ggplot2`](https://datacarpentry.org/R-ecology-lesson/visualizing-ggplot.html). **2024**.\n- The Carpentries. [Data Analysis and Visualization in Python for Ecologists: Making Plots with `plotnine`](https://datacarpentry.org/python-ecology-lesson/07-visualization-ggplot-python.html). **2024**.\n- LTER Scientific Computing Team. [Coding in the Tidyverse: 'Visualize' Module](https://lter.github.io/workshop-tidyverse/visualize.html). **2023**.\n\n### Websites\n\n- \n", "supporting": [ "mod_data-viz_files" ], diff --git a/_freeze/mod_data-viz/figure-html/multi-modal-1.png b/_freeze/mod_data-viz/figure-html/multi-modal-1.png index 35a512322a2407a4d564b6c385728731121bd025..654ab7c1a37d6ca8983f9849a9a3e43c0c5e2b41 100644 GIT binary patch literal 200775 zcmeGEWmuH$_6Cd(2uMgM0!k~=(j^TdA>BO;NVn2Cl%mp-(jd|)-5~uX$gB9dk3}WM#iSnrgo+dmd2{mk0Fppp`jX33v!Ly zB8fF+%n$kkMzXRZ@h*6x%WYH#Urw!Fn3maodl+R?E*{ajF_Uab$F*?w(UMW*wm3mc z&V1DTrCtmDJ*)i9V3rT%${)}I7>)-{rPGH2)(zd?PL`9l;VZL_TeCP3 z)erSzXK7I%k`}Qn4gV#T(m1*F!ghwMCf!)Zr-aP^U}7opIHI8}QLtos(;cd1 z!P~L>=D2#9*bLQsi^behV0?f-(=15*;a2d%?oG{XkuO@gDRGa_VJua*+i8UZ4U%p+ zePwq(J2-XGN{+|;5y?-2LmEQMrdjdx@mC+M3i^)(uu3PN@E_m4DojUpp`ov zNr=gyfGi$QN9=t^lf9b35qA_C)l|OV`ruVp@%9UIllSxG{cW?EhOLa3z*hKlZ(Ll5U-LmsYgX|fxO+6Q zy=(3DzDVy0E1HO8vS`>Cm1TU;;n6%rrRs)w*SOVTnq%mlEgp4>Mp}0sMN-lEEX*#7 zeD8Bibx3_N=lc~p&*}hkfW4q{OqI48ao^^)-YU-cguQGTr}8cJvF-$m5QE7p;m@s~ zP2`kTi2`3@8L@e|@Z}SRl0fS)I!f$K^1TauNpc-G>xFtcolILOOO&z&PSi^pI)4kOuJ{A zI;LVgI-?gn(yipjpD(yJfVAl5^)+8;*oG8{WVs!xy3v1mHi!CXG@!LbONq2Up9*oH z=VnJc3z_h0Ft|jXdwB0%8<~Str(IsVq;C0N0)z=X$AqA~wuGR8 zXDHwgDfk0{pr)bz{XZDZY1jUKe(kRx;^PT}zr`R=pNOlwp=``t_tGCtKDsQHYS50P z)rs`9cc^;NR&6AYZ>vB}P>Oa-Rr~QS!B;!tNNDMUL@DZ%>u!`Xk2~6l7W_S18VvQ9 z?(0;V%EgX)re5;L$;eEUnwAvA53cNxayxFhZ-#@cFB8j3jc-@WmNHv%7n!rplO|KXx!eIf6a zpo$m)f4U}WF-XYvA8zgHGg5C+LzdOzqFVlZdjyaL&p%m-K&&|%4b%5zWXXj6_eQ$< z3}7b8|CgDG03zOvZ8KA=L9dizGZ}R(gRra8JN&+~vLfVVqJ86D^wN*L z8wYoU#M7`);Qibvt!8L_G|ETt-08OMcmb@&`vc`-Gsr-nes0?@lxXW{>$7am#V5jQ z^etLw3B);`b31U$sW-xLbkFyC`Ix!6_lNL4@S+;fi9rfpu}yexH)PYw#g^~rSWy1< z(EOIM8*oG`^7TD``NGUB%Q4H> z7vj7+@EtoTMvZ!Y1~oG?lZk;r`PsAkT)J*YnIDg|#M1KeWG5#l`{Exzemv6XjbwhV zsp#kHigTNoIIq55|L4!2v%&-(bL*)2TE$}v;i9wa=V{k4^|65eYl%%}(OtiO9p=a7 ze#E#D>V;50+SwW$@I=!!Xl~S6M>RYPv2?fZnW>&_q3R3Kj^f_<Sq$$3GWj6%;g7RNj4VZ5_>&BRSZd*1EX3FyNyUxbi-)yWrAcQlZAb^H}`b z6AhkK%YDiDTy&7EQq`J#>jyq@XOABf#k|i+w6BqjqF0JdO-;Rin8JDw_)MIVFJxk= z>?7@O{48x0&1T;EMg9JQyCukcJwR_9Zc4iwD-Bcq}Yc3Sb7 zsePwg3tc`v;y}eLuZDl4l4XIFh~1ksKPY^wWv#r@7pmyCnf4%a zWqG~A8Bt-qsoP+TYK$+mBbFghe|4x{Pn(zI0*oI1##_Yn2)}lB52Hq%FLiZ|G$}M4 zCe3RlOUJ#xBMfgn2ztmm(vh!)*!kY-Tz4D<%z2yyJU+GV;|Gr8+ zp{^$3Ti8W8BPfpgwzjrqVd=N7-oy3^6*hwgVZVM4u{1w17ne%^Uc|(kHyC<}@bzs6 zMh+&`7*kWGhYxLP?*mkYps*1GG-x9UFZvxCfUF6J$1NKokew`^57EI*???iNYIM`M zOZaO*Ex+qGF`*rk7=L3=PA-vWS(k9Xi|L`Nm`u)n8k$7s8!=a(YflJ1*H$a!;{IQ$H#6n&%0Uj zLN?P#;)<20CO7u*1hcCrCR44mM96u*Tj5d)9oHXByt#4^e=%RS8~z->b7EVa`7u3v z4FiJ|-a`Fc(E{zvmbX7izhwr0`21KyS0#5bHZ zVq^p$^5hr+=)J8@#26zD zjRb(7WI@-l9F>gyvea|=h4xV7k9!7}36S^Wu*VpGIztk|M_$Jzvv8!X=|}RIx@zsb zg0qc?H0nZ|am8fh{;2oSOty&kseVB&bEzmlzmA~mme6)hrOl*j)4Axyg6Mue>|7QO zhaXI@rvP_fPk&80l*I3tq)-`#>ve)eA{INNWR;Z0Kfl5`_(~JLni*#xU{B_3>prao zN(Gm;Rkq*z7p8usFYcX&iz(|oUX)Z3m3O3~ZV>_%_H0!r+fD|Jyi~$@(S>^Z9ctP8 z^RA5~451;ws& z-de-iO7&W=HxD^xENn9o2 zgAx9g;?tnzx5*dJcjjCAPA%*k4m4Y#2IomP*ozyzoPOWEp`az)DzVn5kQ2!X^Yg3C z@@dM-8?{g6eC{j4YyFC1vPtRg-Mi}(pN!8%l)RpmP21cR@v8O@2td?XjGvXNQBOTA zHc$1ggnOzfE%zp_$os9XE!gH=u1qUpOq{4s1pWc04A>%)k{-G~eJgdaaZiwFzAdPb zt5h1lCRolS{9w^xXI}h#+~+)Qxa~EiM~;e?=*6+|%0aW=H6nuNKlK>|H`CeBn>Oaz zMAoHD{wkIJg|~l!?mr*uC7Kn{SBnZ%>pNds*}V<7p+EXbSlZ$7*nKTCZvXphsxjBh z+S;kLL|osR8M}tF-R?f_zyqS5C~&NTY*CSsq;_^{{aC&gx!t{nzrkm}J)DWeb>z|Z zIKYpjLjUWlfp`>oAWV}ksVnc#e@FE*TbHOgQ-`VzZ=`6Bk92(u?&PnWl&;Wk zqmd5#D$=L*SVBTWS9iT&Zpp!0WDTaM7^c<$+;K&)K*5FvGDgSB74Vz%3EKIIGV}0R z&uw8-Dp9*EICE?sQ!>QY-~ZrUXW&{4x6ItxGT&&fT6Pwh5k4v<0|6Bko9Lc*QfzV| zM}sF%{_Dyj;dV*{*3zIf)o4#V7ozUj_@ToXFBih+{G{J!ObNbA*qgx9-FYiGIM{K$ zt-W0UrlzKb2)DLY0=cV#xzFu)w!uJU)+!oFa-CJ_%~g|KC&(Xc#bD-O(jBef5r~h< z567n(91Q1M<)c?zy?%LUPs(m67Lef{6DYYl)2o-nQ%yqRvDC9mS< z=BB10Z00#$1a+Rat>Qn|Y4C9NZYtt1u52qJ)YQ|{Yih`)-&^S)SKyoZxRUDAn9F2P zm3*UwxY6Vg*O-bYj=jkO$FX$`PVjF6Ucbl4+J#9FT1* zC0ZlHz>aIkOviRS8}?69KS%U+ct4Dn^XbtY5dHJ;^oijqc!Sz)7yUBQRe&!iQ}TB=s4CW+2zB{FYY0}+SQ_V(p-QLqtH zPoA@?Z-rH}>t|4_@#}hJW}!t5hf_Fs-|c;niLtEEyod;b^WzTDMnkCm0t?ox8CgPx?V(R<*LNSx)xZ)sEqr7a- z4>u-x?IED!yRU3ym$m1)tokKqdJLWQ2-${#rbsFB5ydRD7*02xUr?~&|7G~}XuQ4k zb;XNwj}}jMBc)2#kGBE{WF{Jsk|+k83)UL`LGT&->$Z zQ{NL$o{|gTitlL#hKx-Y&9tgAoeM*%1A#Tq9Epd={lVk30q zbN1f)@+R3RM&ky4Y+7(n>-y-m%sotfIuzt}xfv?vvydA39PWi-)I&P>C!Md87oW_1 z_|&e*C=?_zbBn6|a9c;v6ccnotPc(Ce@rI^s;H=#_(I(%=J z^ww#7%k}m=No4%7QEnAiCyLwg?N+~Qzuy`H#6zJw#S!kc&5exDdMnrDSX z;b+x=&WK4)%pjy!Vj3zinw&+eVkO4R=e}uh1OxiV;HonWg z>#63mapc|vKd;Z~=2hz!Sr5WhrJZ0T7j$;u80;0V3L@xjhDqZlN%ks@f_{%0Jv z)rxTgM}K&C99wbmaE3I#@cI7m=~z|c(kGPW(>{YAgs6sbm{PxMhMV=!Z3e7CPz%I3 zsME_!si()I+Ko4!@Amh{b@#kCvd?kdo^9wP``%Gj);mHfUOL?AF868Dal7fVF>>!4 z(_B=6lE}%&PD8({khkzkT)*B#K3J@u%horYrt?kvzY>Pa`fzOl8yQ;2+qX3jS9ZQg z0Y>kKM!M2(x`t*`V(l7-%;oyMFe^^35f&l&h3mLO*G)6K*b%g@m^)fwIw=nt>h(|P z?(n+p%sY;ucXJ%znWU(=4N(K0>xlEHXemLj!Ybx4^-=!Rj$0;C@+c_MxlO(j6n8(f z=D(+F$BgPS)yfL3eH32FZl^8Zw+Z!+rUdDUDbLcR9Fd0)e0cRA5%`}kAB zBf~UVDC=e~+kl2}EpyjW_OqxZ=lNz7P)H(zl23O+!};>YOoHW`wrc6PzO1i%c8xzK z9MLOxdpjg|#=X%;#qbefNb%@hArZ5tn=Hv1prXqo0QF1kpQ-b^UYQV$iHT{ysPoJr zRiyAWy~2owu;+*?f|3v|_>En0#aj0MMT^M+w;EX`mw3tJGT!&eOueyXuOvNwW-Q6( z=BS6Kw{jb|hny}5pLd>}?8QVvi(VS35m83vz|}4#*D_hXt4(sTv9X03ko$lJY4nGv zQNz)N@FaiJ0|-G4JHjZKWR%9g_9b(FEkGYB`6BgX_4PP3KTfM~aCw8=+rccrLs~lj zW4XLwsUy)h5vkXGDKUb)nhotCBlteY66hEhyf&YWQ>jGZ`qm<|1shY(L#ceqhWkqn zu@u=K(z?fh3OM^ZmDhkVQmF&*3HewND2JnGk4S&R5A0a@c_R-yq52;|FOpi7Idb^Fn-Q)b3k z*j#8p*p%fPQF6+9nDD&JVdkc8)rwrIDig; z13F~Z2N%$~zYMDK7k)o}JY&%32>Nt<>tE2hX$DPalb2CV{bq5J}po2+~O}x5tN4KGLljd=S|nT4%-UX z3ieC%?!T@%Ink|fFO!v(O&E<)b9dk6YAMq$JB+J}G3}0JJW)@%?-8nwEIRL-^N#vC z>N7P%CqqZCt*N5&;=;oQIBGEw^|c;rHVTSgH?Dq_tjg-@A4yDR_c$h&v?&N zVB_Q2eHr56mGaTv@pH-inB%D)Efqp;d8^f*bky?rGe(mlF)^c&Smcj|1+uN(oC^0=j8}27>sw@I zBd%=ICa)SwJqi_5dn2diH%Y9rkcmlmilgp{cX3$1>Yb~EURQ`6cg+ob1H{99e%agbiArXGICwCc9@3xkGjy%dAE7<1g zDmWmzb!RnX9HScEv^2xVrsg!GkE)aLq2wk zN5?q%Lq@b!X{$@hquljX3uh`G*}2P$qouhJ>K~Htij~o;I^7a^tYgcvw6?A7<9rMk zEUDQ>LMe^SEcRwYPu-?j6_P!)&rf+i*-gA6HKGDHk>q>w5bcUy|7Yj@?|(K=qm`DG zDFkX|*0!|^K&!;lw8o3nR1Avf5*Tk%kqPR0U)((_B@(C%AFEE8;rV1u0yU;2%A`Fl zj6dBD^}b$SQLSEN(BOPN0~-f*QSR5TGQ~e8s$XOrk_$7s+Tc6OlX^b&u4*Y2iQ0PZ zZ_fM1jDiu+>*uH@zV4SiCTPELXbf~KE32NI8HadAJlm7SW31$_LK${O=?U@N&7G$X zE@yi_=u1RXW~Gh1rY3nz4B8bJD0xK%>N*zhDY@^$GNi*bEi5crCF*~;YiI!QhZ&-3 z+l|Jij%JO+e`cOSb=SsHqMdAK>vag}re=3g4iqEZZbXl5nB zbzo#B?&sLkXWX$^R4{;h4P*Rz+d8RUl#6~a9+is9)7(63D1JLH)SR6Uw6=kOO>$}Mz;w~za&l~%sG1icB!#Zq3KS|6Ez;)hnj_ zB9)5H{nE_#&9ToC(k7}I`v>8WyQq-)g^IWnGUV{3RTUEj*dRH6}mDi&=K(bY4mTQn)X0YOZYyP$xatcW!Geko>JIQ^l3 zZA+*hL0#5MM7&tw=bSQQP}LcBnVX$5RGjLmC>=(rk{N$7u(@5_g@JI7(a>ZP=2OyX zqjiX1!@GWiJjd6!-<~io;T+ED^C}_aUlY<^)YMCpy=9dF*WX0>9_m6zI8eKMva+^L zw0-yf{i<3ErC{=RHBn@|fyWxHcdcpJkW$Ksbm!~a>d?&fgsFaiBeSVi>L?0c!)8bs zowgHAKFF%7GH(Q2x4u~j{?+CEKg=bY-O+1Zw~GKxc3h8~8ft&}5>0rohfBA?qv8Td z(lp?ACKMWzdS9Vk>t*~8c2?Fttmh|*8zq5D^5TAnW+qHRmJx*H*45!FR~D|q0v4`h zmh$4akX=!!Ow(6$zRMM#wlJg>(#YRU)a>~oqM@L7oxrS}a`n@ICx=IMr$?FJ{Ve>>7gL zb`Gl``)0MVrrYi3|q@riYLD4XEu_wTK*V^UL#4GC9e<>l4mSfSj= z8P~_5K3Uk)E$vdC!U zU~_v6>A_*W1^rzsoNx>D198he@fjP09ZS4?Z5mJ(uJH8Rw+px(hwDpAl;@~K$H4H0 z_i=!KQU}R2*-U;mUW!QJ<{Etk{M@7WJ;y?@hFd5Ak$OPZDbg)wI>7;bSG(6I$hH(g zoL}em^O^8Y^R=$tUgtieYq0av12sU90d8VJ=xXf40&CQ74a60+7iE%RAV>f6^+HLB4I$GnW?^2tIyKB)7kMsKbx#lK<0Left%nAmdgTaANx z(pYo#qfi;rw23z^dEDY9BLxq8g&nll=xBP(Top(M<~D+>4ehgOaYrhbm=8MXQwbn- zX}1L%o*G_WoOcP7NWpn}^|t#d2nx>4l7#?g-Wp`w# zL!3|Ln6zU8);h5iq<~*}4M~K~Yd>2cNw>%R01-l&sk-m*mgVlfU6kBEeXP4T%g4)*B zHdSgGyx~FZq8nXLIGHeLS^BJX%Ew?0r{5nn;P#A&Zkec#g$R$q-m8ZCzdcmx76TjG zI5}5gqWtnz^PmBm2gh3TOFT?YzHa;NacQd5&Y|>zGF|WaG`l+ji%GS(d}xV~c-y(* zpycHjqwAxmqM|2nPY*Yzb3DO7XcTt2JvWjqO>%y=A}UHRGmyi{MzC|s4>XH|I>Y@K z&U0|&CS6l;2AHaDl=WMISSZy=ccBog>y7Om42*l8Jlxx!{@Kx`?0>=^sMy8|bTTH3 z8?tMBk~5*!t_<{C(k=WKK|NTuyEhZ4K&2Nzv`DoYDv|YXp<6kl-{y2x{^~ws=+@Xl zeG6Og%{vzG>_mvRsdG%Dh=uPN6La~!lZYsVq1&3T;sric=gGcPQBlyajM;RNT`^_W zV&>1*1c7dl!2uGuMNwx6e>B@Be?k#18E?5P6+a_toLaM)P$P1yPB98hJEK9y^J-b% zu3o^K6x*3BGe-o#;xa;UQbt-`L7?s6JURMjps~>?+u9lAR!e0_ytY!`H*CYG(5fl{ zBb?cT`tS%V$AF$fIbHgV?)~7uWoGlcnBX1qVgzC=V=hcq%4EOv`gj6vLY65asP*fRb6$h8JSbw5*`jY&iA5R$O7GE~*P7 zk;4%=76Zc7!a;?;x6P)6GIDZJ$&CZC_as=8Dl8GlNThYodyZ7}#!U=0;DxCbi^-St zkmcm@P=kNgh`qP&yr0lI|AAy+Do_4FBaowJ^241<{6f6fDqQy8scil(D(L$oJ4a+g z8lPurK701zHu;D0vPs!Dm~qnu0nvJ@4zuFXiHF47vouYI90}}3pWVULx6jnMS|evo zVqUdbA44yf(Q202^jFbp92^|V_cMqDpp4Ecd~JIk9rpPofx=HMW);&FO+(n3a5ZzLT1R`rMN$~m zVuGM+k?LOFh~=U)Ht4let?kszG)hEtE2QG%;|m8(udecDNefiZhHj-TOVnolC2*On zsl&icpuzgiG${R>LXE?(^z89cs*v-bw1%a7@WFRgvasObD1U!v&JGiQXxeeSVO%dI z7KeCAwEyExE*;uj9rGC5tj7pS1~G-#85g5Mon@c3QA#RM*&bj^Zu?UN#ggFBuLWo!&Wsxuw4AItr90bC=yTw=@a zJ->_o%-A@2%t~Qw`R77MlX^F}6*lK*`1}mb5~t#P|8V~Q7~hYX$&icr@!7TL^G-NX zg0!P-TD;~2MvQLV1dm^3fqri2YkX_%sv?pNMH!hl&uX33wqkAb!a|0I$`^D!b|T{O zK1soOqk0NXmQp8F7K&@6=8oI>Zj+OHwHy|W^nL#Pm_@h#$r>$dXuY-BxutDH{`7mY zioHrs6MPz2 zOJ-r-ZEQj{ZE$<`vVigoIW+u^=ZL&F zm+#pZOy|`}wd*dMBbCwWp=&ewovlng$7c3(xi3G8*B${mg~!_38WV@LpX8G#IoX!x zAOzQkGUcLdK`5W>tuWKp>0`BKF$%R8n2x1Y%OC2AD@6Ez7GPpXR|*g|>v}{uvLUzx zi>^I3H9!3#9)OW+*Y#D3$lcHPI?lvVCLq^=GH9yFr^&8nDK_|UY;0||p$<&=Zqn~~ z@=sN-J{7J;1yR!Y$O-p92le02bQP)T=-M-~kaf}TFwoFe6X(vd4896vR8`t(y^yTQ zxHLn$`Vr)kM~Cv z)e|jd^eYWozWycAYQe##YOWuKoEqOBl4G;{`J>HaJvW;LD?^QPZ-@xM6%DJGcdao1__a1r=*L9El-@hQF-mD0<7(a>U zG7pP9+8UkO&|jjEAB!(Dzmj<}RgLhoDt2`aGW(om+$SV-4vxoov1>vowCh{Z)7MdJ zJI%W(@cu*I*S>XblV=%BneOIJUgM-OOxn=)*>qj4G>5Pa)dZ}=wvG;FcJ`7+_!E8l zHvIU4=vM0r!xmI}<03e{(o}SFG^1KMGiT|fapiVszI?lhH3`|`#m{h`g1L*sD#c1G zUEPGOx@{4l$*gO%&dtsB*dLUxaNjcnW7c&gpHsDS5#%E1CvwEuJb(-TbcnW5QH5{+u<7EUSVLJk?Mh|Vblk?4^cvSUl$?L)B;^w5zf9-&c#n}l&S1SU z@oDA8oTB71z>I&K20bD|>< z^KQ+gRq{O}HA2=l;PZU#$0TKytOn~a{4z<@e(TW*_vy%a*oUM#bu=s#M2KRJRoK|zr)eP2bb zrS)=$Ug-=ruM(Jg#QpmLKzBs{>J^WH`NC)EQV=mkAH|G}(m}(N5fXAECMKq{aoo02 z0`w_M+xC0Z44BkF5IZzHT;Z^$`f7P;^=G2Js!ag3vP2Y+q@NBRaP3dG$yKXUz$jAsXuPiWXCr%j~dra^HsGTZ`;2q`%#pc#0(K&plL&3PPJ8vfzK{bV6`RUI9j zzFJpdg{*8`f1YgpFx~`5N~PRkAtI$Umh=B%EiGc30qiHATv%IKC1}iP25}?q6P#J` zk`Udj{(wqkMv>}nCR;L!7+bf|X_U3<(pV_ch+q~|@%z3WGca1ixQvB}(O0>n2-I^2 zsQC&~S_+DlcAPD#S^hBR7~`JkH&u94S*}M^19=+x`@>5^2$78SD?3uijo_V$CqyerW0T)dO%oZZ7f4N-CM|Cx6%unL_Fzu% zSUfV`fE7)4l_|6-#)R*&OF)MKl%qK4u$)_`;*pkPR`R1}f zuRn~u-irw)rIF}|+jDGMPv5rhZ;cih7q5@1l~rv=2=@_o0rhBwQ8!*PW6ZNMMXT)( zT8KIgJ+(p_#7h1I5u@v@3(?uy#i=)LYIS~ojGA6%S@&5?1OZUC!CF=I7RFlk$qS)% zWATbb-$haY?Y(`)S~XaAz4<>y5>Sd;^+v;d&JnPRwN*O}SG>`ef_7Sl+Rrlvo$rn+ z@#o&1618twerS*e76d>qS))Q>0C+HaoQaTa|9T(`kdGUVt4;PR?vwFSOn5)_5; zc5_dVPbbi_u=HYS;ab@HYq}#zo!c%Rh7_?VTUcslY$`Nh7DdzDP*RtX`O2@cp)13S zNCtq|$OhJbj*2;G{%#=ulReppfF?D-g>ovRRJ)9_Pk6j*X}A*Rh#`BOlXMNMzojYVwgCIB*K7(`^E#vCCNE-AMR zA`0J1Pc{!!PtONgH)VvgNcZ^sgu|{f(eVc`pn=urg<$=+Ev7mJ^NclIJHQ4M?dKaipp(gFoq{%!)3q>~-IMBsx&d6#v5bX3%^?#yyMrrXEn(*wl8 z;H;9mj^1#=1$y;a($PCU^ky3vUlBB)7F5A9V8^JTnLDQjFs0>4wG(=m5ulZsen`l4 zCG?-A1I{pAbm#nkGAo0jNDo}1$=Vq)k7%l1NgB7`bc5+x9dRQGcUHKYF@TMqq4Hj9 z>rd*1PS=MJO>|E2_|bYyJA0@-cvHnZIY zBqyu2iFQNMjw|a*-LZ)8j8P-PtXoN+QhGW^^ATgY514d^&3crNwr2WsTfs+(HIrOz z+kVzCx$zHtxJrogj;Jq|+nsZKN`m^5Y!4@bI30UA#7nOxA@llvB@>5q5MuaN7zS9T z?9uB6vZ!=_e^grz#Pth!8xw7ttFmggq7?~g-j7`e3huk}NX~wlpID^A?xk0St~~|Q z`PGsfE0Y*xPN=)Xho) zzud;gC#qi0%jbTvsO?>Do9t5GC@-;{P*5`QwY`DInk;$BTb+?=bHw*eU+pE?5#ysr z)h{Rv?)i!(67KNbg*|=xv@fF?KbyC8Qjjogy!?T-H z7hTUJ4-}P=c38W7lZHztZCwIRswgKKq$5T7l+{P@tHKFHlq=D}dZl@EgkGG>uy{aC%NO z8WV~5FnTq9!g^>>uRZNjsIr@WO_AWCh{)_p`&Zw;?pcIHy9wTKu@?_xDCSjfNycEJ zt<$~|8G}Wu!hYtTH-G_kuQvB#1{kmZyLfG6onKj@KaoQy*}kCcy<41b9Feq;pcXHH zdF(5usTr$Ta3Y77Iqo&R|XuUy&Q=dxSILtK|X@k_th5Y(mo4GX$j`|ckYhX7(3#9Oof8G~c#A%xD_$D^mW-^ds-j{a?SaKWns}Ug zYF7%)4%i8>8V(cAvx>_p5z#`7OnNy}a!5{J9EK>K@9NMKR;y(zrVJZ-fRksr53W;E z+w%55+^w1um4k!BW8+gtQBUV198?1au$UoEKcCs`6-p|q`{F0iXg!cDGw&bc>KCGk zL;P^`2Z*aq&lRbmo{;Olw_>^Y_1)EO+GE_>5V2{Z>dne&6RO*DBwu@aMg+^unDakM zeV348{y6t!i8wZA@ryX1><}jI74W0FxwVs()Fba3Qg6h_NUD!v?zZG_zW>xmL*v`_RG zpv|p^^0C00bf&B9n>Z;J!8+@J6-fWOp(%v_=gP$7unY96`mmfTly)3ZVa_WFZnpMI zMA@KwC(ir$@uS_Ids83K(3pdPov`rTv`-Fw{zpqQ}aw;5V)Ko<}|w#8)=7c*&$%l`k)jO?AQ}8QLhJQ}IrbSlijz z8G+|p1~wLhuQs(+3`qZvJj-v4pu~@bY`%-nujuVCl6(DeK|IE%%ZNLsNy0YkUIP}e zR+7%n&Z%1OjBh9>q#Rj^_ovXUCsxB0#d?aE{~7}lSU~;U_mo(uT1o76VS{^st}6AniHXr-$dZdh z^#~n)CqcEg6bBmva-~U!7o3#Izs{-bHWLX9+9v<-#9q|fT~&8F9&3NiOG z1Q5;H*#~R0wx^3xO5A|O%(k+$%q4!v&8He1zE)XGo9yV8ib7FtKh8+OgHn2c7`N_< zW?UIDx^^Wf^TquTJ+7HQPQk~(`EyUqRZV#Au*y}wQrlLo&fBgIieTsC8^_x1Xlr8> z5STJOy4I=wHo_|p1|*zW#d8xWMDvZ3VCl2Rg(JOFM|QPmaBDr zEDKxESiKDlA$1eSiR)Iyue!aDGt!x-!|y3Bgmfb7suf*Bg5u5HsJDY(9}f}BhNssi0Qm%_bAo=ng! z*(eHgRB1Sns7&R@71eDCteB^yPLAnje)#ZskL01U?mN8zv)RIDWy{99>8*L8+w_-x zcX#s(RY%{>FbS56kMh{P!mlwX8tBh@w2zOiq9GQ+l+Q#&d^fu!oy*B;+VgZ~8=ied z85Qz8=$j5gb$54P*8A3E^ApQCanf~*Tq|`iHUE%kz&Tf2k z*dwvvRq$Ne<`}coTWCPEi$jxUk$x_yP6~A!^fNW?k_zUo+4$mN-CXC&Y0xp>DB18Z zzWSP==4o^r>UqAVlTfp!!-PJY={&_(uWFed$j}6!-i~wB?BKH-n7h~|Id2o{>h9LF zt(J>@IG8?+9<ST+A$wYsq6aJ`6nDbEgD8Z_a{M~(luTA|=;zt@z zFECz7=XE{UB_c_DSn|R(uhbjAaI#d+WnE8p`c?_z>-mrP?1OHYNJe=%cQKh4_$G(( zgA-ZT_cuzIS^7@?Re1e{N&k9ClSI=j)Xg-eKV(doi?A`wuW2*$>pkhnK7Ic4zW3=` zb{C_}=+~dn+r*SkEp#@%8#5Au?%Z9Vc zW5RqhX<@MUyF07#Tpsq_nw!ujMYec+^0QTG(b^x@)Kivs<5*$2!%KdoTqlZ?Tgi)?S^mVq_GBQ3>!yxjWyoadYwW?eJDKI1=G+N#lCTBBdZ_h<9im+s;;ABmDS80 zFKWC!G`X~Px%QT#2l2MV;TFq%4|k@I-)Jz0r(tk-Rr=-bkINa$81jAd#wFK9@&6h^$1r z+Gq)(N>&v1@CI%tP^JBkArqLDEjWp6Ao*$gFN+fS%WGcqAAB%=9K&57aA-GO2sp7^ z$5bmW?`N06l}6(vWqzd%dSRVKf`_2GhV9>By7I#m26iB}HKm*=Zh5!4KVzx3DA0fV zwfC)C`(G6J*2CpQhgvqLO+=jxv!B#QE`hz*@lW+k&Ud}@{Xb_oS@fF`#m08B)d#3U zbG!B%@^Tqe3OkP-UOnluEp1@}g^aDh-8&X`uf{{WH;S2!TCqa+3v4O=Nn@;ToYnQM(Y#KGAG7k`5{tfCuh<@pM$ka7m=tYKmPsEP z1x$bqlJOFw)^&|d(rn)8pQ^H;e#=xpTCD}6hSS{v{{J-=fkjv|SR!zyTY2+g!9Uu) zcEFu`rMgz&`@5E|u9?Dmu8IVg4($^ppZ(~YEAXAq7Oe2>=UrJI9!QTP_R0rL4ghVx zv0Ei-RCc;e2I;hI9)|7vdl*gYr^L7J2J?=Ly?aR@b}ZUM?oO3kMnw^sG26`6Pof=L zts|!BD;ri)JQc5oFo5bEw@CFKEYz!VtaL|U0Hwai(kkEofv!iW8X6kQ82xlB?vh4= zoBFP6*3TA~m-9x*iJzI_F)>+OO!4w6^TMsz3Js!}65}KOe{_9yR8?K~H7O|_(%mU7 zjesg%t=;yW5V2Oa{d;I;LRpfG!Jwi1v) z%G$PFgrv&T>Z!>ceEJ=TsiLasEZtEa@3Uyc+Q z5gbK0T(Dg!rfYf7bXz{2VvhW>fNGxaa+)+T}N1edQP#fw-lnSQr=Q zg7~XKMQ#l*vGDi<1HY;P^ylKbx?Wkw6c$&5qYAX2%DqHzxk*XsGqL>&q2lR@WeANmyLJveurCF;x+OWvr07Ak5Kdb*S9A!Y>R^2em zmNOjo?uP!A(GP90 zrRo;HB(Z}3rB;dHCoY!awBn0XA0A9Jt8pbQU2zKfH&=k-b?-$n+M<~Qu z79$`kI?JbNQ)_gM$_AZ7=(={)?`>!Tyf9R=JJ<8@R`c2F+e-G*?G*F zMhqxOy~@ml8V3}ekV)r?fr`uH*q27V9F6TvM<=NeDGwrM?CpgL+2ugFyl^oy@vO7$ zJoETxcQ|G>o?%_q_v`8{U{@gTl9IW9rq+I&D3$!&zHUKskL2-MV#$3cQ?tpbh`sN9 zv&&(I_dU=g{cY*=0SHiy3K}al0+2$OICH00u5!=Dg+6yTL-=D-qH=YTlQ018;g3fk z?OS`en_{WXEI(S5GWwpU)I3_9FOQ3DxnJ~S-k8|7Y$ZoK95W%7yA4}4eeTn33vQfvFOzwN%Z^HP`#Wx}R{EtXbL3}a9q zbCx(^ zV!DJN7~L);vfwyG1QG-($yE{327`@d&-g9!xsOxe%!%emJI>gTdOP!?Bxd^5)@z~Y zkF9)4AU`pRWG<)4tv)uwZ;)RrmrkfNbBLzj?c@uw<)U#OjNWDRX(`=_;+NlYE zztEGB*94f{eM3C_ZVR@F}+?d)o7U_80F%F34|?W4i68%lor4CTRoZm zRg|)IJYyY}Th1I;(dg55y(}-Jj)nQ;F z-%#?oGHdE&HUeJKf8i|FeWzpVYp3=w#)p=#t7wPHs;cyyMZwtH`bVgtU!+P(dM#jB zcZ=V)k=-g8lyr+~7naY&cNafPy*8np-&}Rq7wB)1e8^sMkh##`z&WV>S7-yepWX8=FkT44b%aFI{Ujzx{}= zgdyF|m3xuh9m9vj6H*+X$>*vAT89O7=r>dS;wSZ;RGrBzK35cH@z)>Unj0)I2LueL z#ZId0L|pDke-WVW91$u$FYX;DA}Wh!#VS}cl*ZY#{S1DHb z6ukb#RBn5_Z#K7BxnW#6vPl}3wWcXC-o8v>Ik>hS<{gu$O8g>u-cKdLZDkTtxn%Td6<)uUR z3j4-JEAZHe9dzuH8@%XSTFt~^DB3vOsB$@+?>uMg%YoTkdC4{~I%HOgMqCp(EDU87 zCOorZvgg~( z&x4aC)2zeo+k-bkEDpL3Il-Vn)CtI~dp4d+c5-nM>{OdX!zdRKOr>6xkI0 z;L}h+pHL9iiy?8KIJ*9&(a9sEp{DjZO1O=iTDl1x-KsQ9?`*l>M@CK*|4h=>Wq-Pi z;8)TB|GUb&ZSMA>?k0j7zeLkY56A=HTN>gu5PB8s^DA#eUC&`kOlIhUCF3{oWK7Hq zw7?~ejmC{<;5Q0Lu!jHl=LHlLz(-iQRJz&H`jq-MDOhVculoCG`b85zEfbb0o&|)+>6B0!n!KgXtx&dQQ^JE z*#HgPFrOl`W-SEh)IB^W@jCG*_h|)8v?VMBy0F~~501zw z8k{w%Cgy8i_Hl$Ob|lx0FF~PX$dsQs#+?+~?0Ok5>y$hCiXqcn#&0X= z@9DYgIh!45M$8b_vt+9rv#N2XxB~o@@Ccq0JjsI3u`v zk6!)9ICNG6qw>D%v8AjmEU17CO>EJP(>2&PJ&VGkMZkfDOQxglTmbA}Ovx!uYb(6v z&>M~M%wOc-w8J5H?XBG3iP#kUoX$R(et+lg#|tU6}<@Npw>e82~g1&VwY zfB4mkFNxm6p4#{1ChIM)&|!ujpTr}vs+zSz5Ipn~p;GQ&yB*ff*a`D@Y{K2r2;Cq6 zD1Lk&$J4`uzD@VevEEehm{e5?IpL`{ap9@)>>VQ$lZvY3uk>(SEbvFX9Hx-x+!Gm6Xv14XWHJz$AoE!$z`{0i^@4J!pj@>!kUJCV zQc9}%^>O7?B7^XQno4xygSzSkOe#(3gj1K}G$7vH9x9Boee-5y(knN??)ZrjBz3OJ z5E<~y;)mxuj=@hjWECe+BN?v(L4(aV=Rno@JO7M;qHje%kEVclr3=|!uTEL8?THuv z)#4R4Zh+BNp@g`2xS&AjlhczM=sm^7#|N@_@5-d`yE~ZTeyy%P$vpOg18!m}+fyrU z8_^?{nBIMRFsyW%PpEnC)=+XH3p`!WcQhaKJLy0K{rA3K9=xl5snCpmfIx`*?@xz1 zH7V{-69M4BO1M%wD>;I8)B|xfA`CL zY5pw|x|M~W(5L%{lkwEAJjq$lc3;UfDW&r|(;qAf*wHazUvY4BJP45)G^vbjp zlc#9r5j!U7>6pv7cG@B#jV{=>6ZrxKo85Xhf;jXC-Z0sTvKs(OMr1$U?<_HXYnpzk zcvVIX5po}R(1WX|Y~z)GV#2V31|6<<7E+wiBt{|iFsB4Y+}i37`2N~f9o^)1;Lu4awydMWucf)>pGw+b3lE^Nm5G7?0bzr7)gXn){*#Dhv zgYT@W@crS()t|`S78W{<(5`~F1tpMN!-}_YdYsevY04zp?v{?}ZO|*i)irGi`A}Fg zpWiQOCz6Z5)@HC}7v5S6H$UIr0BC^nV9Gwh_p%Jkk#eq%F_iA#D=fDM5mDL8%%wBN z8Ix`ALaUy?Dx>a+{)AlijiU-as_#TXUS2-p?%i%zg1J7FDw*Fv&gI0q3^ktE1(4h9 zyH2jHt-0-Fx+d&9P;Fo;@cU=qZ_>D3OsOT3+OxUkVNDo!@a(dmhYH=3LBF_?iO2o_ zmQa2$Zz6@gy#<1rpiNs_#dW@5258iVZJ1!06qgk94uRZuBh@rZ#L9{h@G_!WN@bN_ z%qQ&AqZ2T2SXHmf$@Jpj%S^9Lz#A4&*FR(Vc^Xi{kw<;V&i^?QHVn zJ59aXGs36sMy8JqrS9Uwl@R*i`$a~jy`6IEce(k^*_PYG#SAv2Y#Q=lvg`!g0hqlY zDl02HbFlc2H}gM)g2Hcis=p5Tvi=yh?O~OuP-WmqdCd5W~|nr6cM#1jFP+t8dZ0RWT;V5 zQ6(eet5UeEW+86&SO%{3Z>E@NZA0u(apBF^s5aPqM--xv6V`~fF6m;CZxPi0VL~1vp(ICfpV87ciFgFb z_dW98KljFngGvca>0|spDk`UhsU1wEHbcqO_S@yxi)y4ueGKx{*ku%EE-t&1Q?XXn z&?uw6xYUHOS29WKfGNN4gh>lqQAO|JOtd0=2O8(Ijh?7y1XE*;*Q?TRAc21-tU21# zvCaAtUOgCsOSxXU{QW;N{Qt>LU)aKI8jqMDshX-~2u9!LI}D>Dc1=_UZ&;@su>2Jv z{IJo0fUXaa(6N&}MAq-jY9<;TcTPInf5V!SnV6rQpLWWywEk+u9Z8@%}$DVP{TYu>4MFq(hR04!n#ie&D0BREU z8L7pslX3)qy5Z=?#D{OJFOaWCT-(kY`=;kkUd+w4dGoO`Xs!}{#qd$WT_5bLFj)AU zZMfXx9>+|Q3buY}HUuo|etRDw*&oLx3~8mug@p;ZeP!P1;ZabCC%72P7-?wWia4>qgrHk2t^UZjmSajW zF%$n!Qgp<(l za6>~koTTXwc%!c8hhNknjx&_K4)gxxT^9fKUB$Yf#^g_Z%gBlf^%%XwxmbZNHruVi z#EoS{E8rg4Uv%g(SlN z_@8;vzi+MM3+7LHAMp^gN}umugs&2Vk-~9%gW$Oq{j9_m%YZ%bb1n8C!F=Crh3*)C zpI)_Z$SRW6*~|x|4I(b0uf`g$osEr=E%GDg${eqS1Jlk+2gsj0dl9s@3Y3xRx#s5P zfNGCl*M{QS18|y#jEr!P*}Q^6MCWIuhdgrd6aejFjlAW#BKCPn^BtBW2 z!HMcB=C5IMXx)lJiz7!R84ou?=XLFe4 zYAf*%U6{>E%lrp8(R}KCu&EaGM5j}APSTqolx)1wG@2g+K*7o|ZI{%oK~TXzYm1Tc z6Nc7L?w2b>pThs1@vOPPhOsD`R#mLm<_0M3&<)s~FM;n9fawpu|nS#DitY{OE2-3x0jf;zoVDw&)yxP&4EKnHt| zVJbz~*`WXV5bs0% zu&@vvD>}aC!tT!uV>4efN>GSbPDC)!GD<5Oa(Dxu%=(s-7ejNiv$N(^Q?-`BAtuBXfi1v}}q2eKD3{BfsmlRf`?^e2zXpkhCuC;iJM$cv$}xKG{W#Ok?+o@p<|d z5z3TYXTCH4uod?{lSb+pI^Ji9Lp~9i4Xu!5<8*kW%;cL>y z^KE|OIG4=G(biAc0t%lN0Ja;LiHgkHzGjEQDS=u+BGHPh2!ij%suPsm* z%qi1%`$tlMF{A0pSw@DC-8Jpy0BDuHV?i-*gladLbM&nCx%uMBcpMEa*_15k9 z%jAjrJ*@J9FXnDvo#E+*koeD!o^g&IdE_|)kN4*SAI?%h%*}Y(@*}pS&3{Cde=Ua& zLQQ==1ptdc-qod(#&mywze-Qtk}9d64=fiTK{PWbz}Z^nLm9kxJ2MvRrthlod! zw7WvXZLjFq5Xd$tCbkL(Ehcf2kmr*W8{lCWv*i=BJL{?~Q}8F2Fr@Z>M7hvVy!NhW z#|FY`vW*>qlEk;rFI+KPS+cZ$DF;M zvi6_y9cMHyR}TCDK>APm{$Jl;h{6D;uSywNzid8|$iWc&7qW+MBYjacwZuZ?I-8vL zS%KB$%Sj`bo-_p6EdHAivC6D(FQi=E!XVGL_xjsgP*;`HWYV%P6tAg9ahNZ``Bth?5T_;Mot&L@wYYcD%Qi^Yd~GJyDvjOi z`&9%DheBH0fRb3y^QD|IkWL?kgXzsAP1b(>%A0zrFVZ|gO2_vs)gwn3H_H9!^-cgj zP63kAo~!zmXKwyrCk>;4o%G%Ymhbid{bFUu3QJ2P{yw|Zi&IUFMO32De%+Gm>MY`E z$%lA06z5*>LJE(r!Ey?pA8>&OH?MXq!wV!k9H5dS^z+Dk1{4{ATVqv1!Vv_MoS>j! zWqo};$6e))jI?yuTC1l%OLxW2aTBKQh-?@5{J2p1^-;1wgqf-KS z*YK+Au~G#f)ylsX+zxBOxQ@&36@NHD37#yrI^i zE)3Jp^{|dcQY7^$;gP8Wq)q{y0;=Io$JLVuBS~-DJ~Sw36$mR?Nn(Z$!TvU&XZV&E ziE~NkGaw>W{{^suQ!Fe48@AsoBd!f;gc-_v@L}O@QgG40;RXfI5_L)c0<171HJrC# zNBGzHe6Qf_VsUy|xP0R$epThYOqg*e@jQSK{}dYi4vcG?{NynFR9>}#$Vs4g71{dd z`RGqZGiGeoyE}X??0w8b5pv1J!#VdYWjbmytt8XBB!lDR>gqGZv{VouB`YsX{?=gg zaa*M2R3m(Y^Rw5pr2HS=|_<<-GXa~_PVo_iwuiY z_(424>8asFr)rh1LpU{cb?bn6v-`EY%B?y-9|kU#P}Dq~35U6PW}Vr>`Xa??$YYc6 z5uSkCrW=$9-|VUGHTd7Yr}RgV_k)^+@rAa#{`0}-prO4K@p96=R{f>c3~A26v1Z|p z$R8kpj*eHPEkh^Q*Zx)TFew%e{9T?X@RlQUyMh&vqqyycZVMj6-VbN$7IC zEC)G^9I}8;Hl_r_z+5Jw>T~`@-6y)=Pa@|7(1xaHrlB8gEZJIu+{o-paX_SMV3FZk zgsTzs<(O>a|bouk&vtmqZ=ot)7Xd-+*N0uFR29XST(iz1lJgUblgm z3uT0Rvepl;K8N%-4D)B>H9%Pdc;wpiMAP>mbFWeWw>sM{aqvfN36eN?$Y|;3r~nbQ zo-4Lm{p1)>l8MAmw3ui);?lYi>BP_t4gMaTf)utAQ8Y8i4Rk}_l8dpGb^3jg$|b+d zM)V+a>ok7-HMdL)&(_q`%VgK^erpoXYNd_HNad;h&IbMgha^;58=_Gic`^5|2-X^E zimB;~*0M>2FP@k6+prq!(`4%dczbNuDGAGn7+j+LYjBG^jfw}rfpLJb#;a{|e=>4X@k-O|2jcv#5n$y72pv&m#lYoxgHenDUZkWcw`#OgO;z=i zXqet_^5taJJk>L0&Uy&L&d9G}k5lY&Nbr8`y0PR2oYdJ&ZEbA=U!(Tyt*x1sZ+4o$ zSeXSUC*$UON@Tv6c0$D-TYR{R>>-*8%V4IToZ+U8gwod0VPYPP0-MeVKn0amzOl2S zf1X({-a*Na$bUgLf5ZPS!2^i_@xxRKTznI>mh|N>8<}B-#?1!uDo&k7)wfDPg{oD1 z8K6khpG0`ky;(Ot30XUGMn@doZ^=>1Q04Q+8{(iNh!PD4Cj?j5&ER#6caBJXtt=d4ph+Wy#3M#$LUmpyHDW-=bQ=bCG-g{JCG) z=@z*dlcO{vOQ`f_`eN)tx1^9BmYOP=kdW^%f<91NimIA0&Zoa!_^Y{f&`d9i%9PD>R2t|RD=a1K%h1iS<@#y4UAU|QK+eHA zSTZ%XWf25X7(DHbVHLa}dXpyL%4`@JX}4-YKK(swN<;pr2I*9$iEq+8KNa&Wl{R(O zAWP8nID+GA(L_gqN5_OCtY{)L3yVE>bLCh^2M4=@U$2zY2-kS2Fm{7+s~mUa`Oo?( z^5$8%1b-v+Uo~l#g=Fg3mh_(r@y7T;y;y<%Gla2r!b>JgMJ!Ohv@sW0r=;$Jvi7q7 ze2ge|CyN*N#>ePo9nOIY_2jeR{wF+-;x#E78wPW-LLSGf8FI+df^zieV3iZ3G*}W6 z(7&P}AoCBG?x5_Z$qO5IUY8c(hG~4gYGG!FwNfyn;Og``AD}nFTUa zGZOx@RLZWPb4Q3%*F@0M%W!}O;wD+$@uL6rc@cby0IY}F%OXKxX+Af|wz_ilCMP;& zL&Lvbn`?MUNa;hupw@v9;%NBJHxH?1N;B%A_n)+DB99HZX$0;CJe$``Xoq`q?~!Mv zkeEJ9y2xM8jrtbB&&7Ig5L_>0KSufLw1ky(Ij@zCkHcS_DT|dV8|=2Sb3UK3uo$P* zCXvUqfAr9dNadpmF0_OVe(UZYTXgNWGn(DEmB3i@3XWXAmx(+i6|pmNhHPh1Rs4nL z<(%nv(bXgvo3M^70na$K6d3Hu*?L=&=?Xnx(XCYXgQwEq@)+Gy^8ZfgyfIErPCDlK zj>M)ur`zp*6HR&eNk7}ncQwL5U8jVqE$pa?_a3%yHHPo#XJ)7=NT5OzI)I2k4v=_J zG)-T>hC(@-g)yBZC7*p3bGv*L7R^h+8^oD9@+S1Y+M_0=D;QfNko?I%j$qX5@y-pj ztL><`;SgvPIw$Ap9|$Sh^I_z=iWwM5AuQkbuC}>5<{nAh3#NY{(?(81qbvOmFoS1@ zSCYnYNqbsCH!!c9sl|j&;>djlTms*(M05pV@?^&$57+|20mV2`oq;$^KHo_o+*NLF ztz=d2V!dr_HIJB21)BycNc*N8SzclM8AG4S0DlBY;c|5r^!UY4BL&hU+vYWHSHzaQ z0BlRvyj-Fw_k#XNxe)7UbVo1plT2G*0*_W7sE_M$Qu@XSlH9@X)qfl*Q(})&LRq0>`8Zytm-clK4a!iYOgKTDGx&N%)YHne-1I@{#T@MNhjQ}W`2*Qn4V zKQP^yE@8UyNgfS^AhW5^X~sbny01LZ^$=Lr8yyA~l`nvI^89(tI<+$p38m1B*_tc1 zIcyE;y7%582x9l%5g9-J39kKxvQcwD)~1XlB~d$ISkZZI6nM8WU>Qp0=_wJjB zQK-I_>2MY&^pg+J>ed!#uDksSCF?v4=%OK-dmG9FA6HR(sh83joc)6q+}THUiP;OC zFvPnK_Q1Jha7bHSGrAQFv}*9U<(I$64Pi9zj{rg7Ew@J%R{D3nbqGPG`K>7 z>U|mskA%dIOEP5*5Hy#gb=$ln_p0nt14@?FgMpZI*K|Wu4!z%34ayw z9U#dywY6)pzV{TxB_-G|CO+@ANF*X_S9v!Z zxnyFNc3RKh-#^oYap?op4Io7a%XO74gntXI)Cki_XVD30t<15&#W8tij=sH#eAiFQ zPu^Vt#>{AGUR3VHPxIii$>bnt_`NB^tafz3f^08Il<6LXv)>KW)9))^Qm-*X&&a&9 zgXB99FoE6(_rnV2wpk1-ZenC?W{_HAsAPneFr>)__xJK&3i6}*+-o| zC5egAX?I>y-dz!o~ z3$wEw0|Q0pvn@tiC86cZIJf$ylCCmOc%V_FUq(X{kkTg1%QGN6Z^73PS6su~e~A7873GU^+pUFScws`O9NFt_=lZ z1zWSY!3|Vp-{doRAkCk@W4fhk--iw|(U6rzeqzAvT(uQgBRdP;ENrb-DbZTbe0zUp z_@d;Zu8#?<$$Y0(Hh7_rV#~Mo`#o z(iCT0>63SyTNQi}9>U=9`x|#WnSq6c=23Om64kPIr5%6*C2}xdi==lmsD?&BP~z3* zdAXve?tY2QqRP1Xpw-_o%6=zm<)u^WRNYJ< zf4lPx?uY41|H(Q{M)k@ZHJ?5V+@N)gBwvF>L&XUO=>PQiooGmLa%3-^yD|3-6B@Lu zOh>y$)B|_p`4hbgrsVpWONIR+5yhQ5V)9*h5M(S}Y52&z!Wl2Jf!~(yX*3ZmEbN+$ z-t9y*Hh%D;LuX9QSomm32=lEC4^QJKwo4<9sRW9xVZrMgp=G9+wN0vhyGu>Rm<;|r zcz2Mh=8ai$y&{I^;N%3bYJFZ4863e;=5*OmR~){KZF-c>qddCd9@j7+nJwIvYHh!C z(Pz+3ljaVaa5nPt;s;a8om)=2khmvh%*B~uS$FBHYHP8pB?M|hKG^?<%JR>B z`118Y)W5s{grN{2=^Rqicu(xEJ^1-{!fh&!M|=pC1jtvdkHhh@6bEH2v3%(-xo{}t z{&ht8r<3O3upgXJ(adkjENz(xCM|?#m z$-+Z@|Ni~jc)>-R!8*WZxwTS-Krs~)HGUb|r89oa;Shd8;rNYyKV?H>qca;qm_|sz zHRe+XtD%8sI9sUwY5xmj&@wakuI0gvEviNZ5A9qanT$tB&0ddC5%Sk4*c0q+Pe~TyA~s+80J!3|m8h1$ruTR|WBcVmeN-0Gs&s|yGWO3x;~808YHIlA zEMk@|N3)<;!QpD~gjWT3oi6wlqcL(0Rq`;+b$g zre+qY#kh1Te(;O@C;8%*6Ygc@b!CF1&MvAvjTHX2r4W39t1ooy^V93wcp5?M-mqIf zrFO?B?D}wVz8Se~WeZMnRd2|m?0a<35DTs`9N)bT+$yvU!&a#@|3R%h_1cNFH~CSpjJOk0 zkg`q0-k|@J=w`URKV4Y>@&38%2cy4Xqj?DRFpM2K7LuxhQ>GHf_LP7IHY}OC0~NtUO8p z3OJn3wjM)Nmn`HUgbPb_Vsqe_3U3i1WTJgDo=|9i{@nC2x-^gEn<(S@>BYk^uY{!$ zT(ku1JEZeU+ZDn7lsB@JqzKxL4zS6XhPhMR`n`+&BDupEBw@y6)w(TSI8ss!FReAN z+zbL$S*7tnJ)M|TaCg4As4XTT;kqy7;laCEQjtHik3BA%q=FH!Av*;S?x0}3xZh-tM0bzylFS0WbIcEgY&p!6Vvkp$6KxKZyXyC z+8=+A@Qyk!eelJ)(pH?m;Jx~H9$=FpEGh$6Od$8goNZlG5y{HgVv-a8t-6u0q89d< zMUQJ^8PlL;I>h(r8KG$zLVh>2a1C+Ry1Gv~>OND6I~_j{Q14P`Igq%+q$iqR7km|$ zJEPM^vrnUzkU!+oU^(pE6SJ#*fJ0e*;#0d=m&Rs45D-I63 z97({oI_$p|R~Q#+A~IXLKt!$tz2Jg;I}xpu8Rv!K#A3{Cp$pc`i#6RPl+i)1*;&c( zJ|cWRftvZzw&rGKZW~u$_-MFKEzfd<}?4J{Ki=Ygg@$-u3$% z7adJmA2%+TD_8yuk$?4HjAA+l;%O9BRDyv~WK?5Yl_>bxv}nhV9^LL0NZxMW+Fl&D z#(IM{o0P-RCnBnf&toWO47NXVrU?1Au~jiiFyj^l1qG?2!o2(#iObS8Ed=~u7DF-` zzKz|=i>%In&e=6zBF2BNWP*av%#rh@M%&(*wuhY>!Gl<5wndaYvIn8 zNNjR>T9oypbQi#zEA7_0UbgcvQc;QH#bY(uaDc;=uHDr*gOv_nQ|1%IdnVV98-EJ~ zPmlTq8x-IfxjpI~e<~}(b9iS4pgB=hZic6(ihnswP@mv+!yLnBL`z6+3C_Y$4}Eu8 zQf(ru5ATW;4bh~Ap4-|Umy4IZ*Nf1@jh`4_Uqw#(hg^739_Ny%N%T5f}4n-4wV@QylOd9#dC5#sAdf&qg+kb!^v7nukW?>w)Viy-9k z4bZfRt5Xj9$v12n{&i2Iq8+p2{_`QKFXoZWu*y*g(&p{B3QE;fc1^7N8Q9l*c~>{(TXl7KPc77+ayG2WzJkw=UZ<;;OvX7!Z#~(k=2ll@jEszc zrrc+nh7zK6yjs&y;)2EXo=DqtdA(b3STyNiXD-*rE%*MP4o zA}%g&WV*Svbq~14lX=UEHBdQAAGACfw6C8T32byIe5hLX{QHS53BVy=nFln@S{%;r zoMu1GLQJD+etCFToINjlKGu;tK_1oYBUa$JegZaO4v;P%CA zDcXIy2m=Y`xxzd{B%4G0h}Gx=I`OVCV!w5F(>5JnZ{Gmj!qSCP{R0^CN-P8t7`{;{ zwa~M<3C@SJrGEYpLr-yC_)oBA6<|-}Lx_c}3&c%^T z`mKLan`T3EiKj|V0MYAjuRSw&5HKoLp}@Ou=ja$17zq8usVr>~yM~!35^XYGFKCeg#>i%$SJ=zF?NbGiIATuZpJqM|=Y&j7tu zbQ(yH;P8$mdby)wV%oyfY|DTCez;RqUqFnm!d4Ty@intY3>m6nRnG6#P_5ELjgLni zl$JjAJ#LSE;u_aLeMKg6jAntF5YP2WjdvPi6#542KH^Rx3l_eC$*<>Ig7pIXaqXzU zoDkRv#sYKPbwFRL<5l~F)p6^^aPl_{FiHsOEO!1f=BPwui&j;Fv9FkJh!?Krl>ywy z3+4?iq1*DaKAW9jX1)y8W2nGpRXI`y2APHjckJ9;gTdlmoo3fhbuV%21z!Akb>uGd zIxgu&D7$Iw#F5X{2G|Y{^jZ{kjnzki&~*A_ZC07LSoxkd7 zl=^Zob+#)$Kfb*J!9{N7r7iO-ef=g}SXe&6X0ZW%tG38YzNF{a=ZY^u^*yOY;pg?+v0-7;yw0wz+%SQ(f=%Dk z9$bmVV(Cj2M6Fk^UV&*>Z>dMXeo8FUIH0efm3aS#deDO2z~O7~{J=9_n$~JVZJiG> z=NfAEt0yA-w__BlTg&}k*$q_w!AQNx2As~x&%WH`^%6(FPjdFib~I-<%Nq1k){Ud3rh$4hnub)hGm&^Pwg>pyZD z1{ySc{D@V&_%Sw2SO1^}I^5Byz3Rr(Gu1RL$pC4T%k*%5lgds08F6OU)uQr5wt5DLph zBFN99Y#w^vIdak+E4Qj6Q|<%O{ozh=U@(X`rLj+x*5=QI=>N*`{*ml=z)-C2?IGe# z&QDM-7zs5c6(RE}PyTdSckRxvc5jX>C3lbzVx2imsYa#7C@Q_6_V5M&l|2>y|Q#CUY(v{t1OqTbNj zX?&~On~-;;sAh{CCvB$GnrUf@D!P%IHCQzWGN^QJ3K{}%y6FCr*FWL8e5&!{gVU>k zvC0}D)IPPpA{i+xoq3{(JQ#}LcKq-dD55TQk(`l{(bN-DU0>BR1ZHbskbG`u6Bt!U z-wcOKM*W>S?S;^Q% z(c=I%zhb()#@h(A0$-+Hx8et2VPSE=)K(&bJ!eTac88l;8x{4p)XN`1Sh$?ncWX;+ zY~Hx>B3biu=hqca9H5EE^g|1kg!0CG;vbzwDxqN<`Ssqa{skEkXxH_Sn9s zmi;ws6Yb48k`n3A3eZw}s|kvkfE4Ii{JoFO@F{%pn|8_^K_t(JpA#ra{q?S*CVx@!x4Q1hPtPNgGc3}+9 zE*8S;SDO<^2xyD8C?+eFN120+SvvOqeGu9HCbqVWg1_-gZKH+D|} zQGC9d(q7c;ak5UU3X@>M?lk^ro?0RkRI=uDPT4$+#A7etQJAknl^i-wvTY#(U@>G@ zv879x$x%#!XScqAVwT#wfU`3PvS%LA^<-b$pVnwB5_FPvCEWu4L^*m%{C9oE-K^H~ z0$sfH&I}HU`Ps|_--GQ)T?u*&2{XoI@BK+8KFTOEX$Rc|@1ec?rqY1JDB&(nXJG+_ z6nH_znpEH<0}^oW)^C%E^gaEt;p)OofHTboRQA=9T54v*UWcE>P|4TfytOaR%m# zK{Y8KPwhPl=a2$EI8ezKnF8YYH$%D9l$5Z&x@ThFk&manfXW4#=&;iD_wf4Hyd`++ z90p{K*MvjN$4Pp`v#zQYxTfu zs{v&#jhLAA*H-*q5Gc$e-s>sG$5F6IILI0Y#exKl;qlT@ugRe5B3}4uB5h#T)rFr% znl9V*akqpUXlN^Ij<*3j#s;97|EW7}XfZ4aH7%iU_6bRo1UDw^&wcWgRdx)*J2^S& z1CQt@JYijnX6=5LkfU~zg1ncwu8Xx86rC|6T$NKLK|RXY5s{I1lsQvHiD%VW06KU* z&6hA>%!a_gL1w&L8$s}`+_^q{-~a`EPSWUnF717-Ds*%nl%JpZc}qfA@h|G{>^vV$j9m2`-P5hW zKRsnn;|;n0DwJ^5IWwb1yHwYYF>RGZd9yfx&a7D5CGEth<%^AlMLT+$l+{M%=0!IO*gpd>zHW83tl3eE}H3yZ=YmTSis&wrk(g-6=6p>F#)@d*A>2+536#_lslr#o<_UUe`S1IDSXHxv<%$?7s@@ z|1Pip{bwFQ@Ml^TDrRb^N+VJyOj9*I_2i6RhY-&o1Q(8w~v6_^=5G_(YQ^ZWWqIo@zJs{lCYYQ*K^fQxuL>mg@_77DjUl9usq(bSR$KkckQ zZ1hp3TtA;WN*TKD?6;paG3;`OA?u*wO_vB);dH)1zKBK&1t8piP91f zKLJ>n&iW&&^9;@n0Bw3QOY40C;9WYy71AfsNiV)0=l!bEFwk8EnxB=6i@!aFDdyc# z`gkuPtIU7v!2c1Q{_9R=gF^bRc{}Nn+DcM98gU8Li5tsee1&SJos*r+7i{t}35e)a zDOxwfP*nee2ce#PlG_-f3TstB)E^V|InY|WzXU^&+{WUV^ODt@^4dx z@3VQu?MNP#2}7US=51Uwtr;sKSEtZy1he^DYV7g6Nj1JI+g4pn`(Y`T!(n~g>l_&P zu5Uz^)aitoR4F6+eD!e|_zD6ae!>JA9V-lI*S)d6$}@V|SiwYzXgM8jdj^;AaBV67 zP%5&fXsciE99gZyvwAs9o||{NrAqyt!2q3sj5|KN2Z(3fveClguvOx7+9ph6HAWW| z?HEJv@9*C)b2F5?*8`b1K0Gu!Q?i_A5ay1%5^3cL~)G-P7}jlgUHa*kl24T zh`7B6+s?7zekn1%(9rg}7ofC>r~*f+GvahLHSu?U9eyL|t6#^HxxT$0^<(PE`9*_e z`Z!02!}B=^=R5FDo-9r$9jRn9NH(|H(&4kTh7c$pmrwQdNO(!tNl94-d+)8}{3C;@q*dQCWv~fV zQc6k)@9UL-&CnF!^9cuB;0`*QyJ#P82kh%@DD8w_?zZmS<&ai|JV;(o*G=vO1U{HG zVB*akUwNFlyGm3$y`A6P2^E~Jknr-l%OG(zGb_~T1><<5F?h+aN-EFt1Kcs)*4!K| z)3?Cd{Iz%I&dY$_Mm&2Kjc5aXwrd3;pjMlb(^>TIkfjR1NSL|U9sM;hoM8cdj-Q<_ z=YPDr2`MPdyh#OG$(8K;bRd3}zYD>LDudb%4lHggRI_PqoX`@>NnebhzNm%%m|R6f zb?^C7dkx6Wya4U8jN7}$RUbz5T)r4v+Ny3w4#M9<3N#Dpok1VC_ACq3A?#PE7#|`t z_U)DEBSYmFmw6A8U#n`*46u+|K4@!Lq<6}d$s`3*xQrW0jGZ%Pm~Vw&d#=X61#r&$ z8T0eK)FAv0h8R{e^&rw36xIzDanxEZET->ysJ_ZvXg-BiJ97m^Lu9nd--2)_A{lqn zVq;@5Z?8)U&T936 z{SKy=CcE!ftup|+e&5~QEp3ZvmkykK+`m#%xP7~n|MZmrZ6nGu1`jnejTqq~`XXeF zMdpGfc;k0XQY`Z8&)GpkHDXl8PGx#KQ%-z1ivsS{t`Ry z*qZpgiY)l)+~hPWb*j&4iVqYa!FJMO1w)8oRdll5YyZ9MuHGe?6@L_|cOmwk58u|G zDnq(3Sk5eT>K7iLdwpj?*6MVA&mU@EOkCVB8FK&W0uJ63Igzzhze^>q7^{VqNi`js z9QxyUG;4xAlY>r;^Gt-A`ehJ zz!Duu>Z=~Vi=pPgN`OcL?hC1iOu z6E&cE!i{Csmk}0<^GSWDSwP87P--t08!G%Yl(?_A0qN^3I^WEjF(slq8w2)VR@lpi z%WRx9@PH)w4?gF2Vocz=#l+fIcuLV^%^@zTW7DJ3(br@qnsQAQ$#}KpC7jSQK>s2g zPqXn-x3WrR?|yx!$iTaI6ewshgj?S6inWcGCa#|w+JqH4Fk1#})SKI2qzkKGV;nb! z04`>7u9qf6aAs?%u>$`bt*XWEOC>Ct>4q8=u**E$z;Bm%A1*%r<`yDEJtFR;gG-_6V>~1F&wuukIRUFUlMc+ zH;m@l3*p7afpBo=lu;_9T>#*u+Wc3_#veLcKtxUG3M+l04dj;z2>w@7@;`f#S&6)c zCj98WU!mxwb?v4TLIApFRvb{)QP3M`&RB^T+xMiswoLPLeIiveFn}%mA%zIK>88qS zy8K#Q2s4AH1Szr)>j-><+MU86$chDI&h*$fuFPgSLOjh;a~=J*h~*V?R6I{Uhz0y` z`jipZkl=fvF$`8rL_d=ZweWjqU84v)Ma1@c&(b)TBVnHoQi z*>dTf8|;!G5^RGF@f-X3*O!uN% zRWKNGQRpM;=Vry;OP=S*1rXcx?e&&6hL0^^iF_8{`11!TJDb@c6vo$*oM7NWq)tQI z#3M*_Im~_k1HGsy`MtOInHU0#m6gdH!fDUi00KjsFiGFdiTfegX6C2b{R%~{?eB1V zWns3R90WjwM?T9#3=*|~BC2JfI{fWiS683kBaHiV-FpIpP*})8pj7|Gsk|z9>FJsd z_x=J~nbK(%d4tGVL>+|{1VCjVr|pmM-QBjpq0kZ!g3kD|EQ=oczdsisQ|u(hx;YWB zawcYDDzI}A{3`6*P`r?t2;0vLw8Dql9X@Elf?%be!VuIpb%9TFSU-O@$i!Z~bX?;C z!ma$p)8>n%%tLv1;+Z1_sO*kj%-;SMygkcJ&IR0@GI}m@huoRlkfP$qZ%66)bSEb# zbc~EW4fab;{J__yeFhnYm)b6@UI%PUfzN6b2PE1hHcJhhO_`z;igl`=4=*=vV0RTQ zeWZU-{&Xv$=JJ&aT7p^cIWg;_>!V@g0lKi_3ClhrM#SZnwIq^f{uckQy&x}D3a-wo zHIqR&XmyP$Y;R5G(J1_plrXReBORtfxWa&KcsS%+YE=Br&LHM_SCOq`)4t`qDIhD^ znYoIj0qSFZeZ9;^Js%-Z3%Bm5rCJ}Ud2gTYW1bIdyr}Iv-Gpih(Q0<@*xH(K*u-#b z;IJdGxSmKq-;c@)*C0*J2B0~g8g;G0M1t)WIHdMTew~Avfxb4G7FZE&XVyynKeNrh z3q6;O;@MJD(ArsiBX&(Gm|XSuot#KXbibAER*nKa7wbD@#eIIf1e*<@ z_r`68LB9E|(iwa<9F~_*%8FW>nuMF%1-R3uu;lhTQJ|9y8$-hn~+n^3f4;GsSuDoJh{^YaR04!I#T;9B;&RH-T~xZ}`b zx#%n4PiyP&i2(uOV*Z0(dc2>D%6=EhAz5=Y6iU*DnbHbEU>`7airZV^rFm(mjpZdh zETO)dwY10`=|VIi{9#k`?)h^xr!{>YD1%kLc6|Vcef8y_EUz z(z~@0O=7!g8LPxMEYD7561NIK3~+Zq20EnbXT;98I_s7Hf);3IxW0zUE?$xyEWK&e zKM9v(cCcHwBP6oP^SB9KAu$6+la&8(<8<^g(SXq zRzmd@wmDy`Q~s{4Tra{*0HkE>@BAUj^5!k4FNgl-(*)hyCJ4Eo^3R++BPJ&jezs!2 zd$(7x-N|#b)QB-!!n5tEL1}&OrkG2P7C&k=C8DJ1E-UDU^_>%E;;vF&H=S-+t-DktG@ML-z&Hzm&FmY zfa0pVrd1_m9Uvfa2W;6UdSh${myaw($?`#%0zC7OO^48|ZK~meHqVMWM*FFo({EUe zw#8rS#8Wh%$X;h#I9&ln@9R;5$GGe2o)p5;E3lt8!|#Ng6T&7K)adi<95kcM<7S?7| zDYd_6X2NGnXa+}VMl3ed6&b{`v{10J*w)7jI*;DR)L)J+IrKT+koW$gj{GY-@9piK z=T+I~Of;|1xT3J@@^RzVmYCE)VT+Z>-QH2HezA=a4S&ZMEak`EHN6tKEs5ijhC4&$#p{PE<{`)V%^vXd?A*8}UpDclsRu%90qp^(Vo!E~b{R6%p z65pqq>c+h^8JL)Q4;|Zr#od_;cr8HppY=Dm1yL(XSTsC%#Wn_9)|imRlB%WW3HGTUV%cH)@kRR0vTJdHjufI#5gUSf7OJmvmV|K zN8#@XaVQ{`NIF-w;xo3L&w0-UHt75^Owe;jzBYsM5gWLdBB1NiJ&tf&~S)J zO7@0r0oer-x;V9`F(~3r@?3yPbF$2&A*)0+1h;E#OS8Q}{%M$O-0{nPf5$VsEf(T}SLc^$EgBVb;o4J*F{R=>1;5 zyfJ@#WY3b-v_kOe#T~EUQ`-Tw9)t9iVUx7TQ0(+{A}^v73)@jct;9arWCx4T8TyPm zSK9B)U1!-T72&H@=$MA!X@a~_Flv>N{9WP9gOl0lY$i3F(6wYiDA}PU&5V$zRRu&j zy`V)#Jf>z(*1tsTBu1BlvG~RJ(30y9I2n+ZnQKu30jmFW8~uGa@L20G4T2s=AFMpz z|K^~YSeb#bh#cFA%pZs=ng+RfXokup8zy!iV+h9ENLswhZ*3(xws~*2$`Mx5YkS_Ud=t6d(B+)Uu;!Gki5mAMVRP% zV-ZBV?k|wJ!24Am73h80*>_9DB&})Mh_T-Z$aLCCT&W`QZ!3pNP|!EthA0pb?h9=0RaT`L#B>s=MY z{M821Vg${UnV9fo30Mf$N^pshv>*!9BD(M0in}}WJ}rcW+^{eM2~J#8DnEgnz4Pkb z2Q1LkWwTmU^nbkap$L$G-A7(YB`VJi>usyJ7X)&cXQO?oqtDG96&p)jXD_4;%;WP0 zY|Op=wyl3u1dG{;>G;Kgu4i8Pco;<)Mc9}_n7T18oQz8|SG3>PiW^)FGC!Byw52r- zFXVx*oA{goOGG0nNyS9++omJJhu=iwei>wd zAM6!KKg{2V#F2>#EZm7MmOsG&Z|+pq8zj)w8H0_o>7ZAB+ zR*(MtiB3vNN@h$@K*JKyO_fgULp((c{GMiK8cj%blZf%=lIq$ya>iW3SCd-N2gaN% z)HVAW9EJQgHV+`elBOG}Q%iyg`A!Yi1*Cm!ihk!S22Xav;l+&cq|?{{CK-4tFA3IM%Y^8$T^v zVuc<(qD0R%Wm!RrRGyC+z^v-gB))7YDtcG0E0C*e^400?%)~9SkHTvP-z)5)u(F+u z$Pi&U1JpOol2@<8m(yowek)|=2up}F08*9M9nNqmlTpA3J66Y#72n#}r3wF03(w)9 ze|00*4YWz$^h6TwB}@Mv_7YSm?9P&r;=F|aU6%a&mc|jI;NrpsqPudVN@&ZKMCeW^ zlM4&m@n~34h|Ar{=&6!rDJ(D?PAwefK8#S&3T?ue*zk%?l}L6nS)%de)uqq&XzxbD zTEs8EJ6#0<{zD75y zZ(|hznbX~&o(=%j;BIF;AZD{f+U<`UqUq+WYO&NzxC9(rQvXOdfQya%$-{D3SXg)t z*iY60qCKB)F`*igoQ~f>dZ#O&d>b@o^Ubi6nRsD*BzBEyZN|i}0$(&9cACc}xedKJ;CoY`h4!v0@Qv_DKy`6Fnm}95AMInd zQXP>hmhiKEo62yxcadCl=uVg6sXKSr;bRf)g>d${?QkVwUHIQ%Yqb%}IWP!Upz~3b zBW9^d(l26E4VHY}w)E1e7F8zmvhgj+wX9^6O)BPRZDLS;X54GK{qQpywUeYSBdk&~j@Y7-b zNKQU;_9Q6ij`{jk^;bcMq{)U(<0qUGK<;@6Z3%3(=p3iI8EN7d+T|xy)xyU588Vo8 z2A5Yb@rXw+Jbsz55vqRQ50TqM5@q#_Vp;ueKHrn{Ax&G?UqEG_NEb~UWduaDp#Lj3|NWmenY7M zKJEWB7M*eLJ3lLQygX!$=m#EsAEas8{4V;81TjRy`*#@zdO~!v?E!R`unXy|RD0s@ zaofqFbkfHzmifqf3OEa{RYjHsnVc>!Bl$XEvSnhis4l~`Wvl9m#^GkQ)FTh3yb-w| zl&&7|WWGo20}8?YEFl$o^~iw_Tc?KD9Y=<5V-eGy50)l4d_@cM2UfpLS&=sp zV^bG^nO*U28p|fuoy`a>oge!E5nSc#}B~n3I)*SAL@JDgs6Dk0p2(!J3TN6%EG|^5VlArkhPsyTN-jQZWGxg8UVn0`D5Rkrk4%f+blyC5+n$$lZ&0^X zFm8fyvHB4qFg*M%UAK8S!FJRYYWG9@VHk-Zm%d*_t6a&iPi~%`TRZ8mbWz5(X%V2l z6vA}hF)6ka5cpJw%;suA#t% zGG}Q&pksvt-N=)pi^sa%l^d4w3Ca*5gWA4J7qOB-lW)>xJ zKq;7NQg==f$QcDpWC>=6JI6KUkKM#!B+uV;0P98{mNR(*(4(zC`@ zIzvkgZT>-$Uz?W4(rrd?@(MJ4j61vx7yER-niJ6%g>H9lF|0uUZlK^7d7WwZaMGn!$H?yBV}e#MBY^U&N*TZ!s7xR zwnL3RA)fG0#ea4DAUD_sy9Fv#o3@*ij{Ff@i+jhBiP|@KBFl_;a#&2a(nl#8?~t(nF?bmEghmOs*N z5q9m(tIrA9M)pv*v z6t+#)sp;s@AL~0iIf96MgY$lYla0%v(6B(nVc~|do7iuE4x_7N{dL_ z`U% z-EQpH=QaRKF2=$Hn3*B3FOOXn=uLN;A3ufyyLptq2Jl_>$bZSOv)R;NE2H+3*nBKc}zugVwpy%Dglp?AA$G1^~veT;6 zVWk-kFbSf66K;XgKYeOYFyzcc9UNOlOQR+RoT%v$B$!S=*ndj?dq@Y73>se#01h+e z;Dy7tw$$*FS>@^4Cq8vYPaTn85hH`}UP`7>MAJ^Wcu`q7O$wNU54;#;5cL~2pKHIsc&t7wGB313LLzt9jH*n`rC?3vq-BJ^A{!Z=~k;& zVBJr&iF!K9XjCbuRx{k*PD8Uamaxi?Ncx(Wol132a7a!q^;=e~;>;MqGMg?m+cMqw zxCxaZpC@WbWG!i&!evHq3e>zbkXdMdk+@sP;mcbPh|(S>%&Wz^k5;9y>9@ru zCw}VXA0A*Xe!hp7DAc5&Frmjm84E#iSo;gV1|@!K{2?E-2Oeey-2^T=EVue2nKNPg zaP{SJZ(gmrU~ZAJZubz6wr^>s%UGR{!xH@TlGJx{I!jWz+y|ffpZd-N%v>NFTJ#DD zByqwvgyx634v3KXy{_g}p$Sh}6Lt7Dj#JhLdX5celj*ha6uEy{|In&6P6rW z&@wZHPDid8g{@Ebd7Jv>i`YVONY;0f4U>%!r2@*nm|kBFglOYu-#>j+ z-MgY-eC=0vT=LTu>Ji*V{dd<@GF8!?l9+^aAoBKuBT%mMVc$WKRs zF9<7k;&ZZ67{uj9Cr%0~-QRp0sD%8SULP8hJ03s0O8E%GL(|}E4lvImXK^PnFQSU# zSM(L?NKqk~pAUn6LQtzmLmX(Fz7QDUh4}?QNVT{qzf=4$f7&EW)_5>)ZqxZ4v<^K5?b|d{&r+sk#1eNoY_+5tC z^c*p``}gJk(b1bmoK39{b1i2Eq)%P3pss%Sh^(()&z@gzC8C~NF-3R-dRgwJ;Cop4 zf#8@?V2nuQf`1!@rUplK28ZZ`6}Lngu*2wG%ET;b;ZEFsBTJLU8gpW6U%p8|@biAJ zAn?z@f%r`yifVNE86s#l?-}KhB)piWhqp$1|G?yvE9&uRQcWq9gQ)b*1J+-Uq4Hi@ zH^~T-ffaLD3u|o-sIO?QdvgOr&I@@&Q`h=wpI{#yQJS&V;)gfNbGA%t5WFPqCUOXJ_S=VxBnF>sp+mqX8L#3g zW)nt+6i0%o@sn&VQ#4d(;XtwN0W?o%=h&12ibG6D++N4ocKO0AA@Q0|9Bi?zC?hOP zxk{@C#&LEOO`{*-V>O_&8m(THoIXcw5RgXKF%yShqor3Bw6}KWk6GhGta*qw;sVpt zewi2Y_r{f%ss8{!Znc%FFHq=vb3Rd3Aw6Jc635OWhq?X9R(0nVFQALGN`zKR>Kb3d z_?o9WDBMGFN@u|nxdM*KIL=q8^5Vfv3tthOS#eL2-PyM(4^y*?_^uY#y;uZe{_`Ie z0M9o*XJ}&Hn|2KN6Pe+u&umVK8;QB_Gcpw%@?gpS+1AtD4Ae1+)l_dWg^2Kj^(|700S;Co$1!h6Zxkprht|N zVI#I)jtTO)A8=BN*~|UH+ukbKlJGL{w)y0j$F8cuWD1io#zu5$RR7wTt)n&5u4c{!Bf)5=7A`_O#H2?>qadc-Q=Jz9*c9 zF3$!zzU~PaKvE+mzT3cI9l03ca0xp^Wz@&*p!kv*bM2+ayDQ@FCuZgNv{uCu#lsW0 zuoOn>62rgmy{?$koRes}^;)~oG5$+M3D12+?I+6~R&u7?y+8FQkB%NWIfZ9=BJzvFHdeXyK-lgezHv z1xjL@&yMAP`Z<(2`F_1bO>9bQpqJZI7CI$4H&+FEeqy@h7{BHD^~Z0^x>Pz$5vzS6 zy{C6s$V(s_lhcOj`-~BT%+b*~7Ze&=TEU%Hz~6}9O4iY_82fzvm%%bUt?_u^?7a&; zuM%*9Ly4giNGus}x&?g!u=~5H{lCjoLfQQ^KsrG9SNj*tj=gTo1F?W1kJ@NJ(}?Pn zg}p0x=HX8n4Y;xl^|Z}X|B*k{R7=M?keTviTQ>=kw%!*{bK*r@S(yunI=HqOw1Btl zn-+xeFEab>@z3f0EETk5GXwrblI(nVuiuvSEsH8Yd0i2& z2s%LJ{bsjgGI<>c4ohG~J~@_CMc}BIEzaW=h9TR7F8T!g8UF z2oH~UlQ*{Z)zrj8Clik5iKSistkuQKRDJkA5^8CqxyeLg)hotdTyjEuhT3AS-&PSni*DM^W9jZHK04M9_k? zOIli5KEW@Zm0B}mVIrQ1GTeyY5(-8W-)7Rfy$l<_#%%JuA2V)^suF(TgVd9Lv4uSf zw7SFKYcZUtui)J+8yQjLkvPO{Zc&8TF5Rkqoa*jnW*k1}CO;ut^;gvx!l%RiqWE;- zf~g%3Pl{t`41%$OFV}BOXg|He%6efv$|VS(d3t1e&dz+9b1UUszwHG=YOi$5P&KPw zeR3|N_AuU+mAq)YVI0QkL3wSRSbHnO^U)-F$;Q9VmNxd~XaA^Ybt!Zc8vP}VT0lK4 zP(*i}0W2TyK?j?9y`%YiC|o->-@{#gYg;ucnl1uu)|qNQ@$U<3zv0A1;wXtB=)_*_ z81#?c6M9C8XMJ)kq{(w%bba8!hyJ;zubvT0Glr*IEwE7x@r+Obqn?1W_0lTggK1@t z#M1Beu4S2PpEjp&S=wWf;r@Gk$um${#eW&ifXV-w-~fg^^6<}O&vrnC|Ej8rN+KKI z@af!>&P{iqzP>KkyS4FE^3X2CvQY-1hpNU?&a(w`uHcUJrn|nSy(-QEKj+hfWYNR{ zmxHrU@mTW3F7xBij_I!7_Qw@~0i_pNZUnL=)Q>6gD}KhTxM=Sq!4G=U@W3RUrxJ_o zsrBz*$DmT{1IM0+m(06vWS*{sopTqxUY9$TR+z7@r`^jlAs3ds+dDgy9i@bZ7nn@V zho2z#5VGKu27AF6kJcy!JGR^<@{6dAOam_wA)|-2SgVWs#qoF*T%-0gnz~#^X?HCc zY;6GoUB>abo~89fyfUquo#nbl=eFF!!YW;2V{-G$Y6o)0fOe?wWa)_)#! z*cceQnk1os46bTWK+7*lDud4NYXS@eImON11;xLyF3-BhAg{!3IiMGr-BQ|nzE?^Q zoLrBlpDIaZNh zMHB+9`uklOPDd}{znJ5R6_J<*{gT_W-%uC3N{Eeht-0=j0j7s_s=6sblVXcQQ10{>XXDj9I-edMKECC@QlhNnX;9pjHOlP`0)4< z)ek?`_ja#?`|cFII)gMf9wDKQ+%a-)3qz#Z95PNtu<5nbGxK)7YLeyBI>A$yE4rx( zsrtl0=WA*ZTh3#;za_E27l^86e9Bbrh6PQ^Gi|L_Y`i-QMcfT!VPPajxDnCkl+KIB zyqlBGU|rNeu-#Ql?4t{jDWU3QL3fs82aMVW)Pi)N4thPi@qk&nPK=UKT|}PN$c0P1 zYXiCp1g%W9Vcv^(G$PT5HexKuV}{aKAx;o)LFrF7Zp5ltGjaC;3gJk`pZz!d{rtk@ zc!c)w2|B;wu}GsaEq`^Qf@337x(y|N`heN_m&;LoN{+_odsB{SF+e-4R4fiqe~r;z zA`@cypL_6V?~yxO?GuXQiBCpG5;|v#86&vgu1!ac7E1_ z(K9#a37V=CibuL4=)AkteGue~XWd0K+*@n2NCK^tf6;lnW-G?M%tCIsl8Egb@5EYs z8%|B1txPM4;l%PTh`%QD=)FVPm(RXXRGG+`W$EHHr=;taThlt#B3OF~(_8CZp??R< zLi>y7P}wL1<2_93>m}|AP8T40A5@|YJn4NdwWyfe4+_4;Y;=A05j3^S1K4AY?UTWq zZF|*6(Mn0ag0F`5>|fRD_DD>(#UN%=yR6^LJ~8bL6*3+{6TTWr-+r#1__p|Xaf-W~ z1FSeX(k>6H5%7^4A|x@sU&5sB-v0fWo2Hw3iWtNVkbZ7?*hRHjdQIi>^cGRdQ6O&{ zB59b&R%KV0ZbYO>?qv}hjT8q@ubeyH56C4l9{cts>(t1duszq;UcH*II;xydy5fl) zB$nDOQ@@pVhY0TR8ANYxX(E1>GMnmO`aa?tB=aNi)OW3?+H648&~P*$bU^URn*-mA z9QgMlPzYEMw~)RoXJga+5e3?@&y(3EiQe|jJ>VH@>VH*%z|3O?(dqoKFWm@O6yrmR z*$zo*0;(OsQAVlJka$nZs*jCq68mhyMx1l!+V%c` z+ew43S zkE57P>aN5c&7pTHn27Cui=(&tXnx8dJdO(Oaeg>`?$4BeRR7JGI@{3C?#;(8V`kXi4oiyqLAeK zC#jtRPjGESEb#~NFwb3K#bi)K4NEG19oUbLvXs6O(`8p^>;kN;wZVCRPw{pf!Q<#k zZu+z>?zc+K2Hxg4m67CxhnqxYEmc1?;V=Pn|7R5uddYG{t-wDJ&fE_sCX~v7z%^w- z-+nW3mTV03$qCDDYt)&~sUQfgygd=fomSM@p6WD189rL=RKmQL+M#o_Xln9%{-$-< zp}zTNOM~mc*y?BdF3hgGMRC;iRVM@fJ-2$3?=Mq>du!gBjGec+idKnqM4vi>m%PX7 zR1!!e>>J_mv#?8CPeFA(H(H_imiTY?ET*XG{V5$5!153RbfL7eVexJ9*R2m0{n+bc zW=jn#@ey-}YHmf)jAX`i-)tv0kbAvo)^l1;GgZqbCqs6Wnf~VIIJ8}B-RQR}V|E1q z(Z?wFoku)HLKk(wWZJRy9R39eDMzAWN&J9@j7)UAVNf|P`UrqqTb)LBcP)j0p&R-Z zJPR*Z7oOakEKujP&_6H2ISpG1@FI{!gFLXWj*y{FHVjqopewgYIvZL7d$GN0bt?r> zYMq@4EVCAt){(l$a~3Ei<5qxmX~_FHe_~;re{In->7}F2u_pnaK&QbZFp`X!Fu z0nkQOmVzG+gObtNm0`V%+U|P!=_czp27e$6LF~zf$(L-!1b$OCH~Tv6ij%fJZDz7% zd#{ipgkW!CV!ws*|AeR2_by~_f%u10*>y$+VON12q!}MS^HYP)<4>Tax6B$0+0KwE zTp_A+r1tGPE~Q=U&W`=U6AzzRo4qG~v_}52n7ts(0#XC%G7S~g)oBdA=L07&4g+@a zb$+g?(Bh*;8PYsY3sze94C0n$URc}oB9vPb2X{a(iYfz|`Kft#HKpR*3Py?#w+GjZ zQwq#K`n-rQyG<$e91GAd0py3oLYc(PNf~k}f8pWcyFqGk(K}&UDk^9p8N$Y`1_H93 zo0hd_^jfRh(jg>`p#7hJUtBB}h=(eXNgvk`OIv#m1?z z){Vb9E`Za}aC16mmL+6m4PqA;^#H0WaufE^Bf)mAxv08+HK>ck+}wO#dy4a)SBG8( z=>54HQcbmLc)JTD#xX_O(qt9J^Fd(zXYRPH+(?94Hf7_sZy-oWmAvr?X#@o@SZ_( z-3M!Z^1sFg=Z|YJ@9%Rro)a$(vAZC5*RwL-Yzn5;UJC-t-h0`F_;&ZFB?7Vdvz%5B z%ISe*knoJ?=#3R6=DAr;XMKI6@Z|&eL2u=U@fnwneZ;WXoxcutFhCA(e=zf&EM{MS z6@dhi75=Eoq)7`*=Pxff^n8uMb77z#w7~Wc7ls?4E2gy7*qjc1)r50RyQP!@)TKbOT z%qq<+TN}U*kB<*90xZP;2Ab4Df5$b;BKasQ`<#~ON5o~JG$`-z_8j( zfA2vBWhZZhe}MbXbLfJ-OXRlV?|`3kGvHK(eca_-V)r~T7?7`%ge45F&~Kz-WF+3% z*_-z@uVA~3AAi?K&}K9W4|_|^q`~yW2Cn*U_0KrxV|}eR%$#X8 z3LX_4vv;kN{W==Zr_p#*E2DUB0II|>hS4q?e%%BJ2Y@^W0Rm`t@!)*Q2qA_C4U1dp4inl zx|dXMBdl`$@t-!z{qBd2;R6*tns^G>7liY-{<$_OpnI_SJdOxsi1gL;*MhdU@9k)+ z;7p@!1BaRrOUf%uyQQH3x40DaJh;elIexk2wX)@;x2Ⓢ75sow4&DBgPb?I$-E$k z)dA0X2>Ozd@^v5nS$iO3&NsQzE^2Eu(aeD&QEOoK94E$FhI+E-s}wTN8}KH7I~#wwicLF`=m5{P!}^ z)fMNfEo-+8`#PZF=!Dnxl>%>?=F~c3oB-<{$_gaiBmz4I?yG@ui3jiP!cn{6{8C&V z@>8*Nsm)^~`dYTIL;?*(F3w-RQk?9u(*oc8(VIrwu<%1Y%o$rP+H&1+doM=x^j@A-F$hFLV%Wpz=2FiCy&j@DRJZ;xaq1z7C4o zcCZE|gIqJN3mJ=ZX7HMn>=G!oAH>|FI&~2LTK5NkwK3_$U-%BWF+8~rlLIre^~KVK zGy4gxe-47=x&Y?7htR9vwWf7O{OF9tqfUXb@b21esREUbgfdUF{3W~Sgg*ETg1%BL zq4*K{$-lpC!qtUARV@HZt~MRnsz7zRu=NPllGYoUhimHUs&whUuU{t^%yqeSRs#W) zK@FXZg-6ekFH>o#S19{RuCA`Z*J_x;!jd)mx2@}@L z0$E>xp%*nx7m9m)AD`Pk=5yN-MvT$g87xGH)8O5_>gHSsE(Yua!scBQi5Tost2ZM$ z*>Ys1hvl=THgLx7?YmBDML`bXypd8K(^lD`fvRk+A-)v~%uEAX780C_Y9?p8NBdFv zhZ@~$J)d`e>BGVd7Z*n0x!`Y1fBqQO+4L!@_1AryBt~(1Jzu z$of%IBK%L5L9@fC?`tKd-zr)ko6DFbYum@qsMZT2?^P)I)N$)Nzub{&Y~LYzl#sn( zg{U?FnWd2p12vtb;x`;7Y-I)5hf`A&FU#;Qr93lC1Ev9HINQ;35-~{<#|nqhof=e~ z)IZx_SI7~0$^fR@#~-%S>45GZh;#`qvix@$`VJ}w(;G0oCPjwzOfljT-GmmNO|x$# zd0;7)S@(RwSPcriVk&ZMBQF@od38B2+Z`kmd-CykoWgzPPA0H{uNys9fh2k=jh<$b zZnE^!nu8adIKBzCz(7IqP>PQX58J|jqtAx^?;99MM-rNw0n{ZyVppwPv7#W#Xc zUl_GRFv`^c`C?vrrz%y3k>kPeeV;wXak8<&(_)>22YLHS$AelT(s>$&7r*h-X8Z39Mak&~H=B}l* zZY>#2?OU3XY&h_=S@nqNOuUrFHLLxKrR1G1n4^#QM|A+*&p#jdYwg@IBk&?f6&3e8SF%=Q06)>}qpwSDj3(l;P^)7>Z?(k0yz(%s!iBi)VC(nxoA zmo!Ls3)0>3Z1kM(@BGK}(lHzgd#}CLoO4~*=h~Qkfe)v)d>Qj95jEKS8cq6a!7wN= zzfA2%S=Ne_xaxs)ILPpZ2csfIHp%6>+ibXXm&6Nzk=qiui^vXtf&_|+&MISEPA^_; z)iMH;E>&o5?!gK({eYNNy}S&8G~&OnQbrdN$jlHIRVH$m(dKlN39jy^)#Aqm58Za3 z9{_dWoCsAapN9W94%!zicfomn&ZZpi?#^|VDGYPFCpG*H^z6+2-!Bq9yG85z%X0R@ zavpAOfR+jqmynQMT#UGs&CJV7fQyUU<&U!PkrWavtCtD7TIs))w~PR480OR)O!O@n z;b>SJb{GR;;R^dp-Z&+5I~GebG^^|}~5o1gBd;CU)44vKJvllTo#w7Lx*&|9kU9H;R_^I|b$r&pqC z)>&2BX8)-eI^qC#?p$08-PGHE_sv%5yKvpK6+yL!-C0t@8G>*YTXY&bG`i z=wIzn^>HxcY}46wOelj*coIO+2k{mZ(XHqPc5AViu+I+Dk0ILkE7RRE11+dMXTQ&W zFnRbLOw@=v0s8x-JLoG+g1)Eh9CGKlos62^0kINs%#aQvw3v!c4c!boSC62GCPZHB z=;2{#riAnS)>guVajNel*WH8DtrO(kY)Lq#onF?s)iQ|v35!Hxfae#0uJA80fU6?I zUDN0oqmxDBZ|?8?#jVfHB3nJKH#)N9#=k4SH~!hB$0pnb%~%fULcsm`wR!cI7mhwA zfM^8&WVDC_f?(H3{4i64l=iqvZj`n2d+t5X*0%HdWobRp))xK!{5*eIby8mk%5kB| zP<_=0l8heZ8jqcT8z+SFRPh2;o-bd%j4v-EDU0JT8T0w~l((hzo%pRHzIZ1M#iKeh z51>D`>lwyYk-f!7H@OZj2oT`~qOV?XQX(SI>0ZBZA2w~hyDJy)v`UKl{N`{(Ztj`) zMe$un%V{q!nOyJ5tOQ|)2Cfboev#sLoBu3(IgET?Qoh(J@BoX<-0sb|m*Qt)1 zt1E~sY2HX|UMf~L3Q$EldH+isb$8?PhbRIMW z19zmd>juAkn#bS996)YOpKS2ihwM+|ghDD@4=VfbvT3vy+#=!umzF(X zQoz8#6fzvIJP>7X2sM@ni3Df2T3|#Lu)2_0n6v!K>+K}(z{kqLEedj;GRDryy@1s4 znhdk`KG&4^@PWwC*35;Bi3KnPU5$CWBZ5nSFpP064>TxXr`<^Wkkc^|-de#lf^SWz z@ve?0uLi`<;u~7tNj|RQb!)7&t>YUIswY zLa)FA?|(`JVRW1i4^>d>)ozq@c*B-Jr0rHJOnt2x*A0Xx%WSvfx)jbQiw2LOn6$J= zK#2HpD~gbSuUSS9Py=)J(|veHr`rW!mUhm(lwyGw@EiE7cyC5*y;whU%kWV`&iZx$ z8RruV3Q4-hO8-=JE?h5 z$wG0OkSRxrTOre=kj#ZhY;OA=PY|0EzRC0<(FAyOLzsq2Wa&d|du!c9#$NKA(=bJ^NnFB(hKOa{TJhX*6MbB?m zYW}ws&}=d!-#!HH2-7us;-Zy2kFlu2HGUoA6UgYLLtm0U3TjWi z!ys;XZez0lxxB2+sSMz&n8l|)P@7Dt0>8Hjm2c;&yU8!NpXzA;j)}0K8^rH3o*zts z!#p~+r`*{Z#mKkwBs$bTneuHPA&QXw+GjN$4boCNk(;7vkn4q{mis*>dS_nmbVf?r z7As#N(1ZMWqzww7gQZ3=;j(f?XesB3{u4I0b+vMIQ=S0v=8$5X6^g9_80SAHhG4d7)wX@4XhPgC|9*XUfhIB{Odr{c5e^cSZqI*Pb~`WB{Lkaf{vpm;}OVL05?T-{&X^s5drS zvJIy8Bh)^Psr*Rl)qggKe%NF5Iwg)!48`bVuB<3+f=Su*$TBYpEc*yiz9vscidG^Z zX-; zMyRkpm>qopV=d%2w0%(nZ*ZZjG#nV49SBU-U?mf**4L`UhJ0BVKk;I@DTTP&xcg)ox- zP+mi3oQ9j^`k?lJb?<_yGI;tUtYtI=)>TtE6U32wg3Xmj`S&s~UMcJ)A#8yW)f@$% zf75swd$rRn*)*)OU6oL-Vpe4y-a6eE+VpwcCdCA5O0xZG);Dx?6cEFhdP>2-zGyBC zYyjqtg#|hPYyjk_5;8J9eV6;0JPrgf&8xj#;%&&sF-+@I9`5eYfSYF{F)dtl0-?Ln z5MuBeg#4mK-O57$`;{R0IB# zSVE`;H4b(Cf=LVsf45$p^zt0l!Otz|CM#bS5oX910}S5$L_GG280TAaz>>fR+Ht8D zo$8Ip#Nwhy45-da88Stqd7{Luf4~ZJ#`q2~R{u&E1*84p(Sy9v_5EBY*w{=}QvIn7 z$JN~<`*6yw=5t$4p!$f5>P!~_KCGcOI8%P%+reinX@W#*z0hYYu;f<3$AOc+{E_kN zT?O6ZVCtb&?7d^HUG$+X)bnvpY-vDHV(ef@JF)# zw)olj_p|JXAUPQqMuFcsWd?VawVaRfqm?u_<8%AR-i|{-WUZHs))ET_&4|4__5!yJ zSM$4Fx>J-kkNCo1cXmy(j;A^yxZU2ase~I@Zu*4$Pe%s)+a#U>RVaCK#sol==y=j4 zC~2BUVNhL{a=0rJceql}+9dw4mrC^3H9I^M0WkNeJL#v*>-q77>wMc?G8T6BVP9f` z28PL}XG7k4v6?HcPq)X&sGWGHyREpc_c>Zysx_=svmzD1j?{$!MYfr+&V$1Lf;a21 zsJ&v|1|zK69*Tr&Ky4O=qasyE)Ku`QN2*FsEdPW}2S>d9QBg~)Z=;;4W$Q;n(Ogb{ zj|dK@sO~SoP2PjAW0Vf=jC7lae_OAgu78k4(Li`N95xYNnD+2zS|t_g4;`|@;9)L zy_W-7&5So@A4rI;yzk7>+o?9Qe=`5ivg-R*(nnLKc-kT$5Q&~jwl-@2(mhH2Q{vLH z^^Y>Q;2U8LQr|WOAfqcE6{gRT3WuGF4t=)Y zsZ)+^QnpUrB}s2w&5FktxN0~bf2|9V9#>0^SfTT|JEn2jqJdt&4hRvAWP8gKT*PAb z8GN7i_RA0&#?}D=oN_+qp9lK?`_OaH?5O-Gsi_fh_Mp*9aQKHc@&Vwtu$b5BVC{VV zgxwif@ZoNzdi6aJv>6%nWi;`BW=++p!oxLRFy+=NBASSjLM(3!u^?TVI%pq)XM6ULO9-e+ye?flyH2Zn8&+Yn=DjpCJ~V_#?ESz@rvdZj z0SSa+q@VbR&^`-zuee0MV@}az)rbmFIGL)q2`P;EFW+I}hD#tKw^2zX&jpBwC&Cb| z@LN5Oqv=Bok1~DPKCv{j5TUPrm&zuieZO8cNK0bi9QXV8?wBv$P1Nfw1i1A5An??yAv|IaYd=il*rGG9R>HoU1% z@!=OoduIl4pLJ^H+_eapxq>r>5hF1pD=K#}VsBE|>K@m4In*Sqp!bR_vpBU8%v0M* zQtVHy^d^emz2S-pq!^P2VmOS~Of;wA!36fr&$+^~)Va?(J<+ zF@a+g63S+LNNVL_%t31pAe{4Ff=w6bi4pp>ZllY$J3rE_+QCw3CLJn+;Ns$Pb93{$ zXva!aOz{mCHIGPp*0Em>^51*fEMDF;(o!Uw;@4_6f zfqt8ej7;ocCL^{VVK+T4$ z_jRast;Ss+<2{dHYF?iHTtI}Dgbdwqxt`8PoKsl}(lDfrO5^p(5wG6>c=%~&%R6K$ zeM`3m12&?My#_rdd0dC*OV-7iZnfd+`m-Hs6rCl;=Kf^j3sQUX?o9X-C;f-}996q1 zjUN`X<3;r7paG+P3m@8~4M+J71GaFuo=P!9GObNuk|>|bySUBq0S7A$4tzMdlh5vs zUh~!{{XhFFkA@uCfe`ncM)hF4#+59tTt#@NdITbLR^Daz9Hv~R&TsG^_rDmOoW}7? zwQEen2L6z{SsbEAa$D~r9K-z;hFTBMb&QOSvksQjA8vP;1v^MdqqjZ<97@7E$Qz_u?@<3qy$U-k@1}Y(e9uc9zHH zgZj*TJsNLzyIN0bYv63NY~1iu&&M<~S}$~NdK{VmgTOaYFZzMDcDjHNW(uj z_qHw+XeHw1-PNlP2_z2>N4eOss?HF#k=PTZj;1Q!Lje!^A4zSsb##a@tVhhsu)7y` zlQPE4-gQjn_$I*_Fw+X_L{O1Wxuda6lEJD|)fLsqe}bQG8~0vHI4#Ui@5+b}Q_>#k z9lJj0-k8IosMHF!q027)n1k6C|>v4t()h-OG6`Q2W)YVtgDe_gdPe{5yAhO`->LB+&>{ z|Gk8;LjPpr8@9mvvf8+L8F(UPf7)?8GA3F0n!2ix^3C9h1@$N{=~r^z95qw86Y`pV zsyc2kFWi7LQ!FsppPI;n;Q8U&^3Clo-NLUKauZ1ePjbJ20=&^b%QB&d%nVk0mJ@piRg1*zI_yhAoa{3WBjQKb?LBiCZ&|-kUIE`2e`u?|^HSKJsgf6FSdt|0dMVJf@m)1xIkW7L!|JxC`x0y_I7Y!G57 zq8i-^8>t29QivquZSWN71Fe#zE|8+;Wg5QHon@S>7|0$70{|_+Rwz3o-*zgN<6($8 zAD554umKEEcW91yY^frtw8AjYVD|tC;72#Iz&kSCz!vpt5{lac$|5Ghmk|zOIo@UR z-%Zy>4N?nwQ{c%JatE;3AQHSdr`MCi^{`iOSif6QtpgQ;=l;Ac2^-sGkg+KBKXxc# za7@M9T?ff>n91?l4%i>P-PU>5B6f?LdtGXVTf|?O^vyNl1fukw>i&D-MW6c_354W$*!Rp zlI5`g;tl~CcGRO^zk01arMg>b^n z@T}1)ci-4nBVisNIGC&2UR{6jcGjZTA!32P=R(ln#xj4)T-n}gz-naRyt)pSwo*w_ z3s7PE)=_ZfOptADsglzs81~4cBRv?fe?&r2kXV0_NWS=KXiOOisWkl*Koa zKV1E(d_1O2YKS8syR8a2IGU&f94JS^$Jc7RsN~hz>kEXOfuakS)rqi>X6MHhX6k9P{_^fS)f6C^M-?j#w$5y zv)KD$_DvI0Q4z00mHLns96jt0b(@mdFoj~_l6l#Xh6ggpq}I>m)drqBsgUVVTA}!H zRG)BrVkDYLE%rMzyzU_#WoxP=m}q$){7$Q}z2HZaUTAQ@PySjCs3F#G|8t}Hf4>dz z7xZ9J<%_3*yl<40g|OI$_t1skFrR$=;hI=gptE?}v2I1ND_o9smYim_S=SDJV}x{E zQpY4D1lf9aJE09H-%JZU&t7u+|A=UmXtZ7q38uN-;*?k8GH8vZIdQC0rw3ivkuItJ zROJh!x0Gc!sr*6nLskMoi5X%zXi$vuwHV6or>3^1i+?J`pqHa#+{z5`@6Z14Q~&de z0DoW-1|7_I_m;EVz~Yl!?h3lb5leTm-fVrRdpq`})|a!g7|Q#moA@Qv63F{RDwD7T z79dWnPchk?5^B2`FD^Fi*7SAe$9#u^ky1n_4f5|~h!h8;##kZRe6vO0#edj2erd54 zRsIG_<3q=S=nJypd#E%pW+_pjUE$M-KoHhta~J$OG$BW|8?N9KGX7t4_WwN9m+tMi zSSBz9$>-;ba#rRdQ!7%k6~N_fEkj~Q;mw#g_X1^53Q3FOR@>d5cZ>iN@DO03%!yWH z{tz8TuEfPId&hae_pyhOYY0A%ZAiZC%cjHEsdf8Gg@^-pe1;QoLc7Y9us((kD`U5g zRM6@FZ8~jC5Mi(GyLQdQ|F@d|pC3(+8}(+*A7y+&J*Yy>gqmuCl#a#q#-2R5>Vhy?omum zPAa&+#nQxrB_sX0nstm22w5AS>!W}aI}LJ4#%VHOZ346;mV(t`^Z0d@_mexKlihL5 zTGb*o<@q~0FA1K&=G562iT_rPy!GwN`$_<|mhO$L$p3tHe;0snvUrCcyO8yGgsXu` zI4;R*3>ue0V4*AtH8q?YOkyT08yVGLm=7xB4L+_R zB-8p07AR8no-8=%zf~AFQZXJZ$Hy=%r`wG8`ib7Ek5T`zup-PS8}FfZ*-1TPcUK z(PQ5B{4#~$a5q?JQ3}6sU3?B35qAsp?+&WFm+mE6jXxp8X68jhMIc>NO$8tDh<{Dz ztX_`FELk{-7SD8*pwinZNM(M})T;;?U+5$@FiYBA-|k8e@1cWB0%;*M zKeono_Vyb0fMl($tzGvIUF2t%U;#AGa4Ofi$y$~rIt1Ao3DAr=#&Co+{cD1KSq&bb zlWE||536o-hf7chxw>-l^758dQ*x&P3pZB8$GiU4@9Z&9va372E!2OT+~ns_`ja9d z*?5}Igjo+@EK^W-ffqFFL6y8XxiW5SKcEW@_muyl*>?bL?`w420k~IsJIP9sz`xc8 zy1?q}K{rM@rKR6SP0e*0XIItR|1}msWkrTl&oXWHF6o`QqONRo(r1q^i1s%R@zScEx zeR0qMF5~qhZtZB$k`m-Y`VT91;X~%>VCk|w59byIyd z77P+J<%2;~4~>PGCoxIMYo>luq(R|D&n*F;**K=|U5%!jkK4jarZCj-@iGCX5em3X z#)@|ubDQ&{Vnp25N42K24<+i=zw+p!3X6)mW`mCkc%7n3>VAq@QUgUa};*dH97g zKf+DX;L_}A0X-b2pQRrCzKK5RAqmHI`52BO6oxjR&q0;z=7a zHdZt;b(FCq3ekM7-WbvK@&35IUDeihX(&6JEbH_oy0l^5Z(QA_i_rvO)boV;fX5CM zxa)($W`w1ASn_Lfp872+F~-;(3Z`aqIa>1lkY$(?_p1JS3(@E%r=<)5z+2#$1S-;r zhuCJucmsNbz!<6E^&uC+2v$~BKtBdEHLcxPAznPa;jwz3jtVoz*r`p@aMD|<0pwM6 zkXzikY=ly$cWBzSPG4%bM;%4XeH6cH<}|2O=>Jpd{oS&eSm5o#lRoBuiT%15#%zeb zzK`#vOi0f-CiG#317zxkXw&|9Rm`18^RM^~#~d24>gxcZhr-$8%!?V89O&msyKtta zatEvJ8PFB6S5zOp2@-sYi)n!f^-J0w#zh}LDD6r;6OfMEHHFr6$C+qs5;_O6?$Iz_ zCE1+b)jDre&pZd@$S>q~)xIjp7853grYLBO`Bg}dXDWq$aocO?tR6wW0kuQaT8peU zn3^f+Q8>Af0xi~PTdh=2##TQ)ZLD5{dm&HG+8#MuY>E4#2x;d~bQUoW1;WL5T5g_o zgrLHu{FQv!m6@ny)j^daVQT@#Z)EPJlD2lhw~v_o>(3*xn+4UJq+W-cKUPj*tL2s$ z<`z~gob4W&o0P9OfCzaY7U#XxifG*aeip)oPk4DSDz36>v7n9uf&&M}=D~dROJ+zQ zuVYkSUQP~_RtJyw6K_jv>*)($3n=e7Q&ZJ&XlS(vJPvDT10IiC&XF4yy9J(a<;3yx zmLDl0hHaEyE`I;1T}dFxGkm0S4H77dPe*w&)@~{pD@+c{4|GP1joU)0qrBnejN_(m$D=oXnSNKXU$q@q5KD$P;tZiyN=veWoEysZD^YW$}?5Sxx7uA5QA!!yECCL+y3Cz8-|(0>sA#L99%yJb4Qrb8eU+A z2EO(Ske-=h8d6~ww5H`npyc7f=e5XfYQno6CUO9ZjFZz-w4#%CS>7+dH5_y7^?$o6 zqAStU85oT$tA^;-d7YF-7b6m-v*Ozl3HSos7Ub|(b%CO(nD6l9m9*fB3!xm--Njzm z#9io#WVaje1Oi{fM8h3AOIhw|biHl?STk0@m^w^(NT^+OvTsY(HRyQa1bKDk+r^zHd;net@l9q2%zb$O_rt8}f0 z4R3t@=LbhR^{VgU&$k=?tB*%6U8^y~0GQ+@F+M<|g4wX|qJTp~2k6)}RwVKMb!NZ3 zr#4C^(@_G7O5#e>F-%g%>vskQrVG8le)f{@GAZf13%$0VX}&94?o|bK$CpUJY7>>cgF<2i}6l zUZq|W9yl9gn5F_=H1IYgB!b8RE#jFcA476{xa3r7t~)(jo?H8}9iT9gl9FnV0C~FI zq4aviW>$TL1X$}mQqIo&5!^(?GtFuWDy9KJq7~nXmm#oCbL9moT(F>PEipS90#lM} z8|`G?r5+~j5))&DkWkN5Fn=6dGotA6n?&H1=`&UrGTKb#?u(P(mW6(n0O zP6}aj>r!??M%Oti(eTV3a@WwVeBVI69&60`JTI6!G=?hj(`kQ(>oroO7`cecDKaob zZ*LNM*yBefVjFM_(QR=>%ka8;`;tF+6f8dCh~6PKqA3aWAB^C8YKXR<`+&4nSbtr> z{_cXEU6KigWxk_$=HiasX9R2Ux4w0C>;&8nIBX{LCZ#&dm+HHg4FLJ{>&$R>A%`_uh;PK5U=VU``w$F?6n;%<*X8HHOj^l;xvk7_%0?K+P9S{MesBM5>;`=9pk&l%G# z>$8YGoD%Qz2sgU!xyMg2+ThW_V0S8=JEgzg<;JIyi|mR{NCDisW`UB_Aw7y${E}@> zlL+3~kGB#3;EzE6ID$07#bQQX@Q-XWf(CLaf%e=0VL?;sWVtr<;Hf%4mZYBVK>A?= zH8pJubfKElQUp&)*_*`Y@u59dk2iZ=LLEY;Vxd>=ExMty79^3|7FzQjigIp`2-N}8 zqCLO9HSJP%X)7;gbXi38A$lEE4!lV?jLwlogz2rj8(KJzD9L&++ml@Wq;Fu%Jm#Zu za{w#&NRQb+9D$=KmDdg3yfN1dwNY>jjObGTzUDW+g06BzSOgdp_HEJWchcgK=pFWR z2RqjK*O-{``2F1uU#K5+mOdZV2za~uoThi=y3iBWl)eKe79 zuc32MNmvmJGwbeT)fK{u(Mawe)CC?pSX*e4rPf>$1dFU>;o(M}%tP|x*>=r_>>s13 zr~(3AD~&pw<9=N^uM@0m@$BhZW2P6I1FtE`Su^^~e8*^I)H9iBNY=DU}lVCbz0RMnUZ&1H$uzYo8D6c2&i z^b@qb?Ez(p&c|v0Al=ZY3B<3Y$x!UqB>k>u zAp|<+4tog+w1Z43RG;pw{l8D@-!mD!6l4Q_MN7eRkE* z)%eUJyEYt4m4$v)5h@l@;L%Kr7IgD|+72s8JoK#@C}4h3LpG+#N%ZN|^Xawl z6|wQ=jS*q+PfnZU6LS0xwlFh8ZNkCBx-L^*)GX6;*CT7232|5um-r{!y0h;+4Y znxrbqttdplYoF3rD?KOD8X<>GzM|sx9mh7xg6d=cc@Owd%^}Af z5_sfZfDwCF;@Ha52xA0M%`egcy-Q)y04M7A==0k`W$ysF{)T7#OB0lv7>nHVYOT@kHp!Aj8kNE6MgG-I0s`&Ud>9!X{*=VOx*MekJr_wObNFwI2kb{{ zRN%FN+WzQhk>;{gSsa4>Jz4h!!Gz`}jENg<6ro-42hhXHkx8KDfJT6A)>E z;ZUeY>dDkhDdxVv4Tt9G8EIy%bl&C+2!zfqQ;%Z2np(4R&%fE;qZ3mqz!!p?V%o(A zL?dl&PY9>9oSd#^aQ0A|o2#nAhPXYog?2Vl$L>9x9C{^*5JzSQr!jbVnX~mdmC6EJ zts0dEFdHUgVv$mT-UloG-%G3j8Ie|<;^b0#X{juq-e~wjfu+hEYYh6{peC&x{k z2=R9%rSbvp!kMlpL(gF3! z3&~YShwKG%3G0u{orZdj+^;`e#ZtGxaBQ9kIkaM}z}I zKR(JA55SEfI@0H+Zuj5LkENqI>}e0(a@-Lx+yHE+cD?<9M~UM%{tOH7KYv3o$z6z4 zN`*sl4Y>XuQ9T9Bu)g9UilmSnU*79(Bz`MKM+m$?Ksx}ijQsB5dPhg;oK9It)QmVl z#1ZMZs_&IEz1|FI&#F3VC;y@+LjS0Wi>Oy*faIQ~eiI2pfT+{Rt~)Oi$E#IIyS}mG zljBnye7jAoqL5yLhh1@juI%O`F&=OO`_9(JE&bxj5F`7>JW|FcqsP^{gf*X1rQzuu zSfJ+%rF5Bb0aSXo0PD;6*DDA9ONF^vwZ!@XA;>X!JuEm-UAT$5M98rjKKWii(8H1M zVnM8iO(}M^mS zqsfAyxBD)3wez`e<}$kWU_m<C4T1MXa8W?oJ-{ju)P|>D@ z{e){t&?ym_Xr!o#*YPX1MSU0FTVQM9OXtooYy)V&mj@%q^>{dN32Y;!#CmF z!+n54pV@dY?y2af7aV0r>1*P~XHw6bdFmhYo{;g(PoS{qpoLlCidVL!)0rDW{2J#L zOrva|XucZ|K#I2AGrOciug=Ydd04$Y;%atV>Db$|$g*9gkGgeaNcjv*id?VeP3<3U zPQBWzfjE^8RbA#)b(M724S0PXyk~L9IWtdHbD6*ImdOI%F5~8Kv3~QY`Plgl@t(qc z5a(P+gWPkYW8pB%FQ$7mGtpt3OZr&2-p&=OJqEM}edwAytlfxV?t$@uy{jR~O>68D z!vw|1KB^NfJ*-^J!xDzAx||wH$SC1VU)i~%#df&TJA@K7oYiUjWG%m*+jR+kC5kOr zkEh!mxKxbouOFn0O?yNwENFf*ru(n-IBc8Kcipz%^fERrv?a@Nb90|e^b?ayYX&7X zn{;^_h-56EMu0jl`V%(~?ma4wPi6rFZsz>}zvRT|J}Fbujj`poaJ?Zckxi86WBd@k zj<)4mwWbZPh%`=Y4`*AWRbAS=em6)ga+-a-)5}8S65YWA6eKYMX>&1^$^b6h8=?#(dIhWb%p|bRyBTKI-3lj&(HGh?q;?(SQkH;98%rFuEU_#;D|V3{U-pG|ZRKlk-@zsca~MeSx)qm;4&g3wf0)?nDs$3Og3K zZ8@3W$;inp!RUoC&E#m;t2navY{$f$9xVb`Nt+S9WIwee}evK$_AE&v`3 zT^Y67ePU(~`X|#{&^>ttR=< zgBCKLm%t~t=IfruF=;n!vtCJ-=mIx+i1E-asWyJ6T+jH(W|v7h~2Z170Q=x zu77BHu&tVnvtoz!21eE=VA~!KOxEuwf`aqOi8Gl-(D55|bM)@|=37CNEL<#!3WF$o z-d5p?V>7&!&pSvDV_*RyHoyvH(A!9W^=qayZfZA1nX0e&+sC!N;jT5c%9ED&2L0_B zJ6|tvow!|PlfuX6I0KyVul3}#dG<(YIQ#fH&?9g;AfEgZ6kv95{nM6HnbpAB{eiCf znDV^`L~aaBN7;hL7ieqQ=-6Kr6%_-1(6reFVcXi;>OTVsBcxvk#Cke_>6CfP-CpsJ zp&{^@^Lg9pn2D@xWJSmh0ODxf;Q0Lh{i~PoiS41p>KZugm+Zy@H7^08<$-o;z-+iy zH`+3s#-kd7RZT7BAo8j)=tImaXQc66jK9+;z=ST_LFJV@ZJretBIvHb@N{{%{}`q> zLMM4z_%S~~nV-V$^>s5_Re3)>id+VNw~F`MCiJbS1EP<|z^cMQfRqI|w#k$@kSSE z&@v1t`sf{Pm>t~l;xE?Of~khAX|t{4@$vBi@+-0uhYmVfD>@jgpy&nft1lHG=3X^G z&U^*rt5LGMoo#*NmrMY(2|>v82Iy3aK}>5i-kvxPS80MKAF#snm_fz;aJ;_A=ny<0 zwPe(8#{mL?DmX>Qw?sPFaOvm8{DVlxNa#O?-s_g zOKCop_Y-@kZlCib62*=p*4zKW3XbMIlgQBiC4^f()LVVM>P`Hx@(Y+0X*+dycLVT@ zXCirHk4dkomqRNlEQDul zy+8*fzk%@syd}%?^KPF6vnI4yA=1w`fmy+_PlDX+5@(tOpk@f6B1s%X+s6y!(oqeA zY}qUgV&A2sgB11v2q)h|romi*M-MDhg|`Y=4Do0VF^>0XYWZ_1NG&Pu<}nOGZ6HmYI6+$hUc z@DI~9rn@7V8jHZShlpXagtV2z+%Qamu!O6|gZc6SYvU>zK}6gYzJHePw5!gaE-Ak^ zpjNtdeVaj4FcelcQrXV*L+E+l{rX(;rbI+-e1GJT!Gn$-LaGzd z0MD+Z#SCN>UOv9SB{g0FtoqjkgDhSTH~im*Y)1f&KOAsDAi!O-Htk`jS#Kk1UkyV@ z2%GA8eBFw&c)2i)br9D@BQn6-o2poV>N_FgBVL@ww!tbZJuj2-P)cE~Oa}98X?x+x@R>+C5$`@g_0O)Ei%AOISUSM9zb}DCgd+ z$Yjl#y<_3wq!n3Oof{=H*&xAwecK2q;p9=*z7_((ecitz59bo&U^uX`m~lUGNrmcm zW8EsNs6@RwMvjO^1t^#hS)E4(#tQmUCKK^YSy1L&ux6SO5@U(vR6rgW(|c0v&9-fR zzwHc>VIeN{P=s?-k&2>0gw(+|{=utCGEo_1TaD@f>3fBgj=B+z%@cB}kLtj@OUIYXiBj|^U+T?Wi9 z)9&A`@1Ut|NCBV!hSgF78gMV;_m~rjY5iBmocYxUnoOei^Yd$r>H$ZFm~#n*N;R^% ztP|p4r4S!-R5M%KerFT&0-)5=F7B%wJL_mwX?}y6ffG^^2YK?KBgcRHodk3WsuCj- zboOXC(V$=!qBC?ehos}f`{bmmzo8_6Vf0CZ^aas#Z2utEwpJI&#le}*%99}EQnOBY^Zuh#BW!k z3+Gt(--dJ7&2@cKCzcw|p4Qt_V`acnlUFJ1hFdH;y}|ST@tthlgZ}r`HdghH7k}BD zeL?DWgOXqg*80=^G4@@Db^C31dY8Abk`|{wGBxEw3clXeWx#Op@vf&qUR9{x zPWyZ$VB)2YxrY(rdy8hjPo!7uLz#4z)^Gn5j$#83T$%nb2{2_L_iE#$;eS9Mu}W*G+PDV2x*(4*k0$#3lVu80=94wDa*wWLOG; zNR(5ctxvc%RH)o*L6tl8R9UXjk`htB%($-^u&RuR+3L{qbx4JFrxjg1A`y2%7BzLS z{yTw{@#fV+u^y*%AJVz~0Cy*Rf=Hbqhe7e=Z#A1LZ3%97=*uRm&Z`*R{K{;s=@BOh zE;~rp8G0$nL6f05lm0sExL4tAaex5s8maFu_&VQ&jx*yC~)v(YS(RbHOZB3-cT4cj5qG# zYkNi=`59g`poQiA`Fq~IIyXP0Y$s0&&&1K=`*&2_s(B_P9V1Xg8hj;%$)9I)TL0-! z$X>ipWGk?MFXzvWh3Y#5eSq0+WeP2m2I2L=P4aDOk!8hJC9dC1Z?t9-_FXB7pHrwV zwc`3Lm9U{OVx85E;(HOYq{FX5n!tf(!v?9NqZ8vslj*{eomGZ;$4&m)f&3GVaQmAc zx((^V79IR-uW$KKB5OgYFM702+biK&h0x^l`?{Nh{VsG9NA_@aT%u*o5Fi2OwY7Dr z5cGVJvSDn7vj5WX@cpOfA$sX~0^ba_l{vNE-{9y1 z_}=gF+nIgYfwM_^3H{+a&!~dR*&_FK=vx^Q{y+NRPg(OXQTLj37cDBuJdSapnLG7X zT0Tf09`ys)FhH5LELss=Z`5JMQl+I`>vyQ{a$4Nj`$~?35(D9FiW;hNwF(?IHJaap z_IRebc$X`)V%6}QGPB%?e6_Fy_~$ z_?L7DHr<_cM)B{MPNVyiF162S)kawz30GVn1Ux=e=&r^(b@PH79YLrra(Q82AAFYu}BV?pd|`N2TJM#i)Nf3h|A?0%?ASLhiW z!+AeB%P&;2uQr^*)V!MIh%fS#DFjUr`l?=_%$A`iE?Lk{aj3j#U!ju!heTn==~j7i z9JGwYr@;Jr)+^g)X#YwW)}4m#ZcPYTL;QCs?o-p#o&Ej!`*@Y$NBdJx%BpjQmsGzb zzUJp+)mcf^+wY~NNaZZ2`%3nBMhZGft$w0zT)J3A-KKIrU-Jxsk|!1Z({tR{4+ z-6DbMq-iEO@QKdpJ`GaAHGmR6*v(Xn6*oZn7!^m>g+L^q)UYbVl34*$mr82_Wwd0lzPd+W% zW$=Q5{S>B4Q*W>4eJt#}NOZM(AnQ)5hBBlqL2qg7HYT-hrLUeU%6F*Z;XH!iVvk?M ziY%0)*gf2~&pyhxKS2w87(m%xfkpx`EavxTv-(r~z49$@f5Hs*G4+;lRefLgHgSMMclV(~Lh0_3mhP19knTfwx1=JCAl+S( zDj*2b(%toJ^!vO2&wKpSVfkwH=yyM+hSCNjCko_Iq6 z0H=1)>~+kA(nwcatPy%G{i@L2EnI#DR!JUxbaJb^uT0ErUaAKY|41akQ0V;XG_&S0@A6wIepOj#Z3v` z|H*u$OwO^UthwwZbb_haFP+-6;$lRrKYvTx;5d1WqC$z_?m*5L_yO;&9Vz`}E7{KG zJ|yC&X0(>)d!WS%udS`EuDC{Mg$VsXz*j8!S#O02z(^N8_I=(T|6vEtRTvJL9m$(K z2a@+V)jgL(RXLbn-Xkfuw~%L>dk@#}^e|(KLvSQtXhO6OV*uUib48DciShTIscw35 zf2>k6Ia=;uWE+=R(&5gv6w}pw#S>W7_j!RiETT-8HWTOK;)3;79R$c;hfBwCl1fW) z>n7biIdNM@(NybIYXv_HPnmN39+xKPIh^gA?ISq`)|OuL1mP{Pwa>W=$H|yhOB=ZH;Y`We#-^v$W6wUq=Jmgs z4e)dM_+TCR`D|urP1UdU!jl2)$J~k+@^TM(^V8j|%Z3=+&kMc&K|xNzfqXThPQ^dW z<2EL6I!yNu5578R69!mO>?79CeOuI10G(`B9Zn5Z^m#Q+ zyC-3u_=fSX3qVr@%nZRd zQ)>TqE?{1v+)aT!zPL#0VGk*mpNk&t&^MWA^U}Rn=j)>nIGGshc))IctYZVtu9ct< zg96vB6zD45Z1_(cO82RsY6>c zgM7VOxJ_;OM-aU^i_Sf``DX!h`nsIn%T6ccr)X2x2m2#ZU~%#c}kthAOjL4jGs< z469AJh9cZmAnd7GS>a;P@69_=zao&0D<;#qY$twZA^__h6(CLQ69k;u{Q0-uIRGXQ zeaMOCklzfw_d3Irfg-d*11%|o-Ms_YRKN}pot=+Pg^#}}6wxmGFasZXBkDozdM9H% z0c;_{{BGPvf}IG9IHLj>;{Dd2S9$1T*QEavGh1VT9P)PL$qnQ_iGS3V1|V<_yzqv|Mn zUJX&Zs)KTc48p8)(syN2OE;+aA8<&PA9*{krp%wy-SV^xY^6+`;hq(jj~DFdpHU`z z4lJ+tLd$S{>!K;9c^CuX2z1Nwv z9`b@|!W@!`_$Z?Mhhw+aIh=jwbbyh0xcBklt`4v&_XxV<+8!U71_X$pjd!}#s_W~c z`nOS3y{H<8n>MfAbR>=1Yb3yv_7FA z9>IVD?R@98jg@eSQ{TMpoB9A9mycRoGXr?8mzo~^hpz!-l&3ACls*0*OzmaNFA|pT zjLkE_o^$gH^ZC9w`c3vdAqc2GI)n?qurvWV6|nhG1&^dKtmn%f!B0|}IjMcph>VJ= zYZCZQ1BW3UwEg2}iyIPG)qqahTeuuAo}wh(%8^oKlWgqrh2pN6A`fx%DvCJEab`}Q zAn7Utwob?VUQ|czU&0dmu5lHWJfs25795nN9;Pa}X$R>vp8$>m#s;WW9$Ic)m zr1joXTb)XrVT)*L+?d_ZX9crdJsLVFXTik}@eLi-uyGT%1X)a0gYgf9dCEv`zBGPP zU%)3G(rrCq9UoD3@p5*eQ%G#}sWf-g^vwr$W=B+q1q0V32XGWEZf=*B5L;*PzfNQP zD>K*W-ROXlRr?}_y)zQ9W}-xCRg{+>173%<(hMV<`_Ghs{_h^(ZVv#CK?F);AML+a zEYI6!Fbgs;UQUyK1UkwHi4d2% zI^g;Vnvuaf;dGrq6>@_ASfN$bdfH3loS^>6f<^wlDU19QwD%R^$?dcf-~*wH`#%2Y z{7TQyvf1t{bEQ#-4=6sq_5USFq9Fy3jw+CexMF>6b#1o&b@KIazoN4;Cjuq=Ct_}w zGX>zl=rTpMYgSF;@Ma!WAs3sjbC-G%OIce}YV~;v@Iz-+pg`zfTBL1CTJaS0*@r>x ztj7D)enP}GCDDPl3CcM6o5|i~dgS%+@Q9S8Bz&7ngXqe9P+noER1h{`qMRvx_nSAO zOWYEQc&Ny?wnoVQjvYB{!5}{M^$Tn3&)t*yA(= z@zixq5jH;elhtMEiPu_$Xip~Re%KzP0nuez#Z+n%(rZof$~YzKwp3A*7{oN(Z~Lvb zpF*OXKf~_Dqh9oqM`1ITJId%e^7n`H@QwIJFNE{Fg=cE))qESaG!LqihI>`t&MW`@ zH!K2e5PD?h^D$ElbW7(f3F4sGyx6U%S%VT9iM@HV(z&qY8O6}Qk$^c)_5ajqDG_5d+yS*qh< zVq0r*RVf8GH#dfK0;;DrO$gxW|8AZ$wN=yKONZ(VghmN|^ZHc(ujT?jq5A?GK-CfS z=K~gHB*@m2wuS}>b*yT0ormC7Ml4Y2t8Z}Wtb|)i_~t9KX>w}nxo^?J3?~6r+gnr@ z78w2N$dD~X3;EktQg+Wqbx$7Bft8Y;uO(tk4IF%R_of8o-(V@6t@7~4=JBv%jP|DL zge1oms`X~U4+CiFRyIO@gCI%YI#?iyiZ*EB?1L(K8?r&OGi;luqjY2!h#b+_BBP;B z$X9PXe+?yNBytCrR&Fx5mSfo($9z)=KPCWflzCBY#b8bnv`+Zz-La4-*>Hp=MYvVK zh45b^R_D3)iQvi&27MhPi^%(|loWIA2{sDoAOb&BBKZ0`7`n4OM^0JR>;FpNK;2=; z=&3av*@9l`mKh80H=R_FCWroy=(qalKmQFQy1!sGQ^OG%r+uaA@LC;7zY&E1^FE0+ zS$zg2`V=a!6N}6D{+C%;Ay3P`M++FU>Hi8U@VWkx4vI19y1tpLdv?3J43SIFW(=GX zbn?H0WqSXJG)r*{i$(s6E`^N`|ZAct#rF{yP`xcSwI-D8Q17eLMQHMH62 zYS0GNM;RZ$GP8Er@dMfk-cux5;lBm+Z}kq374!EO7P-BKZ}`HCobvtN$OK6C>U9%g zcj*{dSRfV`7oS@k588IC=%o}sk!(M?%o!HQk0_6o0M(VfAQZ(E2isLSNN5VgXnmSB z|LhAQi9>HLc>%4jz=LHS%E;UV0I-U@DS|AkJNpTqa|B#z-yk_k*LO^L)25t~6< zbhLBM!@%c~1%SuGsq4YXx_q|Q(VOeu@7sLsCftNkd1#XV6XYI!KCm61xfCv+0_Rx6 z!#82iWljQLB2x}zK+xmZ=D>+TIo5v(?`>`0SX;)<4_rHf9Fr!xBIe#SXAV%dAFt^0 z*1KnBwUtLPeP<=940VXx_dvRPA1r+RW4zxtY4i5U^Vc}%%$L6d*FJfZVT9V)STv!h z`?Fpmt!jfPAlvVaT*;`m?O7Rze#^tfb%$(~I(2YyarRlf|L)y+QMZnoe#oA3?PG;DL z;MyC8e0@+8NW)5{Jk}W*UJNa(KlZZ(LIUVzYYpO%aX*Ai_TBGqPrl?F3xg4@$zRj1 znJ5=BG}+~l(654K>CPfVm%!y1eFx|l&2Qn9mB;B5HyIt%%T;w}L#~lwy-W@){Ei2R z$pp27@t{#r!s=4tt!oodqxrmV-Me2*t<>9q5^=9h5R;?}(`Yt(Y0j5F6S<}JTQ*@r z%$v@wtJw>!rXB8n-%7O25qm!yi9d`%uL{!cp@>vkj>53zMw1 zU#@!hH*AI!91WX)==>|03=i#2lnrumGB)zUBYK}oQh6-k2c2Pf?-$bE-riF_3!tw5 zt3CyUusL9@iwIGZzQc}TbMJ15cJ>67CU;)Ed z%?H#8_mGn@nSVDF&vnm{9F8Cc;aR9A`I>s`MU0qOBx$D~J3dm^LIR*$3SC4Iutl~) z&ATqH>2>ppGQ@LEYA&GbAft%F#$vgP!ZW>zAJ;-HcvaM)4$n)u4NJ&*nlX@{RUi1thd|e-X^+ZO~2aBrD6yPdZpVo%8ve%#a>}cn|T&9E3O@{-i0d;W(NREcS6(D7UYE*t%+Vu|0<(6_!H_sL@7q=$iF01||uIa*Wi)kfpTrK|Q4X%Amwu1*S^a&4vA?(Tb*|ka4 z%5mZRNDQ@n8yn-^m;0&O5U{jk%v!koamQUU$SrWaw18`bi%_wfoq5p~<8!nMLZ0y}^`FGpGi>T@vV$A7nIt~u!#BBG> z|5p-~g0@%g`%nW1cSe26-m*i-2hazTg~UOE@9r6u1Fs2WBqwT?*AE|l(uXLunTu$N zda+9fTp&t4TWP7DpZgL7bPZQTm^J2D$eoo<)pBhq4b;yBl%b{Z!8 z^qN{`KK5i>CQiB~%|`9<#^w)84#=ahDoQ-v4DCuc*^i)LYd z#@PfWHwN5z$y)SbSvrqqnhAc!v>^-Nng1GrTL{MnK_n1WItb)Xp8yJCuy^-utM!q; z5GD~jE`##Ef27g>Q#8Y4)kYbQ;Co;VO+r_S@VnR282t{Q%GmZ}LqN=M6T5yaP-n6q zOf=A@vezlsAoEZ6lTUn)V%DJl6|Bf^ZrtDAoy>Nw*xbwHB9Cl_2XeG#<9c{eB{c%Y z-2gXORDzXa#-qn@u2omXE1%a}NaoIH;MKcca|^{ny{cfCe1dkIvtY zXHpxS05r^3r@9!)J)=NHr@E)!zNjALajEs(`JbjZcv`uD{bX#U4^_EvI@0epA@P%; zX*IK?zPWHiRh6L@b#WeQT*Ch?)n*T54Q*{y?@sqYh4D;E7lg=RA}24TNHx`5YS!fg zqhOVBL7&w#W)T$L)clDM4W^-U*`w~%Hp1@NSzH2w9ZP20u=XBssT{^t8H^XbjQGI; zMvZD9Xi(OC@Jd@^4*?eo=D0Dzwh zh#FUt%N+*k(+9|B>5=06fm7jSA|WkLN=cys+x)s4l&-xy%m?R?uPr)t{dzc7@;B;+ za9rgx^-edqYVbC$`MIzpyOBD#&e1e|#<%o%Avea{yU;+Uk($x>Q$2pk4GFM?=93B_ z=JO^k7W|l->$B!sp0ZP0|GUYp(sFFP_I?g9d``z>WMPp6?&=i8{VSR=6T>8pI0sYUmfW-MLuT;SU|ME?O_)2$2EKF;&$CDEoT%!JXx34 zZozL~EcJYo`mZPhHq!ZK7aEcVXy(gTM9LGVA`!gXjg0np=Qu!0336v-#z)tAKs6pu zAsh`5`!jyuM3a393kx&((IE9LUnnR>Fy4G7`zEAT^R0Ez9Ur#=d|X>qK0-F3Vcjl96|zXJ$}jv zTt^S6P731?&7~go-3k`VFV#ZA!Xl8pS`p|%fRh!GWl2YB>*t8;aDT=f`*}u(r;DaI zeVW)TM8cy2Jfb=N;T#e99KdM&o-iQZ+2N>_PSL7BFv}uSNHeV*@S|C{`l3e`o-6Fh~edW}bTQ`# zf$92xPh3*o1h^XTe8CoiAbN^t>**+_WCP0+;7^2@KDpY|b%P#d!Xx+pYlUUdVFMCU zeO9?>M+y4r`Y#39*p0c(=zWSH?1aza;3DamTB{W{>xR5sNVE+9fsJV@4b+vp! zKHrd(*9B9^&d<1Kb6J&)aX-USIOY9sB<+QY!Wt3FQ9PlW^GQ<z}97sMrJ^kz5tlR(Ih*SY4+-06}PHI<5wsp)S=?s%Go=#zx76!> zp=I2*)R^v}j|Tj)UtT0;4IG=K98IIr)A*R)<4c~$g2f!(l8-LX@*$gO|1Kn#6zc|> z80ixmrii*dF$@YF!ibs_l+12TWMfRu5?AizSN+EqoYX>PZYXcqQVN>D1UMv7lZ}?{ zQO2S-BP!jCT?pv)W$bJC;Y@VBAV9z7cgNwkVlo6Ll^n)t}Cs15J zMjGpWN;dtukCn@p7~g!kDT z35e@F1||>GJ4?v)KJp74#hJgJTKWV9z}RjKJ=_IOS})Vc=qrd1SeXHIcC#9&S?*%? z{Ue@JJGJy~R_OIZ%nHX43VH6vfld<_io&)SnTT4Lq1nhI8VrmWjJ%Y%hAfxPm+rhM zlRP-Zci5FBB}DpH_>wy!6mN7k=H#%uZp{XZ=bup<2CeSM16E@_FUW3m5Sbl(XlOsa zK`eLuJ~T5k<7|Co#AioxvVUrM?BQ`RjhG-ox_%U~t;IUj7LlCj&nX!h?+@VReZKv} zlmY@y=ROdJh}R?188D^7rih0Ly#*wof15xPTHNMtdMhrOt&zhmWKCmM^g`~7$d{vT3haW;` zNWnx_)0q5BRvxcI%bINta8D%WNUH=)LI@$-1L8Zo{REQeI37io=}~4(3;FzOq(>(H zTM#0Z_os~dQ$GqgyXUIN#|WY3m*bx#9{Hi_!Tu{2S4<%YNIt(Glz^;ARr7r*E1{c> z6dK5>T}c;$;Jd@HFO7(pU$Z=**WWvigdqDT2xrH=jpTgZ=U^ae1cQ?D{h{Wczn*H7 zeq5sKc&p6TmFa2fO6Ik~?*G7a&*za&WMHUEg1XK?tw$#8P{9w8q0+TEc`g3CjwRC zOfe-T^ifn)tTJ^|Gw0*(v;{SW7BV;U!R}rHm3+~_fT>%Jzl24O88DhRXda-^K!Koo zzpJ*DgH<=dsqPl^HQ+|=FC``Lew`AjT;c!dG)tdB8vC-NuO@}y@Ar)7-)#v+XM*1X zgMkd(j4Qf;FL^@Xs1JzKueQtvukt09d5x*Fq{ALk7N|r3w%76TaV_^I$Kv7I+!cBOPNURTvzwx8pUm7#h0^3n)66c!c|;o<}*=gyXWWrH5gPrD3tRq&wMv8h?#ZE_? z)q#wbmjPt_v5H36L zb96mHbmGwWr9FFFqYz5Q$MUtS3!&|8pvP#9k4tBeWD7c2?jZHO+{bpr5CzaEfE8`` zL*2}ddv8g8SC{Cs?)iKH(`fq{UGsj(UQoN*fXT1hzUKdg55M1H6Dfj7RqqvYW3oiA znd&=S`bxMlsmvFxFjzv+X#-x6@=AL7%7NN+->bK>0lzj)T+o1750Hk87<>Dx$%*ou zt_n+6O5$q34BHEXI!kf+%6Xn_z4TLt9L!!0#sRqu91;dukn#$+r5|6ujb*>n)zFZf zBxkq4bwC%a_{TjUXXlFAr(JbwLTmLyt)P7ZCy5|9Q{7i zkz>}xgnsmWg3NcC$0d&DVH5q1;UNdo!?82F{wP-(wrO=+e9?m*u{zBz_nx_(95vV1 zKd6;lHVP;2o_MQPETCCIdk-CvJr+8o^!0OoZL3N{IB<|RwzosKwYsu<34KrGc5`%_ z+msX(qK`AV?a~tk>$)COQ&Yhgz%%+$43JzJYrBRfoQ#ie@by=3BwHc}-{{AszXV!V zbhM}9;i_ySpPkZ)@4JogeGQvW^A9(Zfdr{aj|n<~(7OTLaog<&^o^3(pTraO2wa_rh)Q=A`&RnY6u$*d@67`KF&*3k10AWj{ zsv2oF!tR$RFf7)ozg6~o+$5k4{7eZuD@)_GYt47p{Mt~Bx1oTQpOB8>6;|PC&x@E0 zv*mX0xz-%Nc5HZw$LYa%OWi&9CFS~}kMeO!$W&iXFFHYFF(}$I>CQ1| zKvGC{+U*BmJ4I}Udhob;ct91$Q+UAC_65?mEbV)M6Cbe^xyc-^FByh^FL!(-%m~~H z&*W*gY~iTDDu+kH2EBEqn-IW*wf@!Wv9+pdTi)6#FjBblY@|kIZ->Fe#C#qa@HEw3 zLYLDDS;&YVtN~N{bJ;6!CIKDmwLoTFm}-z)5HGbAZ^YNu)_fk4Xn4ig6oZgCVClO- z;kd!o5FM?13Zf6=q2o??xgmWexyBt)#vfdCiorZsCu3Zl)sT4=$~$6U$f;OyllT1p zILyFuYe7@xy{Y2OR-@$ z$3fLduqX?XH9e^SS{GX+7=_iC54~`@8d9T#e(&xHxzEP^{E9`onp(JbXiNl7}8s!Kuf<-0KYaSPUOvk?&`&C$T7eT zSK(;Lq*RSwG2v!qpWkgF#bNBkus@$-d3paCuM3PvZ0#P2e8{6h7&Z~Vmtp9j46ElF zBO7Ka3lJAKddFTAH=rS!3Q@Y$k12N*m{75gF)}bRf`NI}A&=)N(HY=e1S(r7V3eta`4Ojz zi4O!qhszP%ajvbei+lH+o}JYJqt6P*7th{#pfWxI7;~sQwrX=iXS6&F4JziS^dR)# zyUGmuf>n#xgTezGVrsQKGn>eim>3G><(o6lDq^;^=jxKc1ip(TWd3JI!>m<_0Zlbl z*K5@e_4h(VCu0qHrQpkn9H-Szq{E1W;TI-*rS0toD6&dF47BzB2;2c9309~?+l>Fo zC9pOOjN#%HkfgjKF6NO{7%;gOxVyhE6FbOol2-535o^DMtzZCdU4j~7-JIR+hOS3~ z|#ON4kUOY_Hh zPXEio#!>`^(^C^lZV~`&a8UoE#scn#1eLQo>;ArUf2c^b*2X9RAK^;_(vP;)nHgsE zhi@Xt62d=E8DJrs;5jBi&=xQu8oR2O=4DqN--Wsj3=C?4-0bI^xA}JOSBxKoS|*PG z-={VUXu4@7fCVZ{E0@~f0LXQ`c1kmDEo7*%KE;v?HcooBC;#_0yPauGaC#WWhc0Ot zrfQw)~x&KSUaHQG_Z-Y&ekqcKBvy=0#+!Zox+oZT>Lr`xH!FLy$186V$GH$E{k%bju(dXUK33_4MH|O<0!Z<;9H5#4MS5P=z zFro*ph+rE&Ngm?Rsj~R@A~8Mb57!PH1=YnAGN%3WQ&Ur)jx4-Mus!_GsN8q40!`pT zsl>vmfk-@QLpO@_75uo}Y};?V-y#hp5ZxR0re>Bf(%_{Sr34c>tJ)H?II|b8m>twrT5@B>flchpCg#vc$*!~)o#{3Ai?_Jj(N`k_} zp}>81VUd0IaW@x+^=qWn5Aus>nuWP`x;r-WK5W=VW{;kJwbw;^g{pu0-d#}!E$fF# z8`epjQXLc$YjeWVJI8@ziC-ZCx#~i|e4gnZSkS5MUp_gOq40C46}qc7LIS3}srwDw4A`q`q9q(>Ss;hLE!+#Q97iJUT~ml3TYUMvu~z26 z08Cpznu)Yp98ZpvP*DAuEcn%y=_SqH>hB+I(3erh^7@acmNp_=u^4@`Dx$*nl(xgQ zgCk!NsxIS(tC)yWVLsrlX~WW;I5&jm)Xq~Ncb%k`@$z_bk&x7Lx~!3Ba;%1n8*f=n z9!fY)81)Y*r5|o0cEB$KJA1IH3RO)b%qJ)W+pEoJL#!N{F>MH5R$8N0LoPLfb#M?Q z$=(x4=u^XR5?#D1IS{gbv|s^U9{!3cnn4Sb5`Z|Tk`9w97CXp@MZR)n*MN5wUJ}}L zc24f0dPLOA@raDnj5ILOV;m3)9`BUjz*vMm<)aQId9Er3R_3iZH2-w~w&z^B}iv7zN} zTn*9NBT;T!t~%ceVJ7cP;3UON9&RssGp%w`Z=)orlOL%w#%jDK0sri?!M^lC=>E7* zyS;0*A+&-%C`$r!hL8(N8ygwf4Rpb5PZN?h4>1(o6uU2aBQyu~lOlaU3oSn4yi-KY{smxtQUZ70XJVe*MT?2nY1*j~=eJTs z{as^h5-_N0(h6Ez86B9*iGORN1ahWS?VrF0)FG8TnPBB`pOar?e1+Q__&;a^rkO4- zF9+X96!>m#X1H(&YtD;}I!}J1d&qad>3a0HM^gW0d7MKAMUrxLNGm^5MS3{=ra+<^ z`C)u~J1MsvbA6CKC^Yme$ai^0J>mR36ih&H{7g0uYFZBm9=Z-8?9R>WrO?Tf=+J~b zLS_eFX+l%dSdr0vH(RK%?%C^X#-S6hU}lOYyD2} z@CXTuzf{4pIeM1WI7qgEgJne1@&iz8XkAY}XiSGCdnw1I7t-ouT4~ei?Ut7ezv7DM z2I^99g=|3u`0^&h#`%I|)@i=4Vy?LG*Q>ZAAA;;04_DV-D^eaRqc9cG>@G}Ch-k9< zcvLx4UMz>01_;sgM2XiF$fB2g*%KfDM1W{cF#%@wQ2mlmh{&%$qDqh<)WoqeA@x7gd;omxaqv{RfQ60KG&6HITer>|c+ zf`>x>%)QmR?IjP9L=#Z{!t|uArj~>^ma}FsG%!$DP$1?b3_L`>cZXyFz(Mt5p<|)m z>UI8WVQv!OPz%@S~7j!xK;#IG2dr&|rBu-&;~P9?7#x*kTJ@i5zKM?{b0 zHGvbxs^ud2-IYZF6E{^3hyn@M2jlvN-Uo!ovWv>L)h?e=uJHFTMH$;HiiQ?(VdD0M z)H?nW*lEfNcpu?L4WDrb&A1XBKwr$yb(?WI%1a@*_-TpLCBpaR4NjL3Lzw&;A@sh` zfN<_ExPDFADGPv})fPRj6uMrxG%I&*jr8owSJjK8*r)dZ?%8FnAKCOiolThGOVpjnyH#WYa&q+SBQ!KL z(8-3GK1h+eT4^rTA#UPQR??>rXr8GyVSww(2jfbE7T4~bJUL8G8;GHNk}0<^OQQL# zs#luFucwP!zpo$;lHXNL%h5El@p25%rUc0> zz2-KIG`3if{PFO^6&)cmOhXi3hWiOeOCG8IO=;Z)ZAeys&DpNH*Fd=ZI_*h}l*ngy zGA5|d0xoV!QW}0jrVHxt9pd?;DHfJ9RgFZdUP(dIkz_h8H-~dU4o7EcQ01oJS~Q|f zAE~iW;L7A5OtajTJ^sALKdpqFJOZmpA{;4)iq-GXiCgmdec@glq^UODn9UJdO;2gS z%O=gS@KzC2z39Dry+mQX7{@p^>#0q(Z#i@baj8e`^mtGLii$TuR0gz{Tz#xSI5gA=di(dTC<$xd_&DG~`@Y#+zC)I5`=|lqfafS>80t-Uvh$jJ%G`h1wSyyn&4k z0`3IRfq;orS9joEN#3L;1Lm>-94bYie+Z!jZ(_W{|BTWFkjv=UIdFP;!0&!{f;fPv zO(O)3*5l!FS1Q7So$DXh%4iuc48wQXSrR>Jv1Sj6d4HR3P_hTNO^~{6e@$ucihPo* z9jD}=avTQvP@AA|zpik25d1(0GWdlQ3}7A<<_2U*MWna#4`!~)k1};ikZU*ulg{jX zyf}shb6o9o^^q|!1}70CPz;Vnffe8bQbVZhlXNO=AQ@OwAaxS7UfSAJ*5eRS3`4y zS0>=at3clI4V;wKr?(pKR732a6=C)}29{i(va=oNmta+Yt&R~heu$i99tq+5=m zvDPMqa&t*hs1Vl;9L0$Nv>B(0 z@e2!!K)tZn?ADHAX&s8=LpFujJ1RhiC(7y`D};fu>nX2&d2S7$Te+NGW6K!}2)pM# zdbUnvL8apW9?3}zg`OKe?w>5NIlHoLEd#+^a(1_B-plx22^Rg( z<1T-b!|^pRKGiKwXh%+Q9l0rB+6A5-2E|GWPB1O$n7r?!-!W+TZU{(`3%u*fNMs7D ziLPbj0xOl|L#2?5K<2m_d0IfYT|RkP1n6wCayA2zuYv%C$6E>cam zV(XH(MMQ~oXNYyjxLV9mY9t8;Iy*aQP=n&)(kB>*$(g^5Dqk&JUS1}$Y^X+B$j*el zS8cL*nWBUEX<1njEgR(vewDNiAJWK7u2AF+MZXl1`Y;j7tTbZs{eySZn6pk1)->J; z^f0fo5CXo6_v@cc~qxfkN|GM4X8?tbTLvQO`WmdI9m0?bG#d#$B}}->zxV!$p6B-CFQL- z%jV4NI96Pw&?iO=I$g3WO}UMEz?mnfdk+Wz7PV*h8Yj++Wvhd@yx)Gsh0ei2l;RUn zEtmKBhv(m;zn4?l32v=1|9CVho#@Y~wftEfr_);EJu0pV9MA!xk;7)6f3p>w%72d@ z*GSI$HOus3Eh7p;H(2E2d^Q>ENhK_;C_}8e_>_`-*)vHps zxEWJL34s$;wJ7ZD0}GT}gO6Wgl@*Q0AkX(8(;(7_P)Whp!S$C`rM>_~07af;)9gJh z@=dzTIv|96Y`sXr90z3r2{|SxP&sU;03eIza%cZl;M^F=abhyQh^iU}%dv$g7m#2+_k)jzzjPbAIa~*MStG9}| z(A_W;yj(Wvja)V1#^TZ4C4q*mcT{Ww1b{5!;0t-p-?zgNJVx_8FL?(5U(yt`=Obh! zm(^o3Gl_C;mFn?&FiJndDj>8*8-J3;#VS4@XdQ++r!#T>kdBA{PMGL^40?Pq6%WF; zU+a&8pn0%jP%2A~-W+jQ?tXS8{Jv1%X7bUxcd6wS7C!!W2VresK49T9$L!MY9oc+V+%I2G zvwZOO(ZR1kP-@k~*^%|K{=q>{{UQ7gT&jwb)YvmQ(e%hf zd05pb+8)CgwIzd+z?>M>jOpgP!?sc^<`1flhKr|Gz}m_LeqkZllEXi%+mcfMh+0*A#uCow#cUq~18q5>*o^V{ z=u@VI+}V%C{axlyj#M_bsY$1h7cv1dLbt_%Jm##q*uw)*+)aP3sB;|yOJ$aKVl)OX6Kw+J5@;Rjz}>rWp<3> ze+3A39OCK7n@F`Rj=jJqp-C%-jRj1<;8j8+4;H7}5;Bynu%=(4fw|{99h?jMc< zhX6xbDZsB1`KdK(iE82=YBlOFO<@p2@FF=gQFacBO2`&73B)XbeGDr;d!3l=Y|Q8Y z#d&Vz=M50AplQ~ijfjdO4JyIIuBy%x=VnH!JRFY=i8dvdKk$68R51ZE? z5#a9s{I}qNH`u+2`-eFIK`|3@V1N1~L>py7`p%EezmXez|AsI>qnB76$RqW&W(rWc zh)SjRkQDMOJ#Lp)(xg!ap@PBPO1v)yfw^=9(Day}BqMz#B!{v;CgePIxO^tKE44j%Xl_)-nm@msl347Vu2u7;Ykt zLDyW3U&F?-HpEkkvauH8mJ(&TxVRK(D=Ww2Vdmbd1QMA8Zh(DN89(1_4PozHTIG)v z(zCGy%{OhCS?&!sNxlt4>xLggDyLiiUO;4aPhtO2x-X(>MvCZ7m*!|q07P^_!96TrRO+YJ@2;miN7`Q6!jH^O3+7C zVAYuy4Fw${wMMVaeSY#|rFVV-R?YYExzN;rXlE)Cg(e;ktZ5sUc#-4QqQjl%nN(1> zzmTL>m&#p6f$}$?9my@;+}+VIGRmP8%kdA`f7LBLG7edmTaXiOk?Oi$izzrp0$ee~ z!+SAC8ZvPt2m-4XoV}LEAjb9|2Nh1T3qu2PD>*R>mY;N4ESzsTY{5&Zi^gDA}6UiefJ`;+<^+)hfKKSrTvQ20Tw#)6lMXY8no z(-0~UXJQwDnWOygr53lcGUjlFjm9zl8tj_#vE7dnV5J*`h2hcB)-A+(TKf*bG{TpW z6FzgP$A{?M`+f4wDNI#)S>H`V@qGLC>*2VUe6n{vzKR@VVbh;m4ZT9~H6aY2$e46@ zbXwe@c#Udo8O8+g%aQ-h&|v`ND@f71g%sDmVU-G5qRAv*fZBa^tH`@`uZS}eH9>~G_R9t^|=;pBAPB0Bhl z;+<|;iQW+>Cnp~t-W~O&o{iyM)_$bz};x^pze^d;%~RLzUMZ|-7z2E?&v zWiKq<0SDgFmF-tnZgqCs`cu}U9Kw!VA5`8ux|wgnat0^KQ1uF;73M#XW-g2zh;^C) zq609AD9S<70lbX#q>2UWFcevb2KLma)$+zR>1?}b=v!KB&-64l(6nfOjuCfS{P81v z&d3*i=q%Y3?AjZvuiZN`_VS|@t}-QmypN%9>Q|rZ>sIp_!x*ZjH<<~VS!$7}ZOW#L z`86eEl(}u=vEhWklKR)nz>?Db-5X*)YR^gV2LI3*Mo6ZB*%`3>1D>MQSHI<_JsJ%& z55gHSwu+)H>C6;y5CAZGr5QWccuoxf;yhI)h@$cZDsu8h{K(&%Esh!F#Op6EcNn=N zEMfJ5Se=47VGz?)_1XEz*bf-7M&6jWI9-bO_VOT_t4D3400eF`35MYwc=326u=6Pw z`j^RfDd6-u9SjFaQUMLm0y48kgW;JR39RkSj*#SJt}xJ8t>=NBKE=y~IACIF1XcxS z0=72dK2GguL)fZgB|$-F!^<<<{xyN!j;Zkt=D_-qs1tixybm?g{AFfe%Lf_h*D!(`nzs{#ttHKU38AS7!;*nNS=J zPmT3G-b+dS|Bt7yjEbuN+9rnX?ixZG1nKTBN$HU8?(S}Bq!ADaY3Y(KDWy}oJKrB6}OfIU~kyc8Dn=Pl$P{Nu<3Z|A0{tN0Zc$wHGIMtA`veuQJ^z` zjyA?w5RB?admU`5>HH*qr(?FNkbU+x3Sj}0O=MAkXu8tjr!9lgRtAov#5z7PNs(EpcVO`RRQ^olyhPMw&v*iSD=0(3p z_mqDH;N?XaXdl)B@z>PuIc>g|iAZ`OHqg$4bo<>1)AIyR@~Yp@%DMOG9i2C$hyN{4 zV=t^0Hv~7dM@?W3-{E^J|F+Wc$1!(~lCURQGWw{58ZRPtv72ypQu^?;)*8HTF;2TB zU*TMHo8IS3l#4>GFQqy=94k@^P3TxnV%oN0GCxFK9yX2OX>5kgk(b#W-v(S2Q@9Jj zLyBzRYE2I+y(>LK30@U|idVXJB2Y0UyXk#w`vxEF>)9K)OtUEP1Jx>1+QduMji20* z5l$LTU(tuZ04?$Z^#yjo#eVr&6^Mv4fX z9a{9&1Mlzq5^bX;+&l&8$S%xS>JZfnBUD!#YII$H3sPWZE?N2X>)E9 zI}#gu45#Jh$V8zVMSDCY2!Hc{X&g}DMs;gqb-VGCL6CbtK!~XWl+2%)KbT5OQ*lFt z50{n~GlfW*fQ5q&w!#>bSI+^GDr;l7D}LdDXjYWMNzlB7HYnlzXIvwW4{6S3JAQcj z{S;G!DnBD_YF^BWeUkO@28UQM<76*+^Y^cI^!DOW5;T7P{w!7AS&-|&cn$=#REWqj zLwOiQT&FNLT`TB`we;_BcZ;|L(0mOaNj|+^+yYjrGy!?9o1BCwti~Rai1~nc>tHSwo6f@`GeCIPdDiOlZy- zA3e!RjvYOhWq%9K{$ulwh-4jSBYN;1S8n*e@?tOhn75OBm*4?1N%EC9w=_b(@%~Kd zbJJgV$QcK1OF5N#hey-T*4hE9|5NEx1iKE_*Ei>*zg&&^zDIC=f03+(lEHAjm1(;( zhwnl8_ZeXENYg`)lxu2cz&4hV13AoS>Sc=1;ajJ=^lYjwF7-4??r3vsjmL9OihMDs z(h}vW5ubh*x8fDwS!NL^mTAZqu3wSyyQpEO|BY) zjIuKN!NI}QH|vaDa{%x8GCUHQy4t+-(I*jM9ny9@&?M#L#Fmtl2rs!A$a#CeQQRY? zjh=~~swx+6g<|+vv36Ui>}%8K{@*K~%IZ!+F$ON>KpIc|2bi@c`<(m1WCt*Sv(l2~ zwTCEze*^6iBr55&^AY|0{G2(+U08(sJenN5w@g01l9>JEGyJ2fIXL-;kAXj#%T{cG zox9lLvT+N-)#uSN`a04WA77J39yFjiI0ASL`8qu+S0_*Hff8VpRmdLCUF;sYl_k^45-hIN^%{5{U$olfHLWVqlaI#oX zJ0rLKpRhk=`eQ9rs8k#Ii%(+MfP-Z8(UDzA>1P7zblVb3or_9d=LAe|CKy6v`+g^d2k4GCJofXMWq&T!Z>%%5x>A|=WvU@nU>>QA14EM4;%3=++ zDfR*^AHEFz#IzO&*9A~ye)1RSd>}rL9!|Ok-7o}X4Vp8Aw8BLev%Z1NUUnmWoYdR& zy+vC5z2rH1Z;oPup|o}SpbTETL#m5&jb8u#ll1WXnUt?`nI$ zHTl2u!tQf75Y&?V$s7HK)g~?nS5_C3PVrJ^WcU&wCV?OGXh| zCW`86*&z8_(aqo2^fZI3i!?k>5)J zC`rcie8>su>24+E6civY;%lm?tLt}M?%w*_8`Dj$AtsPF5ufbK15Ak7GqbkBC~i%A zhe2o&tfrIoWtbwZgp@Q|ZyAC+;JcjcUxJwrw(BeM=D9kcbNHQ~@$Ycx%Y*W_v=Mk0 zbT@A2>L#`w^P@D^==nNFl08%`GG+WVxJuTJd}9}+jkWM{RDRmmqqN<{IMRC*g(>)a z^{dwWJeD^Z&Z2k2Eb{~pDX$34`g%atktsa23T1btC2FANg_X__*k(bR*j}goZ&Vq5 zF!&sZ({!sMgfz+$pC&-ksL&&M68D`hD3iuVZgAN-IqSQ-5!SBK8$+G~YHAhXy~D6O z13OOYhg0(oVNJczhA-%~p7y=>v%l^h4r|2rNxr!!Bb7P;3<7#TrR86K?&2eBJ2-fD zClSTP)Y5Q~1;z9fv}>iJsqdjf*CA1762v zm%X55a8$rPD(@iVpG#ryU|?7G;6$O_21myFeN|?SG(0uJXk{zIMp#R1CA|WoPKw&e03$g<1!=I{hT` z%shlV6j0ZUgCJ?&@p(d|A-(GzPDBSWDC=gYP^HDLpZhrW+^6U@oJa(aV*KC+6JudM zpisyw<7_xpN~hg5`YxgKrd8Ix3{rG|ut*f~KmUdZtr1bcV0H%Zyt7;luDogg!;IoZ zRaG%l`bejYtq2Utwb!KJ5vI8HGrMJ}MI%~2ct<;3srL2km>M$6kE-hF#V1+u5tF?8 zQtf~@m>GYDt|qR%99*VLhI?}y73D1LV8dbQZrm|~wTNLyCMtTfZG}ZGKBBg`x6EFAmA2{H;wYD!hg8=bpb zIT8#C;W!{2Ia*tLlD)Fz$w&h*t19(;nQTLfBjG_mmy1+ePL3@=$`I_>GrhcY4Od(C z^5o^|`(1UZ#N~}?qoB8rzXu|yMvg!ZkIF;c&SMt4ddjbfzj7RSODf^KRNZ|3vCQ3w zd)XJb1ep;PKMRu~gaH{ek?7ZOSxzLJL|lAXXVD4!njq@)Dmnq!`WWz#yvth!ovUxU zjmQe0n1e5jUc~`75+Ecb+CiEgKe91e>f(tb)(GWPx@AF%?Xqj-5wyzN4(DC1ba&Q= zaSoW*_4f4i&<&1-{=4y)10Ii47T{M@FdDvS96@udYAsK2BhJpxn@+A8(AP|K-|=xE zUK+gPo+~UV0S(X&oFQWtsV0%$6ZJBUEm8W@B{MyxL60|5(IRu0UPQUT}?W!t{t zrcY8~6#we#>Yd-#-ytS?F|ixDhC?>u&-{oIUkFPndfShZwNitdQh^M|nw6+<{7og)^3_8x|4y3CfBmpvX62^{DQ^C+VnOAD?5;1K zfUDCd5PSW%GMZ}K)uhdjImoR?Ir6@~0;}GC$y`y70Skm2t@9svI4%2SIZju5Y8pFf z)W%;i{E8)ER!l+CdSS}Q8yTqp?c7(%C9dV?10euWWak!QV0 z!Q}W&f&MMeV5rLVU`lr?xMfdg>g|8vq+K{v_?xG6MD@FrT6SRrors68+rm=p3g-&0 z516}Qccv>Vp7Qg<{b6A>$*Mj$T~LlHF1hn)iU*8Hst=_Z;Zp;&#=vrJ&hajl~L7r*l;>uWsy&IGP)O z@*9OR9Pe&ir&~O4|1e9<01Kww)uTuM0I*ATbE_%8V4&YIu4y; zH^5^O^Inu&F2;-&78cIrbuRiC^pqIb6@l;AO5cmZG}|YwMksN?fPULR_hlaIb)QG2xHPW0dhoA+JDE(m5A- ziOOFO`r7|q5FnMMyji>^rWkL8M0@zSAH#n^$js>)EIE=m{NIM?OYKgog2xqk1lVo8)QN&H)mls)R(rTr zJ^;nc0vTXRu}2#L)At$vCf#>ibr0bHTLjGN0be*6pbCl9`g(DiKhA6F(ic11*0t0{nL zL)ArtR4hHZv5VB=M_~|lkvY0eOn9hL z4Vu4$yt=xYUamsY;sZZ&j-S+~ajUmJaXkD%ghzu__WHJE;{1Cw4PqG$ns{68*aID`%$ zK)aBfoecHmV10+Jy(@$@V$adqMnQISpOp6}Zs>d>^W72|(wgQLfIpx#Ij#vWM=+1qwEX5Pq-n-gi>1T(64>Z% z?u57G#*y+7Sjb+6ENp}68J-qgsn?kalY8DmD(EWH?O$gEhwVQ4QuDZd-yqxz_7 zwr=#pCY{$PjMFbP1uRm(;qPM>loz7^ZZZSJE4q-zcpwD1TFY$mM2H=qXY)o*0rO;T z`KVSo4Q+ih}A3Ina*=eK6k1Q+wK$-QdjmT83s zUOF<8%B^3acXo5J=GSc)8m`TcgLr@mgs~2e4u^$U7mQ9vn;MYJd@b7Is#t&k<5qLF z1EjiYc#s!w;htSJhK{T%bD&nx+hda(k@Zq5hG!#l;cgPmt*TMESc`eZ468DVU-Hb6 zs2Ra|J&<7Xi6!LHu%smLn7$x8e#~PNwa_ZsHfREjB*+s2o-1YhV^EQ$0&c8lchma> z>K}zt%ufMPoocF9SeCF^DXHnFL$54gj|C46$+jTCQ>rW>m!_tAXWHY@PI*0wZn#t= zXuW4d54{U%DVXZ{;H*Z;iD-uXvk4uLzhyT6h5N4@AT$I4g#Gz7_F<_^Lw#Um9<90N ze@n(V@NsD`Cv=1CBv|4!SK3@1&JU)2eZBY{hECNqFs(r*+brzsnMPvShX;lG#?bVA zL!64(y3uUTsXpHi1~ZA^5UrI#k-3ba>IMO%%dFdh|8OV0H&dFRDg^;Fe>trGyWERx z7bUO*?4CX6fnf^3)H_p`P(4RWoLwoRaWsoH^ouD#za7-veb5+AR8Hr4UIj2l&UYqa=jm1E^`>v zs7L!4$SytT^{y_;prco?z3D-EHWpEOBXQO8TXA?90JL~c%I#LiygK&@Jp#7%;o#iy0ifH$-jr^Lmp^Ij`1qbxf9rz?Ai&OJHjOG)ak z(+@N6pIjCg*;j;Mlp4{zF88#y=RR#{QbnZmH&9!=yc2x6+frp@BeGh=rq1qfd%l@F zyLtx}4)|3NA&r_CZQm|c@&$+s#>uU6SG{X_``?$0xb^6*87D_6pZN4%JCo!La&2Ttt@jBrc)32d1{h^Dy8dsT3a_-@ z<{gCAR^A>whgsmH;JUe6c%TD!p^QsyV@Mx^GphFfPfQsp*K!frxY;?PODOgYz3>c_ z2oDbyt6Hd4`7~u>&9{xW3`Y7h_cV8p4_gr=4(z9)|IH4mql4r?k9)<5C74_Q)7G_o z15i!QFx9NvS#t?8Ru%?6+CM%06>0tu90{)iCd9h{QDYsQ-+OL;49MQTm)cxN#c>S% zTr6<({n+UXBP@phgCKY?zNg{-vMt)EX^*ejxCe$**pJY1au?b08%Q#dT<6QrxT|7S zz>eVJNOmljPu+Cv6#!n?1(9ZFXJx&k;LWa*va^wmwn|Cq`dmLQGtoKS|4V%-N>QY; zai`qRgqI?SGqtq0a+ocst|pUDR~vbjGI=8Fr><1=@|TlpXE+Wg$o#FZL)S}75{8h8 zDh!4ZK{7+rJRVZE9%#>bK9_Q4XtGmV*xya4Y39n;fw+<@nW`I zRh`(0khKvuHs~($xhgkUJ-e-BjFua$-U2a@oeq=h5(_x1I5@TbFhG@8drJ|id!EOU zRD_%sy@%QfYdvEBU*EKjj93fh0e9y0YZYnP|8_-IUQT>u7!;<6-rbS|R(t(G&6vR5 z&9w~8@k3#QjVY14=l292Yv%-e+8>kouY5wgMCX7ZhSBSVHbaA?w)5B9{Ih?EbI>ss zP~h#Ce?H*``kqlf4`Lh1sQ%FCzrk2^Awr$kns*Eam^7~G{cM3hV>c4!CC8-1>;Y7M zj^4Ja4`s7?WaiQqAQz{oUID$T*}-hZ%AKxl_^^P0$g5_qU@O`NR;r0*`GG4hu@VV3 zwnz6*JvyS&tKXSGO-$#sf&>7_lp9N_?DL24%NZk;$h^^3XaGU28XLf_QTb;X2^*B7 zvIt{(+jW!wDxkZ6ACh4aC#&qgUYnUkN5InQ|2>Y_>LopA9k0p{WI(NvQ_hm4TYN zHcIg-eNjzI?Ag#31a6R&l0vHr0A!}Aq5%P&EnSL9;>}AyC?*AU&OZB8kDyL;5bpeS8-zDA+#P z1d%ZYAF2ci>QK_15Fktp9G&g2@Tcji2Za?$ss&E>?dV5mf@-RRdz52ebDtW1Np@Lp z(VsdolMAg~C9h=xmFAbq$`GR@PE!lX-tj)H>-cN>amVoT&!}6Q?Mdk~aZt&yb`6*f zA320(S){*&-R;n}bKr`Td>UmVvY%Pk%ei7Fl+0&u#TTZM#p1=Y43xETqUJ`eT}G@hrQv)n=q#-ynOdM zF#b!B^8?h6sF2rYaTtIJyZFPAM;B0uIsNhSYh*w|Lqk(-wk@zp-MH)2A~e_airO>7 zB~BLX+6IG)LOFLpIN*#sbt*N*`JI}rO`{~8nhX?24k8;VQ0kDC(T2%<@XGXmjn&xk0!Z#Du?y*t^zdcpp%yGGV4^bMlJ*yTSkxbN`# zKAc2-FmYa?yI-5QLxg`11+*|sw6*SsTC?xAO=reLb4Ap_VKsDN@g9<9_6P?|+fyEd z8=HmV1`GuRtEvaw)ri5e3s1M$8HiZ)H0&ElblV%Ja25)mOd zXfnmu5@sX992&EcG=@*(z^pt@2jX9OwQN^VC`cEE07@WM9%Q`T3A&N1C&&1+QT;3FAa3UcOu zpO;4j9PF}g><__ulI)EWP>E0!s)V8ds4pGAN^Wdcs55J|X-%={TSpWmoHb8{J(O?qo;H2)?9*wa}N3x5T|=~)@>L6I2t-N&9! zy4l^uXsBS7KgoTcdljrA0T3&(|Jx+lzkz^Vh^Vx=M+T*%s6oz9dPzlZSK_NSaj{n5 zv{yGB{8(tMH)P6e8pmM z1CKs(;Y7v6^oIP7RdxRY^ZQ&X>H}$T$~JNXkM)$+Lc-e#oLY-<;!9Ajw}jz;AZ9Q* zq}W?8)i&(;J!8|NoE@FHQ{`*7&{$uENEdi(@C4Yl)*hT1aWr^cP0O55kT{xndcf!( z=hc2x%=O+A6U%Vf*t_>2E(qIgW*g7E?Q1WrxB0ldOu|-~|Xerq4wi z13vUrfSR0Phxsi=XJUBnGU}|x8%hz84`BlBRuzUuMhR_gFmV(FBZWB+#y_ioG6pjV zX%a6h7UnvJ&RP~+g;h^{QSjCc@w^H~$|HSofiTdzg|WN35?LmJ^z8y)%;{uH3-2(> zp#Ct5;yNH_seJ(GCOrV(Ivi{4r)pb;(`e=SuHit!uZ>uh_v@i*Y8eU?hDR z7~D@uy_osRyy55)cv{QEOfI@@*}jtdEPq!q!%Md2fLsM!h)=;KKV+eD`+2Vd*{bl)mM4Y4RTL)`gdF)E|-@f>=q+5Vi;2&$?4ZHh+ z#Mg$sGmjJ}6uD!#Sy6d%D+NVdTTgCrBv_C|$O}lLXO5+Z*Ci zze_^$#?Y~D#0dYAKo?^+4@T-=WbgBD&R!_uMxE7_Eyn=4&z5vfLP)@Pork_^OnlZ0 z;|^XQV3+!5^Rnn{!!!N7p9KMFJRv9lFhnPF5No*5(aL4jA=bBYM0_Hm=($lX*^L>N z$8;iW)*|ri2d%6OhIW2`BVJzK%tLzc08k;*)TkS{^7XFpukZL+ceI*hgg1l`yM4*E z@1?98ow}oj@J`$)9&eT$w~RNv8EAM*RqlRI25=uq)&gpZ*eJH2qUbaL`}@P^X@VEP z-Dh)@tt&16kiz%q?PV)3KuU&i)P}$N1623j2fzVYHaU0<4<7|+4+~oHeX0R(I$-F7 zZ(D7d499-Rr>(X2%Y5_`)r*ecrYpv}q5&h!u|%EAfP50~|AwXTGTk2qYttobRn~r{ zJq9pG8;_@(%dXTB^{AIoyX`(H3(%fd9i5Q_EZrA^L~1&^vR^*7uK-wPi_fYiZ12JT zML<D&Q5+J4NWL^=L`9Mi!N+< zUy(NBd^KOwGq1j=$#_WzGBcHYP3^C7u$P|XSTar=I|eQ+priY=sU5#H(7Jy*K?4KU z?03THssH}n`R3@uK9}VLX%gkUUSTVSjl$*Gr|`JYiv0F%76o5xrBs~iA1Uz% zjM)pM`gP0LmEEYtPk|U8oWk5?-(>h$pF`bxaf{B72iHK+vi*GXZbs9^+_Q~avhR#4Vzz~RddTg>NpUNnf4yR<; zM`dK1q_X%*0i+A%n{d&cph9XG2v4d98{(E@Lww2Z?tz$_>x*$oO1u%?J;i~GTfPDK zFAv1vF~R%NJ3x3NaKXUz6w93Uz9lImVY!!IGTi&a*JI!Fp<$SYLsCpk+=LI`Ev{EG z7yfHh1li$rM_v@GHjH!LwqG`;AW9J0_S8ljb1C~B7KfIYb{a-L@QDP%p?MR7H z>XYgM%Ue5Gd;_u&tm}+%N6oY0{j{@?y$-ReF9!0nano!_uy%0b!e8o!Di{X|UsV}q zg9plHiU7Qf{x4?0_hCOS%4ElOhjY(Lc# zy104RJzQAdn*tBsC`>7_{cAZAp;lzn@A>U1Za$LlaR;TkH4cxH2l$U_#~T~dr`*es*O-sQtlJ>V zVfS*CicxE?&41EneibBHdz!GjVW-OpBW@X$>({ku1rplY(FImyM3krsfW^v}i;FKh z;rGz15JkTid6-TVr9!htH~xZrYku$q;hpSuaf5xhimq`W3&%WsHM5Nx%kNsP4~T9 zCWlES?oz`NdP~TE6*xauCZ8*vVc?ch12>l_onRMCU<0N@szt#?l}d8$xdiWMoX$I{ zLApM24;U(8ARBipk7qSRK?xzCb)CFD@QLZDS}>6&U10VBJuK?>!?z7~UPjfGRa6c9 zYsL_%wg<75vIxUvbsrfx4BPY3{$Q6>rNo@5Pz3d)+$WYYrNgD{4oVR@Fa2iI)n$^} zkHgr%9PS`yYE0Ca`1tS{h!}g)(KL}lc&*sr*9AvxJZ8y=IfM#Ax02G*dP~}=mOcx!PqOmH`fB2QzTBB&(f>jU&(tu|K zOaR>Js)v~a;)%IzTF{~&`+%*5hP9{gRAiEZ$(<`z3$eWN0x(G$m1>BkGGLUm&BZ6+ z;NU<>jWELE0r;>0oePq?@K7zS=&;DbHv=a9Rn8<=ZZxd#;>P7gGIpbzP&{iV(NVM* zwSUml@@fV_3oTUf19skShUDHLAoIQ?S11i5iq0n_Bm~f?a+Q3-rcPu`^lHHbj7tXr z?K8qhO$`YmKJ;E8qb*a+nHh?soz{)eIG7;qJ_ zgPj027?uYMoUj;jn2aw*i{-Ds5mW-taYNevi0`~r@FyTF{uVzNSiGJait*#^`+?1c z9nQZjkQV5(Xy9gsHAWwx)B+GrfVt_t!SaX;Nga?c#H={cO?!Is+h?CAENco({bDPY zlxU(3lGV{+_r+SiULA^NBc5G!NyP7y!4JxD&|!?m4gHfz!Wqm8y12MF+l*z~OziT9 zKkd4CEBv}ZaJM%;d6sBt)C*AD^){%SG0T#~47hg^6}uTB<>)ig-vSI&i9G|1URnzd zRI=kC9B`&q%V6Y63-*oLmebjt+|g3=B;%kvU)~YWwLOoa#}24FH}{QB<9WCUlr*JN z4Narm5X*uyC&UsC5-dIpyLto+g$mWFT%R zgaQ)G+ds$K+t2L8HthZYk@znm8fgP~%)l!*(W>*Dm)%DGNarY*@uFM!fsH-@{04VFDs=x+h#O4MT z1uy#Fih$lECweY@vtxLZX$~v*B;SF1lQh7lzz+<(GP_lA2*2E!^M8{ARe4;cxrB58 zVvMQi2BE&fuSQ*lBLk&cRkV}tBXjOtDBg>|mTdrj`bQecdtl=4geYk%V5T)=G$QB9 zz0aJ2tqgVpY+nS)9*@1%%~2}VN!~JfacEWR&N=!5bdLuxL<8j4X@WCr4(~Pe{Fe12 zwqS0@)HluIT1JG{Xre{cCM;`fYlst~sb=bue|*kDJSmL9Ku!13X{6rKyOtR34T4uG zTMLeNKW|M|Cj;Sr0F>#r9kVCTU-(ED=fcLh z?Ostz{PV~7Dez_jR=Dy|@i9?aT|>G}$hiF1|wtfk%DS2Bwep~*+# zQ=l%e|2=n5oXLR#@xcdyKspAK1THYS-9vP>KG!z2T|EHdc;O7N2a45+JkQMFb*xpo z=T|oF$6Ea1$##$(vrgYE;T%wW=1HCR(frD7kXygqjhG;H(rmhy7V>k>_#@)K`?;*_ zwFTl0-oPQG$N8%FM7*GR85?WVXF1Zm4g8tpO*bePcvy_`Z{`KjAQlT}tBQNB#c5`+ z&SbXnSjd@!KzZ%$D=_5<_c0htNLm@~x|&f4MKjkq)L_K(iP8J?UW&Q3ptCa{#;@Z3&mS&Z2Ah@N zS`j0Ud@x8QWo4VOiCrO`@^S9b7-l}R?EySr-K#Yh172WD^2hrlQ!vhf9K?j~^3lMD66X$dm#Q5|PvkfaMt`{>u)?COpx<={a?YS1V<>QpA0Y@5kp zj{E{l34wf9e29?Q&SmEBkLh(^XPD*)Q6nto#Wf^tPCd-eEt$;ihVTSk4rrDExXM8| zgqct|B65=xV&U?M42KJ73QzT<@zKW93a3&JHyF+=)Hp!=A0j$>)7MVh#G}rq>jfhM zo1nVGRA6DfzYkr^Il#a{R>>FSTYn40rTLInXt9FJSo1e%;yY}1}N#_lG5uCs$ zlNp(?<82c_U3_d}gtMt>qD(nJd&kXf3h?B#n$7VwfLU0vtVeeARy?-r$0Z{JS;Df1(G!(N04iB#e0Vs46r zKBU0m=<@MMi8iLX6^UpIT~gE53aa}2((xH_uqa~jwxXKn#@B@2aZs^~uyB`vL`CF0 zHjFL0gU}F6@bqz$xINYXiykO8>3;n%=F*D0LbcRhDsnL@OX2TQ?r~w`<=uKH#wK8& zCvDgXOdhL}c+Ep=v!s`T4=U~Id2mwUOj$TMCwf2DpTty`+NYDnvwO-rbCsepFc@r_ ztlgT4o#&^0g4|LtN|Ucl8;9-&+par!gKwX-mcC;+iKSkV{K-X zxoAM5A`;xoBGHmC7U7xKkEr@9o=6vnL{M7sc&HVV zQD+#N+Ug%J7Wd%#L9r8$vkr2I>jp$X^xcZ^+O0w`?Z;{#e9%tYFVQo-j@Q|pYTx0M zy}mSiQqfHx$%p_nMoZphotmzt`F$z0FZL&dE|#Ub7^ns;ABVQi(0HMhjjdolL48`G z2{Cwl`rL=zlX;hZv%@VZKepI%V^|CID|yyOm#V&X3SgI_021#D&y%8#jt-|hvtuYl zyT={2z-VnFf6cp9LqOLE_xnM2$xV`V3uZA&kfpd>eT>5GWh`8_^;5{a3#~r5)afeY zhPXhwFckKg`qIrg;yZyq&G9}s$ne&VES|9ezha$uUFgg6`$_-1%Q`tqPEH(nT5~Ez zLw07Ux4}3YrhkJsL$Cp7X9rsjW*fiaA5uYhWJ=}PT6vwpXNs9dID=me%f&!7LX@(wqvZ@x6BL&Oc^92Bo;{3?C z^()fg-a?%tLP0-4_StVu5cvz@g)RT?vkg@RTV%zmG7E8$6pWwRh7%qQiuXs>b6MB+j%zOQRvN!GK02fxb5_Ss~ zmR(7#5k9V(npy}RxJqvX#Ap%j_2bXwq(~w}a8DdBEh}sc$jRQZ2K^q%s^18xC#DH} z{%cgQXBAwS+0U9D_BBn>fV$|2Ep!m^QMumafU+w3k5*M`d?h@~5gIn`!R5cy1i)mY z_r=iAP^Z(cbdq0?@D?jdLO~9RM60X z8~Ad#mZTsC%SrJ}rQ47G#nxWJRZY0P$TSkSXQcB<0M~lhDcNK5XV6hdej^n#sa6Wk?;$#`d;5J%g}BV|USz!(j8m{bzo6u8T|7B==d(Yz;kB z$5VR3c%OG%+P%Mrzb!QRdmFabqPP&i0)|ZC5PtXBKw z8L%kcaF}S`0!hkyX$A6K9J?3XL5OV`6_UpX-pwA9Ef54U+f(7-r%WVOgGS-_emQX(Md$?S~Dbb|XJzj%32v=go7!rw(y{44Dep z_e_^2mhj|=xd}MlaJ-?@R3UXF=aCevaTvmj!J9?YQHie%fbo6XZ&@0Sdi;Q-uf`&h ze;SW<+Kw#0IMRno+KB_&Bhvj#i|V;q+$0Jw#?7r|xFuQJr89G&Ml{v~iHk>2h%RB^YXiYp#0ulEZU}D`6juds03*ojggA!X#2&KUBCuEY4~f1 zgQoz1p9?SD(;{;+{kIZGfM$ z@gfKwyK51;TW-_v2Fd<6W&JKUX^Z7c73yf!cH)NYf8=-v3(jprc&T~wwKJrvi}ORrr<8T=HBYW0GmGRo zz=v}LdHXL#akk<0M_W1~_V1xmet9Z?l-F`F+o!y@ZA41^pn>fLoWDV6n8r6V)aKOTU^(Nq)XB`wF0SS7m?~(#t zy4gHx(=7^Y>n1VJKAAQ}GGA7e-*H*$BI1u-&teNf73y*j>MVGqRY&tTAb}sG|h}f#jeOiMqfe+H1)y#WqV_3#o4F9 z6e3F{C2NOSCdf|CDy*!kWsA8>4oQ$Rp1R#(l3@!5vYej|w7YsqYF?0hh55tOf225H zjee&et`v24heoG4jxbbVh>&PF*mZeldx(jR57O2)9QH@l_q#Tr>*la|J$_Wd&;uRf zca6BhogZ>B>Kq&&`)9~F`>qB2t~Y;{tUcUZK$1QF!O@j@w;I3ae~ioY|Gxj>Y}>Y76DHd=lkEwUZCg{5-DJDT zXEr9=QG2Hh z3%J{c=E&Odky&hbR2C5_$l6F#4%$WV2@X*COEihCWEm4qC_Tlh{wgxLMRp7Q?wr>22usWVg~boR6w z`?eBx(Kojrulc@L#pWlXVzS!(z1-w>stKJ2URQ zZ(c}fd`^3C8?Up5DsE-qfni#ek$zaplTYv@>35`B0^96|jr@AUpCWfJu}NBZYJCWgqaCR8qrHotaIGhCg&TSEp`d6uCBw(^dTeqK>?B5 z!ClJg#@K({vcxjw0<+@ePGp5c`$9(^X!iNu>R-h>uA*~422V|ows)fB7e6V(dSW9C zQzOjbDc_~B`WZr-i;DTVl`f#0r(hnK)kUnU+{$T*-dz~2IVCynQ zNg>v9P?-|eKnXNYt@||VzUiy!uzhuV+hqHlGrj{RZ6b?d^}IAb9<^EUk;}mUyhWv? zMeM`^b%+e#h1iQPFpMT{pBMjDD1riXV0QL(7%kZ#6A&P))ncFwrtew?Y1 z-&T9q8nS5<66(}OP-{ZJL17%STtG#kz-PV}+7tQLAZwcRTI=mLKW!xI`j}P6H^{ln zQ8r^IC?NQL@tE%gnwIScycWmzw>uDzSrRctt!%A$;~bRDuu?e3BbAvc-F3zJqX`Ax zFv{i6)OSR{oI6vkY1(z~VH>H>Ve}UlvzrOfgYt{|isbPi!+A9H#e{y(h%sQwVZd1U>!w)`3^#_D`_sq$AVNgTAb>0Mx`wZYW0 znK8Oobcc01H4vkQetCiTsdAZn)u*P!+2;|2eE;$RsnZdwssi!m`SeKE5B;#HQH?a6 zc=+B^`HGw1|B?ZCaAUMmR!l^xOsVehD0>4`S?cvn+p#kj7Z)7*S};jM%^uYEA7Bk5 z43q6}hP=o;)tM4aS7CAO_SV*7j>aTs`TkKHk#I0}uyLvj`x!;-kauB`MwPL98Agz9 zIW@p*DiB>VAkv5lBu0f1xcnU3KM0{6HcLTC38YpUyUqf$eqH1+?cYvC^ZOd8CWVN< z!7K`CCpFq%0rd{yHbc#ZwYM5XLSfH7jzWXyA{ABNa88nv-v{N3X&!(LCi*$~wTS{F z`>)a-5wlc;GZl~R64@^^xAUId72{3*R^Px>M*hOu$oDq z0AkHeX<2WSztk*&;SIQrFvxPphzV+5=oXsXy(KU^z|`~E*@YVqc^2UJFF7kX<9T0_Wg=!9dr6%x#1`%;0GN=SnjnxoZSZ$IQyGUAZGNLf|)1ncQpHBA_{$VrdAl`Mk=jDLzR^;-JBR1jEs3h3Xkia1A; z;7*?RA9=qQE{d+M93^HtM}@HBw~(JnhIMa2N%4TEpiiC5F@;uy&*}o`w(X47k$-I} zpR-;cc(MYR(H{P0+N*5pMZiaXcDO%!iqX{2ClWZpIa%dN4`K&Nl@HgoAUYGK@G+dh7UyJqPoa*Wns2{g8&*f8F@u1 zac9c%V;Kb$3U<$gLBCul=lc9vuGs?(IS3b8s+(s_5{|}IAh38za3-Cx?BRMrczb@) z57yoNUUVb2X6L_89e*Iec+xclrN$vZ=z$Spm&~}qfI1}n?;3Kqe{R0uT3^c>7Xj3GI0QJ<{N$RKkMsxM}W2Qy^f*1Av%HCtFP(8F$^M+I_ia1G^W# zcqcP4bP48iE=gxXQy=Nm4d2>|#5y(asVnR*im6eHGRtL|LtbukWdr~kD%$DdiKmY1 z((Mj@u;!uZbeLyd=&A6F>#}blz&H5`-AK&y&vpYFFN!*S31x&2m6#iQgEAm2;VD9C zE!+Tu?`d})lP96s(0q2s*79-)m?VV!Kz$JTd=bIi*Kb>U7E*gu(jL4JP5U}~W?xKJ zPE=&}t<3<}gT3`AbGNX0Z-dZl4k1??_zZ52OrTgz(DMu3u<1*_-7(oQpTKDAZxXFh zH@7=S@tyk=GS^k~ljRm%=fC6N-S7BljE0U$P+E6T6$70%LzYWj!nm1>ig;h3Frx>) zU>ZdIk>mX1gne;=(&NjxgNDml3tdnY{dZDnj@zxr-~>vO*Zu8xbHH}|e;-c=h)y^M z@Jd{iw;d%kqohPab>7`w@%NDYM&vssnx(N*Stw4C9lEgU5inE8;TqRy=9k zHz};#ihhqrv;2+W>_b}r_(W+kUS3?Oa_($C^MzXe7Bb)=w{A!*>I%@VzyNjFEKe91 z8IjQ_=ZqKjg-IsbbNzw8!2;C}qo}gr7;{G&zRS)GS8CDQgEefK&n(kiHGHl+Kx0@T z7$KRDc~@FvD}A^LH@ez@|>CW5ePIuck-0_*6PWzqP+Bu{?dpmdT&!CJLUU9Us|XhV=l^3J`D7^Q51& zsg-;(!@WrrdLmy3#N76Q_~!3r;<7*|tJfb_%3J}7qjcs@;zCTog{9diaXYeB5`(#5 z^V3>J;x1uq|Fzlu@;*dU4~`-FI58&&Ft&ub3(ud60G$pS5=4b&4Nfyb92^bx<(Mc* zP{pUFQu{>X7Y|K^xCVpp<^x^MOPi^DJ*=pyEYXiAv zfXgT}$9^Ces9Zj(t0g{gK#h8UZ=64~-{`>qroHIEb518AnhgDyiY8H>QH$G(AZFSO zf`r8&LK{!%-?@nyDN7LTORX8Ges!nrdj-*uPB)NrXH?u{ZrBoFjJm!_iTBmbzzDN{ zw!~f11I632i=5Vo5{ijpX|s$bttDIlx4SZn7myP6N`wahg^rvzBR>JJuJDS&Pfk*3 zt*G-fRO@f5w*85`pdr>~{0+iqKI1 zo}XjhWiqEOGDWX8Hg_lZyW9V?qhC6hu=a?9HPpk?8B>ezYh#(s?kDneW{d+u$v%ZZn>2^$ zY-LF5T&Q2gzo8Ym>qul#R|AHqsHj#ji%?WU7aMpaO{g(O>)vh7dAP`o_&g==Mz=3J z!~ZBM^{}{&!Yd;U0d?@TnUyZ@53K7?m9$u_aY@P7Fe@gl@5OT5{Te>p|LdBo(6<_J zo81PbWc{`Cw|>fN<{w7;a2;r>!KA^Ppu+PrG_ig8;TS9iRL_2J*#uzbg$PaGCeqG? zaVw2T?M+hy1}6K=c#sJp#;%Me>d$N4_W=2WM+&HP01AnNt7{X%qm60bZ{~ILf+^E@ z;kcMzjHSa&B1L7y_LgTu%9snT(MFJwhLz$a2x@thLtlQgYSaDzs6R=u}$Lg**KOc2ir52mb=4eKEPNL&G}%m(;NlUw%!j&VWji! zMo4d;GI>(3*qSaRNz38rS@Mk=pBF6+oYT{Rq)o&E1&U%p&3^5!Y@d4f+Bu{tX)~dD zTGzbMA24)=XbJ66p}@oJz9Y)dXxzZ8+y7@-HfW0n09KHH?duo3Y2kvxFuFN7I2;}> zjLDCWulC1Et;>VopY|PY2$%4K%@w62CAY^j{eLqPJ&1k+|F0oFV*wuMb^SNfr}gvh zcRwn_wuoT-FRM0sv?u*;)4Z9XCpBQG_We~6zq}OPhJ`c~Ty+MwKbP|Yd5|2%>9MyU zP61x@iHU_~jov9C2Q5Bp9R{+3wguarOh6}U#5B6g=|ENrFlVi6BcOQlrI(*72?{vG z0qM99YjAcip}QLhL?&^kCk#RF4x}1j*h0Yi1CMITJcPXLZ>i9JWpWljaK+w7NIURcfgAZh`a4Zc8jksc#95$UE#pY`{zt3UcH+n&7U z5}5NAj8L{Bf-43xVDjJE0S#`P zch7rt^F?!9sWz05Cnx5xGsX~p3^9auklf>lw;4&SL{lF8yX zWP)Z0C6>xo0rad>a)B7h z7pFKxjDA)cq}cwyXJ<-x@)bbRuy?83qS9LX-$(HFve`1n{ZRu6Y0l`S{iRyt;IB4m zfHNk(xm|8xVE#s+BfviuMJ5?&gU&t&0uRu3v|Vm~63xjokHw{ifeK8MuK+@Ye4A!8 zmD|HahUV7($xLUA@HUhbBe(|QL;U-_G=McfBkuyDKe5<$kSMRUON!OX@~tKUlfrR~ zF^AKk)LPqG)wdMRZ4ehnvQ>CyyDKQLO@?6mCcd&nd|ij_l2itk=_%ke!j*}|=Rr2Z z!Y=re$&uCS*3&vF;-L-maPV13hCp@$Fx~9-yE~DxkrbMba+MJ2xj5bPKAQqQmYY9; zLf6)!K$xX?&fU^ayPws~2WQ$+5sEPG!Teut4v}dAi5BvZP5QLuK@TSl1WMonWFNsb zB0Xv{@Bax^finsLh(?8vCo4TmIIIRtbEfpdnh>f#g<{NU`2OuV+&LBW_F3$au!8sw z@IWb%x=YJF@SPL-G4b)zRtjWMEI8!9x40~Zmvl{u;ri7h*Lt_-Vs5_&hpM zQB#vsQeytZUWS8@ZK*uE^|(C(m!T#W3tI%15|?cNZw-^y{){xKsFM>%8s6ZCni-=9 zVckTKZAT-K7L2y_ceA2U4wjnYTYC5~a@VB3t6IGd;xx-qKwQ5+@I2Ou1w-Bc73CEc3owbtxMx<_pK7F)vp{Y%?$K2^oWR%CDF~5QqDa7A~eT(up{qc1}raNKQ+>K$d6a(~7cWm7w>l|E6c)xo&t7SLvZi_HQN^l&=? z?vm0%uI`BMU%0(q3b&A;9yUb7(x36b@{BDttO0mGauj*WhHObH%HP7S58REMhimA- z^4#aVbZv;QoRla9*TBbl97}mj2a#%pH$F5|9ykS5#r~k-<4yC*MS#yd%x*?Gy@88v zup1KiLVb_fm?>aus?w2~)nf6bibUWE5gsY+$p@zJw4Ki*!|!19-f|b*5d%V;X5xlO;DZp)9ds7E--Ko`dBIeKR{O=)6mz zRlojCuWWdrBSQAV^|Ds>a?Gi>L?K-)$+IpQJ|%v9TpiHx%gldXcx?A7wf;^AP~={% zZEbftLx%sEo<4quX(CvIT8>reWl_{vEn+Ta30Z3+~WM{7<&G3{j zti8V`I$3^32H5Z{bNHt%Gt7ED5c*8%td)c8m>?8=zDW6V(*B3}hAm8So{8jt-@lDM zY3vcVK{c(f#x_wfWTS)pw;kbY72eC@BvKed_s zKargV-T{1^MtODkv=KlmF08fVG_6+Tqga7%Vi&4XpW;O33q)6P^q`dQ3xm2$vV86- zhz7oUyv)#Uoo}x87M%k=swK*v@1OnO#&TyAk66r#%_*B+vG_A1AlHv0iNNmpV3`c- z2u^l%#N1#z0PVNukZM}}5$^V)k5AdZ!}F-(oF<4^D7Hn}eh8}Dv2{r+AkpBu z;y}BqrUrHw#8IcMuEKu)^rT1h>4h8#=96T#0Q?x?QFKC%UzYVxtAyn#t$xP&JVEInbJ8HNHYic6Y3Qv#Oq3*7)VCgMKlJ~EP8ZBA2rXWOw>=B^ z(k~2FO<3AIt_*|3(IQTM)a4BXTila&!{V#*5@}B=L|oPwy>5RgC+#_CSkr<45fSw_ z>!y({hS5N#A+StRTW|)@sd^~^?#u#k+pr5B+t~UsJ(Lcif!plh5_VMta~kNpbY^0J zSa_CG9tH;W4GiMOQfFt@dJSB;V>KX(3XpJFNhH49IF;j?NR9sNG*j1ummY%4;?K5U zZhTHulvWY>jg~$}JJMEnLHoZ4u`msQlXBjjZU$SJ!_%n$NcA_0Ke2i;zCm@8l#~^a zla~)YN$P4Uvt))F(uAhx^V@;LVbXGthUC+hS_0DUp(=KOctx@nGSYJXs*6lg0HX8& zRWDKWxvo2>X-ewy-<~*OC=?OhbM*i3@V44903((=_0avXWL|MGO!wQSp}46E;uz4F zsu)fO)$@*SWA4ubYR;mckfF+iF&*Jgl`p02pGGl(_=JauC^Y}Y!S4!y{WjEbx9y0? zXL0x96%ZKR-QCB52p==|ANO&XtpiX12PfC|U3s@YVG0 zp^Zg70U^KcP_Rj{39*1Xsd(mmU9?ffx2C@z77gb!K7TgC=E<9Ixx5;r@@R~WEnR?u zMRCdpC9GeFkj$klyl>|#M|m}Jc=Rk@<)Mi9fjd4%LoO1Nx;jg9vwW7>xTSY z={{I8mBM}yo@BoIGSx@|C?OsVRm(P)6DOaZo2J=Si8-!bK|J#S>_5ZpE;Exd#a3?Nl zeSU6nIH@FK8|xae>VhZ>7y7kJi;Xr3cndALC;Qd_wVzp_Fnz$0~n*hX9f&9)Lh*!Ac!WV1LHeDI6LH;ITM| z#2yZ|aV9^Qf6CxSkCBu?T}Xqwd)Ng=BqI^}LqVTEh{{({Rc1CGC`mM|g)7wHX#)0~ zlUW;ql`-oobT;IvNs=d5mRsBC(>6@jt^!ftu2|Xfj`=z@u5dqEdtebbD~rrArI(Jo z7c7du(4ru~i}IV^5gX>pjbgIsis_$Rmh~cHHN7?$+2O_(m;l9QS}H0n-GNj`2p-A2 zvNCw)TdzX}wFfWti(=8an)f)9a##HoB-cWrmy)>fVDVJR0&{WqZMQ`fi-zU;!)n2L z|KpSRk9)LLsb{)6PYr(I;8)D4+>?G|czInq2W`6_Y0dd3o}9Y#9sr?OSXwGmt7?hw zKJNA7w?HhZu%E5SR^f-+SXJ?(ZhX4g_zhwQw_VS0Eq$#gV7QkW_uvC3CRUY^?Tq44 z=Y-g4^BG=_<<5*~S|EsX?~%D*uU6BfJEV}2*^dAHiZ}c`}Zld0BeD zOr873er)I!jfDP+xLtuNduF{iF!GcHd`Gj0fEcn3BDG5iFz*%M0in*~4pM%1dViK5 z+p5lC6@TLvdezN{w#@ZisBc{~HT3Q`V0=XUIE-DWem9CnA8)Uo`>zYUV$uWsoPQ8O zW_AKj@_3A~=27*P^Xfx!AQ@-Z>|{{{NPRdJWRMRAXi~+lpANyrSTqEHo|-}i#1>coK_5TwE2=XJn9#jSMb>fqO+J`3lJmvl1_@NRHM^dJtci z?wL_vS<;KzA@^TM@gE$#(ZeBLDabmTb2ve_q#dRd@f{0PUVqQ-?Y~6yViC@KXNLa5 z@qWkB;jAC?Ze404^zJKuq%?>3zS34Fl7lcb5Yx_LWW}NiZDL>Lm#tUjwt5>oMJNHk z7rr<@XRtVKf)}%Gnh>uC_T3NW_vk2}?tZAdG!T|Y4KXuwnU(~?pVEZre@Fd<=^&T{#^ ziLM%|x8z18nV5B=mX?-`yAU^HfMVl}D^SUq7J_yNR638GvHbPVbQD-b-M2fR?I&R- zrV}Oi+v*O|_5;uRb@xPPdi?qY4W`&wkeL(WWrBb>5IJ4i1x{P!Ohv@~ z81;zOpcre;u;0DZlGee*iP40~Md@x<-G%BaY*ksVCO-!p;~vtOHyX&$Cd}T zp#Ea*;4m~kLmwXlVFyC@oCkO}Wb0dU|jmt*sY*dAm52=;O$n!n7WgFzvyleMdD9ep$$VTYcRZ zBsWQG7as0Sd)DnQL_`dm`#=IlY&Z^-9)%$y9q$i*w>?iNik5Tt`u%`B#UmSFKHz^n z@jc&dF+-6(ILgpAz$6mR87FjiEfE8twm^O)=Y$^rNaXOi%k(>4;ogp)KU}bGawzmN z2TYJ@Lf*|r-v*z`OGZw6d&O$o)+_7A;{aw`K1z9Qz^{@*cO=%53uyIr*ADZq{mw9Yu zq)V(SiKMQf1TK(Z!Odi6an-&P`YX!o8Q=U)_9Tn9tS%Nv3rEw5WB!!BH6$!-fa)+5 ze%WYYP+!Xf|MZmFEE_Mz^ZKv$Y6$DcC!LTo!5}*p$8|NKvs#@Ctdska{C=AdIQ7UL z)rNOZ8N_N}DyLkUygIE3@mcjXXZ2nFZucnlO!;={gJw)%HNSrxQ;aXDA(4~>X&K|v2GL80{c`T6OU}SV#W1#0HlGa>K4?X#)NvtaHo%+gRX{H6jB(cz8Axcn$ZaO?qjk*C3=kMyyTOVwc zj4CmVf;zJ+gs!;vhc#1BV?o*0*(!E2Bojhh3|#|zU!5yK4T2b-!fz(9VK2xcxjAky4gLCyawYNZT0qx z0B%`K040&Q&^9(P)F<-}&dk>4CyivBgF+vB;y-`>T)bRQAueC=LWSCo!>@sXr-*t2 zyx>Y*chc9Y_CTTV4Hv+mW#r$1tgy8+5dFpP7yQ`H4lr5vI(yxU`6sj4LW;|x2Q)d# zY%iaH2XzVQV*)i^R-YZm0kT;x*tZh=68^$jB-6?Z0|XtZm@!OitcL}_e|B(l3M=c2 z2;*c(JCe9Sy^&fQrL*G7I179l@E|AqgG>Er&1HBTuptVGTi5PT*_Gs?gE2IkWWzpU z2Seg@n15#aza9l(k>&$F_g1ecBg=+s2d6%m(i6HgGQsNHhp&c)sCPq^4m3+vG)oQ= zY@r&V!bI;#n%5k8wY(zM=kb6DFd0<2XsUc^^KF_+Fu4^^C)faO3b+SMm-Q?svp_#9 zsLh;R@Y&fLykavuTT{H!PN-JdEGM~iNN4SZfU4dBqFFNFwcRq%f(g;1bz*%z8@HL1 zg{f=3l%nUPn?N@@k&P1I)A3q(p6s3n4fDCrCF5)XJ7PxQdF~Zu^qjk$-EYwU6sa$= z34@12BRku6_Wnf$$fDE9j*2vJ{v?NUlzhk-NlQRuwY8i+GFDM-UV@;EQYs-Ydm!L{ zf3fTUlC0hJmyWs7-c07ujCK~cgAEG{o4~x?DncCBtoh&kwe7wzQvaSe0k2c5%OeJh zWmd3>4y@p-;P&x8qFpe-s1&wdhyL_TrI>d;f=LnP-`7R13MwreE%9%N)=G=LmVrcF zSR;S8Me~n*b%km^LqShYt^Lq4j?m7E~!~FPFg{!egp{Z}l zH8?nl(uIMW1^bsVU}W*D#y>vSjVCK%+asdKks7Y=*~7r4JNgnd4;&kMHnPd^`iOIO zeQzZF|AjZhSM8MaLy`JCxeod_gBzn=OsH+TxG`S-782IF8u8Ui4?w^ z92{FjT-EYW>PMGvw7EM#sHAw;qReuDc@VQf3kHL5x(a4xTPM20%hPL}GW|6pCZ^2G z3Wp7|!R7;m4;IC*9Km?I)n?a4NIaRB`j{z-pwP^S?>1+$RSGwz^?rOD|s|NcGhrk0%e0WTd4+F|CvQrfdhH}Q4 zGut2@l`s@kfjNeZY2yWo>}udAUt&Wml${UOK6uBeK>%9?Lp%x+6cg7z_P#_eQ7q`T z?{BY#E_?3oT0`-qL&xvWgujSCA${|tr-?kN9iM`|_URC@#$mBM*lH5;c5G_2osK)A zZ(o~K>bFaRE$Y|&*pQW>?&0i_I!ecfe${DpAfL<;icb2?T{>ZhB~iVVhh{a(U8-mk z#;^+adyp8E6r_^731wI>*tnn|g{c>7fHuDh%3}+Z%ha>STFQ0)1xK!gzI3q#^JeOk zJ)Y)k!I2(PO4YE8#LLS&>X8tmjpMV`EF0h(Rz?KLPmiHVbygnDHzyONOe*;{LOI>; z3G&<~7^a{RaX+z1F4jkX6FL+xoUu3w@%F$X(p#U=&<-0+p%H~LO*h)bsggJwvA=x8 zH8cWX!3kCBBeZV~=`kDvGzEwGY*!tE@oH!K*bi3O=lKNSt2?oOy1Qk>pM0s^oFTIygvgTaJ7l~3 z`il#!4;8$IX`xl%A(%7b|B8uoiww-1!D$O&t4$q$2e-&Zw=3IMadRQ7B7@EKfBTf< z`8WL-c6W*|@LVRZ=3)PU88;4#Xp14j$ndl%!%GnJMBopx2d_nV*Y2#cv7u&kY-AF< zzYmjyOME(D?jO8zlJNGXEXpSdmmj^H`Nx?MIUoZEf95WdLannfhvewduRrE?x=7#k zpIbVRkPLWL69x?McCtg9SOh%g<94bBbYOa3MBf!yIe(I%@#f0X6kGcn1AUEKkWMkd z;`bvignNcfMXo8$&*k*AoyPa4qV<=on!>|2v^G(8@_)HyX|;k7Zdgq_!W)FK163jo zP)W3kARxclzB1V7SXDAFpmH5Rp@n6g=pD|(sG)*+!=jrE4H;eT49NUy^aI8bx@0|3 zWx|i08a^-cC}w95c<`AyA(mQm0s8h84AS_^4{1Wad*5HbH6k{_I#yRHw9*DQDEOqEvZ%S* z4L$ax)3p=Mm)i)LlhXqv#12DBS7y6(#bDy~K{4o5BpM`1j`X>ZbBl0hlxe7l!)I^` zTr6AWk_!qF0RvBBBb%-?OuTfcwEf)u9A~pTheHE;EQCV=WCxda7L(g``SpT;x>+Pk zL*6;!;Q)h@VA~(UdIv*yz^;iOMLJ%NGsVn5nfqe_waeR$buD1Y5Xd%sCP@KQn=r+X zveXHTB}iLlZDok2{IY3xjFpVy^7xTD8g2hz@fGIeEiduNlbd0sj-LU;Z*S|y(E{bO zB7gj*tp`|~ZxA^vZmNi4;6TzYee@T%b+yjtfEW1xMk86_=$R|1{7Lxs$Zfk5bfeIX?!3yM&a{)_i%>F1N*Av1 z`BRAzEJsIIsv6^9ofD~#w;#J!9iZB?a(`XcRF#po0WM1dwS^LcQuY%bfh#q1ti;E{ z0%d7x2P4p!vSGkwvB-&O0I(Z^R$hw?g>x#+u(%G5tj8CqACy_Y!FbqMy23D9?Vy%? zk=qa0tzestxRM&ck{Y<@tET5#ME4Gm!qfAEEvi0+dGvlFM?*H3BNFsL(`~l>ZD+a# zOR^3Hct)5C4=B$ zY}V@U7l=xkw59dLwq`e<>>mh;{$Q-1{wRh5Q+VLO(j87vLJ=BSdaiQ-ikiUCD}wjFYM18jAt#<9fo&8B2=RsvEGa1g`14Aq zf0k-Oji|31qTauu4D-j<^p0AiNwJD1kS!Ya`4uvT(agw4EC>rpw_F)hoEBaKLLE5f z_Td)Q5@~(uWbcYCQAZo5rm{uC?XpG(5R5IBvW?t@y$Sd%c7Y!ah+vvYKcHCa`COuj zP)Sp(;kK;XJj*DJhbMW_uAWJxhgw!=B-|n@wI$em)rG!LMSUWeL!0SN59+1J*GTK` zekUe9cJ~#h@#}p2n^~$jGPJV=C79AmbXNTxQt1d!hAsXUA+Y{8au=3fA`t#=5zLhZ zhCmY(dygpj?cdigni}x8pKT>oEE6+;Vxsvw;cJ|MgiI|lD$!##DSsC!E!!wO(U>r# zdOLW9FD^*_=<}AHJP>#9hfMxc(73x?Fa2JWi+Wr0Ur{$B% z7F~P0hMG16rFOr7BFZl?l6KMa@EEGn--9qB>bh}rjW!;=Hz5<2W4HUghMRs6ec0Vh zLfik{0-g%;^1-Fzk)dYZOD%QYxl2o0W>!sfMbA=Ipa9Q&@>Ut}{%_dWL12S_b<{>j z>t3(Stw!|CZhh}BJzKiS9)TCQ5wMcqYrhVOD#%toH~>2<3Fy%;NdTdNZ7k&Lm}b7Q zGIIIUPtgUM@Ze}-jK%npq3WA&^ooPnDF@=Hqb3&xNW=*0>eKzj;kg2X$`OHUk|^%2 z1xyVXcU2sC`s1bY_F`F<6o4r?3JA+K*fr{vmVq+|Pei7)FH8k%5L|m5FyL!`GBUq` zyje~tENzNm;uv{Tkqgav^#OK+f)6y_d7FvfF|X939JhMF*ZkjK3vm%hmrE(w5$yDxzr^RB zCfG7#WJcsbq?;a??~hmVwb1A>^H)H-@2Dm$$onoDAsM`=5Pgk06s(S8to)>XLN9mr zR#|In4uEKCXP{I!p8~_3Sa#4N4ePc11>XyCrSy2{$HJHy7$CeG7l28`*6DuCT{A&F zqr3?0{5_C~Uf*1XW`;pPVC_o$Us+z51opg96H^)RCofO}(j!sXI$)Np*^+DOY=1X) zZih(geE~l^>!Fk{Fi|1^x|OP5SppQw@TCNs!oymj`^+;8RzzwsSaE>67BL}7HvsbF z^aQpmF8|rVwF2Wz4R+XSQ^Q;U2-^FNY$9Cm12TfTJKv1&gY6Cz{TcxjpCLtm7mc24 zU413lq#3}vZI|ow-FA6{@Rw;nh$Q5?IJ9jZpgS;z3WWpi+7z`7%d_t1=IC-hgW0*223O)Xxo)Hq;LTN zAh!DYlZmu5z@FOa9YD0Iwvxcoc9GSS92U?9N}R}aFc>Uz z?;S{KMRx2I(2&>&2cuWK8q4|zJBC8gnelxgL$$C>+D4vHe`#QMCiFe51?2@bj`$M_ z4Fg}&6u14yQo$i(P$;ujTtUMS=)c9bKjJSY2IrDYWFlt>bxNqgI`Gbl&_una+Z|_) zrFZ^OFy*)lPJhu}#0SZ(?*#j+wWLL1vxlkiaL`YHA*FxZLu4GlI-N_=fUwgMto7MW z0M^O$SmV>l=RQ3j{O(-(>JjLpJ^*lTF`!tPB}MC_4GTdLk!G);`dDq#oyE@><7p4K zy~=(uR`tq6-%j^Nk{+()GpzD3A|Sb|G=g!;pHtGc#Rh8L1*a_>oa+0IYr03H^-yIZ z05g^pDYymZ>o8?JT>9-dZx++hn=-h2B0V@uS8vsq+rwaxbU*wOS|C6@*^%>NaEdEZ zTLvR-eQ=%kZbAZa_aGX8{7OfEk>UTuaWg?TOP>F%g415#1#MDD+~M14>_pD{B^2M5 zBb!%gOA3zbM^nyBX;6h81aB%Kqv+Q5UohGX*gyDRjE1&P2jLHI&7!T(LLlc)2`YcL zx0@2o=UTX3#WXD_9MlzBnI8wO&+AG9Y`po`^Iai{ngBz zge#nb$H+BGBeb3_`zmgz3iZtjD@)^h$WD^Gk?r>Yo1I3M%V^{*TQ-xFn!d+gy=1Ot zvJZU8IN^u_1_lP3{W1j=F@Io{rTcc#z-Dl6^g(UgiSjD`Kfl@)(a@x--)S5C#QD{q zgC2_PGO>o4xPf@Rxj8oq3MT^WQn&h13Xg=xlj1=HI;{O< z^1qO%@f(;{8mNQ+!Iw(OdrlIxic6CbHlv4SMaSRLdItXUidjLNcFVp}M-OTgT-Y&qJ6t5N5wGn@ZSYCp9&jv(Zx(mb+leWaG3|PpM7Cr@ zl$M~Xp+RFtt5tB#Y_H-`O_cppZJgYjRbw%a7|J!gQJ!i=$v8*QT~+iU6I@?*P@U;=RDYdMO7Bb3XNQ)GUwiA;iC6(6Ib zZt$xr?EtIF#=|H--sktjn4^9)6D|np>Ofnmu$+qa}Tt|Z&b+=Ta49DQB%iMG$;lt-OSgfbH z|4t8rar1xmPB>0yTkfe0@%;b$6#^5LjQ|Ly7_Iy1GYqzLgNh3UcD3k+3o~74aTVmz z3eSW0bDq*HjI$`G4lz;LNdO~N+?Tk3GY~QX@fRjA8JnQgacs;%<$$!xM(*2FIJnq6 zo8WC|yjUN#j&S!Vu!aMl?NHncOAaEuYB&q@?zz8UqAy+m)YOA)&|T2@g#U985eot% zBNStw41+w!m$CBlu%!v)2B7&%{EjPrn_F`L)rAPIq3*L87+m@#lI<20`$c&+q;bjs z0As<1p^27xZtu!rm`4$fe4*$!{0vsD4p(4Kgo!6YH z{vP)E6ap!Uc4i_foXmp#86h!=2&4Y$*w!X!W(Hm$1k81l2_b z15GwVbV;TUDV)sA+*WqXeVrA!2nFR;ud7?H|MPopK@Bp1y9xDIqG`JcR=8)SXNTT} zp|rvB_6$v?OH?|9FnfT7E0IWQcyh0sLh{=Y&SfWGDNW=c(2YrAjK1CBRJ&^}NKd{Y zUL^fi6`5e-KxhU0nHu=lj+?-X3>^e~Lw9|`SAy3Mu!>5yGb`vH{+0j4B2x(c18b#o zr6Qb`f|3|s;^5EKW8uS2%XS3E1}bfpMcfRLbhtYjjl0X?Bb6LDsC7R#0Fn6oLiPeU zVjqs0wF+IBC*JWR*SBp{f3*6=4X++{wO6{&LVuaWu-4ZA1GBi-2&QSFheH5M|Aqs2DtA7UZ4; zX?>^OES3#k%_)iHX%Ubui)61%TJK(eK1pZlS1D$F-_0_tt;jCyh^E18NG;L3`9t&~ zIED`bNLZO}%0weZlQz6yRynG+#M3#hBwnBPY=2PeH>EB=Vibuq=HdKpZzl+_&1y{s z5C9T5!ndQLA2npRv;Jgf`Lrr301)UW31Iqm8V+Y6%nB) zw;8*hj2I*3E@0@{SYibn02z1_w33--htK2HzGMKCFv&e5@(%3yH3^RVm>xwSC`64onol#C`>WxNpGlX!B*F7)AKMc?=qRTpbQ z#D(xXOJD^$NbBXv{hxtfJK)3Fy4db3<@ba*C!Oe_ygm^96xAm#d^_>_x<&3X%_&I$ zzj5-*#k0C&&JQl)1|2ruqG0_b=F{~85;v2@8!?~ik62%>3%{8HNjUgV=Q=!%y^1*( zB@vVK>cufSi0npBK&Dlg!-UsyRZCYQ*x%!0fxmvzLTRxRX5iz4N=qIU)p(wM0B!Qf z3NulcGCE?x>F{g8odTtv$xoHGF_{~N&_t=i$O6JS)s_kp#R;#{&i!OF6%<%j*wWpJ z>6Ek?zf#h2HrUcUk@#l9Wy^*zlmaqpN*0SBH|BF+YeX$oS^r8X^o{;x3}4uWf2Jhs zb!AST<)&vs#utKBS;z^OoWb?0l(MO7p!C$d2OmK8On4TFq9U>r${B*P4GlQlLhBma z5pY|56JL5>vCq%(CQKIH#`=e+IFYxvfAhLzpc<9anQ2^I6XIGj+{{;mPLRlto$ zOz@@lblmu*$OtzE1*f!A@A4f-DDcmM&I#S8(8ae0%a@BH+`()(n(B{o{m=QK@Cd>T zq|M?h5`bAZ3oVtllHu^YH(YL}6Yz%Ik01LrT!W`&mu$OlA9xHzef}?PSx-SE0tGNc z;!gciju*Pa)6vm!sHZ@k@V-XtB*XT*G*&77YW8ddz1(9!vZG|^@2p|e|Zm&e;DubCKl z{i8lEaC~QZJ0tbj0+$67aQ=(bFM`KD9EymHV*#hIwTgz8ZTtT*byi_ zEvTu0Uy*C$Y+{23k66h*nCc`V^o@Ll!sn$rEcpY;;&m{{E2lU~<#GXiBps*6XGV`j zR~=iI6iL@Qn6>-|1}!onVH`nc46Jp=drzV`&ed=u(;=`))aYcz@*&3aFN*;oyl4Uu zS7+B6AB-P*?y{m@z~(HAQh_rCo51&f?v1L~u**XDRm970bx>$7BCE6KR-g*$bBN}B?^#}`yG6Puwy-8)^Wi(0}>MCoB++5U&IrYHZ zpVGaJ9uuEF*E=g!)PMN+u#~Op6rZ?BxUg&9o7l(lzXLPR4sMV|kGi(Evm-S&R_=k# zTBk!*%dUJ0o(aNmjd<$B6V0A#X+xVIRC3rJ7&UFj8i3o?S% z)Ai=gSi=TK-+)!0&Zo1EnB`+SD>~|rgNL!lgn`bGg9MP5*+hRcIs~gd(DuUM(UM^x z-;MQ~vk^Wp2*IRO>x;(sOk(D2EQ5@9FjOr3SR@q)TRQA}`#Idj-P{^gKg1CTwQ7Lz z_y}u&wG3evlo9HFhrPGD6mb^;`4>EPj;xu8#($@mAGe*$&1?8n0M5cox-!0E6|89x zXqkX>i2=ROgP<%=DIsw4Fv{_`lE{T%m^}mGn1Pz(`caTV;7oIQCflz@+i$&7A7|k- z{LyhlFa)(Yeu}8OwP~IM+#XE=22!GHa5=?sFM2L~tif4MFc5IeaVn}TU~Q1vWGHdApvY?f(rxf7l6= zpxT=Q7S6c!7x7Z89j}EaE(FwEl%3q(>)~2CG7A8Gf8~XL%rYNW%X}wR{5~9tFy?LP>DAQ_<`G(u?arxh#9S^t` zzO?5d;c*fP!NbGHTg6uJ2`w#?uFYugT~Xw!MY2w|w>(|ChnG9Dl^>nj>Uhvv@E41! z8?kKuOjl23a2D$9vl{~+XAb3=MUZy}&^ zAN*I!U*xBmed@U`zEMy=I9jds%em46X)?F8|k>D0{za>}7^6=u%q|pvSzz?i{^Cu!&`233!qDN5Pd#Jg+kbt+~Taf(Uc5KQ-R{ ztTVnywBYcGF}-@|oyo1bea1I3OFCz>jXuwRtKGuaTho=j=9}A zF!D)-&N4lddh7NuME%|atXOMVxHZw+a3hFI zqv*s2t-aj6t?v-NnQQN)U0+}KP8Y{l+4>;8)6u4>Y4dr*tjpMptq)|lSS8@zMn=es zesc;obvxYyqN;&4(jEm)1(jO1pszkUy)UJ0-vu<5de8FEP}VXiJX{_;cS971)bc7= zt;)p^R1kv?9Ju__PH0Yb-eT9|;3q{8uR;J+e&`FRbR#kQV9| z%vl){WDRjKG@wfbN?zy5CD7eCg^9cpQL`6Vtf@PfbPmwv@qHD?#|ucXgLmo$6Z1Pr zh2Gd!@Smf8R66@XYGH~cizHradfmyU{4wt6kJK*73JmN`+cApsF?Q=`%TX&B=cP=p z7+-E7vrydi?P<|;K8?r^?mSgampGgy+UGM1Wt0$;O(WkV4 zl8)o1R2oh(R+t++$Nl7KF$dXJ+o4uNDVItXrZ4akla&`J$`nYyj`SdY% z{-bRB6X4kL3iGmZVpp4-O>sdRbg9jEGoSZ^KD$jvgoN>@+c?^gHYKp0_QD^U)(w8y zRZx0%j@|@*aJRr)8viCuDYO`D<#xl}MphZcVG0QCDFG4(6%BXjeV{uG%z2?wCOry? zhM@jmv5}f_HoL=Z&u!mH;R&NV$)A#2Wqg=w$pXuE5IsId$1l3K!9Dl`(CEnZ1^Qfz zbm|M>1Vj={UUTPn0}t^QC*zpB0Pbo4^AWNr73=zHRN&3eQ6~{bH#YEmA#D4Hmr#Q9 zrG0>cHVk)3s$%U97XOnWi&)H)WWYyT^vsP8mQCA`4hb9dhcr*TCT#{yD>y&e&vE54 zG)aJY8jp+q+p5kvGD3WAPp=@OQgwB8emmua6z!uAg(1j?_PZWCXzaPpwgzjReL`AY z9n;0T2^x``tAJ70wQpCUd50j~y90*D#fy3eba#HsRZ#>U$_7%@jC0IYOxic}QGpn-yW{kZ9!glzt-^Wk?i z1_>`Y7IWVq%H;H?i~Z6Ap}ht6`&h8u!yW%u9KM|bdtaYiMs6rzM|d{$wv8K03oJ{J z@vkQx{>(yD5fz@GOaT-#W2(Zf>Sm0*f_tads59}1ZGFPHy2h4$3KnGV&8@{sg?+A@ zKzfFDC8ztxNh8w@;{$^P*4@0Ch5#G517eG8!Vfs~yL2gjZ)fUnp4wuT+7T=>i_-PE zdTX`wPp*IK)=kcyfSpR>RgW#_=k*{=rCE)D${ovfToBvz@p>;hC;v4*`sqBRiYxBq zOWXPLT~<>LKE6m|lMq@`v_2{Z(b~hQ>jkkU5dj0WZ}_Erc}Dg9Zwtp(ZbIB}BoQ`h zur>EvY}pj%o!|8>akOD)e$Gdy{St)MeFw*kc2Cu2<1^&mU%#$z;X+8F}MpD{IesJMBM7BX+x%|;f+~4Ybj|y@;0J~fO3)eoi1>w-Lkj^ z`;`{Mb94$7ZP$4lKpamqn}XQ@MAE{n7HWm47R?Zcw%7 zf=)NRFU1#l!=jsSZ(WLQHa!Eg_6B(6T$+eN9TcLz1hOr}iL~U*q311(Gwf7&0M18o zSy?e7Y5Zb(3Tz`&n|+viKOcj{mqPf)rp#H#r+oR_8Ef`xz8ft9>);Id<|Hc>p=U|M zqe76s){rO_I%xe}Zs7zw6UdA5+Xv{s)(c*Xtsav+g7+OwZi;ff7#bP9o6>Klm}Nfn zckGmWDf?01;?&V2z{3-fGgGgr0ai0|#eQl*FMj$rg&os+AXU-670Di+M0|Q;;ScOY zP4|RPEg`)Gn_}PB?CRJ%w~h~gq2PH%Rwf+tkr!LRsjlTi^{_v#nKM2rJBZ4-^Bl5R zpwyAT09IQ+tVJf;r6G%Zsw^I8+BFG;mfFoiCx+Gvk@n}_9-)xbzWXg7(#b}oL=^c- zg(5{t$heY@s%H8MRfs4%&}F(kUckCLTCJ&rH`uU3kW3!f`nV%jyqq|1?;3Y6{DC%L z8QrTuJ-TyMrVOa}6b{wPy8vAUnry=}wgq-PKd>FsbJLblh*qM_e6-6)%C3!V1f{`U z90kIw(j&v92h$nXeS$-yE)W6|HLX+i-= zT!n8Rn@T?bzU8(+xz)87iaY?>&jUUI8VLgOuO)J#1wpe6;Uw*thX@9Qc~Bhb?5N{s z`q$Y$bC0(82L(pyo(X(;>3A zT+6`*t5@Ha+K^;fQ8z6};N<`r?F-}+g{(W^0|_~bhB6^15p9<9rwg&htfJ9^Zo*N0 zLv%t4kJ|+TphC4Dctt4i-Jt8{zD92Jjg2AaW?e5ONW!s#iph@aA`zyl=6*V~l=Fap zp5$NVgJPphHLo4*EE?r*8?ie04O|m8(l}gGiIbNp&jp7LIQ-AiffqSt`Z!v7Ds&qUiYCUS%0x{@}d! zU^CFWq4;ihso-|jOITHC_<>u;5=DR!LEv$SR@qSfsi3RO2m%u(R9ju(`>AIBV`HAn zo6EF`i8TC?qkQgTiM8a7OWo681w5;G7V1xG#jf zLh9O{NB?g%KF{t~UCr!-Xt5!K5#sPmco>uH((gpT4orfX#g3U0p6kKPp8JfBImT>K zFvZYK0ZynvDB{Nt)Fm64Wl=3FtK{yVcn$9wuvlY_a zrPP_`uUe2L<>mDY?2di!1hdbOfi>@|T#8fhqKu93aEV)Lgc><6!qd+@iYt!Zggqyo z6&{pxc4D9x>c{WF9qncJ&WI%(Hm-dj{jJ;SdELIi&DQ7o9rKmN(A<|q$&RPitgG{% zRH_bSbkV{$md@il;P9C;LS7<#IG%?S4z0Mqg(!=>5+wD~5Qx8rG0)p}X7_{Xxuc+G zFgwlOWa}$%b?X=o9)8O({>=4^#mi?80|^bSa@`$6WaAM;?q4~MA|Ot3rl73+1O1gJ zuDgkHG-dCnZkVqcWIRciB;?WWlHdZ?tiB3^_3oY?a?o*zz&)}24xU6fH;j$^=>7Jz z`lJ@T(WJiRW>TH|!NGyCHMT;WKKb>ETPR6kVQvElA$?gQQwm}NsTEJu<1Ao4eIMkE z_Ku+YZMOBnK8?`+&kJCYA0LAEiHUC_=yH%lU%wMq zW*08)jwYRVAWpyDDYL0ucHI=j3G8-_CeCIn?85z8#J1kqYgbZBkT2hNw(C#Xlm^^w z;c84r9>tfghB>mp088mNym>W5VC955iiAh^JB$wc%#K)naNZ}g8r*FhX&34GU)__H zPfp0g$m~S#OMyP@ANURS;?u*eWsuc+6hM)mT*t(T7ZgprQjuNV3=gLdlKH(J&4D#6 zv`Al=nNN)WczQ%k+K{wmRQeZ35UDxT#OW>6c%YNxkWb*>?9{o8)c%8nT`B zu*Yj1gjye7(6wsKm#=0!X0Rq?%J3n3+cm(R8crv(P$MAN`~6bHgS+OFj6KutP`$ijvo&Z9&rp=N4)RGrqDB^KD!8~2!C}k~0j#K-II5UhD z1%(Vjm;B67C~Y`cFC1Lag>a0AAQ0EPX1~c{FT|EFxA>OUs`Jb4=uWIVA95&7dw%g9 znLM#k^&>;Xk~SzdN?BM~7zG`jbr?0X@BI-ZYe4qi6}Z_q39bXM&ECVpgl&krAJvlHK^FHl5uiqvYtktNY^cZQ(j#|+d} z7PQK*5CBICg^|Gs&F$r8?qc?EaT9~ED6 zOsA-TG!vBs;OTRtStfax5d^fHrRll6C~ym9qDd#|G5ET)(Al>)$6F}ZLHpt0;q)`? zM;k}np%u|SVQrKyGE2#La)Z=)y=~VxWrKLep@@+K0Cj9>0^^GJKc_Ic6Ee`GYh&eU zV}724%o`TZu=aX;T2%@%!a1swHXf}qE-)`z^swB{)~mNzrWu|adLTav5}rW7f5#YUkZ0}=lr`b7+|&~hc`Sb8r6=- zKePc8nE_dsh6eR?yap*L-pM3Kumzyi+pX)&&2V*gm<+1xojI2Y&x&gCM&Utjbq`)}FzB;%;aM z%uBMj=5)dk@!hh#go18%H{1_QCc_CzKQHOte;wZ!G^Vr>^KaiasOFXAZ%$WQE@tX8 z>})wUq2Xs`;`Dmd(fpMjr;AZCWEdD2&hyuPHz?RNd;?hx{UCq0pdQV%4n}SwgniQg zNZyLM6XWwtx&vF1ZNI)&9tvurrjDQga4CBv3|=PtJX8s=@2u)pfEN^y8gG!{GR&8n+tDRg!N_$Nk}v<@aZeWhm&B z588WT!u0AnJECYV#|uH&ET)pe^J38J$g=A=C@3#N(ut&xvJO_uy1xFK+*A`0k1lA2 zJ)b1QRJzvwgnF-lU8ARHs5!o73q!IfBeisaqZu=Q5`{yu)MR3Tc<3wJ&&6v8piVt6 zcWbsfy=VWjgEY!Y=c8HMnP3*P(a7c(2|_Q(i|V^>h~&1cjkI}w&`zQpNY|`JzI$N` z)RyRxKR~txP=E=4&j7mSU(w>Aa2!0m$oJ)U37T7B+wm^SK12|>QlplSS{!5P=N9~l zu6XmnCw!5Cb=dcxrBjrch{Qv5|8c@7#qD|{gyPsWXPw9fWov6IPFhJPH#A+3huy}H zMGWIjf5MBjm+^>=`B{jYjT34uqprU|jLATcOthlEe@-%Ay4l=U67ce@uA$AI;^|rw zm<(UXZEL7iRT>5H^|m+u?;_TRn2Qm#wx++un~#7pXS#|@v~zOBzR?f%6DsnFzzL$>?($VC(Jza?? z%<}e_HaeH3IJbrg<`;G8k|JTRk_{zXz876Nh~>K>FZ+ zN9Fh3ETR=-mr|F;m6jv;%$V29Zbbr&K6qb5=tUNpnwl1Dn=l-sMcpv7+D;e}i6gly zVrt!HiT+(?5|u?Js9HJ44$?~`?WIO_EpJl6X%hI3t?&<1f1Kf~_Uz%>17IwUMCDd& zf1RntL0gb~>e{IAd7%xD$?5{XYjZnHIqP0p!d>EDLYQ{i;9)rvjI`}z!qV33hJhIWm#*5C!G!WIAvPn4t}wgHUdZ765hn4|F81}e0_|Lp zWUnc~$yhV47l#N<$v$+79yfeIm*z^JgxFxFg=p?)NN#~9qO_+YB1gUXwZNH_#B&|# zUAEZyWB}psYgdZWv~u%;4|Bm{!2fGkDoUO7b3eGP&uT=q*08GCXVPB1pmQSV@LT4* zx|%GKR;cye)3EVMlyBL!SoQIc2*0kRS^Km);I{m+}vKZ%y$r1E%c5(&MOnMV{+WBF$ ze|OUbfn6&$oe3MLLdueu4N;`O%U7nA}z=r4`;4goK? zrX2F)w)pebNW;Zod<-cW4W6@g+r0YEsQg&-q9OJj3+cCP0%#e9o~{Ax3HvHCRCnDG z(=D5f^Ye3%vxBSaz~Kz-MHh5hoVyk2m-5H1P-g}X5v#2Og92WolE%CZhio_O|9wGA z!R(?Un3aBSGKJ{&<5fI9?VSB@A>I3epw9u7R3>(rs1Ow`DhBz3jLGz|O}}#xoB{&; zQ4wMZObM8}nazFkO2SDjzfWvz1+RM3KR!O@^A;!Z5iuG=toiz)bchnPAQE;UDh#;L zW$p@dbqKnI1dLy6Rl5C_MeWey@yj!dB=e?b;u05c>XJ1-U=~(z?-o7P#OP=TchXIK ze^yLgumA_Si_!OJ<~wM}8L(W)+byHP+H^Y4*o(XmbgPzfAh?PZW}P~4nl#T&PoFYB zFb`sYwsXD6hN+Chcb&ygpMztH`yOZP)o;CSTja%sokrb|Xc&`ds-i$1# z=N&x!d<;wKkBHfkz=l%_pB_!)1&x%+&==x{jLiGUbgQI1mmB}ljnoJ}^J^oL^uvyA zK;Zq&;>MU_-m%!5_{TykfJgdi`itBbSpD+75#bP{3j78+V)L9jSh%XO@~Vj#kflCsh( zV~_x>(RFhzR-K22Bv`yA-ul5N!_JXQ_8<=Y3*}|512DP@@3gy+F)8nx)RfLAe=xMO zLlH9>!6LG!?VJNwSSju|dxBC@vhCWK5)Z#bbqOV#oqWqk^+w111cQ$$Azv3j+EG5U zu<`OQz8)ua?s$41H>@mI;|jhnkVx{dekwD=-Pl@p)V#2Ca~6F~tgN(CCA9J@)`@Op zi&Jc(3aN0IaM}s;Xx!&x@;qHG8O3y&K)U}Y6X_I(jim4nTE0S0Akc;Bcxx$oK}QpY z#3lCrC@T`Xh*pkqNaPSg=I`YnQ+fL8Srr1Mg|3u)G8bdhUjsiIP3<2=;axmqf!n6s zUtTYq6AKkH&C%*-gpCXG@F77Yi;HOkr~jN1$(i}Ur70u4_=Tb4`7vn)8#sxw>%k~R zzkhcF8({fpuKP|)NMg@hDD zn?7&p@i7gPFnOI?TY27>jUw0ruvaj&N*^HnEWB~Gf1^i7{sWwS!th~uB@a&IE-7!t zqKeZWDo5Hxmd=4G!s~qg;?c@A?9yvIzK2^;$;Y2X4g+gO_gw!aOl@e5>G$M|46ZI2 zC?xado9fB&$P^y~MM5>q$*}Yy^l{~~^!xBj>}Mb7!p9PTXDx?*@#m4_$IqYQaoVt) zsY;R0tpSo54kgYLMQ83!PX^5Oh87;ofF&RAAvbS4z~H_RBci2+cZVMB%x!u!N74*T zwNh4hgucJ$PSj3=l@{dv`lN)U9P)#CpUo?21bp{+Cph^*C0igi=;tw1QAWCcHdSS2 zx<#Wre26S@A%Z>-Xl(`ct4vbHL2E;Sj0RLyY*SSLUJ+ZR!9c=+b~qwxYy@3 z(YDDT27)2_)Wj3!u2qclzyag)jkRK1cW*bx|2;5nw(eDGyQgwozrjICy3`gdYv@!Z zmwc$OhKbN>T6C5uVOl@#yK}&Da8>%RCyCDX0lrc=vTBzeDAJ^XApq;w@V_dW z2?V1p^<2$d=?XelzObYh^VaNRyole(b?DvJe_VOcg5RLrt3QG$L7SWa; z_%EcNiB7}W-mM+REe@JCJz61vK+h5Fi{+O0$<^>YEJfZi=u zYb>lby`Rs^kshbL6>`F^$MgC@P*E|B&{s3u)TLCa&2gBz$-p8-3Dics{JG9`Uo!s( zk0HA?W=bn=!1z7`_DTLQRxc;eX8dE=h2IavO*7oLAYWgw_YsCgNzc*ef1Q?2NYc(8 zcZzw~4(t(~78iOFKBI})spxhj>1NL$Wv^=RQEu1LW@k2xzZx*`3q-Sfb&>PGTXHZn z#M(k{A_jh|m59^rVgC!Wo@91a_iQk6SAQ%#Ikh~?NRo|)om4HL^V9VKw|`t~wX{@@ z)ca4b&f9b67pPxAR@K}vGRu@X92H2o{Er%k)37O-p@!kcK+uG$zDXIdeK&93O3MhtZQoAJ`w2$pFQ?vC15R`cL2n>zo?;;CBkH{#}P!f~4j>8ug>0 zaB=i#zbIvCLdubf+2|Bp#|IH3nC44R8n|Eb%3i{UG%LltmWjmzKL5+m_tYz~+|=(w z`q6cFow}gg`ML+u)d5M&tOM+exaQ5U`{?D z&^^wA|6h7g1!D&ZtZTWwvXQQDZ1~O4s^>jSL7|_PbVfcYi?1{kh((b4j0vlFLw{HG z<}CXg+@&wN)x?gTMK6Y;Ni}RP8e7MjRNFJL9vzO@zLIa@Ar zR!F#h+;K~4Fk*F285qN9VWx%#S5F|)xQvU;TP0u3Jo4vH{3T-sJ)_`U3d6lzw4mD| zS)*HHPF^XvVcO4Snm{Vv&phrlnSk&L^t6L5yS_$#K5rD0)6jrvra;&!k>+@Dz6i7z zSmK@(WK`&z2Q%luJwzdSlvkyy9VkJ6hbCZG5>SnLGWpYK*+@Diz>!@kwf2LtRs5@* z6|nL6P^(;C0+b=QI=*tOM!!GNPRQf;w(G4=^ne(=ZOaL}v9a;G%=^8$RzB6bHn@!h-qHNdPq!T-Omnz+}(S?^G($%4{A&aT7Bht<;h# zKyr}wsk8S?>a>c^Tn6_-bY3_Tod3cjf9Ko(E99B{EENQK)$eX@45n_&isrvVBziZU zN5KXU&!k*_KbTNj2<+-7=xu~F?4xhk0K}m`Gb0M+(t_W0CVE~spX$$+{rYEJYR^U` z&|Y3?sV4j7HbRRDCK|mCtbv9Pk2w*7>nEM38V>urBIV+DlbWpxpFS~?;DlR1xGq;`ajh#1qyh3Ns1ullPSB}vS5W?2f>U_ndrtUGhO6mZtN}m zgm$hC25N;W2HUi_1leee*yYLI6&eBp!n*1#wG%=u3DZQ#kK5sK)p;ajO`^#{*`QFS znxC65iTga3{oPyZB#vFgG8#Ym+cY>>hasx7A;{IQAdS zUm%L8s+>YQ2t_4A%YGIWQ6;d?2+maW70)s+bF+BWvc{37?&Z4t_+gwyO)C+k9Dp^4 zjlQ!T4UiYMI*$4fer8mWHR301^3Uj_ z^BB`}SvgA0QTt~{!3|?%-`Z|(4b43chcd-s1>RUyRpGj1Y3XY;=xWP85)-$Pr6WpT z-4njw=l_2H^+%LOpkPJ8#f<=5-Ia9nY0Mq8#-Hk9KLN|os>fuZ&VY}tx6(&Es#C_yIZ-B8ocq>`lB+nqFBg_C z)%w>*2t~FQ!Yzw&$A9oSFD(^6I0zYBBCJ*H82q!uJymcz!SJ&yl1W!>4?Ja6~X zkQIL9yIUp|Ta2oP|4;X~k_STl^m9!Mg=2sXhTrwIldMdp-|Vrt>J+Vo>0r=r4d3nx znpl(+QGsz9){J6_Z6|!~)(kTE-{+Nf`@bJ2#N0a)EVxun*&sdyFraokayu7|*@!fs z(S6Hb3d*=be}5M`0l~CVQ@8N8Yeh%cW82sKU>zw)v`zIt3o7G;t6@|r;A>vz{NEJ) z%alR*?Dkz7R7E@yT89F^JD&oFhw&_2K|kW> zjFg|P(TO!A^TgjGGc%=SGhVUS#{9!C@$Fo0dw={|%Nwn!L&?CJb>O{nR8>3aYjv;5 z*EXDe`2VJr+Jq3Bx?-(3I5Qph_A1Do{2SnMWBh9HHf{O(~uDnqLnRU&hO3NypYHetmMWMOJMw~)f>u+eCWn< zjg@0P)1|h91OjJj1w?wHvD^qHUrK|M=4Leg5!W&c|FtITh1YXJ2)bi3y`XZ$={=+% zEl2q&$bAXQUh~5AfWBR$oWSwlG!A4H%)deo9yG&GU*lF(f?OpbSyad-HFgx`7<5No z5e61zvH|;WY)t>!TGsE+tI)A(cltttBnT#DZd;5#BHrGw8ZA`^54kn8)YNb|o32jDc|@?`B3umhvoXQ_q+pPwejOmGemNF%Um&0+`71yaoIKJZ`FwcrP0XMR z9P&O3b3WK*814VNUMW#fQH4J!?n=e!bT1xDZ_YVapb+}PYWo4>Q<*ducn|0oAU~5% zj7Rs|-^Pf1DU{{>Y7rPg%0$}t@ykU?YcsAxB4w_T;YDwNBM6g1MDD>JIK_seqKt`& ziTBU(3Pgo4O=g&ky=^t7NMU;ZQE7RkKgq~93rp0AebYr9SXin6@KjiI)Y2GG2SO49 zU2I6MiLGBL-Ii_#xnDQ7P(mj~W|Zq>11J*kiiDG8&`(>h{u@N;-`5+Cm7Y;4Rf|^3m7s&_7mI^ZtAqZ0Po-2QBaFgxwZIOHo;EYh z1d^JDbHIJV-+b2o1SBWEDwGY~f^?!6w@3v_+Yy6cN+Hj!)|0KZwVv5zPn(z3D^&Cg zALFGYHs_eDx$rRBCd9bt_^TOBxh;;?vwI`M+mVw<6@~r2PEeR32T1_3PJWA7K&)){6%+ zcLOsy9-c2ZZ1KS3z`ky;f`VQiZSn3dobsoqRdGfN#b1YHYt_}e*S8^(vK5D`sPaVj zNv)Bs`7SLvJ)5+ni&YA9bMihuwtK6xPV0AC3D(p{l-lKmc$1cHsyqAO{)Ei$6gdb) z%77{xp@(5>H1t~LrISsiC?h<;CnNV0P}lZNzZ4b~6_f6e=!vmBnj#TyuE~qCLZ;1` zN}<}5_%TiP1~y#sB~=I`?31w(XOKhQN0n|XKCQ1Z-`W^N#eb=c#+wEp$EQnl!u3b52L1X;Wk~d+5qZn71riPpPSL#_PsNLpbdXQKhLw* zOhSf-TZ<=HxA0ItFZ&l{M%~GvKw`);npuc0mzY@90CVD|atnnAB09&%sv zI2lkY03{hyx^IdD>doDqOpWNMox#YU$tDWQ-6aX7y|ouE&2Jn#k`kC1Ip#n}PJ`2H zLArEoNw04&`}@a1!Y3a~?oG(a9z!>cye4MFqd(3#Q_&U2FXIP(D&oVuz%^o34qUE$ zsxB{Qu*z@MTfM%!hFKthV^5-GU`=j~IKBAI-GYJwA`S~r`kIxrKnZ@qOO*?j<+cfX z(B8-Yb*Rh`Vh`315=#%3Bi+}HsMY8OAI3RcbTt}vI@3Zt&=VmzA~07)WgA|_B3;R` z*}f>cdry0rEG!z!67;9fLnXdI=1=L>L&kgN3{>lp7_jWwC)hx>mN|&$36n#IbkT*Z5)7&xVQlVXeUs$nU3~jQ3@>x%omyfb(V98~O$-b+godwXXf~Xj zoVp^hEeBV}{aLth^WCMSN+hS1B|)w#g$#Lvp9yI_Hcbi!W5=jG5S013>>1uuJAeQM z3=+s;Seaeb2i(Z^(&6nUuWm9@B~n)Ac`` z&@=612q*mYJ>z#H1Oj0-YF)U#?RUOBx~x`}qUnrhEOMO$tW1)Qz8RcD@$g5IS*bU2zwGVc1KR^=rm|Uq z*Vk9A3|f8y??-?Ew$=3ztu^W9Kyjs+nb4!mmzH=oay0=!1Qi`ZW}c{0xDYKLAK&)nIN1Vz zsyUw$_cw1DBT_EGgy^B4n80TMMP5*+q*#}(R{sgs9HeMyJ(m6NlF!0h%Q1j(>-k~K z=cjjU(2bunSrba5hs^c-O&TeT^fmrOK|!vI#v*$1)irOm{U&rA$kz=D$P3YeJMBQF z8{0y?jlYj|pQ|&MWyC6tDCIlX{Hz=i9|0AB)>816@!n3JL*HHQPm(aY*GnEt@Mg5s zP?GtwhE*Cw(V;(of-0F2;1wt?6oTEa&id-hd`K*NMy4;;r61T6LqlPj2TjE)&Uy}x z@zKR7wz?>1BVyN5Iyoud6LXa?ew3d^Ju;d$^jeRp>su%AiI4u)3E^Mh?5+YL4!c`7 zmM&P)lFcDSp|5=2#%v7=%;wegUJwI+F}>epuU-(s&mLP*9oZ!hbYB>wd}BiJQn*I$ z=XO~DrWYwaLO0Hp1JJth5R#&GAt>@o@}bFm9c6)m!{K?G>CjucX!~UxLKUx4z6CF$ z8i+zg(xdfNYh3Qd7Vhx{vqw8&yC7ecW_+qb#8#b>8K_(=E82nquFYZY)62z*VLASL zYQ3HY%=M1bE;j2bK;WM}%A2d;DD?Oq6DX|yJjyQCoR47Yj(>>|xrkjY*%#MyWb}~D zh^5^YVPnFU8?Qe=(T5>jD!>-Ga`*oiiDV0#dvc=VK0ri!36M=UuXoYz^;ojR0y&T=S`v_bOd-BN047jEG7>4q z;H|(4%=^OCtk`>s-pN76r2X9NB?MC-kJLYnPONPrMKT|fwI}I`PBOm}j?z(oRuUYt zX3YDtVn%7zp-8oWt&GZ-pqP+Eg@}k)l#iE%Naj6L>pG4H?@!I4^&^Hucz;DQipXjwP?}Na@kb?dh*EE^nGxGa^js|)lXr~D=F;Kv|Itej1=;lQYCi%oK z+gM;oUbtdT=rnNtvldtT{p$N&qz@1%^8wI!=d5CdN*>C$T?rCqrBwfgJ+wGWoH z-{FVi-roGpEiJ@9$auK9SC`>dvh%*uVSER8p?0jCL^CW902?7LCLFt5no`wLYrWW| zi!4bn-);QE=A76DQAaKeJBI}`QeuXXqbFmvq;?G^xGWJ$#zt{*e5`#b`h$B26W0w! z*x&S~Bm3>a!UVTg=mt6+(x~XUeruIuTsvbF|>;%Rshrf26$e5!6%oIK|O8|M` z2v^Lugt#G*jozw(P+BbA>akhZ7314cR|%j7uDW~m$71;HgT zEI^gIzPUvv7csW^DIV`*2a&+wHWJfNq@pJ+lQ~{sy)nvbHOlj`nLtg|tfTY45@#-2 zP)R9n`-tI|s+0j}=LitB6o8%_9msj$KUeCLBmnbT)BzVlUx^mjH}^S~b36NxZs7*5 z58|!7Hw};$nXt&p(#1 zv%ouTe>%-MIW`gAUb?=CBj)ks39QSZ!aupOb94?AUX42ZZG_3jMr!ndmv2O)#`Gre zPlX)_=Jmh*1t4|;!qxGd;@-%IjpQaK)Qq-7l&ju#+<&@6KEEm;lOQN1nO`pUGe3b4 zCDWK);6B>SamDBeRz#3nwjaP}a(CZ)vswrr1KNOCk)f`xJ@_DlMga=Ah9F2%|K17; z)Nb#L!~i&JfggN)<7>0QA*Lg~d%j3q#-w^WI()SXng341JUlo?=kbp0qCO8G^k?#! z6>h}!QuHo$`HKNlMaRvJ!vN>%#9@)Y{L$HSaU$N-TST1x(^HoFWot_wZ2q zq170R6*pE-3$m>MS$qZT*Zq$inW-pDl`lO#3y1?PwC4tL{XjEp$S`=q@W(MJY1@Y! zg4*m9apBj;2%qfnasSwFux^0#e`9yo)HUg6rL1PFTf6Z$s!~KbiVx9*Z+Nh<8|e5L zVsmO3T3R%}T-4Pvg(M_kquh`a#M_tf5o}QY`#Si^0DKs+pPgcb8{7Q8`O`$bg^=Ad ze?m8_YmAB0-G>&R)xV*PjVn1QvX7CrIAU#TYHC8{lrYNJdS6mW&owY)bD_P(v55~s<=JVv0LTCu|Uk`k(v^4v#4$Oa7bO1-!|1=tJ#Q~S5 zlat6r4bR!KuOU-r3R}m;@i21mfdL-ZP;Q={J!=DmkF(y>W_3d97DBmZc$uE3O%ojJ zO)d(8qRq#ESZaHR~U|f zS$G`T9mh#zBanG^3_|9wJznNuj7|?Ny`;vWbOF99(p9INnyuPWZjTz!K62*P9JMsJ!=3eS> zRo!Rk5dw;Q*);2h&7_+OGXOT>cDHh;sG$)GU{-~uRHCHfh4Y_vgMuh8#$0uU4-t43iT%M7X=_RMNGZfBfHzBiktsgMfr&c(p6HpB-9}c-Pw{ zbWf<9(!q+s!%yG;v+OIA^&)Fadq;JgO5Y6L-n(|qsCvp=65@N8e7;gf38{_yJAEsDlP4;cs;AA^xOh>+Qmk(knBcp5R+6K@i&SP z2uGS*XYBC72cjZDq*ERsubsyQFS%EfKMgKM=8eLTP(Z)KRZ>!qxnDZ-mCf-LJ_-GX z%*U5Q^3SF}G*BY}m%FloTyGQQ_bm=;*P>_hx$q@TIHY5tWTivvS1+tRaZm?)!*=$B zGYWOXV#huCD?5-#s?wN+S5kaUDENXU$1r}~oL#CdGnL6&mRD5SHh%QCGTAFOZ4Ri# zccLGM|Nh;DHuHSy-njjjT2fH?t+cr2)#tz%OfbM8UE$@TU~NVZx^b|wV#<8D48}Nz z1l~7N7#N|QD2`YE^4%IqnM+Yq7y8x5b@WB<(XX~K^Xo_0z$~p1e-rcR_xc70wt~-{ z=HF!E%OLwIR8G5fummqHIym?~h(n&5!R>LCgruK%j;zy^hGZ|RQpKXa@o1=g%b(ZQ zvY?P}H<-dqt;@=?qtP_>$)V5zwZnVY02e$Otfqm*BH%=`-U^#J+Jq?Y8LZiY4JPyk zzwVR#wa5wcKDc`X+N4pTkZ6p3iW_8VYU=r(z$Ph&YCjWx zRd|J$a3xIIEa-^64WILd>)}B_~!c>NS2DBq@$DQ zK^Ljxw^XpqMdK^gMeSy~pm4sM`6k})w1-gb1ajtDXEpuHYp_Q5*nvm#z8Sfw+LSITZXLfd{wDSv6+%V+1oz%g+S=<_Rx!fnq$)!n6Ob*)^XN-y}{U5g8GOWsWYZsR8 zn6xyLZj|nBqy;1=jdXW+cXyYxgmg+HDbiBXUDE8EXT59hwchXh$uH({-B*lpjx$uI z(+|&~Y1k5w)YaA3hZDrZ5tEIzBe}ilhHO0xDOHVd_x6QY1NAJ!=pqcr%w#vy$k6KK zBAh}>zd|p!*h%~Oy+{o5S}upYq0#?=a|aiKrRU>w=IRm>;pq2q%Bf({IxtC05jtZl z{+8NxoBWwFU_5NTx`eJy$WyVhwo=T)GvFeje+^gS)FX@fX8oLe@eM z=NpJLaglB=##qHgDPe6|0s7acWl8`K65?xsd8rbNZhtH@GVp?8XsnKJEiWM1D?<=R z;S*rCzozh7)!9)bSQck2+BkJvb-{0FP%J9GWlL1Un|3pcZ`9JI32V!v#F4pQgNDOJE>bG`xJ`(^oTD5_3%*W4}=%0K;Bjgqn`%FzknSzR@r`u}M&7 zU=>KxUovYI3tfyV8gcE!BI349)|Cr3C3-~?09E1Uue_H>>Bf1}ts6;Dc$~9@tSqSV zP*zvbxh&<165LC5E-&FrQX7WZP5Ma~_cMLLt%((B63+ick(I0YNM1)h4dArjhTqF?? zF$?eglC-M!e;8aeZvxiV^uM*{xCPV??JT+0rU~YP_^p%9-ah~Idq%GI!C90vJTk^T zubW<33U>9uNB8!`5&DdO!PjJZZIrJH&wXk}DCi6eIANS=wQtb0H=l=a%0`l#a^G^Wn#`;dgTBhgTKE-!ha+!;%?(!~8%-%lp zamoFgNo8&Q09#MG3tUn+E^WZNRoKMKBIdEJk^UVrS#hiMx`|!AiNk{No^%_3%nl7o z{F4VQ1h{qPfe_H8(b4~%}#hLq} zb8Er?d5ng$zh=Ti&rxaEL=nq>T8`ve26CijjlkB(kW*5^Q2*IoKQWflRqet375Cvx z+*c81l)=dX2rPdMzKC2tLso!*F$q|FCvtXzLcW4SGX+x-Z_e5{te+dg!QGa!kB$Jb zCS!lM#cRE0geedWM9fQ}OTZTuH6`a`OMwHYA(7^py+8Hc^ceKg$i7L{z zb=JgXfnBC`kCtMnUM7>uutg?X$0@M@;r0#Df zgYti=Dk+KcB{uNz@HL)9`UyzU_|Slb z!9b|ZvP$V!puHfa*uVr!a(tThm$IMr3KP4GfT?rQR2A|Y!(DdD!22f80`^M zA<7kbIGp(-*I1hGyGcq2FJ6Y1$uCA<-=*rsO9j_Th2U|mO9SOo2l8d zVKgY|IxM<)Q2e{gJ69B0(Xn1_gIir<7;op4YJhLgl2K)b69%EhjpZ6 z9maD6*6whBXR;=r4fGBpch zt_@ViIJH%}=t>77acE~@8Q^}P*bF_+p#^%1Z;P-DjEoLI29k;aK$%gk3lCrO1|(QO znWsR^V-Fc;I9+K~JAk>Hj`I!9&L;jWm{6;N_y@49yIq|0r9m>>{gFfDMAqEwPx+4| z%EBk4Q%+xn1^M9zxvv9`I?E`C83cmx-4|RJOy7DS$mQ2gQLZa0sgv3TeH};B-{?xu z6~G1?5Tr%OibR{mPnXPA*o0*z4Sf{dJF*FavDuR4b_L$HY7k~~j-njgwl*Q+)I1mo zQiujNNTU1m@#kJST>&xV{HoK_;agjT#P8k{^m4k(`l2%Y;My`U5#`33T{e3_3>2(H4i^srmp~7Zt3v6 zY~|R3`*8gTLOi=Rsj=vt<%!bX5A?+dE-rrJW;^DHX!3!&LKm(ok7i_J3vcWkQ(qZ@ z5;#b&xAi{+k7fl3pn$QkU=}s$;Qxb&dtHVY!O4|$>j8q-rI|9yEcN>}qGYupR zmW9^$Qx+re7KKU&a&oj3G$z7}VKCJ1+9*}Pg|lQRKI&{SZ2$*cpB|ee7EuV z^3{#>Oy%R}mzNUSB!B`Cj3sv3i4^EuUu~-A{84?NyQ>|bq&##Twz>XVBMO}Hs-;19qjFTrSEZOe10n~lPP<@U-VFE_ zSC=WQ)QL?YNoz_~)Xw{ZFNhczrwaz^>d7YTj1o-%KSwR^&2^KqdU|u(p%+JEr=+f5 z4AuJ1PH2Xjd4aPXNLA6qa9B^2o z8guUuwLUKh0mIdL`6&cLuP4NELXsOP+}Nx45qZ;zrFXgh6x%+i5dD+s=QqdVCh>aq!+OHrK&D)DQzJ!2lpD9{jHU4fAN0RU{}UF%(_4wFo-BzP^4Nd9ITh3=u)x=!^U$ zFc}qI@AGi8QBjtApv5UHOa?&RsrSmI3K`J7C0-{pNZ2&%Qw8D#+_u?N`A7jrz*#z7 z2y5=z^DLd6*Rbzw3he0%L2Y^L9@{cwLy-P&`~F;C(k@ZlTsF(+Y#ns?WsrGxCDP6mIgzkFNbTWQ=oQhPAMKNO^xjN~rZd zRp5JVOt2jV==|~SxUOgg6-Y9tilXbxdRAD2`u9u4FZ;?trK=qYZul56@ms$MrfvGZ zQF&5Z@&jJ8&GLLn|Gw1U{zG;s8|o>kC*JBSBFi#jSgBASb6pTv3vKZX32JFh0)y06 zLwWm9tsC=Ll&VryO$+w$ZM3~3ria&!O5-5Z!EJ1*Mf^dBH*uLa#1))}FG!oo!(*hs zKt2wz3F#BsNh#s3i;5E^3z16L&abYH?!?_$u&2bJ#c<^0AE07Bq^9$n59S$s&-aHR z%>B0)!1C}g8e7^=V}`{TOGl7L2cQE9desC148;>i;n$i)9^Bg5DL7wi;Foh-Vwq+? zbq^Ff^t}CuI+6+e`pbGvT5r@y^m3nfU0(c5~;G0H1T z^L8`onap5J&HY4;DM^lJZ%?rGUNYP@i;0&cWqW^o$C%~9LWVBqCR^}4LDWZPKZF>* zB7J$+l8 z0?&n@U{8Udv1t_t&bNjf&Q*o3-?dQC*^i?&F=ebMU0ol7ur{l4pMC%~7OD>>S6&*p z(*_Md$F+Yf&)NwKB$@v#BrPk%kkFz>Clk`8{ed9(1xb)A$NYUj7A?6L(|5^REk}yT z#L;!a14M@;6N5ef!AST$mGI^e$_R+({h5wQi?hbqpr?-Hb zOt8ZS^=Vh`OmdJr(XIMy4A+PWpC;c1_8zM@8Ry#lW?9K%&V*(~I`+E=2zXim>mE?C z9z{z%bQmKR=tPtf-w)IS$iGAbF}p`M_BUxHeUZ-PaX_6^5_OF}9t$2|9Me5`-YY!MxRPHC)NI4=`1O%CrFb9-2O|x1VThd@1g)HEVS7z>P0SW zdwy$c13(3oKR*wv8=IR8Gmai_NG2q*x|ml4f{J@{oL5lBy7bTfOJha^GOHO3hn~lB_#t=9*J>Ep^8T5CA zlG$wU*A&Dr5=?jymc;YKwj{+mEj@56BN-vK<}?#l40Zwa^+?3o#~b|GG4DSZo`W;T z)pnvvmTj{Vua+8+v+F0ZQlZ{a=8nOdIPo^l;sAgC{K%y>n5I#9~1%oxR z*oKQ;LPtDDahdeSIe(zv$zF>t@e5JHZ-yZ+2d}fE_&fXeO{oL5S38dGtsao;g%H(k z&0Qa2`Sxg-2=fhFNX!F*Qa{1z7g?LUoJQR*5_P>C4eVfWw5>ZfCcYQ!M*ps}tyx}e zk>V4oRTkHZ&)-M%-!GC;c$;Ia-0O1{dG?c=_zrxQ*dDQr`vIO^^ULH{NWq%_GvNK# zr~hUM%?%EuHfK=QHHCp=Dp11niTS|X>6<%D{*q;m$>u-_a|BF^{D&o}=% zhCdK{C!(m;QKr}7?{~Zs(mHo8xH3efVP=-Ntm^uSH1?*RDaF=>U?Mfe$_XB zbJO*Z4JyIF=V$@30j@)V>*e8{=XjxS)6GmtXEL7Bx4h`o`zNiQ=C;|lUh{k9pj^Qg$5rrLGl0!Z5*6Yc0vgm})HRh&h?S>nwIMy;U!%Z=FxY$$gpv5p_ zBO=7TS15wfGi7|HJEJ-sJ>Ftr#r7Z5p(f?(?%}; zT{DPT)2nEYesC9DY;7yKbWRMbjhaw68cf$`3ISmnK2dY1()XCL&96Lg^JwvYfrFS> zM0?Q+#)Xl9WgqC8LFZ z;f4<$tTWVfVzzX5vi)@3*Vk8r-FuKyop?*_=$k=6AQ~O3nsT!FD{|zLFWBiP-x5{e z&!0ax9!~v4Ts|-Y4ON}Zq1PP6dVyu+QUw8phI+ab_b~F3q7)3uyc#Um@UuxXhn5I1 zNHy;cDjtyeHuYIUx1TpsgIN)9VgA386et=ISWoO;_06=ezhtWP+R^(aM2m+O&bGL3 zKmWQ0Ynh)j))u4L*02<>?oHqS?q3307eFvDjIXfL@^a$V-47ThVgfBHz+Y|PtW6+J z&d$jxfl9@uQCNR?APN{gw4{att@5~?Fj=pB!#7v78$9Cu;$<_$`8O$q5Zjax1PXZd zqPWJ}{T>~SdbHev(MPCJ?Ck*vWW64lzkvCdmzLs_lGxWBJ-)t~Kh(^@f7XB&|p384PVsVF5&XAqUH`O$ULyQ_V zWn|doE5B!T)?M;M7q8SV5yqfw;Jqoh=y*3^~ zCeY!{^WO1`{kjCI(A_PJ!1?wpQ=kj{?|v(a_mRRZ3J`K^%%BaWJfWE2fw38KWzCiz zQittb+Jd@u`_B)(Pupy46n$?g9fNhl?aHLWXwXhDQ&aL`_CVk01cymQV~ToYD)47H#AqXC?ZO` z&ar|08S%m*+$`d1M7rCrqA5ywSVl(3X7IblPCIFA&1&rsUMHH3UIenE^dZ*&I5@y} z5ix*O2d4+=z{ch=?yEei9*c)VzMOawyk8c@K}wCD8&{>@O{){cNibkIA6pv50R`#3 zGmk`Ih^mc*Xx8Ye9EXp3+J5w7O4@7kPkaBS6l zx@{jn8BR9c=6*kmzXP9(z*~IdI~XG>Ffc=2cBSQ*XP8=ZeTtA#mf_2W;>K1Q&*LWJ z0&G#m5Ao=2WB8zBVq&gHpcG=p`7}nx$0bz7-XDzO#5XPW^v_`U-V~fQpa;hUP+k#6 zE9>jhyXbRsa=KUpE9Bpdg{dOkH+aPgU&YRPPnBroppWv7bPj`37!CZt1;10f7h=K#fC>_(GZu2B3|J4w( zASbk5WAV8{+^A25hd<*rK~=9A^lJ%#sYMz}_Z$3w{~H$h*P9g}gy@Xd+a~Sn@>`uc z@a-qq%`Lmax?Pzi;P_ht6|g#$@2a}9b(BQ{;5LlQ+I0(l269B!+fTcSR(eLdZNoZ% z_XirfNIFhxfn1n_YkQjgsp8NwJKjDKT!Bpg*M(dvme^y}@k2I+qq8u;id6ln?|b*> z@FF;(Q4|HAB{}tU@oU*6jc$iK$fz?;=pIopBG~6dq+es5lpT$q9HyFYYbb80u4lu7 zm0QHu8c8+QgtCVPhn;3JxM~_}5g!Hc-OT+I(xw0xxY|$dyRHk*0=02nEiF~Ud7Xk4 z{P^g-y{in(i`5#a7dGW)f(TTchE^wUq#5yVRR#)@Vi-}EP*oE3rj51EdP+u^&7s?* z^V<=dX*W2!1X_rsZg4iYO=5LTIS;ZPSc6*zv`y)lM?%ZJ{?5#)BeZtF?z|TV7)4u|LwCBQx-%qr|9xqFqlXrQeE@_e zX#=T#SD(LgTh7rDQrbjsBIgH_{^bY0N-ZpKfkj+Bbrlvu*LFYi3FXTqurIez2p`~B zvKWZzZ_7&9E=IQ)c%cD;lbU4J!qd)kLIsg_M8tP8+pW!CXZu4g`7NgCKFL3+W@a8Z zocAd%%)YOW<-$GB!rT3nuzLt3QNaM3tTuH&ARI}RL&wCVn9AAh)53mCT@LPH|vQ>dz0sGYQyJb#{j#C96iQNhGIVG?FZk#bFfVC`em{kZXA0gfc$#>(*Ux&iUG zXT{fP%sQp#hUJx>msj&xn5~_zM+*U+UVk^mHs5-GJLT-?+^W638{jUlsg$Z)hG=D6 z6TeOWm;&oY%leJrZu268F+3uowJBw*6Cf|7oC>pOs`}^|jmhN&2#u>$N z|9h)cuBO|*@ln%*g51Z~leX~iVx8b54lq{gotaTxJ#cgG3P%Qvzu@d0xWpVbctw-~ zE-pF_57c$DMdve2C=a)S3JcXQMB+l?S_CmzT4Ql0|Erq&h6_tfOngslL7n^eHxQT8 z$cdjgQGh)dho0i-i?l565+cdwVp5Z5psgMG>-n0sSX#2Tx&S>+3U!EREN!zJV=eAi zD$Ag_cvw$GDKZjm>=q|h-@y5LAFOi>K4r-W$N~ zJ@#(tOpK@6M~Zh-(!fhY15e-MPYTOVyw*<1dsG!tk8KeBCOc&i@gg7YMx_g2og%W0 z%g)Z86nv%_)pSFR0@3!`b1vHiD&}+!S7c&Ty>+t#=DRPEN#oJWIdzsFZzDfZn^%Sg4zT2R4>WI9B*&(xC3Cy^}7oCZMI4e4!T&5VOnth#Ly1%5&g) zVT-|QUlU^g`W`8RGs3>nX%I#&*S`=Iu=5-Yda1y~l=z9E`u_u}fl7 zEYUkw0cZ8XhK|94B5k*sRvEbpWEDV-moNs*&ufg#RMBw4Ve96c&jj9upPh$}FKB5` zC-O~6KRj%h5-Q-^+JrO=z*dNN9_rtcU{9ez_uc%CIC@FKiYsVqgH1^ZF?sO_RE18L z1IKeUJ6>#$sQQJKKLEDu2!^H$t%xdK*9SjUlNWwJ-*ATud7RS*ss(eJ03wH*<7Ha7 zY{0IW6i6l5?`WOxGJ{gE!pO1xs7r;5>WbXIu(3ss^I0W0Bq>iEQ_koMIctTqd=b_@ z?dAICjWQR*@-~;o6J0CUqGALM}*@y-|&sI&3b2g z;1k|H)KP9^vEVW%n;Tm_;6OjAa{FN;V$_OlO3t@(W*!Be5DRHFeO#yG$|qs0`PXexg8PgsTIp8eIM?5gQ1iQ z>jvXftdd&?IQQjhYjWE8slinJf+o=_WTB4BfO@j z1~agQ;PHz(13wq`6KtOM1rnFt8Y)F<1h}w=%VsGUc#UNG764O|Kfd7R_YNq@7^p^urb-Qs)uB;G8o#?~5v6E4eDoBr>Gw~=(q=%!<&UG-7r0n9i zWQ%xv3%J@Y#@raO3vl@`??^I*be1HUdY^>7Wx`amhu?LK%m2oO zfuQO)$es3E;;SqP4Ms)?#TsQvJjC2elMdI0W2Wdn^;eqPFsv`*tA65dHyL?zA)W_Y zZ~dMYUhh59@|3awrU9c*1Z0W9w?dmvqC0^yw5KnS2~*x!SyNL=Xq~e>--1mHBoI`>T%;!le3Dr+y3aeyNt*ZNI)=cHeZt zmmFIIUHokCG*3iXSvl?yZ9AiHzN)hlIciu%cA@xM1GS>aThyJW+JJts;JTTG)fjq7 z4l-Z$>FnG=#KA#ebTn~&_b_;Ww#-V`iOYIHWVg?izVb`$=;FdcJ^(~$8-)K_eGnbE z0b~9F5YS|7-Pqt@SYm!x1;=?yYwH1^A(pXy|Lo%vt-)*#rSJtk%~rFyjBIU#s_q{v5@H2)3n0`X99e_MdQ__d68sytgi&q!A0bT<0yUYsF&96;BTmFXd-}db+R9`_7SvmoQ-u7ejDPt)|u)P+PIFvkQjru_j#o z8gGP}E&Z_BODzbs?eP!u=igXiONxI{5v9Jn;VM9G%lTeAB=bnf?|NiY{gul*J-lYD zPzFQ(0QTI$c~3J%e03EvHAf{Geq{xtAAw}?7HS%wdFIW+GL-8TZEhYK9kgRUPQov5 zV8n>|bP6zV1S3t`dIAA>k0+9E1s;@1uPCIy;|tWfMii345oEM!0g~jvD9Ixu#6H1; zV%*6+E(RZ;nZj&;R#?-C6?PFdC?|ql;p#()u%NbOxgYbrcyu-}6bcTA5NfPfYnMh8 zP%A@n#_#SJQV6VXg_9CEbd>%0-G4rQr9+EWqlCYr|WPs+hbhE~TosPLjnYofzv#A?!hdpGa z*FTK~Vd5z|mHiHp1*v@wzHj8$z<6HgmnLUt$F9B>34e$uWu!>`HAY0vUKjjz92;bt z#71im7y2!__i^gs-IyC(vZJ82U?-`xv77vZ41KtGktlPZco5`KzV8>Q+Yf-#2#>iZ z`>6UcN|FsXA}h-kzt`yCvPu6@74)eN0=2TrJiQPv&Wjs4f;CM5SVn~3f8X+kqVC@5 zrDo9nUk4Eu`sp=i?cgQhJ@E4)0;lIzW?o_}CZw_zQp z;WYgLFrwPuaroJ)5wQZXny-4RY*^0chG|zQ3iR*Eu=sWFzqZ=4nwpvpPcr)}OcDy+ zh}Lm6CxJVeVr&Y2)gVy%7>}cAM}&peh8QNjMVCAsOgTbX@Yy#{=YeK78~vcuOQN|A3E z#cXXks2kRp?y{p)Y+**WI%v5B;s9RXzfZS8a5IzVb zN*Psc&m&Y|upEAURb@@iwD~W|Xk$uGQ!vlc5_AnW@L2LmlLLYf;64x$)c|Q{`DbPl z7Mqs#HeRr0qj|3%9x^VvMBK%=AYT+Whx}okHmt6;!Cadk`IDwn7N10F7CpM8O1GTM zX3ZHO8uji#=gS7?FdXo**Zrv9g5O66k|w*=t2H7ZX}-6_Lxsk`qd%agBq+(77hU`@ zF)4|Hl>GSk_+!@IwmP6^@VR+glg)=AuZd9QW5^7+FgiC>KiMSb=8^)g;j&0ozXwLJ z4-HwY#3cqG^n_L259`e&kmqCd0-2K^ic9k(jV$4>vCE-uS2 z)kc4LBOI=H;z|uMHvJ_kyIud;vQ|1$fUp`!cIRLdllzp7vjMTNek0r>!qJp$2vm4% z@80b%#2T#`_TnAub6a);?0SsZg zz@R41`TaWbISOAAVU(m9nqfxA^BWyIz>m2G#m~n_78(ZTo$U%Dv5+HDGZ*}k!x}!D z{{OT12ewlFyVu?)1=rO5s$Lj!v`|l|CajVWL&OcTmOn7)T3E83BmhB`~#H;_N$`DwABdM;dhEw_V{>UK^nL3%k9>Ro1;!F zx*sC2*Sc|m!{y7^&|v`VUfc8@fY@j6JSj_{p}o=T&@CRXqctSNq`+Lyd4Mp0BWW1; zEKocf&Wgh5l#j;7)j?xIpB%H0dB&T0O^A7fgAOAI6Zvc@`b8fl8ufK7EA1eOETR97 zML{G92{EDBV%OGWKtJ>_9#O0HYSt|$csm+93?lm5O9_D%m52*)pX_@0jGbti1yT7O z%`J;LBMGYddi#ftyJyKG&d_i36EW=}I-s+B8lQ!E-mqD0FdhE!V=x#An~pa9E6oe+ zH{1WXYk-C>ALTVJ4G`QW=o$$K2(&TH26JzENPN|spl_@#2%%oOc%9aEsolXI~g;rHOu zx%598J6l0u5p4z4)+k?BG|+!9Ua)Q#^NL}--t#7^?yK`rLmI%Ch*!yuPH&Xw`S^|K z*+qQa@)U|)5T!A((~$p3o*qCd&V}>yxlp)DZ^G-c;ECAz_mZkb)Z{4%x37m&%_+Y! zU^6p65heUA!FmiuT9JbOyW$jg#n$LCDKt>^0=@C~rOOT!Jy2Whv#bJ4K`8~Z>uis; z2L=WJ+F(UYH9_;iyJQ;Lb=4z}9PHj2?C#Q4`EEay6@DWcd!$6fn>GH^ned^ojrDbV zqPSu+pX8hzlIQDMzt0*a-?;wQs^-B;;3VD7Zohv&3};5u>DJ1{$w&=#q}XUYSFUpQ z_wNvpTIFyzLv8Z$V<4YKKJ8cklor;33xMII+Q#;UUq$d);q|KQVWysXSiDX3#W{_B zy*=}Ddjen(rgY&~1g+0sk*F_!ut`HI>fipL&g)xDgt3^gOA2O@H`c-WMyR}~O!GmI z>Cr)H-MEg^DVQRrt)Z^YKO;jy+x2FgVwm17y)Fow+d0#BtICpJ34ta6m@mnpou(nq z2i@Z0qAL?UdY;L6)&0Ln}ihlQ1SPSYl^!^(3rI6z=IWXo1Z{uAv=^Y^TdnMn zB{FOt4!zelr)9S#OKc`5U@df#82J!WRVpn*!vf2M`40~ODGLn?ynwUI#!-U11^8nI zaJ78;Nt65ktmyy!!EHnRySloz<2R0v(Gu7*zSm|IjJ!G}=SKhq*6cu0pNkPROtc7^ zqzUId!@Bh`Jd7OlvgWx1@&MWH6Go_KbYMaS1(S(}J=bLq@$Zl6CoMRsW?XD#*@j#@ zH#2$qU!W>28S0*A5T>}ThKks8Lp~*uuw=Ie7Zw(F3B{(d2k4HH>z_oz_4jEFrNnfd>Fn!66~5N4D->|F<@J6+ryQA1hoy z4ZF%U_b3W{x^ha0iAk_Ox!=G}DP=ncTGtQJ1C8T!KOtVMm{Is~)LhmH6h~52yRXWP zja-zqd?+q1A33^7?vuI4f4xIA?l6m=FrqxLA|mGt`4Fuzy#`N*6B+AsfkKmy_WGJq zzx_2(3s^9Czs%B5jSxW@tKsbK7F3qMsb-!yBjhajRf+^+itxox+hpB5c3bN`e0k;u zyxBZgcNAGsk&y!vVt5ba~=#69jZ2D&z{h9RX?)r+ssQce~Vxhhi=zr3J|5>~V#PpJG03 zps45>APM>i61*koAEQzJMqk*`mA*#QyE=*xlr0S&Hl&-$+?p1dhKOBZEr&ZB1A&v!Hy&h>&JrGKitr{A9HG| zn2L|F*LTb$AI`cqjy+)hW7qum7NURMLi87pt-!K5Q?PknSxBJANjz3?c)LA{S7lxy zq`y@g7Hv^g(TO~b^9y#|?oHsk)3XsY{A?K`=7Da+&Pq8Tlnvc|Rr5UEe9^`z!w}Ky z5VZ<$k^`#B4cG#{+tLdTjH0>@xgaF|`2je;}`NVR^Y<%nV)-sCnue1NxKi z*wnd-od1Rb`KT1Wl6JR|Rj?@U9F#EuIAoTUNcHFL+Q?@a8AOe>7UW6cOD1GSN5emy z=-8d`Zt5d|X#oR>iI@KE1LGzr#Q`oLwML8Gg*P-bbmMZCA4HN=kOWvvrR50wk^0^m z@=JC#8?D@$l33qgifPT*u3p&1+lEuivqfdPp~YvqYks{_DQv?=+aRM?~FehHto+X;5yT&L5MbFIdY zXwhNB-D;PSfP@Xn_eU=qcaz8mAKXNgP0So8_)tYTo`h^K+olW;2{Q(+#m{6_cM-#$NlH>7>nXaIbU2CxkH`*xmwSeb-&WiM(-G zTM(4PiFa5PO43;6hJtI@vFJv5VK_@h7Hu|Ol`3PP3w}4C(B=)-Ra#PwjEt;v8bZI- z;VWvZfHvQLYtLxkqwv?N+lnF(uF-3=cJ~)$8v_vq7pc41==)#{LNMp$Ie-^w!)5_P zOfV%Sy|6CRg(LR*{?8WyeR_KHVMQ*YB59T90f()|KUTfhfI;kk-dk?TJ=_SJh9WGS zohkEN8+!f!ST`>Z0KezkK_zLKk}&$3-p zQ**DvB=P&jaZ=$Rb4xNPd3pJcs~P!)#?RP_IRY{4B@Ye^VZk#RGOi(DqcOe#Sf|s> zxYAOZqsM1|9A*twm1yNn+*;Eo^V9{A%zJOZV&66;nCFpe%7_T4vsF>(mFfi7Nd25N&=Tn^HAmDaz)rWxGx| z4{sK{+gP!IVu&|RE3S^?r2N81aQg7wwd9f^)!cUf?{Uzjpj6M|O(t$W@Xk1Yg2m>= z1I9j~fq_bZl7gn{47CJ`(#OD9Co{|bd4+31#j-JlQS*X4Y5ClE=fCGHQRVX&Hi=m0N=T1=cSrq=`yr8e7V9K^cIH%6 z=_ekZ4u;SzIpBmAwzp@oy}K))5Fkc;d3h-gXcC#2n3Od%LIwH%ZVpec5}5EKW5oK9 zdRBW5(8ry(uAUjdY?>}r9g90J#8{tfA5j2N#0(_TlK>U@b(#LfGoVEgZ zOx>YLx27Vy%0%}>0z3-Ui1SM`fftR%=*!|Z(T}PT2G}`b_P@gyXHFk0M&!>oMwm&! z3+lgm-JeFeh{q$O*m5vP(Xv@SL6w%3%6j|&950KFs}I3iK%7h{I$B>vxq2ilzf`UH z)g4K<^JmsSM#;vHwfckXDgFT3Tz6<(+%YeMie*(pQ@^rN0F$i`BIc|_ zM)n_sZ=3}Tqf*cX+21l{XS4bBLRb6_CP_P)18q(MaPhhZCSacf@SJ^C*r47OZs$q@ zgyCfYD$?Peg|M2%kJskrylKdWRB04oBPywAg6X^u?j+2VzzgRODE=y{n*g-a`>_+` z!~pt1v&L3*#+ET2$Y&iB|1mK%bH`C#Gqos3N{^EdYmY+Mx z0z8#$$nzB+Re$Pj`yBvLVT;$4g+hK9La{gL{65HyfTWn>{3NI4cLKPZFUp{BmEy;m zas(vH@rR+YC(cYIGA86Cnp1qoxcn>GsO6Ug?S#0-WOe$oly$1enkWc3l9BiB#eRR& z78Wk1q*+a-Iuo3hMoF=O*wGnX4=Ca|tXS!|2`m)4|()-~no?^|BdN08&~9rmo_7Fh4n~V|4#7Q+2j7%TR@JHzyU`zJm4j9u5CFx*M9)gTxT`TgM*quzF&edI z^~b9dMN-=ONQ_DF zW}tkQh~-d`+bhsm$ao`{oJs}R8uT-?PrpZ87)x_2Eak&}_6@4DS?1c;{cLJVEja;) zfFPf+G!uI#{J$5wF}CMN4!^@e50utV^^N+svm3kH3nggS<+W5PEACyf0wG)cZZh$Ln?ZX^mqj*ByZ4ryP(BCVy7fkIGrtuiFQ!d+;Fhkx2v*QG7$TS} zPS??iXZ!9-{3Sj)!VPE#~|$@{4*PQTMwtrMm5`toi|NMpR{`21g4A z<;^G@R_*w-6s6#w3(M{8WO7dGR9KOu!=>q!8JSH@mI=-L4zbuEgvq|%UUYqZeJXD# z087BfW7c+Fbp-(5|0=H7peAUQb#v3X^r*2czF+6Dhz&ztVUt-~eo-ag%M|mKvxr-`NhpCQIQ`ZYhuo zNby}SEi5&7;(S+!B#*(K_#dXs^=Q#ugou*WHMNMwDfD1cSM}LPXw>E3j&u z%8TEI_&Y7QVZKK1iv}z^H`n*2{V!)f_w~Xn<)lR__yrWKWgi&=b4!rmReU)B_R!x` zfP_K67Po@xz6cq7;4w#d4-(v6y?xow{=U98>wdfcUGYzmLIWzf zpoR|A%l!vrs@2rvcy8SoRa?g z#)IR#bkIz>5_M8Mydixno0(7cA3jhBKAQE6&P2kpc}12_1b+F#d~i6cqJtgfuq6d1 zCl}F4u=!zHgQ6QGCRw_QatJv zYa0hm3pPbnMYfP}VqXjK!H~<_^&qZCz{KHYfzYSNFN*-^%h6x!<{^ifeQo4kqm!yu zxt(RMek~HjRQj=OO2leYx>y{@B#ou67h-;L77^tq53)%y`PdD ztSZWu%1<#QeVC5o;kvc?Kc)&Ivq<5H(uQyg2B50N<^rvV-R*L0QZ=avrfk@PGNJokUVmGZ$dS*Wi(=AM8?nld0b^T z61Zumh=A+Cv#5W}mM8$wJ^%nZ$-rVd6Ao;0&yW7nG4m*I8co;IKD|i{wKBz!GfEB3 z@$qqp`LAwoql25?%cdaz?rvfDkTCKct@o~a3*nMQr+`eT@*de*@UNhPog;IkB_ocL0Pkys44~d6nh^(M8LUwn#$H^(N{EO%(Oy-B~WVMh2ZJIgIw}>pv!M ztt3vfrVFd2i-KqcnaV^fs({88*L?C(Vf=v2?F6f<*|lKoe;t4S9l-v>3>YuP2Ui?y zg5~zVyoqsyiAFVq7zS)mN~;L{_SzFNh&Z@F=2%?rjRJ&Bv_H2Y zb(ofjn5yzICOQ#|gR4b^3YIH>uP%&6ap1Sx$v}d582SB0baXTaLLVz9JrmQwbdfZ5 zfT95Ft6L8Qf<5&aswzNn9*7E0s`~MK|Es%FaK=bt0H}^DVD`b((Y0bg=v!ZZf5lhx znq+C20 zr;dv<&GqBs)h)-|3LV(j-110!ceJ!b#ljm2w4FFzC>^JVSgA;mB9h1Sf{O!Xa`P+c zAJiX;cBu8^l{ksLCl5PodM@vBvIQOci7zDOi1i%5n*zVy(RC^E7tA;*PKYVckroLkI61v!S0vv}o=o9x4% zm;cZG%>)gO`$e1Q9VpR2EAi zjoBj%VbOcZ#GmWdi{gYTKOsa}^}n*8X!E~9r}Jz)ZIQi*=;{)q?g!V_sH=Uem03%d z#)nan*q(ZRdMN3hO7DdF$!2PATaI=}tkqyStlj@tpIVdcVJ30(-A@-!sM>a}0f>j_AfOVTv*Z z_39m?Ofp@@#^NZo*Cyshn-Uc7eh!WSkj0fnUR<_RAG=lNJvXLy^fv$Ga=$2Enf7QSTc z@9n}|$S~hjV>7>jxU1hWQfR>DxYemh;6{b#?Sci2rO8wf(TbJvR1jlPRzk&-7SgXE zBR8feMya4rbNme5%}tpi%vW#de!6m(qr#yR78aH-rJfCeTjBseXI!0}Bu5JWwV$zn zhdjj~{%Pr~uTQG=cEuw!Jp7;r>lJX#pgOl4Nnvx_PSnU>A_Z*O?*I;ZBp?~I0XRw~ z=MFq9p;g$aLN1wI4P4Q^i&IbcM?bVmLADM!B-P$ig)x=1;bgQVMBO8rI+Q%@3g+SA zXxu!xcMT9vw>_{kTkhLD;+C8&TM)|W6l{0)+Ymx-HFTf$>tTQWhI5ib}z_$>cPKPt*0B^6(6 zH5=+ZT+75#g6M`JF7>zRe+v)%oyRyLd^u20o_HH$d+!%ob~qw@fIyMuRfL%z8)@6{zKikjj)fHg(MLN(Nk(S0;0f+vc)}t!emHIg3MO z8m$)_D1Knuk#ON@4aVK=|HwK6)O2LWrN66bln-ankR}hK;o*e=G$uf5NM;p{Z9s_e z=-{#yMh!lVH|zp!uvkmVJVlSACt+$^B&Ut2R6+_wOHHX2XcGF7{md|sbr$j-H& zuz_LhjCDoxtNJb^tD*7T!(oLiZ*HI)`IM5gRnaoQU2jt;N#PiMORFQU)(qp#=vBok_6{RY}BPtGsK1B@&^v&pyXY2$4@&oKg& z%edhZzgNA_vF0D@4EnmtHJ1o$yW3=bc~~7x!$Qx&qo6B{m-~3OTW}$unY#_CRdB7hY2n z&HAAwQ&_-8X87NhN&gx-e3uky)U1m?UJt1Gd`0W+Tz+h3%M!StnU^COreaF4BJ6{9 z3k}_oImJRH2%~##*xW2Bsw>#`;e!-Wlk}WQR=DAhT|X6cWCJq^^?j}kSD$`;fXfb3 zEa@-@87(z1@{va)Au1)o>Z=qVP@DEGpFh|`afJ8|pfy+zyG&s*WK(uxGHGPFB1D$P zx^WRb-kkJLn#!^!lEZQqyopSr2%Lc-85OLOV6lPCCFKeR`O>^d=!ep$NMD7xFpEK; z@hvbHamIE7c>j0Rp=AjcqBFxN`#%}r{ z?0lo!&+f(hh>hM2!M>=!dPs1nSyADN$UAjpdeA8Rdg|;*G0Ak^sG)c3mlZksQbIQ; z*tBJI{W2A(#rFWi{bks;((E1`phd_wYhmWK0uaZNpY+k@n|pRI#SdEZTJ+O6=Z%_i zp;AurxuHacxTG6foyN+5z=}ANAW~q1a9Cy}Tw2eDk}x9}!4(4^(?EW(!jksX92?FE zzFNkKN_oT|vWm~5l+MB;JT8kwl8ZKh+x~#i?#ilPBM_i${1=cAMW&>p;;-^ye?#=y zyep|v*p#+_gwEQ^KdO#%3zPFkSx2BXG743yD}_L^b?T2uQ$mSF&quW!-7PprCttQO z!-p=TkZ{x?BL~Nv)1&a%*wXP&&-Y8KUy0aIA3E0aXw`5HM-qB5CksQ?@EBq%_nEAg z>ygmV(5$d-wMihE=3jy9Txt6#T8oKiWYr-j5s8OFW2wOoZ7ppskH%aez3zGO{^@KQ z`YTs?CmJ{&u0(_+CE{Sc(x>5%maQno?E{)SOX81EtVbB~X4FJ|<)JesV)Z zz(u}3$=DL+rIZ}cv!B>pTT++W}cfR3vr<$grfLikfq*QAWErlO%CS5I}F!2JgX zUXo~@)Snn0Nm1%4&u<2AIGJTOxX_X4fBQWO(p=t!G%QN*`P+FH0HRUV4RAWw z(2F)Ym}_XL_v@NMtg3Uq<9$bvMMY_hA#Yr|oj@{&I1G)_Md@HXNz>xwfj%$6E8(xj z`HMurvr$UY*LtNDPiy*vTXZH2Mnq|xIoAnzWuc!RT@a^=op@HP8&n3(+tBxD)g6_V zt$*{~jGq+IWw!NCmuKB5#+!XnDuO{BG!m@Lv#@(=kSUdI+RaYI3o<#Dqiqi#wjR(WbqC5V zyosX<_kO1MIGVwaGpO~xF~JU=xm?|3eJkfCWO3=Y ze=P4$C>8?9dCutkizam!_dI)P6 zi{B|D=PNl(!By8JTA{2`rS))yLn7EY^>i{+&ApGK4HohS5dY@BCmTtrW{RGiLh`j7xR}oF6>01E zUxZJUR!7~MJ|`5g_RfIPjVnh?dLNT?Kz#i2!k#Iy@T)p z@1)eCeeVfflFF%gr?jxd#I@vF?XpENFE@7zo?`=sT!DHcI9m1<_U`eKg4_x+d~CZj zcLl5qzv^<-e)m^CAe*_k6B=erGXd+uzq5pg+)k_nc}Msz8=0H?1_!jUa0Zl_#)l7C zhH5LJ%Gi|N?g{}va}bU9IC*hp$mhQNXxO~n#3o;20z;`A?Q_*Y-srb&ONK5A3<;6t zZQoGP_03xX8<*1o5&S95Bgn#igX|ZABO~756m?WQ{|yU#Km1if9(*s$XTZN*WxbtR*z6dpkd4h7?DMpGjLf1;g8aOAc-_-em|v$tw3d}(_P()G z0==4nxt|xMx9K!@N5?V-gT~KqIt^;qdp@JK&);*G#ij;|{{)Pe1!ddSF*8Pv{9H#D zf|%F1F(A+)6T12>-q42rsZqK>r8uart%?4}j~{uIKL}U2L}8tXrgq1~1cJNXO^`^= z;|&ZA`DbP6XDvRdGr%G!#yz~xw{ipW>^@%Bk@i>Gu3iNx-bbDgR}CIMQ@^=7^Iz>4 zG&0kBuyGvaKga`dlzFTwY53ZYTLJ8&qn{59>%2~B>Y9;wCTS*5pbc7&%Uk7Y)>fNh znO=$eay&yDs?ZB)luy{7{;v84 z-IfBZ&`K&UD)nf9E-ujv@S!22bM{yiIi_|g7%pj}2(5i9rLD}(No9NJF9vLL0vojk z2Khi5IL3^ctXYB=oM&?miT~Yp`-c00hh+M#sHDZpb8^FKctfJJXdj9dO4smBMk-`{+~_7Hc96kF_(&4?dAfXN3|- zO6SIu-VS@{szI^qgci(<*k-fIK%=s5jOXj2v4UW0dwQ638R-Ye$f@AGreog}`regfKvpes!i|3WBNR?}SwSQLJPhGAc zPS^6w5`xF2B_P<7Lnn9s6#4P)G)#U#Zs1#PZ>IYW-afHdT{N{q??B(}%4>py-?4Aj zD{C_xSH~9J?Rtx?9(hD}o=HSwLSV&Le4@`%eQk5m(N z)b%lkSBn^^v#Rze4B@zjwB*Nct}eg%!Pz=gG~waCpbLfEjD;t&gQIyVNfI6AhQhie zLZHM0K$nQ1Uvo2GF-lEM)ar;LMbD;)0`a%~d3ODY$j`1QjJU$YU7X@h%^`t- zW>eSlSC8qGeiiw>nBrenVXke63Sl$bK<*)J{%+pHf?uzEap2UvOdRMXe zNd53>egn7BGS`BXDc<~IbNe>?dB$?h7Zjja33~;8jn~&!VZYIC*6XbWMTA@K>JQM>7J)MaQJRa5HGx>)-eeUiS zM{}pg3xlI!b()iV&~p=?Uw|}~<<(li>gt)p3ho(6pG!=hC1v48_XFqBluV+Q;^>MW z2F4hWtp8OwWaHa2c(Y`j{+Y7vluhaApz4@S_47Dt+HTMEZ5(rIXGs(G5z5WWLv!4Q z_vUjY9`27!a&V=SF={Tj5>1?> z;6`$km=w{bP*jxDuf+ulS7}?X%=R{_hR?&mQmo3AHU(i9b=N|{t>1ZhfK(po4kUUt zd%>6YhPOho5eJ#3q->~utY5c+W+fm*9<}k)2=c9a>Rp>hG6e~8$1ei_t~X$2KqXH> zLnA#D#(a?*0hTHJysvn7sQ)^2o3o5O<69?!NvOu#ttO_y%|+y*Qje)HvVDVrN|W3h zk=!A^6VVfZY4IQ{ESFY|1gJ)=nV*DUF{oGGIX|LA^?Btsg0)`kwr_Nd_bYogA=h)R zoLRvII48ouGpiZTK0vr_vsB3BQF4@D=#WvLF~9+<4vSI36a}JThNZ*-<}y6YTUeW; z0Mub_zF)Cax?>BhPrYq>vq>vb^!!G<)CtbXIQi^=KeWeg|Fc_f8a-}&IMTr&s`{!= zV6jt4ZQAg<&nK_49MD;Raos2Jz-m`}WTO&(;#5uQZkWc`+TA!K4vQ+0v7hGq&ZXPr zKy`F{WxqdZ6Yf?adA0p)=juvMHUi5zO~Hi>; z{EZX=+2ywxS#0Z7F(F;DP-YM$L zRX|`#C>!uPchnQ{{}#E4)q%za2Sm7VW1|zb(kD-Yza-{=SPZr1%z;#*+y#Ie1TsAu zDysKJbCU(KRNS0#@u$-o`iY48$%u@udnkhD0f9AwMc7< z)PS(1A0|3FI;qIsuMb$LNj9wJ1=982? zVN>$WD%zOobn5tV&O9j0_=I{cnB!aI4R#OJV(w z0S%Op`eRX5b-W~)=+1;n!!tWaX?I7#$&!l}B>yP8lMq${6v!f-9P6o#^5TdncyBy! z6%$me-T^HUbBO9c1yhgb3iVP=Ct(Tae>r^NWS^1LptJ|k0n9v3Z$A!!>J>ph^5 zwr{INj2dM!QeYuElH@x~=?owX(6f#h>7s;zy*gbbv7b=3!vW_^&7GPWez-o2Q!Nmm zy9=tsaTFm?UFxM}J534W(a?#hXnPn+k=N|^hlGyE^1&g2$;YJ3sXlpBJ6(3JXsw4V zmZVuf*fje#n`Qbz1@CDEV1GMcT%GoO@f{H(^{EGBq`Oa=KRX|m@quw!CN+jqg^#>0 zR1LU018c(tj7d#4S`Shp$j3>pl_B91dU!1kN*XJ^fXgkcV~kp`(XD)`o0=+GtSd+B z?U;wz2!su>2g^=OE7}I_7WO#}^J>iX$PKSC2 zK(FOt@W*PA3yDNBQq?88J+JMGgz{65rC6xM3%gE3r7G~y4&dFp|5!ViVcfa-{&Ncn z1>qWdUqOLPSQse;*&7<=7JDR17>DqBC6O{&_1|wwsly3s>(MVV4ntXeb5H=V*}td+O&+PYKvaTQ&a+^qSiMb7K4s>pbEHryRy=X$?A z3;^8Lf^hVd*w>w=4dxq!U+>57fK`g0YH^oARZpY_)8LhWRv(3c0O6PM%#Soo;&Ydi`u=UVjMxQXt z{L$%wN%b)NQGd(@J2O|(dy{KY#=;wYv;!{@y84V!C@$tUeRD3cVkw%inRRHZg`N;0pnFA(OVy>^@ZRmJM&q(kU*uDWK#s}P4NRE7?VCAR z_-1!v0374cPxh*LRVdb8oSu17VH3tbhrH$LvaDw7&{BZ33>_0KmU;lvd%R%*FxQzA z5eM&~kuJ>Ew(AO7jR;0MW8;UX)2*W;tAPn!_QsDI8nflUh*k&WEULG*wrIznR&PNT zmcg-k5hs9ggsB4$-g`lyHa4lG;^ozHXF6qn9kul~#>Q!6d=_BVu15@`!I)6ei}534 z)2uY`D{xJHekVrC#)if;&6UE}+Za**CwN4BArr$$C|!bWeDsz=B#H;IRH=g!@++b@ z-lzLjQq6ETAA#q5cAB}0EVtbLmS@DYw1-O|uAhJIloVME=ZnG)VHgKb?HJlK|A2br zw9cEr?VW|pHd7y*Mi7FNBO~m_Q)MJuK z^V?C*JJY?$^tGn8Hof(sPU^Fnt#6@R%<`Nb`p>` z5=(s**MoHg2=z8giK&(Xf`alb=T<~ew7)GuWMpJ#XvJub=oogQOKT(tn}3e0oHx2& zUn1p-1O(VT@6K!vZ?=bkfk!$EZ6d_9l7m&m#a-1gZ(O=RP!4UsyXrp)SWHd!mAk)> z(0>Bi^0!6B8`+%RqP!xMpGo+BVKq4`ZES|UF7gi`Vg1}?w*)9Euu{8uGGS4Nx<3H2 znmw&6QFpiPuTg+yM&WQbuyNx$#OJ=rFpL6HCjyH{1h@gc0IXbwNR{B4I2^*bzHxHa zPwasR%#c-AE`7Wh)+oG{Sw2<5{?z_t|0l{QlYp%@zV`xZD?`vhrQdOPET^zg>b?U7 z9$uv_axlwpGQ#xC=`19qM&k#r184w{@HBN$ZDR58@Ti@E0IVD`W_HEkQ%Za%I$6YF z70`HUcM*F{K^cwzcHAL^ekuD+-b zn%n{tA#vy?q=K9ra6p)ZP|xf7qbX?&p-k6BfUhJ>hl!mf(BBGQ)kkPtz1e{v-7KHz>L>czaJ06syfCFd8)F%a zvZ6Cdj7RP!%&cd^SnL#%4FNq(F=RAm`c_!oc;W|jWl0ud@_kErNg{S+N_q-msVIEs z`Fb4@a8-=h`_WnNYP}KH>BvSb*X~8uT@6K~XIs>x$i|3$w?4Wj@s*?V_E$@^_Qfrt zus+C6_7Y|zFTa$hCQ0b_CpM!Q$R+37n5Vd@Kn%dD3xfu-f;meM}w zzbl2SDqr=|-gZD5hEPFXLR1zDv`K+RjE4w=Lt_RQ^<{nxji;$rdf5JIVroiDAOKvq5D+lm?Hi1ziDNAEbK771*=W2W5%?Ikv}gU{rxXO<6Q8;!WW%1QW|M0L~SV`J5~)m`s%c)a6&>PB`%`_ned5Unj97;vX<6G9Lp zk$@%bpr9ZgpBDiqKKm%ONKEaT&q&naK8K9Lu$oR_MygF7N&(P`YAHMFrOy?id!R{Qd2WhakOWybd=;!Sy zbW+3!OcslN5oJ}R(WptkJp2&#+><}Y`1iX=srPM7sT1DjDG|KJZ8u^gTuNu*oRe?8 zh&l>6gA=#A__O>$u%gZ~ai5s^;|2n}QU^U0qH22?_Af=DuaG{gRTyM8mn2BG~evZ057pq`gHIqj!HUzuu8R?jv5<-M!xHM zR=;9|&TQeA;UTTTGRVOKqWcP(Jzn?gSH6d^rkAc>HHBp?Ue_W{9O#a+`=gpuOjb6( zg98hzm0Bn!oa3_&=XQzMS(JM#B>BPqSK>$7%rQ+B1HG3rYtR5gA$N< z2=cF7QqBIg~KrRI2fNgZ}%Fg=~cni$f-SpyKW=-$3IJH6{6gez;8#q5EU;IgcfpEzXAkOw?kuCb zGmSmMg&byL@rw$f*!ZHTF6FKJ{<|<9Bv4mKe*B2FwG@95BZ>X5!tl2S$b94ElIIDCKna{eTy~kW z)W6?u8)j_YZ0Vte!RZNP;=)W(;lS5w?~m4;TTcTwHyJfUj?h>5SPo3bVQEkY4qY3leC!rSgM{1d`5NWi#aWLrEFPn=#J+6b-+?S zy8&uzN#Nl#!JT_;JHdWwprYiN?8xOdD{2GsD-_|BLh^U)E#h!0arJe#Z5Kk>)tK;f z70QI8z_bXMPrW|$b$a=NkkH7XeN6GE0f7HVf%UUsK#r|gvLi^hWB><~mCUc)-(ZID zi9MYd5AyMzg+yKDg;Bq4iJsa{2q>HDt>(#9?ji_Z1QP%aR$!yXRZAnXro*+LHS+t0Fe`1qfuIBU zu*5O4U(RVH7%PmB_Jjm=X+K3}Tib{x$)q*%b0YCdBd^<%M{VTA2txDfb0WM&AXOX# zu#u^8)J17<26Uw`;ghS2b}k%;VN2OqyBk!(PdfIODOn8&-d>7%AtCWNHpHGR;S3Sk z7~Hh;;Vs0RC47O9A!I$=gcQ(`l1^v8hbgbWo`>WQ*z{q3NuIFt3qx>1^snRM&ZFE! z)EPoxGqvMjbRVn8REVmLL$Rd0=<`Fsp;uO(zx<-5JmEeyH5DFVXp?+CAQuxI&1q&K zw6ggn_pV4vDH|VST=p!0R4R(%A1o-yxg2edG^Q*ah@- z{)C^Cy<4-|{fZ|SP*2RY@u(i_Dty=Y*wxL8iUHc~?5l&qppHJhDevzJ7budDLUq>x zQcRl2BI(?|;A+?6%FIE4{*DWBg_Osa(eFXaEw|ath0q8%f^--6(k@Ml2M6e$7w4Bn zKdpP`>~Fi~=Vgo2va*e@A{oOj#iw!1DoNhB6<1WHlgeBh#gYA43&69wp7gKv#? zBWy}%m*jv;J35{LG#Qc4D+W2hF*FzgwGs^^j3X~ffiqyl;#^Iibez{bvRn(aGr}}u zlMd}(pbTKC@!AY6I2an&OTH4wr<_P zflzP6R|#-Q#aF5@HwdE+`aT>DEQL;18|DA_v6(v{zsyyTYlu_qP7|k+4%u#}7>W0; zuQFvEla(JF57Gl5pZ3Np3eibj@ucZK&aY)Lvl@0m_VfgXW&1F*vm?8 z#Adn5c`9Uz+uL&y_6UT<#l zfECto@^YZe53uwnI~eA|EvM!T9q)~R<6J#bo3?Z6&rN%3VF+|0eq)dTT$)n>G)+$j zR$l$4CkhIHU!thBR-ZbF6ghISAybUgZ&_A|jn+N$<8x-Cuk@St9NxrRMa6pB%)Cmq;s6Aqz zwQQ2(mBwiXK5-Zbn@mYp7pazmfb(Jax1m0-N?8HaPFr%8ZY zV#nrHS*Xvf{fQCk-uT2I!uw>V@5-SlR@{- z)HE;P7BEVPM6Z1Z^fT+jBMidARt=s4O?7Xqltv^dGv>no0O&h!H?+Y}LQ6{-g9Zn; zrTx{w?isc@wrRFJ^wPgzk3C&K0-Hu~z&>IQ=&dKmzvS-7LDfE}7g<}O72HV4>D1yD z{8EdAkJ(=iJ~zY6_MJzdohOoLcC>axNRMh2)ucG+lA^LH!`P?%>vyJ90#}LJ(_xMS zUbD0VnA!Dsa^%ca8bXYv%SftkxE~GPqpvW(44E;}knjfv24>gRVzS?DZ?A<=Cqipa z&aO$UlbL-Y$VShy(W?6rJ8OrG%%xz4h7mT8HD_Sw$q|f*S0pzD$buXG>^b5Fp1&Um zh%4eM-(*g|S7QW($oaLT5LDHWLuex3jFId6gvt21f$Hl^`AS{PwZ6d?6FW~{+FBI_9La8U4D=MND03?4u z2<^XhIy{cF80Oz88m&jh*EWwME-rv&kT+mKBhSF&GQ9JS=f$-1Q!Qg;eczF&mZ z%!ie7=%gqtlor*HP)N9SxP7Q?eqGNn#|&e=wR0mMJBEi50L8~%X%e3>aTww5)_WA^ zg~SM)8YQjNfKYHzJLhRp=*GS%{RQ*o@88+ed21kY$XpuF`v~}db_f045?+=2W@V9V z1q1<4p<@~q7oaR7pUg!s7n|eaVb2)%g;1jTRsn{um2q^cfsCm*L2q_`+~>? zHApkPZnPrPg)p&h-0!yb(s*3ikb0?l&883R%FJj~BECti4HFwbe_1c=L@}67p|L-m z`*C)cfKz3$WsADchuCRjNXx`1E+qy1@h!P&mUS#OIig`7M0NSA8WH!@4nNSH&Iq`z zsoGP3pWSW7gz<9`C4eO(IytK&lPrL_Hedt12o&~?zQ6DJKN5syQb(m>H!2@3G%r%r z-Okj$4ZnuEcEq)GOt%!|I-n)|@OOq#MFlD)%rVd+@YkIf)&QM8V0{5yoJKlll9R+T zrqzGe<%#9RaqD(2={_}uq72Pm2*aSl0n;(?BUueAl!t2lhD@K0*+E?ji8}%47^Xw2 zv~QwYI%nFVMUYrS1vp=<8sPjbr_vUK_sz(}nkc{YX3H7y@IsU|k{Nbj;JVX!9f5Fw zXg4UL(XSGl7483oK~D7($HXA&?XCPX-vPg-QUE-a{2AEw_j9}h^Gu5iLqQ=H`ziIO zJnurPxhI|QX_Z@hV(mCU>vGp|tkXrzQydu7lz^#{em_VxYypm^3&_UC#%&{vCN?D{ z6foY}5Pp3@!7r)9J>rDk1bq>&s6j)?ZmNSCT~IIT!QqLDf=;Z%va`37EbGV3$I1|S z2_giD+@!R$MawVu!N6zpEr5+A7R0x=_pm$rt;I$%Cxl4Er@Q~DMh3&j~39j$d<$NF)kJ}5!Y{3{;)^3vAt;>5V)x?l2@!DgWjv7Z?T9k_niY9$i+j| zE);zd9mA|fa?oQOQBdS=T&RGmEg9Y42*6)2&tM${P+#3nP<%Slu^Kwu_gK?wYWmUQ zu+JqN84mpY(cU>?6Z<@vZ-3@X3{OsAdg}LE&hzbswS!FQP5#=>hA&$$M6X;v4^ypiAOxcyvWI(o9M8hpl8(J^Igw&vQjqN=vaZ4?7ljJdoXm zQc(aTZ1{9{o`ljz3>-#v22^x(f56FkeYgTSlyoi`m&-E=ce_q?1m$y#>2Ok=V;p#S zj`%I$rsg;lT2)k3^lhwlym}G8lG{N^g^yGcgy*pS{V5`^ej|Ts`ow5Ja0E~^jwQ83$fc8 zcnx4N`u1_0Xevzx(FysJe6Oy4(b5h2wRj-xjO#Vx9Q(~+Q`^p{ zs!Fb7Pcl`y{f!^cG<2Mo%($=2Tf(Olydev}bqoXU^K8U|3N<6ICOgMwz;Eb7WC2*b z-vo7Ja184AuzTnC^!YmKS&Yh z8%*4r6%00g2;9uQ)kQKqm^;RgyJ=^@+=`Z>n^DnEFc%D}Ow<>g)skoUhysdHB)heh z&{Pr&)&a#YIReFkW7=$jJEu_yg=<4Nqx|EQibTr-rbl8PtlNHX;LoTty9-feFJzA9 zuZ=~^u|V%uNUi@C<|^ghygEpYe1)H$vX`3ZcApMP27HqdJ;NetG;8<1vKfv1DlO4b z_|JJ5d`fW#Air?FjHBOCRfeeu*tZ3pQ}j4v@5=tN$1q3vf=nFwqg)@%mW}=BZbfje z&T_Wk`mem*kq(Iq_U=bbZ@o%KI8;e#DH*g~=i)TZ!e^wTo+1{QI*g0Mak3xG@_}tv zW>}(?>beoI!F9Cb8`xB2)H8E?%5y0}tq+V-4O3!&vLzK1qOmNi4|kZA=pAZK7aJ6x z$___Mfk zCmI8df=FnLYx1u%G)M1et3Ic>P|PrWpKGkzPu9uKP5y5Yv>4tFnN?NrofQrMc`Jx0 zG}Bk(i!J7(?ITO7SfDMzUq*^qIx@5kxsz~9KZE=fa;*D&!}!S0sQwUVmhwf}I?k?( z$x{;^WhlpLgenSY#sE$$I)JSv;Do5tVtkGSC8w2LR0Qkevrj@>R3!1~6aL$jkCmi_ zE~~1|R$D~Gm)*RMjR}v}f$|?!z`t*Jr2x>xGiS>zp%NY(_&3JB2)a~er3kx8v>T+O zg4ZM@V7~nPorlW+lu(7#KsQ~SfmR;^!rW8qn4QJBVYb-9+7_|4gyH_;$}z7_G2;cP z@*!ruyEd71$On4Ev%kBaO=MnL-5LI7$5!g=M|ApHD%>rs7@1QtJJiRjAmn*zW+UgD z78wPFzGc@{BwiK&AW$Kr0jxI_W<1mCJ~i=VnT+Masf4?BF&m|IGjVriMdZGDAKk|_>Z!8yinp1Um=5_ z41n#l0Xp7@apa$!Imi8mmX-`EK0goI&~#+|gfPtro*any;KgZ1MK(f8tD;0Jg#;J` zIQ}!10&(mT4Uo!#`lP+K|J0qXFIcy-JP2oIE~=}mtN5`g^X;$v(ZB1>;xwya;xelm-|f7YJ8p-4UM62R+4UbOi7&FC@5g;jy&Em^ zp%|b;tdsv}ckLvqshQ#sYWGU}BR<&)n$kQ3!19e^KbAziCXPwV zBf&ODrOD132GhZ*CLHd4>{>u5GGp|sNuziOg6v+^mza3b)*)K9?@rMsW2ZC%_H~1M zJqvvY0WXgybc}VnV@Ic31Jn_dF-(I0E((=8YN5o$#0)m~0-e|zYOAR7Dp9$NFg4B0 zimK8oiQnEwJQ`39nkYcM{Uk00tNVpXQHHg0KGr_cM^0Ls0cjUq_!Y%Rh82Z-3?YSA zoce9g$i5_KNS+YVnhi{Pj$X6+2~u`G5YSL16HynwvNo1_+A&`W0G6J~(`z93t0)my zASXoNLcaSqGWxIQ*^Chy8tU7S@_J|`Ni~+7;G*$SOM`5Zn7G{z9C<|uzBO zCMze9N#%|s0rQw_kp>VQ9o^n8R`7QGiJ07!L10i|XZJ8s!Gl10(Lz9mB!OMw1#NA! z6q}fS(C0*GVw=pfTO{A)Az++E%T5G_!;hgn-Cv(1=bLVi)2G=YD=hzv304%W_4i}9RySuADyQ&zxfDtAjVNs?Q)kdYjhU6z5xTs!B@MN6;$?IAYn4)V3{*MgovB#~8En#7 zXWCG@fx|sO(?>}Rziq~iyp#LaQ~j%A=~#fG26T(BuAB(xtB6>hBDO+_kzw~0fjnwvvl=IH6 z{wM;vMCU}@qBb_nHm&D}1rK3sU==J`y-cT`XS z-}>@!$dh4*j+%@eMNd!A6GIRNL*n_n=s=@o@@&0Z8Q3}#K#P6Z%ZoW1c{hIL7U}~m zQOU^3^@L+FEKDc*wF8q`90D#YrKI7BY<2^xSL@;D-Q8Vdpu=}t^MfJHSor+=TVBtW zP!t`Y@!k1mb#%MxLlPYw-Dl-6dU#fUW~ngpHb~P@6)9v>3+wz;Fs-XNhmVI#aWF?( zaM$Rt2hH{YaYKq41|~!y?M((R_dCKF(oJxCyU!2j##qoVI$#_JAq>N>ravGhE_shd z)J!ReXJJjznn4JxvbswyRGt%7>~L?)^1-PD^ZwLny5!@K z`JNSrM*k&+v%#V)#T;4-rC@^eb(p;M` zN9j;rzL&oUwwaAc+xNLVz6@^gx?zMcVCPBv_RSyQ2^axmFx`679T07rN22Z6*jEsdv7;3M2~E~#ojwKP-kZ+peQ3%a6*opZ*%H| zax#{g!e%9O6+P;9IDkNFJ`&JSF>5MrtL@A*=VZcEZK3j8n0l>RA1vb}9f9sQ0_Xyv zi^z4!Px9RlvZ=!NaQ>VL$UP~T1}?7pKMy)G{~gAFKX1lB`U;)s z$Kr0dh{>$kU`?Q}+VLIAB=##Ht_a^5LgGC~6_n1t18}01W>#HSy^;?GCv+=XFTyWg z-rTcq7JeXpMIz$EbkJWWI-DHicdxRVXE!aDTXuW6+E=i+dLef{CHW=|R7tdFe;`eW z7cX~T!jFs73`?tfds4KLy+qTv+=!72k2@y(pJ~m<`mIhSD;a)w|LKKUz(#VA9b1># z(VLo2q%aHp!Ok(evhwnSGg4c1jUFD?-@qYRHwu#9^@Pyj0zujF2GME~osEH74DY^o zZT6!8+7j0P-mbR?zDEJh>pBZC1I`9%vetlInn1azxvHb3Ce_q_1`dweWt}sJF_u)2 zGIYbXOm*fE;Gwg2a8MGtbx>^c<#5}XUfkZrg{$@G&ll*9hlDH`06q1N6agqW_OB4P zhSzDfn=O$@JaJx-?8k-x_GaoTndmOV=YuD)=b)SR+LfYT(<&d#&1NUM)?dGrm6Uv!TU_<=(PmtdPupMH zF4POa?pw=^js;i!u9a{3d&mZf2x|#^{koJB%e~)zjn^};%HHn>260Wl^V}OU z(%2>PngjGEG<0;MYA=&+T2d%J1b=t=%&6Q2se!d*A%3D~Ned-sP2909ob|d{y>4+7 zL6a#Q0$H>q@k9NGN7*8JTICC)VZ+Bh&#Ib)ml~4)z3y*7U!iYnG~JKE*kZ)lpW3ck zU}g37*jYX=P3U*cyO|yfWSH34^rwJ~{F2Rmj2DF}MRQ|g?V^reJg?77EBgl~etvE9 z{=4%>w~GceR&OCI$| z<&1$B<_3u`eZ>RQ7VwZFP2qjXuNnRSd@m==SaVcUFe_k*btq~)I-0z0+GqGnmKUI3 z3QcvLBj-0)k6WJTqO(#zme~5T&Ehj}eGQ1tF&wAy*c&xme|4mSF!)tGY7C_0SqV~d za?nMm7`QBUtSGCtw)S3Ht=Ui7pifKSTequg$p9<-!>sh``MXkhe}JVRe-_r2YaV|- z{_7AAvS!N%_K4vBT&Ip9NI+;H7*M{U5S2T<095LX0C}urNW^`VQ-EbJJB-NU>FJq} z><~CJO&cR{0c&z5Uop#|Rp0h_!oRh%^QJLnJ_|^VmG>Ez9|-lx@%)=!)Z773F(a84 zT}6dSp}TRI#j}%tuwM+LK7BiGaJ3IY^iSk-r>E#N;p<7lW;@S-N{a5RNHpH*9G*vt zk=F-*{d9dRTFJdyJNKf!MrWG*SKS6&|NnqM_rye^LGhA>WE_}ya|`W0!cWi7_T5O_ zxY+t%lT<69k(Isot|+6F zy(1%gZ=q}vj2z^E{9BK91vkaMCxwL1k4s z9{(xuy757wej9E5nP=ma4@+$qO>z@gLy-A;r!?QhH~L<&D8UbV);2h(cE`(ZwfQ=u zFt>JLMb|nPR!=_Z#KB*|g5Z52Dftr1&9jlI9ke#fr&OU0AsY%{d{iv({L@@8aFX5t zgz65XxXH5!$gX>~PEV>s^9=&QCZT9$9^k(yw?3a4%Suly{Q_~kf^gyYl; zohUO}e%N&}-v8m)`!mN;;(l@g$WLfpPqAQ;d+nY9kdYQ5jFFm3N=l)8_H0^OT0F+W zK;^VhG=^U<)H?<^GoaIz;B$18s1z6>j%9k9BY2!V=^dRPp^$qIysc^&YwOAHxG@6b zH7i>oqmz-cIb(rY=@1lw_1Dvpv(Q{(jN;wp`Rcq&#a8HyN+#%9gn@o0|Cq7yiPzVi zeX$j=e0`|f#zW4<+vuVNWzcWQ#Xr6BK%f4-5gEFp%4aXA9$_PzIuY>?#`^ar3Yp?U znF%F>r)R<PLOu#W7t+d6jrz58l@fozvBxEp7}^rEX~I#VCs_UTdE_4M$_GVErC zwg+mYy#~@oLB-YI_cT|O?^#%|t*ovFsnxeVvd9a47`?u*(cjBsKQD@T0O>w(Qz*7){Q+rI2I1Ay^!Tk#(teamLa%!4gI79mN3V*7o8f(N2vENVq*R{N zYg)bcKz1mD7iua{O@_=0Y=u;1&d*U;a|MP!iCnpnJV*tAf18QQf1u@m?c@ZlS+|<4 zu=->GT_KY5=Hlhu>{)|~RayoH9BF~=^I-eidV0K#`wJLWond)_i5D5^cftAX!$*oQ zdFS8Vc=YHz!uYJ?C-GiCG&Su_N_j#3UZ(dt1qs1l*vu3!slJKgg7xQARX`$LrlobS9Gv@Vs_$K6hw{b$^Kv?>E@dmP=sq$Pv!duW=!H1{kV+q- zDJE>)6ICpqp0lG@t<0TSsu(WwX>VuK0f5~@BW8lqBi)PWY>YMIf$g_rOZ3K7*=krz zK;Lr>TNh>gcS@c(O9sFfrnTxEl_T2($tI`;#i9Q26vg;N#u?sIKwabZ8n4g7kwB-t zud!D7QD=~*;~h3-(~H~>yb zZ~ek()4$c@FFrKXMu=6eWCkj=6l+N%wUZ4E4R*7Boyje2AMcWQpFCFkahUG*9VHv6 ze7%(Oa9i{bt@>8qD2iw1ow@EQ<%|FPn=n7W|oCRb_7ayLdjCp+xt8QA;Gvhbk>C zoyw|x=|5q#Bhli(h12S%7v-Gj_8hN1K*RLI0clDrr?&juiBO(L%nvvPC6u_>t`{2D;NUUq6}uwmQqY@<*beVw}G zq0#EEMMzn5TKwD$pYO3bF`h}7FanO~szEM8sn1$(yg|kWp_O9Uv#XPlD z+HZtj{h6JDgI&Fx0rbMlc0ZUZ*4jpuRfC~>%H0}Ib%I(g9_9wE%bpc9d zm?YLZap?|rT#j@+cG%B+^1LlYnX2Aj!MV*9p8OeWVqwAL{{BO4$b=ix#gZ`)3T}-Y zkhrUFzXsYPI#XZ~i9 z;Mp!-K*|ZD@HrtZep^yf+JEKy5y9UCm)oad{8zfra2>b_I{3(_HF}9 zmgbtIHxa{0>92F{12%vP-h!8AQWe6ifH>Zhu79I4Rn@D`#ED$TUWA^7Nz~#Y>8JEV2pID*x@5f7m>lJ*)y6@pK`j!@Y#{k+2@mqg5DfB!=r1@bL~=> zCHYqShuKbyb_#L~(EVml;gL{VP<8JodHO+a`H@>{!COp6DbuGPg#E^Eo~`x+PR)0l znngXORlO}F57pZg`L>wTC=VEuIa+S>Ig*ofj#q=4PZJt}v#%iP0zpuNNmHvonCdNb zEFtYov10FN-kl1xbsxA?2`o!9LZbaF zIEbW-Bj3NDpIB1H=|nL#y4~wVv>g_*E{9nKUN5V~_uJ%sHY6I5?IERSH6Oh-U_Z?NgEQ=7HWbqDWXD zZA)%Q@wf)H#umrUO_L^YIsw52M>c0;-ZqG)$9dG+=}=!ssqXN!heiH>RL?V4Y=qzljS=`N5T}&eYiUd;%M* z&ij&nr2CPN%&HSpFRPm9qOtkCEsy8lc}^1=FYoud5ADI%;Ezq6WK#Id;zDU}-=8J1 zPWbX(Wn|9yrOf65m!VZlWRX5M>|p(3 zN(klmdgm_Bi9(cQL{;O&e?labO_wt#Hy6mGpXDUpnBo}Tt*_F&tZ)V$3lquc4M+98 zGomZh0Vz&{5%NeINH@e4Uj_z@__^l(T@v$`jAdE4N5>zR#|J7>UiZ*YatVrSrxSL4 zURl!<97}pa;mUednaR4J`w5qrX2a_QGw|tMRJzy=+(EdDcDQKy=h-P#W@v`5i*UZp z$vLH*E^kl@(-q3f$~5)#yfBUVr6$fky;*c+baYgr#MFoqgWhZ%0kWRjVbEK;G~{_5 z6^@&?WTu+7^8XcA|J&%ul7i?3w)&IgN&2JlnqS6AdPymE_zl%9$~=nsHiLV{Ep)zT z$Lqc~yz%Sz&i{Nv2s2-|HOf!&g2pJ)sx&lvmM9*Lija`TMSv$zUX8Jf+6jwiQ&b}A zDiw_N40-&uyPFQ2{-fS!%s0Z{V7!>q=7INhEQ(zV!= zeN4k>`;_$sio7n=64Gq(cankcNh*5FNX+A(Avvo?`kcFGmNCFV=mPto$>G;b?nou( z61I+`Gxj^Dr&1c~B)~P(QyQAw2YyY^i7LhY*3;7hQZtie5K-ZVH-m7ZTUHH~S<3tXdG(%NBq+w3Wum0@*`Q1}72|*1K z9@q1Uc8svsk0)EA^o@<@^BMpq)4$v`wlyKEZAxSv4FemDJ?f=4)sk0%K)qz+;(mX=rX_-D!ZhV;U_4JI=JS5(PnMx$1nal@ zN=KvezmRdqvA@GsDKxHu8t0Mm@%5WjA95bMLwY&yE9xgvzpIT^9W`h6`9IxLb~+0) z&u+=cOw!f*_MQH(RO>IEuG<|Pc#Qi=pK{Jz(`h7j+8r#cf6wca+YO4g9hUt)IkK;5qF%vAWFJVOI1_FZ)TjC!1nA4ea^V zE?E-@J>TH_UI6iZbM1>?4)L!av#{L(l&ojc_QX7q<9(1tG!JUUaIwm^Tc#4|bbZK} zYinNO$~a^Yf79-3T?Ley+1;P~{_u-yj%+eP?}VqcO!FRFC-cbR*kVcdaQD!CraJ{^ ze~iP3!PeLxjqT~hTV0=ndh?Sror>3K5CZ#6k1Ir~24eI4qke zTD+^bsRe`(!C$Uyy!8W*#F%w!r4JA4uVlk44WuVH10%0^xdyiCU{SGVP~*P2D-$Z~ zR)VMebzEOF$@*0G36&>~1l1q6@wj(p**KW7v9X0pZsM;r=3M1^^yD7R_IR-jR)1w{ zYisfj(t`GTS17c??cCPWIq`AKZ4T4&c&o-h4^3!e2<0>mZ|kves{|B3XvcALp$3Wh zOifLV`U{$;oE4Qt04$jI4Yzas`k2F{Wcp&H>o{e*P2^LJch`*`Xkkl5#NAINyvB&u z8PgeOrbnVYyJlv%{v_8EFU~`Y6`0pJ^S5844dcOF*yC8Xlr`Uc2SqU|+3;Fpl@*?y zJBB6XFZJl(b(z0dF(wTm%;(>c*3}Zz&+A{Nt?XE)RYH{s)AI94osXS{)g2+-2LY(8*LakI+rPpKX-f=n1yVBJ*l2UQC4pU>W+_8AGjxg|)mb z=z6fnJZlb?;-{Pa%jrk+bb|{w)Lq z<>255<#)8vlS^|reN3}^YprQ2+xIIcJG<%_U0C-5*%jp{ZyGMWDmXlk=C$>4s2AS? zpH=Cd(%rjwaeB|<*$dIG_@yt|tXI9N=Q@6x?Md*2b&6wtKBIhEWd*KAl|qt$u#Or= zB0(6IWB$!j&PO4UHHmbFZ6nIhaEQd{E?;gma!veRL2~n(p0BV+EAf=F(1A&SNPJst zaqrePTImoSh>Qis@!_UN1mlZm4x+OMESsa%_}vpPUad%@Um5oCliXz@Q&^-QAi`j} zDWrbsDzW;tt4KZ2AYOgN!>#f(N4g9SDWTNP%Ow3Slh;k|u4ZVT0?L zy-UKVV;JA{upqM1EOXlvXd1p&>zI8JoO*rvL1=+#zy$7g< z<;CfxBOz5^i9I!+6<9k3qkaaC*wL-!vBotZ9WiqE@)E0iEPogcg=BklnudmW+}o;} znj`Rj+w%whybD4Oq7?o*ltN9ns8shOgGyI)GVXILgfek&i(p&e*X;G;A>$BNXV0rR zNuq!MG7RPKw7aUz${bq6f#iJp+E`l{yIPGfn`0~{rVg1X^6m*yeLAkx^p)BkkshF6UorA-DDEOZe z7CP_`=w(WC7Kk|;TGWS)%EH2u5X|AHc>Us*PVBW1njSosGB?QxPGSF3Z|Ewtv_4Q* z=}uwAR6xh)9L3@JdV*7*#we&8=WJLkLz#;oGT;uA#aDi4`dAe7sc-#lMjGnVJ38bf ztS9e=FppCQ%v+_c`juTn7*6W~oa^3RK^>42ARi&ZR0*Qc;ef99BDWKhg(FjK@pfC^ zyv}*1!7RGQ!6#7Kjg5<&lPq+vpXIW#hRUQgmZKzW`<7#`9b?x!H(|F5z2m2x5{0_a zp1EGyG;V)0D?n7Y;u zf&WG+q1bhAj5^!w+l`wY+XGOroP|&0)L)8*ket9|6P9Wt=_a%;>5wpiT4UzBN2K zJ$H+NTg888K)3qY@fv7Q+s$z2%W1%_FXix`m-e_DMSV7Wb4{m;rB{afx+}sLc6^!2 z$7ecd3(BY@68DX;|D;U6IHFldF;&n6ME7-~;*hnsU7onkSx`Tspry_;@|fUoMCc89 z?72yHjhsg~DJdzeMm?ThZte&~Lv1IW$27Hv0*vsFh>J^c)=%v#^)h;zwdbwoF^jrg3|qr4mfZap0h(l9Eh zgj%@q_;Ctb;WmV*(9hc*BFW~XeGNsIG^t_bOtI6p#gh^dnnlH{Y_^k?Z(B&TM{6H$ zEqx0fw%^DSb1y67Bc2R-Td|AsWJh#OQN!5KEBl^o+x-(j{1aEVllFVF1m7w~Ls{J< z2?mk{E-o$I$J#Cz1Qfgc2T{~7KFBPwJ~evNqQ&{p z_m~wQwkOUY{Ev+dySu}OEexI+BXyyuS&Wrhora7^F!jVSH<1psc!PrOr+T3aok`s zso1xfs@~IK!+=I}0f%*SCcBpqQ!F)b6lOH=r4*(WVdP$F`7~PN)11C-w8%mH>XWaW zjgXvkahurtG)6D+6K_Ie+owcxY5#O$%_Px*b0h5WQffv9K4>R|DWTS%$_Lc(YIu?2 zid4<`Luj$~uY_C^tH}Tt7dLm9RD92C0W#?4qnT5cyIo->D)FEfP33g&%-eV0qukO% zvDhwOPVov?mHd-er|9W+4tt~hF3UGSEX8?lMct;e7L|Qc8PL|~^e~ZiGK7x|lF-+I zXM~WH^w|}af=Vd)HqEBk|STiQHB1<%9fUly1Ea}yJ=(RDSGc> zR{B2C;(!WVQw+emFuPrBdV+Q?oXW+w?hW;k!`RYXoXx`f`t8|;>t^f^L+<}k>Sl>q z(W}tJPTa_{<6EvKy2u;!BJwabA)zMOOFgr&g{V&?)3JrFtLB+1Qbrql9RuyyG4sX! z&5`DA1?|;_{8tS4bz;&t1s8$%eqKgi)>oFh&WzoV`YUR4D5u`3MCq8J2+8l+e{l1J zP_LbQC7dO=_l-c@(AdaGBBhgU@S8M_?x5MH6M_bgZ$56}_&o%KV8mP&?rw`A>QDK+ z&UK$LO|6u)c=C<;f*qTcavP?(wx=IetPRFM+1Nk5T)F26;V9aFTk`^1CLG|P^_H3ld@WQjG+28H?_!iinpYW z3Yw0}MVBiErnF3@U+>Nx=}vrFTE|_dM!dUd ztr_I!u+YovKW*#8cg zGDh-DDZKxD_;O5I(#~bWGZ>s$BX7fnKIR07hI2-y{C({f{n*t15(&Rw*4j zK|~`Xr#bWW{c@FKQ=Dsvdz)#dXA0?r67P6cM_g}s*YeDhZmq!*Yx~Fbw7*$QHx&09 zM1&}edDjxl(XqeMBW+f8ndhViRIPsgLa?(`Y5V748YH^0$CMp7Ig}zIDY}Jg? z%4JRuKOdMmSP8~C&q2l6|JAZSIH{PeB{E=MN44CgktEXVbuywW_*Uu(W9@}TrRC~s z!Ma$1OVtDTmSZ3qzL11`&}(zkUWsq1-84=oL~{APv~x-yjZ%m#R!Q`>E6Ym0|>`g7H~xlDU1>IPf=Ux-jQ`*UAQ6!8j<8q37f@DF?AC?!B)K5!Go;~ws$oaxhlpyy(frXsa@E87#nNALewR!TGz3-RCs)?kD z_Zr`BU;Ldg2oarxg`~Z7|2j5tMCB74q8wY71Ooy~x{3l${Zf@(F6X6?(Xa6pgm3gJ zC~WzhW#{q_qs=?zhF_|eJJ?TlcWHM0R-b&ig)Sa(ThPiA#hsQY6b9NEDXW}1=5V%T zF#5V_7p;HaV$amDv1Xu&38`9%)yciSjBwoFygxP}y(%hd)<=F@WvSR?4EYSz)9K>2 zFG-tK7P4-7bx~MFBng_#(WT5-_}k?@VB?25f}q7{3o zJ>?43+>_4SJT?PtMiPHm--}nil-=2sn#3lA^_}L|9NuZsxjGXPKN_W!SvP#vk$3ZddtvuP$K`Folq*%mNj7N*9{1w=zez~QQRxwVYee-4}|fced3bnf&L?3 zYC!>Y`|%(<*Mm4kl@IiBlGW<_ypI<5#5>M{JbXz~8vE@j%#sBi&0CYehKlBQFBUkc zu0r3zT&`|&I6^|02Q_qk1;%GCT)5zu-O%?3jcH9!gCg%$WFzS&w2$|$Qeo%zZ_g=> zOc`22O}{n&rQg2Qk#y_7fX9(sPMab6t*|3DKh_%qz4FD@_gS7l|AYY-yn&kB6zShT z2;ASDmJ}@9=$p*hKmQ~V-m8naHAUd}5`gHV7mA>Lyrn;65dBt2ke`Id!3mY+OWpay z>cRc#wLpSt1-471{*kZ$(02d3Fn`$X|7Q!sZu%nXv61u6h$q1)V34Lp&5y0-d-Dur zfvAU88T#QMP%^Tb@1Q~Rlcd>qymM%Bf2Jo}TiY;tgHYcJ4Vt{XJ5v?JN^R!jlyY_b zGMp)H#a_iR8Y<%sX3>b|U^x1d^JwUgNrA)4hr_Vz79<0J>+jt}qc{$6ab>OF8Ewcs zhTTAX9l*jwnB#WQ3s|y%;QRI_J1`3jyXbIWf2CRKQZen6N%0oyI7Nt(>b1%ZczV%H zO6ttF+xWtNJU=utIpfnU&EzvuH2j@`XWrhk70cFx^LtMd+KjTHG)y1*w7}JQOOCS^ zhMc{b+cf-YeEMHvjTW&e&4hsODbe5mQ6VBNGQMa)59p0l3@dd@bB_lWO(#aY2oR0| z4f7=y%>X^)R_uE`a#Z;`Jub=9+TjU2q6Lm<)z41Lf^t1bO1OP^YKrr(+t7t% zTxCcw?GEGdC{O_{PyAk}#H&F~ubrdLI`Q@)?a>dh36r(m+j)j62LCK-@EZk?B!_Vu z*54Y7BTvb_1SCJz@Zs+NpoadRUB-Q4-GLnoQRD8)L`4(fhFZ|i)drxHa(@!lky>#- z|E0`+b*uGoul0Dm^X`VsQCuUGj=KhZQ=>c`FiS#turTVU3bH`E5K;JU;Dza*Ul0}T zDYGVS^$kJKKI#AAJcxki!u2Bg^L5;BfCtIAb}f1Tk17%sO%7tB6f?zhe^%-AR_+r| zhfwEXxrJnj6>}clK`5ZFI`6D0mf9FoIxSV1SDzHCets!hQQJ1gej?27+fy>Fp)&hy z$Q)0BNIA`EQ~StDaVONy6C>7J$Q);G#(DyZc71V(E>X=K!f$f?)8Ki`myQ6b}Vu$V9T$>IG4 z+rna0iR(VY+by=VgNlh?0pOT(1cLB?}C9yZhG(DU{A%47f z5g!dR%Q<)NEkh506A1Q2K%~xGAGDH1oQbo}Yi;BlkH2!Ji%ZKUjwqy9PDa_*0&#s# zE425W1e{KK4+!s@U5W;DZ=Ob$O6CDtsYBE@d>MsNQDQZHsshL*MnOARB0!9J6$V)~ z=(aIv-1UV7I!RDM3$lxq{y2McdV)9Ik%g9zc* z5TWperBH4E1`oCAx|+sm82M~h5FVW+Z8G zuxonq-&GLb!+n&fCusIFOm3!Rzz!C@NI=PDb|$Zhxs>U{db+93V(LJm3?_14;`GYh z#flh=tot;xz}T*pk6(iV9M9KvgZrQJ zUzUj%9;qN#YX<%sdltiGrz1vJkUbGa-l4mZz-7r30+B8JLZlk|>f^7;X$K-?YQzkR z7k^lT$Vz`zJyI6+$nsTa2v;)S9uLuj!RKvpBBUsHrPq?MuG~KW7BGv_bxC9aM3`P> zoUk3T*V}kiT|60Gs`i|{*+jfpsJrV-p6ReaQ)6d;$-qxXIhMygg(dWTMz9m|M*4*A z2!D7RJi9(<+l}Z@Km6ePQk@6TOE)wL#(qk0ZM&0S4Ow5BZy;Cqo zX#mnzZAsEGOft>1Q;SR49<&vy1U}UWY=z6yH6Pl8O zY>6Q^UDs1J{GQ^G=?aQNAA55fbbg4oQTrVUZ0!YDObG|qM+;bW9nymQ?d$rpYb=K= zEJ|H@1|0>=IY-fe$a!P}wO+huhv&(d-GND)qRDCZ7l9DxhuKD`esByg<74{~*|}4o zmmN#Pq!d9m*9h%Wl+}Gs3Z~}N7#9BhwGML_2&A*9v;8$mAgKJ!1<4rt9|w-=WIe2v zphTjr_{LNl-6FHY#$4EV5*7=%sz;UzfgES|5WS`%T@ihzW_`f2wI&HhQQam`gpd46 z`&Ln5ytXF!s97*x>d`t6OfxGha`-EhnS1SaQnR#LmkzG>)lZ0`HwaC!vs~=L=6B~0 zv0dOlkIkeaO)Fqz)KwvV?T@moFmZU83HPS?qKE-HqXW9#(s&^S+0viBmeCNnvoRl0 zsEWutHnEowM5;Nqt_hk76!~)7Z}8lLLPGSmQ{~fBgusE+ZHZJ0P_RE2;tHK_@+}1t zf9Z^#3^$Y>o97P@XJQ`*X3$-QWZm=I(xMZ0p_8^TXLoZ^BVj8@LA_z9={4xRy_Kdh z=)lapZ}RjgEIW^wZzx`2`s4`4c$Ib{L(=Q4gZ_}kk?SBsiG~#Qj)CQA-rcUf{vqJp zJ45l5H)pN{vsreBB}F?tK9h{-yYO+^1kl$`5f(>w6j9>gU=SX+-vfuk-GQ`za4+G~ z+GJ~t#`x4UR_#^o+DG4q28=PUqJZ%aP{S)G5^%}FW8Xl>VHHl zwTQYy*(!MGr0O`Ie@|YX6dD+Pg*`EAUSGI73e8I|*Jwm3oaaz|FYb|vb;F}9xe{-I zkqqU}vt4Ppq}NekT3%`uwf&92l^l3XeAfgTUG{fs5X-q6$9`fD#j(7vB{xvy`v3bAbzEsnoj()C+@iwOQi29E{LQ z=}hUmx7-=nhyaG}T7*5lvl1=m`F)YT-D;5CoO^W)U{p3=B4RDYJD;b@@dF~)JmWr( zQQ3}ZQ0v3)HwtE0Biotc^al0eGPU{l05SRRN{D7aR(dWOj;O!*doXiw5^l4211PAD z_^S@xN?n zNm23|URIR<+K6NwkOXDSeBOV)eM*~UzdFHVFat$2)WtIy!W*MMrH zg84iBy1YL%*I(PFsfQlED4}Eh>w%GZBZR9JWnO!_Qy%eS<{%!kZ^|s52rmLsggAg4 z?VEIvh~S?y<4#4wie@C1@bwq`bL5Z-QmNOg`|=HQrGY-@uyr^0*gwyUoU2#TV3K2A z@csMSe_hyvgcU=pDw^lSKf%9DAsUd*zC@TWcFPVAb`d{esykDI(&~#a;v5;{AApIf z1*eE{WwJSf4Ds%}Uh>Q&LtaAZa}@7&n0*yGeu9FXUR+#U3%t638D{k|t|sY4oF?OL z2;Y$yHdqk6xh;T}a`xZt%Y`6*EO31J00D?7)bp;&t9)>>_dRo+8H%xv)HbEWt{_1bl(D>kG*vaU zNaU0`xkkcBnE($RI#J(x4^@@!Vz5I;At2Pi$uo$frV$hOtBa$SeD$K<8^H?L$+F(* zcHQqz8)iVc^fS&w%#EASC+xj3ll+JNi^;#}4t84i!|j;E=dI2QmP%&ye$Oa)tn`uG zTQ+!nP`;}C?w-1VRVRspYCz|ms&}g5@z5efK@xPoHJV==LjT6#j;YL<`ze+HwAX=S zAL6$p9{IXhMkk3$)vQy(lPnG4nVXgr1e(a!Z;QL#L7HgQa3Q1`9yR((_nH`maj*ci&YKUo?9TTlXA$eC z&ac6{$e~xXulW%0ugQ?UyIXfBz9ge$CMB8jOOx$hgh|PJ+O43zh^`LrJXH#Q2P;*x zDvSWpHZ_Oo1SLWyg^8bD@DJgdk=CrcRFql4$^C1d=LF3wQQ>hkX~Y>S#BoWum|%ck zaj>4LmkvOx66fVK)bNtKUJ}mY43gitZ034ij3_kTjc-Ay?zv{ua@!vd1rBKhCrU{$ zV4%8gH)+VjHTJHx@?FClO&DhnUyyd)kyZW?&HYpzvC&}CW=;=$J32Z?-?GcQAMEee z$x8E+AmkZT8hfK;sm8kF2K$d4$Q1H<_1!$c7pnoFD2FZvBB_8x=FcG?4)+)JsRQ8>AM;!J+6JC|4R0_?~s8?*^l6*{YUUb8D1swQ*s(9ZuFM8Pg7|k z&g)Q+^|14%95HazDKK51c#2fu$Y-iYaa{G&N@b=3HZX9_Uc&IZ15P<8Hi;fjIQa;a zhd(=7Mu}DZIj=_@!V?AWGop`Tx4#cj@)^+FCFc1G|L8o7E3bwQ*P&nn0GG0pK%yWI zM2>Wgjg7bk(m*Sfp}z~932NDqrVpVYvEd8~u&pY~JHy#+14r8_!sJZ|n(6eumca{= zpE23bO$Na5AdTvuZz07T%BCZZKvN_g81g1&aH&z?vwn$B#ZwIc4kH=I&N#7mV*;eP z=n6wOOQDk7eid|R$n}cE&v3U|aP8utDvCo+_Q+lNTfp-Xf1xSX=4X%(9-0hBJF(X% zFM^jyG^Iuk;?7AmTIqg_&KW8bE9kFW@mo_*eRkX}Vbi`d@ctKr*}z2G&j~Ol{n36y zy6Jwo)pUD$Zq7O)yH)M%&xtNd)T><5&;{vU)oJ?0&VawgLkvQBQ?>LLp}dWT_}d~& zvwDKNN~$)!I`$EV-@K^hzbDyxb}EQh6$Ww|^~lVh>#*|N9%>FsvHg`v(0F5PmtqVp1s#wvq0N5EpK)n}lMo*cAx=y2Et4{DNq8L`F z0b(_7v^;J4VOtlmSB1*sj+Jhxy$(DF5aKH*q^*ckytQe+bEyXBQY7K?M zJ3OPi1CnaEr0?|)%=@JW1CCVA#SGPAHkkerSO{}uEv5OEDvy!qsJyIA;k%qt@*bU$ zbLjE8LgUAEU=#{2`%MA+lgJS0tjocccm^|{kmwzrh8Lrvg=@^0X%Sc!UL?i1l!)7# zjC%?0J$yjsw%e%i;S?l4T2m6+%u^LITa4w$ce^VNtlw#luC@wNLsBN~zV45(?58c9 z5rdQJ5=r&Xw{TL)`RvTPYKh%;fGv$0$fsOHB0`vyPBBbbEfi16;}9D24{E~s&6!5+ zc!t#Ijs=LFeGVfDFIrUPzo42>nnG}{1)zpH!&0_%ZU(=QNo@gdP*Vw{68M*a|`anh} zqLC3~&&O&Y5Kd(srKJRfquFtdP&PYuNpSm))4ZWZ(+ZEx969)L1|r26{|GurOR^w{ z|4y#WQLO$$DflUml+0%MAR=ct27Ra`Fq*vHgU^f7;hIJJCX`i{It-mGpNE$1FQ&5LmW(!?K1%;lTsR z#pRts*WE%RL~odb&D5J0ch7zJ=@~KwBrGUt&xrwJ5>Vxcbwgjds&La)rp4EHnV&<|DXWT9RWUKUgf7%yy#$&`mJbk4>MYXVrE3A?n5^7k zJ;g=?_aB`O8y&jvc-DA&;t0Y zO`>tu)9n{hUk}vie6_`H@*d1|Z)K9K;*B98J%YNpN3*|XxXxoJrt&6{Ut;%mRSLU>k=}it z1HkHXD&?(ZDH08^JLuEcJyNA(1bB0I(-!u&S93=caK9!5XcXJNn*Z#7|BUP_6YyVq z^)n%PA&N^X{>%LUjmd)O>%{Eh)W5#{p}~G`O+FJOp8g$3{O7OTpCNT4imqb+c;NrO w2K>Qq;*tNo%+FBzzhkxkU7Nqx#{H1GH?Jv9L`g>z1^(R;k-C*}L(}8`02fI2F#rGn literal 202336 zcmd43Wmr{f+crw4h)N0~C;|e~-5@31B{4y|k(35eP!Uj(X3{z7ZYCiOlkQZSNjH<) zW3FdC$M>ywiSPckkNpEg7>sd`>%Ol#&-3D=s3JQVT3n_IJ6s#r`6m%3E zEZ{rw)@k-AC>Xb`B_&mDC1oWY?HrvooJ>s3Wz8MTovlsPWuK#<@P~zI8CpHiA{I@k zDPw*D3mVDF`gZ$@=X<%G`rx~%wJY;7`ZH2O=m8+&<1U)?+K***Jk>SE?xXS7xt8E*4X3ka;cl=i8T(;-%zf?cd zkDjALdqz>nvON5&tYhWJ4?)*v-vI*O0T$-)D5fZSRLaE`=+xE@P5a#Van;GSCA?5o z7WKpI`-y!1?#I8(QLaeZN(}=rB+ff=$6uY|{t&WDr1^EPwE{(lJc)3(v_I}E=LeAV zv!{^HS3jmdV6W01zr?_H*_{p-=E*5weB2}5#hVweOgr2Vbnag%J!|`8I!Gm=jNO-b z9=-ZFhRHelXULsot)&^)4@IojWo>t^&1+0V-Y{=Jl!)i%r=ECgh43ND!Y?M4Vy_e0hs8?POdoo}bgXzg z_CK6fuaH`x`EIjVS__U32xMA(l6bZqy10K=J6rUJPHs}n^Ggs*6>&SANU%}jE!R$V zw~Hf$yAC)G=hrs@GJJ~9bZpucd(S)lbt)J#?tm&?{Ud(ObSh0pcB3XklPQ)}BdH^| z3`BpB#D;Z=H^1}TBw;rulKMhJXzinKoSuzGv1XOJmT8rGI=Zkq-@zUY=QeaA%ew0mO9~qn*UqUCuyg-)3utw`7_`dw^41l$AG$w$TDyt|&fTkx=UeNj z|3mgbMA=L5yMG_ESt#CK`*uCu?zs3tib^@g1$4 z*!{=v?d!bWZ5?H~`zeS_6$Q4t?AZxpzVZS2aElU)wKxUcpAT(R+uR#*UN7yr{p+(e zxV$2M<5O$b(&8(;B2FgKTgy!;PvNy|bq@x!Zdb%uEu$2BbMm)1%3WQOO?zgVyCh@r zUoZ%rysQ);T*yZsIJfHI4X|8n*g?q`&3bdJ{)XYlt9dm3(V*659TkfHJZkun{u>9n zIg|;X2BYiqOE2F8^pSa#>Xd8rYq~f1ZKh*AK!D&k*OjwSQbJ(@zQ#d84Y5YS0KP&6 zJ}7_>6cn@+wEunwt0@Kjzh9&O^+UqjBEY}mC~{H~nr~1yXR&<@M!_f7MYZ{J=K8G8 zk<8n(?U*<$cWAzUc_*fKPj{bz;JE9X*`(bBokDlyyLV>jkG^2wv|-uZ;>Du0TYT$X zHKX+`Z)&5V-vwqy_aq;Wi7UOo%Qh{}yId3_%hos2F#|Z~8=o(!TS*JWUq&4$47J zPnYq{cR}7BCty$Le|bGMG>j(U4QEdBKP(8i6`JwI?VF8+e8!vaD5&fawTZW|{`79* zNHd}T|IAE2kJbzWXX`xR7iR|z-uDQtXMZ|%iq_ZHClV6gCv6o^8ONZYq)eDTK3+M$ z)kiKYkwS$kYB?2X!iBblvnw!s`&PnTE7^elICtZT&*GG5N;@%Gh8HJ_kr-lktw1U3 zzOBAx$6*u0B7yz>R|gc_R;nkr(|FMw=*3a;W7sD0iW(svbDr5T{m&0>wv4Ni@!qdC zjHX%{X%;?G8k8|?GROAM_r0#pR?im{w|zfNO}!)`CVom!FXQLuw=F{M6(LTIv$V9t z-~S5Aa-W)d9@}z>7^k{QTuCVm4{G$%h_Q%>$9 z7U9F*UO6g_k}9G=r3BvfW^C%s4AQ#0ySGrcuMV5t%K5U<(a;dcwc;hT*n>BNP-4A7 z3SRSkXiEM%ryy~Y#9L?lyEQ&Z1e+UF$GX?CqKt9KCA%lt4ZOkM7F zsv-(snVVehFJH9M2|&@UtY8$C}DC!T#e_0YZ3p zP+U6lWRmax@u^cJz=pFD2>sdJXVd~vVug%&5TE}TJ_ZU=P>bEIC;lAL7Q%rSnAw92 zTK+LI1iEA51XL)r-u~lNs2;>LC@bMQVWPBuj7R0nfJ+!sxc@oSRHp->#-rZm@cth| zO^P9~Q(HsD=zqMbi3)gunnkVj{+oRdxtuLvr)*<_ss4P`_ER9l|9^kf8n?D@K4}5d zcq3zD8$Ui0b@%mY0p8!z+8Xu!d&VQl;Oec|HGbfP4rl?UGK@-lc*AL3z(R9#bMrcx z^{CiuIMFoJe5y6xIXLn34ACvbubWP1Co>Xd%9hp(n>N;SFT@YWvnrK%op^qL1&iR z#Y2GeADDCp@CBkz6A$oSR`JKQ8rHqq!?r%WxJVuJOnZro@4+d|CI z-v0hjTE=U{RLhx7XZHxO&}TyCkAb*&XO|nWX_CA2eZt=x_125PFiO7e*L!J=wN8N- zided`=Ixzd<9Ydw-mk)}udgntw9Ki2-(84Cdg6ZRp!E$twqP0KDq(#q&gKWp=7R2@qlUEpd{CytC*v@V9w|_`KH=~{UJF=p(=yV*F6h#NZZNEP2ioA^ z;J4)D(ohA#!iiUo%4#J`J4HDTW?>cliP^w!1t{PHVRF4ka`lD)7zo4swYW&a#Vv84 zj;?#_)g3zz`A_xFpL1IeWx!O#36RU^15RYkBxZWuP0P?n>xp5DNi#j{E;O{Uva`#R zKd6kH6qxc5RWQMA<+g++u51Y-OE+O@AkJnR$hF^a1kqnMVKvDp$EAITP+VFMTWF1p zl&`;DGYGz_cgw2}zC9YVU9S-s7-&CcN{ifI7+T}@j3*-erl`G3hzbIMM0T6d5r&4~ za0cpO744$I8MTiTRAkvGWko6V2H5>z4SBiT`jFehEFLow$7E1p;Kn7$;_i5ell}Jx zhujM7>({S{pM?H7T6x;m)6pzb+Pe7#0yW=SADx|cGxO_8uJn#l(Vz0PnT4t$O|Jsv zp_N>n=+}>a8^)Ugh>JBS{Fa$Pg9LmtDe1BiOniJgJQjjKc6XiV=;*9_V>$MZE!+}O z&w83u?vjfG>%$sov&i}Vmascvw6wI8TMecKav95f0?bo>ho*NPEeLrgZZa0|gKah- z2wxUGqWxn|y<_zVNH&;QDt}JBGaz8I&q0zhe|jPiL7xM~1OBa~$A7%68j5s5Xz=G7 zqT=7jJLNr)p-IVwrT=)DlRvT&@h7;C`uhO$w~fT{fy-Ff-yrRGv)LA|kU{2}f)nlc zXXS5G29g4oy)@n>z9EGCb=j_2U~R3SmX411{vELC!O`hmPVqn8wF!rtmv`fIZ&6-Z zSzAIPaQF#O#6~ixi~%WYd2371Q}SfSB_jNxh*y`}!NI|B_3K5zv7huM@TmYE3(bHr z@Y&7P>-CGC=ZjwKN-i%gb^dgE`sk5FnML0aq5r95U$WSBWJvX;Et;`|KkofM8N!v` z32|uMa5$vUEhoS%>sh0ze;y%uae3rtm2d{_sH&lqgaA8K;RPm`DI=B2?hGTkOu zF@~D?V9S0c^ApxfdS-=_8IR^pU|mbg^M(VM5CW*0 zi#n+N4es8(ix~I69A6PU{#a0DH_1sQ{6>|xXKPTFI;xKe9}mxYtgN|&A=ziB=2FEU z5&8W2^TRyQB}9vI`{FH7Z-PW_j$D8IU+qc6sQ-dn&%4VJyxg#(W< zHYs$h8C0x3Bqy^{Cyy9atZenrD-XDrX-~v0nRa(|4Ssnf8Glmqizs2%qUUgZnB>l# zff?z#l^8JEpoOQ%@q}sL^t5-W%_!3<@rOduxnsoqm^)|Fl)4G!ADq3qp$Ry-%Q-m% z`D!0aHBjBP?Z9y#KRSQc>3j(9UIx4l(SLoRVo3(w3v(>r*f`#r1{X0XCq4zu!(37~ z%R`cgFG_=Ud`1d7-`tP_{}XAdLju5f18Vx&{E@h$qxZ;2nL282O_~~}3LlMVysUG# zchT8^!@<5~sO%xFsoBYXQvzWoE|>>TgoLKgoQE>xSojEYR)HKbvTRZO`{2bsfC0(N zbGtZwy}Ys_m@O6h;e)EW`nTZQRDuIf$#+nQceymEyQdY72nY-x`ua+yE7ry)cDSGA z%;mjV>AgQGt8YG#A_2+IcP*S>-n!h&@W;f&gl*8ue71rYqn3oBOaxT--n#euaEBF- zY^J$v%?{3Y@+zw*%$i3!XXs*Lig0l-ux&rbQEVMgJC3gnrejkOEK-T1y?*_Af1gm* z%dc@Ie0^%Cw!k)`q@(uv*>M$g%cqF5{R0FXBYlhCU>GwuiSVsf0t=E1RgWJ`F+FAU8D;M?Q~263c1+$*1=)Yj z2RnSj)QHdIjNbXo#FQCErDk#ohBYU1-aP(SK3j7Q`eNsk{fBVS^^{(bwFq0nrGRQr z$C(qlR>13<8oq~5jK{GNi`3USpV7~GU>*7qa~GJi4R5}ZMyNPjH2L@;xJCBgQ@X91 zYxD}ru6ymV`KviQmty0x6Jv;y;&3$=Va9+8yHoBol)A*Y^yNpKRQg$r4||=ZgCV%M zgUdb9aZ+6iVO4 zQkGh?7GkLERW+KfH|4iA1LK30K6xQ$sr23{SwaK1%$7b1iR5QS7QVEUcj|1fgn7)& zH9W`&xzk_vy6lr`e5l2&%_WFOVP=n@pKeLpY!d(=G zPsVzZ{^@Bx)o{OPh#omXI>uRWHmVC1tbrISv8y zXiQH@NvWiCDJeMKX%*V3U)&shi-{la&G%;47@5Yr_S(q6w(soc8w~-pr|(=}RIkh; zY^*?kb;kAj@}z2Jd8{8d;AnrjjQYjg+#KGmTbs*VHU0+#1O$-6!f0q*&}c_T$0ha; zoI7jgfuBC%t=`|9zkk(!$jRI1{Vt&4qa#_sZ5$jYKVI)2DdZSv4Nr`(O{@3)p?AAh z71}kH;eo)^B#JwJ+f3itekcj9@PQOLI^zkBmmH{D2ZU{H2yu^ zmT>y_aLmn+YJ!0Gv1eMBie9{bL3&pAQ5GE)JG3{YoA4(5$)`dt6HAi&08|iwGnd2U z{_oSHrKQ!E%;l3>B=_U`uaW> z$co}F<>E!!8N;*<{OBnW`7Q@YxoY+2yq`U;C#+K9T8PTR5-OX-^=pi_BrsXup+?%> zt7&9Jb8@oN?qhiP7!Xbkc{vt=2-*11@AT(N^Tf4987>`5kN@?FzwP=^CJSGea+npD zP*L5sAeEqN=lz*7D7w2Z`}uvJ!!8C{etgyE!yDs)WWSH20RiWD z6O@aU9-p(4seXu)J=xKaR_CqbU#_F}vfJy_)0= zyMH)cY$E9b{_#WV^0Y&2Y`pQ9Cx!rp*P}8)1t-9b=<~)+i_F8DP$WG&XROP~(IV^V z=`k;B{iM{ckdFH{?!(umEjns3e-ds%O} zTSO`)6ivuud-GLacplx8Cu=d#i%h@cvSdaEhHL?N4;E^ATAHBfMA*qa6qm(N+3_2? zrod>lV;K*r3*$^G!JN=c^G^i$ZgZQ;TySrJ-tP^5;WE+$c$^xo=Q%Kp7I}Gv3K14V zh92gM4C!FG9$v$vnm#lA28%FQ)OC7@V|+#U3Jjc=l0u`i#d-NW1%wII9<9LyE~1DE zT3=V^eTj3N_~Dy+ng$l*Y#IGVpPC2sd5ckob>HKnGRw;>m$lRiIn=(!n~n)kh`AQG z|7f#m8F+-p4S!-e&C?FatC%*^04m_(s=5i&ts=uZZA<4K5+0L4&u1f(^>_{DZ;b&r z5N`s|z{QiBYP(&Uz`Vj*`BnxMHRsnK{i?R9Mq{59bc?r&s%ihdvT%&QGL7D7H^Tv*^2C9*v(kvx(^+6g3>j4}GHKelslcE{eg1szM06ULAApo*i~?@4S3%$1y2&b5EO)4=3ch~)2shK2wL-k+=92$0Z%uW=_*m1Dh@2idgY<$fG zo(87Akr4cj&x0q04KK0PHKGi4Gh{*}Iez|>r{L86$iZPD|07~ms$Am&bq$;Cpb)|@Ey{IAOX$mrRu?-l;v`|c!gTYvNM6`R|En5KaMBtQtwEZ``d zx!@)!;F4ZdLn9qA^)f0ZYUA{0>wcq?1g;wo#h_}@z&__FDQP!Wm#7|^VpGnToky&A z9CI|#=`st5sQ5|JKXsouRH!BAvVdF=(KDXSVA@ZWuYbF=18~tm0Pe6`FtGg#DEW<3 zpi9v(Fvy-OEieyXhuYIq5GgWp0hXOOJeo8!1=lTPLrF{k5q^e5gxy5$6a2P0LsvH5nyQ%k z<|9=RMqo%$7g1G3nkcaN6}1>F-MAJOt@`W0gb;Yo04$!S9At~yzLWcC;Qq;yxxZLpp%CCLH+Gl+Ve9+N_9za`1 zM2VrIp`7}&X` zg+^WUly;I{!TY_gAFNl-pdio8o-i8#^q%tAEQbVP+fJ0z8E%7G%6pwHR(eCe3~R!P=(JR#X>-WAOF&Eos- z7srl*>{*vB(ZuE~z-1IfoB9 ztke7Dk)XYWjIdRfia~={MSH^f9Tl)(=C^P6&Jub{PEX$`C5sw<{P@uq0>EQ3RAQi% zm-EIXV6jCX*D7>3zh=QWy&RxnQ{kI&lJp?V76tf#TmX#ZYjAR<{H@Pm+&Rq<*VT=6 zRt(nBvM&fXx5yn)4B;1H)dpl9?E25q&_z8z;3=y1_QvE)uk?+bvt1ZMy?^2MDawC? zFwt=-rG3omVx|#?&l?nc9JX95tlL&sRP3eOI&2R{$G+X!@x+(lfX`Q;@k#^yxcQ0i zB=6?o7mOxRl1JE&{&&hlJ+Q>IgXv_~gi0him#e1cTNhe_{n&*~(S*2OR2LRLug0TB z|BH8@K82DnhtS6b45m$%5mQp6O(C!{4eN94vz3`W59mewwi6%vtUU@N=SXX?@U}bm zJ(4_ zXg!ukwE1+2iQm^6oial+_((r&JGmxh$YEOh>(^CDc&rF72d|6*gR+xyZuRnr+~uDC zRY#4st`oL>l?QCwu1bS#3L8nitQ7`)T-3OM`2g4miKfC?9{X*B<#U8}|D>kC1A5$E z#gA{~IHreLm+}Vku~*FB>I!jkE}3Tu*Sq|B1N1>rReG|Uk!^*{?$a{@@ur;Jcp4gq zFrgivQoJJMTgKe2J}ipuaSaRqO(OZvp?1|-Nb=qu9O%h{{OFKffB#yiB6$St0ns$i%a4>P!p=|?{9m=P{db4fv7WB@%icr+vZ`P^8z1%Z1y9Q}WvBW`jQg2<6ALS0p4PcTwP)p?E-*Kn=tQge2JI z_f>*;1pek(PtU%M=@ulK!FfWu1WVO(w;B5=v|AqhLLl!Ao8v6MeLR}toqY3LNP&if zT&{drpA``~C8yboo>_a*qB0HYcsow9zsX1^iNG=cb%Wv{SKJ$%eeDGefS5-P4=-79 zS8S%2-6N&-?iGLPo;Z)#_BhJUP8y1KSuI};(~1{%&3UDSrPF}>Dsy$!)?5A)g`!jb zkeGNRy^^yscp@%L1jfv50AO8UEpsto7YEpoulgPwdBdx!;QDOYf`)Y(_etCs>U!q$ z19?=t0XE>sl#A+Df@5Q@ulKPP$oivZv^{5aeI|;3SR#+Uw4Ppfp`(-X*NHfVskYE= zrlNr_T``QJqK#!6n@G?VAq@~R&|(+$-&e?Ma;RCqYzwT%uSyB)=5$`X`0$vP)(sGc zt@VJeN?}PNaD%#@_(_zp&abpbr;8EHzN^ai!qNHtN^vu_C8R}6wzfwnR6@%nl-agL z4V7&V|6+a$k?FId5WC^G=@VF$Xq0%S9^7`ZJ#5Ofw0_^k`AMB}j2;n7i=qaUhK$nv{X2#)GF>wpRTv*-m<7=;LAgci^Q6ODR$z3T2{~(0Cy8ceC;6xo`s{ee3%c11Z?7a1F(}&^G230wiA;fBGDWS%J>w)=qae;$ z<1lptkfinf9yW&4(foEI41{b-e!gNVz4)b`yh{0=9&CQP;~om!-el?q06*rA6?>dw z)RCKw;sZ!E?xNt9##nPYg!L&O7jAqmhGF>#P%gw-CaG=Jxca`{AU5Q%9-k(>vBrq(v@M-@G#qm!ni zhxg3S&(~EmjcjFs`6^yXH-KJ+Ra`b+ZaE5#pTl!J zL`Q!&HLtUxzRg%irW<0wvnha{}xP~c6<9my&UzgZp zL0=yG!q(K}Yn&u(!x@Bv4DQmiujCXWI$oNHHtyr$4dnpt+Q#NL7`MmkxxW-0a_|7P zisa2!#dxCaqMId(dzjFrtK)`VYRYI!?S&ueVVMo`?}g2?C;w1FwATe=cciBAI~t zLm$v9z&Ds*2yWZ2;fw3-Nt@{A^1>VVK$b*m{(?sM8{D;;7M}f|!h~NAw;MkzqyX zVNVnwVLSqzJktzFRtaf<*1urWgquyi3rh{qH zQ`N82?b@dRstAl)N(S;NsR;#M$;H&+_#H}Yf+73gR{!trxN0@A+H5_wR`rq2_$6to zW#71G4;7d)5&$J;hJI~>F=}plQ6FcK+4`1ex4Hi(DkDHmeuX+Z+D~WZi-uul#Kdbs z=^ezHI(cl$Nl&l8sE&@03yq)H2u}3$D7eT(XlI=9HBs~05848(NVWz(R~&nxrg9>G zs@U~eKNA}tysS(AdapBr+j;4Q&cLjT`b<%*qN2XvZVY**2(!U&2>erQF6a;NqGs38$XO>@tjM+AutC{y^QE6#8~jBsqb)>lAsT-!hH+jvz2)k&x$KSr^nLX$D8XhS5E2}UJ5j` z7?oRx?<)=5C(Xf0?HwJmE~_y6+kTViUTI;|CqL7d%%RJl56yJTpU(SZI+B= z??~|Edues?fL}_{IS>{%J zPpz|`?0n=EJrGBf0GaplLx1i~4NU)I<7|!#BzK;(M^wK0D$uWHn~F*>CUxqXz+g1N znve@0|9c`h-6b<~p8!9ykG_O&RETMr-ZZ~7RpP6f8kH8j%=iBN`&0E>v|uGMN6xum z=2yoAc+}Au6h$c3r;3@Rd{(oF6HQ&+2}|E!QaFcM;^lFLBhfMk4@Et;XR-ygfTWwN zy!jaWtoDW28rBW9{{REpq-=QJ+KkxP*!tQ#lC$@C(I8nZgIA!8IJCVGF~rv^fE|GY z=7mrCo-bF}*P;2;YuUfm#e49CE%Hc950u?>+Kh*Xw^2TSu0FiPu0an(kCNe9Up15E z?kK@{KZ&{KL7UIyS;cY+PuLva6DVs+yBFLa$H*iFk5M|-r=D-u6|vR+5V*6Z6$r#? z?kazPc@BFhQhVm6lYA&eX($rIJVuMM@=52t=4>OiFDn+G1(O02Z@i#LUm0M+2kLYLC17kGnjI-n}hO<~8+!9b`(Gvln7TEkI=v&il z#GEkdwzTPaCPfh0zne?(ec&;3p8}o=c|qEDPE5k&!jVl8B#QOWbLsBn`Xff1a^j~? zbH728ReF?Sgo_}r!`yZ@cXr4ccQHkfvV$rIFfv*i42?%q+NDg4l`SVM|(!MV$_=% zv-nN9epN}k&si)68Pkq@aoU##b6k41vg$6HMp}8n12}E0)mz^lo?*Z5786z`n7d7~ z40gxrao?OsZ3`s>smBnY>of_`=t8~E&09Y^d zDF(zuwfwP#(Rg7|L^AaGtB^cis6_Y&DR5WM$W(=`+q8X*#pjA?93f$CKq0!>O4ZZv zcgz(Pd$JCw#uzS_7{p$S+M(D1!Yx=ptl>$&MZnqF*?MM-(NMkT0qcDWmGake=?=w# zz?&L8sr~Q|2lSx`6mVI|X>6RUfpvYQl|8(I5PWj{JUmeRXp5I@jWWelfFVjNkwv;v zKkUCE_5~R5w&tAS=gBjGlU=Jr!&qn2^`D>l+;~I%2CA zwu)076;M@GmE?!8_J!qloh(JF*ucjMN_Pm)i`)Fxw~!t8L+{NJ?Kl$_>`5Z~a=U)R zq^c|CRn~^Aj8+%Ox|co1G`KeN`$8pmy|o+s#g)bi8Gs_herV0;q8&H%r_t9vBz z+|c1+?>KRZiGe|;o*eCgfi8I#;X#}g6ddTk@_`|dS!`7x%AyL@6TCG?7moIn19tf1?jpl}+ zp!q&AcC>Kh=F1+3o)m!r+3hL=#k1_U#T#W*KWooSKn<7b}Ye$syva%K@8 zmt2BxuKK2&>Fwhsa_6|x1DMU~746ID#C)JN@c3y3p;_?|FINhA)9&W z;!JAf;eX@F-*x9g8K4SL+x-3kFu#*yV0X9tx69E5kkvRF!+Rypviz)*HA-5L`#f2+C?%&@J6*ZfISaNU zl$pLmJ(7HVoCt`OxdG@#+YPCTLsy_-V^e_?4(3!BDn)FVa~9k)O>_I9Sl|VWafuQJ zOuB{bhsu*|ud2HFMQfUj(OrITo1cD0O`7l|#5C5EU-6#acxk8T@;d)50NqgpmKSPx zr#9#xKL1zdEvEz4H|=RwB>$X}`(zuwyg1)(Pvzp`ahiy`kB>Lf#peOYFRu;7d^b5J zc~BYuqDg@o7C`7B$q_0ss255r)eu`it~IF#yk8RZv&J;l)Ogy~2g6KNQm0JM; zD1EUUK|D%gEHoXD7FN*gFBH1^WTj4iQ&18TPZXO3aWWmaO>XH_QeD_^tTaagaA{&z zS=_4G_;s9^zh7trAZ-o|d~)aR-P8v}82xO_8qdwlVuF)&UNT{ZFbNCSD;#;3e(M{G z7jUsq;g!<|)+`PXJ_G~!{l)+HGmVY$hylLT;_HN?rborChr+Q(t+k(|MKHYN01c72 zwchFy5mm;niD5#!-9+z`%WvfW0*8JxUT2mOpY++lP^Lni%Ln~Y zFuTcHWxsOC^WmX7Vie}5Y~yFY?n#H;&#tcKucBZEBU{6l$YG$|W|w~y2~Dk-ybd!t zgxdCBAhi!s4n7?WkK}5YXWs>8O#g$}Q)$y`X95SB)nKNZCBzt5pPB%9<&nvh zHBT(#b*(yD{WS+m3Ilf2N?$^@n`1*JOSZCY^F#@tqp*)@FN{1wBsHjOekWx!nsE^M zYffsL203z6c+XerKRwUiIo#_7FbgNaWIOV0&69$ZqkLFabGVKccXkFyqqNInr8xb|OmZmZ&`FbV3q4XW{u2qz_F^h4b8-F8Wa$K4*lLju4m zNB|f&Y51Q2@FgSrYo~9ll$<5QtagHBTILXV`9tKwJTTJ~d1X=}p!=-E9%oV{XKi?g z$6OyD81_@ZofkEwn%XeYo=XZ61k{|~@hbO5igS^QiJex$qtn`>m)sWRSz+X<a|d?f1{PTVe(pAFF(AhQ_bQKmbgScr zU)R;PuHOme+}CtEOeDRPln-S1>hpnh098K>2q1WC=rQXCt7t%mbuLR-PDX}Q5EjAE zLURz0Y5Of?SAq?bA_5Xn&;@L-@Yx0@??2oD zc^Y@gQN_f>R%?YT0CW$1RoQBlTHAKFoZ}PuLk4PpF)fmq^_Gt=5Jfgs^$)myv#xo_ zIM1z-f%OPq`lSw`v;0^qp7w1J=yFj%nV1x}X*j^`qVF**V5If<$lUc`Z?StH1BqV&;}S|>!6UBnA)yc%8JW^xm7Reh zCV-+%2_QI37V|eU?FhfrY?v;$j=Y`RrX^Fn9x-ujoBlD=Ex(#3%C--R3dm9SL8vY5Cc}Xp7neNye z#My%@wN%y2HrS(HW~*vCQ3e8v3nqYH0?GvN@nm6H6euw*xZMjWAtDqxRL$;C1tk>W79)6Xt$zWu0R9*#Ud?SdgdY>F(VmY0{U zk2fd5#}0k*Jn}WVnbkefUDM+#j2(P7QP!&fknWY;0bsirGdjpw9wQ;Ie~wzY>NhFJ zt&r?wCt^`fltw7a$5y>Jt5gpD%)obXBO{FjQ z#MlBAd#=JI*?$gsnafe2$5YC2#BK@Ho3mOw3JVN1kDc?*kB_@;xuKRdv(x6f`Qlj+jP) z2|EER@3JG;ZWHMryf!l#{9)u&na!=2QGTL&lza|$2jskeWuL=(adY!b?SyrF6|)uF z-K8_UTcO&UKhETpdq#3YN7YAOP$~2b0{XWp;j>0x0OvdVZ1Xn622i8)xdkih^)TfX6e`?jrz}vC_h%E6{2N=jio^paQR%n>lYFVabV30Pf zKc{MM9tTmt*Sdd)dFS0aMdbp$`FA@D3&mR$O0A*^=Ozr0^k0PR1S=n&&$VmfCsbsb*L*XgKgbh z%E43w8}X(bV-^#Gj_w^>*t*YIhs6=6PBa4w?8j+X*53rKp8wQ~6sfHB6awr5#7{3- z8ldmm@h#S4fMq~h#9p3BXA0JDM06`CMT1(NFusmyXd9LWB%5E5_E~CDDNl8n4 z99eQ+SPJpsvDG`bZ3)vyNS9?zm=yRsrUNzIW+Hv9rK|*ibPpsWYbK3kmw#=KcoAMc z(qS$7iUbs7pVI6uPSbF$gb#GldRJKP+%|mOMuJRDKp@Dv_@>zxo1Btje64_3re4Ss z2$i+ji6)+Or}jHFnRGjP3cTSI8X#;!TvA`l3TPE%cfxbVB}5x7rW#(&dfA>gM}~)6 z&RY`kdmZM8PWel%a+Yb5@hAs!IcU%B5ilu;rF?06ikZ;Ma^3ArlJP*@!l*I0uR(d; zd+rNCyvX5zBrKOu1J9P*tepq%(`VZ%u0Rtt3f^;y%!fwf8~As(9yTF^88d2@xNBaI zwB^4Na7rGU_sO=_{>MuSV(oM}jgWG|-tm6@gP3r22|Wl`tlLuO-7T?d1<}CFctN35 zt=VWc`gnJ&_oGZ_OQ-;Rn!U`LL>mwgiZ?Poo(D+f&ivFnhibs04~S%Qa|&hL2^i() z@wvIV`JrE&c1Hw8LvBU)cj{4ao8@odmo4K^KHeET(9-fz^OJal_u&H{oNFwXgGo(G zBB#3l+MWz{hkBx%?E!Yej&*d6Ro%gxDt_b9(ck}%&;h?-nUY}(^o@G9l z=t2#OqJCEtwpq^==Xc30u_kFFgdaQZ9E)KK+JC`w02)^j7mH%=Agy%@`5^$psx%3v)?-M1I*Y%$42apC(*+MZ|FjK;8-5#7WB z2;@~1s8e%o;V#}s%EHpnbtZ>fn!Z0?fJNMNO#9I_`>@gu|#O(r7I;Dtyn~ zS@cRXdOFkxeKFNFG>)bj{Ae8y+G4P&FUWu(Q#)$Bc}~a8JyL*}83)s$jE(cO7 z`#+VsmsUz~u_}->G^~$24eNYlLRRr`@imp~qbPj3zw`U)hPdf(8&vz0YhNvYne*MD zVlk@M9UKH>!(Kk@N4BUbMHs9`^*Nx6i&Z88mzT6k?Hg_ftAOl1b)QF1nZUxPQ$wJ} zYg`dMq1Ge9&qy^T1elp6xIg4IHPN^M2jsV|{B*J;M{_oxsUTmh$8()mLv6S(Z^gb2 z@_C+SmPP{UW71xKe}MPtla7<(TP8f0s2I>w*GyAg zf_N-?$$igyKw|)0Xa7Dn9~Rw)<-*L&tnaXKcm@ow`mX~swmaoXLYlyU`fZ($qW^3) zf!r(}gE{wRk-+eo@}PMc7W1?E#r2|NI|U7^3}WN$=x8k?ZoZE)?(12WIo2nolY4?L z{G*B$r@>ce$=5YKm+8nqYB8XL#_M!=IINPlbZE^&A*c(A7gL#YXC1$05v9lBWi-j5 zzfG0Kcf36u{k_a43{sk}@80|4(tQE7I!_8p$m9~S~r}L?yu=SX^-Jw=ftS1wHm8|~tstKK<;Y`6J zZUiP=(V@nLszhkJI$gg*ebN`>B^H#l_p`cWP-{MQl}_OU41Iv>K=z)KY-ba%&_Qx? za==XKSHq4G(WE!OaP5I?3vjcxjpRV-stNp6p96TIFe7`QlM18nxlXUf2>HL2I@K9X zlVz3_F5}ufZ^*1x-U?71b6h{K`9^5oJzey2m{bOmR<3E;pJfG9vs4eq%Uw1SO$1+? zLEX3KvXB+*O!aG}x>IE1Tm7zSd$!K)Y6&;&Hr@!B{?vS2&?awUs|prZF!SvmSkR?i zfMB82yk`nBWjn2%ubT=KL7H%C)}rbsm=Os@LmqyaGLd~G+~>TOyFE!WK)oNn`v}6; zqvJV<8Hvg(75H<|=6jb_UY_X~=D_uv&baD#kW@x;sp0~Y^S~<3UU-`pG)7ZEp&Xk3c z)tbL(G*1WqGlbfqCTZ>p7%YT_xon{Fx!zCxBGYBSFCHTb=@L;8xmmpZ1{#6Q1bOXMt@Cq zb;0uNQ_{7|dAliShSa1MqQmtw)5!_aSTRxcNEw&4$OiGLTN)WPHb0z~x$<7n1bqFB zfPf&cc)cU?^cB*a?yW5+zwgXZHv3YF2i-p~@V<1={4uKXZ(hmkMH;6{Q_Uam;+{}4 z7$n^7_1^gJe=$o&H34i&&s4aev%I5S06#m=i)B3KG(h&o2AMU%Ro?1dcJL>CZgq8Vk65cdk`I4KU z`(~_Y?k;ir_8nVst=PCWcsm9Uo9 zr@Osf_LmQ9EzZQ}hJe4dk0`oRSU6-If5FzpYq>VQUwLWHZ1%O6ocX#dW-`l9I5A`K z1@n?H$|sMEM#8=HLEeFc=x;?F!}9W^@d%q%D%F+|I2yWH!cMRBA0AFF&qq@{$Rec_AKDY>c|jMjuVZxvMnd&eSCbQ$p`q|kD)9RI zi@9^4HRg+{n1l5=E_n8&;7Kp^1me7*)Bwe*8Iu>o4GK9BL_o{PiW0Zm~aMy*IRW}x>@L8UL=#rwAQhe zNaT#3zA1-OItIour(uD7py0IFohWR|q+$W|LahChc>OzW3Eg83o)d*sxpL{rG&l8w z5Jj$-AI)F2Ty$T#ry}Erq)wG&D{xHaogkiYg`pqQa4+W(#P?P^O)q71r(ZgAzk0tl zlrdK*aw_At>Hw*W;KPA(6E9L8LUJQSz~g8#{#)?g;%^^;Egl=|+{%HUfIici+Qb9U zvv`=p>3nbJ$#{3UXN(LCMLbu3kwc%1454(I%8#*P@KYMgs=*?kH zz`zajqToE;y<9jEu<-$83qyNft_Ok*|w+k*{jy>I``6!9x8p-*<^fF!F);iz&&!wNvimsrY&tp=-=f`mFXX*8^2=zu@^zjnUVJA z$PXMLF35|P$Ab^Q|C-wg`1iFN5qmg0dp4xiOeZAvzu62RFvEmUQ5k+wS+3MEFrfaK zQCl%{tS_TsLPNQ<-zbM;fsxjUTu?!z!XcbMP>wW1F!#;v#ja}|&rQDPp%V>;*Y~L5 zn;Sr}@YrjT|AzehYg?tD-N%MTY+l2-)6oeR;3(B(Z~0jU&PqvX^}Ra`bWtgpnDb*y z)9JIwqS~p){DnldCRf=c#$Ji<)R)nOwFbytXP6}JcL)gNfx?*5LGMM|p+3ewW2K5w zBB>>ucc_^+)mBR`d0BL+SG^H76=o_t2a0_GrA z3kLos@8c)TWSX6Ov*Jq75x(cWl&|}qd;(5-+x$zOmYdv>x6rreWxf%F#46&tnNu!G-i4~pn|Qg>37mJIjJq}wK1g=FHFkhb`ZsA zBk?*C=}dYghrlHJ|9F*?K&p_rR=c)x$HS+ptD27fv^mH~kXFWhj%Jwq{FND*_I}!L z8M&_b2t1chJ3*_M#X|0wyI&>wd3Xp|MX&nk=z|s~ak9yCSv}6)$>Ck2D4LtDjgLEA zZ7)3rx3oNFg#_Gg2|b2=kiAjD3dizhB5A zkZiukV1fk4UMUk#vl7ks8()c(ApY;)^5eja&3>kT-$APMhEnuSQzN)P__RF;slxS8 z?J~P{eFOxCTPyypt*tI9mF*oP#p8wn zpD}_6eWu_G|M%WR5$ZxG`dUHq@?!YVLK%sP7@zWW_@2dhvP%%#{C~&Gwv{LHhq@d zLGpWM>hAlm-XRW!J3t!XBG*5!F#3rJ6qEyg66K|6n>mp)tw)A<$#d;*T!}p7||M zjRVupZ9-=^+s&Hgm66Bs!j^?cSTI9yvn%Vkf5E`efST(ljfvk3edZz15)piOu+euz zR_=akw!XP}VK`bR6RaO^q4QOK@8vV+>V2Qh?d?!0QeUFaUS}B5*m&eF<7Peg3eQ=s za(Ygkd1>KBX25Z~nyHA*(R}mfO^tufPqRC93hSIR{?D(>6>2Vu3S*}|*4u6qug@lg z`1lB&O<`Dgf{fcsT*L1Sc5Th%OP7R)E(o|SwQ=AO9u8(vl=Bf&@nmu9%mq~XgvHCV zDSUc8DYuw7PCeOG5fLd*jBvt&0kO9@*ydvY*+? z@42*A)6!%`inF(07;=797iDOb;I`l)_Yu`sA2;$r(989Cc7P)`A_Q^m&Wh(_J3w`% z0O^gje3;sNU7iR%0G(hmOq>b7arpdIYWA(ILLbx8V7K?6+SjCHm&O`zWde<~YbUX< zP-~KVaWbuU?yO~*p9xZo!i0zibAWT-U5!AW=#A{M}DC+tfpi7zMD(*X=qpJ+*vGu-h_y zS3<7(we4K`5jZD)GQTGIFe6by!5-*`sXBy zEQDPBaLdK4)zxlMSpxIAx}Lj)1IBL;*QTZ>@>5ShD0j6{85%r4Nfy&L9nKVJ;%otx zTmS)H%`_590PHn+qB}glfB<=X{C5!f&*XtFl=*R+sCZ`1TR*OZ%3uL{z2gPj&yYxj zC63G{=d#F3TK_g$GAuoteyFYJ+t3P< z(nvk}{i3*3Butu?vL7fy6I`!98&!mSb9Qb6L(8&T!N4^1WU7ux1+bVKk0_}F zmz6X>$yHEs1Pev*`+zzz5R>p3Sagzo~G zf?eWs9HCEF+Q=4?jNP7-o9o#)e2-vY=BkH=7DIAd`SCyQJGlPEiNeat7_)t6@2t=j zowPMDMfV-LiCyW?2()H`%>D@gVu5}GK4ZXe4_O!EWV?A=;xx57cT?dgDx5Zt7rv9qY6#MspSsw z5K;aWC|kqftua@VW!$cXI9*YVerE2>R~S2}!b?0Rr-#$#D+D9QF-8PrFi`|eyM91# z0~5_{KV3|`|aAYxBh&}Au`T`|Eb7`sd;kE>x)XZtp#9sFy?S|8KLK#O^rw}^L&jR zDT(*lsKVDV;zwYY@_HeG{ta){ap0dPBp3<8&YT(AGji9Jr`q!Q*Ty`3K7xOWH)H`w z7}0?j2?|irs3`=PIdVt8cgF22YN5kS^V}{6c=-8QF*Ef;R9vd3+xtf`QYj+uKDW#S z8(}QEo}E3{sOobR*|ntW5mNA)@!7AA-C7sC;B%na{S;#`Ai8FTQRN_T%mf(1`ha zMFp~1LzmmkRnwQ2uobTWL#oBHD{8yavl~Pd6u=yTbvc*~N@P&4UUJ%bL3gSg6siyZ z{<(E`JJg@m*Z->887Rppb*nhI$a%R7yf+oJn3`LRo-Z!cknWgnT-oe91U58Kj&we! zAs5$WENkUaE<+#+s$)>2Biz&)-SSAPG#{oA6d-(CV)IV0j<7EO0{O~OQWuFp{E@ll zu$8H1e$4Kn975lZI2Z@QO3s%`EWS_a3l{}ZVb#*UA-P}gmy0nNr6gdlG!Gj5rp06c zRs1!$ZIC;=oOq+S0w9RCC+l;nFhT|$U(-m9R2GG<0WUyzEv2t8DZ?^^`j6WCpGW!k zoB3yuW5&?1Q1d~+scLm&KN9ayC+MEOko#k1bYw8B3A=2}A?HNZt`b#X`9`G~z2_IG zW0oDaD1RctD9KvOw<$YTmyq#GD~Cmc6OBO^ zgue(K9P}(HK0dehKGiPqE@!_4|0kYz_jmgh(Z=&B?eJ$VN6R8T$&6#K_Ji>BgyrPq zq?}0p)e-#DJ+#JBk4%{vMVFF$x^lDr>zUP7Km#9khu`x&*f==!Xm|SP+hFe*cQ!?OH20nA4{K?Ei{yfYI@Pz|T zE+esPzd!}n{p{>q8R>Omfwy6jTvyTpMn`pFhah}(B(#>Jx6;U^_|44B6g5i#(coK5 z%D&hTHkYquWz2W?_t)61Z?R0J{_{ln6DkO!+<6@CCF!coJY}fAU&xA(V{@`!UBiT& z5x>@J#kE($Fv-g+O->n=q5pgtCNP2{y01kdP#7;v)tNI2S_|2aGJML=gm9kFFuDk7 zZj_Tvk4t{SCd8V#Im^gA>` z!=tPeJXdYwww)a^AiK$j5ZlT^B@AJWzo_~$>^~!! zfeA(|VfYD0EI|7{FWX-<`@hd-`uDe99T?#y+scxZ2&P zS4A4_9%CE&1qEnfcse>dj0bE5E`zW*de?MAIjlC;8MjsOh5C!`Tf_;pw1)LHcqC~; zXw|z6wo4KD3L;rksDtOup-fCtUS_z}b@i(tLh?o}P@I_E?;RQ9^YfRh+v}4UeCZnKyc8*^=;%d0KK=fPZsL*d=jWGQPc;+)o+1H)(z6Ch zzkP*2pWiNwPxJ1@5FZMkGuGHgt)t|Y<@!*AF{JuQ6Y-HTn4@s~ac?$VOUx-> zi1^=|8$;=Qc7osU?ivQZUB02WmH)(E=n6IJs1d3LV_|{6w;pQuvNkWxJX!7F8oBp8Uu$vJU49P>I816oofdA^Q>hT_%t#Vb2!?p!y)| z1RI`7kH}9xd_?n&i5r0fnnCk#(h`1fKdlepV2pF3g21drgv4$wg(PVB=T`jpn=^&1 z+$#?1F$CF7`SvaC_&|7<_4+UaM@}vBb#8ah3&Dyq?TKn5?THUZ0r#ly6JcWjoa^Vo z8Aha6MG>N|PtD1RYeUhv65u+&D1XOxNhsvQKhe%m{G$Nguyk^UvYqI``W{ve)4loi z&npxPqMWE?E+ezS_L}B%rgwqpoD;Hf8vR3M7vyDAfgp9Q0Jfa+ZNjTpuP{7MkiXgJ zu(}Q%9Gtg}CuOm);^q(Kld-ZoXau@j>l2}YT4ELf`QEPlqVVKD`>&@e!=N0&%YGHB z;H`Q)q@v z8Cn;brW5z1)qWZALb@v9h zzBd_H1dm7)Z}l7h6~q1aq_UWg`@P-9^^6L+=`iHbijgtRFp!~ zcavQBFk$l;81?meQD$v^Q%Oz0(Y5P}_>Bcoq(JGq5E3ft@4W4=_2@gQ@#_P( z?tf-^@bUevD32Dd70G%8%9liZew-)z*u4AK!@ENOs>(-^-aIFGv@wIu9QX)4YuvUe zA4ga`6$K-ZQr(V3FiLGxaOz3}8aN|}bO#sI)YSE3KBmhRAj_9e78u^lA~gIoi7-tR0<8Ep=ocJrEce1qKFY8PlzT`j$ch-iB~(h-{#nFmq^Vq13!Iw0}s_ z%K?EuU#=~k$BFTR;MHZzufs8IhN2u8pLn~V_w7))s z=Dp4(gJ!R5(I_OHap8)-eOvyl|U^S2E^rm%a1`YAyvtQ#1^3K_+aD+Q`V9-RgSWDtKNB#hJ22wd-b zoxsIu*w{iE8;?TEr@isQ^p_@`th~s6TYyG|CC(EW0=e@LZ6_~LeLA=trii?$W*|>~ zon;U=(+0(()pg2Jl_M!9hXt>*Fdv=8A^KNww<}>H;+m}|Iv(dEOd6++O1tRntls9B zF`U}RX9yZF5Qqk94r$yn|8=4M3JKe3erIf9QBe{)Iz*0+PNO+?K}{vN4k3QvsIRByM-RimqV}(ClOiE z;auOlyktV+O;(1FY|a~>09rofw@c7a6D;9}LLtXU%5{gj>AIJ=G>Z@1s-GTUn3^&e z#t^@in-7!B1Kc;GYYXB3FP#1zUHuiHfg1Y=44x@USP1h zo}JFr_zGn{47nyWie&Jf213Jz`Q=f}oTImNMH5HRKZvk|q5CG`0{bQI%Q)KpF^Axs zg&!8g4na7^@i0R3V|}QM-}7z;wxZd414+{qT)D{pRS`3@)JmT0(fOsico}U__cpK^ zF#rCi5haGcgumlJ6`xo*#z%xWyw3a_n|p#0>u=*ysRT0tvE0( zO*{W$Q32@SJgkfaz~jex%BB)r5pVuZkpBZt_*;>e``mjZP-BZw;qFu<;NrGx6pE#< zF#aIm>#3Cu;*fCjIUX07eL@14H?d;VW)GV#Gbw!B2N9%xek@#|*&tXBqDNZ@-=7ix z^7HGOuHobD(tY#_dPAI1rvb*cJamKJ>a@Y2CwGjK()u~C=Q_OEFDs0dZ9g(av|2FD zARxD{3O|f{5l31qyiUo$P#gI5D-AWA+Ls!xu?R$m<5dZ{6qc2*t8sN>F8i|qA80tt zUN)C?*KzamiZ=5m3#SBn8@HoU>(2su952^tB&diru(F%JOi(-$nGszT@W zz&eVcuZ0+|+C|Z^h@eA{{5Er{CdpbCP?}%9l*U@Yv|?ZYn_v73l>McL`$Y8g7Ul6; zIm1Rm{^@y2-D*Egk4$^^6mM{+dLehOu2z`z!bFlTEqE4}Jy=tMVo&d>*t=YG*!;8D zvwsUs1E-+cPUv!At!*68cjGATDd$%z54Y}Iwaf$lpGuVWzcI*POwT~_5s#Fd7R}0qPx_cBd%UM*peMHEeS49u75Lg2tFF202L8Ap;A+xAqe?5N zr&ajp)<$mQ7w8u&5L0ICPs`tOf1wB^)%6=GW{SwXA-voB7TbLJ2+JM<6gEFBeo(mL zmje%?#=Db#fd^vtED;Miko0h#l3u6HD%GD!4*(Dab~eHhg)mIPY6+9-WKHb3-4SVl zt||51xKH7Ko(__yj_2KIiue#SbHvM?)h&pCCPepZcZzdGLN0%%9LZcOaxw{)DlHWH zb}ztp__8%!)!NwC47MrwNxJ!8*Vbo=lpQ)@(|>^R^q{ExN)VKnSB`Nx)Z1G)EI8eu zyvfW_1%Y8JTF1aEkkvmqF_lHe2X`TQ7ZDjLE|Ps@UsMv*&~WS5mCXunn=>Z3Z5fwQ zmH$<{LSV!ijsZ6pC4Jqn1q-;JUtKSMU&CA%6cu*td^>%|v8yzu5gfVd#ZJZU%9X@n zBn0vTsv&^CYI&f~j3^uGs$i<4wN^$)J3{J{Eq?LvWL3Q_TfeURML5?a7dP1v+4_-M zR(!jTy2JWyKtsA(21bboEeQy`?sYQ{z}cP-^D8M~ zBKmPLGF>fN>!iu}VTT~iF#p^?H+RkX)AJAz+P1@5)y+@r`SS|ESk;&k5s7eQWa2~d zcQv;e2zfor2_zgw?ev-u25eFXdcVTJkfhh^`)o+Y8L)dR;s}ql?fB7`mtuMnFdzFE zlvJ1egz>gWika1a0V6OwYf&wC9*X?o+8v(Cb=L*3o5M51&JLx)&-0>zh-3vnp>!`UE;>B3Dc?I|w@PkDZGHdabW5T_1hn~a>Jqk3 z=CY0o5yju-&R&jtlZuu6Uk~~3)Z6BGw8uW`yn`KG6v=F+(@GmYb22rx$r_W6Yi=j(?Kd#Pz00m?^HIj z)yl*SD`!HdXWIHr-#1i@zEk1P+TXVx{i>2%_=^qA8V(LFbiw=BPnJPVA-b(=jrK^y z3(|+h3U+=!NLv*x?_1&Xxk@KxN@#z+!+IowsB;@~+2NW4KG*ClnV=ms{^-?&H$4jT zVZi2Yr7%Bp*Ko3!APibd{M`sZ9b2piqj~qy@!s4J4;Q~< z?=?XGV2k`@OuCv9Vpm3fmo5M`Ix^z-sa`T+m!O;$DV*!|SCMd?WHB4_l zMz81DE{^K_>Ni?JAt7)?It#J1jcuRn4%AFS8}$%UHLu2x<23|q8Nb5abX@G&ckyDd z9a%ddaDD;%`{1FB^A$RBBqDAIe0r*F78{>yCYBtfc@Yu`+@bS&!YELYO6~W)(r9#J zT^9huMBpP#(ZV9()u%esF3E|0y^i`EKF&6d5OV1Dttl)PhxdR*7K2Faa5Kn zBV&8yRwh?L)RK$u$w|n{jAB)Fb#+yqcCUwvLE~}8`_)>hXflQ06j5P)i~OQp;;uyk zsY1lfg;-v;JAc)y$?ejmmguHLG|+G&0zCq6cXtJo8xh!`Sl?FNH1?od4Ga!;t{CSl zO>Rc=oT|}RzL~ScXf{A`ynKWMB}@mi&oNSw&K$a!&Ez;e9=YJGyUnjj39P!MGL3(be#9(*oz10U0YwcoY&N(ikjFGflhEb$w$Ojcx+U^V>rO!s=7??#a9$~N`HXfF7K z@u28G30!UGUfp$|9bY^I)H2WzAkjvuH@1DTh(CH+WM$i>M;@E;g7JgmPi|_(q}{!d zP~G5$!hBFENvO<9@Sp!&3$Ia_?tOyhj(2{IG{6|DzQ^&@Z4j(x+qo8IN=nA-9kk zgdk>*JyPHD^pw&21x#<`5AjtWr+}*S9VR>~YL}hQDIpP$BQ)?s?7>+X(t2Mq!aGA@ z;uZSwRCepwCK@(-)6^j6EPKlGJA<^1+l;;zyi`fO&Kd}s?{yew$T$@Vg%9q30Lz?v3U z>&;_$GP2HhJJcF)lcQi5Iv{tOnd-qEEY8g49YANcuMm3_>93xEYOr|c%chjs%p)6Qu>K0UAF zP8UKeAy60`xW=CTSr|YdA@oEY!aBd&jgt4iC}5k+=~9j1WzVgx#XljPnP)%5dQK=~ zAdMco5^LLZuC_tVRS%X?4Un3QchDTgrm)b8KvD*_0#4xQlvd2V5#@`bxA&#>6}_RK zVB^2nEtDNULteiWGclohLQ;PQLYssRIuYg~*t__fYXo9KK5d8!0hPov$)<;|plOws zll$op1(z@qOi*T^!}wBZ%3;a-vP#hagsoa7RjA~03wtzzIlZNGGNToO^cz?Hi+w)7 zASAr@>FJsxgy9fYi0dqAGZzZ}JHJo_!2Ait+}4%J-``&#y27Xv@#^Yo8DQF7!6<}ue$Fu6 zPXCo9K%?6_IEbQPSy4Fog3bFgJS&kZP zu6o`E<~zG|Z)-BdIk^;l|Ml2(=v}7h$%#q|d9rmM!CakFPmbqWmO5i$c%m71E!$CX zU-{m8pyQ!Eaxp@$`u?fn|a5yw;7b}O`i8SnSiXIj^l9x~sjk%>cA4|n2M%m$%aVLSf^)QGcfc$mBRvlrkKjz$5#SgS1d!iJrY`Z2&wKc z_^-^Vzxd%W!W0ugJ;^E>X0i3R9l#6T&U@=gG&Xg1qK{p(|$gOMAV3LwCN-`AzN%3Z+tR zS&$x8r|r~S)|67z*}V*%6aBWC-CA*QW#ucFa$6G(k$=Mj_9f6?Y7~t0cp?SY425hF z&bC^FAFef^<@MmrKP!j(B{pn%Hv*5M5l;y+qa6XIIZLMI;=@{tT5KWIUY_UNZbxLc z4F{YNw__tW@I7$#-z=S2Y;4hqh&(TUymy>3>PuIa4(nI@Icomdd&&}b@9?l}fCEUf z5$t9;J})&KHW#fUykeFP7A+|)4F-IrFYwQ8RX4E+6@nmV?CaBX04DLDmC%{6aIpAj z?M%)<1ES`}^P@MO(+Q`dd*csj3Dx!Nrgjy+#|HzqThKMrfK1ycs97KVF8F^E`6D ziuxR#kntGNH|F7XUe8CLV9sl_qAX7q;kdU|?4416{BHh(G>0=vH^tgl6 z`j`R(JApj66EEs44T=$DZdS$UvhXo$0*7kb(j;20fVf~eYF>6#MaOfre@^`kr2!@h zZm`doWw?Gf;|IU9DVcgzDCccPcvog_kE`s#1;TTS3c5iVDNRX!uXfrF6ZR~TFcIDC2$&1!#{71r`i3;LgvEC*RWuVn26uMhMs{lYrr|iJo zT*Jc9Cp3VNoGz@*|MSWJH7aQpg&aeYdk~u%UBvi}Y8zfKSy43lE-=4|Z$GN~$o1%8 z=I7tTNbu18MaYyiti)}s-h8Iy*tx;$y&*EdXQ!c}S~l@wdyR}*j*^LZsZ23l5OviO zMUd%qD7BHF^!<|?P@h&SfmI?iw!;B&^1_=n3)T0s z6P%ldJNK+m!K{x{_SV-_Fmu)3VE{vBsI!CP??Ks*&CD#-kw?^pVrpw^s|9AIY}EJZMmVJ(;&byT!fZJ!i)ygOaEDK|2~yaLAS%37%iiCv*4iPT4W6iFcWVtguD;Mq5Ht$yW zlEmKbvT4bE1ubHee_&?8(Ab_%`%+SO*2je8S9lmhOW3kSgmQ`-bUxNNnHZfyCVw3u zl}7SI1_cG#Pj~Ujb+SJtUCtzRkS5Kq!}za6sj>`xOc7Q^Sr-F1WmH)4}if zouijy5N`90BbUn*?K+J~l7Hj~D~bzrW>r@ zIEZYDxj@$OA^+uvB44_$m(8T9bBk@DzDv+Cpp4QlE0?qnID=(mW+ugk9E*3NOH4|_ z)v14vz;-zOL`x*8J_3sbgVNH{I^WpX*hpAfz5t7PInw#vG1*8LJJ5uz9Uil7xxY6{ zgxoAF-~IWBmSNGkGw@;}PxckmKPq}}Zf%I=zA^MYcgQRGRt3BLu$8}FsRu>mFLswcw>J~+1lji{To&#UbZ=PB3Q5l5&I zU3tpN(u0aY$a%F!o#Bi1l|F&!BbHV0h`YMdRW!&d>mDFH%;qzmZa6cl)PCe-L({X{ z0E}N&J&&Ed3$vX>OLl^};f*%Zha)&ftsez%GB`dfJ*(Vo=&t)OCg3x8H&0Ko1H%wW zS^)y((ZYOXcmwnA@!2+2K{BV#K9tAIl8IVS%(O8I)gV<`(JH-tn->VbJIrSQC@=-Z zgI#CtBIT~-o%5ZB8kc2m`wI<(z`d?j$&+Ay_7G`H5wat(6%Z~%zIBrjj+`_$Hrj^6 zrOc};Dk`eCHv#5KTW2Q>05`$7sHv2v7c2k_`O?li6FeBCmnR`Fl6105_=&)om+X85 zH^H2ywYQ&?CKuwoi_;n_8++sOM{aJYezz!Oho7I9hJ5ot=hPG~oB1@b4lW0LNa}q< zut%e=uKTjoxF`K9m^r~CBdfLt_Z@x%#!VM9(^?Glewuq3YmI-V5d%Ny_07pNUvDgr zrevd`e8)H19wqbJ6fIU?rYBy;K^2I2P5-E37u&Dy zkSqy~%Eh|-xrBbX;JW18TLL_>LYyo#S?XhghiB-~)p7b;zK<^9A*sjR7Ueu(5*0BU zZECm{bb3MX)i%_hO-9*J8E`5?5zNh`2ZZkH#X^W7@{um zrzqZS@gcOtQCJeGK=rO9gL$V*0!IDj`BmBebn@Qb-dIrsF|mN}^voS%@|pafd34EQEVmrzS=}(gc!`w4gI4CxyO>oS#sa zWMGGkRmg4^4VCT-;h~*3ZnQ$A4AaGJ1jhu$IpaI!4&yY`XNCj1F|EdDR}iQIP`zJY za3|9mZXDFa6qLt7K}$Gc;>Ur(O;pd^ki?(2OxPUK#l;0a3O|4%gd!rdPAx3Ih74$A`b&04F`3Z^w5NLP z6(5FFG*KHsZ}>R%Pb=o{eC!T{;#A|IF2maB8e(H+gFV=5b_})-5u+-RthIFvYgMBn z+cEFXn_3v$MV>sn&)jUD&su{edEEBAxOIaYN0G=MmS!qOc=?2vCO-7-8>P$liT;`z zj!R7~47K2*gw~FUH>XBG_VdH`kd>9?+vwey)eYuKG%Ej**txh!!nSXqFA8Z-)tXI7jAuqxIud>+^&@=ZN#gN9#**4ga>AsVN6#LLxs@o8&pAZ}iPYVs*u#z;_UU9vK?RC9@2S)NP z4c)C~D+@|$sbK~5g7%mDBrhi#Q?6?ymLn8ZjymzQ6LoZ=qoX&!TKObj3$lGfS|<7( zLP0L~wrZlo!jVjLQ&RZA&D5zz-cG!;S1w^+uHEzTR***i)5O}AQ@~8|qUSD4cvV!& z!^~^$+gjI8x%#Sf?YKY53_C=K6nJ5_9-X_!6@?ahgK4P@P@n;b9yQEm7ubx(O1-x# zp<`+B0k8?fA=zJWKte)^PZcs`D>Rfke?y-vm3X$tyQsvg{fy_tM* zQG?%>@gh0l#cInLe$x~tF)_;ZU%-g}o(KR%&$Iei{|JxLt%&f_!^vNoj)fyO6m`ud z%`a-(Y+?F0u%%cac9g5khZWd=U>i)XS|B1Y>zhZ$T!2;Tm!8!^h@hkaO2;lPGv)Z$CPVYN2_#$4gL%=W<;I1Zje{`@<+j6H}&4f}02R{Y3H}wk@_%_iO zM-wxK5A$^+?>rX4vYDDohhOT}+oiXb<}^;ex^jvEmFUREfg#C2u{#65KRlq~h=3aW z%&eEgLhR|psmo?<*cA<(U6zh*e0aIFtqG8ovOkW?21}>caYMVzKGevWF%tz0zf5m? z2+%hK>tf*H;op(oTB3cgtW=)Vfg_9}sH);9N4nfq226%#&vm*VCW(}MMfhV-n?N8Y z=~+2Y_GUyiY=nBWDO}&oSlg2?QdIFQVPKb+d!IsPU-QX7C41;ko@>tj<#d z3jOx>KRVCq+N8PTqNeqIi6^T%?ca}+vvskN%U*oub;3U02m6=~*&im;dx%VU z;~|*NQQoocF7r9bwRF7uQEeZ3B%$kqI}~;&86$Y}?$*o&l+KoCC@J{FQ-|DRt>!xD zDwExDFB(NnqN!xLI++O4C8_MO5VnIxux_V)bIwF@2@F2aQ)&l_Ok};Sb?xl@ymxG_ zYm+M)kKQ{Hv9&I*GnA#P-2vn;-I19lX5R>J4 z#Q#lEc?wM|MhGc^=G`~vd*n^^n=eVAq7D*m(4XU0eEWttIxZzrQquhZ<*B6fiP14Y zA~edrKd$$IAOtpDkv|F$n)>-Jj0z_$vvINBeft*D_7&vG#C#fT?;rg$CeHgE}k5gfmfyz9Ww#uDU%&n9uU8O|6aXSj4;CBRq@f+ zQ0d#-HB>EH+1arOyUzO6Gaq_be9!fyL{uIGn&?g?!Zv#rHi+qx1``;f}yAHeP?6i-;hlC359y3p1qxGM8X%Z55i28I2qIy5r>t9_lJE*+eW0WzEnV z@>%S#*;$w}1`_(v<27Yw!NWociG^9uZ`^UI2;2?xBj|;o+I6!#fSYl;f_B;e6`~VY zri`2V>bOei{&|bSU)0z?5)}BOnqpRAA%b?x%!l1zqIqvu zU~2~kjoA{k&@{&_XFrg!Hz!-$Iz#XF2^!Dm%9bfOUCX|%Z|1MgR2aV&EHBz&zJ-Q` zrC=w+=e*&>C(tq&d!B`9IP1YAT4=>s7F47a!I_J)Z_Qzk!{af~IXwsW?Ihw|p;cg0 zQG8wfv}>+jpg#7ZxovZsz2oju^R)jAfYiyzoRl-T9U#H7p5Uj+iO1NPCA@SQ=ME1w zDQpUxDdO00xAEEzcVNkhe*k)?2Sd=RAaSET#%#GgwVjcjOMvbp@O`6Xs)2{SG)R{n zRQ-Q!on>5A@4oG&Bn1|!q%69-q`SL2C8fI?Y3XjHq`SL8xj~!T4)xac zf4+zQHFXfBfoNjXRCb5hP&hrm=m48h=Ae{}~cs07?E}u-8`PtAK&`mrW7m z79b+T54@0#*3Oh{OxuCGHJ%}qhgzeLu@+$Qp&9`pW^uuHzW%J!VI-+*Wbr|=u;K@k zq~t{faTUY;-_3Sb%k_W?IGO>D2jmf3v7LBO?LUoe+mY{S(%;R`v& zAx4tlQrG5Aq>&g1ghwwQVBsM9*Wm#cZu#s{)Bh<(`qxNpKZRdH0{-@7Er8665x{h= zT9Y_@-PzvG7Sa2WEg$nZFcZA_^ZXUEC-`!ymN?N~3EB_f5QouM# z%=9Rx1SjmU<#7Gz=0m$0C-ehfSJX;a}>q850Bme=X^4wRW!}QM_Uv%L2 zRMgc&{svocd?^t|NE;#0RY8s{DytBV8T4T`@`huY8e8A55|X5-2Nv?Y<@V31#;4_P z_=ZN?4ag{Mj({h*sGhGE>SllNh?bGczI?>d+MwtraOd(6&{~1iF(4)jk~_ijR&f8x zV|(^o^mt61{_4m_w!#YAjxOqw=IP{38OuIkSUZefl2|4zE-ucFs~X*@j~2Z-4`hHs zP4tivUuaEF_Cfo)<<@;0z1_F&SALpD zz9Au1yGj@3e*KHy6ENw(JIeEXjEezNyZaJ;tBTtl?bym{xnenYeyilCF)5Ovs1UI; zHpYhISIk}|$ZjF&54d9Ft}+?H9;Z#PirF0sUxAE24$mhRM{9wPUPtU33{+^cf^a2e zgJRu%cxJ?q8{Fnp*H#uRrdj=JVzT-#eo*Sojt7yc;0mB9W9S8JaJZZ7A=AK(@;^e@ zfBsyV&abG5HkQE$Q|HBFXcA#_Ry2JOos|Wfkpm&%mw9sYlBXS$rVt6>S?(Vm40dUW zLb&4dUC~;vr=-HxXRJM;P`>r;Uc(Z4X1jLdNpMCk;U83nBKZpWa;4*)Mj1jIbFIM2zeq*A2R-354nR%uJh?|5f@4Y`R3I#O*Bik^xNSQ^~cj-|-p>k!6-e@bpTz1W4B;kK43pzu)k!SIY}1e?X(_ZRi|XMwBUI6_?szfBMAEHZ>y;nM{TS<-w#>VDM zx3EEU`#3Ztq!X~bMaN`CHrJ3Av3|+{rM>tWrZa9Zzkx?^5C2Ra880%FzFF0VM3xtS z!HPha3=WD)c?pq6m2I{bS8KY;BcRZ7efY(G79`=yTO4#=;9!49=6n?R&ZNlF_s5`9 z{$>3Ei>e+ubN>VJ!ky&c)Xub!2g~cIPgIYIG6#YCeh$GhY{{kXd58rCRqxY8G(jJ| zTgL9L$slT!H5h3V8WelsYEGG&ntI-gE{8P7@}7HaFUu$D%6!pRjl>h%E3eI`1Qmys zXxczYD|C0szLCSDr@juYY*dE9EojTr9gFw*T4{%?stx_keF(f(5%h^~=NvZvvn;Vq z={$mcq04MNd)oGp`*tnLbL?TEj}MOof-EyMYH2P&$fPqEj)=HPA}8a4XNn8YW?uSC zvVwlIlfhmiBPxdRRV8mq_9;T{TEy2BZiY&0P}`8bMI3Z|P?**KW)?o2p(OpTp`(MJ z-W%*CFOOOvmlXy0_P$=}y8jwP4Q)#R=uZOefOR1{EiD|-UVTkpU)K)|4!&J%XkgTk zgM=cQNV@+@u%{`^5Ge|UqUFI-V4SPkx^0?UnJw?Ce_()NsODq`FP-;H^qj10MHcfY zNxKT+QFgxY_iqg`p8%7inWf5y)L~1;X)aVLEP6J%@x#^e3@+H2Im5+nt>)1-k?q&v z;bB1)H_j!b_ZtYN!bRs{%UaPte4cJ9iptCTfk}J>fRnQJsxS>gQhVGO$jVjSLOx3= zF*w>S)E*<*Rac9Jg_a9wra=il6s%Ro%%S{R*ce(q;rc^B#R7)+X&?Q~#sM^SjhWNi z3vZPmKVX}}Tp^K-`On2mIV{A%hgeR9Z>VASa*5Ec)z}8X#&9H9XaEWd4-fBriSYF& z9M>-A@P~7sr!7J=Gs>=bw=G8A!%NX$O@E#;Q{^MJnQubE(*55sL6XP>goK0+S28{* zh=>86PPFz{e}Ugd`7%glR#^z)oi}fLGPfJkn;56Dx!L)|Um;IafWFsv(-4sHh6NkO zEsu6{mWL!eLWkd57_*ayrc5ji)iWu@7-Q|UPW@a)rP*{7U}4$H?N18 z7Dj5q;8OaA=2DjgW}y15Kn7$NTMEhuAp)l^7#^MMEf1vX;GXbPxqST%DYr{)$K=4} zV}D=xgK0S|jYrL0TUH`tg+vAw^0{`j+$ZrFU zbKQ1i4ghurg6c3T11@(qZqt=nZ%RtuTH&5%wl(+Hc@5eEf{~Uk5J8j%T-VwndVMqtYBoq%!*axgm z#SBoHf_JG-BOhzxzBe^JdyQ=}Y=6xXkkxGlBLV3Xb!}D@1mX}!TR-@)azH<_i<%JX z`pJs$PxQZ`8GN8-vi(90^0+(e%R}J{1IFppQqgG}Ktvo+=)=;fVB}3soMfd5`QEV6fB%p83k#zS4vXBaV!gP}^_jb)PL7NwtR`foeb^IVUmHXV0 zkB6Mcy+J3~wHV4fI{QIVDj>H6rh{i}7WHj>sXlxMk5(A8&l zwntqFHtn;|=S~zHBAjB84z>-2;xnQST`K{QG^>o{OR06hhZux1)W5ah2#e8tljOi` z)7{LtI4Cb69!MQ+(BU+vOLlDu2r?xDwNq6t4Lp@!T*X6-$V5Jm;&b0~)XWrsdbH~O zXoWF{I7w$grA>^j-JS`gKRPJoFn@#)qfZVrv!U*#)7rSnAK@q{9sS%rw-mY(;=JLI z-NJzRfPF{FiSY^`5h*jP=AAb2(t!o=Alk=NpP>&L+UXzvQ`qsZmcI`5O&NU|K81>R z8tALv;9bb$9zhN%icNHG>{tD(L*;CbQC_AU$Ka z6@Sn5_0tGDslp!9V+vrK{<`Raa$ocCrC+493cp|h=?ZWoPN@KvFc&7m`H2+@X9k!aS?6Bc@fb;PA@Xm z2Eee!9*1%E9NM$aK%P|(Qj|UmO~)iN9EXh0erfAC>$CYsaf9IUb!95F9|h}KXRnAI zw$6#Af0~sy$)(q+;Nft);(tQNtL5ip7eRmw2-q6pV{`)k_B%H|{?$=p`2hv5&wQ$j zn6dk~?fb_)o04{bbQGd<9yU&mHfTxTKce@uTs3T+Qb?(Rj+BGS2NtJ%APhISQt+FZ z*}kShJa>3ll&HNf9n;XIBh$_$nt}mq{FRhQjIzP1&Kj(6Qq9p}HgU<|kaznOoOM`u zeu&~1UQYP0#)Ta551s7~ySTksQh`IXNL|$8`%k(QWutO_1=WGQ^l}TF9s;}T&<bhd=H<&Kx*2=XMe_Ti6Na;YD}N^Eg?~X8V!XNSt8`KAA^uIM zz&LvN^9*1HU{~_wjot9_}9Xi&)q+QhDf(JpC_g_lyyOuS(ZDCXK zr~6;9j-QYfLMTxKu7kkb0`sL?-S zy=~e5|8MBOzAgqc=)gz=Coj6q-M>9Sm#+(&D?GEwOg*qFPMpSBjiYlt!L6C72&2HwUq`sK3oNd}G0|Zb@o(baK!;aZ z!Kqbx)!O6}(eff)(!$Z_+j36k>4}Ng36bM{*tYk00?vTng@`Y4XaD@o?j+Oyb~qYO z;enyw#`PjP!>|R2%ZLrheHfS7m&r<&fk}yOUHAzEB$7Y^ssbnuce7vCD~&Ps-WfuX zB_IN+f=(|Bh8-6BV~!N&owf_-U`uU=U+pp?PNJ-I*4!a5auq>2bss9@x{eD1Gm_$D zG7wXEKrk?-fW06Ps4|uvk<<(!>rqj`$rQM}o+z(>uQ_%HAbG9Hwzre>PXq)6rk=&a z^b}#+U&pQf8gqyQxvV-q*b!@xW7vt+LF`jfK zBE~O%7(gwn4yafH@s^*Mkr_~zHDkeEUMh$vxM!~Nx>`Kg+Y`2;;m<`O8^~=d;)a1s zF*_D&KSerQ%frZuI0+Ou;^ehs%CKUZxqxvc(VF$yVsCPg`PPHajZhdp=O9XXN+}!~ktfTo4yUvlHF- zn^R0F3kskw{0WyP`o1<=MPc>jM{1ZwJc)7lY?l2K1!m=yqkjmvCO_m!pH`IT7o{93 zHW|YLG|=wIvN<~!m0ArfZey)9Erq|R;)+8bq4FH!oQTp~S~B$i9bM!OD_6W-RxVi$ zIU_yxKGL;>aS2;@f7nT?0@!ePT<5g0%*=MRjnD)n{t|BW;ta9H$BZeZ$b@YM@Y=J7 zbcHl!B}um)3cR&B6o7`x%wn>utB)|M|NS8IomAnZd~XUwyw9lY2P4_m+Ojqo#0@VB z2m9~EB*l4LCxZN>U9#pVLSvK^>1{cqyio2?!Yx`1_S8zifah%eO?f5e^^yfV*>*zN zdky(^%#H*U|9Rp6*L7mcgBB7Mm3WEpp??9;^}Wwav9x&eXHZPf5_!7eX{MW}HpGCN zBiee?qIr_14@!1UPJM{{nm%vzZL-P691pWfjB}CtK|rD1nG+ZhB%+qi;sh4KwU#0#1_*@@KO*050Yg z)XI~z@y4z$DBzm@_c!N1KlFbAe0ig0!yvn{{*4L4cF z2Y+q7Z7+{jSY?(Gp~b)mHT-4~)T;5p)Kqc&8wD*Z&caB-`FVCamv(aJXHaE`UeID8 z>L;WXyCcE$4&~tI?@YGm(mZ6rRCY``SfiporER8k{jv9?s6k@Vojx7+YqqwPJbT$F zq53^K(DPyh9!M80&W68|$O$xCi^l1W$u^3jdJ8|rOZ^@nuNZGuYq=^&v-Vg2_8FQ= z!SM_J>fguZqKIUqit|1*Rkv)N1!$f*9q*^-L-G*3L>p)-mGjW(jDmCodLvrjsC3WV3D^~>CxHy)&={8JF#h5C zqRri&CqCANMlCFy*OV0^YR;!|q+#0g zw%&hOXtHdGpP%UK+qrDIhX6jdd3Fe_&hoRPkvyaSdE9(eAR{9qlho8T%k;NItGZSG z(2Ds+ez#}+?~<-l6p{*a(KE#yAQuS1+tyZT`9eR}%fJQM@Y>_^tQjnLB zg_uVR!zx=ucD_xGV;Y48!V5iwRLIPHddlqB@h;-E)+T|8{1Q$ z4XRc0Gzz(7!EHFp;(+2NWFlZh#bv~4Z&_32uqXRwI3yO%lpBG5pX57uLJ7U*C!C3F zjUwt}N#h&&V%f`M{6C*E;N?Vh6XyN9k1qcu+LF!!mh0)|uA@huw6}d1*&gbj^Ii@G z`4V3^P$dr|W6ruF}R-B>>q?p7VpAGH!SL$sJsoCrxl%(c7Y zarsP23NuiR9^Lv3bZt7&LDhTl(FH$oB=qnF0+&!+%3GlDNbT@jT}55to`H}4bTorN zn6v9kZdY&p>7;uhbn zpqX%Cu2kgpBL2@09G-C@LBaD8y=R0sX6^wxIqKR($n7tS;sB=MGEx+z?(`-HJ3!Kh za^&({(f$|V;R9LVSjz`*0Ru;Gqnd`P1wFX zxp80pW=<-xS*U~n?0!jES{{`^AkEX?`Hnpxp_+!^QrenSvT7}cNp_4MtBcRGpWgG1Pl@iqti>{3u7XsNC$Ixp) z)gF?r2T~lXKwcKo5+x+qjPkGv?^5&l0`3yNj=AUhXa<^d2h?ce-@ecQ+C?iLH%y%B z8R{Ww-w3ksuaTqCv*x7)^u6??GECte167*BBnw7Ac!FC_n-R^8Ln)x8Y8GkWb-N*F z|NQY%|DKqd8a9osM6s8IZMaw7?rB}uLf;B0xGun!xLA1CvoeFk46k(}3WebvbAE9f z?;Tfruq3amAkoyJYcGi>PlWN%cEm(TIbLFDWS8EB!mQ&7`w!p=pb?TBE3| z*{PxCPq**E1;jsJAaFgm~#e626R$#SQJ9$v^x2`oTdp>%>`+<#e zu6d#KABgCUs*r*aPR`B++)l`5b8OvaEJFFUIy4q@heHoa$Xr~rDq7Zvpws=yyea5% z+N~6JheQkwWo&}b3i|XQA7vJ9;vNqkWnqnwzOP1YA@J8g>HlM&A@H0mw*vuZ8C17R zH`wuGA1%WkC9L;}@FlFb0%}lNDtKxQ+xbI2!=&YicfV%x!Nb#RNlbEbNdFgFkw}@q zr5-h$A42!iEO_oHKwW&l9)RL5+#j)@{je=0yB;UU8w1XNza{KI3AX=v%pAngs^0~9 zP_s5Zj#ANfivy%0N3H8?>hR}B(!KE&;uKG! zUVGP;sViC`8Tmf9<2up{GMsAn`a=?jP6+BhZV|pNxTz%#2%0U|n93zt5kCaNRL1-q z-!u8Q`xXw=2n=+9o=1`o?T*M~3arU!{&+aHl73ZGG16bE;fJbVOTEUtYq{v;>!9U99QE&nHi?yr@&}ySWNJWtd9I&p2dzpWxwv$7! z&_$%}`EF*E8UcfBe(2`qrTzXOowTr@u$Y(>b+edz{_RBU_T2Y0-{c6dD;e~uxE}^v zQO4Z1V(Btqb$fTxTRf`3mNS;8{gk~yC;d$or?vof>P;9QH9S z&{Pr_TK)=QR>7T+tg6L03|j`K9V7aYUQVa49V3t&UOco~0|0$kX( zj*h)@K7W{ zBw@%g8#Bxcr=|;_PZKsoSbSB$IIK3^<1Z#i#+wq2DH?cngCH71|Gb0~3eK$wgvmku zB`mDva#7-t)w}m7Mc)<(m>E^8v+LmWGmf-Jwg6c?f(0SMPNd9 z>q^C+hHiKqbqv-isH;=Ji*yfLv0zlC0$>M>y_&cjRmz+qwr*_|C>W=49oZOt1%nW6 z#sGVd2Z=wdpCSgiQ^IMr>c#8D^9+zyzF&yxe+H|&Tu^g!Ylk_jH`(W7bS>cWFb@5_ zZ8_`aPt42+PYg`xt;I+#5be&kJZufWbfMhmI%yrjGB7NRu@{GjkXFwC1P|rkzp5v_ zZ{`89<&G_d;9&V zQFtjsP*!qe;LmVH@8}~UHH9RI%Q!}%PApJq%RMF*mZ+QcOIY=$BW2inAGTg#NlD2p zkPVJ-`o4ESUKJW=0Aw&|DCq z6K)xX@9($ZKM|czq5&OH=r|A~m1%}u+8s`cuFJ>~RghqEIiDpQZT$G!rRxQXe&g!DIfhO>s|)Z5q4y*Y>Sj*2-*80l5cV-P7T6Yg z#J$!GF=}G)iy_emb&(Cwipa#3#G^~gH8}}ak$Jbh{I#SQF#qGeH~%rjB*yY|?}#;1NnE znr9UxDqi#S(IFvKEKWBJ5@XZ!7KV!=4 z`VD03?A(81IZvdHA60APV0G%b^nwS?6}%PYZI zlzM<}zk_WKbX@+-dd-LM?7>xngvQDLy`q>vx739xz&!x%MEhXLN)H44AvNPf>vZaPpze5 zUFt_+aSgaQzcEn7aZ^F-wvoTdRQ0KVJqj>Eq1*1 zr!tSa&`^KFppR@kPmE*Cex`MgBif^NJ6aeW{lfw{*8Za}mYbYUVR4v!?ieJrB3vZF zK+TPZ0`g-_7lFG4oJu^hl(|%a>LizHPA*o-d_KIrRy2vgB7Z%j;nmwMw&Ut!?^(*t z(Y(+GA=G@OdUEXzi9X~>v0lLG&owSVJcj8ymieUIHkE*-YcB>Ekts-D>iOBJg0LtH zr7s)TO^6&0yT6mantVc8Tb9)a2m&0=#~-7N6Z61%?_c6tAGsT8>JWkv^bfbq0Tq~e zmeY6p!jS0N_|;sj*UO|1{u#@{4k)QAZf_u^_&tx4A8MyC%d?uE*KFVWH@)?DWc=7& zMEA$2uLoHY(jDY3rLEZI%KN` zL-Y+jolnav^M(vfZEO}O6Oz7vqsjSX!_Iy$OSIg>$xI4yd^u*|eh&ETa#(gKin;J7 z?GmRE7zd4+B%c!$RCT3d78mpC>TZ7ZrVkHG4iTupS*-R7u88J_+>u+JDZDx1-Ocfh z1rSjkub9n-BMMKw_=e?c^}b=(tpPK`WG>P-jcD|rYq4w+z7k8h*+ z7G-6Ye|AOB#1dmOGGyFZxa>VO>q8&k7(mrlC#|EG_W(=llOqJpmoGTof91DzTLI}L z&0GW+{M(pm1JEQlq1fDMxP!@+@QB2 zebr~X&KS4$ekCO)Zf5vAK_x_Ll8})}uI@Z#nlImAXAR}TEyo{aq^Qq&-=8zO+c0e1 zI=xCYq5W;3o;h0HXLW+tEW_JCFWBn6Xvndl?mFgdAfmnDs7nafU=H|6KmL?yv}|cj zGhbc$5yJ@8hs$$xU~KlAxp^00D*hSDpygp@V?)0fAA-4~U7pFmH=c4zeO1w}!h5L+ zAA&i#MsWXL)9aJT?<@!y8RvYRX5St$6mfJm@MX3`lj8x}m|n>D?i_eVJB3(V7PkWS zYC021bZKzalQZiHVR5+u;{Sq$gsQ$#ZgTc??Eo!=xZ{U_tj1Q#z zq2hAHJ9kVUKPbL@LgWUuk?y7~jbac9cULuHS9C<-dX49JUCR1bE6WLoDrkW>C6BQc z)OsEoh{sH`#S5@cL*jb*M$~&bdS#@ywRlZawQla$`OB}HY%B@;xbtlzP#gXtpglnw z55uHpuvu*dy^fNUAMm|A5U@Dw!l}P-#Oo5BYdd~5)LG6A`c;+UTC~0{ zNPV;4Jf}31#QT8DGN)>|yZvB$fmpYGWS{l3R!mVB;f)^z!WvJ+&Q)oGe1XRL(} z^fVUAW`w~oW9R-==Dd_k`fOiyGcR^pY{=2jWHcGum&{^56_N{I&>2jK0`*GlM z#hyO5DzHRq?{Y-%#a?`^A&`Lrnt{k&1De6WhoZeOB;7KvxDt2>)p{;-EG`c5Hm4&G zc4;A#?>EfcxC>x=GfW|Sg!x;?&pm!KbP;S!!(AK^p1q(C)kT^mQ0MS@2)K860?DVM zPXj}6c~uA3iQ|bE=bT&H+kq-2iAE!H&m1n+*gMDW1$v?ub+NG?---A-0%_E41(&t` z*VpBju)=B6d|=tt(5*Y3oh6VC8W_tNcHtdg7=LFm8y(q)QH&uDj0vg!ahi42%QL%E z<(GzYlr#J@&)8Z)4l%M9aj$EVl6K+xSTHm+T({l}Ed8k+ zNS6xUfjBy9buUA%OSw)t#Rwi898|b!@#8c7+ibCAniUJK$n?6^(DcnuXg7L!KJyWy z)UXzV8L?eHq>z|EopVkU6p&Q+u|%r(&hUHst>pS>#q13z;-tXI*Z5h1)DoTGm_fx} z59T2qofnPU(1Jcog{K;f14f<%nw#FLTD@Av^p2+dar7@04JA+*_Bql{7{wK}ZC8f8 z9{Jjbc^^Ayy)eLwVuW6Sio@_lW<_E}CDf2Be|}yh#>!YMKG=JvB#CtpjY)*l_8H%b zW5!&S377bK9iue{$|UrgAB5}Atds`%(bWZnQ=g!F7~7`05Jq686j93v^}FCQKKkjY zBOllky7jn|cxOi0-1<<)x)*hOd#BaAVd@H`;&)`Z2QPO%uWD$WZ~3qS3t|HU z15ygkhP;L>m{^^`_ZA`NIv0MtibTCa+^qsux}~MXWn~GfPjDu0JsUvT?t@ha+>v+5 z?HW2L9zpqkf`9M{G#6mDBPJ3XpBkpW9_w+Wq`%)gS-t?WhnyaW9C;$#DhDR!=E%yz z&~#_!j#=M7dJ_N16#tSIA1`dE`sG*2mH1%lluZ&C3{=u@~x+C|Y<#!&3 zk#uZ(z?D;b{SnP^VQ+74VPRqIm)9lDtvuKOU-gvqw5<1H>^vs4aD``*_~=7+HG(lf z;8t!d4IEdFNH~^6;p1+ajl4?Qj|p*dVVPz$9hG=BSOl!kTD@p*B)xlJjzJf05->UL zS4;FFtF~H+cQeo~+7&}0t+PDoVrd0O^x0||9(vKkaFn(IOhGZISqEAqBdZ9yNu>b( z)8Hh1`gbfzW{jOOn6Evt|NxPCwm-49v{Y(!PW-YiI_K`;`}< z0ukn&Gr0$>2-4PPflO8<81Q%dc|SBYKVsiub=h#!Er786*~^+gskDs5*ioQN+nnKS zBzQBRcn|BZ*iV-|zsP);58wID_e2Cx9!TLlW}corrT4Mgc(tGRJqadZL6q*d#~MTL zF9u_XW5u64p->#l*{BG#*qiY%b+%;xXC~zDi`t?4x{NDMQ4zG4-WQ@<_Z~UE)NPEI zPmr?g-OKHaMk`=Q!s8jCPojhFHH-zM_Gx5>esLFilSz}*)CMie$JfG73pN(N!PxOk zngUA*9)A^*Q{7S9+o>oY91DLvMFUGj1=&uCosf5G9o{x$*Dhm0@Y+A1dxfh)UVd2( z_iU=fB=ylVs)9tSja4v2f_Ds-Z>^F}MsK~_U{e#pG}<7yPoXKoKktP&<%@6RTk1Eo zyCr~v{_OH4#`Ux6PdUVhV77duq2(w6VK*oGFR6I#jTc;lVW|ur-O8mpE%hFl-eb_B zFc9JQD$!nH0GAwyFZ)7UDlH;g`T4B=%)$TpLpNly)Ecz#jHQS>jv?jJ!RsGSBa{m% zJbY2`qfbQ>Oir!-=eHrDmusozpwpT$VWrG z^@G7-&O_jQ zYbsidU+0uHP-MJGArlCaS?$;|feV}tL`*zs%1e&EAKCdP-OpXQI=`EUyJu~a8Y@|i z0xi>m7sS#irkJlH;-N|_{x-)V#Ctd4j;hR}j-lGSs8I~EjSI&pUCM&=MXP6zEJEhg zN{mdiLgb_P+%b?@C|`PT)oz9sGjqo?)t#Zd8!S=UfifRh!pE+Z1zXmF*%uFbS!x$y zh34d=a&>IfdFq;ez<4`T!47f<=vT09mM>dZr?L-C6(uDl?GUZ~K!@DCdIVyag@?rY z#2?vU!wE7dBe|`7kOlFGD__QLYCpycc8E!UP9d|rZa-PKUWfvtDv33Lw_%r5#nHrs zd~s6u98*B44ByMW5#^MUaClhALXM|bI1J|-pEKV9A&DwxULoC~7I+VdE@N)I{ByAH zGx_y4lVH%o^|8=A*Ki@KsHl0KQ55Mai}u}EAm%)32ZruPW7nPTbmTUBL#K;Mr#{on zptG-eI@pX}IsY>l$&2zut)ZFJW-m`(wS2{*`_Wl|LT|WSKN=4)hy>M#`&+}Y@3f(= z&-dyAW&$^j!Key*Mi#-CoE=Qg2Lp6C(e~IkOZTW}sLsJ14>uK6K>Pw8k2ex4>4yo- zM5Xt5ol-g77;8i7ZT1P*pLkQrE`mQIvh%%%|DDB0I3xj2Q6eeetV?SMZnGi~u`pq^ zIFzxy`@rw$$TC$j;(=FUZyzXG9GEuwPI*osO-H=ml{OmYUeYb^*xYbYcG1aQx%~5^ zK*JF=xiSFfl#axq2?fLixMW3b=@}sGkG@jIaw#b(lslgfAzVV(LL@w0G(8KMSmN-9 zV@i0#f6Ks_HUS9W-~9aiohl7Iu)0~$i8LDic_avg(zHV67MQ=gm10?ADa*daN>p~TP^mY2&8TasM915zKIhb&nXy|%BI0C+hZF#k>TZrxa0Xs3s9nP@JN zPjo(Y4Ddko%vumy|po&lPkNx3KX;}1d{MhAX z*0&m&7Kc+?yFwjjPL-9IwJOkaP7}m+Hymi^n9e+fh9fViK;rX!q<67oS#dLWU1^1# zwG6Bs25E@oU}Oe1vpk{)yI@5*60Y~2szAY*_Su|5xLzM z7(8?0O$F7d`#j(s|NNnmJ0c5gnnG>v?#xmLPLErfG5kfL7!#)q_PYo59cG2F1;_j+ zZ6Sp~d>pr*-R)YnT6%lc4a6I3-x4oLZdGh5@8I79c&U%aCQUqK69K*I(uykD@Tz3=m!^!8-{sP7+5w`Hw!k_g2 z*4zKiU{)B$C39Yix$6+G9r?F$*JmXcJY>p|YdjM8qKsYvG_wN+^CLtQ9 z`%It^^8-SxBQG2+2gbb2r;LxFT%>zKYl3VsCsUQ4@*w}FXaOxLTA5!X;?QwXbvK{O2cY{uLV!lSh!_RJQroY_xCWHZM0y(< z?GDla?V8usM%zE9A~>THkf}s#DMWU!NSj%fY;5lsIS>xhWdyVp1L-2CI<1+wxVRU8 zk1B}Eil7ZefCW)I99Gx2es>*Z3}@(pD14#Ke5CC?@>tA4By;UA8LHKy*Ea{1zKwM6 zM*{v!lJ93H!S6XBG1B-BXotw;j`-HXLz-=@i^;=MP0U^ac&M01f!8%VZ{X((fW=YsXT8z0?;;yEyc0lV-r)k+lGkoRPB zK9LiX6vf0-CP5g&Fq{3fv7dcx=s^MQpb=U42<&<1E|r@rDk=u%VLODNc4Dh-%mu^4 zaV%_HC?X5!ZhGUK|C;7f6gy`S*bl>yl#~aww!XI}5|H<36p-*&-vg5~7JIk>Jf!F( zJNO5fKC|L6k%Sr^?Ik=wb9kvu_PwEkPuV$)dO3JMCd4rYl-HXJrpdsVKGW#+_m@&u z+uaBXMJK=2%A#iQtCHjwR(>9*e^IMOz0F>6&@ShLg(Oi$*9hLtlO|9r=55yIJ!NJMyz~Q*>4Ab-Yd>$Jx00~QCt3E`p&^}Cl znNG~Xgw7aH>%65f$F4mm`}btDSwJETe|_6l$BJ|>b2!;E3kn&TSC7e}b<7~s#*-hv zA7Xy`CD#f2&kpaSz`t}&^+2QbT(_GU$`Ui{Vk&GbE+N5qElT^;#?9cA`AMvPe(SzL zqf&`#B6nJU!(Qzp5)b*X$Vg41Yxd`wK&-^{zOf9hd~R*8te89CU)GQG6?vLcCO10y zWh)(dA)+2l{ql4wp)E<;QK!XKU7UYeWhYXU_b&G%kkKctT`?aPsZ-8J!YRcZX<8m!_ryY2GInD8N}h=(!V ztBs*oj!Tz=QzZ1KYLm0)UFzg&uf-Oso=t>WcZwa?WI={xaVe{L<T&JdM3^qNYY@JEATx0#H^$q$Cni39%aqQvSD^!w0GfR6CKnAF;Y1+DQi ziWA3Uz{~LeYCsr};X?vT0IF9d%VoPT1CckW=L^KJ`DMTD=I8W%@{M09c7VziEh~IV z}a1GuyoUpM{0QL_`o@jb|yoEegxxUp(8c zwz>gnbbMc|Ya;?4?Yz({$-LpIyEi4F{GZaZ=j#lHq0#YXN#wICMB=&#nN7ypzAbvO zI`*y=52nqxB@dyg9LEjqkT6gh)RS*Sf;(3Hzm#IP%%2@SOdWJ_{IU^FuE0SLXc3Ll zh?>Uz-I%Oq>9(|^yM1kGWpxtcFd**sJ8*$M_s$EzvRh8cF1Z1|Z|CvJb zMDszDqt^=yebE^p>t(s{wj=6zr#JEVoe3=Yr9jN6)$eT=!VnAx3_pYEuv`-UYcVIx%|Y->>#I z?LJc>u&=sqIU&kFA=f@(9dnBl$(pm^+{f9HdebAfI<5qkw_RTTEgFkQj_GCsNKmtm)u#wJ; ze|%XJar`#m&z~f!3TXm3%r#FRV(lOI;s@)#VOfLB&rG@JW_sHI6lQ{0`ZqFqnVr;9klF z!>0pHO>PEXg`5)?MaoGv4*p)*v#uSkdqSznAwCHuN1gL%6 z$iGJ7*SJ0DPuEDyve~EowVTyIT2RD{wJWUubU%Oz^LKra;Ggrh1O|1p#diple*7Qr zcDIj0n&B=37wZz6&tZK!Tm@g;BfWIsR`nliT_S->I`Z6;&F}7$fNLhG+JXTh|KIoQ zT3#jwyc0*RIS9z$unC}`jn8>q4`!fArW&jb49G4}YBx7lqTt?|ww)YpYhRhY{!E=r z$+?fEn9Uam=o<3lmn53XzI=gy?Q<0_YeygCQ;R>t2{RSo#??5NS6U8j9V!g-_{eel zi=5*9rAl8hP1ot=fq*sgJ zw6vkZ2PYHDOG-D@6Xwr1%HkDJ2u__yAr7jZRGbJ73|+#Z;E75GM?Qw0K5dhKSG_s} zIQYzQaRcRN>|>v^KKhYI2$>f%w7S0Sfjjzr@fMM|U)Q1CH74d`+whBq@FztV3_&N> zfA6pp3?5kiuD$d6#GBr(DcE<+h2i2YRwsQ`Qu`ZkcRT9qSJ&Ea?+UJ?y>z6Lb^-&b%A=_Tsr zR|fu+g|(rrE6I(o<2vwx;Sd3&139 ze&UHV#IuT{4&pDxwVG<^8SV}Mxa>V&R)eX665$c}7yCw!kBn$H7vdlj@^RLD@OM4F zCLu*A)4rzf;l>1f)9g&ohk=Q-ib7)5yCKMI=H0BPJf*=WEnZ(D9!s+=02chql+ly- z^PD)<+mAu&(i?BvMrboLoWyWz4@j9eBYa0@4wAnzA;wFoR-H zEX=6W;eFnvKT!me07DT%jaNNx9F&|$0)p7t5TCXP!yG|~=FlcCA!1HXNYnu%AE;HV za2_Hk*yhX)C@aH~yK>yuBAW@{=G)x%`ZpSgd%1D57NzzqeCZ}-YX6*VI43{>Dt#>Q z!s^6vbiM5bg3!PKH(WP0ys(BAE{oRzVsi>_6Tb9KLHfooxlgH-Eot<#j~}uiRDO<$S1?Nl;ym>yL9Q*U*7#OPmoLPb!YnmnTv`^SIYT|OXC@0t zFIm^5*0M9eyJ}EBb8E1CD`ey7N&Me3IIdzf18_fw-BP_6p&=la&Z{@Ip#t()B;UW| z&W5GW9H?+Olp>5Ysth!+VknhrZv;zw3UU#}Rupe}$!}yLEi5l5q1|2rvqOqsJ;o%j z-yyyx=N)M%>dogw)ABFmO?D%NB=4HxotJ%;lHN+|H0!2Yn&LY6c?aQZkuqBQ2IuJP zTu0(Gd%HDu+{07!gU03NAyivkwQW^ON5r$BU=eUo(gQ*^j9KK#+W&T{$kS zLeKuryEQgC!ICRxc?(2zA=^RBCUP z)2!itKh4OAC&(KkBpJlt;E^8bEN|r1hU4#Q-Jl*|}nD4ZLF)@v1RfT3I zQ)~}~8EWfq={b_RW4=ujd9;R6(R<{>0VD@uQ(#{_Ah5CZPB3so!ZcL{g1bm>oL@03f2xjNIZ1l()tWg?8s zbzg_QazpnCmcS$gPZi&BFG|rw8b;%>92A+ajs4GIK7S)x{G&V22`092TK2}=j?43O z=P>Z*z{`q}+ict-Y^-uWi0gO+vUyFYO{$a4bHuQpV&jNefIlfSHqO*KqqI~@d=i@Cu^EbwNl3V8p1b?THwnFL zKScYJTgxPKW$&+gCxRH-S_NFG1(p$&ckhDn4Sj0f>W?M?hCDCcSgFOqPHd1pq8((e zgzbnn)6-5i!QApMEl960CdDEf4_8?KzAsT}0JT}zN7knCf4<5F5fc6TlvTh=9pT$0 zdRzix>RIls9acBdsGRck?x2-JY4{%rrK!@9N+x`%Cgwy#W z4Cl`ee+HgzUkBhawJL_hU*(0;7w7%-<5c+kAc;^Yl31YkF=|s}^mkudQRgwv{cN&& zpsaW8{1+<{R(T_OB(1)xt(6~$C{R%&ud00y@uwNQvAY{psTax#;H0=e+F9}fGMRq1 z4=^%VZqKLlor%m3k{xYJOG{)D<6~oefk||EQOHy%_m0zrKeL}Z65g35`yYKKyM9qFMYdb zQkyF_S-z=A=d?uE)6*-TBaMs&TAeeeOG&vmVA_g|_}UqtncSg3fg#u%vu=wD0-dG2 z$;!aeawLvuTBGO3`?=C{9$J9oA`oquM3vlZ=ZY#fN?h|nQ>O}&?jJyomE@$NVm^SO zA>|50=whCazJdY^)7ZD~-U|XzOSRUF8`T~>J@}J#ByQ@rO{;FithCvbHe$Y=gH`D} zAru(3Yz^*x$qPRX?6eG5J@DE`O`kq}nMdxq-yUjOH}Q8uwvdqza3*&vIX*sKkdB5n zyYd^3qil!llD>33sxMv7G)nL4zfi#UjDQMAI%cP8(=!5xBC^o=)g+}~x$T_~m1>hu z*jM-0FqA<^=S-%91Ec(%cQW#p(~s$Y-LxV|B(MEQfQN+S`$oUn*2};$&$Rg!HUSIh z)vH%$&j_bI!^3F)kG)J9R78Loxy%&fFZM>>{8AnSLARqWs(7JSI$;#orh74jvR}uf zeEr}|G$y2%+$~iaf02CC=<9N|#-uo9*Tzm67l4(!1?VvOI6r($U(^%glI4!ZFcM;2 znRhgWX{L^OOk~Zza=_=UY{3+$6SS^>^?7HQ-KLS0D^@@+_Kq88`=Ct0!NK9qyQNoM zGn~QLK%?xF8c%+`8nk5!q1zr$y_G`ppK{b5>sg&CDz>WtyB%pH`f1NIo7OYzq3q1J0V<-`{=keYq>UBCs|b z1gN5Nl_vf<-d%1mPmxCYeO2LPMG7(j#wXvUQQDr2Ik`JvRYygKyP+w6v3u&&gph(5R=GyZA;LeI2&^Fq(`Qu`Y7I zXnB6$NNuHy%FZZWW41^n1dMGooQZXp$k1+|KPfib!v z6U8{Q*Y8DnjvrXL(afc(^a|*K?G8;$!Mz)Dos&y8;8_d7dx z9>}K2m*q>1e@TJe2Zp#Mgr6e#JkL5F@}F#BKX&lNATe1QZDQLbcrzn$dzyd8x1xNu zK;tGFvdIJ!69j7hxuctS;Vk?ac%2T+61NWzJ50Yu6568&ROOJwO_+t26r|8-M3uoFryb_o*asLf74Y;cJRXlq z%C2(bbl(9&g5h6#nIlA@GiXNDhm<}m^EBkLbifJqqARj-D z`r!Zf(_*xO{gX8)=jl#AJ(Eb#7NZSlEVRTy*sl}%koN%_sN8-w6oQVw%PpXz3%T2! z%7vPG=ZD zQ(Xzte!HZCkwNeFnTt=Z*Gg=H+m^iI+6^!c%)@mU!pn5TAg)X-;x+bxi0_C|?L@t- zUp?PXWI2BM$=r~{Ik)cnCr{EC1$?v_PM%thHkr*I3<&hNxVgnDG^pt5WxYDn1zfG! zs5Zu5_GI9cQUF}_lz(h<_`h4!3-c#)5NSJ*nVNW6hJQZlbOz8iTNfZYjhDjzmF-=i z0hGYiHuu7Bj3|jotDD=WFb;;6SdAlznV$lj_b1-;q*lw6JB;)eWc4s07f~@vNVnAE zRmJo@+qEpTx-tQl8mv+!F?jY6HYIs^)E9p7=VUXhkxT}6ZDOI_%Ck(&pq9FQQA892 zgSktr*whb>uLoEIGc$>R1UMV|3E|jbu-rQ03-(Go`1Abo9I}*MOG|4HK%;8}Dpaf~BZ%BPdmdG@v~MzY{PGEfr070LT>0skC=me;vpYYM|mQ`Nq7$@OQ;&1oMPl9bxor zHib<_Hld>bM72%y?`KT!;-ghIFo-)vKbDEkC%H_UNR^fCnJbk&TK=feGl$1;@dEqk zwU0-9I+_pFaBN0^=_OUx$_#sYxP2R>nubjENX^Gb0L%-ST?BHH=Vi)fds1i*`uiHI z3KSp)&_5xB%2tm1FA)V<;JDx8{hn zezY9jAC2GJ+q>-~ShyP)92{KFXI+04{=gxEh{!YnpeEG%-#4R=hdr#GxcC;9d5+T) z)WR-@UC&U{DR(o87)&fDKlMMOSWa10a}dP{2r%@*9?n;hb0xN@J4hzojNI$Gz;iX! zQ`J?Y|5;0kaI2+cE$7Iqy~oZL_S6&KJ=zO={dBLsu!+>2=;d^2+v`S*J;|hL-D(Y%&*m^RFK5 z!h0}fQPz8FW`HYiQqHV7^GfsK6pnvTLPWHoh(AYnjRO^@qdwRO`lz)vTZrJ_z5W0G z&+<5-3#SknewQ)eGtEnj=KsvF9Au`=F%B8)0C(gY=U|+kd))M}n-yhCJbo?2RAEa@ zOcXvigQsODLqb&p0hkoo^AC>cEzSKy^V0(unaSruL3890u+@Rj=i~S+m*Wz9dzms7 z_#{H*k@4^EOm(m}b>HXKwWHnudZw=l-H|YD;@^%wtotb4@ikC2mBtom(1|YaKAR3o zB`^EwRQ`p<5V*he*Y0WA2>;W~{okcBdc@>&`I%tm@DtlA{EVKh3^x9CGOVRLr`{9J zcP`d>yj1BGpn0Mi>c|u6;|Fc&_J3;am*ki}OleO1Po$E=CaJ1|(fZ_f^VCfQx zeC8e+2HPN+B~R8e*qK!OU7**!G8m?P`w9p(#bw$45xy@$MaS=4<>v*M@)04~_NWUl3PC;JxJVgKl~Fe~-t@TWUi? zfn9fB|NRlLBvoI|EXbw?e0!|u02NY9;UId!xL#<>+dZvg`;k!98MXOtLD9H8Bj>c-ICQEVT7}KF07dTYt24ki$rOk>TP-_tre&{0a>4Ij`y&+S zq{pyvfE~T;U>FaOjG)hJA`DZc_J_%;;r1tgo-0`!15o8lYH)6vf9&4gp2%QIc6Lk- z@#~NQy~s=4o09U9@Sd3{XKE*It|l=FUN$;@5_QshWSe>uOn zpzSj)Qbht{cjdEq6tFILFFtu7J_qX`}se+>F4; zjmJTDa;M6UGVoLlU_%FYLnRG!o!}h-9HIN2r1|g2IKsXUKYSIK523hlp5@syotRc$5WLEM z{{9P<58p6=#}`{(8g7E5ymKYxw*V|^D;AC!!RGd{v#0khLMtAuf)@VvGVDXO%P&`^X4>!Cptp)b=$3343(JDW%c*Sn^M@}E@>zEP-VjyboKCzS23PlGORI(aSI`|DzLHgs!g&+UaxswIhsqZiQ*P9P}SMg0akqC}1QU=??1SW(~C)X}N zztxpysQ>=t>SxFhqVQ0xES+yAcC+RUNaQ?TEG=5_*I zn;vU<@cSPYVqCkAUOGUIK=9EBZNv(!O4o(ZMJGllpgv~sH&qft@3{dIG=S0wTY7JQ zAA!{jBl!X#kRyxVyx5BYr02Q!P4VQf;{eq&!3%ur>Dj`LkBNsj;5VrR_u3vfz>4w2 z@`8eXuq5(@g*BMa5K&MB6qo`=G61n2i=8O8q`W?_MH{B2pb$;XrkC|3UEZs?MZ0TK zDijejuv_9qQ@Fv)C}~uU3Ao{lT6@8mF_KEl$aGu%L0eHW`Dl}l)$DQk@ zQ)uWcL?BCXnwYLEhP3yn%o-n_;`Zz%C~RO~Q3tqoNeD&+oIStLuWHQSV%~YzHNJwC z)`xRDW=C>q#qPGp>BO6~16z`LTbXp?bhmns6c!YmEgcerpcDLypRd^r1HXxoMsGnJ z8R}qE)2oFv;y!fG+o|uWf|nfvrRRIf+idvC_M>E$9B&QZ6cmsneh*m z)0b3s58KQqq&OlGYdyv1dRGpDy(TI>%N{)9`kAytOKkDPx4J^X+Hb~rHXYW88Z=JP<*O5gA~THx5!P}QGsI^v)taHH>c zLPbHvy+C2rs?J+qD07p325h)w6&0mjU0uUziDIR-s(%C53yP}`m;KX0x}6kQxmiDqSS zFMRvues)Z4!_sU`hKFbB{KTmU|^_bOcNBirYe9PkDKbni1GZC7Am^2HjAshB7 zWV2{FxcXs>uVzDBju;I zhYuve(K*f!34Dl%&ypTaq^g_Jw(EP`RQuS^mLp?q?)FA-eXcgc@1Dq?yE<+k8f@m% zFq-W)8;_T@`%+1$+HPLfa zcB6+26VHlg^aLa83#7x{>XW1xB6l0G)E340eaf&+%H$rP--wASv+CDF6H1bE{wV3r zBQ%cywKFhH^WkIn6;m^h-NSmby#`E>dMwdSU*iw1`QGZEJ(1Vmwzjni z0u+SUK=%5q11cmV-q2Wd$jfp1d!4$vsJi^=S+#2u?$^n?Ls6Eit)|`y6G=YylQRDv z4WNm}N29liAVd2|g#Fi9?8>8Hw_o8COKzR*KEwS{XE_NUzmO*mis%{#Z}8Wwc^$Ht zeaXe^t2!;u?t(D&dF*r(*sgdsyaSgrosUi>1=0rTo2qCmE-uEMpe)yHo2VF)Gp@`> zy)x$J;>aIB1IS{=xqw!hV)|@FSjZ_<=5s3qil4Q>R3qHm{WPq(6zg^C?I-lDv_jH# zRs#IkussVB1_rzWiH_U$(}s$Z9}hxJCK@g91}dDG1l(}ruIfWdE5i_^rT`fnU@1P; zZR3ATC{7U{fS?yE=u?wy;R;}_nOacPQImP!o*cUcO1RJG{|=N0h%P+^AnXYH6rl zdS#ZF`{J=58x#2b__oXi`4}`%K2vduWRk^bqGKKX_RXPUhcC| zFg?JO4zP(D8GS~Pf##l-8XD&x?7#s;KCl?gE~wnV;u&p#=~-CdCPO!@H(!?n01t`? zzf36N6&;Ka8$2#>Wu6EHPV}*@)m0l!DoU{f-l~H?z21SaqXTySG-`$MPpmJd%9wSR zU6KS?*@i)7=0N4)*Wh+l6VDX&pksf&zNyaY^4O$FOgn^GSwmU%v>*%&Ls zi7{{&Mn#f!b8v>5O0TyQo#)bSZzP(g^E!|{DBvvoYAG=}!`w*qNAi4M9a_e5oG67^ z^XSftcMIhp@4ZDv?du9c;{O_SyeEMn>|$Ey_t4zYVGwb$Md|}E9R(u{?F1*JssR2U z`TO_pU!s-IfoR$qyX(9bC76XgPsaBY{~Zx1@Xt?BOP-g&I)uF!F~k;_Ze(d7Z41b$71i&KFLfsd}Qtn(#4&}C`kSL7qa6Z zy3qLX51h(_vH(Z^XxW8;sDiiYBHbu}`(T`Dww*q>d*YKY3M;&;Mzd65Ndcb7n82ip z$7kHhtRLE&uCKIZh-W93P*n6XoyxObkGxNc9-*bqz#S^%q-t9V5vYg+Ol*w_MyJf{ zmO>EO?OTv(XBu!YhWb<^gwUi3|MddMk>3&2Ig8*FjE2b)dLLw}nWh`676lOTWKDS9 z1Lhxqz2`~o({afP0K;4yvJmh*L3cMM7)mfFnp_2iCv=`lP$C}z#Pjxq$)Uw-3=+No z=dUf*a8kWCS9)YBk&KUp$Jg;{8;fugStP+di3U>|hzF!=IP}Ct_al<>65nzlz(3zMW-MY~JQ@qZ zEUS2YVwF+k80=_V9*}OV?zax?Y&)D<>Humqb#BZ!JMUVN=G?@@z{^O@yu{xWZojA9 zG*L;g!#^fiea1BdR5LGJ`44Sg&v`(oXxq(^gkK?AE&t`_>rMNa_aonxGed%+5jd<* ztGB$rmE+zRVLtH+vd&1~(aDHWgRrgO$q7Tfox<(^tc55DUI_X*_#VI+ zr463p`wUm&5~;Qq(o7pa&b+D785i$hH#9^XbZKav8Q!dq28uXnSy6^}M%R6~t2f`O zD+$uJs$7=ih=Afwy33*tDQVJQj!C+ph)pLfNQTe8N0nurHh%1j1Wbl=DEvN%G4k;>=RnQr|4EkPS)A+>3eH%hY1N8ho+m(I z!gh*p@sO?J+Rm*K8r1B7Cup}t>zj_7a#cZ@=g1oAG}%89PUm<2g4sTOvn2$xr>Y1? zoyeJw92+#IfYzyvk?&$dgv?4Rz0k+6?DZRIpR5-$f{gxH#$6EOcUVuu-+ZE}p;54C z9pc}7*jQvxZr9r&pr)pVNj zqRYK#aiP4Nzz~VX!9x6Kv!IFp^!ve*AmBSMOLz6q4#~;MwYI0!4F?Q_!Jnlp0RiAm z6|Y@PAm4Kl2~gqxGEjW+$=b)>eu{*>4n=g4+93aPhw$o$64ioC_{$@Yr{gO8G<4hw z@W0`U#+Y|NC|A=pi1fr4qcnGub82Fs!vn9l)+BsnN0G1Sftx?h0CqbiAfZv2wJ$?I z3--?QPvc#)sK#MU=b%CHR99!kU}&mEh7^foF6)Vu;1dD#Ipp`A<_VxUY32IM2;$oS z_PhAp@Rb9Lvus7Xw;CFF=(&8s)@8^a6YDk!h4e~Nz3#f5d$6po#&jEI%T6CgfH|jt|%=s+;T#d=KoTs51pUi{+#-xCAc>B_YEHxCm|~4h0R#c`ZXZPDN%0)N_%GY43Tt|HE7J{LYST5n~E0;7>#DJVt*fOE~S&` zXt##I7@P#vgp^X}CVYsd*aRJUv~nj3WcCZ*~tPM0+J^ zL5Pqzr8&~RAd^!fPKOhDvQ3%Mrp6n!h%O2`<^BRXbW)K3TH&n3a8$sY)byUCSad2j zMaRKp6Kn-|XUXef_6`mrIJG#VY_ENz&51_5O742m&rll9ru3QKIq4#k>2iYYN8&Wx z(9hrgP=*d23wp$2Vg0NoZi6J6*GczHO`6LJR~}S{1X;j7q~&XnAhFugRBuPGvWq$8 ziBfZ@rC_KdqWKLA8`kGW{p3ZHf9t!G>cb9w79?3#?Z~M2YUg0gujEa8rkX9%$g#kKzxElxX9n%JfuQ z8z<``lbPw-LJt#Kv0+KMkjn+UF`vizF(OrJy=X&Dy3@bLB0xg#=TVz?yIs^>q{`pS zt#J%ljH+_ZSMXSMc9TCyYIC zcQlR4B%L=Z8cW5%&6mbZN{SU3h1Gb0ooxje7D%LHqU9i>>ca$*K_gzHo?PMWkU_B( z4St7{q#}0$c^G75hUxWUz!((DY}>ve6cO-)aaDxmOs7AcA9~;BclW|_o*DqT)Y-b2 zYc6!HdRnX>;dpB9HfdV$SpgHVi;B%sH(3s8<6EnIy+l+`Kffe+QO)lPZ z75p;&**4-ZCW4R~51@6gHg>L}W$#kC9NAiWN`imXDZD-zp0D8hf z<*$H>Qy>2I08%X1?EgJrUFtACv%i7yma@S)(nR)ROv9lc6^U2s>g_cg(1@=VGcD^M z^EOcZv)tdaK;;@tFMb<9H~mpcgzeDLS|rhh;hY5J(YkAg@JGI-rx}mG4`Dd^F+#3w ztA672O57okDb?3sKT1D4-&24kLAud@LA97)Q7*aHZ=xsH5<_!|w^cJpX4;Z!m08@I zBqMl&2px_HfLL9ziAKCph9)({Zn;*X>cOb_fhM4b1F&EZOqq80AX#PvTjv`wjeMyH z(8IStE3e`r7GgA2+I7T$pSI~@7YNI?zJ1PYYo!)LlN#_4}p4Blo zP67n7 zvZZMp8X=5F>D5evV#)5kdLUq?r!d%@zMuhS=BOY;VsITAwzFs{hYHO?uwb9s0;c{w z0a%?KzO0{3z|==`Kof+ux7xo~>smwk8rg)Zvcd>^~=PP;w`Md9*5ra|Q5Ajjw$OR6JiDZ$d19Ze_ zxzzo#G80vV{EdV6Tu-}i?xf?eGnec%QoSc8@No4cbaaRY;@!t-tI7_Q_5VP+Do}Gh zb~t=)X7Ps}e>0C11c7-xxmI9(L2p&<)P?1eob$#j`}ii=g|2^M6O2RXPR%yw1{D+( zWR;X0=mI`GoFk`J#iwm-9Im6?`LbsEp7-N>{3=Qdxot&=K5)@!E@=dK)%gb*mY+V% znJFJpLE9<|#d0D3Tb4{9ow2;n_1Vp`>&8#u2XTqehlL;lxEK#395Iqvsk2M*O*g ziqjX0O;0DFgj8q>h|*EQw2S6dAf`HQVy-=(AzH-Ych8Ud1>ezFY^5REr2ZD8#Ft>j z&aW{Bvhf9rwGi<&GwFuSKxaIa4B{d3Tag8A(7Ga&`lB( z5ryalJvNyv(|_zJ=%zcrhXe)D3aV`e(nq|W*?v3E`|CT!dFaBI;HsVp5vG<^3-8f! zVxHoZ9)-uZjBs;h9mrEoMhTBF7M zRXow#H*dNh?*XlZdB^UlzrkJ4rx#DNh`X(Q9l%WT$4__5C&Gw_C4Xdr5>t`W&b#|jL}X;ArBpYpz@Z^ofW(XaaAOFSK01=k&xx=jr{hHn zTm;RoFtO5Tjd;i1i(#OlzMMa1<-FXGB3jtlR2}7lW8Hm4%VuX#^Y`S;59nfv%*6@n z(aH61W##0=f!eVdZnIKfa4L_bt8}*JSYVi{+k%ZC@CU+W2x86p^G%bjc&EE+g$)-G z3BMoe?ws_PL0f6?-E1Qj&=|`5Scz2ncts$2dM~VBw&1|c;bE?JZiAW;2UxBFE5(Q? z6WAjS&Qx}>!?Sz%dP%-*#O|7SORhJ7z_P<<1l;_voqCY4v0KDMk?`< z^#L3pk6}Bz?d*(8NJtPA(a;oBR$q@fQUwzc(d3Ysm}38EDm#<1&z{e*&;T-cKSnX$ zF=(gjDhMzi%_a5Vvvcj?i<_iTI^Ge)AgNK=Zjq!Zs%#ibyxUS7%f^e&1yIzA!L!K^ zwMlmU9Q3D-Dv>B}A|oTk&D8DeD%ch6Eyr$^GG4*G(%!%M?2EJ;im!9wYr%P_N|rM8 zprNA^C5P8Nygv8`@Q(-hnoF*comKbZR%BBdzyJ+!1d+k?Y%#Bsgopug^>SJhuMO_p z;mKHd((yaoo^8I6@_32@&9Adknbm^@Kg?^_ELEJ*2Y@NE)w|7Wy?v_$~S4;v*$YN)m5!4;WA@FaHcr*JjF)vaL_}F*+NHn8=yBDZa#S;h7H2%IdfoJo25g z3yzhH-_`U=&7Z1^iqo?-s5sY;D2-LW2PktxS>uSf<*NC70FrvPTLy~xHx@&jap&y0 z*;%BU;<$lCTWIDcN#BcMIyjZF8h>{CH&DbelZ>&Ef7e`L%z;3v@VAal1zHnoQ9=@Q z$(ZhJnU3%yD(^%`?nudM%7+n0yN$KAH9%Qyh!}hC%LtU&sKPMzjJ#Y7SJlL}9_(g> z5Vh6vFP2HphIq%Kf$i2my9>Y=$S_$vcH#!-=BTD8p*)G`w7M)oJu~$5iU?JLuD=OT zQ<2Bgc^iJ5Pv{E~33j$dkqS1UH2&bN>^r&YVr9HLu&Ah%fAP6e8V*6@aidl=hV#0~&6fu+*o&&P z5Spv=c(m$%n3tppGO80Z6WgZGqFHm`Oq(xlis%C~32{Cg()l2}ntIzCqOSkw^; z0Zc!W!`Pl<^_E?&gJB2u_ZPAucVF9tk2k+nUTV8~2l=f8$m9Sc;^Qw~Lx!3PJaB!s zn?aS8s?90XIY(11%ofBeNP`zN#&u#u>Tpoo7ATnm5 zqNG)Y$^QyV$HmxWq!I;U-u($zE}+8EOzSXT5I=%ikb&neF6DU}YYTc!w<&m=be1K$ z!u+AlP691L)Snu-iUP2U(Zx%x#rw4QlkHW1a>XK&9|wnEOWn=O{6TT>2Y;Es2+u0L zq>l|S!iXW-kG*(gK3Z$n9V+uI+nznVuI_=)$U00qc~wN>)pq;^!eHOf=^_OF83h#| z6BFMV(Q_ny^g1vF2ZMzVb%$mq{al`E^?d##;?U!yu^JIm>-`G`ME z&E=bGtG@~3x$%R|m^HPI2vQs?t1vX&nS#!Rd{=mmj_xnnm&+^k_dKiR2U1ByD+D{+ zFZ*}7jsR2IciUg;_tf?Z(3)_zTmE?Ic+5lsgBPQ!fy?vekDcXZR!5%<{Mj4!6GJ~8 znR71Y#*XN?I7?t>Z%clE)u#JNDH5{ifKEj^%M0?5i_Usu(`|#9TTWQ15G;!#rZv5Dm8#;<))J|}ZkLfI+NTZzE-5VJnINYa=UpQd( zmrG6(5d;q(BojbHL*&6Iv{7~-LX=Fq7d8}ant+xh<7x|dj~A)`A`m1+M7lLg`lY(V zqb5;UChi!5e81?)Ui~u&>D{2a;O~q}8KZX^kT^0yYU*rvb8|OMI$YWzDOCa@SAA|9 zP0R0Gb%?}~>@91RLBboJPR&nXwny*$Hg6e0nf#B>G9Y_cXn<}3WW-Mb`CP<9qa6V` z2vx?{?c3>y)ZM>O_!rzhquZq3HAWfP^;ab`mr?kQ6R-uhO%GfgaU^Owgby?K|AfK? z8nYw#@^(6?)kq_NUp6?F!FfLV`UzTFnkCubQ;${Ff&(;DG{{MVNc&3jodT>&%lyVy zJ5P+3x8{g%-?lM8k1cXrt&fAX8mw{g*bJB@XI57TE6Mw*+o+ahLDXfUA2)0b$F?l!aA&s0W}fwEwT9T#WMB($ zeFzuN|M{8bnD(VNL}h^qMe9${>65Ve&b34-v9Zw|+??zJvnFP-M^Iq`Hei{edls{v z(Xcj3;TAin3{=Al$dqG&kq~M%D^wljFn8lqW!o`6oU7P)bLddy=uM@Rfjp~CgP6fL zQuI#U64&(em~OT77J_3Iw|#~S(21Ds@Anv)V?U1l8CNLi`Z<)SRcjV$w8)o^z1{|j zBf3U{X8qQP+ue9Ni7CnM_klIY&JwaQ&wwOB0w3l5OeIrAVXB$KA3QpFhEIyKO-TOj2i^4#5N9ZlP>!CH-?J&ct29gz>7F zmGWe*h(N@fD3h@)@%c2?w5B}ZJSI#gwfG-K&bC@8V8a} zT;%NCY_B8=TrA|+C4g&2FZW{rmM@wu8M7DdSa_L9yJkZy9dMywCs%koV;{+R&%~2JqQoQG`D*%tq|kq|ObovHC@3-0hCnqwnp&vC#&6 z%`ku8J`j){q(l4jHxfik?K81OT-vr-0kn38CfwU~z+IpvZhg@^`_!hrMsU>^ldomz z1?&Q1R_2IUn{oG3H@AJ^J$#rtnl${pEsxj@$8s=fg=%YVSzd;kz=v<<2Nq$KgjXQy zALOHpl+N*!=X~eY#VA0R2&SioiF^68_>m)x(6LCH0Ax?4v=?Kqvb%ET$B;;_JVZ-I ztv(z8X(^>&R4#nOv`?l3gA|N;4?{^$3Uv7d-L!LGff2R&`?K2f-Hqk2txFQ z3FpB&URgi-GlmJ3^nh2swy9~57FsPQBN!KZ$7;`oDKUb>dsCKVIfqFyu)nWwrM@T? zU>ldFU=%!FGJdw@4`hgE4Hf7b{pEUiF`C%$2g_u}7;u$mVa7vQ^F9^MmX8h}_!0I2 zgC#&o@}XlMUxUJQU5Z4W53X=3BKPUYl6O_JIHw>-G7lRD9V}lh;Q|S=6ILDt=`l%~ za3nhkCck3B?z9cMqcN>ZXDk9qC&eNeYutIr=}1*4f<;Qo=ep;^7a_lOH*pDK%FCIr zWT6f}Pd2qa94*!jeXGWBix%Tc$uEuCwW*ZRk*@ghZh2wNH%-V9QgN{CMYxjP_&Eo- z8-SbnJ1I0)S1Xm|_1`~=f__s%v-&;Vwd^Mh`jISY^;o2chss)bFK9eExwJ8XnKNk8 z6d6Y>ZYXj$G&_H%Y$u3y+A;aHvpcXz{|-fNYErVL1>1gq{lFy(tZ^$<7qYKE=j8r2Wg`a ztl{F)>gzhml?F$gESJ+Yycz3x$|`y^py(N#E_%ECsYB+rg&-$4H@=+?opLkj2{9Kp zHUWt$|KfekqjhPy%=P=%%$Z1YU1m$!RaFXdxTy(o-|2`)q ziy>TZ!-zB_>AAm>j1hE%dO>&Nf6K&1c8z15U0=7`hc8Hjm{@-p{h1P=wPjGLYT)pe zAT?7(NgcEHxQ0nPFx!E67u%v64LI|RIp%dSWei&^HrISXzeBNS;GJw`bB-v+`izw(r)|A6?N`@JcGSa{OE$L&@_PUV$HA?YkJPBi?G zF5pb~;z?{Vx%_eY)GfY$_R_nJX`@J^TFBCv0DIBI!DG7yuMF*>Dic#=w-Ox%Co3z< z!B#Q;Z$vJS_~og9jVS#WpaX~*0O3V^2Oqz4#gbXF?_f|RH#6d6#UJU}Ks(zW>5v5V zc)?>QT`=?c&Ij37Y*cd+fim-zExW^n#rQ|zf*7O%A$8$OqS|vSiomH#kz_|eTKXHz z1BzB3&>f`Uhb4x@Wg)={LGZB9 zWP}Z#qZY~Y1`3Bf}HrGM2R{%?nPx!{| z4x6>d#*=*`X+j@2f+-{bGj%&|+n}|uIqP{sU&-VX2+g^4pw`|)j^U7GZZDUzWYuUD2wX|Ap*|vMm?)!WHp1*t5 zb#8HF+j1P2yMI2PTOyUCa3o%|>XJErF1F!6Oow1sASk|b3{v3$;S7@^=$L|lZ{rqcoLwl?Z`wB-(4`a^7L+VF=S z_ffbJ-?VXOxY>OnvIrodg-n{4?ZWA-vslH(Z`^)h0h<~HXq{s0x=dE3ks z{#*|Om?eLkx^9Oe8F3IRG|bHoiyRQeqX+30GA=EZ3|dkODMo+)BzuP}yhq$NQeg-! zT!$b36(+*|r0Zzi7;@zsf=0YUjQ$BP3nrels;bJXTEpTj+}OSM@uE_SH3WmzB!Un3 z3FzJL*!)gica~)_f!vHv1d2BFZoBNY)02I~Ko1e1P}5YrB|wkx)lTyzof|gBlgq*5!ATmZJ>VbfTiD zMan}`T)B;d1Ae**3FX-sxRp=fOltFgQibPjM9dfyQfc~DkIgDT#tUW8sy0GH6umY< zaGYFoV21do*rwq0)OH7^qzmxf+h*_NI(yFJk$GLMB-lcm>S z{AxVx7)=_4t(mEH%1q_c@pXz-K$qc8QoIF^jA1f|!U-^wtVJl)2tTaQ%a?JM;gRR} zig_7otHn#lGh@>YxqS?i#N%LK{g&(m^jO<_dwiwfO!{q5pKXyn1T7NhB6!^lZ8&SEBMm0cwhNb*B*LVr9Hz9!L=H@(38s|e$ zy?`enmSG6UeA_J4h};*R4`3>-1D4t!0T4HZ$L;O6zq4Qey&Dld-)NLoRLB;Hn_{0g zG4bLyqu?~UyfM{y=D)G;LoN5`Q~Nn;TZ68cVHo@l zE95VSDE`=b9*Y!KEE>(K!un#T_k!@rb+-v)eRDo*{B*UyG6P@|-O!!WiCqGb| zoWW|lLPm8jGB=b@{tBEh4MDNUA*NwXPpA}#D2tX6?QB>3W0bh+M!jYwrdiQM+W)>V2|gXO?aq(Y>8g_bU%3bc0dV6a3|xt> z5jegxnF(z!MHr%p>qmlIKVGoR&-W$e27GYd1JOnqq9_7{RIqUy=Z{qDj>F{`MmCopu`k59lP05;%)4?j9xdH6c(sA{|fa)eqznbAd7P`Sb?lPg8$X;R30k1r;Mvttpw72mvt|T z6!k*%vj$O8H*Z*J{xksH(hrCCO{-C+Aa{s{0@jUgLssH2+ULJ%QbE5b*eR(kM2+Wz z4@6otNR6GtaK@vPn<7Anlo3kxWc4FRo!SUoDcJo|3g(;hFCG99qRod9U=JC+l+CTr zyM6_L1m}a{IoM2@z{C-riLe9NpC9lva{Fb=NsrY;&F{;MsM((X46CE2&Q8WFj~qG@ zRc0+f$FWPSjr`X^)=T=|V_?E)VT4#%w(n*X_qq^tAh`!TmcEsi!t?O(RIY9s;&l+XGz_wJE? zEO?NjyXBr^{BMXW>VRlp0S=_^32mumn5wy$bO0W8JzOb|JiJ&m+)!cH<*1ElNcx&( zP%mBYOY)uzz2C*F!^Pnc9BUWLn+9UDTv~(#9{TazhMM5<%e)pvV7uxBc@Kl-SAF-J z{i$@ff6q*~8e`uToPmP_`+ZFpqIenk-5+oeiHL~k0hCs0#Y1_#x&f{a;QF))ND@FB z#tu^S_+3OBf{qtz%ia*Bl4BUqI^%D#atQX;72p?>$+m_9_O^O{?{d7zh?j;l@} zWWZ}R8dCwvzk)_WFY$2K9N$=!NW+2lq^ICyp#c=HZb;d{WE9OAGQI^3eRuMG$4*qdw?)~ zHjz2w_g{_{O9CcI?O)_6O?Ma*L7X@eC$!Yb@BovtPmCCT_lso^En{w^A#X>&lbw?{ zG2IFLIdyKj;`YOMQI%fNI_gZ9~V2t6Os({+SQ=N za70HMVTMMMM`F_!?|U_;ewLJo6$61lhVFlNj$)(FL+wGd&q3k6zq#-jF!5yBG4_iE z`vkkXPkwRHBGo4PG3Djq|J%a}??HJ9Krry|Xt&_ULu@sLtzxf*NkzRw0Y)Y!POEt4 z{@V`YIYAMU<6Xg8;lh8nKvYy!RW&l#&Nms34d#5JE1Mu1th-Q2=k-iV*a}qobA6PO z;J&}O&C~0#SZ~%&8szIggm5e?5d3cvZT84}c7r8W`#5oF4G3{}*q?LA2-&5uiBfntNuU zzLX+UH7){W3zn_tTgH*KJ7nqyZ;-JKlA^xs0PeCI9HM*+oMbTtI8zW0d9NUMd(J?0 zT!bYm`hOKi58VEZ*YQd#qREb67iL1^*_PaW(hd&6(} zFew)$Ai8mL;!9%gyzC&odc06K=vzV|3G#I3>^_5VIO?d&l96gAZWZkR9YbKgJZLgk zfVOgcV8l`hD$E0p79vQ1#X6g7w7jl|=IqQg@uw+V1YDqkh4^43(pO>`i9*R>>Q0o* zUFum5&ToKUUo18U?Xus8CSzoNbz(RKL};N@1oVA5hMxF(n>b9|@}pI0J5a=uR7(DBb7U2o{oI zvi}R+q-guDhy%a?dgK0Li`(n^M*{3P_5?pF9`n|4yn8WE<9~$@iz&S2f+^4|1$Am8 zs@OtjDxL=Cz=Qm9i7l(RHjiFlH#p7t&Hc}`rFUd0{T`S_ptgDdM44lqn9Jqp2C&A~ z$0sC2Ds7nW#^-4zKl%SmD;R`8Kx&+j#|!P(_fO-Qf7cGocc0N0zNSO~tY0Hw77Ygv z-%8QH=?VHWS`H%FKv8j}82>QGg3BSwGm2_#PgOyUC~p>9gTpCmmS(h_u9859?O>lk zMoYUNcb&n+%MR~waTLpCAmwR9e(5xmn3@{au_7PwYtX~x7o`@m8G~^xj9aQswW5i6 zf;=3uO|2+W&Wps07+h1!qt3uQH@|!ZB^%6k{*UVaH@pggl9iLY7z*P`)F~IaUZR;&Gl(jSa{8ez zdp|fp&h^S*t<4m^foa}1o|NtV_+WHsB1DNk9y%`y*ufxH{(%3#JIOK&XR}xb%@OMv zk%bMD;yotb5gaJXIQ(^SL%K5|{D1ak4r(`xe~fEM1qF$hhN!g~dKN6w3T{Q28ZxkJ z49Y`8>ESD+(uV*ikhtk6mhQq~~tvHcYOx5m-hhpV^Hk`fv~HwjDr zx-0hp6-_3k#eGK%@ZeEscb4SN{tG5I0sJH;+Bc0P0AHeK%8{Hhv5ozJ# z;A{f488*R(ZOnYT)0Nh1k<~zqlm8;wlxPp2o1Hq}<2)lT{qd{4n|X8sf@r`Q6tU9N zWpt+q5sjWbom*h^-~+Whve;0HxSnEH#MiG_^4Z+eR_)7;SHL7?FY{v0g5*m#x( zi_`-;x=5QH<#Cf&OA9Mfn!&tj(JlejGyhPBOw~PSP`ueltLvEfTnz_6fS*uLX*@>} zRrdHs7pxB(B)7iBhF8H)&0fvYf&*}DthO1tAZ=apUTXrz1eQb#j}|UIMd-psgvyDY z3bdVbR6@aT1I%Pjatxu<)RLkiY?%W$C+E7hoTi$b6snbj5n#~d=|^P|9~bfid4J5g z;ar%9>aCH-T|l-9r)n~H7VCxV?Cb>LW#r@_pK|vV%6ewvpC50iNF8TUV1p!%0+fJy!uQ_rpRsb_k)OJ+;6b53tjD69J%r^u#H$(8Tc6=7ys` z(B6kh|I9Q_;JE=jGh*vf&fi|U{Gz`6HN5l+67=cA#_sN)4dF=fm_dCH;B6g*O z;6|TAe1>1hF30Se#en2HX_$S6L1914PJMr0KNwx>Zn9WscCvH@S3UY26`h=JXzQ*) z-12hrYfTNyxrqWb<~@H|p&osi*xFC0;eKLUPpAW+_;Y+*VAVQ0>QQ6I`Dl$Ueg#1t zlw)yVOI}cT0boSl;&@(@7fu(w9(G#(Zp_G zjK*ai0`lp6rMmI5$-dzKB1>St=r^{D=y;*01wYdqo@qM2#SS090tq!*9zL>!*_y+Z zNW5n3?Zr}g4{9s-UZWh_M+D{nhO9uX+9Dd@|CiVA5{~^B6+>N-Ogh@`NBm2K-|HJ+ zBn#st6pA%i)3Oc}aZe@VZpDNxKVC-7g!t@pDO$GAglAdxQ*a>2Mw4EA?zI<(i+(hW zDvBLe&N2+%j8_st-#~1w&fNK2*RNF?*s2HFsDO~HPq&R&uu{S$mJ@dX|4srO=Fwmz zp;lF?iEeaflk7>i4Z{eowZh`cdIgzNrk6vAsnRM&~N! z`4;55cIw!1-wh>tL$B@$#zODw(+nj|O-=3fb0>gJIKZW3gJ_lj+v?n-{|->d!>hYr zbrvpJbo{o;u>oNZs5~pCua$bOb(1i1xc?Yo2jGZ+B8=7?G>|Lelu$9{_N zaES&Clk?;46p1MYk?%-I#V(}EVl{sdX{VJE5KrRgkR32eFmXc4b5R%CJ=o^8P_YSF@>GT@`{$tT##Y zrA2@Crp6|B;DdLe@#CpF0ft}dpJCy>k($NhC?R>lQ09qrOCh>hU5v+tN^xU z+A4M2=6V#eoB$}3{S!ifpIs+_HN&KinI*79qeTZQsMbhO9>>$WG|r9ixkJRt@5Adf z;V>o|GH#}=^kZVQ62d4@jKkpgu8zlj9?|h#R+on&=-zFYetyp5c6*Fr5n;ni%HtcT z{$5UnR;xa7*73j@2@>$>RB`?6cbtEysD6%C1fpE2s2b%aB@?dj%ZL1b%(b9DCk z#-YYGW`#4?TRDqn1ur+nZ4Az^=-Ddr?Q$2Xm>#~)l`o=+MoM~|(qn}32X_~$Yw_SeTfZVFXx%OPpasi-44Pz4b}{OdPh<*A zdJ`)ZVnbag5z1q%8H;)=mgUp>d5S4~^}h@Lw{j(g zcnJW@3tq3{;Gdk&mMBc`%!hLd4f&<8Z(sWRo~byEu>{)I<3&IKH2+YPmF+_Ir;@UtG|f`r?%E36qb~m6C-dMJ?MV$h(NA9 zKZzC%3v0HCXO#w`Q}SDs_Fm`3&=D4ekMnal#EQT_a(sRE>|HFccWcOd z56x2~@Yii%w?L_aIVtqUe@{oJiP$W%4Nv$Ev2F8$`#;_auvof(dfLJjdh_%~kBRl^ zD?97qGNzQ@k@&IN{(*tH0ibQt4rL;w)5+zVzn+x!pt0}AGT2MdU5_sdY>dUP|J4f* zb;P^?Af(3M_WW8@o^?o?@$TaD#K;#CJlv;9fu*fOC|I-2*vN>(LuzueSh8+aKrv@< zP!BK~Q&G;2TQ_nNkGaMka@ngeq&)fBsc^GLubNJ=o(N4<}$L-(a}kGYIC^g zn1pA|8H>s5vDAN?z`cxEwUg$L`XYMl(W1LUtK@1st)%kIaX%)&YBf`|yeh!^hDMbD zAc&XBM0ZEO^#-*|wg6mXWk+y?_73*6=d%C(F~jHt5Y<0q+|7 zzk)uG2PF$FjugZm>wzMW(J=pmoW1Y&fF8TL`vPV&RGNj<;SYqCHKa&dJ`8z}xcE4UQ*BmWibq&TW>EKZHow&5qYGe?}o$AtU+wA)L zoBzt3llW3b4gr}4HksGx{+F1Y6PL`qf+YIdI-m-j_M>TXyBlJZ^Dl;kwNcgbD6+&u zEp~yi#^O|w@>1dFMDvi45Ens@TTZ#j7IL`Bt)bYX4t?fX2E#5w++>UdPj2+Jeh4i= zmPy4)o!lfuTMcMx)I5d-ko!^sT|^HfnpTSg+05_XiP@k5&1Kc|{me+^)}MO3CO%Fd zGm$Z3Vloj@RD+nCi$zXx0;aciH_diqHVzgaj+ToNeU?W+ETmjq&CA`9wv!+5@ZkIi zkRR*dwsZj4k?dRuce}Oc2f@tDEXDEf9}dn1ds8sCZ|ZH9L^iC& zwLV-B#~n*v&@N@2CF}nhZSpHTaU%^vB>)Lnk#=%!EG)WO38A-OLN(bghn2MvhDeSo z>uW;AHsMd3(2yI8gCQUyE7N98X?+3EINCB=1{#HmaWx5`g&3WMSOd7byrtq4_U63P z=%1G=g_1@)m)Hw+?tB|b+QqwuOXJRVC+V4mH3-5`r# zU!uU?axQ@N$HvQDuXv5geRVWB9&UGBE@QCa2-b7no26h@A2016wDwoOtTIuBL=dlLflwfFZJYT3`<`EBlu*U&su7L=^? zh>W6nUis?Jb%(CBe_x?}1~?bAPg%;-gq~O*1@He_*a`8Y$!R7Ua>W5W{@Yi+vyECB zk`nA_F42PF_u9q_BGC_B3y+&%kXAflC2t4G$(?r>Gi7gDt8%d8Tf6e2S}bpL+$B37 zl>{Fw??3K?Jl>)s^E)3v;eV#!%2O1;<}{7h1khPXCPmh1Q$MR$NV+8R0D9>LV0lk1zs-JSa`bt-0?%&bWf!8Z17n?yA5XQIP%;+9)pO^( zjt+fK2>

Kb;Ws&(2CFu>(s{tVS^;R{7+AeG~BeV@D%3%NG=H(^& z_3Ia(!!inBjD>)t=+U5T3Lg&ORRIxe3u5@N2Od{hqEV?@Oqe;9mAuRUdibdPYf!ct zFPY=Nt~_FQwR33Fpqz(8Cc0KDr9Q5QzUxnq*)G zlhpJ9TXA-!Fgk3xm7N^tE&u?GjSJgJ(~DFAXEd1pDvaI+v;oY8>+0$@@S(0-6EL0E z=Q~f;z{`Sm*?j;+8U6j8KrG}|Ty5jxyaOto?`YX;02ItKz6MatD=?dFYH%XzKMXT4 zP~6DY`t;h;B*2XNtIKUN72KU)1)2ki+t7_wEoY?{!XhLMfk!rx@7Mn9%hfvlV#DB- z+jtayOFBeE(PwtWWL6af!_R<_xHk|IFYt!D%|4xDMt%HsAi_x@94(ZJfoVd4eWHG& z+{>ltcuKM$!!LZIn2O;g@;C9EFYn%qiFxc^f95yf^?d*EtJTKMs{5F*+3U}TTCyd( z$;n9&{IxcSW&6OXc50jJ2|!iLXRihJ$$zfnXTTv&sV-zHmQ75kM*_VO7@Y-4)Z~zh zi|eR@3q-gju7={x^0R?}84<4=OaD4$;0I`Ocev>XN{~We-r?TKiOrC=riKP%>#Euh z&2I?Nf3MP6h`v!@ zh&~g3A?(AdLZN}8oPV*mK!ira$e25~=z|3Fn2*o*WTWKqTG@QptavHT$3E7ey4qL2#1?&9UC0P zaX9k=*xT3C0warLzpyo)O)e$`63b`{`S^+h4l-)*vQ2HUhz#I4%Mzg`7W|kq#V7>T zPo`ddRR08&X(~CCHqg46_cqdKCQ?4;&AuodnEgdj6!v(xIr{1|)TqT3wkjP|RKE0W ztK)R)Ja~>G{TQR^9R1(`#=!RH6U~4=oO0h_F6mXgw5KuB0i)7){jS)@$AwQGxP$0) zXsrA3l<3VF4M3kYty5DD9S@d&$v~aJ8J4*Qw{IWlBHek_9WxAcAX62~o>)P!_i0Rx z08Z|D8(S=`p~jlDIbRXBH-3s}WWHzg%j7SWXFboRhZiR*O;MKk_y3-R`DS7#sa-x% zD?KdRxp8xXA{6vmLcV?SXi&P_)Nw+FqJoyrkk@%O2fv}VZd0218&zWPu$Lnu8W2R# zrJL6%*7rYonOuGiz%Eq}+h}4=$cjhJ6uVvnS&3!C_a0y)39d#)x+$VT92!(wTKXLg zHp7k0(I4`Iae=>_UH^?|&`kY41An@bR`Zj8r->PbmUTMN5xp|MZ_B$C4nbH*HFr88 zC0cNNw}9xXi~zo0E%>x`*lgF`8Ubyy;9(44QKGh6#0#t0u+A>)+m7PdT$GWbhyg6>Vk< z)TbcF2Z7HAhjri)gaGl$@4EDaAd{)#&r|C9P2tr!*t`7^EW4EC;m@~;KYf~|3uoC) zkD)~K_0-z7|98fhaMhiT$~G21_FRcnx?(MfEzG_F+s_f}HwR~~tQ>!G%>C%$o6jD+4qSW)J91T2;5_>2^`|Jg z&oBlz2x@5L{deG^@N$UY5)R!qOdXQKBG^Ka%@53_AgL21+M01RMZw1FZB+3l|3DMH7hcjd$mIrd}P_eVRns~($lA8 z?O0aUbZLZS7bB>LT6<1b{@2IJ$eA8KhAZlu5*sv=kKTP|#UHT0E zGuNnm`)mDq!~93b?dzRV3(c&aaY~wnQ)9rIyecG%{xY|d=W%zsGyL{Zt>k3eG2_VO zD}@-OkkvnqUS1T|qTd-)g#~KJu}=VOX*v@UM*B3r?Fs>F4$S}dci_ice*KcJof`;9 zIep4rRIwUZD|owL+ZBEsOq3&tW!~}0D2GXjKAq!A#a-%tk&yfkwy4QIo+IYTSo#)y zJc1ctN?611>MW`R^EE zZ1TbnFS7Ea+%D$l9l?08!w$1WFq5gEqL_;y8e#NQB_#cf!?w>f0jHE@DJ) z6s&En;GDLn%JY^0o~p8`{RL`?#&Fx0eWIfCXC`wy&-s22z%wvmd+)k-&A7dBR-cJP zVZr*TIy(9B^WqQ12X?MS6tRsT)bC7=A7m4Bv}j=XjD1|$T1}2 zfDDWg97uZ$8dGuvS#(Q(0x35jJ3CkalTU>4_6qf^76CvqWRqC``tn2tIjfAc8tSi# zyH{(Jv(aObADuzV_(G+V;GexTLhQTRWvBJ2Rx0=O_$t_tJgB3x%EWc%BjAG45&C8K zpXq{UjMdN%cg>8><)GFkxA%N^;TrlD`Tpg=Y=R>sD3FuP6n^gH#0~JBZ#v^Tu(lav zY-J??>!YHhqxZfI+=WSG1F_7kSs#!!=X14K^4bAg&al8$Cul5&W}RDr%*LQsm3a$AYpx}|MIbMlN zL==Kd;RY>V4<@xy_~94?v3`PWw`)H7t`DlVlp1^;5$wn7j&rBF%dyAG+IGf+r z2hkJdM`({$o%;O}=P15eo=$M+J$N(N$$8hbF(Gl~sGE9xohI@H#icA8>n=DTehIaO z8yOuX07zKo2`)cB|3^N1>R?Rra9CS3puBayL&a0pn1P9rS&dUDn>`j3Bn2h%=C#4B zFQ4JoVbZz)Y}wrERY5C@@|*u5u5YS$`L~Rlk;s8T2>tFm6f!s=l0Qj8Kb?@|$-I5q zp!VtW!eu8C&2Z&O#O6y%EHCQddclar|7~#;R9hE*E<&JCLuKx=M$kxAJvJ};-|}+7 z*TXBg{|24e*Z>n4POC2|!Sy`ZYHkTahe2V=Dx)XwKakA8q)@jTYS4igdlS#N_5*fy z_8CAr7?8^|wy+2T7X|>{Y)Tn&h&|*5i=N#kd6sluHFD*B|4yTyi2p$f^;Yc)Ui`Pw z*T}S|T0E7Me>|%#v681|^=O42p;*S9h5j>x88{3qMNLY#WKc!PZ{N{WO~1SOyrlfd z^*n9t9Xf37F&^JVN`FDs+4m2?5$e{Q0Z;NL(0X_C^xT}a-JPrd?f$#la`kDP;O}?R zT!CIiy7fmCoib=;3DAZ5Eajx%>b{U5vk`FxWkI|9Z;jN=$g9Iu9A_6>w0EwU`|(uF ze(eD6M}MP4;;9+i=EJ8-rXJ&&|27}rU(JQ`xtGrunWaH^92C1^rBVC)kI`gtMlj ziW&6Vjal(>th?v|*v2;t`9@~NPo%H0w?9#dY7@P=#EyD%XYcg+;14@eWcr2~^-7jk zR$9i!DikN@{@ysO)dG4c8!?AI90eG*&~+{*H}Or+e$iJT*iSY%zLfd(^vAY#L7a;N z{Er$M=9~ZSGJY(?Oji&6+*7QQ_Fc8%U+?P#rpd{}pvT^voLy?JL~M6KBOD^W|12ZP z*;W^0g4D+dnlv)9+`J{U`}^`bLlItKwi{fO^+H_-chKeaQd)WtNsQec&NTxnh;lZ= zhuul3;~&0d_;=$kNzlu4PBfp&ht?rud++Z}xFTYocl86#{vU66*t@$4=S6);m{@z- zQmmU^iC-?}*@)zM|BFz;T^!`r^ z0QmFbCdrT{(m$Fdl{w#uavt|}I)Wezp~s44QPiwzW5i5XD`MmKl(Al3l=i)w$-?7n z^kR?h)54s!{n741cy^Neg~6rxT{7|P!2t)7Xw(iRk}Zf7GN3ewGLnXix6?jwz_Oil zE(eyXg`5V)q{`9sx6Iy$sQi#06mZtBv)nKcr^GX+1Cv_Y>P~#aT0U6t-EPmFn4=X* z-vuz_vQ&SQ0{Of;$Du0L>8WV@g5oDvJqP7#0qI4d>|pfBuX3y+AAah?FOWuT5GUBvq!VqE_%~M*+)S^!gw@CogA9 z`LwwM76xYc86XagODuu$N;!o&~CR(tbKtHs)-yahM zV0)r_!#2LYFuJa{^OQJcbBw9Of%z$N4d^aeMhGUkn4zKJ=ZZ_fpS2j?+&9D@pdVN| zmwtKS!JF4@*|1EQ`U^UwR*PeIisD^c8hvMhYpS3i#+2w)58Wu43NcVItop-_^-!XR z$`(9CN7oJUlG=x(pAn8u>(g3g<~V?bE*LnQm7mW{$fk{Q-Uo6P;>xBsar@rw>j(s5 z#aY5>q(lGnpXS2AE)D?tCYGKrYp{E{K`x;#rsJO$75dL_ ziyJV#*DAh?N&NOvjCc7WuE%}C^WLLv%XK%Z8dsbewR$BiWXo!DPW3C*rlO~+>4U;& zXms?mwTA$9L52+$S>t!-0(preYJtFKb@rv&kSGF?DlhCUmLnhXU2Wd3G1$Vd(L%1m zrlmy~o|SGRh^3{5etlA_prUygRLRGN9v0zBwU;(%J&vLqcDF!NDpan%kQ2EZVi5l4 zxm@iRiSP;0L32HIk7f+^l-h@L)$;P*O8E+za#5twVY1dF73H}s^1aF6H;Z`P5y%Mj zf6ZKc?e*xUQvO=M{W*g0s-6h3ZiTGJh=r*7vlQ?+@em9@m)|i_Q8se$q`pK5J)~i0 zh+5@IaU_e^D;R;v2|*RncNUvrfCup)hSp$L1zjOdE3>A#(I%qp2KBkjtoU^c2QU@o zsmwa!R6T9z{&cLQ6+0+M&uTgvk>(H)Ke5|!KSl1};tQ2Yjf+nF{@u;4Hfjm>(7mKLZ#w!~ZT>3x5A0UNuc5*5he+T^g60C<0jN1h>16W{O<-Ps=+)7Gj zh5-VE0`C-Tw|U~W`bKF_L9;n^|uT|Hej9(Atii%=r|Qr(g~PJXVdU^AlNX%fQrctByFQa~*|t zmUYN?QDhEHtS<{mtyE*k&{yrzM4b--Ki5}KzS)E(kuKMz)6r#;V*MA9L^oAmiOp16 z@D7EKD&@=&kOZs?L++v&*u5}+1q(AK{oo%U!NyZcSXr@D*v>}-b|uQ(+}u>fFX>rc zFhRuqaE0+0re6%u%4pETzb&Gj0W~8mK5)PnpJP{>ZCO_v&t!Ne{$G^4HUr`JT>9zR za~X?ONFt7^txW9sSZM#Cf?618mnlai9S)Yf1vnnQ1n zv2H6Gg>~Z~taxiTBPzvqmBVRY^Qy=Ov92<7Rv~xp&B?~UrJ!hakGuLU!o!x*#+(7H%7u}sMLne8$vD-L~aK8Afxqf4(&-rdA%#rV=g{eDE? zW-$y9`_mypo#o994E%w#%FWYJFOmkHhG3o(z-n z_RKS3e9N7UECkWNNauff+Nw{U7riyR59s|A(c4xC7vKjEpXPs+{mUVT;pw}=g(RxN zaOds@`XERo-6q!7Q6x`I8QA%e*MQ*CK2S+FnyqnWV9q}`p$hy8>%u>HAC>?k&^-C< zb$gQ;f{2bNPz5cLH(N}B9U!_@RR_HZ83%@@;%r03IG*Lhf9_H6)JP3-Y_i^22wT z)J-wS-bltvx}K+fzD#wNVEG@QbJrr9i=P_X+7|U7*1R2MGWmD~*xml9cyvv&7>|eJ z0b*NsJf*PnR=A*9cvedXjSwqhypLy#Mgwr|rB* zR~$ZIi;GL6dVb;Uyg0J7wBKqLXqJx~xjuEjR&MV1rzmvGuvK?tYc{%%P$E!({v zZ{5qwD}EQIr&Gu^I~|-wk3H;bT)(h7LPd!rTH9sTKMf8>O`2cxgNzyp5N1|g?$jn^ zn;b+;!=zRBmpWT#>MEAj!|>=~gK;(}I!w@#|=r7C`@qvuV% zu2cu@l{n|~uKJ4$IS$eqmx6?I+;Z;fxl>K|$;}!zq&%)m)h0S5B!ufl4wy%B&c6iiuEAdL?qJc4cSQ-g^g$S_dAg~%(FopB9-!pUY zsua2$IF5E|%d1~#c}N6Zca(Jxo@bKD?eDc>Id-M07PveY$$5`efZI5ssKuBlYt2%8 zd_KrGGW%M~_HjVkEh#@7+gRe100?Jck=i~=*VhIZJ7YOiDwMV&>`o{O= zlANcW;Si-yY2imvmI_X66ab~_0By?OWY0~7-qtDI1!#I1qYpA18shbQuL*#9e-lWm z6&1g{em1ZlHX(*Kg65`8j=D>VO#QcWie-RDE_XKW70l0mV%!5s8k1ff zkTy0jwSG{PUlcT5JXpL-ktbjzX#ih>wI#VxGukCc67_<@9RV;6+rh&=5Uic@OWUBL z(avRHY5y|$D)!_OXLRNlx`Drnn&YaaT5s( zd5WBnrBGio@#;CpNi;l5n_IhF=?vAmNIt8tlObVa7lZ>M4t;zJZb`7G6!3@UBS4>W zaqWTP*aA3{ZvxU15_rOlHaq|)hbioXu^25OniHXaD3%ShW6B8B_mh3)8z@uCiCZf~ z(C|uCQ&locv45uCRXvt><<12aec&BQMUa|Z=djS}ubcTC1wGv&ALZ9^omzEH_KPY| z@cqs;U4HO_l=QP{Mdl`%YRbnb5H$v01%~B)E;%X-ISsGfn~&!M?(akQwNSz`L0)S0 zJX*O!Ef4R@(!Ye_aihNC)(bFOytIL*TFxrq4UB24N2 zy?McG_ORw5Ntj4cZ%D$Le+l49n%r#HToO-v^v$_#Rc-z7XqMlE5@8Tx70fa9Nl2<} zL#60np$qX;CH>^w_qlU43jt6uF^DeE#xY?%nwRHgJj;zBwi_n5qu@h;_1wTn)4ag5 z!8UM8F+&o5pr=Q+i)B|$$2sAB25h3CM_?M4Vk?j+{0TbKzkdI2xIXj^2yNi9v#M;MeLJ6Ef{>6=f~EFQo~id!Cj98L{7@AVp^k!8u>@k2$))^ujUK zyz77a);R@`?LaxB^0!}^6+J6|M9bv3Yf;F8yDGe-)XaTnB)2|8U)@&WH*gvz2(f&j{U3h&7f=be2^NzFDXLv1R1 z+b4q#6Sm47`09TOTM1qNPUk!RLkYEBece86=^=Wg^$a;%Cj6)J(bPCAa9{e`ew6^e zj8fOvIHs6TsboP_W#NickF9*|-$zgNg4M@yvH8n-rv99tPp%~{O~KX?BNrxl8TBBz;R2+SOqn^Zd#1(qb~;~ z)-?L-HZeZkyKQOqhwWqzJo9BHwlpLN^;<>A<%jU)a+Yqu8fQF3_X2H=GKYVQ`IQb6 zqoVi7JPG=WE=d0MZGiTkxnzq+vnOD95LvY+$@`?;d{^1i+gh3Any+tU+4l}AXkvFi z{b4Gp7Zu?UEB)P`*5Mg;-Z>~6K6IzBXu=h7^^AtFAk ztK`}zZ*B4by-@rG6Z`;}j^+fJ$to1AT2R55!7T#l75w0 z3@@y8rP*4ao$tTdW?67?XYi=wK+V>>P?~<tlh;0JRcCgO9 zkD3}0R=Zh`rhW__(|tGHBNoP7^|{h?ceB8N9~c@DlF&KjcW`OXO_62k9q?SVI-BQ| z&DFDKqA+trSH2W-)Uf5#C@-i?-*~6JUIv->xnu=Mm0|How{bd<#N?SccZ0p=F#;x~ zNI{CwTC^ty?o~X!V_mH94Zta7V&2G=d>yMrPrJW;Y;+mCa7^h6zmifu$m($t9VSPT z^1fbQXKYljuBuv_j&jBar8tUyd9krXrQfd{TeH*j9*%Q&ld?V(?qhzM8gu5bjY5rxDW3?Cf;zuR0b)fkeNs zGvBQFO2_zfX!ldJ1fc(;ya+H5LlK3z_R01NBc6{x01hM&-!xgubEr6V;rdj))~ZcA zqa-CNYI~^u_0SDp-c`Ym9(I+<$rm0*19_MF)Ac8|3G7+&;z-YV+Q()vu#Pr!HU0|T zP z-Nw%F&@pJ%s64_KN`q|E>2Gfq&`M5;{#CO66ZQU`6E2O=tNZS}Q^p5;fQFYK@4 z2tu#+q^!w_VDaY~+Qiftf=Is1E2n&|GHuq9$=?gvUT4&J_%7exd`gQEuVe{Iv8}d2 z_1bfg@UeR!+cScdN=PP@E913nWEuF`6KPa9WfS%U4Ehp~c{0%mdCbnW)Wgzm&GNt0 z?JmA%k`NLiJ$?F==KbFvcj(JU$c8Nnd1;gt0ir+@bb@qvfm=d2lE232qBS)(KK@tj>@%;-=0dfUl&>lITKZHqZ-#;j8s&{VZ+SP6MSEqj&-dll$`?93 zn)|OUlFu%^KK)k3^sX$xqgC1dFS0CwAQ}q9XID}-b$SG(uTMYAWDy>vY7{Iz*-^kq z#`w{Fqx2^V6!cSN{on^)pZNc*Cf8~jX_1w#F(6+kc)J$X=_~o09r+);M}-)SL6Qkw zCW{W(sb-B*K`j~6ad*2D)oEM&-a&19$ zC!{*7uULum#q1ZYBx+AJtJ?AiqVM2peTVf&iHV6RXsG*bq5F3yR&X0#n*POC%6@IV zo;!7YR#G6gm3~2Dcw_s-IDmilkB2)}q|hGsOh?eVs%4UEX!We2!$J9`!{Ypd7Fbp9 zE>kOHr=kq?_5_1StY)H?TMS1Z3TkRD1ab!MZyf!S@!q%)RR9`nQ`x!9t3H~+X#1Yt zBRPu1B3^DvSgSX82WS8d#wQ>oBrGl|c{{AeJVPok?eBtTx3&}lFWMgzCb@lUsM7v7 z6j$$0M4xGyR;`->u&+jFGl1wT)7t0Vh|kq5|4L@1&zd1p_#u|OJQ%T)^(6c4sga{g zr{Sd^FV(7&D#4&q_T{jeL05-w>r|(7xwPxw=@0fb`5&_3zD0| zQy&(-uW4ACXzyzI41XoGa<)yA!!$@6Z;shbxI2DT>IH1IPo_{)ihVE8$?ds32mB}Q zReYh~yGyD0JmZxX7?y7LR9-e%@7b z?fkeUB=OB;U3IY{bmzZ{Ds-GMwh`Ip(o)ccc6`^*PO0BUo!f`_YRmFHj!YPq=oL1Y zTqY*{+g+mZ3SZ{w&}W4B3U~6?z6XU4JufJ|!DW-CU+BBvmA3$&ta!vKFALvv+qa0~ z?(;r8lY#D`;;_gI6N~D*ioV}I+Qj6f0f-D`Y~$L}lW43D_2fDj`n%hIRqBelzb2%U z{pbhODjGqwhXR@+2x&Ghg7|K_(3F$%JzE8^uvCNVJY_eWe|={w8oqSQ$YySlkf*#` z9ei*-R1CW=lLoW*7ZQbmNUX|iG1wuIo}fI)Om%N26*x2Qm#2{|By*=R}Nijx!7glX>iug5Ag9j zO&6|8K7FM#tjK5tx3%DU(`an;X&`vfebEL_?8Wq)$3kd*p;K5k3S@V=4VT)d+RzIWHV z0$s+ANlfeQPt?<63#P1H)MMYKGT*N_4~<867Lsi6}*1rCA@}T@-MlO}WH)mr%(r+shjN_yCNlj;ps#;tFwfmgHw? z-`xV^vdA1)dMx&?X6`G<+cZsZ=9g86c~W!(tKC|P@8bgtdb7)1(+uO^QTcp0Dubhl`tO6Te*#wR|Uq6~XNH_eW z$mnrvE|U>RE_LB%u6q9Z&bsL+>j%1>9gp1@bulZ&l;JA9Si(9eZ*>q~JagIKczfQU{dXL5wsB z1_PFvD}L6EAp+Ca(LQQ^GFo6iEVf*(W#CK9taln3()rWv^W=i9{yNif(gJG%;E)4< z5se{8$s|5IGu3=9Tf~M$zL678(wm-fps1$J$rOD!xDqlk*9~$@YVYa6h!HwKfz8i* z;%{Ik_FL40BE;qr?Hn0zlV{pq0S@WvhUUefle^@1rZn%gA*nD@HpL0d2opOYdx(iSHaoJIl+T=Rcar*$@3`fya?)tJ7VDlrXQ1Flz?aN*~u744y@O9z&?+-j5RHrafZzjTQA>-M#0wY~QhMF_&(>BXAcncrWMm^1 z>?wcZ!g2s3V&oD|W`B>PdN))dMoBw;6@F=E;Fwu++8?j)tA4ne7mZB>%fXx;`it!s zZ(cv>{3-A(ckZD^sKL(Z$vP_k?b8=>YX=anS65(Cw!pQwYJ@vKp2%D{p8R!i9`*5+ z)xnCFDV9`Gj7>mLgmvjoxr8iZ2@SuF^VS)h?Wn*VA`j})Os$`Whet)(HL0J1eg{|? zgOwjK#yX%WEgqw;W|d(`q9v>5$xJ&AJLYzTKK*6Q=c@awC&dCwMM9=YJYIT$K~&3i zreM;qtMCpW^402+gDOk{qE+Awhtp=(1gl2?VS4Y#+}jQI5W9t7$=@lia^xqu!l-xA zr4!Wu&pQ^fYD*F>-n4>o0d!ZvqZ=#2x*|#iTP-Cf){SOF9kV}N2Crvi+5<p`# zo(&vdv08M_2l6Q}nePB#C=Q3!*HX;4VqyV9n~v8P$IcaScf#_Z|Nn`nZhaJ>FUO%w zoYjU>tEsB_d8Z zZz<7}j(RH?JUTs=63#Vv=y-WP6(W5=X7r%$lOq!>kR%TCV`ag%87e`3VljFyl{xBD znXIQ89V1!PKi@|;X8A5GrjNcEiQrRF1q}twqIw&(;4I*OTIia-%~hFv}&pbS>{fdYC^VhvQeF-0k4DC-9%s#r%uC&pq zr-PM?fs6~hr}x+ggP@=2>h6LDYo>AJK35?Op|8grmbYIu^*y#V7|8QnoDhX-^&{S) zxGGd&WX-p6vzx`k<4@+MW?{vZ-)Z(529f%IlP0KsR(`E}hUUv2sFyYO@Mpk%04Pb7 zn}K4xEsAt*_ew9BkSuRYszFw<&9VK`UD4tRsN$Bt(`##41LCI&L>G1Fy2uml^@u)5mHzfGBzN`Qd8oWOD1k z#$s^*#2KMLh>T-vx6{sU%&^NH%JZ$H7(;QGi2CZsCiYc2>ZiAN#AkC?_~bJh`(Lk9 zybgHd%?QUVW2z$_w&Bk0;xfqcP1_#GcVRl$WMK>KV z1J;`=`@bt@ETMiBRA)Z)NjvLj553fPw0@AEXa8qk4@{RPs28aw`A$f7pcK&x$|D)V z2nU$-jIFbdE^Qn2b-UZ**f)I`T_J)$G;j~lRStaxG&P&X`zmvXKbNu`f(2U<_ zwQ`;v_wd)7kJ+u+TQ3UVs>4e2@4m%~BzqFMwg^so>}9}a%f)7RRI zi;xUgI$CCfp1*{Ean7u2(m-_{yNdJ70?s}Sc5H#8=Vpex7**3-oWH%P1f*`5a5hj< zdOMR>=ZQxA5=M|HCOG^{4v4Hw4&miyy;JysT{YgNy-zxfEgnoYD* z%Z^6wB)q5Ct5OL(McyUni(ZpG?zip5T-O?-F_@&bV(aP@&nz+PPpbl9j6Il3N6o>Z zP)d4_i26aI2+iNX9bKIt$`tj(;_x{{o9e%WEb7H@TgMtP!36l`aekNY>@eof`A+1| z@g>dPJ~be0|7wtp5^$p7#a#FO*m963@_GR@yBL6WQY53|;$kRZZ;FXee?R@){HPC4 z5m1Ax6#>tkdYZ?N{oxgaMG7{& zAd>dUq3$QHw7!+{wl1!psnjIcs&LYjF3P5RzeTW*)I8}L)OAhhzBjLZ5R(LzGd4QR zS23y2?Ilg>$CSewQx9S&{euE&sMFjO%j@$p$=nA+;wskJjd!MkVT7RR3hP#ovn_6J z=1(QvN4rq}U<5XjyV32!r!#>n`KqIWodEDi&`l@XxZ5<2D!*@^g@di!h+nL-z5rAv zl~wsOH`0=dRH@tD-2m~gF>guL94uf}ee^Xf1Z_r|81u z|F5y(6{>bw{&0y&_WsK`*~`o zS)Sk3wqcF`tClgL+r8B>*Yhu}HoWO$_1s?;MLt3jjleaGD=~dS<|)-QRBivuaoh9V zm;O!4md7l(VXJ)x7f0(1tgP1aq<=^6iwjC9IW!DJpHkkx4}ggra((u6ajDplGVT5J zt;d?)cM`_f+NOaSdE$m8`-zi11)dBTedE+CB!`sni%|B<@~FH0Wuv2d5GLJyqOa9| z`)Bz>V=iuyMYQAEZcO)>aA^0K2ns*Xix&(*e+hDie(9MS*&sP8)%(sly;~`H2s>vr zbx<1%lDSD=jLMsGa!;kFoAb8PIEIZ^#9_>d&1OT~AG!Cg4 ztXEIiP>JaeWUJ=9r}DXC>P4p`(#B!IX$CF>j-#g!`^x#%iUiAYYi>=u?9~l1$QR{z z8y+0+{_?|0;=XhZPhsX>Zs@ez`kTE6#)~d|388q?kUv&WH-5Jp04#!0fO

lVsgZs(96k*yV*!>01f8mhQz7549gKwtePwJJxc*4PSd*Eto%2F$X2) zP=we1Rali0Nh#I~ZMHjy6XFkQjEDsBq^OU>)OLbSkLOHGoGTv2AC)%^B>Us>~ZJqInGu7F5GopUZ8TTY=)e#Y#~z zv79B+s|Z%&)dyg+hoakTM+H8fVsX3b{&(=*o0iRDB47`CE7F+Der<)O9QJ{v&V9`j z7TEQjG?fRPF@E82Jyc`RX&AA3Vl^4!Uj6?pLtwvP$5ufo2=l6dpDuT$;%P~+-35-8 zg?xT-9uUK5Te|wjf%PRXOqxz&ktDH z)ja0Kn4@1fTod0B`Rhw{e9sjO(oOe?kK+u0%k?sS+FbYCN>!3&&}}*%TD=CEw?IGH z9=U@^VR!$M&Z}vUeyN>U_V#A5vu#*-GlasDB$nwssZ#}wEFxR)XWHd z#SQmJJ=$BCD`$cS?bR140sTbe8Us8-x%*SOOM*o@{AFCjZwpd;ouKF54ZePMDzP zWk3^4PC{5 zu`urz#nCQS-MqQg1BTsxFW(+i_!M=~k?&|fMqkupad(dOTn5%2V78-J$xvxMI1nP#vpH<emxPKR$>Y#b(TBG9+h{Qb&<4PYm zU8WA;QR@TLyWtH1w(VUv1tTLeaMuTb9visFF>79SI+M=#=SID^31{o`8Gwq?yC~Qn zFAMh;XmV!$icHxv^z~E>z<5}IH-FF=J;~~_=)0{g{Xhebd19 zG(uU|F{t|HZm9ktY2{3lptN`bq_@1zE3lZ&me;qt(;~i`HD&t%@v}~osDr_wIPXZz z;kfqXg^zD8Zx}R8j?qNRj?IwYTft!D?J-2*pbm*{))dG$~o(}8`z^{tpeNn4oyZyhReQj2W0AYmqhTI-JKn1FGpht$UtEw@`Z<( zsM~Zy^X72K{4TCG6n5gzpslz)d(_|MFiiPuM*cqMK}(I2N`PE z+1aHLN;pchJ6T&ao&?{XowG+H!q4YZ6z-UFrP-?J+R z5o$mnI-n6#i*_?Wg&dusQYAXw?1=Ir`jR&NVxL1Os52NdhS6(atpkSAuMvSu!2BW_ zA`@ju9FsEF=ka65<0buq*u{jn4xgJ{cV-=p0P%9YqA>bFtQ5&x>#wIOn&wa-2H_+p zq2#yF5;^a#Ei%XLic(Q!7|+O(NyeSyeXU~ub+S%{2?78z<}hG>bR^;ti@WHV$_$A5 zV7P^tnQHhkIV@GbD=yDwZZ|r}?qG%5S-q(pDBuz-G7+h)&WXvz$_v0*4mM7XnAD5A zszg@iscA#=zt)$JVB{gzffZu$$%M{yQ2#j61yz6A5H+BZj6)95(+>MB<0vho9RS?I zChXmAKu^|%VoDv#`|L`b-3oiNo*SIO&$T}k(CGW0<6KGTD0dsy{w|3 zpbTO5MQz)nH&J(ylF&5UNdc~J=M0+=v0c!uDJhmXTa8gYpbEWq9D0q4{j<{6ckiqD z1h7bb5})%xtl{WBe1M2cKSR0qYLFzVo``7SOzvBxq2-kXCb={Y09az*{ZDsSHEKJ| zA+$7k0Xt3>z_`I-*|u4x9(zRT_U+~G+XevqpHoLPmw^nTEOTujKGcbiTL!$clGa6; zKFHK80+!%q<>gd!UoTGqi$%Y|hWnpNe&jwOM|`<$zBzW}7_R#rO%Bn)Ezcia!XCzASu0RuuF`5Yusk}I%VPP7AqJ*F%zOop&wS)O2V@N}7>r?AW zFw=z&k(MLC-(R>SBF@mF?NfrE*zmw6^PGvrE1pRt4z34#^&-lYde-?S*l~5%T+2oe z?sY-X?loI+>!z#oAb5qW;J^ToIE^ zCiu}6{}79W_x^)k$TIDrcj;h^ikD*@69f^>cjVhMB_Q4 zfO4niTcPY%r6JgjZp*8i^Q%gQJpvL%$}xjS$R_%X4Sc9wV7_9_i5gi1CAXsun^JEG zDl;<_EmI$hw~)P<^ao%sdMO_kxRR!CI^2f; ze!%D2I7>c~c^lF{0-!*N?9}OAr%-wwJ`VG%cd|_4)pt)yzM8=vaZ?A^q~&E@Nxh(u ziEr&2ccl;RVO^6|k_C~8@rYlNe4lOuMMsM?)kXf0KHeAy=(AFy^@kL54L%mL&Xl|U zDp#==97d%P-hifceP(*yu-IYKV}{sfJOOI*1ml1gV4;-)b@=MlTKXcod^re-WMWkAsZB#?bs$G0YErm!ivdo1KqL3Nj7S2P-5Wxw zu8_!hCf~9I3r4r<&r5nCB1}<;^W<=12v_rp4TF4c0@UYwk&dSc{~8)Tndgbj8tWo6 z^q7I_YNt)We*ukP;L!&vn#kDsav4O^dk@YUSc_7~%C8MH_5{6^c>ZMm0Ph z##9VIXyoSf+1Z0&QxfznGh7ONnwH6J&=X8H7B8E*{*4qk8^x7WdmD#PLab)ZWpjnQAt2}7Y5a4ZuzSyYpE(&aSBi!`l#&P&uJPC#u4!eL&jW%EK4fB z&1~`yEEOX4-1x75=Sh<cDQK}Gjqh8$YvPWRIgpA7l8G?J;g~D8|q(N^d#lTIqHLg zAp~~{Z*e)8A<(D>>P*tg2E4|;nR;!HusVABlX|AiDA|3;$34dUl4Re9owONJXQik1 z-OoZ<9gKR4Hq}*oE!nnhxqw;tulA^JLUX-T_!VggB7L%4Zbt~_?&nVc{bo_HTzrch z09V#(oxY2676nh?fqR~M1L`OYsRT9|XKDs{sFM~u6%m3tkvGl0>N z;wQhdx>~$owEJZ$O>i!@qDLC;2>>B@WB1&it}a_?6g}`aP#VRWE8jmnxQ;hPx>`Fs zIeqHlF^EoLQ^%uQ5xY(u=|m}=F#DG{9KS4^c#v!yhlg&y3m5L z{*%oka%arY-u5;ecwgs>Z^>TZ&H#QXyIH&nRaPKK-=11b*9}f(Xh|RI51RMOY>ujg@zR0i+ll59Yg9N~orce^PuVeaJTTREk#L`75$% zXY)adb}8wH!lbBEDAD#N$w}eQD55Qj-|nTOMM+(9wih3U$A8TvtyrqaTRnE5@hL(^ z_f-48eEE1SVp53TCDMh*M)?c*_hC?i zjl~hlSC8K}*3u4x)k#(E{9yUlIl*Ss<9G6(p<2V*bHHEiDMOA-V%HFjd^DRC+8ThW zf(6fn`diiz{5wMBJVOxuc0#e!plP#REMZ3vtveJ&lfOyorOf%KMLx4tw8_d5H1p|K zzLu>Z4{Fw7|L-=jk`&sOiOLe)^V#V=DHz`_pk#eW^gs3nI_kN&G!zNEE6Lxlp84Gc zo7||}m(7)H25*KWnwu~l%p^~$TdROT-`Fi@=jELrzo~%fsNx8{-fxw=+>G%dzw8eV z4c&T7DR8Y)!h=j9q;uH3Kn-6hO{aMT>=hP{wFG7c9yt2|OVg_CEAUa6yN#_g5GyvD5=w+M z5%wysSxztx10wCr{Ez#E7zmFY=|XI#v9GrT_Pyz!BUueLxYM))nhpFR;{_f7V)fKs;5ogXKQV1h{Apl#za16iJa^9ol=3PXTV}^9Iu|-ccyiM;K06!Il zVxw{W<`7FO&Qjc0>^qa_Ny50QukeGBKD=q-gBU-xTynByOyJ+AxUXIYXZM}PuZjn) z>}N(0n(+Y-?MoALAIiW;X!X6Cm%f%SQHQ@IJbiss-&UMcM~W1vH&V@u{mJ~LXE;2n zJ(ec(NbxTlk&u{gZUQ0;;0iTG$UWH~=u@980N9srWN>^b>AIPzf?)ew; zwqTqx%9#Q>0rA=wJLU`l|LbX~q-t#h4<5kX*^k6q03kD!u};?7+8QVVc=Uiu#p?xB`@rEaYe8JWR?S0t&;S*=d`}pha`^5`ypV`&Ck_g- zP6wrUUK}~zFO;Ice*h8gO#lYmq%&4sW$&q_g_nzEOd1LDo=E~>iIn~3FaxU9X$~Ye zC+HcB-&E)^#wb(vm{b_uYhEYFcy;s+Pa_ZM1OmkEs!HoeG29)mhut=^P=$+!KYxDY zcHZSfiq9Zh;c~@Qj3-&9gt?0`)>x#oN~k`1>RAT55hQ*H(@HAmgu?Mw9E;HzWfgrC z^%Q~Qoit)d4n`ez>)G@AyRhcb&*I#l@@ty#Hc)y4`g^`$&PTxOeFTrPP+3igj{GeK zbaE5UhgyL(aSc*hF*ISZ=f^JA{u={S!)oxMKeaQlOiqPd7`C?zfbkt z5Qt3HZ#7OO&nQ@*0gN*9P#h6`C>JRzmuBXVUwrPTUs3V|eIHYD&3lNrNV`-QN_REg8`fMd2aU&N9O@;Oo$`&Pr24F2fs z4_T|8>jZefaNTTJzvCYdy&6vl^9%Y$m3Z`{GtrY!eoAWMGwPs3=Dpl!9$hx^dbr!k z?rvJW2dc(sL!mt9=}Ts--4Dp`daaR$1ADvZwmJN|n7t#hi<^Up z7dyKG5%J4m(1>SeF70|P4#Z|=d#vUSs3W4enD1v44Cc!v_&7V6uTky@kS~0?zc7<- zp18?Bw&ABE%tbh$_FGot$q*vCA%@5x&3rsQ;cbmM@gS^b3+xpSc!aT-%n9jC#xJC4 z@(fhrI!wscg6qtO59O((o%rqYMKyXYWGIZB@Tm&LPZYi0)N#6ii=5-7VfO7C9_an_ z@$R8U9wn%#e=c&c-{>!zGiDDApu?(9630FY!l%7k;E|${y0^1d8CX`Hyk^;#`hCKk z6xk?9(<4(IuX<+(J0)9}GIs79xbP-80Ik3aG5|bJawoeSH)lMEf@Zmdd;kQ9`*x18egAAe`biOvIfpeMU&vHD~%cTtX z(BL(KV7<%;%qkZ&Bx-j+P%l7hxjnCUr>IaUV$W^#+1XjWKhwcfk~=1weAZ*BjNCr= z4%H{#w}1g7fbgyBM7n_=pM>2Ve|Hc$(TV4=XJM|^0xRt0R>k?K!e8C$yc{%Pztm8= zs&?wstL(6I^fBO9o{*u}ItA-lRr{$ZFvs>z0VL=}aE!ka`<(aQ`WUo+f-vRuJYN^_ z@vnlC_}Rxew-BI6Lnl)jkS`y$kzbq6@kOJ=x1;x;`rHg6b%2>I-iV74HX3Bho|%<( zF-l+k?NZN<>mpiqf&TcsKI6;9vi)Cl9JD92-SybvJ^O?e;!ICBy`=&;Nl5oE^Hsfk zXFlXm)v;RbrIdBIWW*eVzp*?1`CVUME-p8$chvFt==H8LRCml!T@{i)2ba=EoM?m&KtVINk-rt< zX=rFrL)QxdU~0PZa$&b`Ys9&Dtj185dX((!*E*URlao*|Sl}1|0s|HS9RP>^QeSV~ zwn0i+S=ms+QwRtkklA1_oVR<nea&S zZ1e5!6jwc-Py>&BilqsM)}PW7$TovRc?RE2sgg2w0&Ji(qu}4)svP)hYQK!~8AJ3K zbC^7CN71O}gJXv8Y`Q9+q788szmw|h&i%YBH({%%L!AC*C0g{8=xW7eeVS5APA(dY ztI~w$s*N!U1avH4sVgY6N5xWWpb+7SKElkq^-OBUAro+>PoF0NA4N|pT%{<_0VA&- z+iQ6yJ>t^0lP|JnGM03xnrOz505gF;^C$kSmgNjMn-H*HmTQe)Xfj>YSvUzUcAoaC zf0!d=U^K>3CGEQu>O)XbQmzNb3kv&*@sznB*W9^<AFScNpZg!u5%w0u2OI~Nt>J#y1pEWRbulv$+(<_$R3TmbEk+{%d4SoTOGSSwyj((DDlrqlBdu^5}6 zfzO464{!&+eMm69kzkf3BO@anOg2|?Y~zL67vuI`B2*)0khq;jjL@Fi6(rviC)0Du zO&9cq8L{|C^FE3N!vuSz@XO%2p#oTBH;3iboATyyE(VV;O{oj1FWB9xW0x1V)3$x_ zi6}xOjYwUeEM_(4$ar`@Y!i>Kb^TH=$SCz(1g4mq2PM4PFA3lR9ts?Fg zSUBIKJYaum)qsd6y&6-wGM0Ox+R)O81c-D>zS~tPwPA_LT64yT-X+C41+-^H&rRf> zEZMCQuR!f*_>X=qh5Pujful%?GB{-5gy}^wGB7;6^~?RQr1@v~=MtpBYb(Tj7e||F zSrcvKx?8GE=?N?Lj(hh3l=$A)`~H#Nnq)S&mIHCoIF3_GL-t6#0LE`FE>w?L9-ZhxctCV&Ju+8zReaju>zX1%opxL)ezAlplVGSr6?{eepfDKGX_guROE6QZ zUj=TE?l7z%{dP(jN5U5z9FE#+gWYHVv#nsK`Ing5bXmTiQgxV3(_VE@R6 z@@zDu&j3X#mcWVL>3mt}8HPQ*6obey<2dOE8l!PMFp4p!+9bDAG}Ms)3eoDk%0#f5 z?fq%OAO->gIUR2oG}5+HWNxcD#>&Rx-Q|kpT9V}eTsnvRC#5kgi)@;uzE>v_RpZv4s-|jbM$T*SEBHJ_ zVk^GNx{2}TuE|*ZYct!p2yjmarK_|ov%M75NadQ4-BSDMdx~qQbW5oAa_- zS}F=qsQy=SwusT!(qWCKiYPs|3R zg0t9$?9#&r?z;N9___uG!Jk342P$SEqOnU}%o=J4DvQs`xtY@z$+8o<;Tk4A!1^e_ zY%F!vL2gQrejNJWQ1cBlkbpb@8y8outKLf4YG*X%Sg`Q7t}zk8yG_m}hLSYfaeGt5 zK<)Fa=s6=XRG(e!SUxRn|5RYA6cncMmKed0gqdnUE7|q zr>>sYs4L4qU(iq15i`mW5~o#g7`uOM&exX7d&D60vrP(849?Ve01VaZ%S((u7w*BE z1ru3ZlcPGrzl`2%Ki~^av4JBTH~@>(k6;}oB^Hy=)my|~VS+g>_;WBElJW$Ldnp8S zvC5K9-x`Z__-S9nf!9k35g>%gAj_of0wmAYss6>2;jirZnlL*BKGz6FPO~J zF$&i&*`(yU6q>~@s~TLOQS=oWz4pXl`X{=8ATR>IkO!!QCqh)t97nuY!~?NfQ*lp# zqCMm*Kq_dmPbRT-k~s(HtF_e!Zgj!^p)eEHE-AG3AA`!fott5JhDN6{6;XbGr6t}@ zaYms?P?3@ths-xb3;t|`q1^~L50D1H-R;ogqwb4SC3H2Q2?c;iaqK4WZ`F&013`H& z^s=4okzwpcv`q8DDLLMD&A0TKPx_hnYG-(=h*FedpcqYoJ83doc226g{xMq8)rV!oAav9ftrBLB6 zAQ)PVWFHTPl<7FTRm#fHW~}}ABXvC#q~@$&?G=N2F%Wcis?V= z(bR)U7A(B4z7#nMGOz$^he#~<=mA#6KO!Ii4aZIFPN+_`gXen1(xQ#lJ7J>~{?oEA z#Fb*IfNi~pz>)U<@pKhjRYltx@BoMI4(U$mM!G{95e`UqcXx<%w+MoCN;lFa-QC?G z@iyLj-ydKM_St)_IlnAn4rvY}IugR6lP?L3`hx;^G+MGqU`$qX>i@R}if0MtZhKyRNpoJEj+EgT1{AJ0eDEWXfY4JAuf z(qzGvxjL#h5yjDz{Eiq@Si;1f+RrvTFTq0n|6>$V2Qt5~RaM0>7tO>(Py_-&(8mgP z4dvJ=*x50F$n3AsQXU(7tjm*?|lo37WI>? zZ0li7(}PY~&o9_nYyv0i>-)b@c_V4mzRmvV3dEZ9eKxSH?%7FpxGbQKG6*_O=(_%2 zSF8Ot=rvA(SjZC}D04$ERLlrv{#c(|%>o}l&Qe4-8donJP3Y47*(ff`TA-1{Ioi@f zp`fWr7xR#l5XJlhd+!lo-c0F|S3$P%d~XMZOhcDN`~WMn`aPtz0j}@q2#9znB-vzVylGl2UErLC5BBQ-_w(R39OZ-G>6p$(ai~!2cf-%)lrjl-3ON~zN z1NQ>1FcS+kjATmfD$0-u@cftV#ISL&JGXMB-L3Hf&6$@_p`h9wCo8AXLFGiMNW0Bh z_5Xk?2s*|d5H-My#ph+{h4eZ!BE3FQ&FmPzKA#ZJx3EU}T~*%qj(}|tt0G)hz!by! zY0l&BYRY!bXqffVr%V1n0~Q3PUW`mc-JS;elE ziVqI&&Y`%iWjcYL1O~;mR#j`d9#Q`vXz@-?arT0wD9j_{>AF-UY(Eq9Xu^Jqg=>9N zTsJhdcXA5KQx7*8VYWxQ*w?BnagtxN+@0IyaBo-o9Rj=)4HPghk9{KJ>KfylRZWP93~F1Qc9nExEP!s- z9SsZFELpi9$~no&kA3`L74Kibx}42fpJEh9owDWMQ}Ij5I&481p>;U;+vWUg8vXaQKgvQ+oSD)fLj#9*&sWRvOui^P zJUNH|P%Yq)&dmXtf0zIWHV5JBMEk#5tV_3Vfx`M0e?k}1!Vv^KTy;d`h}it1M`ERH z{>Wu7S2C?|WgdpOB>3KlGS(u|nR8=v65Vy;9hB9$yfY#a)R;Y}TF;*<#U;{GO1N^7 z^0_0W!KIq;cy_nKCClm>G3m7wpvY$`MbjN%8S028&z-a+vt#Ng9mkETm?wuDM&*+E zE9F&UXigWRG%LdMgQrl^TjSs898GJl!Q4@Vm6n~&QfE03Po1SJ9bil!CsXBqzM=HD zh*FxKTC#{VvosQ@;LA<55Wx)jff!pYK(wyL->0)bH>at_co0dnPUjTx4Wo!ZKeXhh zECYpB|N7DNJdV-IfDtcuv(ey`1-?+zNUK)g38rV3?eJ@@VP&eGi)sgB47mZomZjmMb2VEgd%$SW=X^&M!D@`wI(~2TRN*F;J!FUN5TXMhQ zs&!!g4=XP&I$yljOrIsyB-P-bWnpWD$(?;1P{4-X$-|AUL^Am6EuH3GrdhjSI2-(1 zCWHFN&}OeA?N_*~UZ&%HaYuD=9Bzh^1Jbo;E4D4ouUb;GD9?J_8h4mc(`UM-=vX$-r%UNR>?>~84KUj zm#ox0&-hg5#)v8GgZC8`6@U9U(#vku-{W%US~sS%x+Wi7Xfe}8#?x6h6I^9$*x>gd z*YfrjL|Nn#86YYd8>Ox;FNv5&wmjdxt|sNFP`&qYe-|k0sL}S8RmB&%V4$Tj zfH<%`oZwfR^3a-nO4HIS0 z+v%=xPGuQ-{NPZaF*&n{v+T=h-hQPt?qd1~S0}E>BhTU#lzv#J5)tcgLdvl2*hQK& zZOKXZlYEk4t>7XcDkhG7$+e!WkcgCn$cU#VX8S=VHZarj2eX= z=nwOq-zU2DwJn+$h}YgxiS{5Pbu~C_et%y7)$I1o+>hU}O$5b@N4h^awyeY|Fk={6 zoaD)Va>=6ssARgo^jTifXhZ+c^bK5W_N`z&AXkDx2#5xih5Fn5_SrvAJWMXD48Swn zT7|H&1~ZE&Z)ZN*$+rEXw*&IOc0RS#!oYm}`c=Ee6Gcv#L88u!9r(g?^#oP!kexyX zW#@}S;98+gio3XyvhWHKQbOelFcH2bAhLUbL%UaDAejqvj&;EdU$^k#H>H^#z3>fr{Em zh={C^6t}tG1-QaY@!TC89@8-m;;^^*@5PPsR+kWx3cn+F9_Z~^=Fb%J!j;lVpo-+< zU!?T>qS_z>;xtmS3##n$JbiLw_j~pRD#U-~-0GHZZExr4QtEL@p#+Foz>lf;4f|S? zL@Jqm6Rd+eh}5m!)6K=qb`=zRMCav}!)&7$f)hs#9}Rx4D9qan=^@gtqSuJcz_ z5WJ4*0PMPzkWLttm-ttH=YK#{XK}f{4+aJ4gda{q6kOf4Q(t)oL zl9gXM-CxfMD+n1T6g13;)}o>MmraU> z%Ov|7pc22Ucc`#*?_qqNHyTHdOnJTgI0LdJ*lN7yTrrpgF|Ye>T&f?dsn6wE5~2o1 zjhdWNdMLE<$XbAz-(Leg7XgYoO_G96#caL3?>D`1DyJFZ9>G^EtV&IoG$W(&6`$3- z`aEL8?ya*MEehshQq)lziD*~s96YQejk57W1qxdi#zY3#D`x(Hd>jnQBbC(Ru)->w zkp*4r`6h%C&9eBG6+J6`PNi??6fRL`>3>#@(Lb{zPTb}p417u`Xel>w&p?<%3`jOb zL2bG7%o(x4VL#0J%hdZ{UC`Z&vgN)c6tGtpb8k6V;i4aJT66N3{Flccn^EA zv9N$C#;??%ofL?2XMI)m__Wlrz+CB%rE+2=Nz3OIad-7jpj=beQS?_r^Ph`OsEi|r zNWr##Kyw^`n3;tYzGIGzndw?YrxS7}p2v@&yGGwSVG5D2V`cDWI0IrB8cDF1?T%2Z zNZ>18kATmh2kg#w@9YgX|Hf)zi|cDJGz!¬OOj*fk@`Mu_eWBp~Ls?~zvXX!;yG zxw`f^jS10SpFc1R{RLisfNbuUw6zLA-F(w`<$;m)J}mKJ*tLShhqMD0IQ^+aag{mu zxtX6hTOPOYxu!{Wm1eq)Ej#;*8N%&%%q=-PbdYuMgMS&@I5DPA>CVxA%&Sa@dV2RI z(=_bh-5UnPk1Be}K|Cw&Zz_N*n=2+mG9f)vQm-QLTcmG71#2mKUlFgqYkP#l`;S!U z9NxUx^N9G-Qd`?QU1Z6LS6y<|wWSK?8-+^R2O;QB7KQ{jNAr|zSlIEijFAS)1HHYl zuP)K|d7`}R|FG?j4hnlk5S!&jk!t>TE#;Bu@bFlsj()N5glnu^vQD%31i-Qi)ehS> zKE;#YrpK(uXP&1|=79Yv+vLYiX70t`kyXe0|kGOQ9NEQ>C)feQ~=|U&i z-^QxAV$)e|?i62z0__vm;#;`?CLK@m@dWVB>2BE3X??e_2ugLKltK<0=?sG}MmAs& zCZ68C#IQMi|od88r||S93Puod_I#W~RR5J-zGb zoWMcEczf)a&^bo|s-;;sl0`aM{KO)}G6pI~fqr=QvR}yZ$!_^pg-IV~a8N^x2hiqe z9f0iGUhpIrdvQo)PAOcEW=iXDjex8>_N?9}OA`%#{d*6YU@Rw+d_|Mxo0c~+kCun> zcOi7>f>U(IliWNT@nFe{Qa>J*V1rJ^lOm6|`sL!dU>wmc zltp|=0)3qZ-?=Q7U>lr<4OIYUQ{b>W7&}~JSW*mTLF}Fq0HRN4HqZPpRhHEeD$L&w zdzi~3hfdj#*>*_k;_w1_n=(hH@?N6>t z&XbC*nI`t9-^fnsWe5647`x{;tiPa<&YZ$g>&bNQJFZu2eF{(&UcVd5u-nX&_O-kY zk_PPey;}pHtL$EM*b`iYX_RzV>`Zq8;nBRmR9Am)uvixyw-f#^0|or6A^uYNoq+f+ zHYovd6@rm%7)i9zUJ+1L2b=#R?oeEmk|t2*yiG!ziPewKc|>f_lC5)E9B#T-MTgVP zZA)Ds5pPd*fo>yU8gCuI3_#aH<#p427u!Q|PdVP1tG>u;qzyWP0tTgtu>Azh%3V6l zSna!;=%l2!c_1_3>llyom(}*%@0G{E2(GliQccf{T%hb*7dAm&`u%lJTV2=g=g-hA z-5_@Mst@fv+lQwRhVm25uN?n|P=MA)Z!mo6Cd-SEV0xZlZeIJ#rc?B@fwtGqQ& z9AVoJ1~br1@1^UmtrZ5o|15DmK;sTBDW4b$>ryg-$2%wF7J%7@|H$sF$*cGD>AR&F z35!48BfH&<_sN{W7*BjfFR&W>Q@obeAg^!4GTG3ldT3z`(=l292!&tQ!6;e=q6H3U za}=`pao~cWb6Q@=SLEjc-bT5Vh)~Q$JYL@IuD{4`%;LweF-2yE=kw9K-`a4w$)7(zw!OX8pC(`J~|i%{SjCNsMh)s@Lc!>Lc9n?teo`+tba*OgVmrB_cu$ z(oO$AFwC4?3yj7^Vp{_t&Ejur7jP$}VYa%a$q8Zvng>)2CL zK2q}MF2TeK>sbRX@{`&e9^jGhW%&n{MBV9yy#J(w;zcV>xBO{+Z+H1PHyKxGoI;R* zk6&hci{eY)v;tukbOxH4u3&+z0^tucl7iH^Il5pkrT_dx<2(T+f0=<+&4yqUtlWTk z7pvL4y=5dikrcF;iEMin-pIKnnkh~7?)gQGBnL}6l&|0nOwg|=F+B39Yy!kndo*`Y zlN>1HCl{SkX&0z{ABb~`Mp4@?1jr}I5C{+(hq{S8g6k;gJR|HI&+j&4+%P*|Yl#L@ zo;Z|6Q0z`m(ClUgZFh##K1@C^2D;$1`wCkL(S-aaTVh1dp*gKXz!+9;7}RTVCvSUn zdMLI?_0{g*^DM;Q_n!S`#L(R(j~dPLtAXDb`|v%eKn) z)*1lCX5FY->qpkfKe;O@K-e(pSXuji3ABmHT0q0(_c(3S;St~c{Lpj5z9D8>;$ECL zNhra@#-7S7b{`8xL#nbbGIj0WADUl--buW;M27h$|9uZ^&5t=qnTi#|`MUH=?>EV) za*Id#i|k9F@lwV%^tO-W{R3Xu#6fwmS1jumg@mY+%bO|5X|Re1MC?vm@S&fQd-|5> z%Pq|;kbFt8ws(YFXT}--lj#^|RYVNR%EBT*F10Y;+-|pgbb)r72x$KE`XalUFAuI+_gNeGF-|^RW`9a;gQqJXH?+kB|U9HCC zmotJy`&3Hx+Ax7R)an1a|mV0z23HQAAXAS{A?Ujwv{7I-viJY80A$6q=0h|&O9IC89La+}9Vl~^pj2uNLhS6s zRzjR#8SGd4p_5|q+~VS9UKlB6dm~mQP$p!IxAl!r3n}4lmt2d2%$><}ES=g%zoJh5 zx7C7*Fdx>QST0a;N+_*c2??0)^Lx3j@C%BJjAQ|^*jNI|a@^cqeSOjwNAEY>ot>`= zU|N2^{&*0WWpE|b3<`poo;=LerZ-?JBB$gbnU@}=biYCm4h{|vsCUFBEegyZMG;R4 zlRfY7{3&hXijsbDor*8?=jZ0!)s*ijpr8UwQ`qfBGb_+zl`>SmH~Ksn!!yCGO&r{| zOW{vlXd;BQKmpGw6%b!|Zrbw@T)!%s)mt&H1TgHe%n=ad^(r7Rz2#}NkJrq+o!zKSHo*{lC~CGi%&Pgcffep(J53^%p9g% z3C%$3g^jm|&MsE;Q3WiIi1m{Zr7IBe3_-m^39MU2WB5QF7&x$H^r`gsDJILq5YX^4 z9!ZZx5|X67X!ogRA^1%Om-$aM(Gu0o47+;SY+0$V#2zDyikfLx-^Qp>5)-(>!@}SL z0|PyumfLo4HUTiuN;J2riIBDRN*UWgMiPkg$;Pl<-uL}ABcV^QH?3xAYnzjytbx{UNsg8sze^;g3GGpAMFq2%B|?G97^6)Z6*FSo!< z(?1J2m^u%Sh}f?p2GlyhwTp44>0u;0zH}pJp<8>F*x1d{Pvpqzb;R2YU&g<_ptuMI zQq?oX_ltxK=JneSBZnuI(#NIVTmaAw&{2bBhn43AX$3T>UDJ}yz{xyD6B^c{y4 z?O%CTy|HGkU@&`q3h-!YAZ4K84K3dpgt?)ExQq$99yaY(uxZ&T?XRd;aKb(xTS)2?7uP!yiqP+#7Ye~5i{h`@iBt^{mF4Z;%V6#oP%8tU0_ov<2o z&Ov459oIIYYenXMA>Cl@+4H$d!|rH1(Cj~9fIZ27q-lmFfD!zhaV~$vra#84^oT03 zti^pw06mv}%_4Ji;&mt=(Q%-BHy9BIgwFv4AO*1>;O=x}bpx6u;j zDn-3FTbjE_8stM8!oU;dNf$LXrkW{RhQe%aq9N2m&?#)t2Qhw_vM#4j5^9kOLzSYJjXGHQu53eB%+-($}8l?uYM>%aN<&dZ- zc`pKe0N`A^06XKfYqCbi1@C@Lj_KJUp$A%9NR?PdUJq|T>zsuHm(kY=RZbfKlxYOP zIhQywq90&T!(J@|LmVNNWt>X`R8`bBehu19719CTJVQ_Ht1ErCuVIf~9_6j)qpEH&{{zg;FCa@ zx%bT%f7vf{gX7~3J6S5 zSV)PF4@@!ez!+&tzp1`&l(ctc#5nKr->gJB5WSwA&lpGVAdaf9L=_ZPEO7b#=IKJA1a4o}-t(W{ z&6x6d%pvXVBDQV!rpCqvo5TAF;4E~onVod8FUqzeopO0^V#j=|aL-cZ@3WA9>)C90 z$9A~E4udvkXEu=OG%gy;Rj|^->Lef#35gi+2%oh4j!yzK6LSt;)Z|)nuFWRc1qIzH zRwoR_85peOgH728&QH!ORKun2Vo3QRdM^_l%Rsp_IA}c#&1mw=wONjjW+Z*J_526# z=X)Mr2XBsKCLe7vIhJo+X;Jz~WlWbTD+QFv=>@O@W+ zK`X}3nw^yEx(?X2MfxM`uhRh9a6IGAqi*C1&>#2b3J91ECJ0Bb>GBmOg1sNC+^*%_O`MkvcbotlT_o53LoCSEW0JrqPeM~?vs zw39Np;E?@ZCbvjS7{Fo)l8zyjl#|1ZjQWFtO&s*)YP0K;I&`5DLSL=LB+Ds)l3_3NNS8N~U&jFkC%P zG>vsK?p|(rX^I|gHIB0YA#l=&q>B3WuwF`*CxJCRSx_|i?#=0#rEMG#kWFyto)7*= zU|wF+>*r!R;`?l@JD=Z9EY5$rN>e-gO=Q<$e$-C3b9fj9P!%iRa&6m4UjgcddXTwb zI!z)2HX;zwB=qRfQr}?LKd%GTOg?x2jRXOJ?A?}n07qAFLjy1I=rd^&AD5A4G{Qr? z=9e!7HX1>{vIs|r{_TTEAahYuK=M;f5$`_zCuBu)w$T&TjkFL=)b-eFK$`YN4Vr zkQTSainEF0Q~RG{$e3nFr>Y<)SC=@ZcDQeMk~Z0Z;Kt~4*Ti^i485A2xb+c}Je}vr z;_Z{4>L(mlBTwdFojFQ*8^;iB-G^*@ofd?WSwY4nVcW1@DagN;U5WOdw_b_!4k`f6 zK=pJ{0P<|-T5a%_h?ddu-5pd!KEX-nV@`~&S0>Pp8~+HXh!W7-X0fkv3z+$igoP1yb+eWH04uik&xQUhIHUUs8v@wl5?P^5uv_IIL7t1VM&Dexg^14`t5!~Ll@ zbDiboRS#Rv0u8vCye}1rW#p0z7TY){W@=s7tevEb1hDqMNwD4c@I6?V~1b> z9I}y{j-{CFdsNjooRO7G^(Ub~uvN1{U&wsZ53SU#2FqxrQUm6S+G&XS7uid+-uIJ{ zz=82d@Gkzp1|08``!}{FEe*@Sd!fw(?T8|e%a>y_AG-D8i1R*M~byL?8)y*{&4M0 zO>0p32o0YA(|zn+DQwKJWPM0LhT=12^mov^pkC})3B%}3q1JGZu8y+J!?TlG)l z_(AdRUGpZlH#!XrJ3;t=?(xuL8Qv_$jR2{M%0Cm`V zfZFahko5W$gWG?QmZW^eaqw>TyJ5!PqOZ3IzrFw;pP;{iQyu&|+9Ou=Zla&9%InC!syuaUzS4RXy zM@L6SK_O9p@`o}!f{{=`U?5B}v4``1>cPRmcW=xmfx^&GIQsU_`hC}@ZGV*(yn!UJ z%FlhM7)XMJn>o^KX!29FJA;S$KWk97ZEaQtY zQ_z)~@$@B_29tA8L-#2YcLD5-`-gsr?_}f2n?NX(oosm}35#vM?GFw0MinRpM73B1@3J*;{2u2f&9XQsndXx@66uieu5yeCkZB?)W}E5rjW%eqYM{B1uHXcS>cL#YCN+?#B;KS*O7 zd8?wX2-l;8!lBHBDo7b=pvE&eu2GeK!^tlKM?){Jza4rOxe-g_%)cog$( zy?tDAF#qu@t4FO8u>q_ItKS!5zCXB?4n(iBlp)O2K!ekhF0xCa{0X`fnjfohgiPvvM;j~c5hPK07pPtw@xw{&{D;kH zZuxi25*S*D!x`_Cn8^#yeGR4(QI6W8xaoX3%`jb^H&^ButZ#I;q(sTFuqN-~ z93XA^dL<>Y5TGWlsC1}vFXGwW>C_fw`C7MI8opy*Jf`*w6<~C6TON6J3+JlRR;K*J9kA9pTsbj$eEl2R+reRHhDFOwiVu3F z+EraY?VamD@~nfsUnj#1D{OilU#e2~@L{5rqV89!?!Fhp-T&t2)!6N%920-YA|ERJ zI=~dDs!&PDj3bc9EC=b2;(t>^mHQlGW%v>P7+Nr{Y&JvkvCxfaf9 zQ0uw+iI;4UfPsT!GFsp8ebrjNrjrSGM*$c(;HaQI?VmOcp6PUI(i!h^ULN61q*)K)(`4Hg)g)MuMjhTx&dY2lZGWAo%z3`|Sl~s^&P;2q?QRTTv|Lmp7r$i()ZI~^w z%hwPBAblf1n$#DhEb`1m>T|dI8k9f_TV!jE{f09$)jtUxjHxMGh%01FDjX{v6Rf{Y z6=CpvQXYSZR*))_%bu=svIRR5D?>GFdDdHhe0Q^GR~Fn+V4BL@DW{?PXfh_v#=T!j z?v7}SijR~f6&2i~&dbNYp1044flc^@*wDqBZWkHR@xp$+D@2~QsX>ZjjswXzxVZZ_ z_`C$?LDXzmrw#138ANxp4iFuS2ExL^e}c&|`+&%h_2Xldf+)l?QpY_#o`Jwp(? zzy-;uKmPpt)5NyHUGfM92CMJ;0vU%vj*gC0c#m7ZE2Xl+ik&DV6!xF{pa?OKzE+x> zVJ8rjxgFCND+0oIno(G+2{&Es@UExA84G1IZB6YC+MR>J2odKCq-U$o2}vOBVHXl zz`(U)=)fbfO_@07*k-W&V9|56$1M8!Gk!S!QY|VM&v29Y6;`J4z2h?n!Q&UGJIA`LVMcH>%nD4omo z&kSmzq`Yvppx?^~`(dI};8R=GfG9<#()e`1l}^3bF)|95q=?c5i1aQYMK-G4E1d*R z(hB0v>YVy3x3G>sO9C>fh7-(1BN1F3G<5f(;l68NgU^R7R3Z#^xK#v)**}M_?^P*e zinyWY_c4je$`;~pO{PVyrK0yK{6@q;vS9%zgYQ~@NA)$$T?J6G)pG4$?{e+8(13=J zo;0@r-XZ~gDs0L-DvpC@4e-W@du<*c9nD=7WWiMc78Y!t%}HeuDF}qkF@(rOT1!Zb zLNL$5wy~7vO9p1(aZXP6$+7^u)eKW?(J(N&1-zBYrIlyHZTH335>7{$V}QoA|L>L| zm1J)t!-6`(-tPI?2mn%b1LKgol&)!5hw&}s0B!iBQaN=iy1S08|?LEMgV0mn9o7L1{xVJeT6@Zc>r<*hr$ zZ>Fey-S3O8H*qixM6K**4|;%i&~^GzVhw<`Bh-QWj_dbH6Q zp4koa1O1+Yw;hXikML+uNADG~mqcc1>c!(e@@s=rdn+#M6u}17aGje!avK`(fiyC* z8n>D|jYXrtvG{9y0&Yab=_GUPypl|5#VR=rYoO=pZknTAv}lIl@Ea}p3_W_Z9U}=y@so5vcM#6_06OpqKQ2-5xQR1%T|R1 zc}%PvyyU8!_moDtBtq2~-&R?v%jU|<8PKc7kE(9H1dbNny1Y|+O%~T&7mu(nxE}!q zv{FDK>N{@&f;K#~XHoqXL>UHuJE4Fc!32QhA=ILso4@Q}lOw6S6q>O6cD zTbC0PAhCzPcrWLmhaLIcTv7?$!admokaB1^X(;%IJa4QDS6~}UYs|FVXqB0AYN6Mk z#jDs)K6&4*#!yWp>OH>7gd?C3c#ZF;RUf}SBnU`&CS@zf1G8?VLG9{%^9Zv>GdS5u zC>a9u-gmQx+-+;o_SqOtI;@6=6sgaW=Rd+YN@S*IC7G>TpF-IbU~sS#?@ENZ%$*s! z8?D&5ACJ>)C2DjeIuRlTvnP~<-xg@3mWn-oyUhxo`19u-HqLBrX)U^E>BziCgeotI-3>@og z#(_0q(2hYog~}PzWnB>w5wCHb{&pq&e_xd2O$Ootrm|`?t(8`XiaZJ>FXY`rGE9NW z!Y{kNm(^dVIlmHAaz!oZVFwY(5n@TIH+yux0uOOeKAGC;Imao8SfJ`NEKJ;y(;X<< z>tvd5)oSQAd-d{>6#u~)pUvy@L@5sBnoZh@iG!bo#kO|j(MQQ23ttTK_vSc}tMRSX zOldFhJB3*jJUbVGHC6Pv6&~N=BX<#^8^9BeerewT@dlLqqsgu#x5tk~!UMD3LJoHN z)=2TaUHULq_HArMCw4!^?UDD{l-RidCnWdq+owHg$m$OQ0)jvs3NK>%@R1q6Qfb_^ z-+>;X++2;jY|wGAq^I*W(%S8QfQRaQ?sE*v@y^*;Bh=2>4|8wSSOkOiXOVWPv5xJK zU#2b`O$T1*y#~siv`bD?5pLX9qTie(gBnTwyti*-WsMtlZ~pFWSd0|?Jww?{fWiXF zT{5iEB9f>+67nBzofT`(H6s2XyhSir5af5FBY9pT0A~WhT_`@D4R9gkoIbxR$zy1I z=lSVb(!FQ$2O*r;WNYeJofs zW%W+_L{t6PZcHW=?V28lWy+*m@xXQD{YS%AfaQt9vWJ5Kx+!_RugIK#n=W1Ua~>_E z!>9Ek0mYY$fRaF%7uNJvO-CaV5``pKENpIML;)yG+B%OP5okLQyt!Zi9YA-R&|=;` zQ)l_uLvYX+>Ol4Y58$h6I4o}t9%jYet1}rwQg8kv-|q_vu>LwV)#}OVqL;o{_uGUm zPN;jY>7mYbdyyAfX^kIY$>(;ti&Q`cH01m8)-?j6@HlvX#@zD(WvO{j_;sFYA60%b zZRV&NG7U*hOOsuV`gjIxcs|=`CeW}^xXgqOel+z@z%c*&<*(lhS}ELoC<;+gwtGjo z56QVAi$C9n4uSM%BOcVFjjh4o2Z+df@G2h+Kt;nU?n8zsm7-#JQ2Ys~LfAGZ8~l%b zYA1(c69=Z@pODXDCj<5ARP^=3sVe9N37_OTu(q!6FLw=(;BbGecxvp-%7AgH%en^Q zSz~?JiO>W^=Z&uV0(q1=fPn$2exZp!v%*SZz1&7d-wiKq&a&&_(Iro$U>5h#<%q%B?Urf*Ug$D?7xw$tIj#KRuCwpO~j6Js*#=RC5vE- zmL0IaGT~ljeF_8}&DiehZWOab7jT$7PT`L9E+b&0rW?L#`5WyA5h29Ts z*xO4Npuf+Onw1&eNSJOxf3Wz;=H%N8%B)E2m=}$f|(V}AaPoGSj{s%D1Xf=)_P+!)i4g#TPRgH!s$l<_BDIFo_jZHHY*4oE z{*OFsUU@TOscA=&V*b9k^hbCdHH&qN+;EcWT%#u@I3!LI-3A78!>=o!qgfEht|{$q zc##6XwP3vEKxj4bY)I1x2S;K)u&G8P5yC_wT#6)zTAjde1STA%*L*1*?`q8-mm+ve zLPDa1QwEs@dHirt^vBV7p_jaT+#mD%eP1nz*$MS>H$b^Sst@NN`&fPD{J}Ia5cGw! zf7`LYniYts(J+49>_` zVpcP1qV&|6;Z$c#BV8-v@RioV|&)8)pdvl5IrkUGgzK-O;%0wqE%Jf~xx z0{(3(r2Z(Utdd*vG3??SGG^x*FznY@Tiedff`(ZsvM2IPi?P?zh=a4)RB+GB8R}Nu1n4n2bOw?_zo9NO=0e|ztKt83FT2s z#1G9;(4LZ#l<#22v&7#T%R1J1YEaadk0u-nbh`1AQwwiZ$cdIrcUfX|j<{T*OGl=5P7Wu53g_{dS7SBJ@FrV9bOQ8jRtj zckRP_N7jKkr~SQNBc_xWj&9?<%n_x&80MD3(p)^WNWQC^At*WUz^$T3(Q&X>?Fal5 zn^{mp!4E*}_IPn|5A&fB9r{)^R!_@Eqw|5q#<^y@Uyz`N&%3AxTG5L$G~+JNGE7L4 zs_(qSmU=Bj)i z^!BmCLmB%mA1&K^u%0}~iTgY%>s@_lCyYK&i9-){!xS0WZ*xp!&ZGiF$RkN-0~0yV z8N%ukIjh07k5f)(j9>slCFc#DT@%}W!Eo8YbNEMmzB^VA9YLdhPcc&Z0IEbpbY-mf zTx!T;^B?HcsO|H=lg0%yJ4Y7nM1v_502r5ex{*igMrQf zb8}GOuK-m6`=>!Ag;r8hcF41ua-j#G*O>&`1W0jQ zEOe}O-eCdr^cpA}s=2-OX(lN=??JzmCbh7RV3f{xUWo>{nO*^?mX^5HASNiCGR@5_ zw-pSzfl4&(dTW9?!blD-V5FIxckHS{oY*}CJl4#ZlJJb7BmNOO!(t&gIJgA8mE$gv z3j1<+Tkim`4SE{Yo`4w!#l(_Yfqhc{8s0u^k{GO27-;=G53G129sZG5ljY;KbGvsp zhhL*IZ-I7r`zr=_(u?6}*& z9%fXcgC=m~E24O846UBCkG(37AAR8d5oWuDi%BXG5&2w*yN3tgj3cp;dpT^&>FZa{F2#_|123Wc_kG*?sflQsz-Ny(;JURjG$c)Za~<7Q*k zgYeOYu<%OCf!{VI14I0QPw@DvCda~pWqEp2c{@@8^{dzfM=?Rw3yI$P)6OS%p1z1#hf}exw_8eDR3GXZy7_cOGPZ zSlYEjof2qlJx*cdjgEa1X^A<3O2TjKwMhQf?Q|vD(M9OdtcSu#lcHBrXX1l)(2`A` z{lRn=?@gCxrtcGnI`N3q)uTG68tjR9-&gSaAXGkbVY(mGY{_>f8FO_&uQTwvx9Ndu z;?`crbNg0MId-R`us2c_Rrs>YBH@zn-{V=YhK!Y$;vd;vfRMQDf_Ls3lQD+<>g`8m z$e}V?Tpfix8)Os%2(jidH8=^{exfK|NlRZ zXSb|n8_Ra9Yx8I)L+-nw|3h^MWg*ON|hd3}*UKbZSbc z{ukOyD!q#nSLoGwvCfs}74@{ku*|64_yhhvONMltc3^7%;Z);2*>$vTbM-0u$W?^$ z%Bo*smdh-60_73;b;HRS^@Yx3gL(+5BrIH?!F=SFttk>6b1QPiu}U4mD~0QkJpmN>9^(kg!DO>Mn3$Kpt6DPlX3S?b$!P>he9rubgjK?P z!`}R9v%MU?FK6-nTiWHQ4)s%jd7jDs7gbj;p{~v)vlQHID1J2~68fEmRB+4x9E?&U#4XErjYE8WJ}WGZ-Mgb4;#FT&yctW-jK ziM;gHL{j?)%_kcv@s*vVoTN>~dvJ1yKU^1@$BR$RXea=Db- zDYGI(YYm#qo*_#$%K3PnZ#Cn=FC!Mb-r-qeJL!^IZ=$xt2rKr+&_-i`)eUsCVHKpE z(SY@*ZgU<^<1hUM{xAg?$AVv=g^#Qo13+nvk?;!#jZsJN{ru!9B5ifzIhTb?NXVGv zOHgtS49Y>yNsLVuIt>{+RSl2#uoN{R-nA`t|A;9BhcVqR5 zotq!G!(E8&?oW?(6M+!H=h9>9z;7St;1koy6GwAyG;+4yk0XAmj=E;wCAC~B|r zsZcRpwxM^Tmkv(?XH$sWLWbRPQCR1%rHw^@H}p9<`K1Ov3NC(>P%%$(Jarznd&_x# zetw3exUGj(F#HmR{!nRN`Esd27ZMT@oe{<2c7J4O$0Se`GyIG&cs*|DURhihh3HJ} zdol={73$Z9+@xvAs-2HDiapENELEu1zP5tR*^mPk{8R#|E!$PJi>2h8c|M#?gy;w1 zIwn7F^{8O>$9&!V9zgoqk>9Exo&4z@=FRSd$8T zWSbtXLpQ9tl2}PwRpjn5M4qSHm;VQ0Sl6nuB4Vuj`H~^H>bvENp3DA>Z$WVrwoNw zCjMC`dl2)rxTXY}7lq6P{{sA>ACUpQm^rUB)WH z5rLf_%iI_z7hApJ$`IGa!iwBHcpJW0U(a7jBJNJWZq-8VR~a?I%l8|eXql}pM0LsJ zlw59KpAK*FGV+B9 zM)Jm3gcAM)2Gg6E%;$yri1Zki=$Xec>e9+6f18&i(8qlUqDSW%CT4d2TtDx$WoSxl z%G!#pX13EUS)YbXVdoE13*g=|&1f~&jzQ9<>Qorv(>ki5!rmg^jl8$o;`*voK90g6 zYh?zI7*73}i%CepWJ^H>J^`>c0j$6!%*<#VScnEZSFSI>#6Rp3#_p83g9B4pv zPsBHkJZo1D+1Poq)5cM}ZGZ|FP*mJ17o9vK6nCVHQ9qn$qd|wvL=O={!ogqnyv2y- zSEh<|4@o+vUk?(OJoP=M9YGYCnwj~k6~)}498LBEo5?u3kodq4d9FG9DneYgk8>=m zK9ypaWu-g#S&A~u3am49mjC;AFs@TzhTWdmPzD&Vwt_JdgDkJvE^^z&0RoBLZeuTS zzOnIeLrn@lGqtz3vz-6!o79@>i~QC|Ci^!yRR(8Ft!q28aK??zm;}_!`Kd-8iF8A5 z8_(0s0YugK{RLZE04W#QzE`B{pWu6DVQRpugqh0+;Jg%CRen!OY04?RqFWMpq5ZDg z27LO+qtaew2BA+T%h2L}Zsx_T2?+@y-rzG`Uw@w0hjM*iLahr(Ti~7BY=+SyCQGeA zVDs2%{QJ}BLw#*N9Bu1lk70{itm_!D8@wX88xfF0TW!1{@^h&--^TEU^-dWn3$VB2D!QEb%ufjsDFp%u(c6L zQR4--YS)>LJK%pj$r_L-6+Q0%7)Bw4;eraZapw71Bj{AYJ<2)VKzMiygei0{X%{PEp3D)`gy@b7g_f(i>h`rQ(lhv3AUHlBb^zb@Z z+G!45=*i7dWOKmcHO~1Dg=t+#`%j3Lja|CT`9vY*FtxB8!nkBFu0P(U{&O5#(OA_O zAiNSib`inG7g~f16u8m1xUkS~$%{|RB)wkwIcH`0R$<&MeoHwQgZ5&gG>NDYpnYVy zICK(jU`EANY26LR%b`B*{U)wzzn}EmZGTlSgMO3{yZdD~tGAEU#ZaV@Kf2c6C5+BP z;4=={>gL;#W>?-X%>}dZ?Q#(+raaHJ*13n{GTYo5a%5-ceRNEI;>Q+&VUily%S7C% z?IwKLV4UgcUdnY#3x^+{a`35$1N#AfeWG=6$|5~U7MQL^c(h^7kQLLp<5 z4+(OD9dYo8cdy4Rb}xCSlz+c^0?e=U%nn(o(QE+k4Si&5$sa(lCS!J_^|#M6n`o&2bI=N5fsu z6X`CIz&4-Xr&0mE%jf2XV}Y2hV%q$ozb2pV{J1gQFSaa$2wS`v!j$LJdOR=_1JiUi z1Rl84dk^agYYFyo05yH;!38?vgNET*;8!?zd}kW1kKc3H-Fm5Ja;fv-w=BJKDoAsn zI%r^?VL_snqH&h^8dcw+_4S`A{NP0VLpLk#)dn zuJxa9j7nShh`h44XjU;9;bZfvH%=~JiJ5wXFu?f8rqW(l#&GN$k@1Ngs99OL<`yGV zD;ykB&4Ut0hGIZ?L27V9McOz`hHwMwUn=^ONeW%}o&Ot6h<-wp1vc1jkT~7W8Rkeu z#r|^^8+v14vBd_Gn+TIye!_2t|u ze@Ru}1H6l1MMQAcW>Yo~;zbe{sq+w)T9ljVhv8HhE{LC=6S7_LaUBH;6%=q!%tLH; zCix&0!H)=9<+ukWpSP!c;p)VOvdqN9JUV<9MR4iB09GOo^Yr|D*eV)@p6%q;QYdco zMe^&M%L>?rehsIl5rg@%oI|e7gHF}< zYrys9#RYc`$&&bG|F{ZUvJr>`60gt@K*<|Uq@`r%gsP(V0s3<> zV>(F9ax4D=T;L`f$nmQiN&|Qym_8%(m~Pag7z6}-s(5cCz>0?#5CC>W`*vIwk_FV{PDG_f<)(pH_s+%h zH0dUsH2x@XgzJ;r^RCCCnW*B2>nl-aG2h998$E6}sM9+0}3N6#o% z@)`R2`jojBcTuX|-t>&k&ZU$$e|N26q2pz<)3Q|};E#WP$@caYt!i{#Z0E1HP&2of zzPaZnTK6?y)Lfswl|qA^5Xp#fn}!*RsLaK|Ly8e&S)HR0ql*Vb-v%~VF7U%(Xpp+gIxu{cQxr3a&+n>1?uaIiV=z%eR*-l88eJU+k!u&< zy|RWkJUl$rpgL9ZH{oZxgnR&+;rI6oP{K06T1SW=`y{uuUG3Vk_;gDKt4o7sF&eka zDtofF8|7y5D7S$Bh7!3K(^`Z|?<`IsLV`AYOz-pP=6%d5@5_G|Hvr`BVQoS!j*Gx! zIl#!5r;0d2E-P{8Gt_1ERj>rLN*Ti)KZUs#g@+rWm9ipHeQJO1zy`DYw0#R%+*)SED@i(evuCh^G9ZUy*DSWwqU% z0|}W?c|$p%?d%O>P7Fm1v`H;6oJ;$7l9ri7Q26eS;a1U?3wm%=1b;y^N9D)t^4R>t z7*G}ErO;(%6~DSB1*sH`*Zz_9+ul*stp0Q?z~cU<`8I^e3kzAI#xRTi$D)DP4kW_a z>S5K+;eOEy5J^?$XQ;Z7+z*17T4^|w<pMJ>eTy5%k>L zTo7a_wQ<@h>ovMno$cP@qgkvzlx76Tw%rvHDK_;}KPzE;ykvFE#9Ng*z$Q4c6dGAo zSAL3B^~o>8`>f>ppcD~Y_;WRS`MJ`Wt~2_*3Ih+51_OAQ3Q&_GxbL26@tK|&d<14& zG%6H)%}<1GS&=SLKJ+wIf8o(1~)jZvaY_wu?{CAp&R3&<>Bg z2tsWxe=`BKewn7@GjY?7^=;SXdicU6xQCRCo$Rc%g>5DK74Hk3C0}G5CoAA7w^u6s zf(lFwb|g^12v^1086`sMLE7~Ao4iH3yju3j2-lVW^8&yEkcei$x+=tG z`@cJU7%_-0;9_Q$6lEf>q6*riLp@9;3P|9X-F()*ZF$gqobO%$=ntD#FfOOd!G(A1 zeMN4)tfE|24?y$Tt5=8T51IRLFwS&i!5_R4-0|$vy*3!&NunJkw1NbYNMa?jayzIF zZx_d-HSC8~#2)8agEV|87stzk0Syh!mOc&=hj})$oFjN-n{_d8SaqsChJHsEBEZTR z=` z`NcJf9=L(aG%LhbCQM5&cj*xfxwyD^GcI9uZcwM(Z$t_9BIaXEW$0xq21jw=9bH$1 zUS!mXw5are*s!(divn0U zqH!N+87_NieGqV8hY8=WVjjpwbQ4@{<)W`Uk%fX`V=1NT8wqi6`oXlqWkc8&KCgpl zq@l5syHW7_S+@&>LRf8Q9rt>e&2M3j`@&LICr&cDq3}`0##X_xO*n)+Hk({xog_A$ zJpX;W46p*()a+~!r8JsqM;q>#4)0neMl*duDmM>`d)nMPD-Ai`GUY<2B*EaVR_o+n z0g-@<)nF3+*N|`-7vh(1Cw*rPq+Xi>4=Y`7{yi0g=#69q@pXx0xjACrX6`Li?5PNU z<y;+eEx)AfIXU$@FNu;1~vsXwFw*J%=#ogL%^^x5{u6ga?JgI zjT{K-{ThT3evDyr)(9=g$q@~!G+F|zob`rsWE!4IE&3{u0zJF%Pk(7Nw|P(qF|Sn) zU>9XRd`c3cj~#{;hWE02S>|+l!J2TAGwMH{mTcX{+2>{FL=pd z_+bl;WZya7c0~JED(MK5Y+BfhIw4kfYH*Qi3AUAB*OC2$&;OnA!1psg;O?@hUCU1e zMDSJxT@s+ckw5LntelI!9ygfNRf;YPeW8leh9PGpbHd)(C9mT+k~e2O?K7Ukepg0H zG+q>80eHYmy0AAtz;#vzQdP6ENW=t9U~6V(W?IHq_ zv1e2;1LozPcCxzuxYl?&f>94k?dM2bU8%Q|Zq`7?_`|-)My?Lc0*}O9Q&TAesq<}z zar9qUo?sqgaM)4l+C{!d8_*RsDHHyEiW%{A{2xMP2gQ>jYGcFVa5d2_1$YK&Y&)Zd zBots^(~E0=foGb(FnFHO{TLX#F*4BU?FojVk)C8SGcumH^SnRn@6WJux^wECRp1UY zL<^Es!em0zto%OA(Gi+y^43FuU^gmZ#w+I)_1|yD3kL}xhy(7aZ26$bpS~|Mq`3>n zmA~q`&0$t1C?+i}EvKg!LP{f@l^=$ts78LjoowvV*ke%s@J3k;#16F8Ahv~>Wx-Cv zV=;zAwb@jnA%di-WJmKnGZNWS6w1E90cNfl=WmhU?GZw~eqePyg@ox_<ioK03lo zN>n!%@A>HD^t~Vl64l#R5zoSL*_-UYiIFHqRKgYO&F$f5xBX$X2R3(#I0F`WQTP#o z_maFR7FkWlc5H^9BNj+|8nCekmLfmVzeGJe=;KvdiUWY4vTI|nx=rv>o^*FVO~E7z zR?OBGJ*^ls0#3`%#ZGSl63>&GWB$AGgR$?dKme1N*S^382K$zFhqbG#Yq$5VX;i2$ z&HpZ?|J}SBG{A4`onHTFhM2YO-KzK3*);w(0N)U^#;T&K3XQn%gO&oW)+;lcJN^`7 zjVLW{pw!mVuQ{kGZv-kh`VkA%xF0O9_=Fx~=EGB>OUJ>bUfWj zV?<3%sP%kyarzQ?4H?z01+UWgo1Q$&;&3bkav?YjY6S zc_GbcLEP!|lpYXoS&3Di3^Yh1<*XeD3S9{CHDY%0G6tNfp~Pm(s|ZQBdZK%2Sb(oj zc$|rFd0=2C`iwVV$TRR>LF}OJ&_UfWK(z`OZ~dH)!7c@a(*>8Sh6hQHFZRaWZvkD( zAQ>2zZIBmj#|Ilw^Wl~sEK1&b!ySB!jw{l_;NQQ3ZEhOInQs+7pz@A$pNS692!bSD zqDUZ2v;^;pR02H*_9#X-$1p0%ms&BGqZ8p~t?yeT#ggCEh?-gmsd}@4y%SiH3fxzP zr|*HdWGz&DIO2cXun+|ME7sK~;vz>>@cE?f;toi(otPWI?6Rs0zeL5zg7bXXhIgDt zrdXVJ6#*YbHyOM>$u(43c@K{6SU0VO0O?G5^-NgwC33|R1_iob6Q{QI%nzvk8k3S8 z_fn=}pN9eoycbgJZ3mZ%0EmDYH@chp$q|lyxd9A8|4pAeaL(^@JT?mj;$K6}6+qn) zF}3_BqPe-5aebea4|1OZOeE!V-?jmGIA?Z(ImZF?31ZpbB}f6O|K%H9YC@j@y-6aP z^~r69lnU>w6wW)14;V~Ru&eo~3?0qQ-ZI`hqjUdAkiwB_SX*DhAjB8#Q^a z0@E7Nfr#kX-sH^8XuxbWCweulywwLutq6jk28P0L3j;r>!D*u#3gW?agIDm;OEv?{ zO9h6lws%t{wwJ|YQ4ae4Xn&Jl&z3%RU~wF>Wk~w;+DdlY?9>gm8AsiQOuaEpZJ>&S zor*Q%|i#av-pt##miu5JGWFO`$qMe?oD%AOyi901~2|1_rob55Fm zKBOXI42ER&0*6R%(YGe#TmD7V6*o(yU?`yJ0E@g`3bV5}8qO4EFL6ob+e6&mN?qJK z|2`3grJ!i91}4Y0yWLEnT?W@6$~0gyA!ZV#zGycN3m=|-uD!)0~RkcnD7 zWk_P10d^w;*Jh@odkIwl1okI+?klg3sp$NOEm`+t)&wm`L#&VpgRCo zg6Dew{oMm+k)m z_$09MC8OOsbC`B>2E}zBulEstdDgm2Y7N4{si>)X+DteID53{twxLB!ID|9iBu1PS zVzI81rxNA*2=9!>n@zV23+8UXq&X<8|C5XB$hZ~c*!A z@QKh{09L9Dj23rFYtu&#T&XxWnvT6$?c|}C-I~Q)14NgzRevC>@O#sy%Q)IY2+y6e zsodw!a7&~q054B0HmvNto>i9KU;+LDRLjuc}n*on9A5e+8PrjK~RMeCq+mZ;x zWMB9M5f98vf@w}P5<-)vGljtqP93>9bMqgfz8#1p zH4d$7jpCBpbHed@Z0e~wwj5x}b<_}>E4u}J&|wiujZUrbjeX!o)F|FAancu>qg+yx zOoUnc`vpehP#Q!$<@YRURtI!L9gE(r!R94z8qk z+kv5BfJ8$>gBe=e{QX;sp~6ERrPGj4t&%-gaXvMZxLyJe@msd)T}pEWtWw4IaJvn; zr5()=JV{nJ*)7gEZ=xS4i(*K+Gx!_%TUvZ zK0wv*a#T7yI3*<|4z}@~%zbQsieTHeDWFB`0?qGvr$)q|5p7j^%;mV(sv@20{)FY$ zoL+4@&d=f#EgNpT2{V8O`RR9s@({uDm_@tZ%d{uc(>ZB^Gc(DEp%)Y+YGTrfX@q8`6Hp2Kt8R}Fjx4GELu7VI=V3zm!^En< z7_>)pRxS9(4i&d|m4>`cvb2SO!r!cCxBhDWYe(F{ck;xcc+9jIV4wGaBv9nNyHNxQ zVoe?22DtT>^8?)l)PiluH?4ZVUJ+po>)*k2RV7l!P6_CqoAPNOqyN~dFz#gBL5jVw z+SDH3;5WCA;BdD(67#aS#w|v?7{FLcf}dEV*l>fh($izxSD{k{-G!G3bcwr{7L@}A zLaeB0Jp)Y^F5l6%oov6*RpRu*a(XIvvnB~rPywQAoe0qd?`7iM@z^D{J^Od6S zfpL`V#W}0#|GGMnE%FbJ`1P1HvaIp6)G~YDFVn=*~Qs86aK;tj7nO`3X{9`Aw zr0flHl_iqQqWPzX_at^>~YpcgKK&lOI4sR-i}GkzjHa(RVLbj&sJ zg}g)Hfm{kr$r$X}8w37FlqsHHpmN-9^&-C^2B^^|2Nl^IOD@r`hUk4Y^z?^|_4bKh zA1gOb9J@ubf%hyUMN~lnRqtg~?&ALXT_Sba)b)>e$$G7N?!~Y%v=QjcxAD?HwY3=k zdy+Rz(p(_(Ap#sON-60VNmTbz6kaT-JhQ8F&|f?T^4)e`_TBDn{{>~I!($4HD`&(9 zYaVu8(goB=l>zM`=y=|HNU(#LQA3#f1b*G3VRN=5c1xmlk=Qlv=U?8jAVO{fpo(q6 zz#*2)Xt6(d9VEUKQkWYm>@3r`9C$rCD*9m@oSyb~zIskoR0MYdSks&PTWhx#lasC# z7boj?K?SacWb#j+E?Z+Son1f+&P?EaKqPPM|2>*SfpzsKgTZCiPYO^P*LyzRM9OXt zekHhoRzozed6{HsPCC5W(=`9fS;QUIuG@2mYd5D9hl471s|=PWk|ajS7(H>IFZt>i`1_i&(Yxh3USb?n6!xlyLyv zPKDwEnug|7dO61bKqfE-q8;QI#^eYu36>nzjdx z{*@($F&3Oh#8D<&V={{SJ|ukbg?2QdS&SSM4wwLmI?7$MUCN0kw+@%k+c`V;JDd-j zS6T+m7x!k=8z(i)MO$K6-5$;HdOvYSM6l+xxYGv&?!ftCA|oq_R$2DV($^Tr^EzI5 z1GVn|(Hh^|mCqdr#yF=OyJbr0c!7)MAwsn9dK|7EuHdH|;9@&ZPtuf-} z>KI(o4pVNm?h2{Q;*Q zC>~%Hd1={=;>(nZ<6rB><#Mt^u^4SAxNNUQa{$O^Vp#9aDObLhTFzI&x(g=ECE!!D z^F=T+l5?TtbZA2R#*SJ#<`c(uu;%?6`UWVZ+BHP}w{G@N{{rwrcoN@WUQYmbMOa#; ztiGDsKy&7=ls*PjAFsbntFEF<8dD}2I}ZJ7tRq#eBPx{DNY#uA{>MdJv@!A7@&|{` zUm+(ICEv0(W|>TG*z(TBpWJ9z*sR&yK)(=W1gJn*H`0)&)bt|uMi{hY<>KlepUs(v z@z}dsOXA$8LPi@yQNk5o43Rk~Tv>~IjHWcd#`xcv&dx7<4Vm&#BYyaf-mT|xNd4|S zVDFXy?4{qc2^(#nQ93>7li`by8bg@??BwBRN;v5qZmY7c45mqRgGql`Zq|e0V|o3oY*dp6EjeLiN+MZ*KY1 zamh8uvT`c(ZBv@ccwg1O*Vps788x{4E|i+%{T?3vREyN`B@0df!c}o*N7KhP%gd-F z=%0bwA`x=wumU>A_9E&0h9|)Hh`kwtdl_;6T%X539v~~kefiYc?9J7T;Ibxpy5S$L zU)A--@1SHND!W0cZi+_(TSCsuA{9a#KxwJLuE5Y3+r;H*pL#=ymbSwY)%_l&86t9O}AsTxQ*^BBY>fGYBnH0$Z#|Q6C>+ z0PUij75<|fl#t$o3ZNdT{LZDyMLqq&z8wnX*)S(N+t_)sUE-AiZ0si#{T!%*qr`n7*W<#h#8f~tT?m#%JuQ4nJ0Oh3j^AQr5q z0;iG=hl>@QEXxX;p?3_^92_rQD<75IsG+yAuaon@-G=zyw0raEP|B@i7oojLS6X$&+wLo7?{3{ zC`b+!Q6;0XOg-#7J(}-0I)tmSL6r?&mcs^LF z2P&%Es!lkZ>@Kv}c^xL*PJCdOUjRMkaYBA`e~+9TZSDS3i+}?aS#YO2h$TGTYw9k} z#77sis~vrx5UH*|itZlbh0dK>kENX|fLZPrZ2f$coPZ<^dM(po#E8iD$x@~eO<_3tDXLCCj zR36YCKN72XDob*eq>WG?jgm<;VlRi#o%QO()16?fm)lbCb1#$B2hCu!oXL#=pOvW#u%*bVN z;;fOp9ndj+{$Bq9P|7!}!n8fCrw3AKt-M*sQOW!Fi%0?_PT7~MTmPUFwx>zbP&rKj zCODy|`d3hJy!Pkc`AvYIY+i0XQi7roiC)_Wh7uyHt>h(wPsn+>+nVjZi)^+}=9t|Y zp$h#It&Y4pJ{Xf#&uHxqUMtPm-CAtm7x%hqAgJNw=|#%TU8g#RLPV(8oa_DPVtYgM zO72RbXvQ&8RI>y%L>7R$Vds62U-;KBPB~nX&5e|>eqkRSkKhwJvb94473g~+xW#dQ zs`UKxNh@D3N7v^?ImIR7h(5B~j`^IX>S-28E587BDH_HXFK_n}e~i?trW4wCp@0Ad zUbhvM%_^ni#3$!cdrQOZ;%&TV&O2QU#WVElL;lTdyxM9J9<<{^#g-!hQH5i6-y-=M z#TLTC$@JKyq$ENfEg!0PH1`3J(-kKd!K8Crw|=4ABz;!E1YjzjFRlTkZEbCyi-tQt zdEc|TGqcYmW1g7YKcK}S?4e3DVVw4_O=Z-Jo2pUp3!xkyV}dvd?cRFai{Y0yNzRbU zQZQLi&o1VJL8eDmB-S3(tM*S%K-dIkUlp$j|C#PZF_~9Q&z$QL{_;wQq{jxuD*eRn6DIGIy;%*pWB*MuIT)?G+Oj!MR~2CQ{T)fv_^>2r|&M? z4yajO`~Ntr8QK2=dGe+{-u`}=dpYbP|6wV)T2G)13q#p-K=Jm{gDZHSXhy6Dlg?#{ zYr>T&mBvC?mAUFq%Gv>MZ*MOS`NY^?#ptVn)OJ)|CL9?g#&?$1NVINq=4N`>;dXf- zt;iTaK!=(5_#8VsU#S~;yAw6&N(8Xc@YQWH`qFVV z-i0h&YX8uE69d6{9;A&l_xn`OGZw+H!ZR$RyB*?pKz#tCxU#a4Ohw46cyB=~+8eL9 zU1(~DCb+Q#3pd|gp0KZSRc9y~6M#d^%6qR*Gr+|#rKP18S8}X*9iNbqO&^T9%fnQj zKdjmPMi^jPcb~de7GJo}2#?dK@KBDQbi3#+{ppL!RKrRD@Iig%pY4#%g|waI=bK=0{c=@gvuUR@C| zJzo+9g$7G#=1ZAbqpeoxP0fX45THl4os&Edy!;B>B+Lu|TyGGsYLT`KkchCmQV-O1 zRAnQd>NhRXC8+NMX8UpKUQ+VgijeUbKJ0`KW`kLMU*(v-H&4;;4Z{y3F21iZT0*77 zkM`i4O%97Gv^*yWGJH;z;-ve<@@9Iq>_~K2^<~8KSpelCxex4Dnx&g@Du&U&bmW(m zjQ8bV!$6gPzx^s;W2!rjZ`BkxF@QuA#HAl@auQ*R3d6#t{k#E*reup1o|eKE5E85QH7oL9f)di z;P4$sTvQm)uMXg(puH@_W{A2AV=|Dn451U1wN*Ra>2R+RoGmHAkDociKSiKAQ4>#tc?n!)(R9(b0ZENUX^mQO(_i?bQidN!q@!o6to~#%PYFP zj=e)*Q6f-h46do3QPn!-Mml7*OSDu}IbesRO4R3RhgNK{CMvPB+;${+*DsVGh531Q&%u4%xS+G~{;!j5e8R7YHWIJ6Gc-7`MNQpvG=SE2U&SFF zze}|J-?O8XwN~I3rl_>WsGc-Zb`GZV0flOsghFh^Xf?=>k+5v$@a&8QrizA|daG^Y z^(^c+h}(8O>dzn57a(YS5RkR}a9vmLmQHO8X zs*EHT?N`zhuJH|`o5AXtR5)`CdMPT6$O1LSONIu|n zE&A#HJi}LjN#``-e<}r*L)PMEUxbtBIkpnKkp>(?a~(Csza~*GlIYR<)a*23?}>l= z7H-VS1Bm&8I(9F{X#!qrcT;%Z9`VdMc(yu+S`P9E8&*Rc3Mlo^yS4TfHby@nZZEvn zX{hck*j*mu&q1x|IH9oELq2T9W)WTXOsg0GVH)Jj)|hJ<$>e+{pM<*JkXSO<|V*w)gJ~#XKy~c4KV`{@QKDMGVSEtk6g)7~dA!Rbw3#WA2>F8p4QS zF=eTzc;dWBWzzDZ97yPX>x;w0&8%D3&2ZW*{Oe|*o3UetvQoKwcgxmugIoZ7UoMXg zRY=42IWEWQM7Fw#76PmW8*_5V)DmZ;)xWjra)&WFwzqabv;6Jpo4-rF=m*%mB#3MV z_t$Hl%2hZ<%crg?%Hafj=R&cQuBW(6em_UbprveZLn9)Pj7>~*uXkL}R{yi-c$Z^% zpZ~!-kJC_73(CkKs&8sSbH@W&Ru~bx9(+(xP>|5)A9n;Gk?|6TO8?p@W>baK>y4!$ z{tU?M-O8L=`F!s1gba;a4g6<$^tNe3+Z~8{>{AKP+}wC8t&fnwlCWI|YBIlY!Y+{p z22cry3y1AHqdsR6ZP-a5+^FQL*UZ5prKwi$OT4i$S4loR)qgt~ z`Q707tYx2M#q^yK$OO43(BZcJ-MI4sbNOJ8(rrc}Fil?qUMK;jFxnh&r!P z(Z9CmT6z3?P}+t&wh6p&=+h8O^R(-Y`L^a`Wi0IF@uf6rH()3c;*%jE!pg`fq3_B% z2d>TgX3S$*lR`+cUAvz{d0XeQ`$jOz(h6!HHsn zPPE9&7MeKw2#|6RB(E^Pus$%4`F7m(W;u2yuk1Q70Lx}3XsO6WpQ zm(eleemDj~j&D%kUjjB2G-iK&h2Pveg_cr#QI{ZM2rxNa{^0J-XY(a$Kq<@pGacv! zS@uVjb!8CU5B5Ji>yhn~!U1N=2cU7*6-e#91z`I|gTv*Ltf8#zx1jHilgz;7({^R> zms>rjyb=k185suSM9mxHgE|hT1P)1NtNqv*s~L9JVvAuim0U8g!OhMFx71}DPH&G> zW{Km@X=|qivy?*i$U@gF=;|sMM;}PZfgp!?)rGcwC_4@-p#lN}TOhbGiTMC_&A0n+eaBINU=y%2TF;GLQzv!7pq$6 zot#t(bJ6ec;6h_3Mv4{bB8NG_>27{-max-zjw|_EM3Dr=#5kGR-~9qrF0as18haZ3 zVI=JF;lw=*I3@9Ud9Q7HyFFKGYkN(2*Lvrmi3xZ-#C1$329pi*9Dn zGlb|50^={?li=BcApKGSmNOR7IAP41ni^i`BWx&W=**G4qf)+Wo)a1{u8ZZGj>?kw z67%b85-N><`tiXMDl`r8ujiwj3w82ES|VW^Nady`q_?>M_{O#DF#!jt4HifmpV!;l zF7m26v>#|**?%sW=@#$lmtB0r{8r8g)nO8gzfC1Oe(*Oo;JM9;4vUDugnR4ycP-*} zxtkL2`2r!1Th)_C>#KMdF3a%C=MfVm7^Yts;j)_EE{WcR+gG?Vm$(9pVdZN1LH6lm ze3i9iJr%m_^{zXI=nna{9PhOLn{96hYEri|X_B6zIr+g7a>30Z>?DA<8ZXE$oW_T?qU& zQKixJY$6LNdFguy6Q#kkB&xwrP(tKWE3?-H3Xhs)%l+# z!cL^WOlC>3Gr0&(Kj!GphJ$CvP5PkP^T@E6P$|r>hSsO5NLkbZhQ&Ke8$WmN-@wiN ztq(IesTTick;h8v!KxVBI^#XXnVx%tnM>D&@2&Q4l;Nl0N;eugMGcb-Mug_SCx%5- z2-}@>h`L8!PeJ5CxeV+XDBArBJDw8oBRACmA!^#NTxvm#?xN7$@rVWg!(wH2Qy!~d`9u}f7SOO@& z97n-E3JNcn#C4pSoA=1vcab3V-*x~Yk>8gzgMPT@c(AD_;Eb{^e@9;N1=G8!t}fBR z&3e5(nyIbH@RWEmyuUYM2rD_!5emqyDK3_g0)coDO=k?Ot%b_ZWRNE^9^qm>cmR51 zC_p1hslH_b_@V;7l|YURfDw}Q5H~b7Qn1i``02?o?N$m?lz?P#dMrtDey4n2oMU+& z3?CeXm!;qJcx%h1C$wsdF1~)MKxW5>ak7}kKJn-rI2k%b`0$rwa`W|my|-ipiRyGQ z+|Jdtd2vZx%2!F!r&=QdiJbC0ftpcc8(3+#b~A6Uarmc8lW}q3>^nx})ZT{Jo&*@6 zUS_)>Td-rsC#11TC1C0$?fiy;fss*A2WTOZWVqm`Ng`y>&Mp#3{%KS}D!9%FYs#te zL#3k?2LGV1mRy4caZG!>n*Idn)NOIH}`a2&gvT&Ux_})UZv1hdxWji+=84JG*Cg=hE zBW6@-B@yReQ?TnY6J2@vmWX4_>|Y=ifZWHqh^A+J9rr_)a2%If&*+>KlAPJe{;nk3 zSM07RZ(NE+EM%VT^cy1tm+S6~9Dqgf!0i=ZPwFf5f@4!w^mkbohYjr-yrU#!pc*Jn zL8}QE4DLSAiE97g50D5EsYa$uaKGju-+wKlnp}K%aL>o>I7q_h2%7x38htj#T`NziHmZ#h6x-h7vhyaDnU?(; ze?WNPbOL58TU>p~l}{NV*n@=7JmD7HI0Xi4Lo3Oe73_bwR@O2Z>@X-wf09UFM~slz zC3MjZ>CtkBy^}2zwwFr7yTE6hI5-2&Chv-y$bd>&{>Wy|&A;;D349@G*6fo6dU7a= zA_`KfU^fgmTLOMUSC}6xjT3u0Aq7v+-phSLdQRNTgh*e#x1XG0&5b%9k3wtS4#`ST z=y_-0(~j#V~E&Mi&5}i-ipoURi5;?$`bNG$&LN{7!&=y@jc+ zrrq?xYDmZal-D*(4ea=U?DL>xltXDYhd0YntQO_D!-qocP?Y0x?_iE^lpXPTYY}JP zh_elt$m-aNCXon+=-1Pl;dyAwO<3yuClF-nHQn>KzoBNgpPCRF2P}2uIJmn6zA>rn z5CMVlg4M=r;=@m)$%`nEd_B0~lOb`hd)gi)*VvH&x36g#iGGBGEW@aMlW=S1tRj>@ z*#r7bWFz++c@-g%_XG7wy$SN}3&J@U^mmVra9A9TlX=O`mGic9;TzxQ55MpRa^hS64f_B2y>yn;Mx3w2xJM7#GlYIwd{5y;{f?h}SR*#ljfC zeRu*gKrQueed=GZUVzrNf3Z$VR^Uy@blaY=w)~+j!Q;v%IQRjh%ii3AL{i5?(AJhH zY^Lo-LG0Dq#WmGriPI=JEG&zvP2p*5UFzVAP~I`T3ppIPqTP3?7Ek68LK~S`=b*sc z*B3rvXweg) z)Jelcx!||tTHd8<@WF3Ck%8WfR$sfZk-}$}vAcFu{KP)gj zuxHwtdpuuw*+5IK|1z_6Z~%9S9+G$K%P@Y~^dgDHzr3_mecmlLb2IY6d1~NiZoQns z8R)Odb+!8&IQ`e}m-kMy$X|^^#EQz8(O6gliX2A^dFD<2IdY#(JY<~?-mSX}?)H zk&`qwHTC9P{cwogin#Rt+v2sjvJ$Of4yMC;F8$owl(hdg1j8tfC+lI3ohY&vcar#R zW^PWtYdr5~&b$-RhmAELe%g*=K#{lfpkb15-SyA(bhX(N&VlCVi~MirVW`B(PPkE6 zgp}AH${5>GYT192GczMhObNsdc?+ONp)m(PP@m2E#lbE}nh*GpWcb(B)zzC)7|2lE zm{|WLSo>oi)R6QvWDLXH;CL5od19KpN80nwLSOWvqUw4(TlD+<@i-@1fk;r6z~e?d zUm65sGeaI$rd7TW%tq|^(M{VKiXV6w(*yKd6V6NEwFhP>j?-S3+Y+_~)?iz)H zz!xHhi1xbj2O0I?Af5D-PPB>z`0{5bYNru5dnpR()vldunowQeSb-^||}h{`z=KM~;3{ ziMbpWZtR-FY;|NW6FngnXZ|7mP3lODssXDf2I_L($fKvYpViSRWWn>ewjXOdTf8^- z1xF!Om?FnHMq0}8tt%u7k;)7sfsTj%kO;q)Druu8+^|eSx}95H;|>&9@q2$lB_1Rl zs}q5n$pqdxk$zH6JomxzSI911T324Tq{kbuFn^KihZ>MXt5W!*P44QVHT&W7O4Zm3 z61l>b>gy>Vk%;fNvDaMIjw=1f)dkYrOdN1>U!R8_0jsYgeLwmG4o?4USgG@``5wCp z{++37{s@J+H4ij=QHbR-Q}dd!g=;wNcbT@JNaKNZ_;pLgWA~npmP@E$ca5z@ta`Qd zjo{#vB4Oj96;0WZqNl?{DqsQx$FqzmUp$6pMjjN!*8W&7eOYKk!TF;c=c;291oaCw z0FbVk$;&d;BeTy~-0L|RwC`9O(+#GTBWfNuSLnQ;hqB3XR!~O*+(S zj4Wfdabvh4AOu*Q`5)|>l(nnEdS+2?QG7Q8NSL&<%q|r+2ve}j{#Z<~BpP)vjfq4o&DD5B85*#sAgipL3RRU4jO7LX++8g4|| z{>HsT_|#vOZl%(va^IT>#mYHfvR%nzB;o7(DtaBAYjj92V=NI`87Cr0xW>{!H^vvC zZRCT!y1EMK>KXdpLptz5Y`thnc9-YF&5|R|^{jx7oHWCjd7A2_FP zya-PE5)D?c$9DKhK{Gw#6XwTf_FNSp-1G+ODGt(4BJBt4QKS9PZFeqW1e&^XRILO? z7f~tH204NrTLKB16kV|E@mV*(cLn?@Z8Tl1jH3`ZMtgq7df)txp({z>1IY_gmk@=7 z$c2Zeb5tvjW%G5!I6)oIC7)yl=z~izT;RU9;_r#(iW{NmXn%z`EnpB9XGrgqf<+mD zfwhecA%F7f!QpAZ@BARKukoNFy`@hQcC-@Ku3WGCw_T==&))x9)TgDq4FlEFN*=5KU?CLpgs{Kd;; zAk*Y0Ca?D$6FUN9K4)DJXc|VTA76XjN)bwC7r>Y@eK+3>~0ct>$qLmB{8>l8rov@!HkQbD!~U`pJT^=rSg(@J z8<0GsezN$&u&X#}+#k6d<}rZKi(%5Iu70JB5Z8=JPX32Jth^++epGT@mf z)23G_+W8o$nk~khYpBS22EAUAzB!V`H4ZV-I0+%J64KJr4*O3gOANdzmPVp-Km;&G z7!4c+NmcV0`a%C@IV&q4N)+sh-byR8QQnqh3dAq}_eQ4v+48n>`4<%sHo%JDjnniV z@lIh7{0?2fMKtcci2e)BbyqmAc7UtX^ zoh*`4pjSz`3W1ON;Oy+{LIe|w``?fXh-B@7(l;PVlH= zISGn|Vk7rDyiK-jQ$f^pCG;ZXH!esvJM*iO8o1k*Wf|!MsP~~!0SH(WB2hU_e6Ms=)BFo)CC_REQ$ExV7$QV(J*+i_rVrqNx_+cOcGj_Pm@VLr!tAXpM&CJUkRSSe1u4@4G8>Ypz_x^l zMzgg%F1Vp!I$+krEL?;~NE9fr^peEmv7o)`i3K zDI7O%VO=>geTru_FhhK8)^jZ8LXCiplMUj-oYiF%6wL0c%)qK8#|EiN=_^oT*RU|3 zLe$owQZjgfkBoBUtIbpTx5&NAYPc;5F{dpu;A&tqxQXe{Fx%BNyU$ zKE0j~3pvv`LsM(hAzLH*$tRUGahF)nub1o2!7+s2lxaa{RZ`DK2kM)D><(pHRK+H8 z89Wl&`jb`ltUk(*q>+DP4aLFf3BB6*3Oo<}Tv2^}qQ0=vUHJB=v$ltgpr77@e;*&G z@x=b_x^qg;SA%g&tBiSkof}DEyqujkB6n6HXQ=yPV)9c|MUOSvmEndT(#ZE3`i5Vq z{aH}$qk8z!9YTHvi%bLfde=7$7OKFA!oslpSb6GaBF+<2{xwzvdl8^U&&Pm88NJZ; z2c@BPd0l;}VYh-gOK$}OlAC}e(F1$(pO)S>710r573?@)b01-z8qdYGwQ&J|hP7ii zmVrO9fPjb`V6Y`e${kwV1Ox=kc1Ly}9*;X@1uvmJqcyvGJ46QecN;f%|9>Mo>S~8g zaoblHj?7Vp4Lp8)1g`-#E;lg+&#p1Dv_9616~&Ral2XAXpZyJ=rt#neMeMsJanNCbK2;-nn(Md-O-nJZxPc)RbkM_AJ{o!m@QJ~3{q^5;qF%e4& zw@RP;CWX>IG$d<{A12YB%;0DFerELrz#&6E;h52DGZvLQH~6eX-!l2HA{}9Fq=Ra| z-Ts)B9p($$JA-L{6v&q;Ep`a!uD<4y2>Sd<1R9~-e9A_e#K=jACA&LZ<`qGuN zdnJ%*s`A<26v&GfOoWqqS!v)2*u?HMN|1k%tfI2 zfaG=CU82wuw6wI~8t(cvuFv}jBbFVlCVYAM`GLUG2D4HQo-*o^l9EL1v`m=xUnSwk zrDb%;%*>Ybd2oy0%f68_fg~4{(4`X*{||OuW1uCd20{cn@IMK*s-z|alylA8VYN@{ zOBbLb$LrCNAr-qQ$-DR&f#F9O-(3c%&bt+s{vh^@*q{0{3D;0c4WXq%O5+=`rn8`H zGSd2&u3jLHXoT=_(J~K6m^FOx6c~}=z=xnvsp=4tvaqU=w)4R?Eo?s^=ve6BWKUjm zr$Y4^L219Sr`~P;(>Mf9i|2Aef6F7Uhp@oiA*YzeqS^x%N}ju<%n}amto$=}bPUXr zah|xg^i&d-o~Z3uuO82rgM8;h##VyBav@Mvkibn6t~bL-x4u0>uhdxFehwiacanvl zWv$DrX@79qN~;+cheCvdWAjbv<<~uSwy0qN6_>(n$iqwMD{lx8m5bOQ0*ie$S z*AuU1lkwDvU7so5B>I$jmLX!D-LaL??U=a2a;BW_pwk3th6|jeDU0u;LbDyK7V*^) z!EoSbTZDF#krsLFQ;UXiV~VmAItp0` z#OnsvszVj_nwyX_)wN*vWax+UT>yo(LuF+IlaH5|n%w5k8!iN1ndU7Zx^Y_OM&?92 z*dde8d2jrLgf#2D?6d8`!Cfl!*$HQ7henccdAXk_7!sT}Z^x3`*;!e2+f;SHB&Ztb z$Re}* zV^v{k9iqF418FLY!54ilmQo7Nv3l#N1I4lane+y%`{zcRBNu7Tbl$06@B%d!D}J5uO~2V80J@s7X3JHZiY%9y z6}v}p#kS_niX`}fi!a%!dl`S@$T&6-A$|jzv*FP3q5`~Wye6nB4V%AJFFxG^0^!On z89x~g)#5FJcq7 zKb8^Hf1S2v3|yGFR;s5O$T?zu8Qge}nD>bxDtYS6)He-I8u`7Q@ZQs%_zMH8(})y@ zB#6e4vL64sE;4*Hy#6W}1z5#THAH{CyL#q?*5zEHDbmrpULUU*Wh?Fd?UW#IzP1OH z=J|T9gd2m}M&~DRrBamWY=4NywW4z46yaVt5snlTIOa9}o+Br5fZWnHM5fN4FVnnXk0JuqN0(}km_$M=I#V<>SB0pN%UZ{740{tqj>Lh0-4 z8&fy=rvxDj+Gj=of*2&A zaKtKyCVE*>s~jgoy@1*Nv#_nV`J&)!xN2(Uw zoK*{~xjzJ@9G`VMhG1ZP2>F+mhmRFPhZEMS0SD*g4KT}M3vM%!5U7$0Ab#%UbfMXC zdwo`VE&RV9te6FcX#CyBcTn>D^{MNPFh|9mqV7%*dGd$xUb%R;uKcU3+=p(_nrzqj ztkEe|h#M~J4}9MkXkPG~3Q9Xr%Z>IxqieW~+P~V(ZC*#-t6F+9RI~|d$UcAzECerX zbe04r2hs73`~=AYsi~*{_q)#MYp$5;*hmR&XG)B^>i|cpKzLT;&4#&s#B= z%NIAy?s4)^Pb^PZciP2;it7QQ9kjwvHyX9_KHwA|^Dmy&6CYfE8Zi~N2VNRZvm~MQ z^yT-~mPB2g{w|UgZbFR05#c=|9XZrkrZdVU0m9}2hlD^O|M!kOp$dzNHnS7Mf7FW~ zn@_H$mWYEJw+^;g13Foy;T?IAHeRXYCJ z*&kP4G!B{&Tb?d!&(OSHJKj2Yt!O*cloPBORKCz2dh*L*5BQVioH9(i_7~QfbDOp7|Qb9BN%wAkt)r~1~oBYdibeILWE&mJo@f3mn z62RK8p=b`y&b?J6f3erz+EEXs_UX|hkUCty@CNIl^5vp+kJX#^Nu%zv2W4M1J&-Zb zn373KoE2a^=)fA!MrxwXebxW3?t}IIYp;yA502v&y#1&njPvYQL&x*JMY=J1FDT~5 z9xU^S{#2tp9gQ2vfwQo-ruVGu2$+g8_O4YnHTr}Q`EeZ1@h(&$4C(Y!%ENWutK`~X zvSeqw5=IyI0COtK&vmsjNe%QKVCoc9TX$ zKbQJc2|i*^>_u1cWYDyzuFMAcG1CxXd==r!tnaKSnZ0)m-MtD~%v*+39#MHaMcW@( zUCgaoR?ETu)%Wi!;Dy@p`_Asz`gDDNv(Xcz%8-J&hOO7myKguop=3kN$$-qCmEw@` z1rfiWic}xzCVx#=PD{v3FBm%PPj6gZeQQG@PUQ; zg)$<YTI@ym-gSUcOlBaQj_@HnEZZmmD2s*G3}9uE*@TK z_ePdteZNreeV^9#3>`A)q?W42-m2mY??qRQ6-G?@;!1ut>f_@xeJ=HE-^>Rm`&zKvpd_;jtiTe;XBU4t{;ulig#kL?ll>_(7$A7?gI^YmBP=%ejnn*~ zsPrSGiE}Ave*OH%ID#dfDF_@KmrHamh)qkT(e(6!xNV0w9ovFOsoRc2U$5Jp{}bD zSp1yCrok|oJ%@b1cqh?1wOk(fD+{B5zomOH4C7uMy-?>jYCl{xat|OwB`10Quc*-Q z2TX+<0anfm90(|Arp`avB1p@)T2$;C1bN-@m{~-c6^n2;nQQs4!QU2 zhvnOoOFSf6M`lu{n^>u;EK6#diHenVxn5(B{Kbk?>rDFirk^SHcpSy)!2^V8 zSvU1Q4vwc)BI4MFK*j6kr7>1sBXjK3`1P2vgU%o8J*;F2@m=ZNwa?@7uU}x*5{l2T z=4AHIfKOVC=GHOs`}Lq&LuhSASjoOYGJbND8XMg1{&`3e7yel;)l>stTTBfW&0 zT|99+!5ZIHue*^}CVImU1Q6n2$A=Zt`KVG4@J=b>n9?#T0WmoOtzOuwcNs|m&Ry@& zTGa~1u8PD! z%i@49hq~dD$|fPfRo14}Ss>>p0^gTqBcn#v#yM^vrd$LxI(l5|>H+=UH+N&feS7`8 z&kJ-Gqai|AK3KU%(wC#`UkC(Kf2Dv(T0^DsOPWP7dV*ytrk zDXyN6aJ#ZEXPDg!3W}i5^HyU>FuGDbm3sy@_+HYW@FnM)8WkvX6m2WyJOfWx+i;bX zX2cl0d`ORuooepey42|Uq^4_o-imY9rCP8fH@8n!OwhVW+Y!Vi+jdj+VA^5BDt1`^ zZ{#u|ExL1$W%7mR>k|?7YRp9ha!5`o;>k+RhVU>dPUD<Iz$GuQJ5B> z!Tf>I(Om$ginp<5kOoj+2Zq(CtN*I5a<>4D2t)m`7?The2UYX81AXp$-$6qIj?+!E z=h)cF&}6PxegcuLfAQrxB>v_}_~Lyr&8EV*Ph$l9G0Eeu%fu+cQl z=MQ`jmp%*SF({o<0Da2-#x0qP$@1?JY0&0t{VXWK72|l!vg8);{MGpzh05h|12_Q`o6E8kAujyp-cm;L@@TevxN7O3xRaxS$ zL$)Y}x+|ZF8iipf2+(+==!q|&&;P46MtY|B)|`=xDjC6yL53{tT*W&4S^3PU`S2c+d5C4o@tRw=7rV_kJx zT@G(;=+3P{7-$2-tnC-59)EYcDB9>Uc#&{>eh^xngyl}H8az68hd}E{C`)UHXu>oL z9HJAllzyP(dvwtXN4g#WJ#?E0hCJN<<-r>6*$B$G9K8TGfl#)|OE0TqqidiS7Z+va zJQ{HtKOG*tbX)OjJlTqWlaFnNP zwq5W2VRbK%UjX#Kf>zfchm*tWXdGLRlL!3};+&3`Zr0@TDl~G2QmfQ}JZp++&iAUc z_O2^Ya0j%VeMph6O}WVsxIOd__*@aQ$&3~11UfYh4P0kliYhS_3&)}Jj>c~R;1jIm zzKfr9C5mRdXXntlMhvkU<_lZv@O@G20k8c`IIT5x#10VBoLi%Viwn_$J=28Fr%ws< zbK)mUKH2R{pbySC5gNW>@j0$Ic9s3z-vM;O5k^V4~1B z)waY$jvrRlvn2akI~tbk=oEM(J_eDVK=S`MAVARw;&=k^88rnZWsUa)$w%(l{GXi= z242Ki?9LAi73c)Yfpd?y>B2s(GjNwD_do6Jk-x~9)CBt`hmN1dOybym^J&o3&(kK2FU<0BTK69)~f1SQFiOY4)>Y# zR1UY6E+&`9UODAaPu6yJ{W7FJ!CMtfc>GIZticq~P`bnMwd`Ejw5X&5SR<$pLzF<) z)!OS;=zkMEydo$dLB5D~AUW%^nOTs)+%E}N1#ET1L}Tj`QOhWV4!)f^;lVHY$9A(h z)m%*y#-baa`>XDjSn7A&$a}sL1XusRdl;@@AJ}37Tk|EPj4);qGYQwPh4?GeCDWVB zy6HIx$>x5GN8t;&3dMx~VAw^9FoH$jKJs4|c@>Dpy&PS@@3ELJmZMW14^?ObpIDn? z^^Lf-VCN7N_Je%AzM$p(m>scnXCcVtiXCY1EuZYCwOTOT_saRg5~TqzXt8uKCG5{qgAI^#83-ygy?Utascd>B7*|T6GKW_=!NF)3S5wapzq>Q# z6?ih~9pD*8yEn~TXl=zHAhH2m%ZCcfH?;Kh!jpIGutFE$o%f#cwA3p$K>t$-{NYzr zApnLd1=Nv}oQ!li1NXgs3FysSRML?7Iv-fBoTE?y!0L>K7XzoIwj8)BvmBpW6;dJ| zXH4o$rFdO@5pH31u5P+Wq(>zDQxZMocIA$k3SF^eu6BRkRXBjc_EEv_FXB=Z!luNz zOp{WSL~jTjDgx%;@-Ep`R#Y36e1#`U(Spk?*9he#|j%;2LrdO ztmiTC`U&q~k&T)P#{Sot>g*x;JfksVoy;Hnl6r%$mZbFtUc8I^M!xY0f`m?fbfI}l zY;$UXhBh|Gj3i0_jinSL{MDfJumtdb8;mE?rhjj;%#Z51#sTJ)MT%id#93;7cMJae z^C^8*Bb4iMTk{%Senfz%gk)jeBNP3!(WOBqs}r|8C7JnYu&@-*AG`8a)td8Qwmd;W zuH`MW6BG6-QFVQhVfIMtwy-1aOe*yXylP9KziM()b(>|d|6_Q%&HOKN$5DTPJ6CTu zR_LRxp#6X(tFn$zsTGo$X`b2GJu(s-t$Z?sv10V7qr*Gdosk{PETWjhPiSxd+AI9) zmoMFrhyow6KQbZrq_6eh|Frx6IDr`d3m8OH)WJk*g<{n{3O{YM5Qeso&yR})eXynw zX=rSb#+^NN^hAa7BkLK?K$VY)Ei$x z1y5c8P6b=?6grd9P*w`gQ(4WVXHG(j2G1C@1|I$|XT$5g6f2l|PI22FY6P95!XdCC zZz|6d1rx?(lu;ro%fVIQaK(glyhd1}UyXNpnfTkPkHy76T~O90W=E-%O|zRcR`Q zc165~@$ap(_jYr{zp7x<$b`xAfk%&e{u`eCubQPBDjG@}qlZ&h=#HdZQY{4{_6}>9 zd{P{vt8h4fe0&_4goGqKJeq_eaPiX}*`=*83sxDM-Ja?1zn`I=F-w6s8fjNw5|FK( z!}bqGYSePhLQt4+?PI)9%mT)7u z3@w-j>6o#{Yb@;rn3N}aqCZT8&3}S^?>c!M&hTRq(Myz_n#;+WjW$L&jytddlXyJk z?m&Ohjk2-D70P-Jkyo3VWGMtjZIzi?~(&a<$7UkHZCFa`4A z7g{w0Ms)OjW;5Du*IiX8Ky~k(Em^M)IUkf}7yJximL)0_!py+R8YY&$9;Zs+m3eS3 z#V6lYF0Q7Q@R5>?3~WO1dJL|?6a9aPuYf6ZG>V1mi7tO_L(dGsp2P$VhtTix&Agw; zSQwxj(Gk2*P!j>^2}uYXjl7sIi)UbiDT9>LSNo&_8Pr2zw-)aVNAKL&*lr;6%EZd1 zmskPEVAgvkdyQFZuzxeic^GXO;GpeXU^jLDRl6)vf1z%B2p8cUDF_Pdy-`0O==)V? z5s(i>!7-=kr;5-{-2ldrbOt^RXxK``y$M6I5U=KDnOw7f?WiDxG`ejyt&ZMj{OPV| z5vUC7%QhJjP}GD-nB1Jj(9$D(AHUfLDd2kIArDlFjFnWZ1dk1PoY#L<_11JBJR2Q)s& z^5YTB&7aL(wxS;RD1ymDq!phClwv+7mttzSla_uz2?;4gM8u(rFT*hx#Cku8c-v6? z{QUX$r$N8uq)2ZsRH9%W?TIR+kRz;>VFh8aB%1+ z|6vBbD=3AhXGq1`CQ$2g>K0HbiH&b|dDi7A7G{)Ci`oi)?+A9m@XK=m{3EDbV$gW4 zRHj}7?CdHVo>YOB^UhV&UX^TR9I?@*%9FwY-*C=4epW36GG#x(J0GV%)!-tX4_-y+ z=>{7*zh{sT_1b;Sla`&snQm-TOj3Hh7Zh^evo1nVV8T(d7Zx@lt$gSBBS}#_pV4BF zmGYeJ3B7IZKY<<9w@;=okGGU_n<2pYi)M!KSrUFeFPNVKRJmtBaMx06@0Vap`v$`x zGF-nt>4A+VXHrnt*_%0%@2xS2GRV08++ig8w^J?h7(a{)VKn4HDr-<-C zy5WpBQ^T(?qP-HQN7jT$G`2NQryOzd3izaQT+vykbp{C@dkv53l}pk_7~63g%>W93FoDFCRVoQrBO@c7 zPinm3jLC*EFgQ0X{af?cWS0y;r)AF%^zQbL^cyfAKYlbG<ZpT4!;b>I5R(A zz~7PNl*B3etJzAb@bpi?&jMIwt>3cD0#Wsm&Jtj)pFDdO7xA~mBL9gNj43eCa*f}0 z_`CMycxvtVOo5BKOX=+oT(iGXyWkAqzP1Pvcs!Sqt_x!=}h&{%zeEHt)qg~Pk$uY-W^PJ&Uq;{Bo&@Tz))}Ig?{gaNt%ki|ZzRdypBP)~i3c zo)8&qAuhdq(!qFK4V1R6KrOkW~#zLD?PjB3F=%Amp zBaE)D?!~mxYv5d=xJP``+*TRf(2(=bILR8V$Ku(!f=DN6+$PR#2VH&(7?}KDjX2~5 z1s9i>_?K{8JNU~A6A0utoqhj}C{qC*o;!8oH3q=Z^#3To5g-9^k@utE{^ zh|NerlX~y)yraLQf)rtU5iuG87Lp}yTEK3Ot3jEPU5|N{j$)p$k1^lGeCH1In`5FVBF8wd9)$HyTAcoHIEROrRpu*QAT*rM6Q&tnd3|b8DpTFx0YJ z+8UOGw&nUqhaEkuQCd|7kNM`RkrUBjq$^ibzu&!Y+X_u%Mr*)FDph1n@Yk>)I61nvlZ$f z%A+}F)ow+OijUm#+6J1Y?gd)1iaM!uO10p@YH@oM!bbrs8ELTh&j>H08&LhK=_G;& zRS;I@lmN~ejIV$5-aI!t`L49u-&LnJbuR&Ilnme zNaNH-UtXv03H?SN#^yQ7!jHr3*|vk?ipZ4BrPCc$5xP4HnT{(ODdh zaB#=d0Q4a5=7u8XN{%v4WZX(s3@}Q@WzyfRzTyhfDCQblMXJ876*rD1{mv8}e2x|O zI$@4lKkEX@Ycy)I7y48YCv0)qS1=|7grXX^W1_LLY&kPag&TdyOWmo~?=M)}lD7(YwLLP}z=QBq2aYAeCDy!gJiLUrq!&?7&O&<*-mmT>9Y z4G)PINyvDIQ3Ym93DLKd7?Z$$K+fH9? zb1$fDBWhabf_7uk-N7{ku-ePLlEw+geFXdp5`>Z83)hmK&Bk7Ze@>9VUCx21S(HUS zy*SkMZMB%3C$aFV8;Mtlh_^5DWXt0iK=!Dmq=l)DoZm+My;fDz3cWh|axye314su^ zobWML^e?jxWLF4DB0o_D^Q8}~DUg^NCGjk@j4{yzrK?b0QKe)@&SOkH3dp}t22`-dplRJHbN7UTJwnGQ23&PZn0zG;!-leP{jjZVzg*U%Mi2e4=6SW5)~yvKU0FJB0F zn56%x7F|Z5wIxoR21*x+yUjV4eRK_v$e{Y<6ge3xCT6b=3`Fs1%B z=kDU`dVwOsT%-znGNNDi!imFlf&Z2xWHH1tT~Mxs$}HuC=Gom0?Dzsz-Pp>ezZL&a zL1CA{=aiI%ENw78JK@n)Y`;t)%zK$&vy3iM?kSg=Jk~jwM@rBd3IaRIN=i;lMWFbh zu{?YAGYucKQG#QBiyxTp?Lm?DI3{8F=7m5_Xit$BE7Y1E)1P6U!lari9$9$ z#DG6O8+&`wIHG3z9e2Ei~*9k;Lx{X$2Ss5c~R+Lxy~|_xE{f$Q{oIv4Q25j;j$yS%#mfOLqBx0nYk0 zG9s51yUSwbjc$x?MssQ2FG-S3z2{<#V?uTlZ4f(}ky9&JX?S8pDt;wlYe|o?NRYHf zEI|hYGfo2c_eN3VcKnpzAA=w<=~#)#kQNAwdV9TDzU=eL(9%2}l58IVVU-sWgR{N# zwr?L&S&f9v2{j%lkmLdW*74bc|L(*SdsedrYfe=PLI;VRgQJ9HU|>Lf>+*D=>71F# z=n`-G4@!OBJ;X=3%A1EjSF^v^CA4vZjSmT{TNwOUE%Agd=7(_ila}=L> z{N*R31CelP{zVJr0oFwD&R2KtfJJ+G)(+bpa77~z}BaXd^-#rK?Ph^iNQ_o9b%ULo0zFYTpc}c#-*9c>bmc^ zu?kMJ$=T(BK?Bg&7<*>Z#RWWP@R5$TuJ4kxs^1|D!#aipeMkg-@M^papDehq{dG!`1mu7&V6Ae;Im-w|I8wd+c6dz z?FcgkhR9G<%3fV$oVhC8=URq`d zboCNs5}2RqiM9-|&K!!jOl%-k^aSdYJh(@f^+sS*KhFDqvEx6nTCtCeV?EQagLxkY zT00vKy#HD}=l4#-xzya0$Leycf2hzr+cHWy9}sk~1jJrk-a#Mv=XaCo=;(j@dvJOn z@tJ*>%rQ4c;Jn@0G-!M6512i%q32)CdhA=1+9xC?qYm)jjkvuO2RFJ@xH)}2V>t`2 z{Y`9mNIEs9&`D+xW~tpJTiRoxPsNQEXAeiU6bmVRtTi>+uc4Wla5Q~YP3s<*p;z33 zoeq>MUf4T)D*?+wW?>P=BE@+zIWq&h-RY%keRS|?|8v|Uq2r%Vagk*QMl?qwO^s-oAx_EWe5OQ%E?08ppMVInuOo~9RJb{V+7WY$nuN8jW%Cq#alXYs!#&-g$3;% z>plE-K9`W~KI2~%8-mCe6gT^vNqyZQ#Cbh;cUg+fNd^o0C@+cxz1=G_A}=4?*iM_|IIt7(ULO1dZf9$|A!?`>qnU#Cx2^TS zU^tq>Ix zm!!0Wgml+?aj(7hde(lwiH9CKbH)EWe|cKmoUumrsY{ZU&32v%5LSe>jbkPMaZ?qJ zm+xoYB$+py`{GzhcfFOikB4&g(i4JAH~J$eR)g(Rq@ITq$ge_7o3xd*6z=U#-I7Ot zOIyP0c3(xF5QJc;5#BQ z`ktTHRI^$0Z#9gJg3fEqZdE_x#2dxG?!DV#0k&BzgmZ?ezAHM3jyWwc7sf)WdNiD`YO8n!0}j z?<3jVEN=XPRm1hAk$ju)j2PqZPIz(3dPIm{$36r7z{5%Bq(6M6JC>&q@ELiM->lPr zV(=5bm4WF)bpv*bA=U|r?r;T9=~fA2u6@-)0q4KDCNPVk;O)@p1lR|ehfQQ_X1szxrGH9VpytV0txHy%TZ{SlH3m)=!hgbPQrG53Z>UN0&5ot)Xdr_>!*kJ^+^J5%FKXc1lR*6a z`!{%Hz7kW$#04W{O6{tivRE4e4)_*}JpxpK7oBuvH8%GE&Bf80VGvQI_w9k6YdkWFL?hD%TdDBfHJ1`6u^;BO87g6H)GwcPOMT42$dFdd!X z77kM4q6Dd_+oi#$+ zXyAXNh3E-ABSPR)UR~|S$H)7WS^zL-JAC+~p(e&esL|wQFl>W8Ym4zodpR=Mif{dG zc@GuV@DrfVtC~VMR+Z6#Ji)+#it&Kj#UF2rb+BoR+XOwi(S zb+zZ=oHcG2+?P!+8ppQ?k7u#|ecfN+rGM4lGDkQT$Hc@y3fBuD7B2OVzX0y*_HtI~ z?5CEIY%@->T;U}#r85J7qqlo*h0}ti{mB#x0XqnTr8Du3gmyd+=K^Ip1)}bkbf!sb zoCLo%ejk#Sk=E(2t8eQa@WzhEa~ANYQs>{+ax=Pq;M{b5ihs_Y6?P>Q>LH<#*A(5x zhv#`ZAtz_*METwAP*O#DH9wEb&8EzcT+Iw<7-|tAr(Y`D&XE#EN-2;dM91KeoNi4g zDjDk)=hD0;d-qkYbFTG4A3L6nny!pqpl@UoE9;iJT4K`W%ua5hXfNLV`xZF`XY)rPoXZ#Q3VHW9`qn z#{VsX@A9NkHc)(fLK~B#q!p`!Y!Oq+_=(gVA>$}&c}D!4_yO$(D7XFIuq~$m24Ukf z`7vjle9@_X8u-r6UM(|fT^tu`P*tr?$^DBI7m38nFHJ?ePjerP?164H>`BQ_xY)r1 zFV##d3-dM-oZpm#%c3t6guBCA(EYcg$il3?eq%HU#B_cQ`Xa*TU8jpg#wx+{(}^PA zRd1wyem=jCzI5fo!#6Ap1c&(Bn*>k~MS5b2wwKo(we2k<@U~h=6a|#&?Umm;4_Nvy zATqJQ(S-)s3C;TMK_pQa!@9$NByo(pIIL%==2Wt!Fr}nF8~Pk`Jpy{2gdFazj{{5M zz~Z%p$kD%x6yn`q&ZzQYkLvBa!gg3u@)q}o+Ojmxi4H#0;@#t3Yo2iR+lm-&h&FNP zVm;*cS00oT@j5d>OWBF5~I?{ZB8Ay)%>%XvCzj>gm`BXv|0g+2t6Q>1il|GYH zX^-nJONfYRoYZk?Ug+P9E&dAKthsb}ymvYAns{idIy+a>RJmj7tMsf9xuk4t70@j2 z1E>p)c_Mk_kSQzzg7Cz5VU8>!1dX@BJ=2+jw{Y}wm`vFg$)qew;u$3)dl_KtIxa#|AT~4O>F~=f4S*5VFztiiyg|f`+>`o|*MVGlc-|?NA zweMnXR0vF^_?M0bKpg{GurZq3YzST0yh7oZ?q`C>jK7wstSx@^LzIpo={s7gr8p;a z0!}wkuE4-PRuzGJRr(RSr<5P5nG!U zOMwy{t(-h7I*rQpJ$Ii~-j`@GBu#$aEpOW3oTcwrG31*c!lYy;z2(_1nL0{WQTDi) zoH4OQlo6ZRm-J;qW%U$zS~BFtl@)Xy;OGdPrrD6`xIq`WeY5tkZ%TFYq>0s)8AOY6 zSYh2{V}NV2Gc0t!|FOLtctECu$%%Qfo!CsS-+Nd*sy zt!8}$OhuG`?us0d!0x`v-r^dkYgIH9vvKXtU$E~vbBZbmG`JM^mGQerS9BvDv9?lb z_Wumuhl3|YUIg{x8K6*4nPx;D541Xaw)!K_XhJ5?`6xS%_;|FRZDRlunMOc8!x)x$J!`h^x1e-kTKQ zM-7WK&iHp003$gQ=Ti|9*d(a3H2e|NBw0|4A?~5AjW5}^2y|tg-EM2qHbhw6NRtAY zU_bf`NV*GeZc1}FOb1Yi`CVn?T*zTKDdK!T1~s?f;daeEp_F<(Gy)~NvnfjA6;{$3 zZd^7OM40W{ApKo6tE(5{WPbe}D@Hr%>3EtU{wpwl2kfv{xR6k#qSXg^+>QOk0?blZ zrNP_&Dny9c5BM9Bf#JC?srwvROxSQY>mv9!LYHMYstKAMGDO@K6}%%bl4gl?2?z~7 zt(V@wMTQuq`E-a?C1a94?XVU;Vh4tyZ_nBrq05uUYeYnZ^XF&4!V*(H$utT~@kRIR0@C0$ijD4FB-8Ezm}Q^OUT}_o=1Y z!Ey$I&I}|g*Z-to*@RXE$x2#T?L%(5iP)eNY1mS4e=e1D-7SJ#>v+Hx!33EkQPWFa z)JSU5#i;eDIr?YisuP|>866Yo;xKyrlVm4;?wGL6){xj(+OQx!PJ$h`!;B&9%a`Y8 z5Be7+I9~^I9+l_HYj>PEp1kG`Qem8N{!D_)SsN?G>2t+(X4Ta7!rBW*3zm zpzKD!GS7IF3?b8j#46u)CXrzLW%CDGU#mZ@@(*w6z*Iua5A?O)sO{AjL)qhrv=p;^ z6E)UD6Wry?Kd)jxAX=HypM`eumnaFd;BK*EVi$F3h7Q&Au1dFfL$b>F)|Vt2Y0iwZ zyKEd6CiT znpzcZ5e2*9>I#>D6-&u!XE^B>89Ae@N*o)9LG&lD#S*sj-%_cZA{a^1`IZ*lMP2>s z%j544)+A01X8{!oWI!S_Q@|5va9eUH*7u_DiIcWk*q?>$ss52JN)xQ+9+ePns704l zrW+GmuzfDPkq0FNN|%F!&(6L@Mcq(wnn*U4(;6v?UL?7OvvKI`C%P=#`URf?)5-N%j>@;VXR4K+ZO7vl_}eid70gLyb~O%UcFH4Uss) zz7DrQ>J9c_EZJb;sFboRBgTe^o20a&ynuv>3dcb7MG`3gL*1F;_e`nytUREx>XO(X zD%x(PLDBo=(CCE_MCe)t8dwZTes$SU5WCH_SG_6vYpFcGQu&BgDnlO_l230Th!YdJ zk}#dB@^<0RZXAI=;#-d9ahU$FgALJ9h1E&~)4aoG0mfXl%2t5yzq1N7cUeMxXC*=yZPwhjUyCn^RdcR2QibqR8i ziu2)RiYg~Rv6J>Q$NgAHEbeE@&*Y0*^=DeqE;fgN3@`ey-%+v(IKZhJVlPlC(Nm2T1jINu5Wtvoc`=vTf66< z&XO`p{ceAJelVb`w4yb(fnq8r{_v~ru?7RaPB6jIQ|m9!rDm~@E&{X`s>ph_ud051 zYOt9ZjEjr>+#iDcI`QA(STSf?3z1sb5*5EER$O@B`6$hSm+Tm2VP?6eCEVThVl>qS%F`5PhVPNJv)$eQX0$MD0)6u%cAGaA}WGhP0? z0a&!AKNYFk1!{e%7X`956i6tGn1;D$Fnx=GO@rgZUZ{JM$5cPM99zD%^EU|M2cUTo zv1GoWfi$ug!xk5`S$wXKy2G()3uursB+Q3LN552w;+XTs9{_ocgj^*X2bEz72*9IB zBqMr9opcwAQa~mz{+l&ulk^~JtO-{=+aY+H3Auy|Zr=OU&|#t2eB3NjJ7Wr3FO7rL z|I(@DdrJaOnG&^MzB|mw=;#^4YOtmycdyDiwhJwD$hTR~{aM&bjGFgy2~JE-tO*~@ zHyx7wdn=6|xlo6eMQ=AEmgGTEH%BEGym41t*8~31e_2RIisw+%FmsfqQP<)>-csxI z)3ct=p<(c7Ga{hS1ff#<(L)(92K^aw-IVr+_bAg1PiE9&_)&6oAqq% zzBM6N+v=RISE1Vs^4GE$sN|FrD~jPtUhi%vo}zb=xbcg=};8{o~&%{}+mmT(lO zJuq~V5sFxCi!;>?*tk66oR{fn!PU21-j?c;i$a&Dy`G2l3sqT6)EA|tj<(u##QQ#~ zIuSNDRkYlsvX^#S$CNubSYLJkCRUpU0|P_US{XEi%LZ4?4v)d-_OmM8=k8p1{|~qF zr?MgTcz0TAYFIM=N5H+_nCM%A(Ih!h0r1xHd>xXdsqkcFshIkK^-b%m5?LkUd!h8E0S zc)mQ{oB<91xhp{M1^n9d`SA{;;kA(6Du=4VWIheRq;Xmp;Xc8iHHT-dUGXqRGx)%aCq6EXlX8u9yq)jh!Hxh1Zs8GEboH7Qz#NQ4O$QQFt~9hQ78 z=I)jN6XubbJ8_2`Kb(PxkDd1xJ7m4ATJv?|oYpC(KaqjfNiB%$5_v>k?v)l6mzRe^ z)H9}XtXKQ3RXW$+)g_8gNGd7*D7LurKzV=Zo|?fGcQsa!&EL)kInjxCP5DPR^glKt z;C|Rw6xBCn*f;F#kPVDHOLmoaK|x>V@c?5UDKj)2GfVY(QrV zuCjY66e;694gb5WGFdBeml$Hw-Dl5(lH&VEl^@J7pPD!W^imCxlxDm9wzM1sTX2SJ zEbc$3lhR3Os**0b8XRo%9dB+zodGJlxvT&NOs2&VgjiLaZAagEK>TXKHM-+0cgwG~ z8votyr5?0+w``aC%z7>lG5Tar<7x314*$;qr3URgt{tkNWHU-vfZxzPv*p!Q@e-xgObQrux_0N%os-cKYk8^~9_kWBFW~QW~N}q|U=6SGF)76d6+mf?; zIDg5NFqBm>x#wIt`XTI=oBMx)?*IBeRxwGjQl7B1`Wp9^&fRCdaF zEy+hlPERdCk?8q&`gj>1%#{Ew3MP7&L>_XtG7EtLL(HCs<;VbD_276thQ_4(IF#cf zg#CmXUFe#yJtC*Amk7UwPX~l(&J_4h^jhepC4a`iV{LmEvT_mcPKAfBtP;f{BdEB#I>)f}~Q^sv|fea3p? z7k+tkvBv!p8#|eLr=t4P0w6X0kOJ30_)|ilGkGt;9pR)Ct}`9_AVhXLc6ODP{X=HI zG(vn?ypSy)@wpN@hGWX0_OzzrSdy)h%v<4jBU6n>)oghu3SwN`fcYSE@9gPLsIu&}xsX=jJv z^oHYW8MaUPR?5GQ@&ECx=_&v@j+LF)tVy`~l9F)*xF7CJbb|$+Pbc#kq}VnuYfoh0 z2v?%nozC_5M{pKBm?D9}YV*Ksg0J(_siiEhAd^9hx;Lcz>g!ir`dEH}gI^oi-C?a2 zFp|P29F`O0eAIA-)M<|HXIjlKd3owdID=~0_8IIu{4_LNFWY1<`?I@%;VhxbgAo92 zgoc0a=tfGaEG^y1RY;xbF49FfVcIh=1z686k#76G;|?`(m$e)4noH(1`SqD>&VLqC zgMD?{J{SlS&qzWiV1VxB#1muyeahyA?J|u5z3N(@k&+0%{mG?i#S4A8WMo9ndO+O8 zl!I_o*U9z8i8HxTVgUyqkC;4!ruN-J4uV%2uJl&3_|ZQ!_&s^o&_4c1 zY8;&uNq|G%mv;&PvFzKH*2-`q7tq3kzjn~)w<(vD(#dk0DI(-(1 zIM3by^fX_%mBrKi1@ymn`+u(4F=dY2BNmRJKJ(cK@&&M&Z!POXjAWON9j_v&ECcc4 zl9NM}UQQ&Aetko5!6x|q9#0`cKj4+2^)?$G+!?^c(6j+G*gSs&K2y=K$-$ruw@5}2 zBz^T&U{Nt&PB@9K-A1=2os{<*VgGL#? z&N-<^tS4s=kHJhn4r9AHP&jy`!|>trPSCS$4JS@R$Yk&6D6W)LwJUq3pS?!J0ijZO z)p}8ygUfyS=ykefK&)G#*x46&^erG3G>)ctZ)ZoP zgBxiS&`NSf1*C$BiGP2Y|0fK^=l8~sRN~M9w7jtKN{(JaoKDHkg#Ci+Fa?x`g z13}(LDYOa-@3otq3+wz&*4ic9#R}3fuU(6B>G+;YB6cU3RjWzy_cxPv4 zWnFDFgJ!7DSzZOI1W8C$Lo_@A+yhe@^AkxrgB-eb^r5n4f_Fg5duUj;zw|N(CloA*3 zj@8{=EBo8|&#L*K`YM^E#uSs0-*1LHjTTdrzCNc|H2X%o9r`PAY}`-8wp_O_%4%uR z-WB=3)x(;W{H=2W)B1!q^xs#fE~ra+D+Mb^6-I5DqG->J!ekkV{`H(40UdSpnORc| zyjF)3-xb2RWMzkRQD>%J|6Tf0G=k*X*r1Y-!?Ld&oD~y}s3>>y6$G9Wy5HvC+*-9l zfTDo2U#tqGS$rYEV|{!h(m8B)<-l}RQC?1v!tyt$eKnc?U~cC%4p-#)mU3oxmU^%b z$k)v#j>scZ61u{9epr9mnk!Mmzz%!O%uim$Ztj;fPyr$qLlLoQzLyn$M1F{I{H`|I zTuvza`e+4mdT@}LWzz+OCu0gy>KiW_M?IeWP)Yd(C*Vsfcy^oU`wZd23jOLPr;X3s z%$sLu7`<#E7qRjKM{CdYh7YSg#y`N+x;1`1Atqr!ABX36uHrNb?IGjawyg_}et+D* z>D7qG{qH~D^`|5#xm6mR$-my(nE_34Sv6o>8+%tE8KF}7r@n|)Y;ru6?4oEF=>YU_@kA5I1tE8BvOj*f)gBsaS48&~1twG?0& zUdy(DcjO@vfJ|PESv$6dbna&oxEPmLN$(DShMBky4Jn~cKc^p5%c9twNR>8{tmR4# zxOTS(tglnF@O{{_@Kgx(iW;}KLm9U>*`1q%F0IP?I*{dgn?BqRraq4kw};b%$_pyt z7_kuHDQNv5%uV*UZc@y=Ugu`t+#W*_t&jr&R}WwZu#)%ixu?%dBlX+}#0&>}H&`rc zB}1TVxn&T0M)W;2h3uP=%cZ3+CXnX?dV>OPaE5iyg&#OV$oSN5^PKR1pTg%fg zJ8v#5gmQ9vFt#jSq0B2O8$$jc^Uf<;&;?T$B$w9m|Mp7Tae3If>U88nSg) zD1Um&JPXz~jPPCei`*WJeK|5cM*}uK5P6;)2{B51x-*KK+s^kAdNJK)L|%S;S{4ShWPH)88xI{oVS_! zoUyNFTJVlV&}c}p6Ww3bMq4g1_uZG9eP4xPq{)zM@QcR&wtGN<1;mo_4;wPb*0ju60_U6RK99w z4lOmfX|wR>jyuSy#bvC&Fb)MV6Kx>@Gd$2QJZn&MIbd@;uZ7>o?7ap(f>M>!8Kbk@}n=!ekf4pzf9I^UNyuf42GgG`2{35G_s8%s@00NJ;(dgxD? zIxigEG}#YKl2}r9W!O00{L=cj2_EoT6E55S{aR8&PkOuacFN_TOYh-1IH(6&!8!Fy zSn(q%Hi>0-?+lqc4GY^sZ4XQfT9}y$LZx<;T`YYYF)R^=<^h8u-vP2krzA7(bBV~viYtcni zN#e{ydO#{aVcVxKNackw-jo(D}{sfD9-;F@b5EAEDg3G~FVQUI;sO%VrwB{~Uz@ZVB~vTQ+u zU9${s?K2v^5jua7x_E>EG=VSiTW$w3V1P_IJ+G0_(ZPx)E9w~p1m3&jL@owPfHG8a zLKyQWmwGiwthJy(vXqR7=*Nt(`*=DRuhUjAl?}4k<6^aG5;pVO^78WA^HJ8#LxDEB zwcO?Rn)9`9xHTtW8uQ7Z&Zs#jnDNKGoPTH2e@LB_1SoK!r-L)-aS#R}VMG_Xe=0#n z?=0xewfA2c){l}-1}x5pGjZB}q#E|-Y-}@wzq-tu7$0R_r|Ylgu}`*JFTQOjQaWt* zetGG+-j7xOsT|nAEw?oif@p;pxVEJgR*MI%StjA^OB4xy8AZI_J*!lC>dXW$L>K{) zdUH)m;SgwS?vS1eHom)Hvv9gH%n+%(QWK!pVP`a@XGB%3cD+MILkerUth{9y%jDhO zmbxqiuyOtG(;)hQ0QvmOW7Q20mq{0vgmT?D5-k46?^X>+m}NZ={*0gA+$;!t@-z*C zZnQq=5-PmQLbzzM2^ZGp_(YLW_jiB>pg&PHqZinr>H+8nxx%1T(^;vS0{&gRgEDWa zl{_Z4>6f4&I;kGwKMz`JF3bnkqli0xrDRY$Davb|E>pjjf~h0F&WL7b-Ma17 zvAv|Mi?yaC9>0pNMpWo%A+J>KQ?11~p^8A$du(MMXyE(WJmeAofamhwK#1z(UKbEz zeFH{}qE7}7I>#Ez6cJgOBA(ECAN?b*FfUG}j*RIjJ2_Q=QTg3XuL zw?R(e+5+QU0>3v1ST->6zC{5Lv!{1RxPdG+<+k9#AwX9r5cwA~8M$#(b*7uWzxxmW8S5C z)`#v9XlrgDfnwQ2AjJb6e`HCa1Tynbcg9bhkb;}53+vjd>(L0P=^eQ*Lz^p5KQjeY zQyi)6k6y-W;>IF-ORLaKb{CUaphvEJtR6UfH{_oU&gbJ%mnGd6}P8wQpG_l%Lr=V!<`J z)70+H`n)eBePvBR5P>L)#MgjR8xEFZYE10;O6z;=xwAt4d{2iwC83~j=lkoUHV!{eyL3=Dpwq@=W; zwx5^7kOAlXF3zKY@(1U;tm7<%D#R~|Y4B41OFe6I1OxXLYypUQ&?y_k)Sbn_<-5F1 zdk~zTQ?+V--7uWNU0`6Pi3DPH^_#^Gk1=>Y0ZmLg@uC4FUn@Ja5>{-ok8{s68NVqi zC=A8yehdz6lJ7}&z5`EU(MCRq1_lHKbaU^y{nw`k z01EmbRI{+5t%dO?N64e-dqN||@*j!LFsQ{`$*9r$AU$3^y_HS%J^t6f2QY^LlwAXJ ze>Mhy+lxF8T@oyYcc*g0f`*1hSIa1&s3RohOP%^Mr$tPtc0F4 z@(zl;c(EZ-rX9(K@kykQg-8T?>@)WI@grIy{qT=z3|T`TvJ`-wtX2T?0-ybf7!i&> zw;^?j+72@-GqWJL(gZiZa3}e@Dm9~`5AM3p7JCBdJgl(bk$|1uXX9uqHaPf;qsMH@ zFJW2-$D~euH>qRw%UY6|w;vU>t=INS}qH9`Acy$cgyRh*-Z*FeN$jON- z8gNPZ5#;BVqgh6vNZyPRB}BAvHrnEE*_mC=R@zz=MRE1PzIjWdOOzQVE{ zKU$TzRrXunM)RzR%IffCdmg>E+-;BZtoNmj3A*z7P<(#+Fhu9=%)6tDsURtW=%kYY z7@61BYGB9d`}Z|nx5pRO!I%pD@Iax&J3#pY?R=5Gx%o*7da<0r<_z@zeoFw=mM-dm zc{G%}%Enb|d={ykoRYH5TKR|>(ct}aL!Id-JGt$J4_OzGORHW+vXlo zMKjy6&N+0*CtsdBJg?59L9Don$Y8t+UF4`|)ngqM(;Cn%eSZs8cl0L4Mg&jFsyK#)9ms3l}Q81#{_IyDXhBWHw#(CYnH(Y9mPyF)>W=cB} z0{5eM2a(g7tuPOZ|Kmhzs~O_gsW1N!8lohCUAHL4GJ0j~b^As55gQ&ZX5tDb%zMdN z*1F@ztIdCNsS5$0Pq-7aE=Sj;Ytx@BlxKVNu8w^}c^5UMz83`DNE$U{#egCYh-y z7;fQ3$%JapK6vG>Q1uN;0O-v!<5%s|vi0FeZ%L<)oTn?D3``DTjxvKA`Z%0u zhXcgcrE+8A*TPi1>!kVAWE63oFug9l6)$vqJ3C2*8TD5?lsSmLzw*rDKH-)NxQEzR zUJh5;jD4DR${Wci%eqV^A;Mb=rBjT=+AwO0S^ca(^*#Snll0G;q-+W8ZI<8Zyd%e? z-=s`PNzcOa0a?mFJiJd>SQr5XCHQ;qKVN6g1kA0TLe?Z{V-}(n8x&?cwFM;!rP2YJ zHMgp)4j~ewnu@x1)Yy^h9M{jDr9SM)tiPefI^;}O_S5vm1S)2*mR4XGN)FRGo`t;n zLev&-)F%DUSOvbu){BJnmZ=anQz(TmBBC~H^}rrB&3r@ODAP^^RNWkJEdY7yF<!i%1y%kfb)A$>fy>pgVT+g`XX<^TxM8q2V&G^bZ)Uky0FyY=!!tj zzG~M@Qd^yXnU}Rx~}GZo3}tn@lQ;|Mjd2;u=b;GWtC)$y&H|!o~4os zVm4^One`RQQFH7^QxS`gZAXqYs?6sr)CS5J(W4KdizaO;f|$?JsSwQ_I-cRq&(H76 z_IUnlO8BoI4kQ9#*{=AseyP&r)WZZDc>+gDM6GO{4ZCP+FQ?$UE3GCSAIeU8l2#Yt z53*{cm#(Pb97_b(w=G-glZAy(Sbzn|DN0a_x~24iv>oOtQWNZW%at@ z{dVSaKUt1GXrsAZjFt$n$G;q&JZ;!`aH>9qaVAnd)j9a#$axkJnZ{!_C8c3oLKWjS z{CjFT*;w|}dof)A=$yv~DUZ#}sJ4`(V zh%Zcv+;qX+3eZ;rdpW|21m+!eAFGIVet8iT(pGY6DuS=HeCAkq?{fIXhQb|`Pm1j^ zhSBpZnX$bhX5s%b`Tu^u{^#S(g}y-s+qjfAeDC%VkwF5O9*j8&GH_tz8k9!oY54N{sg;HRS2&78-op#os>=3%g}#X(7KZWHw)Euy1nQ>~Z_^$~*8m4q6a8<%yN>81IS&t>0W*~@AN;2HvGrWNCPmRLj?EtePSMqHcis%| zjqIH9@1Gc@i}T)kvjv3YzKa?|iPuU&i&rYEt|u*?MeChX3mqC7B3eTiCdbBy1Yl?r zW~5JjY(9))E+fyMJd6twexUd1l=?UZTbHqt)8727vLkkW3=y{%2Y;3#2BW^e3l0Wdy2JnhFH^7dM^H2SpeUTOVlSk}2Fo5LgJ^x&A z>_z$CW6OU4C7=)q$c{vhXhR47#D`>2Q&Y+o0SX=bUG+i=E)l3_ZeuFHKRNQ#;RDxj zpEJ&&-rHiIIo^_iEsvtOy^zy?Us>7MlOs#oWLH3cBlRI#oo7xJ@Mj?=A(5^hZ*!qx z!-Dq4IlemD9lu?$t=u%RAy%b}V7FQ#q)%7b@$(Khx>{{vi%5kVO@F^{Nozd${1yn{ z=#FIVK%lT_YH2C`6;i}nR#8zT+T^fdy9*f!ec#uJvze9jR^|tDaB~IwSAh_UpmT)3 z6%o=3<|?rWUzIaM$Z968Ax6*7&*?FxN-v)}pWOQfsPpn?-w^%1|Oouz4AY&?7tr-e6H7ll~8x>%n=d6X?v8lGkC~<_SY{IH8nMQ zUCXAGn~LJf%2;3{?3$>kOx8Hzf7E?K00eG1I%~mX6aG-`EE3v@g~pI8*IVsq+C~Wi zq(cs;hFkLU!}q5d+a6mLImPVVel$Dxf5bm35=QXPIAz*lnWH(wNW+1F=9&ZeK)o`c zc9HJk(Ph#Avs=|9UR+3%KFEO5kKkh}oB)w6%*SX+ODB9R^pk>(I$GY zsPqefheh}Gl{VRpxgia-eSbJTUwhfwj-tf3JPKA6M#eBFgS348P*#=kMwzVyqdR*I z(2y~Wfk}3$rTnZjZmG5Tf%c;AN4~|oRTW_M71|wZS9-Svm159Vg&Z0>4v9#ILPw8< zIy^)A?;9-V7-qyaV>n&NLk0zQ2UXqD40|^87roZM1V@O#MpXA(GIr+^(ATr|tB_;J zMn*y!US~mtDi}>ldN&M!C~Dtat8GMY0{%(6IF%&mK^BLx3qw?o-Z%K&I8fZ! z;2hgva4UW4p9T~WvuCo2|hrg?vhbW6}NOczJXIfA2#N!f>JVbWtou0M8ENi^w>YpLHQPK-dsu3L8_vjxVg5xNA+O)PqCb_pX zH(8?$`iaLwSEQ7&oLPX;fHs>?!~ULv6YX{c86YD z`_OdUu#(jJANlvc?=WvnHa50^m=5*>{!Ax^&V>jlqh(PGJGAQ6XePk<1r+lX>mPB= z826i90Q)x4S`5*Wr7$jV{-*}Kn2$w1qS7=(Dk3AQ z3@YA5Xi(XK2Ojy|WX%KB;nBg^j#GzI`v9B+QWosFZe z&!AhWd%h3z!C6?sW?X8Y;7^ah(AygkS2AFPgvY~6sb1Illsk`&uM9)^Ud!3fAh=5C zDEv`1!svx^PuQ#FQV!_t3+$`B`FI@cM0DzTik0edkXuz4@td`4cFYi~Th$f`({3y1 ze1b+j)nmIgYds#aU0%OV06K90S(<#ipKlMS=xu+wKOnz6xS~Ge_<9R4F~0*yb^<0Q z&(MR7I#1^R!lN3nFif$PL>{1nDQ6rbQ<+%$Xg4!Cm!;}#Z~m@Agx!!xuCDg~#>w9( z6@2dQnn6-=*#*C%J1L*uFDq_7{Uf>d;|6X*KW_20?kfOMciu|ej!SYT*xtMA85(n3 zY&||u&S9%S5iSx_b&a(P-dd_c?|bKf1h6Q0F7&vhntrJV{O~%3C8r*UMUku|>T}5y zjV$j%+~C-_*iw!#Ru!MXxzf`l_x!2c9!SqmCykpM8;)wr6Y%EHBFUm6m{FikItqMe zlaO0#BIQR_QY9dn+$piBPUwgPIQlCq?lpgHa=oq19LVn|L@e34qXXhaa4-opSC>#= z$6K|xxH0jTk>b9khAkNRkW32Sl+8zgRa8^}JQ&8rq$V!e3)AmeEs2aZdGgK+@%wif zv0E5;>z=?7xwg1}kv{)n6v9Ov9Zw;$b#--7o*D6BUaN|lHeo*CTDbIP48qUhLYrKVJ}v=G zHt=S|lSC?wE)%#tWvhgZ3Tfd^+XK;7dKH4#hy{i5TozxT)dr=z*M`x``4wdt~LUZ`{Q#x5YxAK-wk z_B=Bg8C{Cm-#_M@t;4jpw`U+5A6Mk{awZ!4{ueMh<_Fw=1=CLL|m|!q94Wg#H zy3E39uXRXMvA54Ct~xCZI3gL=P{)^udm=qtDwjf=7&qb}-(pj#v9bq74O_i$aI?H| zOrO^JHS1#@valQPHl1t|NL z;AIQ>W)wb1BkLt`Ye-la2Rm&KnueC6*4(7zfErQ4&`5fGZ&r(csWWFMCMFJdoKm{F zyT^QQH!v`ua?c9X_j79v4dqWw^D=q3_87)J$4!1<9O{HbUAm|%#V85iiI-9aRXeoW z^DR+SC^w_Bn*E<~f8!n-!<6o|EYi-7iSQ2>4$_n&`LDJ%Hs5>Kv1}ot`}=za^fthC zH)m#U(h|w>H#_%!dHvTA0{TQu4=t?{Md!58HNydowh_T8yeOjZ{0 zz~*pj-mRZwWl;fre*PLL#F5ZZgK}k%3?*Yx5Pd?F{P4kKqE6;0Qa0%0$M?|1Yf88{ z6TJu8dcGub*d`k$WlHm|*qoY=sy(UfrlFUrKs$9wi1NNWMFk64SmbtfSN(}YL~}eM zh)%UEuNwF6YG(>_OKZw=ad~;`A`)IXcNUsagcSe-%STp!jJrrn!%I1F<(HvM@vwn> z7B~IKM!;ZRw1~eSvEbbHr=>cc`}b8{K_aZ7G@xJ561^SG?Z3}0e(xGe0y8tI;loIpa z3eP3gwe3ys$v^)XPmt(~Lp70;x-c9zen3~9^(+}EvHmR$}ET?j~*4Bio2BEUp03PJ?dGrnOyK>z1_wUdJPXk)=@iiZO zBEhIsL0mc5>$v)H4Ou5x^Oj_8%+UZnS^G1s3hAoHWm2U0D|ab?e{J*OaSREl#&<_Pznk;7Ik3u1eySXQQ!FsqSw)|OPd0S9nt2){)HAE5G9xhfP-PbcxvO_ zIZC{|Lb33b@IJxezNgyjH?|(+$NxigbE%GC zxRC85)d+mNmGGwah#$~spoPS=TAa}$>K~b>!20Jt!m9I&3rlGfPM`RIud8xGi?>E0ei$zxqsjO`H6vDE@>gPH^48t#QwUSty{EjqB z1btn<36jXSbbI3e0<@4GXx##0b4MUfMkOvCx^@B$Mn|V-ybs^GMy;|?zw>8#`S(Iu zXQ;ryCmML~9&n9T$~y zC2$7LW{0nqoxZpODPqOTZ?Jo4&Y2SZTsdt6Q_WpR$Nuo@5=6JBG?Gq+)cS zIaM1aV$4U=$1_^L4?+w#T9?q$l#Vw_+*I;9cufZQ&lU^(z;^#nJX%bIa~TSvCcPiM z9y!hT^61QX)&KQRhDwO$!&U)WTEUmc3rA(clq#nB!<|`F3IJP{+Nq<#%TjeEJub#_lULgMt)kohhSOiMHs(VOwF0L)W zm#YzQJ(*B|$XQX(Fir0?k+BzZgT5Jj$fJ!hXcH6$m@a&J%e^G@R|tz3yxBnZ*ZQ!! zv5zsM!pQeTF}go;=DIeyC?BI6_cRPFE{~eh&@=jE#o#s{@YNR;1qM{{F72sT^DK%U zZ5o-`1ttwYIh-ELtHzld4Pg+MYKcyU{6BdR&Ua8cMiC}9KD_@N*5Iu};RpS(QE@6t zcsm;UAlgL&c@|E;pd5PT;$p?W3&D-uWDAE!MrmOmBU<*mX8)w5{Q#!C`oc))xK3?? zdFDWR)=tFx2o)E#>Q>mJpLkO>GtPL{9<(mxZP|L)d>rQjpqhOCpigLm%Wgxt?Zj(D zCneF0D&p+R&c#5NOTB7pYSO5oe#AwZV1T&*R+bU;y&g}K z!(pLDb8R_864cVNtaf0D&$fmigBT8=+5tRC8tDjMkgAPsxeQl=)bK}O2L6T6=U zG0bKYBY(!3iF~L+4g7-cPj%ZpXa$~Zs4;-P3fqjZj5Xq)2p!puMGZ4qb&Z|ObQ?L0 zsTIu2hk`XDa#0;iw$LP!-2KDAc5toCfmDn70zo;>eIfG#q(xfJ=%ig?&X9T<<_gI# zqO7*lQan7^cgdJ^OI`ohJObFdCRG&gR&y8q@Vx(5!aMl$vpL(xi@qX@q}bUFSsiQH z+J+ukK?810?abibiZw!kKwUg)TMH`@V`0?pPb*m!1@Jm4*JcXCbO)rzSbaajl6r|` zCcp63-@?E(^LE)exxqF=AH?}#Pg(ocZ_CmK8v!cXobTL!nbC2sAOc$CBqb#~W>~(o zZ*XP#;?erlO$m+Z*th-M-`^*Vd3_l7PUvNhDZR)`)c4EFqYMoTyWktDBSVu4rHCdA z4IiF>P%qaF`G!4~mEVdMI|tn}Qg81!l65;VHxv2gtsTteqkSo$4Z#!Q`<76esvMQ$ zEd8NWj}h2K4b7H`53bCJoGsBnu9MNP5QYdy->~<-(>Y#gLf;$L?aXV=QH(P<$P8z< z;)gbz>NTVdN~@`4tbj2WG4?xnaKpe~C0YaeldARLPp%oX=BD;_`(^2DRQ+ z%s&PjW_X>WKBtF&o?z|Fni4Z1ViNZ$evhdo=B)IadFVx)2P8(}si*gW_cLKh$!J!p zLG)}er55Rw9yKPtwGQ_}%a?g%W<@7WQnT^jKl{S4WsdFz|2x_Oy}}HiMzb?2co7x< zb|#*AqNcR$x^~`gNYx2upV(Oma3LQmvj>>mQ`o6PQ=53YX7y=un4%8aFMof3)^EGAmLdlVi)R*g6+ zIVRzuy8`naz1a2!sVzkdDlgU4oipT&4T3>)Mwj%AmZl$5ah>%uVnkCC0o$Fx!CB}C>o zVsO@JkG^V5npO4i8UMv3NAvyQ1)vD_muvU{aO)?~+iD(fXK5ZQ!DykYS+;y(a*4(Q z!?O4i7`3Pn$0GK@bhS=kgTWV}iGB#73HBk!B^^+-=8JFVCwp)_L_5MY=Ep>8*W>;{1oWe{efafGp{}6dSwA_|K;I>kHDQ%pr&N zGa9kLjCH3mKTWWW7YW9XJ&m7%Ng5PJVwfG_}|L)GYO^{TbE3&Mbc*cZiCeHTuu9`%uco z*15zBq;7e!1aw4Qc){1%KXaT(v2xkCdGw_2FphB+eMV>4R!sqE$(>azHeX<)X|u(?LsCVHj~BnFfG}=6 zQo2f*QF+YN=&k*OW{oq)?np8@BQgBj$=^17VD&kQJ)d-m{Pp(dOCb(1W)z@7t)@=e z)8^vK@`af!t*&-taM}rM*ce)D+Xx~|{}8$emda%hL>kn|o8y{sB)3dJ{3>%~KK^?Z znDTLKJYCNS)%yc@;ji1I<{4k%kJRx6=km3=0uy3cdnOLk{x2hW4y zoSy$MY(74-^u7Z__+ZDZEp4qdzY-i*t-FF+1*Q}!Y(W9cmSCyOBv0;l@jYI4I0tK9 zi$=!{H-Bm2;XlkiO_Q7X{|t(vo-D?y=VYQ#rrqCkaaV}1teo9Hwh%Y!f|Yh&h0der+(zVl&!k@kPr+5bt{ z3};}2_vzvJ+_z`vGONnub}qI}h?5i*VMHCBlTLPgwW}k8B1;X1aFDA-TJ1$>oM=;y zRmCCe^UMtmB&gV|tzy)4!tO(*J)95N+7)Die506&fQ2%yZr{C@vvyJE?|f<9dtJRt zsNmn`vm4crpKjlRwXLuL#5fv=%r z4+^YhvU&M|57@T^#T>iM9EM^j7$d04?qe$HlATMomS=XTuv*>#%NsFdtZUwHp0v^C zGkDqgd~-D5rTO_1DYJ|MFmKO$f#w3gW)o z=|2N{Kok1!Jb^FA%^i) z$v@70p_NcYV`O<|@9%W%WjdHs_XFj0darJd=6lA}1DBf7s@?1JVV2&PzhsE9#T$^s zU4Bj;ahjN(o;<*W)EaVd{G>?wK5btE+r|_#myDCz)7uMTO>Fo)7wqdx)ChFu-0J~B z#D8kxfZqVVTE;7~cpbI1e!-W@YC{9Bm7+{R84;MhLJp9KX)aUFO>y&F&6O|(QUo$& zUrQnaszc->f4D7S{8(Hft#5`h_ytx^Nkvp)K(RsJfl|aiw#WAk%j;!%jf~yNWK5xW ziF2m9we?e4us~Fb-Nzpa26Sp%_0>UFy{-f0F*25HflJU8<{)VhD=RB^3rY&!n5i@h=zxxHr1Vru3dq z?7hjku=;=Zz?rugJa1y(AC^?)nu@HWKeRUKQf4Y)1(Q8m0Ak!1GZx(wyEP56`9vDB z6frd=CT0c()w*W$WJL|}DIT@F-(DsG;yUD%u`>yqTU-8>J`QnzFi4MVz+1^G?jez9 z-Y&1LYB8OSe%X_V^(HEo|MbkO4(=MH-B6kIXvZb>GJ03E(*2f3dLDs*A|vsFs5{^@ zBSC7_Z2^pg;bID|wogSTO6>bhLP`D-G=p$&DDP=@Q2sK%V43FSC7}|9*d`)%qz+v7=8WEBE8-)bwm01-3cE-CbV=O@k9tgyobhw=5D~RNN*Su##&u7|8c8 zce^LzlOxd!{)>Zx4CQMr=uc+Mk`1yI=FhXXQ=l}3;6E|3lfTe^HZ-_o2wbIh!PfqY z&!cJU2!mb|d6^G&$0J3miwE6}{y?#-*Z<%=)DsC}$$tSEq}u(dOp|7 z+?l!P=xFd|1}D@D7ZS0p=>i4O);-6cJ`%8oSLz$srHZN^nwhF0uXcO-Jm%(3@ zzs@P4I4T&8y~z$`C`e~G6XvVe?_}4tK}+|9-vHoKg&sujSua4d3HtaHCDKua9?RHv zA$!iId6a6L*3PfaC$^rN4ZrZ)3AARC7MK%)nZMlhCFMKGHD7qu1SNuCe*eZvL;ZSJ zkQc2Uh1G|y!a^EAbcf{nz*AZORf1>r`j75k;p_zZMM_QMxl%-Kk9+jqKD+*vhYkr- z@!uwi`#U?7T4CFJiBp_S6p^mvoV(UDv!-7zDZEEgxuU#NIi5dI*kL_9%Je9(xtukl z;T*nRG!uop{DENi0ec&>=={9R1emrJmk)EUe1#TxI2W}kps7?%ITuX4bah=^(#Nf( z&MIkcf~~sd76oad%PmxLumy1K++LFS4*w#P8ag1N4=F_&-@_(JwwYCrg7<04LiJc# zsp96w2@w}}xI&JGnpGq8W(NbRC|uSep6FEfZYD!V0Kr8lBZ*Bkoj`ims}3Mb5P=@^ zwj3snM9y1IlLBj((z3L=I@Zut$+{&bnD<^LX^Y?(Bf=?dJI&~@gqc+_lBnVhNOhFe?=LyFy&>W{6GA`)(Nbq#6X7V-jO=pOAQYN*6E zDupo}0yQZ_Sk2|4{1^#hJgwGYnxDh+SnTSA#Lg)oAgeTJz;=JsP+5vtw|n#WFjSdwjP! zciMMs#L>yBX~$%8{KDW2ALMt8BKUaTG1Yghd3Rp>|7#2b@# zPe_>m^U}?=dzsp=H1}mNeR84)KLtKGC<@!qjj}|vnh=j zNcd3~=W*xO5P#SIpDPvbUBYu%n1EXDMNETL>eF^Z1lM?8UjB?=@ zuNUYRYz@6mDT?#UZnpPj_^xt1TAZhA9mL{{Z1)mnF=R94t6^LBQZ%LfF=;9YqzQmz z+aDE}oJvsl>4v+cvS`)`9Eq(mElC8MbJdjUw`n`JqBnJlVJs}$cSMAThkMDP-JLRc z-SmJWLcxyK+I$m(k^RpHFi0~#!0Ha|KtVx$I18R@1xkxwqgln$fCExCc+(?z&WNgQ z4(;%JGju!7L{~aOLdOJ}fuLgd7R}SfIxgp4nNp;bdIO+R#y~CfGN6%49V#zyU<0)n z6tyC-HK~Sz#RqKf_-pLe;fZHvW`fyILPnhzdOa3VZ|v^oE4|>CegwuGoe$*HSt@IR zQmJ5{@SA^`F{gY+l_qq`DNrcCO?TrBE*}$$Zh~Jyv}G9_`S24n@7&Q*)ZHA~3SVbq z&jfn96Zf9ona z-64|4!_Kgj++xJyH==|jnk&0dT>FcQs^F+)-?;?ikqj>A;Y4QQ`kPt_u$gT6;CUXa8^xfDy!fQY$6XFR)N1l!!W_02NC0fq%gj|2~PByF!pq> znu7e%P`2WX2~<|zN~)I-k&z`?#I4hTU99F^W=feKtPg7;UWsV~?Ukf_@UQAvKgb-} zR%H$|!^k?VtgQBb3q>C0nU}WDH)F{jbU=-(77JfpEnI8h zcfTd+s3%khkEy&P5+q6d3V$jz6gls(J)ktmF*Gn>(b?R(Ne8YH`aE4vMup-Mt5cU3 za#PvDbVz5Bn#%v6x&5%*;=(K^q!(6O%l_$RMn##pS8NU-WJ#4XkU5x;B(@m2Xo}KX zm4Q)>6;#S%JhVYQqR=NI-{g@%#X`|ZMiRs@dW&f zpBrT0#%=s^uxWaJe(uv~gaJy*%E~fvDYR{E?D&9A!^|uRbfm8-4Rmzq5$i)&9J-C2 z{)FRB%)zJq1toEM^&>ooOdfLpWc(j{f^Wjcv+*mx&RU;{KiwUhHMX=^JNFm47-x#D z^wu@hc4>@5xwwu7q*pWCv?D6fA>`nKu^!L!UP-hG6-s zT4e`N3a3OD;bQoXtK?D@_4Pxgpe1J}?c>XuaIa|!5&(AdA^zc{gpxrwUX{*YB#mmx z=eOfkXhV$tMW}U+siHgm7$FgnY*-36YE^%5e?*)+yDhwB?@c^v=jPbPMD8ko{L??H z0e{6&MGzH5+wSx$_A|5`t0`2G5JX(usAJCTT%28YwZ1kl+wb9XeYIthT!o?kioSCA z{ixF8;Zi@UNN}vrt+ceXmxsOcd1r4J z<6S;Rr%@}s4t#rjca%&v|BYMeU6NScf&FbwloK*rY-SucAKB@9(Ck{o_Mz1elg1~3 zz{ewiXJ&kSeqljxC;3&{6?`R`Z4_7_8Z z<=gFS_Hx(<$&#;|9Yy!&W#i}9Dhj|}Py4jGhcx&!W;C(gD}dAqW-ba`&P{fR`0d-c zDgC_l6FVMRFn{@~rmN!JwTFuq@nfsRM)Sp$(zI?=Ehq+EAEH@%g8Y_^sA3BOPr91k}}zNTLbCX zCbe&%f-*S#*zJpqq_{v97$l7lIGmyn2qO~=5-P$+Hr|X%!fzA?Imo{RURh>1T1!1L zfwQi$(LO&M@9W)lfV!^GIHqr<+t)sCT^~WJP}kmsFeOU1qaGnB>Gu0gcoVgf4M9s* zo3xdu?*u$9h@m=V=WB#5V#T)^<@~oTr1S{1;K_4qBH;>ybe}&qt}W&r3~@RxcPX08 zr6lUR!LrdSN=t{LA+!KroOpbdoz7otFQ+a;7$R+byilUVOp^G4PGGW;Br%lq^d*w+ zFOIHR!8_d$+v&PTz3F7Ww_Q@J@r;2Hhd$2h*Y&ax3lW`%CF;nFlYH1zyiWDMTsU8J zWY-^i_LegTtNyeam7EV?dUw_w8T5TSGU#8uBI?I!v80i+>P)R@&kIMv=jA;65@75D>N8$`H zv1RyvX!3Lmn<7SN`%nE+rWB8Vy(@G_k&Wc#xN!swgABL+@Ld2CKIM@vT%V;B+1vuc zt}OqIuHj*{hQDq#Qeg{`Nun9F7?=DH(J;Rr9<{d=ydVA925XD(tGwSI*gwyX1rnT+ zhC;$Q0_$LkOe|C^J`7%3Oo)MrnYf>=1|>1we~YJS0MioXd!#%J3ihG~gC>bY2v2rZ z)mI@{lgiWeD)hoShpuy483f+?h@28MhT-x_A*gXpnu*v>z>?4C%q~LZC$-|L;-pGE z(=gr9utPI9IMOL6PI#5C1L{QkS^HDm+Cz0U+xs_S$l`Odz;zF>q%oD4^=d2y-H8>< z^-at9bj5Y-%$LBm+WpSXv}q)3*l`1NDQE0alDT0ED0G9z?lQUN0!D&VH?uTIe_iJYcAPu+rO>2Y;UyHLnBiU0)VG5c?Sk`i-=l6Jt=9PUZ=3XbNkhFD zgqZPQBZmdk9Nht$drAKe28+6HyPv_M0OC)Qq_mNnNlD_L{Q)R8<2=4mTlGEq1>L35 zQkt4N*_yeWy2LD)Na~ymYRP`%VUfbb&qFqJ{t_dISo^2`2f|)w zdnE^3vhWT_J~d28?|shdxXV!|Pl{ z)ug?Rh%)a#$(wOZY9=V0ZTlF1pU(^wi|9HRXg4&gdn;pt^_-nh4NT0<>2q3497=$c zX`FvNjYmZ*BQ|#0ye+F`vl-p6v-k{lNSXc6H8cF&;S7c_&p%$zQ3vP z9wm`8uZ8zunwDeItHm%iz|K-zrhDxi^BvI{67l^Ffrf@Qp*ZIkB3!e6;N%RrupAyF z%HeY$L#sHiJ2}E_uXxJK%Rj1l3+O8)prbTP-cZEavKQ$;a|OcO9l)BTY=)8o;~v7P6u~ z!(E;W7xuHvaB^izjG#if$2;DANnD@zyhB)YE!&Qu1Auiy12&#sSAZyw)M^vwc-cjI zg%PgGj!i^FM0V39B>jIvBTkM_YbO1h{;;YfUW7yw3uA@RLg*wu!+Yp-LyzWa)e3eB z{0!;ZsNL$K83v}gM^?ampl{qK?(z}(30&rds>Z?NW|V873_E8mMaE&@K;gT+az};f zd3GKmcd}YQQW6ejttt4o-W4BKGWO!{M&bq-_K%i(+8hcQ>#)gS zJWY2{Y=Pt^Fbk4YcoTp^lY=VQS$o*sHG8kE-ZA-Z{H-JTl*3bbGVWy`F*KY!`|%)VDrd4P?mpjCjEFD!+R@|8$q-oq=voC{WppZ?Ucq`DJbhI4GxUDmBEHByh3!2F={YL!?)A(`{0{3^b{XAM zZDp^~SCjHJ+ zS{}?u#J9wvdigZdhe zK%=O*Sa3YHzgp|LaYCQVUnfM7H<-O2;D38uex{6Ftr-_s-U-~x&!>9#z3e-x=qN|n z)&ln44J{rsE&V@3cbYiJAZ46X3>fSKcx&Cyu1U_1RJ#{AGDi@uv2^SX!xGxCa;ZS7se^8QD};L;z5*O;aUb?!kKx0hEg8$X~xn zczotqp&Aw2_cMj!s&+>?UR-;@0@xT(_Vx&vJ5rT`FAw8Clvf^RR8UHbtK^?^xI~TI z`Q;npNvg8I+Q*2vwSGZHl=1jUK#&^f46i4GQ!mfYYpkGLXTtbsU1q0%>z9iBJB4gq z0A03BT356MF4T%))T(!VZKO_n@id~f1KE_8it0-+UEao#((=IG*|oZKGN~IJdpQDm zOlKZuwD9G$F;fg7iitAcGrXGoqKF8un$Lt=Xr1F2Fns0x zawHC~J6LJ}1Wy^Ee6V;yMn=ZM;$mfxx)}uykOW7v7FrQ_>cF1o2oW36(X~DsvnmA++@P)-n3OBai18JOHYTMY(ED zneNEl7G5Zjo@|~HW6~WBMgR7K;|oJ35nj1)I`h0E z>K=M*O*JQ#;2jFz!)roDANBmJgEG=T`{8YD2@#$ee}&(b1Fm?uvW`Z$RdV~wSg_Q6 zD7*yQ{msQ7zxm!X-X?hm!oyBza4<_Q-qHV@EgFXy$n9VlSFXGE>Hdjh%aFm6gr2c7 z>t+|*R4DV?UUE(*%!<}qy5DY_p9k;(Q3+5BIAKMvWt=wM>9;Z|7usTyw{CuOWF@zu z2yANV5y=&b`{qT6jy?I&%E}0p1z~>GY1$`G@OVWYs+`yQ{6Pq2z$Y;yV@-(Q(426q zs-LI_&~%;;(N_dBFj6b&>V^XlMf#_>uFCy=i{}kw@6O!Mo>Aw|S`jQ$5x8I^Z3c#wPO&50!S%hnbZTjTkcmL_;}?D9Vw3geQsA>nExuU0R)|)K@m~t0 zW5B7yv&LR4-abD$o?-AaUEXXUIo%d+y{sNcecIkSzr8)x{W(s{LIbf)vnc8HAk%t2 zB_qXME7OE-*;{IWGgDvsfltZ#2Q!shQ?TbJlc^l%X6P<`fd@kEte<5qA9sTydPY?e zHY`qQKb`l5(Wt8Rh@bp)C?jm=_C!g@(r*mwV)pkc&t>J*(@^C6T{DIu}whdPong ztCSRf(MTABpiK4qOTEuTA{jh%M8dojZPg3ZPy14yR~X2{Yh&pFf@w+P3y*y@&s;m!Sd^2Oa0D{JBh- zov5%YlfdUk>cbQ(*0ndhwDQP4Y-m9t^+DW`@T91pG4w`^AmjAmO;`=mIucyk&j-@K zSF%tj_E)?GxQKx7rXOP8dg80>p}*lc|6SUEwKv?2Mb92=%UA9@cI{`|ucNaiaqM8; z?ys&hYV(_85iaCxY|0%x#I*ku;GLQ8D=!jX%7^)f?Jt8BRG%7iOdI1U86xMrWk_^A*vdSHX@(>2% za;Fp|@HYJWWX(X%RTxC|1VxdsJ~XnOY516+&@d8mTMx+9OrV<-Y&K7!2&_fP|U zFQB&dDuVspmMJ3>qfC$a3YQfq=BM zG>h}G=8Kuf|H-lal`~sg3FJ~ZxgXqcR|L!jUE(a371h4huqhYq zE&FLZZH$%G)yjv%ch>*9g7;*VsodvTXO-#IoJzt0M|)&sGz&fByW1n5{l0)^6bOSw z!1qgkE)mYsE??dE zufJCjxI6FxTW0ScVQdz3%TIp-#Eff)O@OVS>_ReP#Il`24B~YsE5bjIa(}WbOExS| zqEd>r2OjI`+$UK+?!4iXo%KLd>x&3~hB8)m`7)M7Nf@CX($u^3eLb-U2~`017{m|R zwQbWUYCFa(Iu?uI_wsV)&6Qg`IZw;Th3*|%eWBN$+l&6jWi3mGmfL=2I-zB96($eLCuvm)WTw|lvx(YdHa4>B^K5%%K2 zt8XBNhsQJ<(}s`1h1@Z>^c??i?9KY}`$;Hne!(@n0$0td&2g)Bl!FF8DLlYO&DI>8ncxrdL*A;VQW0aMmlIZWNJAxPMJkFGIw?Kx5W~Hdln3|dj-s}CAalspyHhR@*aSY*y#^|T|Eco&Je8Kt+3A#tBfbCj3&PvzY+TzVE3oFXZcjYB!xA# znZ(R1AuwLHiOOe}ugy)Iynm81eX4*r!mQLh@+99U1P&l4e~`TE>na~ZZ(#I<0Xn$S z3T=ANmHe;xxvB)>a&m~bNqk)rrqB_|c(?NPmEY?7E<@&xj3*FGOiVKCi;*6_Cl@l% z_53|PQ!F2oR9eQ>m_)Uoxz2QXhw2%?N?H-N72{saW6vr8f1YwrE*73(Ti20UUpbVr18In(*<4Zh5{%IEH>KYsqw+wH+swyu>>Q_cuZ_63a zK`ev0!9zx-R#;f%%@hy~BMx;jp1)na@5_FWK4kg(;&d@aeg_l_x`B4=q4^PNuw-y|g*yEQjztrBsP{Q|!!rPe%`?hsTsU8zHt=!*B9Nafn$E9eXwI$Uf+Xk!g@l zkmH=4xDi??laA2Bb&X|JvNaqizo>pCcx^kgcc^!g+#aCSY+08;xFR4GKuO8TuY&u` zQ?-y$O}JcttbCdYpmNDoz?@LG;eme+2XcGAbYWpjN`cRZ^~-Me=xT=ySFJW8Qol#m z2zEWt@f~0ueQ(hoUofz`-6q|WeX5i3d?4fF1IOGcN@D542dIOB=KR?Iycz$_);}<# z;NS!+ZT%ttzNT*@C)<4q+cpkPGfm;;aSt;iKc-0`Fs-AbgY4n&D)xTg5`}?GYuIN@ z6bO*~@Fu3F!pjf=L6K+5!Z~W>H58aX72HHC?h?veP=a zeDhZ?5EV8>YL9xsu%jbpAgIGrksOzhse@{YXc`)TOr|pd5-%J<54OLnFLd$*vfaqO zg26F}^vwi`%d-N{m z7#ZnrvABUJ-0 zktmZd9bBf60ew%9JUNv$Kc4~ph?<%4YXxUq{VrM7G0CrDq^_|c;YC+1QB>WLHkLo` zY#_pJ#WWQvs-KwCttlbI7v<)r1Dr)=D4dF~{M&xHx#pu5?V%)kHl zSwOz7SGz`UuP?z0QsU0)pClj@RJB)URDmDm-JW zNjhj; zm;k7F9ZV$UhRG$8Som*BiQ#5ZVwS5Lo|cZWn}@Xj>3{sYABaN$%8`iZR_S;^Y=6F# z>FxisDfAt2X?RF{YsApZ7IJvqkRgf7V9>2+eSV$W+y)YohQ$Gd;DeaP5%u2VAYVT_ zL;j`{>=a}vMgJ)zAbiWg=B~GSzhz^KyY$NumK-|L-f6ExKg>d>e)x4tz@U;!uLswc z^Q~8VhbdarEnTzDc~X?JlhMLbX7Qvq?uSX_(@1>_G)s(+p$a**g}Q_8gs3PX3)yYa zv^cn6M~1uTK*R`!+C0DPN)J8 z*OI1@m3=a&>L4>`GQ)bqF3|n`eOe9T{}n6zMLT1Aw-f zxw-7RI!xeaq1F=Jgr5i5K}-x{PnejXAhPVe>Z=028HqSaEkKAxz)m6w-10uW#J&B! z1ZM$`9|3Ro`xz$9cBT2)#Qkxv(1xbZV%}opR?pWjpFve!l~hoWxGzt<(n7;1Nrt7H zpPHmpaiULh5l$8HKb6_P&KV^* zPgxHDn2$+G35^nXgrU?*wy3ZNo3S--u{8auhZK2^jM#TGZx9tjE}5cyw}}0IqV`~t zGHL?g-w%~86}X0m#>h(_0Y;rkh>$w#ZltD{{CP=z9vp0+q#>h(*XXSdB)zpYge#=z z{Ck!llvn3l7Eh?nK3d75VH9A^y;QeLQ$B=!<#WL6W@c*%Z-9z2U0+My!GdhIxe7qW z5d^=y%O4EM0OHACz<9@FVYx0lixfF4>otkr-MONz2J5TV(ve+m9|*$#^k+j6$dZ!c#wk^F&*=~!Hnk$qe6-dWEpEY1jieb*?E zl2PkF_cHn~w)&ZM0PNKJ^?o1)3J4*ZD;ywQ>H=METDR2UVJUVg+lD>vrZ6q%$5UsP z)j^V{{|RwzV=yvj25Fsw55bh<#q2GuE1 z->bU+opn+%MkgnSAg_614~+N7#b#fuZ;B@|?1I6gxHJuBSdxpLpHERPu#loHtq|C_ zJ=s#GMt;tz{03xU(O8CXy`-Lxi>`180^rFxybLB7c-o^`HA>`V6u1eMes%F>Y<@>S z#)NANQ#XzXVB959SI~N%c6!UP%n~Lnmp69y;fV{3Xwb;p$5g4mBW*~rR`a6GoMYi4+1!f?o3|Zjb z=hu~ieB`PD>&K5hh+HdR@rUTTo;R~EP$Q4a#PdK-_8*u1f%+VfRg36p`XEj!C843M z9Qm3D{>)a(`I8XGK41$>#a}Dtn|1%yTT6oE28|?*6D`YwqXV_Dfm#CH4_7x*<71k} z2fnP;EX2`b?OM$_QR~&>QkU_MSrF~*e-W|kVn__g?adEaa6nNksb=k~$m`PJ2Q#U?ot9odg_q69)WdV=s<5$S=1<3BDgQGQ^1PwyPx!>^ z-faptDme4Ek0xonkPrN^?=OH4Et~^H676rb+YpchL#A@k`sQxKoGqL80sG`Y~ls&r4*xbw*KT|m4QMzu`kyA(mHS#-yfHg z)>-jik;oNw4E%G`xbGVudb+0&1amzC;!-ph5V?B3&+D1@TG>1*qDC7?zg~31b0=tk zB4F-cxB3##_j@`_GK~wLL6sT*aVhV=bVHSrlM{FU@*%q~!sc>!^EI@o@T1bqs=BHl z&>J1#3D9~!rAaA3AkBA{SR{r^^63O75{fB!{sG51Ndi{wI6Ha5DUD%8?Zv3&+Njr7 z(2myAs5=}Q5Ir-6h0MY#G9NBeK1QoLV#p`hX%f^Dl`UbJs8Vj3%%bMb|LK_?LuO2o zn8{bq!o8W$suw<*2;e(&gSoxEr6TEPN~#vxTTis@h0x^VCVgEHfa;Qc$gQ#ezpeoR z17O+-w`hN0HZ6wf;k-=vesxLOVEV7v6bh2{wFotj_ssT*foIUH1e^P9wh;f!KV&T( zot75fTrIKCg&QVOfrZ`>DI`3Nt6GoQ^Y;&UM2hxIttv^@ePt5k2o!u!Sx<3H!^ToD zbetqIpMBw|j$#lTdnQy9ViAqim_%DX)by@w*X7P|gl_mK*LUC3t_oHSegBxN#Lh2= zpL!G3X@yzxMcHrGK)3Iu-Twv%j7*_35L*tCSRg^&K=v1lnD09ltTHk(qFhG&@9|B7 zvJ)&xDV>Qk0VFC)LP7!-R-qZ}93}DmW!OGOFB+Dbo-6#47~0*MngK!YgJ$j4blkiH zfbv^nbfTHv?i%X`|M*-kRQU~}7@uD0uOh`1wbgmpI>!&UuA?7@yJr^@$AX(&SFfqh7zvoix&Ttv;k$9M(6M_yqUWC zKv^NnkA7)A){f?X^$*_<01IUHzYhQvNhN?ZkBg7rFra?*&z=&IMb-|clY>1jOPg3w zhbFA$R&yoji)aeIP7hpZc4}Pw@g;kYft9Qj3CPrDI5mXL%Y}We984ow>hcje93KO7AF?Il^Yulx7Gd0`C{7|8NTsNjF*RFnU$pY5xcV%4c7kt zVT5?g)-eW&A-@tlYPC4J7yF6e^O=4I`6G{|@q7VkAN$e5y(>TNQpvln(f*@-|F0K> zg&i_2J)PC_)`oqVZ#^fe$HE>Pz3@)i$#^vg`pX0z$n4whcu;z}9y3Q%5|Edxf!AqA z0Bi!46Me}JZcwSep8`-XbWtfPD}!7liA$5+=m~7nl{AyX@;OwMJj#N)oE-AwlM^Lvm4I6nAy3a2 z!z0dRQSdZ{bD8nJ7(+K$f&UJY7&s{)AkF0f+D8s#Jg|rS?p0n(s*O(Yc4{Hk6vX`m z!qT1}Jg8+mfmOypDlB*YZ3rh`!o$PEv63}F78A{NU=WE$OdQ2`)~dpD{cf8JDBQhL ztAMUbE5O9S;jqHkzB)A>|IPp7>7?Nub&G+GP4DaI?d>flDYOLk3GQilBOVb!Jgt80@NH>v_fGxF|6f_pdua9XN!#7X3L@%$ zAN$8z7D+uk>IAsE_l(bqH%;hi>P1{vJ6h=Jp^`BD09FoPTA7SNd+B;Cvg-iXT5Re9 zVn;hiSSKmfXmf_<-PQq0tz#kH#EgN5U)IY@;Gdn77oyEs4FkY;k>EXd$ty>AUIiGy z?|6~89W!G)SKP|#%1Xss14Tt2djQ$_o@(dD>Q*p2#wD#3knRAi=ssk@XP-Z$RUca^RjI@K2UP3T>WnNd=Cos^g) z=J$xSB0oJm+^9ty(Xgz8w9Bv<%WNltZ7}DM$bKvzhz5~HR5?3fEZ67Ml_A=02K`qu z^?oV>BYhfRe((bt$wuRz@0F1_h%wJleA2^JDiH0;z3H8tIsJxxh+K3z{6BiOCmz}i z6?`u*9rxLEP!%_PJEbcGyiEx8)WfPrcdBE#Zr9$4M;q>VqqA*#MMXv2Zr6nX3TQXg z{#28(0&z5zO99r)^~VnyaK~g;8K}8;8!blARtJ!y#v3lYLk8XvEriR?-egDfb92rQ zG3Ac6OCQHS6~c0LQ>wg#Vdyy@e9Uvi~bN-b3|hP2%@~%nNT2VF#u~BPd1q~guL@nWT$zU zipAsJ;nU+eGwb9sCMMbc*WO!(Wtnwh!+?N*5(b@$DBX?JD4>)e(h|~0w{$5;gOVbG zk^<5xZ2-~&N;imfhvd619B0PoVV>jr^S!^``7_9I-8cK%d+inHI@elQjz;Yzp(Tyy z16EH3KhG|*o1xkx6LK`4{^BP8Grs=Y5?_-QOXGV=oIh150-VN2m2I}=-?+&1*5_F< zSOdK(MT&Im`ovLTn)JQg-N?0g_vN~iO*=x1Wy#4eD26s5tfkMTcAd|+^UIz)!q@w#w=ah52<~M^0;o zR)wu!6tJ$D?qo&NlzZzE`ONs9>SoN176DAs`PsXHS!*k8p~T@T*42)$~UsWQyATS*x>rF3Z@hQS!KZ-@OBnkSIw~M&mOCABU)n$VvBvcVP9kYh5 zq(y4-Z?A?1NCU82TqcJVMKQLjr)6xBvNt~Te?i|k6UT?%^I52oblUGAi+DuVT zk38Y|m8|&tn2*x5;WM&BgM$icYLB75$37Tq&TSGD+bw4XrOS>~Q_jny;)c$EBizY6 zxF#-~Dm?G@y!u7YmoH=G(=U`V(b3WAY2|k_MHS|DmNvwSA2uLjV%dW2R_L*uL4V^u~3*O@w1Z!#u1iV}wDNEsRl7Tlt^rn0KIr`L^T*v5v5 zylbN8fqtGZkb~gqW za~tD??upZF?;&+Ue3^if^3CMSkf3v@d0puhvZIV_J5mx74uhX)mQ=z;Z6D`8|E`?; zx%T5njL^_fVn=IB3GZflnx*9*>xJ$2W~kApGs#TsIu#rq@ZH(O%aFN>v)iQC8$$gE z%zksP^n_rMENb5Lu7yea8*$kQOGCY{(0@8koP9zt#GK_nEvka<4Xj4)$F{7wK%F>i zRk?x*?c4-+Sge+#);>0O`*ULGO<&9nYU*@%o)(;5+k-)I^Mx)31}UA@q+CYp>Q*$W zoG85E{a1-F=$8x8Cn@n)al1SC6rox3_Rh{hw!W=C&Ruy_(!HHsQEjoiHW-yg%(V|~ z6#1)vSlpoy#Y!OQCmEeS8K1}g{I$iqQ*4!9QdG_ly@x&Id%+_3_quaf)0=+W=u#ao z+}7f+vxm^ZG3V zNfz}*|9wH3;jOFMIbnywT4{%*H>tWI~8jq#BV0GoaH zymwn2TYBSbES-Y0@*b$M_WP8})DSZ^`&^WPp_jCN&C))f+Ic(3je7ux)u4m{SO)OPL zj8xoN-7Tkq(K|Wy*X+LP#RkQaD4b`LQpK!u;h4}nMmb{o*ZSd->zbJb0Ym>)W~Q`h zH;fxk8f;{m<}EEMx|;qH=HaSDr>3MphBl$FdKXO~rWa@j%iu^k>U^0niz>TJcf!_g zx=q5uf(2xj`MiVmftf?hcy9s=cnyaYq{sPy5Umx_Cs^ zyNiECh3=Fklm)@%Mfe3--KO+Rh@yZZ$6Xy(_zOw#Wb+p?SitZSsd5rW*pc4%ZS7=}~zFGFv)e z@D^&EV#B9TH*flvSq{AwkJ9qD^CES*to~=P%8r7z`d;StO}cD9e+E;0T;UWEEXQpf zwu`J>#3LTNEIBm?F4y?@`dp-!=jN#IsdoCDG|qjL{>Cn*`aU=^xcN9uEGB#`PJn_x}Ryeh7;7NfiYDPwv zjar{Qf^tVH@=@rbfG9KU$WP7__*7SW=LT0s+}3cRcgIOXQMhPMVY_n{_Y@Qev|VQC z1|0`1dZ7{%=?J9c+^1cdH1IO z_obx~r7x&j^(ehwwrjdSucLQ0hEdz|*OP`GyZiXz1%_OC-&}lsbOrpkUYd;XUJo4Y zc@JY&pix0JzQMTJAxt*Zn?98E_%s^xd`%|ni{tuRyXEG-7BbpBsnI((dAkXOSPqQ`gp(Gh!qmAu;wL;N13?Tn`LBK0fHZr@yLu6lPBxYs%XPzi|Ok0Z?{OshbpBM!RE13R*b-uGeetRLO znyVd>naM~o;y}PQ{{ua&rG|)=7caanx4ZiecVLmf?PSbpkGS}RD(c3sO=NXcFx>vT zKmU++Xoupt{Cn^HUf|cJ{+BvNvQ%EcXl5?n@j)`P)66(~CaS`|he$=8JvqUc$)j)G z*D-5c_$i^kO;bwFkZRTb=D>1y|HZ~HC#?(KHGdLKDoE9S{Z(nXxpRE}{A&?;OKF0Z zm)UaIKA*7ljxgQdcO8+G;A^Avgj5qIOqw_0lt(F0S}XXH2u>>xj2J*k?0ZW~a8vEBCSvxl3l z`D4-KFn)V5w3suoIPMMIAHpX_jpiP%zoMv&-J`U`*LjTJxy5LTzd>8GUdOgyA0=08 z@+IJtA&bS>>};k^%kTN(Ukb-toE1)rMpWFK3S96CF*7qJXr4m(;p3%O#pjpvsr|&- zb=5S~UKUiJj(h#OSG|aO<;r;e%5Vr4Huj9W@!e~vv}a%*3DG2e^DVTU(HMt&wA>4V zZhMO-o21wl2NP?>oiUT0tuQLhHoc zRtvdaqC>)Qd8%3_wkn^7L^q1+>guY6C(RJZ(4PNJvFBpF$zyWgtWGq{ZGaBz@-%A2 zTu#NZpQLXu<;kQ>QF3%4t;k0l9(7I6b%vxo9Pm9_6Gca`9C{7@Q{WKu#BPrPO0+|G zYhC~&SYs3zIs6w8t=}UoLnU-&n@-5s%+Nn8O*^cdHskvz|+HK2IeN4K<9%{!#9=doh4*3wD6 z<*Y0&9fWO#tN=WizjsOhJkUl-+Q|lFy)UpTK&6AKwlhZhM3?TpoOhcKb0cEyPQRIr zrZOZ7)>l1PpyE+`gAo0xUAkN>9}x1qmgK4aaF==N!8H|8 zm&;0VjLU&Yyowhpd$O>DIX{Nq3o+m&QtnT`Ndz0}my1mZP>SZ9FdhhDR;)p@dR z-OlIrm-|o$jJYFFcRGktvBXd{++wTecU=yID1Kr8{X(OvbZ8rwu>Fjl%MHdRK?~Ni_>F=(h}~ZEzGFI#Kij-k?ohf6do1zXsOiAck}uEK`wVfpP~KJ zmQ|;HhDeG@2nIa%r&J*ks!;OfWZf$djuYRKEBRNObrn_qX4d~>g}ypo3D}esS>dZNYrH~2 za1s9#T{cwgSsGd{_4e8OZ_-Bmnn!zNTty|S9nSrc7TT~-I)ZoVR1oXHr&>WNf5!`B zKp*?>o36W#<~Jm0u|8yQ5>r}iSo{#cv!wj|l}daN*UvCqM@UGh5yc<6wORt!5KJMk zTDkQRS5_+N?OQTIn~;I`pDJY1(N3I*;x-~wPE}CWxKJy6Bb!fYwrs#TS!jnY}3Fbc%>s;3`4?K%rVYT_C3YT+Il!ZQF<#YEj z-MNRr#L@V|a)U*-UbEtbs*ld8=+^g@&TE28KJ7=kpPd`)tMK+l?FoGPs;Y2u?%?nM zIi%+4Hw}vH_BCx-;)O#dYa3MPzr7MveoEnOF;=EEPd)1?o^2Gt1U#hB??#sRK`3MJ zDHsqGa1M9)1)-dh*nxbU=7~qpQPRHS(dk=KZYMa8*~#1!XaDEh1_1Q%MC~ zG;ZcJDq6*Xb-8xyWnO6YOdUD|*${~r7xO?s6jIt{4tnUn)&;+@ifzWPfYQr8rlA;NBM1oykA}cH>N{q`b6i%kF{xi`wsphqFIF}8uvOl z$g`d)Qs@NG#1KG3y!s>yHB%ETAfw-Fnm6I@Qyonlcp)WjC?6Bb$IVT1fdln$c$IN@ zWCOD;GCk>KC1LyF0ri9V{DwTICGAFmZ%BD!*SYU!(YsXGw+&s=V}<{WJbu5X#FuIX zYFQtLkZ)YkwX_TceEP!NBQ9tQf|IMj7pu8WX(cxlhrt#l3e9UVVMfpaUfsqu+0kfC zOK-_!@xhnZD~oq?Nxo48)3isJSelD#7~$A9wI{(8SCvhoUiLA|?Q*-ViS#fLGOx$wE{8pa!RUPe0Qk8N2oR_s zlF;a)un<*wpRNP0GMnHuIL8Rr2 ztP~U$ctGSHFYgcISC!Uj#@(U zLFMoK`a)`>RJE~XWSXhwY%X9Ru7%lG=DnF)p6@2w_i!AwqpclQYZqiZ17a)>QTWS_W%`bC6{AxEmLl=FUZ)xVZSEvg#YjI?q)bmjTk2GKt{oA3APK?mBptEGGy!zha~| z`nPG~=3~8)8 z(yLdmwwY^@CLF|r?tF+msokYdAZTtpmfkd z2qw@9+H=SjO`rCP`yN1PKJ>&25}mnH^*UYQCA;U{B%MdsH)Xd6@EV#3v)GMVNDyr) zF|Zb4LxZnR6vkb4O4+$ML*D8`%NPs~B~&G;q+}v1^@@)tFJO^T)Ye4v+@kRnRMJ*O zv*QDw=1bB}HgHytm*UtLULxC%=I_&K_Ej9ZNX<;@;`s3J^6EiVAm)UohDPA)$9PN| zQ@W8-)b8h3$!0ef25^V{Un?7+v~;(9r&+uO-8+or5_}8XKR(`{ywo2!aOMmz5fIZ= zHLC7^VWbEYcpP~yfP`uR-_>d0z?D`KC1E%s;GQFGp3w%Gnx^Jf4X#=2Lovt-oyAY1 zxv-ldj+*!6*&mU-L>KuxHJvB;u+bfT{!Q-r*-%L@J)x~JSe^_0)cmq*CU540%IOlC z*cLrPVq#+dS*GkH*tn$^oletLJ2NRZJf>uACFNUaZ5?m5Gcr+FKsN=Z}6% zjbz~F1haeaX8Rqs7keA>i269%r*i_dof--M;nV*1qt|C_C?s+{Jh&yb%*05Oos93x zg7s~5)a&8HhoPGr>nTM==goR@Bp_@8l)hUj>Oxg8YMzYS&cMe@5}ZESDp_nR5yARf zKZ;N+yNXOBO0JxI>8Nkh z%F`oe+0JQ?Zf;?bKi_1yoxLMt-oKBN6YoHbCVZ_8)kL{;bS+k{(Mk@Q#h^gW@%^Pk z;n_oI#$Ag67>d8w!O}ywNY6xoI=qK51P{q0+J88uIWgtD^wJg+ZO`5lqOCcB6hv2@ zvOnN^SHo2$WP^(#c_JVNW4cgmi=Nnx&k&ka>WDnWt0p6*bx@I*#<=jjWzW|ybEKW3gMK!b@f_DQwv zXTxb98O*qK4Yc>aj4&TmU6vVnR7Zl|Rr+Q(`Kl-1@B>;sef zG5U@#HdR;Nz4w@9Wo6Hm{9%@4Eo2`t1*m)ZM%;q9oUd4~FH>=U;d==o=l!BreXpoq zj$P2RACrd6tC&2dDWim^jl?kSJI@nowkmyLG7L0Jk=ePqsv-olT}#?c6E-8~*^N+t z7)nz!M;~f$OEkr{Umg%qKwDiizIM%bwOrOi1feav5Qp=v%vi}s_Q*~?5(398$;$id zR~2A9z5DVd3J3*hg5RP_H%bgU8z=ph)>b!zEvZCDczl zL@$))_~FjESLY>u32!dr7|~~?6~VW#M_XMAuIzXR+I^k1kLDaBc#J7FIAyH*M+F#&^5vgE`u>GYpR$Ns(ZiSk8)X0+&8Y2)oH*;NVB&qyRa{u z2)Q*+Utz3GC~igi11dh-1$xmK1Z&-V31LI%4>r=tv2^h8P=RyB99uzPs2LPd&395i z=8+?J_SHfhT=h4xB%IIA?K)(fUbqqv5U|}7Z}t9FC$?FAZtWn4%$UJeS4M5;vy0}p zw#jBzZf@_cc{b?Zx{4_(TJtibe)xAv@n4Fm_8|r`#A8?}KB+nL!@|?(3yd@^qr*)! zv6AT@T~`K{4;O~8pA&vbpda<*+pVT8IEBU-IEp#pL)LPY+a{hlHh`2PATG9%>l~L& zfGb5JDUaY&l9&}kc8A9BL3X5-;^V1Zd6S~UpZT=`U1JZ%I5fkxws#n(_6jfMVIRx8~%8IwM&^? zS%Y`&aIp-_y^?19l#~aLyzpu^XI*a5mb>6)aPITvH~f}0`T6&U)5-wAf^$<%VN`;rFu|AkhyL3zQZ0pw& z@Z6|K{V5&Qh8)@=)+BXdEK9!SN*YP=NB~FnVMN73PddJ^@J($r3RTW9{KZyWW(lY z2b_rL*P_$U?V!!H`sY-cmg7k@EA&6T=Sz$|r*+E3pdE8=sKQC#h-34R4HTi@_IF?9&sD9}pGyp< zSg)UOIUTSNkE0J=N-41aKFq&1%w!bJ&{sEz@i}&MZe8w`L^sZy%GIWMW@scMxAOXy zd!~0{{w(*hmznD(US(-S>ha4A1o&c%!2co7my6j}C_ zbZW6N(+5Y(X-eE@I`h{b`um5DWsHL>SiB|;wvQ_?>sEQTd5qhhZ?DGGPz*R-tzl;I zelYfkn0N1~atQ5tlq^ZvicA4-HjkAcn-2%c)egaz#rTPMo)33p5_t*5Y0` z&tW^+DEw8``?#0&qYFRf<3PPK(tP2Y{<7KHx7HPCn`O8R#p=N#?mO3^ykv9xnV`e{ z_k!xm6d@>Tgs41M#F8vAVDTyYq#41QY#^_@tC0C6I zy#|3W-glDe1Sjo%>asJ(EnztT<2J!OkdWH?(?`y%F{Do6Wu0vi++Xj?>J~|4&g>*c z&CC35D&r;g1}Eikrae(q6es|MP<jW{uF0NY9d{iVpyEPQ}pQPlZsa^@Ox}GjCBy!`Qit?8PaNqZKJLSeR`&99S zFEL6M>Ya*9Vo5s~;M)-c3Bj{+tL(7QnBoQT;7+I+``+f-yDCXR%t7~=P(?^v4GE|e^BY9|L(AwMly^_ zYth5tIPEItB{o4ofMP~5-A9kzt?A9Z#spn+iEU<=*%~5zLz%du>xA(`KcC#seXvtQ z6+%3GaBXkCB=O^v>RrpY_nguq#~qqjk`4BG-kdQq7F4v2n>4~}bJe?xK^sCw9tnbb z*;+n2<@wwU=4Y=^ZT? zm@UxViP^PmttEH0ez`{SIcY0zNrz5E0zXxjSqM*|v#qMH- zRf(2M9R4@|!%PUKNFd~CoV*INk0%{Li)^}R;t@P>FnB)Dl45iyG)-}40I`Bzf3eDu}n zJdc$oFR^5TJaoNe{#MMJNgmi&w?}}4an*yZw`Zf%?4x5EykCJIXyy5>;DO5yEyqNa>55KPn&isn<1jv}$6G3G0YosM zAhBk<-DdW6I*tcVQO)`KYsVdy_bnP4luhgo?h9f`M(Bt#qGsNv-*F;4d%~4M;Tugt zzkPsBgNLw)>xZ(~n!NE0xzv%RnrqU@SYYg? zTr#p2Q}wiX2HY~k_zbDzHWw8Q_b%K7??m1Av~Uyf1ivSlj8r*P7O&Tgzb|gy?A;(V zi)m<~>q^!tRk490^~L;ZoZ}_1gCEgS4bHsMXd8bdLFY@3&B^@F*}I!%Rsz+ zYdH?pBKjS3og~V4VjK_Q!N0 z^*K$sVAX>?jjQYnZ6*h|?L7fvmp{B}<%T3M2Ku$Wf%1PK3<8 z-b-RMq8PVpvt(jefAk4d{fkMcJhwdhyi$lx@cQwhft<4fAuQ*}RXJV_z$4*KAb!qH z()N%4^%1coMtDZ~b651gKSHwv^0T@a`JLc+c|h*1m@+(*`6?YO^zlLnj|^jDc})#5 ze7^9nkDz#GVqtkHOWe}@ACFX?BS2}cPjh{>ar9GT* zmhyOsINH^WKzK%Zy@>8ulYI0@qYwzi|24sp%j#4cXu@`b@pUVszW6FRwzN6L76T<} zH6HFoW<9LWxJ^4B)16Kbn?i?JDnS|Q)`I~lSIvpgKqt#b3KR69LuL+F$o7)gOI7Wp zvNC>%vY+zYIAoPM6nTn8&|6gt;1@!m?W<|-tZBc5hR#UTTkY-&LsEIOMg^7fpPofa z#jx#(!=n_u4)xSC7YEQcr4di8KWm5aTt+(xcd_**5wjYB>RlTN7K%?7{>O3SI1Ub} z2fzG=9vB#Saj5SS9BI;93?L^Gg5nQ^DV}^UTxMU-4p;3Xhb=}+bmP!ZcyIYVm$&zc zll9l|g}Wc~oK8T_GgFhM^?*#+#lB z{x(J0>%Hi;ta}C(S2}KBblO^)&UCCVQnQ4tvw7bG4>rjUy|X}YhV1EW9KZ6mZi1pA z5Tv9C9BzyLZilpp7q0IwA4xnLE3^|aY|65;U zqA?+5je;yVxWRhm;ZG~F+K{N#r*dj0J|ztd-gHzlq}iFqbqalPKaHmLHWjp_CFqg zc|a&;3h7l`f*{KAkz+9z-@w%SY!!|EReHVFIy-X(3@HYAjsi0wH2!z0q1Cp2j+!kFYG}!FUMMipBXcPxR({z3q7o=2i!iF0;zD61ETeb;(Zyn6`Ca;JFKuyafawUH}%7VE&JNcYi zPMr>ODNp5_npbAeBkSpYBeuiJ<*~O7dxT$LhQKq$GX4mJGqUrwR1XRT)zzol67;8< zqWU(0dq`VsI$ZgKb-@6!UdU!RUeU$=Ff@a_ix|tP6t+h_ckSdhlgD9J(T@k+=v@!}J-7Xt=^cTY8tE3t z+|_#$c_EWKXo-`8PrW=2))I~Da=DYY(?$19k)GtOIb1r2J3WQ*EFSC0)P{v&*csJcC$t^i5f( zVb6^$oe;>b8GM`(W?_30Ehk&}>9%kH@|9$zwO7j9%zN`Lobn(9Ck^U*DdBBRQWm(^ z=D-pA6rCa66E~gcS#2{xCXB3M{^BysyJ(1HP9WHx)nxahe9NBLP#UUXS`D4yLnz*ojLSE3xS*YlNeYeHv$(6U zdacQ7XwX<<7PTv07*-ffBFDy$p~aTQX%ftp9G+YJ$9dKPQYQxb1gQc`WZw!()nYCE|? zlrq2jRoy*+rc*N1?|6~phnLX}1N-ZG={5pEI$W{tlH)TJrT0HqyYJ2`5 zf@VkcTQ!`!3jRk|qNMgk4i`C}W$-#9J^kbL?A$uvlMOI>pgES)Q2Od>BdtvnL;45! zwsAjl%g5`%(tqA0v1iI|3&lM~d`nt39|`-mVuu1!QqOidTuNPk-n4of;)=rgBCDf^ zVH@-KkX=kpWrdf#uRf|;v!i@p}MSOU9#)Ir>Bv#52%_$=0$7}7PqB|RNs!w!2OTJal1rx1qBY~Q8*8}VYQ-8>j zU?3W+9M(Ej&(grV~`NKTs30j*<6@nzRc)>PVYataTB8u-}>j zJvg&R60D7=Vu%u8nO$J~8uv#ZAZenWR;#k3Y}P2kK~;!q82PhY%88T zpQPKOU1HXHV;3$V1Ju#w4Dy!m@mlmpYO?;yswhNw8G9*TqD_QTbEdb30wJTqx+d4h z_B(p9FV66S5xg-snLd?)Coma)mEsZWk$YE-6X}$2Q&g5!z2K6|=DfOPsq4mU)D)Z6 z+oD0+I7EgZu%?1G>*h%yl7U_FLT_4<>)G4#Ab}d^Z*Wxrto#-yG5Lv+Pio`~Y!$&&wnspY zZ1mAVP&Uwd*Os%m!!3~e)rcWOpje*BW9YWxEXJ$)&3|x zL)aUL@pK3hH`CU*di&W>ROe`&uMEI#K@(48SOcK6_ha~lYXzgKi6ZXOO%_M?)cZMN zERdMm>;S;_Z;ndegqJh^H98R z6*1(-VcMB8A9|=CDv%}{hb$FbDo&&+Uf{w0ZUf?)y-t6@*fohsk6<4!fiWx1u5u_*rZLE9H<*2S!aqP|;|qsgcZ}b^<>C7-riUIfU>L!Xiknu%0|&3-6Pr7mbaO4zqM~oZIL43j*V|M;Bt3u|dJ(>^`;pj=&C0>< zQkc%cYD`~PV7W+2@k*CnwLJgW!@DY#P=|5kL%IzQ)z^BT&>#O0P?7a2_()TF*ACw` zmbOcY=ddgr^F!LTO)kEa;ZuEXQ%`EMGm{>Zb3gv*cq7mBDxokd%QI=G7@m(|(NyW= zZ9To4?zxlBKcDwrde!_$zS=0*M^RAD2;96beg}kt5%G4^88%?iAQLmFpM!{40zpMd zK+?2uD*b->DZC?)4OeaSskSk6f*=7ZR+x5`i$uD70$E=8%cBdgg_S~wPt1CkWCpXA zO4!AKNxxe{l;Tevl5q%%ZR3!F>|a2!KLzXE?E&6jx9CWEehTf;q7musb_Jit#S;hj zKYst$1Movq5!i84+=1gh|L1cX+JGnwC;oBb*RA;H=l}d_dKTCejrYH7@iBi;tO|(+ zx6(UMex3F)e^MF%(5SC@{{q(|A5d7rK(C$MXW%}H3FYBfG;}1^Vd|R^aNztVTj7Gq%Wh9nCn< zqm4TcKs0;4UZ6ZO_kV3traJ8YGx3Gcqje)&FXn!Ro|QExB1Z(m51U%llOJH(L3#6m zlT*u7b4-J3^=cFemr=_in)e5HmLI7M?g#3BL>v zjlv5MGiS!tzh3LVf0%vaMMp5V@Vd+9?~n2@9F?mpoyyE5P6}NQ7oS%O()6046@O-!g)wA^!GiOdeOz+~V%y~PTxjnp9O)M9G(5@N?efk8EWOe7ak^#%0b!pR$*)N~L zSt*06;fV&(1DohlFo-vrX%4164~*|ZG_0<@qYt42l8m2UbE;lTD4UAbELec^!v(s?hj4PIQ886N7rV*|gx{D%;^%(Vfp>deBzgFz_}C6UN8Q$lG45 zA!8>+x5ZO9=AR!xb3+kwC!E@!!NadRuE2+OxMopo8M>75k5`_IsHKKMf!l-AA~6{L zo#q)}B zrwU3@*jz_?^7Ng~sS-sreCK6VR3*1TBQB zNld?~;Z#}-mgT>5PoIu)-z-cFF-~;n&4TG%y9O4wt*w9{!vXBmG8j!c-tKu|cNKE- z90rZyI9sW|UhAjm-UkOD`46q>T@6VQDrJMMmrELrIbbqd+I6HG8)$M$tUKnPM#@!Ml%ivKc=CeZZM)1 zN5;EkR{oao#+{qXOe{6qU}p`)-`YRt&YBBk%6Z(0TD2PL2{i-J=ry-JyQaBmSz^G~ zI*-Khw5_k=AxlSu-W+ulh9Y|ZNbbD*&p}Vbt>T_5QvZN?lkm_xnKSN=&Va~{&}~P( zjf@sRP=SOEOK4otmz^j0efEl=0hyAtDRu*RHI3G)W=k>r{YwoyQtR;2?u4pqwE&11LR15XMLtG0>Zs$A1gt>TMbpSAMwK^Tl>2ySQqHq500d zx;jFAxVZunBwe!RtUp}oj8vCr2bmISO3Xk(Qt%65*LrAvL8gyEDZLc_Nl`eLN)nRc^pj3U) zHH=EQsa2=QM6vuwMf2YFM3_DxgP$R=8B$h~{Ou}yVFnA|n&kL*0x_u?Cxs3VYAVRV zvu*XWZsBM!Xug04G`mi-(OPds#QiK+aUeyJX2!ae-d(K%XxYW z!AhpH=k1;eMEmHP#(e-HUHPEklp%7VE6snWK`8jbtMWWH>-l>FW}nuwln&lLyZ83T zxA=*Qk$ubZA)Rtt!qrt` z>6BrUaLG}Hi0Y#7DcOHq6Nr!Y&TlVSh5zW|`)A%Pi3p>UtLCC4+&f{Rkhi z@*-I8pwXuGj=CxXNiiQoPsuAsKpC4HV?xWpbl5s2yx1rPO|8|0#8f7r7@`jW^)!lE zy;~2JPY7OE9%eX<&p+xlZzp&F)?7U;d)=S=+QNr(w}Jtl6+yz5PKm!lrlUPbibmuN zyZ_$bc4o2yNnsSQ=4h6(LIH}lr4z)2!^lKWt=lDQ=GmQvuNXC?^HNsi$;(S#Q-dsz zyykudV_4~&OGi9hKRH^`vbaA0wKOdf?R={)7s3TY=k}{>vR8NK3k{p>s1Y_^20hAF zajrW{f2mcXvRvT=kXFLYRKD#4cIlj=I}R~JGQ6u{)9W2pBJi-iOD`OGRDsK_{sm(AdiGr;!>0@lyWkddD@x{>XzouoTw?Eu}#! zlS2g=wVkJ;o-iVQyyt$8XNZ;ioTj=qNWMNGoODw^BfVe#S<)tQF75ph|2UW8S_Fbj z5OQkU;kOz8@Er^w(WC&lZ^=kRt(y+Mj13Hpm+DG8+~*$yu6SWf>}{|6>)$`C4fGO2{M21MDDM9Y0tb_uw21-UCK*6fLEr@# zdB5O(%D^X-RZw-anJUdaMOXQmKnWiMr6(j9)pfI8(cyM+J~Sm#c~I%Ho|u>Pc%h!Sh;fSUmvuifr5T#wRlePDvj6S3 zzr{q}_lbag#N*pf`IqtiuOGyQaIn0-ys1h4AMw(iLa^cg?`w|T=>NAjzlq#$@^z=M b_VC1*;_?_e>ETlp`0u8K{PmnabbbC0m)VZ_ diff --git a/_freeze/mod_data-viz/figure-html/nms-ord-1.png b/_freeze/mod_data-viz/figure-html/nms-ord-1.png index 82d11bc74a9eca3ef63ebf2efb7c57cef5102d7c..15e8275f574d1cc27045adc9dea713115c49c4e8 100644 GIT binary patch literal 100104 zcmeFZbyQUC_Xmn2Dk0J(sidSxN*jQPfT(mSjg&BSBO*#FNJtHWbPV0Ypc2w4&46?b zLk~G~&-lJ7et+Kk|6S`{3)dRwaL#$2ouB>L`{DI{H6;?Fi$r*McqGbqZa=`oJ0p*W zce;V#Ebz^?MgIsM9{vUETet4p+)}#bVC&%Y@Yxenb0u?ob0=%l2TC{b@Fc>+H4Uv8 zHK}A1t4n#=y8;I?GovnGMBkO!KInZhx`;6^wHxP)wk^Byrge29*_u;$=IEpK6pSUhP%|nLrrp4*R&__$SQY~Cs8(n1sYaidAPeS=FAi!$~ z;x|=%`mqQO5Z}c@{<;3?(w6zgMoH&SK0PEpJ^Z{8F+4Gxrx<2F!$*-Q)@H=eZY3IP zxl(vv>C~%MWMSf?qzl;dgBrI4dNm)jLu#qtu=*!p3Chkg|b6E=-wf$EW^ zSAD1^Pgm{6@i@hP4<>_X&W=03f(lrdHj!b?tE~w)*zgQ%>w=t!He?%3HNEN>7oBr* z#d{lgvj*NyL^&vYhJs3R4!zkMC9W8AnE)(DKp;GG!K7~Cz_49NY$F>ReQL$ zl~~K>tG%ju8Vium8o#Bro%i*y%5&)%VErI1({i9iQt6Y>#D>{mW({Bbb zEBn0*3~_=uPWpc}oDy&{b~1Kk0Yj9Y4tStMY}TOZ#YFSgaT?>0n|j0i(%fc2GM~SE zHd9ewq<;B=@QI+Ot9Tw|IIUsr*_I*)vpk=Cf7%@Is(pUx;CH!4im_+8yfbG|!l%{j zXI`)=%h6vBva9ucf!xV-^;MA>g$6W(H*BTPn0Om|i^x*ReSQ{5KN(uDvAQwfv{cf5 zVLHq@xvV^4Ipj<0?2HK!^coM%`MHKv?vC3 z!H}~OM_f{9y^8CkQ~9TR4z1e7{4Hl-D7<{xO!wUf?p%!~lOTz2fnPppt6%KSV_DhJ zcem$2;0=4jjIf8to<3Wr2PW~VQn9D89PY$Tri1N3fRHfPQMOQ3#p3}!6X2Z+w#LH; zKAi&oT?GE&;ek>?KfgI6pL+V|=hF?iixuQwH1P1`@sw}hc;tR+b>dtSlP2ukI(Bdy zL44`bN4!gyobew?KEQvJpiYiY#Xzd@s6kCbKA#|4y~sc0E-mq88GOn&R5uCvnJD_a zN=7E!EU(s9uX37|CYqsE-DDw$YujF3=rE6iG2sO&&4t(^GFEwiyi*{2f|DOu13^vn z`O!S&f4xXe=I=@N$M669gs=Jjcp{~$c_x28?Vqaz@-nCY_Pv>wJYK>`rl!#Utl`I9 z_^B!XyS5+SU*l!P>$L69rTg1e(EaZJvm@MH?}3F|R9I_<{q5?hY|DSvc~TDkcmW+m z1oFaTWju<1x$2KMAp95U`O90J!B2f}r$^29m#gx4%k+O0YbglS(;-$6@;9Ng;_aOO zb7B8%{EhCZQ&tvK4nh94++65e!bF#$NIQjRUp0XmgKSFBk?8wNeOK&u~cY7g9WkROJ3IDh7J#*<< zJAY+kzq%Y!byTq;7O(8Bzrf_R0QRH0sprHMt0vn2r-at?m#6UMWKo&HzxRAVLQb3V zFBN!C7*Mr$Ryk6I{}I{#(J8Q60(pjog{OaGDH4kP@yh7MBhvoQdx4vKB!Ciq=Ey+x zf8_=&L0*`K%wKG#SwrM+kG;snif5Mu$@@P707Plk46p?j;}Ri%<0a&0@bOc_x4@!* zV@Eo5Qp@P$RC|Bhst9jDyC7VeQ$1ze%>3eYCo==jr3_V9Iq}OsJvtkpTMBGUNPj&q zKoX=9%l8`UgS2tk6}M<%!S1!}jW5X3GD*4GcGwxH{M>92;i*%LoX;44J&Lu+D7B>I znxP$&x#NxA4=ZjC8)(v8EVYe^VLlv1Gx;^>m!Tw5k@?$4wcPkEn|)A*1|#YUa}*p* zTv3?kEkPn*IrSj!mw|}LH?ZF!X>j8Cwa^nEf;d?5`15xk=>2})uUFw;WCdi67yIyU zvgYOkH29=Eu8-nxR)YszJYAG@_vem($_%I8CxFoq=fU2^{>`G!5&#zEKlkWw%OO&c z$J5h6W|aM2;R$`o%j1pW@n!vOIrnuy89|x`863be0y?R6%AacH>y}!8)hA~?nwpw= zi%nZpQ)XCChy`a%5r8qx5A(F$25wR$l3s5(?Kq&Jq@by($%cC5wckW*%&2YTVUE0D z^XGa}@c=_R({m@}4shd3qI>VJagfNhr%AM>J7Nc*p}9JgD<;#8uSxh{Zv83P23K0E zaV!sFtw|dXxBuaH4MGaFEgnth9G2g?NsASF>UpOx3!>apD1Yx8r)GID)m6d=U|VYQ z-xl>-;;G8OdQ?9za{jr@e^eu)bPC!a-lC?1UeKeo9LIEPh*id%s?Oc-!8sorkJg={sn<5{4S`U%!w7ke`uX&i?#AUckYb z(MksvH-v{JAeamcWyNFED{{6p!&+DUXBjHL_9dnbysG*^*DtcrY6HZT;%>suLj3bt z^4|&K@pGPy-c|lnjQ^OGRVomE0v@Yz{T9S8V@!PpNF=CV>+WCe^E43gNbtduKi_lG zOoP|}cS)*v@ZhJkPX5wGZopkwW0z@nW!MyX|mP4eJTJ|g?_ap9id+rCowd5E`wd#j^+BdtPVQqDG z48QdH<3s8joKObz+5n*hbW#nQ#~G$&XLIx9@}{S~26d3LDca6MrT!-kh_C=F-?Ap= zxCB^XG$EtM?Av1Jr9K^)r%UvB>Bo=kHNgMa(N2_qF0hCnknz}qF$W+wJZ2y&bC{Hk ziVXU5gJou?qFKvl2esY)y!VGb>1hGhx-LL7^3QvIID-n1J}UmZ?|-Ex|Nku)?oRjP zM>uhwVY}*cupNILY!Myfwm$W#H%lGDSqowKGmQt_1o9A#N!}G2r{Vh6efE++27)gyB~b`|%;*CeW;)dD?zO9C#_!BBR}m7YB>2^F#!+PIgd1 zGaWT^_J3@;=ght5j~4l}kN*7rANR6-1jHDPi|2j?of99podLuCP?@^I~u)JauKG|A7x-*+p z8KL=cvUOh}gF|CHS0to;@<4#6yC7?91wL7iEkO3E*`?Ll)N=D|h&O|Lyr~j@Z4ECV z49M%kgQ>ZWRW5+;0&bFJx9wCI zh=Mip6DnBIx^#J&3$ja|_EMGgp1aep+K_Y>_Jk&dRn)}VnUI)=;;6G6$hP#(RQq1n ztThI2w{E+IU6N&H4!d@7lw#K(X){0?P(5=n3C@9rFM{$G_#)=IVdQy6i7dR}OAE^e zMb(?|OA84@arTpl1pV~S`p>%CsBKL5OQBhi0|7zQw@b7yTtmgG{hfX8jM^_S_dYR$ zbFYk7)T^xiN@qCJ2}lBZWu`yknJVX-I?sCBwj>=f4>f_W?LJunh@>(uv*_cm^W)rT zB2Vjj;WHWZSirIfl8VVFL{d~74j1i zCw6#w8IV4Jd>$2c>}Yvo!VMa#pA28;8tgFY`zkdA_7LdcLyded*tixgBn*ZmYzPlI z40$g!58WHAtq7kUXIC7v$tldg8`u*v?*7qcVWF9;)2FXkrGICfF2mzct1+QkCEaNx zfnUJBFI{ZEqt@#|r_5cc%GVL;2MQD~3rUs6P@`PpInmf5<3O_$$svH|(v;ts@>dpG z${>6qg;bs)yJt^!LKJB$uSl_Q!dT;Eg@utGRg15+O_%&En%lGyW4niP6#9I~vU zSBi%@hoXx+8)i#=VftmvJsfxSVnj#H^bxG@Pt|KiURbgO`?=ZKU(V8m^53zeoFp02 zUTD5kP?$YBuAi~M0lQBtyXkbG>-R)M*RgD}z`@yLD0T55<xvAcbox^n@g@=d+H82?5b?rR;;tD zMQH6gA?B2pO*I^OTU}$r%^|c43Y+iuOI!LMj?Qyx-Tl%dLd8U3=X9$4gw9Wag7^UB zH9Ppw6@-6{?G1iRr0l_>7F2yMx4d}0Sli^293)b~6ycfd6YA+UZ=~-aRpn*(^?I86 zMb`?3QE+Vptoqxj#P6i+rk<`BVYd!+jq8?|mvuz#KIuPM^t;u|PG=}%+I#R&>;Seo zC(gFtaWFT$og`NMej$2j&Wo{Be_O(Hh#T4CnLZ?WLcw?e^El8KQIK{I@F1(p0qg`S zo~OI83&cWLjKSjm9fnH&%g6 zU0cdRD|-!$%fA2M+X~NWdLEpWX&NZt^Ilh1DYwo6SG?s+XlKSpurRGWEY5#?;3~`!eFLS1_=o(2!7KTE*5Axclonpze6s znR5-Fpo}U&vy`T3T#BoCyp}T)TCNQ(7+=k;9LdplmNSVPw{&;*9a1wgMc7uaLUj5n zq-$8jq>tIl3k(d&Lm}S$h&t3rychbT&VFAagTkE%m}}?hrK~#3Ly_}Hw|UG`5g`NY z>mPa9_A~$w#N>A=aV9|&Kn<;<2RlH>9L9w%`Dk8ul35qCyn-B%tYMPmtC<6JuT>*_ zk!R7vN$p9uSnuk^DO8TzuD4q5iL$&h?}jmNIo+_Ls1u}@lM&lJ&-q25<7hvAyWd`? z`_A(oU&~cmz@alw)-eHg0<3rw^RvEhfZUuYSOg>$cQ#tjuFjy8cFrzDK5^*XP_U`+ zTKvmoW%cf^Lz1vG3MiWcO-i)bm7XUAG#|{TySMxDV$c#6{j7lrfS1QFDaTul>eJuqhG)0K6RbiD2OUs_aGV;=*W#O1bEl zDa!6;*Hy-ny=FgW7?5@wOug^uGaM}LoHv+f1nKj=xfdon6@4xX;?l0wlIZ6rfTTZ| zH_mFTY06thRF>3OfHF!7?b5rKyp$+jy0)oSnFBS)FZj_gh~q_64bedOZLlDCq^;dD z&1IzO<;e-Hgr?N5CqsEc#?K=I)6nO3Um0;B$?e>hFKkZvf3V=(k9`gpOvH&-dR;L*)W9G~#7=2JNU zaH8+Dk|%WvG^Y7@&SuMW*8#&+>UWIN*%^KACZM9M;q24>X7Tt_Z*>p&U27;pw}XD{ zMru_)yC<{M<>P$ABI)^~wVe25;#)ul%R}_>L8Q5i_C7NS?L>K{p|Gozr{T|Z4ibD| zTw}_=-V1m*KN+y|Qwg}*xhMk!xb16k%+b9L%(v*`^|x_ilCL`s4rycJZlk{hD?v%; zJv~#dUTdo@H}?A~9$+E(MjCor4vp9@b)L>I(_T!{Z&7MGqIPLBDz`A~sIj*7REtC- zX#(rr-JPY(tdh4@Pl8_gTR`R@p$5wWQn<+drf#}^y*IAEN0?BKHx4>7%M>yes}Py) z2B=|I!v4jEw0O?k;|~PRjNjy3kH5Dr*=}xOu3bk`AIwjtgf6>i?rUa8+-W9Pm1R;w zs|WTVvR3JvGC1}ZMo!|h1`D9(6{l>o#(7tELz^|R6SPvUS{GUc0Dw>+9)%!SE+)T( z0ML%%0NllOpA6K>y9oo&401w5!Ewz=`pr#Y49uZ1&<^NB%xwAoYwT8@`v$LDQv+6p z{Q%;XDn~G&Q4;u5vw(!%D9HTlqu7|4=+e~_~&Ld>5r179-kIuH&WaNQcFx36N zd-u1cR{ZHU!%fp%;9U$!Phx)W^c2ViP^gRQD&SfT7KH4kUR~Hjun@9^wB^W8y0Vop zyia=^EL@3}17jCmhl}ojaCD5znVLGAtVO35E|aGAEt64F=8m0A-)p&ATC>UVVGNQh zw5EMS3vNf_UMB0>5g{c4c3WM=f9!P!NLV`*^*{hGdJkHOEl|7lerZUCf;X#cY*sUw z^eb%IcM#Ki_--5!_NeWq!8t;*Y-+1YaQ84WE34^^14~>4dBV|x9dg{foNIP4S4*?e zMq7?d*iV+z^F2If>8^PXwwDs0 zvJwshua)m%EaC0>8;MU=sD4L9eg%M?e4wW%Lr6A!Ibd59eF#wxj1k9*nMXsbxnMGG z(pcD%&db(&Lo^`#1hv$Un!2mcx&o@1znyE$PfU3gSHCl)BkZM&AkFvivA%a5MY?2m z_m-fXtS6+-5~#E1l1)z_pmE$#wW0S*-$f=2d2DuEQQ>to(`&JfTd;Eu^-G@ zJY$I`mX$)nW)!=@fRSaAbMst@VCs_LKJYc0j892)UyVUw*arFaV%)Z31Yz`)YOpJk zM#tTByUCPkhvc=hOqC~;DBlh^Nk#symOcm{aU<2FHP;|}^O@xS-KNhm3^KKvkci?h z9_L#8)tRXZJYB{YJuU!}cfr3q`mROISn{&#USq(`_o9MCyP-~#EyN{92=Y){Kq#55AYB(5JCQJK$Pxm8I zTJy?{z>``9Gj0U72C4brv9dVJYe$Xk)kD*k$D6P3SiGKm_TCLuT0HeBXX(Y=(^&MI zJDzX#ewN)hkH3zqhWZDe-u0K40olw8sJ0L%IeszKsP5=<}Ei z*hw+v0#>7)>H<*{nr`kl?V-I8ic6V@y?(#V=G6j&VzD1FT}u3I-z$ipge1!JYY1}8 z#5ehu7dBJ?mT50#1HTQlLMU`(yd3Ki7+9Xo3pU^WmK)-!qBWYKdLD2l(}t`I5n3>* zZ(b3}_yi(?0j{?^n-|BrA~>ptW+3pX?&8G_?M_cbEdKX zQQa;OYDR1xie8p~ObO!6;{O^>>+WYyvsA1Bt0+MTw|kXE2GBBJl=7XI(HQ;U0$UhX z_LtwHSu7j!A4c_dx8!3PCs0;yF@mQOK~;|$Q|6@-tS52ti5&@h+;n> zl6*Sww(x1N5P<0~7fok)Y_z#y$iYrfc)gEz!;Z@xPO*#IOYVr}_Rd)H*XJ_{)suL~a}SNq-DA7C+=Yj>E}obNo$Y&0b`X1*mLC)=kS<@71%M1c~p3edEr{jpDP z`Zt^f+0Yig(k|Td>eyRK*C@A`IIHwjdu(-pABU{A4~H=p&QSAWMPPtvf|P~%oNB$u zapfiY)9zoMTm>F1>*v*hZP^mq9J0Jkk#N!!zrF-eEdw%}2A|+aJJm^5;U2b?&AOr3 zek0e@aFhf^9VmF6=fKw;0(f#Eno$a1*fQn&+PxOdTe7hGe!J1|P=&}j^(YZ6s=H~1 z4f^UWd|zizmGz|DbfsOkU6$e-Zb9AQ z%+vg{@9(SrjEVuPKq4<4t9YuI+Cy?`X-Fcr_0p$z|$&p)(vggl0Qbymlo7 zC+FAllFAS4S0~r1Q&K49HuH407u(CPR$mGs{JJMG*b5j%aWRk z?;OCBJ^w%lh~ZQ^lGd(a-8+33{TkJmN#s-jw&963gbtN*5Wa}+J$?aUnAYWzPgP9q zykKYoKZg}}#mozlbL{)E>t6F!45L3zj!#~BVt)d7rd~a>jmO{N`tT`;U+?j^4uz$T zwNk%CV)oqSu2my|LObg{KOrQ6uJQ61a^)YrI|n2H=>Rk7sY{a*Vl_-+e4LnW=;q75 zY=a6;PM_@|pln<)tz1zL@7pa9M`fO5rc!kQ6H8PO=EetbJ=^{2r*?JT`;(@lXe_m6 z-pczCnIXqK96G*cVLTQvA+`|Pg(Kxtc28fI-9v#=ZFw8vsl zj8`lF)s=^tjkB%Mrk|dl5i*kL|sH@xPSjVX4*^=`Hkj^IA)#LxK210epOstz^X9I>b=~9!D*e} zCfid7#7F&V8O(N;TvNC>A|BIbot3QRfF}vi@)B!RGt(oMkO?bmBkzqSiH%0enRwHP z*R`m6skyiX1PsfK&er9ze882=>Qwma6>XGGs_gDtGiuF9ZSeYS|ewX5mV9evte? zX1nLs#`|t>&J;h){$8uPFNGJfT-$rql;sEENu-fllgnd)@xE8*-a1iiuVI&6R4-sY zG1^p=rxu1`kz)-Bae4^DNuekizlKFt`U2fyllwsxgUKB4md)K_r{y}j+DTENg9!E8 zmFV+1BMv9?0aEbsR2_2sEc_Vh&`-hsp$j(+;Q6I{_QkCW*V#$#M!yl!t;ZgtvL-R4 z0QIl4385F-^+#nu>xKyBz`B;$Qqos$@V%Yk@2hTva>TIW7>K+g{wOJm>72BOhW+!8 zpFBi!b*ip4vuw=7yKD{D9gCgN*Kj>x;mAc}UETLxyD3!P-#t`&9Y`wc9-Xl?Qfhg$ z9xQjSTiGboxIEWhy;>E~a0AqNC-o_^#(ksDYx=bhG=gI$bLfh0@LL~KYzv&M7CWhw z2H;=G79@~Y8+NI0h5Vx`#&E%0rzeX& z4lLTW{TF|wGXRb$009U288aBUyZkM0J_hm8ZDF2I#&i4CV1a&C7AGaUE&)}Pyo5GbdHMKyRSi?2ks-Z)l3XEU@8*>2oJJ=5 zio>^DM^6;Rj}rk{T6x_ANm>E4HItWQMuc5g=fi0$oxJElFpq?eG91gB}O>e*?(etGRjN?r!$S zgbs)GyQ4NXyFjxp-(rF&V4mlV^iaR=(S7Oq7B>@sdHi<8Vlr}#0f$LE>l`X2TX6Nh z{3Svhb2+>KsU|A}C)Sz30S;{AD5!D)S$d)b0D$Y&vx^m_5$Tb#l6O^9Oe-f`Lt(q4 zyt@c5)UY4+INh`*V&gNhU?VMjuW9(|)OeAV@~iDYmLpdCs_!qI-LBDCIVA_lgK2a` zeQw;7T&XNx{}8uKRSJ-8!^r0TbR0!`-QpauV_A4zWN=eB&H`O9j8;Ef6SpZ_xr}((%n}9vXd=BKzac{i+@G=CPR=EJ>j$Noi})YsS_yPL;)sIE8STRzK3Cix|sAqIP4ZaXg%%KghXHStq54vyFUvcpj^x$_Mf+SHK%a%C& zvs_jc5EFX%(yrH{pXvA$xrlNH0cg%Bjy%aVAdsB4%p$&F^C!7Uf&2j`zm#V zy$@+%!K?>Mpz5|vTBA6Yz7N6aLtko4y6AGbcDYfj5fcoy_ zIUer}weVMtQ?!y1>z8yKu9ONbg-y=xxa;{r)|8MGwb)HI&g*)l(}}>jHDc#snDJ_@ zEPirrerA2CuMlTQNT|tedpfdTVR~}1R zOzsD7yXmBRO=l34u^^;K zZeWvNq?S%D$3O;T_P=N(227-(g3;DjsFr`Tm5*x)M3DJhV;Mxg@9+Y4Kz`W6$(bk> zf!E^=PD0zn#$GM!;A59oH;NY*B#<2z>d#*J7d&m@Mna?)$>k3 zdpA0|=28HXt;nJ?VY_^1HfEPe?O4Zo;?SR;=ME%&l0}A8Mp6rE1nV7|%FD~0&TyEZ zor$aw80^YR9%tMCq;)KrVTvHZY`(1)T05dFRS2KD4D_Y0yxGQ|*y>Wh?U*-8`z&`z ztGS;G!xsx*SL9sDfAHI{69VOtey1HSWN5rpC1MwD>`s=doKM0W9}eX^bjHN+YG=xv zihf*Bta5EX+I)rJXtxp`zSN(kv?6<)HhHLSKM=XP_ZDpOy-G;Pb#-`e2#%o+W^&c( zJjUTHO6RY;_a4QrJ!G%RL)zp>R;Kns^KXx=~5&E?G zo>8ThrnF1Pj$=bKtwLT3=D0&IG4I!0W5X^mm|%|;8(yVghq9qyXoTp3xB{y#xKuUY z`)GzGa?HNgj?z-(s3+MCK~HP&?8X;C=O)E7_VxSo+dA-Qwa~PvsPL4^>*wx{l}93~ zQ&LMY-Zt!LLl%Eds5R$|PvQHAQZEfX;p4W=x9UCaov+`vg2{@X8@>)SZ>$39vrv?I znfzfnapMnXik8dIY;13cH~Nwql5yB=aox_p-b!>L1?sZ{&5z3!u`a#EEzAtK;FvpZ z(;aTesg^{i9NAY=QqqiQ+hHK!(=A~~7SLB~cU5kGLzCFY6}nEKQbq7!Sj-_C;aD_1 zBAQSt?HZixnFsVjaSN&|#6EaXuKj2XW%Fa^;qs3rNk>o$5JNIwnzmLg30)fc;@dse zz0>mXBK3(Neo+vprt6-}qt<&PeMO?wAMycM*;0C|Ls(-n46W*jjrELdjpoX_WL}&6 z@^+tB>6KX5(`OZ|gVl*WN92BL*{zk~!+Le470f<% zP!|1utmFKoyh_&ZnVjKx|7{b~YrB-(tN}pfmYGqYQ&OdT7q)P^fBK8IEIl%Bv*juB zsPKm4lrXIl(2B5$nsodzjP}ZE0t#M@A-)khHi${ouQsu}C$ST!QC$Q2HM|I9f^eX# zBu3FQho2DGu>So^ze#@l?p+{k+k3pt&7FB=EXUA-F?sN=&aNcg*K!Tgr8b`F7sn5W z^Wu7usYU@2Y==$uL&*tlwj&>}Z``z(zJD^$gi``zn=?ljeRh(r1Md7eXsKhdb#<{o zzcKEAK5a? zPT3QdN+5C%V5zS{6(KO&k&?Ym2Vcrcgr|-4;cDqyTL)X^=A&_78o(5Oyc;?bAtexT zeLkL>r>#@g^wRyvhNUbB^W=V)+w=~6VqyxftUr-=W*`s41hbfzi_A;E9P9V=)p{wj zLoKeJ62*6|{a&UcZLgQ_z0{d7Cg;^v&})riN=G)>iyodp6s$%2Aftdu?>xoi>uzh~ zdmWZWOW$&T{w)Am%HxBb&~D$|Qd}HVeFpHrefUUI#N<^p4O@2F@94iwG=QmGscBiYz8s(sO4CHH#7G{^rCwNV z+814_4v~FZS2GTv46FW3)v~AGct~Wlp+-!t*f$VhB<~(XRt6Zdi#Ceert<-kC(DqL zmz}}0PiI8YB{x~g?|j|~6|9%2-zjitQjx>hulmkKrfXI*CpOY=?Bz>fUJ`256!M_z zaughzD|1nyESRa6+an!PR)O-uly_c|R|AI+rY+?Tb@TPgbJdnI@4ctM9t^>!&&loG z*hWpt)o;c?8$A}1X*oDJT)N5w<-uf&BA&r?uARyoNBe%cSGuD7)I)RP5R_k_vN;xz zljXXYI#8T=sOftThEVfP51U375XzweDx-|6?@2II>{$rPc?LkVICOn5dekXC$=JHP zAcx(jlXSK^xO*gJh;nbI*c%?2$ra)Mqrtcm-u>!Et8Y0@Q4`rgCCLlk&z&?iaODJGz6h`KXSMVaD0{d)Us z>t1r}tWr~xzFNw?=i4xpk08u=Q}Wmb3LC8YTrC+5N(RFa%wDU-5_9h!ra8@pxFa(m z%`b?%B9?Y2o+`flx!_i*z<#`t zY2PN3-_eFD1-H+7AWMD@Y9PnO*tqbzRfYm_ncbKelJMYg5-aP9@L15{eb)9$FvH&J?v9ooe;!+w!LDDtZ!;B}z(*~!mlX;t-HNm09Nd`}x# z;@Tvg!wdzgFOQN=a9GSX>jEW_+<3>+Yzh3QXKENFB+mNnERHf&m%oS8QHD0xCscNy zuUI~7&CPH)!7_5buevg_p4_`yUZm&24Vx%o#qX!P2hgYmYfyF>AGZhu?2`9G>h5--iQ(E531@W3@n3%|PGf%v5g9x9KQx z-+C1J>&M%K+`!{AkO-7niB9>p*)wcmt3z5gnwO2=1d(4jo zCh>`GP61Q^_&_NYS6;bFtqm+*L(5KAaod06X^v!J5rl0&NJlN4#jl8!uS8!6ueV%H zbUTcLdl=^@6km0Koa|#Q1HF=!1d*i_TzvO2_t{zg=J=VA>-M$8!_jgZh9d?_&U>%e zBBifIq0FuW1hp}HE`gXL?C(xUAG_uz;LgRFbS`nWoo<8ve!7qE!va) z$x>g2iYdShYygx+3&0m$?3O*WRec-9x(ytE%Nwbhe)EbwkyH}~Dpb~Mhi6nlb>I3}y(PtW+y}kC0ZAbA&+!!NI-{CtM(*b@+P;LLmOP+)duj zPkE2h^nafZ;r0e{h(o%V^h$oI2{4uE*vb)i8-;)|Z08j?&Gd6XyC=O+u#HOiF@MyC z96$atrOx_PsC9RDWiaHpTjnYd-gG~AAM~v9atG}A9pw}jKK<;pkU7xBuTcFuh7UiX zR(>-nqH1HCo2TFy`V2RiV)iFK-638)r3Y`<8W{udNEMFR=q|JFpKj%X+~bdcwmDl*_pNpoZg>K~ z23K&_9mLZEkT~uDR5L+-KGXXMCO;(Lo~Lbnjq!Dm{3QZsx?b_KM($R1t~;kio02in zN+vB^q(aq(;jd2MigJK*(hIe#5qmb(a7UH~q}D8A9IoRHv_(C? z7!DFrl{Em=9&iFH&BJNjwX zRUESJf>kQ*)VRET0XaTIJ8U11`!`fjExI-P_PQ=Gi^}O5YEVNBb}JL?PD1}W4&bjT zv6FG%4~T(wMs5USB?h|;W3>a51!S+e)(_aAJh;Vxx!bVFqv~!Sfr9R`$s@j8p3Pt% zt2+wXFj*-hI4WX0%Ivu$allPVZUO$+$#6`4ok=&Q6K>kX>o9+kC{E45&B`56ree-q zDSA4XM&~vhX}Xi&=H=_9CXq6`uOQf?g+#BMEdmj#fV8M-Nz;I?Y8P+zTOV~! zFi!3&@}&3GK>;8No!ZA9VePQpdZ2ey_W(CaSUg6OH3tZXwH?H!81@R+$CGqfrXo)9 z=HoQn&e6HL+pD^JR`q6|t{X;{(`T-<_~~vK9dI0V9Ps_6_%z*$2bHVD+t*#7R&agD zXo!|pOZ#LWaT!hHuMUX33N5$Y65g_U>V15k*!&P&NNtDs$UwisT_*B6JAbHKMX&GX-5E%-V+x9iLt{8{iSFoIdis|_2K;nUr4!n) z(goZ%_8a1Q>%2G$4MB$BR85gr0nU0rUwN zrblg{;q<~5jpK*dDBtBRTx(&*#`sVVMmfO(;RG(+DiG1Gr}Nn&Fyy*zdkAE;(b zzE(zrW0Q2GGLgUxEM@CQ54!_5>q|trzr#tYFTh703+z%@iGd!%F>&t4Zt8O}S?xCh zO`)~jp;21qxR#n#E2yb@$e&blF?G8?ifI5gM}$jeWxy~7Fekde|6*Jx6-Q^&hA|I0 zX$+j$q*{YlDf4Mir_N*CKq}7j>nJK15WKrJ;Y6r%MHyij%<*sO}Ji;Mj(h&Qt<+){RZE89qa%qFYK#d_5P&6N(!!gtmm5dmjs zfyNzhg7MbM;^AALO<(W%@W?s4vEPlho<5*Y&R@{sb9{eu2TQ(_K;$4yYr)dNN)k58*$+6~8h-dqnB!8z#c!=MqBP4;mp&ZQAn- zF)oYLs7TqY)tH|EJ6k<sgDS^)eNe|YYX`))NL$SB9^YT{GNd~djR|i!Bz{+ zoST&KKJ8WRbDQu5>e}N%nCUj8`}K55)X$D_gB0PrGJ#NEmN7Y0*?VzM5?Rv5U-g81 zD9})-h|kpOo#**6N)8MWwF46D-Yp*Tfkc#amBZXN)r$Vv72?L@29I-d$+*EYlew}p zf52-iLY%Ajv@H+Yo*CpfX6p7{r`0GgaW$xz7Bep%l37pyrdAT{s{c7dkN}sj{QygA z2Rf!ilIX(M9%H}S+x(i35p^CL0}s`kOe1AzaKJZ1C9dL82#torDXUN`ix}HN5Dxe0 z02AwV(?|gE_jqA{Sog*1sz!0I;%hG-ngQeT)n{wlJjwoJ%nP zNgHb?kJspAJe{4F2d*fF(UDK?HH(|uKEJl2ON&E4(`Ef?p}1KJb-#mtpw3i$ zS7yzCJd0#xdKvku0^ouHV2(RQUJr~Cw4~k$oK`VDc#v-dlUxzu&-BgU&&=exn@Uf} z$1Eq40$<;)qi~`7Cg=)V8~FJqc~kt+3(L%{SU+8{4LFRBC(jgNoU+qW(VSZI?5@vv}1xj+>_tNvl)tYRBiX`#v7sb6osvJicl^ z!Djkfd+K`q2Wc;oKz~&IM72YBj%jh!F%DA%jP1zzMdcFjQ+k447~Q zl4H+15KN6xlfi%9J}K<JL|Xn*=nNH z4c4UU>F+uMhHsbJE^lr0xoEfSZOj0SBs$U&i*efqnk0yoRlf59quI4r+w$Jw6X~Va z)JSrsY^(rBeV!P}xWbp%dhHidei=h1+2Vf1(A2^t*85SR-WXM-E9uYbzmz&&)J7`_ zU^J}vMrL42jfKX?U5a12E442#G1VN#$VFWE;u@|iY}u^Ll}u$bnD23DT|s;YIEyJC zo{=3SxW253Zb_`xCKWe0YxL|6E+dh+0pn%30HDg*W4(Q$*R!Qi6rHE1ORh}t^yg5S zneJ2)xa)N_ctJ61LaAj+-H1$qgM%d3Uia{V&}y24VxB={uit{Fzme;RIY|wowG;TW z2%~kQx2&)66-~X3Gg;O~b>=-;IDHlvWKnPX2`NR&8^*Y9C3p>e%)}6)e~uf8Fa!Y5 zi5s&K>37?4Wl(4qY}^&TGg7n;_EEsG5Xo}KLC#x1zEwT_ZKuVXHWmnQMWZU##E%|# zj%NscFH5}a6s?|b+*VT=Dq}At71Ca@$}+u!-cbD!<`BxP^yi$vm?Mlim0Zlfy{yd# zsQXl1VjN?sk&syTQ9}HQ@)O58eRLlvgNVZV102}#jpZ7Zf+B(zcXAR|Bwwu zG59r%(h_Yo6$RB^O-EQulypOS_MkW>3L*9J&5IXgl6?vxMBEDZSD6@epfC*eO1W2o zL9J2aS(6*&O=q89R``fcd4o@J(-ZaGub!}WL|N>TdFjYR-GawU8O!;OD{zw5H7VmO zH@gz>O%tgpSWi(Dxq$Mt3T|p@QnP2)b3`O1G1^YnPYhOe?!3h#Ec2JIZkw++Yo$PT zb(wNiJbrOd>8v4iJL8U`JIx_`1K;}Ye!b&2*W0Y-T^G$T`F=Hw=;7;&(m7Bj1XS^Owr4bVF#!tu8ZCA$H)Do77*g+z~wSI-1zzBRXvK-K_Rr~#2^-wlm#}U->#b+YIH+Y0^{51*gOw05~9ilJwbFKi-{`Y5! z44;UYxZFp=Yw#tuxUo42_OzLoho_|R8RM5W{Or#<^b%Ps68KS*SYllj^r9Fa{Uod= zco8^$@b41_S*QBB&I|Z5H|Ow_gxoK~#us<~^Ge{Yc!Uz5Q&7sA=9<>86)GPZ>Y0w4 z(M_fWf{VnsiM+Rttx+Cc$8~Ghy zOO6ul)gilTG%xL^QvISeMuvYM_t*}!Gwt688W?prwZQi8{wSz z5&{y^-5}i|N*)@dyF(y{v>wirM=uDh%i$X`h+8Qy2R9AQB5W4LemZ-y4)e zEToA%oEMHfn0P)5b=Pf%q`dO3v;Ft@1xqAuABE~zw^oe4(KL& zx&Gf5VM`WfNuD|589n7HnUwn98^Rz5GagDg1u)&>%8K7;$7_T6jiI!3(q_qn*vuzi z2Whs&w|{wJJbvug5rWq?F;OjfQWj%@#gljE9NTlKOaC>U>fqyS?)P_& zss9~3QUrc77Sr#Pm*9sX16}lSoyw%!#1T(}6^T?tcEarQH?1C&{m!UbJIw@G7fPa| zoMz*sOiY93o+%Yi5t+gmkG|f2+D>2G@Le5mOY_AG#B90bU;tGYPK}V4T0F6py=ptMfQbqF@2Q;xrGX^bzK@nSHmlape%&zoiR6SY8oWT?{#EiC z5zMA7=SLfGHgXAEt!S@Ws%e0@NzM5j35u~Jx-u?KIu8{JPVJA*FkK){{UqpGMV&$@ zU#4FTA$$zwwwUq*NVIGnx1P{EnS~_zVP$mD@T zg~0>3XtDtOgRLLs_a!AIHPhdJYiu9T@l$l^s6aY9emYWchi(rIZ01jw*Krxty@0+P zf7)N_WP5fGFp7HL(#2}_d+A?kYJSl<2L#9B@85ZO>;6q{R0!e}3Q=02jpeu-RX83U z3XtSE5y$X1s; zYkk^8A0FQ*NiD(c0qUY$_#r4-UAp8;P{;MJ4BvLX1qn#MrdPrP7!CLXSU?t?FAxps zU!LxD_xF#=37@ZJ1O^AUf^isa_EZY-(Q}xdFNlzFqQu zzNz!PhfbgGvJG9>gxo|@G6)yd4nX>AAn&d-jL5NU+Bq4r>EBXR-_*I+n}AOpARep^ zbsY=|7ZdzY*AI){r4$gD!4k24c>7fNNx_*9izr6(X|`RJ8%Qtbb&aY?D{j8#+itqB z+^U+H(UwW(!vT;KH=pCU7BPph_A!TaMPt?o1)%WQXAYX{~(_fx7dy zo>`S560c35gco)g<4`OSffSRf(24WQm!f@0sSyW&E{#X+uQu_|xM> znc3O;0|}g`S10qOma}|@hAcrJD(CMGio>q823x?u$4;7A(vVo;3C6#%@6R10 z_V?02uqIM3FGaLn9ptP^RC4|qr)IU+fk*T{>JwwMau9hKCJMcSDC|oOG0Kg` z7cJ{*{;t*>$4}eZTHIBjhQV8DR8*9v+w8Xt@#W3(A%DNrweMM|fct~5B*^>2-v3@O zLh=QY?c#dnD4hbc@_&QdZRZSs(Nb_{s`$0Aw#V7ePQf{wK1Qd@y{>OFPB6LNx&`nQ zyMP8H%%Bi7Ad!Ga9l|%41_E6VP$utsQyN`qcocB0)s+|XQy$b;O1Zey9pBi1#3MNU zZMD*kS#x^YO4G!Bd5-Smc;dw4?6kD-)z|6lXFMlf+VuP8Cm%jknpKT!@}mDa%3fR8 zRt3$;A&87?PZtXTaH2mb;c#8e>4Gmh=~5ryOLp6_ zB=NlJ1L-GOnVAc-j`J4ZuQo*ytDvy5+=oH-1;Pf~gs9bePxBex zgGuX?!}S@D%_$3CZ4K#Rfh`8x+XiY6se95QykOG1)S0&0jL7QNZbSsE;G^sF4Q2-2 z27FdlR)0%S^={8}&1lexhK9mRy8n(Npw28r!i!9=oEQVCb^sZ)1R{<&KvB8W^)By}okxC(d15dPND|Nb00uU^2i z5@yyG0jWS29IeJ6A_@r%48%p)-QRD$xjqqQ(5}I-7aPMx>rRAEKGAdM(7iU`Wb>&wf&fQ@_f)cr8(qsBS#5}DxmgfP`1*E{ZXR?mCn8|A0) z*}u#0TppktR)x@^N=K=8owNT+v9>}xe3$<$Y+0(6B`9@cIrz$X=d|k6m$^^ms!}MaHV89_z8gC4 z2T`8JkqMnvuq$Y21Ocy432${7^X*&birMTsPz~={yz2Qxz$XbEu?N+08mNDomoN{Fm`>|8oa9m@XD&UCZ*Z=v6Fg5~6m`<- z*+Ob(W8V?}gkM2jJ@MNde9yXl*Xk*tcuw`MeKmTwfB(0fW5MoBy|8K+C`D0@QzhLy z`dKF%yiuRuoCE45Cl)q#Wm+S71|!#pnEb?E(x*jXj%&kYAz^*qw|#jt2C9SSME$&M z=p-ute`aJX9$o}6BpA1oJc)!(`JkOors`igpYY^Ifn2{*0sG^Uf`#=>mE|URRE7M; zxTHY%!7~2{rOA=2t9&9{WLEb7%k3i(FT~uohPatPDiq4bMn!Jz(Xfnyw^z@wU6V2B z@};EhH%u%tzGR9LwN6jhSBmU;++HLY>ll4e zg72Nwj2+cXR_h>OM23A;%>Hg#HMgU^-K0D=Gh5xfYQ6VC>=WscP9SDU?CAUny?tEA zlCUqMb%<-1X9rP{k$SA^Z2{kXb;NJy^u%!UvHnr-&rXfZ14fD(zR*h7_|JVUKPdMzx~%7 zx<1!caS|5MXSp#UMPl4NOW8+L9{!HeFDe7lQH zhl|f^qmrpqY7=Q_M9ZCFZGI@&3M!gDcXt~6m&e8!6AHf+Jib5YLZI*(|9#WhTRWg_ z5L@1*zP*+%as`W@_^@!Xi2DC+rX@7sDQ^9qUDBM6g`TGu(h4S-CAIySM$;*NpBxu# zU4pABbh#_y=jTW2cpxlsxwiE-1g7%SXj1qB=`jrz%yDPV2}~(kHCS^m|BRFiCXjUv zFPCA~?r(Dj+X4Swd?jH1-~nD!Mz)|J+Q&Q<)3&A4z7m{{i!e010b(aI*5@PfOkY#t zQ=rb44l(1(=^#hOZOLQBTsxqp1qyGn*Voswb@9o6k3%YS79S?JZGm@pbkEc#;D1Y1 zl}rm=#5p}ZJ=nWE7r>CQdwAFmjFl`-yLwu`S}W~Y5Rv#5qpnsFW{ajN&Y~<4Xb9nG zy=sBFH!5=j27wlr-hmD~2A4A-UjzYq@f!XtBNkra?t+kE!6{nT$D0C;(9W)|<$$Li zUJd^qotO6=odq;Xupj*Qr!1^NG#R}}_#I@h)xM~>y|C?30fLd0*H)Ilj132}tPhmVXIs_t>$6?hlzOs=(U-Ja`LcD- zEIQ6d>|O!L1|(FXs1|B|01Say`depJ>iw$sF9-cAyXPxqv4z`OTRGC~z+Q8@5+hIR zk?_JW`L@MMz{sm7)#HQ9tLy(1E-!a1j0jTRFY?4HV_mQELRB7uKw!?5NG{aB{m$+& zB4gDZcR+~41$0EbRFstX8;ViFb@93`&lHuE>{n%p01^VJoP)#Uvs!Gn4-1a(kD| zBfU(Qgf&vc)jET!zE%fG+bAF!si@G>7DJmpXTTb;vepsY^VTV@q~g>y-kRsvB`IyO z+0PA&uQNUyOy>JK`$O`{;8gx}kebN!VM0I%>g9~~B6C16fwq$4$ojRVrRMaHJSFLm zUtRw`71#hug~T14Ql$c6S!u9wtyn2}USC2eakV1%c7UK4<3kN9Qu z>CWb+xvT_^qZ2Nc2|A)FD%W~8TB8X@l6JEFs(C3KS4t6deHXx36k&E-u}drvjl^y2 zQ4iM$VSh&cVLL?qSD!UB{rjW8V;38B(>@ca;-euq)+sH|@-Z4w8?8uD={3nx56ZJq zPd}}xOTv(t#aP?+XL*nA=i*G>ztNumF2Kpj>qQ=wiS6ws=lu%Hb5~B2j&d3BLE1d0ZtH3c~u;2`*L% zwfDOtKf$Yd#=2*1#rR+TyP*JRI0ebX5Jq7c<4W$Wij9HEUYIM2&B5udgpx?^5Zq#F zvidA428p&HcQl+LSWQGMRFvPMfi2FN zC6GL^4v8^cl?zdwnf|TPK=!(Qs|hKZ(YkD+!LfGa%IiU83B^pj_;>eV)uQoZ5ZaS- zbxaX?m%swWM3XI;evU&0@4X>PKd#R4e-6;tZr zAgXlyv@F`~3t0OPec(lb1-q!HXv&|7!RS_CSj*rZv_oxNp3{qkyxU9;e-wf6p zF{FnwI;bR=`S!j{G2$el#=2fr^=ogbJHh-@Shnz1BCg30q#mO|U#s9f|NL(?U&k^z zey!WX!$>ff1wld@dm?Xk0}$EHgY9+S7O*acS&T>l4|?hE-@i=@%3mOklUi+dZROH^ zxnkMG#PMqRW&(P>)4h7=I^-U}5?QZEdmwBNvzy)Z03u z*yARHw~|%BOySmm@nuOhIF}$@H|#|Y7EcyQrf%}OQ^V)*Z#t8VpKV zWbs~&l7@eoSn(mJ{Dng<*~mrnJeKm4rYnaM|3h{MCDk-?|3{m|fg`OihQ&JYjs6}l zIKH2jmiF=;R(-kUGX{osU>*<4=@^*p>+753Dx{>QCQQs8hjqpjL_&$MBrX(FuLL*C z(vDtmRxBP8C;Xixro!$pLZ{ILX)LJ`F#AN;W>U4h8hOvwRl*TN9wZM+)x2Ye6Z+PjHUlm~-2Ak-&FL!N&%qZ97(NZb)zN!wy$!j;jVby}X+By+kG_ZBliI5DNv|#{u1ZTwlewdPK+N@qye@nWD}y~I5Z)S<#re9q!$7(`xSj-2a@1et;-%M zSqVRKhRJvc_s?x+a&T=n5;CuDtDtxIbJ!2WT)p4Z)y(5XIw62wbvyoHTmvj^oSHZ| zI8NKOt9C0)V9V5*9zF&7Yv0jP<-*@vg+ih2?f6W4ageDb_~4k!RCv|0f|KKS^=5X; zO;_#~ScSqtmd!w+yA9%M8^|!K+?JOcY72Wp@A2tlM&IrgVCMx1nv1@AnE*Ol)R`Oh1=zh4Z;ah}6e3f2mrj^9s4z z{ytVMd#t^Sf?O7vP26xHIuF|0;lXPe3p6Cab5*|s*H?RWJPq(MZ{P|w7);`wtSKEY zbOK~gV8}?9^5g0=;7`BzzVYZ|)R)jMeAr83FbcMmc1+us`{@WNPrSXwcj3r1)cQrX zrK^Pn!g>LlhIG$!R$ad9M9b}uTn`PZ4n5|K1rei7FTKjvXVTi`eAVO9*-ERkCH50Z zuA6#&hDhzk15=XYU>)CCkppSzaFcA3CIl6y?Mlm#XOm+B3h6=`0+aUHCq5ySQocLa zpC{W2R94Nw&S+zO{-omJ7QV2^3>6@ryU2&h5e|Fli`$nYgo})P@f=UU_ZP-8s4e%v zh~Tq*3DkzZ-1BbOW9uT3X--aXM*#^7$C(tOh1^2T`B|T4TD1^Ly_H_#P#?oFlSahprSg&)Bei+w^OqI`E<83~D!$x|?^R!piBcb7pZzlI^p-^&*$-8Glf zPpr#FIoTH-g1)vtB@==+-a93e(LgMVgz6yb(j{IEGku+6&3KJIAm-ujf_ts1zgAZv zlbM$~;NSa(s@Jn(1Lkn3kiPC`i7E9EZKRcw!w!Au79*HM_r^tm6%`e!H@Wc+@~#qZ z&NnyDS^hMgnW?oU0{}Pp(u5d|NASi4aPMQ_j#oXclbvYcLs@yx8yZ|dPJs-_uT!)j zwknPPX>V>UcU3+Zbu!l7b8Ta18=kZb33lTuAfnD={I&RLoQ8Bfxjv5m0@uizW;~hP z?QPunwHv3gSahfa!;VXEcV4?3CZE{up~`p)1LZIl;+z!56C#2yPi&eE6K}Lo(#QiX z1wM|!SUjOrdVD1oPv=y+S|aagW6h&2kMc0l`TskZ?(l<2r-d_d3|O=Up!gYYSLj(| z^7&O`y$r(Q?cm_+QQ+bB?6gx67?R**A*&TB<4&l8dg%l20wY)Z14zV1klZCphQU9o zXBjKyBn=;>wpn4HK~R(ht+Rh91|UBEB48LFgj40`CHN-3V}v#;!` z9ou#)1g#oCet;~R4^g+#9xo@7Qc*n{GbK1>e1>B-Y9@B~t4Q_RUoXN3Q?_ez#W|SP zNIpQ(tE)80yKd1=6ADB|T&R&?H0CvMpQl!FNSaVua98`^;6HpSL=2~_n1C$L2eXHva}0A|7aLETv@o z)#<3FnnZ4u@E}E};ZssKF?~uo8ln(ge`{@N)Krv`X862PxW1{)m$zpcjvw~#EaZsf zO<4!qL^P)xhyEgAP7_l9yMM%cbwCuRXUWxGB343dCMeVEGFzQmDd_IZ?dgPC{M+nW zzM(uGwe`+H%FSqA4cn5}ciL8D6`(RWuFKH)l43S~URcW7UT15yO8Rh+O+^j3@@`YQYup~CWbS$UcY?*9r=li~y^p3*Dqu3ODoVi*{ysBF$ z(%cVh?nKTMH4is1y3dohcgD#EH2(hDD5$e4nH2geVf=W{0D+K262cr00X>(&i)MkNPZTpEoZ_Z?A2MTf|Ae1^xm@_DPvnNcRuyzz13kH4l}SxtMToU-l( zQu~uU`;sc@@!Aq$TdVdJmCS?o=2#3m^t=CVPY7U34gm?#-P6Nbq&ze2>*Eu;1-OW` zl9J)cD$y*@o9mDG{sxgC7mp1nThcFFGiscoZr9JR$fIF5zo(gr<%&Yv*7V-nHf~9{ zwS`HoaRjFPf+V?ce@+TiL|oOQvI~w|#h`NYr$dPS0O7+zu#Rd@R}1|pAc;loUdfDd zlW9CALb^y{snH}?4r#Q23M{GEF#laErs0fMGN&1?(AA;R@`HKj)h9C!(-xh_z&2Oe zztf40n@cS0F@u<4g^XqL2`Z+Gvd=~Nn-3DPllMQ%jttE5Qhwgs&eO5|q+Y6n_OY6S zwJF{=CGD9+>@N1sAtI63Nev2@d*tmgd;CxKR{dPx2_ewntlG+m;%!MNZ|_B8n40c? zbGaKNLD%KCW;Mf^8{bPhbR%*^x(9&|H0qzhjU4SY0|FN?8UOIy^gTa(pVXJ*$urY- z8TP;DY!nCj`A(1ZoVBDbrT%I!MHj~+)oTrRu1ZK<{o-heUJ@ap59WHDkFiou>1%On zjQpA9R03Y)iNnDz*XfhI8^_N_Rt&OoBiF)nyd`){fgxWUlWA^`Pk}f9K#$aIvmK}- zE45Kz)E~BYs9Dr+OUoJtWS={0?f5r;ATErL)L~pgsT5f)i=^$f)Zt8QRG5^}*^k(z z^-;+X6&`XK$Ph;R{Zk);g2Wf$?h-H|@08wwt4enNH_|U3&COO->;7_b=Gj~FPd=3m zN-Ot25vMfxH6GB-4&S2mkPzP@4SQZ~y#B;y6l{H-CK5onq3FC>+GPdS!VS2pqe>38 z9p3~i-K%OcEQ7=X87DE{v1?$MOgg#MV6G@OvuE6|J3$uZ?gl zUzI**j)?FsKY!Bym#P(n|3plRm(XzkUeG59>)H<}^xmM=A(p$uZ2b}lAXTwWSdhd+M`;(DM^b} zQCV)*I@bRjOmu#lcNb%0$DpQN?ayZI%>@@SSS<7M3ZSSbgMo>f?m#;1pbp~F%FfsQ zm+3e6_7W)205H<8fy4kDdR0+5g%1M=7*AnB4g)S+H?|5CfTZFGGMI6<~U& zyKU%|%OL)!MeBXz2ZX!#@Groc7KYiattx&;Fkj7J)_I^8#4UX%w>IMrOu>;2Tzy7T zaaRPfGIW%@BJfTxc-ngWkk&>)){Fi(>d{V+^6E`*QyaF@d+z32s+)a1U4=P^eu3Hl ze!MtLjP4hJy(55BNe?I%3NEtxFqx40P8O8~!;FR}kC1~WJQ$bmFQC@GYn7))2{ z5iJ1b^a3NJDRv#wJqo+;waA1*4|LZaWW4>C1rRkG8iKJg+T2j+CbO0gkVGuO zMdIR5-TmM~pUDUuEEaLtS_^9|wGk0BuX7Elm0eQlfYb%BJ*DRuT7YUbGm{wI-aS2r zM4>`F#EPlIBBkLPoX-yibT13+PlVGrInSz3)ws4>gWFto!sSmWUo3qP60{+8JG=Q& zH7`*V4#a9jaCl5sS{jabS1_9aLDA!+v@_aS~xteT3s2ET}3m?#VHOa$W!t4LxRW z@TMJ=Q}`>@6knBe96TJnPhg90X+nm$)+~Sc=DR1-`sUJjS?-gjbFm83KKsX_4$sbq zZ?qLIT(oL6!F$3FAL5md)e2|!QrwR<0KNJ2ldz{ydV0b~h3tj~Vjx4J)pb8o3FJgY zMN#*K3h68qdVhv5_Z%@^H6YV9{vXOFUjbLtW&NxiADa=V|sG_?sYb)-?FIXy&Sza7iH zk4Z^AvBhd)zc0T&;nuk1&@r2Aw|%7%84z16G;-avJzb6d{JG5T+ulI?xKka$ib2g% z`2cdiJz;x+sq|NAOBaQL3ErvbS^9CoTc+=}#)jRn+HBnnX$KWsjRnMp{8{z{P$mVv zufHOkIxR^EG`OHT82%(YQ7T~T6h57xZg4xtP=Ijizxe~eFuZ_?U9;<-5dc6~8|vV| z1z&q-Kt2O4eZG6J(HgjMz+l-~@}_M)C>5+%Ac$`A&04P-Rtb-H-JY=fxj(&GzN*;` z?rWO+fZbkGhfw~?p~Tr_^I-E4Px-bq5Hj8HzSyI?x3^WAxe-naWq`50*%wDctP>4h z9p6ig2snmvmwq=yoMyA#eHwS!@GPp(*Fhmp9siq`15uK)R zY0hTpK#5RKg|YP8+p#M-*tC~?v?4Y?Koz}k$`7qqGwBvuYO#UigyF=zSXx!)UxAXz zo*q-J3A7Wvow+H7jJ&+O1nD{W*7*IqqQT9lPxuU4RS$QXZ-g!{*TxMXZNVL2u@WpG z8i^oSuor*}a!Z7~!PNC;Z>eMbO(;=1Q)BiR>Kd1OHkbdUEawSn*%-FfoqB48p(}Sc zW;^|dk|9kwJxF4wdKVtrhs&uoXD9Zj$9lm`)3&oc`j-u?<*$i<$&-X3NgXygLj_zQ zr{i0l0C@58#kR6{Xf0U0vL0HPt4o&hKfSn_3JmL_^;vJZ-5!7eSf8B-lTr9$lO5PP z#QIiJ;^T7|MP{Z09zA)0DdO(#PKKMJPxg*By8Ss@b4XlpL`1|Q$7@O00PpB>BIN4Y?THOc?qKB2iccY3xHO+DpuJ$tsZ(2RVoSJr8{KMjBM^P)wlo;rL}^#(gl>793Z>lxXLp_ zf|;vXZB{Zq;ur&RQr4S4qc@BUF}~ALMCHi3p)~6_O~q6NXIrQCm2KS4la=}5LIL?d zyw@-Uu@J0-;h>M3{#bbJ{1Zpu`gA!`W+%HtvQ{Gq^^$!^;4U*W^N4go3^Wm&R#2Nr zU7PwK)U>fuC6PG+8x$8F4nWX6GPyViE(*(;okAn(?;X+|F1R~9o- zj;RVxU5>0(8m5Ha?UyeYgq-k&pYZtz#C>ZlUCI_ZNtvrerKUF<0%MESD~ms)Irw%6 zy-SA)y6W7FCXr&1DAh|BQ7Y+k!|$}Gc>#2P8!97(9wvfvHf~h;>t+vR6cq3Fs{*Z+ zI%0-DG}u|yrb&~f)CsbX;H_AtWgYV1!Gj83od(9$duI6b!mn@mht?C9{kX(PxB9~B zj{IJ;%1pe?c!QWFWvBiz+P?!Opp;SRhGlF^^?{m~G_sy2dK)2aYm5 zrc{B5=jr18{$zfvX_Mzpt%8$%NWOG)>3o%M&C9A|>=0U^W-Q68#QjL$} zoXPvF_NC9PlPz?%fBMHP0!r)?dgyrY6!UOd2f)br7UUS~vjpi@!GelUz9Y{wuIQcsxK^eOpyapVsK1;@o(&OO&&mgie_`GqL zS$(JW`1<&2GWu1b-Xt5l^<(%nyM_Cr69001dPAz-B4fN<>;aFW(X zV`>I~>iop=#%Hfi-M8s{Js+RL1OtGnW&%SG;DVg*6fW_c5k=?dm=>%we{@gYF_`#V z1)XTTj(z?y_7?j`GliT6XitS2dQM6*46r4%6hN~cPqm-lV zVPn>+G`dBz1-9tjL(`mvF4JYJok*cIf;HnJq9M3W$V&`a;&NeC9PN9@RcyOiMA47n zO~pRvNOi+fVn6uevgrj4fhb^dU!3ejy!O0OsqWZY?lJ&+s2UJg6uL@5PF~^hzG3g4 z%-$mO%y-3Pj2^S{G7qSeizBr@oPs1 z5;l#Tc@IWI*H`3N;=$*G@7!N7)HP2u^k7ZfMwRV0oX4xw*q`xTy3yho(bd=2Vy!It zo2dviz0XeZcI-?J4x%fKO0*3T5=taWS^e|yUm@!2T*tcX>T8z8o}PV_jHVnhE4T#- zIb>jBsS849E7fc$=KM+=FG8j&pj*jK6 z1L-wEc=tSkI2a6UiV!)$Bjv4~H%1;L-#v6k*WV~Loj+C}!3B9Pr5{5}#_rnOsi(0z zr2l!XyKxonHKMPkI=HE8jnv z_q>s0|1SB_={G$qRJMF#YdQyg_&uKxK=ud`EACjJ3C>9)QVxHFz$ZtG)3*6RaqHeR zT=}?iXF)bqrjY{SRJZ*e+=w=!Cn3-&M0ckTC65v=*J^C=bR0Rb(Jw18@t8G&&Ro&c zVCmmC6Q2|@6k4v&-whdj5;8G}AIwf6LdKN)ry>r|K`hswMr>aT)fx4xqTBt2VZb*v zI#Sm~TN+kVTsiUwmE^u9{^aIrOiU!w^*A%S1S~L6Ous9jFNUv=v2=LQktIw5y{TIr zz*N{DQwYc}!X3-XnmkC&A+_8bVElk`QF_zNOe@{I>~#J+2mYbVUyr_C189(k4BL4l z#>_EWhOmv9NO5}mP9-^&jYnY_&Fj4bx|yrhdlJWn*uLr%2&KowIevpbj32eN+77k! zdk7A9^tCz7s?q_GfP}ofxGo7}llf18VLE0s=y$2v+D|iXf4OhDKGHVG-agU|}fTJ9)1b$?7yHaZR!w zc(>g)5Gxx*CYl|^<}M(Rd1!3BdXAUi{8FgJ9Y}dF=<&e>f{K`xS6lbS6yZzTS1?XM z`LQGMrKt!r{G|6xN zot?M~ggw;RW%=FL7((hVEmdzYZMAV`I(}rvm z&No2u?h7&j5KT=@I}3Sg>%RWzLsc;{O2El&pN%&FVQn=#O&1kv^78Tqo^UXfIHo1J z_krE)x3Ng`UByTFlb!P<5?{gE&-=^2mL6WgOTbKp0{e&|jOHo<#N5YQveizkGhJ#} zJOl`5Sg&-n(>8uBnYmh8)FHw3dBcmy-EA#Y(jOOq{em0B!iyOeYh^tty7u1v{e#Hz z`gw0#BjL{c-cJwA()NK^CfyE+&v4fru zH@Qm=x-blEGK7|Hq>3qtVKa+Gfk@Lur$((3Pm}7 zBZio#wjRG(K;4EUTns+S6;hF+<5Kyr0@RYTdK~CC%=WHr#y1$x3{z@qco-elfIKLC zVIe$utiE#Q$=-El_4TL)y*8Dw4eB?V!D0K7wwFO=e%C^aEKLfyslPk2(6&h@+n(Tc zikXU+J;4aVz5<&Af(%N^yktD*5RPWg5FDa|*e4zZo-Mp{=>ocxPubeOiD|Pr|H^D{7 ze*jzRrLYvlI$)s6I?humSM49992P|!Gde;Vrk?Ay{Bw7hAa(W|{=gZ8R@`(>=dd#u z21L(f?CjWbblCO)KZ6^=wrA_B`=)sEwy+~t#0i9VzB5)#KQUgPI-cpKV{Dtah>rcK zWaImJ#>n%^_b!K-eCc^b0K>LEGJJXm1}$}$V~-ULCx1ta?^-|S-;>@G;3m4z^?dZZ zm(UJGxu0I_G?U>j!3%D7n zR&8i-C(ky>`@Pp8Z~fC1M1@t-jOD8pTb}tj*ciLG@KNd}T1C+u0QsGT@@1?=P-Q{8 z9TuhrWLQa#S1cs?9gUXdiIHmjZ){TD+WM-CgqnXkRZy<;4#qRsa@u(+%mTSvnV#yK zPb(P##??}%=@Iw!HTSg%^zzWwwj;F3THS72U4Ww?RVO%gKP52hZTLQ9KWHaxVMOlz z*z`&sgSZHd%vdMz&N<|FE)9TVS0&pb=^WhM_H?9dWf=26eqnoL^>#`*#A@l*dzkB7 z7E{W`!Bg4l94TN$){G9wQ_6WU^&~Bja}&UJc0aZ|8d(Jdh=Fo&ZKjr8Ku~aTd%Gwo zB`YhdduV84DU+ItDy^X4S5Jz1sFXQP5Sk#uuyB}?^7;&<9OZ7&320g$)Hq7c>|_7huWog z#0+zjkS;oGA0G2tHXgh-5fvQRMt*EL-=s5LWw_P=I+E46hOueuYLB|IvNE*b-G@Cm zOKnkF0GxF>zpy;ZF8KHD>slOF98+C^FN@SWT!hqh4B@dkXX&N&w%?kTma(jftcz!k z(Sj#Gy1uw-TdQ#~R9HsMa(kJgZDErVQ+TUc`B@rp86}h;d5g$bcKbJ^8!XK6K0YJ( zt(ltzc4kdOZMd&X&lN(sDemV&8>^{oaP>M1>&sTpU{}kacU84|(XArg3tEmeyuWV) ztP={@LTW{T_D@vMby`Hyw&|ggUHk2%1GX6Yfms~~hYlPysJ^EZ*L*gvnwnZDiJ&I` zpzw_wJaHfa`5d6#@U>`wFh9?hPpa7kkv4{uw^ysr{T3wrgpq4@MZmZ9*y-i{&xa)g z>x5KpgFmeA@RuJDYA+z;tjhTfEv9`%1K}sA0q9T18{B4mUM#HR>6KbIb+xC-puoV6 zavw<{z%@K1HdWq4!=-Q~S*3yo~~_1eRBWB z67X2~r(jQCKBfoNMLku>8-cddkHm=vU>Tdl1<>Pp08N<2_v!K8TEai$h>-V7oV4}z zcZ|Qo+mA1Q#=j=Dum!*W9#HQkJG6IpzS99SkU^1I0Kc)st{nidpBM0IX)!gO>cjJb zr&z$UDWa{dZ5CiRd9k78Rh81**1FUQUzNaa7u}M^{c+hzY-^Q0wdoBq$$TI6$^Jx? z0oaQKdC>x739-Ksa361Y&pp59s=Y7GB&FXhomWI#y;9P-_i$Ou&gv1+aqB5T?u}wX zsUH1PEI7l-ClFfq0@4pY@^c3mRo?C#9v}M$TEp$BZ;zghr!pL&1iHq==H}*}<7rt@ z^YEx+D^lSC8>2TNga(z0>6xtjNH#_>BOKS6 zrY4OJV5aXZ@FltqqSVP~} z{=_YJOR!b+`5AKh%m|p<25KwYj_mgihPP_VU;yd)CluQJ3|;E$Q&3Sh6I>(N2Ig?N z^3RN0y}fN()5rQFrJMIA?B1$j3c12^q-;jN1-pfBkLBYXZ_kEY&Q34Rdn)8(D&yk9 zfaJW}c;~sVrm${4UONJK!2MVtcUSn2GZ?@r8GH3Qv*G(XJK2ert)T?$!;xauA?F@U zJE?1nq!w7eYZmt!XUF0}ljC zxN{Cp;}XCaTRt0C&9G5$bK`+kw3$moK>_38;yuEYMuXSZdxh4F>SDvF)|DK^1hP-1r9e@)sLZsbS)2-02&}!UpzBCur2fP}b z2pqNpskoTX^6^4Np`$VULZSNmkTrw54u*p>{3w$Tyg>ls8}KG1?3${_s8w(8ss^<0 z6sn>GNRKE*H5=dOdOsHmr)>W?2zUCydPZRRt?{1gTJyn5{Kf(48sDwC@yVwaNu=T7 zVFd}q)YDTZ$h~_GddiH`b(G=JE8?I5MKWxI{3#d1>3@uB+tk8 zmfudW3DG@dKkqcfm43@_R}w?+*8mvJ?$c#%JHZ<^CqMt{PhRaITX=EP(t93oF*84y+EbcaP}*4iY)uYILZ1RTtKDm@;giOrIVt0l2^?i z0{}d&w6cbYTw6+q;duwAWC4$u;ql?2U%OjgLBVMIcVKx1?3&4*o@$YH5E#GkX%cv> z$y1zv5rF)M1IW6T*T;mN?(%;nQykKM&Yo9vwA$8tJ|c2MHL0mZr!f99bxW=#{vsyS zdTo8s{hbMm<N#3LN`+~6 zVB=Ju=NHv|={YHdN3KkR`MVMvjIk#4^W*h^l#@XpEj=q=Z2tV64`dl#WWduFxj!I4 zeC8mu-EbgBtcD)HGFAZ;3l9GJ@camslSl=mO$Ku-jd`FmKtLuYidOzR< zk(rp@3NIx#e>MgOaVpjezbeUb&N<aKaY<6oob9Pb%Li4bEcZg@XM2AaK?P^v0~W$tEtwmBGxE z`fFoqufqVwFd(Rfb@+*XKl4ilUeL|w*jPU}ecX`HCc^N(!~GgqVOl^U!^5DP9vd6P zWaXElG|VyRc)--5A&PmpvNHJ}gM4h(D#h7svt3_^YkS>z>VEBZfL~jUVUJWprG2W# z9(!*W>QDrsUoG~_SC^xWT1P~Si;L>OQ223$()|b4ru$D*Wn8?m9g|<7h7t1D`b=1{ zN4c@!4%`O|1KUVU4ych+AuXUrJ}SL)6#QyYc-yI@XDR2mOMQ|vSDmgM&X>^}XOv{J z#j%H3MMx$h=$Q|%fG`pq;wN#?jEDd#-%NGqUNRJ?Uqh+INP2S>rZM610%Pz|r&2`! zgRbMOy{4xQ4cBn3t+45a|fSh|~;)n2ol0 z->tM2opeLgN7A&^r&9X-`T^d1ci8T)h^f;j`&O{@^l`SFy*ZlZDTVQ_nE4)(+ZQVQ zCFx>TRxVTC718qC*wZN0itCw#DBlV0oz9dbw{jB^Wwvmli<6exQXR4%K^h4!^+25d zl*qO;Q!V5N(YXGB4mWJ$gIanymY}wZer%l>R+dCFpjAaor``-K9e@DdR$yUOucn^{ zf8Le^w6&-4V!M9o!IEX|{Xx28V`=G8Dm99Yow$U6Y;_lDAIfLCMd!tbV8>xh_Br{3 zG>c#l-&C0!=0{5^di3;`0?N0Wc@`W4;vYYL)YXd)gGll&yH&~}s*n_@jLgFoJyR`2 zCJ(7Qvdk=UiYl;_kNy`#(;tg6b8<48ZKQkpd&?$=v5uh(o_`l8J}&RF3}*?C;_#!@ zf}VBM%AjfyA~HTz?Jos_Cri~U9Odk%xy%a_hrh%7mg~Q^Wf`$D3_P0gmXq>dJ=&O9 zPB1I-c2c2oKNpG?y1{5_Ja46^!)4N)qZ6@vCiOj>)N_JMIe%DLkuv2HjO&%B`%jSZ zs-+Meos2E7p--hH+En`HhAX9&OBXx4<@%_bhGm{0MMhFc$80c7v`ktL4toD6Vb z)Ic;3O>`NsK(q^*bNUhJqWa%6CBr&1ABQLeVV?u2p78=7m6zhB`Rq_$`@-wWcX7v`H6X8Oi=%yJ`6I8V(Rs>D5Wah8`ppb%aPA#| z@6~+;7JE3b@#@n2iNl;MqQy%+@JXnt$*&nh72U+%ZF0UbiS=(I?M`>!Sjb3<8f&iC zwI}Kt{Bc_Fv3XN#c_sK#7S+bZ{OzPQBjcCKg&78ZpKX8CHW zmr*r8$|Ip#m?p(CyNwvZ+)2z9V^#YR@eBLwyoH3me|I^ad)`a!GMNp{Qq`Zg%SZ2; zRI9qjj)bXImkRu(y{oAG9Ekz9t4I6p-o9v^>H4OD{^DrJkTl7;CsH~$QPPiv0K$B#`w6DLXp{SVN78?`e z3r3;W{Ov}@AUirCKcfASktt}T@_`|hY28l+%pRZZHo6?WQz!o%*B%|GA5@x9<~V`z zRHyno1?Rzr+<%&VLl1W{cK@yRr|W*#7(y2Y^GqzJ%L{EuAB9I zq!1v5|BtAv45(`Bwxo17(jr|7(s2+4i<0gZq*FQ$NOyxYNVkAUcS=i1gLHRDy>-C* zz5CA}SI^#itr=sEG3INs{)6)6nuP^jFxc*hlDHebu;Kxt3MBt8$M(`zsc%`?XeCt! zUU!=$Js?zxyldTK$v8+K0+RRh;S8^m`DAmSNhbilj#qh^O;ZTuIsxI zy&VRmcS)h5#{p!F)os1sEvC85Rd;?#m_4b3VYk31K`kbs#i%XQ=L=b1p=-ddIiSA( zzutIK5ZaerXJbChJNwjC)HSuESj_Hc@4?*Z_4$o1df3*7!jCY5U>y8G!ng>&%<;k# zi2zu1ew7~QMSGkS?aTf&WoD-GHRhiQ0!T3~N$scjqPwVaja&8u&a~D6=qzBw$*FSp z%TESNU;Ew#$>k)QH+TS*mf72w?bTv-eM&a9y^hQl@($LLFgO>kHu7J#*mu1JKz&i6 zrm_8w6SF#7o#>=w!2uh&n*&ujv^a`QviwDl1qzbhNpw0W>#rHH)C_}8W50Z#FrOj!D7~v*iO3Mi)@L;J2mNhqw@@ zqcAt75t0kt_sf%*F!4TQcz0@_ulE=q82a?-KN%%BMBfLz#mTsjqA&XL_Hk#pf8b=g zH(2T^oPZ%q?MC1NTp$mI9I zVw$z2qQnzT>mT_8hN^2J7NiL27aGA{<1RtYV7=pI0*l=4U7$QQ(agR?);dAV-qIruuTuJ@U8IO^X#!u{hSTWH4vM*dXnRjqbIe`r` ze!wK`lku9FMVnXyqS?$Ymc`xcg+m*~&r|X<9aL*z2#i*JmClaJpU-Ja=lvya=|ycz z(xL7@G3+Voa*1r~VvU&tMzE(T%Z$+F=D@SuNG)|g#@o4|y4xAAM=u|rSc(8Fm^hjH z0f_s)!O_xgFN&%qQ(^rxyepkGIB|3USL^5r>^|Q*Q)X|V<0Q!sZW`|V0h}r-Cqx9C zYoEBHTx|U(b{oYnQ^y>dVpik+4>u+xAu)s5q|j@JAXUY^5Fh#QkrLj;=&~h4RZ|45{NC<4Tjnr*k-gQkC>l4i(5hy;4MGZEhnP zNgt9;Wga*a+<~!P&IqvCYA~1cr8GKaPQ|rZ4+%zDsAt*f?E~F=0b= z-o;dk24oXz^z_wv)YLmF@JqlFj`hR@OpASRQuTpdEzrWDe4UlUi+6lY^-RAVH5u58 z!??R0QB*uUxsOugU%aY1WR&ZBw;y)SYbOh%8T0|4}V;oe{V{^fd$TlYL+}qdOQJo;CPru(iTW zF`BV}gR=|js4adoU;!LSnqeh`>&oKzoFD>|)?trm9AgfURE`0K~}w zgshg8nY&@j!pd3|vJv&{8UVr)w>Q@ZOTnD_K(Yy}ti;gZOMl>Mh3gZT$;~3)T^v~0 zOhHpzj@R95o8MB#a~XzwP9D>6+%*g}J?U_2X_RoXn)ZlE3KSxtUI>=*JCT~#t$mvs z@)*YVjDVP-fiVsW;6t^7?H1+N<#l+jZB4kgf7KYDSuzX8c==G*Lodq6EQd00J7GAp|Q~Eh#jXQr`$CtgoBc z@6M2c=fovJ!xI52xX84$@R%4v`a5X%U7akOWG--j0diU*P&S9#uql%>(f}P#_Bq7w zAN>jzR>%-ssntIpW>s;7%208vUvGBwtIy-b5}j{2O!uq1yYLJB$+;1+emO|WFZASr zYen-nZrBm4jPP=sxCBl*wBzX*#B9ybY8~;-aExK28ABv2A?Xgs5-f2#%iS6cK034h zH-^QRZ+#^M!oVa8<}BcTdHSqG*ZECtIY+EsM$7-~!CZhj!ig~3919D}2pccf(i+%V zW;5@8wpdg#F!FgYT3S_J9tsBbQg%rsjwl~t`y!+V`6_jk1}q0gO`t==)*@sD6n zCxDRn*74EGwNkBvgQp2Xz5bV*>`#kO!}=N{&NoW+KWdFCdpB(MN?m{dzI@)2&+6@L_AngYhenZ99 z9gA7Dr~I8Z;|IET?bjx7MNy9r@SWCiKZ#x4q-oQ(=^#?ymd*WOs0@0R4&XIV_2=qn zunU<)Gd(aJ7{hQe$fky@h8Qgt|L(UIb#Gk&ewhW{vtMry zj%MQj`N(&2{hrG9KZemeKz433XR|c|K3r4$xyiOcA3A=Mx4e&{X#nN`KR|0kOPn1T z82Cl#^Go8*W*__a)TjuO?KEb;wf}I~O)!9-ctrP}EScVbwq0XI`$oamEPinY4N`1ba_7@v?tvCtA0E# zn6o4l5nQdtaqJ1bT^xRPq1*FzSYTC}Jimx#zHGSS<|=NYY%J~S?7v{ppb&>Dq` zUe}NkUQz9T-X#!Mg1m0_&eNpTjkv{Me3%EaX@x?doYz#lt^c9Nc=(fbQ@gc{L)>Y5 z6h7ocEgVx8R`hsX6HJ!Q%tvDcg*>#8N86DJ2U=`A)H@u02oAXIV0o2~e(ag7Hu_1- zj=c^R(~fNFb!2O|%QE9X^vnK;1pF>ue{!wQ@x1x|D36WBC%D@6*8`Ier9R!w&y?(v zoZ}l^s~UTV@6!+>{-paLACpj&N*%Y)R$ZNQY2?$`nf1_q*ZtX+m?>A8(V3JK>#k<) zA77<-1ReDK&Ro2lykVaf7?}bP%f60Z$;k6ezC86`he{5Mhc=?cje_>;a?)_IG!uL? z)`$aOXC85QVp38t80_lQ%`Y$vFSQ6xhFXfx?eG3Ze(4%Mwq3u;e>l!a>trq)92F1} z^4ZXdaBJq>sA!@^KRv9G=-;XT=SwAt_Vu=mQc6@US3_&2jR23UK4~{O!-lB|H@x__ zH=?tzpF-yZd<(pjemmTY3zvt3<#%Nv{Q}@0ANQ2L&aai0-m)Z@R@~HLomC!o50yc`fwBG0lJ(`X-`!nfN99qw69z1G1k+}&f|9^y_$A4+AF05r za0#p_Y?BqcRxAytNBAP@E|}`F$qYDz6Lq^LjP1z4(K4+IEO5^Y)W1hRVIE`1b;QuO zMGqct=>Rdv3%35jL8&)y(6K4_yk>C4I#H%ugF;k%mg%T|No3vS@HO5t+_{5t%mxE z&(;mkamMSOyvYofny?&uT*o!Tldr)~MLvZySHgJBWc?5X)V4MOHfkY-tz4zCVUd(=?V=IX6?~QI=&9)ziJsiYIKV`g z^Fkm@hde?)pOS`Tf&)eEhn~JMHQ4@Y{*wW>1u}5Bn8@q!Y{# zcyxPRDDWtB^W=DQ@;MsOUC*eOvqh03d=&9dJI|(+{K)_(H_}7Z4*2@QH z1t?N05xXCM9BB_SlL?2r-`$83gukmB5Pi+3^{TWWOI~$CA6Xt_N6R;J!Qbd zszqUbkVAS%r90S!#>3#24W2|772(kFxxI{%w-UjVwWCVQT7p$FZX{%?%Qy^f^y4Vf8I7IFEGY& zwnE=_Rl?S*XA_;dxh%~J=q>phQ7GW_o`hQX7aw63>=aPT|!Z$yx9h#%%j)eyqN_?`_`wPP<4z3yNgqCdJ0;MrNnCW%? z_d=`gg*D(XRH+2iAW#bI!ROFb;<96VuRk=m1=9fu?uS;n1>b8V?~ViEPz7XQ1L?Ea zVZ&y@_^e>Y+Qb=?1M8r6D|cNujS8k`srK_1pU#L3{1zp}1~QuPo?sIzSp|SnF^xg#45s??y8B|(W z7*N@_;AtVUuy+T`W7^sS7tU4RfG6!3?TJ7x;3O5;;p5{gPS2!6?9hJ5?iSbDPlchz z7^+~g%I_>cfITueLTCt{gEv&p#$#xqwVd^c7tjGGr%QPAK>6Ei1Gb!x&fjFYZn%Gt zu)X7-HzrtjW&!^j^b?893xkMmbhEmbHVT)E+Gc zj#~A6GPs6Czut?Y5_@gHOf$@+mwx^3-DuHaEJ8nv?!zVcBwfB85tg8z5#MMv>-vfA z&zF}dQl{8^=;uA0LXH*pTivfS_#ufJr%RfW-%ZmRYRFb~gy4SpAS#~ReI5vW8^QVV zxj>B4&+dgqMMCHerIb*H#nOQy9I)yq9nq1_d=WZkv?@Y=#{<-Q)%eK%R=mXXX@@&>VbL} zQ{@)W5vn}Al?RtO23a{uzEr%Ka9F-d16g273v4p>0snofD~S>ZX%y&1UDnE6?A5JQ7-ts#*mNBIh`#%HC2Zj1 zZ7wY#C@3iGr~zDU%0Y`&=+m~G(!BjI>4KeFUc-e-7d^Pj+6NO!8jug(d}n17b2&V9 zxcM4`rqwmi@2{Up2*mkgBK{Ryz#EkmkJSr2)u^B4WZ*CMW*Q`U&wmD=@vN~sTim}FIOx7qqtw)Qy8_10+Gm9G$m#`~Men5d zclR3;#u`|c_vnf5Pg?WpI?^_7+k9a>N(YJ2EX7WNicQM>kD!H}%e)v@k_3j`4f-$O7c zfW%WFC2~j4CtLPRqUPgRnnt63>R?9}%A@v+&7j!vw*ceXx+U`2iN@OOt2+F@U4 z=)q>XHWyLp{ZAjVhfficW_*{B{>G5Upc50eZp494;A{=(Z!-&$fW5t`#)PDzA|Vy< z@Ea?EIZ;;iW1&~Zce#w-S((cTmg-B~upW z!$m$Np6cDDTW9u3d-v(<<3;qlP=*RUi7Y>7)lNCNUQC_E)wOy{efM*RqP+&-@+vC0 z!Q?x);pK-944q+=!9W4J2z*Rniz>NXj!l$^&jJY;SuO#;QMCuutiY@tLCUY{e_!8TU>)Os!V<^SqKCU5RUvQa&FbF>U9#4#K$o#fx-BMF~|v$*c?w)cS@aUMlNV?%Z%g zGMw6VpFh{n5lH5zKL5zRRKg)A*bb=*KUUr}G#+%Ov)fA;&@M3e?|MZbpx`SUbmSkg zgy1r|3iHvtQ|h|EHBe+xMi*@yOS#wyO@Gd?LRWuX%H*fvX{#Kor zLy{+2(q$%>E`ao|sBe+mplox5WPavz(6<7+u!Fs=N;K$U5c-wz?dunY zt9esLuLfx_kFAyd%Noa=azs1Bw^bnJSF*xyag^+*W>6h^ELLGLak87K47PN>hxK<} z9?0)`{r;S5z4iK5uwa}Ri09#wBc}gcD`sIWN#auXYpK77RbQd}T`%a5cfP+)wjCh- zUStqUboU2mWpF{*{aGCnMi5+s6R}xB(l;uTX(|(F0|Nu9^-kqcsFfuKsl0JHbN5%cr#Cu^FH#RpE_6B z7QG!4d@MZ?Aunw)?z24?1vI1~-rBHrjfs$G%T z8tabA0$>-3J30N(rq%^~K2k7?)-LS{2Jdd?Er4?;)yd|C)YNou9Io{Ya*qBuUAjQN z$HB6*5Z;39%%?-i@_lKkaK}|~E z+G*x_%nDKA)WM~Xh|Qt8PWJAYxCq&m;zs3O=;2}SF!dmLsCQDNRy5TI=L?TyaXOv} z^N|jxr^rl0mdK~=jDI}+$8?G&1Jl>KCa}Efo<&ntP4mTJfu3e68xApj=g`6+D@mcb zw+l@a2TP(ax301}l5DZT7~3z@$Py+FwoaAy{4w8Crn#G7nmKr$akG%baOkqrUsaje0@PzP%a#yoit_I-dWl@}SEYV?g$vZ>KrzY~5FFeFh#>nz z5;!ftYfNp=!8V5?ohB-v7~`9nosF$|K2fCRlSy}ib4%?0bS?zJ1$X`MN}9{AYFInJ zJ{>7{?J93}^}d9xw1QSr;`NrDMO6xq`A{?splnfZXOB8ht1h;gUACzk?*7-vj}ivp z+j)rAUemnC7)$fCIymdJL#&+;PM*HM+UN)>SYFCj(dFwWH2w{o>XoQqkONz`Vgs0s ztpg=2!CD?3qA)8GAPHh>8G=zb0kR=@#}|0)l#>dg^3$_NjTqZyoo)WE&xp?)+jh%;!}h z-c>U0LTEh7DQ-_Vzr~x1CbW8ZSUc|mmT17KU%PZ+5|_UE-cF-nn@NG>=|N&On?}Xqb}DaRK(fx|FoCIlG8Ln8U>* zhlIG0f+0|Ek5l`XkDX+p;$6*>?tpy!H16lf2);tDj1sW`3Q#DS9f18qxPKNx>1&TxXl~+j~~Az8)G6s z?&|DhQD{(exumu1HX%$>ek>g;sDR*ppn1#nKn&|B4L$7(2!L$}p}X+v6WqV2rdaHt zB^uQ-WDK`pe5iN4KFozD(arN%P>>vGa!R61n0a^zOB+tVd?3+b9%4W?mk=3Rna@(P zh^9`j2Msnt{Hrte_NB1Sk`PysIGW;TdgWw3%%2> zXOa(`N*6k3yw~o4BnS!~KgjJcyCo3wSki37&KD4kYO9Mq(_X$v(*syA!8OP}+cbM3 z)Sx_FcS3N%@ls~A1&aXy{1#*Na~Y)?#+L-#Rqyf^&A6)omy?f8^ zdE?g2Y^++sVf|s3lb_RNuP`VPZCqFcu1oUo$HLhnwu99OQ7$~F{J&%_47EKty9N`V z&mPx*nH3sh1DR*GC0tw22FOs8&Zt^%E=Bvcx0Sd8lg9OjDWhXJS(SwT6}}eF(^>F; z_w{d#=0`u3$^#m@5wn*bc?KC^H6K^`Q>nAT-S)+DjH>*j!;2n`5B6e-8b#{R_YM<+sRzCP%*T!b zy(v(o*A%mZ@U;F$=H_{D-5^*MELYd#T4`O6SPSL^iiGkUkO}JGk^gjrfdWaclt;H&Xj>}E2{9Ud^pE0#BeemMoS4_s6aLOW{Ex#HxOi{mo)Y_7O!UQ z$+LgQ9_{r^{BdM&S@8hR^k{vhn#If6i!ffC`W<@bHnYq*;v zQe?%Xj%7hBzIVhutg3hExtR+Y=IWLGCoHfE%ccGlzNp@9Yx{)R*Kq$=cb;u6;iJf- z%%4SFdMu5fLoQEHjCO~>oU65~tBdQ&#V9O}PbGNGWj=elUIWN_Ghk@~nEy0`s25VK zQJwi=|5rR|v^ow(x?5fCFD<0Z5^rt^ukJcrfg5MerM|dC7FM(+(bfimos(oZxypWrn zM(|+UXISeG6**7Ire!O zG(iCY6vj3x+*=a7KcktDXoPg2n2%7KvjP;5$Kq{Zt%Sn^5u;2$pk*@HnW}pA=1nUg zF~R%{+urpVbUufkGA*1uGf8Od49 z`HyEjKVB+RX`?#dJWEhj`va!hUA?ia58Xn;Y@fG=OIa`cynr*jttwNZ zfiy9lBV0>`yT*zYHG%gy6?~@;{p43b_XH3d*AzP!bq$R_po^0CI}CluhpnuvjMYFC zo(oALitwB~R`;O3*}NtWTDt9NWOKn!`C%;JN9_quynAZ#ea9R$om)X#7PD_$6W3I~LW8+i7;(re9KiQQ^!iTl1yEfyAu6Q^#dbNSjXZ2n8EC&~ zMPMUnI4z8ScrE;7WFpIjXaY-B*Qi5$Qjv!pZT#_9(lQ!qP;$sZ!qhdv4vjZ$4Y_k; z>kGN7J?yi-Jm_AQvwHj_Abp(!VGZ1!)i^%FfMYJqqfn=`oa?s>LVkx#$Go zS|O4MKmeQRF^i9L4!;9`qFvMAx%t^}pGRWEcHK%IBXJF(GIP}wh`jy711O_Uix;)P zZ9Zj#QRP_rDJ+e>eeZJFl52F)kVhs&q$6!yNy6U8lnA?B;|UF%}I z+N*i#-qFWK`YzDhT|wlF0%nfPiM1Jzj4#rS;844mb-`AP&))sygXa0*0}F}wd}$ro zj=YcdZ#zx4e!vRbbPi}R3Fi`!k#T)5Bm6z*KjEO-uoW62J@m2tmFA6&Z&bt(E<*at z*T4A%FWwsNhQAjeOH4}d%G}sbe4?7K<1R`b`MajTK>J6I5Cz+p@o8BHamz{#cma!c z0Ad}Fdn9fj0qc*6=@Tbf@ULAU8GfTvyD}Eej9%P)v z=S=5@W2syJ`EuZOPGR?=J&3J}E-o^OPmyJ0WNNVuD3LLrr*Eq8gW!QAr-jo zb>Y{U#*&e!BmE3d9R)64eKQb5#Y9uOocNX}dLY=y&|P?AdhG(|>jNw^M&s4wpfAM6BaVt2m++v zNDj?7Ta(LO_^L@m{f@6jlg(J3_B_;3rxg>U77!5FX8HN{Lq)3h_)H2P;z7dm1gt3wmW759;2`Or)`rEb|jlee#*ZNvNyb1 zn`(dP`-F_2b`^1_=~U0;=KLccPP-rIZP^(jk~9S9r(YZs4y!~S1&}8uW+;XLcWJiI z$ys%eV$l!84_hB5SuZx}|Bduq)g2d3+e(l-zOW(g+b$IC?!~Dt#j{ipf1Re z;0~NU5Z-vrSjpfll(n_%k&$LPPrPm`vT#pYcz0&skKcxJ%a%$`GFF@YyA-uvRSrg+ z2ey$ZnmXCf4J(RqTVyX*fA&2hx;KJk~+Kaaa>m0eD3)PG*dg(5tX2;%I0atk}Oj9jXiQy9P=sfc2c5A(9WXK2Q()I7h zHjvR`CpNd&CNiQpzTP^+yZ)`$Vm<$8Xc|)H>zID&J#B_87EYhLIC#FyLQ#I|v`3RL zff`kSy>Jh=+p9t~pHTKvRz` zF1V@GIri^W{TQT4iA=Bg*xV_VEJTFPROoMPHUHtW?_rC*(O<0gWJQSZoyb28r&Q+P z29{Conk8eUv=wXGZpPf$GhEa!57Y`y6Y8cHzlg1-9~n#(Y_9kT@BJ~@OwAP0qIfbj zSTSB+1nLWLi^4c@Q{w_%`Sv=(Ch5o8$b&i326C|%M~wF#EuQHpUXoYHe@XILPDzSY z06s@^32*l)?&m8NV7{^rNV3^|WDILqX8A7%2RnYsr;i~myQL#~cw*+9j;+oR*o}h3 zM)nlPe)}_hvERT$g>*`c&%VC2TwJ(D&NbcqpI&jwMjcLkS`h@4rO}@!>PK`P8_xVs z7wukL&8r94Y!~E!4wI-YtN|`QvNoD*2hXnVCBRDuoC~Z1V530TQw$v-FO%rcr-rs~ zTcAWlibpL)no!ghUU%2L!+C)#WsHfu*OunmRL6DZtixC z3P7SSf$eCv17&fV$4C72=*UxkURttN$JZCMk)OYAdX^u0;~&u8U@Ysrv3MA2$O_&Z zi1Xc#!kH{;>P-a)hOSlO4Uf3-FENY(ne2z4bPs?E2qk%#`e z>He&Lfe{)|GcspA8E818jtSyqc&PCHY#XwaGas`o1=iPE^*wgNzN@r&d>~5pIb<9I zZ{IgZAZtqbr+39~(sFh%*YMjPJ@2NHzQcuh5Mbu5ZfS7bbwqOLT$1-$w9rTlk$VG7e<5g}!AR-R zRnuSw)3a=GNRJTjpH^dPrZtn2v2=`m-TPq43;zI;6dFg81Givzi;L7~TMuq4G9oLl#Q2Pnf>TBR%{C8^dpjb| zO4~Sf2L^mzv9rAwDWf>NncMe(&7As7>D!}qC_pkMZfR-h4sZM%&DkeD@;y}FzB6`h zo#Tyk=LPcK(h>=EW@aTJY%dm1mXIRf+I@hi{Ne>#+(f<{FDNQf11}YFWPX5-+Ouk% z&NLDm(^bCp>)C6Uo~IvaV%*-L+nd0XRmMoHq`DoxK7ikRWOVbSN$5lKI_sFI{BKca3VxbKuddr>}22b~H5wpNNuWt^fD^vQE*gLf7r@NYtBo!)`-671o{$ z-%0%FJi+ZH`=RQ-1HBXLUiZnGol?2WVTKl2eUZ`>Ae>$tc)9HCfSbBnODT5ZI6hdKO7d&875;CYme) zw+G+o=)bCCffm+T^b~(bayJZc{~BB-gTMhTkzPU;`Y(u1lk;N=R7ICgzlI<6=pCcn=I2q zgyCi|Po{VdE&QfU*4xnT#cEiixA|1bMvfF*c)=K^>>{sB`?MSRll zj9-%#Z1`>F(U&R0i^t3_ZyMJxfBBSQTNhWAmINU@IkxWa_*pRP80b8=?5}Vih}IJj z)aZtXUFr5~PQKNiHMyx9W7u68TXQ}hmX}D?zL`sSV()>=<0Av3 zVK6>SGXiYo#vH3IPJ+JKYPxu`4m8zL5p`6pDId9BsekrxDoj9ttfjFyh|Z_k^mH>b zkL^0*+BM8>?-l=t0^75}U)bI5w0S;iDEl$)gJtj7B8+!sUtXD4RoY0@ssA-DIY*M? zUrg{oCO^&3ycT8Dkb&~j>hA1$i`S*Lv^-qR+PBU5E{!%%t?c66D!T`Jm?a{2-` z(WF#{oNm?wxl_zzPEH)476%x%f=dd&pdd?w%#_BE!qJD76xq%mGS-M1*_6CjyVd)) zkipVjNq2n8k0mPC4pQbyoUUqO8NoPg7j{*Gcz4_)dsL|KfzgD&)ZoHd-VgOb0|vhu zu@P?6UmaJ26!3Dv{zR}8iTU0OUe3pD`eLYh6ySu z+&vPMO6jSm2w;33dnfqB#~usK(489r*7o}3Cm4>4+ec<{simoD(xS<0$MEYn82Y z2FS>Q-@K9sE%$egts%!q>_1|1+(*Ca$Sh1+1jh>Okqd4sxZ>SSkWG9Lke_VupESSu zY9XXj=&^yA8sT>k8pY{f#>@B57 z%*=uCat%S7jS0zX7rzMB_K%QHGzSk&2%Z?PUF$&;$i|U#-f8jG=kShX zcv|7~FD7w=nA#mk<%Lj-bo_9m$i^R-X=hKY#j$IcZo|XyTDRk2sW+$f>OyMbfFxOV zXoD|4l>1Uv4i72x)wrlb@7^N6pfldzYdY_A<<%*d`&TZD;A@kJ>Ur!+UwT=mTp>-# zhIiVtv)^bXSL_~|&&0yomU>>m8NK|>a^JzKFB?{^=+YqYa!Ccq|7v7g*kP>gB|vl)Y-To%JHkEU?}2*`)3a`YG^j2zcu-> z7`~`5@>C+ECLx&Hr9~!k@!BzFg?P-g;b$oAdd)IbHz)L12|M6o;jqK~S$1?RvP&($ zDeuJ_P_=`>u$-=)z3?ivQ1Ef)6xp1+MKRz-rLJ!jG|fNWi3pIst1qE0=<6bIE*FPJ z;QazRlG!tuT@1`a-_?CQO3ya~qZ@85o$@K_Uo_h`a1+)1^~eicf@g~m>#>Vf) z4-gHa?n&3z8z+f&PaBm9q|N*Au67B}MFgMksX2!~HqP61OsTG$+smkh|Ch;sBQ4n; z9qi%_lR?~SvJ2mBt@?MlBDalDQPe^dcrfBQW;8>y=MS#ml_rQSiZ1>NfZ#Pby zf3jES!A>i>BAV)tX@iBZTP(qR;oNj9cCHIW0s~VCSgvF?6X? zGnwG`tU>%XB1`Us!m~jeRT3D=WsV?}zj;z+!~5v1L%xUQK!Jtj{Y+w-go(&5-@}yy z{cS5}Uewe_^<67tpWC_|p}+P^gHEAn|LD#h2rg(|ot>Ra8U9qQU@^Pt> z{1KBalgyfA&cJY)%>1-|o@501L-JMyR#UJ!IjoiX&N{RVw+_nCgFfy-8rHY7N3Fkp zDQuhwcwC)-zdSZ4L!L_a$}fKW_x2QZ;AV^DMgP3~SnDBra>^yE?I1(OrLkk{WQc@E zf%Om26IS{`PJ8$PLwr6Ph=HM{YoJ&tj4hIh6RW$RM!!ca#5I zjQiV5*Zhg(TkiJX89OcIFYjL~5vuIZcK@yc8sXy8J$z@m-U0RiwHi_(Tf6P2g1P_XRs}H@E7K4? z{KPOV={GBmu#j7d3zEX8daGd9&wTpXkX?_6b=qj%>ocz7XdvearW7iW<{Nr}%gxRG z+221)qdIQmdc1x0&>nsCEzOn*=D^LRN;jvjJyG@R2Hevg9@UE~xBaHotF(;dTZIXY zc0pLED&~k#mkStdw?4y?KM@3R=K|K$QJlD3J#P2zUtOai)N61KV}CeheK&z76B-J z6zA1ZsLX!`gQqa!zpnjRUw*j{>M)~J5rVheJEA3{Itxo3Z+^*?`J#nvd1i(^`W4|1 zkbKQXw?ue3O3@-yPXr+C3UJ!sqj&REHU8wt>S_F${RgGxAJ^|wZP~dCe2wl-fRo_A@uj>u~ z2UP!TcUHCT6MKDqy<*otP9Bq7e6tc_IvN_96Cfy+gXI9~6A3^nF|rjcA!0UM=8fP& z!sS@r(OfLMwD)_S*75>=?8Jb0SMu*J#`;kkVj0Gf;?5t*Yveh;Mj}f|>9^yDo<>be zNK2cZ3RJJ5yex9bss6!M*U$u6$Bg>CN@dp2K51MoN1ob7VPXd5)vD8K!0kXTHF=RfbucXc`D2Wirt&OB;UsP)SqO_2hm-$|Os+w_&eGiULD$@M?Blc6V*cBn zTiA3MlU$h8ADim)c0}7f_C@&POEXcOV6kQm{%oCNDbba^qhlA)uzG_f5^+Z}TG9!Z z%;{3_26hU^>YviqcLb0nKWyH*+sSqq?0GrjR;+W1%tgj$f;v(6h2Pf}V2uM%F- zTWVrlrpqM=jdMDeSWM_1yKNCnBKy-ShkQ?bYJ?sXG@1pczV^{sFLe>D)p@`AJ-W#pfXC2rJXo9uLmd_RSGG#e(gDe15yvw;tD0T% z^@m1N>btC{`5N*7|Bo_N`)>N}(u|#%V&DYN+V7KtmS;(83tyz|XmVFgFdhT_(8}0h zyjtRy=26wPG(L-r8goQggvaKLP*Myz3s3G8@!W37_*ywNsF4ufT}ndt>Ew;moG_Xs z)>Wc~shR(R=Pe3x%1pWEtq08AtZJ`?{O(E8Az1J!5(_}>;%p=nOWu_U_zVEDZI{3i zQM6Nut~LsP54x}P8w6}bL4(P165N|foB$o?R}T6kgI$phOyus3b<{F#2uhdtyoemg z=6D6&`UUH^@Q(4BY;%pj0<`O)X#2IsG$A1&_Ef*rV}DcJ%&8yObHXw}EF>dg`UvI8 z;2NLtBig3q0=0MTd*L2AM(1whnBDXA?qX_F>+{SXKAobv9jhgK>^Rp00>0ByAY;s6 z!Lz-ddzyW^mn5ik6)P%M&4BI@W^?joYz9<1>aUwe2k=4PhBlG(9Pj3wC1ynt3!WvU zr}<_}^!bEJ7qi{*v*YO#Wf%H6z?(A>hi*sPyjkEr3Fj8vGu<@tZ6@hTKO&gODrQhK zfKy!4=z0TtY3vMg>e^xR934xA`xvh{*~Xpcp}QO~@MZN~6QU7aMcw={3XQ8RZ#KXE zUU47B7w09^I8L3g8oM>#6yLc@41LA{_PC4TpE(QI>y5PP?!;g*Dd{7rhcWM1@ z%q~W0p|yTKvi1pu2(h%kmh4`EyJi+R?#eWmFW5L5YtRAe4gai7tDB)lxOL3o5HNQ2fRE zIgmej!-=(RKWowrw+>dHkiv}xGjYN$YWg1-Kmg=@&EwDeiWzZ3?dvtP$s)|m zS`}7pSnERUpuNFgA66kk+)qe6ze7#cN2ed~of%q1T(teec*5qh&jQerg@`}7g zbE6uKoq1I`{TxYX5A(3Affl~3_~1z3>C^YlPg$+5uT!5?eTUWnbvO42E;Tidj?O%9 zQCH4AY-0xdD&If&I9Y*Dw1JS)`C0EgkOc2$9p}~Glh2#pfaS?=`*UqSVVop`TGl7q zc*P%XOdQ6868sFAI<*5w+4){ zZnGCW%rukc=;3-HCY7+NPP}RKL``zEKWB&BS+Rb1T?rilh4l84kJ>EaqN!2-7OoL$ z{fwTzQMR3@0jqRTF+3hze}BKl_G2fiyMm9@3s6BH>@ytUVJQHT9@Vu+V(8a-nWf`o zC8ql1eD7CR8>C=pDL$kTy>QlT&GkAe$RDU{CJ>nQ=jJi=^FZy{JI7l~)oD{EhM?0( zW?5aiH@6U{KOk#oU_cm5Pw_1#Mg{VLF7D(mS#9FOm|~k#9e+&`Y%K?kayj)46oq@+ zlhbN@nRIZ3XDS|EZI=ieS5{#v2EA8S&HF#n-ZCt!t!o2?M?wSyr9@IBq?K+^K%~37 z8>Bl`P`bNAy1PL@y1QGtJA^YI@I#@W>gXUFAw)a}_GmQS_Nenl0qobG_8X8-Pv%n)llOH(@EtvAB2#S=0((q#DH0W0?1yBq8 zjp!DPP>y;y8W4Q`HV?ZGwQc^*?`8(BW~CP>cyQpKK9^FE^9G3Li>?7{$@U2+q-Z?; z!PCh2C9X8n#E$X_f-~$b7$fV~&t~)O&|1uj20HDbQM3_~*ONyVYLDF?#p^CV;~3qh zX^}W{%vVH5aH`&#)l}36G%b;&f=;10QVFrJHbPmo4+|XJetc;q$4=L-aDD6P(nOhP33+4hYdJBT$ zsW7p~0guAR*MGW6@9gY6JB^r?e1jy+OQgtMfsajeILskHJNc=wXdYz&(f74Mw4pIC zyH`OMa?4OGr(TXV|AW`p{ngJ9qbrFdF+y?b;K!Oc_U*K9~Yz_UaTNk;p;`ZoP1 zm-g)EX2VlXUeHS>0>fm{M^zn&j~v_HP(pTX#>Nm#2+Cvd0DrmtuHm!(%K(Amz)Kht zVoN5Xa0oLzO#}XrQ%vk!&e$iE zcSQ(|OxX6?NxA9-PU(h$^GP0Ax#5j=tA^`k(A5|PKwCA*4r;{p?24!la6uRj$D<#X z{uP0T**pQzt^yL2fS=drE5m8Qk4{efX$i*s> z@yr#QAAc=oVX&0BC$2S{aqMJ00_XR*EJAzlm8PHep0M5d7PDQs_dtH?#?3_qz`4z* z#b=hLS|lF6WM+A@$%`>gwu(b6PAX_!vdTobd$6k)PRZwGfXNmb<`rr9{Ls+VI?(~i zHpFY7@X6*%;W>TN+TbOS>(K^?=)0LJyrDvigE|+54|~rsh!U?CnK8-RQUPyi46CI^ z;=nh8UkM-|Y#?~Ir$~)5&7SaLG<}2mdqx*LOeoO;DU2$5{`5W63Saby{M$jp;gYKx zE8LysV|+yT?6col6yFc8cbQd&%aTIdgg@vk4Do)XlF}d7I;6ZFm@>7{3D6R9`Iu#B zyZ|?ae8O?KlnBi_JyVh8K9jy?SiJ0ueb-6(W{XPw4#{m23b!|z}c>ZSXQ zV_T7G;I|AoTc-`-m?=%SQ4tL>(*6MtJcjUkz^?`LppdDl^R}YbfA?%j@nwPc;80c) zhHb@sX;b{>AReO~KXqd1n<<{vWgd5hH~l8(fbPQL`Os$6-hK4fveDZgI8ep|j@FDW zB5t=jicFuxUcIo8PO0RxawdG-$Qv-3c^%@m`e5YY=cFKu<-4YS$$1#XA(!|}nH1j` z#BygCTgf*M?xoyx008PqIW)IpdE34;V?pGAY3`fI-KlrnZbLp-~TBOpE{amXLDGgL`ebsiZuc^GGf@iM2 zO_sZni#}!0yRV}|wJtv@1T!)=hV}Y20mVzTm1IQL(-%_DcoM#vaRgTDUx%K|2+Ltc zk2%@NB8pyc5&P5()I1zlN3o1?Nql*?^JI0&PyW%+bgSN*V8bh)=I#iOEhA~DXL|CH zPr?exy#^ihm!yYkWD^d_5)hA05uW?P>wHw~f#=Cv@Bp2}Y_1nCL1TX4?o8G5v-SOm zb~P0Ls-qL1k{WoCV;gC740!hSEE?)a#EUJoTA+O&@_lSOmunvMX;fHv3E+jO9WUoAX9V6q_n@%{xo=&6<-2%aA&1wf2PM z9GyK$W4eoLZUj!(sck#LMjqAvN@+*zdH(5c+wpk?C$6?g1pw_fD3zd3dH|+{(Pd z`*ly{RljY5y|Hh#|9p~S?n?L3Ll`ss}L4ja%cgH%>s#YWAv9x9y=_IA)BMG{YljnY%pFroi3`>P!^t588GV! zs`u2-Z5|*aYg<~P;gUwe{^i1$;T|_0ZQkMmJK`lAHHpt-wevfifRC;2z&5dF;Ux z^pPl7!|rok!pz8V(Cq=Xmt@v6bq)scGT!*2`?hDBs~Hav@@pJ%CC{`=vkf}dQGx5n ze$<6Kx{s~o$+n3-bTji%0O|47z*_R%P=tl+`Oc!Gz^CW8*aZ_g72^*pEU$HtT? zim>q(78{3`>i3=9AZffafia)zq!QHFh22i^GA)BW)x-!21d0Z44{AK_<0D=v@_yWg z2FYt=`tN2Qd1WLAKQrz_O1vY=CX_;%uqPnMhvS~3IQRJyS!2S=`jq~qbnm5d1MBG4 z@vG_vRYS)~CFa{SF6it%A2o7<@PLf7OE;V)8S~o(IZGA8L&3qhXRy^ZWjMWJY!ps* zc!j@uOaUSaf`G|=-sB{R^P`sz6@pcY9jX5BqA>6JV27w}Igk3u$7=134Pfa!4x>va zKh{!vNm;%cRWwL61pVz8fGXwn)(^(Cto2uLgoYOhzFo??w|;SOn*&;p(?rU<@x(ZFby)B`#4*8z8t^RMDn|ahQd$1U;N)ga^sVhwJgnH&9lrBSt(t6}&JX(M0;k`lHtGH6qNXIVA9VG);Gu`alYZr6R)8 z;kd)y*$YoDXz1&$NuEy$u+M!!*>C~TXQA#lwsUjjCd-wZhC^KeCK84`_aD8y_Sl+% zgtOmhmv*5$H!7L(M$eKrB7*$Kc|d#NWCllBO;6tCq{>QVq5Bj1$4yr516P+vFaE3; zxJOp-ubA>Y)g9kL zJ`ZS*ehEb}RyZFT?;kp-kh2I;ECMq&&GEbFXYRFn^CmwypuM(}jdWGri7N6LVPUoS z$r=9EX1DfU!O!o9x4&Yd!9A!gbFjq46Z1Nw)S~gptd-o!q$w3rzp3&?r|?c?r&zJ$ zky=_4Yjdp?Ro1htl~}2zZjoYFn;-3$S!d7B;M0Y-z6}!Eiee<$aKhfK8%XDB%Fu`| zJ+bgJ~V(A-H{9{ur@o{&Ijl4oVX3_-#9#=DU45E-$oTASE@IN)$fTMK}%IBM*wWe9N2e+jq|=SqDj) zGP>(AdU_DL3jDUl2C%m*<_20CNf6BHQcr$f-cB+BlyJ$b-HgYp5wR6LN4Anzn$l`M zqvBbSWH&=Zi|@q^GYzwQHqa-}k3a1gcfaw9gR~phbj}ieI!PqqtiJE~BOxGfq8PO7 zeSh2CM9VsbtPIDq3li{puBw@A&mG!Rr6OV)HY^?%?W`(j4cy5ZE>-*aZb;6I1a2a2PI)sNvfVmh_Zf-;PiZIsonQ+gZEFT;rk~LDPKzqy;4&~P~mhwJUo0Y zAONQml3FEndUP~AQ*q+H2IUe&6F7s@I~1Eytq^-heASaiI-jPTKvMKO&Yp)hiG;&}M* z<;MIsR75jP!%jkZOVr&DzLmSAHv5==xd4Q^c&FZi2=kz=pZ{XV$Ll~}C-m&pr?djo zh#yju=CBT6+DTGKa)*HBab9E+T~N#m?nI*CyZ^Sco{ex9s*0Ln^`KE< zkOWehN*E%#co+Cu`%M&s0FU_$duYuy#1Eq=q4O&!IBatEQ~Q4{QMm0d-d?$PZQ?I) zh|{-ce`%tt`vEk}SN2qzK2@jlmSq6vuXNo{u<^%DqUQ}V$)r=JZA~2Vs3S82vyVEx zCBUZ9xxMRJ?BwqECU9sR)P2{rDU0L#yrF$+wjfn+vszxsl9_Tpc%SX#fY8H_RkU~9 zM%vJnK~0)3yj}h5cb-6GYp<5n3?7cF)#tmVta`uUs$5BV40l?pjV_1W3bo8p%{ike`i|EMLo3H7 z$Ifw2aJ91yDQ0OP$^$#vaMLj@0mL>(t3pu4mz+CkdV*3B#cbZQ_lJzf?KH-CjRNfK3Ua(sV+Y=j@h ziirBhS-muhl|%6&j-?%Y39!%5vhma5@3 zD2^8KWxm7uBKIrlD$Wb;*Sxiryc~?L!hAk~zM6Mlp99SFH3oJ;IJq`^llS!EGbNMW zpiTb9F6G4Ez9LK`7wEe*oPVSR@nf7Y-`Rr*|AF3T-2~l!A|ZH5o;xcgosayFsJs(0 zr#_*Ydc}mxoYBcU$M_LuG5d$cz5-3g+IuGVY;f>mYPwIm&39`nHSnLG#ZD`a*Ngd^ zMeAXukkX>H>re{!{9KnWG~F_3Vqscwb|Y&0)&pzL&80olifqr_PNK0U2Z!ISq4}Cn zG1{}_zIP)-aaN2(CB#F3!U!!~d3k@K6s-t#GjK_#lM1+dEjEbZNEz|Hq#50Cbk*RL ztb_^Kv!nW&GFInmj;9YXlOz6oDha49&$7gi2LnS)>dkH6P#*rgi4e45)|co8v%ODL zaGx~f=i_MfkKzw6PWXjclf51p3uBURD#o|A)X{yYJM>n!1eBC=QOb&33uajuaO^}H zW(zqA-@xhh_=`reGwBb=_c_-n6>TN{+FWuds(FS zw-_4{+hQ`Ka>_{_WU#n!ZgyJqWcO>y$faGfdmr+i52VM;rBm;rA%EO?1x0>s#k?5r z;JEd%9WLQJ$NT_$gO5wym-(cm?wh2qI`Mxx7{F}Svt}5hYschCP=`zT1gBTb|4=*H zzhjhVx(+*Qos$2@T79H7CHiZbK41;GkAHl_E~`=&-qlq;vD|xGTis z$)%St(jg709v}1{G#-?d4aq?VQjCNkpVAc1Lsyw@aFe~8mcv3-98cR^7&h1^(=p!q zrL0$^b!4G7!5Itfv&H}JBf}h|ZHCFVz!`_OAZAy2;)~Fi4fF^qhQbbq$-Qzg$c>Se zrRjE!yEBV^VRzSCrMw$+bn{p}+T@G_Llyazrtz9urggszyF-qF>O@{XW0~Rjj)8}+ z;|BwrYVK@C2!Z+G@cc?A=G}mah@L&&-cMaL!o4X}2is%c9cJi}C^z>i_a@m9e%F#r zvgq`Z5=&5DQPS~+Zxz~E{>~=lI*h?!nR!!(+3)08DkT+MZ5FwJj$h@76i<|pI7iNJ zGBl^O`k!H)4?jpeaMFUq!eY^vAGmC=W^*FO#{Ftf>} zNE&0<;=&4lg+;HL+(-J9;^SV}ICn4YGHe)wP=S9;spiio+$e_J1TJVVsPOz|4FV4f zd{m4H8#du{qTo$xtueT>ocruhpX5vT#1c-)PwTK@ymDXDpDy;SUun5~L2Hdf0-G0* zIzbQ=R7@U_nYd)$784*Tj@EES()}KgQFyh9Xb^sDZM6cGjf~99R4^fM9|`F-5X)O7 z%KF`A0>ZlTfM#0Q`R)YIeIoWiz@3^K-(B;XXtq*QIPBS5b+h&nS2#z^g6aRfgpk~>AApSS$w!#!@!Ey-*Dbm3>tD= zx&dupPKhM0oLPO)hE`lzIquZ4ysT3XWbC?SWMsxh1Z>{FfBzCXJ1rK+BrMzhy9*?x z`L4`?=s-GR^Ik{tMa29ZaPA__@f=(=;(T9W#}B^U zha{?ugyIDx&GQ98#>B^@kE^@`{)~I%@p`PWi935Y^8<#HIs5hd4)(08U8?AxOZg1m z)&}g22{x8PQqYvV1`9qJgxtFJA#6_~u>TApXPA8OktO?EhL*HDsh%a$>N-)v=z4iZ zYo?{lwRLPSr4nrsc5f6U&(20Liz`n9xynG9ma9M^X?#_XSG84Q)mNE@VBmNI!A^Wo zR@KEoL8545>%9jYIv)%y96BH~f}}{hKLJDFQ9xsOcvc^j8=l=u0eaG5jQwoM4TpU@zbwP!?l#TiR8qS>O{aj(69#gh*wb?c;ugnJT-!hXX!kL%UHf#(4m00z zW5s^D{PyZBOW&Hcsq5f0uJP;jVlO@>xGfPoKh3Ngt5EkXB7XM!Sul`Pxci3A?0g3h;y<|RLV@aUMuGP)+dQn_eG z?|A;8(!2P2P_^((ZH#iqS? zYXo~%GuwOCA`Ajozg5z%zBlE1^BCfeG01yDKQQdM&SLugZ1?B`*2!!oF9nGY3v*uA zkV3KSy`-l%VuoTjpKsW3y&NA6dX90Q&pr?4K8x9zsVbV(;k`L%4U38*2Sz0aDw4Ye zgq?S4KB3ivmOldv)I{A65n%2^S@~$+<4*N^ARdLwIYKaAflcno?u=rR`;{>U`so|3 zo+w&tpw!q2`dP#+H6SS(**G3(kd*crM{B+1wl6<;goTBnBE+3w4vFZfks`@44JpE; zXAo>nr^oFv5hUeV=H!=ON0#pd{$%}3xbJBj z<5o;kz#8rs!gUh#WC+KsYBG;o?|#kFDBt>k|&N_Y?yQnxDrwRMXos8h*E|w6SghzIGA`h<@;%Ux2No?Bkfs8rD?R&n2()MQ#L)7pMGq<@39X>A{GV6Cui zkcL4MivHQqYp%x0AdSS$@~&Fnw|Ajb-By!Sn1X&wNwKa%^|d7RV$R(3DH;kX9QuK; zC(GWDg42~r77&k*TW|t&L&hpw#yF0s7RJOjfQ`M>e1^qlD~NzoY+rbnHiw(kUv&Dj z(L|woJAr`*r(69L_P9Px(7-L)qB^5^Gi zGS~I>T*rjtp461~U>_6~%LdVs=IZY@(t{u7kxoCB3wKWe#BYe~pz_^X8x;QA>hTdf zLyQ&Bg*txTG)0LQAceYzvqtAfG1r}8)+}M0Vqa$A&_u3d(}nLTC|))8;eC-PmSc{* z3&U#B5fVc(+LqYuZ#+2Hx_x8nbA~l;9}`~AX_uoTE|&p2K&RkvMmOH3vsjJxt!fe*2pPuh!3<3FRWYn!D>y!)4Okn=@&lPWt5A-&D*L zKQ=2+D|x(}jMZ16r>}%0h4v0wMAyWf#txQn){W-rMzxwupqsOczXARQ@N&`)a2*HS zv)Zkg@IxpS5dn8U8t9D-*iyT1*b_qt&d1b0!cfp9Y@AU}a+aVEp~;8J_k_yJOW!%C_U%T?+n#X!#{?&Q&l-u_V|p zFOSxy0}Y18ts3G-qh;e+qtXko({dDgWFA9?u;1lq$exXwzo1veaG;%!GKVDLJqR?$ zv%od^nigaD1ZjqP-YQ=oU@Ew%&Qi{hURUekLgTVAlGm4MQ@<#`dvEqF!zNCa~&E1w7Dh zZVKwabEEfmAbjcPQJ_-rgvs{31r@w#;2L7F6VA*lYHkS}X0_uqOhpHWzjF;Yy_noA zZhcs$=p+dB^n5U0Pq2)n)8l4m3mViWN(wd!?{1G4^kT>C*yM(>?THFhETd3ND~Bx< zaFz%IANG6B)jN+#MHi!a6~mX;U7cOY>-~JjtjF*Fp(hu+AyYLmwX~n(Y~x0=wO>?n z)INdMM2m?1_bE9B;F^O&x(=O`q8D)%7J{gwp#SfIFP8 z%s8~PQI2D1B}`#BjQxp2E^(f}BYJ1Hri{sKvbb8*Z?0%cghNTL2M=MMf88vnR zi@hYw=-V$|ijWjS9w_sZu!xAC?Z#jW%J6K1i^Jwnx-V|J&f#*qH6V_OCm%`m%g}7R zQlz`4$GTd>>%!5@=9=>T<#51FL)u_V=8(vy>nXGErcg*ZZpxeJnVxgu$JR+6i`??q zW{nq=G?J?*Juy<6py6u8hn)i9NP_(U`Fjs1?NMx(;v56~#H z7~3>m58FS{hr-@NV7rEnQjROiaYT>Rh&C}M_!&XY~LaN8FgyY9SZ zlP_JOoz$D$8`Nz2n2&(W+?W0vbv9!yMG8h9pjU-VL{(bgH3?LvyDkn_^ud&_V9@r% zM0^vVoBbmxCnx9GfE9IeNUZN9nloV2dd$XLKgLhZ)*T`p%{7tmy>nDHAMf8DCAmSxDy}w z4qlCdei$u=Zh-q-O)By3IQi~#zmx;;knGnW{GkH#=GnMyiXALNa$P2m(lk?mkNO?_ zN_r_U=k-(&*cns%R~mFtJEYYNvYx2#W4)y{T|`AQPvs(Asqd?dN9Y8OI4Ko8CGT_+Fc3fHF;y!NQ7&8I z0FO)g>{)4)uU=QVfifRwQN4sPqV!_+6>d87hbn|4Fk?@Ic{6&nSsnBV@0oFm{DBF~@!w`dtvvcDZ*ZRfEJeY;uW~Qb zK`OYLe;$m$i~E7)c7b&i#!Cr1Tdk%P1(%`i#yxLsG`6PVbx38z4fyOd(-7GKO3|(H zf_P)#0B(Qod$1P#v{hGKuP_-CQS?Wx+@v|4j#Lk?#O&?uWug@avdaDB78JWYZkKD( zGx0yT%!PC3k}Cvrzr%na4Z{gs>^^odLGQgG@V&i{tI+Bm=S9wnkO^M_P@wvX+<>5O z*DFU*G)!oC1URxH05!!}fbk~gHx+Q+o4IuD)D3N`!oQ2y21ULPOET2F-YyD7y?ab` zx7HKwFQFvvXZ|bu;<@^ibMJ7=8YuLzJAgzv5d&o`CEJ}HpwQ<1gfWo#qivib_pU9? z^mo6*H{uW_2#2&-)m5y)qfO!-`<)2orUA^m)*riR;bf5!3j4FPwXW4NLN=JfRpe48 zimG*#6HG<4rV8QM3L#8^Um`_x@&|QQ3sTN&=Unu53mbmeC5~-5^!0O>nyk}JmGGvh z>Ki=I(WiCNe__6+&OWRqU&&M&T)UEU6Hc-l59mFwo{VtBRIOZ{V3LYKVG3j?z7`38$mPfhPMq!T=%=5ZP_=B3 z-?3O+(vZF2a%&9z3dV)kADbrv@xeW_=<>8y@=3~X&7ZpXO3Uv=T@dv|UlFXW(OYTdE`ri$&Gi=>q)-SdYDroo z;U3#HV6Cbd!n02x)Ni39UiHq3Rj};8dM$x!%c2&mKW=J1hhytsaH%TrV79?>5Oor2 z#zfEuU(sZV#p?WM1rm{O24kn!)H>kD*7oZ5Sz07f!;E31RCP>Pe;ASV-GS*9>(Ys? z0NaxAAfj5JE>M=!G-EobQ1{+f0XTzLP_)bLZdTMJ$=O^hKownD2?{^Ie? z=+C%vuSkgiGekZ+BW`ncQ_1pG1s=qG?As=e5*xvc=0$zdf=~B$9ro#d-pOG}v+iv+ z!oW(Mi}xhxW|#2U4wIUDOG0BpsO`ObVD$cI^6hCaiMjyG-W2tM_(Ps8Re%6U2GRB%=}v_v=&sxClQ!3!~-VE z<>s0}%w{b-Q#uvPqf5w@Dl0Y9GphMNj?b41Z57Os*-h-0 zYB-3ft7|r?GMq(a1{$sEFFt@8YA(q1+;ta4GLxYk&oAQC0z1nSo*>J)wzCVBW1@a19(<61h(^d&vkoK*M1`k zg5i`{=8s+oIf}ipv_y>z(h@4FvUCfRGBOW8KKdDC58FiqzT1l|&&s{gH9(8;@^wEn zX1~2&xQo-$TFS2wtqbzVp?y}d?YOwJ_h}%|R<%wVhcJ%rB^m)ELvh`?FhwG!Bim{% z0(k{;-JG>7#Xgc4M~83a^5}R1{5adfRtl`_E)`L(OGS{?Q>vOs$?SSvt7Pu-*G}DE z3rvJ>(TQQUg4e}pAN}4JVlQ|qRqnU7<3+mIb#c=owqDGw_5BP@BraL8d|A~nS#IK1 zMbp-{D@kq!kEWz&jSW2FPW=VMp>}m+A690&LdC>SPcB#QJMoy7Ms0=r`yo3Q&MA(< z>w<{|k7q-JNd)sFnh$pE>d?KetqbT@rvqJ+nWLuN%apbiUdm3K$LfyfP2H{}B(gh^ zH{(7zpfRtPzn-^fQm6=sQ=)y3r`n3iRA_%t@#QpEw~D3wY@4kl?yc)hp;`QCMQ^Cs zR8o0{TKl9_hMi1CWCK}qPbjgQB)<^s!js=UGj*8m(vCh!TIs9D19X0_-#%o1#l^& z#Rg!~qkp=yGf|X$vnlBjmLANbLZ!@0;V){jCaZQ^QmMH>ZU11*WRgeWg84wrMQ!5S zdU2v7fS6V`Vz)d8&)A-;?0^C2@6wx;RDX<}BV|f~fE{~{GXHwz)0+|HHpPvw%%Zx! zT8VlL2^?pN^v4QJj@jzXqH*nwAA2yar2NJrDnn1jo+_3>637RYmE$J2$4yc5$jsS< zDVzt(-&A({c^53H$!xLPCuJseB0*ur4hYBhul)~-{3|a;ki2bO``K-N#Ds02Y)-ek z0)Yzh5T*I$3LpU1Mg#c}4Wua|sSB&q+K%}NPP zBnU8`C?F?NkjkZu7r5>)d^~mrk>?j4-y_mQqP0og;J_FD&DwDHKE!3}KX#C0ohAq$=(SeoJX2Z~ zgk#vrKgOo#f4s)oZ$4_9sjckEH?OcRP-UM}VaQ*C43jZFpR1pqan28_7_ewpPU6@y z`v$;2%5(a7d)#tn>BY8$Gj|Pe;JuVE20a%5VIdvIc5M*s)!B??Jc&uPwK;rl0clS> z10$noFBB9fBSc$Y1lXE?!q);;=8$Y!9wMMuII>l7uzv$0z;FP+bv9V@aN2EmB%k)l*id(mD$<* zO$$}Uh551C(%o_tMEkZl3ODyx;-|N33q)x-Cym7)+Py4}C8un1m^x4U%;B>82}fh2 zy`UNq95-|F+yZJkPp7jatCDI)}zuwv3$m527TT^ zn?9_HYuInFab|VidkqwD1Tio64$j3d{HuFmY@HVR65JGw2`EkKUDiUmZnq zw56oo{3ym@JlN8l-Kf$cZR+A+k;F)sg&>C(G?cSqLx^8BLRcoNd0K6sS+$-xp=dso z)-`W9e(!A4eOD*RqQZL2voMG?SNDrao}x?s(6?(_)sDIRq$nMD7PKJ6Tv&+)pO4UB z8=cz=%+MmjKZH)srzj7#*4Si8BxV3Ou+(gd!R87e#Mu4%Ydl#X)R!RKo+t`d8I)JO zx?7|7c;z2NbOq$(uDu-d~0Hos=QVO_t6Bw{GK&<-h62anbh3M{n+!Lw; zrRr^tL)i>rio9VfCA$|8anWcxT4tEseQq}k`(Z2jU*`C!(13E&qMZWU<3C1$7(EZd z@hqH<B2{i`T*YUGrWWS18jr*%QozFQW@^=t8Bp}goaw-WM!K!$YpRCwr zS#$9C-13#)aVRFK?Y8(-AOYFyzjyA}a+{MOPvLF=CM>>Yhj1t>!7d1FG`TO4yJHS9 z$NYLA>-;ykVD5~2I=d=B)pP%uG~0$3Y5q!yTA!D_s;ZMhbuJAljSs?+XbFl)8QjjvdDGpT#@ZMna3*dCAw8;sKdFbIpy#}q zci{xR+kZ~ZpSlDue0g4}#3;bR_MgBcsyFC%1X-VsDaKndSD4ypRoN%5zP+dPlo(29 zl#udOgA1r84SflHL$>=Pe{O);?VttG0O3+a4`Rxmi|33`8?XIx4%=Yz@%kZ^<`U-WnK)72N^A5#yqgIq1^Aa+#7e09r+cD`CrV}ien~gY{uF|$Qvf>;oH7rV{;i{LDm+d`y5Ajk0D*>$*ti&$1 z2S{yWJ+rKjnyiv$62b>~N6;8t@3N@^v2$p$^w0ANK8JPd=^36T8qqx%$InL9=d6&FS((&21@ zkAdq-N2~`~*bu#j`L`mE)L3>KgD)+Tv{ARGq8%(wz)gw6$kp953jg1C1qSkkIHll_ z(EUOiS~a*iE%Jy2EJ{&Iv;^xi9d~Ng?^9=8kNX5wWU(8SRPXTs^7LWA7nsuJk4>+k zb}{<7irJFZxAJC`SS>rlYk+gFn!ns1h*dVcNz^&pY!d+vr4q>#p!eLXPSbswbrz;t za;j}6+vDH53S1wo<89AOo65T+%)f6%{mN>-bY!AiicDWG1*T(GyI6=FDweGN_Bn$5(4xe}S_LUpW->(#j zTml|6NHq1@()h(&)C#k%&>163{aj+P-H@L>^wATs=$?m)25yYE|>qYQE_#Bfjf1s1c+UF=$p(` z%1w$BmYk8{LaBYt?h65){@ zehAQ@jkag+_>;!`$MJlo_F^%g{!C{L*=|Oe}Y<;X40?HFD6`Lqlp~SIX366M<|K+s)zVCmY z3-sGq!-MOmqRa(qV&qB$p3TK7uMz;mt z#nswwE{i1TdH;bW|6w`#(jK701RjO&0!Xl?b-SpEPSfpFzGklyBuBU`W-(pq5{Vqc zfc)Kht)u-P|MKszk=DY%4!+=B0s?}4w#o}tmdOcDKVl77&Jz+6KJQ-v*zXk5e?ff(rVuoST2F&Pnq z`esx>07_R+4+d(KfuUg!U~A9=dMFHTSGLNkKDt14GN#s5UU%OA$5H!x zB$Dr=!xW?-x^+JD{Y>}oryBip5{&v{x}j+Q|I5N!AmuOxi-_o3XQ0I?6i{}=@dtAq zf%>!4|B6Wd9svP5KIFm9zBKoC!0|~7*dS|_!G$L{y|%jd4{P*yA0NwuPmso=y!Gkt zFZ?HW__d@rFcGYV(^X5XIscKL6(%{#gzmJ`%W?>Ar%}|JxY;{`bTT02&W?UWZmA)F+;1yvhMp9tlDsKAG%(Br<$}UBp|}6oe?A&yR}em1FMYy6z_MEU z_GuXO>!-Mx5ieGNUn**^Ap*aY(G19MSD_WTgamxWo+_tfgB;bW!aS!i-2dL03n5}^ zYSEK(zlaD-fj(?q$2L3KW99W?a*@!4)4UjQI!Ehj4}40P>Cf;+q{*C>-IGMuMK z&%T~uqXz80ZLl@9iU4J2+ApbVgQ;T3rO}`z55L==TYFOaOe*c?}8?@L~{l71K zCzfZ?{_h>RkNyOBKjTZlvj6ZrD3VHC6%HS@L8VAililp%J_d0E_{wYk^o{r3(hyzl zsd}7kjtDU@FhC&(>z$Gg4hIiF(3D{J|M`py^~GpS_g5H%##QJ1zCT4}a^=xxz8Z8C z4zJ#P(Dj@LPEZE)gPv7bBvu2`DyOwb#e|)5vOK^ZN|6Z4M>~@xtvqeikR)JlP9x&u z|HZC(+}|+U=U9?%z>q}FTU#emBvdoCy6=)Wzs~5tHsljrxxQX0^RlU-afe;bb|@8it-IoQDK50&R5^e2 zu^66jj!<~#bvu}7p_c(@(0Vmg2s>W8&VDxoMBDvViI?+x9Q!+Z@Ln$-0E%FOcDxYg z|61?gZQ(QWvgv0RGU$zYhR5S7w1q!HeIiV@GuPylZkGHr5SRZXzpS5XN$mU~khJRp z<;OL>YLhW4Xn<2jV4@E)%>F&l4b_84mq$z$0tY@zA~f8>sxxle+1o8l3HL84UuuwAMye$V^+%wDrHX42)qUM-kXG zMuPV19>67uGj3;*Z~|zRBRS(iTg5wtLN(<**G?F(_h9Wl>22(PH|Ku>2fhe6+prW6 zd21XMAbGpv+STQ|oZFc;T*NM&TTYQS@i^aUVddV9j6h=MDDk5@rFMF zpc6Z%9r*=_w2PwvH)LClL?Sh(f4e0R>d!gq&a?e)B|g3{{OQE7zhsrR9Y28-u!O^pOTr9HxT3T#!|&{ONMOu5>Xo4F_0}hVAxwI z*RcJTZoc5ATiamTGtd=I(E}=1tFa1N+XEm>?RHvEqyX(K+H-eMhYS+#ZXj0Fdl@$Y zR!R#brnJ=ugh2xii2jtn>wQvOOuZ+yHmd&tl`)1fF&*VRu-e zeD(tZY8dB!{YXNs@d&xo^=_3u=-_dfs+I<#8GI5EcR^i(DB-V1Tf*`o0k}g%pM7RkdXIa4r zh`JQ}TBXAC-U2Z&n*FHfYIc#s^7oM(ka07G&>;Sj?O7Bk5)i6abV@;gjk(Tewno>n5h7WeFryVO4%^mz~p(oCFyhEaXOH_tOaWHV*{I zbK15Z*A@C!%fkdHU6BQFi*wu-tc)Wl@w(u>MhjS%{&GnWGLs0dxZT}P8xKgdVa7l8 z%XEj(b=B1YAk&{-yJUL|l|XoEw~6UFn!iL_H#oBls>n3)5jdw2;(ICxQ2qyF39pL2 zjl@7CPF4^8JqL(+{ok0{|33Zy(piglu_d3x{$)K$Hc(@&>48Yh*$e?Yi##wQVPC3h z<(V7HCi$|z7+-HfXjCA*dq2$~<#ez!$nHeqjcSZQHzOD`xXFAqz?tL%%&ZJ}p6~iqY|e-wc<#5Rq{$rnVTJ49I2ydVMF==RwZ7yFrwMjY zqPeMKSH%<%@TxNLBJ2J?yn{tKw6>E*vI^6)dg|%sJCbmh`-4~XM04fVTOX<#Bqa*y zY5?3#oyFHB=~GY=c5&76L)j-!A;ljeE?Ll>_cH}gf|g@)dv>atRkK7u6VE_Jba#&1 z98s0tsh-tfGa%e@JB?qw4vB_@^LO`W>+n_I`F$NoSWwiAA8pt!xxOVEG;c@+YenR0*Ov%8!l%PzJVn;r;ZziQo+i z@Kv3#!lQHP#VrCrU2=n6J>T$%sneH1l~h$tS-A|g`2BnA=4)Cv+~JlgXGB6s&|7+q zCk~khr2!qMKc=rkph*4IXpkjt8-+usf10kv85-rCQeUqNk^d zhUu!_zrR=QUPXaLLuieqPFp#5rR%vZ_l-px(lIWBIxcEP--eP3Eew0XWbnt23G4UW znR43*b5@gQat6m527q`NqA0ee1}0f`+>z#&YqI5b(d%IEsGMhhFZm^UGP3Yo!F$YX zNG@&a3An1b1L;j)Gho`=+v7OI<;_H?Qyu@&nPGWM<0+Z=h9F3{7q(3(V7B3V_m_Ur zFtuyF&9qx=*Hk?A!}-i|N$LZFieQ`Rf&7^fc8YD9ruEP0iPgu8zeYA#j+84n&Ag8( z3Ys;GOLl|Q_y_(Y1QPPFyYmdH1cvaSwd_DjY6cmtXZ%Q!XUL1v<)Gw;bzEgCX3l4% zptCcxz-O0hVY9sqJMSvwJy)5toF+PA*&zchViK_rFAnZjUag#gBlak0A;o)B&Y0JP z5mV19tVpn_1+ZivR+H2{sI7Oslrbr>I*3ix0Pc0edLQCHM?dXT0WMj^+R*FiKI`N;BY0|t0_|1QB;lFmRKx) z6%FIJ3}0e!W>V!A|HDlLu-ypA_5Zw^ZNB;yJc(;V^)8TxmuSch+^;C$GJVA@P!lhX z6GlGzri=a{-cUmf7d}3|O1h)AY_dvAFoU{S+D%4n^5Yzr%erdyG@nZC)0|}@|8x9( zlD#nSeZ|jylnOw)kfDfqyjF-5ONM7+u7lctmrS~X&P3>`$N4t)U<1Au#Z>JEPzs#V z8Z!8iEZ>1!puGvld&)0^a{n2?PQ|@j9*(|2V99H1gvw_%@$yDW6XS!_G-q>p8t*^m zZC~nK+2qP^ZR?5yDB&<=DJd!cl%c`FDoCOjfbv`7iIV8p@-yYruZU-%y2_`%$rcEt zR7HjZ$^c!SVK~OqBfHHG2e%NM_^A-ar8jwKV0B=YoHBJr zwS?@QEJ7nEw3(jT8Jf1Hg_$n}kp5|%-BQ-2a~f53 zb(o+O6~9A>*9F+r{)R7YIFr zLf{U73_b~61@kuxWmsoA)96V!aO3zJ1DJZAAVfBBppC2Zz62%@az?BtnDu*#om$ed zd5J$*@3;3)w`}@m1;8()YIUGoE%f8m9fPe!oicDZ5aqej*&`A0*?M`Fygk>|u&DaE z;k)5PCR0c$dat{xXI%Kfs@h?qDhg5f8g=%A=r=XEW8p4x0Eh2qpM*2-UC@rEr$$}y zrr4)G`%o5lVz(!fbH;{xE;YE@4ZOe0lwg*Buja*Yp?XeVdu6R7w4>&}$uAvA2=474 z7X^Kmc+TL3Fp?~m)0)GzVdQFvnmfK_E>ibsf=i7&;1YHNoKvVyplUB%7g)Ofy_`|N zTw21gyVR^`Rb3wwV+{-H;8jeGwPa`Pi8(DRKc>L(g;P7z{F;O_S=IJHz6iH2KxV5L2G zFw(x>A7ae+TnD;1S~_T$9o^>qwW9n29^Ua~WGf077#pEYB)>Mb+Pw;JiP#}29}reGUm4>O5V^JzW%rfK8M*Ux92zC3NEYVjkuBXy|2aVhh z=wIFS({5ZQ{>pU&a_6}v>&s&*O0i9vG#A)fQse{Xd~`o&MCCw_iCmM^P)I#hLH>wF zpqpZ1dHSb2kG4g*ugr5yD&1e1DxNy~?e0r5#XfozCty8}gaHW(=3dolWA&69zaJq&Rub1C4RhjLR+rn{) zh)baO%MMWQqcp?0ssiR0W{$O<5|*qy)pXS{CEv$6aJ(P%5tiJ9){_UyLJ85j4CNAcII0eYqB(|s;Ww* z$j;Dj@}1?c8KD1+mq$7Lmg=&FcZ_enpr)42jRZaq7rW$^4;uf}aX!D0LhJhc_*hD;*0INZRRW_*P$cUHBNo@#&Zzvnid?}cGV{Vd30d zz)OzfDKmFo(9rBw8VYX}N=f1cqGcj}9YFhP7O^6dlR3#Xe$$<5L|>MYWxlq2L0fz~#uJ9rmo3pf*B45daQm zwia`lsB5l5D>W|ys2?}Lma7Th*S2l`gBak$6!iPIzm7f@1y#$ZNQCbs)lW`Jr8G@k z4D^jEJaRn%A@u#UndmGyVpk3nY`-)K0UHto=ayF=OFvJ?Ra4J?H;+4%L-$4A$x!=! z+aEjxvCeCBAH@#Sn3J2k@0?uzXvWn~*8yrP3Sl|CJw80JLi^^g5O2*F4)!q$45yK& zhl0ZUe#7=lk0Z=EbH8!Q)9KvYvgb-UFunkH!5tyC7pX}{Uh?(sa@$gKh86w{;_A-) zhgI8I$_i2d2b8jQWTD|o1dxwsk8^E|>Z&X$UG}$}Z;z(!6t)on0>oE1Bon#gLlsCm zSUp9}Uo=(o{pWo1>V1Hb(`y@hSk4Loi=!s$=;7^kx`o`=9k)w2rxnJJN2r#rrOsKl zsGn;+H^#M__C(FQkPUZJd=*W3=uwpb75s)(-samA{|dO()Vs-&w!ak?5Fy-sy(QUM z7Imz%Y$dUlRj9MDIwdrQzYRe32`J{)7%JJh2Mu5kGOgH9Y|k>TbS$9T+(0nPE&ZN- zI>&s3t3KOCb9R0_@H_PM*=#_TER~e%F$+j1GgJ=~B zc|yUooZCO%$wE9u$Jr4q+Lh37{wh?f5tn&Kie4sDksZ`sM zw(WSgUSlj2;8#0nx3w|#Z*UwDTTu#+cFt}#alXZPo4{nU^SNHDqMIha$1;{;U{Drl zmIIIrLDD#4JSxtcgQ_5hQAVG#-*#TH5@a$MOfVrivkcTRLtT}X|M=~1hW`2S&Cl5Q zI-P^^Ud$ohc2!-+vEq<52Di2}yOhjyX`pp0H_8TY7WTmLXarUUwfnWnpC<*`l8<9i zZ94Y8k?!;D+*VzJ*P$yN0z?T#*j?W(st#|h zO2r-&WVJ&5k4skrn9D*;#TkUK_J9z;)G#`#2NY@3wu~qm_wS8y;!4}eaIXb8bfp5r zk2dEO2Svz=IFr&ewzn>kWx1+t5k2l{<0eKv?39*%`Dd8((;189MXVK}EzQ=LehV`C zG+@YGXN8IWnt8?n5RC+f19Pu`dmuZ1A}NG%_Rlve+ib4WZ(mNrih*zcI|K*k0s9;h z=+uRNkGyx-vh79iLYu<3%aivBZW*I3bAkM;H|c1(u8>IY%KVD2jZY21g4^ouA@cKz zK&>oHg~Ht^tD+L1mZDZwmsSPVh+PJN+Fb!LF7x$h|+j0Iy`A7pbd&J-=V* z%|I5wzyKlQpd|6Xi|K=tCPRbm>gsxvyq6EcxyG2L)wCgBFSI^k+Qn>!Y<3AtG6nRJ zYLnP#H76|NSSBsLVR_^U(tA2r+H(wFaozMUww2T;hKf^tcOqQOXzEc zbR6p?$6kGXr>gI*&meXW-Q)R*&XxF9HFE45%0;+h299nl-i2aWaj4o2d`_8QM6f8+ z3+%(t+Nd*{(XOLxLO`l_SF=x{wnhQLBoHYv(9XV=Qj&+`n_T(D?^Pb~e2!8kgb+y3 zXE}_h*gFB#jp&b%>Qp5|hAR~XJ>+>M#{fkBdhQ$!cJbhc%V+rvjf`e;BdJhc%+h_2e;IYj_v+-?tB-@4%*#bR~5T4c85jX7EK>v3H3p_OW ze51oNBinPuh7l$dD893Tig~K&3V}CLa3wfgR>;^MAv!K)(Xd0`o~){_St? zxb@{;IIL4!zx@7mnTv2r30F*#5B+_wli&=B?s(G9b3k|kg0!ngIVjKn{*{|^ii$1S z^Y`)mAb2s4KTXizd8W4?gT@J3*!iR#|7s^|rcXB3s4hJ5IZ%r3CF!ebK;4pf#kxpm zCR1efF%Z1jg}ah#QT|TcpN#>Er5UbxMo(<`gxU$;B}LOe!Ve%yqo^EA`ub!32yU#- zRDOi=h-A%7y;sfmAJD_?0jTSbJs%&aFw3*cQJVfAJVfsWG8Cnt@WT4fsH7pubZQ0c z=K-K&KL$$2aEF-x zvWBGK@{qEs^7eb~_-T~|jnn+!D{pV`|6i^$mZGOQ=m8>4L`O^`XoJ>s`GSuS#0{zf zm;$1rIKV;5`##zt4>j^!Py@r|e1G=aw?WX$p>p5Hq6^TG-A7rL+xG0k^u6Y|TxZMq zvqBegrm9Y$JSe97a^E!=ik0O)vp-1)WJm!!IL`9mIDuozkwIDKGj*ViNFF#@Br5dD z6>`V?vdEFM5nGEwl$(&i(i@-n+>n1j07HY=MhTLrIOcb^*d9@QfL48X-CjHdwvW|| zk;JTqVF|od*4~z3P0P}f(4#R-j#3o|r$>?JYe#D0l~8>G^aODH^U%3W6JRw54xG`J zg}V#-U8K04gdb4Nmm4BF7`(jppgKOPYP)p2Y^aYqQ{#oZ3zGTF2=xl*}f`xxqGf?W3RA)U^J=U?WU z8XPv3>n5F^MW*4iHGvD)DI`BKTYX{>&`c+*NeyBp7*FJ@T!?>^!O&1sYH)**Dkrk& zEWjWcE}&!tfZkI?w%D$lF$FciU&W*=--O-~SoHz^ z{-lj_kzkBG30#(z=4SOgn;-Mxt+Mpa8T-c6th@Os_dg*2RO(jP1yKwlk416#K#wnQ zq$3d-t{82DEjoEus~6pkF*(K+VEHqE%7TkQur_q^mv8(0c$kQ_Or-XTEafftE%| zMPZM}38ahx@T7V?vB1EQ~jGR7v0>=RyXc z@TriL(nlP30+X-*mPnPGR&KOOQ7jk&fp9(L7d&ii$z2Kavnx-#ZZiR;l@(zyBuhjh zrs)>fDWGK7=d4q{fO;Gzy+eX}NupO(?o&|+n-p?4cEg!;eq^?=0%xiU=x34cfVZJX zPOc;5}2>*L&-^}Q-5v2wqe;)Ql+1>FeG}{%NZry__aFt1g6k=t=yV70G{8- zDt4X03!R0Qn`YMjag7;Le!za3&#{)rWr^FV@(rmDVvf3-eP z5|BJ;E<&uBeMQXmgfQG&K_3vnhd&?vwqLTM0j1~Jx>z(m?0m#=K?_~E@z7{?dF6Px z<1uSI56LLoNr}udjn71h?2#+u# zGFEk;ViCoZl9_i$;?l*+S7+rL?}q{(UUmG<1qCA@ot=lbE!X>8bJs6`{dg~d%i5*> z;hgggL?~OX-nj+ouA9n0nL)!6(}*xW03yZAu`I~WCdD2;e7N4#_5^?<_5aB1igo3p z9d1&9oMC^4_4V#XTPdi1X99hyEUtcF;CfI^FMl(CExl6wsthE? z7L_Z;8u-LQ*A_XS*53!qv>^S10iuubI*qhbSV6S z7`w4jNlWs}6*W_SZ`$JT5WKmABzpRVl%T5r-c>A5gGkfiP9C+>2)h-8B?JQZw^ur@ z)_qjz6$0$^kL2Bv_RFxyjeQ*T-dekWA06ygm@!Op3jTpew|RSZ+8K7l_}5JUKa^azAD-pVE)X1pBU&N;bCdWC`*_9R>o5JdDa(11-9!l9%q~caEGkRk z=_^AHA$w7cl^mkmpbAJ@$e&b%SS_j_+Wv%s9~^ct7HEbc$oZ%w%0y`W?T%1ZD>U{c z&BRBo?s6y~v@%9UM)yP0weqWlJ|oG^C`vq%#q$-Z4Vj-CYuwh9&$p6H$9QS}Siz}M zNncpOR9L~#f3F}=k7Q)!0a1(kcofRO>LCx^6g8)G0@CdHxo4(@jg1LUfPpPGwM^0d z3et#;Wc4C0GL>t2s&I{rt3xR@J(_;n{!kq}`eTh7zM8)}-VK)C7UXIub2C@IdX&Ir zZtcr1xs-e83qY~)yD9uy>sB(X!L*P)d-Gfl5l3#rx1Ery-+5jdHzl7gPQ z17%6L5{{)V8!M|#0dXc0GXJm+D}W@E!%cfo{|NRbO3&S6&Pwm|&si^7PMfiBZ~<#If*)fsESbcXHQVQI}Dm*lbiuiTf_*5*LEJbrE4kxLLn*U ztyQRlRoK-EcTJX;b`0M{pVh{N);7QlVzp=Nh3?(lhu3#5Hh#n09KUOM{T}gNqWfM) z@zONT|4aZ;;X_LJ5@12`7;fT#z(8V99Vzn$c^j&MfdPt(-Jgj;^3w9r$PAsRD+Ty> z^aI_QrSn?DA}6kB4MZL|yGZ}_tVV287mJu!`4zw?@u9OGdls{%-R#$%a!-GePe6)! z`D5BLth!G;-)#Y>|0=G)S#vRq|OrN$SS}fAJb&Hn3h4Vw6sp_{&`FEbN z&wWlw`Cz{2~>o?dG_hd z-ET2DK((a0$RT`lw(k9Vf{xb9&7j$I2bG#Hm%Y#jx8mwC#Mhhp`mS`HRmdDXs7Um6 zG4n5Y=(*3UI-HY)Sx0fL@w)bVnI+e(tbM`M6%j+M$qZ#%C%DlF^w4$id9OFIL2KHj zxPG?|-TH}1J7DA3HDLt8`myA!!3lde7AwUJCcBCdR-ka`RA*sBT2g$x4CI~%A%hV# z(_iUh1!{(X{%4*o*COMZQg^Kr+6Cat53h6Zjr##*o(&>oLtnmX_Q*>BB};v5Zb+jG zA6(q&qzYj3NK}qA+ulKhMTJR~fmWnj`M$nB1Sp1*fcqO6rMKDouxn%|0j(MTec>n+%`$7R}<^)jcR}4n99h+Bn!@GZR%x?X98PL zp4S;L$lbP#RGX+By#*G;ShWD^H?D6(_W<{R3*5`SE3PYVK!$gz=JL9}wH&Cuc^6FV zg80}D=+_+)_2v+w4S)~jP*9U&2c6sHot&ILB)q<}FAxY2_!B3ru0cRI3(^8efIkRG zgg?~N-*1&y1-Z@UbZt$9DVh}H@nY|xQ)uvNHFPwT+si>G8KDPw&(iT2T?mWTRAT0| zxB>k4SrGCO67mI^X;7f+v#6}8-gQ8P{^*TqtbvhHHB@0xa=lHA{AC15q*1c1m)nW^ zo|SAD65eyef{IQJ@S`GKky77TaA|})zhr(|%YNS*E#I8XdM7PcB>pp=l zdMM-yZ3cnUPv5#kPJyU}>bnY&DM&gph&NXB>!p{2qrz>|iyn3&K6vu+ep+F{&Y3K3 zpy>+@J-IdjNhIcHB>+~ru<{%Js>H+!l3&^TLD20A1=TOGmrU^rP!JLS(M|94zH=c= z(`Hl-AjCXxCXq?~cml+s_RgMr+@J8n#xvC5^O;q>Z9Ac6J^fWhKCKljL-V zPP&+ijQ6-s5{$7g>!IMEq~>0;;AVQU8esyQG~3RiKR-ELSiXi~kDKc}VqKz~7;eJC zV2QmL;s6*{6(B%#4#@>N%^^|LI-v4MxliGLp50i|;mSElLPKUIHWk&NW2gCs7TnHiHfWVQPPxY{xHq|9&kS4B8GA7~ zwy3pkQr6&{K2>T288p($+-&>SOU8#WF`BB66~}B+dX?BGK!S^#2rsb%csRzZvF1RQ ztAo||o)gWdGs{af{i=2RJ1`b`h&)o<0knL{`yD(}$zsOil~8K*WzrsT5+MLPQd-`D zEI|o$jiqp)IMnL)Nh4j0uX9b)nzF0aZXb}Pmo&>+xHePkUr6b9@rqol3jLKCKtCdW zcIX2XVdSn;0(E)lA}5|z{dk5DF^-%K-L16gf=B_u6ki9|ZkTb#6# zyY%UUZz+eYBfK9c^QQZbT<)jpYjoE)v>O#LeAEHb!m3TcSuBpQ>4Zr4DbcqAQ;tRp z!`k`?fd(2e4-uCjB23j7$WuCDuN{^+A!yj+YZ>qLPPT@|B?2m*80~2j`lF+-BD3j-ZAS{L=74uKuNrL!(>lqhzFe-{HO$-SsJ(KxgP;z z2;WV;xv)5>Fp4K4*K?A!@>7Zj5sR1RGqT^{pQfkM2keK!Hcsxz935SEn3C}OmQ=p- zvQwgs?Vd@}=dmuxk}|!B|1?#NQlHi^mpORw;O?-nw+ifEu)zNh2jt*+u@2a(3{Nn# z#brcGCv~2Gn5Z0u2HdZ{3RA9SdsdxvjJ3GrR#t$}M;b4-&qFD$#44RN2E*YPYg}N_ z{?Vjc-H>j0a4>9kuqFa{`+KnO5#Bd|h@G9r7O8VuYl%z;fRz~%oc%54JR&;yBX5|6 zq7cs^+guwQtSX}tP`h}bYx+GtW3Z_K>%6gp?)n7ibP`tg6^fHyYL`<~1RK~v^E<*2=8R0IN zte`dzA|f(KV0KT0p3tmkB{1uRyrfA$ z*qzK*VvtkM2@Qix`WxEI_?OlgU*Wz<5R{b_l$MjXGzWl z3IJ7a%ke1?<*V&08Hn7dcH;VdS8MDvgiZ+raOBN>w_;8x^#DK_)#_9_g$($N_FiT ze#X8Wz$EW&&l}aj@^X|b*DULjrZa_lr4svlj*`a|eWf+>0Qep{`_evyQc`M-$&Vn~kZ2sN) z9OVd~@7butTD?vRi5BiX>)+-g-l zShIGml}3I;nUr3|g76$p%O=-=aBkXEHk;(+PBX0rm6t^hTHvPDL0RurwY@AFs^Uit z%i)kkiKFwBpGrxMsdQMpyVw>y@EnyM-e(;VLcj`qy>3Ts^%HGw?TM#V9&?8|0WUrR zVO&Km140!6%9Qk<+5&LcL*iH3=fNL|pyo6C)E!g+BpJUaQd7n;QpNrcPMTf_zMZBo z>w1)W(-|c}cwtu$NtWUM8vi+_SiaM1$DdYia*?NeSfWhe_h5Z1=_{|H;>Pq!} z-IuNN!TS?GXpnZqlvWjUyAU N&-WB|mC9Y6<1t$W4;V>wl?Nm&9 z$|u&DZ$8YaUU5m%4^k^MxnF}4UpH;Wf!HiSa`tuJn-$%>{)lif2ihD=I`TCUp1*uX?aCa1vbOBVI}LPbe#pwnD&3PYS7M*;1dvqU@1xop zxcgynfgN1oh}s1yR^KO_lZv)}j^%N0AFuN9Uemqimzq-&!IrGm>D(NCs*@Lc zx$P(TMlAT!yBpNjpa$uJADyUo^;GpacVMY|4-$MV64>VV6@d@Rt10zhVbc?CQs@Qo z)B`C`I}d2K*r<=yJE)>nFGZP!?=ClKjMMjVW*tgn$!1LOIr0;Yb1-9>%i>mUQIl?f_jY!Z zSyFa!YPrY?sAZ|YoP?W=I$b`+QHq#|=wV{ni6bFf9G+VMWtlvY3bg)piDQ)rO&&Nj zCQl4)%xA zEkra6WJzL)O6rwAXqsa`ZZAHd-}`E0W@$GUUk=yM__J*#?PMn}Wf^d=ti)8K;+lGc zbxk9<&rN){z{C_t298fvSy0m4fEEIEV1s?RwFRuYp41w?t@oF1h!#EmD9Me>joh%X zV`Odq;tOq0Ew?x~la3YG8xNuq85EZ}0x?(&zPmo?*=J=W zJ)n&ev}Jgor*@T&&vUsrhp@?%sB36=xLAHpME$H{5*tp6g;HnP+Pror#{3Kj`8h`s z?P2p#rECneynXw&_!Z@0^A}~PQjg#TlZ8fJd*2>Jb#wK`yTcm;OwZm2J{l3=xZa!D zb9mxxGF_@>X0-Cb?Dh54Z;DUz6*XbHAHIdQPTdpVtuf-Z_!m^OZ3JSsFosnhynqd^ zV$VkTZ|qy|IKp@(>D37^(D+#yMMKoZ+I^OgmAY9y2`y8ICllcGd9jnLnM%WzP0|U?|d!YG1VCr7AfYD>^NuO zQd|(EJTFIqgQ#uzV=dX|<9mm58_ph~`DiW7qIKsPlt*c&ucS=1#N^mf+O!C#?m0lW z*J=&o{yn%VhCbUnM(5xaq8prkbW@wW%_LthBGgiFl^=~zz)bS=I4a~@twrhlXJTg*JNFC;UEL|Cd5mHLTZ}HB31xR-u2(s=AhL}iWn?QB;d6Qf)yJ1Ix<=DLHh>30! z2o^t=5Aq$Qx@jlI_*E@cgCp)#&}U_Ar3t2CyeY{~2!t7@bHOkWq@L6;G)SQ?jhfG} zcgWE7S&b?%}YKC4V&TjbqL4+w9Fovc$NuSu5$qk6)S`bRhq@6mQ=CmDrVDe(g zqm+E}wFZD+ikf&0iy^UxEDHGQbr6ztF;&}Ok}3kNY(TZNc;y)n&{9H$#x_nt0coyf zC%>}^2q}_g%(6lTtb9yIH&Ef5kEE$6RVPSn5h$ z4Or(6LuvM1DoxLW6O#E9$C{I`@s@TK5OY4Sct|7kaWb()-K))dK6XMH2`qvE^EclAIo}iQ`RAz-_q>+-cB&}`k)svcw@rW zGFbHFFu0p9UK6>j)Es<%^y?PGPKepY=us(G=9aG&IggZu?j`!V3Tf-Zg-S0YU}F@G z4$R{+MXg&+*@K=75dVRN<}z!Ts{4ZSt#3dQdu$adS_#yDN!lM zS;k$usot1sltKc>#jFY)_qmoK;3dB_M2fVK8qbjr|C}dL(s-BZov)?v1vCvE3OV5s zN{6aww++)k7S7>VArY4n;a`4OWHhQhk|>jp&+YqHQru2(ex6nsCWfQj;G_48orb@> zzI$!A7#~I_xD#4hrQ}^Ug+fuzVs(K=wdDY#=ksH2Tz&3p>T8fCBz^>9`X8XgylqX- zVYGPVOL-KOnekkyz;V9VwyfIUeSlCHRB86-SmnQaW(oeG08n{}a*TZ;8`A=IHfe;6H07-ZKLVs0RV4Kuqp)@@aZn{fLH$16J=pHAJD{dG<>Db3B z?#k+6Q{Bad${#rNRB;ql2ZI}*Im%K7VG7xZdB|IOe9m5;8Zh=}w^wZs37_8pi%`@A zB7KjWT_dZR)0LmzDOQm51&C^RK5zoyJ`RF6NW&_ru&8XUV0QI~MCu@9+T<|)>2#~< zw(f(SLNC&f&TAvW-nx!>o9}r`4*l{RW=6v)j);6jvFS~IeNsxAfa%9KCT4=^1}Mn< z@qoD@Oe{4YEuPPJN=_-Sxj4kT4b8%7dd(kq10i(%mKXc^t03+=m!7ovp3RVYlaKrJ z^RJj;)+~-!FZ}JJAjVD(EKt_%9sk=C(4WBFy<;h{1gwtVzxC&@HZU++eA>VNkst!X zk}O}=o~^s^_Yo*S%J1p+e*FE9S7BJx1*C&+e|;{x0(@9U3WvY{k>29}G+(ILN8Ve; z>h~GFg%lEek_wvI!eD-X2*PYt{_yud{>!!g{j2GFVK#+{gnygjEp{ylKN1BUQ7-!* z)-e+Jxc@Ek)*k$4ABQ&v(6lziT~f8F$T{d+s^=?D*`@-e+DvRFS_(L`#H&gL6?);oc)0oUd&5-n#-Hpn>$&XK9awKgCiOms%cBD--nB$-+8w(Bm@NXZt5Du9f5H?8ICNj@l{#t^gO_NNrS=b#H zdFz#tEH_712)beH)%iu1{m1y{oj1qAB!x0_Z(M7aZ57UrS7G^9<9`e(m7TC@81q*R zFS-e)m^xFwAIs_V_In^HOmlAB-^2B@WLwk5y^|7tqeLQC-$#iT7DyOzre`>L^1YBOjQPseoUhtL zwXKAkw_hEUFHl+Ffj4+8t;L6W#6DUC-R9m1p4}$X%8+i*{*)Yh2W7-lM$yD7_0ll$ zf=kOy*OR>?H|?Z2f|)2WYGT?DRz9tgtvf9c?GpA6q(-GKknoxD7Ui+%He6U8jCTHE zG*k4tzH|fSyUdTEoG)|eu zQ*_JfQt!<>=@^#Ho~$njD@~T_3%;IK*{(f=7SCAv`4#yTqx=u(uSL(@ymB>-If}hi z-7LhcSN#4Zq2{;OoXvJBdiyEoqYRaZ#s2#={92VCJ%MPPYO?_>_yDidYXii~1ngUn9O zNq_HhhB?6;r~FzBrg@!=oQxcqoMG}$`#lkYHXHD?0-^=$ShaE3oyWsGk{o6MQeVD) zF;h}qqI~&+@Cl!nn@Bc!D78V=x#oNavutpVAN42aHT#^R!S6B}a&OPE`=rkz1kR|~ z&%R($l(}*(z^=;c#rj^lo3E18C>+}4yk#qS*2KriM^KtV=F2nxD^tO>YHM5lPRoVu zmu5n&lZs2?SAxE_&dr(-!EbR=U6`*=;aII)s-o{rzf=-yHII|;eM|IofJ(o26u5)o^e?IEwdfYk1(UO`K0;(43H&o*LX$ z@ggAn6cIt{B`^NyLnZ~W>_as&GA;t)d-x6a)s1eSMRNrC4TC<=;@!Uf;yTD`Ibw3v z!X4v&uJC(dLPV#1R83L0k2JDIz_LwhQOwj3wGk&KJ5a&OjZZ+z4D!Rl#XEh`k^*u( zH}*mC^k=`mrEnHM#UB5U@Be%678f&4r|r;Fnm>2S+^N%oh!%s8$je~}(@gf1?wl|_}-*T2b@)UWdV-#%~2@xu`;GRS`N=LCK% zkN`As=5L>yX@PL!NAd=4{%t$(Q#Ss#5%~XY(+{!yw@p9v@IPz%*CG68O+Vboe@^dT zul`?b`Vo}>i%ma*?|%^DUljBog!m6aoJP_AL5Tk##2-n`e-PsT{}AF~kYD{E#XDiU z$NKIQu4-y(1wxQPhyRlZ6Fo|~8+d6e2JXW;T5PQzvFAV(L**wNU}z=hs_>=TCa)s> zN{Nm-;s0jJ7s-dMGm>Wd)m80=W#hL-@Omu9564oMSLJ+@J!W1zZhk)R5X^a?{J*I! z=r+N)+nC)}hd{SSR4KHE6~~S_<(r5qGocJ$_2!pOi4ooZO_iY*%n>hOt<8Nmaif`q zI=V;x_X}8WF8Bn<+Q~QQHUGvBq&xw5fR~4234dWJplsaCICfu>(w_cx*ask-(VH@s zh1Kr- z8ZhN;{Y+!c{|J|WlZ3kB z#l*xcKW*m-CJw%2pj|yOC6kAC&j~9j*r-eZUuD1iQ%<%TAe>4yMv32wGe`VJ&XHbU zlB4dKKxtmK&-IHSky(w5Uy9X#xBj}qUcD2?ueY7L2w-K#;mJWi()&5LQ}^&!t}^4~ zo_~Ay_q%=#O3J|vNb|{F?C7tV{l3TikOa6Xl}~^Fmt=llfSv#wNSs~Q|MqXY>jg+= z6*~X$Z_^=C0^vN?L47R!n?m*91#XgX2mWO za%^mD?5zbKhccS7pI$lvsZ|i5+v&qxEBAnh&Jjr-O#3_btI5l1YHFT=7o&IK5Y(#U ze2jJ;U;h8_ZA2hoDmY0Wf`EBu!s{DelYiSR;tooaq}^9Vjb;W76bLtkGZ{@i6;1wA z1h&r5mnis*ogtVKyWP-GLVERLr9^F?>84Ab-C)z~jE1r-^iguN*l%(EGlHNj)`A-i z)!6?gH(FW1);L0Vf9vV*$3f~8n*=`UYm~0nfH(dl0jom1R_{oN-kzENYOH?qucz|o z0khPb9OnDA(Pp5O>S{?lSy_!HL^fv3isxx>s%)A;d%&-t_mr3bWRt*g_2<~k50z5v z#!4NL6YkA`Tx{0I#-Bg=`cgJQ*iIPfviQrgY@hq#xR6KC|9UxynM`!6f&3fE#h)vL zej&R__~`p?O#ENBTotQLwRjMCGAW5>t;W>1?J#&qTeThC zpAV%(;NzzR^y{qvHVajz81`IDO3TdT;LPGmOO3?q`2ID}VFoVoKXkMp10dyZ%fv#! z27}TF=^Y!$3d~xgbZWd@)AWnV%SFXqSNb`IoXGwZX1)+slhntsQPY@(t^XaAAk6>5MTul5%T|MMBFVuG9jr!4&6bN?%~{|~uvbb1^MV#Rru zb&aOaTpXU!u{Y82Lq0#x z%*=eXy}fI1|jF41e0ekJ(aG0zjh2`yuLYp>H?^a8CY$ZvRN) z|GpBh=MLib9{XPm-Pj*6^oxDNxBnE$zl#H0C2`I90oWhn`XK@6MPTJ33a01(=WYMI zBY+nO;s5W8d|3He>c>=}c$ArWd0{2NT21k1+tZ%zjb<7a38V5Js~_$o^~MHT-6m^; z#@)=leAKKJDatrPXST~~gWH@}m-j0X?y+?x2BYsa-?-b8%d(=IbW+N~pPq7c5&VEF zXcE_O24)-~1kEamDX{43%Y0WSIXEZn3fAs0D)pVJ9_J_4zNew-AQ6=Ms!%02&9gD)gT zePL0W*krzcoZX#KOtC&*l+JY< z=f(ba#6G1s;-=PD>GO?uj^hvgXdx%{TW}oyCj|3c&3`(Wsw2GIsdIQ#trPhoQL+Be zt*7EBly%Ilc;T(priFfy(_y@h@b}PN^`YhYiLn|k`t{3p=iEZws&-Z669!$wZv8R* zE1=dzjNR9iUi>Ke?pT#XN+lR9hIrd)$~r}ai-B;&N~Z7zb(_?4!_ivvH}i~AeAc>D zt5Z`I8#T!vD+>2kS0{ILPxif~gWK+>df92_BK^}57dmUy1u-@65gyG}_C_WV($A+_ z{I!NU3~5^}d!yB#5*ClYcP^ePQSHnyu&;5g<1GnOKdqwB1;8Ba!fNt50ei4Y0nG=E zTQyQocAQN_T#&GAT!{;&^OzS`=?;jSpBSy#W9-yViK`V6I8>R;KHRsQ@bZiuKNgBE z?8;3O6?xrT`Z$X#yN~c4rxSw3K)_){55dz3W?!|5a(`1>lU&;^SXSe;AC;+J=~a|n z#p$UD+7MWr+!7k%SRdvb4PRqWq^abkx2tmQ9{-?%HiBG?qC6$S_y9i)Wj}YNE(;+k z2L&`MGwkd@JDvVS6TQ#75I6npwOyCF?_byrgIO^Nj9i8ZLmuI|)YP)=h775i_6;W4 z;|fh$s3=u?)cX2cu!O8)aN)TG+fL%^0`+A}qr52F6+~XE@?7!u-r*;^nJRm!LqjpU z+Zx&$b}JwIW(Rp6Um<1$v!Ze&3NmvC6O>vB$3e4@OZiCl#Pj;w$ocT zfcmdl*;5K!ya2X`8gJ+p-svqgZy#6<@Qf87wSKru++6+kTyn6}lu~jWqh^JC6R+NvWd97EfoRsNU?>jIWFm{izCaZ_>$SYZxNZ z5s$!jzTE9K_RLpaSAF_!M4cPy&iV zz}-c7t4Z`h!Y+sjTZ(~`j-NMcod zSS||1&VA{eE84BNQsB0gRXr1>= zZV{`tq;)I3GH#bp>+KwTM))(6!sPPElm7i>OM|_v>}=EdH-`=Z?;QeD6>nBP9sS}O zXuIj%$mPs(*m1D;ZugxQs_MfNt0B{uvO{U^VSR2qA|82vgpEIP1-Z)#&-2X$r>h|~ zG&_tf0m&Fop)9QdjN|^CR3Za%ohfO5f*)lcOE;j~_jEQ*iS832Z_yRJR;#f=RGE^D zl*(wy0+9_BcX*?!a%&Cs?)*DUOMAr--2)}dD!5omSj-&52Z88rhvbkMs*k;gJ)e6~ zFUWe=&#(mrKup4}jR&lPYuJrALpMN}W5iznB2J3@E$1uS)Icy-4M z1xFE+BGhjC-2q7sXOo}jd65Z^__s@xz0N~)ht73?qs-U02bLA4uahcE2g8zS2W&V+ zbTDSco7Ip9_hrasc8-_wwd({gtozQiSXHTqUre_ABT$uo1mdOM%CQ}wa)QsXp2c*% zV)N+)N4I$e^7f#mBHMK@uCw_HEW-S4mlU+Ed$)&t<^g4<*tovDc6}JKeUr{L+4n@quAgLdO8jtrUXlOHg zu`$6HbYC*8TG8XuXeOvC$y;vw%CjxCf|bxl{zbluo4d(d1=HVK>3lSN50#uvwi_X@ zTVp3@SCftfc*u;Q^kh`n11GQk#ot`EYy=kTX zcJNHq!k3TZ1l{`6aUFdIgS7+ci3>D%2|N)GLI#lE&Zomll(ZYiAu9KcT5lo|}B=BkK&U-c=TFIam=g^il<3*&n?+!%+ZA;Q6#V3IwrXeCW!`iq4w^4>r$Y zROh%tx28IEHBwhLGU0sg=ni>+1li|)d*I(=$TY_F*u|zS;OSD~DF6v%Ex1Aiqn6y( zHg}6G(1q{zFlagf=O;^1mX`gx3FhRJuc+&G7hTv64Vfkcj=;eciL1nI?MeWr@Mmz$J*InvW! zqrK}6nUzecdogA!(4=L6y)m&)8us$NGlu_0){xe`$b0c?p5a#8)-qYDyO{0oy|Zs7 zQp=04V?^~%iJ5fx9xm<@+Z{1%KDtM#!*A}pv9-@GvP_1Y(K7EvHAME#-c9S%A9^~TwH!Rt$T!6R_dzmQ6U4uG zpIeI1SQR~cdXbq}zAH8eiyr&cjSuHW_~#eH(^;gsV(m}JkZ-}_a&Uot75{Ew|H^@# zSb7y^Z#-Fm&6=GX&+Hku{7w#UTpf~rcP>>e#+wn%&_2qVZPHeP3Hu}i7WG<0KfAd! z)-2aP$uQ;kLi02j_|0lUl`ktyu8SdH(I#!(jITYBs-sIUmo@Qu8|C9 zn|Mi=MANxbVG4f`AO(jM57Mr$3>8@jz=ow$5mUU zW+vYc$c2|g<3RWD0F)j4WC%v(U+b-Z6xK5*<6vXKX8-}&=RP%;t@hpX=SRh+nKr{2 z$WXkZ*Q;ej!dp(iRSXw5a|~-m2^{$+^J0-NZVEWF^hL=#iWC_f_9;Pi||_6&4zgB{~s8Oa=j;#+J+C$aD1=|ZJ8^_uU&oc9H6 zPS}RUU;J{3AZLKb5spnBQ~~_45;M311*PsYHs6q8v)|=MniVfzqORui@_D9lUBEyO zh*gr%C@S*;RK@Y4oCbnUz~1+oA-ZeIbg!e)QzjosKKQq)rxqu#9wBH>YG%`#1KopD zQ&X!=ZBJ=z&k4Zjbi%t!-T{OTsZN!xrv06?j6O10yHR4zXMvzLrBqA%b4cHc=^_9~ z+HQ(U0Pm%Trbf*SqodE?gNLtEsyma1*BtpDo34L}@uW7|+8(EIU+L^cXF(bxm=a0H z7z)%UeNPiff_#Ag+gWkQWd%)3;d4^KI|4YvKSaYdxOECXr|Iq5GL6C9jdxMIxV@>~ z#}^r~KtWF~<%6c@+Q9&e$+7A>6MX)8*9{r(@Wjk;g`Oxk3VcTX;WoX`Py9)EOO%8eaRcCB=5}@KIqi}w%H4t zx|;t4S{RmV+7iS50S@zEMe3Y{7|D>ea&IB@7sf@dF31RqodYipFS&nz9~pA3sm%PD&isNf_8%EXgJsS-Tm!B^jaHS_x)6k-qaxN-oMl`I=ac zEM_d6jJZbxUkk`<;RY!9J9KuhT1jDQ*V7ocr<5pZOzqNf+?h1l@Jp_YX%RTkbk-h= zhh;Q-fzBq58(Rm|yeI11C*9eel=|ZKav>wgz+Pn0`NnCku<*kD-0l4_Y>xY{83Nt| zuV-!V6BQYAJ{gn}NUSHR4?rE&^R1bWYf8IV+N(-KeW@8}C44rfIqZfCyG$uAGxrpq z6mt!(5!GkPy_C(B)w8uU z8iI|gxfUiccR~)kvNL!7R zm|rhCMNEO7V9BVxVU z_1L_emo&0>YA*a-XU(mHJsZZ5LlV_61B9r==!LaG$;jgPUpfFfhkpUu24a7T2VLOI zH#_O@5Er?>y3FEhXnCMJTA+EM;;2E?YP&2I*^H#}6&|9Kts<-r1Z7JnnIHmdNzc@Q zAKX(!<0~kkC3xfz`~L6Z=$_c0+W3^4XG3s!I(*e_@H!tG&+i5-cWekabe_GeSz_tN zXRk&JmT`Z0p+#!LaI!+VsI<=mhyu#Yo=lr54h(ca9E}PCyiT?tXKY^B78AyVtFH=;by(PhnRRZa5@Qd~&+SXm&MSA&R)0Gvy7S@-GK zsH|PjvZ-9=&>sTqKS9Ug>dw|dS8TJ*R9#)yX3@@jf77~*I;yHk-_p$;274_+(#$%L zZpL827)<)gS9@`uAibD*17jeGo^3_JLq{N2VA}SlZ8Dw97~hI~HDSIz7F!9LJzRu^ zAMPwRr`+~80qUjv9-Gq~H~5L|$X9s>Mu_mZGI(Ao8XcH+RAg-AzhB=>=HFe3P6be@ zUBKu8AyD~^22H6l(Ij1>rBplbyaC(>~OIQ8_Z-m`Wy3!^B>K-zfWCQ1SzP z1Jus`yRYZc`#_VEgRLEkQ8Cxi?uZTRs~WOi0dVn=i~)v->C!+ZWKm-`r4xoTU5svC`gO z$Bm-8{&}-u*lXA!m^)$DBDa(xhN>BhF|g5%!_MYf)MF445bs@}QE$zyTSn~75bs>j zNoqBvSqmZ-0A4GYDe+up*=bGNIqiGmll>^{=#Yx8uC9r$gSwvI7c0h8}i2zBm&=ZTJd?Iaw??-l%Gh zvrN{-qHQ9zl*%klEb~p9d2<=(!$ zG7ePLTwlUkaRcd{6lmYDU9~PeS(2BzUx#AC3*gN^fP6FQsFiuZNGG})za;$l2+%-$ z$yT2T`m*Ypt1CC7M2S}CXy_8<@yVi6K|2P_-BwkcTVjNZ-KvkEZA70Ti6LoC_qYU{ zc7Rfo|9j%tY)m^Q`AM)(t3t3r>NZk3x|2%QM; z{)GFKvUvAx{P?K)s4W50lPh(hAGF zw85e4mGIrl=DBeumsBtCfFXKeDwUJ^FNd?Lk^%0f=Vnr*a)Nq6?j1sQK2D-HVHFCu9eY7%l^ zsz=RPTu=+%o6|FCCjCs^p)-T#o|oZ<=x~qa8SkC8)6hGA4-4L=su!roO6?N5c9SVW zt_Y;3`@rbJLcJ%VfT@opZfdPbvoBaeX;;43i)W+ui%L`W?71;-X)F~A zDLD9aqa4a(# z*xoY+A_KqI?h=iJo9%T(m_dPw>~*;DuwygBx7Wq{limmpUi)#}@82M(d*U;noR(J} z3$6P0wFsWr_Ldm@E6F812DjCOmno*vq%M=&@e2>CHhN!;S4FtjqWk0B{f5?xPlc&^ zDxe%?&p%qmoA@G@TL0N?T0^#?QuuL4oIpmGQC?i8ysBpMJja-9TBmVa<#H?C^olcVeSgZtJvE_w&3=9UmOb ztz#KZ8u+jiPN}AJPtb1i5K5cGn2V7*e$rI86W(yN+?LL0F$PmI>GZYosuIN$uNy&91vI}5U zso@5DpoXf2ZBFa4YcJ88AC``n&xq_fg#q1nAPfA_i>s90GEPrIq;@k302V-KJs#cO z+xs-Ii+?&Q&898q_ZBecn*;f(UXyE4-s7RP=GG62+6Jxz76#plo$Ax)We#rB`cloy zF(D<6i+X2UW`&6&C7+Kt&TaHQX|U%4TNS(eWzB|NJ%}h1Ets+%Y%xI>NyF*JIBdx zTD0+OW0o+mo4BraVeNbKNq@ekB{l6qBAMX(!KN*m?CBVPql1d8QUFaO$PN|uuUcf2 zJG%46)xM^7q=IZ`w!2n{N^*bu=$6fCB#o`{_bM7h)R&TBJ64W{y|xvO_cM~OBq4|0 zi#YpsmPqcEeVt^=*2=vv|9sFjawRLv<4^(;oDp8hJJ_KsQz7qU8$8L-H|^XEXC!Oo z&>+U<($%azZVJfZ^<-6f$KXAZX$|rOjPh$2P6pk& zd$KlCWWzh0p$d!N`SIOYvqFaUw`ApegL(*Dbdl)Ze?7z3%Ed9l`^#K-PCr2*kqd(^2re>$DepC9MG8}0{KFs1`o-v z4bq#P>LG%5VgWRj-j4Y4uxWVc6#YWr1x5lnNU zB0ywVu>FGvWO;WJx3YVMc#I_7o`$rC=1;BXsUDd=3Jp3{xV{KqvhJAoubTN4g2+oM0B*wAlV47`Z4xH|+3b`~aO(ljiQ?W9j(D zZisFF;Mmf16Db|xJ(LN#v;7PUtbi6dpg7aS2sySbUW>*!tbf=+2nzT}@i6>u@H4)P z*B5-Tm4p2u2ro&(UByPPz(Xyq->d9=MuU7UP_N8wgJ|0{qmG)<-G=lA;f#YEb+gy7 zG9m3PbZ;J(qJByJ1g;lH$u8 zmJf6*kWlmuc*p6Cdyg9EXGC?xP0{SGC009Ur-{EjnQoxj zEMeLCksE+!iFU=jr`}-@u_lA-szdr7N6MJjynz>ACo_%argxmyl06{s4R&XLmgfTk zo-eX$s&Tz+im|n0qILh}+qooI)mov_8i0}<#~qs+`J3s#} zkZ^I5kzs%M>SN^UBXx_@Cap}pNnp!5mi6Kp4=yhcRG?=70UKj zh*sj6L?y;-O5d(vb4T+<`PsWe$#tzmPRvld8~H#}WBGdN*vz1X z0rR2Ae$^Dp{$z%_?o|wz_F&comZzD*pxsXZ{}m<;K57`ML+^+5Q z0!UZtTckoDgPRVmPwh|u7bmC5+Hk?L=}y_WlhXAuBVaD3o70~Gsu|&0i|YJ=v)Ge8 zt^sK+^xf_Rys-xkW`DdlTC4KHAG2vfzq{C4vLDKuKsBUs!qO#*Toc{t5U^R-2Bq8} zJXw^%ypC3t7F*~7uR4BryXzLF;=M+PbH3#Z4R#c>H=1GQ@3^}4k7_jj?x)W93vQ#{3V%~?r2 z%%t2!@tH0O!~-UGRo94)Miw1q0b$k1$ z)Nh$mrh(x}bXDlx1sa#nL#dJ?w~jt#BnJV(o&`+ptZthiPNsnGR&i+B3$W&UGy`O_ z-A4u*-1-Hz9ki$73j%0)0;{IJWQYi9QhGawN@|^xZ){9ZO@8ov_bU_nW{u&tDEQWF zSwMZ&S@8wA8YzJ1a%{Viac@3oe(R`Hq@kR+`}L4+V8CE^{?drm#c1tt z+cL~m{+SiJR8&40zs+EdsIbWA@ygybI;jyAyfL7O-8o@+FHH5G+VT-*Bj2K{&|`TO zY3o0Fp!hpv25{x0>(r?2>Ec~F<--}11`!9d;qWUpBN0T9O@F2wG4S@>XgD8`yY=M& z(r#m8xs3p5IvUYu#yQCn5(}954bge;@Mhlbj`kv53a%L&M^)~TT!W(}pY7y(2)PcH z!rlh_sQO!nWjEtE@`eaAKBP$muWU7cEG%6&$xzCBHRpI}U4i6=`5rAR=LBTkHLm%m z>WUl7SX0YfCXCWfGyVQN&J@!!kZmbPJRUM* z+N(*PE#ir&cKiM@)-VY0p75>JWLS_?s)M0FHMT}2-7DrRb<~YLSM(v;b)Lc4A4E=3 z4y0QqK%QKomdKKZu(q_*x)`j~ZqgE4d*x%rO~E+k<-Dqm(VSJ$*Y$&>|I|DztdM&$ zG2RruGoJ2qh7KlL2YVcYKj8=fUvz+Wl_v>uBLc?te!JqPO(E8AR$0UQ*~zc6wgVI+ z?R-3MlpQq4vpsEnel~4~?r_~!;DcV`X4}Blgpm>5@p_pNT15{(J`;57XlEh8A*@BJ zYZB3+Di4VhHok#LCjLoe zR0151Yn9+*>swcp3$KT>up@G8S_-XtQxE!LYu^~AIxlwh)3X)k0p&%|(PN1<#5UaL*rC8_^m+BI(}mU_MSAQ0s9?TYQLq_! zAWPHn3nBZxTM5VULN+TXy4j}CV3u&nTTuuzF~CEwfG@HO@(g&Q*3BXscxed_-}45^}suElW?|kdeWCf~97#(nh2$z~Dta6)BBV zr=r?mg==RFwt%S!t1Ge?I}I6JiQfzw-Tx z|V%?D{Z5fQM zJDFEt<2xE(l%{TNkX4H14Fy6Sd^w|5#$&fvz8MRI8u#99mDkpNKlnnWBYHG&fG7}Y z7;VhnrtJfIR!rvXEMv%Q2Ql=231?UYxnVKjZ!#7aECV&RKipH_Ar$81G}=3)g=9=m=Mqeo}Ay{ zdZFz7Du6fA)_0UWio0cgU-ED+dieug>+DAB>{e?_Phnx96+n9yJ4{JOH7~kPdM?t4 zIhj2!mF1jIjU&@JXt3Gp*bV&C!bpN))fSpkdw7=lB&B)L~k|*Rz4|#`EMFeybqR6oS{YN2TYQ zYgFeO>BEgr3nX{~-)8EX^_f~<&>-~B89Dk~D4$z$r_yDA*n`I+7xHRrw)r0fU3%u6 z$bG%H4`^YAmaztIfCs7|faYGkxunv=5y^p0Y_I zMPt;CdVq$FD=R2P&UpES?v@KfcnwL-zS(!n(5RRk0(#$r@7C&6Aq*27%8#o|Jh_t( z^u2PwUXsXB%UiyNU-}ZHxVlYqkZZc_`?RT3wIS%n&Y!$8=~Wz8YElm4f9!bpD};)Y zSZl7tqz4%mj;1Db-hOCnc4vAnT2;k}uyFUIVu4xT&PDrgIma78DH+qgkYSQ-V^Mqt{4b&*HG&a`Gew#uy(7@-)PRj8RN(t`{E(k<%4|W*F4?Sn# zlfHG%_p2$wlSB9xOHZCb)d2vcV5N(DQfLT{g z0b01u`R<@R2a*q~GE5G$o&Q*v)dsL@3%=*AAh(Hf5({o#l+o_gd83VCsNJlk*+!9L z)uSaIbXjWy2&onFV53lN$ksRDFhia@`4F!W6`vt07wA&pRFOgLithH@T?M=yDUhbs zNEccbU+K$D=Nz!fo{2%AYyNasj8JzHQVvn*KvC?6?~83iJ5xKQsEqyaB*^yn=QiO) z>;|s2OsS^K`Dzjs$|EWNDY-07m6DmhN-2j-p=_|4iR*R_AOjWK>@87DJENOSf^?- z@x0SZaq8U^6vOCwL%d;78Adhmwajo?p(?Hz4b#hnRAET6h`OR56N3%AxCuSUmCj z?xctjU0{?H#quWQTTOJcqc-qR>u>3eYv?i=|I;6R3O$!!H7<039F{`{Qd#tp1bZfJ>QnFVT* zAy@Cd6ck}xw`0S$uJA8OG&f;XPRx~qr%rat71%m$1~Sx@<4zGoooWEXK<{mkKz!eL zTLeSf`hp6fFV1k2*{u#)&0(#!-(|9jA728xjKlD(9(OrI_pX&s5#^Hamli9LRBdhs znN=SD@hXtz$x|0z8mYXIcq`Fgpd5Oy-nnSw44$%%@j|X@V|FZt1^&|RrCay}qCEhX z+x{;BmU{x0nOBQ3L4dWHO-J-Vq!p^v}E=27;z`E_HAvVWxu%PGZD7&4qt^ zJVACq0i0Lt?CiGjVzRkr;>aOnT!u-p+C!hUeVDzH3%w2Fe!TZ#Ea*5AoH|>J8bV}+ zy#pxt)v{CNMk@v&~?tYjkILr8$2RKf~|ptK^|w z<>n8rL(GbPllN9(SV<4Bj*eeH3sg0YW=3EE{?n&yc}Qu0q{?~- z$bMHQIeGG}2;mx!`t8Plo;}fcPuNjyXIp(#16k0l>QtF9?T2-R0$s<|PCp{pbPfLf z$8Tj30?1YgTYwU39V`9a43ts<6xkD$RBbGyDRsmhwH_SoV%oEfJj+T8nx|x_%^8Nm zCtD@ze~?)oB+%ML2GYocdkzzYn=SnE3^AqZ|u=OyU-Cw zt}26akpxo6Eug;G1RTIm;F2w}D#Iz(bt#e))`SwmP9WGLCY}RyZeKt3IwIUVCKJkr zeoS}R1-ikGhbzXl`>$!P zQ)o2+mXouqgtYV3Ngg2)Oh%{`ZBqdL>toS~Kgn6zYZ28dtcMsrC78E2#a(mPjtYb)#MXu&%lSvS6#D$K$3mRt%o+AX(juoc z@!VBkq zv>m*)-3`A#MOKNFK5iHWD2rq%??KPxFJDM?m|UslzyndB+r-!|HfPHdVJ#Kih@mh8 zd$dmRm0_NrJ7aYLyE83S9l*gI(ipCxxl=o%Y}^!b#c}2}d-`|FAtVEtgpt?$xv$;s zWi=Bi?+e$?@I(s~rGFj0p*ryo`1arG>1s?Wb>;d28EP&Y=6tYbOK+P+|#YwbWT#zt=i|`EHb(;0@+Pt z`UqXA1lJP({1Jw;ocX{JJ{tGYPO^$t_CX%=v69yY2uYEkc3}k5jO!3kWW~1MO)8}v z6CM_Q1~Pd~ZLOLi>wO^PrnD0HYy}&I^o`H4HdYWS(awIWa9(^TX#FVK8WsdJc$DL{ zK0vhgt(KF>=>~>ftn@fea(f+?wtt@qT6efGj>#*d_yDYyxk~vXv9j^WNLf?7NNf(= zO|7A2tSd-s%>A*E#gAqxpvtCl{?aQ<_oSzvyyi!T6V@6eDV`$1ix<2Gz%Rz$cy7{o zPK&~vXBac+s%@J?NhjR!(&RIqi7u)BhgZ9opx3WMUx2(2C*AI|4;I*>BW}e#U_*E& zuO>PpDl$$it_vVgLC$`Gj=O~+cxHC51WI+U;$CLP)nngmW`d-ors7>zpSmosrA6h2 zNEfQktwZQD5ef!?tgUZo<>`N#p!Gq&NWY-AnjFUYGLxc}pu10)@Fua(n9o%Cits&V z&tAC?m>ftY{a!-=L2`e_#r5Y(tl04{TeaiUC(@HimM}e%MzxoRcF4?-J~J#X=p-VH zz#%L?iw75bATJmnE9rt>HY(WIjV#SK$cw~RNFl@rsxKdPZ7SVgTRrPb6HuNNdB*YP zP^{E`f{0AVKwrPhg&00YgZtQTbeQhT{4g~%b&i|98WLx(j$YgY?bf9)brsykyAHzs z;ZbG=p-}uhBMCFg2|3X?gb|1c;{v%How3 zUxjYc06%u3aQ}Yca5;U87e0sI<8UF%r-AtU9Y+`DD;k@UB)lr8a~g-TaXLs1y}-G|A$+6gQ8;s5)mo{?Z~anWY%nG5_9#`HB<%u6Fw`{6_5*WX27 z#t=RXD7f`ZWyqJ{dd>0qTUreLb~?9ey1YZ(O#mN;E%w3t(Hp4l~_Vs{@A<@4?6qhK4C+r=0u{>UYyuh=~&j=Si z5*lj%S7M7RzeO~7jd}9C?^hPrZ@oksf&LeNtOfX2f|>Nnl~3!b9IPWIRi8L`ZYv%s zx;(rp`sM3a9%0_|g^a-PgQlHvqlCDv_bV3Ps;WgAnT!{tm2q_)JxkvY|G0Ji=PZrQ z@q8DTCZ>yM&iA(=a57l_Q4y?`SmT{Jf9XY0kpK}n`NOC<`^aS{H$?q%5YYvin>hCN z1B^mcR}%eo0&vbg%r;)QzWy52H?|s4v7c|>USn8EzhQ93b#eF~=>s#iAGArP8v#?q&R1=U;`V`%7 zaeuwS$+333xl4fZxm{AMG{BwW=hMFj6GrJZLNm9ygWRi0H7t?C9g@r}HvH~C-_?YL zhR`hn&f=4EUy|SHiKfMVC{2XyRI^g~#mdhFZ-0M(y0A79wK#YtN?&KEI6eV~K$N1U zCi%j`La@k&?%XC#pDQGw=q5DZPQ>=sk^7N&wco|Y|8_)ySegN8pIJNdyY4D)8^odc z{Zo$P)fPn5GCHbMHuL&&^Ok?VtX?3^^FX&PCww~EG|=!tkUu0gycZM1KtP{{j*ecv zlH$VZd}wrfdfL?0g>iLp(!pFc1v!Jq#44)QbJ%(Sf=v+Ed4oj+ck~2c#ln!bhvWWk zx*yDqzsa3bZ~nrqwK4ni?cs1!yj_ZPX`kq?XGW%v0}lT0y23r%&|6zu=K$dF@%FwY ztLW-Irj+_ZMbq5-*F$msP`4Xbq;cNsmtg8m>|-q~3yYoG(_K=PVspWeRS*O7^=q6L zL%51{T5jXxe;O4UI-FA2$HS!~z@aohomD!J%Q~f(v8~Lp4 z*?+g4S3B(A^V}__Jz!xDt+C-g_u-b{mmO? z&*mkeXnn)`T+S_+T%e3ZNpv*f`yrl0E0c{>>vmyZzb~$krqr_CI$#5|RJ*&op6atT zHuRmHo#kCH37wgGClE`U{V3c;afgsXitHul%1eD`JWiBq`JJ&>oAY;`QU6`!F&MI; zmIF7ca|Tl;Tuo75j3+S6u!TREE$`i$+k)!<9XKWt_)Atm?GS?QVtd_AL!J{G(WlQD z%4Auz+}y-~8@sn%yTQVIn7PI=GRcWt2FdYHh@2KzpZ+YG>y9vLUJU(3aa1zdq~3J5;%zJn^MHdi3f|6+q1P*2%m_^v{nghs|r4j!$4#7 z{Q`k36=eAZlFOq}t@dGNjaq!5J!PxJBsjBl1&Y5@(5R5$sEOXzpL)7FBvVm>#Nh_sRu-dk%ryS&hc zHvi7Ih%@A=foExLJ;20A`;ANDf>L|r_14@s@Am1SwIZIq!Lb(#EI_NBuUa8aMD*fy zV5-Np)1Y~Nl&1zR_vMgybV?`J_K*HS6n?GxpJN6fZ~jhiW-*>3^5N#jAFQ%%|C{o_ zFPl=(;wz9i2;jXm?)SBW$@%YZj1;s~TP=pC6y~GJJ6-N|;F6PvCa@dcuicA*ZoY@l zG~m;G{VG)zw;d*xagi2^_IIOJiZ^|Akwh+EuS+T@jN70_B#&EOPB8-CPz*fU@7E^p zao{Pih4+p{ko~!;O<&LPM$(d!l6K7FwZRT$0XgBchg=ul3_xGqoGWSX3G)?lRfL*$>(X93}0|3(yu!H0oZ=P_1f z^u|gs6)5>-1*)P=>?5{o>IV;!*r7Xrv5z41=bg03>AUGI` zWQd84UD)q>+yPQhA7cC7ZEK*>j)F=tO?v(07{*3_8}U`^D=f6IhqFeLCGhWaHi6}A zL8Ot(=^j*yc|y&CJTKc_L9pd`Kq8U<8g}LD>#M5E5c&u8^xZ~_CxB;W=6TQUOjb#k zt(yJRq%mL3D{2S?Al9Tx-_Q`D(mDmTBQ7cFRDAka9MmijF8t;p7T<1xx#Wu|=Lc^m zRQ?-9@GLhLP*IiW1Q-3$JAb&a z&clV9GV!W~LWeLfpjD`~VV)iX(F@e=-o?eN%njS1y=VCig$Uj(ZM!*!%kO;H1Vn6i zomQJ({sozKxRzT>nRW89DmkWQ|B&aO7<>G(DIaN~%0f#W$VuDzXWTEg9QS|F_m4TI zVPRpN0zQCMw-s$snS|4bOs3HkU<{c7Z(?oQB?FbOmVus z^Yw&W$n_qNHO=)|uLq}*nz_08>0So~o7I94vuWy7baHa=2Mi+4Rd;)8WHm||O_}#T zL`Ng35huZ35@~S?u~i^Tysle*^d~*U9_>K3a=gy zaOJm6S~P~(ZH(xg?Juv66eOjSXn`G%xI`qW84i>_nC$QR<85ocKdlY#L@wn44&mA^_S7ljrRF*L#X8GqKES z2!JAadAK%BaZBq8Y1;f6wO-G;MgB6n&DPS^r>OLHNW+CrGRF;bwVz!Z;91yQPc2D7 zXNz%TndZ+bzw`EVd=ekKudeEuxUc>3e*gd!7idEdJv#|4YL*F^ZCqZz4mR)5sI~2{ zAQ{*_I%-=V&O`M3eQgaXV8!UcLF5x4R`E8dUda5dfF+)+ao4zmpIggiWGT&eJTbAOB1AaEK(e zOc4^}1CWzsY&33GGs7%FsxdV$2U0Zifw$!3sp zuRtau=gw+h7I~E2Ix8tDIjrmYYTAl|H2T6tFEFzXRd?G1vaLZ~>~dm`V7WqSKDzx` zvdwrnGm#@G)b{GS3nKpE$Z+@to#iWJ`CUgRr}=ggi&lUPa`a=}LB>lI-a>>%uUn|Q zpo3{L{v})L<3#?RBRhIIzSsVxF3nK+b-e>q@)f_TiTGl5Bn003i2~_d)OFE=L88c>#9`k~aLH{^RriakUZC*akIpX?3Z21j^LBRPhjx7)y<1__m@HasX6aYP#J2LVP>= zkkotmp>~JX<^SPeM2>fY;KRFZ=2c+u1(L>0rk2+8!QN?;k<^Qm9T9g;a&^I%3YR^W zLG>gJSEw1@Wo5Ah&jlKvW#tOrUZMQEnL|CJ!s^Vsm&=ic_q6{TBfw?`)l%w7ByS0& zSKz`En)=#TTp863ycg>pOzY2-izC9r(?WjX66%z^AxqYfzkMe2jt|tLB+0k_+;}+; zL~5q6mV^Cv4%UBBC>&xh#GJT z--i1$JKbhkukXcv>-zh6U#EEsqE2s1DW!~5jQ^_x|E@R{9C=7{Mr^w!0}7i}%_ez# zITE2*b{!qS|EFBG+b}ZE%*bJENYc6d720v}XVXrwcy&fvPcM*7CI5d_JOa{aqH&LR zy{g8B72|ttcB6lvPL$_QE)5Tex6!nlyi@W0yXd>vj77Q)qhIwg%lZsF3132Vi5n9H z_w9dQk{9Rev{%`>n zga7!!_sP9NgG*?%u4lBOpx+YkrP;3&-WN4q-(TBMd~o@NB$0nBS75|Q48CX|#2b`Zv4i6QH+s!(!>izx`CDYt+(~u?_W-PXk z??NMIXFUf$6SQlR_kTl69Ge`gG|=x!=1D|Ktlj6eP2|JudG0Qm(m?qTd5vVG`P(<- zs%d+*sE+xNKRe=BFf9wCK9fM#^SlX@>sj)knZLxNN=}!%SicFOKYv8i=Z*%%tQSs2 zHG}LpIZQ?HeO`}MXS=TluC}Xby^(nveeXj+V^O!+2R(;@r8%NXt9`ArldfCwjco7( ziw3mh=Lpg>)8%Aa^1dI}pWO|Zs6>d71PFITJlnqiy}bg&?TyT^6;`#OC12`r1%~O9 zbT}N^zKBaowD_z`T8tiZGV4L{;>C;aGbJeK6b-{hxaAq;P(LERu;x}rpp(P^Gm>B8 zDSlJ?vqd%|;GYTfJSQFqYzO@Ji?(pU7unA|DlGqiKJ?ms=fqpTxHC)PeH*h>m|R_smo~hL z3nihEVq{BmYW7_^q6jb@Eo{?|)9{t!tb5fRgc?E2C_8#&z53rS-RljcE`r|4WYa{z&l4SAr-bOFK?e*!yhV9_(-~eBT#-3W6f^lLF6MO`_ zC>9J{Kmt_DensUiw5({`A6w?<)_nt;A^7AHLQdwPR=HeLmbbvLF^+D?kYnUu)(<8F zU4`_9A^$4?K@^6&Q1vkpodWVhN+QR4qo8U5i~;-7wh4OF)tCESrU9t~R)blR;WUqd z1j)m6x&x0!TM`m^$#9gnRr|1RMxSIPH7h94-@mCd=240JmGjSJ9yvS?3HE`AGDFnT zO7oZ7*VBoo8(% zq%j^81Toof3WF8yil_+FE`@XcE@!~wEhmCVRwB8r=VtGMXF$T!*3m(-AsZo3JD4pk zA}$`dU2mM3mWDK{>B6k5r-w^SY~&nZ2P?ks6+DZT&h)j3JJrS$^_diJ*H~pmpt5%- zmEx#KC}}@AD?TA%Ghlwe7;^vM;PMJu8~#5hkW(~m$+4otlc~3WeNkv~ZnBkj{GnH5 z;GLzISgO&V5lX6|x<{T3%106IZQ}V%Y?TTcJ-PM)&n#9S9A)8mV3EzOucl#%Yz;5( zC<#s;CMMPXT3m-{5gDH5w2Uf{v3$8Y$xF7eUG0n3-zERMXN{~Nk}$aNCjN%>pscFu z9Uk=|iWd(_xXg!`hVQxw&EYFe0o~W)N~%XYsExwz9Ch%rv&-ShwvnZLfU0`K*9YF! z8Vj;eprIOyO!KIjadUH1($T%*HM4S4$NRnfp6~D&p2vPxOH+$aE$qGxc0vofzp%cu z75YF>;Eg-;x7-Za+b=fM)+6}5NsP*zuELlbD(S(BcVglh{b&Q@%H(3;UxfORl$31u zZ3X@Pc%M04UVWK(VQ!8u&WxO-_vauy4TMKQ`#@bq6gegum~?)t(sKS~H|^+;T<7Xu zfgAOza9|6(8$~ZY(i1|;1LJo&ZUsa`Q3X)v2P$c~5m2udhkJ6M%YUKYG@jiWlDy~K zjiZtDkL)ZKfwXm$+D%$x9E|=>sX5f)3hIH;?_Wyx4MC-7%>u_BLbB68W(r9WQa0wR z^M(@P4+jVa3wcjZ2WJ@h!$f#pY99=AU`)Gyoy*RDt@1akH4Q+&5Olo?W-39!Tn0{# zCP}X8w?_B1;vQ?6=LZe10DWBWyWF4gY_|?pdOHBlrV5A#HFmp?=(Cq*%HB=6TF04u z*<;)>68>C}j#lEy{C{DAAI4qKoD-e=efV8>49o5_aFmNT z3h!v0#WkjWPU_+s+Ahx*#$e^xOL8UW1>%}qeasF2B>ISH1Pgvc`ATsyJ71(OfcU(M z(j*79mR+jkXqLH06cKVjY7v91USC~sKo&;a0cSXpK?PX@X(D(9;M6bIr;0cIF&G2S z5!O~@`H2O^-r(Jr@7ofgCbf+YiO-42`ao&IpiH9jzC}ka(-te)PqmuSpbtjnso3Wz=NT4h6x8+nl6A#W|4Lk#Q8^N0 zKu~04WI1wXX6AhUwdE)iY;mxhSbbKqS4^W+l|)=sH8H(Y308c^E3jEMS-nAxOKEl3 z`+p#+i5$J}YMOjjVOtC>S(c4kcqt#odl0efz%N%*QuP+M$~?g7-lu&0JS?u|Ut582 zuF)?(FX*VVtIn^M(p&k|7^E1X38X26%3;dUfqhK6IsO^+!!A^@DfD-O!WDN<@=UlC zEPLd(kLm`?D)h!97@esJPcV0d8z}{71JBdLz67QH{8zJcpg%o-yaTNG6?b=@$iU(7 z?!EG3pyt%&+xgOf20&8#7-vCrbTo0&_V#xBvqv=Am=Jo8Deu7d>yKN&P=;L!0{xI& z9@M{4znsx)VbQrtu8QR+((D%tGxWIoBw*OzKbuVn~RV#R;y)ly>X zyN!=`-J)VN|LmXmVttswmeEUY9eI5?-M1rYjd&}a<>DlUX%E&iShr0U^uajMKjXHQ z{8zeV0!k`+ih%VBHuMadZ$nvlYZE&0C3&THw(^*SP*G5|9kVHI<=I~SJ8^0K5F=4h z|A@ra?R=H}gM$)4aW5KM*_`a|vQ|ny0{nb?RxU>lqD|BH9xB^7WazPDI~=Z>eqYtn z8B6l-(ign)Bup#tO~K=e^z_%Kwk7ZBX-UslKVlR~5&j@0LOUfEYmLiOm-Ch{-J{X| zSpMDP6UG-s5gP_4wdmChao3i#vYl_r{0VYf^$6J1pl`O0cKT3+F@MTGrv5@IKqlGm z8%Lb6!a#MH+(&u$?(yhLjG^SXcYlt_1%00p@>2!|w1$R;?>~OLP**2y4V260!_)_Y z0tST=?Yo~1_8+vIEiCAYY6+$znNWUHGR0x&ebkhnq ztWPM0*wVyt6f$yJ@%8_37Dg1%`^=tqVu3Y98|44*ot@1cfu*;^-HGb(N>}$$$6} zzzcf8O;tPz)9*GawF7BGVR{U@DD}rH@g^8sm-hHWh>}fEbzac-duwvE_V+dG7}BGt z4QR+axL;tUmweRF0ISVr1EQn@iXJvpuzH!#~qMivDTfv z`H!q6rVP$`r|C?8A-4n~E|NJ+rK6|4Uk3J()?B&lz9KCFCe?DZ$x4Mf0?^{Q|K`nk zeBc77fW-`7{n@XiS6`lx>v`U{D%M4T(8ulccMq(sQ`K_*>rH6)Lhs-W*iG7C91K=4 zPg}|-1Q%ot0eTMWG;GfVOwWnfmZG-+D4r`p%X68GP_bLN(qe#tV@Qc zIeK`l)nDed3@WqW z^#JtP0!V1^&*!kW41gMQ1n)q>lx&EE2jPWhx7{;68lk6(!!p`);`s*>#~{dOyzy0a zofsX(Z2_|D*p2P^^BLX6Fk-f5T|FzxE?78}i<-|^(W&!^G!q|5)L7Op zlb8HcxQjalSQJ7Rxk>^Qwz9FFOf)i4o^{TwPd@3T*-s5r`Te49q-DP4!N_K7WAaRP z*s!e7`gOoQ#dg?WD7m=&$G(a(m>{pQ4s5;FZ{=^jUU}jAL^a+q7oeH? zUvFSD0j@LkP<|x$|(#@eS)9#+MI>CLMT+?cTnzo2~A}3>~yAia0xsMY0Q$a`(l%ApSJ~W zxA$f5Qcf|al)(+%@Tb+K8oKRE} zX9zWH>&^R>(5Qs5bRp@I{d@_Af!SP(0LLM&ChV*Ct`$mR+>2Q-!V=Jm#SQOGrGZC2 zXq2Zl;!VkF?UOfCh)w@h-X|wa>*&y{<66DrPe^~rs828v6cDAlv9(x_h!uTCkZu@U z(r{y{rp$QIOe3G)c+Z#+Q(83uE-ZHM$&LSDH{sKU?BCG&H(fZ!?xhM@MFMgA;9`30 zD?xP3BhxCTP5JjL_RAwI+gnGp5o}xyd8G@R%R6~04#$}J&idBWHlG}O6ww@WSuu$0 zFa|^O7i2GET_JgqvQN8u6HTL0kq~Jov@F`?l7lTG(WTz!T>DReF4lk$;=$iJ1gYOc zm&f})6YtrX@njLb%2L3w-RD1u^DEx>JRM|lB1adGj*Z#W*4^m*@kln8gMhmVy_}|3 zin9#5(>x5^NVnc!arCo`*Hp4A>bkW$cE_UoyIjQd<4#yT<)YA{kdoD1vBH07x|*n8 zWgRm6u|Y0K&80(yjriVu^)igfzm?KUK#hgeE58&36;f-=NvnSuys* zoB_p;a7X^;9( zW#opNV|=wXOplM%6Wn5?wPU?CE$Z6xPJH#Lf66at@`QszD!LLR8||a+GLJHPDwQ~r zK9$$85L|t!XzgAut#qW1m{uFzLQ#3JDjJPz>y{~}X>1yG;;XzR&A(}=@&)L%oC7oc zJTOhN6(Shq_GBKL;4RSYGftQdp=XleaXT+Zc{PtiBLig8$<#|`l0$ZAq(qtjYug;Z zPE)*;rh}JuRn@(0W58Fkw=yQ`ms6i2!Rufj=)m)i8B?KkK7a6ix|@Z^N{$8y5zE&NkNR>1oHuPE-4=MORL_d6h?2p8;UO|0rZHc*wl&W3@U2g0uq=! zobQ+XgzJ-yRxCyC>7a#VYF73So*qk4I*hVS6DA;Q`ejOT(Kto;fAsj2-0B-G_gz6T z6_fe8kcF5K=0E@O{SdC`x3cuPu_6t;)vT$-l@MZo66a;z`4nYZ$E?wX62^xLcrSOz z`|#pyv2Wqg4!>;0Lu@Q<4!{%__$K02aC!xEh$C@4wDsWf`o{QswcvG>;dT5fRnj}i z8lbN`LFgq<5M=ok#&^T*E~{_WD^yc@X(noAzna#@m=ELg#@Skt)XAcn8}ozy#KGY@ zQR*QH{Fq%@Z0V5SNvcIAE>+vr5*4n4d?Ly*ZzuI(v3M}>|KxGkv%9N=O)8l62ot;xtqGk8X;DCOjb_;PB8EFav3N9ih8>I!{nYT$3h#Mm`xK zg}lBCb@TIS8ISv6W_E?i#8M>lLowR44#wYB!C&^N{Ds8v8!@%JEL;1X#6FnK?NT&- zW{}KhsCq(1I1N0>pK98Pa&P=FKYz-A747i(3^SUq;u)R@I%rw6&^G+-Q?r6AHNxoq zF)!A%4h@7okx5ZAOCJ3(OSX-R{8u!ljR9TK=|&JsAd%;T2C1>Jajm}H)&v0_-RmZR zWBfr?v{48obcWR7CFf~{5_6}T0GL$@!qp9gcQ2XGdEorhO=l%76l!hLw3-XyQd(>k0>RWc=nYg{0l3@hdAY@#xmSiFqHI zNaGnB7uWJ5SF9>=I^wb~iA~RU+>}oY5=$xbXk*KhRpM?J+wX&;3kAb@yb)dGi{mZ! z(6!ZVWJ3WDcmTy`l?2h(uUip`kPJ-YA)j$QkOU;;>{u!REGG=!ZqW%D`UKVN8~oH} zVv6@&Ni3+fV4Bi7mV+`S+p!Nz=4JNV&xGm1P#HvNK1qeMz7~qIg5B2Xv#4BrkjBE3 zpzQs;#F$98%-RZsW}N=$KiY@1WwfVaZS`weNtRvARoYY4R^6&9_V%2JzVT;?I9a53 zgkm>dC29X=0nE74K`NkqUik3d*5FZqx$;Hkz)Zmo;4IIQTwS2>4;Le%(jF6a^s(b$ zd6_SGu&5hR6lUxKNiJhMD>iAfL@MZI<^cY{s;r9*Fu~w)NC1*}8XiZd7n1-!= z<$cZ*@2O1i*mlMrdH2^%Z{^i1Yda7j-X$VfrlBGEZWL+v{HqPEoA;S|i{3dwrm5At zwDj~<#-8e~la%3A8}a7pGx%^jK7h#!LJJcCB}yBz$VXmz@`pJ_cI3uP&+ARKz3 zakAYn$NNk=DfCmXK(~W*ru01IFrDR&xMcvIR19$BkZqpEK(Lxyi zO*S;V0?ZENzeTV<-<}|Ds4q-szLgmhvqBkT)F#%3f;go{YEa0u`rc*E3uRo7yEo&N z*jb7*9*jh1YS9vx94tGs{A@H;hkj+lBPnySK2a zaXzE03OmERktk)_nNp`esd}g30tXMz`$rT#N$qY7)D~(|(-@Y0QHLdi!PPo&;6Upf z=yDNT?i-LYuSrf!G$Y3Ae9n%niqR-cngY##H3#aH#P4U{&&fm|v#|9fC2RSsb8#aLNuid~TkhbFjM;Vhw zCRV|J;S-bI4ptA=?s|*%;wwr7fmVLU3#jr;n?JvvS){;zs(krW!z??1D4p*y5fFmC z2kIPZPEJB((R%Yi` zC}|9e-R-r~p&~Bw=`{*WvC{$J(&zvRh^6_=fCszZeLs#RMZBAmPVXeX>7$t3Td47V zfji{?M@Cx=7U!(jf7!6gd^#B8`_G@Ph;9MUc#+W34ojEZB?BrmAVDL3W^2T}XLGzc zJ_ZK2{8mOSK&?%y8-q^Q<_65RW*0V=Ji#vB3Tgt}`yDaHF{`P!&`|E8A4%AC3(p)+ z*8PHm@BT~`Xas8bR~NLDnmGFRP}(t{oz=A%TDI^x1ap-PXhQ&#w3}LR!D@G=%9qbn zPcx(CwqW=P#&5BP_o0_Aq*+O>XE-ul=F*JS+@8h@49@;Ojfv5b5+V^ToMZe@0lR=) zvyn_Krh@w~W?lT;9C4ddmUV$c)k?pP)yH`7e z&oF8|X%*P*;56eE!yW1ofJ~8tWKMwobo{QAP!}+~AS7-5Pij_{mW;XxPmY1r4BjeZ zVPWC_AzRu1ejtri_|F~P&kM_KNWHgNT;K6PoI3a0${pFSvmauOfbgj5)dU#s?Cy0g zOJa2hB~& zK@_*(>$rnn{?zNepmFU(*W0VpuWgC`bfX{PfCjHOXV6XfJYkY#JEJUQ6%?TlJynBE zCMj~BmLVlHf?NJPBNnhxBh_|r`EZ__#`GZJN%ci7x86%32D(S{Wm@5>pJTsSf#9g3^$HFo)I5gXa-PEJYphh!=h5xLgrMI?|tF`|%DECMB zl`lofRRvz2MK9g;;l0)V?$B{WT%6(2mF3Gx+VEQ2Jaen`(YwPH<2gz25*V4g#B(o- zg{TrP@6RJ)RDjJ_*(<5-WD^q3 zuXWe$Ev$YE54ZkXaD>onM_3r<=H{y2KGpAvP0)Dp8J6Dq%(g9%0E`r1qX~dvpeuWO z`ln`hiX!j!W$U5EJuU}aj`AiRU>*O93$;T>&=o8mPRF&0{!GLhU zFSue@AWc6gT?xCS<)f;-1gZjGFU3OG{AepaJ|k{i-K>^+Z`J3Z8fn9Gd*E#mG=r;XDkP`DjN7@VodT}H>JrMey8 zgFwFA;*_vI1;}p^Olcd^_p!JvbwWjd8VujFa+8Hru>>~`vs!NogKvn0Qd3ix&j(47 zR*gLs&Q;L6*Iqf+M4HnL6&+)WwKXSpo>#fv_10abBkOsh=T0|s}+vbQ0iLV5z-CLhCdyLFUey0aP378*5- z?#s-MCxX}%m^K^-vyGMG^i1y^u^BoWCM}rzKAwBqZ63pyhqzVLP&&X zd0mu-l?D9<+Ek>ot}104r?MtpS2@bBhmhZdfU4$c)j)37=tA4O3*00P2eg21WVd#A zY-U0-njvYY0pJIL~uP_4(iRTtwI{C5$y zh(N%B)#HFHy##}@YPWy{9IMV@r(HowL!S?*s0|Bd0rhxWu8L32v=7L@^gI+rnF{WLYM1YK0d@q#*|2un)m42GeR`H97D+Q``rBNCggzvhCN?|w9Q32( z=~OK3?0{CRQ1k139UYwtgD9X(TIp#UDNtk6ZbDkLlQWfB6a%;O9x5KX5k**S*IyT* zY{`fPb{QOkNoNTJuBItC%QZZrS7#TuXMs6xspvU!v>#A$s;ARF$j_tbt)hU?H^^@9 z4A1QBl98i_eQ!Fx)$~(JZXaXsyvMs*S;eKH*p6bsdcXTnHh^EjJt89gTfBe?POKK+ z?Hl4gJRMDPgNuC{M9~YvH}3juW|B$kRWvFWtSJ63&YwptQ6kpk*fcP$hzK#riQV8+ z6w}^-lCEc>K~Yf1FzPB7St2(I1<8nO}w*V7#;r$<4vp{mL(!GCmd*icOJLCxN_rI&XERcf!blLf;RJP8y z!Sq$d-EFY0x_YjSNUPPN@yZwI^UrJd@1s9WLelf(B%wk@pmr+A!fg+r?oZ38CPrNm zK(W%ZF#8cE5ZqEEP{ed4wYZj@`YQFg+kIZ>#Uzzi^QrBUnb?<^geS&#znWOu)sYR| ziT-{T(wXAg-3wks<)KD39wm8KHw4-`VUrJgMwBdo-^KC_yX^cJE`rUqXs4D76pNJ1 z%$Php<)@`FCq5g%yaTh8zftZc8!>(Nt5b9o1g z1Euy*vb;8okL-q6u`C)U4S2ngaCY>epOBqSrrmCu&Pj0on8fD5^fzzE7ndMKj~$Fn z``4UUogS{urac5gQ(-U_I+r+u7GVtC1#(Ev)#vu59uSTw3Y26Ll&YSs2k6?en#c`` z%`2c5v@@`ZllDx_W6!d~?jYcAe)q)NrNo)F6zC;g!Em3kYO4oQ$=o`ZbLZ#hFJZYq z1fvnqnYGx>kCXMi&Ijmba?`HV3r0Q>9L}sP5(mZ-xX42yC&P9+&L`lvh_7)YH)8JgoxR0lazVj9lDkwUXO_}r;lU;>K-EsSl^4CMlBaR2M4ZWLvS-q&(Ws}^5I;mW1e@8EIvf* zF;?z5Fy>AIN?^;TWr_fC+LXgVbtxm^pX$pGWN>=DDg3^4z}=7mWyI*RUF30^Oi@cR z_d>Mp%6TEW-4)i$;+Ge*H|;)ZlK8sF5TH;D;_`p!iROnt;> zr>90#`Ubrz)7$Y@Ep~XKH+BMdu$BDeR_7M{`__)vt~Co`9#?=5P9=NYG{0H@ES>vH zvE*Vi@BEilv_tIuqB1R0Ee0m0RsSg|1v-&CH-L6m?62tRfqA(Ph2bhXr zp+(co5i*nw@ReziD6Z}ygVxq8l;&7&q1ieVl%+DOhZCoQp;rC68@$er z8I#>MI8U8PpT&zcFD(#xhU(Foh=i7+g?ZfA0s89}C|25loT@KV1Y2zkHj029FQGJw zC6^@1nkxDzjOhaQxsGYK-oHCRmc~e*EB<{03=l;Ecw3CflrzmQ6>IAwxx7~e@gK#~ zyr7H{hbo$6)5#=`^mPbI5Wa!4I#KaR(`lKY9;l_bE$5JqcV@#7v>F304Gj$x6s#h^ zO`~RLl|1|lU35QFPGD!Wplm~9Wn5{|Eqwjpx4p^prYRi)jNaq4|NF_zjgQ7a`+fI}ltG?DL!7&7gqgWb( zyguJk^O0A#Kn(JU0KEg~BUdK7HNb@`hkGhO`#4NT;Q9QGH`gxFdD+KI!%(0)1}q@D zl}ls`P+2ZuJ=9XC&G}lclK3hu3u%4MBqq#dkg%%`u@jnnJeBh#UO5=pad|bb-$SQq z_$q}GSWt=<+4EfR!5mlpC|5;xUtTH5rjLdcmE)+}(<{SMgR& z3xm2-qKF&7fEBXj`uhHh62tkCT${|1^`*5BDcH|He7&iccCoY`)o?)Y;>JV7o3Fuw zRidd+@xP|`6z2}VKt1yHWB1r8FK~!Nv^IbBCIP?*Qf!(%-1yu}ZD6ecX0@Z?^k z2fa&+Nlgv&czzUJIj~(dJ!rHySS&1Z)WrAen!aS?dU6-!qy;czuLZo3<8Is;5Cv)* zLVMoaoA*{Q3IlDZiJrjDN9FO~se}LO2jl7Iw6ARkd#`~Qsv2~l`#(z%ZFMPWEM!;j z^Mj#06Hz7sN^f^)L0WCWU%`x1&6t&v~l+$Fo0VJtl;(1SzIzE3){7Pwm!Z)o6T1?L)#4VB|=T zUD5$opKyB@23<^E0E#s#U92qbVqyvl2|?d`d(UrgV<{;qNggt>u)H!iFC?VrU}M9j zrB#qVZq=37{Y4nP!Pch#$mkPDBGi6@^b_~;Z1yqaK%5S@$^aN#a!HIvgnQ112-*(gV$Ev^_$0U41wp z!GU`SCR==4`VmFKp(wag9_L=c1fRZtWmQFlKlMkV1PWTve;eOhxV8*KD?TH$kZ)ee z_$pHT7!2vJNFjsyZ@4&gupXIMERU7wCB}tGz;Tk@E}1I%d%rV8Gaw&=?nfGu30pp? zKh!J|=;Q=4kr*+eQN=)q^PP2pI{t42&FOXfkrj{?Ax6;Q5)%4u6xLz=YL&LyAZIrS zj=dCRB2&?JEq%Pl+ZI^HwqPjfuDf>zCea&bw`IuB;KkOW55W-d#$37nZ-ami9Y6aQ z<<{iK+KNf%{Y=b<^j2M8B&$89E1$LX0w{mF`ySWP?I&v;zQ8hO*Nslq1Uz7xgY(J%E07EXU+ZJ49RmZXUPS|M~^B#@xmq#5cb^i?ie86N4(%+ zqQ5|mtDp-Yo}YH_99Hh4r?51f>iTmS+H`@k!wcb{YiVt72R_{|HG&gcS21n+wO@J4 zOO%aj+y2k|Wq?M*q8v1+KRnwDI1)-D*T?MqYBxg^9nvOj&#?`L$vBN|SG1=ls!*L# zPAjk9t2v^;E;PjTRdR(5p9C8c&t`bru!2iXqV6m5MPjsq|CMz8#AAKrnZj)aq1J8y zEFrMfK5@y;&ZZX%t^Q@$!25x@ry zv-n82kkDu<$cjsSY&9mpF8BTbi^I?UY_h1jUYE^Dd?N^Q3N>`;WC|FnESJ9Hb~8?O zcS~MObHDwL!>qHNK({r&QV>bbld~kEfZaS0BoVHt?1O`S!am)(@f#m(|Lz0s++0fo z=2}c}s;VP|#1CL-lZEWKIHhhDgUKxwQUZ)_U{^A+f`dZFzRdN$jqLm7pwox?eQ7U3 zYAU9W><;Y*w`%yRlt&Hz8CNErc=-@1WoD#JIQM;y!frNy-eFK#F7N==V*9zz7Crlh z`rw&{(CMZZam-TAR{}P=)$uZVEV5@lQ&)BKlZ|Jy8=VIS;S@v|XWQMUsasokZC91F z5RmVAH9fBEx+i6=&Eg;JtL;=ynzw-IICpVhbraAZUGPoxio^+RE4x#n$k-G+hJu(m zjaa-ySu1&I&@(3w>0+sSE%9~;B%#5s{8TSbzOQ)5hrq)=bnW;N%sDlz&fM-AA0G$x zCTckrGBPr&4@u5N=&@Vd50M+;biOc%(Z#ip}oNE`Jdp1ZnFSc zd%tMsqo(HR5@i>rf}XVw-w$(yw63onZG1jf2XmRfpr@@Rji?AAb&u6+o((DKMxSqQ zuVjZ|j-~L}?9gaGf=Z?E1%S@^Jg~?=W%-e5>frdLMeH7tLTqe0H!}(_bLw`4l7T_X z*Iq!I0z_(eEV~TWs;iiq4b4^82W^UfVIN3Xy7E2cbj{VdF#4X4ilpC;Re{g7KcFUkNPh#rx z30Kl|LdkZ&d0eY*O)&NHWm2%Pt_S(WF~*jA$$dmWrI;B4Mn*Qb4Q5fPe2BNPv=dw% z>cV|;{j&Nt%BZT*hx4uYig$RCjn}UwTzP@%u^zmirW!;ylcOW>I<18757UuUF<{al zlZAUy18>_?T3Qr~`aNHJQP!h_9$svaim9@sQw#Nd=-bYL#`h8olJ@f>cNPuWpl<#= z)CQARj{?F*Ap>Bb=&newuUE&Du(7d;NKVdg^H!-cCj+wxOp`^gk$RH2ddCdf_>k#$ zzAQm}^K8Ai!sj0SfdOiCGS;$f*ZDeE<5&G+mbQa8Xq6-62hkP%*{BP9%kx5dq_v4Q zsUX(#q8TApR+MP@toZt_9+WjBoU=GbP+?PQv} zfSlz=o^kF&&jH9#G}iB9E+dK1Xn>|$8nDisB20*ft8esnA|qi4<2f(6>+XwS3ITGi zTpG~@^2ZSJASXz(cqGv@crTUeLPmW(U-#Cxp`&k?A1ivn{=6p<%jDInr=Ob&m&EAX zzl3c`2)cxdsR^)Y{xxD_Ko?eSh+;kr6Fx8e=!WY)$&mN>s}Sw>*)L1jbHxQi4S{MG z)YG}1GH2Z!7+mEio+C)NRC`F$o6`Y4*t0LI)OJR69TRunU7Jl<2s}9I&_`J4VF{g& zKtsp#9S%-Qta0bQTx+jqnFy8h_5({xld)Tm#6k(Jjxa5;w}4mZs}Dd>M1aX2u4%a# zDMl*TnAMw1`=)@{EAS0m*1Qa{k?+|l&Ug0q_P|>{I10ZWPpEpuJyLL`!WA_98Sf?a z@!5;1GTIkmHPq5yCR`%sh64fu95+iko@$27Q^i&-Ja5!(yK@P&h~cNHOonuQFZ6|M zcUjM@-#FbT`0&d+IMwOt@n2`0*9g5#WNgr<4<4(8u%n#7D`Vje=`+nL>TBXc^%^XT zO<8o;pRGw4C;Lr+wbb47iNF`I+fm%|bq%|o_oJYkJsD8E@ezE?AF|*4euAt!Ibw59 zOpEDNeJOcNDKH@f-`96zoBBE62z={v77&DBl5k`6f$WEy>W%ccygEOJ#xH}SRnRRyq)eG~&>IyAv^`Qa`8eTSu7PMqA z^zn2#P7C2Qr?lCi+pJzOu&votg{FESCx12RkSLH>$^5Lfuaug>eH3qS_K+&i{Oi$( zujF@4aPNv#l=DkV-+ZET#%<jC!xu8AFs*G#*Ei(jg3EvP%;Oc19 zR8t`!wLYk|cSdOsa?$jlo&pDnBqgLQhYJzjRinY5H`bcX=ovA|EXxcrN`?ZlMw>byKc^|1lBIp)vE-v=v(My_MAG zclqoI_hyE7Zz-tT6xFUXZiNRgM?Muxfz#8_tk%Xca(&!sywT*nyt=Bm-nVrDuRILF z+Ak!D$Aod;@={HhJIH^~bRvYNX_M^kJ3Q>cb7G5ISg0=VLH{pQ5+MZF5trM})0%Yx zSx*|P(vDUVm+&0jR$e5_jHLeRJl&P5R@1lNS^6Oq|LgSWQL;}OtOc3pNwmlhV7LSJ zzw8g<)^Z8yAd?1ym#R@wQ)lGL?LG%E7R=6Z{4wdU*=MlNYw9m-U@O0n?XctieY^rw zt;$?jSy{P&`zJ?Pv)Wr-tly}?9-gKG$yvLmwTgr~5}urZ7vY((?Cb8S$jkpfuD&`d z%Jz#EMjE73Qb4)|X&5O5K|s2@q`OJ~drhO^7)fX?f6cR5`UusL+|9pmj9%vRtt=GV7HIZSW zV7udm`${#u9M^~+*?m1x+pF%a&#d%+auH1(sAN^XBuo+cvt+X^*VrPK#t{V<#o@c> zzoQ>a*fVu+opJiiRr%W4VqloQkt+3LapU6Pzy_7=p9`U6z{Y{YG&DS{Sm1*$=zf_) zhHx=Y+Y3la*?phjn2=t+d8_FMctDssbAYxB2<by$SZp2W3D)d+KNRL4-Fj8e~$yp1gdh7uL0Lf~(MEgRCLD3MAxX@0foj0~<)N>^3Ib1Z2>r7s_eYN47mtY8Jz3|6b^!k%b_FP0c!;_0RXSb=8sufloH;)w$KP+$2V^p-Znd zj5PcH6T~2Vte~Omf9=>f7WJ1mlT@HzNXxa5|Dk)vh4>XG>n}y0DC0+5g;}_QMd3T) z8s1q8?Gp~Y&wD)qkUxb3Pf1B>UHSL?oN(4z^7*J+@V?_nEA#N5Zr^M<`}-~>DF`c* z-y<_c0@RV4jr<~*{6A=At)>gU0gxY4h3BCp&#=3CrI0_+;(d8^90A^gNc-;tHmKlb(+0SgbF^XoFpd;}{643d)?-2rLfB|&+uTM5qq5~t zswB^{0+o(C>u7n12!cF#aX#!6+|v_d{d%!Oy^oONSC!55VQLCJj=iT{6JFHhuS@FKq#rC z{WI}LuI8KM?Im>;lILb0&vzJpoP9GBRDC-gzJw@QZX*nGV3vA2ZquU;Bzok}XW{CW zQ9^g8+B&QW2$thKdKBKWoW*zCY1^ymq4D||P$0{S;&k27-adw!M%(m&BGUgFTSgHI z2p^Pr$c)LNf|)c_PoLri1qDqluB;%15z##V@kB8eAjL1BGYb)ao<;_S!fnhJmHyLA zgyCek@X|6)3u#6ApTcMJ<>@76+;u-BGA{ji$yfM1+}WC<_`#y+JXF9Kvo1%3v(X5Pz!aBYvJma1wjR^OS2=Gk=+IwlsN%P_`?;1 z(}2>q`K9X>_<-$HpuWfK3pqGA*aLD@`d}CWAYHM4)ltJqa+Rjw0eUDfMY@k%CgJVx zDMom*h0~HGUXC8nD|E^3ot!;$l!rZqJ}*MFZwbkyFU9C~cBPCLt?t7p))$V4`&WtH z6A5W5o2vVb)m#D)*P|H!>~PY~`}9%Y9l|^B^;GqTW)OSB83b9+4Y!`s8)-+Ug;Yq< z#DX!J#fzl+(j^NNmSO2cndjfdvBxz{ydP-<@6KG~w&wcN*u=zO&}j7tj3%+D&&*zT z0Mil6BiG;aT&KFuJqL~(=wq|%1EBQ#th(p}_?R3joft7ZONCj)|2#!T0&4s!!O{SY z=$a~y!#Ny6)b2My*Y8r_E)43pPZ;JVy@|h>{^^svbYB|$ay>vFyYt@6@Yg-Ep$&;3 zWNiV|ik}w4bX)F*Y}B?F$N~MNzEVqx0a^Q@p1VBwZS8M4MSX!_P473|u~NP2Mxyux z(xjCi6voa55Lz0n`@#_&P+a5*D}xF$Grw|Ff*E0Zr!Y^ZeSN^lz$PQZ0NG<`#JkY6 z_V#w5t|aN;zy*wn;+r>_v-E^npy8|)zzdNe9U12%WIo2rNk(nh>&m^bT~L9cVz4ss z69_Wp1?D&UybqduRs3JZyVbbg@^q@GE@boWa#|j*udIbmUmQ1qt_eFQeT)++e-kh~ z4E!B6{P~YlAlze7xCdmlBxk1CA3Oi$av~EXffw&r+)vFHk5$GedhB`XVr+Fr$@T2^ zOUsuODNw`t2~rnug*vMO$~%W5opUiBqK^Eh<1v7j0hhOGu&?_3xxeX83vAK<58Rbg z2Q6~|S$^gCSr5={zIuZa$Yaot?HDFHd|ij>cCwYz8e7Vv_we^_u!?EYrPHQ zEM*#8Ojq!2&h^PlT-|whly)Xv?OrY!6K7w$mz3{3YR2o@p4}(b=y=#T$}RSOWI z{Wd(lxQ~8Wl&m9|FI=(s+}*`zX1g8o#1mh>ECY>v1Ymq0+12%Nn9ewD`oIsMbVdKa z)@os}eI68qwm=F>?NfdrhF34oxcN^}@<(wRqmz>)SyE9$(XZI55nzgcbO9mE~KmqI$)m_?1AE6nN2FZs&)iXGY<_hE1A3z5`xmrIrG{RRpGmZ#P`;Yzl_ zZCQ#1XJ6yY9c`Vc4?*p2D@DR*%)Q(j&MLN)L}o&KRlIaACXl9@vwb1hJE})C6=;-l z7WV>VrXXuE#G_U##^L^kG_%&&NvOred^TjEtlTvj}M?;vLHIXd_&;? z!|%b*h@2n9eAbG^!30cpnc_1`bxt&GpCoI3^PZ|6K-3=vUD?_)xgS4<2T}lp8ri~F zL(jb*?~zi3MUI9NG2t*+;i>>N`mU+9m=gjpS(vg_;@@l8KiMqaa#3gLPQH~tdE>ib zP?R*J49fm^iw@`_%|8O{i8>fOE`@_1Qn)mD*(R_a-e~ zO@Ai%gN(VWjaf}2sekdE)pZ6kBAUp=uTlw#j{k?Zo|s}WvmI0W08JtemWMP8^s3zN zw@#c?Zf9mY6?GLL^z!4u4+pw*?fHT>nF)Ch*t}0=yg#u1cQWl3=4=%lgt9%%kj7or zRyQ2!%5f8B%$uVF&z`YOE?cRLjt;`RK3of9_LRsFM?TN`&jHvXGzg{SOb<;mUDhUr zAOGv%RuBc3>G;s%WE3z<%^^7m)eRuCvcEZB<#q4b+OiVbYkDws4PetxpCvvZTbl9K zAh@+W{Z501nOZRA-fjW-D-sJIi6h>lprZ#GyWwxnI4O%JTldj&awh5+IgYi!;!bd+ zSY9kF$9URZE%nyG*WS)coITvk6EriYU0GS$j}o+}rJuL^eKsXJfOShYN4Yqc*5L?a zRfC@iB6w8q153kj8>QhIt&U*@UWU8IJN*y@O=Y;*N?Xb{Lryj}Ox!ho0Rc@sI7!;K z?0>T!K(!diGrf3HBFt$|1KF%otxUj|Uo zI!@xtNF;>TI;{tkm2odGe^fu@*0(-(j)SiK*Pbhx0eGgKX1;~fU}Q;OsL$K*os=*8 zA7q@4(ycEQ3s|6X5v2~^DYNX4o_aV3KKLObH+UZ+xxCPyT%?Uxx!E52?lLYJB*0(v z>P77|{k_~I>+_6ei22*MOfXed9G=cdl0e{J>iUjBxflg->P5jIuxw3lTA#mQ6!d

cGuU} zVJY^esv`h^&FSGCku8rdO)+PR!l6eGL^HS=T~HLH#h$t7?Zbm=h(>Z&QLg;>^m?M3 z25@c<5n;~Fzdy1udb2*Sy7Zw~TJ@t@iW&$x&o9k1;*ugQGQKOFq`%(rDcb5{GasKN>cPpAk+~NpJg*zikh#Vrvc^|S z7bnJxTu75(ITJybV0`G)q92msYpjKjM17vOjr=C_$BKO3Rf=b@{&xIJstnWU>f&5HzmNd5Z*EVf;%f#ngZs;Z#PWVgn(S1=1ogkQY5nA?N?=&(noExhf? zKr$Y?nHrf3#sy^n1!>3tQIUVL^`)uRPQX>&6Er6@e=<&Ks4;_Y8U6&CB(VbheAi}L|f)1CH&%mBsBjCnHRpz~gIP_XeN|)CVc{uhD zPexCFc<&D&^#uu+w7Kuy#!imZwMM2fHpLwI$w~Z7OlSlI1jXn(k0}aCR$LQ!f!^w>qm@;GgV{5m;7eg68%yL@M7;4PX4H&~sQ&)Jg09Fn#1q_i(Y|>w ztV;8=uT>m)R7~21$14cr&Sx4D^Oxjre1*Th)v(tlJkdL3tTV%ewp}F`pX8q7W=w?v z;{#2??wWA|if^XY{G$uke3%%}HD)D&ZE>8*?w6Iz2?^_k4zMB|RM;PAgRBQQDow&dE#lS+CE>nM!W-w4+@LA5MC(NUM}tEKDMj;IaV3Lg`jK0QTlVRw!S*S zw4%Zf#qamylm85VFaj~0^~>O6{Iz9>Y8oc-S+~&F8z16MiVGYr#`iYT=j(n$qy?*_ z!`N=S3>?z$3o(-z2dWW=+Upv#+Z9ez$a(6mk7lC|i>clT@uH@UJMGLZjIEd!bMr_h zzifH@%3X!Wn9fk6A4X}&cF!3lyYs(}@%-*5#{1>{;?OaKz4qE`T{*8g z@3!k<8!dRY0d!$rMA}FV$yk^3Gf$aQw4b(>HnU0-_gFt?C^{q7q~qmO)Vy(iY$VhB zE3FbHC3SXTeCXNNv=H$G-R1s?O)KK{SInQ2)dR1=0A_%OWw`X$UR-r$Ts)d@b3Hvh zKUdrIQW%z&e*SU`U*Hf^H7^#Id*o7uo2Ps4KHSxMFbDM|y+~eepUJ=Z#$sP38qRtq zW#hbd_u=F9GqsfS8eows5TY`dMQ_}i!mkSE+s`L4Y@?@(&_zO4satGqLb#w@w|Nm! zTyXJP8h4qXC=l*j>Mm|rn@ycm0ErJzB&DVH!yzWo#&Ps>Jdx?SDrz66y3Ltc*=0b{ z9Z4bd(NHFhp8832FC)hGrV2yh+$C5kzB%2s5rEHzhkx9- ziALd4h5 zF_(I@yvDFj4?pA-Jv(}z2t9kyIe3?VC$}wfR(yWQQB>CZ(&hwXhH%@Gs%jJ#PB+XX za$lO2Uc>fpQWA#D4HOFLJ4WlkK;f(NC))vI5QtdcL+!C%tQsl=HZ<^p>V0Z=Q~Z7( z)tPinV{Ud%mHUsmgveUJFEujt<98P;)uRY}iCO}UERTn%?KOOLybj3q21jkBzV5F! z`tB&%YlcC+6#5EZU5P-WQpK95!$K#P)wn{Mf#zRmO3M2ZlrC!`UZ+TSS#ST!t=ALW zedIUa7(~y{;Q)4hrJ=xHc)vmMhJRM@Xw7a5l#i@c^4GRkK1gx-a|Ro%EMyhgMbwkP z>_j0cCLLl*dfO_>!!y&=Z`59|=MNdfrr*6veIS|e@NC%Z1Tc1HBS9kEF5#0Gsa_da zzBlgxLEXfq{Wmah0!lB2)-vA~f7NE1MZpDT3x5z_CFWU+FxuGrxLTKSP?<#F$eUbD z-X+>NrYw?2->tv;l${no-PN3Ux*+>HFTyHQ3K=(c7D=1cT&u{3;6#0d;Pp=p!34EA zVFV?89MPLM{0pz=S>C5O-My!T)h#)1Jp?YeH5;MRc^(34j!?N=%sz*gJo_+i2ZH5wvDe)*WNnAKEIb@)NWnuTz<;Z0TzbA{T zK?NIk{wRBwwv7kys{rySoyeQ5inrmE_;8v^(L;wJEU}eq4#d;Uo`aDZ`mxSm zS9N2ZjlVTG^(2|iyPu-kIP0LBde1=Z2BaNqVgs?drKDsn-+ZlIO4rp3aD>d5a(oh@ zzT%kB1BL?!`M#mvV!~o6=-#S(tDsqid-f4OUftE-^PRI=yBf@tcmbG|LBqkv!CnSw z*2+Gj=qVOSlFO-nBn9=G`DRat{Z_mu^?2s69yG|3-hodxA_|HZ!t^=@W}fr6k{-8B z!sI90#~|Zp0iqj<j)Fl9-`c^&2}rvfzM%qk!Hy*7yubs9;(*4ziP9^DkX z+uNU_=v2P`@Dba!ajlX3)d0Cb1~0PvRsHVcOplKd7*35`O5G2gIHjA)P*!oxit$`J zCavyr`!*nkm^hK0?W0-Lr5nbS)_&fgV|5jGKi+=H8YJ0(H}*_cJC!~kvja?6EQBYu zPq~K~-FYn|BC~(1UFF;MRr6@pAot2j*HoB8Ir4;%iX=x)A!xW)xSsmWFbKzdazteR;0Oe3Jr`j z;|qj*YRg?!d3t{7x)36DEWPgjTCkDu4zl0u=f$7A)})d>+=gWt51$H?lBF4FwQ^PE zHm(a7)zsB}HtZ|HdThRAW(@}M6$6>F*G)pmjMf=pK6gYv3M097x`W`x{u3ni@rpU# z!oa-SkJ>jmRorfV{$sO^l56{G9v0)VhcZMAcVXq!&zS8@bZt!8SbN`c9rU7dM2^-F zURQH1^tNLLotV`%XYU;A3!eePTNq3ZG+?kExoZ{G)3=s5JR#waRaZ|-&w*=QU67NL zWx)s{eC6KPaJm3WXFlDuzZ^CQHQyD$HQ}%-2VBZEqay2tX5Jof4_tSkahxCRwqtt6 z*9x5|Mp%DHzG{j>b<(3p`2O6<`p#x;o$cB&KndUPied64BscKWHSTm|!T32I)fYzM zs)go6#V~a*&SFcqm2p}Ee`UwM4!rGL&COixxGTQXlq{2dH55t)oH3`ZoDij+RR~SX z7LndNljIN(>%}7E>fS8w%KTWYDqoeQcSB)U^a;EpTqTV3HZ&V5SocNERa6v}-F(`w zC*kE=5K*34_Yf@UXQWBi+eknla0YpdwGjUwc@jrb2KJsz0GqUNG=EnAz9GzYW!Z|=r znxMV&sitAvo?xS2Bj?yF1HO?Bt;=jNheg>x}iLOWcJ zviCg@(y8EZL?vaTTWd2pHL9@hJs*?2`ed#2l4P0yHwg2^Qp4w{(zk7|8UaVX<;F0L zzMiz07#blF5t(-eGh8R^cEdiCKUhD*z+x%(b7pwXxz)Yxkf+X<_=&EDh=XqY2pa9R z1!+0$3=jPy+5KLNC(K(P>gYSvzvVw^9z5HdsYZ$<+i@R3w~V=MQjAj8z;3S0roLK5 z_m(ssjw2@3ILex|Sgf!`NcCm5igesXAljNN!Kudv#>{gRF9-8b*IYvUTZnG#Wy{eP z3e!8fEb3ZD3b>c2=w8`oZ6rcE1_nNRdlt(h^HP?Ge==Qs(6VPo{NxKWO&Y>JI-w3Y z$>M@r&QHRQM$8p)V)9k#YmuaLLa3sM<7 zq;TId#@ZzdZ%pz$ZMSjydnD=0<1HV2@-w6N9Zd~K_@%!`&w?aiUg)CUu?qWnj)!Ir zwUay7Jznh^+pvJJH?SJ9*An-zsLPKhOxs6mMR%bpMdAh@VnCIwsNbe*<|JxsAJN=%UY>TDq^IWN5ny5GcIr5{NU`cHT!dde}I@c$I|@< zR{nh5JUBy~#w3Z`5nhDN%V!^`m6jmuKwdFhv%Gg^opSxTev7<*-hI{L%s;J>c@s<< zb0t9Z#Wu~j+sW%!PFZnnLKTG*2QaLSk z9)k%flUX#!UkAbN5sWY->pS+cv5CUALygG%&M zLFdV{ej%i#y%Ud`&+|;oE!l;90oC>|NliA58thdZcyOpwuB7y0w9!KR!^EEtA5QW; zuUHD9Y|ts#!i14sh~BcO#OYfwScZ~d#IVwpNVpUjx3CH= ziE#D4{CikE5a6C-T&V!h{a6EHpm6%AMe@8egfHYfWu1lF1aV~M%2<=KKHK1}pRyJ} zzU(zDGiR#g49%M#pXy^EPYR^6@@k_wW!+fGuJ5ZUS?#|TF7sdMWepJx#Won6(|H_6Io?6I2PwISr9?j*&o7g}61zeGkiuT#Ymuy&ceN?&JD=umq%i!c7^?&Y*zb zvmSBPV?#o_b8txO{3V@tviaiIv-c;1dWDs~xelm@7PrrNN*s$%2mn&7gz<)_=+WUq z>91nNGXx_cN>!o+Kc!MJQO5c+McQu-GnU?K+nit9hz{Pc{F&VXC2>3*xG($4(-FX) zd3xkNj^1FaC_V`1-RO0%!xNLRBfs5qY%pZtjPd>uQc24+ksFIR#7U^;pW+g+U zGB1Pn55HGtux^jc`f~;I?J!FEW#IgTfhmYV^T~XHsAFB#k?JA0!k}eXqHqb0JHIAf zc%3E*fgdDn{_i9^#B&AhIou@AOJX5F8w6?Y{083=5~QG2lWw1c{vlbr@j=<0oJRRD zg&@6B%|?LLjhwYDGt5~X<|{c_PISBIR<*bE1aMwGQ;xKwgKq2w9*GDOUvcd#(PsAV zL|QBtbl;|ue2_}&&4!Dz%Qk!A?dPjEG3;9A74z zS_DxufCm_4GLyje`-}+5=e`jHYz3N}A-Vns@4FOEUg3Jmw7K2b%Jd!IrzLfkm;tFM zE+JR0+0Xp~o^HHDPd>h9^3b+liFP5SEERK27bmYuGOzEoks4k@P*fbvj|=C;T1@w& zQ;UYR#%a;t4lfll1gl)crX9lmK z3cZ!x4{~UIfUdExz{{-lQBOvTCT7AN6`^J`m&f!ov6@vc%!w>hpkKW$pFM@LVY`9x z;)fa*Egv8$!Q*-}h4ys4sENWC`i^T8L;F?aVB2cS$MAc9<>l{+RfKXdxedi5)1jEav^auSI@NY^Rb5@Bnd;P zUR+2Z@r(sD8)`J(7`d{(e7brPDK?=u&+>IdOrhs@pBrR0>SC#zA_(ee!P%giJPCd- z$`T(xz5eT5e6ZId4LPP0YHkw|E-(KgBUg|bRawWhg}h_OX#AvVv!kl46vRgI8R4C? zZ(k=W&CD|}iVpx5j}Bf|P;;=g6&Vy+e7H0Ndll3)e7xS$w9h;Nli7w*^9}>KM_DNf z=q52k)bE@exgOkO-b!SUMq@7r)IAkrP=81jU)vL;|^Z;?5dVMB7d~PIoN&3MW zKH17>1R~K?$Px1NjP749fSqXvMxpg}$LH6B4~Q?TIp5}?I!U7*`F!GR3b0KL377ia zabaX^o~zZoLR{IFOU*|c7jY15e08&H97(ZCE4ak+G8sFrPUbWH@`ArsMD3RSwJIg7AUi&X1?_x2rgfOMeF zYU|*#v9#>!%=c;R`>xmXJUoxfUY1;*S*TXqG|by4O|9ewL#o%_9FKdT*4Ny|RmiOHo*tD^RA|#?!^ScHmL%bc+vt z5>;jX#E??(0F$i{vkJRWO3jmPXr70{G)#_ZUJ<$Nle7qlcv_iaG09D3UI?s1^Q*$Fg%Niyi?U5!$1SaNWp&R zAN2dJz1Q{~a~*kmBqg7JZ9|NBJ#YP`n412z*V8|}YjEa3&7l9Jd^HL*A`tJjJ3d*K zKoQcX;Yp(MO3SW*9|->-@jxn~O=$@`(DWyHD6Lk!uC7tm5l_Xv;}25yFURfdeWgn3 z9PGWNQuF1y8S+ZsZPXD~GI}SrTE>Z&sFiEjunWcwg1&{(q3vix<6vER%GS&T5%V|= zV#>kcn>2u~&whPkKjCa*YZ`86Z7hdIJpKib8FjIblzQg|rT;hkF{-D<_hJ zaU#`Zaa!imvBNzNR07C$q?KLvMnkbS|MXGJU{y|N;Gq++2nlH8$9o@Kkf(WH)fmW` z3QIj#M33p4iSc;#BITW}kkaQ#1x;oDD?zc&{8@NiJ~80$#!&)wa!~7y=9uWAj`rJ2 zC?IIULr5fV&&99_&{_SsT=vDwNw22}b97MUzM(ww|)Zv zJ(6@KYs80ZC>Egmy1(xxMSAff+~?yi_H%Jhfa%rOlEFXYQIWL~Lv(-f@yF}1{&(qX z+cYt1e~sFj11u1!@B6@z4zmx<$|vL4Ot+lhqUb=E@G~hFzT+LNFea_eI#es=r6|@y zWL;_Ti8{sgH?&TPqYPBIStS;3O19oq(-}} z1Lcy{{FwJ?`{=9p7-QIMkLMENEgMQ->Mn00`Ofl{>@H^4b03|2ObqZ4oOnXleogS` zIgWsA9DbuXlE7_%%;*kP`S-aBH|iVi0utd%N8?QxVVI4pCze``xJ;jvf3B$M>wbma zO&Vv%&e`OKnWQ>C%-zsm{_|OX2dIE-FOiTYfIGyh-Q3-YAdQF^6|+L$nHX6bC;t2k z)pI9lHAc>u&l8y)yCcg=;$B%2ltzQRrG!@R4Ghr~280z#O^x-7(94S*i`8D1V%v&0 zXcW&Bt0U>CrTO?J=a{ zIB}Zm7_%uUKP<0C?EV5CAIbHNs5XF*@sSDD<#*q|{mw5JJ5_gQCji}Z>jf-pw2|^$ z#PdpK4y<#5v&z*qG7gYi_}Tv5oUijLc}4dR%K(TAdda{sTs=!17YCKFPe^(tg1sa( z+OsSdCdKN}N;&9pF^9^XDy695#dlv>OS<9L9pN#1Yn0%~kUKZ4 zU)D8Kwd1hQd}oC};1zAG12_!C(Z(&#p@1u%V36OR!<)1z$5%%s`LH$?fl$ z{j>gaq%%C%)ywX2U9fN>Bzn!({-ZE5i-*OvLiaOO$~po$g#&v7YM7!$m|R#jJHXTx1MIy7?Vm#XV89 z7KDb&KWs|Fd>>5U0lC91#Rl*lp3)Su;Bc0=uZm24ol%@f<%7RAHqG*L*vgmzp2*~ zKxrkuCngH%vmM$7IC3jl+ieuk9Xir(vRS)zUQ53A_@UVSfY#YS%W6k!~DCd|DpXmel%N5;BvwyqAzd96| zy87ZpmEqffsK-gquU^UY$u$xu?xeqXH|_i9a*83JNmWI-S(vheaJ~Y$yFjR4u^s?( zJu7kcD$85G?tk06nEk98Bbz{`Rrv}<$y~b%5!SqvevL}cQ6V_x#!*jzY=hL<>g}I% zL{B*n>^zMSB^UIG))%&y3M)sE-=_5Yhkhh#`x1;U{Bc8&v4`{TuZj-U4y6BD!RHC2Hipe=)Y0Vq*}Wx74aUnX~ta-Fuflb41op*{MLXY^|S+=VI;i zGlqsEm*`JqELi@{$*w1A8i1<&(?r9}iC8<+cYu%GlI~DkIi`Pf=**twFrgl(lbtQ9 zyW}a5Z0_PP@R)Vy_XH01nl8)+mpu4VQi3@!FaV?Q1w61Yodi&lIog`QItWqtKNl)I zAFso&FWqc^Lh^0pyW)t@@^tF$WVM7?xRuN<@!HC}Vi;q}m#k)hj45?IH1O8aohZ{M zVuXulBIetY>KR7ijR+)2QG7l)BDVgMQyqy4wthM zt0+JQ(lR=#yo=M-*|`VgLFkGP4knYS;*-98{R$g^)&=09c-r5zq_CPuCxZE?vcub@ z+j4u7c{wE`#J0H)tWId6Yx1v=GV@Ub{QT%U4e>PS=qg()Q>%-LbPHzVOw@OnG07&?uK;zh!XhOFeDc>E0KB`;({tfU|$h9NIX0}O%WRsL`gL%hp6DA%j3MO-SChYA(Jp`R2 zqi3ljL-JR>ZxB5#P=~|C2bt5PbIWEUW0dkXC7k!Uwnxn?4u(@#O*h%b&9|NA;zvoI z*%0!3C8XDkxdBk)!reXa_d#_xx&TcSE>WR^KN!mLL zoBf4WX26l7kOxq=U7-1>2ZpMMqr;(O9&iY#q#lsvNWGfV46G%lPaE-SKV!Cz9uO|8 zhV}0^6^6+u-!NXslQ7_s49Oa+!V^3~K)s?V`9Z*^J3E@hY~OW%)7IB?hUAJPi^ku7 zuu|EUg>Sb5pDk0ZqSM`=n*agz`-f(rRJ#NKX}H^BHE!2V;IIZ#sDlz%Fqb&cn0N}7 z8dcuQB|)pCyA~JgF#jg+wX$+Fn4NwN6tk%6?02J`jyJ>rT}F@dGs!A?)~WTO>@F~) z{Y)yJHS8W^b!z`12c{c5Lh8?#j0gAo_$~`vfEvGHjecA~sd~vdm08QlS`iSn73+ld8t_V-mL*{0i z&u!^b{PF!kM@Ym&KU{bWRbDU}b1-e)*8!Gtm*GS~hL;i`*R7VLR62o}Y$i<6BRS!= znL5QqcV)DnlU=lw2Iywf0F-JD&-KwznS0Z1^#E74IMDT728wFz@N!H@m1a{^Fj-^M znc9(~6-u~`k4`^NN#WG1KT(SSS{=ikh|o|{0DDYMw3{@7-*x_hZx{>7twf&zb0Al% z{I_Iq1hN51zqi2QX@7%&Poh?FrKDDamX_KV4ffr`+mG9!Bjk#8t(#UZPHZ(tvBIMn zVilAf#vQL-k>Vm-6?&|eY1Zi{t)owR zTMg>#Y>)e=qd-jd(AoSsTmLzWmg}+7`MmdoB|rlc!_vo7-rl8Lq(M_PW0xQq&nlKB zm7u&-Ll~sx3E;%ytG!85T|XfeFni?!zN*zmZgwXi(4Yi^`WQCxCYLtmF#y=miNvqj z6&WN>34!~f=#p8d%J0E7RhK2f)bOQ$KHC+~?4;Ozf3Wm*1+@M~=(CF>6|}_FLFm_& zZ8q_=@<$uY7IWpC+Q7a28BoFe!}z>rQ>VqF@9TEm>lnF1g=H@TjHRnAjMvnvX^MR*8!t{FpZ#Mn8@Ze}f z8VIDbHXVZA$oQAxA;vv|=|RZK+5VzFIN&S8;4Z|?ModEGokek@zdRfOqb7H`yKf$2 zj@MXcUoA7aTY9DB=H_O##G-K9YR)~KoR^&A25>=kf&SZ@(NJg3@iCY|&YKGaIEpU}Dl>EM`zA1IFiw_sQtPB%NGc=37cSkJeznNByHo zlS!@vBrR7$qv+fD1>2<5M)n)>E%7;`h{@XpJ6qcY{;lfj-vs+P0H{T+d z{xSrukfWAhDgZa^I+SWX9Jjd@Hw$Dl$-tDoQ0dr3B>~21 zEKVz~4;FB9F#kcC>S6bDTfQn<&1Q5HO*v4Oh>lejj)wcCN+Yysf_{>#$fV)(Qh|@{ zGDZH!cV#1Gkslz{$loSlFil%jW5i+83acn^|01VofU13-Xrr? zNH~*gPI@md@*pjUL)vbJ)#kQsfCS=HersbeA*cM0V4_W=cq*A>Nw_bn72`Yth(Av* zMX9T6&lUpj73Tbqy*S-vVyfTs;uN7y8x_lvNPsb04Q*JTY)wcj6YLJmZvdH=MAw4I zLO6KY2iJg#jdUbWiG_hS+AtcO+WN^c!(FaqsI-mPr1&7HZ2 zq#t2Y({)u2M^E81t-*wY4UXDSPQFdf3sH8}wq(scXuspS|738vTvUKTI8)MEzHN{3 z;J8y0-s$Z`sQK3pR8q`x1fZ8(r#-VVR=iuKIV}9~E>2>H=~2BL=N=e4tNv#H9cprD|yj5Ib)(c0d|pa=Z&t+FvOAQ5MUXWL#NGPR;h(b-kOT?Hr? z`(86ZNHeYLjsnwJmWEUGUgIk8YQz(dun`rks8j%G+Zs!}?HCs)JYccPi0oJSoJ(54} z)`2>8(Pe`pt#)nPUt!cdl4cZrfAe(bfvO*(vP#~q*|ae(Z<=)L!641&EH#@e*&i4f z?LT_?k?vDcc1B-BCF-nZ2`lMaf9td=d1dAQDKC>sBA;Mn$$_qaqLZMmJPPM%$yHTX zn(;O14)VrG{;FZ>Ohlk@!>WB*MOqVyNva&RR40E%VlO6aLUFa}HNZ13ZaJ@|`AWW} zn6bRvu1bhow!H=<|7A27E4u+^77v=OJt8wlJEW za(D0@54Ott;sso6O|hhw3vkIUki}sZ2^R3KVZYO;2%on2`oci1o_FZ3k9lbEm|1w? z<*g}iXp_N{Ldd!Z-Dq59;z*_{blSE+^IN?!XSl0I zKsYpj3_HBQk}IT8xfO{sM5fqy*3gzt$&X*hI0z2m4OTzyC&2QDm(&6s0s?Hh1ox1+ zsoqu_?2SUSKrDUjm7>l^H1@*%DFwzC<2?oU_e}Ig5-J#5TGl;mR3*8Ku_)pMJR_t2 zfGEp0^;>Tv7y`hj+djDkaxrmgy-D7q0nqca6}9Qw%Jkt>6Ql4(yjK_=#)8aq(TxC) zz*LDkIav)NlKRZ!diaNHIGQmb+qNKs*^g z_r4;Li=~Pkbuc>I{d!Yd2&%KeNhwCeW%HbY=6KTmYLO(XBBjV=vz|&Se$6xHyMXkW*^e(aci$9NM}@pk z4y!FlDy!d*%s^7;QE%FwmutL-YR6><-@cqyENv^*Rdpg+GY?XB- z)|ri$w4nJI%PKYzRya1W8=guvN}V4Esx;{x7t4W=>et5mF~BE1%W7a-$z~)k)N3u8 zVLl9rTVQ*h7eR>=HM%oPSL@QKE@F)>R-t}d0f@#U^VN%$GAu_r2ixz4XGNwf>Bya~ zs;(nfLK!?>Zp~0{o^HJ?eo~;^v}v~+!B%(LUgXU;Lu)b|W;|+@gH%P|L=81A?g-}^ z4S&O6qH;~&cY9N$VSdejoc!dj9UemqLp2gCA5)(P#8nZ1A+iPSuCaG2VBJd)LJS6> zeVNVXXb(M?OXAKws&yCqlSKSUea$*<@Eu}-=CyH{+IYn1J3$JwncT(q8@pRv#M!|Q z@{9mhn;D?ns7Wvsg8`LVF?zeSD0(mSUCJT=5r%b z8SjMiu3TI$%d*NxBjy+JtT}lR7vmj;E?ZY#t_@&D8-_Yt47JBUrmF{fujmS zCzr2MKAf*A5L`)>H(-9o^+xKg$)hJT(&YxJ)E`xJw0p@|RyKv{+G|?wS{0o-j*#6y zimQpsGn+B=N+ek8R!uS*%{Swpw-4N|nR}fy1Gv%i78#6iBRE;8uJOwi3(B5b05Y>L8|w(NDUApwsRfmk^4z8=lYZqo zi#y^3sl(=zlU#}sz1L|G$7eU@{D7hqOL1iJ69F47=GE|Z{1?wC=^)cP5ZiaWW}GH6 zUzU{lM0Rkmnz%}>)hOoX^`8}fblpDPVYswICgalP$9qkiN2zYt7adqKJP)HGYd|CKU~YC| z)JK9#E-c9gA@!?Q$_eMo8JmcA$z0gUcHU#ksgvdTDg)Z`U&r*;ePp^%lT+dNGqOVy}k$y+#;B<*15XACiTk@8#eY`9b z!*!nt)AvfdCnd|OP^kuAEgN158K#sS95*aHa7PhhgXHPwBuA4g6?VkfAY|o@qW@G< zkk?J8%&n}RF|+;FT~?c9($9Qb_N|Qf(LB#Pgk%5Hlp4rqnIcJ8Cqe-~nR|e%=(2E(-;HdbKbdHDNc%3^#FW9z zq~ZHjRn7K_C+F}c^cCI)rK5#9Sw$qz+X__5uU9$i${)nRWxsDOuijUdBKx{0)4D`q z;^%cBKW6+9S~VCp<}{igJQM$Bd(2XojrX~JdOe$Ue0P}5L|jR|BIU%d+E>kEU_mXQ z5aU$FA%L=MDVU^l*bar=#jmP<-Z%0lwoHfR-R!fMS5s)CC*Wxxj`cF8Xs&9d6 zporW$4ej4>mL`>f)t0kdz{eVjakEapkZ;p5Jl0^KNj-~}b*~{p?jaok&oJYBZRizp zctO?|+X6%Rs&JfnR?pr8`97^GtqMxB=N-iCMp33QN{&YkMS^oi5^~$}@2Lxmr8_eMgSpi|3u9nCR=$@+O(B~)3T_)S(&5G4w+DfIzJnA= zB+HQs(r7gIE?3jEmYxjLGh)qIpP#EFDsK@|!x0Ge9e{ac+$6DHX*)Vp=LD5MP8gE< z44z2hix7s1vV+OK)Z{CxLLBt#nY$m%XPqJJYK0;dt=WzhmG%z(ty5Jz1C>SF%=yql zO_OMuifC4oY7a#*840%HlS<R0OxH1(b#!z@Uas2UWEy5xC`4C38pdF z;M5D%*{EfbFxNxbC_=%6jK*YTPdwaS$m@HuQ1TP)KzGCMjqq%*e+SU;NDJzax^xnS zh5$7L0X^tUTZKde7IuZ#XJ{ei5CI9x5CJs{3-?t0hH9Mth%1k5jYKWegmG@J=1I?Z zmb=Z|n5}cJ!5!|xjsu87JB+dXs25hIA_r;ofD4%Um9jqp zQyua9Jc364+2w6>_H(WvbE5;y-aIM@sjUJNEJd4+;Mms#HkR~iiMxcSNd?btTR^!K zVx!jN*09ErbE299C~uFrZ95}rtN_DjJj4jlJzWK_`#{gJ+c~`K&ZMz33-w1NseNe~ z&rwp>)u4Y40(W0{{2ylzqWOoSr9YO2(Hv}O7}q~Y5}@)vdOV^+$6;)?+rzmO)RhZA z^7O6E##y4skL@pQTf-{6MV^YzvyQGKxD4@It_CxPDLZ3~E4j*aDi~{7YJtI&d4*cr zrNW21LuN$wK_eRzRUYZnq3)UTJ5$npWW=hp3~2dcwDWCfN`Z< zYrGR2)-6J=<6#9Ui;0OAamjoA2&`Eo#oSRt`!9~}gkO)<3@#ZzU@wq56Z;zr{3+z@ z)1(SCgS@3vQOgi>ci^+HWHOIhFc2xy6|kxAn=-HK;Vgdh&F7=^N>TZwK)!9Cnyl#qJ1S&_E81%(tP_gb?Ppy?r$%X*aaP_&`~I&|uNN!EHU_Pc`me#zX;9N4xo&7ef-p5ILOd~sJ-aC1%F=UwHYG4rh zz((uYC9&huL#%g0jAwW56FzSltD1G{>Yh*LC)tkX)oF$^Jblqd_|~Az-cFT{prjo7 z5qM7moe_v;YDYM0`za_FaK!*e+uB8Z2P*mTx58nL&RQ+fGH@>}{kq%od{z}q4Reui9pwYh)WyQ?Vj=zX`rg`M8 zUSX^>fWJ35G}L2|RYfe)!bY8kAmI$`J+mKGGmN5oQuMn*|95jR{j59z zPw&*1t6@ugHli}BL1)kr=t-XvA+n7`3)oOi04mt|XzUwqbFAgDY>}}1?9lv3`jFn)m&RVqI;K3yRYR<8MRaGzla}<`Yo7ZMz3Rr77XcW)6rXk#XJR zQYGzcJfWZXeK?ktd@=fkp#<6kaM^Z&O(4OQY*tvLn{zq*R1Z{8bx%bLHEPPVCHf)9 ziEJj(XLd1Ta6B}%pswR|b!E3Z%K)fiajsuMt1Z&j?B0mo)>?c;qizS{^YkDjFFMZ; z=S4JrGxp(aQ?OEc;1jWgSQ9wKwwGjoO8bBA3F`M^4|cdg(liwE3Z=I7QPN}-`R&vx@ZU0WIzED;CKHH-^*zxdraP%+@t@kJ+?Pm&?+RqB z6$LtGMhNRb{0d9+fZ9wcUtkS%uvYlL+XdZwjutNRzyXxL)3xg-KWN{+0&G-4hfP^_V0nv6@Wk&d4cI*21(la<=ibJU>s*GQ|Ic<48E;- zH$XlFEv*)QHA%1s8fVeC4C-f6T3_@!!pV2Bx&?ZmU1<#bY^Zq8jOGu4E5O-an9G!i zfn>+Ja+MAOL;4OOm5AF8-67fTaI9#G;PD7<>0~XNJE)*uoeN)7g1J57?5b(2z;&sF zruqxM^{J|2dae5Ls5h7gt9^D$_b61QN)O4KVh1!|CB^L=4awM}@QzNd`4)o`RQjUQ zyTi+ah{NW(0yq#3=LR^}6I~TJ%Jr%PR(VwUpx0 z$cb{pgneyWQ3MR!n97m&Y9nH8Y_329Mk-JlqcTFA3W9eQUW4p9GTY$X1J`yE^qz~@ zFIX-ekb6S)DG6JhFXLj}$x`-?JF;}a$iHv;w;K2aFbAXad7QpMjgr<-Ga zK_Y3zElB*w1ZE3GJxd}X#Afkdr&y-e(Jj%x=D+%{v-8XF-o^XXG^G;(5#^nT5O$TZ zj*u_+oW&w4&s;!c)V(~JZUeTGvL|5kKMeJsGSofKaFN9Yuk53!E5L?h$P8N4Ax1rF zlZ3X)EZpm+8~^*Un9npof9@xH5#c$ocT#*@3Q14I?wC~8wd~F{&3&TpK>Ssl{f|F9 zV{u^M1d*vPMhe!xKu@FEih5hV7>mrvc&sRi&M_SA-@W*s+h0lxB9(5LO5iG1?!)9G zTyC#VGi#^7JgDB@d<#_wu#BSMYDz?r;643~3;1g{euo2?n;;4LcJ%@Ltoe}sHLw&` zvyQ8A6VpIeRBOA?o8j zJT9Rz$eQ2--X+s=uYX_Y-^F_Z^NC_-X}b#=5@cMC17O8M)Qpo2&Q+kl+(Dl! zbubEjFQ!ZW$zHHx(dB=v*z*jFG+bv&}0!2#lT|N8>!zF9k;VH$*TqRp`xai z$P|yt0DaB$nN?_|DUI3Wmlu~Bbg20M^VechgRjxGCIBLYf%JR-f`a@mNB`aNuhigP z?oEWV{Ld4C8d(a*0O>k|N2p`1iho4GW*@xlDUv4e_r4( z=ShwL}9Z;)N_* zA7Nop(yX_C784cqTZosn`Je0WR0h{~;^I#IC75wV+%cs7-`o5FW*z(MIiCFQ<)n*( z--69cAMm#^`Ont`FO`M{+7Ug0<^O3bf4>5001~Q)@W8k9od5W-KmWvx2b=N#=iw9| z?YIp8r$LaFC1qlg0b~E97kRm_KHeP!j8R_HWq061U~vn2?>0!-t&jnZg9=DmmYX2+ zti-;}04AbnSw#5X77L-c=K_4mVAXr=CUPK@A*x4y!2 zejKe02$q$XYckA%R4EgHV z0+|}0fu0;~NeU6F0}TXjSi)@QBWMKErNYZ%xKXK6@=*W$ARlv>Ur%WTvC;!#&zZe} z=zk9f?CX&>BDnz?7)Q{jUjjjH7`VWkQA{dDiLg!`WBuMM{O39ziUq0zoXVqgAiItP zB+C8RKy*@&-E2~C6!b~yV1S#;z$Ck+@Un@(flBK#g&Yc1GV#3VyE+46 zEq#R4G|kqDe|xt+gh`JDkl2I*Y4$+&!T!k5NcxVEJPC8@O2W?0$Xtc&czq!1c<>!d zraekU3;&_sZv9~}A%{hCs>hl!Z2m;N)E~Q=C^I@}D_gE=wHi@U z@khZI2_~dLLz_$~+zNkxdaBjnB!buu^HN^5VLh8{*hb_bw@Dt9{Chm;}tNoKpp)^A(`%Uff#qxAXiDGd2J8xx=2$(1C(?N+B@; z4bH~P15N^m3gxO~tF6#9D#%5@;DZjZRZq@dKRCIlyx9WkZW)=GVgNM>E0bC7+U9pznK`M0C~+uk*&AOM7uw1eViP`S4;=)OgX zjBX@ar%|ZzGtaI7m|X0w#dwu%*2VTzRl;RPMn-#PjMm1qLS0LCR#v=O(){|Tm$MeZ zrCuH};290WD_Z~kVLU>bfr0B$?lCt!-@>{1RE<^@n9YwP$DeoSpuF*j?EKsa3i;s)XMb4;)0bnus=jqV}VG>J|yw( zf7vGp1u6^_hy65D9Rt|#XO>~|#u*7F+gIX=a(;NmF(=d08rDicbUG8n0LWHr6Xn^% zh0w8yK0VR<$aKVaG2xCe6aL~O|GD%^h`fBbT2D_$rG3=eNaJj8@$0w`4q?*OM*ug= zt)l66zWiwI29%Ys<~l<8ZqD9L0IZ!g5!eV&DH^_~zXR*3{NohDzTy&Ae{`7t=+-W$ zTAbBcv!{!Q_Z&~jxhSL*lr37c6+2nLId@j8vC52)NZr{z5tw@mG+Ylh@{1>Z9fX>( zKx&=*+#2h@eN?kH#87m1x~5$9x}DOxh49v!%yc{cWmn=1C>~-K(az*H3Ur@rPj#VD zC(6F_VNm}P3X&4eHro~Fy3P9X2ppuZ&nz+uZ6CjfjQ;1MVejCH46}K0K+41P;_7Lx zrBzHbHBRa2{Af>^_2>fw^nt@Z{w6<`8SY4cWCUEQo)JG`St7vf_1dLIVLV~ys2eNN zqyx;6acc$p6iD@$z}|mkKK&ow837gj@pFjJ0}@7P41IW^LQ}Z9rZW-;-5w*6uk~>( z9>uO~FihiaK?x6_>Dsz(t}l~bnSq*`4H{(w=w2Z~1mPE_<tN01Dwh-zQQd z)oQtu83+?bv~0wPh^>4*7tqgY>8n`Oiq^Dtiv-zy7my7i9JV!2JOchCiYo!=PQL=7 zoaTm0JDjHu@WUvdzyE=e{qqY$Vv%gZQvn0HnpGj1SM>cl5q6iWq6)3E?!7xzW?^&5 z4iI*!8&R9+)%o;p5th4y}>U=#rB{7K$%f-&lPPPj);uAM*(EEY)=LV(? zQyRXu*;gT^{vm{P1Ida~W@V|6!M4VFNjCS86!D*j!H>jK?z6C16s?jTa1UkXSC1!n z^usgk`7~)s?7cFEMPoGR+8r#qga^<<>;N(T`biN3KQ-{{K+5tIy|e1r5Zuybkby@p zEiI+)%mHm>qv`6>0`-c{#9~L@y=KI4z>zn6ocIAE2K-61VlYUPNVAY0J$fYR$_nT7 zeuuhfgyBV)&q>jn@Q44{`4{x5O%X8S87qKHZ3i^gdVnLN%yRIKL5H>(?RpmXwY|e{vnD&hxEHVG@Dmw0Y;r5Nc3?#X zsTM)<`VWy0tZvFTk*lbkKtUfmQX1@gQ!#VUlgM z3qU2U3<7uFxzHMz31W;igVu9YBnHS_cWN%j!~YH={bW3kH^;aIByi3)wQr57bZGs5 z00hvceOx61B^fv<031C``Tn0T{3m)uK>{E=IzZQMXKdAqqr$P7R+I?5Qh0Lmjs~a@ za7@(M7nk?kMKHgI31khLReH6|b?tJJ64<6^8@I?T&yTY}Mpp!B&|fx!jAw4B!^()I=v#hC07s8WL=$tj3V~%pNED6Y+%-48s2ga_>qM!%fs>YDQcWP zQg6n4)Zoc33EvxGLlg|K|Hz2@P_qJ6n@uqW8hTb$ta+a)R5VXZO1hX3J?Pn)tzSQ( zH3{!l)euoIFE**|32%FSx2yDCrdbF7j8VjExXhJrFw|OipgKq%dY#U^=>&f%VXn%k zI#yir#RoR+xjlOiJOczY_%?%5(m5rKyufMO5y{&|!6!TOROZvP=+*X9zATfUo9xmN zhh!aY*v-v`yG?zQ#asVa-ch3H`+ot6>={B)&VS$bVKokfd1YAv7yS-vU(~NYY6vCP zE0w{4fl45U^&zcx*aw1UkQ(WjkV`(&eEd3k5rYYVU9qE)ETLPls3pj@ zV&Fh{p#AATV$Gf6^E2iP>TLVxZ3c^#GNsJ|4~33j)k;)O4GzMSJaNPjuG%?_Wcmc1$yzwJ`~acIB=Znv#v!otG#5&%*v5ttFtA3doTMGEXgFc58md3 zs#g*i)|tF9DHgFGNDWW<+E67YkK?{*~J7d z^Q{IlN+Dc^w7|x+&s7tl35D=@mnhXqN`glyCgUfq<3j7lD9fYMm!^A-;m>t0`n((hobIb?IRAkTkTX?( zKZdmCh8*qce++|#z)kaMPxLp)eIACrJR#~{?vB5o4zLksZc{c%3g$xjn95JNzM{sr zV_+c#w~i($CiYw_)%_Ee#AQRjAj0OdJRH{=P$%vchY0aD%m1MzCDiRmw>GyD`G2 zD+fI>JS8oF9*kXcwPWBX^<@xna~4kj^QGV4$H?gW2LvcbCV`^FSI~L>Go0|(RU8CG z-0Z12^}ULh2124{Y!Fo>e%hUVs2`BNA3vxx?R60gP-oBus0#ukZ77eFuJGZ@BnThgmU!CQ#E#@>Po!4sIUa z5Na}jZb>v74bdgNG@qtDq0X0OKkh#k)=r=5;%N9a5^KElTZ;MXr7~!rfe>n0U5Bzc(3?m;V4^0MaAP6N%B4f05u(dbLdf20pr(9Jt;|9*}$>!cHK#4d-q6z z!D=r2OF2@MlGz{k(EkS9d+O4B&Dj!bd}gj|0jO{g+0nNwMmZNCl5OEnP`zo zCxSS@^+N}&`7WLl1bRLGI*~pG=IK-Sdr>5&SE}dh%H0@8PltON$wy;(z zu3WMlT3`mngZ|fnXPiI&1KM}NV{aomu1sWzD z4M$BZoX}8e$oJ55b1MM@dmzO0eVW^R>+>IdV<_atChdL->}x{6s7{{8!vN4ixletKUt(j_O}?5WQ&R&fM59&LuF z{}!HJsY=7ks{)^V_EE})uwt^Z^lYH~Ca2>8;m~B_&kF|!K-$O#ZQ|l9bup4;bvWk< zu@;am6X%^@k_bfurABu?Ma>Gh4wbXR4SK_5hpxryf4cUCts_Rgos6C}+MA9p!0&sg zNo$_j)3MN*Xp^w{oRi;+>l$O-Hr9XjJTPx672wn=?f~ z6N!SZ5%-1!++|)wXxGf2Qxp^7lB8(ZBwiUCdn~5-`@2m0CZbjj=Z7Xs15#=LD27^o zJQh_p+X46|n)X*lO=6E~%fI_IQsja~cX@uY00o->JeZzAZP*#n$lD)0tpKEYDfT3$ z*N!T?Ff~@KET2v_Rb-p2f_B9M8mG!u(C2%m*CltD=II8{*W=>x-ADELQ)FV}zI>6N zEH2Ib;kVGYaH#7!h?i85HW3KHtkzRn#eW_AH2Qc(%_)W-7gJQy!yU(g!w7Ci_*`tB zL5wE_DL~CBSnKKQMEC6FOD0eTEDV`_-@BQtL`kDHEVTigVpvog{Qe}SuD?aS_Yva} z4PsM@w(k1mn0k&+6`AqW9AwPY&SM#I64x>)txDJr9&H|O1=k6i|@G+ zq5eG3nq(&t0>LzrV!+8*3IB~z&dcf>VoBf6yg1CFSkm#hek`DljXgA zZg-~vJZmplLA?D@z+^=dWvM_m?v9QSO?%_V#_G|%XMKML&PeYMhLSSt{gTD$Z43eZ z69V=LZ_j~UwXIiqf?Ra9@jKU+tGDH?5*0^uW>wCrymEH|VGly<@n>dd`+Gc*LEY@Z zo)ppXUyW_^;>I~)BvJolR5w@nT>^7^2}%Uy3&{4o@V z>l3JJ@E|(mWfhQGm3^xUwq_Cj5&^`V2#qDt_6Xo04UfI26Xo?<&W{ymW@c(u`R&RE zRv{@_c7DAJf zQ(jCBI}`WwJt4d=ixK&ndbsXRtFW6LAkMD)th?p{MU-vc)j8Icb5o|gm%i0-u<6y6 zN4}ah+&d#70!2G9>aoJE4CW#*qhg_>lxaTB+*;9W`=+paGKa9IqqJJ@ORw;z_&r$YG(*KBYIp=cJWRI4?)cg@ zbUQ-*MQIZ0C@qP%M+^t`^W0Ru6nl`_=yIV+V|3=yXl?BJhd`+pM?Ohi#$X1pUry-t z_W9_|s6(Xh_LJij|{hESc!&kL2BaK}EW?p0Sy39mnIB+ni>BH%1Yr_;@FhHud}i2Y14R zgV_QITA%SLHE9rC9mk!IpMi$0w^D70Zxj*`pkH$WlzAZSe^qX|t5p0;7DjT>q~h4l z26GV4n}3AK5xB$saB+q&QuIT*t_#!$tIs1N+CY#ttc!XOYEv&6C30i?1i@eyV8xd3 z4tno;Bqk)SL|C?Jw_iIcZY2#Q9_qMpp~>)SVi9-}p8*HfFLNT#%rNEPqIR0ySY^a2Fox%1qCgJkd#@ zLohr7(WM6)aBi5GA~#6SoW`KX2mrZ$FQbsL^4a^e?Gj$VAyW0B>x)(z7wE}S1>qC=J@{{KOSljyc{cc6OaCL2$10P z3;q_aJK@2B^t<&>u~J<5E#9{S5ZEC5w*Tp>S_>?%8>} z-9LfE8;CFSb@Cj)y~g%y)8JG(Z~tLu;E|thE5rJ%7E)(^xO1TG{s|B5c<8sM*q#P- zHMA|Cwn|8SfIJ}>FG)Go$WsHlC3l&5^u%zmw0Mh}l-PHt@uF^EYHi4cxIDek94MEx zC%8T5Q5CjcOQxLLi@?}L78Uw zmLz<*Al4Y7KDIGYfiY=RP`0V7zR{KW?=imPoL!V-l&iM``z1{yYZ( z86F|DbEfSMzyJBUalbmaE;myB(>9Jh3Y@~@y@y_o{A2Kt)Ixy#`_jLyf&Wiun&#}s zQHePSJnRKBgyV$BtP%WDE!QMSS5L1(k}-@Zyz@Q$W+ga@g8SK_LcBrS`r_WwoCJ3!AYOVb0a4uQE#XNUacNM5e`$w-`s@QdJ-z2jU@8-uq@k zGgJ5RN&D}XZ^GHmNO=3h!6-O1!$*pBAfFnUNzT>oWTB3rw-qlxGBuL*u~ttbR0Xe? z5^=BrJYN#&4FI}k&x$kBwD4%*_IFb3;$Z+LT?EaIAcvfkx0UrY{YEG(0jA6BwbxKB@FNZcKy*yy{m| zep>UncB(XTW#Y|Ry8`GO(j6Z{KQq1Z?u(TlC5nKY9KKeBY#!N9Vh`u{dJFUcqn^w0 zZs3e%??=gN*F60O0kY`_(hkkJ3lgluk*VpK(f*R2r676xzn>*&;RLF4fAvd_D`{X&vjm!%u#loARAbfT zq4=p3NUcgvt6e-Sb)Cgv6W{CHwb z-F5khhz1712lQLOQi%?*q|!(CwIhVxO8N4j@q;IPU%6Kl54GQ6X=!d+8*5yp4WCy& zT6JV^7}l5usM0os~32lI{6wfT0;AbE~83=Z6xc*Ob|VAP)cFYbUu z?61FHokFyU4<9IIi!F?8b<^ z^U|8!$Xj11F3BNP$yTNPWeDd^`4aqA_|!It0D>H)WuK)=ZwNo1_<8>{RklXZ&;%$7 z`6Y5z*1#pC57OBhA!QPDqysLyaG7>z@?x;{UQ`B4cfY~Yh^mYgp(y?5fV6(+9!k99 z77iFw8f85MK+AsFhw!RsIO=c@|0-YTDG@$@fvSOvtc(oXg)>xV8fByc&Sj|Fcb!`k zO9mRlfG`G{d`H5D5`gO_pD0hyHyqM_CllwLJ}dgp0JmUve|Qk(B@&}|5*>PGf~*f ziV80Zyq+87Rki7X3!d*JBwvtP{2lzQuMF_936xXtn$o;B4;)gD6_K<@F(Q8#}&xgwwhq0N4n`Y@N6l zZR<-1Rq5j9B3sYV-47e+-!Z=EWG>hKjdE&w<;*;H;%U_L)9C$QKRX=sM}c%ywHu^H z)*t5&C{{xorxMsU)iGD*EvekH^81`+pb~l_&@q{xOWiav@g@cliXu>b8#g;6qr?$* zZ4(9wq5<_95gKhgoBHztORhZQ_m*_-)e_Dj+uU?uh$&u=B#b(;)Q7{ww0cF5cXpVH z@g@*;2ZZdZ1Cc*2Jcatr>As4dc6Cm-TP!MRCU&jqh0onCN3FK>2{USDskAMaHpD$8 z#Ei@X@Oc@h@!<~&tT61Kt4@yW)aL=XAgPGpc4hlCBj8skaUV_R<|caol%6QfVf-P~ zsmHPpE}x<{yhDQ;M)mZqE<+V4RK*pP)<}z4zNDAQ3+d7s1{?mLsS-ZwI|r8 ztlB05R13%nrDH2l-$HwlmS?k?2o6>lzj%3dgsP;tC4yU*7-R;&4$J;UXLGq;n8Fd5 zLi+!h!bsq8fxmE;zI6O7VkeR%WILj0Aovv;OAVnNr7e5{PgJ3E8-#cPC z@cat@M0^+&f46>lUkRZ~Bi>oKy3Pe?cUHxFrqBYet$h#46YR}dnNRM6Ke)ZerB@QA zp+%VG@9#vCK&Y~yD0N&i6kfo=$TzyA0s;*dID<-Bp^qey2(ESO3ZCDEMpLpvdr!Ho zW&o#^uYE}c_GpbDkh8Vk8_W1a#EPETpYgPu6gOR{%!!wM^23Z6O*jk!axR5;=`{)} z!ync?l!Qkfpkn~yc$&?4+3{D60iqBV#LC8Y5lGj4h=Bh_1d<70aQ9SujX6yLoq*ZG zdjh0>z#RS3)BZrqC(LOiQ6_Hz?jNrL#gwf*HXF`}V;3zVvo^p5%n$mj2s;=9>{;!5 zu5;dIop4CKhKg$m|NC-^A*KC_L1ro3p1j!M8ByhW5_;(rD6usi4VLca1qsEQH?HFK zh?E;~i~fRdE#z)~^g8B92&`v1PIe9sj>#X7PSt1>Sm|j4Z!f%QTreGW(SeuLlaR)A z{&@FjPv_dl6XrHGS*SStnkAT})rmX!{9o=_SzFKWn0JQqx~?u>GPa^;^g(1&pYKz_ z_Q>AdG^jE;(p<7?Tz&*7qm&lCS-FM{41#OS^ADZc6@QJ9xzxoY}8AL(Po;O0t z*dyuoq}KP19PT~5rNW-91dKEo3?r(|)b|(C)}2`nJ(>^$Y45&;DtRlb%r3hu{V&rg zl3qk5*5^0Qo)B_z2ycpNVDEx$vCMQ$Y&<~2tRhQ2ncRu};I6UzK3eOW1!MPs(0@Mb z$aNgW$uG72Z^Lm`NTI;!7U=6si|}*!CIqnZl9f}nZtDsN`P?AX@}cz-Y`)b_z`o5k>rn4>=p{Y|QqCbB zK;1lHxy&Tf|D4wT&=Sf?;-Lj0h8#H^3nbi6ba2y4g?m(92A}DEigp*jT2eq^kJPw5 z3O9jN6V+=oa`6gar=(#X3DC4oyMv_Mz}eOV1X;h?i7GF_WobIC>(K#H4@M+%jYc4; z8$Kqlf;YY2VNR`g(8|>@m(;@*`IaahPjF>&gv##BWY7)Te=KK7EF>{(Q_#-ZjmPe) zi$%j>v;6GL;7{;RRV6N-=3xL`57V{oE`oPTT67<>aEc8sKi*x)?o?S@y|jP7{EzYb zci8_8xIfqHzf0|pMY8$F|J|DmTE36s@!=g%?tL=SZ7>0)0MONA#Gf7?;MQhM;N8Xx zF#@okcUaOziN8GwWE* z9p3{LvekG_OIt{dF&wRZKvcBOmCCdNUWF7WTDxL_q7?wju*Go}6X0`DO#|P*GDgx5 z@$YzsG%=c`6yU&XOi@#4%QI&O)~7cB5)vR0xS^_itF;WS)hvJ_Xlh9Yd%X`P8m zC?==aI&?gou;9&C1yy&b6G8}|{`u*YK>M3CT>}F|s&2iEnDeri%=%`Trk@1mQhj~> znDihwADO77r3GtBw(NRt06~mym|zr1lYAw0WPK$$agCD9J)jU7*6RUBCQt8 z5O_e+LhG_j5y0lJGv62b;oz2H6DULZ9U!+d=W*xpp=!BnOPd(a&hIcP?ls8 zextZp?)?mp>l;#}p1UI7tGgK-OpoF8_D$GnwnB~tj(p5|b<5)=3!N--(%EFsbJ6D{ z3rCWY`#=)%vq({c&8SL-m<_CL2dA4Oto3A1P57X04a?IMr7yu4^pt#Lg~*G8WGG3`=dKf0@d-pOC=X2WIQvKv&Ha9Pe5$Z9}pOsheZ#}GV$F*$G zdM;&7W1(}wv5U7Oa3Q$0HYaVg)H}W3XJEypW!Z>IhT`P4SZLEph-1qY+W}~NdddcI zzkqNtdtDmeeC25~7qOP{U*N@1BK}4;$Gmij;pIai>w)UKfx3O)AjR*v2NHAftUI3$ zMlYx!-XHhv?HEJdy;1L+4z(LKCfq987)f<+S5LbIowALQg}(mHY>4I(N*D9c;M~&T z&H4IliA%93nAcppxhx76;D8tB`cf@g`o2z~jbUZ&Yd5W;sjQXpZRud z4Q&fUL@A3s)<@s#}QMZ=&ytj0G(+@}hB zbBwxzST{GnzBq{H? zZ^xk=b|3l4FGo>)g6rubYRcf0QJMP=F&_=f47oWN5n+f zkmdTUy|6>b5I;WYc%ySz5iBFoK>tN8wMe2)BjO5ocXt}jtoM?!mtLE;^JGD#Zy#u! zUeY{gUNp|rqH;Ro!xPG)HPtXOMwz2p@lnF_>_M>y{aKA`Tjs%7fcrR0yAb&Pdvx`t zCML!noUI1g@`==-`>!3E9e9CS`VY;at*@x70t z(pat&vkK>DVB2Gps}pP%1iP<02z3V)QAwvnQd^tkWt%FP7MXeqlk+2D7J$PS^N^gwL~+U0tc1NeefA z(wxX{E+8tetj$$z4Nb~U!!gQ^QgS+_RM^s$q?B9@3O4uh%uF864u955z*6)`sl6_I ze`WhHuEYC`^EB@5cGI+*ZvX5^Q`IckdI&1v-^moD=#1E#ZZx$%YqLQ6fgh@B;-rtG z7GUS3p#9FW9D_UItuCSP;6a>Bn9=%T?;-nDMZ4^@A)t4BXJNVu+oyD%&a2D#(KHFB zV9xqVQvY}<(Z&HP{zh$m*zXV9yHM!PJvk=`g0!>vp5ynqnzW9kxK7Q4TGw!>yD2r) zYUCK{kz?|taVOw9;9!s$wqJ?u1YnYuAlnUe4S2#*sTIUqdkqz|9*1`w1<&wtr5~tZ zY1=zpS4_eE;bc+B&s6uc+$5*WhVHeoD$yXHI_-)Gb3q1BP)b-Q=FrlpxdDJ3bMCA|(HL4IqQo7=#t`3M1Kvag4(BkX+3i0Gst zV6W}3sk5;YCc7~4pwma)O&+Ty{bnKL0-Bpx2lo@d?FgPoJrwnON0q~?F$uzJC7|si ztvdW|L1`el`AhaVXxJ?kdd4&EYTzGzp_P<0or_$eA%assVa&V?_~lqyT^XTLObget zP4ES0Qw0*Gs*{VgE#}9is9J1X+quGAp@_E&++H%z8sIIgofaJ&H@AO+5KNMu@wi00 z|LH_`;Z~!{!ffeLNux@e6gcVQ%#sZSrSJ#q4FP87LF^{wI6_sW@$dPsIRU>F54R4q zfJalM46}~i(NYQSCCWDPEHhQ7jWlV^v}-X{O4Xdy3;B3eq*Av{FHlQA++|wQegy34 z%n;T3O&uvzf~xO?K|5d|x&S=r*(P4>7lP|t#HHB{U26+13x2EF;=6=gk<+7q2ZM&l4 zwT3!IP&;PFC77!3Q>-G7Y4_LDD0OkPFIXowd|C@b5r8BuG-qsqJEcZk{K;wGj>NAO zv;e`xN)wN_zzP4$Xf+22d$0Wa?)tD|pIi)hOeH;|{QP7)7qOsv1Nj_iBlZfz|FD_O z>t~PNp;FMZz#mY=7y>CqA1?sv)8@G}sU;^z` zdzN7qV|$&OqLZhjYT!%AxY_qqR6J3x5h1EwacTtY!E=h@Ku@L*5VVN*FZndT#;DNPa2yCJ7W+eqd|~k(J_g}a5GUXn!*b_mM~0mg?7ppfZwmJ2Gn2y zqZYePxe%ZKsZi(V?W^m@gEP`HJ}Ien)*f%NnFA9umPIwSkS9l2t%3^XgKctynpdf1JoNhTjSqMnSICFT;DQL-1uoM8Tz zbaMfCPhQRRP5jYIOH1MkT`!2Gc7~S@hWC1OMc&r%Cg5?26)q|+LhBH z&p^fHc(J<^Sez2N3PU=4Ca${ZN#^&8sw&Q;+_d5Lq&Tw>BF^#e1q~B%If5>W7LOy- zQEBQvoSj{;!O@1tA5fDvm5sH9%bMN~2i2s34?;TPTmw6!>9tE$)Zbw9x`D4f$TXCk zgErh=YQLJXdUQ(%x0YHgCX*9*_jU`W!9>E4z2!wD%KS8(s!Yq4KK9DaDK}2GJxX$Z zK{uD^U@2Hj>;HKBKs+cF`I3Rm!-JorWg~=0KQ)vDmOReX7@u`@KG%^doGMr+$TfZs zNQ?9`3eOqW0xIIlHO{oE&DvF)VOtwU!DOx)}SH6`}E}}egWbZ;ftUmM$!lWEp=w2lUoPSLjdbHmyNhO_HQ83R~ zm)Rq^REJQoQcvzs;^o4w)w4g>f9fs=k4(twsodmg;+CPnQ1MyCIEQ+krYFLH@j=t) zavln2`Kg&8gK)xw!@OZ+tXCSf71~mp?=>xq$f^|(^GT}*amSlK6bUT4;)b{78M=bju&6pLqVWX>|ptfl=&{j_|GRKv7TN+1h!BgGIGol<62kjZ|N}hQQ!z|-9 z=elj)r_2k;?iJ{3w*v@Ej!zz1*zLamyOMfCoq23(=RJ@#9HrosSKwac?Q&@VU#}Ar zq{A|9dAttY*nK9v-@K^>+r~PIczaZ%sPJXLv1C$)MIp}1u&cDBT{bq_R(h~DVv($NslJn)@g#12E;ibTkpbo3*I#2-IT?;oFz{$WG?sqGyQgYOMQw9H^@Xk;E zq~^3aVQ|tNlWFiDho)WOnj$|iLC*t))h3YX9d{T?w(yOXeMkR9bPdRbH3w2Q-XV$u zNZC7HL3>9czw!*EacgPTw$l7x*pI{x&sAQKZbYyI%WK-&+V6z~wznBJpzMA5AlDnV zxr)29DQ-NBBI?skUTWy(HTyo~?!a%xN!D;LH%$jrDT(0UTJ!--g>Gi&8UjJ)J~ulY z6UKp5i3hGq_bLa(HD5}Feq-INReO!_h6^W1bc;f!m2tJpVma43w9_r&2C>}XB2P-k zdiA8rkJ-`+U7z5dF#C@A#Q5+FuuIsks8cqky&E~mP-oa#Tu{ep5(4fX#{}$BW^z8| zD9yhP1z(8ngDl>Ay!k=qK)zL(`eeDkHnY+K9OqHX{6pbPX&kuv~2>3_eBZ;pr_ zllPg$?s^!i%>@SRF{K!lf!o>2uP#5JDOUyio4dk{ zm*U16fI6H%L#}DsQJB*%n-k5A&N(s!o-`e?@QMUPQp1=`m7IskT5;tO%M)yi<`P|T zXJ>O4pB=DCCE_`||KbZINw)@rq97i#*hIcpY9M7q|@#m$Wn`Sy4wko-JF1qK-rjlt7#Pqc-J ziHULCj|gT+s~iWXRK2!zE^uWI>_`f^%Y4s}Qxl9>kNy-&N?Px*@52A|@W8R#5IBRb z>_T%LYg1h^00VvjH{y<4TBEX0?S<)v5DM_;!lvWVxm4ifR2?S3`#GGIiRpZj`s6-g z2;}2e`d{a$_X8pDZCa~(&sDX zJ}0K6R|>`%;v#tt(zutTU#N&jH`6QK;OU4vSAAX4jeXsv#4kcDsajf3!>DZo(5Vp_ zbI4}oM=hB{vLUD^_$j!V&z-YbsjzdJA3KJ38(_QB(bYAofaV#AKS+XXNUzXl`0yy# zPIk!0I3>KXbf|peOF`bigPjxXGJWLDS8)z30&))?&&C_}s1GVAHHUB;>vM63D)VCj zq6q@`=e-S`+;dth>P3o8U3%xN;gZ3>4-66>*aU8!kSpD{1!MkBtqsN5piGb63EJ&r z=GVW!MFxb;T+)8$txW3o4=IPvnk~7_KmU@(1S*SnJW(!(ww`61-umlH5W>H=`}SG) z`!D~J2H5ADxc}sikm9%JW&~A@+}D5pC8!-iJ=5+nEx5I%Y~|a3zg#y|4E#jfFv;(~ z{GZ1Q_lAVXe=(SUX2-vOGzJ5sIl?sb`w(wEBl;Us$e4;QQEz5fzr8Qm|HEif;Zyzd zF9|{*x?ewK>e}9Xe*Y5CHzfQ!$=lQMzj%@(!!qS?aBznuMMV_Research Coordination Networks or \"RCNs\").\n\nNote that Shiny can be built in either R or Python 'under the hood' but for the purposes of this module we'll focus on R.\n\n## Learning Objectives\n\nAfter completing this topic you will be able to: \n\n- Define the three fundamental components of a Shiny app\n- Explain benefits and limitations of interactive approaches to data exploration\n- Generate an interactive app with Shiny\n- Use text formatting methods in a Shiny app\n- Explore available Shiny layout options\n- Create a Shiny app\n- Describe (briefly) the purpose of deploying a Shiny app\n\n## Needed Packages\n\nIf you'd like to follow along with the code chunks included throughout this module, you'll need to install the following packages:\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Note that these lines only need to be run once per computer\n## So you can skip this step if you've installed these before\ninstall.packages(\"tidyverse\")\ninstall.packages(\"shiny\")\ninstall.packages(\"htmltools\")\ninstall.packages(\"lterdatasampler\")\n```\n:::\n\n\nWe'll load the Tidyverse meta-package here to have access to many of its useful tools when we need them later as well as the `shiny` package.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load needed libraries\nlibrary(tidyverse); library(shiny)\n```\n:::\n\n\n## Shiny Fundamentals\n\nAll Shiny apps are composed of three pieces: a user interface (UI), a server, and a call to the `shinyApp` function. The user interface includes everything that the user sees and can interact with; note that this includes _both_ inputs and outputs. The server is responsible for all code operations performed on user inputs in order to generate outputs specified in the UI. The server is _not_ available to the user. Finally, the `shinyApp` function simply binds together the UI and server and creates a living app. The app appears either in your RStudio or in a new tab on a web browser depending on your settings.\n\nFor those of you who write your own functions, you may notice that the syntax of Shiny is **very** similar to the syntax of functions. If you have not already, your quality of life will benefit greatly if you turn on \"rainbow parentheses\" in RStudio (Tools {{< fa arrow-right >}} Global Options {{< fa arrow-right >}} Code {{< fa arrow-right >}} Display {{< fa arrow-right >}} Check \"Use rainbow parentheses\" box).\n\nLet's consider an artificially simple Shiny app so you can get a sense for the fundamental architecture of this tool.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Define the UI\nbasic_ui <- shiny::fluidPage( # <1>\n \"Hello there!\"\n)\n\n# Define the server\nbasic_server <- function(input, output){ } # <2>\n\n# Generate the app\nshiny::shinyApp(ui = basic_ui, server = basic_server)\n```\n:::\n\n1. The `fluidPage` function is important for leaving flexibility in UI layout which we'll explore later in the module\n2. Because this app has no inputs or outputs, it doesn't need anything in the 'server' component (though it still does require an empty server!)\n\nIf you copy and run the above code, you should see an app that is a blank white page with \"Hello there!\" written in the top left in plain text. Congratulations, you have now made your first Shiny app! Now, your reason for exploring this module likely involves an app that actually does something but the fundamental structure of all apps--even skeletal apps like this one--is the same. More complicated apps will certainly have more content in the UI and server sections but all Shiny apps will have this tripartite structure.\n\n## Interactive Apps\n\nNow that we've covered non-reactive apps, let's create an interactive one! It is important to remember that the user interface needs to contain _both_ the inputs the user can make _and_ the outputs determined by those inputs. The server will be responsible for turning the inputs into outputs but if you want your interactive app to actually show the user the interactivity you need to be careful to include the outputs in the UI.\n\nEssentially all Shiny UI functions use the same syntax of `Input` or `Output`. So, determining how you want the user to engage with your app is sometimes as straightforward as identifying the class of the value you want them to interact with. Shiny calls these helper functions \"widgets\".\n\nLet's consider an app that accepts a single number and returns the square root of that number.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Define the UI ---- \nreactive_ui <- shiny::fluidPage(\n \n # Create input\n shiny::numericInput(inputId = \"num_in\", # <1>\n label = \"Type a number\",\n value = 16),\n \n # Include some plain text for contextualizing the output\n \"Square root is: \", # <2>\n \n # Create output\n shiny::textOutput(outputId = \"num_out\")\n \n) # Close UI\n\n# Define server ----\nreactive_server <- function(input, output){\n \n # Reactively accept the input and take the square root of it\n root <- shiny::reactive({ # <3>\n sqrt(x = input$num_in) # <4>\n })\n \n # Make that value an output of the server/app\n output$num_out <- shiny::renderText( # <5>\n expr = root() # <6>\n ) \n \n} # Close server\n\n# Generate the app ----\nshiny::shinyApp(ui = reactive_ui, server = reactive_server)\n```\n:::\n\n1. Note that the argument name is capital \"I\" but _lowercase_ \"d\". Typing `inputID` is a common and frustrating source of error for Shiny app developers\n2. Every element of the UI--except the last one--needs to end with a comma\n3. All reactive elements (i.e., those that change as soon as the user changes an input) need to be specified inside of `reactive` with _both_ parentheses and curly braces\n4. The name of this input exactly matches the `inputId` we defined in the UI. That it _is_ an input is defined by our use of the `numericInput` widget\n5. The name of this output exactly matches the `outputId` we told the UI to expect.\n6. Reactive elements essentially become functions in their own right! So, when we want to use them, we need to include empty parentheses next to their name\n\nWe included a _lot_ of footnote annotations in that code chunk to help provide context but there are a few small comments that are worthwhile to bring up at this stage.\n\n1. UI outputs and server renders _must_ match\n\nThe widget you use in the UI to return an output must correspond to the function used in the server to generate that output. In this example, we use `textOutput` in the UI so in the server we use `renderText`. Essentially all widgets in Shiny use this `Output` versus `render` syntax which can be a big help to visual checks that your app is written correctly. You will need to be sure that whatever the 'class' is, it is _lowercase_ in the UI but _title case_ in the server (i.e., only first letter capitalized).\n\n2. Use section header format\n\nThis app is relatively short but we think effectively hints at how long and convoluted purpose-built Shiny apps can easily become. So, we recommend using section headers in your Shiny app code. You can do this by putting either four hyphens or four hashtags at the end of a comment line (e.g., `# Section 1 ####` or `# My header ----`). Headings defined in this way will appear in the bottom left of the \"Source\" pane of RStudio next to a light orange hashtag symbol. Clicking the text in that area will open a drop-down menu showing all headings in your current file and clicking one of the other headings will instantly jump you to that heading. This can be _incredibly_ convenient when you're trying to navigate a several hundred line long Shiny app. While rainbow parentheses can be useful for avoiding typos _within a section_, section headers make it much easier to avoid typos _across_ sections.\n\nIf you don't use headings already (or your cursor is on a line before the first heading), the relevant bit of the \"Source\" pane will just say \"(Top Level)\" and will not have the golden hashtag symbol.\n\n## Including Data\n\nYou can also use your Shiny app to work with a full data table! When running your app locally, you only need to read in the data as you normally would then run the app. By having read in the data you will ensure the object is in your environment and accessible to the app. However, keep in mind this will only work in \"local\" (i.e., non-deployed) contexts. See our--admittedly brief--discussion of deployment at the end of this module.\n\nLet's explore an example using data about fiddler crabs (_Minuca pugnax_) at the [Plum Island Ecosystems (PIE) LTER](https://pie-lter.ecosystems.mbl.edu/welcome-plum-island-ecosystems-lter) site from the [`lterdatasampler` R package](https://lter.github.io/lterdatasampler/). The app we're about to create will make a graph between any two (numeric) columns.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load lterdatasampler package\nlibrary(lterdatasampler)\n\n# Load fiddler crab data\ndata(pie_crab) # <1>\n\n# Define the UI ---- \ndata_ui <- shiny::fluidPage(\n \n # Let the user choose which the X axis\n shiny::selectInput(inputId = \"x_vals\",\n label = \"Choose the X-axis\",\n choices = setdiff(x = names(pie_crab), # <2>\n y = c(\"date\", \"site\", \"name\")),\n selected = \"latitude\"),\n \n # Also the Y axis\n shiny::selectInput(inputId = \"y_vals\",\n label = \"Choose the Y-axis\",\n choices = setdiff(x = names(pie_crab),\n y = c(\"date\", \"site\", \"name\")),\n selected = \"size\"),\n \n # Return the desired plot\n shiny::plotOutput(outputId = \"crab_graph\")\n \n) # Close UI\n\n# Define the server ----\ndata_server <- function(input, output){\n \n # Reactively identify X & Y axess\n picked_x <- shiny::reactive({ input$x_vals }) # <3>\n picked_y <- shiny::reactive({ input$y_vals })\n \n # Create the desired graph\n output$crab_graph <- shiny::renderPlot(\n \n ggplot(pie_crab, aes(x = .data[[picked_x()]], y = .data[[picked_y()]])) + # <4>\n geom_point(aes(fill = .data[[picked_x()]]), pch = 21, size = 2.5) +\n labs(x = stringr::str_to_title(picked_x()),\n y = stringr::str_to_title(picked_y())) +\n theme_bw()\n \n ) # Close plot rendering\n \n} # Close server\n\n# Generate the app ----\nshiny::shinyApp(ui = data_ui, server = data_server)\n```\n:::\n\n1. Note the loading of the data is done _outside_ of the app! You can have the app load its own data but that is more complicated than this example needs to be.\n2. To make our life easier in the server we can exclude non-number columns\n3. See how we're reactively grabbing both axes?\n4. `ggplot2` requires special syntax to specify axes with quoted column names (which is how reactive Shiny elements from that widget are returned)\n\n## Layouts\n\nExperimenting with different app layouts can be a fun step in the process of making an app that is as effective as possible! We do recommend that during app development you stick with a very simple user interface because it'll be easier to make sure your inputs and outputs work as desired. Once you are satisfied with those elements you can relatively easily chengs the UI to help guide users through your app.\n\nAs implied by that preface, layouts are exclusively an element of the user interface! This is great when you have an app with a complicated server component because you won't need to mess with that at all to get the UI looking perfect. In the examples below, we'll generate a non-interactive app so that we can really emphasize the 'how to' perspective of using different layouts.\n\n### Sidebar\n\nOne of the more common Shiny UI choices is to use a sidebar. The sidebar typically takes up about one third of the width of the app while the remaining two thirds is taken up by the main panel. The sidebar can be nice place to put all the user inputs and have the outputs display in the main panel. This format allows for really clear visual separation between where you want the user to interact with the app versus where the results of their choices can be viewed.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Define the UI\nsidebar_ui <- shiny::fluidPage(\n \n # Define the layout type\n shiny::sidebarLayout( # <1>\n \n # Define what goes in the eponymous sidebar\n shiny::sidebarPanel(\n \"Hello from the sidebar!\"\n ), # <2>\n \n # Define what goes in the main panel\n shiny::mainPanel(\n \"Hello from the main panel!\"\n ) ) ) # <3>\n\n# Define the server\nbasic_server <- function(input, output){ }\n\n# Generate the app\nshiny::shinyApp(ui = sidebar_ui, server = basic_server)\n```\n:::\n\n1. Notice that everything else in the UI is wrapped inside this function. If you want something above/below the sidebar vs. main panel you'll need to put that content outside of this function's parentheses but still in the `fluidPage` parentheses\n2. Be careful not to forget this comma separating the `sidebarPanel` and `mainPanel` functions!\n3. Three closing parentheses are needed to close the UI elements. This is why it's _really_ helpful to use rainbow parentheses in your coding environment!\n\n### Tab Panels\n\nIf you feel that your app is better represented in separate pages, tab panels may be a better layout choice! The result of this layout is a series of discrete tabs along the top of your app. If the user clicks one of them they'll be able to look at a separate chunk of your app. Inputs in any tab are available to the app's server and can be outputs in any tab (remember that their is a shared server so it is impossible for it to be otherwise!). Generally it may be a good idea to have inputs and outputs in the same tab so that users can see the interactive app responding to their inputs rather than needing to click back and forth among tabs to see the results of their inputs. For example, you could have an app where users choose what goes on either axis of several graph types and put each graph type on its own tab of the larger Shiny app.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Define the UI\ntabs_ui <- shiny::fluidPage(\n \n# Define the layout type\n shiny::tabsetPanel( # <1>\n \n # Define what goes in the first tab\n shiny::tabPanel(title = \"Tab 1\",\n \"Hello from the first tab!\"\n ), # <2>\n \n # And in the second\n shiny::tabPanel(title = \"Tab 2\",\n \"Welcome to the second tab!\"\n ),\n \n # And so on\n shiny::tabPanel(title = \"Tab 3\",\n \"Hello yet again!\"\n ) ) ) # <3>\n\n# Define the server\nbasic_server <- function(input, output){ }\n\n# Generate the app\nshiny::shinyApp(ui = tabs_ui, server = basic_server)\n```\n:::\n\n1. This function is comparable to `sidebarLayout` in that if you want stuff above/below the tab panel area you'll need to be outside of this function's parentheses but still in the `fluidPage` parentheses\n2. Again, just like the `sidebarLayout` subfunctions, you'll need a comma after each UI element except the last one\n3. Here we're closing all of the nested UI functions\n\n### Other Layouts\n\nWe just briefly covered two layout options but hopefully this is a nice indication for the kind of flexibility in user interface that you can expect of Shiny apps! For more information, check out Posit's [Shiny Application Layout Guide](https://shiny.posit.co/r/articles/build/layout-guide/). That resource has some really nice examples of these and other layout options that will be well worth checking out as you begin your journey into Shiny.\n\n## Text Formatting\n\nBeyond making your app have an intuitive layout it can be really helpful to be able to do even simple text formatting to assist your app's users. For instance, you may want to use sub-headings within the same UI layout component but still want to draw a distinction between two sets of inputs. Additionally you may want to emphasize some tips for best results or hyperlink to your group's other products. All of these can be accomplished using text formatting tools that are readily available within Shiny.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load the `htmltools` library\nlibrary(htmltools)\n\n# Define the UI\ntext_ui <- shiny::fluidPage(\n \n # Let's make some headings\n htmltools::h1(\"This is a Big Heading\"), # <1>\n htmltools::h3(\"Smaller heading\"),\n htmltools::h5(\"Even smaller heading!\"), \n \n # Now we'll format more text in various (non-heading) ways\n htmltools::strong(\"Bold text\"),\n \n htmltools::br(), # <2>\n\n htmltools::a(href = \"https://lter.github.io/ssecr/mod_interactivity.html\",\n \"This text is hyperlinked\",\n target = \"_blank\"), # <3>\n \n htmltools::br(),\n \n htmltools::code(\"This is 'code' text\") ) # <4>\n\n# Define the server\nbasic_server <- function(input, output){ }\n\n# Generate the app\nshiny::shinyApp(ui = text_ui, server = basic_server)\n```\n:::\n\n1. Headings (of any size) automatically include a line break after the heading text\n2. The `br` function creates a line break\n3. When the `target` argument is set to \"_blank\" it will open a new tab when users click the hyperlinked text. This is ideal because if a user left your app to visit the new site they would lose all of their inputs\n4. Code text looks `like this`\n\n## Deployment\n\nWhen Shiny apps are only being used by those in your team, keeping them as a code script works well. However, if you'd like those outside of your team to be able to find your app as they would any other website you'll need to deploy your Shiny app. This process is outside of the scope of this module but is often the end goal of Shiny app development.\n\nTake a look at [Posit's instructions for deployment](https://shiny.posit.co/r/articles/share/deployment-web/) for more details but essentially \"deployment\" is the process of getting your local app hosted on shinyapps.io which gives it a link that anyone can use to access/run your app on their web browser of choice.\n\n## Additional Interactivity Resources\n\n### Papers & Documents\n\n- \n\n### Workshops & Courses\n\n- Posit's [Welcome to Shiny](https://shiny.posit.co/r/getstarted/shiny-basics/lesson1/index.html) (for R coders)\n- 2022 All Scientists' Meeting [Shiny Apps for Sharing Science](https://njlyon0.github.io/asm-2022_shiny-workshop/) workshop\n\n### Websites\n\n- Posit's [Shiny](https://shiny.posit.co/) website\n", + "markdown": "---\ntitle: \"Creating Interactive Apps\"\ncode-annotations: hover\nexecute: \n eval: false\n---\n\n\n## Overview\n\nShiny is a popular tool that allows users to build interactive web applications without the normally pre-requisite web development expertise. In addition to Shiny apps being simpler to build for the programmer they are often used to allow visitors to perform coding tasks without ever actually writing code. These are huge advantages because they reduce or eliminate significant technical barriers in developing truly interactive applications.\n\nIn synthesis contexts, Shiny can be used for a variety of valuable purposes. You can use it to develop dashboards for sharing data with related communities, allow your team to quickly \"play with\" exploratory graphs, or even to create a data submission portal (as is the case with some Research Coordination Networks or \"RCNs\").\n\nNote that Shiny can be built in either R or Python 'under the hood' but for the purposes of this module we'll focus on R.\n\n## Learning Objectives\n\nAfter completing this topic you will be able to: \n\n- Define the three fundamental components of a Shiny app\n- Explain benefits and limitations of interactive approaches to data exploration\n- Generate an interactive app with Shiny\n- Use text formatting methods in a Shiny app\n- Explore available Shiny layout options\n- Create a Shiny app\n- Describe (briefly) the purpose of deploying a Shiny app\n\n## Needed Packages\n\nIf you'd like to follow along with the code chunks included throughout this module, you'll need to install the following packages:\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Note that these lines only need to be run once per computer\n## So you can skip this step if you've installed these before\ninstall.packages(\"tidyverse\")\ninstall.packages(\"shiny\")\ninstall.packages(\"htmltools\")\ninstall.packages(\"lterdatasampler\")\n```\n:::\n\n\nWe'll load the Tidyverse meta-package here to have access to many of its useful tools when we need them later as well as the `shiny` package.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load needed libraries\nlibrary(tidyverse); library(shiny)\n```\n:::\n\n\n## Shiny Fundamentals\n\nAll Shiny apps are composed of three pieces: a user interface (UI), a server, and a call to the `shinyApp` function. The user interface includes everything that the user sees and can interact with; note that this includes _both_ inputs and outputs. The server is responsible for all code operations performed on user inputs in order to generate outputs specified in the UI. The server is _not_ available to the user. Finally, the `shinyApp` function simply binds together the UI and server and creates a living app. The app appears either in your RStudio or in a new tab on a web browser depending on your settings.\n\nFor those of you who write your own functions, you may notice that the syntax of Shiny is **very** similar to the syntax of functions. If you have not already, your quality of life will benefit greatly if you turn on \"rainbow parentheses\" in RStudio (Tools {{< fa arrow-right >}} Global Options {{< fa arrow-right >}} Code {{< fa arrow-right >}} Display {{< fa arrow-right >}} Check \"Use rainbow parentheses\" box).\n\nLet's consider an artificially simple Shiny app so you can get a sense for the fundamental architecture of this tool.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Define the UI\nbasic_ui <- shiny::fluidPage( # <1>\n \"Hello there!\"\n)\n\n# Define the server\nbasic_server <- function(input, output){ } # <2>\n\n# Generate the app\nshiny::shinyApp(ui = basic_ui, server = basic_server)\n```\n:::\n\n1. The `fluidPage` function is important for leaving flexibility in UI layout which we'll explore later in the module\n2. Because this app has no inputs or outputs, it doesn't need anything in the 'server' component (though it still does require an empty server!)\n\nIf you copy and run the above code, you should see an app that is a blank white page with \"Hello there!\" written in the top left in plain text. Congratulations, you have now made your first Shiny app! Now, your reason for exploring this module likely involves an app that actually does something but the fundamental structure of all apps--even skeletal apps like this one--is the same. More complicated apps will certainly have more content in the UI and server sections but all Shiny apps will have this tripartite structure.\n\n## Interactive Apps\n\nNow that we've covered non-reactive apps, let's create an interactive one! It is important to remember that the user interface needs to contain _both_ the inputs the user can make _and_ the outputs determined by those inputs. The server will be responsible for turning the inputs into outputs but if you want your interactive app to actually show the user the interactivity you need to be careful to include the outputs in the UI.\n\nEssentially all Shiny UI functions use the same syntax of `Input` or `Output`. So, determining how you want the user to engage with your app is sometimes as straightforward as identifying the class of the value you want them to interact with. Shiny calls these helper functions \"widgets\".\n\nLet's consider an app that accepts a single number and returns the square root of that number.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Define the UI ---- \nreactive_ui <- shiny::fluidPage(\n \n # Create input\n shiny::numericInput(inputId = \"num_in\", # <1>\n label = \"Type a number\",\n value = 16),\n \n # Include some plain text for contextualizing the output\n \"Square root is: \", # <2>\n \n # Create output\n shiny::textOutput(outputId = \"num_out\")\n \n) # Close UI\n\n# Define server ----\nreactive_server <- function(input, output){\n \n # Reactively accept the input and take the square root of it\n root <- shiny::reactive({ # <3>\n sqrt(x = input$num_in) # <4>\n })\n \n # Make that value an output of the server/app\n output$num_out <- shiny::renderText( # <5>\n expr = root() # <6>\n ) \n \n} # Close server\n\n# Generate the app ----\nshiny::shinyApp(ui = reactive_ui, server = reactive_server)\n```\n:::\n\n1. Note that the argument name is capital \"I\" but _lowercase_ \"d\". Typing `inputID` is a common and frustrating source of error for Shiny app developers\n2. Every element of the UI--except the last one--needs to end with a comma\n3. All reactive elements (i.e., those that change as soon as the user changes an input) need to be specified inside of `reactive` with _both_ parentheses and curly braces\n4. The name of this input exactly matches the `inputId` we defined in the UI. That it _is_ an input is defined by our use of the `numericInput` widget\n5. The name of this output exactly matches the `outputId` we told the UI to expect.\n6. Reactive elements essentially become functions in their own right! So, when we want to use them, we need to include empty parentheses next to their name\n\nWe included a _lot_ of footnote annotations in that code chunk to help provide context but there are a few small comments that are worthwhile to bring up at this stage.\n\n1. UI outputs and server renders _must_ match\n\nThe widget you use in the UI to return an output must correspond to the function used in the server to generate that output. In this example, we use `textOutput` in the UI so in the server we use `renderText`. Essentially all widgets in Shiny use this `Output` versus `render` syntax which can be a big help to visual checks that your app is written correctly. You will need to be sure that whatever the 'class' is, it is _lowercase_ in the UI but _title case_ in the server (i.e., only first letter capitalized).\n\n2. Use section header format\n\nThis app is relatively short but we think effectively hints at how long and convoluted purpose-built Shiny apps can easily become. So, we recommend using section headers in your Shiny app code. You can do this by putting either four hyphens or four hashtags at the end of a comment line (e.g., `# Section 1 ####` or `# My header ----`). Headings defined in this way will appear in the bottom left of the \"Source\" pane of RStudio next to a light orange hashtag symbol. Clicking the text in that area will open a drop-down menu showing all headings in your current file and clicking one of the other headings will instantly jump you to that heading. This can be _incredibly_ convenient when you're trying to navigate a several hundred line long Shiny app. While rainbow parentheses can be useful for avoiding typos _within a section_, section headers make it much easier to avoid typos _across_ sections.\n\nIf you don't use headings already (or your cursor is on a line before the first heading), the relevant bit of the \"Source\" pane will just say \"(Top Level)\" and will not have the golden hashtag symbol.\n\n## Including Data\n\nYou can also use your Shiny app to work with a full data table! When running your app locally, you only need to read in the data as you normally would then run the app. By having read in the data you will ensure the object is in your environment and accessible to the app. However, keep in mind this will only work in \"local\" (i.e., non-deployed) contexts. See our--admittedly brief--discussion of deployment at the end of this module.\n\nLet's explore an example using data about fiddler crabs (_Minuca pugnax_) at the [Plum Island Ecosystems (PIE) LTER](https://pie-lter.ecosystems.mbl.edu/welcome-plum-island-ecosystems-lter) site from the [`lterdatasampler` R package](https://lter.github.io/lterdatasampler/). The app we're about to create will make a graph between any two (numeric) columns.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load lterdatasampler package\nlibrary(lterdatasampler)\n\n# Load fiddler crab data\ndata(pie_crab) # <1>\n\n# Define the UI ---- \ndata_ui <- shiny::fluidPage(\n \n # Let the user choose which the X axis\n shiny::selectInput(inputId = \"x_vals\",\n label = \"Choose the X-axis\",\n choices = setdiff(x = names(pie_crab), # <2>\n y = c(\"date\", \"site\", \"name\")),\n selected = \"latitude\"),\n \n # Also the Y axis\n shiny::selectInput(inputId = \"y_vals\",\n label = \"Choose the Y-axis\",\n choices = setdiff(x = names(pie_crab),\n y = c(\"date\", \"site\", \"name\")),\n selected = \"size\"),\n \n # Return the desired plot\n shiny::plotOutput(outputId = \"crab_graph\")\n \n) # Close UI\n\n# Define the server ----\ndata_server <- function(input, output){\n \n # Reactively identify X & Y axess\n picked_x <- shiny::reactive({ input$x_vals }) # <3>\n picked_y <- shiny::reactive({ input$y_vals })\n \n # Create the desired graph\n output$crab_graph <- shiny::renderPlot(\n \n ggplot(pie_crab, aes(x = .data[[picked_x()]], y = .data[[picked_y()]])) + # <4>\n geom_point(aes(fill = .data[[picked_x()]]), pch = 21, size = 2.5) +\n labs(x = stringr::str_to_title(picked_x()),\n y = stringr::str_to_title(picked_y())) +\n theme_bw()\n \n ) # Close plot rendering\n \n} # Close server\n\n# Generate the app ----\nshiny::shinyApp(ui = data_ui, server = data_server)\n```\n:::\n\n1. Note the loading of the data is done _outside_ of the app! You can have the app load its own data but that is more complicated than this example needs to be.\n2. To make our life easier in the server we can exclude non-number columns\n3. See how we're reactively grabbing both axes?\n4. `ggplot2` requires special syntax to specify axes with quoted column names (which is how reactive Shiny elements from that widget are returned)\n\n## Layouts\n\nExperimenting with different app layouts can be a fun step in the process of making an app that is as effective as possible! We do recommend that during app development you stick with a very simple user interface because it'll be easier to make sure your inputs and outputs work as desired. Once you are satisfied with those elements you can relatively easily chengs the UI to help guide users through your app.\n\nAs implied by that preface, layouts are exclusively an element of the user interface! This is great when you have an app with a complicated server component because you won't need to mess with that at all to get the UI looking perfect. In the examples below, we'll generate a non-interactive app so that we can really emphasize the 'how to' perspective of using different layouts.\n\n### Sidebar\n\nOne of the more common Shiny UI choices is to use a sidebar. The sidebar typically takes up about one third of the width of the app while the remaining two thirds is taken up by the main panel. The sidebar can be nice place to put all the user inputs and have the outputs display in the main panel. This format allows for really clear visual separation between where you want the user to interact with the app versus where the results of their choices can be viewed.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Define the UI\nsidebar_ui <- shiny::fluidPage(\n \n # Define the layout type\n shiny::sidebarLayout( # <1>\n \n # Define what goes in the eponymous sidebar\n shiny::sidebarPanel(\n \"Hello from the sidebar!\"\n ), # <2>\n \n # Define what goes in the main panel\n shiny::mainPanel(\n \"Hello from the main panel!\"\n ) ) ) # <3>\n\n# Define the server\nbasic_server <- function(input, output){ }\n\n# Generate the app\nshiny::shinyApp(ui = sidebar_ui, server = basic_server)\n```\n:::\n\n1. Notice that everything else in the UI is wrapped inside this function. If you want something above/below the sidebar vs. main panel you'll need to put that content outside of this function's parentheses but still in the `fluidPage` parentheses\n2. Be careful not to forget this comma separating the `sidebarPanel` and `mainPanel` functions!\n3. Three closing parentheses are needed to close the UI elements. This is why it's _really_ helpful to use rainbow parentheses in your coding environment!\n\n### Tab Panels\n\nIf you feel that your app is better represented in separate pages, tab panels may be a better layout choice! The result of this layout is a series of discrete tabs along the top of your app. If the user clicks one of them they'll be able to look at a separate chunk of your app. Inputs in any tab are available to the app's server and can be outputs in any tab (remember that their is a shared server so it is impossible for it to be otherwise!). Generally it may be a good idea to have inputs and outputs in the same tab so that users can see the interactive app responding to their inputs rather than needing to click back and forth among tabs to see the results of their inputs. For example, you could have an app where users choose what goes on either axis of several graph types and put each graph type on its own tab of the larger Shiny app.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Define the UI\ntabs_ui <- shiny::fluidPage(\n \n# Define the layout type\n shiny::tabsetPanel( # <1>\n \n # Define what goes in the first tab\n shiny::tabPanel(title = \"Tab 1\",\n \"Hello from the first tab!\"\n ), # <2>\n \n # And in the second\n shiny::tabPanel(title = \"Tab 2\",\n \"Welcome to the second tab!\"\n ),\n \n # And so on\n shiny::tabPanel(title = \"Tab 3\",\n \"Hello yet again!\"\n ) ) ) # <3>\n\n# Define the server\nbasic_server <- function(input, output){ }\n\n# Generate the app\nshiny::shinyApp(ui = tabs_ui, server = basic_server)\n```\n:::\n\n1. This function is comparable to `sidebarLayout` in that if you want stuff above/below the tab panel area you'll need to be outside of this function's parentheses but still in the `fluidPage` parentheses\n2. Again, just like the `sidebarLayout` subfunctions, you'll need a comma after each UI element except the last one\n3. Here we're closing all of the nested UI functions\n\n### Other Layouts\n\nWe just briefly covered two layout options but hopefully this is a nice indication for the kind of flexibility in user interface that you can expect of Shiny apps! For more information, check out Posit's [Shiny Application Layout Guide](https://shiny.posit.co/r/articles/build/layout-guide/). That resource has some really nice examples of these and other layout options that will be well worth checking out as you begin your journey into Shiny.\n\n## Text Formatting\n\nBeyond making your app have an intuitive layout it can be really helpful to be able to do even simple text formatting to assist your app's users. For instance, you may want to use sub-headings within the same UI layout component but still want to draw a distinction between two sets of inputs. Additionally you may want to emphasize some tips for best results or hyperlink to your group's other products. All of these can be accomplished using text formatting tools that are readily available within Shiny.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load the `htmltools` library\nlibrary(htmltools)\n\n# Define the UI\ntext_ui <- shiny::fluidPage(\n \n # Let's make some headings\n htmltools::h1(\"This is a Big Heading\"), # <1>\n htmltools::h3(\"Smaller heading\"),\n htmltools::h5(\"Even smaller heading!\"), \n \n # Now we'll format more text in various (non-heading) ways\n htmltools::strong(\"Bold text\"),\n \n htmltools::br(), # <2>\n\n htmltools::a(href = \"https://lter.github.io/ssecr/mod_interactivity.html\",\n \"This text is hyperlinked\",\n target = \"_blank\"), # <3>\n \n htmltools::br(),\n \n htmltools::code(\"This is 'code' text\") ) # <4>\n\n# Define the server\nbasic_server <- function(input, output){ }\n\n# Generate the app\nshiny::shinyApp(ui = text_ui, server = basic_server)\n```\n:::\n\n1. Headings (of any size) automatically include a line break after the heading text\n2. The `br` function creates a line break\n3. When the `target` argument is set to \"_blank\" it will open a new tab when users click the hyperlinked text. This is ideal because if a user left your app to visit the new site they would lose all of their inputs\n4. Code text looks `like this`\n\n## Deployment\n\nWhen Shiny apps are only being used by those in your team, keeping them as a code script works well. However, if you'd like those outside of your team to be able to find your app as they would any other website you'll need to deploy your Shiny app. This process is outside of the scope of this module but is often the end goal of Shiny app development.\n\nTake a look at [Posit's instructions for deployment](https://shiny.posit.co/r/articles/share/deployment-web/) for more details but essentially \"deployment\" is the process of getting your local app hosted on shinyapps.io which gives it a link that anyone can use to access/run your app on their web browser of choice.\n\n## Additional Interactivity Resources\n\n### Papers & Documents\n\n- \n\n### Workshops & Courses\n\n- Posit. [Welcome to Shiny](https://shiny.posit.co/r/getstarted/shiny-basics/lesson1/index.html). **2024**.\n- Lyon, N.J., _et al._ [Shiny Apps for Sharing Science](https://njlyon0.github.io/asm-2022_shiny-workshop/). **2022**.\n\n### Websites\n\n- \n", "supporting": [], "filters": [ "rmarkdown/pagebreak.lua" diff --git a/_freeze/mod_reports/execute-results/html.json b/_freeze/mod_reports/execute-results/html.json index f3cbd06..42f8495 100644 --- a/_freeze/mod_reports/execute-results/html.json +++ b/_freeze/mod_reports/execute-results/html.json @@ -1,8 +1,8 @@ { - "hash": "33f094d629b511b27ca7aae08eb2460b", + "hash": "91f7393d7676a6032599900675f9f935", "result": { "engine": "knitr", - "markdown": "---\ntitle: \"Reproducible Reports\"\ncode-annotations: hover\n---\n\n\n## Overview\n\nAt this point in the course, we anticipate that you're likely approaching the end of your team's synthesis project (see our suggested [milestones page](https://lter.github.io/ssecr/proj_milestones.html) for more information). As the end of your project and the course as a whole nears, it might be valuable for your group to consider how you can reproducibly document all of the work you've been doing for the last several months. \"**Computational notebooks**\" (e.g., Quarto documents, Jupyter Notebooks, R Markdown files, etc.) can be a reproducible way of documenting your results in a format that allows you to leverage _both_ your technical skills and your scientific communication skills.\n\nThis module focuses on the structure and content of these notebooks from a primarily technical lens, so please consult the [communicating findings](https://lter.github.io/ssecr/mod_findings.html) module for the team science perspective.\n\n## Learning Objectives\n\nAfter completing this module you will be able to: \n\n- Describe contexts where computational notebooks are useful\n- Identify the three fundamental elements of a typical notebook\n- Use Markdown syntax to accomplish text styling\n- Create notebooks that include a blend of plain text and embedded code chunks\n- Make a Quarto website\n- Define the purpose of GitHub Actions\n- Deploy a notebook with GitHub Pages\n\n## Notebook Structure & Value\n\nFiles that combine plain text with embedded code chunks are an excellent way of reproducibly documenting workflows and faciliating conversations about said workflows (or their outputs). Examples of notebook files include Quarto documents, Jupyter Notebooks, and R Markdown files. Regardless of the specific type, all of them function in the same way. Each of them allows you to use code chunks in the same way that you might use a typical script but between the code chunks you can add--and format--plain, human-readable text. Arguably you could do this with comments in a script but this format is built around the idea that this plain text is **intended to be interpretable without any prior coding experience**. The plain text can be formatted with Markdown syntax (we'll discuss that in greater depth later) but even unformatted text outside of code chunks is visually 'easier on the eyes' than comment lines in scripts.\n\nAnother shared facet of notebook interfaces is that they are meant to be \"rendered\" (a.k.a. \"knit\") to produce a different file type that translates code chunks and Markdown-formatted text into something that looks much more similar to what you might produce in Microsoft Word or a Google Doc. Typically such files are rendered into PDFs or HTMLs though there are other output options. These rendered files can then be shared (and opened) outside of coding platforms and thus make their content even more accessible to non-coders.\n\nIn synthesis work these reports can be especially valuable because your team may include those with a wealth of discipline insight but not necessarily coding literacy. Creating reports with embedded code can enable these collaborators to engage more fully than they might otherwise be able to if there was essentially a minimum threshold of coding literacy required in order to contribute. These reports can also be useful documentation of past coding decisions and serve as reminders for judgement calls for which no one in the team remembers the rationale.\n\n### Structural Elements\n\nTypically, these notebooks have three structural components:\n\n1. YAML\n - Pronounced '_YEAH-mull_'\n2. Plain text\n - Possibly formatted with Markdown syntax\n3. Embedded code chunks\n\n:::{.panel-tabset}\n\n#### YAML\n\nThe YAML (Yet Another Markup Language) defines document-level metadata. Most fundamentally, the YAML defines what file type will be produced when the report is rendered. It can also be used to define the top-level title, author, and date information. Finally, it can change the default options for code chunks throughout the document (more on code chunk options elsewhere).\n\nDifferent notebook file types will specify the YAML differently but in both Quarto documents and R Markdown files, the YAML is defined in the first few lines of the report and starts/ends with a line containing three hyphens. This looks something like this:\n\n```{.bash}\n---\ntitle: \"Reproducible Reports\"\noutput: html_document\n---\n```\n\n#### Markdown Text\n\nThe text outside of the YAML and code chunks is plain text that accepts Markdown syntax to accomplish text format tweaks. Dedicated text-formatting software (e.g., Microsoft Word, Gmail, etc.) provides buttons and hot keys for these sorts of format alterations but many programming IDEs do not provide such user interface elements.[^1] Markdown syntax is used to support this same functionality.\n\nSome fundamental Markdown options include:\n\n- \\*\\*bold text\\*\\* {{< fa arrow-right >}} **bold text**\n- \\_italic text\\_ {{< fa arrow-right >}} _italic text_\n- \\`code text\\` {{< fa arrow-right >}} `code text`\n- \\[hyperlinked text\\]\\(https://lter.github.io/ssecr/mod_reports.html\\) {{< fa arrow-right >}} [hyperlinked text](https://lter.github.io/ssecr/mod_reports.html)\n\nFor a more complete glossary of fundamental Markdown syntax options see [here](https://www.markdownguide.org/basic-syntax/). You may also want to explore [the 'back end' of this course's website](https://github.com/lter/ssecr) as _every_ page is built using computational notebook files.\n\n[^1]: Note that with the addition of the \"Visual\" tab in RStudio there are button options for many text format changes. Markdown syntax is still useful to know for general knowledge reasons though! \n\n#### Code Chunks\n\nThe code chunks embedded in notebooks are essentially a fragmented script containing runable code. These chunks may contain code and/or comments and share an environment with one another when rendered (i.e., if you load a particular library in one chunk you'll be able to use functions from that library in subsequent chunks). When used in concert with the Markdown text in a given notebook the code chunks can be used to effectively demonstrate a workflow while providing as much human-readable context as is desired.\n\nIn Quarto documents, code chunks look like this[^2]:\n\n```{.bash}\n\\```{r demo-chunk}\n#| echo: true # <1>\n\n# Round pi to 2 digits #<2>\nround(x = pi, digits = 2)\n\\```\n```\n1. You may specify chunk-specific options using this syntax (i.e., `#| option_name: option_setting`). Options you want to apply to _all_ chunks across a notebook should be specified once in the YAML and can exclude the leading `#|` bit of the format.\n2. If your Markdown text provides sufficient context you may exclude comments in code chunks but opinions differ on which is \"more\" appropriate\n\n\n::: {.cell}\n::: {.cell-output .cell-output-stdout}\n\n```\n[1] 3.14\n```\n\n\n:::\n:::\n\n\n[^2]: Normally code chunks start and end with three backticks (\\`\\`\\`) but to embed this code chunk example we need to \"escape\" the first backtick (using a backslash) so that the notebook interprets it correctly.\n\n:::\n\n\n\n### Script vs. Report Decision\n\nSee [here](https://nceas.github.io/scicomp.github.io/best_practices.html#r-scripts-versus-r-markdowns) for more information.\n\n\n\n## Applications\n\n- Static PDF/HTML files\n- Full manuscripts\n- Deployed website\n\n## Deployment & GitHub\n\n\n## Additional Resources\n\n### Papers & Documents\n\n- \n\n### Workshops & Courses\n\n- NCEAS coreR [Literate Analysis with Quarto](https://learning.nceas.ucsb.edu/2023-10-coreR/session_04.html) session\n- OSS [Reproducible Papers with RMarkdown](https://nceas.github.io/oss-lessons/reproducible-papers-with-rmd/reproducible-papers-with-rmd.html)\n- UCSB's Master of Environmental Data Science (MEDS) [Creating your Personal Website using Quarto](https://ucsb-meds.github.io/creating-quarto-websites/) lesson\n\n### Websites\n\n- Markdown Guide: [Basic Syntax](https://www.markdownguide.org/basic-syntax/)\n- Posit's [Welcome to Quarto](https://quarto.org/)\n", + "markdown": "---\ntitle: \"Reproducible Reports\"\ncode-annotations: hover\n---\n\n\n## Overview\n\nAt this point in the course, we anticipate that you're likely approaching the end of your team's synthesis project (see our suggested [milestones page](https://lter.github.io/ssecr/proj_milestones.html) for more information). As the end of your project and the course as a whole nears, it might be valuable for your group to consider how you can reproducibly document all of the work you've been doing for the last several months. \"**Computational notebooks**\" (e.g., Quarto documents, Jupyter Notebooks, R Markdown files, etc.) can be a reproducible way of documenting your results in a format that allows you to leverage _both_ your technical skills and your scientific communication skills.\n\nThis module focuses on the structure and content of these notebooks from a primarily technical lens, so please consult the [communicating findings](https://lter.github.io/ssecr/mod_findings.html) module for the team science perspective.\n\n## Learning Objectives\n\nAfter completing this module you will be able to: \n\n- Describe contexts where computational notebooks are useful\n- Identify the three fundamental elements of a typical notebook\n- Use Markdown syntax to accomplish text styling\n- Create notebooks that include a blend of plain text and embedded code chunks\n- Make a Quarto website\n- Define the purpose of GitHub Actions\n- Deploy a notebook with GitHub Pages\n\n## Notebook Structure & Value\n\nFiles that combine plain text with embedded code chunks are an excellent way of reproducibly documenting workflows and faciliating conversations about said workflows (or their outputs). Examples of notebook files include Quarto documents, Jupyter Notebooks, and R Markdown files. Regardless of the specific type, all of them function in the same way. Each of them allows you to use code chunks in the same way that you might use a typical script but between the code chunks you can add--and format--plain, human-readable text. Arguably you could do this with comments in a script but this format is built around the idea that this plain text is **intended to be interpretable without any prior coding experience**. The plain text can be formatted with Markdown syntax (we'll discuss that in greater depth later) but even unformatted text outside of code chunks is visually 'easier on the eyes' than comment lines in scripts.\n\nAnother shared facet of notebook interfaces is that they are meant to be \"rendered\" (a.k.a. \"knit\") to produce a different file type that translates code chunks and Markdown-formatted text into something that looks much more similar to what you might produce in Microsoft Word or a Google Doc. Typically such files are rendered into PDFs or HTMLs though there are other output options. These rendered files can then be shared (and opened) outside of coding platforms and thus make their content even more accessible to non-coders.\n\nIn synthesis work these reports can be especially valuable because your team may include those with a wealth of discipline insight but not necessarily coding literacy. Creating reports with embedded code can enable these collaborators to engage more fully than they might otherwise be able to if there was essentially a minimum threshold of coding literacy required in order to contribute. These reports can also be useful documentation of past coding decisions and serve as reminders for judgement calls for which no one in the team remembers the rationale.\n\n### Structural Elements\n\nTypically, these notebooks have three structural components:\n\n1. YAML\n - Pronounced '_YEAH-mull_'\n2. Plain text\n - Possibly formatted with Markdown syntax\n3. Embedded code chunks\n\n:::{.panel-tabset}\n\n#### YAML\n\nThe YAML (Yet Another Markup Language) defines document-level metadata. Most fundamentally, the YAML defines what file type will be produced when the report is rendered. It can also be used to define the top-level title, author, and date information. Finally, it can change the default options for code chunks throughout the document (more on code chunk options elsewhere).\n\nDifferent notebook file types will specify the YAML differently but in both Quarto documents and R Markdown files, the YAML is defined in the first few lines of the report and starts/ends with a line containing three hyphens. This looks something like this:\n\n```{.bash}\n---\ntitle: \"Reproducible Reports\"\noutput: html_document\n---\n```\n\n#### Markdown Text\n\nThe text outside of the YAML and code chunks is plain text that accepts Markdown syntax to accomplish text format tweaks. Dedicated text-formatting software (e.g., Microsoft Word, Gmail, etc.) provides buttons and hot keys for these sorts of format alterations but many programming IDEs do not provide such user interface elements.[^1] Markdown syntax is used to support this same functionality.\n\nSome fundamental Markdown options include:\n\n- \\*\\*bold text\\*\\* {{< fa arrow-right >}} **bold text**\n- \\_italic text\\_ {{< fa arrow-right >}} _italic text_\n- \\`code text\\` {{< fa arrow-right >}} `code text`\n- \\[hyperlinked text\\]\\(https://lter.github.io/ssecr/mod_reports.html\\) {{< fa arrow-right >}} [hyperlinked text](https://lter.github.io/ssecr/mod_reports.html)\n\nFor a more complete glossary of fundamental Markdown syntax options see [here](https://www.markdownguide.org/basic-syntax/). You may also want to explore [the 'back end' of this course's website](https://github.com/lter/ssecr) as _every_ page is built using computational notebook files.\n\n[^1]: Note that with the addition of the \"Visual\" tab in RStudio there are button options for many text format changes. Markdown syntax is still useful to know for general knowledge reasons though! \n\n#### Code Chunks\n\nThe code chunks embedded in notebooks are essentially a fragmented script containing runable code. These chunks may contain code and/or comments and share an environment with one another when rendered (i.e., if you load a particular library in one chunk you'll be able to use functions from that library in subsequent chunks). When used in concert with the Markdown text in a given notebook the code chunks can be used to effectively demonstrate a workflow while providing as much human-readable context as is desired.\n\nIn Quarto documents, code chunks look like this[^2]:\n\n```{.bash}\n\\```{r demo-chunk}\n#| echo: true # <1>\n\n# Round pi to 2 digits #<2>\nround(x = pi, digits = 2)\n\\```\n```\n1. You may specify chunk-specific options using this syntax (i.e., `#| option_name: option_setting`). Options you want to apply to _all_ chunks across a notebook should be specified once in the YAML and can exclude the leading `#|` bit of the format.\n2. If your Markdown text provides sufficient context you may exclude comments in code chunks but opinions differ on which is \"more\" appropriate\n\n\n::: {.cell}\n::: {.cell-output .cell-output-stdout}\n\n```\n[1] 3.14\n```\n\n\n:::\n:::\n\n\n[^2]: Normally code chunks start and end with three backticks (\\`\\`\\`) but to embed this code chunk example we need to \"escape\" the first backtick (using a backslash) so that the notebook interprets it correctly.\n\n:::\n\n\n\n### Script vs. Report Decision\n\nSee [here](https://nceas.github.io/scicomp.github.io/best_practices.html#r-scripts-versus-r-markdowns) for more information.\n\n\n\n## Applications\n\n- Static PDF/HTML files\n- Full manuscripts\n- Deployed website\n\n## Deployment & GitHub\n\n\n## Additional Resources\n\n### Papers & Documents\n\n- \n\n### Workshops & Courses\n\n- Csik, S. [Creating Your Personal Website Using Quarto](https://ucsb-meds.github.io/creating-quarto-websites/). **2024**.\n- National Center for Ecological Analysis and Synthesis (NCEAS) Learning Hub. [coreR: Literate Analysis with Quarto](https://learning.nceas.ucsb.edu/2023-10-coreR/session_04.html). **2023**.\n- Mecum, B. [Reproducible Papers with RMarkdown](https://nceas.github.io/oss-lessons/reproducible-papers-with-rmd/reproducible-papers-with-rmd.html). **2019**.\n\n### Websites\n- [The Markdown Guide: Basic Syntax](https://www.markdownguide.org/basic-syntax/).\n- Posit. [Welcome to Quarto](https://quarto.org/)\n", "supporting": [], "filters": [ "rmarkdown/pagebreak.lua" diff --git a/_freeze/mod_spatial/execute-results/html.json b/_freeze/mod_spatial/execute-results/html.json index d8c24f1..cac49f0 100644 --- a/_freeze/mod_spatial/execute-results/html.json +++ b/_freeze/mod_spatial/execute-results/html.json @@ -1,8 +1,8 @@ { - "hash": "89a9890b585bd7bfc4448edcb321aa6c", + "hash": "760cf1e32e85a128291454cecfdf7125", "result": { "engine": "knitr", - "markdown": "---\ntitle: \"Working with Spatial Data\"\ncode-annotations: hover\n---\n\n\n\n\n## Overview\n\nSynthesis projects often have need of spatial datasets. At its simplest, it can be helpful to have a map of the original project locations including in the synthesis dataset. In more complex instances you want to extract spatial data within a certain area of sampling locations. Regardless of 'why' you're using spatial data, it may come up during your primary or synthesis work and thus deserves consideration in this course's materials. There are _many_ modes of working with spatial data, and not all of these tools require coding literacy but for consistency with the rest of the modules **this module will focus on _scripted_ approaches to interacting with spatial data**.\n\n## Learning Objectives\n\nAfter completing this topic you will be able to: \n\n- Define the difference between the two major types of spatial data\n- Manipulate spatial data with R\n- Create maps using spatial data\n- Integrate spatial data with tabular data\n\n## Needed Packages\n\nIf you'd like to follow along with the code chunks included throughout this module, you'll need to install the following packages:\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Note that these lines only need to be run once per computer\n## So you can skip this step if you've installed these before\ninstall.packages(\"tidyverse\")\ninstall.packages(\"sf\")\ninstall.packages(\"terra\")\ninstall.packages(\"maps\")\ninstall.packages(\"exactextractr\")\n```\n:::\n\n\n## Types of Spatial Data\n\nThere are two main types of spatial data: vector and raster. Both types (and the packages they require) are described in the tabs below.\n\n:::{.panel-tabset}\n### Vector Data\n\nVector data are stored as polygons. Essentially vector data are a set of points and--sometimes--the lines between them that define the edges of a shape. They may store additional data that is retained in a semi-tabular format that relates to the polygon(s) but isn't directly stored in them.\n\nCommon vector data types include shape files or GeoJSONs.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load needed library\nlibrary(sf)\n\n# Read in shapefile\nnc_poly <- sf::st_read(dsn = file.path(\"data\", \"nc_borders.shp\")) #<1>\n```\n:::\n\n1. Note that even though we're only specifying the \".shp\" file in this function you _must_ also have the associated files in that same folder. In this case that includes a \".dbf\", \".prj\", and \".shx\", though in other contexts you may have others.\n\nOnce you have read in the shapefile, you can check its structure as you would any other data object. Note that the object has both the 'data.frame' class and the 'sf' (\"simple features\") class. In this case, the shapefile relates to counties in North Carolina and some associated demographic data in those counties.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Check structure\nstr(nc_poly)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\nClasses 'sf' and 'data.frame':\t100 obs. of 15 variables:\n $ AREA : num 0.114 0.061 0.143 0.07 0.153 0.097 0.062 0.091 0.118 0.124 ...\n $ PERIMETER: num 1.44 1.23 1.63 2.97 2.21 ...\n $ CNTY_ : num 1825 1827 1828 1831 1832 ...\n $ CNTY_ID : num 1825 1827 1828 1831 1832 ...\n $ NAME : chr \"Ashe\" \"Alleghany\" \"Surry\" \"Currituck\" ...\n $ FIPS : chr \"37009\" \"37005\" \"37171\" \"37053\" ...\n $ FIPSNO : num 37009 37005 37171 37053 37131 ...\n $ CRESS_ID : int 5 3 86 27 66 46 15 37 93 85 ...\n $ BIR74 : num 1091 487 3188 508 1421 ...\n $ SID74 : num 1 0 5 1 9 7 0 0 4 1 ...\n $ NWBIR74 : num 10 10 208 123 1066 ...\n $ BIR79 : num 1364 542 3616 830 1606 ...\n $ SID79 : num 0 3 6 2 3 5 2 2 2 5 ...\n $ NWBIR79 : num 19 12 260 145 1197 ...\n $ geometry :sfc_MULTIPOLYGON of length 100; first list element: List of 1\n ..$ :List of 1\n .. ..$ : num [1:27, 1:2] -81.5 -81.5 -81.6 -81.6 -81.7 ...\n ..- attr(*, \"class\")= chr [1:3] \"XY\" \"MULTIPOLYGON\" \"sfg\"\n - attr(*, \"sf_column\")= chr \"geometry\"\n - attr(*, \"agr\")= Factor w/ 3 levels \"constant\",\"aggregate\",..: NA NA NA NA NA NA NA NA NA NA ...\n ..- attr(*, \"names\")= chr [1:14] \"AREA\" \"PERIMETER\" \"CNTY_\" \"CNTY_ID\" ...\n```\n\n\n:::\n:::\n\n\nIf desired, we could make a simple R base plot-style map. In this case we'll do it based on just the county areas so that the polygons are filled with a color corresponding to how large the county is.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\n# Make a graph\nplot(nc_poly[\"AREA\"], axes = T)\n```\n\n::: {.cell-output-display}\n![](mod_spatial_files/figure-html/plot-vector-1.png){fig-align='center' width=672}\n:::\n:::\n\n\n### Raster Data\n\nRaster data are stored as values in pixels. The resolution (i.e., size of the pixels) may differ among rasters but in all cases the data are stored at a per-pixel level.\n\nCommon raster data types include GeoTIFFs (.tif) and NetCDF (.nc) files.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load needed library\nlibrary(terra)\n\n# Read in raster\nnc_pixel <- terra::rast(x = file.path(\"data\", \"nc_elevation.tif\"))\n```\n:::\n\n\nOnce you've read in the raster file you can check it's structure as you would any other object but the resulting output is much less informative than for other object classes.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Check structure\nstr(nc_pixel)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\nS4 class 'SpatRaster' [package \"terra\"]\n```\n\n\n:::\n:::\n\n\nRegardless, now that we have the raster loaded we can make a simple graph to check out what sort of data is stored in it. In this case, each pixel is 3 arcseconds on each side (~0.0002° latitude/longitude) and contains the elevation (in meters) of that pixel.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\n# Make a graph\nterra::plot(nc_pixel)\n```\n\n::: {.cell-output-display}\n![](mod_spatial_files/figure-html/plot-raster-1.png){fig-align='center' width=672}\n:::\n:::\n\n\n:::\n\n## Coordinate Reference Systems\n\nA fundamental problem in spatial data is how to project data collected on a nearly spherical planet onto a two-dimensional plane. This has been solved--or at least clarified--by the use of Coordinate Reference Systems (a.k.a. \"CRS\"). All spatial data have a CRS that is explicitly identified in the data and/or the metadata because the data _are not interpretable_ without knowing which CRS is used.\n\n\nThe CRS defines the following information:\n\n1. **Datum** -- model for shape of the Earth including the starting coordinate pair and angular units that together define any particular point on the planet\n - Note that there can be global datums that work for any region of the world and local datums that only work for a particular area\n2. **Projection** -- math for the transformation to get from a round planet to a flat map\n3. **Additional parameters** -- any other information necessary to support the projection\n - E.g., the coordinates at the center of the map\n\nSome people use the analogy of peeling a citrus fruit and flattening the peel to describe the components of a CRS. The datum is the choice between a lemon or a grapefruit (i.e., the shape of the not-quite-spherical object) while the projection is the instructions for taking the complete peel and flattening it.\n\nYou can check and transform the CRS in any scripted language that allows the loading of spatial data though the specifics do differ between the types of spatial data we introduced earlier.\n\n:::{.panel-tabset}\n### Vector CRS\n\nFor vector data we can check the CRS with other functions from the `sf` library. It can be a little difficult to parse all of the information that returns but essentially it is most important that the CRS match that of any other spatial data with which we are working.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Check CRS\nsf::st_crs(x = nc_poly)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\nCoordinate Reference System:\n User input: WGS 84 \n wkt:\nGEOGCRS[\"WGS 84\",\n DATUM[\"World Geodetic System 1984\",\n ELLIPSOID[\"WGS 84\",6378137,298.257223563,\n LENGTHUNIT[\"metre\",1]]],\n PRIMEM[\"Greenwich\",0,\n ANGLEUNIT[\"degree\",0.0174532925199433]],\n CS[ellipsoidal,2],\n AXIS[\"latitude\",north,\n ORDER[1],\n ANGLEUNIT[\"degree\",0.0174532925199433]],\n AXIS[\"longitude\",east,\n ORDER[2],\n ANGLEUNIT[\"degree\",0.0174532925199433]],\n ID[\"EPSG\",4326]]\n```\n\n\n:::\n:::\n\n\nOnce you know the CRS, you can transform the data to another CRS if desired. This is a relatively fast operation for vector data because we're transforming geometric data rather than potentially millions of pixels.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Transform CRS\nnc_poly_nad83 <- sf::st_transform(x = nc_poly, crs = 3083) # <1>\n\n# Re-check CRS\nsf::st_crs(nc_poly_nad83)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\nCoordinate Reference System:\n User input: EPSG:3083 \n wkt:\nPROJCRS[\"NAD83 / Texas Centric Albers Equal Area\",\n BASEGEOGCRS[\"NAD83\",\n DATUM[\"North American Datum 1983\",\n ELLIPSOID[\"GRS 1980\",6378137,298.257222101,\n LENGTHUNIT[\"metre\",1]]],\n PRIMEM[\"Greenwich\",0,\n ANGLEUNIT[\"degree\",0.0174532925199433]],\n ID[\"EPSG\",4269]],\n CONVERSION[\"Texas Centric Albers Equal Area\",\n METHOD[\"Albers Equal Area\",\n ID[\"EPSG\",9822]],\n PARAMETER[\"Latitude of false origin\",18,\n ANGLEUNIT[\"degree\",0.0174532925199433],\n ID[\"EPSG\",8821]],\n PARAMETER[\"Longitude of false origin\",-100,\n ANGLEUNIT[\"degree\",0.0174532925199433],\n ID[\"EPSG\",8822]],\n PARAMETER[\"Latitude of 1st standard parallel\",27.5,\n ANGLEUNIT[\"degree\",0.0174532925199433],\n ID[\"EPSG\",8823]],\n PARAMETER[\"Latitude of 2nd standard parallel\",35,\n ANGLEUNIT[\"degree\",0.0174532925199433],\n ID[\"EPSG\",8824]],\n PARAMETER[\"Easting at false origin\",1500000,\n LENGTHUNIT[\"metre\",1],\n ID[\"EPSG\",8826]],\n PARAMETER[\"Northing at false origin\",6000000,\n LENGTHUNIT[\"metre\",1],\n ID[\"EPSG\",8827]]],\n CS[Cartesian,2],\n AXIS[\"easting (X)\",east,\n ORDER[1],\n LENGTHUNIT[\"metre\",1]],\n AXIS[\"northing (Y)\",north,\n ORDER[2],\n LENGTHUNIT[\"metre\",1]],\n USAGE[\n SCOPE[\"State-wide spatial data presentation requiring true area measurements.\"],\n AREA[\"United States (USA) - Texas.\"],\n BBOX[25.83,-106.66,36.5,-93.5]],\n ID[\"EPSG\",3083]]\n```\n\n\n:::\n:::\n\n1. In order to transform to a new CRS you'll need to identify the four-digit EPSG code for the desired CRS.\n\n### Raster CRS\n\nFor raster data we can check the CRS with other functions from the `terra` library. It can be a little difficult to parse all of the information that returns but essentially it is most important that the CRS match that of any other spatial data with which we are working.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Check CRS\nterra::crs(nc_pixel)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n[1] \"GEOGCRS[\\\"WGS 84\\\",\\n ENSEMBLE[\\\"World Geodetic System 1984 ensemble\\\",\\n MEMBER[\\\"World Geodetic System 1984 (Transit)\\\"],\\n MEMBER[\\\"World Geodetic System 1984 (G730)\\\"],\\n MEMBER[\\\"World Geodetic System 1984 (G873)\\\"],\\n MEMBER[\\\"World Geodetic System 1984 (G1150)\\\"],\\n MEMBER[\\\"World Geodetic System 1984 (G1674)\\\"],\\n MEMBER[\\\"World Geodetic System 1984 (G1762)\\\"],\\n MEMBER[\\\"World Geodetic System 1984 (G2139)\\\"],\\n ELLIPSOID[\\\"WGS 84\\\",6378137,298.257223563,\\n LENGTHUNIT[\\\"metre\\\",1]],\\n ENSEMBLEACCURACY[2.0]],\\n PRIMEM[\\\"Greenwich\\\",0,\\n ANGLEUNIT[\\\"degree\\\",0.0174532925199433]],\\n CS[ellipsoidal,2],\\n AXIS[\\\"geodetic latitude (Lat)\\\",north,\\n ORDER[1],\\n ANGLEUNIT[\\\"degree\\\",0.0174532925199433]],\\n AXIS[\\\"geodetic longitude (Lon)\\\",east,\\n ORDER[2],\\n ANGLEUNIT[\\\"degree\\\",0.0174532925199433]],\\n USAGE[\\n SCOPE[\\\"Horizontal component of 3D system.\\\"],\\n AREA[\\\"World.\\\"],\\n BBOX[-90,-180,90,180]],\\n ID[\\\"EPSG\\\",4326]]\"\n```\n\n\n:::\n:::\n\n\nAs with vector data, if desired you can transform the data to another CRS. However, unlike vector data, transforming the CRS of raster data is _very_ computationally intense. If at all possible you should avoid re-projecting rasters. If you must re-project, consider doing so in an environment with greater computing power than a typical laptop. Also, you should export a new raster in your preferred CRS after transforming so that you reduce the likelihood that you need to re-project again later in the lifecylce of your project.\n\nIn the interests of making this website reasonably quick to re-build, the following code chunk is not actually evaluated but is the correct syntax for this operation.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Transform CRS\nnc_pixel_nad83 <- terra::project(x = nc_pixel, y = \"epsg:3083\")\n\n# Re-check CRS\nterra::crs(nc_pixel_nad83)\n```\n:::\n\n\n:::\n\n## Making Maps\n\nNow that we've covered the main types of spatial data as well as how to check the CRS (and transform if needed) we're ready to make maps! For consistency with other modules on data visualization, we'll use `ggplot2` to make our maps but note that base R also supports map making and there are many useful tutorials elsewhere on making maps in that framework.\n\nThe `maps` package includes some useful national and US state borders so we'll begin by preparing an object that combines both sets of borders.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load library\nlibrary(maps)\n\n# Make 'borders' object\nborders <- sf::st_as_sf(maps::map(database = \"world\", plot = F, fill = T)) %>%\n dplyr::bind_rows(sf::st_as_sf(maps::map(database = \"state\", plot = F, fill = T)))\n```\n:::\n\n\nNote that the simplest way of making a map that includes a raster is to coerce that raster into a dataframe. To do this we will translate each pixel's geographic coordinates into X and Y values.\n\n\n::: {.cell}\n\n```{.r .cell-code}\nnc_pixel_df <- as.data.frame(nc_pixel, xy = T) %>% \n # Rename the 'actual' data layer more clearly\n dplyr::rename(elevation_m = SRTMGL3_NC.003_SRTMGL3_DEM_doy2000042_aid0001)\n```\n:::\n\n\nWith the borders object and our modified raster in hand, we can now make a map that includes useful context for state/nation borders. Synthesis projects often cover a larger geographic extent than primary projects so this is particularly useful in ways it might not be for primary research.\n\n\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load library\nlibrary(ggplot2)\n\n# Make map\nggplot(borders) +\n geom_sf(fill = \"gray95\") + # <1>\n coord_sf(xlim = c(-70, -90), ylim = c(30, 40), expand = F) + # <2>\n geom_tile(data = nc_pixel_df, aes(x = x, y = y, fill = elevation_m)) +\n labs(x = \"Longitude\", y = \"Latitude\")\n```\n:::\n\n1. This line is filling our nation polygons with a pale gray (helps to differentiate from ocean)\n2. Here we set the map extent so that we're only getting our region of interest\n\n

\n\"ggplot2-style\n

\n\nFrom here we can make additional `ggplot2`-style modifications as/if needed. This variant of map-making supports adding tabular data objects as well (though they would require separate geometries). Many of the LTER Network Office-funded groups that make maps include points for specific study locations along with a raster layer for an environmental / land cover characteristic that is particularly relevant to their research question and/or hypotheses.\n\n## Extracting Spatial Data\n\nBy far the most common spatial operation that LNO-funded synthesis working groups want to perform is extraction of some spatial covariate data within their region of interest. \"Extraction\" here includes (1) the actual gathering of values from the desired area, (2) summarization of those values, and (3) attaching those summarized values to an existing tabular dataset for further analysis/visualization. As with any coding task there are many ways of accomplishing this end but we'll focus on one method in the following code chunks: extraction in R via the `exactextractr` package.\n\nThis package expects that you'll want to extract raster data within a the borders described in some type of vector data. If you want the values in all the pixels of a GeoTIFF that fall inside the boundary defined by a shapefile, tools in this package will be helpful.\n\nWe'll begin by making a simpler version of our North Carolina vector data. This ensures that the extraction is as fast as possible for demonstrative purposes while still being replicable for you.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Simplify the vector data\nnc_poly_simp <- nc_poly %>% \n dplyr::filter(NAME %in% c(\"Wake\", \"Swain\")) %>% \n dplyr::select(NAME, AREA)\n\n# Check structure to demonstrate simplicity\ndplyr::glimpse(nc_poly_simp) # <1>\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\nRows: 2\nColumns: 3\n$ NAME \"Wake\", \"Swain\"\n$ AREA 0.219, 0.141\n$ geometry MULTIPOLYGON (((-78.92082 3..., MULTIPOLYGON (((-83.3317 35..…\n```\n\n\n:::\n:::\n\n1. Note that even though we used `select` to remove all but one column, the geometry information is retained!\n\nNow let's use this simplified object and extract elevation for our counties of interest (normally we'd likely do this for all counties but the process is the same).\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load needed libraries\nlibrary(exactextractr)\nlibrary(purrr)\n\n# Perform extraction\nextracted_df <- exactextractr::exact_extract(x = nc_pixel, y = nc_poly_simp, # <1>\n include_cols = c(\"NAME\", \"AREA\"), # <2>\n progress = F) %>% # <3>\n # Collapse to a dataframe\n purrr::list_rbind(x = .) # <4>\n\n# Check structure\ndplyr::glimpse(extracted_df)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\nRows: 521,671\nColumns: 4\n$ NAME \"Wake\", \"Wake\", \"Wake\", \"Wake\", \"Wake\", \"Wake\", \"Wak…\n$ AREA 0.219, 0.219, 0.219, 0.219, 0.219, 0.219, 0.219, 0.2…\n$ value 95, 97, 100, 100, 100, 101, 103, 105, 110, 111, 111,…\n$ coverage_fraction 0.027789708, 0.084839255, 0.141893625, 0.198947996, …\n```\n\n\n:::\n:::\n\n1. Note that functions like this one assume that both spatial data objects use the same CRS. We checked that earlier so we're good but remember to include that check _every_ time you do something like this!\n2. All column names specified here from the vector data (see the `y` argument) will be retained in the output. Otherwise only the extracted value and coverage fraction are included.\n3. This argument controls whether a progress bar is included. _Extremely_ useful when you have many polygons / the extraction takes a long time!\n4. The default output of this function is a list with one dataframe of extracted values per polygon in your vector data so we'll unlist to a dataframe for ease of future operations\n\nIn the above output we can see that it has extracted the elevation of _every pixel_ within each of our counties of interest and provided us with the percentage of that pixel that is covered by the polygon (i.e., by the shapefile). We can now summarize this however we'd like and--eventually--join it back onto the county data via the column(s) we specified should be retained from the original vector data.\n\n## Additional Resources\n\n### Papers & Documents\n\n- \n\n### Workshops & Courses\n\n- The Carpentries' [Introduction to Geospatial Raster & Vector Data with R](https://datacarpentry.org/r-raster-vector-geospatial/)\n- The Carpentries' [Introduction to R for Geospatial Data](https://datacarpentry.org/r-intro-geospatial/index.html)\n- Arctic Data Center's [Spatial and Image Data Using GeoPandas](https://learning.nceas.ucsb.edu/2023-03-arctic/sections/geopandas.html) chapter of their Scalable Computing course\n- Jason Flower's (UC Santa Barbara) [Introduction to rasters with `terra`](https://jflowernet.github.io/intro-terra-ecodatascience/)\n- King, R. [Spatial Data Visualization](https://github.com/king0708/spatial-data-viz) workshop\n\n### Websites\n\n- NASA's Application for Extracting and Exploring Analysis Ready Samples [(AppEEARS) Portal](https://appeears.earthdatacloud.nasa.gov/)\n", + "markdown": "---\ntitle: \"Working with Spatial Data\"\ncode-annotations: hover\n---\n\n\n\n\n## Overview\n\nSynthesis projects often have need of spatial datasets. At its simplest, it can be helpful to have a map of the original project locations including in the synthesis dataset. In more complex instances you want to extract spatial data within a certain area of sampling locations. Regardless of 'why' you're using spatial data, it may come up during your primary or synthesis work and thus deserves consideration in this course's materials. There are _many_ modes of working with spatial data, and not all of these tools require coding literacy but for consistency with the rest of the modules **this module will focus on _scripted_ approaches to interacting with spatial data**.\n\n## Learning Objectives\n\nAfter completing this topic you will be able to: \n\n- Define the difference between the two major types of spatial data\n- Manipulate spatial data with R\n- Create maps using spatial data\n- Integrate spatial data with tabular data\n\n## Needed Packages\n\nIf you'd like to follow along with the code chunks included throughout this module, you'll need to install the following packages:\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Note that these lines only need to be run once per computer\n## So you can skip this step if you've installed these before\ninstall.packages(\"tidyverse\")\ninstall.packages(\"sf\")\ninstall.packages(\"terra\")\ninstall.packages(\"maps\")\ninstall.packages(\"exactextractr\")\n```\n:::\n\n\n## Types of Spatial Data\n\nThere are two main types of spatial data: vector and raster. Both types (and the packages they require) are described in the tabs below.\n\n:::{.panel-tabset}\n### Vector Data\n\nVector data are stored as polygons. Essentially vector data are a set of points and--sometimes--the lines between them that define the edges of a shape. They may store additional data that is retained in a semi-tabular format that relates to the polygon(s) but isn't directly stored in them.\n\nCommon vector data types include shape files or GeoJSONs.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load needed library\nlibrary(sf)\n\n# Read in shapefile\nnc_poly <- sf::st_read(dsn = file.path(\"data\", \"nc_borders.shp\")) #<1>\n```\n:::\n\n1. Note that even though we're only specifying the \".shp\" file in this function you _must_ also have the associated files in that same folder. In this case that includes a \".dbf\", \".prj\", and \".shx\", though in other contexts you may have others.\n\nOnce you have read in the shapefile, you can check its structure as you would any other data object. Note that the object has both the 'data.frame' class and the 'sf' (\"simple features\") class. In this case, the shapefile relates to counties in North Carolina and some associated demographic data in those counties.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Check structure\nstr(nc_poly)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\nClasses 'sf' and 'data.frame':\t100 obs. of 15 variables:\n $ AREA : num 0.114 0.061 0.143 0.07 0.153 0.097 0.062 0.091 0.118 0.124 ...\n $ PERIMETER: num 1.44 1.23 1.63 2.97 2.21 ...\n $ CNTY_ : num 1825 1827 1828 1831 1832 ...\n $ CNTY_ID : num 1825 1827 1828 1831 1832 ...\n $ NAME : chr \"Ashe\" \"Alleghany\" \"Surry\" \"Currituck\" ...\n $ FIPS : chr \"37009\" \"37005\" \"37171\" \"37053\" ...\n $ FIPSNO : num 37009 37005 37171 37053 37131 ...\n $ CRESS_ID : int 5 3 86 27 66 46 15 37 93 85 ...\n $ BIR74 : num 1091 487 3188 508 1421 ...\n $ SID74 : num 1 0 5 1 9 7 0 0 4 1 ...\n $ NWBIR74 : num 10 10 208 123 1066 ...\n $ BIR79 : num 1364 542 3616 830 1606 ...\n $ SID79 : num 0 3 6 2 3 5 2 2 2 5 ...\n $ NWBIR79 : num 19 12 260 145 1197 ...\n $ geometry :sfc_MULTIPOLYGON of length 100; first list element: List of 1\n ..$ :List of 1\n .. ..$ : num [1:27, 1:2] -81.5 -81.5 -81.6 -81.6 -81.7 ...\n ..- attr(*, \"class\")= chr [1:3] \"XY\" \"MULTIPOLYGON\" \"sfg\"\n - attr(*, \"sf_column\")= chr \"geometry\"\n - attr(*, \"agr\")= Factor w/ 3 levels \"constant\",\"aggregate\",..: NA NA NA NA NA NA NA NA NA NA ...\n ..- attr(*, \"names\")= chr [1:14] \"AREA\" \"PERIMETER\" \"CNTY_\" \"CNTY_ID\" ...\n```\n\n\n:::\n:::\n\n\nIf desired, we could make a simple R base plot-style map. In this case we'll do it based on just the county areas so that the polygons are filled with a color corresponding to how large the county is.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\n# Make a graph\nplot(nc_poly[\"AREA\"], axes = T)\n```\n\n::: {.cell-output-display}\n![](mod_spatial_files/figure-html/plot-vector-1.png){fig-align='center' width=672}\n:::\n:::\n\n\n### Raster Data\n\nRaster data are stored as values in pixels. The resolution (i.e., size of the pixels) may differ among rasters but in all cases the data are stored at a per-pixel level.\n\nCommon raster data types include GeoTIFFs (.tif) and NetCDF (.nc) files.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load needed library\nlibrary(terra)\n\n# Read in raster\nnc_pixel <- terra::rast(x = file.path(\"data\", \"nc_elevation.tif\"))\n```\n:::\n\n\nOnce you've read in the raster file you can check it's structure as you would any other object but the resulting output is much less informative than for other object classes.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Check structure\nstr(nc_pixel)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\nS4 class 'SpatRaster' [package \"terra\"]\n```\n\n\n:::\n:::\n\n\nRegardless, now that we have the raster loaded we can make a simple graph to check out what sort of data is stored in it. In this case, each pixel is 3 arcseconds on each side (~0.0002° latitude/longitude) and contains the elevation (in meters) of that pixel.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\n# Make a graph\nterra::plot(nc_pixel)\n```\n\n::: {.cell-output-display}\n![](mod_spatial_files/figure-html/plot-raster-1.png){fig-align='center' width=672}\n:::\n:::\n\n\n:::\n\n## Coordinate Reference Systems\n\nA fundamental problem in spatial data is how to project data collected on a nearly spherical planet onto a two-dimensional plane. This has been solved--or at least clarified--by the use of Coordinate Reference Systems (a.k.a. \"CRS\"). All spatial data have a CRS that is explicitly identified in the data and/or the metadata because the data _are not interpretable_ without knowing which CRS is used.\n\n\nThe CRS defines the following information:\n\n1. **Datum** -- model for shape of the Earth including the starting coordinate pair and angular units that together define any particular point on the planet\n - Note that there can be global datums that work for any region of the world and local datums that only work for a particular area\n2. **Projection** -- math for the transformation to get from a round planet to a flat map\n3. **Additional parameters** -- any other information necessary to support the projection\n - E.g., the coordinates at the center of the map\n\nSome people use the analogy of peeling a citrus fruit and flattening the peel to describe the components of a CRS. The datum is the choice between a lemon or a grapefruit (i.e., the shape of the not-quite-spherical object) while the projection is the instructions for taking the complete peel and flattening it.\n\nYou can check and transform the CRS in any scripted language that allows the loading of spatial data though the specifics do differ between the types of spatial data we introduced earlier.\n\n:::{.panel-tabset}\n### Vector CRS\n\nFor vector data we can check the CRS with other functions from the `sf` library. It can be a little difficult to parse all of the information that returns but essentially it is most important that the CRS match that of any other spatial data with which we are working.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Check CRS\nsf::st_crs(x = nc_poly)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\nCoordinate Reference System:\n User input: WGS 84 \n wkt:\nGEOGCRS[\"WGS 84\",\n DATUM[\"World Geodetic System 1984\",\n ELLIPSOID[\"WGS 84\",6378137,298.257223563,\n LENGTHUNIT[\"metre\",1]]],\n PRIMEM[\"Greenwich\",0,\n ANGLEUNIT[\"degree\",0.0174532925199433]],\n CS[ellipsoidal,2],\n AXIS[\"latitude\",north,\n ORDER[1],\n ANGLEUNIT[\"degree\",0.0174532925199433]],\n AXIS[\"longitude\",east,\n ORDER[2],\n ANGLEUNIT[\"degree\",0.0174532925199433]],\n ID[\"EPSG\",4326]]\n```\n\n\n:::\n:::\n\n\nOnce you know the CRS, you can transform the data to another CRS if desired. This is a relatively fast operation for vector data because we're transforming geometric data rather than potentially millions of pixels.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Transform CRS\nnc_poly_nad83 <- sf::st_transform(x = nc_poly, crs = 3083) # <1>\n\n# Re-check CRS\nsf::st_crs(nc_poly_nad83)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\nCoordinate Reference System:\n User input: EPSG:3083 \n wkt:\nPROJCRS[\"NAD83 / Texas Centric Albers Equal Area\",\n BASEGEOGCRS[\"NAD83\",\n DATUM[\"North American Datum 1983\",\n ELLIPSOID[\"GRS 1980\",6378137,298.257222101,\n LENGTHUNIT[\"metre\",1]]],\n PRIMEM[\"Greenwich\",0,\n ANGLEUNIT[\"degree\",0.0174532925199433]],\n ID[\"EPSG\",4269]],\n CONVERSION[\"Texas Centric Albers Equal Area\",\n METHOD[\"Albers Equal Area\",\n ID[\"EPSG\",9822]],\n PARAMETER[\"Latitude of false origin\",18,\n ANGLEUNIT[\"degree\",0.0174532925199433],\n ID[\"EPSG\",8821]],\n PARAMETER[\"Longitude of false origin\",-100,\n ANGLEUNIT[\"degree\",0.0174532925199433],\n ID[\"EPSG\",8822]],\n PARAMETER[\"Latitude of 1st standard parallel\",27.5,\n ANGLEUNIT[\"degree\",0.0174532925199433],\n ID[\"EPSG\",8823]],\n PARAMETER[\"Latitude of 2nd standard parallel\",35,\n ANGLEUNIT[\"degree\",0.0174532925199433],\n ID[\"EPSG\",8824]],\n PARAMETER[\"Easting at false origin\",1500000,\n LENGTHUNIT[\"metre\",1],\n ID[\"EPSG\",8826]],\n PARAMETER[\"Northing at false origin\",6000000,\n LENGTHUNIT[\"metre\",1],\n ID[\"EPSG\",8827]]],\n CS[Cartesian,2],\n AXIS[\"easting (X)\",east,\n ORDER[1],\n LENGTHUNIT[\"metre\",1]],\n AXIS[\"northing (Y)\",north,\n ORDER[2],\n LENGTHUNIT[\"metre\",1]],\n USAGE[\n SCOPE[\"State-wide spatial data presentation requiring true area measurements.\"],\n AREA[\"United States (USA) - Texas.\"],\n BBOX[25.83,-106.66,36.5,-93.5]],\n ID[\"EPSG\",3083]]\n```\n\n\n:::\n:::\n\n1. In order to transform to a new CRS you'll need to identify the four-digit EPSG code for the desired CRS.\n\n### Raster CRS\n\nFor raster data we can check the CRS with other functions from the `terra` library. It can be a little difficult to parse all of the information that returns but essentially it is most important that the CRS match that of any other spatial data with which we are working.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Check CRS\nterra::crs(nc_pixel)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n[1] \"GEOGCRS[\\\"WGS 84\\\",\\n ENSEMBLE[\\\"World Geodetic System 1984 ensemble\\\",\\n MEMBER[\\\"World Geodetic System 1984 (Transit)\\\"],\\n MEMBER[\\\"World Geodetic System 1984 (G730)\\\"],\\n MEMBER[\\\"World Geodetic System 1984 (G873)\\\"],\\n MEMBER[\\\"World Geodetic System 1984 (G1150)\\\"],\\n MEMBER[\\\"World Geodetic System 1984 (G1674)\\\"],\\n MEMBER[\\\"World Geodetic System 1984 (G1762)\\\"],\\n MEMBER[\\\"World Geodetic System 1984 (G2139)\\\"],\\n ELLIPSOID[\\\"WGS 84\\\",6378137,298.257223563,\\n LENGTHUNIT[\\\"metre\\\",1]],\\n ENSEMBLEACCURACY[2.0]],\\n PRIMEM[\\\"Greenwich\\\",0,\\n ANGLEUNIT[\\\"degree\\\",0.0174532925199433]],\\n CS[ellipsoidal,2],\\n AXIS[\\\"geodetic latitude (Lat)\\\",north,\\n ORDER[1],\\n ANGLEUNIT[\\\"degree\\\",0.0174532925199433]],\\n AXIS[\\\"geodetic longitude (Lon)\\\",east,\\n ORDER[2],\\n ANGLEUNIT[\\\"degree\\\",0.0174532925199433]],\\n USAGE[\\n SCOPE[\\\"Horizontal component of 3D system.\\\"],\\n AREA[\\\"World.\\\"],\\n BBOX[-90,-180,90,180]],\\n ID[\\\"EPSG\\\",4326]]\"\n```\n\n\n:::\n:::\n\n\nAs with vector data, if desired you can transform the data to another CRS. However, unlike vector data, transforming the CRS of raster data is _very_ computationally intense. If at all possible you should avoid re-projecting rasters. If you must re-project, consider doing so in an environment with greater computing power than a typical laptop. Also, you should export a new raster in your preferred CRS after transforming so that you reduce the likelihood that you need to re-project again later in the lifecylce of your project.\n\nIn the interests of making this website reasonably quick to re-build, the following code chunk is not actually evaluated but is the correct syntax for this operation.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Transform CRS\nnc_pixel_nad83 <- terra::project(x = nc_pixel, y = \"epsg:3083\")\n\n# Re-check CRS\nterra::crs(nc_pixel_nad83)\n```\n:::\n\n\n:::\n\n## Making Maps\n\nNow that we've covered the main types of spatial data as well as how to check the CRS (and transform if needed) we're ready to make maps! For consistency with other modules on data visualization, we'll use `ggplot2` to make our maps but note that base R also supports map making and there are many useful tutorials elsewhere on making maps in that framework.\n\nThe `maps` package includes some useful national and US state borders so we'll begin by preparing an object that combines both sets of borders.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load library\nlibrary(maps)\n\n# Make 'borders' object\nborders <- sf::st_as_sf(maps::map(database = \"world\", plot = F, fill = T)) %>%\n dplyr::bind_rows(sf::st_as_sf(maps::map(database = \"state\", plot = F, fill = T)))\n```\n:::\n\n\nNote that the simplest way of making a map that includes a raster is to coerce that raster into a dataframe. To do this we will translate each pixel's geographic coordinates into X and Y values.\n\n\n::: {.cell}\n\n```{.r .cell-code}\nnc_pixel_df <- as.data.frame(nc_pixel, xy = T) %>% \n # Rename the 'actual' data layer more clearly\n dplyr::rename(elevation_m = SRTMGL3_NC.003_SRTMGL3_DEM_doy2000042_aid0001)\n```\n:::\n\n\nWith the borders object and our modified raster in hand, we can now make a map that includes useful context for state/nation borders. Synthesis projects often cover a larger geographic extent than primary projects so this is particularly useful in ways it might not be for primary research.\n\n\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load library\nlibrary(ggplot2)\n\n# Make map\nggplot(borders) +\n geom_sf(fill = \"gray95\") + # <1>\n coord_sf(xlim = c(-70, -90), ylim = c(30, 40), expand = F) + # <2>\n geom_tile(data = nc_pixel_df, aes(x = x, y = y, fill = elevation_m)) +\n labs(x = \"Longitude\", y = \"Latitude\")\n```\n:::\n\n1. This line is filling our nation polygons with a pale gray (helps to differentiate from ocean)\n2. Here we set the map extent so that we're only getting our region of interest\n\n

\n\"ggplot2-style\n

\n\nFrom here we can make additional `ggplot2`-style modifications as/if needed. This variant of map-making supports adding tabular data objects as well (though they would require separate geometries). Many of the LTER Network Office-funded groups that make maps include points for specific study locations along with a raster layer for an environmental / land cover characteristic that is particularly relevant to their research question and/or hypotheses.\n\n## Extracting Spatial Data\n\nBy far the most common spatial operation that LNO-funded synthesis working groups want to perform is extraction of some spatial covariate data within their region of interest. \"Extraction\" here includes (1) the actual gathering of values from the desired area, (2) summarization of those values, and (3) attaching those summarized values to an existing tabular dataset for further analysis/visualization. As with any coding task there are many ways of accomplishing this end but we'll focus on one method in the following code chunks: extraction in R via the `exactextractr` package.\n\nThis package expects that you'll want to extract raster data within a the borders described in some type of vector data. If you want the values in all the pixels of a GeoTIFF that fall inside the boundary defined by a shapefile, tools in this package will be helpful.\n\nWe'll begin by making a simpler version of our North Carolina vector data. This ensures that the extraction is as fast as possible for demonstrative purposes while still being replicable for you.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Simplify the vector data\nnc_poly_simp <- nc_poly %>% \n dplyr::filter(NAME %in% c(\"Wake\", \"Swain\")) %>% \n dplyr::select(NAME, AREA)\n\n# Check structure to demonstrate simplicity\ndplyr::glimpse(nc_poly_simp) # <1>\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\nRows: 2\nColumns: 3\n$ NAME \"Wake\", \"Swain\"\n$ AREA 0.219, 0.141\n$ geometry MULTIPOLYGON (((-78.92082 3..., MULTIPOLYGON (((-83.3317 35..…\n```\n\n\n:::\n:::\n\n1. Note that even though we used `select` to remove all but one column, the geometry information is retained!\n\nNow let's use this simplified object and extract elevation for our counties of interest (normally we'd likely do this for all counties but the process is the same).\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load needed libraries\nlibrary(exactextractr)\nlibrary(purrr)\n\n# Perform extraction\nextracted_df <- exactextractr::exact_extract(x = nc_pixel, y = nc_poly_simp, # <1>\n include_cols = c(\"NAME\", \"AREA\"), # <2>\n progress = F) %>% # <3>\n # Collapse to a dataframe\n purrr::list_rbind(x = .) # <4>\n\n# Check structure\ndplyr::glimpse(extracted_df)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\nRows: 521,671\nColumns: 4\n$ NAME \"Wake\", \"Wake\", \"Wake\", \"Wake\", \"Wake\", \"Wake\", \"Wak…\n$ AREA 0.219, 0.219, 0.219, 0.219, 0.219, 0.219, 0.219, 0.2…\n$ value 95, 97, 100, 100, 100, 101, 103, 105, 110, 111, 111,…\n$ coverage_fraction 0.027789708, 0.084839255, 0.141893625, 0.198947996, …\n```\n\n\n:::\n:::\n\n1. Note that functions like this one assume that both spatial data objects use the same CRS. We checked that earlier so we're good but remember to include that check _every_ time you do something like this!\n2. All column names specified here from the vector data (see the `y` argument) will be retained in the output. Otherwise only the extracted value and coverage fraction are included.\n3. This argument controls whether a progress bar is included. _Extremely_ useful when you have many polygons / the extraction takes a long time!\n4. The default output of this function is a list with one dataframe of extracted values per polygon in your vector data so we'll unlist to a dataframe for ease of future operations\n\nIn the above output we can see that it has extracted the elevation of _every pixel_ within each of our counties of interest and provided us with the percentage of that pixel that is covered by the polygon (i.e., by the shapefile). We can now summarize this however we'd like and--eventually--join it back onto the county data via the column(s) we specified should be retained from the original vector data.\n\n## Additional Resources\n\n### Papers & Documents\n\n- \n\n### Workshops & Courses\n\n- The Carpentries. [Introduction to Geospatial Raster and Vector Data with R](https://datacarpentry.org/r-raster-vector-geospatial/). **2024**.\n- The Carpentries. [Introduction to R for Geospatial Data](https://datacarpentry.org/r-intro-geospatial/index.html). **2024**.\n- King, R. [Spatial Data Visualization](https://github.com/king0708/spatial-data-viz). **2024**.\n- Flower, J. [Introduction to Rasters with `terra`](https://jflowernet.github.io/intro-terra-ecodatascience/). **2024**.\n- Clark, S.J., _et al._ [Spatial and Image Data Using GeoPandas](https://learning.nceas.ucsb.edu/2023-03-arctic/sections/geopandas.html). **2023**.\n\n### Websites\n\n- NASA. [AppEEARS Portal](https://appeears.earthdatacloud.nasa.gov/)\n", "supporting": [ "mod_spatial_files" ], diff --git a/_freeze/mod_stats/execute-results/html.json b/_freeze/mod_stats/execute-results/html.json index 8534d16..4260157 100644 --- a/_freeze/mod_stats/execute-results/html.json +++ b/_freeze/mod_stats/execute-results/html.json @@ -1,8 +1,8 @@ { - "hash": "228bb41b3c0921de5ef7686cf2078f0a", + "hash": "7653e191b1d2f4ada47f14612a2d3522", "result": { "engine": "knitr", - "markdown": "---\ntitle: \"Analysis & Modeling\"\ncode-annotations: hover\n---\n\n\n## Overview\n\nGiven the wide range in statistical training in graduate curricula (and corresponding breadth of experience among early career researchers), we'll be approaching this module by splitting it into two halves.\n\n1. First half: a \"flipped approach\" where project teams will share their proposed analyses with one another\n2. Second half: typical instructional module dedicated to **analyses that are more common in--or exclusive to--synthesis research**. \n\nContent produced by project teams during the flipped half may be linked in the '[Additional Resources](https://lter.github.io/ssecr/mod_stats.html#additional-resources)' section at the bottom of this module at the discretion of each team. Otherwise the content of this module will focus only on the non-flipped content.\n\n## Learning Objectives\n\nAfter completing this module you will be able to: \n\n- Describe proposed analytical methods to an interested audience of mixed prior experience\n- Explain nuance in interpretation of results of proposed analyses\n- Identify some statistical tests common in synthesis research\n- Perform some synthesis-specific analyses\n\n## Needed Packages\n\nIf you'd like to follow along with the code chunks included throughout this module, you'll need to install the following packages:\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Note that these lines only need to be run once per computer\n## So you can skip this step if you've installed these before\ninstall.packages(\"tidyverse\")\ninstall.packages(\"lmerTest\")\ninstall.packages(\"palmerpenguins\")\ninstall.packages(\"esc\")\n```\n:::\n\n\nWe'll go ahead and load some of these libraries as well to be able to better demonstrate these concepts.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load needed libraries\nlibrary(tidyverse)\nlibrary(lmerTest)\nlibrary(palmerpenguins)\n```\n:::\n\n\n## Hypothesis Framing Aside\n\nBefore we dive in, we should discuss two of the ways in which you can frame your hypothesis and the differences in interpretation and appropriate statistical tool(s) that follow from that choice. We'll restrict our conversation here to **two alternate modes of thinking about your hypothesis: frequentist statistics versus multi-model inference.**\n\nNote that this is something of a false dichotomy as tools from both worlds can be/are frequently used to complement one another. However, many graduate students are trained by instructors with strong feelings about one method _in opposition to_ the other so it is worthwhile to consider these two paths separately even if you wind up using components of both in your own work.\n\n::::{.panel-tabset}\n### Frequentist Inference\n\nHypotheses here are a question of whether a variable has a \"significant\" effect on another. \"Significant\" has a very precise meaning in this context that has to do with _p_-values. Fundamentally, these methods focus on whether the observed relationship in the data is likely to be observed by chance alone or not. Strong effects are less likely--though not impossible--to be observed due to random chance.\n\nIf your hypothesis can be summarized as something along the lines of 'we hypothesize that X affects Y' then frequentist inference may be a more appropriate methodology.\n\nFor the purposes of SSECR, **our discussion of frequentist inference will focus on mixed-effect models**.\n\n### Multi-Model Inference\n\nHyoptheses here are a question of which variables explain the _most_ variation in the data. Methods in this framing are unconcerned--or at least less concerned than in frequentist inference--with the probability associated with a particular variable. Intead, these methods focus on which of a set of user-defined candidate models explains most of the noise in the data _even when that best model does not necessarily explain much of that variation in absolute terms_.\n\nIf your hypothesis can be summarized as something along the lines of 'we hypothesize that models including X explain more of the variation in Y than those that do not' then multi-model inference may be a more appropriate methodology.\n\nFor the purposes of SSECR, **our discussion of multi-model inference will focus on comparing model strengths with AIC**.\n\n::::\n\n## Mixed-Effects Models\n\nIn any statistical test there is at least one response variable (a.k.a. \"dependent\" variable) and some number of explanatory variables (a.k.a. \"independent\" variables). However, in biology our experiments often involve repeated sampling over time or at the same locations. These variables (time or site) are neither response nor explanatory variables but we might reasonably conclude that they affect our response and/or explanatory variables.\n\nIn essence we want to use a statistical tool that asks 'what is the effect of the explanatory variable(s) on the response _when the variation due to these non-variable considerations is accounted for_?' Such tests are called **mixed-effects models**. This name derives from considering explanatory variables \"fixed effects\" and non-explanatory/response variables as \"random effects\". Including both fixed and random effects thus creates a model with \"mixed effects.\"\n\n### Types of Random Effect\n\nThere are a few types of random effects but we can limit our conversation here to just two: random intercepts and random slopes.\n\n:::{.panel-tabset}\n#### Random Intercepts\n\nRandom intercepts should be used when you expect that the average response differs among levels of that variable but not in a way that changes the relationship between each level of this variable and the other variables (either fixed or random). In statistical terms you want to allow the intercept to change with levels of this variable.\n\nFor example, let's imagine that we are studying the effect of different organic farming practices on beneficial insect populations. We build relationships with several organic farmers willing to let us conduct this research on their properties and sample the insect communities at each farm over the course of a summer. However, we know that each farm is surrounded by a different habitat type that affects the composition of the local insect community. It is reasonable to expect that even farms where 'the same' management method is used are likely to differ because of this difference in landscape context.\n\nIn cases like this, we don't want to include a term for 'site' as a fixed effect but we do want to account for those differences so that our assessment of the significance of our explanatory variables isn't limited by the variation due to site.\n\n#### Random Slopes\n\nRandom slopes should be used when you expect that the average response differs among levels of that variable in a way that does change with other variables.\n\nFor example, let's imagine that we are studying the effect of temperature on avian malaria rates in songbirds. We identify several sites--along a gradient of daytime temperature ranges--where our species of interest can be found, capture them, and measure malaria infection rates. However, unbeknownst to us at the start of our study, our study sites have varying populations of dragonflies which affects local mosquito populations and malaria transmission/infection rates. If we revisit our sites repeatedly for several years is is reasonable to expect that this difference among sites likely affects the _relationship between_ daytime temperatures and songbird malaria rates.\n\nBy including site as a random slope in this context, we can account for this effect and still analyze our explanatory variables of interest. Note that random slopes are _very_ \"data hungry\" so you may not be able to use them without very high replication in your study design.\n:::\n\n### Nested Random Effects\n\nTo further complicate matters, we can use nested random effects as well. These can be either random intercepts or random slopes though they are more commonly seen with random intercepts. A nested random effect accounts for the effect of one random variable that _is itself affected by another variable!_ A classic example of this is when a study design uses two (or more) levels of spatial nestedness in their experimentall design.\n\nFor instance, let's imagine we were conducting a global study of marine plankton biodiversity. To gether these data we took several cruises (scientific not--exclusively--pleasure) at different places around the world and during each cruise we followed a set of transects. In each transect we did several plankton tows and quantified the diversity of each tow. We can reasonably assume the following:\n\n1. Each cruise differs from each other cruise (due to any number of climatic/ecological factors)\n - But cruises within the same part of the world are still likely to have similar planktonic communities\n2. Within each cruise, each transect differs from the others (again, due to unpreventable factors)\n - But transects within the same cruise are still likely to be more similar to one another than to transects in different cruises (even other ones in the same region!)\n3. Within each transect, each plankton tow differs from one another!\n - But again, more similar to other tows in the same transect than other tows in different transects/cruises\n\nIf we put these assumptions together we realize we want to account for the variation of cruise, transect, and tow while still retaining the nestedness of the similarity among samples. A nested random effect where transect is nested inside of cruise and tow is nested inside of transect would capture this effectively!\n\n### Philosophical Note: Random vs. Fixed\n\nDeciding whether a given variable should be a fixed or random effect can be tough. You'll likely need to rely on your scientific intuition about which feels more appropriate and then be prepared to defend that decision to your committee and/or \"reviewer \\#2\". It may prove helpful though to consider whether you 'care' about the effect of that variable. \n\nIf your hypothesis includes that variable than it should likely be a fixed effect. If the variable is just a facet of your experimental design but isn't something you're necessarily interested in testing, then it should likely be a random effect. And, once you've made your decision, it is totally okay to change your mind and tweak the structure of your model!\n\n:::{.callout-warning icon=\"false\"}\n#### Discussion: Random versus Fixed Effects\n\nWith a small group, decide whether you think the terms in the examples below should be fixed effects or random effects:\n\n- You study small mammal populations in urban settings\n - Should 'proportion green space' be a fixed effect or a random effect?\n- You are part of a team studying leopard seal feeding behavior\n - What type of effect should 'observer' be?\n- You study the gut microbiota of a particular beetle species\n - Should 'beetle sex' be a fixed or a random effect?\n - What about beetle life stage (e.g., larva versus adult)?\n - What about the region of the gut from which the samples were taken?\n- You study vascular plant chemical defenses against herbivory\n - Should phylogeny (i.e., evolutionary relatedness) be a fixed or random effect?\n - What about feeding guild of the herbivore?\n:::\n\n### Mixed-Effects Case Study\n\nLet's imagine we are researching tarantula populations for several years in the Chihuahuan Desert. Our hypothesis is that the number of tarantulas will be greater in sites further from the nearest road. We select ten study sites of varying distances from the nearest road and intensively count our furry friends at three plots within each site for several months. We return to our sites--and their associated plots--and repeat this process each year for three years. In the second year we have help from a new member of our lab but in the third year we're back to working alone (they had their own project to handle by then). We enter our data and perform careful quality control to get it into a tidy format ready for analyis.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Read in data\ntarantula_df <- read.csv(file = file.path(\"data\", \"tarantulas.csv\"))\n\n# Check structure\nstr(tarantula_df)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n'data.frame':\t450 obs. of 6 variables:\n $ year : int 2022 2022 2022 2022 2022 2022 2022 2022 2022 2022 ...\n $ road.dist_km : num 64.5 64.5 64.5 64.5 64.5 ...\n $ site : chr \"site_A\" \"site_A\" \"site_A\" \"site_A\" ...\n $ plot : chr \"plot_a\" \"plot_a\" \"plot_a\" \"plot_a\" ...\n $ site.plot : chr \"A_a\" \"A_a\" \"A_a\" \"A_a\" ...\n $ tarantula_count: int 199 220 213 206 220 128 156 121 121 142 ...\n```\n\n\n:::\n:::\n\n\nWith our data in hand, we now want to run some statistical tests and--hopefully--get some endorphine-inducingly small _p_-values. If we choose to simply ignore our possible random effects, we could fit a linear regression.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Fit model\ntarantula_lm <- lm(tarantula_count ~ road.dist_km, data = tarantula_df) # <1>\n\n# Extract summary\nsummary(tarantula_lm)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n\nCall:\nlm(formula = tarantula_count ~ road.dist_km, data = tarantula_df)\n\nResiduals:\n Min 1Q Median 3Q Max \n-216.916 -77.487 8.486 59.913 316.084 \n\nCoefficients:\n Estimate Std. Error t value Pr(>|t|) \n(Intercept) -3.3935 14.8929 -0.228 0.82 \nroad.dist_km 2.8140 0.2775 10.141 <2e-16 ***\n---\nSignif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1\n\nResidual standard error: 100.2 on 448 degrees of freedom\nMultiple R-squared: 0.1867,\tAdjusted R-squared: 0.1849 \nF-statistic: 102.8 on 1 and 448 DF, p-value: < 2.2e-16\n```\n\n\n:::\n:::\n\n1. R syntax for statistical tests is `response ~ explanatory` a.k.a. `Y ~ X`\n\nThis naive test seems to support our hypothesis. However, sampling effort differed between the three study years. Not only was there a second person in the second year but we can also reasonably expect that by the third year in this system we had greatly improved our tarantula-finding skills. So, a random effect of year is definitely justified. We are not concerned that the different study years will affect the relationship between tarantula populations and road distance though so a random intercept is fine.\n\nThere could be an argument for including year as a fixed effect in its own right but some preliminary investigations reveal no significant climatic differences across the region we worked in those three years. So, while we think that years may differ from one another, that difference is not something we care to analyze.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Fit the new model\ntarantula_mem1 <- lmerTest::lmer(tarantula_count ~ road.dist_km + \n (1|year), # <1>\n data = tarantula_df)\n\n# Extract summary\nsummary(tarantula_mem1)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\nLinear mixed model fit by REML. t-tests use Satterthwaite's method [\nlmerModLmerTest]\nFormula: tarantula_count ~ road.dist_km + (1 | year)\n Data: tarantula_df\n\nREML criterion at convergence: 5362.6\n\nScaled residuals: \n Min 1Q Median 3Q Max \n-2.6578 -0.7941 0.1653 0.6517 2.9451 \n\nRandom effects:\n Groups Name Variance Std.Dev.\n year (Intercept) 1940 44.04 \n Residual 8747 93.52 \nNumber of obs: 450, groups: year, 3\n\nFixed effects:\n Estimate Std. Error df t value Pr(>|t|) \n(Intercept) -3.3935 28.9768 3.1783 -0.117 0.914 \nroad.dist_km 2.8140 0.2589 446.0000 10.868 <2e-16 ***\n---\nSignif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1\n\nCorrelation of Fixed Effects:\n (Intr)\nroad.dst_km -0.455\n```\n\n\n:::\n:::\n\n1. This is the syntax for specifying a random intercept (random slope variables should be before the `|` where `1` goes for a random intercept)\n\nBy including that random effect we actually get a slightly stronger effect of road distance (T value of ~12 without versus ~13 with). This is because our new random effect accounts for some of the 'noise' between study years. That actually gives us a better picture of the relationship between our response and explanatory variables.\n\nNow that we're already using a mixed-effects model, we have little excuse not to account for the other potential random effect: plot! Remember that there were three plots within each site and from our extensive time in the field we have developed a strong intuition that there might be substantial among-plot variation at each site. We can make a quick exploratory graph to facilitate an 'eyeball test' of whether the data show what our intuition suggest.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\nggplot(tarantula_df, aes(y = tarantula_count, x = plot, fill = plot)) +\n geom_violin(alpha = 0.5) + # <1>\n geom_jitter(width = 0.25, size = 0.5) +\n facet_wrap(site ~ .) +\n theme_bw() +\n theme(axis.text.x = element_text(angle = 35, hjust = 1)) # <2>\n```\n\n::: {.cell-output-display}\n![](mod_stats_files/figure-html/mem-explore-graph-1.png){fig-align='center' width=672}\n:::\n:::\n\n1. Violin plots are a nice alternative to boxplots because they allow visualizing data distributions directly rather than requiring an intutive grasp of the distribution metrics described by each bit of a boxplot\n2. This is allowing us to 'tilt' the X axis tick labels so they don't overlap with one another\n\nThis graph clearly supports our intuition that among-plot variation is dramatic! We _could_ account for this by including plot as a fixed effect but we'll need to sacrifice a lot of degrees of freedom (can be thought of as \"statistical power\") for a variable that we don't actually care about. Instead, we could include plot as another random effect.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Fit the new model\ntarantula_mem2 <- lmerTest::lmer(tarantula_count ~ road.dist_km + \n (1|year) + (1|site.plot), # <1>\n data = tarantula_df)\n\n# Extract summary\nsummary(tarantula_mem2)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\nLinear mixed model fit by REML. t-tests use Satterthwaite's method [\nlmerModLmerTest]\nFormula: tarantula_count ~ road.dist_km + (1 | year) + (1 | site.plot)\n Data: tarantula_df\n\nREML criterion at convergence: 4553.6\n\nScaled residuals: \n Min 1Q Median 3Q Max \n-3.0902 -0.5352 -0.0117 0.5990 3.9323 \n\nRandom effects:\n Groups Name Variance Std.Dev.\n site.plot (Intercept) 8165 90.36 \n year (Intercept) 1991 44.62 \n Residual 1058 32.52 \nNumber of obs: 450, groups: site.plot, 30; year, 3\n\nFixed effects:\n Estimate Std. Error df t value Pr(>|t|) \n(Intercept) -3.393 58.233 23.593 -0.058 0.95402 \nroad.dist_km 2.814 0.973 28.000 2.892 0.00733 **\n---\nSignif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1\n\nCorrelation of Fixed Effects:\n (Intr)\nroad.dst_km -0.851\n```\n\n\n:::\n:::\n\n1. Note that we need to use this column as the random effect because plots are not uniquely named across sites (i.e., all sites have plots \"a\", \"b\", and \"c\"). Making the random effect just the 'plot' column would fail to reflect how plots are nested within each site\n\nThis test reveals that while there is a significant relationship between road distance and tarantula population but the effect is not nearly as strong as it was when we let plot-level variation be ignored. This is likely due to high (or low) average populations in a single plot skewing the site-level average. Still, this is a result we can be more confident in because we've now accounted for all known sources of variation in our data--either by including them as fixed effects or including them as a random effects.\n\nWe can create one more graph of our tidy data and use some aesthetic settings to make sure the nested structure of the data is clear to those looking at our work. Note that you could also use predicted values from the model itself though that choice is--arguably--a matter of personal preference.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\nggplot(tarantula_df, aes(y = tarantula_count, x = road.dist_km)) +\n geom_point(aes(color = plot, shape = as.factor(year)), size = 2, alpha = 0.5) +\n geom_smooth(method = \"lm\", formula = \"y ~ x\", se = T, color = \"black\") +\n labs(y = \"Tarantula Abundance\", x = \"Distance to Nearest Road (km)\") +\n theme_bw()\n```\n\n::: {.cell-output-display}\n![](mod_stats_files/figure-html/mem-final-graph-1.png){fig-align='center' width=672}\n:::\n:::\n\n\n## Multi-Model Inference\n\nRegardless of your choice of statistical test, multi-model inference may be an appropriate method to use to assess your hypothesis. As stated earlier, this frames your research question as a case of which variables _best_ explain the data rather than the likelihood of the observed effect relating to any variable in particular. \n\nTo begin, it can be helpful to write out all possible \"candidate models\". For instance, let's say that you measured some response variable (Y) and several potential explanatory variables (X, W, and Z). We would then fit the following candidate models:\n\n1. X alone explains the most variation in Y\n2. W alone explains the most variation in Y\n3. Z alone explains the most variation in Y\n4. X, W, and Z together explain the most variation in Y\n\nWe might also fit other candidate models for pairs of X, W, and Z but for the sake of simplicity in this hypothetical we'll skip those. Note that for this method to be appropriate you need to fit the same type of model in all cases!\n\nOnce we've fit all of our models and assigned them to objects, we can use the `AIC` function included in base R to compare the AIC score of each model. \"AIC\" stands for Akaike (_AH-kuh-ee-kay_) Information Criterion and is one of several related information criteria for summarizing a model's explanatory power. Models with more parameters are penalized to make it mathematically possible for a model with fewer explanatory variables to still do a better job capturing the variation in the data.\n\nThe model with the _lowest_ AIC best explains the data. Technically any difference in AIC indicates model improvement but many scientists use a rule of thumb of a difference of 2. So, if two models have AIC scores that differ by less than 2, you can safely say that they have comparable explanatory power. That is definitely a semi-arbitrary threshold but so is the 0.05 threshold for _p_-value \"significance\".\n\n### AIC Case Study\n\nLet's check out an example using AIC to compare the strengths of several models. Rather than using simulated data--as we did earlier in the mixed-effect model section--we'll use some real penguin data included in the `palmerpenguins` package.\n\nThis dataset includes annual data on three penguin species spread across several islands. The sex of the penguins was also recorded in addition to the length of their flippers, body mass, and bill length and depth.\n\nFor the purposes of this example, our research question is as follows: **what factors best explain penguin body mass?**\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load the penguins data from the `palmerpenguins` package\ndata(penguins)\n\n# Make a version where no NAs are allowed\npeng_complete <- penguins[complete.cases(penguins), ] # <1>\n\n# Check the structure of it\ndplyr::glimpse(peng_complete)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\nRows: 333\nColumns: 8\n$ species Adelie, Adelie, Adelie, Adelie, Adelie, Adelie, Adel…\n$ island Torgersen, Torgersen, Torgersen, Torgersen, Torgerse…\n$ bill_length_mm 39.1, 39.5, 40.3, 36.7, 39.3, 38.9, 39.2, 41.1, 38.6…\n$ bill_depth_mm 18.7, 17.4, 18.0, 19.3, 20.6, 17.8, 19.6, 17.6, 21.2…\n$ flipper_length_mm 181, 186, 195, 193, 190, 181, 195, 182, 191, 198, 18…\n$ body_mass_g 3750, 3800, 3250, 3450, 3650, 3625, 4675, 3200, 3800…\n$ sex male, female, female, female, male, female, male, fe…\n$ year 2007, 2007, 2007, 2007, 2007, 2007, 2007, 2007, 2007…\n```\n\n\n:::\n:::\n\n1. This is a base R way of keeping only rows that have _no_ `NA` values in any column. It is better to identify and handle `NA`s more carefully but for this context we just want to have the same number of observations in each model\n\nWith our data in hand and research question in mind, we can fit several candidate models that our scientific intuition and the published literature support as probable then compare them with AIC.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Species and sex\nmod_spp <- lm(body_mass_g ~ species + sex, data = peng_complete)\n\n# Island alone\nmod_isl <- lm(body_mass_g ~ island, data = peng_complete)\n\n# Combination of species and island\nmod_eco <- lm(body_mass_g ~ island + species + sex, data = peng_complete)\n\n# Body characteristics alone\nmod_phys <- lm(body_mass_g ~ flipper_length_mm + bill_length_mm + bill_depth_mm,\n data = peng_complete)\n\n# Global model\nmod_sink <- lm(body_mass_g ~ island + species + sex + # <1>\n flipper_length_mm + bill_length_mm + bill_depth_mm,\n data = peng_complete)\n```\n:::\n\n1. We've named the global model \"sink\" because of the American idiom \"everything but the kitchen sink.\" It is used in cases where everything that can be included has been\n\nOnce we've fit all of these models, we can use the `AIC` function from base R (technically from the `stats` package included in base R).\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Compare models\nAIC(mod_spp, mod_isl, mod_eco, mod_phys, mod_sink) %>% \n dplyr::arrange(AIC) # <1>\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n df AIC\nmod_sink 10 4727.242\nmod_spp 5 4785.594\nmod_eco 7 4789.480\nmod_phys 5 4929.554\nmod_isl 4 5244.224\n```\n\n\n:::\n:::\n\n1. Unfortunately, the `AIC` function doesn't sort by AIC score automatically so we're using the `arrange` function to make it easier for us to rank models by their AIC scores\n\nInterestingly, it looks like the best model (i.e., the one that explains most of the data) is the global model that included most of the available variables. As stated earlier, it is not always the case that the model with the most parameters has the lowest AIC so we can be confident this is a \"real\" result. The difference between that one and the next (incidentally the model where only species and sex are included as explanatory variables) is much larger than 2 so we can be confident that the global model is much better than the next best.\n\nWith this result your interpretation would be that penguin body mass is better explained by a combination of species, sex, physical characteristics of the individual penguin, and the penguin's home island than it is by any of the other candidate models. In a publication you'd likely want to report this entire AIC table (either parenthetically or in a table) so that reviewers could evaluate your logic.\n\n## Meta-Analysis\n\nMany synthesis projects are able to find the original data of each study, harmonize that data, and then perform standard analyses on that synthesized data. However, in some cases you may find that the data used in different projects are not directly comparable. For instance, if you want to know what the effect of restoration methods are on forest recovery you might not be able to simply combine data from different studies that use widely different restoration methods, data collection methods, and have different forest community compositions. In such cases **you can use meta-analysis to compare the _results_ of different studies rather than using their data**. Meta-analysis is named the way it is because it is an analysis of prior analyses.\n\nTo perform meta-analysis you'll need to calculate an \"effect size\" for all studies you'd like to include. An effect size captures the _direction_ and _magnitude_ of the relationship analyzed in each original study. If you use a standard effect size calculation for each stud, you'll make it possible to directly compare results across these studies (even if context differs among them!). Note that some people disagree with the word \"effect\" in \"effect size\" because it suggests a causal relationship; for our purposes, let's consider 'effect' to be inclusive of correlative relationships and ignore the possible implication of causality.\n\nIn order to calculate these effect sizes you'll need to extract the following information from each study:\n\n1. A measure of the 'central tendency' of the response\n - Often the arithmetic mean but can also be a proportion or a correlation\n - You'll need to do this separately for any groups within the study\n2. A measure of the variation in the response\n - Typically standard deviation\n3. The sample size of the response\n\nOnce you have that information, you can calculate effect sizes for the various groups in each study. Note that the importance of this information to meta-analyses should also highlight how vital it is that you report this information in your own research! Doing so will enable future meta-analyses to include your study and increase the scientific impact of your work as well as its professional benefits to you.\n\nOne such effect size is Cohen's _d_ and is a reasonable effect size for quantifying the difference in means between two groups. In order to perform this calculation you simply need the mean, standard deviation, and sample size for two groups. Let's check out an example to demonstrate.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load needed library\nlibrary(esc)\n\n# Calculate Cohen's d\ncalc_effect <- esc::esc_mean_sd(grp1m = 50, grp2m = 60, \n grp1sd = 10, grp2sd = 10, \n grp1n = 50, grp2n = 50)\n\n# Check out output\ncalc_effect\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n\nEffect Size Calculation for Meta Analysis\n\n Conversion: mean and sd to effect size d\n Effect Size: -1.0000\n Standard Error: 0.2121\n Variance: 0.0450\n Lower CI: -1.4158\n Upper CI: -0.5842\n Weight: 22.2222\n```\n\n\n:::\n:::\n\n\nNote that Cohen's _d_ is just one effect size available to you and others may be more appropriate in certain contexts. Just like any other metric, which effect size you choose is a mix of your scientific intution and appropriateness for the content of your data. For a deeper dive into the breadth of effect size considerations available to you, see [the relevant chapter](https://bookdown.org/MathiasHarrer/Doing_Meta_Analysis_in_R/effects.html#effects) of the _Doing Meta-Analysis in R_ online book.\n\nAfter you've calculated all relevant effect sizes--using your chosen flavor of effect size--the \"actual\" meta-analysis is nearly finished. Simply create a graph of the effect sizes with error bars indicating confidence intervals (included by default in most effect size functions). Where the error bars overlap among studies, there is no significant difference between those effect sizes. Conversely, where the error bars _do not_ overlap among studies the effect sizes do significantly differ indicating that the studies results' differ for the data used to calculate the effect size.\n\n## Additional Resources\n\n### Papers & Documents\n\n- [Understanding ‘It Depends’ in Ecology: A Guide to Hypothesising, Visualising and Interpreting Statistical Interactions](https://onlinelibrary.wiley.com/doi/10.1111/brv.12939). Spake _et al._, 2023. **Biological Reviews** \n- [Improving Quantitative Synthesis to Achieve Generality in Ecology](https://www.nature.com/articles/s41559-022-01891-z). Spake _et al._, 2022.**Nature Ecology and Evolution**\n- [Doing Meta-Analysis with R: A Hands-On Guide](https://bookdown.org/MathiasHarrer/Doing_Meta_Analysis_in_R/). Harrier _et al._ 2023.\n- [Mixed Effects Models and Extensions in Ecology with R](https://link.springer.com/book/10.1007/978-0-387-87458-6). Zuur _et al._, 2009.\n\n### Workshops & Courses\n\n- Matt Vuorre's [Bayesian Meta-Analysis with R, Stan, and brms](https://mvuorre.github.io/posts/2016-09-29-bayesian-meta-analysis/)\n\n### Websites\n\n- [Bayesian Meta-Analysis in brms-II](https://solomonkurz.netlify.app/blog/2020-10-16-bayesian-meta-analysis-in-brms-ii/). Solomon Kurz, A., 2022.\n- [Meta-Analysis with R](https://github.com/wviechtb/workshop_2022_ma_esmarconf/blob/master/workshop_meta-analysis.pdf) slide deck. Viechtbauer, W., 2022.\n- [The `metafor` Package: A Meta-Analysis Package for R](https://www.metafor-project.org/doku.php/tips:rma_vs_lm_lme_lmer?s%5B%5D=lme4). Viechtbauer, W., 2021.\n- [Bayesian Meta-Analysis in brms](https://solomonkurz.netlify.app/blog/bayesian-meta-analysis/). Solomon Kurz, A., 2018.\n", + "markdown": "---\ntitle: \"Analysis & Modeling\"\ncode-annotations: hover\n---\n\n\n## Overview\n\nGiven the wide range in statistical training in graduate curricula (and corresponding breadth of experience among early career researchers), we'll be approaching this module by splitting it into two halves.\n\n1. First half: a \"flipped approach\" where project teams will share their proposed analyses with one another\n2. Second half: typical instructional module dedicated to **analyses that are more common in--or exclusive to--synthesis research**. \n\nContent produced by project teams during the flipped half may be linked in the '[Additional Resources](https://lter.github.io/ssecr/mod_stats.html#additional-resources)' section at the bottom of this module at the discretion of each team. Otherwise the content of this module will focus only on the non-flipped content.\n\n## Learning Objectives\n\nAfter completing this module you will be able to: \n\n- Describe proposed analytical methods to an interested audience of mixed prior experience\n- Explain nuance in interpretation of results of proposed analyses\n- Identify some statistical tests common in synthesis research\n- Perform some synthesis-specific analyses\n\n## Needed Packages\n\nIf you'd like to follow along with the code chunks included throughout this module, you'll need to install the following packages:\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Note that these lines only need to be run once per computer\n## So you can skip this step if you've installed these before\ninstall.packages(\"tidyverse\")\ninstall.packages(\"lmerTest\")\ninstall.packages(\"palmerpenguins\")\ninstall.packages(\"esc\")\n```\n:::\n\n\nWe'll go ahead and load some of these libraries as well to be able to better demonstrate these concepts.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load needed libraries\nlibrary(tidyverse)\nlibrary(lmerTest)\nlibrary(palmerpenguins)\n```\n:::\n\n\n## Hypothesis Framing Aside\n\nBefore we dive in, we should discuss two of the ways in which you can frame your hypothesis and the differences in interpretation and appropriate statistical tool(s) that follow from that choice. We'll restrict our conversation here to **two alternate modes of thinking about your hypothesis: frequentist statistics versus multi-model inference.**\n\nNote that this is something of a false dichotomy as tools from both worlds can be/are frequently used to complement one another. However, many graduate students are trained by instructors with strong feelings about one method _in opposition to_ the other so it is worthwhile to consider these two paths separately even if you wind up using components of both in your own work.\n\n::::{.panel-tabset}\n### Frequentist Inference\n\nHypotheses here are a question of whether a variable has a \"significant\" effect on another. \"Significant\" has a very precise meaning in this context that has to do with _p_-values. Fundamentally, these methods focus on whether the observed relationship in the data is likely to be observed by chance alone or not. Strong effects are less likely--though not impossible--to be observed due to random chance.\n\nIf your hypothesis can be summarized as something along the lines of 'we hypothesize that X affects Y' then frequentist inference may be a more appropriate methodology.\n\nFor the purposes of SSECR, **our discussion of frequentist inference will focus on mixed-effect models**.\n\n### Multi-Model Inference\n\nHyoptheses here are a question of which variables explain the _most_ variation in the data. Methods in this framing are unconcerned--or at least less concerned than in frequentist inference--with the probability associated with a particular variable. Intead, these methods focus on which of a set of user-defined candidate models explains most of the noise in the data _even when that best model does not necessarily explain much of that variation in absolute terms_.\n\nIf your hypothesis can be summarized as something along the lines of 'we hypothesize that models including X explain more of the variation in Y than those that do not' then multi-model inference may be a more appropriate methodology.\n\nFor the purposes of SSECR, **our discussion of multi-model inference will focus on comparing model strengths with AIC**.\n\n::::\n\n## Mixed-Effects Models\n\nIn any statistical test there is at least one response variable (a.k.a. \"dependent\" variable) and some number of explanatory variables (a.k.a. \"independent\" variables). However, in biology our experiments often involve repeated sampling over time or at the same locations. These variables (time or site) are neither response nor explanatory variables but we might reasonably conclude that they affect our response and/or explanatory variables.\n\nIn essence we want to use a statistical tool that asks 'what is the effect of the explanatory variable(s) on the response _when the variation due to these non-variable considerations is accounted for_?' Such tests are called **mixed-effects models**. This name derives from considering explanatory variables \"fixed effects\" and non-explanatory/response variables as \"random effects\". Including both fixed and random effects thus creates a model with \"mixed effects.\"\n\n### Types of Random Effect\n\nThere are a few types of random effects but we can limit our conversation here to just two: random intercepts and random slopes.\n\n:::{.panel-tabset}\n#### Random Intercepts\n\nRandom intercepts should be used when you expect that the average response differs among levels of that variable but not in a way that changes the relationship between each level of this variable and the other variables (either fixed or random). In statistical terms you want to allow the intercept to change with levels of this variable.\n\nFor example, let's imagine that we are studying the effect of different organic farming practices on beneficial insect populations. We build relationships with several organic farmers willing to let us conduct this research on their properties and sample the insect communities at each farm over the course of a summer. However, we know that each farm is surrounded by a different habitat type that affects the composition of the local insect community. It is reasonable to expect that even farms where 'the same' management method is used are likely to differ because of this difference in landscape context.\n\nIn cases like this, we don't want to include a term for 'site' as a fixed effect but we do want to account for those differences so that our assessment of the significance of our explanatory variables isn't limited by the variation due to site.\n\n#### Random Slopes\n\nRandom slopes should be used when you expect that the average response differs among levels of that variable in a way that does change with other variables.\n\nFor example, let's imagine that we are studying the effect of temperature on avian malaria rates in songbirds. We identify several sites--along a gradient of daytime temperature ranges--where our species of interest can be found, capture them, and measure malaria infection rates. However, unbeknownst to us at the start of our study, our study sites have varying populations of dragonflies which affects local mosquito populations and malaria transmission/infection rates. If we revisit our sites repeatedly for several years is is reasonable to expect that this difference among sites likely affects the _relationship between_ daytime temperatures and songbird malaria rates.\n\nBy including site as a random slope in this context, we can account for this effect and still analyze our explanatory variables of interest. Note that random slopes are _very_ \"data hungry\" so you may not be able to use them without very high replication in your study design.\n:::\n\n### Nested Random Effects\n\nTo further complicate matters, we can use nested random effects as well. These can be either random intercepts or random slopes though they are more commonly seen with random intercepts. A nested random effect accounts for the effect of one random variable that _is itself affected by another variable!_ A classic example of this is when a study design uses two (or more) levels of spatial nestedness in their experimentall design.\n\nFor instance, let's imagine we were conducting a global study of marine plankton biodiversity. To gether these data we took several cruises (scientific not--exclusively--pleasure) at different places around the world and during each cruise we followed a set of transects. In each transect we did several plankton tows and quantified the diversity of each tow. We can reasonably assume the following:\n\n1. Each cruise differs from each other cruise (due to any number of climatic/ecological factors)\n - But cruises within the same part of the world are still likely to have similar planktonic communities\n2. Within each cruise, each transect differs from the others (again, due to unpreventable factors)\n - But transects within the same cruise are still likely to be more similar to one another than to transects in different cruises (even other ones in the same region!)\n3. Within each transect, each plankton tow differs from one another!\n - But again, more similar to other tows in the same transect than other tows in different transects/cruises\n\nIf we put these assumptions together we realize we want to account for the variation of cruise, transect, and tow while still retaining the nestedness of the similarity among samples. A nested random effect where transect is nested inside of cruise and tow is nested inside of transect would capture this effectively!\n\n### Philosophical Note: Random vs. Fixed\n\nDeciding whether a given variable should be a fixed or random effect can be tough. You'll likely need to rely on your scientific intuition about which feels more appropriate and then be prepared to defend that decision to your committee and/or \"reviewer \\#2\". It may prove helpful though to consider whether you 'care' about the effect of that variable. \n\nIf your hypothesis includes that variable than it should likely be a fixed effect. If the variable is just a facet of your experimental design but isn't something you're necessarily interested in testing, then it should likely be a random effect. And, once you've made your decision, it is totally okay to change your mind and tweak the structure of your model!\n\n:::{.callout-warning icon=\"false\"}\n#### Discussion: Random versus Fixed Effects\n\nWith a small group, decide whether you think the terms in the examples below should be fixed effects or random effects:\n\n- You study small mammal populations in urban settings\n - Should 'proportion green space' be a fixed effect or a random effect?\n- You are part of a team studying leopard seal feeding behavior\n - What type of effect should 'observer' be?\n- You study the gut microbiota of a particular beetle species\n - Should 'beetle sex' be a fixed or a random effect?\n - What about beetle life stage (e.g., larva versus adult)?\n - What about the region of the gut from which the samples were taken?\n- You study vascular plant chemical defenses against herbivory\n - Should phylogeny (i.e., evolutionary relatedness) be a fixed or random effect?\n - What about feeding guild of the herbivore?\n:::\n\n### Mixed-Effects Case Study\n\nLet's imagine we are researching tarantula populations for several years in the Chihuahuan Desert. Our hypothesis is that the number of tarantulas will be greater in sites further from the nearest road. We select ten study sites of varying distances from the nearest road and intensively count our furry friends at three plots within each site for several months. We return to our sites--and their associated plots--and repeat this process each year for three years. In the second year we have help from a new member of our lab but in the third year we're back to working alone (they had their own project to handle by then). We enter our data and perform careful quality control to get it into a tidy format ready for analyis.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Read in data\ntarantula_df <- read.csv(file = file.path(\"data\", \"tarantulas.csv\"))\n\n# Check structure\nstr(tarantula_df)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n'data.frame':\t450 obs. of 6 variables:\n $ year : int 2022 2022 2022 2022 2022 2022 2022 2022 2022 2022 ...\n $ road.dist_km : num 64.5 64.5 64.5 64.5 64.5 ...\n $ site : chr \"site_A\" \"site_A\" \"site_A\" \"site_A\" ...\n $ plot : chr \"plot_a\" \"plot_a\" \"plot_a\" \"plot_a\" ...\n $ site.plot : chr \"A_a\" \"A_a\" \"A_a\" \"A_a\" ...\n $ tarantula_count: int 199 220 213 206 220 128 156 121 121 142 ...\n```\n\n\n:::\n:::\n\n\nWith our data in hand, we now want to run some statistical tests and--hopefully--get some endorphine-inducingly small _p_-values. If we choose to simply ignore our possible random effects, we could fit a linear regression.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Fit model\ntarantula_lm <- lm(tarantula_count ~ road.dist_km, data = tarantula_df) # <1>\n\n# Extract summary\nsummary(tarantula_lm)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n\nCall:\nlm(formula = tarantula_count ~ road.dist_km, data = tarantula_df)\n\nResiduals:\n Min 1Q Median 3Q Max \n-216.916 -77.487 8.486 59.913 316.084 \n\nCoefficients:\n Estimate Std. Error t value Pr(>|t|) \n(Intercept) -3.3935 14.8929 -0.228 0.82 \nroad.dist_km 2.8140 0.2775 10.141 <2e-16 ***\n---\nSignif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1\n\nResidual standard error: 100.2 on 448 degrees of freedom\nMultiple R-squared: 0.1867,\tAdjusted R-squared: 0.1849 \nF-statistic: 102.8 on 1 and 448 DF, p-value: < 2.2e-16\n```\n\n\n:::\n:::\n\n1. R syntax for statistical tests is `response ~ explanatory` a.k.a. `Y ~ X`\n\nThis naive test seems to support our hypothesis. However, sampling effort differed between the three study years. Not only was there a second person in the second year but we can also reasonably expect that by the third year in this system we had greatly improved our tarantula-finding skills. So, a random effect of year is definitely justified. We are not concerned that the different study years will affect the relationship between tarantula populations and road distance though so a random intercept is fine.\n\nThere could be an argument for including year as a fixed effect in its own right but some preliminary investigations reveal no significant climatic differences across the region we worked in those three years. So, while we think that years may differ from one another, that difference is not something we care to analyze.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Fit the new model\ntarantula_mem1 <- lmerTest::lmer(tarantula_count ~ road.dist_km + \n (1|year), # <1>\n data = tarantula_df)\n\n# Extract summary\nsummary(tarantula_mem1)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\nLinear mixed model fit by REML. t-tests use Satterthwaite's method [\nlmerModLmerTest]\nFormula: tarantula_count ~ road.dist_km + (1 | year)\n Data: tarantula_df\n\nREML criterion at convergence: 5362.6\n\nScaled residuals: \n Min 1Q Median 3Q Max \n-2.6578 -0.7941 0.1653 0.6517 2.9451 \n\nRandom effects:\n Groups Name Variance Std.Dev.\n year (Intercept) 1940 44.04 \n Residual 8747 93.52 \nNumber of obs: 450, groups: year, 3\n\nFixed effects:\n Estimate Std. Error df t value Pr(>|t|) \n(Intercept) -3.3935 28.9768 3.1783 -0.117 0.914 \nroad.dist_km 2.8140 0.2589 446.0000 10.868 <2e-16 ***\n---\nSignif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1\n\nCorrelation of Fixed Effects:\n (Intr)\nroad.dst_km -0.455\n```\n\n\n:::\n:::\n\n1. This is the syntax for specifying a random intercept (random slope variables should be before the `|` where `1` goes for a random intercept)\n\nBy including that random effect we actually get a slightly stronger effect of road distance (T value of ~12 without versus ~13 with). This is because our new random effect accounts for some of the 'noise' between study years. That actually gives us a better picture of the relationship between our response and explanatory variables.\n\nNow that we're already using a mixed-effects model, we have little excuse not to account for the other potential random effect: plot! Remember that there were three plots within each site and from our extensive time in the field we have developed a strong intuition that there might be substantial among-plot variation at each site. We can make a quick exploratory graph to facilitate an 'eyeball test' of whether the data show what our intuition suggest.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\nggplot(tarantula_df, aes(y = tarantula_count, x = plot, fill = plot)) +\n geom_violin(alpha = 0.5) + # <1>\n geom_jitter(width = 0.25, size = 0.5) +\n facet_wrap(site ~ .) +\n theme_bw() +\n theme(axis.text.x = element_text(angle = 35, hjust = 1)) # <2>\n```\n\n::: {.cell-output-display}\n![](mod_stats_files/figure-html/mem-explore-graph-1.png){fig-align='center' width=672}\n:::\n:::\n\n1. Violin plots are a nice alternative to boxplots because they allow visualizing data distributions directly rather than requiring an intutive grasp of the distribution metrics described by each bit of a boxplot\n2. This is allowing us to 'tilt' the X axis tick labels so they don't overlap with one another\n\nThis graph clearly supports our intuition that among-plot variation is dramatic! We _could_ account for this by including plot as a fixed effect but we'll need to sacrifice a lot of degrees of freedom (can be thought of as \"statistical power\") for a variable that we don't actually care about. Instead, we could include plot as another random effect.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Fit the new model\ntarantula_mem2 <- lmerTest::lmer(tarantula_count ~ road.dist_km + \n (1|year) + (1|site.plot), # <1>\n data = tarantula_df)\n\n# Extract summary\nsummary(tarantula_mem2)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\nLinear mixed model fit by REML. t-tests use Satterthwaite's method [\nlmerModLmerTest]\nFormula: tarantula_count ~ road.dist_km + (1 | year) + (1 | site.plot)\n Data: tarantula_df\n\nREML criterion at convergence: 4553.6\n\nScaled residuals: \n Min 1Q Median 3Q Max \n-3.0902 -0.5352 -0.0117 0.5990 3.9323 \n\nRandom effects:\n Groups Name Variance Std.Dev.\n site.plot (Intercept) 8165 90.36 \n year (Intercept) 1991 44.62 \n Residual 1058 32.52 \nNumber of obs: 450, groups: site.plot, 30; year, 3\n\nFixed effects:\n Estimate Std. Error df t value Pr(>|t|) \n(Intercept) -3.393 58.233 23.593 -0.058 0.95402 \nroad.dist_km 2.814 0.973 28.000 2.892 0.00733 **\n---\nSignif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1\n\nCorrelation of Fixed Effects:\n (Intr)\nroad.dst_km -0.851\n```\n\n\n:::\n:::\n\n1. Note that we need to use this column as the random effect because plots are not uniquely named across sites (i.e., all sites have plots \"a\", \"b\", and \"c\"). Making the random effect just the 'plot' column would fail to reflect how plots are nested within each site\n\nThis test reveals that while there is a significant relationship between road distance and tarantula population but the effect is not nearly as strong as it was when we let plot-level variation be ignored. This is likely due to high (or low) average populations in a single plot skewing the site-level average. Still, this is a result we can be more confident in because we've now accounted for all known sources of variation in our data--either by including them as fixed effects or including them as a random effects.\n\nWe can create one more graph of our tidy data and use some aesthetic settings to make sure the nested structure of the data is clear to those looking at our work. Note that you could also use predicted values from the model itself though that choice is--arguably--a matter of personal preference.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\nggplot(tarantula_df, aes(y = tarantula_count, x = road.dist_km)) +\n geom_point(aes(color = plot, shape = as.factor(year)), size = 2, alpha = 0.5) +\n geom_smooth(method = \"lm\", formula = \"y ~ x\", se = T, color = \"black\") +\n labs(y = \"Tarantula Abundance\", x = \"Distance to Nearest Road (km)\") +\n theme_bw()\n```\n\n::: {.cell-output-display}\n![](mod_stats_files/figure-html/mem-final-graph-1.png){fig-align='center' width=672}\n:::\n:::\n\n\n## Multi-Model Inference\n\nRegardless of your choice of statistical test, multi-model inference may be an appropriate method to use to assess your hypothesis. As stated earlier, this frames your research question as a case of which variables _best_ explain the data rather than the likelihood of the observed effect relating to any variable in particular. \n\nTo begin, it can be helpful to write out all possible \"candidate models\". For instance, let's say that you measured some response variable (Y) and several potential explanatory variables (X, W, and Z). We would then fit the following candidate models:\n\n1. X alone explains the most variation in Y\n2. W alone explains the most variation in Y\n3. Z alone explains the most variation in Y\n4. X, W, and Z together explain the most variation in Y\n\nWe might also fit other candidate models for pairs of X, W, and Z but for the sake of simplicity in this hypothetical we'll skip those. Note that for this method to be appropriate you need to fit the same type of model in all cases!\n\nOnce we've fit all of our models and assigned them to objects, we can use the `AIC` function included in base R to compare the AIC score of each model. \"AIC\" stands for Akaike (_AH-kuh-ee-kay_) Information Criterion and is one of several related information criteria for summarizing a model's explanatory power. Models with more parameters are penalized to make it mathematically possible for a model with fewer explanatory variables to still do a better job capturing the variation in the data.\n\nThe model with the _lowest_ AIC best explains the data. Technically any difference in AIC indicates model improvement but many scientists use a rule of thumb of a difference of 2. So, if two models have AIC scores that differ by less than 2, you can safely say that they have comparable explanatory power. That is definitely a semi-arbitrary threshold but so is the 0.05 threshold for _p_-value \"significance\".\n\n### AIC Case Study\n\nLet's check out an example using AIC to compare the strengths of several models. Rather than using simulated data--as we did earlier in the mixed-effect model section--we'll use some real penguin data included in the `palmerpenguins` package.\n\nThis dataset includes annual data on three penguin species spread across several islands. The sex of the penguins was also recorded in addition to the length of their flippers, body mass, and bill length and depth.\n\nFor the purposes of this example, our research question is as follows: **what factors best explain penguin body mass?**\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load the penguins data from the `palmerpenguins` package\ndata(penguins)\n\n# Make a version where no NAs are allowed\npeng_complete <- penguins[complete.cases(penguins), ] # <1>\n\n# Check the structure of it\ndplyr::glimpse(peng_complete)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\nRows: 333\nColumns: 8\n$ species Adelie, Adelie, Adelie, Adelie, Adelie, Adelie, Adel…\n$ island Torgersen, Torgersen, Torgersen, Torgersen, Torgerse…\n$ bill_length_mm 39.1, 39.5, 40.3, 36.7, 39.3, 38.9, 39.2, 41.1, 38.6…\n$ bill_depth_mm 18.7, 17.4, 18.0, 19.3, 20.6, 17.8, 19.6, 17.6, 21.2…\n$ flipper_length_mm 181, 186, 195, 193, 190, 181, 195, 182, 191, 198, 18…\n$ body_mass_g 3750, 3800, 3250, 3450, 3650, 3625, 4675, 3200, 3800…\n$ sex male, female, female, female, male, female, male, fe…\n$ year 2007, 2007, 2007, 2007, 2007, 2007, 2007, 2007, 2007…\n```\n\n\n:::\n:::\n\n1. This is a base R way of keeping only rows that have _no_ `NA` values in any column. It is better to identify and handle `NA`s more carefully but for this context we just want to have the same number of observations in each model\n\nWith our data in hand and research question in mind, we can fit several candidate models that our scientific intuition and the published literature support as probable then compare them with AIC.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Species and sex\nmod_spp <- lm(body_mass_g ~ species + sex, data = peng_complete)\n\n# Island alone\nmod_isl <- lm(body_mass_g ~ island, data = peng_complete)\n\n# Combination of species and island\nmod_eco <- lm(body_mass_g ~ island + species + sex, data = peng_complete)\n\n# Body characteristics alone\nmod_phys <- lm(body_mass_g ~ flipper_length_mm + bill_length_mm + bill_depth_mm,\n data = peng_complete)\n\n# Global model\nmod_sink <- lm(body_mass_g ~ island + species + sex + # <1>\n flipper_length_mm + bill_length_mm + bill_depth_mm,\n data = peng_complete)\n```\n:::\n\n1. We've named the global model \"sink\" because of the American idiom \"everything but the kitchen sink.\" It is used in cases where everything that can be included has been\n\nOnce we've fit all of these models, we can use the `AIC` function from base R (technically from the `stats` package included in base R).\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Compare models\nAIC(mod_spp, mod_isl, mod_eco, mod_phys, mod_sink) %>% \n dplyr::arrange(AIC) # <1>\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n df AIC\nmod_sink 10 4727.242\nmod_spp 5 4785.594\nmod_eco 7 4789.480\nmod_phys 5 4929.554\nmod_isl 4 5244.224\n```\n\n\n:::\n:::\n\n1. Unfortunately, the `AIC` function doesn't sort by AIC score automatically so we're using the `arrange` function to make it easier for us to rank models by their AIC scores\n\nInterestingly, it looks like the best model (i.e., the one that explains most of the data) is the global model that included most of the available variables. As stated earlier, it is not always the case that the model with the most parameters has the lowest AIC so we can be confident this is a \"real\" result. The difference between that one and the next (incidentally the model where only species and sex are included as explanatory variables) is much larger than 2 so we can be confident that the global model is much better than the next best.\n\nWith this result your interpretation would be that penguin body mass is better explained by a combination of species, sex, physical characteristics of the individual penguin, and the penguin's home island than it is by any of the other candidate models. In a publication you'd likely want to report this entire AIC table (either parenthetically or in a table) so that reviewers could evaluate your logic.\n\n## Meta-Analysis\n\nMany synthesis projects are able to find the original data of each study, harmonize that data, and then perform standard analyses on that synthesized data. However, in some cases you may find that the data used in different projects are not directly comparable. For instance, if you want to know what the effect of restoration methods are on forest recovery you might not be able to simply combine data from different studies that use widely different restoration methods, data collection methods, and have different forest community compositions. In such cases **you can use meta-analysis to compare the _results_ of different studies rather than using their data**. Meta-analysis is named the way it is because it is an analysis of prior analyses.\n\nTo perform meta-analysis you'll need to calculate an \"effect size\" for all studies you'd like to include. An effect size captures the _direction_ and _magnitude_ of the relationship analyzed in each original study. If you use a standard effect size calculation for each stud, you'll make it possible to directly compare results across these studies (even if context differs among them!). Note that some people disagree with the word \"effect\" in \"effect size\" because it suggests a causal relationship; for our purposes, let's consider 'effect' to be inclusive of correlative relationships and ignore the possible implication of causality.\n\nIn order to calculate these effect sizes you'll need to extract the following information from each study:\n\n1. A measure of the 'central tendency' of the response\n - Often the arithmetic mean but can also be a proportion or a correlation\n - You'll need to do this separately for any groups within the study\n2. A measure of the variation in the response\n - Typically standard deviation\n3. The sample size of the response\n\nOnce you have that information, you can calculate effect sizes for the various groups in each study. Note that the importance of this information to meta-analyses should also highlight how vital it is that you report this information in your own research! Doing so will enable future meta-analyses to include your study and increase the scientific impact of your work as well as its professional benefits to you.\n\nOne such effect size is Cohen's _d_ and is a reasonable effect size for quantifying the difference in means between two groups. In order to perform this calculation you simply need the mean, standard deviation, and sample size for two groups. Let's check out an example to demonstrate.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load needed library\nlibrary(esc)\n\n# Calculate Cohen's d\ncalc_effect <- esc::esc_mean_sd(grp1m = 50, grp2m = 60, \n grp1sd = 10, grp2sd = 10, \n grp1n = 50, grp2n = 50)\n\n# Check out output\ncalc_effect\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n\nEffect Size Calculation for Meta Analysis\n\n Conversion: mean and sd to effect size d\n Effect Size: -1.0000\n Standard Error: 0.2121\n Variance: 0.0450\n Lower CI: -1.4158\n Upper CI: -0.5842\n Weight: 22.2222\n```\n\n\n:::\n:::\n\n\nNote that Cohen's _d_ is just one effect size available to you and others may be more appropriate in certain contexts. Just like any other metric, which effect size you choose is a mix of your scientific intution and appropriateness for the content of your data. For a deeper dive into the breadth of effect size considerations available to you, see [the relevant chapter](https://bookdown.org/MathiasHarrer/Doing_Meta_Analysis_in_R/effects.html#effects) of the _Doing Meta-Analysis in R_ online book.\n\nAfter you've calculated all relevant effect sizes--using your chosen flavor of effect size--the \"actual\" meta-analysis is nearly finished. Simply create a graph of the effect sizes with error bars indicating confidence intervals (included by default in most effect size functions). Where the error bars overlap among studies, there is no significant difference between those effect sizes. Conversely, where the error bars _do not_ overlap among studies the effect sizes do significantly differ indicating that the studies results' differ for the data used to calculate the effect size.\n\n## Additional Resources\n\n### Papers & Documents\n\n- Spake, R. _et al._ [Understanding 'It Depends' in Ecology: A Guide to Hypothesising, Visualising and Interpreting Statistical Interactions](https://onlinelibrary.wiley.com/doi/10.1111/brv.12939). **2023**. _Biological Reviews_\n- Spake, R. _et al._ [Improving Quantitative Synthesis to Achieve Generality in Ecology](https://www.nature.com/articles/s41559-022-01891-z). **2022**. _Nature Ecology and Evolution_\n- Harrier, M. _et al._ [Doing Meta-Analysis with R: A Hands-On Guide](https://bookdown.org/MathiasHarrer/Doing_Meta_Analysis_in_R/). **2021**.\n- Zuur, A.F. _et al._ [Mixed Effects Models and Extensions in Ecology with R](https://link.springer.com/book/10.1007/978-0-387-87458-6). **2009**.\n\n### Workshops & Courses\n\n- Vuorre, M. [Bayesian Meta-Analysis with R, Stan, and brms](https://mvuorre.github.io/posts/2016-09-29-bayesian-meta-analysis/). **2016**.\n\n### Websites\n\n- Kurz, A. S. [Bayesian Meta-Analysis in brms-II](https://solomonkurz.netlify.app/blog/2020-10-16-bayesian-meta-analysis-in-brms-ii/). **2022**.\n- Viechtbauer, W. [Meta-Analysis with R](https://github.com/wviechtb/workshop_2022_ma_esmarconf/blob/master/workshop_meta-analysis.pdf). **2022**.\n- Viechtbauer, W. [The `metafor` Package: A Meta-Analysis Package for R](https://www.metafor-project.org/doku.php/tips:rma_vs_lm_lme_lmer?s%5B%5D=lme4). **2021**.\n- Kurz, A. S. [Bayesian Meta-Analysis in brms](https://solomonkurz.netlify.app/blog/bayesian-meta-analysis/). **2020**.\n", "supporting": [ "mod_stats_files" ], diff --git a/_freeze/mod_stats/figure-html/mem-explore-graph-1.png b/_freeze/mod_stats/figure-html/mem-explore-graph-1.png index 78a49684e36538fe76024ac6e9ca381fa7398e7d..8b910654f22d051dd070c6cf3fbe4add7026ba27 100644 GIT binary patch literal 189011 zcmc$GWmwc(_cn}!gh~oX2@=wvh#);mi*&b2x6&OV(p}QsCEXz^9YafZmvqCshjX5D zp8xSYpWhE&*Elo3*?X_O*S*%d*S+zTkrKUwL5zWfgmgz-Oh^t12?d6P1a(Kd3Ett4 zS>^%%!88&Slra$$6|^w5u#|uIMn_Ln&s@*aNJmce1rpM;z(7S!LsG@N{BhN#jMP1z zgPECOn3qo@%1q_@K8~$i>Xn*JFom0zJr8N$m`pHw_+5k#2qAC4^=J@ zhzpq)2j)s!m%e}Jw*Kk@zvBXDW(@(^0 zGyho!<(wYUCH{{RO?T+?vo`dRx5wz;c}(Le=5W82BPriY#QIgz8xzXrqbEO~wE_bJIR2@b$ zSVqlyV<#ytOj!FAu^5&9z`oL}HbUQ_Lei|Q^RmR);s36q=u}6#Vx5!A+1JFFH5g-n zJ?&|4^Od)wWlQM1{IPaChK;LS{Ou*I`$#i$49s5~kErf{ku+lP=?qjhAFHg6j$Qq|pdbui4cfurlcqZ)iH&MF?}c}z+fzVL|XK^jFEZM(c~fNmf6OU$RO zQIuA?d)0k|4#-q;Nu2(5{wsFRTP_WBKhKo){&<|h@il*I)|HT6!Taq}S4)>VRUtjMFqYR8lp<*l=}GdurXf#!FfMAQBavKu>tma8S5m~#O}31#K6 zYkuF_7v|q$6tOW7++J)-rdzLBsU_{p#4L|CTtxcf#P+P&Lj3ZAaNHq7&noHWvvXRW zBee=HteN~9@H4|s4tIn3`fa3q{!IHrIeXggZ>J&8hCIK0RhB01%_CnwP`5XynL_&M zT(5O?cH!u8Xh8OG+X!fY)e9$lgXsQ1A*F{38bc zkdPqBkpH}c0!zN}pVv@##D{y!dC5pfFeGuI=L+`78wse+q(6q5l9*M0gwd#PO}5-o zHkFrr{Fz|j>mX9fVj9hZkBJ}Q>S*Z9*kX@8WCYRcJw34f0*jGJCLZ6kQ#TJA(sMgg zO02IQNU#=OJlwi+Z5-I@>FMe@KQd=OJM4BctKgwTgHj+Nqrs3M*zQRG_oc^OYB&eV znNalq`T{aq6;#H3SNNX~{`GTmQAil@i28Et+rR(%A76qYci8>Q(0_a=rV62h$sZ2) zKmOPGA-;q{H~rUW|C}a;9(pp8pwsFYs9U7X{gBG<@wX_)-SuesLqOK}UmF({Bc7O0Y0#c(6#rMQHHcz4@` z@=|!A6d}4$4%43$jpvK3;rWW}sF3#ONGOJu6W)#mh69;_x_Wv;612z=>|X-N7cR66 zX5$rl0b+M47axQ)L)qBa?or7dU-mymqkifBm{Rsbgg49**Knjz(~=`V={W^D8a9+3 z-5etMKMzN&H}H(5*_Q>sdEK9f3x4)U z#KG|X=NVsm;0`HssFGv+zpnq+38x5wf*pVORWcdhIe>|kpG@(SPV>6Rh5l1xB?uEKPKpL6J3|X zJ}Mj>2q8$cmSavtbaW?kQ?q(3&z z{qPqqc#xZyS8q(fq(b3#K2NUKpDu2>mXqbUT2L;pp}|ztaGGm6UeRsVe06ywP-#{$ zbieSHhc#tLfX>GtReu6zKYJ4XvyCT56L`|RvkeogBSqTZ>*B!~ouds2nD?BFlNPv8 zFmJrDrKhKVAAUe%oW$=Uw2Uti%@Ua-pVeX2%eTBl#BLa(8R3EkOGZbdp1&unsyaGD z`l6Okr{%W8;i$plj3-u9(#3{BSW?n7lsy{)b_F(oI3Vn#IW%|BNgWnKLo>IoJYge@Ca@yO&ZwFhs4dIC{7_ zu6b+c4hctFSIm>eK+dUm1+4(CEL`{EzQ2*V0nVVuLrUN}zA!n-Q=Dhpb$ekA1Kh~6 zCyiHU0x?evf}9VAluJxU+3b8x`Z*JCzGzwo6ALgLuPFFfI!64|I3l}wzHC6sin&q0 zH<1~f!>rzIXty^2F|pyDHO-+r10!^g8W9~)n(>Ab%fg$yri%f z8OW8Kox?VCft=pli?e8WKdqMgG>gSZSs-b4;cUAxdb{CF%GlUgX!_k+gDOMC`+1Qg zmC6)3H8lop=N+$Fc9-FN)l~k9SV4u^9R6DC_EfzmcMc6WF zH@&Im&GG6}DZ>5LJ|Ph#Gk=T8J9cjhqmh`P(KhaW?azID^@G**fxu~B@P;m@V9Ux zog!ZEbvYbuEs%&_Z2Z-HE46d3yr2e3;DH29r42FG5fbFnu6HbVay)XA&T)3<|BleXiMg;jGHs}t3w(fNCRDofaEGTLRvNMRO@7sGj?cI+!Y42(mR zj5xB3U1dfC6AV)PC%UWs84{jDtLHDGzpm{bcfGn#PyF#mdl_wrloeY2s5iN5l?hp4 z%Y)N?{S|nM(G?jINLApc(9Z>Bmwd@6;xHKbi~ZT>31{T*zvGmgZPnI(4P>YA;lDlc zzLl&H&m!&BZL^Y7WJS^GH3NAUoz<{DO}k0OPq&HY^y)#774pP3N?!zMA^SH8V#hjuj>Hj*1WZ5pUTnaU-#~I# z6sk8$IwuyBY8nCNwJo%VPg1-vzR19k<@p@4Bx_`fnbOz`Ja_scaIHj6^YL4uW%}HF zg~ePo9b;Ikz3}pLk<+7X3)@kUqoL!9m2(M|J=?Dq6%ORZMu0*)42e9s{T-p9U>@G1 z%hLU?0)wLM3*pi=dIw*Byt?^H@AQ*Gf>&O4YB~MF!HBk99=c{l z5eA&YY^*I+*k7nX9Q+~$HzbEcB%A#Cfv_pRw(Cj8`jtv63_ozgPr7lXfn?!>JMJSL zp=^?b)Ktp8j2+_##8M|jcla(`a`7WsQk0{)Qc@X}_c`8@Y zb-L5>oPrGf^Ueds#3!ls-fN61dPBL2eFP845m>BGo*6Epomtgk2#@CJ)Lc;5CR0@}Sa-G=f`yL|EOY4@rci>I^ zPvbDpuzg-pM7E%9W}G4^X@985-qqP0E8jg?PU_rW9T291=O(>SETjPF_6=$+UZdtP z1p;F#f0fPs6oTZM4S7$Thm`u5eQBateMsK3LIp5F7L64VJ#w%&RZM}ItA2#|<``6%ovr@9p?RWcD z3K?oNq;IgtRF*FBf?$x$@1a0URc3d})*|GHwOLTJeY`AiwR+QjK(EqiyWWnk^8n69 zDo8Pj9`Aa(YA!>i*l-`vH~)l{3TaGxWC8m8s-rSR*0&zsnPU4lAI&?|S?uhR>f{^w<;>}NKy(*N?FJ<%oD9F%*L7}`%PGpO_}x7~$;E)T6lUlZ zHk=xBz8(t|y&v=Lu44g~2P_v?EQ{;ol%PuwT#<%>4yJkg9IP+Q9VGVDRp{&w9MBAf zcekzlka5mRv*4Pp}G4@gxxwFMtVBeQTAIW_Pd|T?5Fdnb(t* zV5I zBW`x)X5{toD7=g8qhOGF(L@u%VuiMu&+nbx_}-=IcSALLxs4TV{OIQ4v45X)LM$M5o6rSSA=u3`GoyHqS;gCv)V3D>jDSt-#3 z{iw!mkdWX+CaI0S@g`du&iP8*d_o;~`a$<2EEZcV@-gMuc*aYTwDW+Gq}OF5l;qHJ zDOf=~OZ5Woh<6nkFB2nUXL%~`YI^ttOKvdy6Lml7+a`Bp^T&bIWW?FB(h3R+Rep*I zMbxJPnvs>c=`lQZg|#og*T4HLz|EiSe>-yan|EjObI;}lN^}Ue3}ltVTLpw{-h)B; zt!{y(&pOncoYAa^3d19v0_5KG<~ITj%^=2`>0dnu#|Dnq+0FV5S7W0R5;moFB^|2@ z_&2f*e<7*6?uI8-c@v{3k-s+WtHa{`~yqsnDhB?Xjh(@LjjX(A>?K!CK20r zCly%_g72^{d^DTkjt_*Z$o+TYZYiu_e*4Cvi zEON2nGBJ;JKEP0Q4SRL%DRZ>T}|WR+&RteJXYMQ zn#Xk5?-QMD7_~9h9x|w@X@zrKZ5RlYN|%Dj_Yqsno9iRkYQ-vki5wnmO-Rx3`l%es zwU07ERUk=Bb{}lZ7(6_&_U{TSO>uwu1&z3*kYc&PhDp;=}Q$h7qTs%g8OWMT+mW*h|^xm#VvO+ zN39@;_-XFfI!Czz*mmtU|H<^#j!m5n?$ZF}8vPX7fR*RODNUY!r z{!q^C5oxV1h;rWW#5Jox=9IkYeLcShp*}@x1e4~W3kqs> z)h+0zp(2F%-b$3Zc_^ZGLgcypGAn{!erkHTxTVFYgxzOpYq53^{qXf2Kl|RT}&R!SAos4 zfr-Oks0c$);J5g|wY>QAS8m2vk#%lcT&iJ0jGqz|G6H$Oko-n5?X< zigtWWmjYGSy6$sY4E<5{{E-KZ0_ZFK&s_`^Mo@Qr81igZ8rE|2=IEtjpy-c8sL?{^ z0&(FqKf-CHqf82`GyCdh*;Z~9hEj=kNV}a5yvq_d5rJ@Xm^4w9M+{!=xSSo@;D0LC zC@JO=GBYaxC5c-aw>P|6#L7=Cph&&4N0a&OT~a!Z*l_!KoAq4kgd5p&JoozucWD#t zQ?uPf()G|Ec^j%ut=d1mAh(|NBCKKU+W$cej15J5jZFOd>wTyO#MmpxuUAMYjh1tv zSH1v7TnZImIF0Z^g2o0T=?Lu=XXv9HC%Cf+E7PR7&PBE?Vdpx(vQEU67-v7N(<-7V zB9b_3yM-)(^BQwDt5#V30g{IT6%0FFp*C$FZ27;FOTWPIYbmc0gcgjd+C*qXWf27T zYIcG*9K+o0!11Zckz0HgYH7>RAnigkYk5Vs(KD_64>fb$CT&K3$bGB>sW}!{+^<>5 zw03aV!*M|Mu1lAL29Xq`U}a@3)YJEc${5E+$=?FC1}^j-mD&fk5qEbt^z7hOl^dU* zE>yHOQX6ss^6#ZPmk=FJ5SX_$D3;k1wm4i4P>^k26_En~bca2I1mMhOhulO)y9cSE zgd&g!I|#|j=beHWpj`XCkoaK(i&6v9|ARG3e*h>UBm03>HsE;w#v6fOz<^s<5A5Up z4`UgL0S1kD%DM&+Gyg`jc~~R7S#yj1;v<;bzl;GR#sw_hC(Z;Gkb?Z%hEQmt>1IlZ zUJUmFW~pPda(b&$e?SW)${3K&Xx3=0dPm{VCzerwz6y zRQKZa;7@MT7#G25=v|(l71k>|YQlW>hOFh+frBVz`%^iEBH*R8*((ySR-rkfOYz4j$+0pT` z#V6bMgeUDn^Jw6xt~}=~X8g;WZ|T|DbmkugKiz&Gnbbu!C$bXv;qXX7<^i|% z5_p;B+9**FJ13ybY|U!f+?>d#`GyvlagJ2K`S($h?~tNlM%yUh=5*Zd(m z)yYwEL_|adQ<;V&?@kt@MdO~nZQwhakTW6_kg1sBy)zBpjyLPfkXPhopx|01GA3tPA)(x|5S$x5RgR%!(dr z<#8ETfC-ast0~ABi@!b6w_W8rqBuB~_2tVWqS!@$j+LLeI{DFJf&tScQ^B2Ww3{|q^SaXntu>qL3J?iHp$Z=?(7(7X_ys!Hz`E;J?R7zQ3doMz7GLImS2#sMic=#3)3I&K9%LQrIcU`i$&mJ_BP^Zd3hk4$R7 zSkfEG5nivV3Od-a7sRs_GB@}3$bP)SOioji8F0wPA-nR@Qy{anY(}ST10>z%(jwMx zc1WfN%{8-^v4ber3uN&CB#@Yv<7x&4ad~-piNoGfGP~6yvoZF&$v1EQW@ke=2=uwJ zT)YCneqf%LumV2O%{nXW0Kg_d$;MGfM@J7Ioc+Fnhg6`>4^K&_yf*41=-|lm$lh|T zcdw>%=t4zjn4cd;^<|D*XDPXxX-PF$(|E^6bi5D!&wh`U?1(*<6b^H|J|In0U-Y|) z?F1?{d5FeJvsQMGc6TkSm*Y#*DL6&%pUpJNDaunBpN!T!+2!Qt?{}2k_-$VeUqYBX zEdH{{)~+4idA9nS=h&g~`ApU8S<7NO^T8dHgjy6J+lh{kCw+`DklaQ3`{{^9l#mbI z#qh7!k4^E#e3E;#!KF41UXc9gUGrF{C6gek?fOziGSY#}m)|BL`3f98yc-$Z2pg?R zH7V)aVihBmkNzA^$x|+>kS<2hUjH{*8L|O~h`UXLf89lK$NhqzJc;okx@r95hLtT5 z57xizgb<>)8bPsm?Sm9P?gd?Ck@3%C;%}VAYd&NrZ~cAmgb>ZBkN5SipGdxstR6E8 z$34#PdbRqqpFMR@Ix_k-OFkOJ>Q+jTjVk~1K-f@FT=@F^WT>f$6c1eX8Mg>>?y~6*qpG^}zZyn=Sz`_n?C%>#Ao36- z46{uS5Une-YEZz8fR#;1CuQ z+V>in2}bh$oA__`hCpH@Wl$Bbw*iKM`xy`*=d)ODAsG_U9S9miS=*(2A5fZK6Qd#` z#2g%|<64!>jhhK*FSNJoS#K#7M!a!3ZL1dGW~algla1%FN0$2C25+gsdPH99dLu|B zOsIa>=Lct-l}QSUiq(?|{4Ou|9}n1QCCVcwf%9=?>=v1!y=v z*IXMb&(i>C_uME2_1bbCcBh1*q%1K!cM~2KgJWa4THGoYeeTTJu9O(dYx;Yr^2Mb< zlS_rhr8tA~W5ocZdfhTkdkWPqXSPLiYiq{Z<&!4%jGKstUVRDM{s5l^jmDIoiLr6r zMIZ^MAH%>o<(Tb7U@N1Vk|*BN2l!m`07@9 z5x2R5l2XgqW?7d-Fi!V{+k+t_5?j{vqutK zd6ZcNiJQdDhN~UbSE4ac2yjkQ4|vs(_rj$yE0R?WUWLy0lo_T zA$?S*`ue*}PQxTS!3!@huheOtMxTnOq3yo>Wo0HKEcA`6xlIavi(7nte!LmsK{&>t znW-+l?d|W64y1#F$m(6nrDp7AX5Zo41E9CXJ3n-L0QVmrRj)KpIC?A9DDTT~ORa>^ zw_&hv_mpIBz(Ty$8=TbY&sJ7e8G_!J%)S$qBXYuxAO#IOxO!32JXEY0NeXsvb(>eK zkAHo;OBxU$E&!3LGv|IQI@2JVELcV0h$+pA~7@l?se3Hc|uJW^_U2f)pF23rpqpvn3#~%VbcDNF~y+q zPJD|QAfDh!2<|< z6Q-5YMbs0W|M`6+4>9n72&EwRb%yv|sELlnYP>xBJU;$&hb`mE%p&K3J^IE!@(nO3 zE`nc0w!aI=5CGBKkf%?dmc03P^C%f)OkFAgSTf9p>R~M!I4Hl{;sF@(6Vfd9_2l2u zCcE;0LU)UxEHb8VcyLlK_ARi~(?2ZrUyFmWBPJ^7%X_^9I0dr0k6Z|j5)*It zp={0QNH5Lb1DpnUs>Lht#r0FK@xDYC*ll9}S~glXGHhhzhmkaYUfX|#+qEjX9rzW7oXIuZZrzgG6gOCf3CE&KUxf8-}PXdrM%+LCih<@)(&IK6rE zQR5EndjCm5WMm2G>gN;|M%iwTJ=Jjky+-n&Lz}t5jv=P|$I1SL-WA6w%^%07-f({D zMBm|Wf$l`uiDg)vCMNv*-bf-;RPr+Em@G zjOKICznj!x)G=|s_sRX${s*tQYFsZ)?Q@fFU!Si3K1CpbyqkOw67om-P>lT(B*6*h4tfQ`${!H40D^e|c~yyj04bN8dEJf(JtO0-p6>`-|A{&_fGT6uO&itEwYQY@*(F-7X$N*Rbvr-UW23{iH_KyA&V~5d~!r z4Qt?kWk(*bfuDOiyn|l<_+wnq2liA>r4#5K54Oo_E4dH!ZH#=o0J(Bry#^}0-O{9s zSBO4$seX6v$Pg$4#SSO!wCU;S(6~o>-b+}v5E-;pFDH^j^dSE=V=+!J#=vIKfHuO^ zlCPSSl*V3msj$jN5XvNuy&O_me;|N3N(xlJfD)zy%oB)`DBq57m`|*z&X31z)gNgA zLcxlfit3w`KKWlr0BT?7zpLliM0b8`lGO#e$ACO}xELAvkbbQo>gfl_V90AkX$xBb zP45s7&zmo$&Pa_0G9^{09lRXEBBM85#&icI_wsUcbr(A#-*kqc{b}fHVvits)iP_9 z7BN44%GY<3iEpnK^y+uq2|kzTAVpa*_eAZyAdIKgu^Xe$Wwh%u*uSu)7?^D5qR@gm0pHHFBo%iMC}fL6lXyfFH5_F_QyGR|sp9wBErJ=|PR(D^$> zOeX;cXVtcK6{rw_h=)Q3h+BYAva##)W5Gc+dFB8GqiVxG%A&)Vo*QRxMyBbsFHY|J z>Ki$0*;eRsOR$^7yT7D^@daFCT8Lg}l*3)j`jvM?RqIl0>ya^X@~X_9heQlI!DB-_ z7Pc5p__3NNmuK7cE`O)GFm6PO8)TQjf1NhF0}WfuE1-P>N}$c5R0*~r2)hEuGrxWV z{!^8ntG<+!tmgZK+`|$-q=@budt7^gdzH>Q(3xr8ttK0nsExN}n``RZ64eY^`Ao`C z{_w(&VEiMfJK1ZJ}&$nxU%Y#*x8JLIV#%#-OI1y@rl8+depbx>kGEZ3Xh~YaT46 zq6oK5=RwxF$RWk3A!PSU%dLCK?=bhx@njHd#qDx0$#IqX+D_Ne#hRiC_t?6svwKkV z;H=VmRc>TgBKma8H?B>|^mCdMYCA)xYuk;3nOgV&^DWGSp zy_VvdfsqSvC;Q6Rs6LMYFX;|KsV{}{$N&1kMT+15ALtQ7ML6;-3#TAC5ah9Xli7YY z^?hrUw^R!_pl=m(pkab3=NkJ8I2WGg?ohdja`<;ZJBi^!mXq8XChVqNQs>>7HEXOx zIg2g>>vm3qOy4XfVi{d9=w`8novGb8O;dDXGTVxquL`6sI z9OdilYQ}C}qy4pyu8O(S{93giiA(MUfD|@W?^KHuN0KMS%mu2CX8pXvLd>{A-#ZEZ zDn581ttiC<6tK55=rk4{RMZr${47pG$Scdw@52k?&evAb7Fq_R0##3P?)2K^w*&%8 zFi!Itnfs&M3!@%~P18(oh)fvE)XWzN2(z(yDo@A1Yep}bc0ChjYDNDmxPm^@7bMBW z+VKFFvpIiO4ewJ&U%s20)b%_zyAko+J3Vb}5kyi8efT7tFA*vTFKd9h2WsiVDK(}X z3eRhRXz68}Ujpqq2v?T*q(5(O=-T5e7)#A5aB){e- zta>ABw`wE~0S6|On|{p&YU2TJ#6gp|5^1OhQ{Z)k#%^=b;^;Il{g3H_@bh!$>f<^8 zZ&5LbN@WTKd$KR-`m*X%_V~HT#;{ zehBEod^Uq*7bjMa=q^#r7(h&h^LJvCM!0&%Ees7dD@_)B8SjT~FF7;P1WDi?4BY1? z94JEwQ@QQ7m=OHGY?_E~;V&Su`xB;N3zIlA{1Iz>zzW%y znfNCLQdLamV?+k=4?3fm!wnK0M5gTL?~IhN$Cn#^$9M+P`7#FN@?E8N3{`U?leEkI zW(hfY64EL`%{S+lM~%Vd42TGbEk*Q z8O`WYc6;5NwH}O;b3iag&)RUD%4NN5i)i$9$8kCt{heKUC?Jw@#Vy@ihk}uj#*)cw z6FWD?D$I_H`r4*EG8C+qd(6Ig0lI_mKr%zZ$jk(O^VWM@Fb51si z^zxtOcg1rJ9E7s#wqb)C0r@Lz$pHbs2qbqR(Cy3wGF?E=L#tGjhd+%NS24u3>U9bU zn$gz}D1~l0pLVSPjQ=7>E|-z(Lo)PjiV z?6A<#_-2jL@fq8hvy|qA>kBuaQczu0W%jZR5@W=Q=`E!h1a9nFvw1kjd@ka#`VM*xYE| zi74(obD36h4_YHpS-M}SUGZU%g^-2S2<)LF-pW8msm4-O^U7dO_yX7Pd@HF{z=}kE zg6$Z6{a3gXC^5}I5eJ{30}h_BySOkeMR-7DEQBAY(-IIGAJ0vm>Uh4A7W}d`h)j|| z`2>*cb{+5(!Fu-p3hcSqVn%;lb$A>!Pc``^&b&r^kTy)sC%2c%=H}I)ymYn2kE?YR z%1NNJI`C_e&(xyfmz44P!eXFIZ+Z8wi=Q9yMmx*Jsl#&P@!TzoB#=~6*r44LXllV_ z4^6Z&oYq2ULh8w`n!)giBr(rIPwB2r-3w)xs|}6+u)+j;dd1=$7_HrCQ%~wPADp2) zt(8iUv>D$NS@G)gqvr&yT1-H=XtP{POmnH+cp91D^=&*ISpXGsSX|t?njZlHL5=GESM_Q4_1z<&OP;)SoJKoOkLRcc7ub={mZ|-% zbNXqp*#kd_f-eQO7$2lXQLMXS25oC+zsz==qew+xzr)FB=d=S8k$}B}S`USWZeRwA zx^p_%j29cwj;^P>q>DBtrq1VRMz?2~i}Y|UlUA>!`s;!vbF}|WrNKII#Kcin)`s#l zQ8F3W5E>ZKIwPPkYH%_K&C;VNXiS%Ikly@#r;S+_4`rI{Q2UP0B{^S)7PCcK4zb9; ze2*V7^GMT}%2DU!Tm>S^Bg&XDho68hJ(p}!!B-riE<`Gr!jTd81w{MN(T8CGSH5^Z zfz)4F+}5?ClDsoWHhy?;Pz*4FOv3GdcY5SL!8+(+m0ZJuYe%JZbsB}$ORX$Mqf-7m=~2v?&ANx0E+2_~vjXT%EEGcxqr%2lYZ^gQ zQC8cL*oc4hJ00xFctzxFz2nPGOfT=vGwCDAnhLi7Jo-z4q;Bil-BE+2cA)KwHk~vc zI=pAw?%1o2FLi@(B$`ZQkh9<-;mK-y&5)DXmJ_epV9D5(E)f;zGqw<6<&qK-5%nBn zGaO#@ABxs!FZR^4r#1+OX9QcfQj+BnB9*KE291E9i+Mel{RYna> ze@=20IE)u##=jUmRN0nhI@#1)kExD1Xz4L*8MduPo=!p_b#k@JSO})W#j%ToZUEeA zt#Uf1V#H^9%zSK;Sl!9=VEoE!ct)?#=&Heb8n9YPW5$0_bz<}=*&NX#6l;3u~avkvs={%n5OQ2Uw{C!pw4 z6mv#+pBP4yp<1e*4OZ0$m>PQ=YoO? ze)(+Q{OO@PmzSa^v07h&uU@RB*$K$hkNU;F z`eW7}Tq3H}EI}!>V}Do&9n9uY<4z07WRq~se*NR=_e--d3uSLivgWCBkO6T{o~QX_ zz540r68p?xY0e$%y7m(BSdQtnFTB^u=nNv1c<8obJ1{JFlp<0f zKt%gu^&OCJfU-`#eSoF1XHqQx_J*acOYyLv>tAsMBt-OSNcC5jXTqKv8qPaE`UU)E zcu(IIU~Qfr4$wUlv`81{#F_=+uV8>wI)+X7rG!NH8@Mfy7;jY{2K7IR(VePu(6Va5 zx`&;P)W_MbZ{)}FQvP|;0?`NZX9eEM?fR9Wfj{b~OS(2*Bt*w1>gSn4nrp+@$o zTc(k^tR&bi_IqE_kJ~yo@;uYo!izqCc}Ub?RsZ-!XTw}ohrQ#NQ>w(OB0_)I25LOU z8}U02HGinA8eSdn=rG0&M2eSPRJ1BBMzvY6jG1sK@|&4ZZVr_&^NCQ8*(C^O^Wfo| z)y6o{s1;3j##Lt@QdJ5TayG=eKIORTcOG>>cI!B$G&$K_OzoRK2=94|`q8-kDogSQ zZ|LI}(J_nt+ZTNh!0RP%}Uxoys>{TKS4b7X*&--!Yup-!Wz0N;fsyerB1j692@bl=|&#P}Og% zWc;ZBGV61@=Al|R;{1(Oh(%EuAG0$yBg2pijKNndSZ1< zIBla73+zu|wAQzbx;6BI&&I$dtwnH+B^sfx#Tt;?K(BH&P+LpPf|x31#t>R{bW3AF zPD!K}GJ>LpuKHY08?u|n=wJsKO6fyfIlm?YGO#}RMzu3`h38`vQq^#`JNI_GU=;|^ zHuk*K)~-tew}jqKw8R5V3ql^qXQrM1Ihy*qyCPzkbYQoX_4q|&BVu{?1?VjV-NKn2 z@;ju{kQ9FuCv7=MSYp&iB`s@nSL7TUP2qJU>7cO_dPK;$H-I)EOAev8zld=I~Z}_KVwJ9Z{nBF_VE3HDU(dRjm$BL(( zZF`_DW{@EVv+t>*lF?yl*n0$!z^B(QK$*I=Sql2p)B?x{1*B4$QVDAeFten#xX#Xr zeDY_VQzi+hdorrkZ_=2oo#BBbyw6q!Nu*v^b&y4o9r~^gWci95&u^?Qhrd}a@tl?}htiaMviwbvIl z5Tu_E@?tgeNM7z7jvR{5+mgPYzgoFF65ngGHk`TvkYa8GV&6RG@fx`z4z_Vgq1iwwU8P8JF63ZGwc$PU)IEbaADsS? z_X4;K;dW%Kqd^}YDIoN=+UjmSNdnprr4yMDmp;ay+E`e&(?qx3z~aRoQ#?}99{#cW z*qowSNrVy8e3%F!}Ae(EAJvf_4nM2 zA4&!_Ji&Fq5MJn@!461K2nUl97dKrm*0pPmDj?Pum`NQDL6OUyBO1(9XDC{qN!j(w z&MyoIp+h>Dj`;ZPlgpP1W2z0ZqCuqe(!4t#+vhq9PDh>)XxO46nA05A3 zCFrxhR6s6Bx`FwOVkLfwWR@<16;RE|H3-p(oVUAbdxQ@}p>ynV!;Bk zvdlkg?eyMCaurT8zfR;{lFe{`0p0a=%7^J6V*(8=^-x)u}fx zzmTD|%0KK_qBi3w7hyvV;ePfqFu}G86V2>bvC}4flJ|SscXRw_RjRFOe)5`fMObHm z`w2ORvcuigMjL^=Ih7y~!GBN%^Y-6K-vM+=&83G;yv$Y952+9cRaye!tJ19o`KTgR z+q-dQG^_z$S?zh_#X;Ke%NwYEYP^!M?Ba-v8`A=xi!^J8w^A#~V+%bJ*-tChrJD8Tjn#;(52(R-5J7 zKDO!{)3&p{8ripspbsbgI>Dg>nA!fv4gT`OKAZA`>BWE;35#|Yi5r4hp32I4tpLIJ zx-j1Go>MT%GLufxl|_bhbk7#9L}|nmIo1&~`u9}Yc^P3f_hqkoT%&=zD!-cHt7VNf z_h*2a7@JTt3aFRtSy4Y>gxlG)%h8&V@|)GY#aCxYca~4+9RelxD;al3Ly)V##D-5d zs3s7z z;E>L%I9K%c9L5%5ga{*g+Uc|6=ja8>4^ERWL+8huwVLmn6b+k?&f%|4)z?;TFDF>_aJ61sA#OW6 zPQn{5z`cO|ARAHq&wzTCQ&(ebopkWKyHB^@6F1|+-I9wR9lFi0NofIYg*as|2J>t^ zv+_eYCj@Gp|1=EP#go?T*u35}34o&+9$ z*SDaK`!46kwr~MzD2?S4Ky`(l5wPsXvj564-yII*6l*NhtY@EGovaqSd9+Fvp^qtY zw7JdzL$nE5@{%6C`M~u9FsQ%Kkq^-G=bvYBNyd25?aD8jGHgSYYK~i2o|WpcCtf7g70shuG}^Yx%pgfamBbZ>i(1!h;O|pMDbRDMXT2W?nf3Vu6`V4hXb*_!>fC z#ArsNpI=ZBnMDWF$q@akj*#<$*%v6~?gzP2zk+$<2Lj}O7;x_a(VK|mWm0FES0inm z?V&7W+*Nu&!%fgv!k+oE9n>ae&pg1N4nW)<1b524f-Gjn&1)q~?4ZKx)sJ7d6PbN8 zj?dUz2kO@i;*V01`KuI>k&%T0x|?~m>Gc+qVTXoiQ+T)FhO(hpkOr#Pj4;IH$czR! z9widnkrWR<|7(5}Gj^btaN&v?SAC)Fc2T~bXSGvfvs+7^L|N~@cE6`IKBJJ3%;BSk zOTX{q%GZXy_qAex^xWwTfx(wlkKWC|h}d*}_AMvW?)kO?a?Q=$>H}R)HjPn@&&`-8 z&&3@Bd|XD|&cGF{FH~>O!L?6atN1O{elO08kM2omhOfD(eCzODARE(_{Kt7I5)eYc zrcnVRE;Tou4tGjb3qQ<86WEztk{GE0#QG=E~*Xn8A7EnB?HoDf=rpsOId4V!OfGqaz z(2jed$IL@;p;=QwuMM<{sX3AFs6{NQ_~d=pL9S^oA(o^?Spk=XSJ^rZqaSG6uB~mg zo@TZFYU)1$9606h+``{U#I?VK>;kbu<~egvqbImrV0Vk;$~xsKvu6E6cKY1;h6Em` zlLk8`3E<^TS5Rx;3|S51M`^~bmb?V2k)JSRT#(C*L>JCjJZr_ry8=`~92|JKJ;KMz zi{$xxL6duc`4U~Iz6HZu^VT$YfiH`o$p1jM9F`B^-|o<`u7l<|Ei|iK8{{73uqmrHvZXn517w zkALdan(nJ>QygnnxDA3HOOw}a8_UB05BN~xf&+CBE6c;W!M*8HwybZ-K0k~Wm$cLN zUWVq~^DG0ULN^D+l^Na-n~Y6pABVUMLS;{%B_J?(AGWQzyZcG$^#}trt#e zJG`Tack$t@*=0*I*~e!AhN3@oL74(b-#j&S`@rDwKnGy!T&{Ks2X!nnXdK)Ah9r*wPXXwg+Uv978{iAjGqY?N+^4FAjiUH>8JPIM3r7_90QgYkYq|ot#1brc3)u*@7 zheZ1*&hh zSXid!HiRu;iZBNdsXGqP^aNs}x*d}J5o>mL&48G44jbCT6u3}|t|Z#QIIwHX^QQ|* zfIKVX3V(`^TuHRIBgi|6SqJAomEfZ=FjpodC56~4NXXdT3EMK-wt2g_zzv0c{*Q1h zi1UHkQ*dzOUz*H6Wb~SLfwX3d^+eZWn$C}o*wH0o;}sFH>jZlOV?O9&izG~qjWgH= z>-myAq$YwPnudB|L<%}W8xJPijea^Pp!f^=z}wu?h3tb#z;J@^Y-eMiuEum`Vr=|c zyh20=ykm3JZPn92LcsM?`Eq^?RD*u|>a$`qF5 zzRJp%Ykt=Qq?XW^2?T>8twWKIsRYhjjvg%h zX2$=AtB8A~vZ{8}@K&Kv9xJK=ggFEYwJQ5Rh#G**dS|M+mgKE}y4kyFw7;^FX$?8? ze}L>W;;OV32GVQdL0hvR^I*Rhd^HC8ES);5*oFC1n*8zBBIn&66FO+Dx<+8?UCKVt zsw^!vj0|SAja<^sEhRZJ#1lIDDRy|fHu&)Sj!UQbUaG{+u=5xxzuL;mvdkvFkv0f@ z@Qvpr6~@yl!~wcY>oF!7Not1aLzP8)s2DlZzljh!#+`}Pbu}3Wx0qdpp!}EFzJKSp zTza7QmA?!pKhg^WZ6xfA7)I>NHx{1eLN|ZJ9@HiJkSPsAqJQ)m|4bMZiNG|OD(g|3 za%EzIb|`2Ts_%JW>@t+;Lk*B0i_|RHcJI@;+w&D85U#c=m{oc|`jB~&;w@Sa&}&y* zWt5`W->7%HLteEk%S6h7^RVGd^@S2gq34eZ<5swg=6Tps?7TK|0{XVtcMS){}~Kbef>TElkT_{b}tY*wZuLhO--sB{qSSG(2Df1Zxi@BcTxue zX$l)WmwjB#5A=-Q&R@Wfgx^7IPmTuJo-hb0pb$X><9sONKZDuoKkV}#CQx7e+g(}4 z98QJuRd zlS9sV`#s|=fwn~HKbS`&6S(omfkfpnsyKTZ%CyCnOfx|^y}Nl{n>;Y!2rrY|%~*^b zqs_IS?Q&t#sz@lz^k!6L9|w*-YT7SQ=wUQrEMELzhxWYHZs|Zt8^118Hz5S~xqq|8 zaH>Ven8wKLd+}Xj5M%T1*}VJStcv&9-phoZ|D(9W2hFt_ zO!2Ll(4~pa)Gwo;?J)!hb6J$S#Div-99h{M>kxPeUPL(gT_`@8YjVz= z8D9L}#ZwyQCwZ0_>3)h&YX4?IwsOs{&1KDfo(}@soI&A16X*ng`w-fPkg$tD;UL^l zRhq?%qRvZ$n5EobhTDVKRJto5|K%ZJ>#fZ48Q^FkS{p^=#{Vng%_DkHRk=&0RSDV9 zyw98bOEZ&TGGl2}$WSR_06x?~+}D`9-mo}&yez#k{)VGEnBKCB*5vk}WX*Bya7F9z zZUUQ>0T&Iu0wzZDg;W0_$&2`1^V!(fCf|#^1tpX~9+B1>cJVV1-Xna3^LpJxMJ0C6 z2Y#nIR<3;VsV__5b05lmi*`=+qjF}np|v`}Q@)!AIOyBBIySHQ03M)0Z8t)Q9ljX< z2no9~5d;x^`a<9PlE++c~k(dDybvk zvL<}DIgcAW|Du2YQ$}$OXVxRGqubAcslwLpO(We{|Pi|?RGYMqBM{70dDLahL- z+AsM(&|U`(_>dbLE4+vV{jS`&O{y3LqpUdB*pmF|a<9+0WY8SBwZExMmLPx@Q8hTQ z{O6+XOKta;#UActQHtppIi@ACLwhanT{ma#u472)uT-k>3pRc8>^~o%a9G{E-$`21 zVC}dtG2Zc^R?QOwP=Vb_AHE)u2NO<(7QP;wNqFBE99mxCP2B{5nT$mcY;cYLN9ieq z`yQ&k%+ zD-8yWvT(e@Sj|N%B3wqWt6{y<|Mad%;=d#sOPLA63 z^*8BurAxG!27j^&0l_e6!IBtYcrjDNQsY+kpsfyg5S!sgBf2d=F&YfET&4=(*^~4- z8jvzu4-cVgXdT@2c!uI`@HSBl}Ja;&j{ zk1=FqS@A2$A{lG-h}Ykb^(gE&TJcr$*RI#~K7%uWh!4j&?_y9`fKKl0XrLsuhXt#Y zg-N2-&gG=QOV?OW-QFW1d(CVvS^K9{s4?Aou%m*! zP=kkHo(HUv1jLq^bdI49kFx*oo%SH$^=~yDo?QlVsDA6hj0bvV@D!2&RmlF@v@SF{ zcu3~6xcelcW$0;jb>hg=qe%{lMp!(U$B zz?r1j4%JTlmCJ=%BK#5i7eZVI7pO=UOBK zwIUAkjb@sERJCfwEE%SNd(9&h6v2GnL6~Ni4Vvs-=bnk~k$KMuCk2$9C=dKp>a+5Orwi%}D(a zUbq}^?>ASKy#})`9@Dx45ocWu4rLJEQ7+mA=J;;6YVEr65}VQwOUOko#~DwSos#M$ z&sy66Ea|notOc91g_eD8EZ|TESC{uo88E#2iU{|1gKQkR&R4g2HQ$}nuqBCf9~(7P z53yvUPk#9D;dQ>u2VomrPoN7kKVR}7gtS7&M4t3VzI>zF`s3tZT`%Q>MFm^oD&qpc zvftfxlYcn6(vA$`DsPlSK9JehCrvF7>eiTr`E6g{mg_YR{E{MPdC*wM#M3qt5I%@@ z{Z`m`9|c*V7A3~U?#Ghor#>5*c2~1buCK{kbhtj-Zx_@^*7CX*8q%RXUWV?HayPs= z7J?2GXb=b2XlNnBYb$XR!jJ&Gp@qzuHFqA-kzD4kW4y3ruN+ggld8R`9Z0k@U7j{) z{uGFx|A1M()%6HuH4b?9W|;!U%%uzANne}S@p#U?8xRG8_woU)>VW5~lrH)67q2Po z?uT85&3C5uxp)+m=P{vjTmQ* zn<2dUO;0daD2y>fYC-i`JTNCA?mHf`3y=?N5VE;2a*AbK3gi=kyMfDI;p zR=8jLO#Zzf{$IC3vBJZD)GxpkZg78h3j>0MBoMWw_?Txpt0c1Nc!W=&lvyK?meTT_ zzlY)7(v<3|_B%MWVP&?QnyUSrZB*_U8ePjr1qr|4=oN@>6SIgl5LBb{ZazZvr^s4y zJ*}@_82OP?^J3HrS2-?qC6xRsMVgoog05|6l)g%u zmLcZv?AhKo3a)p&EO%~v-FHE0W3jmNOFtnwO*0BYGSOO# zc0=~alcG#hG+!Rkyn0#$KWkl0f{W;o28pfjB{!2PqQ(vt=L}9~QNGvqx2x|r7b{v1 zY}|tzV#k9yoc|pq7^Jd-JbD;rv{u}t>5y7}=H^ho@1IbOBHw0~@B>U}+hDm)tt6T6 zACdD5J-YWQ)UZsQxSW>|A=QpoBjX2T0zNIL#c_G{S37UBQtpcGN2SQT@(lL@nbqW@ z1mWP`J?4dNeqHaCb2pAo_RyPpdwEl>w6rb}K~O*7(Pb+icQ~6A0=J+T_h5b5nr+{> zKyFKX>X8f6GmW$=D~BVU*N(y|zaH{B1$r3m-OZbd3*R#2435WDD@C2Yn?62M4S23> z65tWjejvW{rlivMc$vg{SKITzW!Cehw{Ql`56XUP_A6kkHFl+!L{L#uo`_Ukb0HfW zs-fOr(i!1KaGUw^*I^>mu)%s^|18a4yp^kyUFbHmB;$RUV|ZHFR#Ygw9BZ0 zHF+jcJ9a=^X^i~cOTG10rL#l1^V<_Yne$5{UypRv9nl91<+61KUJ*^g?PyDVY6|RP zYY?(rQ)2Rap?)`q=!RHzh*`gReD^^dAD0tj7@TUgHPd-lKLiN<0ww6dbWxIavVnU_ zB)QP?ekJ2u17}X5AJHT8=IU=#M?TsvT!b>5RO=!6dE462CJ$zJGSP7AS9Q(Ed<&y52kM$zt zcG-#iHh6w16-B;;V{Cf)bX=0g>I_0YtA1c;1Pr{-CM`m<)8*%jHc3slm zDN&k=^UvfzL%tEF$nyFtW;>A=b@e_Scz`+O5+6S&qUL31`1J(G^84f*ngt#sZx>u# zRs!Gu%xr0-bzD_FisT)K^_5j4`!7EOlQVE{K3u$J`MG~Ix-C2XZx?M&zll$%^SZHp z`JN-{u@j>xu5rY-A=fLIVfJJE^gu}Wxv;sOe^biEI?h3B(*EFvm|UE=Tx zQgcXtz5n5Vg|U6{-q2#_oL^{2pUUD3gR1bBv80V`X}hz=#NK7HzoiR_+&e0+#1T2{ zt?p3S^Mj=cfBDz?Lz+gjJ1e4jJLD(aRr-L$Wf(BP zLt4!@2fbf+z5D@LHZ6MzA)D&7CDcMFIal5a_PaVkix9gmLZ5NNH#fyI;T2pZ=WlWR z;9of;=a&J6*M=DVq!+|}Np&->HIEjg4er0FMD3>=+?dv3{5PEDZ=w*@cW*|*kXOv5 z@?5To&K{NSbNf{tH^E4yHYGvxoj!(?Sxl$Xh)X+pKG$g0kK+i;gZHebJPeIMyW51W zyq|pa^tk~>LfNhTj{Dq7wRj>5%aVJ0=F**KMI|iWJhE!Rk!@?lnM_k&Xn1;qtM}bL zA^|b4j8cza%3E$&L*Y8QIDQtQTl_WI|XG37x2`yb738T0!t3TWf6KKKGDFOqM*!54$yJIOmG-)9+wdHPy!V^}iDQ zdMDYcfpP`XOz9pR9nR}L{Qz$W*5=ey;s&eLV^-ZGwlmoFUY#2guu-BWS7324jE{N9XadsS z3A<9U$*EsI!*B(jvbXwU{ zYMKAR-Y614aD*}?$qd!&_g#dL3uWATx|}ImTb?X~#Cjq*E)=@_awmiCbK`FRsncix$q_;Z#4>=LJO~C6 zA}8%9v870`Pwd`bKiZ#S&)VbUnnPFC)3F(;{STM@5hrxT$K6VSq>vc_H%f)Ci9bkY z1G75_tq2W0p=Rv6n|v_eSJXd5kBa>Yh5r!p`Dk77oxKizQE$aPaAC%RKuA3h36&;&`e5kNjSW=k zvaejC)zD+gwUL1(1=dU^#FS)>azOTG+3g}A5+4|aTo(vG3Xm)0=tlb=_0yf!NTQP^ z3{*UZi%7o~Nj!R3bOC!vqXbkEhxBE!1~Ezzk-QsQSpcSw?zsWgH>!VG+@Kx+h0niu zHPRZZ4RI;PM%J=he8G*~T6XJ1=J838i3bF0vOQ9b!G3RK`oOfM*!2q7T$Jj#;u?=a&%KrJCLuVW!So|j8H4k#}vcx;M5 z{0#700GY&!17FVWL+4_rFlQ@-i2^B1w6_O&-z) za=(+XbYtQR7ckd{)^v1zYgK|X3abe&Ks7p+Xod<-$oAf6diQ}ZK9F#DHd^b|0 zduaP{U`&qH{e_scV;#)-ACOa;o-AbJ_r)kSpn5qMWG*ZC@3_L zC9uR9RP#MVrBTs#1lBq`1uPs79N3!mMVs3BW1vm7wQp+QtOy*qL;cY4;=V-IZXQZn z&U}R?DqW_~ntxWV0Q(?nQ;%@h2^cWLJl`9E)flyfYyR?qX!@@ZYY=homq}yWI&gTO zs?yC)U#SUXkv&z7*h(Zp0=n>=)JE_?uMudwkloO#5Y-8vYU$#JhKEx&;bic8#R(cV zhj-kYQVCqb?1(PL2NR66t(NpXgcz36vtO)haL2AoOzmrTPQrTYN~O3!?s)=g(i7D2 z(eKES!or#+yg84$QxRF42lRvFr5JJwNGSKioDwh__^o6izKs&@dl zLyPHr>5TnQB6yxAL5XbDjBhCOC`mztb9uxTfsi|d5FNxBn6y_dA|>1W4c6@tYv zy0bB}A@==jS-!hc*Rvt}>a3{bAIc7ER@2$q{(SIOTqHkSY>xLnh0}1axiBCDGl>*X z_ho_d8x90j3IHpa#+%-L{Hfr}{5M*f!rA1;`e6A16?!2ioLvpF_`!c)&T0p;HUhk3 zra@IJL|dxAzyED`6U%3eiJ8uF*KPf++M$=>``wBzRuo8YMfXNQ(qicwOh@fB#P3LQ z2A`C&FeMryH5b4vQp7!-@4gx=-z9uu#Y@J1>KDD?XkJI*TU>!9 zk0>nY?@ltOHE^BB_r^a7rktFaIU%NRJ^vDJ1kFIr!`;I_aO_EgrW|dLdH<6lNaBINYykrjKSqn|Th;dNA1t!c z!KI|GG2!v^{kPpbZRUs-(%`7kyDfB%@xsSNyHne;NQ}7#h(OrFQwf?)2By~;DIVS* z=AsNEeSq?B7sI^Lu#sJh1A%rG2iXk8!ni`GT4#O(-E}BjwK>4rZxJ>Nh}a62eCyU_ z-Ddtz7WTup)8(#v<^f(kl@<1}2}(-JY=&b7UtBOg8lc-}%}5E-p8v8Pg#7nI)jF(+S{w+mOj|b&*BfX@dOSP1t6{d8VxqaWNn3^ILD1eISs`|I4vdJ! za;T1YBH5>Q@8IQZuPH-wbbCuaWBI=aIoRIAP0?sgcnCT{J66w>byOIclnrc*lZ`nj zb+AJ!icpIB9>r9Dv=!*E1Vc$VBO(yp74$gy|LrFH_3;yUqG0SV_+O(Lg(HsS!Le0- zOWL1wi``bx#%b7|HBu`{^mx5k*|7b-rMKJjjG1td&jI2X^A(D!^`3=$$qkp3*G;+{ zxXQ;co+RhARnm|Ss^iqx{8NU$KEkDsdc-6wm;0la6S0b4_|a`&pLDo-nMn2p$#^sF z?b}nXaEzx@^NE^$9SOs~jJ)ERmj5J8FB?SP0>|)tlxOW7%V0foGoX|Mj4C=7zoO}* zfqFZk0xGTIK>*FBQQ{U}xNLgucl5E^9r&c5uWui3j|(taSpio&9IXe_K_&{yqb)|lD4ZlHiIVmDAOBd3}C0T_% zTFVHujGU{r#Epi)C&R$no$ON#Mw*_I8^lB~TX3-c`y&HM5i{N4&8cYGSXkW4O(M}A9A;&pl}#W!oG$t$yO%Hhbw;=>X`LC9Z~eA z7;FQp!pYS~Zg$qd1qX7+OF-s*TwH;!Oi%rn12st^MTQoofDBp#4?lX!z&(s-f}8>? zop@SMGyBJxc9j+RkBgYJbC66m!`Z%l#M4qJA}2e0ERb)NP@Lte0louNoz2`_1rSO>6iB>-^8$z(;iQm_K5RHpf(~eKieG&A_z_w3H``5S z|2b+-)a*d-#xi(#Agaf|hW{nt3`pJIbpf8`VF*9yDdJn?4TX8r)sDI6XXaNP$iNm{)s_siy3y zolp~@l+xGq4WIf;T3TtQf&Y6M^_u20&#RYLoiHY=`Z(#|@z8*R3MGDt&$#lK0{(IN zRuI)dL_FrVuLvXL==r$hSdz~JlrzzdxjjTaG|G3iqxQ+l`x+(C(5Mpo$S@BQ{jlnF z9<6s;6Sbcwt!&m|(#jjl?hU9vcwuz7tZ3Wy$pyQwHGy2XQtg;lp-3$FYJ+KsO1rVN zn7*iyTu4N(!I6$$LFeCJ7>Jf22})G{uQgL51Z;W3(pAwEWRQfEzd-w@yR2-?{=LRa z5{-!8E!>*6Ys*wCi+Wrz16rqUG*py=vPoadzc{ke3iV(0Qjc^pT^K6L7d7dXS%_Ny zMwe9$#o?vEbX*QT*@asX$Yiv-WdmJ9Vj)wbIih?Q1@f1E1L-g%wj4H)78EHs1?xTz zc}`3(C}`M?w@}3*F=hMei-uK%vk7s7TCXrYW%_4ueEL57^(|p4v23e9cP8qNHyDoM zB43w{!uPG;j?vT8R|yVX|Mx;+M*VtNz+B?>8BH7K%ZmFjrz9A*!n2fuXsDB(cfB*j73M@4vs4m4OD{x%t(9PJ&fBpD#LQ_`6-Bm;biC|e2<>z^kA_fA` z3;uJU`Pi$}(*K@j_M|Y1$@M#`AldXx#BMm#z3N*}%bwS=P?Ve=>~8yRE6+_~JLOle+dScebDUymxm0;trAc)cms`>=w6Z#QUucJ( z9{=b5O{eN2$Qp4jb#D(({3!*v=zLZMA~jMPrrcV@r|ce$f4)a8fc+w}y)op4MZ7AU zt3lt~hDqCZLM%T|tP2Uq)l;4(N}03oyQRI0-vNZ zB|AI&32=VzMB88H{M<`An~j=@-`k8H@wD=~vQYDVTbXKsQp6u}AhTG8GtH*#FIuBk zlcAeo_fEGkx}#>UIH(yH6dP8tE=s)EIQ|0p*i`KkmfBo);?)jm=|=zDzKKD-y(Aw! zb#>`uA)o02BTb7C_rp`LgkXVWEBpxet5ASo*w{Ol!yn;^g%^u=ZMWo4VRZAi)s8~# zJRW3I%6GWm(Ul)c$v;!W`xtOj@5iorKSlHuUTJ|a+4H$1pofPuN55-XKQwK5Io0~h z+3gus-L+wuEj?njRAY3$NDO9fTIN0ZDbYmo&4-(GL(z;)74=3nr1}6|T_Vem9;wtG z%$PyFC|(~!sT+k3AmO~ zHW`b;-#N2V7q4f#f0vwvD8HMbMAXCRtNAlx)1SW6<@Vw)k9!Ip{6%J9+2UU(3cT1w z$pc$YEZ(QW`Xhj$DwS$$^DEX{QQ<8oscdl)405S`%o&`+(6zR$NCdhja~kIEPDxV_q`^n3qP{|rp6$S{tT39s7X|YcXU=@Ruwu6H#Kv(uP>5Yuhm>O+vU|0wch9q z5Cp)QeWZaw!lT?MW@bhm*3?M(xCkOfEfx1;O*2`$ahk*v`fo2vLihgiBivKSL1!4X z56$mGx_aQ_%M|>=v*imSQl2u$Yg`OeP#W1f{pBtYCG)P%jf}*z*$;d z7_cowPSim>TFm!2C3ZALf#8Mar>RiDU7G@5g;TmPAZE6A3ZV_#f!fnWgdAmhS}Jw3 z%j3sHjIlL-XrGEDsFsZp|NYh}u$^wuG~hqG;Znjsi+&8)US9K|P|S|*-90CC@WADN zKa|QPr>D6UyZJRk*PI5AcgcZ_FhzQ4DzyPe`7c(Y@M=Mt!#Jr#rJ-h^tZ}p3^q7SAo~J)7pn=p`R;deUpr&^dF_e z)jZS#S#@fD+TG4EhM3GxLjq+l)5JN8#ZFs({@SspLnSeWc#jI4Cj?L&d_yLocny|Y z@;d)Uwg(Ff>>nmgs)Fgj_`ihGViT&_#*Gd;5oKjCd zPE{NZK;Sc=E4z9=X^wgE`TwVpB|ulEb378w^FQ)*SF*`M3bp0+IkU6xA-$WCU^0pp zm9qBLChx0oPG0xrw>YJFM7R&+HNSx8MJ9rftpKDgjRPng5~TferKM;=ys<_c{kH}w z3lCN+uBxiKyaWnNjzjBT`)yA~299UDs}e?k-Fd(;Im+#5FAl48n~gwn7~g(p1avcnhS_ zkf_8DsNkN1|E*%QUFgawkCXRtq!kpVqCd%m%a@M_eOOc?8x5*pT|q|A0ltso<7f6* zil@B28df9M)BD-?#D59eC+#Wy_~8P)ccUs1e<}5Pl+6aM>RG`=0pM0!=X8&+!sa-E zCbeLD6LRDE?5uXH*biQH=~$EMSJYzF4*VIuhJ)MTSBTxIGEJ??3192;k6z?-wmKpq zA=T3mY!XN{6j`;`uZ3^x6&k9Ex*!SuZK0!_|B!(`_S-G>4a%`cnX ze6Llm+=iYcIQ@z2_ALKgKXbM<3aw0R?n17>L%3Yam3EdJ)X?WTyj>LwZ!>+y6|ij# zKiLpZ?R2Jk?z0a;Z#i{GCw&)AmU8*hTxMEk>emOcpO3mBOhYWpGccy1H;u`?3^Np9 zmVJ$oO!q$f(J4FAXW6tS<-qT9C>6&C)F6jAA4tne<-*Gr(toG!&bepT5=@Z0PX8)bld=_i zySMz+*Y;Ay(7S>Nik*}Nb`H%uv%U^0p$+^fC<^+6?czD3UyCaNgc@$L{Ylqz!tMEZRGF+jBC!Q+Kj@at$ybT zc=_8gFLDkjg>Fyc8xls^9exeaQ)j%B)thF^`TbiIK5P1d%W84tchl@U@Aps3_0C?J z7tpB1YcBQ8a0jZ6LbF?(%Zty)TKi`zPnt3_D%(G4`twPUVzBkJLUw6GHtkW+`UZwX zY4{@r7?|JQTMYQ^)o*$K`}q_tBa^fSmn=t?+`^6j(sM)&I`a=crKRpl7g#Pr9|Qf% zqUa^sV_@2+k&u#7Q&z^}YAZnN$0UK7e>}4evu;@$FNX{8+cFMJZ`OJ2DAwPO3f;?l zEJ2bTj4I3!HAI9p4L2E9h*xf^e`QZN%wMx{*(`oty)z_D^sG@uWaAs4a=C{0Rq&TJ zvEaVuKzNE$RGE)rrKtP1+}rXc>CLUtk#dLB0_KkRjF9UaFWLPuA!_QMeW-QKRKDS; zx9e*v@$+uPC8WW>kAt-wbAMCDLvGG%GYebEuD?89x~ZCe{E0M8Yn*R%IqRbdyWiHq z!r!#+zx6`Hd0K+=(g-4FDG$)}mDim*L57gxu`n*jg|xX-^nYVB(M+g)Ck-~Vy}CVv1UH9BC5 z(OGt`FuT?$)!fM>dztP3rUCEPy?x`RijvQR&E=J^uNcKT4_RKa&aeH|~0<-)C~QIsJ?{xl(AQmmPS0 z`d-AJ)sJ`IP~non+xN{nyR5=nAEGS(rak)anM*T;osK+OlR6cK;o# zyA%L0KdHcGXU@t*MI}aDk<Ccphj#X|p@PQ2QB2wy%69W|6E;WC`^Fg1yV zl-Kl&N!Tkgeer@Miu;8{xZQUcL!ZurtniJG(gO2+SBPH%)H3>dM-hDt;PlqJ_TyJX<-SMT4g>u) zr#TFo%&3?Lb9wc_C_5?fAO_Lr(~tYq@5FO_(N{g=moTry>CZ8V3c~*E+ASzWm-*R? zy;skqHPHu^Y*XIa6TO_G|KZK@Uu?x3_w|ZM;!={Hx*8(TeasIa&+~gp~Mg9%(X;R_K5)jiJ>B(l6*i-)Nas)V(}CmG_Hyc%h!6HpY6-(AfAYyaje{j@s;`)>DtLu-GJi z6>=~OW0Lab`CL)L?vVQHc-{AkeA1ax{(+_dDpZA1eP_#F{9*)GW62j`C#Nd!x6~ru zcFT>E;$i!9vX{21oq`aDrBn{K=9WP@o>QV>CW{;Zgj#CVi2n@2jA_x~v4|z8Jijvm-7dsij zPJXtuXnhq{M9}7CTt8;lj&+kipo)&L2grx`FOvvCU{^f8i2B#vUPJcK7n!hre8BjY;wt2a78Di8xp*+qX8V-agF!<_EH5@o8B-^oj zqsD;(sVLi|+kz#uyK(iC<9K{bT+2h{-8kAUAPmo=jk}HeJOeho-<%IW{ighgfyoh&o<|jBG~c1`Jj_Wo*3p z#J%19K$|Y$FsY$~D(#d-+pJsK`u6e#=q`I`C|0<=uD-S_h~|hDuZRWP5xD=T{n5`h zo8GN|;5ae+9v`f+heu}TJ$r`H`1RH_o=~}Q`3HV})ed$cr10)!XrZ)kV#4j7f#{sb z?Mal6F&UTbM0^gMt5B2lTGh-B7r2AzxU9y5Xxr;mho7;{o3{Q<_Bz?vCg42JrJE8; z@VIN&mEkvUAu};_A!p^262{n3bz@LTT$Uyacb2%FUxWL3S-)~?@iknGNN2kOe^l43>|_)LD2w3IGX)#o%t?bl=mizXLoe zt#;@@jm;pzyx$FSz3mG+Ek6CTGfuyaNRrYcAaTgbm5P}+I2Mv?RL-qX*duH7Se-Tc zTui@0GALcg=fN2pr<9%>%2S(P$P-d-6x!Lq6o+Vq4^Bfv{{3Q;b6tT9?v>hQMHPYX z=xptn@GaQ?Bz1}}_X|!NDn%zg8Yf3`e0EMC5m8=6k=R>p)y4PF+GYlN%6$CPN02pp z=%qzoGE*T)s^1B1^J_1@T~$=Cv`3o%`$LmNQ{sRP9@!+|wEE~a*Yg=g5`RLfZIZ7j zk|R+e3rfX7jFOJDbm+%JaU-Ex9Bo^-;<4i1=O2Zn-)EWcoysZHDQIq@q@JGL5`@KL z>A6konHE?7hAHPHEPbkO*k;m3MMuw>4xf>Bwqcx7Pe?zb)MeB6YqB4a=oQeZpN@>r>yRJ+DYG`n-=8_Db1Y*KLWMDLe_)nhmBoF*;{+Bv z9d;5-bje|A_of488f@NHn%0bf`klR80&$@PekFs^WlhWq2=i-KtZu%g`-6no?p=K= zrBg87W8Rd&T4Xa=YGKU&t!S_!$VskLkinp$)C(WD&B*wXcmP&L$LfA#^k>b*y2LIR>86BsofqxN&yhT%2L}d95-tkHJn$tjtB4%&&U!a1UMMbODS#N-{FTM*$5=H0kFdN)SfM_2qZtp-`yN5 zT~-YyR5ouC9Wi_oi7+hBNVSbW81Ccln<-!`1)Zq((oafWr@`V6kgwMg4dTk&qx5?T@m zmKRC=hkF2R1ukHHF>>Nfb*e5%21MaGHzVGSrZAcf%V3A!U`xsPInoEw{TQ!bX5HjC zI3}IX$r%U{-W3~FKQw{jf|^N)>L=ZnsB2{pQU1C>1ySUl#uCR1b~Z-;Y2yF^wHnsV zE-=}`nJ4+|q6KTPx4-irwtl-qsWJr7Jj@=@7jb5iC z!6%Ewp#E134=J&4xgQyPHc}wPXJWDZS(zlXbh>tHGc;O;r=~gcm7vg{OWYKbla{XfJny~?UTgOU+Q`O0IFB4PQho-i-!Z|ZCNXC z$zF&H9PY{O$XI`z4d}h&ktaS&BU{~Y)(@EC7XvrNfSB+{jJU*9^V7;5`;K9cCn!bj zJX?gSIO-nfgO@l>&Jh)OGj&uMJZ(P^eU?JX`+FzMvR~lhnmR{%q|;Fd(cu}XzTS#PC$pF_c(oqmq%IH z!YZ^L1qN^&`w8L$U0|@wmMBwAlhi6gEc;^t8AxCNPbpOFpk&~AW`+-ad)r68K@6)WX;7iY>vV> zm@7K7`E{oY=i9)zg2e8S5sYFoCjwH(A}Qf3Zk=unYwvB{`4H{7NV<#8Wti#1>PH=g%vE+Xm%tz;$ydo6B~(h#+&?uc{O5k#D!D&&1Dau>y0^ zYXm~NRZaA@te6g=W1?Dh(^`HDeM+3wP=@J%%}6h3UY*zxYs{xBYc*{Ll-P$amI|$A!UKv1hqWRY?V0pEs}a?I zd`F>Obw?HtrPL|bnWlt&A<7BR@zcM&uy0(FH1a#ezXe{UdCu7S+YzainKL#^1ZUwG zve}Xh#A{>YH9XBiGOiMG8G{y{Erh)1eJD|ZH%CUAE;i3u@nsV>!bLX(a}<|U^_ArnMr;6Ne2;Wj~Abpj9 zDgoMO*x`L@GgVHcM?D=#vkvr$#SFaN%2lhuN4TtyK>Mv#eu))jcITL8u7PxvT~r;rxfS=4ylkrNCkancVluNO!jww3@p}hzROi!o&`z0e;}n&bHQ}S&W|bKIeZ4;e z4OHO&2vvEPnRSzq2#4E$AS% zia+BQh`SZy4tUn#1*_-ms!$dzkjH|zf~%we(R{R7mcNHw7>)eU4N87LWYAE2VV}KJ z{LGGo^U*4Ls#PIFFd#^7bu^s*^BoZ%o5Axy`j+(6DwT?v`r780HFOz4FL~=vZ)v1hx_*EaP=(MRGNc`SE6rgk{>>+& zRD+6M*i^jOnQzGqoVdT?01W}u~2&tOKK*XEb}`o6V{0A{VmVA|aDVZPJ8u~P>rj7=gO&s5L4 z--tXmDp8hJn{I@EZf`F?#!>Q$#gVZ6Qu8bsrCt;dFhB8EvAjTt{K)KYY^DZm6){` zcGJkr*;=$=;pIF@@NlSi=RSvGHl@N^nfr0U@pU3|Dwbe9#g8r!V;a|RA4h}mp55st*L z+D{r5({04c5)RGtVvR?<=dPRSjPy)*9^MaDT?E0gKeUz7BMZ07&f!b~Q**BC0Y%R2 zO9MnPFHMGt6UOLsvx(!72&WDZSISljKauV>_P+7o%*0vSKA!Q~7Z%q3-4lJGWZ%BM ze8WUzG`O+-zKQg=FG%ok`g3A5C5x_adlJ{pqxl!xk)T-NdZE}@Ox}1tr7_a zU9q_`24s2_9wPw`leaK0uJ=Kv!pERgu;;nVp@keojoj6PllS_OpC#bsYf|hXuGOp^ z=GMIx7ax~pZp(c}-Z$@U?`XU@I(4-J$E=#3QpK9*q- ztJyunt<4$%C;r7;q{ev5Oou$R3qg&!&xdVVYlHM4^#OQrVCkm&Df*E$4NNs$ofLLrMm){Axrh=6jP-UtYkQI`U*k5Cd??_u-OWwH(? z8F*`Vq1t#()CQ%wJ2au;XY+m&7QQ6pRn@`3%oG>Tmph2Bsvo|yE-s`UNL*F>+=foNDW5VmX!x>>lP4No|q@glnJmWZW zcnaeraDDa4J~jZW;JzokF$-0Tmd-#v*G2yz}D{w zkHg~XSA){S)CNX3W#|lizsHu(E@X5S?x}MnkWs2#Wnn&!@jCA{PTqF6rp*qXswi0V zNgUej=OVA`4Fw&gJfN06oS1Recvi0U&h3}CPyYG#fg%p;>|J`{o$H~^-1R(>H38-e z2PE_HjpN<#?ijVVu{x0-HvtK@30cKjDGzfa;Fa5VQleL~yK8JgQQ;ctzV6gt+AHN< z^Nzn(Pm}nLt`=M8HYL{8hY(4UaehtAD7bC=X{pHfgBWfBj<5AK+A`Xn$fj@4ps(o3 zNtZ4>vN9$a)xO)LUubkY`H5|7_A=T@J`>8l{(Z+n2#l76uAdMb1@KLT0w6+jzWXrucPuY%^A5Rmo}E=3zaDfKX1eK7;NR9WNkJ~~|u-OjWZC*l6&!(x8%H84skX)auEwJm1}f~(|4;qOmX)1|wBv%##San8Hk9hn&C zrEvt|vkC=qxeB|d7ID#W^8RQaK9-#{A~av7@iBafCBE2z&3?Q%igl<{7c8L=rDdSW z{T7m^sV~8lCS)A)6wCsTjL+qSpFj7BkU@2WwDY@)swMmVU3vdI#M`wR<4Iq!@~a2+ zBAl7v@7I;&H??ZvgsJMJK8;$?U z5cvkkhePNZmXGG@WOd#jesHcmMpQl0ulxL}d|8sSjlXCl^J=ox4C#ffuFgPlK7s0i z2I`mLuNg+x-_;aZ(1}@RLij(bLMscxV>FV%E%d@K5eanDekZoMPxn0HVVtH09zi-n zt$LFfXufPhLIwf{a>Q-+;hNkI8&bD+y)T0b^uOgHFDR_YtjYs7#hc!PVW9+@lhIQ4 z%KAsMUlj=*#rNGwhT zzZgIcRQP(~w-Hm{*#8i{p_R}sYBVeO@Qt7HZQAc?)%4>NK01Edw*vJ5(9B-%8%b9_ zV|8T+WwF-8$IKK}m%j_f?)W2csM8X3MHBM#nCLEq%eyVT^}el0eNnLvx70dtDs!jr z?^xB358dt~(of(#7uJ5Wg|BvV1ElGSEiP1vqPkKvZ_#R=GZ~A1xA68CZ)>!0%8$HE zGWG@eye^ELQ3wbk`KvBRbf231raq$x7V=Tj1O;MYlwx+efPer*o{x&d@wRKPY>td5;?pl`M2JSmw*Hi{zi! zcF8Zm?01Gin~a?NkS>I2VwawSp`DoSjL?5syL+7L*EaOiE7V@-Ph(AOi4#~bA6G`# zmLGf8LN_V`NY51G?4qv3_{K-E60Ws5`7M8rFfWNB^2OogCBmh1pKU|4KBaqKY1wNo zl6}4ac|Z~%cd4$>bu$f6+LY3cdqOFlOs=)}cuY#WCgS5iG(MeOH)QP4&=e+IJ+wNm zdu_JNnz$ord~}1>L;T+1Xa?Uw{fC>%mnxO8r23ZhrYIege(W_hmh5KX3Cq+ap*8C5 zOP&|^@pp*#?Q^li!*Yq*Ic@IhwM1h^CzYR9Of9Wyngfpgb8~`goxgubS)K7dL%PJ9 zO3Vhs|67aLKBBlxhr%~0@>8WS*9L&6vsY}t%t#H-1!llrlCMp}`4y-X^?rY3uG-?eG~ zgyj-Q%=Xit9R+TWCNES~)t{W$H3PFJ^T|S`TH~Gk~Td>$$pp3e)#;xVyQ?juwz zak>2!3mfOQ^IAWaOL>+qn(tSH-Kh7z##+@+Z=ZIoTVfv5l=#@<`{VTELm>g;pkb#*%@AOTDW70zwNVdZh6u! zj<pSRE}s|2$ysHsrNMRTER5$m*^HC8YztoZ7hzxpzbu$BgA8z>*E1 zVd@ohGz`(>-DPMP0prD;vxJz4Eein*1I5|Q`MhwAx!^i!I<$hQ&4Kd>$8lWvV^>Vg4*YY^jF0@*WO-L%*>lwUX9?^XGV zqg7;)im~Js+S>y@w0iL68L(*i^l%oqW&S#(1|a!o$Qgdc^)Lq9!g9Dn^OH=|#6k2H5ZYDhb()A)A7jYq^#B6p0 zAU6(^I)=4pRp4l_smR@*DoJ$l9y{RBGPA^r*ZsO|6v0lH(Ga81)di6_h`>m>1w~O! ze~KdQxIYUGgfC=g6=p48P*5+_&WODR*Aoj9G#c4+dYb-vHMrffQj&ShE~36tu@{{V zN(!jYC(WnDK0ibLmEd5wZL2+-EV?i6z1tO7V172-VLo@T$Z>goh>2!eFUQ|F#NY0v zZ#LTFZH3s$=Kdo9vG^1bGsVaVm_5DYp5uzT+T7b6J=CwFSsn|?7u}l{h zy%~b&@ni_1(hwpd;y3&@ZKF$}?yi8DSq%g4y!PLEb(UDGO@DxMRlVgv%iqpuh7$Cp zKLKDq4ybsO#DhRFJD&xI-=B)S|MuQ|MOGKPl_y}HYi)dGNFTCs4VId@{dacEN2aJQ zO1{LK<;0==-}kgY4ieqj30rQY|A|dNy%B z$@=~cy0_g>82{Qx0fhe5q%(Fv?b|XV)3p;b!yM)*AGgViyR42 zV^G-XQhOmk(ZAxqTz}=oEF9IXP64k>ANTYfY8Q}rv0EVB??2e-D|ASsnVoA8$$p7I z0}(-A*A95H5!tuut_Q!XEMsy(_Uo^$k(8-}4aMB-mouS&Gd{O=$jsm}I4>!H`&6Ng z>X%I}s9*KiTIhuxnVOU2lBYw;2Ma%_){PrjYrDLZOxlP7)}NY`ArMUl$bBc1z=`Pw z849kJ)o&ogf$BsvkLop5q6+@^c94=xHw8-9KX4;|GsL)PzSYT*uc>1JF?q zDjeWm00bTflRO4)`x`cfLwliG5MoWo(4VsrK#r+aA)YsezmAQal(gpcwQ11d1|fhZ zWaQI=?%8E6@ywh`)3e~+-M51v`esUTrS7ZmvOeij+_9Qey5Oc~>owA8PFKfaL5-6{ z8@Kym-3&(vNE=DSO9VN$WllgcRHo1LyCiY`FU&->LI*c^jl4@&Q~&Mxl1FgZ;hTW4 zjjP?TA$Py44#G+4I0o2YV#xv3P0kzyt7DnnKpj#no5`#8T z654}b>wVDqUncY%+{qd;^E3qYZW`<&Sh#qbjXrBUp#{`nfY4Z1baDWo7#bAl);DMl zkH2|j)b*=Hndk;1Tlr)23MAi|baPRsT$F4|j?8fv%t5ctxkkK()uCaEy2ND-CM{6N zpwbNd0kr~BAT{>G&V;)6!L-xih>aE;p{P7UeueASn`nM)?1wR;DlU!&!#!R7MC<_!y0qpzm{4rdxV9aOWW8&fkQb2sz}r{J$bN8ppy49aG-( z(A~(qp%%CS*~T231zrSSpe0oN$V}p+YR^xzHH+qv2Da=s&(-#%-B{xN6!+dssyk?} zN5|((DM5z>v?1XzsXh^|_gD%y@|3i2(tF-HWK61p{Fm#Knq(; zO2SZK$HieiD^<_{Fnx!*g2{?oQY|E=|-)kM0*H~{MS+rEa#(E z@q-cIWn5_grb^zxz`$}V$2Lz3@A;FW?;%T=ACoQ%zV=sGH~-K#-GEZS-F7*xQ!FV9 zNL*^R(y!qmtFIF-OcY3>B$|}@6~s{eHY=kJCasi8C#@qm%DviU1Q1br;gW>oQT=3F zQ5C?qNpIiEaB$~mWwM355zE6YMIDjBD+jC(_Kr>{=Nzme2iYioZKuO<)p z?0HbtH_27`R|1-e=b$4%Qr-+&^e%%DR*MFu2F?}E7KsjF~hzcp@N-fw3wqzCM*)pu04?rYdp zXeoDVc`0;#wU~&ZwUV{I!51oUs`)YLHLK!0CVxI{zRXh_G;%s=0-4L)r2Nlx7qd`% zkT6XG%>EMm8({$#xVcMhgR>FSzq36c3J^dd}f9*Ib=~ z2r66O`IJ9o3ybtH;`C-D&ZsQh?_ro^gnjx;b`KX6s#8zZh{nVdhbN2yCMu2~Y=IzK z4JeQaQJc6-We^h$-0?PiL`A>{EQw(p=B9Kmwi@QNbTbztJLekOBv({a%tSs?n|QR% z0ZPg1ao!e-@98i^!9NG-iPAaim;wU>Et$eV2!}|?It+N6a-C@izn{eI3)IfElZ3I* zQMoB8L&OWnV;uJQA6P7y;xUtL^Ie}>KYPJ=?Iz4f&b~y$#*weOG%k|QjtxtNqY6*JI~gFbKpXrKJ|q@k`EKt7PtQ<$9a{|5gU_f7HpLL8MH$fpwRktg1aOa}iwIL4i|)Ajfv&DIYn!uG$GF63+2HR~ zvd_oUA%y+ck$(4`RP<~^HRw$l{QB;_EA@X%9deBR;o-^r<8_?eczR`!F_N7~yeATT zD#`FgGC5$e8DZ&z-AU@@oTBZ>=tZ;!=#NqXe7Q-yCYSeLhOLtelp7ZU2`f z1gRTvp=C!xQWzr=yWq;kGcAZ=0E*q98+6v9D6 zD+P3EZz=1?kd3qOw@YW@GP27_8vLe5w8>qjaxLaw;rx9^Hd38U1&rKq-(`(qut@#j z;D8fC0jO`ZoohtY)g{+&INN$tN7T~ACWuC)>krWT_?deUF|`cD1^ym)M9??>Bgwz@ zH0_a5&bvtIg~P9LS8ze`u4oTaR8%xa0xe>s-9#4H{imyW@pNtP{^q>7ud(_Jags(H z+Vb(TMu1s&`LbKTOzg>NDfNw8 zj5;8JZMO3Ne75m#8}%SmBDIF;NVL*NLq0p*ODu@p=AlNhQDfrgFkGz9^~3EdR-D4cW7ct6zt^ zn`Y{6WY5&x+!}580R3F9^AfRByY+|iH$=I>=iAVKUvRlX=MNZ;pQWh9{lgW6UepiO zf5>^gts;9^(9qB*GRD?Oj#Xdye^YkIS&KoBi6u-^iyJezzr`n1;FY}GdyC3X4CXac zAgEI62;7yn18M>Gl3+NgWOk7TH2&6zv$R^OQLzdckq=31u_I%o1#xZ}N+a*%;7SDJ zGOX*-FE37WxxD8l{PXh1+|qcRYpZ|^IIkD?oq|FEu*;tScv$Zk0UjYlWN)Ync?f&) z>+UTt{Q+RUGfK%fbb+mYohfg6*vN}HlOH{bHbj>p?C3J_QhZ?Wg`%urYad+LTj2*VnDLHt=`MR5Wn0mImi_NWEa0q)rkb+<2=^2c;?pdkL9b6N@vq(Xbxx zzpq|`A}T&U@7;VKvm$+tAH$JsN1BA00&({5XNRq`#!#KIToZ{W2W!3onAa^}Efz4c{=#!0x>08+!f0s#=^}?{{ zD}@M94(0%pmU$ZK@zI@6I7QZTwOO^PPJ`!6nV$~$TV!9nM9H6DCVFr0)nWXAxJm6M z5*_xsL8aiE;JHIro9Fc}kejqqKW`IT(lq?Hw8Vs#|721))D!D04g&--4awN0m1}do zvy&PUz?tPHKME@|UL>y15Jn@yPfFJh7ZV4Yk25-k%n?+^1<%&y|6UTX=)!VNr0M!w ziXw;wuYNgRJLfwO2^GM$aQ&>170TQ0b;-{Sc%R+WrK6y5j{RfAvE1#2f3gxGJ- zTSGYw#NNLMVP{|ynxeq(Faa34YO*~*`RoDW2#S<)By5WAFfJiX`tgy(wI1X6@XsFgXegkLXNgrRFQ(&Yh z$B>`({&Jpi@9py}DDh`*G+}5esuA#Xb4K&!dVKDQLw}!m7n&iCTbCYNdSwY(=R20Q zjTPIT5Uh)C4243Ki;F1*<2bK1g0iPG4H*IN^z3iFiZjsyG==py5*4MF?l<3_X1BJs z5~flk!R@i~1I{F|4XB6;KG1=(#_^x$11PUapj*ALl~t1z;BVXq4PNzREKnCS)}K;e z?2psXldCzWM1GCP(aze+WeJ4;P@({80oZf-Km?MeZNN)=2kCMc(XygiqTY{g2pauX z)NTQFC2I`oLQoe1v~?nro(#2oDr!*eA&8lhhJ;Ez0HoRpc#2J#-jDJf*V?Dd@V*9N zv^WJLZv<9hoZE}=Igh5}URlp6aPkY#lz${YD>1E(n!=+COr5l%8>Ck)>3(fj6!gmt zC}gd6PmWU``2?a&eQK#{S>&4x_3wc%FggW*lFobTEuS%q8!zLNC;g?(xM*2(u@<{F zhXq+%n+Yx9JGor#kbUmS9JiYw3P~_C$S&!rZ){8|Z@;yak&^nkXEvHD4MwQcvvaC^ z=9Qlm^^NCn6Lsgr09AedE771?Q_=yhTy6$uQ%6n)e{Z2uj;bveMHEP#R;lTXB;-p_ zR#q+qwiM>0|C&z<=o+e?sEKW9cIX*0zm%7xVCwQUyZ*tn_XSY*(A6ljSxn6ATb#ts zgcd6|oHd@^$Q*V*hm7r{VZP2&=92VT=~3(v%wlhS1N({RSK|EN#&5yjuU|jhgj+I# zVLTx&)>QrEs|@inRI~jgyRpXnUpLW@~P7t(P{H@jv6eI$G3ufIT0G zr7kHyRg~&l^yv-5b$Ymr5NL?wLA3cRk1Q00A%LxYVRT-!kl~l=e`6fXqaLT}zGifu zVx|;iea8C0J?((6l+YKL^5QQbssP1I56HziO@qc{0vkeX2HcQ4z#(qBdTSODN z5qWEK(;|W4t9Si}5j;p=2Q(QO9NA$;H!OcEBo;WswrBeao&%)#T>EmD_;ooYcT1uw zA0EZ4>=~`6>n%naxSR`7FQv_1S@fBH4a#?(YmvKnX2{GM96i4(>t`hDh z9_)*arCk-pOH*R&v|VhDCnF;(1Xk}RX{nf1Rb7~-Md|mz^uS~4OG#*Ouu@qZ>*+AX zP*Ou4ql8W9J-o*m0CXnkD_TX#jyu6q+TBXA%}?WV-Q(Qm{aX*nVi+PRCbskQ^W>mK zy%)v8N-7iS;r+2cmA|pR-CbUfT!vpBKSYO#le7*uxJ_R|Obp0fD}$(hp@a(qqp7S~ zD4bTmqt@2f5pgwE)q}f{rQSqkSjby!8y4y3HnYq;hY{DvT7W zR}qjb&bCWoGkxI=%nJkF!~m=?Vcl`L5l`m+MdslExQ zpQ0xroyW%jey(_UHnc7LgV+_7#ytz_3%rwyOhT7z%)DrDF*RUP7ro=uul^xGl#ux! zS`C^Zi0moH9-G=k>^~}i*Qy^3dt84;4?TdQC|$+l*3!^{a`}33Xti$BECs}JklI9d zA#Hz=>-2+*!9t41w`u(TgKqcF>H3^ZX7}3!ZDwX>-u9b9bh551Xx;AHz-r9H0D@VU zuaEl1>OR>%z{`rm?ZlAR3XQeBIhVm~z*y=Um3?t?rnEC|ug{1#3nfb40nFcAi!!?@F`BA#i{MKN& zJNtQQb(hin^rURGBWNt=8zaKa7Z&<31Jjt^yxLQD&DVm!a|mkT5k zN%pO|fUl~n4+H#VYZQIv`0$DL&|T8hpAPD(%5kvsrZ20{R0Qk2ZXAH-S-25YM}XIV za&eIh(!WcjwNDmlsPkSTAr;szwdD-gt+g}J&}bsk8oQ867&k4ZwCA*?bZGnt!`z{Y zb%%g>z}j6ItI{+*0RK0o-DKtt0DkrMeaUDa5U@T=#la4)tRb=!x@?NBRgzpu}Lh zQ(Flf`UqW7A5ZSSE=#*&%+kw^rM$Imb2t+RIQvza>rxh@PJPrt8_ZY&>-*mR5y+DV zyA}?>O?Yc;rxbd@oxtu+-fjX71}6$)dSW;61)rYgtx_VkuYXjf)aV@3t*S|i%iKoD zrKB<0sjtgx6al6N{*;(!DjWF1y9hc@f`+#;dgN&7+J>~X!xU}e#4It zH_irWd8qwDG;#?|?oIE%$Jab6Fp2VQlq?}Ew;iEkR)5sTPUk?kG+|4_Ck2pmU)yehMY8b9oTdW0|;DQKpI47uO-Ns z2uMJ@Q2ab0M}ViEg8n-!RAB}(fn*-X{9-8U>C)tHi@4mx0G3uX7kC(s^>`nr7G$_- z_tBq3v$ZWHzPfjH9V}v1WPi{DV6@|dp9S4f>;mSU*$Au~0m9O^;LQG*AdyX^LnDlV zfiXo#!-}>pi%?xjb_57M!Oz_(JUSnD#&dQ7w62~)CilF(;q51c@r-|#4XzX?hYicT zwG3VjFFp!O8qjnRdV=Ud>~%-GVyAzVmO<)*?K2@SI7ERiS4A!Wc92D-Stc`%nH<$@Cdah~TauKj0 z>9@-3(>#2X6iaR!(ULpB;iwoMoq?02M7>P8gB%3HEBuj1w*eWgL3e*;)Jag|KLU{u zwR%qpLyy%Op=+E4eA#p(gn0z+Ps#lL4v-ZI5h`J7d38x&;kY*A40@oDG13dW$G6ay8P#-j4$%{DRmAW9Po zWM=u8KjkgHQ-ObjVC<;QoUb;@$`rxlB8gYOVubW z((H0r{2>VEs62Zl*;io#=e=P57i$JA^q!s&or64_3QjDXnO_vonNGxSiiZ*wn0R=S zqejjkB=ikSroSWLzn=nrEpWuFY{9Xyv#V`VC+zO3+@1#rM>~=R{gN*zZgH~_DJ1-d zZ}VZS*z&lxH@DUgi9(6=g3*-*2w6cqTx`*(t8502zmvl`Z%B1@Wo0gE!l?-PWcA@W z={YKp2J-Q-_C+2at_}9-Af)cUpZFik3AxS@#Ep#&bNY)jXNHZ01sgzZf_T~~0`DV$ z4n(E{h6EtbCvZ{y!_Q2As`>9{WqP3X4-WpqKG-}iiz&Aa6amg$-Ksix04yzX0iGs4-Y|mQym}<2+6~iYXT4hE0lo z65`^s1P$fF+2opuVDPbT8I#9lJ@*rEpWlY#bJyDp-u=JN8(jZ9(5o;8T{+LDE*uWf zmWRetK;Lrs<%rkoPi2bA`X+=0v54^I>QFO^hx_ioUI2@QCmzYdl&G(PT4@z!@ErAzBk%>Ty^QD<#V0YZiP zJ^P}v$jHc%7#J8B1Zn+M_bb4loFmwV;KoUQXoz6)f>wby2I2Ta{zRCQ%I}h%9ig3pa+3f`USm0dG}2Dm4ql-C3gZ z+!#7Iuz{{aq&zz;>}B&_aS06-Rd2TblC+G>uec0{2VkNNX5+A{Z2x&on9y#F!ZION zdb5Y`Y5|n}YhvP%u)YwY4Az3S6bXO~9azYqUHrM%)Yf_a=eKZ#R987WNt({_H#axU zC4oAv`U};cKEW(rUG<`cv!8W1A7x10)pys|44qDW7isTf|bN=YyUD@zp8cuWqMnk zMuSIk%D4{>oktnS5ifbQx^ECl zj)CD}Oq$|DV4Ky(VYBe&&@`)m>v1bngEm`_A3haEKj4TeN4igMoJ>09+S_`&N$$q(c@Me%`nlPdehRF&~Tpv;k;TX9=DBd-hq7?{h9J}@6+R9sSF%`!w?+uucq6Fhm1@}wz8xo z%f742FMUw72*^ovg7Lw_)zlO3AU^^B^ZpnAz<#HQrqHCsPm<2_qWh(d>09qTnaDPcSpEm|m0AihMAeg;S~B2`sY z93B_uFHlf$3W~6GZXSos$LiNyGs%|Grq;;L$x0|8quLcyoWW=?>&8hVWjY$~Mx!zk#ocFdXxOMCdOJAASnXs=Zo}ut|7iFq&adHjLZ+c`}cx22Zt!_95+>#ZZRlT&I)EygwUXoUZxA_?B^IhoFvbv!tDc%GJf z%PEwOK(D9G1xg&|%y5q)5#p;Ab<8s7rBnw7sRY=G4c&SM!acn|k$t<8Ge`pg(K zp?>yq`GeYs5WL|(b;rUNVPI}tTSu_eB0|rn(?2kp886hTU?A_Cr2h1WbaV6D;(A%1 zoS@$S3*67;urpe?n6sE>mewxlCX{1$HiJB9Sl>okyuYvBAosXj7ouh9pJw|o zvz1ZnlJxvP^&$(j2HDRF1EDE=1of3XY2uzdy~c{J*lu1YC}Y1LJ!$`69Sa0J_Ao$) zrMCWZs=~k^TqFR$7Th9epWGU(=dY@@48DuQA;zs(>L_Wj;YbcoyodQhZU;&W6)-wd z#pzH9YM5wZ4=R$kQYo0O+9??T+Q*kMc4ttGX?IlN1l%LW!)w{1w`S|2Wk*2oH#ztZ zO@LI7FW~m3N+f~#GDTx?Xp|ztHq?wYDF(HCE?|o&T3Qxs2UbFV{W$WsYbY#-`widT ztnaQ5XcEvcjnXZ^n!gDA+Iz?gpmu_wQpM<)(vq1_!4VPpAp2Bf&&odM&ql0*vXm48 zr}-6mZG_p0R==YN*1x;!w7evlZAV&|cwj>KP? zIvGj}*It7HS=vTQ&EBF;dkwZMzT6~TO#gRPQ2wkm61!Yd8%B}7`zM`~`>)p~gijJL zXE;-?#{qI!=9(M}xXyt6cAbL-Je-}R0a{mKZG8m6XR>K+y9h}3x3}jbz(%!*w;-Kx zITUm^6%g2Q1ET=jxQrBW$v6`!M!r`J*K%LNl7w3x^-aeszQjy9SH>LbB{1}yop*ol zm@EeF>!M+te{b;rUFwD@`01^L`y#u;%XMxdZftG+0@Q5z7T4qal_=ghFSL?=rReu~ zJac{ee4e5%73mQ|GoYH)`H+x*YU$Ut3J(5 z2BOp9G`_^qRGu99<}Z@(@ZK1cO%<++oXVibP&zKVn6p0Eyp~j|3aTz^a~(-*Q(3Vxu# zm3Rm@pD8CK#pd{*m;P_%7FvN)u*@o(p9nl7E;t|H>QHv%i_)raiRb$B>Z2rgUK6o^ zv^!67vFUW(=iZ-HusXXJr)`$_DB?yQ^AEN^FUhGby%qh`iuyeedraMp{pt@mzUIsA z$OfMV)vIS-pY?Jxl&v4IPU4KU zch~pQitPG-`;w_pP;!;bTy9OV`O{HH#l^h^#=#n6_-GT|qBp^Vk28^Z`XOhmw5v3M z(-B#nnTm8wblDFr(<&0$Y@F7MX=K@Ws(Y#jxO^~P%K=>DKCMrJWtP; zQ4K5e4Q{pBD+0|&&Cl_?No|$hBDMj^l*scE9d=qHY+l`dp6N&@W zx_9&(+i`JEX2(pOxy5r(5gos(oLmzF{A#7>Xh=w4K7H~hrZF@OCCMZSp(*THyPW>X?)yc4~w&W*X`HcO1PYsssYZaJdP6D z4K~)Fla9VCq)Dfcj;Dv$ahPx9`kxs1KB&oO2r>eHh+6Af;QcFbW~J{vn#i1CZl)oR z3{`>s;R771FcdW8hkqs-2^1bjP)HaVYAOtvcRQK38lI+XMB-)))V~{|^c;mz?wIY0 z;q)Yuj^#?jq*GtwG#FGZ%Kb4Cu9tyfEZm?~U>}4~!lauWvVRURcHOR*Bp_EA5oGf% z*PZgi4>f+lFeY*R{;Pay&jmEeN4EqruwXCOA98(kzRZ(Nysh2W)}BzoA1#niL-%%< zfp`#Fv(I_YfzwCtUD1x)!#&vkgFwzesgf^uRO2CV52(U}l}25;4x9UR7_ItSaL@@>(o*L~$GdlWim7fk4S9NyD zd|#LjC%wo9Q8ttfUD8`UWISuugGy5b@41S2fV1+!2x>Bo6DJl>kn9 z20)^Vg?RC(KVBx9$fZnE!`ObKw&!v>T@yF^d0FnSszB?2sH||_%_Toycr;lQYd%3u z@qeHC2CXpS&kc9oS$a4a2}96_axr_Hd+hwoLb&1eKKm-?_Qd0UT9!c;rcjLHk&`aq zipJAwTS6e{^?PbmS9vq&kPLMXMw`yIFM=?E)Ybo&a`lfgM41>;h7#4Lyl{r6b^P&& zX??dvb?8b5W=2hKpUn4ouU%Em>NN0avb)f|+&$%SY2 zT88fdtCd>Lm$1WGy?iEZhQ+?H@E=o!#_&Zd&8h*XuF3U+8n!Ld>9uzOAGkHb*(&ZX zet+4IqoWiMdeBX52y%O(8ylxS92fB7p?7P z)|Dsk4X7jrlW1FN4{sw7C-qw1{&JzpfKX9>MGpw@d{7q|yZhuR;|WW1I%8!H*`c;8-mokicX+mwcR)RveaA;mt&sx>#w<*%VqCg3ob#FREn$S>8-elu5T78%FU*3i(F7m<5e<^Cx&zv#2Q z<5nZrw#^FO!2f!lKao(OkZ;BMUy-0D!9esv!|4L^ADFCfSuB39vC?X_r+W!DCmzBJ zrtHE$=HtCHJ?nee94^XnX}5INk&SH6kiC6rzdwFG8QgVtRB~wAva6=Avd6KzkyRie zo73`WCe*jBz#BWWhkP6>WEO$LI562@YcNjqNhLXbR~msN$_^48P$>}kaRNiWcqucd z=yqxO#(Y^J?G0JM$dk{~r#!hJ@ggfpOItzTAOaq@vJL%W)%>^1j>J-z&ko5Mwhx1f zosX;XgbR|NlYwZPDip_~_B$I~R^tHGBJ`@s?1ueUlQ*CcV5XFbi}ICIyA+`MIy(Tz zE}1myKe!S3S~0zLMZw*1&=t+){ihLBE`k+_SpVy4NEZ(2975+2h4MGV`6(}Zd$GGd7cNnj9&>?PBenaY&h+PV*^Qwl zjojlCEjHUNv+|XK(N?eT*Jrz5sda~XxK4U@sP({Yd#(Hj9{0(Gd-i7R+i&b}dylpc zk;!y-O$YoX8WjmZZiL~`Aj?zR@49d+J8KH~Un)9%L3T!Y8UFv#bER;_8%rs=83K8ER` zE{F{;m+*~UW-D;FaOh$Dxji8PV{YXAGbc~CCfRc5+Z6#H6XCL_0acP;fv2J4Sq758 z)|@wy*%we7+j@<8M@PqnxV@1Vfe#e_33{jteUfOj zRPHxB$B}ALIv#v+ZYMi|rNS#%z8k}nLCi;3ckNQ$KowH*(0Cn-+?4@(-rZ&Ld|Q-2;KU5L?CSqXhzq#>UvHk;+L$oA0aTqSlI zPq}Hb(sqK~z}5{Q-nFXAh!5u6*%R}aNtz|WbLPe8c3OL7(A?ac-H~zPb84|ap^%Y! zs6k&sT&V1A<==(~rHLl40%pGrox*u>_Kr_n$+hD0Ll?5!RL_(Wuc`+fRzBQn$3Qmz zC%=Xa)-ujT$5Gk1S>9ZLr{dG~#JHay2J=Ji;Wu{oVO+q~=H>g956W zG@8zkD#UQPpDFC7Z{u$w!KU@Kb&Gawpy*NUpi&uW1s&98MHLKfIqa{+dCUA3dXNj; zyp8x?mI=D55%0mZOXCOD;EL7j6TsyVfi#)Cs4do$hUpx2K+W^xZZ8e6nW?fW=!YK0 z6=X(FR?k^r)h&%XnDVExv%#kN0-)yV1+4~~G)*UlVQw4(+(<0!k#?bb zi?~7A6eDR1A@tq~wJB*Ts>rohD+LmXVnT;$FMyI2AuzI^@c z;<a4_X0acqx`mF>k_10h3j?gB*UTi_#=s3`*(9y7a_17v{Cd*=v*?&&%G| zgg?Tbq6XJ~ddAiVQxls@q>7X4TdoN}+UQj+#bj2fkoY0mi`q2Tl+gJrcJQTwFBMY< z@5=*vW0`CD3h~F0b;o4~%mFJ11bX6`SOajVG7<+9HK|10OFx8!^cNjaZk8w?SUMX_ zRcSOz`WKy;Bmjvo>j~ZlLfM^65~5v}YOatrY58?KLa^`bcs79{e4-(Ya>-pY^iYnB zTeT-ou)@qgi`mR-*y(a;Rn;C~f%YK=`j;6WFmauv5eI2gQY)SLz|2y}`9w1Pfm(*} zl9TaqB(2y`y8Jsbmc*K@DROG8Qwd{Uf93Vqmipqw@JF6q=mhi~}Pknk_ds9TRAlK?yXMkJ=_pABq zQJ^FrP7;?%ZN^48jo=W)XQddPgmK4NBL~P;avtB0XZc~2Ly9R%ZmQ8PUr(>awo}pN z=f?5+M5rl5t5MqG_fy_wYz5{~(m#N0>@v8rgAFNw&E)=2O}J#?@%vulyglGF8e4`D zNCMCVk(ll(OuBzu8)xcgqoTg;-3pH^c4eUC)HkoLeABPp@a{A;yy(?TLKzJq1*Q&c zsJ4tLS55OwT|9Q8zVNl!CqM0QbzjUI%U4F3X%=ZPTaoYlUGqKQ$c9|l*{G>f@vd+q zh!J*Qjat0Ra0~BUxpHAs`|qV&cYB>layzr9w-X;dbc`DQyTTjrpq(uXYq%@P*og}Y zI{T6qJMM3uQ`FAZa{^xTDNzus2hUz@PB-e$`flGr8B`B>AYCna%*Ni0YJ9Ds@zU{D z;!+4rC||Ec+Piw)zqFOtYjK6zRkaDy5ba?ta#mn6|hp!7sb zjZVTgSO_^9mw58u-g;tKG6Ihp$=-bM7A)Q8(Wuf*@uY0?Bm#RIr_fUH1Ss)eA4-6Y2;z?CdIKi_o* z+#*(OZBh%7k1B68MLuGDe>Q`;i0rul28n;Yo7+4o@GTYpbo$V6hyG%mML*MPnQHXS zQxK6c$UsXJKuRk=j<7Uk(g|~YzoEBP_$qGlMQ`+jV7f_(kRwRK*I|AbvAEHr!9dL) z?S8aBzzNXBaeYs&+Z{a3NWzCf65>Ji-Gpota`&td-5*0#;!C*Q49GZZeR8ta`a>c> zWct$V{a*}xQu@d91=TTrB7!~Mn(X|Id6W`^!bdTo{HaZrqmqH{ksz32598=B8KHDt ziTx*S_P>eRdTT`~?`k51PZzF0J|n!nd$m<%IprvvyvQ~Xf4^a8?>K)#I#|_(8sQLq zCuO#a!E+vaG?uxSbB!o-J1uc}f|NTIn!3<@{0kl1px82;r1HjndULAW-g<@n>|5x3 zLD&b(N-*Z(9g0H%-1an>-3=v2;NQsNdA}#4Eib;vGmMSzZY4*`r058=iZthxj(F+~ zISxUmbRd?s=!hh8gl-b|EBO?Qtm(xITh^=5oYN zJuhpNT+`~B=BT>MJ=?KuJR9F*_>g=G;Gl1fbn(9jZ{r&dujRnEl`jk36?i{kt{-cO zreT-L#{6!TZzn{fa%i-D=y#kf>ta@hGGlkd=^jg)*s?LSz>SVJb$0~#uY$EUW*L#(lmY{T;P^DMpI7t5wUe{J2Zn3_w}rYLsQ(ucaU?hW?s296ePsXwG*@B9jC_ z1SaFb8mI7t!;C*1-1R6HXU!N!%70peo{mz z(6Vw1jy-BV)q2WBDLYAfnkQY?Q9j)IS2_gM4Y?X43r14pi83+TvE)85~)ZO zhHx-t4vlYpMoW=N{WtUSp)$xpEP)w<*|hRp@*Fm4SaJSe32-b3=dKImb9VpOXws zfiRI|KHL%Kyf#KrgxCxi`!=x@*k)=gU(`EWXj+DMp-*C)rykHoah|-9Y_LFR_e|>9Ih*a@H&S<~qsc!3!0bbe`d9 zsM6pWWGO46-~6CaR-+OLiJdz9i$Tcfw4(TnadKyT>$T~-51{%ik%}+MB90C5iAYQ( z+I}V`TTw0bt3T;SzHP=mjpp#=k^))*qOTPo;o7?83=l*f%Yo$Ns-Jwgu{G@g zj)wHAw{(o;bqxD7MTNRF>Bl&^+w0jd@X63uWPzz&e{5XC7Y=k0PbX zJ%g58mHqV-3#7@DmyACL(Sj2+I{^XVwEP*m*6p=^=SQ_NYEkrJA81DIuO)-xm;g7S z9o#VX-!y|W&yba7A>9{t1!ck&$wTLS7P^N`aM-CNh0chIAd}(Kwem{{?w$=dPUhLk<&1*{?>bB z6e4k^O13sRNt|F}-uiTZDIZXfHA=KOi$Kzsr=E3nHt-jS#_tI`X=N352dgpn9ZL}C zrulx72(TS zZ!5v`XmixrM~!CP2L`vDPIiCOACUAss*%3wcft>f=Bb(u{UW77M|aH$U)_T9Jf+c` zTd9}QtZGVOLBZze0Ku}xN=oE3=M9k=hAnqO^I?`Ou}Rg;lTE@YhoOFI$%$nG;E~K8 zC3ckTx;i2?234im+E@9z*Fd0x-9kBnWC2{OSkRsA2MxEJ!9w@xq&90j(7!0K&pea* za5`C*%uF^HASIA0&~?1sJV{?adB@U5m zgQq_O;-ff8>t(UBoL}WMVL#RA-g>Ki``utgTfV-_3lof9jqVm;_iF)eG#6mECFb0^ zFj(&+xIysY8z_!Q#7{Pbab|C?c72K~p4gfH>*rbPQH37Q$rQ_e#p!PZ%m+1BW!w7~ z)&8EX=6k-7aO!szPfm*9Ke#gEj`3%?a#?p`xyi7oChH?6BFfSJojafj96kJ5kWj@9u$63L#qR^E zl8MvI`EUEXRV!9o6YL-B9?G%#XtoesUn^;Tv-V;$Sevv?n5;;D;wU~h@Y+$raV13z z0=y8$rtqKF2!+F0qhxfC`Vy3(GAB$^Fw53v^pw7jaVU2%F~j!iZdLfUOGv44&5Owj zWloT|$NhypZfvR21eKBoa5TV_ax&%`3Q+9vN#bStq1uA#tA_*K4Dtm2>qoO^4#m^9 zN@ILQ8bZiHfV6A2`sWmcr=N!T)>v1H$=j`Ebp+o;r45%GH6dfau8n!kiAuw0EC6c` zpd{}J>;;KCSX5J)fSft%WtrIiPv@m454!p$&vvvl5A`pDA6$otrd>9B4%Imy!PqFM zLtq|Oost_KSwuV&i|0+-jsm3@(%!8+9pl03Pl&d?U!=NiK4FdAPJqzd>phBqny*8d zo!3vK5Q#ADkD9hxA}So3&~TXTM4$UeyqJ$W!WXor0W%{4t$aD6Y1J-K%)h9@<)@bq?4g6I@HWz2z%%#^H@EGU~~ zWC-v3j~`2831!PabD<{t7l-X^Jlkd0y|>{`sflNgO=>DdC(q4LayqUw_pgu40RG53 z{NAK3+S@~vv7)u@W$Wp;D9%^(4b&}}Var#CmD6r>2ScXfZB{+WTA&nQq0zpuMA9~y zqwq^~D|3c7RC_3_=QWiNY~$!-N=OrXlhiXI?girl6C(?$0^3t|Lmg1X54y!=@^!6Q ziJZsJ*R0@h2Nmp>MvsGS2w@)mO?eSTwkJcwyyl~mW$s3x;*G_?9O6LR;sxl`i=aVE zhVzv^h3X3i^(kr(rFgQr6Ei9K`64Zks^9jE>9La;oz4hcv%3^IRaC*|*r8FbwHiFB z$k6iVtn>y-fHTQTB%x*jRy{62d=%Hju=+@Aqqc@Cbc#ba9tWfMgCaOTV_WaW0)b9L_ zNLKBwf~{&>^oenc_f)K4k*SRpasmH7=8-YZ>tOp8o&gnQPqWv-rnhnuT_uir12 zi+dsj1!Kk%sFjAAD+P@-OWum4K!jg}r`wL(Jqj2&VX|%db2h%+W=Cx#Cn8_`I`>QI zrCn`NigU?D%^UqiE!mS&I(9#i)(XxZr+x^X5!@Nzl4r%**TyKl#a6n2G#rL%^GlPq zU+eo;-{HS*_Ueymr2(A0woG)dlPnfKZq4{0FmMM4SJF?=oQU8iPDAU~SAbC2 z)_5MrvujbaSt@DCCb1(LH4hU;)(UP_AmZYwNl{%(86AGrq)w)xiOSygwx_ zM3qLjAJQuyu*45(tWOYDPkInn=3TMnEvE)BsmffKjAu-2f~<&MXnJVs-j&LV>Uld- z?Hdc{17pB53Mij-+!$lu1~7Da+5R?t9%lwug($M{>$HRSOP#uUs4^7ZC;+fOl!{}s zn!c6N&NT0dsvf`XhrCc}o6{}JDTkew9S-{6Mh9c9+nfO9T#`nUb}hk4XUtQk<|?Z# z`!&c&Ch~$tfUvHgHOu#5Maxm@R{F!I8)o!>SMk@kX$3Q8Z00qGd$-%<&yThey*V3| zCk(b@>>F--)x6qzgTA77O#6Pf1>%$0_|A>&!nY9dxEJ6HDzq*w%OUQThwYlV<_B|O zDpqArAl(Y*Dl}i?Bw8n{eKK*F7YqwcFGd_RYE6yE4#QJTw#bVBabRymefDEJ6kvzEw zJFTJNEHSpM<_Sf2h|%32qzIOp>p8~Jz`+zTDi<1n+0)uo+LepHLc_ z#waHYB(mCT`-eVI8NE)p)=jvc;<9~1hAipjW$fDCUeg{|PbeOOS_u)B-0FAE;w}D* z5hsXgq-qJR@f$!WU9NZBgh_|oX+PUG1$Wf_RuD~X2jma)$?XKX{wIqrsOI1aYJ8VRLs}CAZ+nT*o=~4CT zEF_n&5Z6X;vA|g!uid4ZUk5i5hbp&G`t~fD7F$ z5LE5uvpIe*%6}qgGu@bk=ul_X;S?5qvZFftXgF1$c}L|J)9_!ZrVPoot0(DyoxLf7 zt)9Fe=VO*dxqE3jl+V1N_dfIUf*r50aG%z#t%*(RMOY8k4hCFP70k&(#x`=_&KULo zj7=RNk|@KthzskC z(2fA!ZMj>wbZLmL&8@BABOdBScHSbBSiuFa8(%=E_<(@_aF82C==%&#cwuzhwz_W3 zq#i`N|5N^nO3lO|v|zShBU_^){D==W zA-|F?)RMWS@bA#B>UUL=v6{a@`Q%wlu+xjAxD4L!Z~wTfJUUdl)h< z()n5mCK)%hDbiK%v;_#NOg=rID)$C-f_-3L+|vy;lOzQgnR>em z?4kdLDH|xXu(2V*a*oMp9E%^Hh@2;IDJ%mdG$x;3>TPeHRFQM(bmzX;cH0?U;f;c) z$LNcpN3OP@@+Wb%iv_&KbA(BS)+R&s2P?VSj&4v{e*|Hsy1e*r>PvJ8SAX1%));E^ zwtGZae_*0FU;-KG@%X&oFqKDzZU6{nOY!;BjW_hTvDK2m8`A>#+}27{wZu$SLlS(S z*SAcg%wUUbFNJ5D{5N~VsjR%;89^K#9!jY!=?~_i2p5>`BAd(IhSGV2K?t#9;obf5 zLC_xSbvFgOu~ehzLFT!6+h18CE{4!#(mw`>A(yh92_2IJQf5;}p3WM{hksdzkUBQ+ zeX>(Vwpv{acqlI)t{E--5qVjnV$+-YEI0kR52y83Yi75^eb#)Hv~+$Y`Rk15u7C3Q zxF@!GlXqS(3P{3}4;;Gd<|<1DQiS4sFE0+j!qChJ{zkvc7|UO5GFI>FG4W3rUgrAN zub%$=6>vIWM6XXAc>8huO_!R_2*f{0pD%c-*XBiBH2W>rDC#cMcoH3>tFK)ab{)j#nII|8Bz3e>U1KjjSRcbe*`?zOo>g8v8(FhYvWZdITaMp zTHwE|_W!!YZ3FP4A#>2LVx-5pyAl7>43FbI7S31JXjXN46w6@C@$3fjRQkNO$7)BE zBa1jd=E%Whftmpk)XP8|?ZD}xxFcWOlUwH8U&nA-Fg`dc?X!3=+Vap`n#)$)vpUR1 z=kUdcy-~k<&)=m3=W92pM$ZNGL^oUzU24G2aS7v%cS1!+yCAJ0iTX|_=#XyC*TJO< z0EK9$8#^%{+iY2HJWGn*#fA3&<+t{f=3;zJU)XUeCu?yl$zbZORm!TcfeTb*N1k}g z9iME`uYWF}=ls$DAi%b__I$P*Sfy%L`jxB(lHa24!`-GWlzdxVuKoOP`LCaSS0OQ0)Y+xaxY%5DjPDyBCrB!`b7129j)*@==Sp0B>0s2?>|cx zN5qeo$_h4mP6S0#f4{0-IqdPn>Est|yXKx;D6NJ@@X|pY;vYe8(K%{NrLFfK4OR-? zfNqq!fZ-P9ufEo7fRakp(ih#>e6GYxWwR{?!?ix5MF>k3Yj(RxkvZ-UH4}WO!eW+O z@H*s54fB8`djb3;hlt;gW#&OOMZS0X};52B~>#!}crkgq=Yuhf{sfyQm~>9v{g z`G`*M?Yh3fy4obeQ!}nT*kRO;d0zn{tIF7CUHBzW@@%G^z%DHxc!)WM&nC3^GL(~M zMfF-b?#N$Ne}C1@%rNPUEm)FgWi39vIdH65CMTyn9I6_67y z;=ddY^Rp|Yy{;Hv)gs%xGTRci@ldAo+AqIXg>J|8^6V^p+;2m2I7~#7uSjw2`X)id z#S$Sf9bx$BUr&9vDe%hDR^AmDpyZqywYnQsO!RxCUEv?1L#9ka$=R6Izg9u<<(IC7 zNs8@#w4yk9fveE*^Du2M^|+r+`ou!cl!@IYDUY)v>y4WR{FabR$pTHeMfx+7`_XGJ zdo{gp zP6*;19Lc#gRo%V7G3O^NnK*dXvjD4)6a}nvShbq(u@mU%D#PGzFB<-K$l{)kst=; z$l@1d9c6BDa*A~?KI#rvw-@;_^!#p1&=~^#3PsY?cK-7G6cm1__1&RPFJ6f__o%<( z1&F;7DSDnKczP%O@>OqshB&ibc*AKaRJ)~s%CnZ7Q-6F)roF`u?XY!f1n&+Qh$BpT z+N)#^|E71 zAUL>t92o$6`Qs{MW7BRe{fmGm)%PmJ4$=fwO+Br=cSUJ~(vwQBkO!@m&NrqY!n~pB zeL9fG4F{BQJP8g7s&>ornY2qT)bPlY_R28G#HqoKr(`16iFltVazMM>w~NUoG5xN0 zfJDG{{H?`uKDwSOi}#sNAQ279%hV5{ucuvqA%HO|>eX9C2H~qP8yXd9#(03I{9LfM zv%=QT!2qlKKydmCs}I;i%9hj*rU?$$20*b%NpZeD%+Bx;J(&+4tA!Y##0$k+CSpgI z`VwOD!_0^SgmLw2oVa`A}m0TWuK~8S+4Q^ydZgTBb-g8Xrm;?`v!vaUclhxi%E`) z$(pXdiS@oo)sm(0FS(%aAFfeoO(o=X@y(0GWgPY(@l_}}7i+)}?V~nTAx;-d{{{Ua zs^kkN2_gSC!FmG0yzG<5mrVi+%cTuQ9m9vv^}5762p7|A1i>gH#-zxQj@{P=C!#Hl zG>L9wJeFNHR6$9TN1%Ldamr=vZ4Zf%`&qcrV*4H$&wPJHp7f=6>{PmhdWyi?8F&lY zF+oeDBFVME_cXNB{d`lrA`#c9%A&-asQ>Y)%Dn4!0G;{sgGKIqogXT3gxeb~&@S|4 z;M%VB^rlc%m+QzP6On%{K)FtPME*wF9^M)UZN9_&V;sP@P6U0BK2U#In1tMs-Q0YM znwf3(j`XuR(#TgbyOGn?H_qorHlB#n<;Ww}L>!GSS(YPspID@;lUma@DfC;D#ky!K z(7{k_c>8OuApNQM;wKu79yoO6gumZEp+dq_5k@z9NX46`QB)*TQf}3+C%Mf}_h^DU zqs(@U1tw6DVpp>5I9D$NhW6dJ$Z(gdlc3%$1!SO>;}KoyR^QWUC=Z4zcbfzYt?>>_ zusFgk6?_hQPNZe{rA3lKI2cO?FsL%ZI!zxbr2@q*R0D?8DReK zErr!yG^niD-6S?9R#|%iW^TsQr6tPfiAM-gEDrdW-9GptZLoEyLUri8lCz0w?em{> zi7Utx6rs_+{Otz6r^j1Rx57wtXiHq?mCJ@+wMkPHQ!AlfNB054g)Jd;G#mf*R5KpW z{xn)6+Nn2J{LuM**7iYwoziPtpWYE%rT6+uSAH{HUrd6Uw2BMHNL#ijb-$?{TI*>W(FsGQV-4LjEQMY#1GR2c7952Rm| z+Z~=(7i{jfT>EITYqbpyJgRaXN~&@=RWTUT6bXxXy5_i}DmY`FHVXIf>(b`sY;?oW z$p5f_=9+_AdBgHhhFu8Yll~+@T(@N0etgpD=d|y2q*C18Q`YCeh5axDjJ8&nZVSI2lRf7|iVEZ2&t)zxa52D%)T{*(j}85Vx_JtbRN;e>8*UPgQI3 z5T$<33@V<>k)BG(wBGvWosQ(gi$VlH7-rbWX{0$5-*^f+vafJoQpzK39QuBitrV4j zFJ-3uEFpOUz3S?>YgnrLj7DVwy=K)SbtI1rST40jE0=ZU(*La(2x=!A(TF)43yDlU z6n{a;R;F9|{b=~BpKCf_{#iyI^E1KH6V=)>6LTdCV;`(Ws{R(1x+x(9+ZVCHMRM-O$+>B1fQLe z#IKX_)5jNHvEpyX@?^)vm|3wJf!mb#rz70X=eVK-tAl&MwS*D*X&ab}Y-!FKI5c$q zNZ+HGXz}sz(G+?69akves+rGmHe6#h;NM;VhJQ$tBjzS`Zv(uCbR;eD0kjx4`;7^t zcVLh}sv1H0uhLcN*d9SDeOV*aJr{M*8;Pql6FVl~5YIaQhCgrmL~&a@Ao~s=EzQ<0 zm9D%MQ@0ez=W9^}D!>FFD3;Ze@j0BhX;{h3Ia-wOeJDyNw02|p4xO+J{T5x+4yG<=8fHhFD2KBukLf_8uC~VHFjFqTZ=c@v8nV|Fnsa5aO&PQRxbK6)`Lgn7eiTIPdsS6H~3 zvvJ5#8xxjakzOzAg($fiw|})tZ&OlxSJP|uJ}}DCZktGob4|%pI=lXJC9wbzKWrkk zzqZ<@j1i_q?V%@`8isb#8C-(f%O{Ot#Y z14NTHoEW<3+s!}RanuBe0(++> zit88W&XFU{-N&>sug|xPyYz-BK~wg_I$G!V_m8qBXS*sp4K=R|Gv^2?BU=RC_u1}h z*yRYRR$H&Bwy2G;A7h57lKXOwdZ8e7Q+wW!+%_MiZ{qolnoi+irX?{o^80d(ByT`w z+9JxR7URh}LgnSu(mS>|K%$6lPDe=3kqcYET~}44 z5*#8$GL)sY=}y=|1_9LL64KFo{^(w5TS05)qeDZ=(m6_^U6RGUtETcc!mre#?x}^N z>nl#GKn0uxeCmIVWvhYpSQ$|>si`f9gW-QX5}fbEPTvErL7+ZG&onIWEz!mlQdyc2 zFy(-b2-i0>)1T@4vWSit<5IiOa@PXNG&Ajg>yH?%abbDNz=C0`KS{+-B>fDK(J<|R z=y;qnl3}s6mMbu=-P#0ssQiVLd9l1{d%wuY`54U`j)imUi6quIp-Ia(#Jb^G=d(sE z0K=5bcE!U!NV}>(t&tb;Ya};{=7*wLCisZ%h~yhXqDrZFn6$LtJ2ZKhX4tB9xDLco z>t6??+lo;Q9COyP?xzFQ&d~K+4A;_A)a`23SUL*qE6DpM)VtDdF$SHD-Wud#l%(b;_g3F zEo^qWqzIUL({{Cwru}+t*5MjC5*s$;m$)oe>r-|Tz#EH*^)B0-cClp*KsxZVxd;6! zZ-de8Nvf)8`Irpd4Qg~;B*Zqs9A&n`2+&Q*9Hp@F_}T3qcLvf9nI;QJustvzuU*%H z@#6i$C32|0DVY)b4Lfl#a3HdAEvocN;rG2`={(BsYFMwY=lBZmaVXxbWawf2Ag?8)f;r_?3 z{m1fMZ~|SQXQ(!ar=ZhqUeXLG&;~FUCZf+!h2L4pRYyD$posHae!52qKT-#ZG8@5k zb}jZ^keK|jdZezxQjJ*|_2=<^;dRU90T zByW=BHVF_yL{#oDyD*Y?W14xQ%RBuy58r2gxBKqV+7n0}M#a6!W^Lc?)hq`r2@X7!Ev zxI%+F0j6kcHT}l}_x+4&*7xqZJXqq<=$J-F;`_^F<^(~9y*OfbK~8L{x!5uNzG=2g zYKfLSJ@#KwdSiuOIJ;Ku(>A!|A08dQxo)+wG2!H-%gW2BZQKd)os1#xf9|#Hvz^hp z09AfV>CyDY1FkQvzYz-F9Rl1B#&vH9{&`2lrUV91m#SDk-n{caHvivSn28t=!+RH= zS5f}&XX9Txd|nEitDEF<|GJ|8`{@B|v5#Q}ibg*23HqOh3-br)L3!>i`4&)9|JQ-} zZ$I{b_KRzH-R~}By7#NzDR&dh|M=q4u?%3W2YAjaL!1=suRkML04azWXlTO_RUF_+ zc071l|L!fDUX^(>Pm|}KsSwbVHYZv<0BwppL4Z)G44~zOgsBFXl|CT5T&Uq71Y7qV z!RzAJiq6XeRJ19~Nd!!>?L^n9@d_wYUMF%-;PYkyx(iO}X;Xj=dmoRQ-u=E0_RMLt80f%W_aC4GXSm=RG^*FoDZQ=KYa7FWznPXaK6M2B96)8lJ#J-EZhMaL8K(W)8<*Vx{1+ z#8IB_{6aQX=Q!8iN8TXCC<%5IeoxTn^axNoe9?|!1`DxIrjQ|AGc1xI0MQ@1EB>Sk z2Ny?^QTFy+nl(^kh^>5X7j0g=9TZ4HKuEYidzVE1y$a*JNf0>rXC5C*iGs|T=FchI zdA`5V`MYuTKfW+}-6FwKHSUgUk^1NBTM!&1hx>m${NK;W-#cD%pA;OF=15i_K-c`I zv)};kry1`1{~Q{E7MVc}>P4r1g#XXSd>SlB<6CYFvGE_yS+6o+h3^YzZW8_D|0bvh zpUmyY8)>^=N%ac)t@>T|MPFlkk|X3RsrRUJZ&%QzkiMK zGp+{SE8|{v9Bk&@0p8&rz{6VtlBgu$3fi>q4v3te1Pi1aKR~ct1kRFC2g_hnFu1q088?W5O=+yLkvN^3DoEkOF$tR2h2QsKy(6Vl0!5|kumJzd_b-pYxZ(cjCo&0 zzU}bWqj|0%$>!nb?wiR#dEdSL2dJ7bXY-4@$MF`5^~TLRFn9e+vA|n`g39b$SJ%@NO^i2!6 zUoVBnf{3dGd^=?rW>@Wi3*kOMEPH^B9ushOa~XUU^&N6mKi8dRaqF_(?8qasAx-kU4*l+atvmB#ff?WW9=0iNoC&{63O-eoFHF^ zUw~UH3N-)DAXxVTPa=%Rt}|R!P_f*wfukLaQ8Nb#8V^R&U1OV&aSS?V(c?v$=ft1+ z0Rsp3%nv^V6WIvlg4*kE5XuSd)9&K{g(abNFc=Z=Qa+QIR-7t2)#UoyQ5e^ z1@3k*+hpyv5%&w2Q7M@5-VgQzL>uOvmY6d_6JkGACEJskhy}Oh1Tv1Atq+NhU%i*K zz=&%t{@%G5hT1mpSw1MgG7$~%ocE^%EX~$e9?Qr4kJ{X!o`tL{ssfm zxX?ZJMc@zF9xNPeHg5M#U6g;p(dW%fV1{Waa<4DuQHUMPoj&Jaqd{7PE){UPFMyrC zdt)YXy340X7y2sem#)A+rwW1+KzO@KD!T%5M1So;3<22*IA7QgqG^YSkV0`#;F+rs zgo9`5)&gid_N>uC#$#fWbZ_x-Nu>hM!SD}DGH?${a)3M44BOP%n!uV-Gni2Sg|ZKS znQ=?w*0eR)zdSxi_Ef^Ig`8b({hW24xTEfpK8LGsMeuZ++{-}Huj$ZwYRRTm9B=2p zb;}9F-dfc%aO9>mO!WJdZ{$tt4yx275^>LIH>r=!~r=_Oqt%B7k^~Q z0L28J|GI%B2}^+m87Lc6pAbntSM+f~9F=2kPK?-Jh7t>0NM#^r{;6$;z`{_{x2&wh zfB{fVhk6+gi2wBj6Mxe9Y)pd-4pxTiq{ulHm!tBe3|POB&YZ!5?`=&7g6pV~&^lWa zUOg8erIi3!Wr8{WnwDy<_0lG&R2h&}?hv z4d(y&``GQiYtYqXxmVil7M4 z5mE%V;zCAUMeAAL(~_Ww9?1LkT1+A&9EGocNr7bq$a1BpWm+Zout#(glCgvAaE82I zpw8VY_EP|B*I+^nGd1g~ZNawA8*Pia`lkIA!Sd`OD`!eMvpk~0#L)yP^+h69-17^Y z*y&;%*l{?f)HrgCLg_SoSnEs>)U@Uq@SWfMVe0pWOvol`0_h+Oo?6cB=PREVU9-NP zgCGFXEL9MqztKJ0Ph-TP9vGSb=x(1UrBwF4d}1D*kFB(o*O`btX=}# zA=_cjqzKG5%F#M7Q;OgJtQcZLAVWSVilkfg#Q2Zyz9Un(vx47(?;I;IOm_OV zObrfwp6-&sTX9g1P{x&{BbvSY`EfMu4lhRF8ntrEY8d^t(x*(=N)#iAl+?KW&MGec z%wFUT?&0?UAO;)*R&XfgYIodhuB{oP8$=-X#sk#0`$-#e;~WHl%_=~>33I{#nTep2 zjVjUz=l%IMCy~jGtCu=NHW?2?&s)WK*oYYGYxrP5QZxwjsB2zGwPt~xF(~wnd9g;uEW(O z5sM1C zhWgbn+oc&cnMwWm4N9NXG>x?l*&+WJ_7u-fp;8|Ggn< zW=>f)*$4)68_=+rwEAMRtJabb zsDZch{-8d_4!j+49Jbiu@^7pJohFhZjO%0hVb+TH$9a_3`sRSY;kN=WIMW+uDdCYnhSf?#|-|;@0 zva$ekOP+bJAA)?$H6T53>*a;g>TcVmWXhRjHUM>NU;! z48WDx53{!x)&^T^+lnB8a%z|xHn$iUlz3@;81|+g%_CNfq!s#;3e}oJ=R?+r7{m>q z8aJ<|LT_L+BNH0v2P#Dx`43eVZUBCJhNQEw;OdXrBxCR1+JN#PBck`VKC9aZK4e;r zcHH%RF!5%vnV6LgT)dG3+%KYp@3MYReXPVBNmIg?c;b~^Q^u0Fr`=JI+SGH9!sP>x zMU5I;EmU=N?PU>j6`##5E>_RGGir;VFQYa_pHo8?1n8bU(!^p+$a6E`0kJ7?_$ddp zq%^>d3F2oMhe@j;%vA?`y9Dm`2!{#dC`SN*r6-QkQ`d~ VTe?~o@nz)(j8P!2Kk zev5!7Hqq<#>u>n-VC6XVKZ2$Nu_Ek4f_4((NX9KiDj0vtNx$&`Om~g}{sszl4m0+> z8bll%`mmm(=2EYbJF(L~c9Av)D$-a%A1Du%@|yL3;U^I{mWVi3X|;D^{>M)eK}*T8b3~=%vI$$d>NQn$4z?A z_8~|4v)}5AwkM-yV=Zye)#bOK#rc`fcD#qj7wo;-PB5^gY0JuwJBNHO!oq`|n%qjZ)kbOqy>QM5D61Fqdrg*4C6z z6r_(;AdrE~$J*2}b&CocGOSeMpCRt24|qHe^&wp-uprDrpzPt~i~$?qahTnRv`;cI zAt4#WY8)7FS;)yTDPyYlFNyaip~QdUwgHxaVkyHPQ~1J_Es;juJZ}5-Ov146{p%0`ttR8*ZHG(xAfHz=_9dap)ky(Uw$(tmCzEDF$agt63?+CRUds zHmU8c0{PW*pzzjGmR##kBG-JWc_X(qGsL%-rvG7J;4$WHwDX4@&*3FXG75ti$%(@} zSH|dFpg51;{b`T=|4}z91d?Ve;tpHh{jBx|6qvQ$E2~g>!B(r`Oi}~l@lnoxPw-Hm z28sKf$*iUbJ#9b~pMjD$s=ssqU<6X^?F|nUh-s(f7-MkjhZUHt1;D>zvw=*9k(e;$ zwgvXqZy|9llNu!=aL`^zOQNDD)F_%2pYiwq9`sJ81R+Z&&_#Nlp2qo;BqPDQth)sA zL86%WjDa!Oy=udN{Q{D^r4_^Uk#xn;|C0uCIs%z{^V1}HA222bPuu-IHQ{5V^|m-L zY-(Hi^vl1+*niUi|Kkhe2LL{N8tJgq`CDiFAIQMpKSaO}BGXs*&MA|B;{g8+K6#-4 z?py*odv}U|TE(E309^S0@zO!Bzt;ZfsA`jsqEB#I_=%_N-pdN`+z0RG+p&AIB-$vI zWis4=3@!apP8BzPkokjm07%O^HUlMQ`XgU$b36hjlORy@_5fvB+|HdUD}oP&?H>s# zU$_oWL2yke!JBJ(l`|Z}Z>QZ0Y8z1QF^|>_;scoPH(r%_??YfB7rwUW)2{5jp_OO_xa{GUS9${uf@D4`az`t-8jmr z?P_dH9AH0p)-M$Y25ugpbwGvc3vC1i#}cM_=q2{028WZ;d3#!@sR5uRXRmPtD)(s# zP@}P}pn+Md_xtS0wmegEMH`GwEWl8{(UPRIEG}$Y}-gZ^jT0oVDEs_k8tqqNg>=cxd$9Qj ziqltwhx$`q(Z!bJ8eh^3{?!L@LTN6)V1(hW;)%l|1%1N)2`iLT0Jj5C(T~BQe}pkf z^uY#^-terh1DD4C{QVE{04s?Wp5Em6Pb~1ie)1~@9{ay9{oj`E|1S9d+(wN5=WesS z+eIxzu^BDSOQEY9U^tpotXT|j)kY*B`jfbBYRwu^spkX%e+wjM$y;u^1ExRDkR#Cf zDgTYp)r69F9Ekouq`h@mRqNI_49kU-N`ojV9fE{`q=b}oi-bxVl%(VWR4^z>rMnyH z7D)j~K{`Z`Mx@~#)BU{Pd!8-lukX6J&b2vvvtrJB-uD>48nQ>SzdS{1;repKup7XFe{b3x(bFr59r_B+4?>~EVL_fhhky-f$3Y^LtUR z+<7grw30Stqta8Qk2^BJMV;=_+G+)5&88y6bgGTsw^a%6?8KQ?`s^Xn)&VPAABE!E@5n*%FzZ`ot9 zy2K{j;iz`Qk=9?d*40o5w&Zc4feN&htv;N8T-l~No zXScUJ?*ZtF8=mPQBuz@lFYCrgxnt0-=CXi@LfrYK-Pox*kI%(kGcMQ9rS!zt+ z^}yfJsEY)GrU^4wUWMNjww>}7bwlXuTT%Q%FG9A?qKh@7Jq9n&uSfzQaX<3C{4(mP zRsWMm`G%f5AY*bLI4UyURRU!HH4=YF*8mCT?>)DsTl3AaFK%WGcD(!L#Hbj~9cym# z2#vi3nhhhu)FX|bwxsLkPN4AFiA5Z4k`ZIHn2y=Mp%7drNBG=8F4t$rJeTk% zH@&dfjZPd}$q~zafYagaQi5soziYsMszvzI@#Bg-1itwH^`{<0UIFK?DkH29$TI^4 zhU+jVQh?B0hrIufF4ShnlWG7fk*&I7{Q|%tD40pM?f25R`` z;^Ot-(NrKBX1Agx+!=f3Lg#e>zstF7DiBDgb1%xX%JBH zES_7jWB_h0gzx@M4(;s7V(W23_;TMZuY5Y5{>MVCk*>l+tM3MW{KfX63hh^bCj%!Z z^8!*{kG*!#e{9oWy@_4}{!S5|yJT4V!E9^P|K%IYX=2ynN(cilm@ZuR>G*Yph53Qr zr^9j>hl$KDOaSacN_a9x$|7$|(^f73BuLs>l*5ftn zfB-aetj?QMcpe=h!I5)J|D?SA{cXHdP_XV|p>Kx9n**fozd7!6|F>(u%{j4!^#cZ* ze<$#G;P4w9N)T-#!(u&FW!HH6>c_}&01`i4i}C;aZT{RnudB#=@#yB)S=9nfxeh#( zBJ2>=twbBg)@g9%J;>xHOD{)J7?j(|KnVL|oguJ3d&eDqG=+%qzweA18YT}}KBRs4 zc^XNOZL{}$qg!T^P^;balvJ-S54#(9as`+!yTk*SGMFfjd5L;%8=%tWN6L$kc|4I$ zOvxL|gIMVb1Gxs@yqCWgG}dh39(R{9#uR5k6aEJxg3E6_`Yfocbo@j-$g2y}0SF^< zxr255onx@Xf&1=0<Rg>pGj)htyI_L>6a?QC^}An zhc}8+Zq|_gj6k4(6On5z#~I5BSbS>SFHzz z<(QTL4*FKq^>+&_-9^D1RNfko5=$&|r&Fn~8by?z%K z(K^g~YlT3mq>fktb7dD8os|NN07)t`Wfl|{MuTMl%Ik=^!sgoeg2+kfKJ>5qhB#vR zP7x!=?F#73I^cX~6JF4TAZrG2Luta{%MS-`6?cyS?4JTcX#~1@8s>i5>H*R6x&6LP z(5p55UN8jyOMiU>c&{GN)4Ray<~uS`)?>i-83D!1OT6?b{kjzv ziQ1j-@7YOq%Se;qKIRtjLx}fZ3sZuh3|@+!*=czsaRuhk*)7F{Pe3|z4vn}6z?VNj zzrrl_{8aAzm(vJai%jkl)_ZT>R~Du*tSJiu#9m<09_+!y6;4=WVIb%Jx>IRCe_?lF zth&;|_=?gyL1Z%(ej{GbGO7I(UwWNnha*BcaOjJ%+2Q`qQZe+GuF=izGaz{|0o}8sCUnTjAg6>G5h4>w z>|}y*!kWmQYjW_Iv65OeZwa`!xD|HM`7H z$QmYbs zBtbXejg?%f>$WsE0^m4K-hcI!yTj+MY~6v@)SN1^6cA;=LFU2k&%Aao>=XN$?F!5i zsAGR7l_d72?61BAIZ2k3zrVj7Y=#)V7~!E(>p1sxi!X_e!&)d?OW0ge_Mg#-KJC*Y z`TmR#n+HQQO%!OrcHFgKz;sAsDJZEFI;|M)+1yr8>V4OjZGP35nb?4?K7@gfkM9!+ ztH242m%i79+XL`PSHeF-=GrVEF@^4Xj!_~mg~Q;?Qa~sp7zbM#=JH>zj~qiU>DOx4 z(&hr-pcH9&)e?Q{*wWbm@OkO(3<5Ry=nW#V;S`x!yB&m&QP?_OA1uB!{3Icwxu!Z? zYMmF}Q446xIy`1i4I*&l|I54ZQ(*waBDlj`LCP)*^M%sEslszH1Ke0kgBz7AHE$uf zsUnBpW8Zo2E6mb-losq|t%|qrqu%gVLN%hUaf<9v5~^DcpbOsu-g7egLZY`QunKJr zeFoF3=lS5&ahZHZl?)FMKj;|hsm=^!E`Nl0&f$Av_(535{dl#&!D|BQcpkO7Ab8fF zEPMse>0;sQC_UJzQ|4}r!C^8BP8xoqZ}TuYybZR2%-;psHLyqj=3}0qC-;8R+|-mI zctpsHx}oKTqUZ>-hHYQwF%0R!PI21(W$6NIR`VB6?^=!vsdV-VFk#3rYWLQOjY&v~ z3J`^kBn?a#a{nt~Itg+1Mmh=$0V)tdDPh3uv8FkGfCtM>`&GNx6*A3RHFE0KJ&Z50ih1@ zw^71Qxo)7<%UIacVAEpiIQC(N|I>y68Bxf7zZj``Ktz!OY{r8L0Su<(hI!{knFC;Ye!%O4On5|2 z5;#NSG-o48D$tUW*w;bOA;S(SOxR4DTjQSXfc5o7d^AYv9^f^s(nTITZkf&360oXG zAgGX(goM#^f9-{F>Wx?84)z9C|83chkXe8M5b|S`s8jEj@4!hOmckldR132+6b3aw z2DEj$>>AYG5A{GdGUB>Y9U|;xm15i^HvD3$2*!W%LojZsf)cyPU1|cVfZN@?3fshZ z&cw{%h@_8SZN$njs;$ZEHwYKEneb9KmJIgi?ersIx)a{}QpaMwL>$+E*eDzHE-l_T zgrTIGU0B;1vZbvT|Jyn-N8w4r&{4{vN~gSR-2%yb_wR?Xu%G&RM;@$#J*QOBirx5R ztn8)>Kmj6|7v3pLe&j1keOqaSg(>oICcg;wXW zK;ZG)vRs`g_Sp}Uyg{f}D(s?bbAh4kY7KD@vdK)4EhgaHWd#|>Er&bcuPhvKo_$MTCoQ(G zF2j6$yS4yLm)<9I6qnlBpZWeR?S$d&hWScwv7{BVqs?noqYJx}f>s-J82?P0h04-6 z#hal5d1T40pT*lIp8F5jkQ~r%mDZ-yTa62hG41aa!@scVENNI9dvGXrEwK6hflZ}s zeC=(U(HGO^se7I*r8Q$r#N;Q{@{v)n`xihrq;k zF)ixIHQo#x0|8yuEk6+VEk5kmoV~l`?`I4X9m%YRg`wYnAjNe(<~&R_s%`6;%}?sB z>9e(>+9MKK^`a4W?C592CL1X`Loc?t?woEOEAA=wXpipEhsoV2-1pyLF00QdPF!c; zir|6eX;_Fb8@HTDLkN3QLHI8V9 z0#epHc~r3H(l=d2xgD;*iQqXcBYVw5r{_svUF4cFS-|wCxs#sm+ezD?&@@_|Q?@Zc zT-Kb#`8+-;4KOw?VFVu5>s;-R8<(p0!DG50pwK0Hq20ZO#wMJR*aqhZl-@aX6}=by zyku1`H0(TLK=bwibD)WN1&qNqbX`$+cOa%NMqg;gYJqTdOVe7bq8y%JU^(LKzZ)^b#YEk1={trfG6*t(vy7b__2PSUZ35}mK*M@ znWJ19-~IH6<@-?yW|TcKXM#KDf1Vn}PwNu-y1nX^W`2b)!F$EL3~%>c>}*;3c@h#5 z?Lk5@t^?J8wP}H*WF@byfnw@%F3dRds?Wq#4$*NTr@5ANw_le<4^B6`qPe3t3q?o9 zd)~Elvj3hO;8`I#^t`8cu){SH%Vgu(on>z0DRIQXfhOKV@eJ$-%e(BGeAd&@q}uW!d?eWu{&u10h2BW6H6 z?kx(j8o4fAuD27FP9&nLKhX;3hNe@(SG=?u-1lq1oP}K=8ppNUIPCQz#6MU^JzTeW zs;9i-BTTp3xa4}$G}JJ}XB^wfO1i2RRd`cg#Ebr#jl^lnIN=qNG0$b);S7og6O`%Z zXI5M%JmssfCc2DnjQ;rXBfrA%65|`HM|iC1qdA>x5%Oj^<~*N$6XUkUnw^lbxA z56^1V_oUK;`intZL~%LM8wbtk`Guv?T^UpF^$edRO-Omcbl zbFE;ISbbJM$uqgXZ+ecUd|yK98W|7XV74AvgQ&@1LyW|8P6-meYJdw0?dR1Iitj|$ z1#sNM3Q7i|Q5{B}y|`bqybncVW*cBKNOO9ozr@=FY63eL4$ooU3EB%S(;js*81SGF z=2NKsUr!(gRkX25@_31&fxfntTVd3wAbFC1#DT}n+c!DCb5>V#IkxV1nPn6L{y>Rl zxyk(!O8sw}!CtljdR>zYPDnM(@EUL1+Hm4-P~;nsACy*Ll-~)q)-b-;GRF7yl7TKe z#A)1pP-N!GIf3|XpF%PLpYL(nUWqC#S82^27A-(MQ7)3!Sqsyc7pD*V3b+LRJU#!a zfw3>-!=Y1|a!>0%i{~Tk7qwMQ8-$1JdY<14NJ|ukHsht67A|aKyMqV{Qr6Br(A$Tx zlYOq$20{2p6M%OgMSTzZI#g}?8U8GkNGJZXbihR z@IX=s{8go!9^lQSD9?pek03QK>{|AquQY+VI_uq1L_#z^wG~h$W{O;UJ}{+1YY?G- zT(?1v0Wr}xB?|7CLOV5c!@{wLbDjdS;fKX?Qz|iG3a7K$M)MqlBv}aC%Qxm`K0<&3 z1ce||%~GI`ScmBE`OM#xl!;KuRT*osXiJL$XSrT`bikWE(*D)wt~Xr%c2c}D6glgL zs+S~8=jDbbS|>a{7w-r}RGbzhR>Oz_^mdI0%Rxi^Zn4|Wy7&2A5Q#E6jo7{z{S2fv zho^(;@RdhjyF0hi0t}-5T(5ssTX@9S0^4es5FCw?>wB-}`M0u68tG{6h(CO@s9!2O zyjiQ`xct=;_muC$3dkOiucmr0W=4X>$95}GvD>C;dZ3bBya-ap%nd#t=IJdPX zst43qZVB&7F1sb-p-B%8-)ABCkvwczWur#<`u$4KizE&r?*Q=~q)rzebV|6E=09UA0RL{^^JnXT8V z9NFJVu}f5>ha5e|_+p(zO>Bv6jr`X@}4hl_!n=!{_@I~n=T zsxN0Qvc$@VZmiAGtpDnqZ*0Z(+^iY9;8nm;;VwXAi9Yq*RhR%N!h8uaUOX7tP$B7i zF9a>R>mHu0$hNFmahMe(_yG9n>YTiB%5~c*>5>^*0Si2P^fQa!i_4phhka{(xO7a% zw0+k2#?-sescWuP$!w%%hF3~haokDuqxt48*w4qVNH$HXJ{tjcHO4M`Tyr{!_~a@V zjksR!#y_otFDdrcIWh64_|NW|m=r2g)ZD0|5PhN<@0HYAQ@VLVD_1V~g146}NE&{n z?+bR#sn8aQ_5SNwMt+^(!XRX~mK{~+{j|^0S!0h`F39>uvFMa$GnI;kkse9XbAvP* z*b%jr%!?xfC6B)iF!El%e!FScre)^Lp9^yh<8|AsLEH>dzf2)jhu`zi4XJf?8sn~7 z$8tKs2wzvN*k{Ky>#v#o&or8|4TnmY@tM_T8 z0y^W`ms(zFv$I|_D$@W5*1T$Jz33a$@!HlKp;omAjyD+Y>v*TH=AUx-aK=F^+y%L= z#F@x(3wutU8_dYCTc44)94^kN0`W#qh6Wpj<*2U>^o$g4KR~sw@nz2VU5W-e(RXK~Hqvg_@oEi~aMGC~;2@4DTJJ z?18yASFsLbQ2)h67-(nmNP-25jef4b<#qI)h}0x%ZCXK+m(F|)!$JlQwP zJwE8gC2`{yX#3Q}qyc=*42_3rsS$wCXNsFo;awMW+qRipcQWT6ISvgOL_-t1bRzhR z&|+~Zeo??sc~Q=cRFI}ob|roHVIu>=`*u;F1lE$#wy2vhkg*j!30@pry8Pq#Z0GmHux>) z3yTR0rR7LlF5$4>)Q^|IM$~JI7^4b%d19G><6m7ITiSXsM6|~V#{#?%6&2u7saJ+$ zw4T{UG3Q-Risf5l$NHu6Ks_tAbS#HgY%@g3BWwKWS@`()j_s-IPk<`;K%`oUMRv@C zDuJ>jy+jfJz0&a=#T(R6A=`QOsK?YVqL^X9U?3len(tiw%*Zd0G7n~6G7mt~u5_CBI9l=s023Ud^ z3A&H=YLDLjPqqm{y>PBq`d{PVkOww_!>0Ikx}INbiR04kBD~`^83~jX9s>JQlil6f zjUqEQnIt-DIdVA9Xd^oJhGy7N_4o1+*SAvZ2#{J6ypCi?zJ8D%KKc$eqF$c|{585X zx=|ANmna47m-|K@J2Fsiu|h?Lq!!2l$ccWb_Ggp+H~Qn{iUM_X1b}5}ocZ^6t_=Mi z57*r3q;CFA6Q;84sxrkL_uSb@9Gp3VfLw#CX8KO(-k-PUz=OBng{8)U+XjCg0e0~g zz}pZi=S+V5uSbynf}k(_An^8o{W=thKPt1GrfIW%am-F4R~K9dJqL}z>uNIGfb*`o zZ=cjgap2o>tX)?nP+S;y+fW*$h*bR3ttc~05Ol*isKW>Z@+*io5m6F}G<^z(w3blk zl@$M1IDhY?h1W31>b>>nMw(q9R>fgAdz~Lm zyAGR8dd}6OQTtw{%&*XXRdaOmAG7F!O!2{{r$Xoc>x<@e?PPGOdX=3&a{ZRukfywx zeG36o7&M6S(ouq7s&7kUYpEkCDh&w=di0c@mUiheMffpCABJ0TBGz>|81cFSZk;mH zPcajGH&Fgt&DPVVB+H2JA-t@$QbPdKJ;AR!L9z@V--H!~t>x8FY7V?6MI1DGPe0E? zjHH?=lxIs0w3NrJZA`WTmUg`A?N6URwUu04kYDL}xEaTzNbk%1Vg7aZdDUF((Y6tQKr9`n%kL z&*%W7(jN#pv8u6EkKL4(R*3Z*Q3rLbB_MTmdF;=Sg~88-fz1*8M~u2{9&iF!_N50i zaPC7VMYC-Hd&>@iJFF^@pkE0I4P}Oa=&e@kbSa+^#=jn11`M}!JS11CCrab{cb>oK zZ%)Hh;p)Tq#Sax|`imTPg>E*iQR>5~sVP*?5G17GCdb%VFvH>kD!Uzoda0iPWU-we z%!?!EHM|2uV7PP)9K!y%jAz?eKp7YmD7GsHn{oePB;S2zetv$M$b!IQRtXFqRP|aO zuQej&mNA>M9KR?*rjo|4L`cbfwy$7y2L57$mry)P^|a!+dx&q&_kI z1g0G-tRc?8*<@<%#O=!Y)c46SV^Htp72M+vLJWf!!^ixQk8O%ao7>a)%O2z0gG7lsgjgP2(zSt;fX>Kjk8z z`P6hk!0O+-caLTm^Yc`8=W~XeT~G;YowMTl68B3^Qqr5iuLiLpg7eVM#~xTZxmg^?1J%aE7s`O5hKs4bt2=dC^;oA<68-wz}_yU z5m}Snc`z&}=_bgQ?j_Z=b%Y@#Vh6u8D12#ZL_K$Asi*}lf+T_onwZG5*>TH( zGwJ6`PGF$D@+fk*mq2}Hk5e^Y3>XcCgl`aJbvYa!e+4+b!?flByVp8$dwex$R%#d} z7<9r^|Ck3ZC<-q!{5?Kf*REjkHd!lWeEzmK#kpr_{u3Cq;-t`2pt#0q)4@rm^$l;F zbveJLgd0INXksJa`G5HPfu2bYdc7HNAZv0Z4(8~mk%wiRbPzhxCZijy4Qi9agDoRO z6>=R(P(<)oK_C{uY;h3;P}(=zNnYV9P6L~rH4N#RX_cwfZOw&Fzp&Sb%;Pw4vU0bP z1ZIIQk)+T#06)UO#Uf8%F{OXo%g}a4gz*?OJDC;B$PLLTt^%OF(0OgD4=37gzXYo~ z)Lx!cnT<xoSRNYc_#y+Vf<;;s*R7oWKAeO*OX=dYgr$pV2-LD z3mY(`OqnH+RyeiDyD$aELN$LQ`M8TOX$*DldJ9t^Lc6qmT<-@C!f0njkz+QJKP)0u zMWtCiV=HuBz~ypwq=M#L$d41D1-DP0I?I5JoCo{bZF5^i6r)VMbG^6v2aDHrQ@wiU zv~#ASDMC|qNVB{|stG{YJ47Umvr6vT{=7+_dlY_(9sp`qE9pY-Q!3$337AY)F|)S3eX0G8FTS>i18_*STVqux3Y~^%4;SkRgX7b zy&9Gp;d2l<>k|_Y5TNZLY_ORG@*YlCuz2U5$*RiHFC{=x*M~i;Y(1rm`3~)c%)9aO zCu!e4^gO|O*$=z~UTBWrcEBD1d0D;*h=U$HeRoFJpqcA}@#jLuzdkr}^jYIc)MM~E zI#DSow-XO!{5EVgTr5vsw&R(rvJSue?Na;aIPvi_@9ur4ziuxVPvc-7Y!+_%4()wo z;&+9E>00-e>0wKdB%N$Dq=SkZ4uEeHvW z&FlF@Efd;>BjmfpI`@7g1rlU`WrPY(TX8b4{QEhixg}WbgL3az^aTco46M_5H?JJm z5gd}RLqylNm$3o!@f_)kip0#lxBDJ7F8T^TL@iiI=^8^5-wZ zI4bS_NKJMa?2vgKU`BNp*zgq{ZkdO2=_qsDAo2S#QSU=~{(KhEkbZFzMNaH8n{S#z*Sn}FZIn?X5sA8@M&I+a_O0u5RPBkGLghefE4dGfB2qyzDL`}#uY#r*gR zIQpWCmfU;!xhFNfW8$OM3*ot+{XfmV^k1Lsq^>fHp}i7b=FTLj%7 z+es$W1~n5GKIA`3L;yh~*CFgpN)arn5K#MMxg*p8@mZALSZj&d788ajwfkd%8doYn_`L25v28MdgC$;eO^YaH#4W zLQt4Hfs)hj`Wdc%V}gh}+_z#~O2-`Sl-S(#v;tWit$_}KixRemPU;2sWTN=)KmXX=Ot|}&f&F?>(va1r296~a4%_or9QB@ z&#Fh)SB}tACH=OO1aW&#e$FgIdY>~9bsHl zVR6e7m*UVp@>tGSUdPB(c_}lXp*n8ug@kV)=OFLE@b6# z&a#E3BS8kxF{M8E4tW|{3f+LFay|PA^^8W?8fy~d>8yh_H-nE`|HCa{)p=o@AEj=b zB;QL3tE*$qNe>7PC4E?mu})LE#6}$R!Qlh7tK4l0vuBj=WQ&fV`O>PEAbR;Br{d6)KY2TpyoTA{jCX~uyB$5CkLfh;uYVoAQ81n)OSFklsF2Z<)(2JBFLfuK zS^Z**gz#E~(@?TWW-HT!Q8G9lKDLG)0eN8R zI$zH3J2C+(SpiwocrEBs>TN=q1e73JoI-liTbthFT!TzU%U!t_8sk2nb-N~fznRlL zvig462}-M3Hjc0E>%aH6vp46KAJZ1mEIj|Vbt*16+~#NHveTsx-yOUKS8~QY30Q?~ zAjt3iR+!HnQunKSD-WGix2Bq2-F2vrRJ+lbvr^+BhT$gCzU^$?&+In-e7gFw=oTI^ zLI27SSSkDYTcMo)O3xf_1!`TbogyjLWva>TsG$>WUT8OvOP6||m!Z04$p27`m+A4C zz%3khx6R-TR`xTs_1n46$llv5JS4D9m78cIjz|4!v-v&Gz1utw9HUr2VOWDRJN_UL zP%zYF3qSgrZA0zL8<{w!@jF-x53j0H5#67BTFuPLN^rq1SC!B8m ziF3GeYnDyyG5T{|PaC2&*j?o-*DAL&rI`GxiAv>+Y!$fWnPA(jqG*5XK^(?8;L=Kt zf5b(Jy9GF(wv&=siLJVf01eaekpp)0GR}7s3KU{ozTL3g?wqBS#VUTdoy4o;srap&?l$68~&^#{LG7Ceosc}>|GguLH( zaXF*KHjF(bb?H*fuT7SXx~&pAj`v&_uVLDb-Q!9>PeDPUy@gj5$uvce);0Ehxbx!slHcr-99UX2& zd~L6*?d2(JvHJ$kyNurXuIRJB+N_j)>K41>@x87Oz8z{yYTK8v2|RK7AIi|a^0d9` zBiz31%@7U;#<1bYAw5n55Z|~~kwg@J6FagA?0(znto>;tDVz|WN|SW!n`#xMy|_=o z>A5-H15tUnTxUCq7K8vN%45x`K5ZZRC3>Q?chLNj&RY%YQ5DZ`U8*=gpe9FmY9=Ii z6VTa{UnO_rvz7XEs~5P*{=(;=N>qfsaH)VnAvA#8)Z7f?>mlzG!hGjrRej>oWZ#zR z-CI^Q?xee}-nYxf8sxEG*1SL1dOVW|55=bUUJd04uXA9xpoq1S)#UFeS312<^p|Vz zmyLBCycT!e?ewZiO{+~OeCBe^z3Ug@`Icw;vP3 zFg>3+E+Bowk)=FaKISV5r;qXV=q{W8T#lBxtu3cD6n%D(L|>Nqj%B^Wf0odnYDr#j zZVqoFX33j^8y{ure9Z%wf8EF7JEgK)Lq;C;4)t@zZwQDa=dhU+C`OQRh^`QzPQ;S& z+tV?0kwIWB4bQyF!3-9BtkCWb7tRHU5b3_*fX>!uO`|?h7Gp-qF z_3G#m-slXS$X9I-5U@$^+^c`B2^RCdN28FZloDROGE%O=dMlCv3G@YAfzy0YaPapW zlL_s=w`EW%=Fj*(V+)O6%FE)l9_ckat+40rL^s)Q+Sc;AIko-1qed?I)I+0ZsxZ|Z11wO+0g(4X zGjkvlL9FG}Vs2%SAjI2~rNjGT`WUpVWdw1_vV-+P+o$9HMV~y~VvBbta`h^TrA@kc zsNRNr)byH)>3co%_)0ODW}jTuNWP1m$ufU$_~8UEHGyhTu9c5IDa>mPJc?yyN`3KN zT4s1W*EnRly!RhiH5Od;iP74^UCiYgUWql7M_%)2tN z4GdC?b=v!qa<;l2u-r2p9dn22x{O1$p0yy7k~DsOLh5~4S(&C6fcoz_c}E0pv3zGy zq>c_})>E4(iF9zFD`h6Wg`&!CCEiz)W9UYsGO=Xlh+6?S*?A!33XT5ap`k;+7H2h( zBor`G;4&)hb5KddL1{YeF}j){7CcCM7@?cHY{>07_)LwGKoxJ;3q`&X;m7v{pbCRq zj|+@IZu6q5WQQz96OVgv;%Gc2hTW%}CcF zGA8Vlh8JXDsK2{vZN>B)hdo=8(`g^tBmsSQ2~SG%DJN{F|X@IfuT5V-aIF ztA2X->-@DS%ZtH*`C}Z`c{(~TDcR%O?r7v@yOkOzC?uVhc70+0b3keJ%yrO+rYt@L zkX~f9?!-RxU=4s9r5$^z{N9zjx8K!+9|5>j-2^Q5ygPPB1O#ADR8J` zKjNEy-}Bs!+LJYwQ59kG+`;%_0>hj5f#IF+PVjlT7+~S1OXmr$juxo1XAm3;_=v0_ zA$D;f)MYQn+Eh-Q*ksKAYfEWqLnD`oX#A9wZ%0&?c6+0gcxag~T8p)GmdSqPRduq~ z!Bh3LM=uNeG!qjU?&M7Ja{04Kb&iWi&cyHc-J#^U!;Va%IWWxKC}+WN$rwZOBNCa& zOL#9N>7UVsJI<(c4qCaYNbhs>>$xd9OxjLqNZo(f`hhaP<#p?kj}on7x~k zoV+r8nALn6=5ClURs77YoS9FH3x`{yNh|KW8t=a7&lXD@DXQ_G+uJ|?TzCe%cXwlW zor5xlVRbzGqFsl~r6i4TqIR6~R}|VS2HQ#oLUt@44zS0CL%?JR%m6l{bsmu8J%@e2 zP_}Gq){y7^)a`vTMz4|lB!VRh}qfmpR$3(3)1saxQEyPfSGP;2|0?`=j z72JC^N_b714)&Qk&25mfX(?wGEFK{42WU%2c$(3FpQ{55`JfwejaM~$7sAEaHsT;R z*B0fLotq-~4EE&@h3`$25kfEEZkPX1AOS-}h?I8*50RI~ohhGb?Ah`x# zu@z(O>-mnqLJa_wYZ`1n5b5=xvn%_N3+1! zHp0V>2PJ}Fjh|l-{oz(fAE*96XP(-J*ZVzNTCI0wd|znt>PZ%-uxm5CQ*2mL zh`JVRu_N728r)coDwbIDUT?^APoD11<(zWmYaf3ANU13p@kzvQPGYwZ6%!5JV4RJYtNB`4qh!(ks=I{A67Cu8T>zh$2#4GtDczFAIg%HAfei}760 zCFsEkuT{-ag^aN9+}Qcj@1)y)>|J@=!2@eq`UAmDOmkd3U%q^aiY3C8L(V3KiJj#i zm@lB$odGIpO|@Z53icGr(^3B8cQC0{?>L5+seG2Uv$7Bg71aZ1(F@ zuXq(86;v|Cb@9awV?arMkCkAnYG27PG)PQLWbRad-cTr*#V$d6QKH%il{YDc{|?xf z>h?d{Y!d~P2bY8Z%jq*O|M&H0c!a+=TB!rPEE{f)#bgjjomv4hL?(%-kE#c#Nxs4g z?ce;_7kq#3ciUy<;L(G5%akb@)3VJdsrWmOVW$@V>}x=ai(|pFLLM(au@R+G3cq-3)YB4VE`9xmio{>P8WV-(n)3c}*veFV~ECiT5l z2-HvW+=$6LCcw`xzgJFLzlI&LpOub1&5{yxLa7*o%B#@4i>_q9LVLbvUJQHR>jAE0wCImA9g+{DI9Yy@=>0MZ!dpIZ6CU8%kozOAR zzVN@<%a&m@IAf>_ zn5E+9G^g9+mF*l!JR&9K)y00|$nk6l=`sgwknREas}{@&+?kXJ!w@*u72rJ6q>f-S zPVGV?%mcS=y>sepBE$Dd*_kz!=k54RiRE2yZd2rP>Pc!z`R)Mlw$m&9wf_r2ha6X7 zWoiBEbga{rfl;)q*fPe@4tUoB^q>3+Pg@fFDY8fnnCoai3);^tkGoSGnk*BP@c?Ot zz~vraa>k|>$VXHZP}F2SMuL^0)(-`nQe{r_iU0h1$Ujf`*Bu>hiSHclt;lfs+7EsD zd|htv$UVm?EQCz>y>?f`mXT#jqVautUaM6_z2VydZ)S(^;?cEa>*rZQ5S)#sIf5>hPyIhzUi;K@mR->Zd2F9#hw{r9cP*4ukbq zdeN5Rcnp`qv_0PS&*ka!DwaM9ej3L0e#kJ5i-OB6;xhqdJnSfiLN}=~Zyw=Uk*4CL zXLZ#pT{ipE;(kS1+&(JO{B7UbZUPF$Q|{e$616bfh%mXu0muucec5 zc#Zt`eKxS?fYbHD`Kwl;pp}*fIGx!XYad3=QM};1q2-mxrWP(9`^~CS>1H6^MVJ|o zfa;r^p8TT*{C``ImpFQL3dg@ASmWM3uChmo_cni;e0$!M5dbVvJEMA)UR_L_TG6Rq zOERxOWgSv#5=I9uZ99jStf~+k!r|G~khmcn0#&q(P>7A@g;%@8a-8~Cglc}K(?le- z$rzZ;eOxmYMIF74VS2^7&$N~{ouV-PpeJ*1G^A9nCJ6NfD2NoY_YjXc;<2~Xwo<&z zpv~)cusPV1`sOFasfy^ad$W=-!PW%k*PN*}A{h{+SiX~w(&u7CfD~%Jd%l@3UHL*o zgL<{G%72~I_FgZ6Ei&ACDJD~C9~X!FPK(4A)h&E~2Dx*;fg12`^`85ecCQ7Ep7;in zi0?mO<}0AM_$nMPBI$v{ks_)7ux*BeeeQ@il?wZf4Ly=3rTSMh;2AeF<6FXD?Wd1H zRzMGMfH$)Jkj#mcwM2uymxs&TCt#0MAm>RgETz51?%w=5%qlJ*D9^kj^ob$*fc>6~ z%mBs-HL|6DC~`O5k?+* ztHJ{X_hm6*7PnDco#3`;L0q$-lG3RWoy;`tj=!<6wFrr?S_ImmK8RCl;cU;gLF3Cu z`<6Hy>jxCsi_f)wLfOoXFrqN~Obdsi3i5qTpeH%Jaz5xOhf}qe;G7W?G{_J*JhM7x zGF~fCK>5FpwIB&IU>`go%#JtjZN`Tkf{!f|dEu?u97!hQxu?|=A6*u5u_$_UhfgZ( z0Gw@zHt~V{g%QV17)N9UI!w)_OH7xs%NWZ*qp^5mbw>#F-X$N&+OLBJmjl}7kiw={ zt*0sVt0Koo^Akoii=Jn0#8EJ3m>vzz-ny4KeALMwzGGW2xpQAaeC^kmAJvN6moIj0 z>ZdMjg9rY83YZf=$dIE+X}v^$0;Gu7*@w;fuia(2+}iB z8m#>NyZeJr;X)pT`d04G#VS>~3E?}LnjBzGOsHT>Q~~d?TJD87w{>|$O9ay^lsPy@ zH=SU32#Tu*f&wxm+?#mzByWxaye=>GbBv(+ytbj!e2+|i6`kVr+WV6@NLa(@<)tSc zP~PBd?Ol)w84YGwCfa`j9I(Djj4ZMJNSY60y-*gP`ZiXKnCQ-w@4VO_#S$Fr|KCrK z6qZ8(SG}()LK4QarjktnIPzobvW&bC;59zwvUk zS|$N`JXjWi&2RZ=r=rVTkGd@2mxSsoNh!**y*VStK{)P1PTIna?u@<0G&gNv)XVmp zCGQkoNx{AO>#V-FTtKuQN=m+^w13itl-u=%aa1L|#ky)P0%ASbf_UQYA?goJ<;Ck7 zacF`8u!j=3Sujc5&HuQ^ae>#>FG_6bR=)woH453}a>Q%Fny;IR^;Y+QpE4oD>4DOpj@n3w(vKN1e_yHzf#VTL?!m!dU$7uNu%uj<7BPCW_gA_ z{FyhcY&-YX+vl0m)E!Dp-tcFvOER7~6pz6PkFudh+`uWqS)h}DLQ)dAH*eN{kFKZC zRI$Y2P-QM@1fEdQ+i$NAZaN%+uu(}1NAc&~XD>~6bbE*h2aV;RbOYK=J$Tfe_LLiq zjgWs{A835x=JQ|0-k;cl0z_u*nfQAuL+2ZG5Ov)lY7y{9mn+9GiY`7<5e#KBx|=ur zt=u%FYN#kGTFCuQ*o&jfy6wh0#`)ntUHaFExzW><+%EPhXqlQ;0S=BKwwR4e3i|DA zPG?^}a<7^(J&J!i-mk46@$@#Q5lC)zv>JK3G+E7`B;EQHGEtr7q?kKW&a5&iQut)& zjF9WE*UT&|8C?yJm=>|}8<{7hv9=+8`4}1&nS2QmHGame98H|!YwHlOxQ{1 zE?XEDz|8eXC-+XqKBj`R$l<59=}^FgOlNpRg-Eu9t+;Q=Rc|s=%}e6epHOif^&wDW zyVJISVUJ5ibTf3led?h>_&=KEmMSzc`y`ANAf~OSI2@sS`hyV@$GM%G@IU9DmoQ3q zH=9L=ZiU<+N&IOiti1}5Du#jfq2&aTNPJP*W=I`CAUQyPnKEbI*zq5?5$L|@UBc{s z5G&foS)4j6tjr~A^vJ1HWN#xZcJR6w%fI?*#cawb(@bvXtBlJwrpSn}5wnO}L{X27@A zCf~l+~yk~8r)VJ z;JvClYztC-dDv|<-kjwkD2DQE8`hv2_5;-2cYih*1`2HH{{9ew>w(HoN33oi1h!8@ zgeeh4JWs20Co*2~<))Wh9i;Ihq0)r=-WRY)(fT569l!uD=)Pv8$o7((He6)cuUnb_-z=Y__Ev7z+pU+UnFHLZ7i38z=r zQ9z}cHQXC~yUx6JE79#}>Z;v!xss(*`T2R}- zfhckT(~BDWjZeRxgB}?WPe>u$_7StZNNb-`XTRK!RFu%5DqSwh@MMl@Ux7>&p;v!o z$w=icA))_{L4aSjxRFmsSoX|vhy~Z#o}n_bj58YttQ4eq0)VmdY(HVq&Gc{iWJVmf zgi9n}S#RCG+a(qzGEmsv%x%hYxau;+`{ijl-F3^tj@#8Y*Jodbca|0h4Jc#mRicHs zsCe`Rr)d1-M8MO$seUHK{8CQf9a}l;*o>Bbf-|Gvf3xV1aEd!nvnzG1yy~0-5N4u{ zBodM?5r57ZHeW8PkxB+Y1RbuKE!V}*WM<`t0AAk;N-18%Lgk{H5=>;-9gi`Nx-NY_ z*@B58jN0qy5I_P=p7!{O7?_6j&N&kX5@U5v8If_s1|DR^We!joxg~f_H+Sf}fT*JX zHAA?(>OUt_!+R{cWY6tM+@Xgu6HSuY=`AfY;y;^(2t31mJZ9=MPk7Uh0u7y2yc1VR zFYB~pW{~=i3B2$MwnhRONmL`#YjIW&X3>?rAL={)S`Q%s;J0oZ?g5imm{;n$MycKE zNBOA7G6as4>|1m+y9vMOk781=3@*NBgp97ol|b#qs);!88Z^(1ED0azI-bw zVSCqTw+S#VKI+5Wk~>hQHy1IiUpIkq2?`{cjyy*%ysSZHQ!@3>;HyCMe)ao^^SFX*EiE9+ zF;W)Airw*2f+Q6DCMA6e!lGB&FDaoCl#PLOXj5U^9ZJ+z+R2}8`{#ukIDP zetWR<`)%IXpLztB%q3CEPcj z+|foaUvrbzQ5%T+t<0(vdAJ|KbB$Ai%TPxzaYIJ2VIEzYBs)RAUrvgS^CdiS5gF7A z2&(1;Wkc_lLIkp&>x5B|FhIinL^fY%d(T&fz2;&5SKs!4VlZag%3sJ5RHm?=G0ZraEmHi(L(4 z0D2N-2#a1aR4{p8T7sc(v(xu{T6rLLWy^>f!j&eZ1e@Zno4KdT*HK_V2_tngfX0%( z;Si)HDF8r000cAf88pO!-E}G|f8n41lT$0S>WdHy52Kor_g(GoecHM}{&4q+yXUv= zZx_`coU1?e$BhB|4ly1oi}GmxLzM8c1=HXijou|&Iyri5^!1Ogs?T&J=(pcLIJ{cI zsv2XObe=Hf?Z8H3&Hd(-I|U}Sj_-fd=VvwKvlP8D?0ji!C}*!pBRlsd9LSnbdLwGK zwX~3f#Cxq(hMa+OJlnK_-@2=e&mtzF>W9`ye8{OQNWzkEAUGm1zer$?i1;741xOI- z%f9#*vcEB4}3=UUt}XoB`-}Bn>|w< zh>npF6!XMkt-QVCPek?Zaovj)Rx!=}Y5*@32(3D&?(K+%#(~Zy6e6=ZWQQ!Wz&MO# zRb8HbM)-x?b0KmwiWQr98?YLL!)+s{xk_^C)jnh3f{ooClD-e727N z-Ty^9$B63)@4UI~Ff$XytT;4XbJOIEtQ2n1%tC*Ac5CScGs<3UW-rtJ?dVmSY$n{f z53#3j)^=;W1h2=!EedX#7%ohM;0M(kd)?Qi)Cg*wZzI!4^vrPXc!%}E43lZ~`Ly=u zQjEqRk>K(vaS2i4U*qkXR^oG z+MC((3i-ZuJJ`}~7#Pgm|JkWhsrMZ5damFbUlu&IHNi!;oBlAj{8NsYl8SNJgKF?u zdy9X(TenpV*u|NXtep$ z)GsLTRV7M#dKZJTL^jHvoD=uD6uhZpoJrWOQYG^K-DCC}$PkBF6>Q2xc3i>>c>AGK zKJ};R)t+s3P`NdoqP?yOfj7L}ADOP(*0hVm6SU@ZD?7pkf5Oa{h2`m|^JfAK@I%F& z*?ldqKSMWQ>cB*UlPkwC7|O+kV<+6HduC@=$7OK0m(=nz9%YfCMgFe`z*#KN{U55{ zIx5Qdjrtv6kd$uek}gq1DM=~mZV*AbJBE^O7(hv-JEa>00Rc%-z@fW4J@%0-`dKOq`EHZ+~4bDn3mR0=UvRDR#67GS=N|`k5x~}p9$(NeW}_!7cIdji>uV>N$?pDo>9(HAgKq6U zTx$gk{l`E%oZFK$!`!FnKkwwL=Cgn_^E;hb~*(Ga#`ndT2_yxnM*nHHXc!+lM`<^W81kW{Z<3YJVqtlV>L=!x$p$%=mv%JWY(I58UNz z18nEzfBrh2a79R8-DXu1d2)E8R{fq98VsxZK3x36W4@u=?;mCtBPQ8LU6)9e;>1*s zqT4=70+FMEG?OGVjli@FxcK<(46`lyuuq9a-6v<=@9&1A*yGoXH_F!z-WaHw5lw3b zd`5s{Y^twdIUg2Aca*foRiqfxC0|Tn?)>!dZ{n5tn`<}PBE7F5;4>PYnm7(b6st%0 zTd2A`z96x*^*@RYy&Wc4s-o1yz@ckfC7xnD8s6Wg_Y6U}~2D+TU>C%f*Cud`h-?=M7F%l>g^6^%?>l`EIE~ zmdj@FPj+~AW6am`;{0jSV-tpEr~MeeBW(<7mx$^Lf%1O{a~$Uf)f2cuOx7%=-!D7D z7?nFs?yQ>o*zGCvypsHMn=cNflD@ds_7}O%g&BMuTGA*=H5mpaKPA1RJb&D1Jg=_3 z*zC;`RumHAyB_nUOSqj&<9Ekp?mW&^P=kLw7<|_Z3f9!oF=>GZSKmG|*Hjgzj=ae} z4gw6!Nh@;li0!!C0dM%xhZQltXvS?{Lvi zn)Z59E~`TfSslBqf5pzj`f_4*>YBWoD4sjf>$P0Ll!P8(Dv`b6rE7Zp-bBaY!v_B& zcd0Yq!ipm%Zq8T*I6fuK70pFN&y}GZMZf7;G)LyURs#+~V&der4*d1@r}F{iBF23> zh}chG-8f-6zYOc_kx3ApbH{ZVnr0Ul_SDSQJPJM2yi zT{;GAhP>4ieJGc2bg{Egvuf!F!MK24F&_sh= zC-!|kqWNWA^!FcEv;^B0vpMm?A5;%qwfE`fd6qMNq|z;#ItBe9C?7eI0ZFzE#iLyK z!zR?zAPic}ZJ1#e_0J+iA7zKAa+&7dm^7g2>}(edv{1+G)bqM${^=Kk`?TC7b}I=u z=%yRt$zg4Dm|!jbDW9{ynD<4_~q_H&5!yeHxGwZ>^Nn{&Z;3Qx&uCE)=K4ft95|5=*aC@w(qe1h2%5sgw zFQVr_U_LfIrW#7fo0<3Qo#+S7nzi=nz@gvEi~qQ_?y%i|pd zGv7PWN>dMV>{ZbIS2ng$Uyu@(#8ZLAsS#;)ewV4`nXzU?7rb`nl{i4LG=xhQGkS(p zauQ5U$sCI^Y=-zA z;gW2G4j0;Y!k_=9j6Wu@ad<8U4)o*U@Hkplm4^jif>f)n0jLh={*d4lsDHnoz%Efj z3u!*1Q47b*@QTg|5e9BQZ~X30B(dtr&0>=_Bj3QS)|%3yeI#K4cV+9NA;MHcz3zUr zQ1({%&;!0%%y!j$F+5TK?%|nB@}#h#Yhg$tl(0P6Zl0jO15+f2l+mH)q zY5x5GE5j7?XAS*d7Jv}+bO<#?*%qtK)mL7gT}EAZe93d zS;+R`?CEP@SXb&7lkUAG-^(xf?&eVH=(lZ^aGMAdzIcG{@n_P^(TxdkOi+9YD0idw z+jqcjFt?|4itLcA4t8n8)odhGFc@L99p;5Lx>zwhQt-#qHc=A)K>A2aJBnUZN~cDR zR8P#KW}C{7{d{ZkvF-xy)kzPwl!5zFp0fDr2fu*|yychX9jaHqtzt;+LwiJ*0kr%W z*j!3(ir+^X`~-HVG%9W%)RIinhfWh^{#+OMgdFLwq(d9rIC)rF*x)4Qh9Iy}eLZ=W z)mWYXq4Padh0P(0vkrUz;bAm(_tOxFKgj_m;`h`Ttrx%2@aMr+?J6a95y$)u+PoYc z$tBt)Zhj>N+gXVw&&!tEp+6^YzqBB zy@8BTTnnQeysONY=V$=?NtB7B_lkV_Aa1kIJ}o^_6!`-c`?c*RH_}G zruC=snk>@xgiUVj?JU%vqCXMcSCof83LTD8m(Cm=8B>2oK}WQHLf#aX988fT%nXe0 zbgDIMSfQjrxN0zSc1)5uWY^~1@zO(5>3e@Lf@){quE7nU{_DL#a|M3Wv~mZa5OYkl zy0Wwpbaxg7AUwn>Pkw~cAYYN_JCGe{8q)SwwUaQst1d%hxXigSk92!}cBio1hjRt& zt5fW*q$4c-rQcqNz<_WgecVr$Fp2R7sllfNva8GvrSAF`XAQTlJGF;RJ`G2jn#1kc z{KgO%dth!!WH?K}W_MVNa=}mo=jAx*R^+@=!}E_Hr7U3V`e$pjVMRoQUga!5Lr0RE zEv7jC_lW2TaMuYIjLw0=x9$VEcO+I0*}vX?JHoJ0LckXf40ErSm0xb{ac-~NkiVIY zE;)SI*x`GUVH_#2$|~Ttsp<{jF*Ht6KQr8pf_oj4q)==JZrc^a;I(cff0^!#Z02eU z)TwRKbX#b+RyuyWoFpwCR62}(viBJyRhS)C7FuA&w8qh^wu}~WkanNt8adkb%@iwa z#ChiUQ~qqfNL=XUqym71e4BsRDwvXb=}P8mXoWX0H(P(=8rm_l5|<5EBtKeZ8B)&V z)E?a!RJdUkig}rN8LDAYH}SN;=EDP&e?@o;r*sOvz@}czruzH1NBJl2_itB62(-1w z#yvHHx!^9%Gggp>Z7oye%CPr$AoB<)Bd)FUV=Yi*HRlZ-hp!QO8vf84m1Z?cn1-7| z>!cFMBBukdyj8_*Uv47CF!uZn&boZfy&L-5Nf6Na;VN3tZ+lnp!yx2s-B}MYIe8KF zWh+P332+s^P^X>xLfG!mefx!1}ys6>Rxh{B~M^aw~aCsN1+K!{t z1I#IhFzEbD$Cz6l`2j3)%E60MV!}q{pw1U;IY{h!4gM8is*Ok&vK|5$FgBOSk`CrQ zPF7lFuKV9MbpptuT+VUw}dg@o<;bpnWzR5RD9WDU?4GuCg{}qb8OR#ziettVT|;;+uhl)19}*{MpdR=R;#?QuSvCO z5VDlDg)*GGasjZXUtK9rVIb&fqI4%cE5yEWV?m=J!81#XO#t~F&YE`{ih3VR_-Zx} z)?nc+;XYb>QIk*0n{7K&o@n6EEW=$gi*Yo*0D{xVkl<)-<6zPMP9krOP!IEftJTyS z+|Ov}5ke*4O8I=Xpdm!UU;lY`q5RujvgAyC_`!ThvazEMg}d8Zw<=Dn;9#%&9^W0Z z;ev5J&X$`wVk>42oi}S7n?21qHlcG#wUJL{DkiXsmRYR!Gn2fYsc~o)Ba$H6>j$Q@6Ia%)@Q0Umxa!W<+)0AnDyXJaO?My-6x7m$rEWXxtnphp9@)=?m+@CT=x zZMO*c&U;E`_*@_WH374V<#ubp_&7R%=K>e`Hn7F_Y*W6nQ0ngej-=~W8a>GY z@mIkhq|l6%ov$wbzWr;H4hyzqgV*-wxE^$t+5My_v6bO9de_X!kA9(}t19K>IDx*qSz^5B;v>sMHqP zQIZ!_VGo!JgM|D?>?{xQ9#}qIWSZTd+nGEZ)$14#GV&-x^R%q~FcKAfVF;f160+`4i zrG>X8$)$wv-cq3!nom+s0XbxP!foW+GsnewZK6D-En6erz}Yj!X?7+;c$oFvu_aCL z$mWXAj-7v+c8Ay~cvyoWe%!qky>EuMQHj;kamUpRRrCWSM87wY`^w)m{!Kh=+D-sfLCQ>q(nz#%e6oTb#_Q-qfR`Q1G_)lz&__t`izrO4o`;-aSk?@BsW2pvOPLMw6Lc9U)N)siwq z+i^c1*z&dsP~X4O(`cRzfH<_G$5vFlnl1J-gZyN zZN2027bklo;Ws)TU>@1{|5Wwk`m$!^)BZ%&^jmI$@OTDkJ>u{?ApT(yPm7XctBw>d zn^7xhSs}aShI&LV6)FXt_A&mW;Q*}OCx{<|O=QI34~54)*sE!ss^zOcNHg9!iTUF> zOsYWk-K&$ZYR?c1RiwFWW#~2ZwWVF|1oOl{vNO!Hu;{+(N5*>^uzhSm=;3Hj?)TwB z?cm+5LSdkr7E;#A=S=zr)$Z;=bpcL)ajd576g27rjV=X?F;A`;Qkdx&e zEmx$O`1~;o-1hqOY&Z-f2y9BK$g;P1_(xEmqC+U9n99MD>Uh4Nm*;gxiX3gRI^c%pP5g*RT&ijzPJ8 zRMKlY*9LC=X^sou<3upgY}SNHQii&MVg-YtJ!#-jSp(Nxa?dAUD@S`ph%iY|CXchAD}!EMztl<|#Jc2w z*%?xO7@T9H%K@T1Ojr`{>%1p@+_Axyn<(t^+sL(nsvD5?C9q5zQs8%w-)z49_{U#X zuP#=Qvlb?Ys!MtW#8q}3XTEP;H@Aws?aW?(a}y+FiP%Uq^sz~>lubIwc{9@%4AQJt z59d4{AbpJWpkfcSjHER@(;gtqT?3EMsSRdgVxlI@M3K;-RG_(w{W#>7CrPnGHY7L1tV#(;v~CB}=mhYGb<>4`zUwZ( ztzZ6vbHBSjiS51cnc~qdRH96krHAU$m@@8x*{7!8@6hx?Yg8nc*B);Bfa@qkMxfsp z^|C@BFdo6w1diJ0jx74Y2TjOS9k5nS57#@;D6r~ANFLsW7bMjogh<2!{f#PC=J77- z9t1&$>i ze8UjCNN5|q>c4f2wQqmuEc zRG||1tzN4gK*H||{$IMBDe?h8wjw#DZwaZuwW&IA*qI5o zp`bo~w>BE{K8)O2y5NH)j`6-}^8z3=|Z&LUQV)z#qd*K?lep&#Ar5XM$APgl4=u^X> zD?&@f`JZwNH7Nz}bTw`0ak<(T*{@@IuCJ?i1 zL{*7qB1c(&&53&70zri4es}?V&aZ*9q`hQFKrgh5aM*;|un<8dO^^<%o}#O;$D`pn zzvV6d!o36I6Z=A+5Yc%mEpVD<#Y8&r-2AwV#_H&3$aE&eRHJiU%vKh$vdXakmGK`l z4eWw28Ng#x)!#mB8QO)8Y(}o@CdLC6F)A@2VQM3Y$+bC&yLz8d&G!6XKycee^kUZ*Cz80-{pWWuyu_C9ra)H=~0R5p}08a z2;d3<@e}?|ImP+wvK)&NqJ@ZSrx10DG-JwJRpgcMR;*wh$?(%ytoVFI`ST$x(vs)C zF8RwER%1i)GghttYRhn`ABp-1lIAW-j51mR_c=Y%Ht5iJngDjhJ=27xA4Hc`;i61W zSOY6RRW}}5McY0iUhvBP@|GF}`s=65q5#LCwwqp4)Wq2G;*6#6e<6*aaw#N_?liX- zS3IzOLMp+%3K|3)pKNqFqcfafMacD$5P}Ow^T9IJ+)u%H>FLj{{j+^c_<@j% zT{K9#c|a!uo>-ZB@Ry&TWYRq$DHE7pSzAAo|iLQQ5>u4NR{ zX$;P`a}ryvj^Ato=5~d%HWZuyon@ZCT65*u7#FnVqYQccJ zQwopzxhb!~1WE2@UkKEggMg6ae)%wWqRH7}3lKkfz^x&aIS81n%Me^h<5)|;&+ou3 z{i7np=?SCX`QY1VG8HsV!cFR71x0eD` zT=q4}`P6jOvO54eP1oay^^5X^5-ia;V2b^_btUb?S3?~AL2sl01GsRX^VSwW;f(Q{ zK;x%g<59aJjtecCmPgcAYiTA!pv^vIo8n0=ZBVo8Mho>R|1kN}+gIy@8N+MNPfmU< zwGT}PXKZACbC2~6LpJ&m7 zkf_GbHI1~lcvzu$LCEYjUG$5jy1BaFYeN~W9=|mp%&i2(%n1+vOT7`Rr`j$sb07HY zjp{2FH57==pRC4=GED~3mg@JMMArSc|JB+|bC8a_$-xU?1$$7bXY*OGgq>YEE(Kp# z9y6R^2f(E}8$uL?bsND>tq{O<}h?l?LgxXAM)BwSo_zkF%$+4*O8w9Wg#_b4ddu)inu z%koE-#a?WGdNw8;rY=lA)EEepsWiPq^GN!qm=*BO%lE6Ar{+*-rtmq(D}5_)(vYQ> zA%v;d0`Q(kYEBb`oSb~`1?@MHsflwU)sNb~d1R^O`X9mJQ7$_>TO_>>gCeZ}8!B;g z2&~+CM#GBkJE~{P{pH7r&g{E|@G-JcDw53C%&6cuP3uEbD~bxvB6Dlc8hQcm!TNnF zZ1@cL;#Ygf&Jm=@%uR*inwJ0K%eXEOPiYupxxH&}lZAeNeoYVL{(M74vjwc5Hh+7oySaRQpz0mvVi3{f*II+V|7(Vk^liVn69zUZ4|1 z_$ft(RPLdx2xhlbtI%?2uqBwfS0420#>Z2l$ifIsp%X%=ZyF(YU=Wf2=cTFrd=r1e z`$<_22Ld-p6uq==*q@i$HiWteum(fR7?cFPw4VBl$Bb<+ge1_QIhViw+vz&CZZFpM z4+d8@ln&S4OC#QQ`N4^~Gj^Hkgva-e_yx~v8BCDS2?u+p62Orr?T4Cp?kS0W63j#I zd14xj8R2#^N>gf3N^)0qGQ7%Ce%!-&p!yRMczk$f5f!d^`m(S=F=!()iLt<@EWdwI z$i&&@Vo*jcQK+Dxh?_GHd?&=uFiG+>3Y8LxZB(;xX+_d=H?8EUQ2~@>nu>%dDynbr z;GkX6Ud~V63f+hZiWAz=(Lpt44f05vz;!p`aZ$$oJ!afmKj)C-2 zieDW-J>~*OS+;~0^m`^t0K;!uBIQ44c`#HK>yy$967QEWZ)kqM{;8Ij^ywN1If-s~ z=f7R`jtL=Y&$)bh(fxz0zJ{WOmp=I>5>wE-F&f=_0=1w%?e6*CBiV&XLFm+i>IR%5 z$JFU;PlXiG?(vV5U8TEiRKJPazX!MU#cB{0koVB9u3lKy~9^zf|s5AX%J=Y@yPAxbhNFtP@D6J}^M#-0%xK*2tzeiNb4KPiqNy(Ii;ku|Yr^W~1uFNCcf%!E0PAX(xDKX|26CMhc_o?BiYMJbv?VaA0xGcV;3mP9+t(Y`i z_FpPqotf6c)|C6>BKqI^O%Ze&0@EPPrEuyQdPqb<@;2x*HW6AhPJ34cdE%?3sIR8o zsQ!dSltF?)$kIA?;Ov7I;2Sdi1-Id-r%(qN`sSKz+U9j6K?@DFV`{DgpU5R#L4WVd zm(LmhCb3CJU&mt3clduY_CM3w@`>o@^TWr>7FWq*{Hyc^0|Q5~^0gi_)0xX#qbKPT z_KV0Ow<0e<-UZ0-_s^IFrBF(lb4@8z1s8UPyB}c>eL!MCTfT;ehFaSn?tVl+kwjNL zuYMi}sn$%)z*!&N^OAb>^vwU)=QqJJrlzL3v9!-$iqb zzTg(jUrT8fsaaw&VKUJTI&HgX^UWfOb+BZ7b#_$Xmxb>a4^iR!jhT#PDBc?z7wS-_ z2=Oj7lUo&p$e@zRN-#il0H3sV0;Kfb1N$2^TdvLa$0}Ge^}jj-WRXEBs+Rcm`)ddte!Czhj0`K@Cq>>XJhoy_jvVyPu1+C$Xlbf6G-%~N+hlx;Q}!+7M5z=4{Z z)mOM~DBq4xVeb!&o0n{i@hz{3wEiTrm zk50}|JiwatjLd+#wb^$_;x!`T2_BWRG4xGP*GD>$B-s)jVBO&nT0g0=hn`xrn#V89 z+Rwge%5UnVvn0m}DTpFZ3rawUhQj2KBK!=2`0XKDP-iNOKU3~S-}iKWjij!@^#AW? zRn<0S;LbYrYakgwe0w5>|Jy(L7zucahdQp-A93Dg)-2Y!*}X5^3Vj)ZN|LEmU1|*o z#xqP@5_yB&p%yFOVyC2~(Y0&eekaN}9Y1O;7Fc|OT0$B0NV3^Nk;kM%33SRN-hKtS zwwcnbdFM>lGb%a99~ATp>sqfS@41Ve3O>787tqg-B?R3^M-}*X|Ex{tjw4ZMQ>nn* zB)$J;%SEo6Ko%$x!&)8@jcTIe^c!#MSCE6u-|5np)&UH@aRK2(9zvMn%S)Ocs{p2| zjU&m?g_lBY?0x*{Y2D!P+N0MO_}|d&>e&b|HcgO_9=57#by*VN%VB6@;?j8K*|Yo{ zRaI4oXh@z=`_mBl>&pxI;tlm>q2|YIX>r-lUkORqTd|3QAgN+A%M^Tg5+Btk%u<~+ zmo{50xU-$vyXl+rXd&IH`uZCoa0P3ZM$jOsgvb)f=^IUUi6>Fc)!DO8@;Hwz$uC3c zmtFq1fT0AzCP|Tjr(~WkAf6W>LpE>7x>O(+mtY|!77G9#)@Kzao!Pf{cUVJXo&W9) z?kT{lNC1{=I00;!8P%H-ee9)|6aOGGZ57k+<}gP2qIvCkYOyWle6in&Lg>R|tE%^I zD!bj5?^dT|guxGfB!Hcf0!BSrDY;1JbU&)D*ny#DnJ%@EAX%@enWwT2xh6iSx+P%K zEKX3SkC{1iQhZ2eSW7Twcn{{QJ-t-oOAm4um>w^*P}12+Wf7q4oT9!UW-5s6$ecVU zgbm=M$}O~d#?SZv^-?9`mf(%-|8_JCpA_!K*OgTrA|W#TTtq7u{b3-n-Rj#tsmoSf zj06~|vFob_wY-aVE$E>136reMEQHMdtZgm6DWE!rK6oeT>HNRHb*%WIrqZcrGvXVpSw+Pn+!P1!T z0j)la?{}xbHJ^vhdZu^O~5 z59d9;nTxko3H`9ciJz8Q#}Cto}TxOOCD#UmaJt+JIu@UcUP z*d){{#XMBAYx9kvhph}0=;c&^cCCyYZ)xzJgVgX8fEK-X4iTS15}Jl&ewI9Tle7XV zx8Ot7Q*dx5M)hO345=8*Gb1m5{C7Laye~rZSU;1IT@~sAtBnXFeEkCxHuM$@dL>av z4+J81>@{mpNl(99_-x@@tW@0Ep+Dib9|S%}R8>jFbhuh|v3_BUWdc*11%EMzeysp0 zkJj+7zSHNvaq}q)3)flLyd+QGAb%oK)^xVg2m~TNXyK46q)19-?dDVT=}|1ecPTqg#P9r&IgLH4%m zzj1o)}YCWhC@g4iiw+2#Pyht(d>(-)3d zV-$zEUuaNczvKXCUviEv>>d$<&vij5USyl zOd<=ZT)gz3D&>enQWSL94?nT$@Fl*Y4*^fVtx-_mLr&scjn@^ZWyIy>2$_&+Dz z!IaSsyOHM7&%Fn#57pHhrJYlirta(X>e%NGA3h|WS^0;h=%V|it7l5dEW0?_Y0U}` z#fu8z_-Lq6%2#M&E*Yc2_g1A%13p44kHD>5I3?x&n;sGCUu>Bt4$o(B0vO4thzLIW zHUJ|RUc%53!^KGPYPQml55F*;n`dVh3Fb@0u#fBzcS9ZKU^7&U`|hP)UP|3TQ8@^1 z2r4u|Yp#L$e+N3sj0T#MeK!X?+y(vSwF#+t-PGm8%Xg|R#sNwIvF5t6=|^CSXBqbxaAW~Xo;hcCP=Zu zyBqYJB@2{sOmj%1nY2g09Dee{rnGhdiYlc#&|p9X9=z$4oa^D?joJHrYXD*e$;GyRdQ@*hULKP=;)Z5x_zDwW5+i@8TE*92mDUH$r z9QwWIZ&m5ovmdXLjUT<@kjqU$T=)TSE2yPZK2~@V*%@d68r%IcYS(ME(oQDHQB|MO zbFe(~4VJGXauoklc~XjZ-b^GhP6sANa(@!f0tVU~-*2QDA@SmEld(Y? z<3GS1w15zWbwB+jJw0O#mWbg3kvG|A>gBA;zG6riCmZXs7b|_b4!voKm~oiRurJux zUf*E$MKwJMRJoek2JXSXGOf>e^Qj&hS9AD&q|_}(I}k|;H8wUrR!8VPv?vqvzg7hT zOg`fzoCp13u2sk}@q-8;E~E}=u72Sri&_fCOLL=OM=Ikjv%OD^*(Cs71h)O1h$86K z)fGTTO(07R&d_^>DIRq#(9wpl1?n%mk$WAAu**6LF@WhEY#bYrG zIxL{4{KU9iX5Vp!RdvHk9}d_^36m@WG~Kfu$@OmK?@E01sNpjDa3o<~zc&btoY;X) ziBqV|<^q@%KG?2}*cZO)juZ@rlyj!MB~7L^NRk7iZ?#mw(b62OFYDtKR~5ZB9jyS-8%a{cDl{DWKq&1EfV!So1<<^zt)@sQ<* zeeB|A&G0R`7dR@zNljnUZ;pbl)1wA)83vB|=rm8QugP+yq_{9+&KT0gd;E!KKF=8v zRXNQx{`57@`wLVQn<14&n6 zxCb33ec*wt7_A-U55@SdJ$(soj>=ah+Y-&CBuPUP8mYpz_*J^PpRTX36+C_YJ5}ca zy*@q(=lcC4PhlezYdW|z&Z(tK!`#aGbO#jQx@GHw4{`;(0j zsQgtHt~{Pv)c)sW;{7|2d7CP^ZeR$f+?xN_Fbk{wPqP`GqwD)Gl1vR!>4kNCsPD0Y zT0K`D$MP&2nJOCA^>ND8czx@^WF|wwtxvgJ=gmvFgrH2vi==8(PSqxPd zp*_(QW7#F1dN6o@hexASh8``WV<}we*&tLglCg!2qWhtDnU|qNwXR`-^9q>(3^+))7}Guz#0gaSZ@= zzE$cNCGvwkCp_VC$WJ>K(f%zSg`X|#zheGQ9) zo>4vH+5NZx_*Z5aB>TzJU&Bjlj~}$G!-L-~K%sELZne>2MqQ3zr|kY_%*_zur%qjW zG=z(3yYnn)EDe>~sn}SJdaqZG*VtNh?bk$Dt1%D2RFMeyYiL zgy#NIc)cz9kLs!Q)rSQePdWqqZvRK8#$57pko`T(t{~;eiVL1~JW4kb;nINMo9Dd& zdJjcB5vv%a!D6&wHTY`}O-D8?kPFQp%(o=h5ox}G%E=vplgqMEX=RG3wgVVjU)lVk zFb)M^Yxx;nFzjH6Bm(UzG1slx5=ON9*J3m?^r+mr!#D^HW^mjKjTBUpXu}e{%u+}g ztDcOTTo<)3HF^mpNyjr=DBmSo74&k#bRqS}bQsygI~WIMvYl^mkWAHwht`AT;spXE z3eM$yJt?gr)!XB)9Nl)UV8^KkjqxR?eRN*nWWVR0Q#f>*T$^5e<54Bb$|1`++c+s8 z{NHEa)n3rIz+(Kp1iTwY8_R``UxvKLFn>uw{u0SFCwIO^_tmNxIwq}@2>0MhywX)` zd`j48Gf0WY0PEa9B8XZ7f+wr09Z<&0L-fp@8)nB=HuEUtB*h4W;@>@_l;$qe%xg8z zb3-{0G1a;S9bvQpWhbJXEL6P=AD8*v-T zNu<5VcQ}sG>wMnt8mK~sqN__k2>B`&`-r9ZR#RtINjiv@`}eMhpRWE6Y@QsI(Xc7I zSPPp?kx+)}PaRv6xU?49cjw)ZCi(1&o006tt<_%N-w4;O(ywRLh7oNPX+uwW`m5(_ zSu*;CzR3P$bQ4$99{NVCWWeX#@eoQNqb`5{IaPBa&p(Kfw87^(pkC<>GY*RGbZNI= zfn<=!O$!5K%LU~{S>dhpnr}~3-bTE|)_i_sFwS^b)W*YHVx97BWFu-DzqK8>82m*=bOlV;Q z2b4ip#6b!T`h;JncrHB{0;1+F9?X9vFh~hvLqlgt`It+B_OE45lJ8pWK(-Ani#zJ; zxb`lE;iI;Gv($fr(Lvmbzrkx;LeCK{@&0Fttb4d4wt`dO zg`IY0bvk{wD2phTcEdt#s~X(yt2bc~MlgMrxC=am-P@-p;^65X6D(2|rQ2vHr#EC@ zTINrC-jmtNNijk~f@FF!S$ckO_fJEQ)_$X;am=ZjRNKLup29I^`qJQxWkW}z$7eVvKd9+OY7FF8TO}g%o* zjZJu}=RCy&CeWmmATFz_N`Q_Xz}h}scQk4$dd<}=^5}qU1>en^ ztCP+6iwEk)NB(@+T+&i9U1)74xuNFUButWib{`-QY;1xsqm-mFB$oJ8|hh1G&z59FV>}>!Tp8_f(8`p6H$v z=epO({!cRozVqAtdOa)Jo$_2ftp~{Yet|xK+m40<)SiVY~A+Rg@I*{&sarxRnu@%CLFv&wEr6((>g}+yadRm;+@sD-DG!?g3&8t9z6F z?&f0ln6&fhsQcOwDXNOKIyzY?W_>+qA*Xub#s<)iZ%!VB^JfY8`EC`YCHY_%)w{N6 zBWXuTy=q1@^_>Wd9(V#m&H~+9l1((r#b0n1PP*JYL+0@JRn9$s^qZ*fZ&YF8{(v_) zRcb?87Cln9ay?02SoEFwd1T6*_soAUhRTfH5w?20tEcB|@3;4|^wW$mc?_->r^)eB z8@#GT2eK1|yt3AGlJ00eajJgQXZzNwPD@O2n-#$838xzc1BtT2HkAR)Oj39dhm*F1 zsos%cKQ9%fS;mn@;O{2$Qhozk3q*J~?WJ@qwrZ0`qs_F!N^K|+I!7gSBEchk{K0g$YXKf;FcRW|6UOo z&9K?Hc;0QxTO=;5?C7YxUpo)pv`&4VXa_yK!`LI9XO?0ddHC2a3%asnOXzGIe%W44 zP)5FbMgm{`-)%o%8!GV^g~fbhG>gR&gG+L04;uXg*ne2O=D_W&h&+tv!6~h7efZ}Z z(0y?1U%dVf(6(zLdhmS2yo61cs9u%i`5Co&@n4DKNoOGBACeMuQQ;^9Qxd(?t`q}y zG_|FOk$_yG6vkn)u&xJn+sLEIZZDK1v3A|?KoCKsSG&~JmnN{`ss zK3`F(pX1wXV!lroqJCi)1s%zqjGiazrb=n@`n4`2O*2dFP|C6O=Y0kkS^*{8iFeW! zRyr&aaD}}`j?MC=Y3p+zC#UOFw_F|YD;zV$2|l;rgTMq*!U)Gz98RJl_P?%<)X+Rj z&9O!^znX9MXEsTy!G|tu1o^3c=r0C}dD61Ez|6*q7nCnyk(-R~Jo6=Q%4Ey<$dOk4 zR63C2>&4UTPMPf4 zb%!SOf&#Ui+uaUbWK zd9KhSesgGEAeY_JylSPG{g`%<0n9wf=393%M7x$^hLj7ORu0b?BpJ)!=hk4v1{q5X zmc8zczPU+M*Nri1)2y@BJgY&y*@;S>(i~7bT4=k1^6-8!P0a%U1%XF?X4MoZ$~|=) zUL|bK!;}M}MRBpd{=&ag35Z3(X2VSn!_VYy=3RU$>+G#`0Et%PilFy#3^1DuJBX;^ zZ5jZsSWiK~fwmCx%N!+ueR{)gMoLY(B8%cdn){Mi`rT|L4GEP`Ywdn6ncEg zLxuj|r&9v#cHbDqW*BmdUi#3_m+pS=AAeNXFXSL+b2S?(^z%@BgWyu~1e3GIQ~ztO zM{smop3^~36N&|X)3{jKE0U-5&&`NY#*4aopkzgGR@>>! z&(L3Fn4>yGPgHle%bt)YQ1w@@%`Na6PoesKY*+$=U&MhPw_7&p%3pza($C+>Jo_E5 zcZRg;4Y;D#NxNf9H`}BD)R1}M^^(HSe-m5@{Z|n9cCJzIKKhDbj4g>=(GhKmP6(%c4X zch?QQI4v!WsmUozGZ^AE#;2=+f|DDHq{*+wtENRn(rZ7Y#SAD=?-1iKX!;mx``&-~ zW6;6LVOZ0k<@Jl#EBNF=34CVVbQt+QXWK5R%X+H@J@k6=Uk%SmbFW?6mo*zUz_e4J zu!|9!>`8U?=d^jr_iAcN3NcCS8nT4%FY@Y3Lx%aK+v8V*mc)H6Ka~1i;xa+VFTa3e z^h>*mCDBiw+goe4bxNd&tKQH@a7lpFd@QqEITY*bs~!2RW$R2=#nF{BNWyvz2Yn)O zfbh+kmlhlKMMo3UWzeI5Q~oSMG*`^s%V$x4?;AS~RPS6YEolRYSC%Wd5dS7fQN8f= z8y&cX+*d4zQWCY~^>Wn^ZGOk>5dOPw3wwNZo}R>9q~o2)L)t`j^usNeW_2o8c#$4O3t53^X4DW2ChQup1sO@v5shqEr z%`L28HY5Rw!e@JhywvJNFb#e`IKZRV$!cwt#5wSHl6gtxKR!5Fe7)pgHII4>vDTa9CPBpYDp?0lr&Pb8RQqo_g@{)*7^0`<2)oMqaO7sPIAd;h9{uGt zkez(yF!7?x3+0cD*|b}aM2V{kSBq?{hsh6EQW*-v#2@%8=$I-ez8ach6%KZ zX<`EMpl}3I^f0SR+?jaB|Suih9q=0{UW=pt(w-m`3)im2PBz@|5N-V z1FF~8l(+7&CVcANHi38BdZ}*KRU4wH1uBn@ZW$$rptsJ42@n~yXn*LbF*Ui{BL2Gy z?3=5J{hwc~cb8e*6r3W-#~#re&HK|$9%cI5UA?#S)$=OIBwUeW;jk%vu`bF4fp)RP z_Ma)HNo~IS^|ZzH<>I5q455q;-@IApO08s*^R_w0Slvf_2MAzgtj|T5`u<)BcL$Pt z&wkW?yjpl^rnCwN83yusi{0bIEV78@d*GURT=-q1>=Pq6CM7%s zKoETh%lhbUMJ;O~UF8l1>h-jP1&+Vo{``@44#SJa3z41)+kxUlLHi&2FW9{}P{Z7< zXMN5FCkbk()eL6?&sh9LPVj%z8!qIjhLyFp%fC&hxXfok@yt^XSZ>Mt1$=)mgM87n zWm2J!ON~)JPK$~Swa)o(51XZyL8S*3t|CQF|AJuAR>ziIoxA%R@a{y)Kt!HYD}5k! z>&0FhR8+M!?H*S8+!q|3&8upHae$@t7V!?#*6Oen+#C4FWFa^u3cKDot4BDR(_H5u zphNgTT?y(Sfxs3=|3-(;d(hsR?6UALZxm{@czSG+mRu_}U~sM^2O~ zu81EVMtTp1t@(h|T);i{c-0fsG(7mnXFK5i-=HU77K*rwYQk{n(Pe&$+4S0MB$eVP zuixXsg~{R_H*~fX5pV)bgXcsxGc*~TL>ZYAC!;t z=AEY5K2;sA^@#B!B|0`;`wg*IBta|*Z=;f)0A-m0wbcl<37(&6YP2XyNUI3`)o&@4SdEXuNEvTF#L?wURG5SI)kS0ql(OBT#8h8J|v$;1E~ zryB}SKXryqR|sL_KDQzwh(2zmJ^M+OU52(%Bt(OoBS1Sr9dpG6Ap?H64fs!uG2wC` zP-GEb>M-In4)gIBz_Wc5FhIcZ3B#GVOp} zdQuh4`@Mx4UwJ4INu^!Xw6IY5bNj=h>ki*XlJIYpVs^jXWzgS4?dDtEIQ`h7Xs%I& z?Q{KQx(07kzigzc+he_pnxYmgh&0?0KOc(A7Zq(%*S4C_tjH6K)UQ3drTM%iR5aba zKh6qqr9Bq3bCi?5Mt!k-e+q^YmRL(wolmE`z$_cHcLE@wor{Nv2ob8#;R= zQ+#^4z3|vrZ<_g-f3J{b9nsEh{LHH@F%ZNk3C7ivmEDYFafV+Gg1>ylv}-c1Mz=`^ zJ+4fHTbT&IBE7fzTO01-K&04ySh~36yWl*eILyjHu5@D7`4Be8p7#5=3hy^5iSR>G zZ5n2W_J7UGm*F)0@#bKIuxaZvOSw$AE;L(R%%ddGmhg1RA@WobhBrEgQrWl<+_-zs zfBA-zMBu#sS#Ra@IvsT1=)Q>A;{C?o$CuxNv-_`7ws!?W&F|9a{nnePWAQW|%@3!R zTvaO&Mi3g~)pl*mG2omP?ao%be1rV6I6sWrKK4qu^t7@4sY}Uh^fh$^?n&mr6a3^5 zp_TKSbpG#Kxt+s=IOiKgd;#7t8Iq7IMTVy5MDk)oMWKW4lKd$U!Iu=Uv?UqeWf~=_Zg`!(1YiHAL#7UBIZM>oz5;pFS+B=-DTu1vpWvr6iaL= zhYRFQ z8sY-yO-wI(KBzQB9yfi3FAR?udIU>#VU$0%`W|ElT%gQlo5~J`bUXz9%3_`6cAyV8 zBb!|Y45Xx|@bhyVJm*Z8@`{bJz3^_tv6r*-e^8ycl)MkOB>2`3o&B4hvGJyh5#4|` zDyRE4%c%#=`4G~P=w48=JI(9r(Jc>ZMI3p1l-5!@aA@qMg){NxJ#lnNgd1uvyK zi_nyxyjZSAXOHor=D-z~P>V-i71E~gx5Y-w1qSpmZ>EBf@qGh^Pno#VED%J7QWsz% zEkh6}`ZpIvKv|LCoA6_FXD0N{D=I}a+Dy&}6>XPol$MY$4im8&7@R?818XU+$}SF< z*sj&Zf=b96g|HfX=v@~PP7sxUl)aKude_9Si6G!qx)dM@K~r7U>hSg|dNrO{kjSOU z8y`c6A8c+e9Ncld_C)cB7s(=I@+@Lh$OVx9C!a;4um4am{#yU{>^aNGH{U7jMaty| z4S&>oV1+N5DeGtY(H(CYf`|bfk+-=2mGR>+04@{|3gQp)ToYGs;dm!J##Ux|rF=+T zm5N=8ZfLiMY`63_yTsCyD|?n+)#swkLNW^6l(d)C9mK_5cyFJo^Icq}XgRXFX6(8d zsCXaUA?}qz_dCrVz^2*BPB%h2ApiSJ+Twv#Kt!5E5HFp#^<5UOZ0LxR!y;CrnX7JY z-mNA{=9S+HV&)~(?n+827>`|36W{NOb<1sbf|BIfl7lXpPm!&;1Az<)QYaG##2f{_ zfKizJ=;Y!e;kT!xX&?(CeWgD(GMchf+2%NA`^aP-RbxEW3&UEMiXh13@9+f7a&sH1 z${+e!%P&K!`4NM$(^EiHzz~LG0MgU<6d|?Qcj}ixe)Lq~qMcs9f!`uJ4gCRDR|xPY zRT@Anp511_JmQ6R1*6MiaJy+av2XW@M^a-kiMwVHVRZ(gd3ttHT@7515}(aCP;;Bs z{W%vkcE7u_uc3mlySi)oCi}-c;}i*nlf{+yrJQ;9efdFAeEf>$P+V_13i=cw^b?k_ z@^k3xFznF$6;TlamR|TjKU^t_RvB=~y1Gr~>0OZq6WMHkZomd#|LG^*Up^wLm=VkESl>htU?SXKSO!hV=tmrjU+bdt=Up zRKGAXiKIWSKo^`bOkl2Rq}G;Z%zK5#@3=+58!Bsl1&2SH8Q|N)>1w~Sh`vs*Id`tf z-YC&nBR2L`CRGrP0k`|G^)zAw8)cUbiu-QejL7@t_*mX|)x+$oXY!aTI+WU@z7Ru( zN?hE^#h|)cw2*?6DKbq@=CDp67b{CA&4i!@70wy+z=wpi$^<)?VINLE7V5?4$Mp#* zmv}rHaILecesyT<#bg%NNsWZ-b%BKwkdimMek~?oiCK%KKpAHe3aht>^rgSkuk{b& z{0~K%(2py$GlIKgB3ge)ZT-BjS5P{Bx`$w$5WlnRLm$srkt zW8gd|5%43b_te*H^J4nOn)Zij0#I~$csAeBQNSrhf?eOAe(?n{dW~M#uv9IDHPE17 z#pgld5^)_H`nMNUavJ&9LjSjPTzN1Xy>^L=-x)(w!~e@c>~a9cNOIrcFi831K;#~k z+gGFOdJrJz4k|};FzpzZ!2XwmxJW?txmrLUy6pb6K|}og@3-59M%B{(^8Anx zR|h4gz&}5e%zu*-1sk1Oxxq%Ch}R|+4|6%b70cl_Bl*b04_gnwT2KchB)YoqYjlTX|^@IZ8GT*6LOx)k6rO5eP?hhl8o`C&{rqhq@N9Dfiz_eM$ z%bxoGz6My;dlm$H{$-Dy=?e0Dkci0z$b7&Dw4~SN3!D^a6}bsn<(B7F$4cQ}#}zOu zVfX?1jwsTLW+q62unuawlDP42bZ^pB?eyW!*l^q0W>H5W*k>qUL7fP`1+WFT$N<4Kd5GB=3 zHw-5vRAO3~oz|10HLzStTv;%h=WU<{LBqbhr!x1E2_RSe(|Lg(!n2Z@zX4wfAr?{@zgAY%4wkB@j@b(#%ljxSZHx zp-M}@j4E6IQ{Xq>Rv?XzX;D4&4;i4H|HB42zh)B32#r}y0?+1ixPQ|`eC@B0RUf4W zt@%ISe;7*V4h}Iob*2>DpdxcIC~&Ojcm@)mfU%tb`H%T2t$1@St+{k0*2p1~gTB+D zZpNgZYZTu(BJ_f*g=H+rNVFy%6N=@$_RjgxtDATCNc*`{p@?HxT{!Mi%*h*=2Gkko zy8aDswbhJ_^sq?CSm)va5~zCmG0pCOU<`OUFCdw6Tu9@JvAo>`X~kIm->|iiT{c1O zJC0XX{Rrf676qNcgky^<*k675D9KN8Er3mmHkQf!~aN;2;rr$e6ex$`oY*fS0n(*+|SnJ;4PCd)?T5u&aQ-p z5}P5T;Xa?+ATqoCBI+w6#KN2o(be5;zNJL+R}`#e~&?VQu6v?+JS@Um)F?5u zo74hK2F`kkZ40Rug=nCdnf>xVPq3cI$hbnVlE3GyvRDbbb^cGv^9li;xn9&noXdsN zZTI`OdbX#4wVbdY_E}f|`dTteODWkWp+zTybJO1#n=}}HKw4lqxWghk@Z9~jzQ|-^ zk6BHtqQ2joo>_|EaIRpz&gze<+^7HVE&TI|D}#L}^p`=f6@AX##lb=f2)HmkoY#Hs zuLrUzQ*7;pN7irlYq8T{pF`K)^=V11k72(a38B~%eF%|K>Z87sXmFQxsj1ve}KifZW$-wT=Z_qODw2h^| zdGJcEH;&cp0MupOWSsvV9lmd%7pRARz@lul<*;22oR1^ z1~iC))|Rp8xTzimdReixs)`CW5mdA;h#}zZfVfYw6T-(|>2Ff-S#-sLFZt-ntLRtI zmz2=*6dr3T<;}JJ6`2Yi3+V~KS>Qi8Kg+fJY~PK*uJ|M}B-=2tBJ;aH&-P6~aJ68m zVOdyynGngVRXJI9^LuSR;%iUBaNo-p*e&kI>6^UJ(F--QWhs{1z)z;>5*gT~~+ z^~m-DY=F7^#P18*Gr#MGWMysjQPND->@Fstb_tO^d0?P`BXEKi$V%IUbV`aKqQ?|m z5gK)yD0v2&oU7ZR^yoFxYc&IFM{3cRFZN0_oqj@o7(D9r{^#=xV!@;V&VQ%NQ?nr= z@#gz6GWg(L?)>q7_=ho7ykGt!mvCH0meoOe+%FdNt`PMl3|cYwb-LMiRrr6?z|cSr zgTPL=`UROEAb(0D((6V5AM-yVMn)9I%qR^a^0=H$EjYBP4TwsGDm>(TvQI51Gnq`D zoPa&NmzDH|V(g_epkn9Sh0`R$V8#Jp15}jI;eqXHwcVc}ApHMu7 zZ>DsCQbvN7+K;|qjyLtmRZea}UMJv-CQk#=V?!1ol;Xuvrqz7o*x>%6rcmn0zaOd- zWecmbQDmU$@7p5RpwEsgj-I|G8lUgu(fdA5Ryv6 zvGveJ#$PGCbnB94C1jf^)WBNJ)MBy_%k-0d7LETUDWr#vhvrX^7AkG*2Q=kLJ{D#V z5D-m4@JE9fK*Imom@h}Sr%DKdK7Upvd+c^(d$ouIB@3T9OGrkmT4T=t!+D`dxTAX*~4$n#R>WC@2Q^0KRvMy^+Qj*>oU=2KAx~?|&B$H3Zb7acJ_Mh7o%P47~Eb zKi4`QP0HHE3W2t=9>0(DW-*Id@KI_ z?OXROeus>-G|3x3#_wdY(ojLj2vKk3{~f=gaQKO!;yz+ApZ;qpFwWG)@8yDJVkv0xsZX97Sw7`k!Gu(HimW-?6%!%0ny;*^aCiWH-WgD2 zn`?|d3)!4D@K6}~3;pk52L}ZSwe)U6(Zk;uIGsFPu-8wjxPCCIWHGqidP|oOMC9CF zW;47jq&g=#W5Q$6c=}78hVq#**Ui#Q4A(=6*7jty_4r_|YT56zjZucRmF=?g?R~Al z{4n>ot~n&}PDW=v=x#&;Fo7;TWI)rGr~)5L%@WM2K(r7q{_!!Vudl&7TcY;exUA=VT%jh9Qn9j`Q(k~Ri(PRzYjXRtC2?GrKU1ulBUt)nv)y>lG$}Fdtm|wRk{`TCJdxe}sNhC` zEIqn5%GVdp-1@2unY_P${pQdBG^&LFJHkJ|rZB~X`#q!mzdh%HvX#@~v4K`uh_=32 zX`r#yTX#hvo$`1eAV}U2goe*ztHR5N%<1hN$y1S>ay9+^Sh z_o2k@^0dT7#!PR&V1$2|n|%K~s6*ngnd~zNXg9OKF0j-u)o)2?b~#j>C{UU{`H4M< z2S<{AdHB{L>D!c)LHm3Y4LNh;A8^XkUH=BmHZO43=a$j_-|6`3DSTsRg?5!ch@+WLy0lNSivmHCGCkSZ7%1d<@v8|;4+%BI#_#DiLJL);khJX4COGh9~;ly zu>bUFA_7ovUqTon;h2Qg$o{IkAT=WG#Aob4*avp!n`*)wN>0LVD=S~Z1k`IcIO$>+|!YE-$nu&k!_%iqbUdw-))d&g6ny0tAbgJJGZ8SEIR{<`puS6FcD0#rq(t!S za(`a^X;bm}3z*#=xJ+~2pZ^N~cQdk}bcU0h@!&CPX2li^D+_mx<0ozqE6=QIlU`jB z{PQSj)B)dcHZNE{j`=7jHRIkMkp&s>DFUAtYFCMB@FbnFV!n8pO3!xHKhcUC)zN5w zMMFiU6cQ3DVYDL;i-@Qn1lsh}fDaGGyvb=DmDvNayli;PAFr`0cf$UM1HqagX=D_m z)kWZE1wdk(=XZO`+U~Q*QHiK*EieH$WFy^EU*r$O>)sIThBfKKB|N2EuV ze~)r=MEKT6qxE)xgt!gxwVkSBWf@HlowBvd5o^j9w5VX|^5Djj8r+M&{!NGZb8C42 z<;l=n^$crX{cKNtM*Sjo(myS|*^8g{Z&f2wn^;c#2d4^rHp5@&-2ypfq4-zq$$SOk zP<~HX<;oBYTbIw~)aOn?)T%51U1}=VCJjqNJ}C<1}gb6P#t~*DS{x4 zh^gj7Y9Fj_5iS;827MbBZ~~ugw9j$YA91?zlotF1^KUZPAnh_*+bCD|3^YI!&Xz>2Rt}XmQDj3OU<;i_764vo<*)YT8&i*hj^v zpzCIWW{O9C`9d-J;)Q9o*%YPOKczS9D=s3059>Q>3y7JZuZlF;21&VrnX$_%Bm1#) z^%8~cZl4>|_4M@VIXK2=%2JKYex|An`QS3)S4wy=JN#p_7eysZ`0vUn0*q!1GU?$p zfEUD|4TtVp-E?CzYW&gDN#d=1@s#AaVPWq2zQBchosRMI(vvKEL&v-kpM~ux?m|HD z-Z}Zr>rmrCrSioxyz#lb)MaU$2EbNiOSCHTLAanS*;e4}Cq8h;zaWLz^razq$Q!a$pAPMmVB=%py@~dQ=im8F zO0@7m9D;&6y<^rN7u_6rC?R2qJ8_b2D>0UzF+JL&ziCP~FTO!l!1etcxRw>@HWaB9^kuYcLZ&GC~o<h#r@F6ysg6ALxi)j++Q@1zFU>C8fLMh*bu(j0K{-$!Chm$D7NgQn9$2s6dLVi*XMlY+O>Fch$?f6O;T_!w)lRk@)M%{Sf)GisK@v!ow0 z=(l)^I#+?%Z+D=uECh7bvxAmHRgi^>J;yx_TM4gGzF9um<66!(0}=;xaAwh}@Zu2? zO2b^b)QQV~qRfcGb&>u2C!Oiw=!5Ck0rY3j>KzaQ>rQ~-i(F;9u`zTU0s+n#Iq(Qb zA2l()5CQLG$2rz2^OgK7qUU~&VXz#XylSk3bWM=dEuPl@(lXJxq`obnw|K+iw_fLhH8Vcv%1#?-Cp;lX}>G*QCxEIu= z-uRxy%Q_qe$M{Sou2vd ztcf|r(STcwLg_k+!GnDfx1r6PZ*SS-mY6EkU&S$L4VN-({#!<51Gp5&k>m*NHNd5u zSu=boEN1G3;!7tdw~rz2Pm_cVpIM1!QEt|qQerv3VO3sOS$MR#&g$CSy7CL>y4saE zqu3xLooP~am26f^+iX^e#@48HkD%rF@KZc~*D2e9dv5&TyvBUu)tH!K2A>?j zNum}VUJ9?iLXF5CJST+z1l5MOh(O=u02exrD4hC}o38MOT6e_gloq=6qm!I4e_rI1 z3_D~^d+JImxkv;dy%tA3yX+o|1qxTFzcgVUAfWX!1K3Oz3g1{Rwh|G42b{(SDX3rd zQsA2^YtG_me_@Ca@@fOF9m?Rov}WY%V$m+s6PaUgyqCHX>uzb|bE`L?8wZDMWl@AkzxN>bV%8qT(} zDW)|oMz;Ob_8`%%;x*6UM5t2zdH+AO8C4=bP%gRy*JPS?B~1cehmjSQRk=nL>J?Ul zd<*mxU_5D%XuC*tb|02g>oTWm9OM@8L5Wb-C1-Upy8b5p$XZE-SDneAY1wS&c7R;H zTx>5O8eg&%OmR#+-Ux$uXOGNyxMjd;QswXd7Xy!=?f}` zr@lv&Fd$6_3iEo~wsBeO$J+k^Ovj^HcgVf3vhZ`!1EQWh=NM-~pjBNIUEc!Q2ck=> zdaFP0=dNRem5+k1yrcfqS!z9>m#ctEctJ%?jG?M%@JCd(i>*m87?YUCDif%32k3`@ zIlzQ}KLfhi(X=RPh#9Wi^3_ie1)A^b8#9@(MSvRSWV6)Hk`qEzO%L_!0NH~tnoIR{5vY*bfP+=RL@YZYOg0C}9Lyh9j|kv7bdqJ`u}+judElXLE#c=0_aJP3^eg zPN&dxUDPu!^Y7HV#xhQMMQi@LkGQbarzipCbBr)4NLm+qtH755HaoU3C`?h|e)Pb) z!DVaTFx`ji1j4WYF7*=>vt`1p&7zS-`D2m05DBNy6um=L?c2-r?u$a+{@% zj_uJ1nw==XxE2SPPXWjq!1Q+sCuH|m%n)SB>7-uu1c*vqGq@X5zjgvUwHG=XBq8t4r!d9BcCG9HwZOlGHZha7B;)%3AFW4^6kB1?333^oi#Wj^J4&Ie?~-tdWO zh6|BaZZ5q65dKHm00V4+)Kv+nO?i0MVUuKOc$i#S1{k=+E^$wJT?mj(+6i+OUn;(v z<@dk|(tkqJLE%$;?4a*(YL}DDHm#zOR^MU&z#o*}@?mO8tn1$Gx*xH4szuP6q(w5} z9XxWvZkxwtJ}@vynl#DLcsKm-j2acClb`I61Vx~uxq~+z0$tjSP55?8g{hCd9$}}C z@l;Uh;W{8@M$D7bB~owl;oP=Rd4Q3NTiwYt)N3x+l0IZt#D$wxJs30NQ^?<+mMB<9huT3>50yiM?^^?CH} zW@M=MLqL{g!f!+eqcDbWxYE3rucqWa%ll_jzC{9#UG4MCqz;1LZhk(ekEdAf9z6FT z(|zOIg#Ub&TIt!&n?4qN7^f+9K zz9a=*!(1y|c2X{#gW{9soTf}PFIAE3Em4fYhg~REt_*XgkqD9qLRRbkcpf?>9Pxy? z_Pu&4tfN&Frr_WZ98Hl@G(5Rr*$OnkSwaM~rJxAJziY6s7&wmPiqT_Gk z*mce(iQu_U@#~YC)wP}#tMgmivxO$g!#_86XGUu^$YXl=ep&z#G8`PlI7z) z`IN`Bq^(mP^ILq=p@XC?=MPVp@oH^{JazdtEsH?q1G?)Jjepc^c9@{(=SRWSdIuXA z1;f3<&??oYT=~3ETey@H6E}dUKKO>bzM8mCoMNSTb82+MkGJ*_?2TEn5Exxh^ z1cCQ-<>@9XqRe07%=OdtdZ>`rhJQ(S?QY9{&`jFANUc)#9z`ftCDN3n|@h(JaQ{mhq zwF2yMcLtrJ=XD1ce0~^x3AI~y>?~=o(`^G=l>#hTbOTV70UFZ%L4wl7jG?NNyJKAAi|9gf2)KYb?MKTuM; zlW0n%EsL^_B|F~hm8|wqU5}TyMKJBMxB#b!DCyJZZ3R1`?y#p_wl5#p>`*8{N~-f(O}Y0V4Im@8|4UvW_!k74t6WjLR3^##zWo%gZWzfzmooQM!sLv zy)>YfDlv4K^lvU#k#W`Q(+6X2Is073r_kx31Fr0h`t0qc-b3lNMmH#-&98)i=*tf; zD@KYyT_slV$q5NRY9s=5gBtQ}>u{kZ5wpt;6jCuC;6>`fRud+|iF`)=^#U~#Rs1dR zT%_T580Qyj8eiq-2A{m!8_GeE#p*xjq;&M0P}zCyfw|7~ZZ4bC1&h3Rx&zgktRcbV z-718Jp@fwg9}J|^7`4jf0fjgZq$j8Wx{;|7$KN-g3yCtog6p7jWv7$>CX7KQQ0d$C zeIz!7n^8t2TsA5=GEWt|d{b?ah>f}|2YAx)Di})X1X|vmBFuffUmM)Kdcn(a#A z7`k&=dr0Iav?F-WgfI83LQq=!V;AI$LG_1ESs(r%$VyXB!bG99hrFt%-a-!1y(K?nh)i( zyV#Cpbe!R_?mbX(c$#6++#~w0vQz4`ypJ)*Tbon$Ek+_sZ`uy0jQuznT()&l=E{2R zFqNKjMBL@b#QZHJp%`WwE5p}zuU;d*FIJJBrN3)7@bOUDUvyBTP!##{Fy9{$eb5yN z2p2uGZ`KAuHD?q7s_9?$|1_0$9FT7m>HJP1fNQ|adqhLU$U47p4p}!#_oWz&(KtHf z+7-zw5vVv`%v}7A%+9qjp;&8nDx_Oe>@=tLII<}%nDJ~uD1SoAWi(5_q}Cz6lUukd zVC!Ut0=VC(XNe+$_mrxB0vhViu1*@13GJ75<`TH?p%ebAIMMUxtH}%D2MbM1H4Ymi z59rG0+s!Aoc+=)!e_Gq?JID6}@ceBLfomwgl3ADlGeHr_YPQlC!8pLx z)m3_E6nK-#JuuW~^903lPGCXAK0nCHYb9d+fUu zE^S=R%(LtdXf$lD3fDi@)Y0W%d1 zkN;GYSkUSZzhnXYkPXJTHuSnM4TvAq~@H0Y%oC ze8i7lL0p@I$MIpOnytI9X`1?`1b^8_;bAZ&L1x+L$`jb30g%f792%8=aJW)mIJ^MB zH1PGyo6_|#O^)7;CQmfX`L}~pAH}?{zmtH8m`u&<{@qBr+KI5|^(nJdtMH^S!z0~@oZnmYEr5l)Q@x#Nd?YoA5i2%ac zn8KNLi)|ii!1wsQr*9IcnJgNWBR27^@3Vb9M&2dQo+9@gtu+#eB(6Kd+dNd#g$Bh8 znh8@z?tF3~udnqLw{lD=Le{VDZTFD2jq^=IpmEtx-=0sJ(J%nsh_XvXhC&PPPppIU zWBk0WCx>28zHTb)R$JJ;(RmHS7q7oLM1~1Bi^9n&K)wHT_y~0s<8+Y^X9(Gbz|Vil zatxfJL*hb2i_e5eFsX=OEmYYJ=&tR*Rkn1I)wX{)viDfYCg5cc#Z)0KxEl?=J3%ye zI{Vr-mjK8SCn0gAv`CBwsyf9L0gi2&_l2(8jAxgVE4Z}Gs_~tqk1#$d4Vh;4Ox!u~ z&a(STj9QDcH3)=?ay%T0rHix}zBNOEdFVneC&jQ>)MYd;OYO0yhBm0RUEJ`_>WeT9 zum56$$;d37DY<2K+{=qo`ro?te*Ens552|Cky(xdYLiknrn*{~XwlkeZ@kj{F$}Xx zkl9rFqf4Meswh`(!1|Z%+I`=7Fet2(GX$>voi<5mrZifIH&}ZN;s;G-*POp>gbF3? zXE~^ct91d0pOSV>!rh-#(lM)doG8SCoEY5wE~eVpeH-;te@3YJpI zt<<|b3pbn}LsP(Rr@5mS){a{WZOx{lPr+*)q|Ut@XEI{1-%~+}mP#Q|Nt`gGy!5(&F>2`9hQ901f(RuEpz; z_2l07A@!yDkQ(Yyzyrr4TH32teN%G}`c{tj&IMu5$wsMHlCuf4>p)yV04o}ec_oTF zeAgQ>%2|+>>R&>R96m;!;5=b$Y^-{n#Ob2gb?Nw4+-Zbyeojs$YgU3mDXL=ow76qV zc|~-m(DlYmr9Sl!{frS!h_D3Bj;_Pe^IP$_k2xO9JMVG*)RWs|1*OdMC<8Lw{#JDNZNtO29s(p(QH$(UqvFAxVf~1pHNzh%$1cqT zAAdzsK$e$PN`rqgbK&%FhX|jiK+!!N8&S#NU-a9Nv1>0gg`ub<8K}w*dlj#r`}g!K z0`!+Gng@ryqGy{2riMkKK`~6otf;(Kv5q#o4B;4OvTLS0qs?tqaxkhL*r3355`G6Z zYsC<}@`cTurPY)Ouz~3!P65;+T}YiFn+(_m|qkSTlqe!x@Sv zZN$|2&N$o1I4xBsLY}=h5NOjf7%39_jc&NCb2wY~Pe&Ti$LC9<1&GvspmvD$HF0%) z)do1Cz(6I?%f@3CR?W!=&_vwq<-Hp)N%8(ChITp#_F=)9oV=dk1x9J+u!rg|Vb z3^>^wVbjmwohmW2C-scFz@%5nhC1!dsOa(pFQF88G#rkB#%q0JXPefIvnvf%Zj|KV z_ivSUiMy?nmC$Dku91fBD)_tv$&;2O3(eRgE661!3LO;%puLu5Vm8lW(9<)#*}-E} zW>voM`Z(o1@#+0=@>fqZx?lcqA}TfpC^^ewiFqb}i)Vyd&CHAJ{#ww(%$Q4$cqKA8 zClDZD?#Jb{qi&A+w3(oPe;H-=4M^E^*ua@2)PNu=eGqvr@}?$q-HzC&FMdv)nx6N~ zk5$p84LsD;?hpQOQWsElyvRR1KRha*~n!2V0WE{$&zsd0J*0hd} z=KG`xhBnjkLs8d<-S+_>=j>&DCSDba=_RV9EVK9^;^ofjBE{h8Z_gwo}LB-Vr6-$Mi`X zU!D-9Xl=+{`Oc3+&e8p-FIQq73}d#S8y&cI+jo7bFN7t}u9ATEbc!S-96W>^YdIhr z7g+%!-8MG7_(qzLk06Ud>$iM`6k5Itpg`#Q;rww$&^&ryO|LOH@fQ<;f@jDr4sJFW zY{sToytB-xaD-3siyWnGQO@iR@!Y@Cs5L*)ElXL?AyBE1@hTF&7G|lq?)>d&6r}9C zv#=<^FWcBMy{DHre85i=U$9J8e&cg-fS?$1^6>X6a?t~sA!J55N8#|p75abIh`G$BCGq|g3BG49cl^Kf|E$=Ton{;b#k9Nu(mJ8*V( z9^^bs@^Zf9RbnX0E6(-Z{8?c(!|W8S;vgpOx9BP(w8Jf1g~Xy#LXQnWp@?FEUwH9w zUg5L<0E>~mTu{jAFG*p&*StCjz~XtxAy_Cp_Hwz`m+5p6o49I!O|Co+_5PUDr5V4u za3Qpa|1-^oNmjRnVf$7Ih)?rX{${Xjiw+fJ6nN{8*FHRLS z2UjeehBozq(L(cVI$gSJc8P$-hD17rukdduiZ889nV=Mxk!((ksItbP2D)$&pe;D> zmNm>`Ns+7|g}mA=t(h-ulCeqe!Roh!e|m?RLd0ur`g-r!r+(mrSDB97Q;Nyagt)Mu zjW3+DVz#sKLI-~_+@6KTJiNATv@^UNdP?oDYdUx6Ug`B}harXkI@D-|XTN1uFgV?Y zMHZ6z_IzRC0N6H^Y1j8JtfvW?CBIP_U-MdNeeOJ};!~^8Y-_y0GdG;Ckc3%U@iju< zA9ZuflTn;i?PJA+w@N@4)F`LeE!1w9ZuQgsk z>T>?es%-P2#KzicrFML_po(p`Mv?axPy)!Or^HftinkR+;owoX+C0H2+#)mu^#b#( z{j!Edj(fOEnO{H57<{!)3UX;T%bFYNHb|43mJ1Ebi~D}vo>vcK_z<{6|8lqeP=Vn# zAn1Mlouh()g{3-pxOwE_l|_Zo1DjCX8YRzF(=nb4r|C8SoOi|39D262>i$J$^WPo$ zYoC9PWM~c5(XI}`=5zz?6}X!J&!~@P6p`o zwe-;fizN#mOw>U5k-WYKNO%;s06b``aWl3Vmy8a=mT$zlto2rxY!4jPG57l;9rZIO z6&@q)52luY&QyZuvscyE46hXcRGM&psVJCm3NiO-F@hNJ>gc`5LJMIrJ{D*lr^B!3 zfckvPKXa%5Tc9u{Dt%=IE19Qml~q=_+qw5;txU~tnqMM=JXxnhMHk5j>Bab_rslnq zwtES37fh$?nFa%F+)ha4f|66$K#Tr#ugWL=fG2A?#y?|yf|y$NpgF?lm)F-KEz7d) zh>{!PsRxa>9tUhoiH3RIr&A>%d*zmqj)wk>t&R4QkuiQLCIK1AWx8xTIC1LQ5WrEd z;T(ST{l178<^29`Oo~Q9UyjU4&1!%4X#m3c*4RS6V$#kt13$6*p8OfMzN%AhQet(M zN8Rg9Z=czA{$Z`7NjjksA*D6@>%F|LTeS>JAOYn*m&@xk>+rtc8mGY$Px#vE+dMqF z6hvUKg@gtaub1))98ME@8CrtoKhEP%6Xx#72O}PK4I?kL9V~kmJ{khA=55-L7(tx# zBu%bQdH2=qjXMTkt{XhZI8wMQOhy~#L-3AUyvG=8Xzxr;cjFUilC?UFy)Jc%N9;1r z_zE%$EZ%Ovw!L;3X6vJdolA1C+_4!j+c~hv13`!PteC}aME&eM+&1$w8hWEHO!u0~ zm|ORybc9{l&Jhklknkt!Ts9q5-RT;Y@69yuqgPS|Y&ZaKMnekw(xEQ1 z-sAwUTBfBlT3GEHU7sa1^>fS+;Dkeaib8OB%4MwaBXZa4DZkm!K3KQx*W0KMzKh`d zOdjATYpe*&UMDYRE!i#$sMT41RcA#WcaZG#Dy904t@tCzxiJQ z>J}f1>%TvHy|*k`w+Oer2b1M=;#Gq41URzws@V+T(vCqD& zgyasq2roAH)4y|5TEe`I+@^%4Se6CFG&0FY2#;=$3YWaR%yxJg=yZPN_;fXjXaS|i(HU?e7G+I+*6wb@0P-p z%veSKqH&Jb&ieUS!W1nKsBrAr1+{mvrcUeHW{FT|ag6uYP0Q7KqMCKxjA+B%?)8-b zk9My4qhEBS+ph+J>tsA*zD=~KLi4Y~r@&Ee(_g50QG=7FK0!{vjGI~$=h#)okaaRl z(ZvKu#^Q6fAkh$A8}gJ@Q#DZkF!P$^x>Q*4-EXPwPk`U^pR~W`+f20viQM*ec6Sb6 z^=@ZfoNh1YLN5!QzF!E2KGoi3b*EB1j(D@QvE4kyohd}))SDrz+SH*XW8~BwSgp?! zx0GPMwPyo)X0q1gT1H-KaLu-%KcQ zP1a+)IaB7fGv4!dtXZ@5xlP%vX}6Ji{BzSvBH(Hl zM84zRQRqSEG>w&lw%rBdDX66vfc*rAr^99aOPePgTtzMxF2DilY zlD^AA{OE`C2UVT2+!Xm@`wC5if{gYon&hQ^K-M`x3J!1kvt~G?;g(c#fAe<(%g+-Z zEcvxRW0C4KdfoPxf~8Y7s==&@i+ZAgBe_oCEoq|SV1 zUcE)t3xPJ7wf)KjEZ(_2Kw4#+u+rMKv}0dLQ{dVT{hCY@Ny5jWEpZ?H=BN{)+}z+L z=Wt+h&b;GlaLt@+TSQ`D+;ZmPbUmt0#OA!IcjDrB&5 zhibWa&;EYZ9@qk09DT<~u?Q&ct-ZmAEkr$QOT-0?*LyDsj5NYW25%mJinpT~r&Jvt zz9fnC*S(SBHeblj&K_+aiF3Kfb5K?LnJml5mhOFH-m;Y)D$f#+J9D>i=)DO^TeoGj z=&ZC2CG{Hda`16JYb@zJeuyn`Ua%%FnB&t5Ald!95ZKUwA z+-KPjxw%@?w9w|&e)DemL4&jto^Z_4J9o>gyIeKN+(d-mk`z9OpiZ&3xx+v!Q+SS84oR&sy=krQ}0v zv3?C8?P;}-_iCL5tpl{=IJJS$iM`!^Krm$L%+C={5A@Q_TIR9-KgNdnmg z_A=(q?M{V%@Y{$$#$KD*ffyfBQA*;-6 z+1vMVy6SyhUGMAq>-Wd+d;8u#|8#S?U0&yPp5rm^_xr;{3o(W>!+fa02VGv2$1e-U zu{%2GnThu*A{DiCvSX$eU*FWBPB?2KNa?iYm^Pv(>69`>=E2X(kG4s}Cd#fDy=Xm> zZ;{#G;o4?lq~fZ=U1KO}Yw^%9cs<8AY;CGq&SEQebb7bRdr#9lF6ABRjs1@(J15YR zSf{1~c31hS-Z=GRUd_ZfW_qY!l*X8HtZp${*sPd(Jw5Baw>ZSm$+qJ?i&D&{v^FLy*{!CPjM4ZDH{WGUvgHbOYPIZ}kmavaF`1KZ5z27b^ZS>N zXC6M?e?*c6S&}-fs(>hXyL?TbSZXH~<7r5T!j+sDnbwA%B$+00X zeW&BxvLP`tdNhrl#4R}8QA=)mitB1r4kKBXCy}}=sE*U;3?<8t$8y_t4EWu z6PpJ&VoqM)*$r1^dVRBd>kK@cH(;G8gXHpc%E9qPL;QWgYVT9tnea{13$39ROXwCvU$K4nhHs?G)I6!{zKkDNMh~@w1Q{g3~R>}TD zdRoEJnd`DuJ4eOlROT$iy+wy5x7&dVV*$FIcRQzic43-yTIbi3CI&F2^nbsrukr!) zeZpio@ztKU_7$@t(>?jgfo$TR3B#)yx3yb)Ma z?Ip|bX+=<@ErCr+bzFaCpocHZaZZKw)d;|C{7R?Kfu@QxqGQ(v_K>eUKg{zJM0(`$lzF|nObh)5W~rs)bR zAXS0ksO*i1gACY`JNE7sl{*o$$9<&Y57?IUT3U~df^<>(3wfoN;j}!qzEEH9T$Y3U zfpnPS^7nUiX_AD55;k9hHbci@!qHRvkjR;ai_qi#V(+EDnB`G#s)%WkS5zh{eHYs# z$S|FTf8)}Ej!=a~Y+~2F(xx*FdcR{7{4s)5k9Iu9I?ia4ATD5bQy;^QqjjHwjjHUPh8j&`j%OPcJUHVWTy#n@hXp1FOIb@Zef zX$CQhoBMC3S5NMAC}Q7SY*L zTsg|XKxRL6$!D5h`ohtx9BpLCw6{%{CFKhO{^|M=k$j@(AYyJ=YSrNUr~64m0bb+P zg_dI{|86mUZ0Y}xe{4>_ZWVt&i22u@Ze;Q0FX?%+yWSv)-JYpjYIP0G^=Oj^ZAp?+ zfGAG5yDaf~gFH(Cfw___8@R}8c8jaJ76|z4#?ufi+igFJGLlY&;y{^>I@(JXa-w~u zF#cJuhC~TW4G$?YTTtP3(+9lat>uuvF77zW^6M1+zBd?##AnjQGH3og-7l?u31%CR`q==B z=!SN&^1Og>-`-4Nn=KeYm+kKj0bFex@fy^=3$0@FJx^#kmHmTD?E1P_2i@<8ZGKD6 z2Az(L_#OH2J7AwzL04D*-o1Rgs+De8m5q*OAEBhKq_i{;Pq7IWUyUL#0|8kpG~JmV z9Otq4p=2^Grn^g^=Mc$(Q7JePhuVAp+?`r-Uo)?b_iV1CwC;djSCKwMmKY>9cR>s) zTqo2ZVHB?T;zt$h?a$6frS!dt=M8Cr;5qxSHo0UMy0Je z)tktzp}pOSqZvAx5iQZ@8@NVot2sxK)eMTS^^f=^xX%_x0{_($^5@)M z_PuqvKA4R+19~lj>m!FdiFxw~1ER1tX3|sf;_j%NHAsm+K$Ie6gmtoxI>gQCn))Kc zKzBeNyWF8HU{j!WysiL#qms1eyMj(&J8UX5y6XWw1eul4dOaUx;vx1CGl z%i|@WBR{eFhDxwUNwDe`OzAELncW^3fN7Hb2@vaD+D9oZ#R&4LnTS3B7;jCR?I8g{ zTLM79zP+8HyGX~BtJgVH_AsgO zCA&q-tpuSY%VWh1xI<<|R`v{{IZ8*Vm9tpXA*C~FnSH_!#zGaUtm6;k;!ga! zCZ8D83%g_Nu^#|6G=cm&3w03U$6ijnJXp&XX01du-B5tp|#CnVT z*Nx}#@Kux8cqiQW4z^g+vNjp`=4FOvlx`w0d?!scZVKhhSS^o}D;tY2B5x_2FB>^j znnw~prWJ6A)i_(dg>Y7sTP`Z3l2m_in*AKUN2=NKIe^Wrk!(l_e~Y(|1nU(ZbYb|3 zPz4(T8HP?P+pN^l%(lvLJm2(0C{jUD&12)s9bTQ1S0Er(QSFDgzzQMxn^Wo=>E2~Q z{{8=5)B_J?YWLc;wXP-z&$oH+7FrWgQu{c4y2e-@8SlAjAiDn0OaK#1D1Y0P4!G2P z1=*a-Q9*I)h4w3V(8Q_V@KAniPtU;^_4U_VHK~p_@YPr&<~tyf)k0#M?jKRMttreAk4SxZfFu4e)Xj zF{|;2Eosa0yc{Zbyj^;gS~eLLPXDvy-iF%R1E}d%;GH@j3b20!<6j}-Y&S_BV$UYN zz3?{NKFfP=cMjoc{|Y>=F;tq~I#)OH|MG7%Za=DF@g9o`j6}F#72N7J#7eUn$w37$ zd%2G_0d>57x4jaKsPUfdYsq!tLsE4dxDXw;DRg>Pfp3sUyH%YZne|Neg6;rRz>Kx7 zlZoCbA3LgLRx}ohs?D@?fS@89sCHv!S#t~$3B2FW!XR_hSAoYde!j|`H9#1?nQ)iaO8H&g@H52;p&ya)q6bc@yh5|Vw1>0 zDttGf15DBkNzBw??P95$ose1Bf)YisT}f}fcAqS#q@1`l5j(Y43fF@7R^JdI%x92K zQ;r@Uw_d)XiT3fEqD-+_$c7$wE$-x*`TKa2`_Uu~%&x}L>YDbX9vg{ivy^VUJ3I5< zV1%(`Oej!Zj~N6L+MlR9b-t`1=eS?^iL=s7uQnBQ`S0ya*7sXPnv8G04jMp7gt9@| zU9YbjUli;aXALF!I}rA4czQLOJRu$Q+OiXyh;^JH?BgTj>9=q_RQ8994oGvuS6wB= z@G>rsea4??G!4QGc9z{6ffIrfJxN+{Y5{7xSd&@Uq`D{LsB@sKueWakjLh!ScI)!vM3mHzO$)D z8QRc)gJx*eaMQ;RSRd>?_4{NH4LP-hdXm$Zd5uTkHm3f>ms9qJ`o(gv_x4-hZ+~Tc zy3A=IGnKxFvdOu(!=L#W7HH2ML~XW=6KXNe-~%Tra?WW>WW;j(*txQ7rktf~_DlLw zaTY+>>J{m;E#!!#yt@rPw*(;3gSfjCrj3j5K=|?&jNSFO=?EV)DHJ0;!B3NGO!Gt* zYS~U1(!EUt{mTBzK$p!vbC4eLI zQ>UBLQWM8PdKlUA_G4o;^0R8RHnkzy#ktBQKw@OxQAlvo6udeeU84-s zYwv#95zX61n6zBSrYBu(_dun5HZ8{90WqWF zdJDM3b@ihd!%?|o_hilnQ$REq3?_Ramd^n-K?1=>IV&_3OK`eIWEXYLN_!sp^pKb1 zI*G45#XCMm(jV>woW?guJV?HpU*%6sOdSB~43`4&w`3BKZQQ4J{Dk}(*;sJ5a-3Bh-UI%n{ zBd^ zg<=Kcej2-jb^TOI>h&@rnhJul@6V3L<0Jr#M+Yl~zk=xn6rW2AF5@)6&j zZ$<-%pW8F$B4T$;@yp)s{GRP-(El)3BebB_7Tc>+B0`o8MfIB|3aKWuqLR)6Q`rsi_}#$r_LH#6v5d28kBJL;<@& zm!^2|!O`RB$_J3BnE1$u%6n*g6%6>9G!q zfh$$%YQcP6V3j;NMl_O|3 z56*z9$7Cy|EpI6PqJ@mF$(du2wxpba!lK0`-7`+afoC=#yTIcrFO5gtG@ez%DKm+? z6$P@2b(8Tx7`v;;wA$~e2Gx!@@#Xfbb8<mmCYVu0UPdZ`)w} zh!~7q&O>E0!mW>E>?S0$M|kW4jPx7cf8uyq;zJUr$Y|p9YWkKZkj`C-bA^3;FV*Ps zpCyywQ<~aRBR+D9Oax(=xGT&N;&mBGLqR5~<1Slw5b>xFLa13NZo2q0ek-n+!)6w= zoM(sc&2bi{Wa6tHAzyzZe{65OzLt%e=dFUX5k9W0@=5zp#zwp*%nBHA3`Axw0|Kb* zOFHW9VHS~(PD|HzIC*K@AZ@~?)iLwPie8sLaLY(4!Hk(_T(o8*(5|pW#6IPF&EC#@ zZ>ucNS~?3evLMroD!hx-G+81P!&sZ+mh55RCv;-ve4TTU$rMQzWcyFSo~dX*8#}?B zXqllnl`AM>-A;lC#Iw$6+uV7^N|LX?}F= z`O)ajwHGR&4q1@-JzJwdOJ!!1eqQRy?vjYI|CV=$@b)4kcZiu zCm4^OQr_xP2?lm|AE@BXku>ruY4;iJ<9vlV+L#iDL`;hfKyf#-l{A;@j1<$WvpUga z=Uw9>do_q9KBTZ($Pk;JJ*JrgB?j5eWyb~RlUPQEpk8NQ+)NIZm_lV|s#QbpH08?O zB#d3Sw~C5>I0&|UlP&D}mY456YcF}Hw$acf-JJg*ExD!zWnD@Xs6J|`=hI=%= zc}4rIc1s`5xXNp1^Yrb;NNNm%&4_E=dPp*9RqYA~BS|$=D*5!*2>af*f<#|W$X;2Iob_fLVGupq16t` z65aBBrY|aMq3&#~wBKap{rmU!3lhbu?NY2$TUrYE3i4HrcEx@EXd6ZZMCp_DNt>lQ z3B8nyf+kquUX{_7#ZBKF-ylNlem_|TAr3i^5kBxeB$D%sllz}(c-J!jN5hj`ulvqj z=ly9MgD^_M7IJL?tBjq@%aleP2p6d6_B612&n)J$C}LPMcVDige|pbAkD(A_65pCB z%rT-3pA;Q%3Y|utw6!d66||5+-sYe!MU}ONS6tyWI?btBF{n>POxi?wa*w}wX7lpZ zM02v`eZyIWtp!k4?(j%($2>|!nSyS1!PcjA!4`sZZX$=2EPGYyKBi?xtCgPbgpX7> ziCY`Qdv8~v9$m&h%}k^qig3Dm)q4tiMY1bYWm97=RQ8$Lsh6X8wb64~z9BL}ffgNf z>k#f5pH-B0K{(=rt#+&keMe(;?H#SmlERMENK@q0t~9>N{s`vjeW6K0gSoDgf{v?) zc>l34o86%0Y%LVi7-K=Vh8aUrar=O%trO8&~14o*I zzArWl-uL8~FUFe^Di``NZf#(V-`vf7Goslb^%goy8Cp1v@O1T%)# zuwO}M_uqV}OMNzwWCzCXVaLCJXv(0d3PE1bUb&ty<6^)D7mPE}5-B!Umb}%;ro*y~ zPLFpGl_OMT7-=-^zTeV>#)fgrB^(LAq>zVDAKKo445Mw73&jDKX8zJJgZ?!ejG;!# z8~Tj7fHaaVq)S%LvE>Y(Fu(82iAspe>B`YJymyxp28wcylLgb0;u7o+^4v|Wt6r8+ zZ;>&Rl@VY;pmT2TD19Tm!cc&r z0-SmS6lW|&iN;Pv8V{3~ok@tykW@RVufjyT$BWK+A>Xayt}u29?{=NnXy^8`!8N2p za?RpXp`^+KZBT2|@%@a>X>>b4Ur-B5!N&abxK{N-?Sy52bxxto(-9%Ee!0A8cF9w? z_@^q+lDn&hNGdvf;a#<iLG5uGCOzbS?mlX zox-%KUSm=W∋-VCo;MAvbesfD2k-NImBLEdw33c}#~URWH}n3qdHHHjrRi(ukjE z_jRt}CkvOs(qdiRhG3*xx0g%59NE4wn5AMez5?Ky58~dCu_}wo-w2I>l7jny`FY44 zpLm`|Ta(%7Y#DR*@+D3;v*T{)WQ(5)S4(+AN4N##?3$v7D373zI&7qv$CI3-)I9xU z4U9kebE`ZlGUXN2sw0e+ucsD5*C&%anBvwFrPeM~ODNxn#W-pWYQe!@q36-LtLh1} zq14Ui;@#)&6~K7?nkUT(Z6=-ADvHFMuc zchJW7{lF!B#JtZU&XE63hN05cl6`kS6V7=9|0{4>ls9{)5CqB{p7)xCzkPc+l&Z$A zJ%A1jnzEf#Lg!*PpB3@YUOKYY%A6c`qZjHB#)(qJhV;jDLbE_GuSRkIPSmGhW<;c( zcaou;iW)*qa$BHs;R?RIM)tT)i{`*2F8B)Wqb8fM)Dabp%#-8SNDFuqiGnwYx`P*> zs@N)=@!_bE?N>ip33D{_@b0*rMf6hfTd1NT?ib#K!bve9a~5u@B;oknaE;OAmitV8 z02PdZ7(FbT=$0Q>HYq3dSNRi6X% zrM{6aU)8Ep+Pmtw@TIQq+&X)-j^|v)yqZa}VC90}V(h&b)@JZ(i2p&cF*RM)4&_>KjE)TVj2)bkASY4artM|UCZ$#`I)ts|Q3E8m~# z$^U#)j}Ae08F%Iy@niHU&JtcUQ)P9mjf^X9Mw8}Bay{=lT5uIZn^cXL2R{OZ5Vol* zbe*AYuIa43e#SY$e;n)rJKu49#?t~IETyUS$YBRQ+aY0fTN9`}sDg3NLZC5VM zFBx0Sr?fksK%y}gnmcez%q+o%vLDnq-V9tIAnd7T8U#Y;J9WxW&kw53Pc(%b9-ds0 znOzqD3jIpQc{R`Z_P`@GxlOB~xl1^;EG)(MeU=*@ygjc|<_)T4vEKZv8ltblhXWnh z=doJREaBb3C|U(~ROcRe1K zZCU`h1p7o`aXrR^pl+}%SniV%uP#e^QW^T2&q%t6RA_^6Nyig8zJUzNtY18(CdM)V zMZUSab%UX6&!Cbx&sU~`V=tYvx)6G3iNsmn-wA4dXdI>btT{^ARGKY>6 zw)2G#NhtG0{o*p*?Qx)$sT5ldQ?F_olqY@$*pFv1l@QvNXD@yT3ZEtkPq6or3ZyN9 zn;tJpdTENE2)_n#SS73H@+bQ(<}WJ6i(!JAU#ZsFQByWTE~E~)<&N92_M2$gq@3Uf zXpEG0<=9Gjdg3NnKvO&Xx$oU=k*?w8r@bkM3EY{v{n){^a2$zRckBM|;JK-z&T>=p zq6@STRZg>~8`ECp=nN9_%bdRBWFFAH=`Rb|fZ<$a8-$*4KfbxGi%2M?iY3$v)zW!ZZ+Kbh6uJUW?cS*n zY8jA2?{MPG!Fm;r%$;eotasEpy0XuNU$#q8M!K>|v%rON7%di?w_mcV0qil;!3RkK zzQR*w)%)V^)ic@Ct|%SLonN_~;MraLQX`XTXnWvJXakCIp$=(hP={Qh&XQsJ`X9_T zWA2OEY{h~v*9IQq%q=#n>BOvfJ+z%lrH(*b8($`B`{b#a8AtV`qZfHFrJ zCXFe!w(J5Sl2T1kNrv1P>K8FFO`F>rkai8+e%YkebRf4LkjE*Ugr>RT?R{BBOC-!M zrOdVo9X8a^1*3zhKGmdFx7Kfk%8v0rWETCvcE4IMH?57CRBop{HAMBoS5mD{*Y8VWH<8gHcGXX3$C27d6$+RNuj!o_zd_{IUl1RWekU zwwC|wx22ooAER@&$jGLbM@-rn-f7_vW4&_>gC9j{S0i6?lMR5F&sV>A%w#sogb+Q8 zjXtn<`w9tMFaeTj&*bdu*e<=lJn@s*eC5qeEgH4tV~mP{s`nGm?H$SQ9A@rO7A>j_sFdjtBYo0%BkNWG0VCc2(%ukZugxNN z|61sV)+`%t>iAeBIEQb<0dM&kUp4seKJibTpQtqjXdwR#)!62S0@`H z+`|tRq&{7dYwCUEW7T=+N+2KJ(q_d?o7D2wHIa?vQGpS}@J{5d2R*6Rn z9TP;^idO4NN;-kp%c)L4M^|h%2TqLJc%jp>XXsA?u<8Ys2PPuh?*Y)-?-gHtds@pg zhc3TUw`(PU9RG>h+e8mLr`-wD8`z*?EzpiO&qqGzRmVs?_iV4;G|rPbLGm<=cFNmu z;wvOljuubTDx9asHSlEo|nI~C1wC4yscCjP8ir2{V(V?a%O?Iwi#H1L; zL~^SYSpzda=0pST+W`k|>G;4;W0toMJXB32EJJsMeJo6z>{`*{BPEMjHmjXdEzUpk5w)kc`bbhg-WJU+;kOkr_6rkumJd=Od^3=`rWL=_Vz(Kg|i?^aV zE%`;^>yy#7S0}Y(Gq}o;<5OIfY=|f6YP@<$+Oj7{baj|$5uzr7A0lX{G-dqL(1+D= zBsvM-e!$3C30t2W`Pvyxj5@9_@*Kco1ek~_X|EKA8kuG`@zmCY_T6FVBHl+x>2!9c z2eFEMW^b4&zW5a&vAg1M=>}CO+D-;pJ%ZD6gz$RYWA@#v^K1>%`M;q@x~H z*V}@{r8>kDe251pbcO#}*X%Ro!_1i;pcD8fKfA*83d3SC{kZugPAI9s%gX;bp@TsH z_sIIH9eMHozd!&F`u1RDTvYi?L9_8HbXN3^$4ni)=Kr75=*LOquQi^DUMh7zPO_){ zc8L%-@5EGpuV=}rx2=?#SKyAjg+Z6O%o>*3b`3afPxe^v0KTbW*AhV}AY#XJP+S%; z^Kky0jHB10!Gf;lM%GgTB-&)w-4deNC`Cp)UE@NNvI&U>HoHx zT~FNY(vT~EHGmkd2tgG!C}BR74|#3%LZPPLT+%B#W4B$KpC93_7h+;z@3s@`*m@7a zsAJT1xQk9l)HTH>-R{`EceEnrwM1fHeD+ad)3SmP_|tJhCz|S}ocf(wAx9E4)2Yv3 zszfp~PSv`X3o!Obl4u+Ov_xuQOa9OIkgB81P{{Y+h0w7<%ESQ$AHiY z41Dwe58;DX?AvSQ)GmL2IQ{ID}VrPE$$V#du56i7na?nlsn`-taIbiQA+y+~ytgL#k@ z01&JU@*7x%mZuE01f{MYMngKCJy8*maK+%I&Fr-uhi1eEb{LvAK_{btz!QQjP=e6z zn}sn7CD2xe;kYY18-Qi%fm`^>=@LyX5@w6(|K;;bBmybn+Ik-T^s@q#?rpRIeGi4h2h{?EM zSkiFOqA<*gC*YX+MF65UA5=Bj>oT|K%&$@r9h84ck%3SJt=d^gXM|U_6cmk#D&A!~RGlQ0 zj)`D(whhB*P5tEHd`CdV(^jGIzt7^3BWYm09}BGLwiRo%MY02WYWy+EgVgQVxwc$lCPoEaCc-%k$;9@^DUs7e0;c{~v4v-whoQ4AWt_mX33 z05AXggL}?-EEF=Z(#c@}WEf$W##5HJ>iuUm*yKO09gi<-!c%kLsV%wJn7*ERAVcJ9 zd*JK=O_8wP{l8E%FK_;!)e(4oo20B4?wSWV(&Y@;)Y7dqxOXUUKf<(4dZ^b6u(w{h z;hOkD{v&!@vn-rZVGg-Q-`L4{?HHO@Y67=Q&Ye*a@1DF*^sZRv*fuOsj~IQPIScsNbmnSy73`@5h2 z`2pTdrS*C0uc+Yncju6ShhKFnXypC1(!cO^e|(jb-0*Jy_r-r-a$w;8e|g8%Ma$nC zTw)=Crg$*~484G>E6lH8?hRO+?gS7JcR~dw z+hum4zwMK|u{WBlYUm9x8N~+y!hA2A`@y2;M&f%5d8St!{j!2JMg1UoK9(@PlRP+* zpOERAa9s$@-LLBd@d6D|%Uik>_R7t#TE@n)0p-sVQhV4Zfo!b{s&^H^KDB0%Aw?`f7yeQ~>rq_$jhP(7^IBk== z?QWr6!|tE>Y#JV6K$5sVkcAoRHk&pe+Q>F8;1xmt0-&8i?)lI-?FL%OI+akVUoAkg zO=^uKTi-gh@*R_pxP3-HkK+LOq+6#u*BP{Fm7DcgDIl-Y)(n7Oh1-k;ev#NUY8H_q zkPUb|<}@apKsQ|M$l~*YE!p?7wH{=Xv_S zL-IdY>;Hv^WP!Yr>6aG34}bLk*KKpA+VQDbAEv4R*rOdZ1w=sNqb$J4*w_|2px$=S zMJc)a^$G|!MH0fxLzfq%GNBJ1m`!*8>c@X3H~GGxpm!pI33I6~KrHEl-fa2k4#X z){X4T-;+sZhExI?+IT?=Ap<-X3-_;&7-t}FD}v18Gr(wV2-(#moR&fQ8`&t|zZ&8{ zO)?WBPY=;IfRtYcIC2S8!wWRuJ*DIR4B(*~Fr35zam@&jx)Gpw{d@pq5OOxQpy8{6 zQ1Dc=V671V9Nq6B*3THx#MkejPb)YFva<}0m*>hxco_(#mVO(MH%JG%N!~|?p9&vP zAyJlAPtY`I$ZaM_PW;2;Zs{d{AxQ*>=2(1DFUQzg&%{~hQLW{fd6?)i8;{v=uS_Az z=veZtve*GGc;00})4rlpz>s)cz(urQKjtLI z04%j&pF=bMEHI0*SKq;~)qehweQ{%YRxr^yW-PLkPBQM}rLoe?2}qx7Ky902P^r7d z+Z$-7fiN9YY;Jl(!V%a=uUp&&>=^nFOkZqEjae1<-rY97eH^`$X+!Ou%!y_Y706nY zCXUN96$A=Hnxv)!3D)wBTWhFT2g=N?HTEK+=F>rYjfqQ8aq>5~+4+}kc&M3Zj;dvZ z>>Thv{k7MSo2I4qxO@|`!`p`!U|z|4aaTw|otKL3QnkD9@huWamjAH&4ky%|!_IOW zDB7hHn4It@Ev?Xno`xMLWgTVM` zLWHX==M4x4d4HKt%#|SL z0i+L=ElMR#^S7esx(+V`Gw0zc*qd}z5$}bRM0NP+oUdJv9@Z#)f(xKdHo))##H&@C z9vbS2TE20r%IK|`$0Wi7bqz#MoKB|y_Kfn!bz+UniGg6zI3WaJVC16XqY*DON50wNolMfpG0awkXa<8*#EGNG>th@(#ynCHi|qwXv3I%Zy@Q97oc)K;XRQ< z)m1UELKe>wbjClm4G|<$HqcpgSs$Ut^^dn&hp@n+gAbCcpx$zCuZugQgFuKkT`Rt} zb6LC^Dgtgip9t^-1^~JW zPy}Ryju}{?|MHx8sT3QZXZ8S4$Rp6v0O)81v%^aEvpuK< zIO=&PKy`P6;4eyIsR?+2>j(}b&HsPSrkTy-E2^e>Jy8Yf2phd-?snXw*T*?dP(7njqc(3 zb!2_lNdSWUgCX2s?nt0i7J`P|rCq;|e@o^wXyG0s-ZaYoaTqg9zkoJt0f5VfxYXj0 zo$&zmyFTdevRuCXx0{-13DENHtlNRMcEG^17XwiX^s8Bmo173@Q@bQ@2b`Xlu6Mlb zYvjOPTccK_+mjRXqpcM-G@L{bf0R2n<>-smknSoI2e&?k&Xy-V0f*wf;R7|jwhQ&w&!bEuZthtMOn4{`U#|!gPoSFmj zj&X1N<<@c-!Cfjm75C1Io~Jz+fe1c$)3D0T*=i}?WIx6cVjACbHWTiW~ymz<~cff9-_ z5b`$d4lagKg%t;J*+zIMgGES2H!sDWF)ZRa;PMbOb5PE`1r}*iNt!Sabikbp@2cno zulvn29;vyZ4z5r=oR)HT1N8#ZAyKmVqzG-*sh;Q0p79>T4sLD62wR81`1zsgV{56| zMin`49P-ac2%6c0a!5X)h0eQ(bh3~|oG~~;N^4YN%YM3GykgJbZTP;Fs=*{>8vL9D#tw^!t-kqd3bI1zX3E?ag2osD@INqZMBU$w zDzZZks!o_R3v63gtGM2|%1&Tb6v4!zh+Z}|kpt^K74eg9p=SCai_B@;kKc!T{Ap}u z^f)fl`hSb(4Emv%lPE<>5XR9n+65Ygi(FTZi5Mfzz>Vkp{Y$_lEN}SiJl&IXd(PT< z3U?0YqN~|n0lzLWpQ_{_nu@$nweYtfnj=SQ%I@rleNgmId#@htGhgZm9kNy6RrS7L zIzLldvy*M)0Tc4Zw+ik03gf7}>ibkdaWc#Pt=n8G5T*Ble)@(}42n`e zJ%aBHB?KjPZfkS1sG==#zNks93RBkYTaOwHi){=lTKXzBs@dBqsm}dPn}~zZ306P* zhgX!EBjr+#qM4~2I2yiRs#twyI)9ps#r$vg`1d~>!cLw$U!Xf&yx*59RQz4PXE&*zBKR!*&!6FW*&fn-9UgN@L~yq zJ~`}af!}hpb2Y?mKBZcOGg}4WUoz(ZNbwj=KR1Q* z`CAZFDdgB(Q>dPipZ~k(zYBJvoaT=+51rs3~$~0VRfk&ZBWkHqM)PK6l&C zCDFx`J@wBb8DeC8fnEJ)a0NfTBcO{s7{~uG2rn6YCp**kZz*sQVZu@OKhvBwsrD#u zNsyH6IrR6yITa3p5`Po{R084kG)X>jyCthpSnnR>z;;u&wQRQ=1#RcTdM!*lL&xJ3 zBlke4j<2%y;NJrD^Qc-!hCf0>5WqnAuAvRK#=Sap1q`kqj(5%7Ssn|ns~k}4?K@O4 z)A+@;Gir>12pYn72pV_4N0cm}pVSoo$JYb37|YgJrkN`1inO}rW>Il7u5~nH{bixg zT$LLZdCHwGEq;?Q^Ovmi?_p^e5ZL&O%_)B@w{I5J$9R*^MO6wPYGoxYox>^Yv|BUc18wIRNLP|dnq)u^Z&X&1o@q87(PMu?5N?0ZlM1TKAU}y z6Xq)=9}TqY);n(-(UPY+D@H#3_b^?ZSBkn_lPc}!eZ$s*w5-*T$GJ~ORm{wbr@JXu zhxXc8{U>{#O6=R(uf|e*hB^wsaDxBzO2mJ85T+K~r!7njv`?T`LORx`?c06u#Ll}% zHMJ0F7v=G7W@~1HB|g766L0u-JkSr%C}8{_<@0~bA>ieRin{zx|FNqDykGniK3l&X zr*57xQW*%Yn51V}{^qJ%3cBknET*Pfm7fRq+V*C`U$+^>xjBQ055X&TfPR_CnOg7P z4~DN0oUi7)3#5N+#0>$K`*A2S$H7Nbf%<9u-Wmq zM{qR36OuhvFS-zZ?Hfx8WwA3o-aGXx{??M`)*5<6Y&ryYt)m-s_;kxH7rd&VL)JTm zW7wB!p*u|RThK@&?Yn;Uqb!Zz)jvAV9DO8wdN~JO4DQ_2a%tf)HMM;!xLNAC)T&oE zJ;se)nZ3TEh^ns`|IT98wK|3CYd=}Ny?9yfHz$xp*w&1A*PtJIL>v<&P~Vbfn+jXr znCxs84bBi+?=7>S6I=Xzb6_}snt_pVON9VM`0Jt1(@0##A3gc=S5N@x;hO$&AjIuQ zC+eJL9wI&G7W(3{=cW6MLUmOJl$Z~ENbi~5a0TGrz_jfRdqgV<5e84pN(F4D~KI4Flz7N*|t;EK$OFwL> z7v=I`?w$S@b`@p`luI_@^^%0hm%;Ql`EOr=#u*}NjT({*Ua(03`0Gj-18BE>&DOyD z%!~7_dBE?#pyr5jD?z@O5xnUi8DQd)veM&KX#N&6R^C>=&u+=2Zn4pmcOdV%?m;i+Gt1Q*VY=0mS$j#-g%)q}r01c``fQ z9VX&_r#~q_PHdYbM#LV|5-rhp=}y(t;;H^ib6<0ev_8M6wl|UJ9yxLDw{-gif30Au ze1g{@;HS|vX4OJ*YzE@;60oN+?5``K9ob!EttAFJcXmJ&dK%+P%PAX1;e|X-c1^JC zi=Q8*PdDiJT-Ytn6eSgk8~Y>dwq3$(?51Y#x`TX^{*YIb-iEJMe*Z4rAO!@oM>fOG^L=`EeV6{sSKT z_Vxenoj-}=L~C&?t62TOTVR8ZwBkPkRI$PHhjvz$L19_$qkEQ+0r=T6@|#!(?~Q%s z0F&HJpPN6C>-8DzJ?$AiuN$}~YGU=3Jq4no%0s|n`|IbOCrLCW`*GHp==Af@pcJ>Mjju51QP*YP`@lh60d)u4!FY*ccP=l@$tTqGK;y7p-tb)LYM31RHH-x-5 zb8B|7Em7y*Zm!O<9IO94+>tg@_c4%eg}IwgvRT}$#=Z*1*0mc~TQo=9i`)P@t6fvr z>28=0qiEAzZ}RF_4E@u^`9gxk|B~**b%Gn*;Gr>UF0_72F7RB1+4eyf9hp3zP00ZI zq(0`=5Y=gjPx}sC_(H@Y>E&@(A1I*bc5sGzeV6P$^;&wZDJ`J53|*}41~ylC&bLWw z6uKJKe2vELf1(02R z!#KC^^Ux+KdUMHk;KSxjdjs$xD})^OK<2NBRB~k)GIL8H23SRp{mu5!W2Lsuw2w<5 zcV2n`U~d*Y20t~Gofk{`Gl#k0_CGe`k?*L9NPsz&vLJ^G_{1e z@yvvBNdH@QJb|C((t^P2gN5xhP5(&#V|ip`%%MLrCvPN-yyYx9sE%YkzgX@qe&Se~ z#t@^RsqxpljQskyR%+s}H&5`F*7L{cXNFkLKTR21QrfRF1#I~fnJ^E#_=sW9qraxN zM7~#Od7`)$n3dn`msAs*6&9_1h%C2Cq`*0Sl42p))Rz61b|?t#z5Yt5TKB+}uqE); z#5YDm?U%n2i8o zS?BvdNGS*!Vkh(orNT&s7t7R4j7qtLSAKYX zGPc9V8WN4>ENBsk16i?wplUehXgk$~{(~s}nEPjF@zdhalTt!rkT#_zz7jTD&@6DT z4K%^2`aN(Z-#_dvtk!8r4oO09P#7dBB%3vVH8eS{kTC1FrrU}3zEd;ovsB3*sb^jz zXeLA(uGU>4lXxwGfuT6FhJ5eH1tQ;*3ez3eugMu|Hpo=9;e`6RCpAz)f$N`uL9sn{LwL3oHJe+$$^?dKUVX-1n`)v!cS`&NoUY z({ZDNYk?CJ8*+iSazA+^q_p%qAXo9vY7$d+JuDQTOc7^uifK4vT>^&x7uOl-?Z8bXaVR-KB7mwr-ogwQu&hK`ir4msW6Db4p1l zkR)V2wi(j<+~hug-HRxSxckmy!=X(tQd#yTQfB1n=fByRTp?&xxK6}PFcw2wy>-QL z2Uz{&OO&1`R>ji*ghcM$@&B;(6+l&PZQFn#CDI@vAPo|NbazU3Bi-HI-62YMcQ;51 zC`flmNO$+YIPdq*dyeOw+2b&dZuWkjXRUSL*L_9PMFvu!S2`Qd-|E`xH&2guapghB zqt`8HskHW!fa3|Ns>AgOQqHC_76j7syqU&Sz=Ttdr_KitdPY$f+!0QPc0yk#DR=yEf1LnSz3@y6Nq~V(G`l zv1q*4x8>aK%@Bp!-eBr)0T9x01Jd~9-V$V47*?em1=d(<$*PJn&DOE0vLzHTYXCRO z9#FG+0iOlMnSh&4FQE6f1QOY@BlWD25G4HIU#jN>;sp_K-3)PmpVw7Wp`=qQP!*d9 zIo}sK%>dUMg0u4?2){@EiKh}m;-NlE2up@p?Du6FITF16ZQU4Ju)*O#xv!!`iwUYAd}qhyNXZ54 zFaI2TU=;G1CZpP+mU~m`z_u%GI$T5ag;Q3tDBkW&`~|f81g74udLWc&$myP ze7~P}{+@+X$!W9m+Ug6U#d-BhiP-V@+ug=UmfzyZC-nH5Xzq`x_9bNOL;v zq3A$?ks=*&n>oo_rDBX@?DMOuQp)BX7Q3dJni~1kBpwN4rF`H(SN#6R9ssp_u$k3d zCQF%ryuX$tABpOy{W89g!7TS6KMX}l`1QwCOBNQa>I;Kki<)T5yLTZB3)2aY!&m(j zt|?xRMkBG|@$tk$$K0#eHEN?AaBwST~!J$2|~V zqtL-Y5kdU-)kLam`x}Nv{REpAi+4776RUf&An5Li*$d@V)4VWYN%nKl)ebnK)MkmgQriE27gq4^;G@uKO2O6YKr`NAk1lEG{?kqKBqEE>o0)}G)P?nUoZW|#|Y zu&#zdD+ofuqn>o#&T$ZV{`3YkpCNAPd7KaM^CaaC4kh>_m%jDgDD=sSSNC|gDH%b# zcpciEodC!Bx2pvU2cT-~S<^6+KV%Y%l4Ladvg5S#AmS)Of&u~gk5?rz6c&k(UC&2v zZ||xyPv8|P0`u!*$p`>f$HD9&^qEcJIQWF**=Q=oilND@*5!&DdC_AyI8X0VDGSwL zy_^q9J*tx1N@sdII~&>InxTk;iP%AL-%>UhXy_OV&x*xhT0vUoqG$biq`9x^J~(nZ zdgpND{Ba{8;8al|!)KsIVgqMa;}r}HNYD4@r&nkJ;EP+P*(5)4iuDz^gvPUpjRCIL z7Etc4@9w@0P#m_`O#}|-rqIOSBnkcF`9X^b-U`C^?93sV&6apzbg*TrL^N$@iowkg zpXwpri+)st_l|O@;qG8V*J-W@xbn{lW7%uBx+qUv){-u6?cZP}-@GeH1NL<7}%y)c&bndD8DqjY#eDp|f7iY6J1K2T&g5e)Wgl zP0*0=jgOD7PN*f-I#wxCG#rbsnPcbRC^n0M`PaviMX!WIlzq_iI9lO9Z?;Rco+v~G z&9t0_Ylr(KooyxwmQXkrb1Cv>o=iD;VBcR0_c=7%YKw($Msl*m>yq2CM0e++!1ZwU zhHd|Y#VwZ=q-H9|^~(zX+Q6mmZhtVI6CTf(BB>MmJYOZ%=2@lJn?S`QF@BdNUG#G> zx4@{rCW*$x)fo!yZ_D#POV!5KcX14mjXUk^Yyg0>&B$u;*FEV#KaIS1lSy9A062|9TY#b zTcYN?t{5EC7~;|5A`pL*R8)ShJ@ebpx|voE`PLYZjAioshs@hH zyBrf+&Q=FL+?~ppfq58ZO>6*CedT0*xgp{bu!TlLq)H_wCN2ba$POSb*=(V)in7}q zEa4B~Pja^{K*{QY5&!TZqHq1;S|@kgd}?eJwVQd+o}-BB9^$;CT+3v1>U*by#q})6 zHT?*mte|6GzUiEk%E(ACG;0x^SL_xyP9jRZzxu_JQ_=75+adVbrw-i*!fap0$L#*= z*RMUXzy&!?DesF~RC_5Ft0S%$42VY!yP}%5CyEDck1U8H*dTsmQm$4bqCfQe?)nt- zSi$O>C+NIBbUfUYXaB`(LaQcn;Pf*%Xny2!Pgml_SL`h}qCf$AtfHa5vY1MF`N6s*Vi+jqIB^{`KPf01};nGrKQASnq`TlWRa&9zr zB$Zuok5x8p=CZimpiomsBf%E}4&$Y7LIMUZj|baghj)g6fIwktO34!9{H&20^Nyu% zW(;kQ@%f=>D_QzDh}=qh>x}z6hmw4bFA0O`xZfN0& zh

ZK;upA@=1QK+QhH~AHCB#oew5K$Ms2gaj1y#JhbCIJhyav z(o|4GawGv%A}1kqg^p3>ieO zR!>Ut2 zn>Ygl-BEC|Jx#`#jK04LkG`YnT;V0s6@h-Zztm}~k@~cR-H%0=NowlOrQxp<^RRT9 zqAL4(DxcZ&LFPB}?B*75u}NC~C{@kdf*BJ}V==fH6lKThm2`5e9`$@gxk>U9O%~w8 zDSTlx>}jb5m|jp2Kl`71qAC|8Tqc%xc#V!O*zimM0^){eTkG^vTk%8n>O@i)7L@>! z2er9k6RU2Ad0bAYBU1WDHCrWfgK#~D#hg{ge1UE zzFSdFfZzAe!ciFg8zW)WmZ6^QEb2;G94PLHu&{u-Mi2IX^vmputg4_u5_9%EiGZ-p-a)#J7G}Y z$HXQiQ){XhQsAY<@O3A7@<(T2q_?=nKc`cZ@Rt4}w7)Z{!1{xwV{MPUnItr0r-$FW zFY2`2#TD(OZlnV%%0d^f51x*Y4Jx$e!?g}=mS%TySKF$;HLZ;V|EDi(0$WJkLL@nr!sMP}sH z86GF0PN>8D|E%5lI?dqtAnKAeEbjznyUYQ1hvW5(oB2rK!tId)@);klhV;!{*q|dD zIFvt=L)O1UPN@H?^YUdr<1Adv%ecwL_KCu<=ytuxYHn;g*@iC}KMzi^DY;mbbEJ82 zc)eQD&Nb=YCz7=V5UI3sZA~r6i3ev$?rBZ0nFjBq9?`Mdg8z zcN+-E3jhINE9K?jADsqYCpFyGV~`o9WIr(ys;-V5jnj@~?D6qUSR@@{Y!AA$m6g@= zoFP2!SGeHwVsQAXkJVphW)j&h8r7M_g<~91@=%HE6VT?JG{vp3IBYXE<6z}aBqn*l z%OotAtS9iySH?2FX!Un_dkM;MDUL#!L~13WOiEXcX1BH{?|=;I>xc-NOCsTjvvIWv za!Cj!#ByzI?OG5TVQ69SuTMJ%53cP61>eYxWFj>g6j2(Nt9(JdA`6xdM`Rs>Uf$NI zl4KJ3E!u%%y&hSadg{|feg=7gNh7M;;lLZ<HI!Q{{RZ(XpOcsVcV*ABxRB`chU|Wj(J2C|si@*VmVc@Em&7ky> zWx7=DeI(;`&(COSJMO6eoce$N)0F}OlId`drg;-g&h2jWu<;oF zyte-h)`5SHLvpM@!_`{9ITPurjzZmz zZ$mihWj`>U+{>xl_^vHNhLRF!HQ85fYbP=e)aphKK?IvaAVFTS$vzDQBS}K}f zL*bi_fBn}1V}SZ?jyFju2p>)sjDNNP+2U;0I71EqsYRehQB(}O^Ps0r9z(jSAA+CX z|0b3PZ>=v31&Q*G7m8qvKgK=D}f#NyXn|?yIY*$pNvDW(i+z z#GhhR!Qk^L)O&QMN?)AT&TJ%YnxY`W;r>Bz!|nP?Y*Knrr@a4PEifNZJOs2X7yNVh zyPMsLGa#NH2g!<5A5PYsW#;;D--Ldq@%v4-81XHP2pN)z9=fbX#?6hR=$-^u()U)g z-Qx>DSPU(&Yv0Fp0-QX*_HN_nM}HpwM3JWNOvUPfQFN+&Wn|u8@b}wkTfBkF%!AG!(?dLgb@AJiu{1-kmA}^p*kcv9jv$B=HWZu)w*_&afyXhs4$0 zJ-(oT;)%2GrmiN9%DZMK;rW@`xQ08Q+^FuYRe<%R1*{<1Klgdi^1g$-f~Z?bNy*3q z-DNr(Zfm=PSaq@cY(-!phs`!xTfGfKV1HlODd?+y&LQ_H_!lcVHZO&BV!6Dgr2HvK zq=Ys(3rpMbaiwQu6_APeeB-Wo+%GKTum}ldOifLR4}bs8x1g666UzjIJ^8(O<2XEg z;Mn+Nl?B4dd`zMr*E2dnwou5urrpi;3cr~S=jGHFZl|e3i>2P_W7B6+eVxPA{5u=Bw_jj6l3ySLANXgH9ueKwPj zn~RiU6%Z`y7FD29tY%3XmuYbt|AaICjsGP0diPhi9Rm^}kLEy%WRTXo)~nUC!2(nyaNMi}RaCBH&V` z68sAs`p0uj6z}s96EdJQnGfj8dM?($B{Z|q<+a}yW(>c$(V0Nu-3U@SgjMKFYf&ta z7j6paT3e;O4y_}w78il?-GdVrHci&a?>vLO1?CMx z8$tRMCO49v4tThf*GHR^pwwdJe1*ctd$1t9kIR%Pd3g|?-U^dWGLoXPx*}?J9KSj8 z={p$p(`hx@@Y4j-?oX~#V4p+_Cq-%IA|2jk--YCQ<+IPZn@Lco@p>pnym<_sw@Glt z(mH+?{{cqyj}{%}8z}@NhKwISR4du3KQd1ysORHpIF4C{Tjb=jWrEIx7i9-sQnTd9 zG`#IT_J*l-6^f9g`gEd0?di+wR6E<`Dr+sr)VIvX!9|^|g3AG`+Uy`<1l-f_K@YrpH-6#$aHrXYvlE!LC z|5ZKSf2U1jvlTc@xWB&-L>c5Ie{uw|#&g<7aebfqLfFZe6`3Z%jN&5J{bElI2~lC6 zg$?$PRv6M8e6tDM!{%6XCtKSrkoyyY?EEn!qBW3>hDG}7?nV6F!7alS-w* zge(qq;>|^h`Q;`lCyf{dd3h6%T;XtkS4uu)6u7Zl3zfDAkOkw2q<#S7q`d3f?a%K=1& z%Bw%CmO$-Klq2ZljRMR~Fssb{Tn3ttuOf-elt|vQ#YyXzLCFkQn+KfHjY{V&YMM^> zN=h(jFeU4sFs{7x(en9*I?v6;^@2{yEB&0g(sh+WN$2%4`u_l6q@Ymvm|a(B@SrfM z6;Yrl5yj4z?+Jw^mBr?YTF(N#&fV8A7$nL&@GxnOL;FXQ*iJma)I6jiaL^PGD{9>@ ztx#VH(RGhO1^Xzt86RLo#~ z*Vvt}H5F7*oC*d)PqO`mCWYD&E6s^*il4w{2_WzLzs#p2u=th}Jwl=$errVXzb2Jj zDwj(B8RfVu%#JuQdrdO6WuZs07apI%U57HFkjiG8&8UFo?CO~n$E)bLN~8@Oj>0MR zBxRb?1KqGrU0d8i{TNtR&`j9D3mkZI=mZG3UvB@qqrA1B&~^vTG`7&~6r=ak?n z_@a*u+G$H0wv4c`{8Ah^rP#GsOJ{yq-mA!%Uh39wv91z%( zUFwrD1VS_9K**sGjz)`pX8{N|l&~VOc;VIf`qmh5Z5pnsLIdTTFe_b8js8a&2mxIU z*9pZG4r9lTQm9^6JoCv*s`_LsX*E8s_mhjKql%v(CEan>%#B*RhpVU5i<|DX{W`HJ zQJ;(&U8-s}iV6wGfyuUrQ4}+#ExygDJF2?Lu0e7wH0!gq0TF(-AFUbC?UuwCJzqM|MV+;V|4Jps@c{Wl*%i~NIs7J9 zDjS!tPOm}Nnn8$uMsBNTF6%iU+tgl9IZoF3DPoV1P1BB{Qr?c@-N{fz(Pcq~Tr>6O zSgVRsG_||+xW-`QqE80GOO3YrsEyPmiH4gq>a(iH3Q~Tze0PfI{Qr9D96${;Y3PUc z3|b{ z!BAQb`*{2?Oo|JbK`J`BDCrCy+7rMTk76r+>>gPIs*5r@0iGo2a_u(iJNe?4eI9S$ zsn=|Ubd1#AZyDq0B-?cWs;7U{kCyXX;9v6!0N1(*KtNFvz~sPxQ?@hAr94#1h3l`5 z+14n>zh^!tp(a!F-IlbCdz6_)`yO*nak<9FjPIE4?_sk_IS6<{`2e5|v%dqfLYqqB zS;A8DP{uK=s|8r_#O7Cj9{g$ugcRjm`j}*r-QZLS?-ZL7Y*3$c zU{a8i@9g(O6{fE1GMzMC90PmTT(Ab^Ei7n`&kbUgT3!#+rgQ>&SjZyTQH^pAe?J#( zy-`k7t>eDMXwtLS1EI8$v&C$0-uMz#4k(^DuM_cFn3kjsN?5(OOBo|DY<|1a;rS9e z;IHE7jqMif|Mvp;dlLhGz9xe}{CS5~oBc36JUnPdD50!=FvcF9!JZc^AM=vEs6S?? zHk!txv86@uZsNH<*(N@U5uoTfto0!wym*nT!LPD4icXl`#J}=eE^3GQP&l!8A}fW} z+DyoM)A+@zv6YF@va(LClT$cDseb-0)WctGjl>7>0WYGW`j9hT=>^z-0ZX8oLGXis z{^^%#LZ)B%%pn&CXCYuTH;-2yuBcxYAr{h?0@~%RBS@MM60)V!L)|+ncVxQt-rr94 z{)Aaj)yXxalr_(vB!oihiq~4@IXdhD1@zFERxlQpmIKn&a~34REX(Y-=+yKuX>*PT zT$8WPUtUMyN-5=zz?OLBPgp=(;tlMmNTcu@qCK_pf2`v5V7@8VA9oRc?aKUoZLhgm zyL$8otJmeDKVtr@{HBSj3Z&yR$>odN#gA{28S=@>u8pdllj2>``=#D4P+VRsbhqOKa08DO@B9^Kc_b`k3Szl ziw^&Ht^0HfW6q`CcaUUt^rDNYs96>Xhoxd!l~ix(bwr5~>4tQIP}CPnp(ox`!v(vL z{WX?3uddjPZ*Il_2yCxuZmSZFKVOKaSUM%d-sWmdCu_SBb?IwAg({31|L3c_*d3_Nz-pLb;{w7SZp7J zgoQVn56Aa9AQ)8b-ATpu*pt9eZL>;07Fix(T=bC>2l3^b88BFc+ik%d_M{bZa@Ig~Du}muJ=l)>czeiuT!n?I*mq8I(B!e- z#a8{Z2#a^a92bqZWbz^>WmqTx2FNyK*zx6k035_3q)QNfd)^))#O8I+~)VLkAkLrLz~+(lje%@Xw2>` zmI8E9QMi`t@zCPmq&-&6y2r;AN}~gpfceuXsCIc7MNI*IM>bO+z?=2 za5s<^nnnGtrn672X{zi6_9Oo-_e!GyAcEey$-=^-=S~?O6Zi#CkyvrL-I%I=4r~J- zyG($m#ipdBux+|l!GJ^^r3jjWEnnlyUFi9`i-e* zirpZG;}0d#7XXRnpfHyE3q(ZpPdpZBxj*{Gu+lUTOQq@CrAC|=B#WrT*Y{S-ef={~ zXS>2_At${kCy`#`gS6Ybh-4^zjqStWN9mK~MtW;x`tNm>f%Q(S#?GM8n5fbh&D*Ir zFf9Pu(t0mXGo^wkz4L+7Y2!9mW-d*f0)SZQc>)QdT#rni{GbIGN2}(1qD#YBm(Vg6 zclBzhwiyY{NaG)1rF|R0p6GIgjg||F2ehj|>Gm z;i(xq+L1ZwYP2comoKF!uCCJ!j1MzOyo%@3xo_yDTyr3iwiP3I@Mi<)QfIQtvARs; zQob*{888UtKuSjt+kk_wZ6E5{vm!ullvr$GwL^n>I zOLqA?#d0hER##L?Fn612FRiz*v$M+v9ZDe3n!(JnThdk1SSR5$p&WUC4OG(sTe zY^|MJJAu*cxc}lu%0ybZ4o7}OFP6g+&v}tZ+4COdKrvSmS<&j1iKNEXfc0qwBlpd( z&zh4&AA5t9c!O**X`}(1@zYt&_6NWWcgqC-Rm6gQxDZr`O5>HQt#5|d=#uVZHVki9 zx_BS6&tZ{GW3;Lb;LEhz)T!21$PW&Z=Tq1r27#w&BFi_`m%cz>hI$))2gWmjuLiLK z@F#+vEmW=yKj@}3>@8saJEz*Xj7A&#`x^hd*jreK- zBsvOUJEFctn?2~dA~l6I+`Wp#{*4~2&Yes6r2RGwzGzYvoK#h46%BY0eyH=^7-+S- zI<`7}pA~_-l1=;Y?9Ul2s1MmueISz2-@Gl_w9E%F+bCT*_^Ls)l@oHGCY6&JkkJtn zULfiLkrB;L28Peqs4KK5c446i`ODtvWpBLf7Du^v3yu7zCT;>*3nd~qx6*6xwK$bZ z7B#8D!Uxd=hZ2VnLUNh-M1ng(>y#Igu_NJ>i2C}q6evb_^kPBy3PF|d7O2Cw+b8l5 z<7nP$ru98XpcaFSlu#)o%5FDj{gr8P$N$e0=C9MncLA*vl=WyeD;M~Y+<-Y?A-Ggy zjGV^fnLObatDdUo+x>ZBS_ULjgH0P}{Iz9i1RogKVhnYB{&wqdv4zzJy?z2912Ka1 zC0xrvr9$bp=Th>6*?G3>>l+(n#e}jS{E{z-s97BlU{FFSF`z2t1FZJ?+24^r6`tqn~F1Zw;YvP>z@4;D2Pt8`Q>JG$@Dz zP&(5nm%eHPV*a2Y(SOYzzz(tgwBvDotXYtLYLYiMDe9{ROBQNP(W?&IKt#B#)Idc> z2gVAi1cE9fm@Gv?h?*ng=g#^+A8X-m9?}|+MiX)8J{0AzsuEx`e5-(^7|Yj*0;vYE zp(`>{=Nao)Zt|RG)vK(zwcNqkiRz7-hcx-ih#dM};XtZhv+&=*`aG%LdRZDs@C$(U zAkqbqRi4}zK=9)$U;Ur5#^hq&EkBVGt9~j`pBb?I{Nki{tngA`V{5Ar{pv#t_rE58 z;IPij1m!L*WgHgcFc`X3yzx4zo}r)=H8X}xpzT%#AB}u&bg^kaqHfpZF_H*rmoDmH zk6hSMChi7kAr%A$qaZYJ28+~EcoI;sX@wnisIyz<=h%M^kr>IHicuovH9!%kRyG-*zpW3j5@+?Mi~c zOzd7DI3-=;*wZSCYd6aNkm1u~XmLM~jerJcAnMZ~Iz&awtY#42;ua%wF4Ba>+hngU z@%vTkW>qv$aH*yBeD$Tzw?Y`+=B$P)eu zY~#wq?A;(H#A#0q_9gyLN@Ja{aBTW~fNq;5)I9QqWZ;U~==u1l^NoBHE2z9>HCenQ zlx*jdkrSs{bszCXFFHQuesMNMYowEHw^}vZU)bI=QrIf2t;7AVG3;Mo|NGO(>$KHc z#&U{GA|Vr46^#L^_!cl#BSyFbe?GZW;Z7U8F$GfqumYKG0Ok1G(vr3XAptly;NrUE z(qo_Ofi$(WioEWFFcObG6Oi|f$H~!xQ$qd)CV_`PUB0Hm zYZib13okw;6&Dv*Kaa`Np%4-lmW_x#9BQhn9_nN_?$4(3j`r}=xjdt{=Yj~a?DgO)S{vcN`Z9|Kp9i%y!bq5y2P zrLnOn=xxrxj%XeciVnUMAP&aHf}N!5OckNlnDS{n23jr&uytpLc69JY;=C8hxdyz2 zz>4J&p4$&OnzsSoD{E;;Vt-&IpZLL8$#6lm6bhOq$VYvJQlp((i>*n)*b=j3YMSc3 zV&=%@9;p9K0L%Dnq;kGnEU4@)_s$P(h>s2kvb#+b@X0gX#0T_6S{kuL5N zWt*nHjFVj)>&wetN@lXzm7Xy&Z0tUAtCX~gonT}_G>!1;jcrN#HiNxNlRxNLy4lb~ zKFM2=Uu*d~)!}VWNZGmF5+rIq5$2WQeUkk-sj&V=*w;Jjk*#<|Avi8BZrx~>ARbLm zi2L?Y|EqsP&mf>E$2N4^%hJD$d4i~PIiOVoK79CqK^Q%j_-(c>_v=7j?)=aF%%SJI z8whxK8Ye}w-!mbSPu`_>`B#Mh2J8%Yz#>oP;^y8s_!FV|{02~~j9kXw8KP7QLG5Ln zg6x7U;4|lY0E%lrbm73|tj5o9)m3#l5|g;H-B^o(SEhS63UTO^3_R~4>SJ49f4_pm2mHlm&*+)^-b}gELgj82j`*->Prh-H zy3ER3Uy0Wm5e0c>>XGfo57QG>ml_gy`!8C4sWy1fXtB41$Hm8+0)P1~sURoL)PFQL zfX;vnKKgI_igyYql_pU5Y1be}Kw;{evW^Ce7k^26vOx6Md;jFS^IgpyR*UE7gggqS zAjbYd$~&(a3~73NrPy{j5vk_}hK3|t;5NbII}*bO{E)9S442dG5MHo}WV9{E@@Uz{ zQQxv8g!hj2mXZFre!$qjx#etjK1Fj)DCaBFaq*{1pA%OZ-Y2`&lb&cE(k-6P^%vpY zN58<+th+cBfY{(jhy8=(>*M;wGg`Vtu7QONC$c$MU`IMu&{o{KpSM$Hx6(d`zjky9 z)b8-vY%<5N0H@s7Sf+D&Itj369mr3%3s3yS)rg-x050vOL z#nTmaU4CbdQb`E~2?^PxKCBY2Lh!Fx>Y9~|ku7s~H|w{RO%i8j<`fp2kw|T88Y>@K z-7m|P5wkyke&tUbF0GFNz6=H~Wc>n)6#+!IX7ypR+n^O z;U^9c5AUXKfMuv+T!{|65}?zmA?pXr<#K|S%+FjL{gzJdutWkDG8?yKo3U)dW~^;M zZ*|H^dh_Q43<0U?liVrQi2>dtgvDw>49vK4ydSQERoC&EOi-Au7CwXSRtAVg$fM9k z{CWUW@B;w>K|(?T>EP$r)`_(0ijRlv0PM&E=U=*jpkS}^H6ZgME#SX?s?>1}pN_3} zNKnA`gy3o8wsrnmV|LmlYc^MWsqiH;N8!`k7Z*}@AYW??L}^OY*I#u=%D}Ka{|8~- zSAf{(QB+4P)As3++k>wp906!ncy#jy5Zne_eFZbe9J{cvFk-duXJ=>MMUo71cB{7qfyAl<{hnetEvYsvzbsDha(!~WU+GD zO>Mr9?H>&b!e>y-Lp=}s1#aqx6uynY#BJbyz_Kr{=k%ohnaREda+6x`M=wAGfzWDp z5LaW+&=wIAf@8&wCgjZ@1l#QLFGK|4yH*=L*!zngq+qilPNdurHXKSUsLiD`v;ti< z*#odg_i1fwD{?+u>IxJA1H^BmqjILfFaC+R`j|dl9QDg*H4Nxb{QUe(jysARwyV&t zuCAj2@aR3*zZx6aZh$I1uagtINFWel6_HEhJs=K_0}oz)3*V5xF!U)gB64yOxO9u{ zfkWY!HV!B2;z$^vM8Lqp%9nFGyqYpBM!Gc zHd>u_&BoCcT`NGyLZ8aIdl3Djaeen1jJ~BdwROpj1blj>9^!666~V9uG)+H&9wbos zCx8b^A>=!xNW z$++0)s>8HFdkcyOtPD^|MVL*MiUU0iHg5!|iln*mPyX1ZMts0?Weo8;iHR2q-1Iz9 zApQioSzADb0DhVF6PHxDlECqclhwlL6(XT@K(!y!gWA<&^vWk z;3^e?1Zt&4=qt}|-iu0l8Am&6ZNH~6H#447CUso~g*z;TBlGMm_Bh)#Pm%VtOv|^e z)5;aRt}$($U>yDQf+?-%@iOOq4##J7Y2@wC%Xe1~#3R+)NqO*@wyjpwKE8Il-jfbP z=pgC)<@^AKk3GPAZNwYToXCB(#>G% z3)p7s-ROyL6%?+w6PS=O>%M9UZNI$(+A|)p&hdESxpI%s$h1}RcBFa>hKAxP*QFuS zI8O@Bt_7=JEIvUPl0Jq*)%Cx2@emFFc%orsAfQaXzV8h&lJN0t9Qexa($X@QJuXsx zQ+h7IT|Az{gK2M{ORuda_HnwVULT7}UFb&do~=J|o|8RB;~Z_m%!acnEsm3voSgcT zcyIV7T_-GJio42QX6M@UKv@JH_hlTRl52S~WAxP#TXXA=3_8muzU=|iuKq1R2sQzQ zB{9SkF|!cZas?X27Jqp0q|^&1?c`rspUCVDpD0>}U02VJ&`j8`Y9#fNPHqTnPi2lF z7PL06jeOB2%u9^&`H3-t0-KZnl(S0G4j?&^xfBDdwbfDQNWtF!`NyQUNXTe*7 zm>Hs&J0={T=j1&y`@uHu-K>6Bc=Nu!92z#NXt?`GR?il6-gFUf)c8=w%WR|j3~WQX zJguVHpPv*Ddqo`F_HCQ1*U-4V68Outt%z;VVXdj~4k!yA$3QL}omhc@Kd_fD7sW zjCViZKwS5SjQaVoGwx~u*VM8-tAd+ZvIlik)PeMbnS;BVW3O!Sasz|yx+}Y|sOUJL z+!g{MvR-(rW)#qo0a)-8A8?`zbacNCn>6`>onx|i;!Jh-;FuHKrqe$}t z60br3g;pn#;T`$v}QIQ(sh#KjF*IG%164Ar6NAt51|!096i zG-Zij+$Jycanp8X4Lgbr%qwm;QQc@id%4`=g~1P;$r-DBk9z*yH4WFvT40fPOKLvp zK@uk}kc^^2eF?+kI-RkequWwXP`^->S=6y=)V2Pmb+@8PSn(h=?y~^$0Z%5Q($Jvr zU{zxQBC~f6CbQ>av=t$8b_pl_Xj{#fuN*Vf+hB->t-;jR)|SZgR~zF~18)YjprYc*9L`ITT>7G+ zA*G|dhi>E1KEiw|DOYI{iqaS;2E*r+)Q1YeR@n6-v{bi9iOL%DeA_&c-ml-6XAWW~G-Yo{!7Pj%UEiSBW{($3VxFkI?0YYJ`V z`oAW+Id9Fk_pfqgvDD7%J`Rzqm}+=A4fG8M71y3%W_-RKb+9`}l?Zba)0Q?ISYFN` z@-Y%cb_Dxw5F8r^#~AQ%Dgxd{UDeu42cBzVZ&ssg!eJ-Pr{Z^*lb~KInPxJ4pF)zP zq&TiI*hZKr9@@pdpP(42JCF=^e#oaobH$#HJ6rx;QRq1RbzN~GQX7{eafiuD#yHvv zb#zmsmXJuzdrI*J$>Y{OnYAiq9g#y)_PkitpW_U_s8-IHZKtJ!)J2=^qG7#5(SDu> zF&!Tgpa!9<%lMAGf8zcK!JypXvXd4Ue-pLkwlhJsO4|c?&J&d%1>Hcf1nDkTx180f z3%i3m0<^q>0)X4zOSrKQoV=$vet#kj9S9S&^L<@zj-=V`sCwObJX(`mD^JF9G)Q<5 z;r#8W)__c*(+)!x@+{^Mdt8+h@AX1n!dutr60ys>_hqabc05zNb20c@5e+ns^g4c; zxeyW!QA?Tc5Z=6u_zz9Um8|pedDJTgKW(>dvR^u`piEs|7AK%*;0cS`*QVvMBGB&Q zHQMh>8)z{jaYJ!ZIV7{(#}{(AQb(HmnK^noFfglR%}HmmR8Dw~&n99q|B?!P1psT73AvKpWEf-u3pyVf zd8wTq9ihwzu{ZAmR+UL#1WrhC8fsacevaSv)^J7-P|XwO?KmedWG@ae8O_m) z=v*?3lkT6bQr_Hej%;<-6Efd$e~8HI$Q3*Cu1J%yme5$5zr@dytKd0K6q3=Qa%%;O zVaBcD^Kplb1*`2EQ8WwdK_%RNFxOQ8&L_RWI7ELeaV~do=APOhg9eqhX9UDGK1Diy z`|%?ZKxZ;w?8sm%oA#t_1s1e1!Lg%x@8G!H&O!iVP8uHm%jmeI;#nrHmfgj!a@UZ* zgwjdww|P{K@z%y$uJg2X%tSUW$^_|0M}Yz@Jw-pb^q#3KoG~sh{PWjZymrE-&b6cb zxJ`C1(&pX5Sfq3)gkwf9pzGHaCA8W#`HuYD~SU73*~0u1d#v<3e#%`?WrrjKO{*TP^{`Q^swx>6HbA zN{?``1pgN+`J*yKyV$?t~($03^T2hIQKKeHdpN|$6$PEi#kZe-0P%J%K zbG<7NRLoaqr&g+Cs$w>g8}Ke@S8P+FZG$q=R4v7r*Pfvu zzbE*s;5{9ea?U_m4S%iWz_No!_ebHo_2)F7t&-b6nbnt-Xp+34*3PN8A^qk1i=%0A zEutR-4_`2rwTp_0=(Bv@inwqXRu+&Ta{%Etx66?a(CksL{zuAlB9#dwL`)u6HplQl z+m^=ZJO)UW0h~BkyucMjsC9^#mR2ex>)_zP;o&HsON*7f>&X2x?01(0>D&6!%+=nS z`Z{ChEOp0>99^Yn^=Eq%Md*&(V}WLqgAoK3#Ufb$XfpUZ1D;rB8$V_mtTVPo>n$~e z?-GJn%8s63Eigtw3(tJ^+&~R_(9*`^0$L)GS{TreF*O)nS>S&iJGiEFTphW}4OEXU z&THMc#AvuWzX@snWM80XC9PQON5MKOK1>#-pe|pm7pu0gs8$fE*||zVVR@`jjy$s~ zvt6`(Qe?Njb>J&nA-^H=rF~tKOY_^;S>Hy@G}Rs&rO|ja;*5~YIMi*yffFVcHf6Z< zs;{FM?)1rexyZNllhRb;R1?KYZ$mQXbZuTj5y@skx3;#LfNYA@?re(3+d*FGC%6Wz zNQ(NQ%}u{nPXGQNbOUsv)#fDAd1Rsr00nvS5m;5o5`bw!_)`dh{-i=nV%-7{eKxGM z_jHm0@Jm^63AnMK3W6C$z zxoI~^A~+c~iZ}g0gYEPrOgjUNj&Wd*P;DN>@dSZV%0gLe&lboK9{;*DH!&$3Zeef0 zj);uR1Lff82=61!!i33Xg=!VCC$L-+fuUQ!v;y`mRW*Wm#Qz+24t`H7p zx-sH2`8RBvhnMbI@zZ+Qfl^0Bu5W(o3#avl!R3n^`B_FoDxo<$>H;t_lm`=^z>a3% zNg)r!OiZObQuhQWZW}%JN*O~N`EKh?zm41NycIe`%^voa<)raH^ril$8@W}(pB#bT zEc4ZfB^K^!q3obzyW$?4+5@dyUQdq@YD}oJ@jKC_-YTG%tlHpVy{xOe8uDIi`hqw2 z2PyRxIT+i$!)s%{BMXbCaWG@m;}45pkZ#fDA>?v-2OL`m{$G3F;g9wH{*NeeWMw3K zZw(_Mx2^0gD-GF%?5&7|jFxO6Ga~aQimZfC_U4v7L)Q0tpHH3Wcg}zCJC7dcaa;HO zdcVeXU9anUUe9L+jRIwgK5rYxJ4->qcJkc-3sol!-$U9ZBf7OXJUPmO&G#uHzHxDJ zFJBloifau44?{t#*j(*bbE$?8AA-84at8ENv^sMtbo#l{<-21S3=MSv1`YGJSy*#JJNIZ3*lFA$~4*wA*klA|! zF^l^MsP?}OM6xOT&k7UeRUy&q;n_6t#;WVHVTAO7)!V<4SpM$-^dR^or^l5e1N^|Z z)f)`1Q2i#IfZSqdrULeX8_oXPjUQ@+44_U*gMfZZ`pJG|2}KT?N?$-?j*R7Lrd{ep zyE}!Qk2~;Zf83ESV6TMGXl~f7nnd2=L<2sC`C$w270GJb!x@440G0bOiy7TDMh-;=PtiYjhahq$ZwtXCUN{rgxUlWIUB4<)hmJal48HZ+LH-+&~ z#`AwvpI~A13H_O)oExuOU(PAZi9W;|7!(xSe7jDZoQ!M=5CK%~lrGTIXRo{B`#9Rn z?qe&|QQ&g0!kckT_v?=WuABo)Q-~5=7t|-SMVI zY8@66b8{*eE?h`Iec~fsv(o%IukG5F; zXuX}s;Jq5T5j8(Pj%!y|N)F%FN+^S(uLf8;r}jSdb4ZVUoux!duM<8xuT1D#%OhPM z4uqnC3O7Yaa?GuS=BTPwG4LNCgb{Sx+uKlGY2Wx{arEHkG1b5&_;58%j;ZF_{t^E1 znl}*8EJ|h3`bYlzK=~bm?C^O-_5a?TjHX;*ybPM+YtR4j{`S5|_@O3)qBAR~@gHmsi4)&BvK#;6p^>Sjb?XdKDr$)>CZR5d^q_I^O>N ze)x@Bn{oIt+&Y)s6$F8SG-L9muXK_{dkTTbruz}vF?=-80e2=z9cu6XLb>N!f6mGR z@4$RLzL?Z-o$K=C=X~z}I?3>TYlPtF%Py|z&VLF z{E;f%XF>>(NFczT|MB4i!`%O|UsM6c@ z>K}g;3x5x}Isjn8(2^dS_Qy5zw*mb{R)zJ2_&>)>sSM-A8#3Z0I&fxwe-1Fw#QWvG zIaK|FFdPNMae;w>8X;tC4b9Ej!-s`eBB29K1sIgmr0yJkws}d}{0CB$1jd?nxY0_j zFAQaXY1e=f?l<7^WCz-sj%S=VQ>3sK@QG-fjt-I%+{bAirqeLxJWX<7!t`;GNpd)u zIL91t=gv1U9_f3K%%zv`V8wk2$=1)%>b7IUaT7=HtrO;0CH(9fJNT~@Xu#kqyk>{6 zTEYEK159vPMFk-?2$T9iH(^qdjC?doUj8r$n8tr!7!u&Vf1FEtWMt$xD4A$zXskkt z!@{My=PZYx>)jDU_}^7w_uLvW=2zow#0Msb_laMNGS}t#<^0zfD#MT`L+Fmmy*6Lx zzvH?5nzgRBwhfpZ?k3fzBnokeh=|+|52y6%Zpv1OpuYiEW?qj6P|<>B3k80D{*(uG z6#!6c2kJ7=l5`md!l0I^zw&mV&YlhcK74tQ9!iEps-8Yw1u&GLbNPI}p2lA|p*>6& zDRCRk@){HU!~o5*EKTISf89{kua6#&EDn(7&I~9ys_+% zH?o4rY8Cq9NC-7Rf{5|Tm7xqFyh3UbHnzYZVg`QMRsfK-L7u{5ydm=T%A`u{>Fb9P za8URCB`BN~f#_mET3>zry(*;tMT3|tv>GUFL*!b4UlGX`6bwiZhFzRo3_mqrJWO|$ zmgBm?2;J4c*oqF~<7Qa=7!|*`wi4c`L)YEf-QK(bI5W^a$z}Fg9v&!nW>HX3xTL0* z0cstjzGWqowS7JA&B9yh{RP*LAUu2=4U7qM9TL_NBPd?bBnS+1{XV43qCTk!($qB6! z2*D-k;uJjDD#!#7v`6Ps*$U^;o=JM*qdD-*r|RtNfTJtWm) z%RqgBR7*(4$=Kw}9Lhtvrwg%Sz84oW?=qMc4OUHq?R+4$S z;->B7Ww0r=PEUX=5$@Nd+;L~nERL-w7J~cFcFg;`-W4YO6U~JfQMY5uvekWkh5cC z49yY_ZQNR98N`uv}O}dsnQ5V6?-S2@EWr&f~zn)sKxl|IK*aRog(BVQ9)5m4;M^I&-X( z6y(f+C&C}tuy|5j3e`}9!V$895HyeQtO4J95yUV*hW2(df}KAi_dFc(vvQwVt1oCm zMdBhbRHjPqZpet6+qfG+G&+jk2EqjCP$A%w$`JG%o4x$^KI;5?AC(;$Bt3JK_74Yk z14oW^hRb;6Y#F6OC`&c=eYa4yt&6-bYp)#-RfP_sbbx4*7(;#&Pl&Tz!AI zZoy2-xG=xoou)sWK~j~RU71AbDo+ovn?#3&5zX}!D%gzBL%k5BZd3A~{Ls#;mwq+- z{8bWvbsR{pDBXYXppzP*{17y2I?kFrTjI zHkyN0{uvU9RaJO=`JgL@mb(Rg1Rho^y0=%SG_JNcRKi`DrrmrC3Lt(h9tlfgNt|u6I{c*N?uCgXClZg*HN_sd5k6qP> z@;Jd}L(D2i%Nmq>7RV1(c(py;+#UjVa*fXi?QLkyUxnkwvr8?yLAqtNy0mmUz2aht zdu+d(P4P0?drKI}mYaC5>5MA2J&xHZ(=T&mE2!S;P?J1LI06ne)Kva)k*bn@bBEU# zSm8eM09V<;(qyUw!-3QA@@TiBA-Q=L;lmOm99oL!xo1S(z(Jbd)fWZo`TN1aBz<33 z3LOWlS^~K3CFvoDLPUEmA63$WF}K^C+Y&5W>b9Y9hYHR#$XVS|dpNWiYv3E+`^iv4 z6iXMNz4>lt3>m=wfw)zz=XpA<7|RPYa02Gq--+SCd5{pDIv7SWCb44?!olQeT%pt( zjB3h+JSOu~Ww1awH`d&*Om>WapK50r**m*Xcy=yfRxgUrsAuAoVd*it^D#kxok+p; zIWyN|I5e9uS!mZNn8+@IX5rmztm!@AYys6LjNRO8y;Wp+xq1U2cFxXEx3Z|JE3gvt z4p=4qOPVE*dni`L0{GfQ@eQS=aR78K<$dwbx?N5%RsiGW_Lj#_xD`%dl2{#Y#R5Xp zs^@iDKdYL;ZVGhp5CCeFK>}Ev+-cs&i1{0Sa^>LWz7Nhzg*R>o;kW`Nbq!6;s70Zf z;bF!A`hv7L+F%|@|A$bZ3FdnW@HakQ<;~4CKpVP8SwynPvt%fS!y{S z2K~Gf15FqFdiDhOOqT{+huR^!l!yN9@sSZGpWv}0(T7qgI>ko8(MYe3`$hxPo}7=5 z&!}j4F-5JB0)u z4gX>cOuyo!1u+7#!;dri1kOX1>H&ByIRvvuX``vBS?l8ZJ**T!{!ICWr8TW5B<)4) z2aoSsR5;JD4(8t z-L=~At{z-4T}wJXrXF>KreJy5Rmq%5XO>*YLxX5*k8ONSfjEc8r9kz zumZ7q8tj&nUc&mTDpb=A2 z@w$uF5SQk;e!e*q8#rqt>+bxk5I)=6xE4j@Q}gJa&4Cf~Ud;I*CcT#!ae5&b7`i(k zXnP5Ma`Z@Od#XamUS9< zx)ZDb7FU&fx$r|L&*j!3=ovLu(|C2{+SD6}*)TyB*4;Kh^kcI;-pF4~<8iF-rkun~ z9g%LcpvB2Z)tc#){&7?vci_9sm%`Hitf48bf@^9enzGt;Ef)-74nJs&>h$; z*dhfoHT@*t+sKZWY7=Ylp`e1c*d}9VC>0K%$MQ{7*U5a~yynkO{zkdHyv!>i@^$Ul z4){csdJqB|_jn#exe7Pba=#C-_hvzJsIX$byqQ}skJf{rZ;zmMC5K8xpRaqgnDiTE zPI!G-vCt36f|Oo-K55o5fWEfxYzgQ`O=3FVdhcuj8BD6NVTIfccBogC&D#(!%h4TbP%;(lkBsvSk zK7FqST}%*0Y>AEGJGzB8RYc-|+)SXDBf;%mB4L}Dnn3J%``PXS{^2weiSr_l$n8S9 z!ac5XO}#`qo~T%f`81qnx#|cMxgU;5a1a!EUYz(o+tVSDoyCENL9g5n526L${5rlH zDS;Zt6|eG3s=ZolO-XRZ%l?{ch}oU*N=N*sGc&JF<3&?FAXrYx)EOCa z8zWjzd6k?1ZEnOo7vir7IuK?zUGr@-@-|2shd|bjx_mt@lTn~8&wLqZ7}@<+MH*X%$eE`69F}At^b8*S2#APq^r5xTdPAsy_>*11Zazymr!jK|Q^b z)25|An`S#sL4xxuZ+xhjwEhd#aA9!QcE1&FYrbZ%nk-BR+(VIzOOjf*r`Qm{@9zOTEMSvSEzgl zO~0bwAkU7r7^T!q#ugUZ%h2`E)5Ox2$$WJD!CO$%9w_#3u)>Z~v;_$DrV*9lT9m(!jqo z4Fh?g%G+Dyj^;oeGy8-Lfyn{rVxjL;KdaXM_MuoUm+t2UdOlg~OmiZ0ov;;JFs_RvQ*3m5N5-#m-u+LW07vXA!2_|HtXsSMTU`$EWI5{|1 z{>u3*t`kaZIhBt?ed$LJ6!N9Bovhz_eDdC2sxLOqEJIA1l)Q@IIZY4 zf;L}BL3{^2tkRYGu#=ccG9n+n@dvCVnRzL1NC~v99az>nA>Igl?)nDWqJ0Vobgh%m z?)0@ARS)Lh*t`2zUs4#pNiyCXItm^*_O0Gqtjm=iZ!D8@p!uhCFP&>Dvu@P{dMehH zB;HWOo_u=HlCf$rVm97#J;T$~UB=ZGg4(*HKo9{pS5orG`|0G)WSJwb$%s-}?&};T zKaY@361_A;^9UqbzZJi^brCQ#1I-u(2Jag`I?A1`OM*4okI~h(=y(plEz&Hdr2PSp zJ>ca{Gnn(JN2p(Wg_YYDmBf;W z#1O(-ExpLzV8|x56qSDmXP_MVV~UE2;K6jEkJM`*v;#dn^vv4R4dZl^qeY}OGpzJt zkLYv5o=W;wJ(nH>+eT0k0)N))y*^+LGx1kn>yZOX&3FRp4bjQ^(q_ufgl z&*yWPvIrcrFzJ)p+G;9zl3j&b{whTGzAko#@|QOtsiO+WE7^j0RG#Y;kx2jWkV|o; z#ptA-cTyPSealc4IPrtm&`e5F#}ehJexc zn0!lOdh)`cxK*s72Dau}en;Cn@r8}*s=Zy$4s~gJVB#pZ?nn=_R4p$2be`w>J*rX4 z?e>*PFKyyaR3>gUc9m=f)O&cV~bLf5C`XC zQlfZw^1Ifpi`mvx%Dt~lTqXCm3oV_E@x-4CSx@4mgIP;O&>4B=FLBa(9^O*QdHlvS zGTlIZArD|T^v%E>s&$9@`Gu(Gebt+KSMhKN(4Io4szK`G}FbZ}1vJolK4Ns(}!rt2;=S1T#6 zpu{YH_QO8Z-!vZ^{TNSK{87~AGYjK(`L<%p^+x5QzlU%?`_lnU?1wg>^V&$Mk}po= zp!ZP55-=%tr4BEcyPnTvI&nZmxLj@Z%+g#7%0+n=EEa}$#NUM!y(*`-l_harJ`mTw26cCKr86ND79%M@ zpHoI9kC%$3Prwr>0Aj|HYzXYG-g~_UVQhYVQa1z0)&p?P*~#BEcghE1IB7Fk&q;## zhHx?pMdXv3%21NF*4q)5DRb7_dZWq*citvqf9zvp$l#Eh0iMZug)f|5qSw*M>3vgI ze}mkz z=ahVJu&G^Bpz$p+$Ndo@>f2FJJC)cf%|k;^%Sqs+F{G%1(tX}~{27;p!GXE2Sqm@T z;gnJcbA3vH%r~O$X-vsy^CBS$6opEf`70g_n>#S>-EqpyHf;b?-ycI*G3+ufyH8=( z|1&8gmcjY?y7FltnX*RoA)Nv~#R@g6p8v+$+}Oyad!onfj9rbjF}9jYO}_Jt?o4~? zy#b)a5O=B(RTAWIrignYj8id;bMS3%lW~ge4#lX7#D%SA$%P*e{){7t1i^?hWcpyz zi*B6Ky6+**+M(kXZx%u9sI=@RaxcMIC|JIga~t<%_V9cWX*V7fBgcLN{vsNTj$mdk zWU8Ke@YoJj4Zy1#m^dfZ*nanKgY0%SJJy{qN79Yj#G|3%rf?VKD$GDEyO0hrfNI&Z;YlIkzXzt^hiChvdxq`u1)t7=n&tqr5yH5NC?|XYAX?D7 zdK{j*>#%lS|GswBzC0+b)wgS^Q&t3SlNgL4cxvP_diXDDQ?sU+i!VRgG(0d|H27jv z2&p|+nGp(H$Oh-ddo4HC5VD$w997q=v9x5#C*(&Ru)B&!z@~gtS7KUsn|x;83qg_% zRI2!?DI9}as9{d4xkeubResIXs;3dQv4D)3=t?W)h)(Ri%lqC(TjHjW7cdfxf*W)q4(}4hD}!( zZ{MBgAptzDRx)MjKLHbCP&FV|uG12^Al+0hpd)DjVSWGwihtjzKFh*1#A+We`6pUo z)iMPd4`JV3FP@NLt1}DHuYW%liJ~_AdP>0f-@@ew)S{91^pBx{?fhEjeHha}Gxhta zKt3`BXM&8+Z~NDCVc`Vg!ILl;Lg*{j29YTA3C6;fY^~ZDB{VE^Of&4BTJ2CP0kc&e z7*6nnWG?}7cHNMIw1pgb3GfmbLQp=C2x$aL?i*0}{tji_Ua0>JoF?-By2Aj^96q3@Yq}PM<{U2Ra~qHir|AT$JDL?TXD_P!pXv{#?{4 z-5KC0HooEvP{_82qDVes^1y8pnw717R~%va&IZVucQGsq;T!9sAO`mQ zjM;tP2Sod({EQULA$uqD5cbbOX&MYSawHZ3#ZTW+jyO}mc@&)ebnZQ(V2xazk<7`L zY&`+PB8$ka6AM%tevnfA^cu`;1usav6hORe08YR-;uD z2VkAIf#iqX(BjB@TM)eEubz&7VUi4gb7pO^Q+><6YV$+WqphKZD*LhtN-fi3pl@=S zYGV>ODY;LNyx|-+rBqWX_XOe3t%c$y;Cy(3X1h}YaYyt%8&_Q`VxKrc)yQGs2P8>% z{uUAa7F3<>YH3kE7A`hhbA;h`1!zZEBa+rrSBv+*I}IGj%UWGtzn+?t54!~qQsG0v zPZ+@W30bsN(}Bm=sdXZl2W`8MKAmEl%OXcCYQ_dJuw@3o;7willqfmW*P`}o;^4Ty zz649Ht0QuS^5;@tl0i9n$tz8lWtNFQ5>%E*c+oZZ2>KmwLeTjjHVvA*zhr80xOzdj z_Ti&~M&`NS=jdS#9PqQHj+<6LkIzFIgbX&!odp65hkEcFdKQm16i>(;N{S4>HxE2q zWm|5bp|=I8HfiF5u-|45r3*^tE%&V^m414BlqErDuA#`}1f0L=aE1$B-GB$9STR}~ zg`)pGu?+a?bZb`)wAXK7;6}x;D}3_z5yrm}3bxQD)3WN9yNroNX-KsIBxG}4S6lro z2S+`cJ~erB8;8Kg!huZyDxJ-JbL2S9uK+j<&byt%fG;F2k-V=9#s_rfzEpa;0s(w<;yM(iiXiCj zAy*S0`5q5q*)AYOv{nB`1B%{zg*a)5mFVv`@?|)%nD5rd}jPi=H;|Y6ktIL0F++?258c59li3Y5KNdMfLz|&`9gL`7)tHgMrDr4 zCb82!nT!~oSLjhpJ8cS(j1Uk80ebDjqk$AT!aZ?Y2s}7kYVfG;68Qsfq;F?J*41a5 zzjwYLisoqBA;B>2P(!)@`B4AWYd+88+7# z(w1rmSWquyk#ZmfHNOh3W22@OF85`xq1)XO?@r?RGJt3ZLsz&pqzY^;j=#1XAVMqmEX*v zdOsJVr3!%eraGNsoQxyH^hGed%OFxX*7W}3zSHvp04sZXar{F;TmQtynnp=LWSB15 zs0wGLP&NKM9=EOxX8&K^$ik8tKKth8rpdmy`e(WD8BK9f?)dxs_Wm<+9U_~XHeJhg z|Ne77QIz}PUqA9EBjJQ9KERhgJTihb7+EQf+<|L?IHcYfBFHfNK70Sd84#Spy}*|p z_FVp-;Ah`Zu>9TNynaZn75(?8c=#AhUdQX+Qp==WP$ oa>0T6{~x|@!hS_OZn(R|N^6NbM(=S`vEaXpa;ma<(xw6b2QygqO#lD@ literal 189229 zcmcG0Wmpy3)-`Ma0i{bCB_%}ZW+M&KY(NlD=|;LkRFrO{ySqCSB&EBR?(Y27_T2NH zdye<~{C@a6+nsC8HDk;%$KsQsyd(xX2|5A-0*17dm@)zaG86#;$qN+)JR=md!Vms~ zX(B4BXeKHtYGrO^qhkF^&p^_^(!j<E9AWYks42CZhS5wAB)9W*wnHh$8 z%^6W{t~~H=a`oDv%wn23+`L>Qq;qp7!Gw-|;WFLip%A_Z_LuA*;XiJ)zR=${&D-*4 zNiA1QMe<=d?LU`H86-m6s9`Uw3$He7=$<}Xj^FWEnYG=Xy%SQ+tQ9p&3t=WHWLX-T zFKb_JZRNB3;@*$p-p|4m6v-G#he*2cwP|u|hpKIM?6~UW#%QBZQ4;av%=__to~{S; z1_;*#-%7Q;kVGyz(8gY$qP6my$5YMYwp1Xf-%G^WE$xd5W&5Zr&diV-aNRof@%9S! zu_n@O+uf-k0nY4#hYz~NJGt^=<*A1nd@ejI#b?Y~r+gHG%O1Pq&)up%j%Kup`ss_E z1Y4Z8`&jhIr0g5^jX{kG+8#B6c3u5@8}vP)Ry~+&J=v;VPA>OAGgH=Zj1l&nkCXiu zfv&b~vCE1ly7B1t9?J1|m$4`ie#|qn6uX>IQxwaYFn;U~R5#{o-~V`8y-a8bao=V! zGT|HR=glzm6Jg%=U)aY{%Mxl;&q<7adZo)!h2KUi_(3QBu3g7t`^%$q2la#))cG)8 zqB|r3w5)0sdrvz&)hp=Jv2`o$JcH+_JLIOqyAYF_l1P>m!zshJUI?`kd=Bh{etGA- zN$~h3o>&MKmWi9z%d{+Xl2zlB)ePg5(~%#Wb6+2xT_aSfnqGHotPVVWnt4OL+n@V& z#~4Okf7ip=IM2Qx&Cr^9_=z;DbePX6*@N)K$He#3sKV$wRrCY&2l!+#IoqQi*yT98I5#-HtD(EsT4}XP9tg_o3uUmt z%wLtIr9S6WJBLtqBe>T%)u9@pXk9LzbP9)=b)JM#(s?e-E(-nZwN0`~dNb$Mq5b2L zjjoOE8%o7=Kc<17LU`tJq5wtUr>RahnX@DA-2tBV77 z9$p}kb{$H4v6X(Fd%hP@j_sype%a_xVO5DJ6neMJg&p==@|FwlsHKI8AH27yb9uLQ zlzx?-DbOIE`a>nP1lV>2w(WCwGZ)}f1}do-muMDX*cG4fJs7mMeMp?OXuRk zYxE*EM#8&G&B+WKwX1bx1DTi=(Z)*%#jb2TU#+CCuZX6cGYo8#Plo$QR0VI#za~Z+$%n;TiF1`J%2s(w9fMairyBNjr-$?$)4lb8+S3 zesF7e4xu{v=GG0Z)17a6qun4u@EB-F8_LNcFoMUZ2#CHW2uR=&BKSuF{vjYhk|BRS zgA7f+_2*+GFZhcGD|yKX2v7uRF%eZK#La};@vv_r%~qU@8P}Lh8rH{Zz!mxHJr@4mf9?JRj(_fL9lmk zd;e_l?Jlg_CQ&!ibw7elRKL-CVV{G9hfFt4mn;2>albhHf?Eradl zC5*b+baOpeWu@zB7p3zwIS@4&!GgNKBz}1CZW5jy|BsYndU*DmZYqLD?CeI6tMS2YU{mkMD)`+x3B~h<%O>f-%X`})1^8?UA+}1w)759KI9V~ zJ>BHpBd{Oscea?pR49D;Z>cxL$U(WB$8k&%(kSWdHa$DLUn9ap;R!z|&>dgp_5 zl`MIii>)eIH8nL?2VYxJDhpJa1+piu$L439+)fC^w38N(Y?z_Znt%Wiq5Nu;auy;vSb#-5D z(beLn)A>7ReV*6Uw<(;G<2|p>!jh8aARNXJ;NH+~;6syaWG1v2XwTw#lqxM}7!&v% zk0pwta9#A0(fjp7Zmv(73ioYqu1=z{gifJ4E^8U-Yg%l(DOlLXa*$2|$!s>!f|kHl zQQU@-=QTDN0>|TerEX_7F#^t|0!PCtVF?KqetVdRsO8w+`iYdYt~-J?YngG=p}Cx= zIa!H}{M*%O-22bV3_Epi&ehM3r>vIzmM5R#3?kw&|YK zbyZqSEo**^j*99kHyK(y8q;+wXt!6NzPaAHp~LhcggbIMHi$nKq%gz5;AUW&ghQWE zro-^iW!4vCbaL{;G<#jwjGZX{xSCtg%8_;DXnthc7Y~q zqjBgAl$)vZ@FGnnc?o@qXEzGE7*;OewU@0sKf6mO8^s(->2WGzYGx*$WNkWWPQb*( zWGlSksxndUT;_H(qP{(P$l7$NJlw8k_K{Iw#3(p*E2VVLkdev(Yg**ZwqxA_>7mF#;{Eb3hd|HPqdvk3aX|nJ`{>* z%49x^d7xEk9{Tm`)0+J*rnrZ~JFeK!)E%8hnM3g5)q`ug<$w zKJ?<=$+T^-897PfT}gNrEXP1(hE9E8H`+t=7WXHt(P|uy{X%HJd$XZ?qlwA8h|NFM zqvGijEPP#kfuvf%9Jq?{t}%#r`5rI$o*DB75v3-VV4?Hum2efTY=igbcLbLrRA{KF zTR1{@>q@RK_kuk)D>J6{lRU4T2Q2m`)n_37ATJK!NmcMZnC?(7&|IRR5OQ;>eoY1d z4>QE`=3;K9sqFmaaFpQjxagz0O}4esg1G2{4L^-hY)vJ~a#Mt`oyLn^Gf6xWJF&<} z)6mF|**41wjbGbND5@jDgb6X|8u<4}Tp44x;Iu3U~nHh8U%H6=9DY5y2OjPaRj;F-7>Jx0g zF~mlFyx(CZ!4dOKggGj-8Cxov^Br+fBM|{X&54NZO0WIS8{C+E%8nr-zA=CNL>(D3DPs#YQdWQ4*d zvXcUh#l&D94qKD3Yv2*pJIkLL8DP%mKYb}{xus1XhGiQg_~hm?5_6erP8%y%M=@!~HP!_{ZYaV7XCsG(j^ zCrIw9>e-f+=Z59Y@L{6s_CsF^muylose&fe`=+VXGM&t+ROjgV7I z*d?TP$-oM8IWtjc{M7Bp_4R)=d((CDBTMVW6q8Ss%hZPISv{Qh@Pr5G)Wq>r2WuCUv=$&Mj(6{MvuEe|6MtCMXM$ zWW3gJ=st#bKa#RZP`^?x*2^PWV}C~B(fsGsLs<&ZeZn^{I3NY*4^@}YLD%$-K3QvU zr0YPz^E^Bo%Til18#&wZV)*5|PamI)9^?8n(YFq9+!1U--ZIcyd1{lEWdnx_bC3;M3u(UHN+B0`OgjTlH&|Q|`wTSusm4 zLws%@ptX2BXN%!8$r82Ude@g{+Z*umH|?9fUZxV!TviPEZ10Y|vvc zCv%g#w4|_BT5T>Ak+0gbXYrOTt-w;9Jy$jb3ljj_Fy3%;@^MDYy08 zt-Rq!l5jZkjEZ3SV;eR4)tZs;MNA&TNX^rgM#bx3(W+(~)N~?5_m?icyel279(#BC zvZ}UE!IEkh<`;s6qo%ByW{$>)Sz{=#Rwdair>l!c7G?~K0&~NjI;Ll{D#E-Z_@YN^ z)^qcwy-e`v<)87OH&psUsbFa7LnKd&Qf7@-3vmhXlOLg6gRCFs1yTwe{NHy*H5OCB zc&ZP<$?1#DN_6RJa2{wDS;tbXYHfdjjErnaMOy)}AR!ita!7rcH>K9y)%7B;H!I0w zat4JmKM_T`f3N;P1rilD)sGY22am&cA0|`{2Yx`2cnT`wDq42r%77&0 zR^uV!+uDeJa=Khqc(I-_-WsSHRU9-=88hB%C$oK)Ka%#SNajVw>1f3lqa=4a`grT7 z#kren$F#;(JlTr#Vc}WR&a1@u{WD%cfQxfS~lVk}& zMHp;}QO&*~Y<12KHk%t}2ts%J5za}g$o0^fh_-(AT}q)BAMO(4rsF>c%`L7}6=!QM zEN`)sKFN;tT1k?ewx*`$4?cqi8NfT~H^i%Au6T-O^cWQBqls<0S!>}EvSan}@nnvn zR{K5;8$CgOC|Jb+Cdj`LHpSOy@GcSWz6Lql_7>6M=ERT?zKSM`@W^Y?rDPvWbS=l# z{*>7sKTQ`Z5QR>VoX3*yC|`;Gz=IWnj+S)aSNQS^i;&^>dzYRPkoLWD`u+x2M?DC+m3M>2r0#YotlrQJIS>J3}^{i|TKlTiy zXmsOT5&2h;G9!&cl9<)sdXzJHgKoNiX1US&=TqLjRy>2M83$YHs@xrgv|`men6S4W z@bDf(S)SPABx`korrXYO!}q*hS@F<2DdnSvL*=m>#{-JCsI==Y*u;4kQxO2z z0`rY_7AajYcjLUM{ECW-!i?-*98*L;Q0G$MneC}$?g?il$IMkGmdfFMzwJ~}cw0bo z2l!2=4ehFA9d3U!lzYytOyICzbbnOKs)ifiMI|LwEhe`$z-}Sc01F_}vAHq&B)10u z3@6kriEiY+aNGM<+Wn}m16cjXEZG`Hn4hM?>X6bAYitK0PT7^gHfPLW?b%4X$LVuq~Jgw;C+7l>aCU(!H0Dn z30=nXy1|!EbK?URMFplwo>qO?AQ5+W+Bk2mfPPqI_&C_;WmBc>|Aq~j7QOKN+v(G zu|!5zHlwl6ba3lGMV&UTx9rCMD3wYNm0Dv$TyBYHw^4K0(LCCoPZ`u$ii5k{(0I zJq?mP^u7dc#!0HxL09LD2|#>m>^k)wHE?~u)!7MDI%{xeTOtp+?M+8M;nxfhJPX_E|nB9j~ z4_q261&o-l(wLDh#k@G*B^4671C~uW5J4nFdCQND;@`-YW@<0wY~Qy?NMSx+5;kKW z3ro6?iC$2JKKR}?Nw)+Dh(+pVvQW3NcN8L2F{m%>M_}88mm25kT1Q27+qD|GiiChE zypK4FEmUu8M-1ZY37o0AON$|rY=FABwDhf_hdC5MhGocB6ZqBSTLfD%EyZOr>|IvG zqbKBM2xZomu9vcJ69KxGLo5DZEc31pq&hwvz#_@3$CaGF_2SDMBl=ZW?jbIJe5u+2 zYUW+U85}r##733y;uL$#xCs1t?I{9f@|sNsG5x=ALW3co&~ntT-UX}tp9B~924v&f z7bb0J|3?n}jezpzpaJ|C1}mQ^$cBG=-4qV&#(^~t_2 zJ0Rswpe`*do5G`9`2bbM1$KQGZLrW5$Y#6zoo16*`z->JrXRG)<7(bghNU%zdcI9k z9QE{V^LU9K=tJhe`7hP&H2T*G5x`xT-g;O`^4m-&ukErFpxUO7a!^6y751*JtzF!q z5qt+F#RH5}TXUM=3?~03mFLxlZW_z%_7Y&n46e?PEQL!;OVeeeA8pOjH`zgJ9tyjc zYA7quM_X383LlAhm%LDv%?StyaM)2PC$9z&ERBVvbXv!Cixwm=_l|{52Plgim-+Y) zhde+WDl006m}@`ly3RiGo@mRT+Y`%&gxh!(jOPpvDnIe}OwhX0Z+NqwlSTJz8B~>s zJCCZVemYF%B377kYu)fWZpVk%YiS+v68EOfy}2sP%dSvRQc_~G>n3vVuYs*-2p2YC z3Lf^qBfjaoxeO+AoPL8=HKc~23Q8iN8mEWWNS==2O6fyt)kZnX&^L-Y>Yt$r7d6>g z@&oYRqv+%dXCfG&EcZGiRB2N!!;kz|Aup^nRP7LF3$(bK>>6e+a;5_y(x`hMQ9H$;|l$NC_IhMPOumqy9smK|bYz3L%<;ha4U zALQ)pOj%27gK)HFL;<}&>T@6ZojdNFlB#NIE?ClslY1dFwL3K)?LBJ7S5sd_!&a+J z9wul7LO96BbkDW-hWK6|Ecd`^N1d0j!?=r})gmXCsNsrIXQ|ZhPExotC(V5utYww~ zmgy01p)NR7=x#3BRuXsqC9caKfVeJr#O(tz=KqXq7kE0-8H*eEx415VGuCTk#Y+G}r;644}6UX(%=(mkR{o%atgL@`` z9rS{(&Cehc+KOYcii6(!8&kI{%_ruc`%Rvu|9)^RA_#Z- zqmrda;YX$%(hNvq7HY|$h>0>IZju?xx-GhMbb5Bf?;#Th%QfPO32rHih6b`QSmXsS zJG-cjjm;t$cZHp-a78?CaH)uhh*>vdj5P*H2rZ`$1vm_BL~c;tW-a1wjo1d?9pJI?7*=)(Tc4)aeqOay6s3VOeYrU zcN@^6fzuD;Fu4P_sUOZg9_wVc-7mLGEob=D*Cu+DH9wH#T6%JS33kCJ_`pafh<>X* z;46pzD@=HK{d<|5hnGKjYgPDcHbfrj^+aN$!VKYL*!cmvl&JeorAR;^5b0L(|EBeW5kl_-MRfK2-9{?mgseiZX(%FI{1`f6GlD z;j=cUY^B@3e(EXiW54fU_#-9l;n88^(ySIT={Hmy$pZ@6KThhSH~8GWmrD4*KKB3* zd~Uv8p!0hNsutp%{$i)!Kf(f^9o~Vaqrtx_N9N@~s<_PvOuZ9KDOfG4JmXbuD&=8-O8Ds9q4Sgb<*g6-PQi_4vpB#pI~i*Te9C9kF-$ zgIz!!b=~(Rn}5gwAN@L?uVpQrtt3SUk^2Y>vyUdB=eoWU5z~=C%=H_Iv#gcw>guZJ zQ^|Z53P?mbOrg`4tiiv#Ih7jZ%g#$Zi}(A#d^W@e=~4aeQ-w1Vl#@YO)=!LdqM!C| za@t)93Ewj=FOkx&fdL%pL*WXjyL$t?H`IBK`wwRa8N&%A^*~<|d-#6RdAtFCbOtC& zpP>DNK4GlbqL@8RFUy%)PUARGi|)iK(EW4cP%2H3)YE9idEs_75u>7~m#lTzBC%_A zKR-kRUr>phJuypBT5JowMQgw_wN_SE=0gZDlt#ZB+dl#be8B(!K_n~>e-W5c?2u?3 z7n3ralhpxHdETW6!!x?H5@h3)PRz5yMlsp_gN*EKDS%o?B8T+WiSXe)I@ra3c?!)_ zFDj`|{bm`pBzMJh0S#!D!CvlGHdSeHKvDq4hOoJYpemK8S7f;iqQcq-4qulNDnZhh zHTw|{ywE8rTrkBw7++3s`d)dVC9&1kPWN-81ln^~50G5on5h&@<-5U%lkB*z)_j7k zXybhl%t&1P{l!EgwcRD1E|!BWesUhSMcDP#MR)a^1^wgAiNQQ|R^FZZ1O3hMk^&+5 zq891T#$5#!Mb5YnKGw+|lch{wa!7u*ht~UyokvDSHlA;NZhFsWm_Rv z+uZvO!F>g^7Wug=o+`T1p#CRT6Q$qO7b@H902r3oK)rF>m~6~Cp4J4+Pa$|nbj_a@ zLs`C3o47kfrlhI44ykJY(ZmD|zTeLWPePbugeIKS)M?_ACjJt3Z#lW+T*B9~8mo|# ztXC9~!*QyKaYN_q@0(v2=ViRi&Rz`F+Sf(93al8}*-Z6+gzGNObPOpmQc7{uKGckf zBigESZ0T9>Z=z()U8eG+s**JHd zId1!I?wn$$LZ&>iV7h+;vugFzx%*Y*YJ8@r@71CEz}d9h+*zUIji9>P_kw`~HrlKD z>;0jf$=zjghy^_JXoUv-6yP&iV~V(GxEE_$mU`|xV(t#3{g8A%YZ&r0Udo+r{jEF^ zgY_N6hfBi`X)I;E&f}GO#t@dFrkts}%VxE4abWUoayaFdLznsd_uZ zIpJ}mH!k*T^BE(2_let=?27~OY7ulROe0;@ny$2s^;CQ0t3aLD`4aDug$pC(;1^*s zT(z3^kiy=`DKETkyFAnyJbTsh0b^Uy*48#7IhktoRhEG>>~OrKRn$hkbjLvo*FW`d6m3uqL|#b7sg$Q(a+9Xa~||{WaT( zq=)8-SmT2T_)xj|wsE(0I>E1CRDcrsw?WF&i>XZCtMq;KDpEY4qTPB0`K{<_27qiz zC@L)q&kuufUgxhlwa{I`U?Mg&+fEXZ1P!PCpP@>H2*MuEOY#qa11ffL0kP!%z<3 zn-Zd**nVX#>`<=)1bgZ_$@~Gu~h(MDH;U9bd8#H!VMU_{}uA9m+QJesPpnBuBRE%+W#nUNSZh<~TwI7bQdHwyy23I~Zy#IHOIw_rxbRv)`L z?Z(TKw_YRJGf}X2H2=)dP^w43j~2d4(EeI~G1~alXdMyKrSqi0*NnP_TO4E*znv8n zg2aZGE&Cs7jhAe56ZO_5A$~)k8TEop@uJ0qwzY1@5EN( z$zrNrXPD<)i_|{SCQRY*{inOb=h?uq1es+1a4b{^!2N6v-|tP-dNdp|GomXox>Z^x z%qeTL7#J9UhJYFfpMEk~|A9~b+r-$&h;CjIh@T$(Vco>$=^9&u^#J1iqcB?Kc%nJN zW87}6RTs%nIHmdzhtqru&h$^TJ{S*8)%ro8EA2t$6gHz1n`UtOkGF&ACP0j6kaWIZ zX8(B4zem2e@<^aP$w)w18M~NFyUJ2qKu+-w@b%w|><>T?g2vsy{|5qmeZLr-zXIYU zi#O^vZ4j0sH3Jj_Z5D2$R*4@#2NDe*@a5R+8Ob0Za$g4@U^-<9%OvoGr~hrtrLcQv)U(i(rhm zzc2NFw(z#20bWL4#HR=Uh5ollfEsNW^*cDIx6A+Zy4wXG**O1@0SkZ`Ht#@E?k<43 zSp5lEOu4?FpSy4?P>S_fza8L=1iXn>FaM4JF`u3Bkm!G5XA`(^Yx0FD3OanX_5d@> zXni8ynD9G*bg5|NkeKmDz|mibDblRd0tV3ZX^y{wsj2zZ(xB1ldYsXs_rH7zXYYPl z6DX!Y%N!@$1BKhF924GI7X2)Y!X4@kV0$$tjz~hw;RjNYiO9>f!3;L1j+f6%^*`Bt z{~m%#&hNA^OE(pn_*_S=Ow7{KvP^*O|4!&H>JaXY#>fDnmjg)uCMiW$$B5gw9r*Ur zC;G&YwV|x;_e44jMhW(6(-z7&z@MOu*c#Q=DQ-=xu8s_nGtTvPq%Y<(?JrybY>Dgb z=m>c@CPd4wV`}~|x2vyEyDk%g()Jgqnx_NCx3p9Jqm`jeZd?lh=>9zQLVwon;Z*7K z<5`o>c1#`I;!hjLMo0ZLdDrT?07t&GS=Q%ZwN0;(NLEVmOB;Hc0Jyq9eS`O0KVQ^4 zm2kIyofX%qHJPuDQ;3>xoLS6M&ypR%(gMdD=hQEf`>j$*MDUUpbxexC4@Nrt_QeW~9+IAX5h z%$0X*yDUR8m}0O*Px9nvm`SkX|E1lKAUON608_C3-p0rD#V{KPjM7oP!Mxs?S_ye| z!m~zUnfdwp5{y4euBq{clFjoBot0JNbxax$2A$bxekh2Em%hJT=Z6-ANxd)v5CI)v zD7hR2#|<_u7~CBo)aHzztrNC>_eO4bs^!U*yXeNS30pPDu^0SHu{;<9vV>jZ`IDe4 zFdm8k*t1Tn`~w!{qoT%h(*PDlGEf}EAplv-79R$8@y~rf6p9mT(ED+z;k_7mR=Je<0p_U|i%(kv763=UDSa}d;koYnH4!nxFzu;`5Wa@N z7S~&RT6hQkhx4ZkfIOFZNoyH{C|)dATR9!Fy@NAq=4KkCqeWjKsZM->q%F3te<91q(3d$;dXm z=1ZCeYVBF0Up^P=KFN{8`R43EF!&Ql>>$~B8ojVjqWQh1O7 z?b&`;`C1@K8cLA8Bbpd@_%kE^UYfUF?QZf2EZl zu8@oEx+1tZv8SyB^mm--ABQy%HT(13ssk>!k(JjOpd-qPP>63*{-%2;6AqCO^>4ZanGhOKV8*q|9z|5jm=|tXRpWk@naec;Sx$;UU z@f`VTZ1IkPbsvR}PPUjwmORi_5`k=h)zzmM0&icH-+FDQ^MY=cG!WiIZG(I$;hd*d zurIU{-J02~ikyv8M;_h|6f*w_ekmd_!O857l(&vI%_z41`PHysk*CW~(07x{z0Tn% zBEI$tbPy_y8s$<^(UF~_1zHB7;S^lV$=D{N*u-G0Y=!pu2N|E2d~ei|ytdzBgO+8~ zbAKY)iDeeHw!N(!zX8w#f7JWl({^y^#OQ_7+a2Q5Pdj_byh_XJRsTso^0vb5Z zX_f>duZxu=&qv?BecK14Q~zGGqNeNGx-^NEs@VotZj!))>hDZw3>>5c=zI73ls}vD z18tYy$1we07z*y6-ox-rR#9Z-AGp07XB7oT#_ELp=or)qWJ^Zsl6%f0O5ys}C)=;{ zo9#%=3ibBv`!=jPM0Xl?2$R^2Fwqrn1v#Ro-$P78=;<}Tq&FCD_MVj*F_7R0OCT^7 zoq}y-gYlpfY|it>{p8hma>wbPOQG<<_(25064>>EQ<6HWjtACGa0!V|LJYf6?e)kA z?r_KYNS>VXe(jR;nlY}!nPQ!Ds)_F(02qD{U@|ri**Revm$(|+%ARvO2)qO|VHBVV zHBk8cFalANR>TY`RF!ediKx$y_x9~YfiG+3)i3nC0d`3kjjRqIK z6o#=YpB#SagLAI0bMZOXrVA6OVHOVcWia2{l0Ca2eKPgD zg5-Vr9-4i-?aO%)&J+yHHK+`Owx&5Nw{|pzt7M@jruaFraNGWo?!4`&vB%9mr&1?9 z(fdFuXA7jOVP(C1OsfndebYecMhO@HNAQZce!wJW1a{9Ob6HK1f1LD#P;PG+&Bcb& z(BtZ8OsB~4P30F~JXnYa&y5y0c9`WJ38$%Q^*mfS=_UnbiBKJ86i_PBQl!l~M{ozz z6B16>Zv^U6(x_vkf#~~iWjz!Zhv$4XZ7J5Qa1TE=r8d(xsg^|fAq&fqpFLyO(cl0b z=&5A=Z2m7z^La<8c|}D>cgu0l(5xRwUBOWB+)F_wS@EAR&B5X2#La3HI$Ol60R?ip zqs3vyX?^_YDzTCvIqm#*WzB8UXhta*Vs7e@q8Cb<7+jI#*Q^&TNOz7Po7QZ^i?i$r z)Z$$xSqX%>JZJD<4U1|9$cs5xVE^OB?O~)KRq-tHm!)9vw7;4b?4nmpkX-YI0m@H+ z;`U+E!kPZRTARUlP#G)ln+3|7RYu_!?QT-5`T-zM#5e|XLFVR~XJm`rGvrca<@Qp>kq;m{_!0&3G# zk3pa{v>ZU)yF{Wic z0}d&XwQu0oTdk4Fg>n`#C9g?+DN!nW)O+gpD~{~ZuyT9Rto6krli)w-txBCNLW{-o z1L15S*kcjXVzDFC2HXef29oTF?+v~M<=-EoYS^}YZ50cun)1sK`lFMD`!zs{Qqst) zKne5$Ig^o1L13>nWNZ!W&o(%aQdo!2-n2e@xM0_HyC!R+PB}7JVm?Q2J@aebxhiMr zhQp2OvTgCnQ^}fo(O`=06k6FRr_g6Uk+h%!@6sdA#1<;6#8pGeGavP%peuTMP_A+U zG;f2%WjVdd85N$ikfA4i@0*E*MFAM;wJJxJ{^_q!Dg)H?cu4td359Y~5+KM5)6O#e za~Bq?`oNsOzVw|IWKL7oB;0+Fom|C) zBvMxFod*55GVGjkLh4NF+I#AEUQoBVa+!ON_2nYSRlbSwpljhbpRYqdCj4@$ zX*{J_dk-T6As~}pQ=Le!bKm_5y5g2=j9x(m(H9$)3#&ey;+t zc9!oY-H4hoY4CLKSZK)>RTPZ-9c^k><>q!(%>i1o5!=KLV}sxF4zD94Q^WR`<~dqEB8s!R zR%}(FH*OK?8D_|kkT}^f3d5PZZMQ+E&F2_}5j2+r*V7*5nKC_lzy*a@%^cM1*ZVD4 z!uUrUb$vn?)!vKGZZnbH>6JoPRN)$T{Wj{gRELsn#t3?2Tz3p`f5n|su=m%vGDoag zZC`(htD5way_(dZoB6cJ{oz6|HawPEMxY?#*jkV4|A9M za*ax(-&C?p;+7BK0&)07iL>!l_oc4L?ULT0!N)QBpa$zWI*@WmJ^O*zGfdxUXy#|h zZcb9KuTKB^()D1uZtB|e^ylpf`!Vf?%0a=DIgzETF!M2in4-XIp+JI+^Yf9pEM2j* zO!`uC_4eGRlH=3HY}!*WZ+QS(w*%{w%?Xv^*E;R0TLQFID<~>blvFVr1m;1HmsGt? z)6>&iNz+D~R#W}q>Ns}arSx$5FmUN*%zbysAXDS3#@cFegx2E>Yd?&dZy)sf3c6TA z=U(`O>iMC(=+mM4M&%e8&JuPU!4T)WT6U#@#<&XgvbQEtuSQe(HmBcC8y=WdAB}|- zd4>#TY-E%QV4q5@3?83Og$6|TV4^Nb(t(qA5j?c2NNOv<(ml#ZJrEso#W+E@;iOJB zV`UG1T*cLb$JPR$?W;Yr3qnQny@kE|#5+RK%L9y&*bKVxCqR8&Bx*VpAC4BaDjaDt z0ym}l`RIgMS^M}Y6v&H8Bda${yYm$S!_{@&D~xF_zBXEn5zOL9FxR ztm#WT>g1+3?%$Y}cZ!o5S^0zk)gUyCUcpbVL+4Ro^x634z;BRS zgmb^6h%)4ivUl%8fq*OTw!iA-VZ$y;KRVh$NNdqX5lC6&W!JrzhuVrQIB5tl>ItfXLv8pGHMg;K^wMMe-^vHQp>|9;H1uz}S*ohlq)lUn4-J@oJkE!bYTE-U zvN5F(RA*f_SWafH6{_$X4tn`5qqd6J68RkJJfxeBC(T6aa?$&>Z)2jmFuyR2O1mT* zDY|bRCb7xYwPbD214ailM2Il1>qUBDw8@7R-xcr!Nu{K^i6LVvC^5~ocyVJ-wGcdE; zrp}c|hLYFk_x_YfiFN7&R|EgVFsJ^|U4pQtHf%>LI7}s8)L?(4Q4Cf6={Ww%*V!V+ zs^(~8tPU{}Y=M!Q`D zRi@^!Jp~T!^6^@u;EyT3tnI@sEa`I$is@|P1lOIq9U({gYjE??AS=PahEkD}F`|`{ z!`cVMADu4u*)^aorxD(K36kt`JzI)i3X^>l>M>QYNgcTYw*w_a;#Ev2f%}ho$|N=& zndpM(Ub5G9nRCxQ`3EZySE`*d<7OIsV*wWB9h3gyMjv!wT z%Mjt?X;kEQ5-plkrHv#dam$oX;3GiJmi&N(r0nB$eH<9)uYsR@-fTP=*%X4dLb1vC zbc4}ssk^-?By8F;mGI)fdLKzh%8shayEZsB7*OXSW;YBF0Aj1{D91pLLcd0x2AB7D zYf~(~snEZm4BOI>mk$!S+-=#e?D};EpWL#i zew&Zq&t|iNv-}v`IQoMpg;Euurq@F_qVgty@^C<%8CM}eW72A;^ji=t3S~pl z3=%nTbxSQ??jjTMSy1&FVN8YoC&$%Bb)CqQ8|$*WvxVQszvUF6D+b;Zv|WA%{>VYR zdK3_k{7`^QOiWnzr-GTobMjWlV&akS-~8oD2QC7{=W3F+YICv?1b=wldU4YZXQaUG z+EEOz!#v%quD~t=&eY)OG<5jwRVfw-3AP2$!bcp9WDJ(@ zT+`M`gm9;B^I-p3Z>RthuKWPqjZde#lB|r8G4Qk0_Hx;k#|^0!u=sk(2q{-bMX>Ns zg0hXiSkCTl_%-9xI4{kfJXIb-+4ILVpAYRRvXR2CXJn^0M@UtsAgZ@{U$>%;ZFK?M zvy37%Sg=Qe-)nBW!^xIBe7CKucSH3vllEKQC+3KY(*Y@m@+^>yOpm7eg9$cUx#%CfY1y zqZV{e!=%Hf^CNq2;ZZlZo^W+WcvgY=c(Cbb>NIk@qEPRBpl9~4Bv}H~86|dcWF$E| zNnb*VU6T)N%#uBdi{UR3=cPQ=^+?2=r2I|VgU6oNZhdSfPD#yy4#i!9wMcHG6YZ|{ zTkG!A6H2Cb4wc=H9Npf5tLYayLU2qmhjh+^#ThW}_bfbXN6&Ygp-=lwOaLj@_fMBP z3>UnkAw{~L37bC`^YSMP5ny>8VPPnFiYHLN8zZJi7Vix?S}F-LD<5#i*1|`+VR2q) zUg-_xk)KPBC)E;D5dtGAfOP(kVl71g+2hBTA%Fx>Gs7<#cV;W4rK^a)-+bbvI9_T* z?vyNWHqfp-r0+WNO_F?Vq!v&unO>YaPZ-CO#|9m}u+1zJG z@3{>M|5gX$)zU=wqsS-xccV7HKbz<*L3Lf6@3c9bb>(JkRM|n%^^VOU%mgXzhbLSb z7lCYCt(1Gy=R(gDD>K*lv!JJ?VxdY8PR@Q#rfHC9ZQN?IeUpf99(=Fqul@`oj7!3QUUyii!W93b&It$K*H)1@SG^vJ95dp2y{9bF+|$>%V5NQ)eE;Xn|4GbXfQ6s@(DTy*WwNFTi4xE@enR0#qXXCR-xNIcA*TjR zmXsx=_S+!CBhJrCLoba(U7FZ12$F>_cYVyqDb5=8+bIU0uv6n^Cb!q|c!!traWf=w zCgX04y?y-v57J2e(n{Ovp`K!$+#F&`%U|W-4`EI8I?Mvucy7M%|FQKJY*lsb`Zp|T zR2mdPy1S%9y1QEeX^`$zq@+t4>F$t4qqKBMOLuqw$GV?;?`OZq@qPeh%{j&#(Y^~3>Mq>|cqNKo?;fGz>&`~j8$(H}N^iMg;gsdw+8`nxrXZMX-(}IaM zDUW{XPvs!VsB}Va&Jt^^IX99?)uv;h7uUfKoeozCGU2m+bP?PwEwfgBdjX>^Etn~WCS4(td%@mW}S-)3!_1oLX5ff>itVrV-Yz{1)yja;m2at`+I&?r46p-NR3yo z7$v>Vw#bqdN9vrT!FTWVvlj20;fJT8vBa-%3wLDKfb#LuYKPa*@7_hldfZus(d!eq zoicGMH$`D0fXba+$W)GN>TD!Z6C)Fo3&}UyM{9=mR$O+}td)hWsR4+#hob8Z!=^s5Du3oLL^h{!yq~ zckfK|>hc!Wf;^2juby(X2G}x;DX*)M86~s=38qtU_5(k<7akR7X)M_@`1n1KInEy+ z1_@zAmiIy@uq){^c(J7w&$J!7XYX&IA_n6hl2_fXqs5-bfZF(vUEy>ukj7qXr{@2k zqRz8`GuE2p!F7wmXa#%)k5RT4xZaR_clGjA1o6R`!Qin4 zRC5iP48lPV&EeoiDIxp+#FAWK$Exjzb+cx`bgLku950mwvVzn$%W>*M?T!mu6$ZHh z)69X2XuKg;I*O>~({gX;Ig;F_JW8$}(H4RYq^f(C3x77@$2>9AU}Nd+I3ugbajrWP!t6Og*4lM(TlLYHse8UQ`ik1g~p8 zIr+xr{t{#jp0=r<%O5!(fp>vhr}@8K$9S@K{HyrJ_$rLt@^Zy9Jo$;RUTvD~`oCvMm(G@Zd{0MbSuRg3;9jZ!u>&!bQ^5eF=fo&4GWR@pX3OdvF0<;t=-DNT08y?Kt>WA{1u9m2g;uOQ|n(lYS~_2J>3{oXx; zo30H}%84I!>GSjY(*-9&w|f*Y{*N3>i@MO`ozpg9PHL?3CMgM!vG79xWD3&+(Tai1 zWdCh?o`42oVfiC0-AA(B2%(5X`M>gGV1cf0zr9DJ{__ahJtHoG!XP+m$ou;BX(Nc% zO+oQETkpF%A0PTrB=IZf^(UUrNlNv|2Olx)9_>XgH#DU#j`O!me?MgzNbC7WJ_nkC zNN?u0Y%E#aUT3NEJ4r{b|74!XvN5x~Jb2dOEt;T^aMu-1;3Y=4H{K;j6H7b07)j5= z`J=CH+HjYv;BAt6GR&<9a&QbTXRSzjk=@i&?raKs!Hh2LN`g^GapE3@o}tp$KPk&)>b~MV^X(L zS{KNYe9CDRjRC%X7F6fmsli>hnM5W-KK*PPCo$dysBOLZ7ly-26A%oYGVYnwJ)|A? za{@Pjllx^y!y1h_G3AU0Z}p@~EC3AYf|AbbrjXPrrdn7Ut>XWNLJSEZHcrIX_fyah zyH6N28{efX+4UG8E(%^9{W1Slp{l~fMCz?j)W|`e#pw_AeE4#sS0~{Y@so;Pl6$A$ z1Yl1E;*n4L;$?7Ug&&cvr_VRHa(2&Q`pUpYgHQH4=V+WVHqP~J$Z^FdvOQ?Ex_kti zjc6kZipE9C@Ee+Cx_G6iC!keHNvW)?{P%)>Ot)xrWRc9P0z=~Kw~OEA|L46yF;RHl z%O_!R3>BglF40loVKHLJH(D(1U+kTV>t%wjwumD(tIFRxw^s}d6t1e95EHD$Jh=gc zCt>bmuFT4s_{#4$@DVp=ETEHarjJ=&VYe<$caQ6`1)X2uXml-rjUQb%asTjbRWxcq zE~T>eAIy1a(CNl_iBlVOFNqNygBIQT#x*#j zjZ|HB!$L}O8K3IGv(?iWFX<;qeTQ2xO{hvJuw6Q>=g-kKC+zDRX6q|ll#>4=@nxSo zTV*ElB{2}7knC=_(4*A4!QdRmJ@z$B0FD(44Di2)$U%9Ok`3O=P~dOvlZ7SEC9M1O z!b8Mt&CFm!$K14==^DzPsD=2l#RDa#HLGzR$17dz?>Z4)3i~#D?;zBQ|5vQRhh>cJ z&1bOxqQ{B_%FUyveyi7dznA}Teld3vY@U%oHB>)jkcM)?op~q6OZ#~}rl`T;Hc{)| zhC%+f!@sLeV(|xqPc)*NIgYt?TOI42xQ}Z$IgIDU@!%RM?{lH1bJhjf-V+a4I{o95 zdwAq)$Dxy6)3PyzgN261E{9~~@Bz6r#uZ%?v8O#WQEo>0 zt2@&Tow{nt-nTwoo$J-_yU@5xpio_*tG`dQu=oJcz|F97;2%lff0EH6S&%9e8_wXr zNZOC8;1p+#qY=JR9%?+aUz}-uSVq3S2GT%tw|Qm?4L1|8+OE4izJ*rgpQCx=%RZ_+$(2%c8_lA- z@wWsn2ZeB$XX^+R-it}Ij;>lS#`vc$`(^NlnH`ENP}olZ{HC@Uw*MpIq{jveKyNsS zPpBp%6FNP1bK=NiOY?9xHgF-uT?QX_;I|s_;o7yEuSD=x!1d)AEDQk2Sx|k&k}^e6 z?HzhSEX27lmg)p(bV6Ge7FYoZk4>hp7p=v1k^xsNJ3m`8aY&aO2I-L}TM#Le1FvuQ%DK+g; z5Kb-Pjp{od{xI6kj_;?@iliWm?NRb*%#i2%G_V9pBY~d`SGSaL|D(>l^R9!J2 z433et`rLH>hXc&{ay_>M(LCY9T&ONChdAGQ^^_*wbIgNa& zS58-^%U93b0$iQxdH0tKsM{g)%a=^n@RNI=Vf0E{$#%{6#a!-`MfLq_Yri%yI8^8M z9qx8YDiQ&8VvN5|_gw~vkWYYz%hRJ6j>)a`Z-bTThOMD%IE`~Tkn{swK$abAdE0~M zjNjwE88D`v00UOgUM@HR!^Gf==QiY^2_Zrs>&*ai3ntF`^Ur7dZ;GIr0<_g?Hz$(9 zJm+|Bfc4go&eBYpoSf`Fd;y86P*^Mw2-nybJbp zIj6qkrJ3Ejt@$A5;M+Bd7u8?^~Fj81+1KwDWa*_Ua&qj+lwEVF0{4jWz#iz(YXh5`NyS?Y z|A+fSza*z2-kt#tR6Tgz#_ye`QupPWIYCqs>^vrwF#aP_785i1lkZ zuV>V4RCRsHD&ouQ>$h#AOBlC45}GaNZd{wlHH@(F*axAX;1hXR9cQkYv3m zDv2tEE33oJay1zLvb1>qPL-9-SJ)hg*|-M+abxdIV$|BdhE1x=d2Wcvc zP#m+X2B}B+1D<_#-gEy<=}0s93+Q-#+1*5aT?;HGDR}vpRz7A6<##&&bi|*V5iyKD zsLO74i8vifEgSqCv})*BkU(aKDXvSsQhQ>*v_t$fiN%PyX}>hKhQ=zpm}BQkYsM>M)u>T00-@MVkV9^KQD4#HzF46F8 zvKY;d-Y=dEkZq~?Xt=>RWRftv>G!HOOjl<64eK-9Kym7ipk&LDce=YCaQfu0Qt;6fo zIH8UR=|NYPD3uK@Ms|t&g>A!BetT; z$IsuU1@68`M%Pj+jW_m-d%A&iI1x|FV6fn6sc-oJ*rx{otiI!0bXQ0{n>(;R8{nAQ zRF&0uG9y526-c@H^)6lrX!FI#OB)&*>SeZe4Np%dpH4JnA_=O755t&=X8}R|fr#j$ z8TLQ>+54E}_1`tR$ceyPZNx>>tb?3xWCJ0RZ-;gUe3TLf&t_oS@gH*&`vP&AU4>98 z)lsH>otCC=VpvNIvQOqV#OSms-iQK!0y|7f93sSMLl3){x1Q53uzlD3eWNWWLUO%0l-D`hTn z&ZVO_owHabk|;@@Z_rl|&KU^eQj$he;tYHOxDPHv`wXs=7KQP}bKIfz~eB ztpx+Ki=5XxMl|IC`@#jq-Z%$1$i9dO2_^vFA_<=M<1tVcs1;=OOpuCvdTQu2$n`b~ zR$N*?|$d*nI4HJ^1M{9^(VI zctC{{{-XgAVj=Ud@ljD3CO)@0AMR_%o1}LR#rB#S{$EUR548Vu^P&W3l4t30xiHUY zw#wXyS0VEMRE@P-Q2z(a)}m5UXwkgjlNooBv^T`d#C}$+@C;44u2QQD#^9qQfxn-- z2iz>OoyeImxw&kP6eN1tC#TQ!{^pQLCSQ<_wE6kDnSZ;-oIJ}J@~AA1QbTugQb^o2 zHXnBlDk6OHuX``R0@zm~zNo%RVw)2vE`}ru*p3E`EAlIxnTZcq=v75{Zz3=kyMRjovh$^&o##l_+p#H=r&E{pLuy_5D!!2qU3Rf*xbDRU zZFPqE!WYSNR6c~|-p>V&MBKua#@nTWNqm328X+ZqNl)q(qipu8dbZ8s@>?&??`NrJ5Xu|k@6}^>NMO(Nt{<@iXo0e7bF2QK)+jn|uM4+D>!>&}OaZ@YH=J6;mo9}aVd-hMp)$SHdj6YjJB z>KS!EM&B9#_~8c3Bn+TiemRTQzU4Vrzug!`mY^%_mrWI`t(TM2Yl-HWPOGR}H|_w$ z2J^F?!UxmjA`P@);y-@I*-B~z&vNa!$k;)^>?gSD3X{$J~_fknA;U0HX?(^>pBcoIRgSknxDXxpQ>3d_zyOs4xAIU@F`t3_ZeyvIZ$ZSu(7Cw(+K~VZW6PB5PN=Td7ANNNX8U+oOfr1k~cy?tzH+z>+4`yT{bHK z`c&)e-FKnp9i#b4wY#NvM ze)_{;TrUyUek15kv|*uG7~V~t;vQ5Wk*JcW^i^xf_{i_XO**!aIpE7LoSdePPNPm; z9gLPGUA$jlZthf7w>-)-eVMb+GbO(p(HFKa`!_HnNL5{*5=yK|kWGJ$cs-V9q(yh{ zo8Ck_*x&ES)CenL_1p9W`6FqJ_72r^Nn2lVC(lg@AAGu(*cqMVx5%@5)ibq`H z%N=`snS-At=~&~`+%5@q=j7g)eN0RH!n7209ndb-$9b5{Ry);XXK4WJ$Lf6jsP_uN z)ru319X83+?91d$_OuM=&=QPqvB$T|E|t%)J=JVnDQFLszs_Ybuw{yF1_Ke)soNhV zum&W*e=^=@Y$EGe|9Z+ULI0<<)tYVcV_~FO{7z0qv-eq-|iGGyFcrISxCvy_uA~TV-}_~-{}$3 zMD6+jO8+MSQrHxRStqSdGTldhQ5RXD)c3#-+EK~er4I|x#-)9xNQiUIX(i>iEwL`; zX*TU^=ROm=h(p;|tGH@&N=r<_=aWb4|7(}OIQH0+-WM*_WANHz89a7 z^o{K&@#02ZswlmP$;-BS6LGvOP^gb4M39w*>>>hMD1)m`Q-sCt?e}@m5ZXgZ;_%Us zC=?QC1>ztD&H2OG$bDoeNG=NSp%-c7e8w9LAe_tTfvOb5$ z&|*Rjvr#2e{wtvf$O6VYq%P5X1!Hj8EcSHmPI+y+V~02B`R`n4X~5;CTEO8_Q1M@| zf)tTT4065W2^Zj>iSkGGx4wCaA7x)%eC~&E?_9p3kIhDHzKg8=TK2K8`Oi;;dQxeP zh>s4v+u1)1O4ulMmhzsuIDb!m^Ja@I~P%dX1;aq9p`EjpsTL|7P<2R zqCzsegdt#V_%B(i^?FQ~b$mc?hUQz+!|2JS`ZL<2uAo&rfF(t=OfQ+5xal;2t=ym_ zdBZ0p5;5<)Q{>o5au{2jf^_*jMHomQl6f4m={o-{Hj&bTvM4A}S>gbzgdDcAzb4O| z<`lnPsJb(QTehCtp3n^yFq0U08SD#jR^|;3IAa4r=O;XVySMu%urh+QCM3e!Uw3X! z?7X37m1ZL3*u_$);KWM}v^c-kd7n#GylnDeiGyiK^dLKaiasG)c28}rqc~qQ+0fIA zZ2nHh!3~gX(M%HCvDyaODS~rkqLRY*wM8f6wa0*gN`P1xFdBB)ZQh!Zt(4>sac;x< zxjsH;??m3wMnCMsHW|SWV2Q}N>wp`vQUJP%a1c5fQs2WFy3kK=5h-B735%BwQMk04 z$M#J-F#G!GSbapsfd?&XmjgnShxaeGy<9hK=wAOZF8Z$21`u7gB-GCWasg6+jenD~ zq0n;Hqv>mzfUPd`q|n@{)papa*Lq_0V723eV^L8Ui|B5C0z6{jf^Tt^i2$aw6oeYC zrylA#1I^>3N0>~7^sA}~7l#SaaXN2}WC?0y2HzY#0HLaDLS|z+`lrP3_?J6xecoak zUB%3xe=Mwd_Mq8=t$x31rA%`E82WF*PmTp10-G$dDMq`=d3!|? zQ{PY%pBt;mj&Ct-SCeT~xoJA&=;7KkVv8Y9nDfW0EAXMiiV6-+1>^{`vHoShe$Tz$ z7M#D{XLnur=%Tv(O%T~m+mL$L@ba_w!x_y1QA}JCMcl(BLG>nE0G!3>SC4U5iDJ7X z1qtg4ixa z8GxfY)-W)<>BFKipy7{r2sMzWr-uZr^10dFDSIyOMXi13A3n}DPYv4%`wPr@oFpFh z5|D1VEJ|S`_H+?fy>x%Adgs1f#u-KDgR;817wCVCrT+sjkOtihvFzn`!2L4?c_6H6 z(&!s|maj_?zaEKE3^9j)QBpcz*-;W33C}ly-64Hw0C{+n&$wSScNR`Q-Ql)NoDu^$ z2U+@W1RJXSviYZz<`KA=%i6mWlhfgna|#5bysYDm4ka#j^R4N8N}5dQ+7;^jQwGEO z_x5xzC&XO;nDi7ztJ{|+?wlZ*1()eZ6te^W8m!{TzL%ou8XEw~Qg$a9AoT@#ORn$Qba?ia;nSqZhA+?K{(3n)sV$ z7HZS`cr%sQ-45oxHWv@~h8B694tP7hV`=_8L9e~r_)I7JMn~6`NMv+$vPk6GwiE3& ziYCio8gCpSrv)j&x-%W7w1PsOO_vLPF7*BKwsdi@R{qtRoU=+Oo4+-clmLNCB#E%% z$8fmtcT$=y*^Ktda5%;g`Sy7Pchu&_ii)0`0jIJQE(nOcKPoG{6OJ)r;XiLiuhEeTU?T_1 zh2md%DRs@QH^n-z4Pt-);prlm>!h_q*z(ldFGI*jmsqmyMg>=QaCWTS)jQx_0Pc8m zeoA_Lyg0w>UL{lKt$aHJYA@PIWma~Q#e2-}0YUidjs_J(`lxQQSEAjYQZs(6qN`s# z_Ms^Yrlj>Cd{t=Ih2Kd*v`~b0Wqy=6)uRId!};Q1DS|eK-A>O+`d~ zH1o$_9p#K%hQvHcyg}gwFo~$vTDd|`jA@?{izXp%m%rp^&o>bdoalsaqE$NTnV++9 z6E}Ey6Ok{ncV6EAP*(433CZ9*oKa=b<7PQ$1UGKrvOKjhHl!ND!98AgYijpJLWO>s z8x_0Yr#1*zU7H_|@dYY7><59M+ueT|yW@xOL-bTCh{ ziIHhT?3P*$h-KG9c5wA75~O|>>r_x_l&k)DkILX+!}L_pDxRTH9L}!cv-s!FA-M~0 zL{UFeikGkc?jN6F8^!;h?=KBJo7K>+?w$&l<=AIK_hrN;*GlJwH)uHqWXLa3G7ti= zC#PnbKhfv+eleoIA`4Wv##GSQ6^C+`S8nyI=a@0o?Ey+ztwlbw#)hbTRvJn#n!J{e ze}9;&PJphW7;VjXn6bnW^6Y(N#@UGZ>Q=c{@MxT{kd#{s`|mQ}5R{`r%EYOrNjQQJ z;4~-!+f!Z(FUvzGC``4nTnnT6GhYK3DKqHiS##kdUR?JxdG6Zj-=>k8g05;Up>9j+cGx~- z6^7UJPkAKt66l0E1kS*Q9lbhU-3G7^Rg&Umap8MibxF5Z_~)YW+>LmBQDZWgUq(qF#^`kX8gJZ6H{0|U|S8JpDm5P~YLB&EY6?rdZy-{@?wILTxAL>Wz_YZEb|Tc$?UP zOFdwej*7Z^eD4Yq>}d5zRcD3x;@LC{=yRi<9WG@a1^1?ybbnQ`HfQj?5r`?ZK3|h* zrrCdyH}vJ%D+@3vtk&c7L#KGDt}~a_d`Yt2g1x&5D0Vn;t$$c66bk>_RH#`dTB6)Q zxY8+~^%bo1?`VIxq#<1uysRtcRWnS+@N&w0=r7p{WkXkHvwU3Wi?^oNVP}CdQOKKY z@Eq9wCz8^w;buo- z5$e7@-|d+m!Y4iCfCsBh{{p@quFRNVetcOpfQE8f%qTuPV?NPh+-44dCY1r&vQ~zN z+7Q(o$71^MK*H$~{j4D0N3YNwlg@J}jrW6C*DBjfc!4bu$lYD9lFlrXT;R(1?m43H z7zK%+=po}>a=Cu^FM0-937^IcopJcEKptvpN#ezXQ>`tl4vnZUXuB$8?{!-KY{FJ$ zgOqeQ{lv-O6xA?VE9<#ox%G3xNU39BN3q80r5A`6Yt!=bYRVf2MI}G_ufG&$_{apmI}&-mS}4~ON!dn;xsK^%{HOAr zGa#4NoT!N`|FadcAmA>sKy1A_cuFj^^LfqdD*8F%YjHcdTFd%7@IuQNk|!^MKv3yp%) zi?Nq3^qt}6U!8!#EhX<=*-crIB=P0^^Y70D-}2Y(G^k*7;9(8@T;YR9ARfDY+TGo) zdb)Abdi?Gs)OJ&G*5L&AT2Assz4@~v2F{UQEzN@c**dv&3jR|jE4|Fi{{&1~@3AU> z=hIL9@Tj%Jj$U%THLQ_zu0cI%dE=2Z=h*NpC;l*(mi#Q&ZjPsAkNk9YVKTIs>ubJe z#12-eob5dk-V7XpF~m#+jzt#GQztkOuXR{-=3LLd*xll?QDjHWjIJ8cX6E}WkRjX`@s+bt)eUU~PA>xhR0iEM3PrS-s&oFp_ zrCRF~;Z_2h>T7S;6<>HGA4>v8xtWERVmA`uW@XBSP_@hfcGzk2RRpfTtmn0t{QWan zBW{wE+sN;}Zua~WGDa~xe-#e4X2Rdc8p=GEBt6<%$K}I6pu2{$Y2uLH4z3@|;S9ku zV1)vt8z1pt3=&PBb$%;&{B+Uh>7fO$GBd?TFQf8*NcD`#E&a}!w?-TD?wWi8QR@p+ z`4I*MPOj|}vr8?5P^XhbSlH5z@LR?oFKeT@oJxpZUYyVEKt7$2D|e@>k~J{s+YD24 z7Gw7>0ebqdVyz0rvfKI`cpRkhUlJ^;@;?b*8>K$K)Fn@bVHj)>ci3dgwRI6njpDWM zkXF5w9#`Kqjdz}$9;YJH4a|YQMdPzMG(ZnPoGjDRs(X0LISm-tV=7&N6TR3)v|oZS zYR__=qrQ;qMXqH0{LfdT-AJx#NKHVeid6!Y&^#N0xNisbF4vz-s3{dq1%4wy-5&Eo zxW3l&mJ7dIu^BZ?GNh=b{xFnQl*hYF7=X{DI|2lPmY}IGfkcU&sUzU`)%(_9)_JSa zGuF8)WF^k5Z8Y47fDl5&L#$L?t$v2kR#K1Ox$@%QuqSH;8ytvTEf;nXL^z>?sodj< z6jF3jx```_6l4KGaVpCcBk52{dVV14o5bQTHApp_PCIx!BHQ7&BMp)wu(z~2&*v(1 z4fdo*(++SHcc<{7r)WBG`u!utefyTlX|NpW2*DJc zn}k3gtIuo=Dv4O7c8pv^L%?E~3zC9SKw+Q!`2hb`GfcLHFJr^yfGcf8$P~yj5d5+j z^W@;B(&)k-z4Wf%UuPEM6}GjV4`?=~8LGQq^A!AhqO4%S-bK^K>j)kNG{eBWy$oq> z<9CA1lUyYja*Rvq04egNVD-pO7fOkpE~!2Hi7U<9KsMXQ?h`6*uz$iZv(as4Qb#)9 zm{CQGXu*e}EMX(B1SpqnamJ7m+5uSu*`*l>W;EsIf3Y!2{CZNUf|$4>%C5#PF;-^# zLH0WY_d)ncn)rZAi3;;?aHMKsMH@ZW7;9{sb=73fF()(ETb#Sxon-Kax0xFzHB*FI zjFJzcRf=6U)#mNgn}s(;Q5_N65y`$KTb^k3LH5Nm(9crbL2s*z6l&7xN1uLbHndQy zX?xP((D>;Cm|>y-0Q+u^NJ78jz)6o1xV+#=tnegX9-~B(`!PuE-hKrH#I^5Si0qDMip&y4vVXG5HHC;< zXi*zZ(v%%}Y%f>!7WfnFmwwp+IIJ;&OtAlGgR z`FJ{(X#pO+uu9LHqZI!Ak2>J5H>ZO5_spewPwTZr9nvW7D}s_6$|T@c`I4{{hR2|; zMeE?;Fp;nD$`b@J8b+~AuKU?IQW%ADOC{>zOq0tJ;#0Z!#5X0I?Oe+tM|q$0#DjYd zR~NW-^&H<*%|5>N&=z#??a`^d0B0oN+zH zLagO?E@3hx=s2t#@Zxg7?sRAg2Ij=4rIkKE|MJQ%3vkZ*kA6zxx?f2SwZ`sWXjhLj zx{N$WhsA%C{!~XtmMB{Mt?-feulsY_Vxwh$NSSJ$|8Ds*pU9oN%r5TA@tL3D>1!O__xyJ_;!7iYlnCLzG9>zs3*M}9Y?-1+ z`uT&M6?Ein=$72IangzX^(_-Zr5I~@4tb8EcJZh`L_`>M)WQv2zIWDsbTH(uQSV9~NttGF4&|l_55{wT-g~}w;T%TDynER8m=9e4Mfc^p6@-)9FSLqHLM&g)u?<=m zZ%Ru`XMxB=O^qt+sW;sFq9ZV?M)v$}o(E@yf;6#oS=%sR{kwQYX_{6R<2xMkhJaw# zIjUoAq$OaUv|4O#1fF|aZS8MwzRdq7V%nD_=l-c`XxEbf)&BzS;@hKYc;9Az-hZrtB!ZR7*qh5PE3iq$f<1G_88t3Rs!0$+}nK%+lt@CF7=SHW0 z1M*hQz1_o;t)Ncq)*Dl!3})^u+){cvMG3EiiOY>@Z3tEFWLd^7alh)_b=lpXTgvI$ zX6=@+uz7K(l>Nvk(>ZJR_~qnL%r9J~xD%{FGX(Gqdz; zF=(u3s zQ3aFiTHL+8_*vP4+{IqDDpwq^mgKDP<6k(diJDL;pTWM~d?ga^Ozh{G<2&|O2lj;A zmvqIG^NX1l_q&9u%Hph4By?wB*p@2x=1_VLm|QaQ85L*EMv)^xmCD~=NYwGh6sLW08PPl) zi(Y1G4*URk8AJ)2-eBK;X$SR;)BnjKa6Z~1X;i(` zxElR=+#15ke0%InAfKQtcjnGPCjtark*>c$Ufs`S&BqRsmoWdY#sXARF!vh74jPJrvUkO9-DNte;exMn**BmFYED z^(#3L%|!{_guuK_QAX{(wjO%AnzCz~e7+;&yAApocv+sGd7@CieezF$aP~@|D;E*% zXceh;k$nY5sd3N~kbU3i4W#ClKv1j>u(81|{kw4bImQRTBy^moX<)~!skJfeo{`9M zCv9!=@ReYz>+&IY)`6*3Af|U7ywoe>u9NBS2xy zmH8uo?YDC2x6M**nc#$^JB5U#{G8j1??f`Z^wYE@Q2xZ7_For`QpFaq_ULu@Oc;|asj-HiNRV3`pc*;X7ORQcw-`?rx^r+f_DYmEuFEa0lnS1NB zYeH6s>Z>8~WkzW+DyU~=)8p)@sdMQ&3EPMV4=wp zc=qZgTHgDC#3Cg%6B@yy)jU}+T}{+*(b`S5@&c;A&Yah_~WC!->S{{MUI8bH7nnmVBn z4XmqeLY;9doX>CCcUSYcn#A_a^nT(^SFd0j)kJ`mW1p--HHE!DR)75M0~cMr9TR`a z4=`7L1b9p=_0lerMT&4ReQ7zW&+)Qsce=JO-iq7XF15BC91o@Qm&mQTU=Wg?f-y~N z((q*Q-7h5Hwo;MCGyH`1V|_H}_+|FZe;KuxCHU@>zFzx!F~k{6T!^p0u|Q&RZ#fR) zx)VpYoH#*>IkRZ;>!DxpM0TslOWR{?GAW2UtQ!~~h`ud!OFK=&i8gw=L5rXDXmCV9 z9MbRxm#0*F;i2+F#8kAt)W90Tf8q&N_)t*Rrh1V!LF1n%DB2m1@;gJeaK%gZ3L|0e ztX9yz5V_}4H2KxS52WjxTder(V=u+gm}kMX{072XB9YkJyvrjx{mnL2TU*|jF1Rl~ z3^!a-+!H^qi&dhsny9OSikDeU7IT8}P@emB(_28@pqdzTJUvIoBcyDFAX@5fI*wEEoEs!iJc_4q5p ziA|)cV=bqMfPfU2nb}ejKv<`*x2;y9$<)ZJ*d!xTzwsGy3;G1vMURJ)PGX%88pIp- z?+qpZKW#{+m(%^dW_E%4t=IQ6I94#0C=AF0>d1^VW zq_5IeGSvrBhLbDhXM4HVaFtTKviJg;QC4>9gLV5#L$> zK7Z1D8W_XGe+FX|0A|(K&k`BxL2>d~6DT5mNzu_+SL^*b0rt&B&1bX%368hQM$e5| z*sJN&l@#iBp9I>sM^WKkftl49Wh#vP0C_Yq>=;4Fwb$?Eg#-x$1z01!VpQ}JP^2wd z=zbB@i-z>kQR<@+KDi-ls1Gc3F@6m>3CGd$i3q1X%K9r0&j-=~Phgxu5n3{P$@7tP zIpe_A5g0zpa-h?u9Kg*L?YA)CI`k+hPxV^!)^RR8=MN7-`HQTzo=$8BJ zm7?56n}WF#aV5Qd6O4y7{wDl#ffQT~Slcb0pC2|W9Y3<7sy%6pd{wZL{h$0mIRI2w zY9R(T=3-s!1u#nIGx0#j(b3UbFl`e0aI-rp!$G~MY;~uka^-V1E(+cKvb#O*U_v*c z5I#cc=cfU`V?%yWnz`u8C?sUKtLi?8a&yr*lX8NCTs>ZGaWHN)=G3N;UtAUAY_^{(1If0Fao>y1}j?AxGV zX<+$ffHeS{>?2?psy&miZ1h~<<8#`MwM&&rV$xi9gujBv)BO4g=^6RmOC2x?Ig3ts z-N75E60DSIj$>ss-}kTc48rn1OoTO$xY9qkJ?Hh++3Mx?di8HVA#sA3%oB{*kSQGj zi7UDf;1$A7QQ_?r7E1&XH1rD1QXvV+>3q}*#w=bE`K(_YQt?Zq!aHmhoGv>;pSzcc zrTPfaHa!OCF0xS7+H&Y)*f_6$PM-ZFP_t-T5w3L<$2AC9%DE}}0s&zAdQO1UI$zhS z;?wM4KyTn_kJgG-T8(e)${y6KB`AU@9bW8PRIE5nHaPOM9T6!1Pbcj;2y!RLQo@|z zroM>xjyxd z?)lvLY;8$~OBB8!35?&7fK)i%M( zF7f0 zkYLNSl?>id3O>wAnQy^)TtO9mF&Wn><@@tqziJkU!dcuiXEi%bq+v6FQ z8D0NSRa`6XOuqnwl0vwzB{I)48+<7W)@jI@^;^7P=Y7#nxZz57ACMU2ma5k7{ba@6 z$v$jiOk)T?Tv3=Yl>e%n%?E=K-*1`KcQv-D>R*B3GGYWN%Js5e<0kQ z_X7|hodJ`i8tB!_FxGhka^1knnVD-OAP0@Yrkc43Au!Y!Xwu&$QDcZG-t| z40Kk>X?sum2bh}Qp#14kWk>kTPgI-8QWo3ZDsY(?6?)A)&(j_O-ql@Ji@Ss%#Q=as z%YvSN9++AX2An<~Tt8NTIm;+tERdh`SBN{neW4uG!CJAM=nJr6XqLiPrm&+qUuLK~d{-j9+G$UVmOVZ4yb*VvR^l3C?qkHmR7H7)`sUAMgJ> z7Cf;?GdMuNJof4S`f`0SYV4rhx}rrAg%xI-bJ5ubfn5=jYwR!s1G8CBkbbwyY|n5u zYrw!BHdWsrSi)HTLx)L_u;uGkbS|2?4ubBG1O$ucpPj8vFr&8Y52XqByjfEFo9 z*eK{dj02~zB#J0Xhk(ywvj;4sa#;#_Jem^%V>0m)hxqyjK;sEAL$pWm7=4JQ|a*}SU&HSr$t&Gd8=%DK4AlR8H9aJioQJ8IJ6A#Via zhrYA{uufES=+J9{99qc|mBFdHw3`{emRrj7d0+i!@e|o@)+btt%>%IsDH5X!?+cxh zh9=5w`CY1meHR=Dn5@KY8yR6GWLhssI=>q*g<*gH)9z4E&go1Toj~TupH2HZ?dUf6 zwh>c+$QM-d)_Ms_XnXX)InW&J@7}RIB-lAgK)J{vJMB3peasP&BwH!1$74=dx9!Q{ z(t7NQhQjzVGmEW`aO@LK!T6@VaDE9w#Mg3(QLdGVp$vXYo&C?7$goA$^T5Z|Kc&bU zuieW=x4f#x>sP{6l9S5;V+)FgxwQk{i3c#i{rXUeNBd^FR7b2HZ3NKYbTkm=5~t@f zxvcT;LGcL;EGBOGsp+B2yxRx5;b+ynQLvT#n}8sn2M4MOcpU)lX&~x=YQFYyBeFq5 z`A85%dZxQ^&)(95F=tZldPo7;unTa9geX}lLty5BY&4F~(BnK$arqftB71IrJ~pH1 zJ2J}nHZn4@;x24&F)lUSehR3OQUh(`_s$!fK_adAK~Kk(W*6U^Ix2mO@uz_@$ccWr zn$T=X@1Ad!pX+>kz9}|Bm&!Xk>@Xbjm5g-vH>oE}ifwwJ*k!@2xGY zhQLtpmmT)u!NV`FYH3>GPcqHqeJM0jC#E0@J`^j}4RmA>da%~LpPN>%QB?E1zE4o+ zD*h1q^!sHQhRPiCldYPD*Lmr${3d75PBC(WGt-NhehN8Qa7buYh&|isE-cBf4NRlO zhv$V$%Hd^KRL-Im3~IkZe5+@Rn2Gaa;krlo;hk46=Isz}Lw(N?Q7|6Ka?G+b?+o^@ zfl-kT{3_BcN@iKBsfkCsNdR~%EcQbB>@jx zxd9j<+KO@SLFk)9$JeD}=yR^Yg&_xr&!G8L`0iO!+?ht=@6x*Te=&hyHk|z25AK8l$P!W=`KZ5I;5nfyZc>7@8AD@zML@}Ltvl1 z*IsMRYhIzNxMQBs#^O(z3dTa3-p#@wYkK_+ufuBK^aN;~IM7OeQl0ON3vm55&%gWL zrE$Jvv9Tdu+MIZ&*JEg&XdDT_v>5>M1KZIo*!|g>V558w(-xEGYSIxVk0k;c1B$gN zUN_|TXFF~hKdFftT5(NxpU|_negn9uFoV39*?k^mX@#?aM)K-LX8E$4yka7Qc23ey z59o8Dg@pw9_xX>6?UZQaW&yBAH6iolY9w9ojg!!IVVe4$b6qm;+1(e}?H;$sJl$Xc zSuql?gx_g(!E-J|MfGSKW3g>k@g_%w79fIdXXmSNX8a2!WB{Py;T8(K)aIe{@|#N{ z=X;0O_WF0KrlQ9WLKY)G{mZKEuK?)MxMsN#iJ_qpvZ^Z;dqTY=d2dK8Cy=>Y`Ec&> zHYRoE-2XsFK$DRhTvjP#<_sM@a?Q0}Qtw~N&5%?lJIW|9<|>v^xWfPjo{}vhViv)| zFG4IsgXQ9}tL>Hqr+9HhvcHO2y#X9eNJZ_*%gx%wpfKJn^o^j=ng@Qh|RZcM9G&VM# ztf0=6%UKn7Y9M;-s=KwRnY z)1&#+;&!M2d>LEPOX;;!W*arxN+wvNEZvHvzS>xo`qs_OsGsUAEa*5isTLu0)@0E~ z=+>HNGbKqa%g}sz%W$@lWH^4b@+~qV@pc##TJRJ`gz#Gz`}0$>Zn?(kMbuSCfqk(% zRqu(%Ys?f|NHA!&>_7I#iPDJtcYF7NeV0p1o#Td#dUnJ|r4!^P``PL=IHAPgc*-}_ zBZgT7lqVZung!!mU+fCnZRSVceMA`l(Xf<*=dv4C@@gi*+sqRFX{vAxO`m#?l=A9B z4RLf1DRU<0kfE#-wU4xrf_hq#ndxplE`g=8xgK}gn`@pR4YFh}&l{sWPx9gSaB^94 zNS)E?7#*a(vCbB|%n8AKXu8i?jRS2Gn-6IW8(7RhkF2xlzss-%G0SAn(VA48$6S zDFwn>2W_&=;RU>hiuqcL+(^QTk}9n3t%`=KNupyqCi;=L_B{wonMzl40|2nDD@!f6 zw6d5F%*0Y8ZFb0p?ah}>D$9tr863EEAO6&hb4kyy8ci&qBRKSzVCFZ2%@T>O#Goz( zlSdauuAm5#Y^JQQxzl9C0v**BV;@9dQG!RHb1i5Nl)$PqQL%KTEN?2t)$MJm-R0=p zyzHvLCkIXg(M(j7e7N*uv4^dNbi&F7x)W6F@AlrYZ~SgiAJo>Gnj4svrPWEK5PGcx z`-Jl78Z&JDYQM19)$_8PrOWyE_gUxl!;f=)Ux(1_KG>9V@-Dp%UG_5Vr7dsG9r>!^ zhYj6SC}FcV;CYbyB9`&cUY2wL6oxT52HplH0&vy#Lh?%nQ;<~?323dJV ziY#dKF=FRjeI~7r#PT`w()Q$dU5!xT5rS0Oe(c95{l2a>Q_aSNt@xQG@=<1g0r%_M zW1ik$^gOW3xm8KphXF&u(j+j~TAQ|n zNDYhJ5b~y*RdJ`!CXAg}<`_Xy@yPVW=eKPkz8^NiKhLzd*-bG=*&IC;4@|lZ-k{i0 ztnWUN?csGEPpYAjsOFrrKf+i%UgFw5SH3J=ELQ9P-RSP-#b2lC74k>pS>cWhOMW$h zp~J669sCE?ptL#1h`EJ%k-g^Ho>zyXJKc!*O>hh{V&r1#rNgy(^lH+`6_an@nC%>! zlW?5S4rDBN>6HfL)0)05K^u@3eYWL?|8jL@R2sLECy;B?@o{mVU}u%jDsPujXvnCz zKKAC|;M}2sAC6u4?&E#!VZFK7pHIQwx|C*u8{s*Y7<>X1Ra76D_dZiAme{^NnlI2) zuqIl9xWweb?EQG;dNB{4y}%5>Uk+bAx&V3B3n*FeK32gZjQnvEtZ ztPrAPgtv}c^uX-vX$Cn7iDO|Y2E^KVCd&m+BV+GN5dR?&WFHeDH zuGs5{r^~HHc7(6!(q?V%AeMK^rN;WO+%bDm*QapgH-Xr;CR6cj-EKpH?!mRtz#IC7 z!#oH=&pEACdryjjuv73&r1|{{4)cN6j7JQtwE++(Wg3gfop;p{WlnB-mi--&Tm`%*rH6@WH z7TiinAB!xxCj*DjPLYpitC=>^Jy}pb_PskA&Ty=cG*r{y#tL)unv9eoSA zp+o1LqHR@4Ce#}qGb3kUi1>~wEM;;wY?jEMyGS7uz_$cGjAv_M!#M=}&pGc#->FAd z8PhApuO+|8)33cMOxmS-$tx@W0jE9rr>5R6l7GPPm>bEUWrOu>P6pxtf7#u!YnL9b z%qVouK=*|XGYFQ?<}kJ*>YH-gVAAkJZ)Idw8bf$#+!YWs%_~wX%QS~S{ z5^oHw5yO`L0dj5e7fg3eM57&JSi?jlz!%pK-V1?OQwYnAi$6PcBA*-AZE z67`SbL=*;ytfo@B5vc8)E1%HtvH)F*`Fwk#x8e%S66#uusHTX9slS`G>4^38Jcpr^ zTzCJ$M!|>n2KiOQcd4^F?h!&tzlbkJa0b$yzvV{gTedw4F~*2|ldsx9I8bV(;=)~! z3v58|7BVpb+3ajkJApuP?o36)2ATaN$f82_5n@Gt1iBjO2>ml$!D7L?rN>GA<8QQ4 z{kmqo;yiL?=XcW~Ymy5FZi_BD4I|ck<@x&ZfxT*DjD!S)Z@N~x)aqP0;3lJP5bbhx_4%C*v{OBXOh^n4YDY|DV+>4Kk$%y4#W!!1^!stugQNd)c_S1$C-JgC9DEiS)fQpr0!N1a z0a{Kf*<>~2&}1e1IRm*^W~#`H$*(r5dyfDDauwQI{WV#v3T4&>nrxcTR-HJ%+CtR= zCI;_Ho*V!f-_HYiG6vJ$i08RF&q9z}Y&^9J0z-3V?}j?7q1jkF_SkM2`B84JUx}K( zd28V|5RN8Jdo{(W4CCCC;0AdLEM#_9YIB6}e--E`9koNhaCoSGXbtg+pTSd@?{UMy zt2(kV$LSJ~&^mP=H;cfvA=plR*Wa!c_UZOxE-7Bd5sM=cgDRHtvRKKdUu|C9^PXQ% zU@jE}9t}Z_5dS&0T(H$B6wfQ;xkZzkg}x9!H!6Kdax#G!Yu9}?_a)Mh=JMM7n@c1w z^yP2Q0!Es0lP2=dJ!nTI-qe1NGtxd;KN(}){j zP?^Bv0!uVom)4B_;nzk5$|XgS(e0`sJC;%)#~y%I3M$~RHU=HT51R@DmXN%9t%a~N z>NhRhPB&3CnPwAv?}QJ+uUhAux3JEmXi)g{YA|U32zx|GofX#BgH)p`r)YntHZWs2 z157|ny-pD$KgX-1w;Yj4#dWEZRNt}ugjpH>JWm!vfY%Nvp0*~IEiOdW=MhGx(+DfV zrN0sy>cGOrR!3Q;LWJi;NnKr_hosN79QgWc91sfRp$wT_=g7E zg}&t_?7{Fe@%3<$xADh$Q3ms(#U*ZX3uUeQuz2XIS6_qJEyl$o@2IJ$s#H2poK0?DI-6c=&Ma)?s% z!S91f?{)cRbxc-4pIh2<0xBUOVrt7KK5#))kC5}yY~87X2wRK8~@UaR-6 zTVyUiB`7DlhFzFhdZ5<+^Qv&ep&NY*KvZg0c~E%+5Pm`_vUj8uB5-_CcX`GI(qYGh z7DJAP5|8puB4Gqb3v^^9r$Ud{iZ0BtoN<7O{M2a5&!p{X91&1ViU>UuWRW z&#asD7b+dp2o?j1YD21j$VHV5O(VR9KAUV5(uC*NHWqf)MKuh9xTp#`;jt0@7x+#2 zFfJz6HZ9Ly2|Wq}@KXBgM}18*nF=EKZ5RP z_#Mlg1<#^M6)e76EhXI_LU{r2L>p`U$~NL>NAmu>?_j<*JE=+Bb8sf6R!? zB`V^OcRd;q;1KxdR?)>sWco4k-K+eLwaRP%fXI3vAwh2RSZi`LwuZO@{C(7x^u);E zCCmX?z(quDj{~Z|$Vr76iDVkDZ zZWF!MGtawa*3w(%oKBDjQEthuMR9fF zk>pPho{lzuFzQ8N)K>|u;98U>Vd9p5yc;3E=?yVc>!H@*>)&K7c1md@;#Y;T=s7g} znPtU4N|C<0K1D+iO$i;@m;Dy3{CKfnRCki%mF32)b5WT@qxaSP(-l1elBhV*SG!M) zaMt|}2;Y;T82!}G&|kgbieyteQsM7o=ledymekSw0{#zfCF*Q?n5`!R^v^!h;Lg$j zc0IJKVuhWc5nd^LB7^|qTGf*tMXt30#8|i`0cqP9Oqx&l$c%fX@4x&r4?}S$q`8%Y z3)UmR(N@u`_?d<+$>5SUls!cSn3jYXJ1Ou*xv^i4-;R!H0UlU#y5Alb4s3{ z`c2MjgPlh(=&&Rui$x5VfuM85ogFOd?jFPnSOL~ORX3ex{<9hMg0foFx!12 zH4X3-xZkmo1w1(guGc?ngz>YwuB;K=F5)yWg&X*UPPZNNwsCyGQI8J6=%G<^BzpH_ ziC@e35lF4s9ahDsI=BKdpiw~3xn(3Q78u-*tM97kT5oIw(_?n7e4u9Mm2*KxEY)ko z^jjM2|4cmrGSDLaXo0U}1Xn-~ZN={T_kOqOo{vG3(wcFMo(Pd^wFi(y0#rP zI3sJ(3JYk#8CL3hpP(~5btk{X@ku`R?156wd60ztLruDn%;UU!IN;5vO=Nklo=;(G z`FE2%fQUJBu$4j{GtmgyN9lL!JhVPBG3c#D+h`JOB*++|qMU4>jS*SSS81(p9OegT z+o)-nBD|4Nm}mP$li0`+e<-;~9${d;``m&gbHCyD2U+i7{SF@i0T9gNy4YKSj`RnI zjpRPWV23-F%FQ!M!`s$w{hcn)t>ub$59cFrJX1R4s=o%aE9g^y4Xb(u)1xb@4Y`-R zAV!R3k`8SpUU<=Ig}*q5%VPw{U^eLPB1AEc0-*znHLMuitPBQ!DVVybC%h8{JuhlE zER~#%XurHDo$r@?YnD_*20mIfcSC9cxr`o8WN$M@|I|Nw z)-U|Ek~%q)vnFhE%FNa%?#HE>ML1Ag*F z1BzOoR&8{@aN4Y|TF!mub&a(DM2lQ&J*)CU^Ca{YqkdfI`Wk-ipY_xKc=jw1@ktHm zU)b<8J3XAmY@es<-DO0V$BhI@oY38sUshCAY7x> zT3*mq`CgXSRph6b;6ngKK(X|E0YsAr(6N66ue^s;cRZYiKWNJAXUX{(OTbg&h?oW$~cjGBz@bvKc0eK=GObVpGmr zq`omi)z#0k`Af8B46$q;*Pg|seFZHtmCdf{K|FE0<6m%5ikktTSx4cFo(EmnyW^no> z)zwbq#h>!J*Hqo8pA`kK#p%C6&(~x>ZKPyTSREDoCxR14n^Ep6z3CTCWnYo{uvdY? zQa1n8vBhO~r=yNB0oy4Ttk`&Nf8f4cF8@f1j+-TtxGe4Hsq^ixexeHUp6mGuY1JOC zh{AGeF-`0yl|%x^&M_#Y%Xck`6BAzRm?FO#XVRTZ7}122cf#g`yf-NTS5GS`64vsX zta@B+J;j92YMSB0*_f2HeKDA~0;jSFFftMC?+-%2w>Qal>nK_(&Q^U&rY@%TjpVAX zXY;0h-ZmF_wLKrkubL1!5rKZEP7b&-n2l&SY#I@`s+V|i8GCn;r8JudoMM)mf|)J* z_8#DW4P*)ztS-rgWCo)sb&w7!HZ5eBCpCa?NRTA__;9bTDPK?FS4u@gmj%HMBe}}M z(RiD~&(K8tVT~I2z8U5!wNDv7lqqpoilsIbN_4W@m@x?{ z!wkYt=z-%0RBtk%DB=xR8u2kP!p6j~@ZOKXh?wx2Ic&FzstKvFlRoy;Jspi58j_O< z4ctuPwUiegeE%F1?@jnM$6Vfy|#n(zf6=Dl>fl zw><+)8XX=u3tE^zR4KyGv0_M0yq@siN?)MXPJW)J|MvE%7~CNm2{=GF7mR$#69u|TSLjmkB>MaH_g@+NXjcZ{MrIK|!lQ}zY$9inS zc0r@9Gz@tkL`kg2kJ*&5kU?$ZKm>kU>t~q6g4O6CwV)9|_}vC$HEM==Z(!CwnKTj* znd!T&5S9bj{?5qM0o=H9Z z5Z3ek`6NJ<`_A+~JvuyiL0n+5@ZZAKLyi)k500EcWR(c(EafcV^Vgo7b+4SpyYPI* z>Sh%5IlJllI~5axz4n*a0QyWP=;i)LV@&o_Mp^*n8OZh$vRF)q);hO)HFG{j_9oDc zKWmw#ya^l><`xQ#Qbs@^;|%;EfsbtAWSD-lb2h*-{C6gcb2(n=GTvNkG1%3?&0w{U z|0SbS92&N|E4v}{$LC~Swo#Q60;rb8)`fRFDVPG6y)V&4tGh{HTV}sVp``+ zVuoI`b7<5aj~sq<#tfg{UaL%d!X)Pq?-)HNsV#V~L;T9B&W=y_sI?8+shESgNzN9lAtX=!P7Qn4$r8@;YyqY?9Ksruyn z0n%ynti+Q5MFHZh4-P}n)TI#OihfMJC4tx z`;y-nlKVxiNv*P@Z;CK7c4wd!V2-+$i_)S<-66e+iHUi0t8q|Iue$KG!^#IvVsd7+ zzV{r8Dha`9pnJ4u+`rO1`O5#Y8m{)Pf-*A&P*ujMtDH6!7C-0Z=UcR6h|x>!B+}ZN zhD_Q5)v3dm(wKAK+kGw>8JX%Aw_tvb+rW@`{WvbLONrzSMqaV&_3g2eURZ_jB+Tzg*9te-FTGwUIRw!)qVt`8k=j@p9?OrmZm{ilLS*7#;`M;Uwh+XW zsDM|o7=-+ubs>_(RDJr^^xiO?(!+MPfo7zGpF$V3yTB?ndQXGD^M;L;RoM#XATMxz zeIpe3TS%eYyqks*1U%LERw>AjE7;TuY97mtN)HCo&ihS9mn3{N6jQ%-oWYs!{Rgfq z3EME})H=l>L)Rrk;`kxK#xLQL{e)Ap8z6O#wLL7 zbk0*%lQP^u6y1jp&0iQb7FO+1gVa?;L`3KyWvSKr_uvH+RxMP>2iPCLGQv)ARh|@%jee?{jx<|&xwp}W{`&$%7lR)fW>JCtlA@>Q6-QO`z zf6R#BqhL%h@F{!YeYJz6b2O!PW3mn_C z5LrUkpL1tgyx}L6d@4qnGqgDd^)DsR`qvm}6KghM5D5A3i(sPAGZ+dPT256m3fxRV zPEe1SzdrN{3LhsX{@3*+T?0Sugx&uA=Se7+|F|Q>u(S8UFyGQCK*j(X0uIa>56Acx7SE>ou1e~fSO*eu`-1gQZh z*mmQ8Ap#~?kA4G#`gcNR!;&;JpXZ%4f}v$QbVDFz8C@l&VV?PIjByY;oGGACbugs) z$8uIXYE$vHzI;-m#;(_r(NHimu^&0I6TI|q@R{^RFC>gXJo4G{NA!|UaO8^CnICp& zWAo_L8Jsp{pGPgC<>lcUNNK@9MUT!P78N4^VLM%y{h90pD3!3`ETL&R@wBwc63^}C zL|LCEp;wRC-%W#@Jy|Du41D}T7oPJyO5pCpYShhmwuQeo1q0P9Y0kRr&rNCL3jX-J zr|`pGz*=6(_B}bGf;)+$>3~_n+MgSf58ghN5-eCK>!#FlZ+QTg_G5r0qMft3@&j6B z@qz!^)@rq6<2I7<#X>ugmidKq0}<1|o|?`-Bm}8^cUv=#mVCTA!mam*$=R$gNx?em zvE_3T62qREw}Kqzb19MvS0qyvcKZcDoUdNa$b(6MBiA&8_q%WN^Pqim49uO+4l>Rd zqfTuW2$UZMQY}T#%L7V=7HCXpvG3!9PC`);4P&}<&#`48utSA$Is~t^h0kjsF_i&< zY-l!s@W&xDN~nSb|oV{`H3ogA23D1$UF?-eiL!&vtDsbAySS`o< z+F_s}^+c8ZDdLv<8lxg%4=Yr0uzU|$@2lW>exV^-5kbl|p`vIz@}E?Z$_N4O4Vs=qrC|ta$&eZ?ch(u-wHA1rJ&qFk>iK z3>mA$O_^+k*qF^muiu3>|18sOP~gghZk05Ka3NUyAj$Frh+KuD>#pefTSSRpyV83S z=rw6J(%O3>_Gz9yv}J1e*2P^^&WRm8rlgyCR3+UP%ucWU`*YfWRAmpq+q$S3;4FXLMMtH#j|mZ#os>_xVTHgA%3f!$e1}&^7^m{0{r5I zfhT+pF51=B9f*)DJ9)X!Zo80-fJP{;p1?~R)>61O;CO?L&pa;OBZ}$!hT?E9A4kSz z8QpiLLSN(O8rL<*EYR>U?yWED;~Kz|Fp;1Lt>K+1 zY-W=imRd4Yd4Lm-_OQ8UJESFAHY0li@=h67X|Gio z+jnzTOH^+Si0T}xm)IFULm4P6u-MpFU|6>eO7(n@7J! zKQ{S?F;c~VWk@e5rivuHnyq(pb01u1%&4ow{AK2uy(q=eoKE`kg<>KD^Adcp1Zx>9D3IJWZAy>-(E?36;gw8rVS@y{mcs(S#>eeSp)10}Sj*zVFaZg9TO$*hUnQF~4w|(?#wZRCF4DDOK*= z@EAhl`|XIhcE@R#Qu}Dc>9FmxdH!V*5Upm#BX>5Wl-+&bKEmLB_v7h@)lllmjSoL2Tabn5LSIJ!%8 z8DHR>|JS6_!;0gPfaqFIen_qMY40WbU1&5d3r*mh^MPXcW?M1Qj4ZpwqCT@O8ZR~; zaWV1W;9&OWFk@V-moMd@yvB*U5gu5`6s>|$I$-5lP-B!*#=r)Z9A&qGB4cNwFnWp( zwd_;ON4o51@xK65M8o&vS7wO6tW*R6nJZcgF3PA^4-5shz!x}CM*pstXmqD4Z~bUj zbUN$PUe0%35d3rGiFo6OnnC?^d-2wbK85-E{0rC+xLkAM ze5pSHZeyhb6HpB+6W!nz2=dFGW)p+y>gy9V5%`hj!ZSe$VC4b}ugRYk2Y-4TM^9AfI?L)%p&7kz?=9FQ%QaMkhUZdPsM$3s#GuwP#n;jma zsOU;`w5yJ;_|GvCn-^)yOYaQX=px6M&c;|!e0z+5Dtf2iwlg=l z)w;ki*KWhM=sBWq$#*y6i6o31CZxFzZKUo}8F#M(xtJvR!c8DVdzJGz{v!A98Ri0aV3$`$4{E7G? z9qoPlDu@Gh;)IA&MZr)df#`^9r-YeO%EkeD(fMLVz|iFk)b~0ZQKS~Y6?9AN*~5=2 z274gl&@eFyf||P~1#+MekE^Pxs&^PLw;l!P=jo~)EOiJ)g2|@0yFYpSUk1JsCX!J* zd=DB!EoQHU_S5 zd0OLxTKoawzNl1nc{LjZ(kblz!|wbrp@ORr&B)umDF}f6fUJ<^`VEg$Xvy?f25FJr znIRpEna71Py}5RKwkhx5O}I2^3GcK~A(*edy}b73R-Zy2UuGwW)E3pM7X+#vmu(Xu zQJeI8md@)k&I>x&C1^z>ty_5!{%BytP}AgwE{1cP!}uEiqb>dxbuONMw!S?p!3`UkfQ5z6AjZ$>Skh5MoC453swR}a;j)>VRCf61cA1<3Q??JppcRt-!I>iHAlRsfhg&00-l+d46f81_ESF;VGRS4+VEa*;+1fIZ+4|9| zrabe0{yiDy=-M29C&3^jf;rCgp{94pJ+ zbYO+$LN+Qc$#}_aFFg1R6*U)FIE)sv71BJs=TD#Yk>UIFJs;v}cHgJi5-)i?uRy-7 z7DpRerKTbN*;OYQ9n^~d&tC^UuO(cy|0my=4%v4AV39j6R8STXfsG^)%7E(oOH}g1 zWL=stpcmU20@169=34D*9FfbyK_cyp%^)R>Wyth8L6(0sZw795LAef$YdYP^?KV6N z#Y6%hYHc|G&(H1Ow*`nKEOO^kBgP}fCjEet?bOhQ^ZOaHfq{V>kjR=U=*@c$MxHg~ zxHkXo{#0<*u$Bd-{qJa9;6P>l`X2CB;?r%ke-LA0ia-giQ%xTO0wz1(0HfAIbwO7q zgVr}83lZfEss(b{0HhhvZnn@$KAG3nX9vCp|997+H-YXYbSSG72*)nOm6_;!Qc9Ey z%Fc=Iz!T@;|9Jrfp;`b8&b62O<6nFL0lN8rQ(tF|fA0T`h<6m~wg`+aoY(s>IZyzK z2!@VTyH6N+L0fN=6wNOlDs|>=otu+m4n%a&9!u%3ir5L^zi0pdKZ;IF*sqXwAQ3+~ zcY6rQO!Nw>o?!g(((wxzIVZUB{1v#e;KFXYNvWbGAbx|$d=hSP-QNP7V|tBuO3A~S zLWVW~FjGEnQ!PVidj0SBE18;7b-{J0#Pje?ExD{N;%KeOizn6EG=}W_Xm9m#4f^ zlT-?(&T#Lh6g3-8gPNE8x;B;E04lsxg$#aaux(YW;Y(!nB;lUoK{ ze0I|z0$$&T)(K_utTZK`0r(~NHMASx`L)gvybWwDW7z0NK<3gYpduH6Hu2f~@{|Al zf`M=tV{~DZlM^D>x99luFX@CfWZ>c9&p<|4HJsX~ZYUQU6zjl>V)gY2>HijJa(E|* zvn?{UE+tK%jx%)UBX3+yC8E7B+t{lEEkwCX1cgjg5|e(qSst6`$HX@l3*9JWqvL z;pgSC!$Uil09*9<1d-6*_4W1F^FpGx|JDZ!tQ{G~^XJmQti!o(na;$pKPT(RsHH;P z4nyp?MR9gx*QAq#6FM7`>$Ha2?FMPGGc!jY1#c}&VqVKrFjz!PUY?jnm{4J`ytl0| zMC(H1sDeztQO;Z`B3^P(gio}Yot^!3)7p25bd@ans0}D2PHc0BqGP#E2CKkK)Y3mI z2I<3APv%R!A8kfAwXwDPhnaP+6DV@bS6|2i?+GWyky(LN#{rKI8^l2 z9oBvnUnW;KORe!8M8TKZ_$uSu!VZ*l)NfwCH+(27%+KF$<~YjJubGK$dcN^r~lNW^Tt4e#nsUY6m8*s zNe>5u(qw%VatFc$WAOTRqj!1WD8sCf<#3W|P8?TV#_(_dFrnT4j z3pc{+X8Cce-TN%CxzKY5-$3LJX$3v@DuTSROSBl4RcC4N3oEfoPjp!<2vBw!92gj< zvI5BC&SpjH?fTAjkK4l3N&N;3>-qUP$D{xpwv`?mQrR}#lP5O8J-56NdpnE=Mu$cE zaRInYhrX!*uXDECk5Yq#Lf5);Uwx2?W;^`JOQy4SZx&>`vej!nOOGN8)T^Y$C40u> zqi4oeJ?ubo86PbRcYGXHl(MM~R!~q-F?za5Kk*EqZwAWFs;5Sx_eE`G4lyxHPdKJt zrJo;+Lr8-HQr{`Fm)VhL-zuU1*_m^B7PzN56FqHG%buor4&ORO#d)Qc!ka5m6(kpo z(}4axA1E>{7kz(7?#N`JCnu1)c!_W zYuk44>JB;9kf|xhJ~}iq7m?cMRmnul$az8hE0Xiy3q{HlSPxURaQVp?r!b0&{x?db zgxja&x}VlQbghQwo5Ni(u3u}8rJc!ycm^tO8up|}`5}S@*(}v*tWG1G-N-oTS^ayn zko~m%@xhaRqmX+BXqps3w>RxQM#ikYBxr}dGLjOgj1<%ps1#n$Xfgkz+*N}05y~7w zd&xa|pe4~4vuSu$@qD|5yo0+EX(p_<53)eEe2v}6XSP!PzwHL>Pxc?7M^Ei|@MYQb zczr04f9TYm$)-Ne%GN%NT*n(ey~N7dN8OB~8yI2Ru?3?Dx%(`}A45=jGx%u$Mg#Np z$$caP*N}~$_3+Kk+c|VnU+Sz)J^QvF+S5Hbda7+i8k3FxO@>1K9OMdy#MA%Pwtvqj zZpzU@MUqrLw+bE8&6}Xf;@(Bq4dz0W|S*ChYYEP;qZM@41bB=L* z4`9KWoe*)L2NI8imX$29su_uU;%-@eEMCJO7I{!NR`?lz#mi+Af$z-nzfsq#J;6_(=j8vJ(tul#Rf~MKI}>f&vd?(~v|?on z=Rp36#}n-h5Q(y#&W6`un%RgJ5)x9y8%17$4q1;t4607h zM;q!Bzra2shvRcQ$g`LzD7H)1-^a_DCl>NquyClo1%VaYbc}0jYc9J!fQq6940jga zL3`PRakU1FzXCZbh*8}l(D>*KJc8+ z)$ph|$fHfnq<&_r6#lJ85*jW9+)4jeP5iHrJ2HlAh5Qa@e&!YlE%S0Tx}!&doQ0-U z#w95jUg8tIZ< z^AVH%rXL+(pTPaD{DXcfi-{nIm1i!^>h#4~tc_ZO1GVK;bNPZ zkO?!g7wL=U0g>z6MTYvhIR6=XUYD8Rm>k$S%k&g z0Lya|UGv2-67%^RLz|jnm(7RCx?mTh&y(vlwZ|homa{+%5{+^7ij=gog2@c}-qk~R zU3SDnLOlQI#StR)U+g_Oa8J8g5{!DyR-yF9ym9|+!^KF(f*3WrjmB<7Yjjda3;ak= zH>kVwfpMb}Ko^SvJg-p)9J#HUm= zD*Pl;*Fb8`;O0|6fY|<=EVs9?x{n^9D7bJ&)1*P~VLUS|g!N-7YDlzih^?6sqk$gq z{u-9+D65Ok!eF43T?TB7JV59$v;s;6Aa|7V_xG=LKQ{ibl^wiwve9h?)58Vl2-0Xo z=turiv;NnHf_Iq~00$prqIRKNF)aV@-^G7Mn>kxkG{k3eizM_Bq@Ju;^19uyn9$-G zz>1KCy>@22zClzTjPk~nFirmEBTRNO(7S6su|tw@6I`SLl9{i6_{KzkB$XoaTqxJ} zEjR4Q0M#G{BMC6AmL%oj(dN@Min$L*-8Vu8j?&Ef&1!Irgbm1zDAxcFF$YCI6B=CN z7r^-=*K+vL(^LGQx`EWKWIV|HZIot?gvAI(WEOeQb5@`qRNa$CE?~P-XFPeqZoaG{ zw7{qa=F1$(#BhH9Scd<5R>?`A!2juKrE8=;?$QtPpLm+QZO0c0VlHd@k+KZFuXs-`VvMv5p;zodNQM z)wEVw$2&_d#;GW^%0+Q2lFhb_MBUo^uy}4@^}f@+4q%1M^(JhBu;S(yLe@7$IMU0sTV`m?^||s0B4gZ3iaP7}OW65oXC(b6^ly zTijmkkAog`0)Mla!|K(?DSm;HypN;l{2s4W5k z_dn{lMF*qVAGdE_aqh>;>eeipxdu=JfChw(UUC13+uR)Of-k} znb1;Arh)Iu=_Yqvo(s>M*W+dfxd6L4X*4Zp{c0sWha-77Lx3L4MJ^mm%c5Z59B@Z< zJivUAqjYmfQ#33ga(?T8bzaM} z4*2heE#3)*Pl0fOW^`%)QETFG`|BNb3;sQhhQZxLCdNf#Tk}N5yG;_0wbDn4flKl7 zRF#<$kJ;_T%k=`>t+RN-M0$Ef-Jct4Z2IkZOe7DiimopOOo!b%eD2p9mVB^pd?OW2 z>Vsh5kVyT&$JHkoSfwRy^msUbKipBDg}wN~<;9Dk|iAiqd%+a{Hg&mc%$Mi+55b4k5S_-XZL zARo2xCx_8;r~e6Hk@KZGQjf1t_uB$l$Sc?VOVQby{&=%y*$be;Vw5Gb$|~`m7RwDO z?h@F@{}4LAn6zcph|A!8`{-j2;hmoG+PS^63_{@ zJxX?VHOF-((%Dm-Sh?w#|4{`PaM9XmvXJ^`z&g{tK#0- zdYw7~cp;9%B5FPf6on=IEI;<xVze7Yub0AtcxRy$lu_;9`7tM1nRvnzEv*O`w^L_N`Wyk*aJfRQOpF$Xh z-F}!-D!^|98OY^DG|F`-d!vbCJgw?O>%9_6xDpszf4b_r`qoGOLRUe0EOf3qelnAz=cxSF9OTG&P^crAmS@i%gvv( zy(shW-R{=bu3y%@b~v~Q1nv7VLbI?_jC+=SJQEK67CE#j?$#^+!&r_A#cV` zn=$B9Q8Vj>jExBx-20?(-MM(|Jy^!=cM#@&DSuE67no3eXmAgWNIB14_pT0Cb8oLy z+nZN>_}wZIwjOmS%#h2mHEvtV)2xqLYkshB(t&Rn(F zBKLQV|6Ay7V4-&i1-*oUYo|Q|?am;`F_v2y&qIw3o6;569d+Jj8_4>4nrDB^SNF*| zUw#*_uS96Akn_*a7hLamuG+`TrHjsNH03_j_fg1h`tH-{>XeqL6I3(7bfp6NOEdMIQF=qo|kJz{@hoKX>3Fpys7uH}ush;?SQ?#}2qT8!iHH0W*Sb=dBH z{#>Xm@}*F<=F*1*Q*Ry>pIh5BFyG-T@_5)ozBanTrn-7-*J!#%j1|Wd?9l=^K^|ZP zvM!NGlD|94PMfzuI@kOtmO9{G?q`hn^dFP3_D!asiM48 zhBKZy4TbLqPE+mQyZbz!?P!1MQ&94BM?+iT6OprZ!Riyn^-#gjuYfh{y^&>%D}Y>1 zK*oahOw5ZYFaOuE<2&I_cUQ-8999}mZx9vJ6J1GsUQImS=vGKMweqUm4s|?6iLY2_ z+>e;GnjXlqGadMJk2qA|oK*_gY^k=o zt&5zy6#sYst}>weEFX0|5^M(^x5Z+a>)~(x;YUr>tqPS(*(qKkV=uWmcJIaNOo|O4 zAmKN#^k+(S@>IgzBwDDp^k%&FPJO4LQ%YrdI!r-ySSQpCwHiOpKn8~;i9f0a_O7OZ zhXlyNA%jO*tnf${d{yea#iwn_e^>jC8XRQts=bF0s(tvG?#jMDb#AH5^QZ4l2wmNy zy|{Be>NR;4rd<9e+AYhnG)tNaPbel^MD%5DI!R)`kFlMrl?FUbYJhdSJtbJL4#@1F z)hJ6Y4$^J3f9v!LdJGo(K2Vnb@Fle>57y0yo6cxkQTsTpD(LypL*HQKF?Tqkl?ku7 z8=Nj+W_Ehx;--~ce9mM3$E|uX6~ye+AYb`h%$liyLbqpjmRDXd=!1+>2u zdCG7-mey@sc*y-R{0dig5a)y;o`m0U1LwUpwn?O2XDEsw0qigR{bu06PbUEG>;N!n zGpet%*~M$R{rzd&=~r})h}330or}(91JgwDc%G97$pn|xLx#;c>`RaD$MRO2gF&Z- z`kbE`HOaO)&4G!qVK?jl$KG3pMfrB`!hnc?QY!MJK@g;+r9rxt?(QL^r9o6gq`Q&s zoRKa?>5icXrCVtR2KF^Rdq3~{fBwFo-eZ5+$Ns``=rG(f_kCU0TI*crd9EOk{})be z0iQ})hy;$tn9VOl&~>XlVW9fA54oz}ylF(3)B{Boar6h_>W!O}e6kIT|1eHHX- zBPLIJKBBkik_Z-13MGIhLk~DFRzFPUnt?5>QDq`Fim9&%ws0^<%vjbEDLVSeN#43O zy0i{tqysk4UO)M=2gy;zROujkJz()vSDk4x5AS#AEjLggt&oaqQC%$)NNB%q81K$$ zZ9UMYn2wKv&qg05@iG(kHnq14ViH!A6v8rS{C2qIX_!DB=!FFaR3v~@!?5iNsfK7~ z(I~8?c`C|$3g7>E5~A4ltRP>Zx3AW3&LKI|8gG!pdcR_uhwIZ~zEYxG8kCmxTmMuX zZBdH@f479Wb1hV{g)l_`_%QW?(Ob`U#YgKYEh4+FfO_l$PHD7$r@G&pvl$4y*J2yz zP`j-E^>@-eR~GoIvXkq@>g9vsAb#0+Lu2wg5V($x;~ZyurT%h{VVNVj&$%mnyV%R` z*OjL6N_fpORVdvi{;+^OyY>e%?{doIaP4OA{*nDFb*QPPh&^vmYoVaa;(XROP{McehtY+Cw^5zLU6}nnRPjkvm)};-%@KMicboU|&IPwE}^fp&+(HZ4l&mT z{E}PvS7|kGAN@S%jk~VoIoMt4vg8tXQMB&!D-s;}*0aP`gy5 z{^)YcTGOvw89_9^_;LHYC-z2nb1R#{jbj9aiLQ{F^L90-N+K<@V~EJt5Rnk9UI<(xBv49U|?0vXJM{?H{_2 zsZK6P%Z!$p)OX6ev$sb0ohwh;Jb1}o;nSy z%SUnHma6`MB(qU&DnOE07y!+fHG{Y4&a3jb8{(~s6qi&73#SKp{L`_SN(qn6hg5yo zs+FFhb8=iN%-!!RKNdF9}8n8*XuUb#4Zu3V-9yAf=<*Z4a=Eoizv%@gBv@oy0o!^;`)|1WyeI;x~+a5 z*0fq*?Wj}fCv#ou3}p#3Nq^_^ICop_F;lf8KB_5Nj6I?U>v7je$QzlDx3(dvZm=8y z66LI$N8hE>!|k(Ez4}tC-jp1t7Ma=mCC$MR*Zb0`!dE_{A$@$-)ebg$JU|?g%j9WWYJs zT@1PoXapFRSjATI;qCj6>4YE81BS9ajlaE_-*J{0&p~(gN@P-{>TEm`qe!0tE^5qz5ZcE;0^KeEPuv4C>qC`HBqtEr(I(Klb2<0QN^}Nb!c4#hfncV7y_0jz#zQgp6h|KLmrCDjeMCy_?Y9=98KY zs>|}loCs0N$Wj{PUx(A5j5$H|^i>w}GMDI`#E)Xxt-z|9^fDBoZo|I$J*`Q^;dCK! z*>{BtTZuS%k?4d8duIgQTm8HU@2!d63HI>a6MwPxsn?^HpXU3byrhlJ4 zSEHJa@;0z;mD+B)&1D&|(!ezE{t9QKG5W_pBf>FvG*S)KBy7Vp+GSFv{iU&AcVH1nis;c&{M@)| z!$0e->uygXTk;*CT%r(kX7}1vT@QfVCa8E+iaM{i8~rj$>jSbmjl36v>*b?rf({p@ z8yHb(P(w5eHsw|*1@R;L;NX3HVvOSjcj;c@Uy32%uu$}d{wx~%Ig?%Ajpaqi97`3;c z&dH?gYr<(Yb1K(^<@U$=)78fq@=1U=qJp*{@AhLN#`Ow8i`DrAr!c?rt-_GlQ%hzw zJoQ1gmOtaCC{zYDg}v9*?gl;-O2|s_9Qvk#;5mPxO{=;H%VwQ5fKPN<=ltlAG-rdpo^jxYp zo*~R}v1R6?MfQZX#r2s9;oqSY7#Qt+ zHMMiT+DlgWUI0_II6+hvf76;k%R-O?b~;{o@whj3o%7J^%eK>Mo}fr?oRH=UJk`>A zWc;*mwxE4lpSnu;i)H#=>XAS^C)Vb5J_Tu z$LVS*lfZN9?m^jJe9r<16-}I0OM5u^Sr7FOE=0>ZkKgyMWL4q`pS2{<+ZDGMG&5`; z^?8(P+2_l`>y0Xj0DpoXWy7R|dDomdjF6f~^$uv_vXxaFW^y|jlo+j?96u%TqTS`> zqe~y6klB1cNj47xmmC%76vpVpCKV?C3+_&uIxoCamGzF3!t z{g~>`cvh2GW~BZ-5N3RpCi0cS>HMSU^Y&J&VNEUP`38DQ_&|JOM7uZ3*GNkE>TO7M z9#};Ra%@j2(-^WXed4E3GLXb8RJZQbMx<_m#&O|++1IG3)tb~|l!A+pv*(*O*U758 zG&Ih_`q@uiEgS`;Ot7((BroQCf4n6d=qdI!qquY&clZBzG3oYQfi_P6!$LszBz!-! zjcRvQL|}mi+7a09`Sq^f*sABn_x9CKg$u@OkESSF#qACod=Lo2skY-O!`#7|{U)|8 z%8D*lQ=^kSeqT@GaJ+j=DKGXtDN8CS`{1-L^em!B*}lK~NK>zeCEQjW8YXJx!iEzc zDI0lFZ+pJeAEeAav&+qZ;?#@;JAs~^YVsX0tD7g_>o?!f@;^oOQaqOMuh^@ANvTND zN4v`+zwy1NgWw|VB z)qq^4p47`OtlZ#IVy59hciUuDEqfY7C}Py;txpE;c&z*0N9maY@^Xn$BTp!^5M!z& zohv9lnWA|8r0D3dGtzzF4ztS8d%wCAX6q5jv`5LL5jJU^>N@OMEXEh=am`BFv8{9P z29DP!o99Z&bAz*Na68Uc9QAmBHKnzXmh#_4HF|C*U7nz@$vfKw=tYS!co&%7S(v~cAXDS(Pq$l^8ni7f$2NvErni3Pc3QSp%x-;(`Ua4IzBoi- z-<}hMF?}+~HM3CNaKWj;86GPXb&p73%b6x_XR$@=!@RcS_h?!!&j7?!W&&;1>tR+dOJDg z2Ka(Ol-ad_n=a&kRy`Lw(iww4uWe8|U$W9y6#JrNs@~= zwN=wv=|V$ZlPQZKoGT`i@;V{D`$YI)6QMy}tQf?kp5^F} zr&FrcJ8>UOEULVJ@~eKL1{2h>9S{H%`>FQe5aZHdqH4{H`e3uO@pxCl=b5%v=U-f? z86;@M8Z^HXI&;nvnN2zR>9Zcl-6plUH$s>amUou z*+8(5TXmDiDydeRB6SD(c89jC@E%@>8R2aADqF6+Yjuob`5^YiD2ImSPG~qdc8`IyFUzDl|%y9f!T)wF2EsQI^f3#hU4>v+nk z;sLKFxARRB&k*_rm z*=O_baKvo~)2!vj&=X3Gk%FA|{PY<|TDs(OKk#_ZVvA7DA|85P)It>YzqR|#LijOc zyP)>hnCIr_`1;L$-d(bzZ3xFnrA*sc`gvieFs7ic*s09srVn~M2`nzBgXFG( zDvPF#{)LO*=UJ6!bS*FD?9@T_}nI;+P#FKa{rdJdxrc2>O8`|CEG)Yi}Jm}%r z&OW(E!M)HBI5_b|$^2NiX!9uiD~K$^{R=| z=!;O5T9M-FIhzed8B0s%L|)jW?)rVUVEdwXq-;I)H%yYCI=^_Dj>gC|tuECiL~0i- zClyfE8)m(+LUoo|Y6%07o6gyLtYzgk=d?rV%#f2E{(kk0r}VQDWs6#;yK}k*dI4cu zlH`nkvhaLx;*<*4B7qU3ij-19rB&-5|M<-hItk7DPIQm@B&=|_eQPwyg`nPb`fHb_ z+GV+Kn)4B)Ud|@w-}xRpc$QcrXrHcgXqHrN1b-mW?0jgwdR=&@zHOT^+4^={RKxV| zD^Q1_0WIsDkGoNvUbU;NR>#r?3aoGkOPmPNmQ&NR#!LU1mNoBjqpN4H9C#}`0BK(} znFCm?B&=uosnr^CP=Kwx)QZ6`*(zvt8y7oL5{S5|yM@<-7*@_=-&FM`uqYF@GMH7Cb3}QSsd9Ck&~mWnGNey64j@L`NdwjGSJ{ARvE2({>h@a?e_itr z$yM@ZRr*pCgJ&Q?8!(m~H99=Q?&hYiyy6y_`Y8_@|4aAJosp*al2A{VGyiL5y zkVv}Ls(;zJru*EPxBF$`lny-bHNQQa`1!|mjF9D3D|M<0YcX5!MD1DZ*$c7PWwa! zIg^ay^qSEjrai@E6e;2oyC~=@x0L(rzJ4(bepzh~>g6Jx$9GjnxgL3`md&3>D4!-v zcn;Ydv!%@PrDk5Lc_*LGDk2w19U2GRPvzueuI;yoXR0sd>-5^Y?A%H3Vs*0Tpmd!U z{hM*V@7PP{&y^vRX0LH*l~t1EXIt5~lZR_6F})q4A*zx5LBmN=o*K%S+w}6K6ZWeH z4Up?-)!BvNX2J-rY|($VsffN1zIJ=l1)yGDQY+ zv2+NMxGqi;*{7Biy^!{+tRMwHW@c+RH}Hd35*qR7xvZ7DO5iSLOzy3VRGEImm)Rbk zZS}MMGB2?L5G~S)9@NA{7EFR z#|gTYr_=QfI&S|6`M{{ATVD|FB3ONHCqQ54y`%_!q5Jzb_<4vusSBwM+;_wl4u8WBXn;q`H=8 zz!DUDpd$%wnXq4!zq&lWr(btwSav+iKkj|@Pnr5In6{KA`4}dZr(s2o_2M=$M&xW* z0az@IIj}Mh>}62Krp5+yWl20T1x$Mfi7`V5y*%n7h1e&_317ak_54$I{vL;m(_nC} zU#HYkjGlv{q)L$5*mZX-`%$BI)KGxL6VSYnGH6q}0TDisWc8%oWNJKk5c{oNjXSR+ zxn3>=?lW2d90B^T0|X?$5IT8VQS;fwXjS?*XG?Ue04RVb1o-Ut>zW2wR}wKt?Im#` zf((R@QmrOGdpSjAjIbB5O(X=m48`e>%PCC%FxIIUAs!W%+@;LA(R@|o{BEq@_|ZPh zIq52Kgz9F#PgsaX*=Rhctcd$o%AGBfZ`@xlqau~9SAXtXZ@l@Y(W1hi+(LrbINnGm zuj*wgYbI7LDYroYd^;kAqZa{8K8)WH(e6F&W`T5_{NSKXwn=3DAU^$@$-d)ozh1uI z3*xcZuk-SFW0SYmW}Q>Wd$V?;>SgpyU}?f$v$6td=v9KJmX9|Jlj{2JF{^&wUf^hJ zI{QMKwA|p8;)#La0!70WQ0T3I%#O!nbYtx8c2e+5|OhS$XZ&IqNwo%5lo%x$$M5vpb2fS+1S>`($p$ExDRr+{BWGJ#WtDW*W26USxFaU~ z8{#eJz?e!emF+httGvhpxmr-sc3)a}>+;?xZnzydo7%T%2uE5h&$fIsdGK2s(#n~J z3Mtonvhbo#<>CV{&rE?jF9=(BEiw>$VA+r5#{$$e0_WMW+}hIK@xf-#)L0=LyEl=` z&3&Ud*h>%Pp?3XCb`{K3l`uQ=Un4)ewaffPJ?v!Z&^fkriB z+=whY<7G@g4Ng7y#kP**OygQH=YSQ#z&a^ZA$FWQ-|I|@$rEd?sXtP@-Ue8AWg!_Zx&G7Q6uQN?Lhp$wd1-M`ah|Jj4 z{(FU0ZtFa}k&<+`8HrXxm#0BgGskw*4`{Z}1`-OW>ki${FTzx$30bPZ35yVPowob( zEbXc9%OcVxM|SG%!dRNYLp@audA&~f90TW;xJ{t=#G|v4Ll4P#{c-uNeBI5h#E9CJ zWc3wP$g$P=rG3Wv^YPl>0^qS=%7A8-TCbT*yo^YH|W`W&Z<^X zmrrAH!^m(}?c%9v2ktltjid_)7}OwvH1huC#K-2(}wQLMS26>sI6_%$^E3f%Ae$h(!uXqV=C!rkqAwhk+wHr2D8;8g)&)>tesYR7u zfhg+z>6*Mv+9AS&>Hb%;7?`oF+NCLBi~xCx1Fk!IrdMBAPhI5u+u4F*3#s4fs}A(Z z_8xVYhfk?SK{#BDf2>y_5^rtQN4pstH~7s_uYhorV)QR46hi;@E( zxs25Vadpg7n9jnWgf}upNUpgg&G~Vwwlt7HkB)&lHhCA>mkve_$-i*38rVpQ+7Fi~ zmx`P(XxtX;+145Hd{2t8yO`7FSbG&^S9ktwO8PV9)N%db+4X{vd+ zI^dp6uA-gZJlE?{c6=ILi`Zh4zok~;+8(ifvMa)+ljM{q}-14dr25 zuf$D;RcCs1)Rn?=D_PE7mWkKC8oCd`_$5&qC~pdvc{7 z=JbUr1>G$*s$@U*qD~z|UeiA3GII}6zitYhUDvS}mC(ZoL8rI@nzVMvRdJH+eAb~K zL6cftT6=VTTIZZ%B&fqckOl1q1T`A48=o-T=X$_tw829+25co(dfHUiY*)XK7e3#V zu2s6qB`xoK!KT#t^VEPyyi5THvM5_7LZM!Ay2SM;bTVH|yOd9(Wd{{Ew&gOR2CsVC zZ^$KBBNvfvI~*s3Xk-4$bjjj#J1GG!QM)Hr-bc<-Ar-sh{28A#Li9;lWyT}E`i)oE zX`H6Qwm042y;qr#@|<1YLYW z2VuS7DZ56{Gq&o9p^p=Cw-NH#h_Sm@51{{)t6@ILXV2*LusvQzGiUp27<_iLN8Xw& zsyE<8%UWvP8c#7twsj(DycKVijxa+7Gt$1u53D;846;onlU58CsuyGf9SVzSn}3(_ zFB*`a;J$hDrblGIg#jpedHBCa>vgiJZYN)vIfSmU=*{k8DB7wwRAI(4IZ>pFSEh%x zE@YwLm->On=NWXh+d^5s z3hPgff31Vay@+nUie!N#UZhP%&mjX|xO`nqLitYkjtJOyKAZV%de1Mjs&iPLHeh1t z%OApw3lSW!KaZ{G%%F~&NTMl6u>cEApJ;I;lK|AG9)!eX2Gr}NQMiWGL^w%T%2 zm>yrsX2zIdD9`5rn*?q5y2wMvz_C4S6=30E}8ZovBhTfAf$fRRJ`jWY_(ZkBnw5CyUm`8BRN4)6__M~r2B3R)M zb2{ZvJD^jn2&Nl)H@^Jw+G-%?`*r}W=&R>wdu)ZdQ3N-2#$ENKnx%+`dG^)kTdw9TiK4=2NDP15l{lAC$*T(}fE?Z{Aoz)BG36?{}X! z9CyAWEY3gQ$!CZwQja(Cx@_*sT0J?QpNV%~-yV#(5-(C>a=V%cJX{vYOHbV_m!sZv zdzbcznb3(8<*KI7Q^3oR2FcNtw;Rk9sXw&H&YCKG=gX1##Qe2S&V8HXW+BgBQ=YxC zOoN!;t{6h)W~<@Xa(eLeXRYG${tccpO-b@(mlL%rpB?PgU=HsKtFdlU>@X#cC0HUM(UesDC9_B5m+k@uW zHoPjCV5-9xv~}WDb;^PvYR0*C^&&-4dmljF;Cc9f^^vUdqUAAjlNO>$wJgbh-wVjk zgCg8oyR)et-`r#6Foa**duDphZp|SJFYAsd>*vN&k@crpet>1au~~Ly_l*_c(2N;? zvZ^=BX#njhe*yGTfU~psM&K>*Z7gLs{5*gyMNT@K^n-sI`v@~cU~*&vSk1f=@?VIG z>VYut+WZ8bMEsb~TWh9@3ea{Vt<&>H>gfW5ALVvada=P@PYpI9L!RtHXdbr!Pc)4j zNz8j5J8vS!{+AbkyDlAp*Vci%&J(pzLxX-J7krJvk5`b0HBY!p-jI560<$+K`RSG3 zC??e ztIo+I4X)>Rjy7s|JbS5O_`CSlzwY)aHdH77YC~FH6)9-S#sY={Onn^wWa+VkXh%48 zdOMrYXSXk~^gW;ek|odP%BU}uH`?Tq$BJ&p-nR%?fB&4D!DX?f(8d-hlncPwS*Y0T z-kvw=l)Q?@;Utf|?=J2$&$rSt!DQUUB?iGW-`x96!xz+#?(O*W=#Q@Eva{n4fJQxvlzDl+^ zQ&-eK>JT{qDt{#~{m^KwyPMJpFPF&tQf%{B12P6NU7M|PW7-x@kJ+4aX-@=xk+QRI zakv*eA-VkTGv#rkpt;Gsd>fU69F&62jt+;ujY2x|Xnv6bBWA_@PoX5+(O%oR9ABX6 z!U4tWoU)GG54fYnUryetCec)Z*qXp_#K*$yyDK;ij$Gh3h(yh<+E#{XeX!5pUo44% zhhC0eQb-5W>@?8@nVCaa^hj}>K)ko}?0SYSSA(y48weAJj$o6meefEiBnx6~Hr~tk zgvQfLA}wvI@hHrTTLh6o(aNN?JL zR%NA6XGOdjaxta#kU*y5@qsZ)wckWOd3AomRIIQrg*}COol&~9{5a@fZ1bi-tji8J zA2voi`5f2f4q`5AaJGg_gxSgmzFSP+i8;AvDvV)8&_s7yrA)l0Wu zVu+IP7infyLFW+)*wyIp0M9K2L_LgD3J7!Z#I^zNj(5Rl6ZFG-^Ag!Dklrua=j;M^ z#g%C*XIN?~d3{Y~0O{)#G-wu@UuBd@k*Rv+@LGIVN##k^_iB6JzkBM->jmcBUsUv$ z>4d|%YJ?A95==pyej0uk+d}FNg1eB3?6^Qo&o=u=r8!G| zfvRlsc?%|kR$tFi+VvH6FW<`6M!Ah+sT73_|&B?eaehMO`6%Mlpz_vZjNkj`yEc zV@LCk2mx{`uR=|r%T#`a6tqCv5qS(pknEO?3&oIY`mEpb=Qd^*KbV2(f{)N8FEovV z)%F-ob#~00J?oFA~pet&0F-Gi&MS(qO190^ki*NY=+##_& z1PKL5)W&h}N8HsWk=FXGp7q=F0Syt756!#h1Wiqq)Hp z(VJ5NpEBBXa8Rh4^y7{7!MRSZyN74c=&=S-#n1i4+gk#cn}@mTh}}X71FzPkxLmRk zNT~5w{b9d%KznQ&P9-OK%`^j}J^J3AjYBa~>#J#EtlJ)THqQSH;x|i_3fq}oiMij3 zfoRVaJV^M8Xbr=AnQ4bw&Ngki7hB{ou@x_qRVm=Q- zwrrf&w#RIey$oXY(+I457>L=;v1Tjna(2m&K|u@04Z08;KU}54SgoU}8*aLyhPDT7 zBBbUm)(RVU!j!L&sQ&bQ65mS-$)lp-h%7DKYbkK!e%?idq*YIP)^{8H!= zu#pa796#Pn^I%=Eye!f+xn>W_O*XiZleQyi+l4h9;Jop8@=wd=&-4E;KNJkm>B|wE zFp(dO#GKnrJGpGof_XPf0a6co8U5!E$&8y|al};ulAcYZ&*7S^foA$Fh$bBaRnI>? z%U>P%mzeM0QmOy>&yX%sP;V*k15Z;$P;poTzds(pB}<0c4y4cnfqnw$hV}-WF09vn z6#VZO{R1wXVO{Y$DEypjSjh8ASgbG)(Iwr(rNfj_dsjy_(#C454v8ABmSs!5d2o6vV1oqJb8D!hM%ujPQf*u}4=9v)3Rj z*$zelhaeWfrc=3mcD6|~sfvPnxq=XNdGz>h`!w)RjRCsDRBGG41VxQ}h1j#V@0V%7 zu63#;cQsllfAHYJnlAi3$b2G9;CuyYnFc&oJ!XaHTn-_I_0GLO#h}83`nYYId7l@} zW$OjpBo&+1&s+Rcreolnv~q4;p)OEc6eKh>oFIJgb!Q4V87L-mX2Nkgt1;2-rL z3?G=iy!#rJY&iibGqDUkzJLRG8obMOwU4x>DiU}MCxeMrUH$8HBZ9ez#AA^dRq4Y|ELMl#QUC{YveLmb|_p6Enzryy%x}1xp1i0Qxzh{ZWvFt>S8=;rt3}K@}qT9aadP;;DzU ze^=$QpE@{oPvUd1aNwT3Si49DuDxaLL5tAORZ~e(%41T7z`dGtPT_;+4$q-bbY|Ob zA;g@_T?AY}M|yVa1E7j}gL84Q#ti@OO9D%UE>O7q&XfCuxBfqVK^mMgW*PEPSbu+g z9+w!yGxQ5P`i%bXzkz!PY}61}8X^3@zy1;C#?NxCQCv*UdCS$X{T*E=GC zx=8Txz`ySMe_lSMgBa}If7WC3{`Seh{|TZaan6(5|NUd5-+=YM`HRPl4T{=4s_8%r zSAcnvq7lAVGzJzPOAN2A8VhLfXL!{FqI`usx3n^*tRDeRH&FL;E)NF)AzWwgR3hF%(6 z8TfPb0Y&%n|GM&>f1XG-dF@gZeMw@|Rr~Vw>%U+5NGoANA>g<{`?eqfmoW}Bv+v;v ze{SvtHRc_vr%yhY#nOTgC}{@HGLvKtNc73Z=-2JtYVb6EiG_fw#MihO40=%l^-}91 zAtiVuh`b8>MP<19+bcG66l}vy!vH^V7*@;)gccS4YDWvz1~Nl zCHWu_3uv>FOL#zM)3+U}6wgEe$E_lrLCYayL>o_yp+(BcQIIz-Std$vgVoW%#mmTF z;k;7x#P3Z*?z3og?c>Z)U|}*Yp#*m@zAQce`%0v%Mz7SV^A)?lzuuu3j8S|KC;snI zf>;l%#2efGZhU`#eTfn561-5wzwaSjcJvXmy5y8i6aTV>|O!tGKq1G()V*)vf46ady9CmS_@CBT;0=i#bZG-`M$W#YS8@XvlF zKw>m%t$P1iXJ){{mEl+_gX64gK&lVH$dV+9t21e@ooRaDnidNVe8t5Up8|#0hZ;2& zvY_7;&Zw?tjBhJgNb;Tw)3jyyspnS09SXi4s-Hj>{K$Q6fQIN4Y-Aq)vzK-L11a2b zoW{b@KKkffgb{s{yW(slF#k-60N!tLM$KM>u8yE&hzDb6bkVbH>=A9PmVp+kcKSzh zZ^6N>$n_F=WPf;3+=(T!c_R+^BR>P~Gmn6gcAc2W3scygnntyaP#RG3cV!1xnNaR9 z%-6eQw}4rawLlu}mjsYLsj(t;F61w{AP}-2Mv!qOq2-dbw!?cp-Jq85iVm(q*gi#E zH;7oQ`^BXTx-|8F5e3hgZWN?TOu|AUckYJ(kzm@InXj~25vWf4Ks~O4F_D$5RRtK; zSzxm;PVUWcVM%d4FW2L?UuSF8BB z)KPb`CNQ>+FK21vYJM(&eOIGHw~D|()c_|-7#eJ&?PSGl7 zHwO9WF5&BIJjzrEV9vV)3U*F4SU2$%KHO(f2}Da#5chwbLcV@^dt3GNA3y+40+!@5 z=L<6JfS))nhi13}rKT8==w2#lu+rXJ693w*hWkwlBP8b@h@Boq88=QCb2{IVFvz7y z7P`|A1Uk4^13O3IVSz&^(Fn*3-!HO@g6SPms}h&LLO0Zb)v4+sw{cF$M+ezb^lsQMQE5^V5`?gjx?~u zP6kHkz3lMqkbuDvpA+c^1RtV6&J4uEyVFr>c9DC4d@y-O{AX=2$h(*BVTHVO8tF{s zn?x(vR{O~+w!z#uN1TwO1#HYc-n*Eo#^}mm6l5%sIPbr8(&vPz5OCTN za8OS|l@eKPUgm+Kz83`972rzC4QrrzwS-wuH%nv(tP}=vFaWfY01TMa_gYU)(5xwj z_NX^!z_&eB90n-w)(!#o99#qm4rVg9sNe76?nOnS`-^y-Pk^;LZ>e9FsE^7Za|Nmd z)(D%L0CtLl{3~Gk*Ka*SCxKGbMgZXZ7Je@}sVM5*6H$jGo@~R1jr?f*e@T_R$a~jY z1gGCMZSdY-4+-{kk0`tk{@UIyjfW*4f#vAN)wb#?+t=4 zJ{AY)efWPLKjIv}K-QYFRCHorfQc)zX}33vm;~n!bNlsG^DF$|LdEz$*zb6`@r@ko zz69?5-CeLj$3T72I}JSMjMBad+KqHzs{r>@Y+TW5QoKNM zFqpAWuAx<>O1dcvM_1x(>BjiVgvr|OwrzHRjhy!hl)9BhgLfgG@Q|c|gBTZ%6M*4Z zulC!6VA-D`SaKd4>$kWx?N4EE#Jv}ChD|Z(U}Xx{@D{}O%>iS0FiIW?l7w8;2~ez6 zg+u}-w=x*}ux#}C8nqu}V@%*EvvW`D6b$a59u=jqi>$%`K0y6>94;`2N!(z$G8r6f z=iAe8yxo0xW3R29U9tUE7QOU4g?0otY_8fobTddVeFT#H4@vx#3Op?j9Hjyh!jg*q zi~+yyYNNT_c)@ZvyLhnKLnIpn>@X(OX6qZ9j{xwQ67~(*!DKN;QU*hKfU{BNL3I^L zh)4FP?q}$%_EHt(4egoSed%;q6aynmZwVdqrIeh$R1Pj3%il;4RtOhVOs|KGF&BU$ zD;mnujF4Oi@-GOvaqEM?U^MHCeg;E-R991zG5tq}uD*zv+Zoiqrm}J__!osOE z{B=3hCbtoqgf2s!iiYO`YM zU16{$mFAz-RoY%A7uxMyBp5LMSF#=2bj16-;;19ddN|%i&#`%pdzrAGIGQmdse%8} zHa-ElGiQr^6mboVD7b)!A%7m5JFRvMr^GUzXMg!QSK26k>50LQQ%h2}*LG*?5 zG8oB>)^6Zvf8epa^T0t>V?C{$TkkD9SZZ%2f%Zk_;`dnEEE;^10QDJfyX~xP5hr|f zaXF4}DF=9psUCod0&~VWZpXPJNWF z*btBziY#uw@yc@r6TOptTBI;AbweauhI>-D^T^>H#|rx`;}1O1&_o9uoomjLe-c$> z0_?0pmDGpms0FCm?+Hb7rSJ$8HlhW@0`?l$EsVw23oV4Vc=<~$RD1Ao&)x)!6Z@>C z+O7aiJ6ZFvNyTV|>A@e}Ft|6XVD7L0%JVy{W0rXz41%Ct-I|z>iVJ&-%h)7IXG978 zrhvvHfPvZCk0=qD-RP$lNMMx;R;tVoF)U~7hT)h0D#_?#(UF|zY4f2ICiRXqL~R!J zl19i)20F|};5?8eHh0J4?wmm)@{QIljWL5968U$hP6SF}=#mziWD5ka1JP5{Q?c;1=*x%^MgVhdeA5&*Vah_6-YGx*#!(eGA4J>00@yF246w+>Xt8H zc>utjj(~STS%OMH5gx)2Bl?o?ErS%@9+oi13-Ft=5T2*ZsDmn)IDc>nztHK<%yP@q zF{l_tu2@`Yrc^f4;ec=HJ&r19L_TzBJx*%H2dimG8q3_=ypW4+mqD)?R;>vN7uCo? z5Hr*DXXWj*deL~BiTzlt#R5NNv)k$iyz;H%9%G3wHr5~{hRBfN=0l``^i$@75e^x- z&DBB|MnoGmrkM;Hm_mJ`>@pab+&41avOAeEK~}5a3({-FAi~EXM7VYws?@;>0Lug1 z(QmU&o@J@YOsW(t#!XPMx!p1xX+COTDQoQmaH7ErfNrE=gdIK`6mk72!&|LbY1EiO z#tm+Q_`@93^5$UMNqb2^u$&6$De={@>XiTE8r$#J8f7;Y#t8J!FWJWn_mO!4n`obR zU{1KDS9lK?RSFDh9Mh|SfrG|Lzd_0!r$!e8&j0GuL|qeZR5rv+H#BFKw&~EDf_l&( zy*CJS*48>}OzNNzTH|DpQtsUZD`WqQ5jw}KxoL-$J^gcMnE57%Mz(2<`}2*O09R#B zWYy+j5P=}F4H%T;r+(Fjl4TJ-1l(RE{wDUduveq=v^*IO9A>r zGE-w2R&LU20QJ%>(J06VhHzN|)Z6D^X48jJVc} z>3w!gCHK+q&$+=U6g@>X*{1gngj+6RI}jS)=LL{QgS5@VOfFzDKV%V%UAlMa5h$4l z@1zm)eUf7EC1_$gl=e3E&wzKd*@Pc}d6g79K&RiUi-!fu(*dT{0kB#~8&z2!Ot-}% zr0^KYhw<>gCpR{eJ``}A9yz+z0JLs1t$y{uhGVF_tRLMGlhdBOg7Lk|l}O{G%SGV< zB)@k>kazh<`}@IPVoqZ|F-+<{DWJrE=(#n)eFrVzhE*|eaBVM|E9O+~E_!G}kcs_V4uBrz$!w&v)m5ixEG-AreYB(99r(GRTOu zdw?%a#V}k?bcR`QMC#N#e9b_mk>wbu7nL3+3a` z@zr?bYkEK}C#cw>f(3jgDV`&jRX?y6mxR1?Ct z(MZW%d*d|*v;quWQ(9ZCr6ic7ZxK-nurulRhGJj^vUb2a)zzuK$pH|jVOW9D&I|mn zI14}qnALB;#kawv{vpJvY6?>%00q)(X~~P2Nhwl`_}r}|27x(xE}*Zj#DL=tKfrlL%{;yY=c%>ZFcku$&GK3|yJ}OS8l;-kgF%mmQqr zbvQr^1DZ5Pm=un-8Owky29A>d_(As`jfIT%u08)Nkc}$<&eKk>??NX3{jK0Hy;cBq z6>+8`A=MUoaIpmK5tt&@-kJ zb0ee-*#N;-&Lv?(2wl$tBbs;+4j7SbGk`G4*dhr|H{iI&Hzd-R%3DMw5tspy$+)Rc zf{A#5mW5k^WM}B@#4UD;=_Nmz3_S2c_JhY-fT#k_(IM-J~*=`I15RMcT7m;;&wE_19$f_Y8c;Ct+l zaohK27hWZa>y)J|KnIsVBJ9e!0rJH+1{SVs=>L1R1d86k7i#$7k`_R0uBx)Dw^Jq= zf`#sO>XpYr^^cXO(a1eL*qkf|Un2o_56-e)AR%PbueCCn#Q!+6h{X&CXu%4ICTqda zDGyMMUal>nY1*FpK(+Q5O)P-VDV7s#G>*9!vZ$u<-MxD^POtf78?^zGsv=1{;;qc~ z)EK}%a*3m+8=-*vMjj_m06*iZ2m|2lvpL^b_vh_)9dKu4Hh@_a0akCZJ6m``6w9de#m+BXDh`gU$# zjHb1%4`&O?{ z4{RRft7Xpe{fN$p5CR!fYOeUF@$^IBzTOOg6hSQCWea#rUo1nI-O_@ohClW-AMu(x z92bREY1+eQ4=kch2E;3nH;pbnPUdI3qRQhfm~s|mjM@VitfnAk5;rA}nI`A*2RBfg z6rVZ?E~H9L7+N>6It)dE3``}Kk^1obY)Y3V-QPvaQPH(qu*pCS>W_$aXzg1GV}>Kl zaO$4%4-Z5c!)IJV0(C#r=9iKuYFVtU{plh`a*JrV(GweCeZ-_G}X}_{&ZOJ6Q?r#(ZUu6r1^8!43*gOYAkloDZ6(%m3}bO{QgAl(>%q@;v&NT-S*qI8EeC?Oq! zG<Yni^fVDyLCuyMVX_(+?Vf@a~ zU<)ufk4@JZw;FE48qOz)j8Z%k`+|rdri<3#(@bGO0mA9;jC|FS3)S`P*q?Lt`*nR1 zTt>kLyYCy#BbAxQhY4A+qKUuKgzx^|6l0_?U~#a_e} zDE>OV|9L>DJ?Y+v@!R=*m!7f&YLl9%w5B;_GsO#+eqU&XGt+%_5vHcZH&_6=710se zlImtLfwt28ZjMIwe$L`@DC^n`5)rbm+86KWDKz_(Wy(Vtl3g8y4?4n`o~J3yiN50H z%KdZws{vd6o!jxee^=OLJxy>52JyudN<;_O3d!TiR<4oV7p^H~8%uY7YhrFusl zOT(tr)ioFq_#3+4Ddii2L$!4dSmzH=3UrA?y}t#nTi-WI-pU?v`vnBt3X7L&M?JaM z2NF_`qxG2!DPQwGS{!chPXoB?02!%{1QO!n3Ok7%EIr^3)Su~L(Vyl?33`P^5e7#C zbOOjUR^mht-hRuUf_oTBC}K}6gTX$s;*Imdvjs=ne=a+u_Cl#6^v-MTkpICRBg~WA zspGyA8cZXohE6VKPuCv*?_OA}Nyq(qY!1;e;D#!E`&Kwh(SQ{U<#v%Z# zd``Xbko(0um-ZG0rB__~T5K7n%roiIT%!yP;5Ksl9+-cTYP+6Kn zlidw58*Cu7G^u&_R3lqW7Nkv6Z{ID41H2*+y8lr#lmB0>^^f;N%z}KR>aUrvpEU1( za~&{WyMVPc`mW;VH7H4seF$XOu?tQllMs#-nMfnID3Vk6>Ua9F|K}h2cO}nLcH$N6 zPH=E!p2sfaJ^n{xY9lc3aWRRH$zhHkq1Rx5q6x@iWNeokfOB%5J0nY1|A$Tp3spoR z^AaoTQV1K>~|zR48ifOjZf zME#Fzk8O|h{U+A#my7mbG%=L$0dBn-n2I)mP?=FanzMQREQc0n^E3iO9s3cD8Ia!v z1Y67EN06#swI374KYcn_(WUYE|9E$w<1y>b!{KEq$ny#GO)E2yMRKdWA`b?^ck5rD z2gB&?zV|y;>+8_^HaJNKmkQUIa~f1%tKn;SM5;@7=R`7_TG~q}o?|3npy_yt*1qlp zv`F2^eFbk3QQBK7uWrIOvb(?x@WoDuu8Gq|=&65yz*<9`JFZLDV@2GnN2;A#42drN zY@vqF3;6`+dci9U_a-;izuz+Q!C#i5)@ZO0uyv~FD@_E%d^Ukbmj|kP{h{G&p&S9j4`CK_iYNEjR$fDVHMd&j@szs<_QkS0K$3*1yoff4yt4@uSvo z1rlb)3k1Auu#-oZSadW4uchF|X7c5QTDh%mwsyhm^NXx+rR`QMV4AQy>{i!A?PLu0 ztM~8M?s+>GVDAE(BKfTIYa{~SU1^gOIp*znW07qG-p&LBe`-Nrj;143V*do_i_;TC z+yS=y30n&zIGBb^zh*{s06ZTy!gcdTd96p9-`>mN1uF>)Sh1zbOc)mo zV_`3uHw;uh=4XR6MhbHx68LKeeX7&ua2O77>O<7quy8wXM}i;&T-EC(&>qu)lIYQ$ z#hCtPH;CuFx)MqyXpBk?@mE0mfo#D`Aeaio0{`44CXP@;Uo$h)6g$g~loWyW_!Zf9 z#S@Oo5G!giksOVW9`0X_@7+-*;PiH0Iiazqwe9)4{eCYA5ISj?k=8ntGSD)=+evc&kZa9>9Z@OlgKxo|G7qr6 zM+T_wl$kM*(qSL>@}6K$c7GrgFBL{FrJ-tUoJs{U0waZ4g)_7wwr2rbZ&tV=N5`!Q zdX=b;odD?z<$QWi$6U*_Dz-oQ;3p#nby?%qoqtlUXa8Hq$br5^d#0olQ3_(tfR1HizcImlg2J0UC zTi`E=i_vGO1STBpz~)MO8s=gtw$V;Z*R6XmKZjTsn+$5Eyy;lo6d~)5_qEIujyRTZ z)Q!Ggg99=RTpq2P_FY##upo@?@IBoyw2aqrFS=4Zpkk#b=)1x&g4K{MsEIbIZbjhg z48snDj_v;F>6xhrWSVhwnN`mh1qST#Wa4eybTkY0M{e)H29|w)L~C{{l!gmZ1vdnx zByn6Jxh2v+yQF=R-*|nh36c2Aj@P)Gf%CZ!EmfXo;#2;637ZXW$K5gHHijY0KTJMR zrFva5P*V$Me+TxW)cr-~XL;`2o^yH0?|JAJ%Wwa>?3=>MaE0=h5bE6DMR_%Fbeo?i zT_qx@E9ywRk=gA=$c^&Q^g>azMuDSRRus zirqj*sBwNQ;0!f8Yy(j+G8pfofEL02$bs?Bf;`%1QfJZc1I4Vab$B_hYi)jaz!!zRN`gdDz~6U z?;a#$7R;A{*Z=vV!D|jh>flC`t{`Rf#?KFp%;P>2J7GXh{wDFpW38^K`cI!sKL?RG z%qBiI|C#T!&_8o^8v<)(VY{8`(>82V2|e#EPe(_d44kE!-}+430S2{vX?*~B);~5c!Hy@77mZ&)cg)4pTvhW{zDW zYr>MXl~4sGqV09K|EmM3T3+=i%F3!5o<*~MDt5%9x2{_h$VDBUUfshVKdpoMcyE|5 z(?EQfsBZnSbSgzM_(oh-O#Wc?Bg=j!?AToy3txZ*tDf(@8(ZPFUpE<#eHxs|W7kYB ziju8Xp*G-AQ41TrnK%Fm$&7H_H%fH_c0s5W1~p6z%qr2r$8*ZYYKTDNxycETCKA!6 z=)JdkGptt;bor!IAww5QTSC+g>mMt^OEKquRJ!dH_f+k;CV0EKIs97e!RFzr+t9Cz zx$OW?oF|{k@s&-**SM!;7s-WF&09(o#0 zQA!Y)1Pw)5U2x5&`FM|aGMCmT^uiX7eDi9m^DVCd!`ExwWhp=LzKqLhwUfl~E?fCL z7~3A1C?MjIyWoAdK41{->zz&&y$$hk4JS!ZA19>NUSJAGUnL2hv~_2)H=GAI?fT4^kmZeM zj(S_>cc=A~lG(u}kcviu-n0e6yczC$#K+doo=BMeBreV&ur&O0zp+t9b4u~zp2(&M zeUR>(mIXK4lo%;p4UNd_HVowYqL8(ZSCr zDE38sr7rj9aekH37hc5ASRoYQa5E=vrA<1|iftIS!JO}>_&aW!Xaaz-9p-+ZMg3BZmUx>f+d2{hQ?1WEO7ipRn?JIt-oK5 zi}!Vx%cGnYo1dsD59)s3@VzR>(^Pll#mi5>?0LdLTJg5m59X6T9f`AeTcVFSmwE}V z-~ZKjJ#dd)DC*~?E1+JS3CqF6|`l`>IGSm)H>(4x1GmIRi zJ+NyKS_r{Ejt6%HDWPZ61D7Qpo3?BjFJ;34A$O2IAHM zL#;PyMtj1OKC5t65SHI{d@m+Rr%#a_6eD8UNe2q}GAd!`&Xe*_Ntx1Ma1@CVuo|f> zQjE>K0gFmokZn^cg_q3pbunWZP+CgLpEg!5*_{@EdIkw#}$)IXnhXBU(0>7(yuJId|%r^ta zxTK6z`*Kxc>3y{C{z(5-$faNKUJ#Ynn}$@}m@Kl!vxJ?H{}?Z)7&Bf064M=WZ+j)l zO+=ZTr(dB2;)&$wOflUHWMmA;?6>xN=@;iFOGlwp;Y7;DQv2~c05}<|6Ty^LTLeI3 z@*r}$*E~^R2ZASdqL7p_SpF5$(Z{`GDJ(mG;x+1%qWMX4bWZkfd7AKL*Jmj|wuDR% zGv%m6M&_x>wNpKdc&{4PJrVsy?Z>X3TAHnD6Irx0cjldF2r`VR*R66$ziR&JKxlb< z@^u>-zoWUd!k>@kMFM*?C?U#FIIL+{QbAe|^*876@Wfqq~08P<@fVtmlI?uunGU)4$() zxH2{>&0x-tv=I}u$ieNgzxI{Ao9DH78yN1GB@=E><-ssG_z7unDF$|vJnKv#-B%Uf zU#&;X*$DGwNEEVf1z!LZ;8xU^B)n*HokSVe9_GxM`XB#ptEK=9G)haC8MNG43EfnR zN~#_k*|KPMx*5m8p~8wG*_iE2k!W_7C(NtE-OE__2D+tf zv}gvNjkm(v6TUfZNuC*kceQQ^ZuCRaU6~|}{OW^!Fq(qoS567(;0sJ}>UUfwwI4G5 z>39B^PU$5M3dCGm`M+jDuSLI`jd*jT`?rBpY9gc`9URH+*<#=v1~;gWiT7_119BIr=n9e0Ipb9y^D2W zdNArVFlg~033-%#LcAPLCsC?WsbxZU&~DTXo;PwxO?*dya1xwFGgw8>@DlYZmm%bmKnb@vT{pX6Ds1-qF=uy zL=@J=*{0T2sqX^`{%4j^I|8lW3W*q zJtRH>QmDfwAM|&3T#c(8GB@EQm8^ z?90%r=gge+-nN&p9-n#lc%?nMrX;rq+%y}Zok=tpEg83^LmF(5k(f<2hOGl?ho9PT z>C+yfS1d$JnQo?~F%G_aT(Afc1Enj*C2lNVn!qPF-HwF-3d1ywlh)Vw{3sN`H@#&9T^!1!+K3c=I{vDe+~w1P1EGDh1S-VtpdU;JasX+_)6xEd z2Ot#eYk4i;b0+vaZ$QiVQr-POPwzka`~>38mSI(@FrI=4z4ns)dwK4H_I0FxO0PSp z!Bi}q5M-PT6vr%J8xuQZH&`rHZa2za_&X1z?oY~W29#jrnZG`E%`6^^ z?&UoKl~yhxA-y{;(-e+zfb^V|e3p6aEpt}J5(_EylS<>@b-H6dT)29trc5k!8odUht~8gloU3*Sl2l&iygB| zFur8N&PDY;DM+-EkoOn9YjsktzlO6xAsl)58i^bEaDH~D%Oe#9!p8&_B8@Y6^+W=` zqOFuDG+lUx_qZ#=54>9{n48_P?$2Wbel`LoH}EJTswN|5so?)R_+ZL8y^mmU86YM(RH`M1hkN$1H(kqnVO*JIrq$$_3xvUt{Or$<=e%9n zNkx`w{G(UV`T_!0 zEq^PLpmE<&O&|Gedc$cxjXe1`8B%~EN*yGo6?wdcS8Gq88d#58-40~lfKUyxif)bo zl)wW8THiaq;h><0t{J`L9V_W6`+nblL1pRyV^q5Dor{F`7E6#vIsV7}KxDHJ_x9?b zNJdn}wcZY~SItaeMhKbjW&_pLX%5z#_EpnoHyLdJsq-P*f=AyBkR`s&N{M4`XNnS5 z^y_wnK0EU~FPD>t$(82w&yAD_QcB8KDJ#&6XhEr%a4G4}kwx5!c6*D};P6NW(n-?* zLy9d%pk(2VPKg|v*8-~lSWd{7bo?ZUMMM1@Y+14(s7;8T^NEy%cmMT~P&ubm5)fTI z29EN`JG&%Sn@OM|VEM)2Q6toW`Q^*+N1cBJ=M2Hb`ABxV&@qQ5@l`akyHC@%dx-1Y zJx@)2FDd^wrZAs<0Tw(Do%*P%I(ZGUnYu`nujoK*se+0$!wBj32VJr6ct<;=D&j^``s zJUwRb_ld+h^=$d$k;kTk1pzdWe7Q`CTeq>}!e4T3>h--U+55qGsPG;c537E>!rdYK z>%|}s8FF!m+0ky*Xc`qHI=;ai(B5KT!xRKB;dKKD_GZ2+_i|gQy4Er1uQxHXQ7)F^ zhI0n~8oq;!FUjg{0j848$zW}NYED8poOsIzKW3PFKb*Ah{ABgEXFZU#oga;{nTHmW|5nc2CGO&|xR zt`WeR%kQjDD_R0?_Xi{eO zpT4h7j=TW;WmLh1{bJstfZZ{O;3|xaUkaFJ)&lciEJf`&AZ%NC94Ew(9hoi@&omxI zPUgje&#N~5oG##yl@4{}*uy%$uQ@QtCIA5*@ zE_xs%t%Oo5bj?%>w3Zr}jeEo8hyg?7MTNZWfRDzW^5yK9 z5}_(MeVbu)BOOG^8eZlU%)=4?!tDIWv2&-*KgB**ORn4_AnP3v$H`)@yT?lG<=H~g z;}OYov@Ow|@R4q5-jHEM^qXPMxiu@4tkF-(^XJ(BuaAVWUfmFyo7QKDJ~qL8XIAZ` zdB$_UkUCx%ZL8ZagW{?M;cY=f5s{U}8g&3l{AL&e%jj-uYC@tNqi$Vx{JgTJRkJ(K zaUK%?5^lQoXArV#(t7ScaDhZs%~N#opQuKf;;i{`ahoNM1s&7KSV0R=<=BE4dMuf4)pJ}iyuu@fSz7sPlT`iJ=>*N(^dAcnQ}6F4p0o6E76g8gWP3yE z{wkD_EeI;P+BFE`RPQKDa}-BAfagtsfM^msZ2#Ve!^{*o2$U2%d8kTZDh+}RtC-11 zhzUK}u-xn>P@8gPS1A;MonJ0x50+U@6@p54zcsjFrYQs+=5-O}y-l+NuX3!-X~{v# zV^*9W>0ft)kVbLI_;wr)D`$<7|E){jDs(T6ScOI7_w)m@7ejAe5=7e z3i~sZKHR%vl7ODWTs-^wHz zjRiT$nea?sFqM&$;_;y#1W1{suHAgZD1GXACngej^Y~`vn>B&79y)Ad753xY!07i4 zo~h9JeWuPu0y0agTx5y&s~$BOwxuN#w1qbDI6SQ69U63B9p8QLnU;UboLLxTWta=%41$drBZh zh&l(>suG$KSZ8%!+_mEJNBj6=VpFBk)w07;&y`)Rv`PZpwJ?h&F4Z55te(IzUzb(32cIPHoW<3uD@+H{fc(g zEwSg7cZK=>^?+dBvS4|7;*q9CC1IgF-F4zFT)emt8LEjr|CC52w^(EA0op%j5s`(xS;3qRNO{|Aux(7w!iwP z3DjD}=4}_`gG(7h5TJ!(gi+!WuM8aqe`F*pEN|N=QeoFI>=|mjKL|DcYj*})P>(vM zv1?*Oux|H8MFDODGrB+OCn156t5|G#P4DK?f~gtp*AHN72ZOI{Iq?J(>{;on19x2W zp$V46akhpOYOOPq5#<_QyJ&_RmyTgxR`5W;)NLhUp=Ds1?3+QiBumVgTgJM3zl@QO z^97-F-7C`D==pPUm6KbW_}gMXF4ZH!gL>ZI>(FUXASunNRymlNjXuRl!US1FOiauv zr>;<~(r9)~b_o|PUg8Hd%^g2~S|WTdT~LIk2c(Ra*_7R{h@RDmvnWoWZ#xE{dFgcm zRTyavZDatc4mi*ZU6ZJ3FVqWQN>J<^jJ3;mO2KL%9fZj~5s7!mu zthf4Bn4*!RAq%u_I_g|()QG}_Gh)f0_MOkBUjeCTf6_*$0B-*h(gRr=INQ}h^P(z0 zJ-RrQ*_Va2$)@z^IPeOXsB1F#H?4`Oe3h&@NJ#CSP%VM%Y&GQ*D^(03i3 zI&gymjNO?eGP$Uzs2Bto1?0&{lM3gB?CU4$I(H4Vx!x$^|_k6YVl zp~4W&`W5rf-Qp!79wQS|m042^p^E7+dC@*O!HeEV4Ou0b==VF_sp9X1j8dg^b6yox z=X6|!0(6l<1%?fEi;VpV0{o_b-j|we-Sj_WK4p3B#K}{{?*UPz2R%aspbyVEIF`Xh zDtCGX4h2o`$lEx#9UltYjB!Qqv}t9ggtkE@V;N1NJhYLq@(Z*dTP6 zR)1j&ZS`7z^oTrfv6c%gR+jmHJ1V{y74lFH^eu<&fFNh(`E_>7!r;-Kj<2sa>*NSK z^5gtwhmC;2nx?`QA3V5Gf3+3%h_U_4NK^q3_J797UufB$tqv>n7F_obsX*y@P^ zjtHjg_BEf2(;RW-ru|RcUzA3@(ZnWTx*luoWBiNu#kk((BK(Y^3^I#nXomnSie$n1 z!8dPyie`dN=1DKN3j2>#MxC$A_5o8fqb$Ufl(PmBqGdi;JD*_B&t2|3OKAe6CQ#kv z?^$1y)Xvi6B~@AQZn`f)k-t;ZW4lN)fZZRjrA*{s9#nijZ~PTFq7d96@4%b*$fP6c z*kL|DJXE*q85iYrYuCM%cUAG&7{K*H&`Iw0^MEb}PYXO4G^lE-}nqhAiLLr+GSga~Kjq5B_aXOxPvFayw}w1VU#~QWQ|BH1GuE$&6D|yQ zR^^=SHVe9t&$RClQ?*Dji+9j@exSKOYn=_BEeF57BG#R`dS#^e^vnX!!MmnrLHY`} zcJIqjAi3rQ_Aj3lYF@rlZ=abtU<4}Z_T{@fc}KRIUeGOSPbg5Bf77e;4g2BeL(8nx zxJ@SfXfX!S96@x%B}C%JGEi(LKYgRNwUG(mJz~acL)7zo(QBhChej9PiB{=`D%xpX zh|SzwL0?~UCu9cd%#PY~rlc%CWf-?-<hMP&~4vr2Z$o&^2l=PGr# zMpqg{uW#Cq48M7ElYFLb!@Y~&HpUT_&TvN)bZ}GXE-)1Ru<(%3MoA_oWiw%OAH(L{Ae#RBUQiUH$~x8cJjoj`+@$-E!*FP( zpFej~N6xr<+{r?iJM%(KYDABgY~?fJrM#QJ{`Ae~@V9ScKVz~0A0U$yn6VX^TvP9HEY3Ax}K>??x zGkKpr4T;EtQnG8v$X#si*ai2Qq2F=OD9pT5 zEd6b+bNU9RUgsys6EHD0o_c>jyJe>B(naTz;TcobtTGZkOHt zbFt7jye}In8$sQgBfppF4h7YRW@!ykuAmWNkj~YSJ_Tt=6D-|_BuM4cxI>YSe^B4Q0Y`Kjc(thB z*hp5LM%}Pj!wPX6vC?h%fN4uREw4`~4f*->7NNO^u0UFpl{*c2;V{00)Strxq8-~q zA1dM0y};)EfoTuJyx3o8w;a$QWs)+HI(!Dy)COR zZ-fQ|!TYl|`iWn;@V6r{&nA1|9K+v^^HQcrK1W4g#iL)Yz-3XOmi|m|45`+y`A{=4 zF%52C@a{3kqsLME$9sS;#gj0wEh5TRB;&}j_-=sI>;bh<%w8@(B**^b4IzxbLA)4a&J635QGXJeeON8u$H!0^5&QT>}A z_1mG)_Y^B)3jxAoHSji3_QcubLIO~cg#o3}sK`HR8r}B#rhYo0L$)(l{!Nz<5244P zZ?A%o-6#&oOD)I0owjM&&)FmU^|c^_v{GX9&~!JKND9^xa3&e|e9KVFMTTx`Qi9+j zQ|s_+E;A__HZC)vM5hw1ri8E_riP%yGiU?P@*i*}V_pb+H*N2^N6C6c+B z=)3+46m%ND2+Mig9=WqF0{;4sJ5Go_!b5+ds@$^lB!97mS9+1M=US8k$P}fX^AJ-o z`KP|o)Mv#=Xc7x=!x#f^=HsS`RY-}GL(B~d$|di<7B;7d5~$+QGOyW*xLwZaDq6Hs znZVDv3?3q^*>;?<6-!~Rj1#UXM68H-32aeYn1q70YBWSUoBkmAR#t$&Ju>zIP920) z%zkjV(WBfARF7k;P%R04!w=Kv)A+2R!NHnN$-?RXY}>H#f$qjmyH$MpJ0!B}rS(||yoxQIIut}+^@l26^cg?-I zvlpi`m(dBylIbW7vPqD0(DlN*LXp>kt|40~ZQSiwFk~mRnSKL#y$f^$rr;8pvk$nM z`5@>PIWN0r9Q)3|IsOXE@M=!8`H-sde79LhF_VbMokB(@Koh%LgPAtY+`ykuV0EU( zUr%Z5B|U3rgNlm{RkqO|#jRkRP)0Sdzj~QE{d~s}Pbt)$Z6|!uu3t-NdU zLB|=<)Sxc0OWqiIl{fe9Rh3)(z+4CyjDBvx-I%slb8RaYj~i8PAx#_l0SpU$nAUzH zhjZ)|lK=z|hQJb;>CBsS%l|ynG2E95f$4Y52+p}}2I6izyWURKa&k@`m1+Gw2c0W2 zywDGEK<`=5_A}+Df0wF*LvbwaSrWU9FjlRa$`O%$ny2hB9pIu|5dQh|=PdW!_1?VN zKhhRII&a$|tkYsj3IA3Mc!lDF9o`*iaZ{ggX;v0$u-+M&CK;=)sIVRK#Aq#zSoOPR znc(WWYzNf}k#i6iBiv98LsQ862>L*1gY)x62_L79`w-7s z1^bnH$Z!2FeKsgOXFyXMSNTI^4VB2HOP6d;7~3Gf3Xc>O(h*{J>JX$2VaVWCaNM%C z6%6}LNc|f|!gEI9Qhx)@Jm3Eid44SJq+7Y&-Jw?i*BZrp(}j#NmU8Xy@7E4Bv(Eq zmjYI~Y^=ViO2Aj~oKVY&)YK}7wX6G~Mm`uJo(cKP`S&}7sB}~qye6cFo)n2q%dsbi zRyel4;xD*uQ1jka;64TzaVDg5he0aWVFQ)2680+$>r&O%ALV0xfSi)sV2lCW?GLRR zTJ*?|GGdx2LtO(Mz-sD$Ukv2W7$q`H+6ux}0)5TM&Mt=#?-k=7u1Z9{Ut3V#P*%m{ zU);s$4(2?%_w>b!NXLMyQrOlu*z<>^&ZlYRI$@4#-r$emkB$-lKE9ibM_vN? z-w-fAK5SOuB)4_h*c*@EDnN!IJW@=+o1I#c4g9tUqavJ&7p}d`NtbUb3&63|O4o~3 zE4Cis9)yu8ov2*hCH~IgVOGqqcbQmi@uH(@%o zF`Z%+Ac|f&>B#)O?t1wSpUz>sH(Q+-gHh=N(;ji~=}2v!=}6)M!n(YwRgd3mGbdcN1>7rI=vFZa8&itpbQB%$SjTAtF4wvbVzw?PX#?(MgS*X?t0vdTW> z@^Qk>uqx^e(coEP$Z-x3aO5<0T~HCo;dGa011{RV{QOxGnFHCiZ*>S9?8JUmYjHy3 zk?cP%tUDSaU1tg%(GFc6w1B8&+kH;nZzVg+9j1!YrahL0N;T}?7k25k3w?CHB@3d8 zCs)rQBg>#(mII-Fb<;%%HmrZr`cM(8r5Bmh0yZv6ibG3gYm&jVg~C5;A^NTSLL^5n zJKd*!=y(mZ&xEPv*S@$+|L)zz)*ymT;+<8&;M=fy%=h_dgKi~yCM1d&p<8!qFUu#M z4+i74+%NvP%HHfEp0@=7JFQxH`!1oK@+EZF9In3?Nrb~1%0czD-au&>Bd?jjXO8{1 z$M+@B3im;T)eN;8qt16b3ds^0QIGf4*(JMPwhtwg=YH4fn`?6J^cj~W$L=f{>3*2P z)%5!fb~wF#y^jD|aTs^sigl?haDgE2beJMR5JcXjr>*!ChrpPam^W1A4m5roE^mqJ zqLmx`-q7l%Om_YB^oH6STD&}r%|-7paf4FSwSxFV(Yks)?Fu3%r36}7y+YVJll4*x zc7@>sCBLgsrNG=*uP9U?Tq1%SF#SR>Qlo7Af5}O%GOPj#W~Snm+J@K;Bg_?O65AJ3 zjW@rT4*~U5b}Iu#zO+PQ+w)Wulh1cYU*{t%JV%8cwmFA;TD7K_G20@a~7xXRGKkq(W9 zVGKE-Q-tSyoc_@(I9&1unfeWwVR}x9)76mgA>N{0_2sU?UMqp-e7r7VIN2jK$@wj} z1fqO~U!V*3N5iPD-ljB_Vvpjy%ZC{K$DZj!*R1FjqQ?xHXcG;eX|iAPPfuQtnzy^r zk39I)-Y{JuqoWhko1-azaON+LnO(pc8kBUSIv*O^Q%9b7sKBE}kiy6VRo};sVYO4jImXI%UcCYN$fUP!_A1LN zk#A>}Ql7JEO}sH662Eo%e30pDY?E`+&wkue+J9$0W}+ADAoV;>^FajdJcxFEQWInZ z*htca@fgjvTW)?hN=t~2XqGa_a{RvUKW+TOQXm>nc5~n8l13C)=JJ(KV})HD0yHVi{i)>8 zlEjRsU11tn@Y+3^fWLPSbTRl3e|FYQw)H{{iX)c-xMucT)QfCuol?Jd*EMy%N3!cH z_BFII-qAu_)2_1KxNG2hQT~|)T24PYXCb=nG)LCnP4uv=yU4B(x*CLC-#V zmY`n`Xjmqc=Zgc)jE~n9Lu^GkA4n^;G$%#3SZoDHN2X1A#WTCDmV*qu;qwj{QbpdK z0J=PlhIkWF!i|{XVr!q3!06iido5@cr-#YxZmoLaNzH@GB$qrd3%BY=$PJD2-6}7Q zrTFgXi9K_($KiOlX1EIjXD!O22aK>Fw8e0scIE`Y^LM#^QmMS?!cN+a6(QYH%P{L0 z{(?Q`y)o*i`Tsau#J#a1Z{y`W6F?yoV&E681BgEhOw4Jw9Boj^M^?$Pj1Hw9h8lAHx30Dg4{|FT|1p|b2U}X@rXZrWBeTSwPc9fZI(qxGN#7@s4z(S{gI<$OJ8ab{_~=@tsds93EE4EuB&F9@n2gx2N{JeY5987} zV_-0ex%1uK>Fb9qMek{dR)8l;+jR#=(gdfa2OjBiQB&|iqez*$N5bfqiK4eZg2^B| z=8{jcgxEaMT8AnZuRoCr^#_uSJ<#_2cR3ByXMtGc;RWy4pH;8LLkm)^DtTiY)1#JW zij(-zM!!_jrDP1i^wv_Pi7!j#!*=TYAzLGkL{Ui>Uf}BM@aw*^Or7&4V=?{Y<8QaQ z{w0zvV_lv9n;9qdfwNYeld5-?Ow_&8%f4o@TGK4-MpYKDh%`ry+)FgQgtT9N>);W# ze=-W<*gfOr2uvqwCGq1s;XFczN?)m!^v@GpM$~sZE!6Y9Sb?TmORjTixZ=jcCga2t z(H|rVFNlFkKrk|QcVr8Z2rSwTf7j95$tFVK*g*T;>st?c!?{JO@#>?X#|A&k)ZQ;hw0X=&0kkX|or<%Z#W~M`TG)hF)y`bzJQ$VD(enPYpJ^ z2p1{}86~tHc)K*Ky0-dP!%!8AweM-X!;6xG5sD-x$BdBh8@ogiuibIqNvap#e^4?q zhMT2ZD;bSfE1lRUr~UY>y71_zSG!|rzD{xEqF{8!RC%CUNF)B-F*qz6XkaK52DlS9zUlB>ig>Yguh@i@8b zcz>E^%=hxC9rYVfzz)oj%ExRi+eL61?`I{k>rI>bI& zDa}6TsZn{W=rfoUEqMMv+BKi>T6`11%_5VpS6>`_#g~{02-%XWqfh6JoO^-7>3I`0 zy!W!#sj}^YH$t`|uH*4asq42I-y{kRP2IKVJ5z7?Gpt^%$GK6j-dXI<;-EzI1(CJV zR5BNx--lEA6?ANr1)Ou`CpWx!ggH-_Rm^_?_i`radddht^Ru$xo;pW&`I--=Zgs+! zD90urU^O<%Q#b_VFG2Hn-eYsU@+OYNij5qoToBc>xHkq7>{}{a(Ov5i4Zq(Gs*3l~ z*FL=QKTpT?a>nvz=wPV)yEP0QgGH*z+{A6~nuo*!HtvygxYX4(Wh2^Ft^>^d9%Z?w zEr{OaIhAlMIpX|;Gt~CAxZp*BWRvd#y6S&z(>@VcR95;?NijH)smj7@e(Sn#xGYvJ z@_62^W#&;E)?fVr6KIjZUtANwzTf-TbA>I7rIi$7#~xPU*yB+3d)E#Ydu&0nZotp? z#@MamK<`CCPYTBHdR*8I0*K`` zId0yIb5}lTco@H|Y!^qna;dOSq!;8r^vKYEezLdAo~h?|9tMT2G?8;l6Q=+}xxLol zdtO>A*^1+wjikE)@1h z+;{ICtj$2}_@Q3YQ{mZH#y@*+<9XDo^bZ+gh!-NcPRY*~+XS{#)1WpZ)>h`@UGSa$ zPsa|eZ}X>gnJcsXNk19vO-1TX&q+x<4LIT*(fBkU@?9SlL~!b@^X2O8{JW%gg~`jB zJ{O&l=Y2EkzBJU_7JMO_a-A}7oG$4?*9ci~{7U&i@2g0Tv%$Yx#7H)s z;c!Gz;=}BgHcu&$X{<5Y%|ZJ6_wQ8f2Z?1L65?vEeKVQPE>VhAh5;oCZ!j9+XtA>~ zNysc+`F`umPJ3)^GPo)V`C#m!zKH_mAD@KgltdigJ?q1yx$zHtYE7MuEicAB_7sWf zI@wiLx$nH!NnE*j&C-agxvO+fNqmH=A5`BIt8HJx(T}G=Vi9?kORuppjA8WNrzih; zU@*zy4Qlm5RQn{fSLs)coit4)C5KuXpMP#y74(09zh0CG?O6@f{(t5{02%P&!NU}3 z6&h7DWiFSfmD!}*)t5gRn0(Z}MawQvF;aK{f(HlK7<1_WRp#xnK?UKR^4^>btK>bu z8(T&{`XQw3en?7fX>-h=V_Hk+-SwqUqSJd2O6zsED|?i-b|sOaPC}a%<13b@I%W_| zIiUVZjk_wxSpCsp@#5$_ho-jAM^(v`fa34b`gUO;Xr8A>AO<%Dtghj*GM4H9vwfgB z5V$ixn!}~xV^K-{JnKr~xj_0}cIThG*1fDzn}GDjOFg;H7vcm$FJ94H{Lfp6#(wa0 zp7?{?oU?&_H)%XzOS8ikGWl}KDaET%xl-r8G+aGKF#HeI;EfhB)ttcF1U^Mc3+N)0>r|v=Q`q`ez50yK~Q%uA*3J)NA6}Cfo0y zpE}I6p0AqAi0|DR3JWDt-ZeEKZ6xppe;pZ--Jc9kgM-WFZt=KU8Wy~Dv*eWZ6+NfZ zjiEm^&0?o;G*%-)bMVBcxjy_Fr~m(rZXmbjykD)`wr6$BOIu-E`LKY2arT3xiyU{q z)C*V?TeT%!^_+2zHbS0>23h7=GsuIa0D`x_VV@uJit(*SHzy zLJ~;8pg&bw#WauK3WX#235tOKGaTZ?2`ZQLPAtjN_E)QE!h_v-*b9cMzc`I66rRMwj!gX9Sns>|vZZjh!HXD71Q9 zu27}?_$*+h_T{;V{jSAD($xU0!(IrQ;i=ovl=g-UJs;cAYE7qs_6Fa=g&fT&@txGE z+v2I!*q_wv@(zidd2!4C0y^;&Q95IRN7i#;4|-#CIw$RUKG;=`wGKU6eaU#PUn&HBlB3^5AcyRf z3z#NpZ?%AR)|Eg)LUuH!+h>KTW`XS~n-E9OQhM4(^>1z(upF{r7j=>@>~#m^3BX4T zfRD7KwI@!Z1AB;W7&Q(60+D|C;eZ}>fY$p zIs9k10pKRv9F8KF(;T@derXTtvfTQ~eLq~mPD9hoBh|es!J!WL=Zsa*Ulq<#&g2s- z=w#lZ{wTtKQ?T2)Rc@f+p^yfQs5$R_JYF@`Tq#A=dNZ zr)?~96GlA6&nn-hvk3VtyCX!A=JYW>4f9Zs*x1Iv6{iIvgYrjXtj_Tu^D}PBSh*>8 zD{%)nI;9O9ka%h{}?>BrjWg&~b3;h5`vfi8(b@Cfwn=1DUDn|4R`L!C9(Mgpq zw@gL3RD3=?#@}a~2tBcxEaOwF%qDU!VRamGetXVyDmWr}CP>Wh)lKP##`|l|!+$Ic zbnqrjH^7>}OvN&lfK%uJsv8iYUUZ^?epR^S|3}qZ1ys?#-`}v^G)gzponjHv-5r}0P`XRHyBnlS zx>Gs@Bt#mdJEgnifAf6b-@$wCfz0fgx$bMN^;rbA2OrU<$rOErlh|~n5sIYZQ2slV z`IcdJM$T+Kj^_q{x@l<{&wW@t9r|uV)Q)R~C*}EEhP3&meR8h*g>5-n{%dr@_k>sG zmwvRzANeo7wR&CY*6bf+6>BvoBsAJ96zumAVtl@9u%TaO`MHyY-F`*~@C|xZWo}@n z=2)9Ozt-XmjnAb!1@+&EWo8x=Yw#?jJ52cFQ1Y^rWtPTXl#)2Yp*Y_ev<;(l zQXz-E*O4K8li-3B8PbmzVWH<<{CPpNUsQsb{L7`$Ecu5OrDFX?{xox5$;ghc6xBgs ztMK(f3z|jFF9A~vOciuRL0~NF#{NzFbt3JqOIGEk_>IEf@I<3!*0>i6j@88^R>$?I zEI*$RPUMFL?%7m*(l@}fkU1+iSjnFap`cGZLLVRvSXD2ZP?8GgvlUg&qWG#m;H&N zs!w+j1S1pcnvsJa_#0e?gd_+zj+5%tj8IG__O*^;$BO3OfW8_}DAemOR8t05pS1y=%fkEaO-ECASU-0Qvfr{O%ucxKvr4$0 zvZZ(T%qGuABuU|RK!os~X8k8v#nKZiH~>e#uVuDhB%i$rOoW0bShAn4kYsdxgiZrC zV^cFKfSL7dFrSB`23(1z1jm8%M}vAj7qB^3JGZ>9?Nv?&;U47lTzVRsqGW-s)!(1A z7A+}W5wZRA6SI}?wpq{?nos|ivFg4Rx)C0cU>@`?DC*vtPeT8mhum|zQc;glFg9qy z(j@7g;kT9H@5pWSs}A@lI~&5pxSHq7m4B927P)+0wZP3*?WM;h0Q-H?P`jR)=yO5p zi&WZmo}#_IYVFn+LG&6333xwnt)7B&c&{MLgMWLsJ~*wm{(6k^)ZO>S9%LgHy0IPc zu@P&9#FjdFU7ihCk}s27o1+D|3Q?kDB^ACPOXz5@j!$E~GM{#Zk$11%>mLz(oLT() z!w7fKz>XN%Vfxd>CjM3VlRsH$q(3XUD&!TTR>RkHJ}3TrZ4m`TXVZHD6yo z5*_UhP>b{a;9!3iH)u5KVj&C|On;GFiDB-O3h(5hG!=J#MVYa@^m2K4r_{LSaGtei z&ilk%RU=h20z=gjxS8rZGMFgcmY5LR6OskIYLDDcs`Xa<(i?pP<-VPz$TW7qZe!sNX!k1KgCXRFZ53@Clj ztV>t!VYQYK0VJH!QP!j9*G_3#92`8wWp4VO3Imcp+tU%U3%|zjuMVfLoPiOO{MDBS zlA!^I0QQvC*43Sb)uB%Th97ghq6B|fG#XZWab{x%X8i;H49*@2bI(_)mJGq=2Y;F8 zlk;f`Z)8RJ7=-;VzJC@_*yrCOMLzQz8wPXK)HaR{gHg`x@L%6VH6rs2mAJ7lwX2-u zSGKS4t5DVDx(+lchX3%eeauhCYoz@@7Qm|X3l2mIMN51ZB}uJI8)lWv6n%$~B0nYj z#f%x|B?9@6uZE64psKy!qG^+b&Nr!fM`^Yt9b*%uxk~DTGXq(}&!s)Uqg1*i>-^n@ z6g`lToLwNF!Q`?^we$Gz=u!4#)*KIHaSu-7GI(?(`$FnR+WSjS`z$Etjjqg}TRcZ) ztD$Ljn?vpWloZ(*pDg5+fh7;%E-?)@#{IRtE8|>z-Kq-MjKGCz)51ENDQs0rF_5aKZa-G{iOZsFPO~Su; zT!$~ilFq79HC?G$TQF%$Z5mV`S$YmmzzA5OIrkobP^S9BO_$vD)tkOJg7TF~Be&AC z>T`rA-dXYkK7mCP+nVALsG?I+SMZyre7#tpJHE=2_Z?+vXb|P3Q@}k&W<%~-tVH(Es+&JgX zt;B-AdcvUrUsWlZ5m->NzPHmO;dQAJ)~%N%s-Ki~n8AoyLm#?RYxyFhv|g}rtM!hy zk|Q}a$f+Iw8lKO1;eO|56_3DoYddS7kE+P+Z;(U8?V>w@=<>?~V}a1vSZPDAl3N9B zh6H%3clHz0`6WIFzDV*mn-NqDacRD1V(x#j)zGwiJM_l}G6SJSPiDSe((b)|85S{Y zah4U})+jLMceHSG&2_UnH1qkP!y)o4P}2OA`{HA;7vu!|`akj!I5%*SBaXL%OlWDD zAjf4r(hF&kp2w0l{XZ>VB7NQ8NX&RzL^NPlb&Vf9by#ZtK$^H&_F6hT{OB~LW;~T} z{rT)puTt03mkJkReA|9$U$Tn08kUQTy8gP)iG()8=i(M^!B0x4%n2R*72o3Z!=S?9 zPoATk^MGzF1@lEE5A$y#LFloZOD-ZuZf(6;>!8Vv^+~_r832xwee(f-&)`N%KW_!k zE~}3fXXONn01|RUa0eE#P6_3w*|R=ex1Tf+;IEC_sswqy_GQL%*@;BuAH>{?oyqK` za=a&4>19rg>E;;2_CAlG*n6LA)^c01u`&DX05*L@+HTWzp3yT~zJ3){-r;2eJ6$S} zBl*+r*`lo4T_1v9Uex0iQH{Rjwi<{wKqPKkN#AiT&8A~_q}0QOA@6E^ISo{XQ zM-Aon0a$hu-tB^N-VLCrk1lQxmV0se4czU&!9Y0X68L<|KGuFJObOgP^u2CM zd3J7>`$V6O-v0DEW!d>|ANk<}m-A~qwvW&B`Eh9vVAhBT*P-^Tso^W~$&QojqBoRFaEt>7{?w~-;X<+%j z8}Z?P^HWn499er;Q5O@874C{9>PN||H<>luC5Bs8up4*8O?p%;*&h=WKE{QzjdI!G z0LHY(MO?PnqJWM5314BeuDbhknw>6u*9FSL=>}a_dVTskbjH{5*PnKs$_6Kf=c?16 z)Jne;pWm#4STxV@c`;c``p5lC(c|w4sec9@1WJlqKDjHu?h3uYO#%X5-4nXx;iQYQ zxfF+I_<<5xdZN|!3|&H)AIf?AAH&aH2BIElyjeh$eC4CKX{@%{kz%Hv`xGMP_q@d< zMS()$VKgF)#pEvx6szjISA3B^191EM>0yM^f~FT^&$7OM$)Mo!Z7W-9l9?vl{`;n{ zhuOU=lHMcYlv7_J=@izQYj+d160DuKR zy@h?D;f$r7A73UYJO#MEiY?V~ki5Z)gB~yXAN2tI6yxJ7MMSmIq?6p|{3TDDm|hA- zTB%FxIqE_~g#oo+rBX{CX4V*rgbq&Jp>L_)(q$+l;Pu6=f@4zXi=TLLaNNb&luy;9 zz=uh84ZKbXctv__0l~VfeZL~qx+8c^;?m>DEW6uZu-g0${U7GVo)cbYdp$8_=kZ&e z+sLE8qF$5uZGC(F$NMZz4L{AHua~Uwgn5GSsG4mi(fz2SQ|^}~4KBcvnUXTX8fw2e zSR4#*i{QL9YL;f!f5=Za^mvt=9?r*{SkWY{bzwnohcuD+?Uj0Nj35OTjMFO7GP9O4 zCg^h%!aoPbE}8$R77Mx^UTwK!AWTN8;C{q7Kw)|n^-AT94b=EgQ8M$Ietg=BhKJV+NR(l zWSPHgZ_&V$NM@zOh;W~c_(=$?b$BEiNxCu(%qB-w_CA5n0zoM!oX~nY%3$52;eulU zjA|$M*RMztDSm2XG2ys_ng;xO3z;|6gv$KX*+PCNF`Z1uC_j(d6OkwW_ZS{xkNOHK z3b3#&7}XMMaK8!@HSDw=!!1;XJ&=kw5|dH#Q+?4+ATI&DPK6NN5%_)Bkk}# zT8i6s=32W`;i|}mAJ*c7@PcdTFgdOJ%t2Bm**{g~xN-GW%a09e(+|yB#PuEWxu~L2 zYk{Jha}1vaZK)`Q;DQ8Qlq5lE<&#%ElmOn}mKq%m(7!}2$T_Xkc}V?i&p`+KZcg>^ zzajPx)lHX@J7BS5_7(;me2PP6RhHg>I=~c}b^|g61ID}H6qEiU2th~zZpgcQP(c=X z`SD5e11B`YHyyy)&49p=kfqM;uu(Vc>fU^Q^Q|N0`C~TNQH(}OQ{X(u@uC?8eE3aO za{NvAnPFjJuWQ-o79ndxvcJ%q{bJX(KbD1^6aFT23KJ83de4VWJ4yP4Gk z#oh`-B<>6g^#7?Jw&6aJkjX}g0a>!U4x4Y*c?~tzH#$<%yzJFT?~5eo{)L#u{DFy7 zlV0gxFZVm07x)MBrc&wrwG^;dt@mwt4HMZC@A72B&5jwk^3HESzApW5MeqTTs-}l^ zav~Bt(b^aPv$c=Xntej`j_zGt|A5-LZLK|h21u4nf*c6fI==6=mfGAMSlo`tKBWY_ zi;2DP>nAIH5R-jk)B@8oE`#7xZSI9I2`q>o2;Ry7k$uMo##AC6193?p{T5IQL7hcC zzO$v>g7-lpWK!v2k@*V~Z44Xpa&~TjSWs#H5e%{8w@ZoKBX8=9@BgS2YmMbVzFr}r z(#SJ2(0lu+U8eS{&CC=v}Re*)}Oq=XF$8sV+%k5yfTC3fS{uf_TJ$W+HXM-eU z!5iOf^-_(>r0W1Ex659_lXVBUKFwB4QG9{X;ZPr%r`gW1( zel$4bA-8TQ!>AI8Mcc+Dg;o)N2eU~R$1HOm`ujjD#jalkUB+JTR#tuI>n#d#v#V-a zf$^Bn@<8`ut`A#Ulz_5gj+E$DbY1mZd1=YG2XWkCwC(W(N#K4@1SOPA;b5b+K<}GQ z1GjIilfB}L)xIuiL0)ewl)L$(DMBbxQN6PYyDdlFY9N_CtAfeyMM8da3}^t$25{oSvn1 z0ss(eh$ILQ&()CdX?a{xQ;mD;Jyp~KwUSh|73$Od$09P6@ZQ*+%8U|2`DGZ;s6M0C*)Q*nVg73yhbrw?#L@kzo<;-#-~SinatM4Y1$n~`*oWf5i+ZtZ29`C zJajBiVkj?~zhpsXbUP#weZp6;(~Z*er!D+yg`g&?OaO49hw5oXZ=OHiEkD}7{hHab zZo~(6udxq7`(Ymr8%1o}kh(--0h(10w5T+*tov?WSyZ;see`~)x3>fCTly!w&Nq*n zQ_aJ)mmpjL@P$b)=O^x~-Amvif2JKyXtCps@NL|4s$ApD*#8(YX&-T}V<@`bHKC;=IReu^MS$qaj?P{4cR zR4qA@a=8v3to{{nRc7S~fNTQKPdsRK7>eXvVsPmc9B6}{$F=S3>@1*B`k1^3&jl-( zRO$eMFAJtebM-@x;Rg76Qk|9T5a=9|vIs|lEJLsdsMLN9rkBJ%5|*#oroFG0|&Cetw2E{pEM7%8}uJxn!{dd@UChyzp zM20k0%c_<_<4a2Ivs2#a>f2HLO#IkSCTKs%mY_Z1qVs(G-L-DeGj9e##-?OA7T>4r z#ghb$2~5`%ICQev7a!p~X&Y>ot@G|DMLvLR7yx^cKe5#^jI0f&fRL_$H~D+mF~-MU zRf3}dzenQ3zbk!KelQDdRxWCdyhWMCsbXR)N0qyCi@G#}-wTnx+}0%b$ZWG6J9w7a z^mXeE`SUV($;Cq-y#Fmc>_NVcSv6v}f zXY~^X%X@3-)h#A2O7ZUiPs(lR@tvZIVk#-9vAC~?>N+uyd~^6#>E-b+0M|1dXpgI{ z0i$y9@9cyV%^zAxrE}k%ot}rjhJ|)x*8z9)o<|pQ*(J285(U-~t_AOQ`vWV5-z?zo zV!L`+5(D!~cYfAHi?@?G|(4Ro8d#0V=KoTWJe&rm`JLD!DUopj9{Rq3s zgS)Leaq?zjdvo?MPq{+f)z$WX{q=5F<;KeNgZ#_mGdlif!p7GraMp0Z4O<~z-EZr= zCJEm>Q-y8u&a?{a>sl+f7DHk7{kLcV9H|OfeW`l2n_|i`&02Bb2BZZ>`YBDK21RM! z)1TBtrw4Bow}E8&{r6To5_E66yi}(erooxbO=if2ri17#3C!kd5@uTEC8}&n%J~=C zdz-48&j_XW6`s>xA|MS$11Qsl&t}thCQfS=t{2$0+t27a?^**rHQq-JBzTWa)(to7 zB`V@o=wGBbsk5h1vbXX{5Wn5m-=fUHE0EN_dxE$<$t`nJ`Afsb2ir@+GKn8WKO2=U66QUrj=2E1Z2QeIJAet5?FHYDTIAXSW_HD$l5A$DWaU zmMxg>@I2TwOW|ApDPwVWT9WGVwgp(@eDHl^S}7O-x{}}uM4f;T8w;zrtlj(WI&;tNE&N2wZB6Za-eTwEQ@K`9=JYFP+x{&W;?D5h) zInSUNz90y{CMA&cj}F8>8EkARBhYk3@AzzBa0X2q;H>u_zu0s%dnKnoHqM6x8W$h` zOUnSuY%n90)k0=qFn?0nrFK+u(#knfG+7Gc9y^m$hcN(Cmqez@wic;ax1~Wg?Y&(7 zjX13_ho_p8XDD(%$U2TF z_C+kn2{KQfiRZJL2(%tO{%Lt}%@D?Z|1BT@6XJ_249U5AF)kD)DUnJAQD#Aw#$n8L z6vMQnc)5xZGX6m_{>8X^Fs2(yj3T!AvA}Y`pTCO-x@-ga9anb;2ZwbT4~N#yTff|w z-{_Gx*=g^zmT>OR77({F4!0}`LpxmhxrK=86?u%HkKq%N~)MO>@9 z)Zb4yr(^RP7;c&9BNcUa1JO0;ee`8;iHbjVsR1T$5)7!O^FH00Qs8(!mGjGUfEPYM zjvdCuiH*m%#1K89%nk`_QE95ZyC!%G?fGB=-{Uyx&O*jq@)8%6hvQNre~qQUq)_oC zqWt8YN@iigq28d8J|W=-%i*2tbnfa#eg%nuqwYlhO1)x>z@W~L^`L^_$ZUbZRe$XY&uBiN`2c+ zA0Y^#@-g8CZcEhlQ|54^v!N@~g%JFMf60e@9z{k5IdIbK_~LMEzp5^E02v1Q3_fOM z%~A$X2J?*w7?ealOq1g>D5%=q&7Dz$TWdwW-^eMt{k6Zt^yDalnS z+(Fiq1HSjD1H9ng8V%xxzfO<2$MWgM;mw$hNXT+QOv1Mws|o>J6LbJzQf=Aw5GytB zzkO!YA-u7-MH!#{bwZnXcj%=0(^F?P(^Qd?{}WiIIHrmLa_jp{y=?n$(_VlAKei*H zQj%Vgf42H1EN5BeN+xp*EMu3{BSV#UJO+lQ(s)WkBj zvtI(^AH-PonK@kuElBx2zRuNKNc`z-L6-*3XN+4Q1({72DQ(FeyP{(?dOz~Jy1K^3 z6MkqxJEC{r{1cAM{7hC5#Q zUliKAA}woypUJm(kU1Zt=Z!O$hl?`E5kI<-Y2HH=WdqUBBJ>U5-DG5q?*ns}bRa*V zCd%I{CXMDC=iDh3PfJ|g-O~taxj>&JLOB=$WRTv8>g!XizU+zP!SuOn{zrst1clhU)*v?!2?P-GYV)UASK@K(t;&xoL)lF1}p3+aR8BOQGBq?+;L8OC$-Md^$_)9BuK z>3r#N5YD2xqQ_kZb9$5#i4K<&MYe_y6M@Tu>z5;)bE(H&?0}sj987LKlsT2~KP$~a zLP(;TK_&LB(IgNUO&8(`?j-G{bc3u?ml)p1_BN*`Qt-fo=U#A6sB;xR8(@SDb6*3W88F}Bv70P>)JW^@} zrvQ3jR@4qiKx6w#6^lC_BZfo&w z|1|pL2jf~6lfLKhEVo1hzhV^gKDSIj0;U?%(ny+GT2!LA8YkvGh1kBbV2v^hCL);1 z+K7HsSk7#J$x>Fu^}=^$MZcT_<3!-QM$P^p6GF-V|3c)TDmHQ!!`t{m3If)isH{cZ zYr^cd`&OXN#8GUbrD}KRn1ecI0m}zrXDVD{=Q1Vjd`0}5FoSU8e%Xy3g6BsIhRC9~ zA^&Eam=>FDE;W^EzJx<+#3K%^cg(9v??V;^sf|qV?WBVp zXq2kUF%cV&q!a1BOF3f1#g#rp<#f@u-&Uf*U}%KTTbCE1F{2jJ$JG;jBTToMPkH?l z(6XM}{0+?3KK`Y<@JfMp?@S?sCqK@W90S>>pj(njv}fB~RU#`%8|KRd?}?2@tsemA z@*(k?oQZT=_ITcW;RM3;*mnY5zL97?$aib8lUg^UhL7-jw2&IcvF|zKF2Dc#g-8o$ zgQIc)D~|%A`;jIQ$DB4lcd|nf1UFv87F)Q}y@~ZT;fVy{KSLsF~ae&E4Lf`>2;<*i$a1kVkutpqEW9f zK!GSAx1y=)hdi&_?3e@k4@+oD-5Kr^19)rdT-05E2dCrSy9Zy!dB z^Y|J2`dtw`gRm{e8rArZRM;Ve0*q2=BmYd-N?u%o6v5QQ_d@ob{wOiQ-dMrXrS8Fb zy|H8?U;;0);|Ak|WCgO%7i<#z$91Y#R6>@S;49sipYmC42K-Ef?i6AiHQzBF+2(t;4M2zi?JkOEy+w_8G%RJcWfEOvi$SQN{(%e2|dBDjeZJ>1)7Y3oK-IK$W1XWyzE{A#kjjAO{7Z& z!`S!2ahD?(==q>bo0^HE#;Lst+UU-!HIM%Wtzj%U&q=^bvdPJM!=4;BvJOBbz!!;= zt`Cn|s%Tn_RVT*�V^0aWpi5DpNt+q(j_=oOjG-aoqmV4DwJmyT8D|mTIsV02taoFtS{Psz261Ooe`_M|MpbqvpUEu(@bLA8sL4`+$UAsR)oMP? zP3|cFE~%i@QOSrp2gR7vE*(bsaB|d*Y8S1UW4f~+g4m9c3uekR-|VVQ0p!Q772e`I zuB@k70$!ZrhFX8G@X+VZY`LB9*H;b^EffHOmKOgKJjAAh3$Bo~%j$|)tw1RfqfK?qi+-|cy|_HyuRZzAuR3!26iLj|>V#;=NXUJ-6` za&j{i^U@dF9FS@;IH8xXIG$4EVmOk~^2gd5r;|YzeYFHi;c&M+B^6_2Xl7uC6qlyN zUzo20#gt|HiMmaVYjsG*9WZ2m-a|~jib#IK65&zPBE59F<7bnvBf9=L(wTN09di3( zNLS-Thk;^A*Df)GUO}m91*MhUlv@z%nLNxvkNdjIgis8d$UZ4z-R9Ws!? z*GewyYZ4W*Ev?mnAB|#b$584=;Tl4SWi~NuQ@GC7iZT&?gfhu9#%f9Vm5OU55eaTV zQt0NH4fAnq?SE5iOboJ5Jb$FiQkOpFIvsC()y~oWz6`jCRPK?Y=br6e4?L8YBcQAa zg{L8oRwk~AAO2HOXQ!a%^WP$yk&)o5LtkO90sRp;g7MZogi)#5P)r)t_u0CaiSbV` zk@)jJ-aDi;yPO!BgC14&Tu{||O4zDRn9#`tce7`Nw8i9 z6l(!rqulqh?=RaiW*0g6>G>)SY6pRMc~FPMZbrsmbIGat(pLnAP*R^x5XX_dHa;OC z1L<@05M13ew--AEJe;CuZtPTg=2Bw_g5#*CLM38w>POqhH7x20^sngxwN;@XOoHkpKsZzI@hv|K*CCFb|@6A|8^5kjy~QU4_F z#Kn3+WxF|u*X+KJ%7+5V5Q%VTzLY)sPqU=Z5$Lk%GztPF0!^Emne5zJ%_SwA4ZWB{ zoE(GTC#F9ORQs^?Q1yc+H+Sl(O1_ZPD==s}4Z1(Q3@MTSF1^hXzbp~?=n2c*lG{^Q zHNc2l$nae&1+?Sw9vnLn+k1d=Q3N9vgGPl;onbF(%N;3IpTtukMOm=mx+#*L!>i1_ z<&KWGeaexMgdDvK5hcw?os!K_J72A{c zH~>~2wYFb7m2U4VL1clS->~P9`G9OWS0}}+(`Ks*$HdAJghc(XrcT=Xcm-ZAor@!g z)$U{?0pb@{_8)C^)tDx#R|^AtZ%^hg&+j#H1BZY(F7)J+SF)3@1qX~b zU1*r$_|s4vXIf3&Hzgu1SDJ5{rF-crPQcrE;?v9DjqR~-{P=q$xr$P;n7{-=^AS4D zb|2{Vaym$!^gfs^lTVQOkyK>)vO)&yed`4FzV(S-4;MlDE;xwB{vHA?npj*sKrND} z!j6>J??nk*2jB|vV{6EF`g0B!XWw8?uoQo4qP2oYw!@orf1ddQcrt1i{$32e+4~I| zQjxvweAzGqwE)v8 zpyB6%oxfDP&M~3a^=fc+BesSVIy4*E7p>$5wRBCwFR*fK5dL2$U=HMI)*91_f5&-9 zbqfm~Wv1NFpyPZxvZj(xO_7PRjj+AluPr`b9%87+8eab3xD&JUu>Im2KUDvMh-5~& zEm2nMWZ^p-%W9jm zEw%h1%E_V5E~T9m|18BpD_Zir|M=8jV)m)b-oV48s_w_aZ=Lf&B~A{qsNz;Sy2q~` zJ%PA5wzAhw=xGUYA8m3-sZ)Ns@di_DR2a43qo$MlJF7&{VS;URNzBR`n0j$mpV z8>C|Az&V!h@%snSm`?8$P^5m2aHJAbK5e+^+uAM5^*s)kZ~d1h+w- z#07X#rPb0{h|k}*b2-4T3?;KCr$evfU^iw_isSsj_9VO?R%XX+Li)nw1bjF;aN2|y zc-ix_{5#|a9!PTbMn8Y7`+{=oXw81go|>kkqq6}bF(1=2GUNtNP^e=PuzW}YvFW@1 zn8}UrUlU(Al_hRFig%L2xAjJ8MBQ)w3EB?9tt?9@*MtzMZ-};F`Sil$V#0QU0*$4F z+&7BRlntFlj1j`azC_TlO{0n**lCT6>hDb!+f%;85&H>DLo^3Ts0AfQ^{R)`@M8;jh!L-K6oe#>40!lc`MEDnxnQd<2zaofoac3{}>>E3bMT=MErgb}Okvf!l7QF2`(~4150sX_^_kqJJ+JP; z%O2hGYtaW)7eVSiD4zZveEn3m$DkAe?1%$wIdPM8*>)!^(rUT-whToyk;uY9EbW-V z)p5QVO)<0Prbu)uE{>kybFbTst6J7ogrKsxpg`yVMJCGh4{px{`^hBcy!zYp5oG=B z{2%ijRXlY~&Vv;A$*uN7uUC-k>P<~*DM#Ll!acKud2}J*4ZdjP2@G0z!$LCgBlcjK z{!9$dw$rdHhIW(s)le8xtP&5)^F&_HQ^}We_3RhT#l6`j&>128JFHIkmx%0JkmzV& zVE9U(#=xH8i1rO1;W;+j;)??PRWoBYMXhzsZ^>Y_A>~HiXyRvqp=VXqZ0{VcTgBoz z?=cLHDPX&DRi>A~SQ~6$VvqP^k-E2Pm5YwJ?Nc^az#SJfd~kXru*xNnfbGY^p6-;= zE&~C1rt{$FbJY;>^%{T|jxSKik_U=m`<`;{-#p{27tQy8qJyGT_=KNB>3KQH!Wl1$ z`Sq-{>^=O4^)yu~`WziyRjWamZ05R!EGM@FGn|6h5H zQL$a%sa8r!x0BQM&e86STIo*MF&DSj0nY65#t^XkUWo^%Nzx*P_dGNZJeXwf0le`%G?YT>0J73b~(w1zoua?xnGEyURfQCGCR zvvadbl(2MLgWV7Jl{a;6%U;s#tx{TQ#j=b09Y%z*2b3!m~!30l*h7wW{66ate}P5dF}?N zDK#pGM7lc}(s-;qF3PY8(nd_l3I2WY$iZZ4=Psl)A(-o`LZs<209e!&#cN>;qU3C7 z!f>nGr6-QE0>zq+clQ5x#8<@{RS?&GBg)AA>u$fq!ER!PgS~${!UyYV2|2~!e1Des zp2_gtFLoPMS8e7!b&?fy5vV-!5hk_KGgR^pkAbQve!G)H@8ZK4gSTE6#Hr2JeD$Ok ze4ZTC(kOv7v}e%K^WOq_fe&{|0fc{-+uSv*L;j+BF{1BjD$_v*-Foj2Bg%{w5lLfW zs*s#FmzFRTKYJ{AH&-DK=c>U+^KnhV{z||}o8}!MUW1fwvCk+{MOcUnqjAz zK!aZ>t6JQ$3?g?UsaCVTyux-dzdHSJ&M@-$3Ey{zK7A~5F^eY7Ki!_*C8}0vy8u#*Qr-K5LCJ+=a}b7AJ-Gjgnw>tE$#Y!FF|x zX6PFtUm8E3rF^dV=h6|5w);P`q6d*nD@eKRX=0ZgIOEVDw-osQZif1b=9^9v#|HQ3 z$FOfG)RCc3D6E`dg=iDOKo$8@kXP;YsLY%2Y7}82w{yc7vBZxW$NF!Bh#f zMpQm=xMeqr5{BgUu$99!5n)ApWoqfw0&{=mxI&@Q#2+9 z@Cx&3q3)}Z#!N@$b%K?>B6x7&JnnP4-QM9iqk5|iJ#I8x*^BbCNnrf+x;U5?*N@`l zx*H-c4-|>#5Tnw!1?|hMd0p$@6~m*J6%n^1q)wN*tRiQb#39yl$dFKN9G5P^VwAmD zZFPuv?#q1Ai}UlkY1Og`wa#te3LKdxet(b1dExBpnV*c=~u{Nw$w4p zwKU$)l`r2Uw{q%UegBtUze}EaH^`&9co5Vhu``c=w#fvH;CTyZ1T3)Zuy4(Lh@aR} z0<_h|_gh2YHve5HqVix#4kqwU!S`~emtl1dezab#P!eoxF~&n9Z`kNApLg7uWL%TV zXHDSxjSSzm_S)HN-(NVdnZGG4MkVOyGF@h^`P5tyOW~hrII8G&U@*eRTfu@1$!Xi` zMUaS;LsS0Th0RER^gc?dy;V6N)bV0r#DR~&ZpW#0i1^HWYxgx3|Auy9N6&T5<-LWUS}2k_!t9bsO50N`oe>CsXh{ z>=86_&7fP3&?Wtef)-%$&Sq($h@p_duDx zZD++6orCp@?wcy*=_)r=?7Zg7Z+NU)s`&ojp{;bo2uC(hC$MYIa#i=s{E+_aEl9-L z8)0w;2|=iQ*M=Arvwb`m4K{zLcDmKMT}aAHSr~Eg*pj87!$h+;F)L{r z0zv7!fz{Y4sv-JPBVyF*`eX!EMke_9dSI@dby9F6t7+l%X8<(Tshs+tZt_LW;n%5? zMEDz`&*42xvc<#Y&kr-q5Vi|rDUh{BppEi?+Mog&7wgxKJJoon0q%Nqd^HEH ztA1rS`yK|7)G&RL3knWBGX;nc$O`Xi_c~zi7#zSa9UqINQyEdA~-hM4yF-pN>s{3e((A!N^4_17O9Z3|+&;O2H&QVUmVRN;vP-ojOuZbJRi+OLo z(5R$^;{OKUpGqtvdrl_QU3N0@Pm{{iBi86#7@>r!Jn7)%Bo`WU%$F|+Js$3truAFx z!dq9m{*>liyp7yD#KAM7-70%=7zNHk@khz){m~L1URC|U8j)2;PRCa70hqoq{4F}q zEBNy3Pj$9-bJ8ibvt$${By#&3bT?EdIzc9o2EZ|s83F|^sZt0GgND_I;GRBhD@sdR zzx%@BH-qX9Nq)oZe_6JahtvsbBk%D$#wIIN3X8MwtUGd>?u#?K8uQ%q5J6eHx?5DQ zT82L9U^g-xu!A!6Jngq~I%|V}#TNsA7`|Eh3L*6PU;i1tb=>Nj_u}TS z=C_}E#N+#r>P+Pi5kDF7_S-B*pYsG*6xNu(E*_@;PuZa{fL2jp&{FZ5RZ&306HYadflDEM6Dn=)1n(``*a^74A)w+>D$B!=IOo?Igpq=hhDNj{9Nl_k zu6oSZ!I*$m^P_^y({2)l3Jz@>Xv>&7iYIYBHs2GG4eiCt7_@K*G7LpEQ?!}G&jB1+ z^Rxec&>}XgzSI~Bp?(T5tLgGcp%u*kerz* zJ@*n~wH35=F2F*|85?LVpGOb>jQi}YUIcLnd2|awdUHS|yn~pcsXlRP1|+5!I|R;Z zJTiH(o~j4FtaGwujYl1DBmdO+(OusTj5?F;p6>1?4k<7yxGN{*GH&mkNhK)nzPyb_ znPhpXTK>&PFCjh>&exVV(+_uZh9TxXRTT8E|9feQV!>Lzux?^~B2jEbCwVVB?&A1& zR%tw#q9~ouulF9n;iY6;800`4kh%Ci02`0Y5wtzmnIhWhWkUlEh;b|Ih>p6?NW>W< ziQmPY8w0;HMO}3D+luF#=7#PKJPQy9QvUj}@>#Jo>8@{?bj`@2jb9ufI|3xZ2f2K9HGyVK83$_ROm6jJ94ze_dq z*kS+1v=tLcK}ddHhPtNtnWpNvO1$iO1$)w@KOBUWcp%hYX$f`frvK9Bo0a$Pf~x%7nF8it9%;b&j78ltyY^$7F0nPT9v=!uC!$iy{k zG^3m^1Q&2oRfc6=6&9oL?vfbziMj01J);{{~2A(yz~h!9wmddt%2V z(QyKl0srDm zy@@&$rwWxch?ru$-#@PnHZIy>T@6c0c+R_oGPpp^mbeJA2E6QBkG3_{sck;$7+kC> z{GU6vD`Gu|!o_vFon9y1ZM0`6f4lZ*T>MH^Ls8QUnAV99Ahvn?vtbcMduRNc6#A^D zv88Do>>CAmDrUE{3=009^GS-tLW`%bsoXYo28CEx4}UB!>bAfnVU1rb53+@tUcYr7 z!Vmv@`yTmG>DuilpBbgE(0e*`K)a-6`QhX9(au!!?I21qBK&BTAjb;}+@E>D{kwa; z>yDPoZns`rSh0qGk)5xX)f{gpm6U}j0hXFIv1rBEzlB2P9`q5*jeZsyS!$H(4HVm9 z!3ti9U#2r7XroEUXpzs#c_#2Esj1i0lUIg~7YfS&QudP*VI#mNIhnv9ywnGWkFIq` zJjm8=_t{diZK;C!nk|B zVScufSL|CP&9Ih)3H`SjH9G@f;f-6s8aT7D{E32^A z_jKDwY5sD_WQ1{?d#{FS{{1MfSB>n`ailM2*?eb;0rPE|dI|=iuc&ZmYIoOfRI{rZ zKKI>I=FP&r*O?sPrPSourD%4`mq4--@5>yBZHr7D@?2(52u@uXruIIWM=mcRo$2VN zmj11?;YX*EgHO5lKla}F(Nxa{!!@=i>TrfYOHL<2xWoAJ_t+=8@B*XLt$Zc&*)N=# zRpya_Ggy60wl{tfK<|XZpm#vQhmRHaP>=QSa4^$!8~M2&d}TVQpayq>UL}Ym0-tm( zA}Geuhu2r3DN@^DRalOwexnh1r4`vF!YBwM7Ru!JmcRM&d0B)(RwkiJU^cOAnEb!5 z6!0G)`nst3AR-`0smz&l*GrY>LNpoX^zD$*UdVniFhHioVDaii8Hjs@?87`H8B?L* zV~C*Zsj;=$6EM|V;(14_)Z|JwLhp2!K#Yz@iqEqzA!b{_j!|pLM_I|wo=>{p#5-Xu zDG@*aC`^E%GiuGA?V)$;a%+;X^e0qjL6=5}W5F|#856a;83C(Olh7ADy^Xs6t~wqy zZQop4{8^}Kaeq!%z<`XV(8_skT~`9n@F5ms4!!lm@e-T>b*2gnFi^k<4fBc7WLyNA z8jTC^U$*CHMe=C-*lthLx?%qtrFw~?$;-Pt=8yRYk)?M<+KNpv;vP^rR8?7az}*l8 zFX(!|WXB_K*0Gh9nP~oRUkr1OLI^)>VJBf+3?=5*RhoW54;%XCxjjN;ETq}njjZ3E z!;n5ahTP=fpFZCHRv~E{gFJw^fMo0wA^ye>7&V;6uW_oKKb|=&tqk~&O0LOEI1bBl()`0l!S@=y=awXZ+~ZjO!6x5F&`AXf8*VL6^>1k zzekkIDrDG?@5$0L{>@Y>WcJ;l2Qi!uAYUb({AFT1^e~BR5XjIfSbC=>Q)jREnu61y z=TU%XlRni&j-fD8)W+95;kQ}wGkBLogVk<^Y=2gbFA;^AXyL*Rpq#YO6+7z#Av}wv zf$SZ~BS<4Kw2Sa!3~$plmfbjq^p*gNmeip-l5=?4r~mZJCCtuNluWBZ|MyF5pG|s> z5x;nfu^XD;jk9~qYtvT`68*+s8zq4?f#a5rmRH|5qL}>CFkM;9aQ8o8u{KV;s%OY1 zIfI6qa4lk(4Ih}VHy+|E)mGEc^b)fd@6c-gcY2QIqqumy&7T@tUvJ0wm5Z(mRec28#ekPhii zX;75zknZjlknZko5Gm=B?nWAEcu)4;&pp2H7h~`P*1E1U<~-&R3I!=c6od>g9g$FY z+7bf6zVdo|?^C>&7az#?ULjLN;Xy%`Mj9Y^3qGkD6IqgII!rm2%0KwFrQISVAxK`Y zg|d`PucEw%fzrZrQ};6CvuUY%WHFxZ*B^=%KN5s0%bsF8>(DB&jo-UQ=D@$v)(XeU zcxo#4j=kNX`U_s2Y%OlXP)rZnl6Bv;zHX@sidNiEcfKtpI&AMZAAdg0z+`yGLU38+ zUQai$kOvko=cV16Z~aO-2L_qRf3OWw`tr8myZYK&2CKI~s78GnZ&#T9$$1 z)aVjCZO|7PH{LI6N6d{z|BT!vTt=v<2t7bbPR=Az;APL(U0#-LW4HhF0(g-qgrz4M z;vt4neBnw%Zy7gUhF&N;gUhZD$^krh~)~5r+F)p(q z%4U0+f=+RE;I)c`f|8@Knzc^3uneF~C|4!3G$2uG~&A2pDBuc_E(yBl#z) zkT*nJP=3Mv*1K~54bLXs!nb>WvG6ok)G@7Vov0PP4^_>iS!@i>r@zN7bT$j`Z_^_( z9T$5Q%UY=V2RTWm%$8E0N9c&llZZd?j}de6KQS-U($cC3mYg+YOaYI_G~m*a9EQp- z3k61XSWL(7vzDw^?-5$(HK+Bkc5=;UCO)^j-=-(qh9%b0U)4J3(b#z!sMAP*Hl`58 z2hxNGd2cAW2xSM90{Vd^*CW1NopVsu4BFA|oE*NHYqH-ZH(^<^fa#&(HCtKK58vi- zpdb`*ihQiTu|y|z09}Ty4Yqed$H^@i$PET{6`}ks=>8uA^S3amREklt3d-pVLM~G2 z?hq?rK%^;KvZm8u?Tx;OhS5Xz6xgiTexl^tg=~4om~)c7mmScH%lr+5aJqodb#5EM zD53__{l!`viyRUj`yjwgEhze?uQHGrSQPA?saI!?MNDit&PV%tXhK1M;^ZZO3B%g1 zw_X){%K=+I2}GMm&A_$Y6<`St$*r=@?b_p@{se!9{kFyM+O-tT+}JQ_VN3ZN-H3i!Plw9 z9+dukSh`Y&@XMt2B_4Y!^ZTCa6$Rx}ImHVq-a$dz=Ns{GSjf^HI8vie%`LOSUBrpD zB+9Q7lE{5FxXbT&?&??MnAK}Cyw+q7-W=|zQg1h?7h6@I!ljxNqgYN4WD&8KCJi?A9trw-R~S4kcmtC zF$3m_nU@D6HhBbyQ&LrC4I=HpUsMS(zdVs3^37^jUVUAdJRz93i~Tl7EOExT$7Cp( z@gL7Bg^C(7@4<(*Wq;R|({6pQ?jM5$N5#eX!~BD@@G5{g%a8H_sdpsJf@`$F{+Qfb z>!OU*xpxrptI^AWjqK{bgQ)k%!>dg385szI_!^C2Bm%mcat!rMX7})chEg3PbN6IK z&pa;-5!An5TMer%{uu7|q-2g69f7**MUn_GGnG_`3+U^x8FWFq&a{Km5VoqW!q=NS zopPzo_V6p$&+%i84~zKWJpTX>BHOv5n#m>vOT4clURqL*dh@q@>(_;pcLp0)GO5L_ zE;OqPJJ-iM9vklvuZ`s+nbLZL>GlMO=lKgBzdfWG&a0X4cb*BsYw^;>#mf%F)y+ zXtqxh`3UE1Is^7u;rapZ5$IH-8URrILqy&y5+WMlL&)c%ayW&sFubB0fXM#50Hn|< zekgP-kK=an7dG^TEhD039tkNJO7s(6Klr0hpERhqpYTBCEriI}eO{$TD;?L-^_(GU znv2$cst9FI6?e?G zGv1WBgZS`-4p!jIbt3(vgYrpHml6Xv>drAqOJdZ~i^*@4p zyaoONgUh*Tj<8s`QgX;rajsoe4CHIMc?U9y%%wRUT=v&h^q(tx{rEGFMzUT?lEUIK zXyV^}#*6J^fgj<9%z)HmXIS_68$WZKPvKmd{ZYC(H`BvOJO~N_mMjPHo%GCCmH%z z*d0Dt%$xey?`46NRaMX*uAO~$G~A2w!WSIek?qIkz6H)7g$;hr7(im4!My;^`77|n zhz*qA7${2U1Yt4B)kR>DoD33jLScNg=!8$P5zvxp1uZTJVCBk|kTcDNO>oJn>834BwJ-d2zk zD$HkmsX_*67-NuE;Q>rrQZK|~0PGbj0#meRF+1w>e;^6xKp{up^BoT?ftd)^y^Av0(va_Ri8SF>??g|0 z&>U)w3AZ!VF<0r`83Ixpz9_y9TlAOVu8nvqqE5KDo)f6P8v?&xU*({Bf#HLj8q4X4 z)`f9rMC$qJRJ9=X0W}Wn&}M{!55p!+!~4$FkPeO~=F0}hWa80ScHl!?B>ulDlqcR_ zFh#|e33HxGD~p%|Dz1$WF)VN=;9$?W1W2`6Uk4nPTBh|>r#_zFg_J%YL~*QCmFFlH z2W~~e*T3Ah`9b~oT-8&|LcchFKjOXSU*XP7v|kggu1@%hXP~S|Ovr$K2ivwB%cu

__XiULq?sVXk9{VP>@ zVy?3at_bxpb^YvuauL7=G6;f@85ZA7M(qX#+a9tiSCF2Xi_D@;0Okc!6(PP#Q05o~ z4`vWCvy^xZnH%FA{L}LS)>@j36frJcL5UuIXIeQwTh;45G#8;nfXsS508WhMz2e)8 zYr0ju%ll3Ql0GJ?1+-<6Ju#StVW@82w(0Zu9>5Vm#bsUrgzy(o#z%~E0 zDf&q>uynVGPEQ#0Rj-4qkiVe3RrDZ7Jx7Md`7XZ+UB~ze=1{foFM}kG7ndgGP|@~~ zAN0-35v8rw4fsrD|E8f$R)KW60-}%6*wNc3t-_5%|Em)Jlk3nz2+LsLmnT3BupICs zVqgNM%uT?c02UJanN|>kZfSWry`wRAJJ$C!j>?4WwZoo@iEO6N5C}|H(lqIGG4s@q zgAb_HO~)uJMbpE0BZ6_bw`W>Pn-88bc(^drDl2JE-XA$ELJDkRX3;23BRVp$P_GF1 z&X+S~;kMyYDa?Eh2i!|ri#n)0Cg0FDdD~M^GK1e@5Kd+o!cs{C$dhrP`xXKA5#4J& z2Z;k7d3&k~sLZwx=Q6tDKJIDncE(xF8fpVykOF-cI!D`@alX zK%SX{>dSeb=-6TsTT$j)F;tZUd@*Yl6l9We-GJ@YFI4FRKkZ@d`n>(s0@=lYyl82e zT4^%|rd@^{Q=y?Bl4ImK%*L<4x5 zw;&+cBL3SVda*e7gkal}aWFE5X)jhOq#sR3c`o&J*iPTXiD(2ZVJxI?!7;dNP*wkq zgfV_y@jLwFd(2awxu);F@88wtuC6OU2!|4r9TS>p=hYM$HVEM1NgROz_+fC+NECZk zcxCE4#mo26_AGen3sBT}XDD^h2^9HXRmPJVi_!|Bub)d>6?2(teH6cOCB07;=) zA(Gw?-m>-dB{y)^%>d6+9NWbqXWLEBbt6cu&ABT-Hs5qWeuIowX?gj}c;i0jr?Fb` z2vb#bu4w=NdxigSoxLzB5MU>gy!bc_6`p_B)VStM1T!Lxw+T7nFw_ClkO6ZE^X#o*L|`hQLPsf|nZg z^h1y$wqntA-ibO(AZUodREj_G1_w4%m&Fap8H+}PcQp_c{mL61zSS-s`Ci%wAB~|BU$GI}UmYmtoKdq+?`I zG-~qn+SB}UetP+C{&LCU?MK>@ciHMe0elSoq4ocLPj--u zL_x1M;My?;{N{5zvH!k_SLl2fu0eHdZUN8?Y+40DrDkv)M%r!b>tCQMd|&K)*;HO$&inXr+j38#wJ0Ib7gyNiUdFuLepi$aetSEhJK8Ptx(Cm2 zZaaO9%)rP_*m_Xp?(PoyQvQFROE40~FiP+Qu{#9>z>V5rRH(C5K;YdY5)%{oXOHoY zSfOy&L|oP#6)2b(w~)?Mkcntsi^96FxQGb*Oh-q@B4J$ye8(UG$XX5z{N*^PzGPhT ziCL!?@-O@qanuN<1`=7yUpS}xp?7pQfz7QMnUTRrs@A;VKNVKrS52kcmAt-y`i2f$ z3rvdoaP)Asp&8Zy;V3r8^nQtMOa`XdbRpX<(j5)JEu*Q_iOwm9*W6V-&A}>}+qgng)OL5<;(M5`+ZaE#c}!G`;X>nPXy*D z8(53>Ut*$Ib#-;22nh;){$=$C`)jf&4bWv|0Rwb#bs2i)M}W6iZcElk*nkc0fgTi# zddbkYJL+evocm$ExT>KR*ceOz13@AIr~XnN6kuo3A-%GYD~(02OK)NU zk?ImVa}k5*heHVE@aA|OmklP*wy!uXK( z_?k!<>k%^Rq4P~W?c-rKI=QDW!}Z;KoQ6Qg$$@DsauT^0ycDC_UVJ*OLHIcIWA{&B zZT8l1lM$$^SOw$xK@v8AH862zKm70D^M^sPe+mWQp@iFPG`gs5l)#yQXUD!3)?r}2 z1+lBugKpX|8%<$?gcw1w@Wvoo&SX4C5||6{m%ajbZTzPW)E_XMfa9;(G$!OyS%mh| zT0^gc@Vic0xPp|XiHlPm96$e3DQT$q%eZ$cgGEC5SjNlv^0_|Kq%ZO$ z6Laltx%rVKV<71cijz$*m@y^wy%VgU7viLHNDAk9!1BUYna>xSL_4Y3>Bw?o@k9K! z`>C+t5u8Ft@4LC-cgR%ro^R2b?6=EMG+>9ZDazEf?lS65v;M%~{@-_h1Eu9J+p&Zy zE~ciYGKJ8=r#Sxv5lTa`SG+HvZh#NZ4qz2lbuNl42zh~}Z_WKm)e!3Ds}*)a4C1o?Q}1t1leu%S=r%9icF(?k)v(pq-K{|rej z3Y056Uw(m%O2i{>lABNIzd7;-aN3{ znVqnX6^zh?1=?9W@2}z=t4O23Hu{Ha{xw{L2u#a&C^c+b+1>To1W=3@iUwnjTER$rC{lonG^X-JI} ziSOUOD5?oV@ca_6`1@ta#%fHIH9;nuzZSdp>g4SYafLds?tO3wsg!DRG%+_&#tUB0 zf~04l-$5XQ$)zcP|6hCWeI-f?s4Zqoi{0w))DTR@^RN>#O{NNm?&rkX$G@MnenRKR z(&lnUzrbtzHlR`>p2pv}8E_s}mPGgEkTyH*mHO~_l%?wK>ixUEger$283JAp2JYsD z0XvO|CZm(A=6W9}afC*W17-V8@!;u`uHKv!K%E8QW;SY=A}c`W$jaJm`vxI#|EY$a zCIGx}N{x?A4GrITzpWy{Ag6$vNg*Sm+bogPo_HdY&!O-YHk0n6<>#>g%uT6`;fp75 zJspUb<*-^%UKgT5p`*B~AOF3cy;Z_saDD(sAGxRe4L1Z3(h&yGE+bCz6+-b23hUr5 zicFa#)=5xhAa52fqS^}m-?a`}5f#YFVkTv4EY*GZK&YkX{v3QVTV{RUo#-`ng~#u_9pL6^!W|*bK$WIZimlUIQ;}*qk0z$$g6*ZWhmF>%L z4!BkNj4z?+Jm4lt7VvDV`D{5|f^Wc^z)Xa;{Bi?%H542ef>nZoM^n*nl&3W=uYYt7 zou|J(f0O81U#^hT-0Tq2enxZoO+9vdTRti8%dZWxi8l#C?QbrPK+*2)C*mcpfEefy z(>xx4`U>T?=40;*rh6dkW?iF)!*##e&4=i+PyjEO1&oj}fS<({hq7^Vi`kSAPKzyy zx&N|$ znRN1aQ~8{{-Ry8fIf$9NN*9wGKGmcJ@&qxH*m8#Gir2<+q!dbfc&|-~3FgsYx&2`W zjq^9)T9Y604-Ezr23F0B8m3S30<&VHM&0A~4`(6FP&`9ou&uHl2@c|NxlXbCc!V8P zS_$OD(m5S2XcCb8cf|hp1Dz}g5VsIz1UWx3S3HMLv;ZLBCmiZW42`we5i>SKNkS7of1axsnDkRv2quR!#fgnhE8~;wT*Vc_ z?&nAi+@NsSInJjiD}e{mfK7hQMblsW)zwwm*tkH)-B+3ya;UCaq>>3jabmdBX|j$r zV_$W}oO59KbLK1fRQ>bBKWZ z>7Qz)61~Cw_ASUF4XU%4&vRHUb-FEzLbMN z#^S5bl2kjOfuZxo_=rL8`;FmA87g>3-hniF)x#-S4}uR7uGcE^W32~$u5ND!Er8=D zD-=Gfueqsdpr9$J3fcR@n>&-OL}VW%r=DH=Vgvznt2&C`{Q%+3n2zt+|KWDtKSW_# z_Cca<4j9P}R89aF;e)-1>I?!GtcRe(#a5a|A)W6!eG`Lh<09L&ebnd`nA53aHs$SM z=x$DrV)TZcEH^8i&+zenl|@*S{d25f;=%sHc(__eBLG>x!Ec6BO51X>HkU%|tZX+K zgtEU7=S98aq&+?`Ab8giXUBVfh?;!r{)K?MCvN^FZkcX*>O$5&X}0QTJrPl#K(yNez%7x-8@&xMcj+re6mQB{)%h+%?&Esfp6YLHKjrqt{iza zyW^&t7YM6l1~u+o)J-ic64c$PHRiKb<4nqW5&t5q5;Bn2Gun-|xhihJNm+GY>iHLS z({Tex{MXj#gsmXol3BY^|4n6-lqPFa7Csj|;Uqt4H*-phIs3_**mRS#Q$6F`WV&%o ztb{A5s5Z#&mXuV=6!!nkKIHIJdJ}++Lx3ts$NdUP-Z~Qa7svVd_!NMW1j6aqq7Qmn zVc=R7HJ$cnu9m{d=`s!lepw4s5Qdi@2qfSB0eRZ{L*u#9yC5;N2I#qw8J<$U!Nj(Z z@f8jZHTD&)R->&^ljmcbjDW))R;gaQZqr}yI`9AW*Ahrj2m-88eU6H8gvDKH50f9F*w!$&)^5id z-}%Ps$D~KbJ35ObV*KMO#pTa94pC8WiW#PEn-DGTou8%7yJ<$iVmLnltA19-kteMP z0HY}3zLAt*zoFJY!NExglI$F zcI82s4}j%dLWxR|lt!zH zTQ)Ws@TT}7pohQ5^OcThuJYU4ay)r75ehzfP^-%R2DMryJh|hKiw4y9q_q?s6e*Jr z$Sb#i-4BJMA5Bnnnxq_~l~2!W5V3KbN3O|Sjmg~RlgTY0v_x$vMnr7m z_*!a&2vu+y{oyR*c-Vt6G8nUxT{X~68vzFysR4!{<)wMr?5^c8(SRy{$Htn}wMk6D zyA!7GvCJfhDvK^XH{_EaA@W$^_}*eZaTj zX;%ng2xjPv+GsaB4vAjxZfbd%a_kF<4YZl8;^QQ~NE2vG98SGcfA7VjUQ4Hg%p>XB zteS~CLcurmYu?euq2voCqtxKk*FY&X%lEhScarlna=A}}C!-NPXbz|zlp=}wfLABK z?T&lOGHyB>Tp=vDo$~^$l8&M=B%VJpd2}_M$!Yy`Ut@?WFrmC<{c@4X?!$?2nhrhA z!1 z15ZaKGE>XU+!Czg_WtkOY^}iSTcuEm%DEKSpx|@#nmpBR9l-npD#V+{>nwce@u5He zn^l<0XoH?z@|+28E4q{{)hrzE?$+-9-sGnvbza=u`pi2OfA^5BoYu#ANzPrKPE?OB zKSbVHwUy(i@&yy(4f?}z1QItvXbm!h^vthWzvf2dDV)Q#9tL*%tFFT^n?>XGQ^~nj z=j~teWtPNhprdbaUo_vFbQS(iK5T|NNt7d>?oHo5t4J%doXeN;d7@AxWe3ESR{a2o z#14=!pMyB%Du=cgZ2X@fh9NKvf*(Wh5}&S!uR^52GTgT60??17E(IJgG%7oj|JeL? ze}f?{^2bFk_q($cCRXkMEBnWvVO3Wxh>N>Frv&k_)u#ge7at#Ru^EMP0Vy?z1p=x( z@u$4^jr3XvR96cA{rRs~$7BHF2}NJUU)$PxZ=D;s_7V%dBDtfoxVZR^gtUnEt;Fzm z?n1yU1AYq>^vZc*GV5qhm5K&rWjxq;A6y_UNMDr$WZtaD3@S5|Txc((J3ds#6EC#M z^NYFB885>(y}~kKDr?U^AKoqOiE_G-REtEOWR^jAoD*S5Yl|8Q2iEnY6$afKh9Up; z=V4yYFgngK2IM6~e;^l4>?qUgo#|wrS6g-x&kr6gS9XEw-O*C#9f^hANID;xIS2>m zq`OrbP3CnckYmPQlo|i%0@q`85sx|Ou50EWm48V|c;Cs|tBgy<^ zpi_>n4^R7u^|VFqmnvVF>ksvrHvu z95q;y=)W=QZx^*Z`xI-}(08O)7ig)|AbIG|OM7)JYhqLj#9RGBmrwd|P#a1xy-ilx zdihVyL+A%nvfk;6=)J8T$j{cVe(e4;k#N4AUrlzKPGeZKKUi$npY8d?a;QfpK+2=_ z($*$(M?~qhi(E?P)Z_Jo=UUD#GO zF|$CKkDfvyr=NqG9?`!;UC-Sp@wEEXb?Q&ufR8H8rwPAyz$MvB^C@4(A@m%XEUfEY z6E3XVq#FvI+1cwHwBHt{tmH@PfG0tIvGQAZ^xad{ZOsdJtaJqib33q%;~JwieSLKuc#YvU0y!_3QM+uK`>1y|uMi>kSfKTp`>sO4t9>A~1y zp#9fPSO?Rpz~&+jo8diS>^b$!ErHt1I<|Mpxy(~^b`cV%4{l<}Pjp2(cJ@h4hbJ`D zTZ7d4Dby6I-s<9T15PfPxb$5=X_v-JhOzneB`VJHUzNs-RY)a2CU}yVk+M|g&dDo& z&SNt%>?l1X)h&S?UV_khNhdvK4oo5v)i6VXLAw2*K4HakoeF(e`5zqi|4eQOy#=Gm z#rkGu)ZaSX&V5K`&BBIHUEB`JL!B=G1$f46m7600w&Orl^wQDW2#>pUlubI7>72fR z+?TS$Hm5QTU60%jq^VpD9~3b8>igz`d$99`gZF!BPNQpoE-Vv7DJzDSmOnr9V5|KVLeJP*FR` z{9dE}8nMms6e;!b5{1wC{CCz|v?{{R1FEk-%E>~~=x4&At4}fZR`mt6n9M{i#&I$; zBf8noyO;8)m+P|@u+PTLOmwrW)hV2>Ul8$#0&}bIoTOb~k)@+{6o9)b`p>yy4<)Uj zzzP4JO5q%RwA3WQ)?1S;@Y5;R3^#I5if3z7z{+Ni+buKWuHv5e??7~7fXwgrpli zr{)@Pm5qn45w0fwMCl{7Vo3gh5&JX&%LM)I{svprxy|K{HFe%*uA-<|oH|p(LIn5k z_w4rPzw#b+AM)eRhWBM3Wpw0K=cZpbIueYflV5~RB%#uLay}<{1xEpL{G=b`_5Gu% zQOl!E^-p0b^xLFO&B%8_5%iK?y zABJCW(O0L^WWcTJS4PyHMYNAb{^N{7+dRKMEn_f{WgM3QLGg3B1hTa%v$EqyV#9j# z=v{s>cf|Sbd9)t3^@bHPelY_la-BaXgirn3!V1oEt~Bm0IU3dTe+S}~LQfWSF>fKE z3H&z zJkqD{O#3_%ZZ!JSV8T!dPCDpn1!d4zFr;Kg*Q&rvUwIADPc`{5ku=o}hD^J=Dd|h5 zi|I&l1qK&$@?9ijzhPx}gp5@n3xHNRcJYd{k5l}G%jtQe{RIds#sOZ1W+9_b%HmUh zO^a`Ac;4zz{Vd@9BV@AsG)d_Xb$Vzq%27yUWLC-d6d6bxi5))Vm5n6f&uiH{iw)HK zg;3JVNu2p)Q(F=Bx_DXS;;&Uyvl9;ed=1<+&ahbeH@E=*>@BvHQ<5;$S2>dUshX? zWzKDZ_KE-CECEZj8 z%G7-gY%ZIfk>2r?w(K>cHb#Z<5^i-LLwV%g2*uD4n~VC*@$G8OmsL5JYF81Ju^i3j zWP7*liV{p5ahJmB%!CJ(g}_(wp6#shuH~3I?`*|rh2@n-+i`2Q5pcZqWnUP*=H;x& z==x{4A!v@?@$Dm^J;XdW170rY0Arc`RozQ=K4X<04H{Go{0BY^s#a{H0m8GBc<1!w zwqLp?*K-Cq-26Biez8oe!bJexf#|#k{owcIF6t zEuM)B3rx4K-0tE2%(KeBckR6(qmoX{&knp)20EBR(8T${q&(re=uS^affGt2JUEsr zp25nXkEFg62jlX@S67&zKa~TJ;h1T+=>(tM=hJg0!QIsFg;K3@DmB{fNUnIo7+p;K zrP#$Wv4NZ^v;wNWBYxw+^F?|*9%7=f(&7FI;0v0ROTcJc2^6J;pcPZ`s%655y`Jau_Bb5RsWADtV|bdOq3{x}D$*{kdqq zS`??JDOLS@JW^(Kjdh;O*foGWAm1v=8<~MmBAa8Lbubu|h(+y_*6?{q)S-DM-zncs zkVcIt^Gs`4;Ttq#Ep99runK4&*?>mCugIz0`Px=d&JUG+{f5jpA6~%;G86& zqfJ{9N&VtgJq`XMh&Z6gOC!c&vSpfE$4DWJD@2MKXoL6y0CGk$|yl1a}_ilrX% zRzRFlEKVjkWplLD#Z;qP1ln zi(mSuxNiJLTJMN@H?u3XqIlY$6nKh5|K%%cyH^&w*J-%+wIJoZ`ropARTg2pIJz#orCWM>UuKkr>>%$bKhK^;ZGVg=%1p{B+BZ z(*(_IFQq08pOy6Ccw8X(BbimWq;uvV5fuf|o^;ypPdYT!TxR`VMo-XD?2Jc;5o01; zX+NHj7w8vzW;U~C|DKeJ0%lmR^|X34MXb;~B_ytY;&|1U97bwir7h8lnsHQZyD)0c{%?M&k1F!i)=JMIqTPe zwxQU8+ZPmJc(AVM9lWccYX%0-JY%3OkOf47cr4L^R~RSXBmU?xaRU2fl}g8V^dAZf z*|RG0EzVfMrksgO@6)=UL8~!5tUo_B+Ay>qW3Al@wMta_DU+u{PMbnkjnF!E%0{nV zN_SzYrGTvI$171dvyV(r&IoB6z1Mokx!2l84pmSNRa8JTh{g|l8_%6dNz zFtNfFBEiyfWNtu6$V+(SzjGNlojWvCeEM}DYck$AnpQeU$IHkXBywCj2n49lU9Oj< zxuc(D$P)`>`Lne}`$pU_32=5c81HiC_+S}j#U>gW7C@LeSp487S-!}{Q(o3vEhjK{ z4BVgw<9T&b<%0=X9F08j?$Y+|u8C5lywY>H&dyHAA)R{^GmUyDoz{4|oAj0BepyLl zbx*PIT@RabqNZ8j4bES-{(QQ!Mhw%Uv@6r59w6r_OyGuGGy2 z&_-etCKDC1jXAz1TlS!Y7b#Y87BSD$48U-?0_KRlmZ7nsVbn;`(>73V=Q_4vTcS(} zRwXmnA@o;8l=!*UJP8M~?fFN`qh+yIBL!}D`xh#KEy&1;cS;ZER}UWt(wLfM+jvz2 z{~fOV4u#>BM<>L7wM7I?ZZe&s9|~HVo?@#p{L5KmaKcq|{QgxYiiLdALUzligVqRo zr@Ibn&9|-%Jn^@&fo~@kg57u5JtkzvFKA|4?G#5c&rHS(BD#{lAVcg-~4 z+L011Rx2@Di}R%@;+_RR5ehB&`T1|QuAY!m!my_f7m@aArJ|}8Fk}1S)iVTPci$sl z3T{b;QWa=3&5~O0PctYh`r?}xH55#Ti^)z19r1KJ!QRe+3surZM;E zp_3b9e&`{xOjw5 z%WIw-pmRmI4S)Gq2V&iX5w8SCtoWpHol+;R1k3qaq?nS~XE?k6B{FcKz<&J_PRJc| zb9p2W_#p8E3Zs{T#K<2*EQ)C+h8@@nWTtRbY9GWmR`DAZuwq{vEH?e9xLPU=wV)SN zEOfj0%wE~WfyCn!bhz9~4aa4X(SxPvR@guKIW#DFTZ1`n{&pK#No zqY7aDlVYDM=Fo7`MM*Lba>0!IqR1din1J)0O9n%=5`;lv4EU`|*O~3@kK?|R?Q`b^g6C?R%~FMLX*ARMq)nR#gU4Va?qUV0(bU1Y?c{ehSRkkHaN2D@Ijj0 zt$1bob)}7m+L>`6#KsEf!hJehHS%cf)*@)g2;os&qk?RVZHBFw=OEUaENF36Lm=`m&fuGNfui*VQ zkIKWJ5G(*Uv`%HeDSauVq+*Q;;px}mLCYP9N(-50Ho%0}ghjF(8F{$z`N?rT{o_V@ z^VOf~jlLo(ijX^C3A+2UvRJ9oM(03pr=(Bec=$q}fXkYcL4afOinNw&<={XD^t^Ei zqZzK=ybU(#e@E`sEkE2rINo2wbjBuWy;qP=?6hw{#l1{kYj3-_l%*ae6I2Y48ID1o zr0ibV2xoSnaCUdlMZGbcwR2U9eNa5A6~81N_j|i8Gq=QHmExxR>C`&gb>?ud*sOPJ z7kI3X*LElF4s=S}UyK^L5L*s}hxrSN`u1`GkmZSFawQ6*qDI7HAH}>5EY9MDV!_;% zngx+Tx*>^Vp-rNxXU2Tt&*Q-&!gpoWx!#k&C07KxkRT*-+7Ke-kg~GH+8K2fjlm+b zo~3CW40Y|EHpc@UA;yZgj2bZT(d6_J|*;c)zH>ThAh!E|k8TKC$yiHU7BdJnC_?T20=p7`75b z5n=5NYG=FXnXf)Da7nndH{pzDM~S#c88A^lF(!W#&gi<9rl@L`!aRRLXk26;h=RY? zB)`?uX^RdOMcg4&&k*35q9sgGeimu9=s+@md5m1*5J;r$biU(ox4!I=WX_q^?(_>W z^p3}9u{h<9WBWkKu(oK*6l=KT_hd5m@WR*8G0)nqQl8&N_3Z|iOKu@d$pUAQ7aY24 zTW^GF6=Injc2w-^G~pGdG*vm%C+93y7t+l?uL|z4<$Hb#6b+Enw6Q==>iU(eRV>-jC&G##!=$uX}0J$3JJ;LX-t z+2A;5Leu`u6(UTRk>m_gD~w!$UF|_-FR|T69am_$Z{fV3dOIEAT5br(k@jysx4^fD zS(0bNr82N%-OK_5bB<<5pS@?kNPo0&{N0$n=zh>%qAzr?guHjSl=V@1f9{L;{A$MG z!yn8hPph3E@=ke+9LdDV@w=en$Xd13`5z1Np`}{0abp6%2~h(TzY2+ZFKI{$>~Tm} zQZp%*(_+FwNfi{~nlf6zOK$|xL=y{6oI~X022Jc9+x6*fhU)^Zi$etM>%`*N9*OkL zW_N!VFDs*29%;77WFA+gn>Vcxes?s#Pi|@v&)<~Zp5Ng-tTkEoBy9an@YZ}s(>2X@ zfUfLBB^uWmtTk+EK^S&P({S!5X-UJIgvoQ=Qkb|otOzo$ugCiDZg#R4MKQM1T+328 zIQti?dEmI2GJ;96l>an^n{T^6DE&sV6EB{ra9;g2%3<+g^;cM>XHycNTQ9dHyg&-O zl^HTJjbKPl7Ymn}T6q|6xJQM@Vscj!!I#GQ$YpmWF69lZ`TdTUf95>f&FsF&))Un3 z{cac$|6MGt9NWJvj?G*-~miZr+V)I`jBsjw>gv}R;KVMGv zwB27WFSAdI#8R8?$_m54A{zH)Q$+HPFHT~fgBZIy;ORDU{%nE9;br>SJ)_e^)qpC4 zR`y?6WE$q8ZJ+E?&05>y;cJ(Ka01P%N7DiGc9ZDafT33(5_Ot1(4iOMWHwPHv%k&% z=^q(CUG~AmNzbl04O;vp z(oY0|MV~H#J*4*ivgWyOZ$4_HtN7x_qW96j!vE#?;VaLO$Wj-%8qB^Y80}}o_Xt1TdWa2UAl;i8_rh5d=;FRNYdW_)BVmqcxp;@@kspTE0a`>{rKM% zzkOcJA1F~?6(&yYiN#VzpGs$dk%1%7#x?bdqMS|Bsj|ry1GZwz zsf8T!CXMkLXXRqeM$=VrX0sgZJbw2{GquP)YH5-X2CE8UE1LlQu7=n33yaYm9{ko>j3}M#gaG>x9iq;(EK~| zm^~0;&kTrrZHIa0V2KX^guYpV(;5DqVJ=8%nFI625D2DWX}%EEzJ&@wk|r(V@G5Ts{4Kl`+)1c2B&3S>2LNY`!VST zMLvyp7GAjh?o`@oRrKXoT0u?T-Rff*=HT}t<1Ra`9jFc^7$| zw6?OMxLvf05ewviej1)6ogTuf9@%Wn7FeYbEhOQp5F zPi}KUCZJ+v`Yl6Qo^J|ZX|bcOmEdGn`3}==8&mU_&10|kQ^`i6S=i`!j0a`0^rw;> zmV&W(%Mu4$izGN%{#~0Amg5J9!%(Dxl6W7bEu^Rdv*3Yi@X+i#Nm7sC;o-$J8?pEX zmeT=GX}C5h*h5wa3SPCKGG_;%gfxx#qj|h9&u6GhcB321VW`_nNDR)I<|1fl@q9c3 z{(P2Y43(bgqQma-wJc8UI;@v-Z;jKdchtg>nhLwFdK^+_NcD3IR7HEtdr%_RQT{Ju%M?BdmpgW)StQ~!r9!vu3TbM4@YyMQ9ilP(dJngWE)_%$lZti5 zV_RvKH{qH1I~VV$bqCbklS(hcw>xYhwvt{j#xc38o;JLd>2`Mmk`;+z&wxeFOSpV))Kq={#h8tTZOrpy;ZblUxD^(~Z-c<`12BtsDwUY( zI;-e@VTZ~Yv#hmMf81S`qb&Z3Ke6Q*+$;tuOtN3|qfn-7h#tmev*M)$)kEV~j6)1s z`i!J;hY2)>+8i0`UGY^6al+M~dXqiRWe2isoVSKk_o+z$DL!UDEuS=|X)hYR57?c> zl)OR&+bxL7`q!J@ZJe=@JE$?k?#@y5Y{{DLtRxbD#VBet&#GVC{FkG2a$xLM|dVz#!njRcnbl_z{3}bt!tjiT2vXQm=F(1t2u$1!1m}XJO>rb;y3Tj;R z6;T>Rey2{=r)=TM76mkGG(2x&(0r+kC-=^5uLtn&lsTenf7leuqj?pslDRRJelR`##h94TU-9sMG5kb5m6=FzYK>-Yr5r z7MeG^CB@dC%pRBP#c3U`NLH`kSTV)j-VHs>d}~4Zjk!>SCJJ|}`Hh~j$vmOp(EW5t zj#II6s!yo2A1_zua5~hp%jDH~fi8>CfSvUMYTsskuyd`ML)5W3KVe`^TIn$Up+qTr zhq=U5M1Y#!(HZY-9QOpDa|$7z1nx&>9?R-{I*MNZL?`OVn#Oly+i8vLVFX(tU1o+& zlixXp0-Alo7DK)WZ=OCSvU8rc<(G99{)jG4Ce2Ec95qDW?CRiNtZ=&$O&Wir+5Y&p z_a&kIgJ3Dj7~QU^tD;lhgkuwKYV z({OlDu&LuW zElL&)ARf%VBA?g;)nRY5eUQZLhuAs;vfy2ilc&vE#{JD0VJwVV?qdD{?A?w+Gfp33eFEeJyCc}FZZ|>1R6_tE><@V8bE^PqylA*>%NnGWUxs` z@_pba+WBqE=ev6;$+mH4+C&7st%OE0X3%TGVghM4kuU3;J~5YXo6BZ7REq8-;xXu) z6UUf#=c{2%5^T0&dX=UMo9q%mqDvi`>)f)LpP`zLz2j}xo+uf!e~{pGU9^H@t=|aR znPm2?Q{Eb;6V=Cuo31H@S$$Fo$~=kC;5%vFN9EIT+;uotl2X@c-O*y;>d;9Br~%f9 zxILq^9r6Adq8?j2E`{u0a7X(dGsAQ$8BdJETqXL7O8EN_eX;a#7kNwD{JH6ld7~4mn4-d~th0rek)ez2d)ZEraroG)eDGdSC> zVTBNmEZI4RzMF@cOrDt657nGh>e#e9ESzVyG4x^1@A0HkE|Nb#en|w^2s`W|iYVfw zT|&3dQ`mr?Pa6TQCcXOVZP>i;(Z$RSc^{Lb7o-LE$A{ED@n3!`!l;fW`&=6CIzOtE z|4#e1W@V?M!fo%TjTulHh`exZGwSh**tmu=*;aP2YAPyWajD$52S)Hvg3uc7I3li- zl$F~dH0RSU_>bXCu<7XgQeWytKJ0&&4Jb-RJn=PL z?ZF**|5Z8DFdj&^LD~1P_4xeJ*X%?>_k$jek5AcFU!YDwMxlC2U;J zi!vNHRZmJw88(LLd!J#(YsR*TKlW568u+7Kkbn#FeGC%ZN-{G3ZTV)f$vzSqb)~<< znDyQ``;IR3r;i*jL?=i-yHdM)ws>CCzE^((IPA|C3%PZ0t1e|99_;qF#N3wW84r{e z=tV5v-%qnLs#cb?_X#0vv9%_sR=5d%!7{W!-)rX>lyZ6e|#{>=d!PAEb^&=LV1=vq+UeYq#k zw#$ge_uv2p>jQG5B5IvQq;L)C!CArqmI{16L<+%JcgJ7`OM02$iu( z&ZBzfg2E8dzEDr9srNR%Y^lhFl%d|;;b{-pGW*GQhJ8q;-le{Dco70%n~%*9Rdz2{ z9J*3mfACcE;YU+K&2rS#JAjSH8qxwP-pF zPS`uQOLFxB@alcPX;ztmW$CgP&U(v-3SzJ1CV! O;UczQB^aehZOGY0#v?|Eay zXQCih$RM3aWEl_#2P78~7X1*kJP_Pp3*`S^`Nw?&Z^wD^P5BvVw;f_#7oL{``*9bp zBYp4M{ZE;oQdiZ!`gNz)0?omi`O*5#hL=3CkHy_K=6f?O%t0j#A#XpnuhRl6o&xwv}O`yIH?Po6z z<{{~QC8f%|M<%VY&rC^ECjc28lETYioaerDL$k_JyRB<3Q?}PpnKoRxoa=9af9;Y0g?X|Yk2KY>99(M&`*Pk}AA z7?~fH^L$Pr?;NOHIZGZE{Fr0#Pd_rUT6_O(?s2tryqWW}4WnGTQ0R4C`3(LfsR)x= zD(7@CrjCnS9Vf5vJfDPQ{Ctfe{@%wVD=fStDZ%M&aBdBk>3LR!;vhAcuxzniFvC(Z zKU$2(ROmPr;Sg36%N*H((qaw-Q3ew*N$BV(IBC{Kr zRu+nGmlsM`o8Am03!LMzHJVoUloJt_dKpBOOqvluA1bfBsq$6(b$j(`NO|E(zzvXH z=RNR`N8)%P`Xz7B(BoA}vYw7d(IzPJ2x!-94Av6^N${bLzPoIK&-smUs3l$FS-u}` z)e0g$KAk88d5a;&8xPtHuu+|)qTS=_h?uCUn99JC;NiC4m51r?YDsbF^hGMW31<5; zqm$a;aTWmc^46OMgrJ?@?%K>4J*MsL$7L_C7iJ`$M>~j8$a2+Q&L!g8C~MjUQ#|K! zvfPPU`*P8*uq(ej!DVfZ?=IZRXCMUZpSZnSBeH(22iey`WB{py4a-mIU5=QSmKiOVMKrL6vYx%ng|w?r-`v7-gb>8eq%( z5b|9&+dm8L9H!cd8tl9_Q8M!p;(GLbgf?S{-9C1y{H5~OygQm2mWAJrkX1gK2v$zl zu5h)Ww`wHJngE3<)#zrPx%2>vy!a{<`|BrOOh&!A33HM2Y|)kL_^hQPYA-|hJ2|`2i1P7uemmjWYBzBAOO``As}2qimJa9Pr5to|(mIhm*SJ@W%zm_!Cvr8M z9?3O8R&Fxe{YZTzx4Aeoab*b_H)IPGwnJE*U`f@^47^$7z;lssp-N%sM&Tp6E zEmC?I7JXZ@u=(cLi8(0I-JvqLJCs(s5G00)R`ef>bY8StJnO923>Jk1j%=qF5M)S)Jq`vVGZ1$0)^e@J~_-1A6UG`8d>C5l8p!k~N zd7M1(1&adFX(}2*Mh}&BPb}m?Y$v-Bco4rU&^w^Trj)jeQA-Y1hCn!Umniey&_u%WRSst>;Bn7` z4~6-3iZ{J(hu+q8Yjn#aTKmewCcagV0VLtVOO2MdkD-U^pdbkNH4Q}Z}1 zp4VOYOm<~DzUvhpR&sD?0KK3bQ#fQ_fiRmz)PF=TKRZiou%M#f8{zQN&5z~cBc8^3YerOjT&CI!~%Jk8PswPh!vC@V}y+gTU!R5k^!WC_LP z%p9oCspfFZc-Op}?a8^Q8_vyfpeHfH>bC8Dl3HLOB5;n-D4w4E0)NNic8WZgQ< zy2m3_-W;##uW8M(^e**cN101c_Wg}PsQcXRKKx|MN;MvEpT5~<_Jszc@d_l5dJ%KC zaC&6lMh*Q$%XCkw5q&O#U7!OA|BG*JO(3o`#U*vk?X^BAZs;R^p0lcuUtY67KY!N^ z>PxaCw~0x%d)4;K*FG)Hre(3*h)j0#afT_+bZ1c-fY^Y=|6ByIOJ+0V^0{oXwY5sM zTHJTaSz36)+;11?`tf)9=qBo?vPL>io+G1aMV*c!>5);cfDa4CzeM$H7mu`sJ5wi| zzS9{z^?~YA@YTx(YKG)DZw;)9;#_?6?Namf@Ap$ zrNh8CHIJpDsmSYWX~#7?C?&_lQs}&EruyKv_I}Nsv{)S`e2bXq3OY8#&V()#ox+ly zoR&tvL~}u=6{q-Y@}ZAw8?8^PV~4EB^(H$URE))}3CCg}{Wn=ZzZLOq^1UGG#D%)F z6Fq)KNB)3&Fyi<-0hj_gCbIlnSRQ+)dj<%&Dtx`;glNpMQ9QvcWovR=Uv@G}7H8G8 zjdA`s>6)RWgZH9Ff_zsR`!pSjlS4h)B8*T>kc#8wQ_m<9{x}`4;wrd*UF{(lD*1}P z1~NCRQIARtxsjky2RhSQJi^D^YDJyUEc_m&Za1vx1a8Ra zjaN-~X@;fA*gN!<$IdXO;FlV>&<8WSYxdqP$~^+NAl*ktU(Hz@vzJ&nFTpdERCu+1 ziy_(~S)4qJmB+XZ$IhplbWuUw@-)0Tm%yOc8{v(=_Q4+*Srg0c}+r1tT6V}r0W44MvCEr|4HnK;bNI2XU!9P5H${d z7U=yV;Xfo##vkdrV~j+08d$1!FV&n)A6$}~!Fl&VKFz=-iGtloX6D^U=TlVj#nXw5 zxew@@S7|MSo=e{_7NLCeoZ-ZC5);_2F)Xg9%Gt3o^Kx=rwssIi|DqW z+ukRiXURIr#RX|4Ks4`^Z}QKg=rG|cmiq3uzxBe&?KDm=;rGGI9IC@8GL4bk=$9^a zV{V;5EgL*rG?9~K))TZIIV5EPet}h>*jJXpAXdSaPTsrNh-=a!`63~t2JyL&CfIM5 zCF|Rz#-36B?*SCZa8AYYPZ&RDA!;z9{^;+cY7;4I7R9g-I#Z@Gh>T54XK9 zJJxT~V20Yp$HNO4ywjAPMLiuhJ&j-}$izF^@<8Lu;;eUsec#uKc4?BhWp|G2u{9@r zxhx*49!;bflH?lH1BAMj(CxB{*|=!B2;D!9p85brb6`T7EU1G=Y3#yWVsAAm2_pgT zIV8TaP$BA^K|lf3Kh;wtw}6 zM=5Cw!3&5Dx{;j9s_tS6X^5Ir8PKg3MWV{tEsG{Xu29SavO(WInhk#4QCJG1pz;#V6#K#m0M782CM! z1$G-yaU^=mTn_s$y8Jev`qF~B!y$lC<7LuxoCr?*Mr}JZM;lZ8h#?S35s~plT)Xc{ zY1<-XVFtC0Ir=>B_o`g2)YEm@HHV!tWQY4pZjxikb(ve)WC@fi{+^HU^h_q5HE;Sl zyau`{hT;#$VV6zJ!TzrL8~{c(nNQ-!o?C>zIgSo28*|;P%q4(P=z)a!B>j5@Y3h?x zgsjpD+t2qoW?ytPx$6zIWnA=BC@{fEhA`u+FJzAjvpa~aj2~fR;|4(5Hw^bonwLB} zgO8&=!(!)pvRj)8cA{!D2qT+!^9!h+vx`xd!++j((92Mw-32M6S41*g7|=H>4j+=u z5Q=hMbf{Oh{{x0tLkvE>v#WPjzILcx3 z5>bC(?sYg)vd*|AA4HiPLc-;CT}IN_LIq!F=tSbC_~g;zF69vwW~%jf3UeYwj3Y^6 zWJnSv<1`zJ`9)I9E%uyv77}CHucKm9Ar=59e>ml_bR?Z;4FbTt$;R-0B=_U_In@E9 z#b}RY6L}X4KS`n;t4Wm$kF$7B(akp#A6JqW`79t&mwGY4uDvAQI?}X+rZeNf?Ulgh ztXQ1-vM{Br##6{Z`KO{M6u9(waaUAzhDseXNnfcMCab4P?+7KId)G-SA;$JGG=$wZ zDnPASI0*=hg-KzueMis7YHOD{Ver#aN#ChO1i-z@g-h~Y2A>Nf`br89cV;y;^P|bT zgpd;n36JcaSjuK9O#!H~!|2S6iqKP)xmvZvOuN)E_Mzmcj;~os6XG9Utk7=TEx8$` zqLMrpAzGP6}uZQ;c-FN#9=EYH5M*muPjRb9Mu%((itgmc@>cHZce3^X1ca!eZ})XJ04` z1aLP8r$u*W)U7J1o9nynv_}nWzIjCaNmfMk6$E}07tmNA_(*z1hK{qFb~SX^36ELU zgqf=f+6Hyt#e9fYx+ae2o?7iCJ-d_H4=EgtG_O%X+YZHhy1tAr+Q4J4IJXIlsgI2@ zXvY;cS(vGvTT?}t#RE5777O|LWaP;?EI>v94*9ZyGVs5cSk~GM> z3kxx#mXW~W3uD+-8EpBo#Lc4R8VcPb?>a_hvH5%gQ@oFu^d=n^6}_*smLHPe_Q+fy zxvcn5N;fK1OGf8M)-}Uuy;xf4)F^}j0&VJd0|mCNTK03UjASbGwH%$@(E`c0aOi(m zV!SUrui!WP*m-l%LDO9|<1Gg^>6Lu6>PM5=7d;mGokTN*bk>nf%0lVtWs^u$?Uayg z(Sj--g`{8MF{o$LC}I~^CNb99J2H>d9xC9Um28Msx(eC7aD0+vMQ_iz*#Nr{kE0>8 z(bQ}^vT=(m+b6p_dXL#vrC%fHgfhCM!z9YNqtsWu|>pcQ*PG-u=Rmz#H4*{f~dLu>MEZ5 zG(Y0w^+RGwrn}dO;L?4-?yj415epzbJM-~@QzLM+veOI zJGHB{sK5A{6o(<2It)*1hdK{fdwvaV2TMlt(h+f4WgB5Ld!5{=q%U z4wlg#_NWalz|PL~VJu#G-PBR4e0^#w|?d#==ex*CJX;uuJ@`JG< zvL=pcSivkopG~g3s}C`$PB{~e*EX#Z%h0e=8>?iy!$QI|6We@>cYeg+sj_LW@H~~{ zI8k@8%BNE^qq3@Gi4V#wdqTE#Ff8{jt7O=;6ZBugcZ1kIh>uM??P8l%J8!U(2yU50 z=kiSKJV&?2m@CE1t@Lu@JJKgaC@;`KA}i=Q7Bjm3_8XV+gi7!kmTFezbkNEcowO2% zL8J{GWVSBvEP8YG-_>TanBYB1ReyzZf(_5mOc6S5yKr~EUNPxSpm}QGqxfB3d%lxa zST_r*#xxGNBzzr)@_`tnQmvUSv#E(<=sc+y<$&R2WJ=*p!V%e#%c9;dmq^c-S5`&6 zQLvmI^)5CwIirVS;5d_`Hj+Z&%p-=aU^k)c&3JM(=FF5+x{_O;o3WrpyQ6k~W4Zfb zOv?SUYl4z_g}dW!dW<(yS9!Ho$PP(fFh)L}o1C85sYt9GWXdWXWjW6W9yUuhb(vJA zAWNg9`s$*b zJBa8`N~B$^`y{!fvd7o_ytq@s9%@LEL5Wl3p7$Y(dq77<*hxJ5;@EKln5;Y4YVL{0 zY7M>-$`IOuhi@Y!jk?mAdt395{VkVPJo3Yxzm8}#Wb2C0K z;POs^=GLG*l#9-n&3&LY9k_rjsGmDdU>LVbMj?lRfsl|#O6R@wuUey>_$JH_Nc2n- zIO{bUwT(p9J};B;8B!?}Hj^zbAB-_{UX;_cZ4+bYwW_6@+Q>RS)WImdO*>J6Om+@^ zoFR>}1na(E`yD}yXva1YQ}CG=uQ26s9VEMgy$I#Dy_ddVqp!-xH6blj3JQzE`{?c| z0WGHQdJB&UrG;cP-H%(h#RsJFcD6}OMS7Zr&8V#pA_1lsrCg;0RJE=X*g$^V!V!>^ zy5(ArW9a=aB+y60-kwA@_}eS8B0-DivV1EuY`G3oYi=qtS!o7!9?}aAaoas58wC(! z8)-DksMymiY7@uAIGS}~Cm@qXBz?q?_88tn;iH0a1+v#}?RGhn5Ihyms?~+ioe~>U z)iI?cSfh&)B!-+8g$Zfmu8NqQSC(@f)7H6t0Vi$A(vBh@CQ!+Kvp_@HcD>ECVVux#y5O7DU*fUKF)eJfJ-}}=8nKUi_*Nays%pdg?I!zXUI*5r;1gG zxb7{B>_XjQq2)G#oxU(-PA`9mdMOW1L8HZiEL2a{lUW$&4x1q#zHo3fMIgcVyU73q ze>^{c1Ge^+`4g{44acfOSOuINS7iV3TYrQ%Oy~f&{rwM)H~J;_tCa&&ac+ z#K;ubzr9pD%h(d185U+Hawurf9J%17Et4%JIJB+-tG^#TH)XQ^)=z9bjowkN#k?zB zm1R05WXPnq6C^qvzT^z7RY3*Om_T4FT3zJ!$a(887g&tgoz{%nC58AJt>3dU^9*|2 z-52{Tm1g>Y;pD|wtN=Bd-Q&{i%kgUGQ}vB(`yiKP2G|%2v+?N(J}K#Fm>BF-HoV%1 zPDAanlpV45DlL(qiA1@ViXu-3C0Sl}5t(dfE`BfeC*e0Hw|hzg*?#|2xDMol>qqQA zU^UWMpdB*KaGChTrNw?7=qMJ+a1t$|1K7X-qhIu-m(zu0I%WIv`Z!c_jjtlj`g3a6 zhnjgDN2Wl9ZeX0(L^03u+a+i6B)Py(M(+hNALKX9Q1cx-cd@T9D`UxIR47;)=dxnT zOTbAQbke65yVfn^E=NX5`o8^bGGNc_bth(b-xU{`mJX;`luv}J_0>rU1wd~43re7~ zPD!XV+@km^=G4i=d%ZC3Jma9znLc7@g!`;)7>!OT1hT=hDpkPNfM6co;No8}?TL@SO$M!xMgGk_~v zyjQz$xhH-KujY#Xb)bxpy_VNeAdbAMp5K76UL%g&GVj0>5kz+mmH|7%iec;cdE3_~ z(snD<$QfxIzS4yjhJ=}*l0qU(9ODG}MQW3?w55w2FO#HGm8%p)E{!RvQ4(=- zMX-MDA|jU)4ooyYA%nr`d^-)PV>AsVauLOa#+Vyphh0#y_XeC;3l6&oo~Id&e;iIq zKGyn}4dcyasG1j^i8h&;K?r!}Q7#)+%%?j?Z9rflGa*T>D`jpM0U)ZqNKJ^P?w9XR zCXG}ZyKEGeHO4wy(oJDm`c19WWZ)BM%nR4t>+A`d-C3+nJ`cD-eE+)3k>P6YGquI@ z9o8w(A-eqR%Ax=%FXrkRvw^qY?QjP)w=FuQi)dCh{=RYtt%hB#$5E=Y+^N?f(^o88 zaNrsiuO{-D;yI`F$}&y?~Rtg%* z3=jEmrPTtG!*6aewKl%t7<~TpDxsa15 zlw85wkQ`i3n>}myL|J0&al|9m{&k!sKlEFu@OY(s;8yx;7+=(}?XGkY+}YPrXdAz? z98goJ&~{>3;Mq15zzdna#IS=bFc+V@!^?8z(0h{Z@!$SHu%23YdEr=XIji|$$F+1v zOldF4P2xhLp^7|Rs@n%jE<1`*MP_eamV{p??>i$7Y%5g`M5rt4mpBQr1>RkN1@b_&w?Hf5> z=Uc4hw(;ZDMV@k~L}XI@cuN6P2N9%`vPE#KF2jesPH4J*@vjeo;&|c)q*I$wE9`k2 zAd4h%er|21eK0qpV5s=B>HYnfl$YL+5a!vut^QGWxGL|a3|~oh@mB`TlF=Hz=5#cV zo$%Y{RoelD*!%f$zF+1I!^+^Md4QuxtnPzhRMYtyK(_%!BIULI6<5gUVdFq!lxxGD zmpcn~A#t31QLb8-7uQ+?%}m%XIV+WpH3X}s_oJZLGIKOp4m>t~i1sX$Y=#$j2RG|O z7kKSv^&o!99|T2eq|(<1S@ip$^5!!Boqb~p(3WO!2E|IK{N~xEsdUKFJ>z{vGfE#y zG)k(=io{#ghQKHUeq4kry`;`pSg{*(w*X5S)V+v6nZ~0lp+sknWMjm+c>%;DPT9|z z5Rpp+KMDd*5_0^GYL4W*_UW2{$+n;dO);B1qPedi$8QM=xAxfOw+Z;!Gd91}`gA@} z3Jlr0uAr!0VA|c}b8-jCaa*h){0+VI&d2Pc>u<9q6j^xtVOW#C>LwO_uXt$fc=Ls^ zJl0Uh+!HNT7M)vgj{qj+R?$)@CE!>Wj~U?vK;mKeivDzZUQ;ggVYrlF-g^vCbr+(W zw@H;X-q;%eF68vKRLpS$xKVZ0Es7Fgeyr|(XA@Y0))@1z*_s~aRVk;)RbxUr`(V@+ z>O+9$*`j#9M>_9mha)2rucTF&FqxPEOH-f=h%QlScb4xn+xljvc7BRv-qDvd8`VMI zwm!3_AO)%)GP#lKj}cVG}Iy1~V#(>(oEYiajpf;j|H=ZegVbeMm6Al)Mi@D~&xgaYVU4;MobA zJoETCH$7xZe%q@I1 zG^2V`@y*zf)8ag}cA0^viQfvkxsKDmob94P?X!Ed0su#Yg&Cc|0d1z{y0fBITdwgQ zrL>iK@FIPJ{tiG0gG@QntmG9p15OjSJeBLAGYDS5z=|+o56Ic9!bW~{d=a!nV{tv( z^-^PnA5s5fHT`he+e_#-WXbs+WzSSmT)B?DUIVhHDF~E3(MnM4CB9B*p<;eFaw-6- z+}LP&hw>X;A(Fa0O@pN-+g1vv;7{P8Kb;4$G+%46!_s2S2ZaI)K|XXyF7exf4CSoI zlYlvx*aSoXm!#`0@q1V$pWj>SGeb^Q9H*dT-~%bx9ebqcl5!l1E{#4Je*<e`R4|M}FvzQu;! zOdefDoA@qZ7qI$o+m@%tCUxdcN)Kf9y!Hxl&MU!&oMS^fX#f+vR!%%*D+;UE?h z!qh$x+hQ7BsGHyq(vAWU3w3ho?qkC_SHN;DX4p{!S12UZO*_I+fRuzpWuoq5QDj_m zp+yS4%aG>=x3|C$P(Q3Q8C15)U0Y^vfK4J?Bw($`9{`Z=c+})xnZQ`&_Rtxu{{Is2=2EfJ*slo44`zeSdH*hpg;M8tS@b+aUj!ehi$oklR)lW=uLk- zEzner6}=`p6)wOgB!fr2V3qRvsWiVbJOK+{nVpg{NEq3r-Evu~ zh2NguDtb6yW^_%?EAYot#44Ptu-2tSd^JKOVbuMBa4UX?y3kn4&N$H2HoKpeT&=nO5Ci9-` zkIcTCPU+^kWmkE!K&SVWvsrJGVP=Kf2i?pmj?fi`g6Q5En*b?1&=M$q3><&kx+{a` zam~a7a@`%<5Va2MWy?lbIY%6u}B@h(Pe(|x4~BvHWC5zn}q|5`UsY1`q`T;TE#SYFd zku!B1z3FW^9;%$BHCPr@2k$)-<*?A#83MgTqs|lozujvt6}rJ;XzOjZU#hXDDux9V zp9>*%Z<^3enJ!JKPL~#9w27KB8=QN*I{3t{9F^jnHXC5Ol4g$aqlsWGb{|*&AS-g| zAmwGII!4jIj&C`xp&_i*&a`FrMZ=41<)2PS6_Y<1cFT3$njWY>K7csygRU<({kJwV z2uh6~H$z*PxBXiaz>PaG9ep_FZYC4k6pm_dCoaT8BmPUn-u^$vnHddV<~}bn_>cDNQ$O zg%%5n)LA@J2(i~Dv>ts!bx&yKRhh{QQ;mE%3~x zNd9>DyUl^d^0)^a-NvVkyTAW?kM{^Rjkp?3|J%)lrXdE;o-L+M;J2Oj6Ilnvw-~L{5uJr%%T6K>i@Uu|6|wx z--iGDtkWFDrI&hd1kHnTAVEHbw)leci-Gm0gwxk+HKqCI+5Ymj5)%7ne*7ERkcKbZ zg{G-JAm$dFv#DGAiy=d^;43svG&}>TOipv_=j9*)S4n8rlYJpTc|uUs(NB-Bp5UQG zCf44RusUktvHVcnYSh4+!*ctUrqF4g2hgj~gDFp;?sBI&T*7)_#O=b<-h)>@Fjg1L zqz=B*;21QA>IF!F5wQGxwmFmDTfZ{hp3&BT#;TM+dfWh<&|A-#`x2=a_09wZS?m)A zja`cRX^aX%V0K;wNR7IXTyF8hK<o z_S?Ud6@Mh4;B1frhSrzdY(AB29iiw`icw48xS=VpzWXU%5{(14QVGjC*D^*)0P;6E zcYVm|N{qX@y8SSS2-hZPs-(g~v6?y_@1p>I@}<-B$3SkV-w{f{iRf542!I9+>H8hU z*I>x4Q+%EAJ24Osc%)t=`LND>4odqLctdZ*@$P|anG*=N4EyeOaa#1KSU*dqV^T`b zhThofpk5OfqBd?LtM`ps+S7jMGmn>WC@4N(JuM6?ct2XrJ=IkdlGz{BR08h=_b?36S_r+if4Co^L*!sgz`_xoXeofVxI@SlW{4 z##fyYC+54ev7oL;4dswLa9tXKa-Bz8Kj8}C1FxE)qXW;PS0!DA#Oh4_S13r7SD-}j z*9IWe@)!|5QKf$OJdj7A9xRFr!NrvC2kTD8{95j^3~>x&Vpo$mo)h4q`p-0~5FU4j ziN%w3iv$gk-S(&hy;-@ecyC-p3eRr7=eh|q)vtH+H$=t5*Lw2uk8@+KF#_-y4#7zC zcHc&g?`%pT6d2P7il$0=>uKG+(EsZq^Ov}8gc-j80Np$S!M&Ca(#OsGn;*_Fs|jaS z0bc4S8%Dzp_w8-@j`Np)jGYDl^4l5tJgMa;!NH8mxd!T71--+KHv~@^rH4e2n_{5N ztZOJBa{hUF_$r1})a8tCOMunLpA!;PQ3=duT3+*f$vv!OrB|-}P9D_$Hv0|N!t4wUjB&cUzl z@UKZ7Yk9O#0|*t>HVkZDlsJNB0@Vn~e6!iEZeh_dApstKh%s8F2X*b9yMZc$_*KJS zoml@RSkKk6r`!U_GJCuyd&}o&{Zi?@Ks%rd*9g1^xE8-VV=3Tc%A}!u=SOi*v)DF8 z8{xnG@0%hM4>~1RBfUo1ockC1=_Xfd7Q%%?Or4`rfsGBc%`7*y@dNiiPF-Gk|7*qq zeHKZk#sqQS{c|*D-g26UaaauWfaak5r6}k3AN&>`O-tPg*=F$RGe6~SvO8@Hq==%g zezBqJOT)u(A}PvV{O2~fjGO8;dYN1F&mmhYpfv{!8{XarBakn}Yo@!a->qC{En{?;N#!aI%OkzC^e)m%hhD&2YcK3AVb>=BwSTJ%E(`rSzJ9*JzM4{;#Yy$N$Dr4_+DxaErCm`TVb`Tf z7n1j@_Y65_&5?f+)@Z5Y!p$R@SVlKeT+@cTyYGF6h7|V<d-thOgE&@cWg@ohi zj}n-_H%=>dixV@&rQqtF4Dz16D>j3i$d06!f_jNWzt15uzSpaPH2d|TX z_=Pon4D~-REtZD{QiPbJ6F&Iy-z#cp%@LQ>?$UC%w=FJ%$?K$8g8&=VwJ0!dr$-p7 zko&m*rW=ad^c4I~AM6$fxN~10{rMofqXiJr9%_kTT}0iJ@Yw50(2BU|n9-kZF>Xam zFw2=nrfLpE<~Oy~)TMhX^>;j@oHXaOK{KRUdE$r3{Kqo*I{-4Pv4nQQ{m0-uFI{t3 zT+-|bp5l?@I?7JW?d8Ngt6HCFp04jR3EB&@h2hBwhUx5ZcvZYrLA2(lag7JV;Vos9 zI&}m&(=00iOl7YWAPFRpp8yB(p*DUXJF!TCU1AU67<}0;<^R?3Ss-0?` zv~#Q~*)%JKTTg%j-Uw|K6d&inhOYR0_kpo&R7NufSxGemz}w!HcD^=+W|MVYtJ5PwNm6dMK{xxJ!nye}z+} zOQ!WGMiaDGtO~yUE4lOI+1g0xBq3&f{>LnN5P$zDtzT-eAC@qw%`6yI`2CyccX=jO zZ55rg(9B%mIrM=3nfRb9LQ+DK2T(85#QI>OSdJ5fSL6NDOw;&t;q2{v4UZj1|G8@P zgp^nrT1j_PdJpXePN?S?Ili>(c%?v0b7v{-v$!liaQz9_UqAo(5$@;%dz2|&C*)ik{K~&BD`B`(!k2_0GSqGNl%?X9HZRe4daIV_E)jy(g*G;=hy2(G7 z88HxWxA!gf49#*`>eNp%`~6zOd|a9&JhxcE`JEZ?^2wDa=peWi@M9K2&gHutb)qsTE9W}5cHA3Jaq zi~)ypKnIWrT=Rw-m~gynNYhZ=pynUP%zSnqRafdH|^$XVpC3a1#@mb z?}q{=ks%$nirwn;Ym>eCrq0e&lU!!o>BEKrenZ;-Bl&Uxjz9SdvHPbV7wMPCO0SdQ zPa=ZT6mFn|#u6CaN)wmSyz?Iv^55|#C;mwvC3@&|bfmjyv2byd)*q9hcYdJT1RsZ5oVb8x_g3ton z^~$r7J#Ht&xLm}6rZFyWMv~kD0T(hj*&Jml8P183m*TF@A9i&mwmu}?YAB!sGg3D-5RyYcI!lOUd&Y{U6W z$ZO@`jMe)2ZF?aSQ$!cS_p4rUhd;8MuCIkwzFaPVF_Gm$BpPFxMDv?J{ zq7Xsvx%MRKms~rmEc_XYad#~og@lA&c7Mu`hWPHyZLK`oYOLv%P77*C*yVw0g75~} zKF!d|wB4iEKkubu<7r^%A2X}^g6tXLmt6Pa0vKj{$Edun{08qd+W8UM%~T;NzrGhh zUS@FmRh#rThFl>#y8(E)cQ_w&Rphhlx-wJ^=ep^KNL|}&B=Zfb$sGt08np>or1lsI9XVGvumgJCcdM#?o)S zq&mvBaw}300r3Bj_qCmobBG5dSnlxTo+Rs;W1t{WHcWFm38#aNr6ySC<8c8p|& zE(fET>MlZ?gcPO);;tP7s&79+7Z!lFWtn221j^_1xE3kMQqu#8=s@9!!2%M^`%-Mk#G>$UF@0%v5Kz8!M!gJ)Xp#|`h zrPfniGlu_&K51%s%Ka{AT-R^zf(Ba~95L<5`v-fC(wwGk_e@*z?+@51v#8y-VQ9=R z;_NRrYp>2I0Ac?=(3&K6pwRlqvi`8KRrLPK*fJQWBJc$V-df{na2mILF5j7D+J$k= z15B@gDYLNZ)XfohTK)81>!W|w>ePn{t#lE+vl5q`nIwBer6y-CmP=eee0sqV*_2Q3 zAiU5;8H$e7iQJ-^I>MR~>d-xPVJfk)|a_i`XEvV7_NOYtqehC0w}x*W8B z@&Rn!2jxr2xoviYL1>%Q@qDI$Mf zG9r-6J&^AcF3fL1;&{GHO7DK8pbP;@)2#h%ga9eEMS|RfxP5~UgGz=|_;dX4b9aYX zqbjx!5#)ND_9#dl{Jakl+92csYmnZaPJLsk`tf8@P%+5Mi4G zy=em&EJeq?`wA>A`dQs>hkczmKPKTXGBt^Q_RK+CJ<5gV}X@G-M#p{ z-_Dzv)-Sv(&6N4j$m$T7yJw8`sa3CBBIqF1Vk=NiRrtEDVN_zrJJ)`wW13_4V6azt z59kd2WR}=FK*#L`46b&2yaYlM_}6~WG~hV09S-Bq)Wo9c8bzo4EYs?C zaO{&Q=U+Q3)^6G-0{6B1cbyU=A*ES$1p)?qPGy_r!DBV3sH$zExhcu{2 z2?z`z-QC@#fRuDM0@5AQeD^%>`>yZkdDdc`h35|#esS-;uYH9hzbWw3&4%SmuYC#= zj2(DLmV#mjRmJQzxl~JNFx{$3;;F2u`5x-(oMgf@u}}L(7F@zm(u?vsnK?V=2sA|; zmE9*j9gwldXg;VyFEcEQl#C58d*{$>g-1=Tq9V_$nx<9kMWh*ht2EN*fvlyb!+^_TKL5 z;uA1BM_qcVTz^bO(Ui9zOd;Nek{(D#NJu0WJV_5# zgr|EGZs@o}j&()wMMOm018mn?ZBDCKe$n~OdfXoIZ6)(t51jA(T1-k}zk@=7ghDJQ zL}piP_#>XD!iy*KS#$)U4-?q^QdfJ{n7l!wlW{lTS>;o*`1PIoWcOS6TkV1GkcpB_ zV61xl3P7|4^PWcUkkEbq`Q@8M(DkyrtTnDb0i;{A=sbg`famlD@U2vjqh%E1zTkx^ zbYse8?JJ7(n{JU3D^K%nLf6`bSw~~|K)T2EDTz+4o!?uTj5>72faT@o4!a#18P*!f zo9jN~D|v@RD-0&s#g`FwpUA$mc`F@y+xMm+R*uutuzIHrk$bV}9*bt*fKud=_YT4`f8#1HDDR@=$!oWOub?7!T<_)l zQqWwOwLU1%VzpgzadJ}W{E`IUx|{Q{XBIL|aPuk0LbeSoi!`36Vrp-}b z)NvYR;I^gh9xAF3(m%dZ&CoS6QQ*;3jc9gYq2J!q3#PVUslDm0#>U1#pIiv-s>G1$ zM8wgUAf2pv#;&p)z$0ODn~7kZPV1f99ju;&N&8&%CZbz(ncVsp$0;N<$_HM70v}j3 zxf)Smyr=QFE7;W$&wJcovD0oQ^2)W7JY~iD;yQtYE|$%hP@GVZ`l%qJ@Vy!f2Ex?D z?`k1Hc?D5Z(agxb^C)XUhB`dBp47;OE#=rF&q3*|HD)3_R@XfU44RqnZjAoetuD71 zPqo|7gT}E5TCaINHOD8m(rZBnet+EZj<|=V!-v335U)r6bcq}ANSb=&y^jm~j3S7V zeu|ucBQ7Z*ARy8bN|N;gA3Syx`b((&Aea@nwGvf0eTEN8V9i$>_to2u zo-ffppmWIr<{eubXqt#$^Up8+bJV`?Mm>S1ZJ50Pkr@QgRM)~n3OFGpkdcw^!g>TI ziglZypb;9$`+m-pmX-#MW3YyI0?s44uP+*bzRWMMu7;;mQE%j>Qdp3Zip__=>prNo z?QB=b8OT=H&RCR6T!UzO#uyjKcRO5#3*I1_W<^l4YrFEridtVQYr6bGqvWSlM}Fp$ zxO`#7jabWH0M61nz@}BiF}Y~(o%9w6!vDCqh+m$CU-}43;b!R?z_D1Xrl9xU0HZ>W z@_9E-B@in7tlPx8-!{VuBvULspd)4Nzam$jrVoukLdHcx#-eMz2>~gU!c$7{OFM?# zpa`!iY#5ZPt7Dg#9Px5Zb>~A0sRY$<3s@8$sq0vZ^}rbGkNtyQlyg zKw8bYk7^2j+&{kxX!T!UDui))ZjLgP6(uq2eZu(qO)Ujn*=!)rz`M7ofN;rWNyoo; z_pTD)@J3Vp<73gb#(ZQq=~FQIEj;j40{-Xs1-=3;$}rGN zigDW#ka9a@qDQvh44v*R$Wu~MibV97V;8ak=0Kk6cfB;h#=sMpe}v5`e2xQDei7=Fk8 zx!zHDWz;(e7-SNAICjd@xqCqzh(lHb-w#~QWVvKMhU^tuhMxkvb259!ol;Oal8tXs zTH9Gd(#!5AslO;yH~fF$c)d8f_|XQ;os>pZnR#@wUq3eX)->0I@|1|VxZChUAUTcs z>G%I}R{iJPDvb9P*YF+BReTSgHNhxynoB7dhy|ZKeERgM9O-%>QTi(-1#D3RD2b?7 zoT0ag3Ef;&I(M&dTCfH7cJB#rNp5f6BCV_*ARgl%#6kBJ8p2x)+yfjLydJ17yJ-_w z;B1m(wA&`{%h)USlKAJz+1Cpe_3pg!B6qGfDunOkB@;dmT1_%-7;fuTUymnb+5mnr z$3Ph}0<=Fd!`3qy6V`kC3)$CVLOjkfU|5np|D zq!J)L5#7Jd?Mrti{I>0k%Ns6cNIFVFl6r+T$QNQ|l~W{C!u8;*V46 zKPT2ljFp?8Y})r|DJal0v$KbRKurR?KFWdGE#650VWowq>gwuJrTBcx$;n}K9>su3 zm$j*7OpjMJdR6<}-awB{ZEnZZ^@6~YgzgV644e<8Q8wv)SKew2;kDOU;-1+;Wy#o- zM2@8txlpHz_eYaLpGRMIfkHYBfIUzRf0DVN$7(B8iZPeJ$-BDw26R;$^InG$M7tjJ zS3JcO{%45;-?vU<2q9bS*!pt#oYusC{wxF@7M>aUit_SxKv`@rMh0ZQ*QWi6LzJD~ zst7Zs!_!y0l5auVX%5^kp8N*xi4o~nDhv7m`~Qlsk`w?Q_tq$g-P!&d11woD2bJkg zZ%&3FKH2+yM@qdp<`Z>CPGX09q~zqR)ooz=rL$1P^b8TR;>5`wM(P=QwD#EX96Jk= zshBDALEOI8=#QEVK5%L-hCY_3BEidw8;!iynX~i7R(T8mPH=5DZGcaz@IOYW+h8-ffDg+h1zV0CI9Bl~FZPU>`JZl*nH~4mVCyxYde7^ z5M1I`%x=J+RRVXRTTQ;SR(DvEAXY?z7H6MHvZT^B=XK0PKETO`z?|bxZcrU_1x9hW zgQI*YZ4=Wgm(J@Eu-Pe1t}mLs*Z=RQxDNF-=TP^~9R_w`X-joR+Bv*qwS@XV-yX`ZTXA!_@BAa1vTVeA%;O8io&Yv212I_^n{iKW0>uLqiv155`>(9SF8c9*Zi((4 zZwuQxd&ICv&5X=twoC*^@`k44&r1kaIy1&!O)T7<$V}pO)%t+KK}{{%M#uOn`u-h^ zx~D#frxm|QqwJkVM33Lbq_|&2MTK8X*WJG>&lVLDTCOv{Z74vaC&Ik1Pt-&CnisD? zD<2s`*h(XfX_QXPs;+MF%v#hM*h2L8v(JHXqZ+C>UzG){b*DAkVe@4>I`^>Hgi_ma zC(DgP*kZgxNkTYaUK97xaW}q2rYwlhR4a>EJgX{E4)`cFGdYIWAKiIQb2IPb`%6k~ zsb+akF>y4Oa%s^zhGJu5qlaUg^wVX|)cDPtruB8Rw0$_Qe(!l{>2oDP#=84|_ZqmR z5-6SAPbwtX)Sj;a(jR6DP~w%d_Jg-RC~&`#Y))0_ZO>Qer|^3Mdiutlm1&|~I=%Sa zLqib=8#vcccF;$~;c~fbe_J|{CmK(6Q1eK=4nFXcnj8f4;_`YEvUG_q{HR_(aqPQhHh@$cWiS2#O} z*`rfR%>pZPaa4ohJngUSphZ{*4=J4*-94$83gE{^SA{aV8)0nfJ+kQTR>!zx3Uoyk zSM|c&|N7hbkm-e(Z}8Umg|XE*oYocIN6*g9<%Ly-o&dkibh)t1xAcJ@WS|i?DYDsE zkw2AV99DZUELoJRJ9@kp?}I0g%fR94>!&20~{fLjzAcwR|1;Jv=<>NRrrnM&JFr1CSE-1kQVh z)C$#BnpsdKt@y}SbO*l%jyT{cvmf6x+1pX0u0Q`svFR(+P1OpmYX9=+SyEKSi_+c* z#U~S+A;(tNiak?$>o{AS3YCPbO>rx*vm#yoZdEN3i+gvKz8FOpGZ*q1`KQ!96Lb?O z+&1dHvFBgsu3V`Dj5> z!_`hP7j12oTYZls#&6_V%^T{g_ooLhL-J@~(YCygB?_{34n3OVp?VM8g514t(2NV< zg^kxqUicLY@{TdUg<+lS29O(q<|0ccV_x<1*M!b}b8nPzBs_Z`nSghSF%2wKzzFbp z{*rOfD$s`^$P-%seiBJo)9-b)NB$_;zx!(ZjY#Kr2%27i2y! z)GjxKG)-q!WTIb3de4?BmSczJ91Xd4$!W1qPM2L|Ty|Uc?QF9i3_><>WV5V3srXL+ z8utKLa8Qu|<=@vC38fy@!H0h%h7wHb$R{|b8VpY=%D!R_3R`P!sjOWAcGt!V^&QY5 ztPq$1<}_QEbF_KVOfn z0_~SyO$SfBUO1so@$!bGngcx|-NT0uot-OW5M!$Q9nP{by1yU-L_6{pT?ko=Pn$Vk zUCmc5PszVWia6mRnC1p&xh_0fh?(K9NZ%wgLw`&Kq_91v5o*Q60KtaB8U}}>oJp7? zrDw;no{CSeOn9;iwS>DFV@LGJ8|9xqBmZil-PSCyDML$kWkyn)Saz}gf?FC?5w+o8 z@#rzyXdrep0W<|tRwvy8svxIE-T+HZ^XaS)1>}gYVC!HS^S{k6NG)HE9n11(M@LoS z`P|s5%ts{7!VnAJ&MUV9o(XAL`5c)2!4vOuj%S;7^wj4&FhrRwvlIC7);Lokvm8ww z$O=wIC5XYJEDdReg41t0=nY+t=^)P?Z||do@9hthC-U$q7!(3r%t}@uGM4@El9&qf zH#Q7vkwEj33W#chz~)I@C=(o7$xxYH5u>`4qR0Bs{t1D#+|9fh%ZYFQK81Oa&ycDP zIVmJD*}kf&dYo=6N|mA5Bn|w9v&rA*plry;k%DBS9?CU8DMW0j4AVbJL^XF z38T-jy% zO0jLf=?QOqQiS!W%IL}HPN>Bv=-*GUzBsBrvV}MfY(ZtBP%|U#3clqD65ZN+6gA(@ zWHWIZ7_J)m^>4b&k~8LlNl}_JvlW7i`x(P(wl-pNM+9^|x!Ht#0X26QSW>sp)E2?< zsjQ%Y>*nr04Cp_ryY?(7D9ki9!~no0;MWNWN=sVnPZ+htk+!dQJ2Z7VY?%7GU@E?$ z+5Hu7(PyDSzBv;ZD?c-rdCgTo*bbK z#@Q#1)=n5%KtHt(upkvsU5A2+q{~acX2_a7I3Hd=I{BjAu>lUIjX2ZfG|(XoB^NbJ z;w1Cnmk*I;q<7#3e)^Qz>~*Q6r4`SX13IfA;7acV|01m<#G;m7QCnBX;LY$GO#48x zkUX3`$LV=IB>5HJ32tJ>>m*W+{0y1{QikfA)$bHElh<1 zkCk`=Yz&N{zPC;tf}Kt7C0yss{?v9&% zjkH6s*?TlYyQxWm?9!5|2B0CMhY!F?t~`AOesq$g8I&(i%%Ux&;JZm}$Q7=GS82PT zOE0*pRd|1OXAi&^8IRdb0&!?1OLTIM@P9w?_I9WeN%4t-yOwKwb2N0tSz20xpPTji zY;F+NJ73UcwbLqB#A*%Ck-Xs8%c_IN#Z#7wB{$H{Vyon8k?CilZP`F9{zz*vK@tfLtBs zJ71e%*_}5sDAbwME(Y~6D*8SYe8NztR-()7-AJ;IGNmRYRJ}gRIi{5L30Azj&{9&u zJr4)7 zs(I(H)f?}+YhCs~yI4Ny02PR34M*m&=Hn-LE9Y@}(J2md0KV;!PW;yZyotVIQCn!U z`Qsi~=8+8hagPZpys`QpwPxr+}z zeQQ}rBNKPsV#HGcKlW>n@{Ui8X&17tl@(O+o7zA`Q&Ywlj+T($S*{)!Zle?_-*Q1ob58}^ty!@~SvtA~VbW)*xX8oqW~AV-eaV>xHyLJI zzcsV{!hIF2drf{g3NP`tWPtR}pO{HQXxlV$fhY!Wn6MV2+R}dHO?+m+B&?>2A<7CH zK3&+i%%NS)BSV|8kRGAqli9vhCa!k)h?e;J+ZnrU$vDKJIX3noh;oYr1w>pr(s;d% zQ9aN2#Nr~JkluD{gB|K+d7z^SoI#BHH&IKPt`55A=aVNd)L@B;iR&PhSfSiB&#AKb9c@E9{WlOQDSV>3>oC}1J~LFh}>WvW@hwgsHlOyVx*qbtV`}pg|(jo z!v_lToRF4H4*S$!2ZA9?I@eg16u@< z!^ALgfnSaS@1>&gHxZ_JjoF1)B5AG#qNKDcOdi;CTv&`@+H$LXevxL=O}n_fEZW!< z|Nd)f)pMDXiyh~?i4n=r1uq_7gA7}Gxoms-?MUvvj6YQ}TgvOqT4LuX%#cQfh#{E&GSBR*G?Xqsp0JuroH2o^kfV1GcR zFbjNo%)oN}=k)_@Y&zW>$xn|cJ-rVD-hs|GFkHieZky#Dv%}QSIIkR#ZS&b>Z(bJY z236EWpZu$t4MwVNg^ewHb7`t^(b3SX_jv-0N`$*-M(YH_R^FtSYR2O?y?}U&AdlA$ z*WMu7bitNaHIy)&(!soXybJ1uHuR!J-WOj@D zzq-;hMrbVIya*MNl0v@*_NaGB96wXx;NTqiEju_>mhSfN?VYbjGdl6p#pJ6VXLKsv zTGYMe)vDvDWY*5pe}m!j`?*QqBl|98SY6(a$@mMHLqY|H zCZIn*ZO=O23~6OC>?-&aP5M2Y{GlqzUDcR`N6vBs{ll>EqUb~An04#G*CCMBU>VK) zF-t}O6isQyfhh=!Bq5-e1lD5TN^_cl&hQjvJKLAJaipSr-Nu23Dp3CkYpKP*NXgwV^I2?)5VZb){NKF)QmT zYIzVykK3j4Nqtefv4H^<8Gn3%UDVd1wiBwwc%k%58kv4*a&n~e+u*h^pV>-j(erL1 zb#gwjiiu+7Y3*2Tr^KjPN4rcn(eZVtF#=Qi`>PclMZxz8a`$lle$05C13rPChpJiE?|&V$n7i%a-sM5VMi z1=-9uF{ZbXY`$GSAaD=4;rA5lmhbMW+FUJ~{V!k@ESgK41`s4LQpprBLqm74PMBhmE-&w6VG*cxZIh_Q zY@YCRNjI(4%hnHWuTHFAbm*d57L`P#iltu}VUi->4ElMk{MwL=w&ZyE_>-%@e!cz4C-iqTw(kWUR?vZ3J`53p zr3K9AyS-qrZy?T7(abCtoldnxmw&qS&HHykYTAw;Lqp%7t%q1^jlkv_B8q-|^Kn1p zOS2_^GEF(rqt3G2oQi6U`S1Jg_Z8n4&Fl?^mZFDq<>@i|*$gtlH!LAVvhy?Jf^IfQp*urKRC}}vpG$DuEY8m> zD}G$Fe82vg5%ik)Utk^*vNG6l40BZ7tE_l}GBt(cb#(Ay3>tieG|69n`qSM)rtE4u zjTBX6vzmbrb;0+^%1>S0<2K=FU7u@b@pA^(KIAsJ#6j{z$zsEuHIOj9Nm1)_D*)su zk}Cy*fs?=9AU|_el2dYjg_)P1akb)bB8l(Z;sE`fe<9KUA-YYMYn_k_{%yCtVH-Z|im)q3t*nanEm*=8>o}FB$WY%hzPvL@#l6JyqvQY%f+f=4nkW-o_QOflK!KsyAmo7~^5z+<;4FcF; z$xSc?7HLrXj~2X*k($0@TED*S92cr|oc>1f6;=Ja4?=vlPn^r_lW3U1@kq7=Yt$?( z#kmcy>RJ%PcuC6*6s$Pfdl)3USq;H7Hh=SWdz0-Hee!uJKEEA{OmT5Bt%a^TWd94h z8{-QsSaWoy=?&M5M!=VAk4a77v?ZR&C!;}Y2(R?gxqh1W{b*$38^d8IH|^HVHK`Vl zD=FH9nTQ|d1GuN$TR0B>1aSD*ZK@BvTcRjjukVuqT3*OJ-N2r!Jyt^${U=^@o8Frj z6uvMZ8zBs{IfN=Mbd!P9M8D{Q%KO%yFfh5&p?uZGEUO0;^ z#OC9Omvo+@;`BSKT$bM;-&9xe2AGaSg{avPhLUKl`%}Im=Ftf?SIt%M1MUWa(J4&3 zx;IZ&55?6?c#^p5Rhvh-Qn*TWC6PCGH5K!|K^uTBE26no=HTGKX@5}|q@*S(wa`G$ zblkM}Lc+sw>~Zrz-bg6mIy@un2G*GPJqAzT$HvCK+pH#>r5Dr$*ztpFn_{J%u2sg$ z^LFWr-%;W8;to}9u&t=b$aQeqrT2r8fBk1i$T0}SrME7mP4n&yKZ=7r(R(TB7?;2u zHl;mjkoeED9%Pwd4&tBnO(})q$jc8v1bEa$UMKOWnvz0Fw6j7YBh!;JOAVIDEe=mz zo;E~9MM;p%-%X)8zkMw$JI2czM)yNa7fgmkJA8b6A0m6S;fhb9m>0#JK*x0aW-Bvt zzYmM+imIsirfu&T8tbrk_?b!8jumZz5Ul$h8z+4%AG%}?8;><9>9ekmuWS)DGh52O z4p{(;=8rLLojEw)YgbvSwsJn_T6owz0kYiFUAq8ur*>#772|S|Oafc(CT>zzD^}X< z4U$n*}*?Sd;4e7#_+nh zC^UK)??aQ&=f6z@4&MJ~H~qiGMo5vs7iTb?&f?8=*;sU1tN5FVk~MDFeDm+`#d91Y z<;TI1@T}SJLL^1b70-EZCPu-!PeliavzT6W>?ERE8)e(VuSQd^!6o$6N=L3csNRNR z8#<#L_3GN0ViIceIKF!9hpG6xP?gTFJ-G8~WTd2pBlCdvGI{t&$qofUa>*dzbg1cO zjLSzY!BW4Es%jvkFh6Z8#iG}2m5YA<&%5})-|K${E#=*zR9}qJs%XlBxSrLMd$|$a z50beE9_Ax4HBlxXW=(CwyR4mUx|i8 z{gMBTyCR`tLXlvLSkS(nO@M=qtPf=bUHj%|1*ec#6~Vlv&-UQ?W*^U~YQuKoGZT4!?i zwcaX1B5y^FAhm+A52Q)Q_gJCwvY837;X%ziJDY**wo?MGyNT%T?R$n`=Ug_Y65!p& z268b)%l&`-QqD2^;#4F*%+8*dzG5LXgISrsa_5z>JB|i&kAS$Lq2bH;mXmRqN|`~i zpn9l?_J3(%gr1__w@f}X`#`lt_1v_d^bJPOxEmfG9@T#LqFfRO{L8dh7!;;?g27my zPNV1J9f_hE0~FbTi=2%+n#_%~b-<}~lPQxKj>On-eD1ZfF(1?j4OwdxCsF^QW^%jM zh+}r8zLnRgM`SE04V-w}pBw;G3%%nQfG3h*3>#U; z0Vc>RWV{!zhz0XK zJY5EuA;27B#qcEW+yV_s8Lj4no_8YvXoLvtN)_W*u7axbU=u3^5+x99*Vm8>ONY*|9@4zNv$tHZQB53{mnga!4udG^O#z3X47kKUAq zjvOK*e}&kALv|hT*VX^a#0h4N*HlR4sdR5D`X!p5wRjaScFA0DSY}TQkYDaQIysG+ zKeLqc0Hptv9(Rfv!7~#eu(T+(6p6z3@@2HfImW-?eI&nhDkQAIbSiV? zYrDm`sS4APmfIUfCe~ZvPBXOF_TC)WY;m|98ojCO@C%xJcXVhymnv?7Du<0-3E*Ll zH$w)~o>yF1(q0X9b&7z`Iy%NNJQ-Q&_idA9vi^$~Uo-Hdhet+=8250EoUy8pe2lI* zF1NFDim8@dYxzLizWPjPNmPlohpHFM_c`k!MSN3p{#qJsaSWgs0J4MSMz4Tafl8hT z817X#|2@Kk3yLXA;j<&0X6uLD=a0mg{C1{}30M@QlkvI?0`Xkn_3!SlJq*|egx;KC zDEF56|y95P+Qw zwwT2ucOUglAS5N7s~g%D@%_u8!I3cvkPGyuo56 zRtw9|X#6gNsa%aN8-i0h_`I0lMn8ea9-K_6QzT7MK)j_3dXf97XSN`sR<+S=(m06@`oTF z+nw?8;e;tQ1GiyhWP4goni?J8@=o(I3Q|x2b03%JR{Vz#Aq;{9&iU2rT%RT6zDSyz zn?H%^<{=_C&dmAsOkuV8!I=2XGG*()8xUR}zGhO3_4j7L*LRFS2$!b72kj|Ju3AyN zF203~oHH7^st}EHa~g-F``&O?$VcPcwEC_Z8x*It@+e)Q(Bq~2&>uuGRhlZw4#g}kXyIWjRS{Ypbz3jXst3R@6l-fhvmld(2gPw_)SFc zJmb4WZ=x2rGbyX7fym3t*VS`5!fi!+R6>JmOWjclnM@g&#qmKiqpI7Z?|GHq1(1s0 zdxFP$y}|OQFM!LjE<>x*<0n9vvd`PS$N1lhk5H=vf;Uw+P~FOYxmD8gK_oCb6?ElP z8W@a0AeBlKm@LEuq6`Rs0(VJLKtO=!+L6$eSfJND3EG9^=ZRihU#~P*f6_hIz(&Sf z8B$)s1}DziAuw3)4?j7Lz(fVl@fBV!06HQBIUnC#z; zKKMWEv;gIe6R7?9=oRd%7g(D7RQOH)=#kV#I#0nCAg+ju`+E6&^iRT_d*E})%TsN8 zM|_q~-Wz!=Z*MS&P$SF6&n7x9Swt~nNn|Q-&aUkp8i!%${m7a*%h+XrxwT%Q_a*VF zdoBhB2IBJKpDnz?Zsi2d)%m~(`dk*B`mfQ5o&y~4SfRLNptupzBD>^J$aFx*@~mqj z4zr}!lp-(DoKrUCCQ&$Eyuia$cuANkoQjIC-_nGEfnh}dJcVxs!I=cTH@zM{%%cLN zVJ9Q-VnF;(eWrTonxP1&0cm`yJR*NRFs+4%dov!6^Zp)kFZJh?ccLDhic1g8ZpRns zDQ~bh z#(gNim!(O55?Ng%cuY`lyT!&+>V@XT<%l&k6-_)=o+Y3GhJal;a70HXL{?O=|40$k ztCIkKSn3KM7l=}R@R&^uQ=3lHh=)tb7BgnpwAx#3M>{4I*UhcQi5<(BiL3{mLn^w^zyOSMZ~Oq*?LH z>KoyrhDMt5;YjjRJVHX6cb?#tlBE~DhfOL4xYiErh%p>WfvhG{zDh^F+H)_mqQEpc z<5NAEXt(soi3hd|LYTQ0li_x1YIL--q1h`RW~ja_AXt##WTvvU-Q(K}5&84c^^24R zBSz%4=W`z1QZ5vL%4UNSkLlyxNETgG@HU~IiRhWi=O{wgwd}|k0*_a>a+fu%u}p2l z>%XF-9I89dr~NZTkmDngUo%MCr}qJ(4$xOpV2Ae9f^8Lj1S$(yb^r+-uHn%>cB;DI zO$cdk?F#pJjLvNa>vvb;t@kb+heYqzziBpfcgVKjpX_y{j4Xsel;vDvMl0dRY3;J~ z0yQ>s(5p67ls|hI{bt12zhD{=@!od|Pou$w`2;`?L0vZh9UcKJnxKrbnd#}^^HM!I zO{!x5ZfW})$9c)hi<-PAZWk;7`3%_(w9J7UIJW@25IjhoohvAXP?xDOwbcafpsR;-6_onLMhoiryV#UupN|T^XzzB!I z9D%<=pZWoo?Ffi`O})Pjx`3}di8`u@f@PujKW(C5+&8blMF87U= za(CUj=b2l!n14Cvf*<4F`x)b}Yb=1c#wn)$5$}983Bb8SA#lHYXQpNY= zN=h;j6nlC%RFl3hY3%ejc3<0iNy*5vsupG3ZAPtAt=8}ae;`IAx zQEKf2yblW#ocrI!Zrwi`gOu4yqi*~(r0Bku^+_MbjtF&LoEo;s zcVg4i@kEla+xToJ8%tApFs4#-)|9WTt>vsxgN=^DA-b5j_+wJIqo#UU5qQ0#~(Emq(z3zg-9Vzu}nQ zR{6S3JY^0aE-TB1ZuxV-IX?=je9|Qag3x1xhlrOq2jf9zH7eK?W!Khn85h|(xG={E z&qxVR&L>uWnik-N6U%u>VY`6BZUVwnywF?utg?^y@BbJb!X$dLEUT}7y4VhahoySi}AZEbCbYU-aN zSYW@zQ^6!lNK7mN{JygQu!`$LBP}hR27sX8AULC5Q2qp?u3oeA6G2gVS5m1QoRWvo z31CMx2>ibN@hgEPCzVEmn_`FAUn}V^U`Bt713p?ZI`aH*GtfF|HG6WC33zIpt&{RN z(U-9ifOAM+2UTMVtTg1NftX9dyGOvsxYvgKc9UcOWG}C>$*}DG#!IJs^jeqGbVlGXA2Vk0(3iU{2h9GXR?wVqo17}2*BQ0E zHT1a`%A;P3mny}czVqHkJ(aF@>1&b$X(4YK%qHshHpa)Ipw2Epu+?z1|ND0i$k+v) zlye>sL$9o?WR8md)eV5B2f%qo{5YRF-=Xj0T!L0z`R&_UwgMnKz*ETpTYG3SLlCHP zL*Q!W0t60Bh!A>JS--$n;3kJhw}Q;~zzOl{;}z!J?QgL~Ema-!ZE+T(w(`yiCz-7l zvg-*r9z%KG5~>PUR8u-RbAz-FOaH!Mf<9p_x}L zeHP9$xa5kQ14BVxw}WxiB=j5eWLZ<7p*6epdw&e<(E@R_fdfez;P_&0Dd=wf&4!~5 z01pZ6nx;ODiHL6od<$znWmR^goMIt}>FCMOUp; zHeYRWTsD#lc-EGFi9F&qxBy6?GgaCVv12utKWNfG^F#GHF_B4I=L)==fT%`k&?~P= zWwQ_k@WWwwrB~I_o}SfNsZV}%7gsjX&y+e;3$NzPa#8d9$FQDhD0=stOaC@7`(E!h zdYL7^!W7j{^voZhTqpU~m#R1^9Sc$;GoHq~;c~`Z@4zf_+->prN}kI$z!ee=Sig@Ip%kn+^g-p^@R?LsD)Sk2#~@>fI~RIv}8M z_tH6xaZ@XHD5daH4B&ngNQx`AYYRf53@&N!;SLj%lan7%UjbE0PHHN}j)67;(-SJg z1%)LQY-C4nuNQ7NbXcu-pixl;tj^xWg8rj6;`81Y*XGIG)pNNAv=D*&@JtLMG;_@K$Hho^`V4>%gv9x zghGWd1%0Ro;<}t4s1BE7_K;c#%Oo zFg&cbP1g&ym@fvC^7!B$$_xp#khWw@3fzaz+##n&pduSVXV|cR^8Fu6gXVW zhKMg$>)R&O+iF;+X>zv}bY&x;ikpEfO+?fdy$Df0Vr583 zNg+b`Gaf0C0>>4`#Mz?L}fC*0KBoR z>;O8CKLFvbk6yye5B$-_z-Ya8UiZ? zaimCuRx$vatZV)zt%ctimWe26U^V9rJma#Yn@pN=4O?8h#=ccjh|Ma7x)-g72KAb4 zxGt~j&RS>La<{=obNU6LVESD@atm0$9LQ>QRF@C>3D111#pP8Cv?R8$BX-#{3$iim z!%vYA2==Z91K>C$ro^JCWRyE8rYXtV@2j1ACHQ2`ZF$45QcN&z2N10M*49SBbo1W< zmVdoqUlA3-?rLxv1--OIu3GB?^y~gUvZqZO3Y~t#kt4ZqSq4mVa%kwy#HU35-p#(# zg^gQDUY{wtI1K(zeA;zyp8KaXOLNCfLFcr{XtK=R)Ey2SG8QFNbD<*^(742|?!E{2 zsn!n8z5BhiF2@!Kohe0>;<2;cbx80TwK@B>Ww@v6tix^=gzH&#a4t-8>$&Tsep|Dx znCFiaAd_7$QzR{TetatBRew=81mevg*nns9h6v0x3WvwXqh{g2+zE8A=*UxK_iSxT z#xgC_0Q%$wb9(C+U*!>oPn(jkq%Jp992!FZlH1QhaLg}x<0u~gd61x1caMBsQL|%M zFaxDBE7o7E_CnZTK3ojjW&s9w8oli1X7@cBsZMp=VyuH9a_m7vh z)d>(Q;`_0gnSEP9Tp+?KzWs*NxVyifE(KT6;rH* zDUv1@cZ+TeqK8>Z+TUN|tZzmzx;+yrsLqv>$1xZ6cqmwjJ3OJ+cfu=z{KSL$qeHnv zk>HrKtzPgj$jIJ;pSYxDv({uvu)fM6xmPmm&uqFreh|>cGx^MXw$|PtB)g@hW%hp* zz|Z3md94&z&gCR{F^=C~g=9dBVn#D}gfR1FVytLeQadP(b$TAp6@?_^gdMU2w1jN( zz;1RkEWVmdNZ$2HW`7SmnfEFx+0+RnN#MBTqx^6~nX8xrkF$OQAFs&YphE;6eX<^< z_$xjqRFn=$1kZ70a}z43Yn)pcUa9K2KBB@hSj)UfLW9+}QfriPPrVcPaC|3Q*oIb3 zkEgk@4vfv%p6lyH?187*>W67Ze)ZeCkE+>JC#smOW!jiU+wEDYlCfE*B+c#fR@6IV zfd)Wa!8hL=!B+v>1qcLk0d`h-piMsp4kS^HbC*4F{Oti1nVE$nnY?4>(@9tFw;s^C#q+%D`IQbdCqdLxX8ION__Bb>=bTqFZsZDV~UK(?t^o}L{Y4tBT zB(=8=sAoWItbUb{F-JV{7Y@`R51MF#$EHR`xROr%fIf0S{PXi?+O4AlDzv56k@2=smVD4LQn31^=b7h%?PS7w8R#WptRsy|pek`Vy}!>DucTq@QK?c&P?vHa|hV zzIp!b$rv{kpJb+M6bhU3?&V$Q?UQ_ob3vKr)CY_v7AxHk0S>pC*$KL*V57Nu|Nf_Q zH0jNjgwTfote|tjcn*R)FtM@6Svvnx0kj&@Afe@AlyaN(gRBY(Kx$5Ru5(yzuU-OO z?l5RE%m7n6Fy9(w;``KxB2tKaa^dXo4|(?T1|@xpw41cz;;%O%{GPo-KcXf&UMa9F z?;K8q>EeCgu!fj#=l||1E6k{mi~N-eo$isJIJ8;WT$bBv(Gh*AHI}*9shaow(km~K z(D`OPUKE}bAue4e4JV8`%T1w~=I|iPYKJJ7;9N3#twrnmNr3Gzd-fGs~Da3-vSzAPQk*Xb<{MYL&l77h!zCfO?8n@&}lygd#d5MJTlZf4>d%!9Y1Hm)Wb&331%`4dFMjMuM}uGfx+mL696KR_96Y`yCB z7q&s!s*oD##wa1!ui`bDozJ^dZ=6gw*{bhVdFB;y$;pi8hUle0dd6<2eqY*b@?|Gv zT8-U5Oe?0&5uF}ObtI8j++O8fS=kl?s)%$Th(inZe2j}wveFTh*3{%Y<#!LH24e4< z>)(2S>e^&iL+RQRllJ|z5MT23>q4U!PaqG2j)wRPHZBC0jK&rZtoU0+&p{Pv?Pfm=itK zpG|EtLR^P+cE3NFUj{TrRHu%oVB(0%SQeFO4klt8u37MAtiA$PpneEOnS;GOgZ^Ib z4HtT8?>hi(MK#1Wtn4xSl@C{}FP3<8K`K$)R2k&09$1)dodF)cS?_NWvjN-qwqUCr ztL~h9R`KNUE?3MU_4MO$X*d(tp6S8_EW3a9beS&y){6S@K^{dpfz9`Vkm%#<#LKVG zMn9v#y#ndP<})T4UDr>iqsn|Hj}$SpjP1 z_aKAL%07s0rvXG8J!_U16^!nH58e0=jeT2PKzqtv&&yxeKmusc&J2q= z`xahZ+aR-blM$Y%9TWXj{zts@P4rC7LrEbaUl)S^%%uP5{NW0q*&^O=0}CLD%Eg_% z3o$j{kN*C`$oV)57~O`Jb722H0v@aNtMj8L8TBbiqV^Me!J;~?F=>#~Ul*BCy_3U- zDdB1gY8Kgv#`{)~^OA<<6E^?7!4}#1rb3cA=^ujc4u%%62Rp`a7Q~w$OIFCghI{!6 zG&VI%AxWPS9ohQIddHJd1X%P%Mj1`sAdB8&A(v#;1mT@_eIQd_EsFkN^Exi$N}$WSc6Mq?vzgFzUQG-5Z$PD`@%#kd zGIgsULbx^JPjsiJisC-vLY`IL7-tj`#aYc7xBo#i?{V?ZA1%*G*r+Vsxi5jaw>;WakLP5Z*I; z@TztHSX*RWOaJBGKJoLLkp;pzgc10P4(pXTb=T=yy>mGZ;QrEwvN>Hscar14ZnA=M zfbQT8hr$lu$W7p6W^rlQ`}R#1gaDown2?HaNh<-)Bzt zH;y0pHTay7t7|q@xvG=d>RX?g(ZlKaBrk7t+AOoRG`4llk8>z?wAu?-lzxwT3ZLII zjAxInManJ;m?cBrD!y@?w55FbX_v;`q|8Xtx$|FD4-FB3g4wUP(N9v+Kvz0-tF#-9 z%2iLwZStAVOCB+-jO@G8xR3CcJ5umPQ$6$^&$fF4(cyb#o1C@ajMtzxCQ>?CM*nn} zPthfd@P8a)w}4wlkNx?dVA!D}=q8|boY>DMIV<5x(3S~91Ajw%d;1Ah)swnK#ZoG%EqI@2!_M+mYQK9E7M7|kvAS2c994uJpzytg`0>C%^nHdm9eY)` zom}*ocx~v{{T%1!@P5-Jwyv@4@D4j)v#FWmM>dEL z0vVzvJ`vQ|tFbZCZ$6KFznCCXvIKpUnNdmVRu|ex#3j&^W9*dAV_52-=zY^?I?Y(b$La#)O|KRH>yr5|cII9WLH8GoYpf4GmKs=W z&MA~JPws#DB3(AwfO+)2|GozOl%3Phb7QD;I!P&zGw z#A5#EZV}PlO9X}lAMX+1IeHA&(1BAG7l~*kucT?>qD0*0{+h_6m_3yy2hv+wm0&bj z%-$S%Bgo0w*|6UKJR7jw0a?^fMxtlHIBzWU-r!$>-{Cis?nK^&x(%58zrd|`ZRqyO ziuu@xMEa?`YzFcuOGQP}!BYE4dN*#|@IVcRWoXLD%2FIqectfT&d#p+@vyg; z>-zN$N&VFwFK3XL@>BvE6M)NFEkgqXWehZV^{rHC?CwH2E7>}5v?bQm)CgVAdUle6 z*G}N-)vMa?_WjqbyiSW#v9S2(eE1-$tD$D^$4(YlZ^|-W%XQ?cBsP)g3Z}{Y&(7AI3RFFg(`ONtMoe*%1t`W zy3yw}XMqE0>YDo1mWc_SaDLa15+Afv08-Xt|3P;{d8%%IVR3O5s=-`e313Zf1R|(V zsI+9N+lL-2k|r$Z<9wXmte#?aKo_zd;bU>|=^) zxfJNkuK>vB>%Lyk;qvtYbhK7HwOtw@2pNXz{cN7f!q7_zq2=ZLC=@E>MWrV3k#*Oc zf>lz*prpc57^{xF2w^U;!P7ij7#kCUOZICiqx@~zP$l9uChib+O}8QgO==9v@#Ke! zxUJ`;fhHH{OJ8Xk!hE|xYd%(Ue9|5|8Ss{i1EpRL`j)PJVvOe~A6HiM!=zK#1JrmH z?s4M45m+=R+8w2N}VGJ(P4-e4U@?ej8Ev=MNks_*C9Ie<9P^k z=BNb!ISP9oJ}quZr=aIVHu@s)s2-V;!<#r3DZw~@@&7ux&>ZNE@rXfc?Re`Gnm738 zxTIfD)6>r!52!0O8wQq}+qZ91V_{*flkX5wn9g@+H1_way{Kw#UOXP~PeE&d8`NCq zhyj%)fB77}t4Kv%o!BUC`Lf%?bMBzLBMqIiBDy?0`^)|oozl2Rb{$iYKq}TN<*Cjq zjQ__(7}Ft5wLkvR{PCY3QnM0lPu_@^Z|afS^&^v&htLr94WoUgJTivE%fy6SHj?g} zUuFN#;bhssvZ(Hzbps*7|5zEI*n=l|P_iWdj0gYML3iJSRWWuwcE$C-pQMRAX{EZh z^Y9RleE*xS3T+DiKQ{oWBs>Y%rSJ{;e+(ChU3k!WjIrEy$CP@BD5G;coj~*%;rSyg<0gu1U22N)pePX?or@q|1lk>aB49;@(AmvL z1ROTIlwuYtZdVLG8Y;a7p%oF$aXh>vu*CEU+GfCRo&uq=G#KEbk`i1LkkEW@sH_xB z^?FL{zPG*H0c2f<(7zHN3%S?hX&{Pn++wP&t!)FeB)Jz&Nh%Vp8@oCH&G@MnM{=m9 z-<6PaeMg5kNtpFrp0Fv3LQio36WTJx0HffB8^wFdqX@wOg!lAJOp^2)%Cq{-V^dRN z(4*jiuV#-B%7R&Z9zub1;)DBo>FWzD%#(xAJ8b|Cs59SG?Xq0*3-*NTT-!)F7m|@z z&Hx&h0&u9D^7)eDVxQco^B?C=Ki$T}$r78uJ8vv9<9@{o>7-ikIyO;$+B5ZNoFVU(9?SZ)ApV8&L|q~F|a*!oLfsJ^ggSdNCBcs z0JxH&)$Epxi{&(w^<4sN@)NKXLT@3_L1BU3=;A`DZiKU(*-&YiVx~HSlShtha7Z1n z2^#k0GBii?MfCWDh9vMG?0;Z?*ui7Fwb!@?eWOCV3=+Wu2e6<&df`j7f(v`L*n~Lt z@XjrSdgOUA`A*{lmP0*IfDp z+NyCGrK9U0mUdQz*)0_wtJjx=W%!yvDia9^!(K{_H0NkiQZt9Mf77RJ~A}$Pr z3L$-u{|S!)DMVve`W8Z4CL6_V9lV>8m`HDE9j}<7T2+1tEBG5+P`nW@AiB)aWoOi2W4%r4NCRmF$a`us`tX~Hl@ zCIR;2(9JHZlYYNMQN-mdle69A=;(!^5~zIfz>uems}J&Gj~|JIX^9N8!A&RV6EXi+ zM`Z=9P_3gTA-b!p>(A>bj9_Mx-oAB6cst8oT?qywd31+EVMwRGfADw=uzuBc(k;&1 zmtp`oRNW3u`QWr`c^itXd9G*HR|Ss33PC0=S*q|RBmzk=x2~r$m!hJXgKl6e0b^XA zuf_K_*sx*n*t)Sh3L7)B#2POC#3~QP$fkSf3PQjf$-2 zNuiZq8hX*5hr~ih!kl)Rq!k+FNtYV6k@EeuI}R%&?kBwv!k|*X3uc2=to~u_QVDFS zEa=I=5cl{x#c8{w9>=vbQ1n6CL-qQ5IzHQUf@D5)_+l}xs7xcN?-hD<2RC9aFxgwa z=^wvb5UDW|G~XD9-~_3_c~rXk%)77oCc1q;?^A=Sr^l<5=HTiFDdS1Xepr9` z6ZcfEK^wLE^DrT{<5>ZnQa}fb(C#A)Z#9l}q4iL^TSE|AVA?MR#fH)z_YCvvx`M!{ z$B%UWT(toV%95^n{0 zXMAjl;gifndNNgcV&6^*w?H(4m+81J|gJ51Fnik6D5K;3e;oRMlbu7ng%%Cl%=FJHaUc9JdjdbblLGa>P40l3kxP$QCCsWkCIsHOUC%) z^o<>YJ);viIM&N~Ba92X_=0 zZl7z4g|bQAd2`3&<2>v)N8v27H+6U!M}iFqVFm71DE5W#q;elKY`Rn-lihnc=t=!K z9vew|Mz~{;L=u)g2Q9sp{sNV_PyGQ9C`f`+Vrn9RNg<2tKxS z9EWG0HFNnLB~d0$U^9X(ko{5-pMO! zg(uHkd#d3ud+E}OTq8x&%*lDuMJM%DRMcJI#E>7L$s(}?*525Cra1i4!r@Iq?nNQ| zw8A1SKl%+CA*>@O8Q5}=ZOG`jv$V8SH9XzKTF#{R5fnVM=XoWDFe{iJ-Lwx`Bh@%5 zUGqa9#NrvK2V5aEvS%0UCA#QlJIJ;(q6enUlfCM(jlsWjk#p!K?wtNXqFt{FWbZ|A z_(77BhM8hvX_@oZo^(^EqWN+BHJ+2jh$Q=MDJdlki?E^j>QE>Sjs7b>Ljp;EMGwz}39dzthjv;hefIdYbSEO)$i73Z* zY9}fTMDxTD+8BDi-Qh%lN)X9uWyZDd#{^JE4kjA9FAZG4)$+A_5q+BX((}Y25lI5P z!p%& z5fyTFKYDc!f@t>903fx@q`o36xODE!83tm*fk#kgKLZCg?E>baEImN+_oa2yKV4jO50`MmylVLr^!rF?ft&hl=h zQIbv`K^Sp*vLE)ATEdn<pb< z7gWUb#RH>dFN#cEVCA}US0JLF^Gg-N{RN`ZOwe`u$E8ne@x57E!UbH-PsWKeJ`on+uPeTT=1%ih+031Q*$<* zR59LEgcg-%NJD+_t)@e4z@5MaB+@`>Z zvTm53L33NYh{c}l1ir1Fhb$H34k!EdlKAFz@f4fb49*7LtDI{em=Nf{?D;A+mqW6hW--`1ojrj|ua3{qKzJ;zS4O^!(aE}2RQr-(^(~NV# zT)30JFLA+d`IId!+Ym9CgTRigvU(C$a!jGb?xKw@)?fNJeyDQ~)s< zFdu_%?EYw%Z-%Zc0?WK*26Lq#h1rTK?MxG&PPVLK&hC?8WkQ_~0hz6NkBPWTM*e(B z%ILRH*Gi{F+-+D>fwp0TD~)edthE_`qZOKN!`LzZ9xp(qU*ss)_V%!5vL5) zzKh4zh?Sa(K*)P%V$(()6amd}@HPp*^0S6?i!t3V?&N#)LBBOWEwlJjw8I?3)UmCV zoxhs4a6mqyp-ab5^KqCJO_Fl*O7pR^OP1NKdB_NAu;pDw%59f5HrP8qfPL1gO<4zP zprxQPg&&?Kg&PcHkNtSU+tJraCQyBwl1u#stVX>1*6)WrFw?BNlubp7vC)2e96Y;uJ9$ObIr)n=@PJn^>7YL`C-4Uq=B2kO&? zip$XOVT=3OK;qrAcdZeetP5xDSY6>Wm3dGz(+4fFh8NoBo!w6%rFeJ0B>!yA4-ifH zhB9m@epXIKLQ;fQ>>AtijI=}MQ)458O${}RcH)pXPA4A-Wty?@GiQlhzE9QtVJQz^ z(4#9(iDgJkM<|OsG~#24zIZ?Rw~J37i5t`01%DL7THuGRfFD z1LxV6=UeG?G|J~2S>-&s+6FeF4Ap45oOV7Mn2JDzG?29c8GOe^eXc=jx@pbi>jz88 z^+zd`8fIyjD4TswFFppS`G0~17?|q_z2A0(&zR6&@eR~5i7J{x6 zs(6oDo1590T@p|}LQmE%tHqG)NK<`(b%*8owq@jNH-SX`C{63x*ua5c-~7DCJW&Lx ze$vam!2Z@pa>cOyqT_Uw?QLQXaAI;&V^BFEqE4SG@{bMGZ>#=Oa7p)+Cdtk}u6tTRdh~ zY5W+fu=Al;g0Q+H&kp`zV?0m>TvCARMw1P~tlUZ^^m zYk>=uy+QT(%4MI(Kuqc`1;lO}F`Feu7b6nX5w>|yw(pXSqi7s={ri#ui{>Nc99N-d zreDM`-+vji(YO`Zhuu7!#_z_Yn%j^o-3`6O0|&_rGpV&3Lr&{=x5fGLNW#y)d3r<# z@1h*lbVX(GXAA~LdyIGCa4SmUX^Z$kuLBpWQmR{(m*@?-FSuJG?K0@py;lvX%9^3!P^oAfeZY&W}bI^$&;h1#4& zg1}o1;=RrJ=}f2k^R|T)>~pibopMsUW%@ar;LS(y6G`jO@Y-d~7xhB>PQhB9NBC`r zkBH(OL@AarBp18XW9KZ=W31;oE%r7kHCXnkx%6BG5=gS0S0LOxP$)gZ8drP zJE{~1GoyQ=o1OA2a(p}_J)P!ucv5`x=^qv#ddfod=zy<#)WJ)`elp~TseV5gT$e7a zzJ{hID|Z_dcUi3u@#oHIfD8+xs{{D9-bAR50{oi|{(A@#`$oD-j`$Bwp7Q~KlO_PtB*#4+Ds`T-23VJ;BhrK}(p;$SBx?=xPQrBa z?pumT1oId)ug2>J$@%gM=rdMpjpiHdJRbi3{oA(a3g9-c$w}cgOu|&g zKJ313Dk*L1#k5TEvjiehyocBNrHT8&16khQBv}nAbW|AI!u?xmkBlW8a}({!yGq2+ z2_9X2u4YzL3*fGH_*rwn%i41E23!UR)7k+wE*?pazL8|m8s&fm@#5EvrOeu*QlzS- zAg5$1Q5T&^D#^)(R#hpNAU8wT`5KYw18crM@0@p5fa?CWiWk+~&2S_HQ zkrs`!X6I5RHmA5D@Eo+jb5v^gor8+1yTYn_8?!_VYIRH1%ywOuUuI5=wFbO$I3<23 z+*0rsV~rPW1|~n`q!7}n&Y0x~F}*an-zN4Y{~!B{M1~K_=G(;V2=_$}2bw;RPpA^i zC8^8dHqlq4#(9MYT~#&e%T49Ub=)>P2nOPV9!O^goLqjiF!fe2+lew7WY*>f4?zoF z|3_!Xnks5OEAFdu(|#|!r7L-eGX7_}1I+(px|zV&H@=e8zM)z;1L`uitS@lH?D=?o zA814Cm6;L@yy~RxJQB172!`=*fYkUszD;>DyRSY=kgu}Jh~?^M);wRJS8^Ab_u+{5 z!QJS{uRC}{+xM}~^-{~>kA{xpLPTS7E^!|Ie;>+|AT(O15GWhIDpoMatlK_?>=#T>35*+ zR*WNd?<~RRyG2>JX5#k_$1-n~-6M}-$<+>FCEt|g}As_Kp2e9D3vBPU4SZ00ixiiMyUN;4(W(x*TAy1u{ZU~w1mOMb94geBNT zLDT~GQ4bUSu&0vMdr$>5h=%STzbYRIj^5`(!2BnmB#fAxoGelp1wduNVVL0|07Zxj4;fKK&VNT46xIB97&$x0{j2|yvU zZaRPKIfl<$ah)vkgSr*GwY?!t7A}jxkr|VqbN@thGWl`ws7a|3vt^<{Flnqx@#EAe z=dIr4p;!nw3v0KYTg&@S804+M*b9sFfGlhYV>x@xJY5^}ZQ96EEp2kE=dfQhp8HUGZW z#0nbdMm*m)I?muw;-+H+$S75={KvE&WWY!dx_RQ#7)Dg&<#l#k#s&ST)-~0*&XXbSf9X0UKasz+!zFIY|q+zBgHz*>hkV z67ENUtrP%ZSFxsi!v&6q*dwIJQi6raw}t>f0B9`iK4`vw=2!Dk5~WDa^qBAXUq1XC zMmk2!9DB|8ql(eLH=%$gdYu$jGR&f;-(T=vZRV}c z@zrGpdipH5tZ0F3-4M9;v%!@Vxdgq{ho}`Kw);CL0gsS zI_05Nx+V&fo7{PE_}H&)BYQhGE9(oWxlAq1Lc;wRy&lPve?U!i)*kQFn}-YVT>)es zPZtBGt0Ai}f$IbgxQT;C{RhvPAU%oZTmO8$J67=6K@R{yCZL44&A&WY^0mAbjz=GW zK}(bme?JUBH+M+{Z#$mXVEb$4+=EH&wtDqp0t^v2tOuYHGYDWK;li*65GGyz{go%s zCD;n0wsS6Zd!odee4|f^KpDoG^&|Gk7SX)%Bt z_XCvY?AWHEC3!^BMcc)fFN3C6Hbg6Zkeg-zV4QE%bvaF=tk~5F(n1a(X&~!lx_o(U z+juDyuApD*zcP z^p`mny!MMZv~_eM(R`_N)kD@if5DbiBq-vB6L>Cj6adAU2Jo(Vm)vh|+C?PMt>*#k z*m@F!APi?Q-VY_k%>rLaz88f zhRbjuR1^lCRDR9!TWs{^)icbg)5%k7^6ghP?OrK5Dshi_)1)JWtsl-g%)}uI*|tvL zIO;x&%cOGsHUS_!UI{kmdZFd6(XV-VNsa-V{e?XZvudt+Js?Li+95jfb^x2*Fq>)_eH^_v{Fh0HLn4@t?p(Z9wet%Z?$bV0!ic7-$$#HWJS?5;IS7DB2D#m=?fr6 zXJ&&5_mHf2i(wgxzHgt~wt-5EKPKKONdKjPkkt`ZIFja3se2z z#QN)T5@+0@5T_4X`e=q)-^wJIKvcBUbLq7@;cDaU%D=Qm( zy>AHnv(@^X1x*JNw5ZRQNZ77n-DJ|tiG;0Pwf*Bgh zEdSwfP{;x00^_cVv5z4n4ULIeUOsA^2pVp1EzAleJDvH6^~VNWMr^?PlNUgCmP^8@ zSQrf@*xk8cEe9Yg(1j@4w!jf&nV~DEBVD7+L92it{8hviEf0Wo6GT|=ZZ62bTnDO( z)pdiDZIIV{p@jSMjHjWQp@HW6dM?sruME60RnID7tXgi!IKF$PJ9X^$)8PmJfeZc@ zws%ZsB)=BBT~5Hg7K#-6WyH@vreBUo($bh6Xlt2RA)czcuII z%W(-)$kA;Gy6TdPZiaA}4H!UD$$_GWnE}Fog*%_VLk005v~ufxg9P~w54etWxOTD# z$*7oxv>+X4>%*YQt z;LmYI#~g$6N*+KseaM^~dcA-u&MCCJO-AF%hm6;sp#gS50Y`F`-s0q;`8?z^5|hBJ l{Qrj^ny}vsK1Q&^uY2;me32Ff7aIJ#C3;UJQ%KX}{{bJ&=^+3B diff --git a/_freeze/mod_wrangle/execute-results/html.json b/_freeze/mod_wrangle/execute-results/html.json index f448355..d626a45 100644 --- a/_freeze/mod_wrangle/execute-results/html.json +++ b/_freeze/mod_wrangle/execute-results/html.json @@ -1,8 +1,8 @@ { - "hash": "86455aa43f8016526e87a46fb94ff491", + "hash": "fe6fea2f836f1b9bd58de160ee2e9f8b", "result": { "engine": "knitr", - "markdown": "---\ntitle: \"Data Harmonization & Wrangling\"\ncode-annotations: hover\n---\n\n\n## Overview\n\nNow that we have covered how to find data and use data visualization methods to explore it, we can move on to combining separate data files and preparing that combined data file for analysis. For the purposes of this module we're adopting a very narrow view of harmonization and a very broad view of wrangling but this distinction aligns well with two discrete philosophical/practical arenas. To make those definitions explicit:\n\n- \"Harmonization\" = process of combining separate primary data objects into one object. This includes things like synonymizing columns, or changing data format to support combination. This _excludes_ quality control steps--even those that are undertaken before harmonization begins.\n\n- \"Wrangling\" = all modifications to data meant to create an analysis-ready 'tidy' data object. This includes quality control, unit conversions, and data 'shape' changes to name a few. Note that attaching ancillary data to your primary data object (e.g., attaching temperature data to a dataset on plant species composition) _also falls into this category!_\n\n## Learning Objectives\n\nAfter completing this module you will be able to: \n\n- Identify typical steps in data harmonization and wrangling workflows\n- Create a harmonization workflow\n- Define quality control\n- Summarize typical operations in a quality control workflow\n- Use regular expressions to perform flexible text operations\n- Write custom functions to reduce code duplication\n- Identify value of and typical obstacles to data 'joining'\n- Explain benefits and drawbacks of using data shape to streamline code\n- Design a complete data wrangling workflow\n\n## Needed Packages\n\nIf you'd like to follow along with the code chunks included throughout this module, you'll need to install the following packages:\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Note that these lines only need to be run once per computer\n## So you can skip this step if you've installed these before\ninstall.packages(\"ltertools\")\ninstall.packages(\"lterdatasampler\")\ninstall.packages(\"psych\")\ninstall.packages(\"supportR\")\ninstall.packages(\"tidyverse\")\n```\n:::\n\n\nWe'll load the Tidyverse meta-package here to have access to many of its useful tools when we need them later.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load tidyverse\nlibrary(tidyverse)\n```\n:::\n\n\n## Harmonizing Data\n\nData harmonization is an interesting topic in that it is _vital_ for synthesis projects but only very rarely relevant for primary research. Synthesis projects must reckon with the data choices made by each team of original data collectors. These collectors may or may not have recorded their judgement calls (or indeed, any metadata) but before synthesis work can be meaningfully done these independent datasets must be made comparable to one another and combined.\n\nFor tabular data, we recommend using the [`ltertools` R package](https://lter.github.io/ltertools/) to perform any needed harmonization. This package relies on a \"column key\" to translate the original column names into equivalents that apply across all datasets. Users can generate this column key however they would like but Google Sheets is a strong option as it allows multiple synthesis team members to simultaneously work on filling in the needed bits of the key. If you already have a set of files locally, `ltertools` does offer a `begin_key` function that creates the first two required columns in the column key.\n\nThe column key requires three columns:\n\n1. \"source\" -- Name of the raw file\n2. \"raw_name\" -- Name of all raw columns in that file to be synonymized\n3. \"tidy_name\" -- New name for each raw column that should be carried to the harmonized data\n\nNote that any raw names either not included in the column key or that lack a tidy name equivalent will be excluded from the final data object. For more information, consult the `ltertools` [package vignette](https://lter.github.io/ltertools/articles/ltertools.html). For convenience, we're attaching the visual diagram of this method of harmonization from the package vignette.\n\n

\n\"Four\n

\n\n## Wrangling Data\n\nData wrangling is a _huge_ subject that covers a wide range of topics. In this part of the module, we'll attempt to touch on a wide range of tools that may prove valuable to your data wrangling efforts. This is certainly non-exhaustive and you'll likely find new tools that fit your coding style and professional intuition better. However, hopefully the topics covered below provide a nice 'jumping off' point to reproducibly prepare your data for analysis and visualization work later in the lifecycle of the project.\n\nTo begin, we'll load the Plum Island Ecosystems fiddler crab dataset we've used in other modules.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load the lterdatasampler package\nlibrary(lterdatasampler)\n\n# Load the fiddler crab dataset\ndata(pie_crab)\n```\n:::\n\n\n### Exploring the Data\n\nBefore beginning any code operations, it's important to get a sense for the data. Characteristics like the dimensions of the dataset, the column names, and the type of information stored in each column are all crucial pre-requisites to knowing what tools can or should be used on the data.\n\nChecking the data structure is one way of getting a lot of this high-level information.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Check dataset structure\nstr(pie_crab)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\ntibble [392 × 9] (S3: tbl_df/tbl/data.frame)\n $ date : Date[1:392], format: \"2016-07-24\" \"2016-07-24\" ...\n $ latitude : num [1:392] 30 30 30 30 30 30 30 30 30 30 ...\n $ site : chr [1:392] \"GTM\" \"GTM\" \"GTM\" \"GTM\" ...\n $ size : num [1:392] 12.4 14.2 14.5 12.9 12.4 ...\n $ air_temp : num [1:392] 21.8 21.8 21.8 21.8 21.8 ...\n $ air_temp_sd : num [1:392] 6.39 6.39 6.39 6.39 6.39 ...\n $ water_temp : num [1:392] 24.5 24.5 24.5 24.5 24.5 ...\n $ water_temp_sd: num [1:392] 6.12 6.12 6.12 6.12 6.12 ...\n $ name : chr [1:392] \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" ...\n```\n\n\n:::\n:::\n\n\nFor data that are primarily numeric, you may find data summary functions to be valuable. Note that most functions of this type do not provide useful information on text columns so you'll need to find that information elsewhere.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Get a simple summary of the data\nsummary(pie_crab)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n date latitude site size \n Min. :2016-07-24 Min. :30.00 Length:392 Min. : 6.64 \n 1st Qu.:2016-07-28 1st Qu.:34.00 Class :character 1st Qu.:12.02 \n Median :2016-08-01 Median :39.10 Mode :character Median :14.44 \n Mean :2016-08-02 Mean :37.69 Mean :14.66 \n 3rd Qu.:2016-08-09 3rd Qu.:41.60 3rd Qu.:17.34 \n Max. :2016-08-13 Max. :42.70 Max. :23.43 \n air_temp air_temp_sd water_temp water_temp_sd \n Min. :10.29 Min. :6.391 Min. :13.98 Min. :4.838 \n 1st Qu.:12.05 1st Qu.:8.110 1st Qu.:14.33 1st Qu.:6.567 \n Median :13.93 Median :8.410 Median :17.50 Median :6.998 \n Mean :15.20 Mean :8.654 Mean :17.65 Mean :7.252 \n 3rd Qu.:18.63 3rd Qu.:9.483 3rd Qu.:20.54 3rd Qu.:7.865 \n Max. :21.79 Max. :9.965 Max. :24.50 Max. :9.121 \n name \n Length:392 \n Class :character \n Mode :character \n \n \n \n```\n\n\n:::\n:::\n\n\nFor text columns it can sometimes be useful to simply look at the unique entries in a given column and sort them alphabetically for ease of parsing.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Look at the sites included in the data\nsort(unique(pie_crab$site))\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n [1] \"BC\" \"CC\" \"CT\" \"DB\" \"GTM\" \"JC\" \"NB\" \"NIB\" \"PIE\" \"RC\" \"SI\" \"VCR\"\n[13] \"ZI\" \n```\n\n\n:::\n:::\n\n\nFor those of you who think more visually, a histogram can be a nice way of examining numeric data. There are simple histogram functions in the 'base' packages of most programming languages but it can sometimes be worth it to use those from special libraries because they can often convey additional detail.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\n# Load the psych library\nlibrary(psych)\n\n# Get the histogram of crab \"size\" (carapace width in mm)\npsych::multi.hist(pie_crab$size)\n```\n\n::: {.cell-output-display}\n![](mod_wrangle_files/figure-html/multi-hist-1.png){fig-align='center' width=384}\n:::\n:::\n\n\n:::{.callout-warning icon=\"false\"}\n#### Discussion: Data Exploration Tools\n\nWith a group of 4-5 others discuss the following questions:\n\n- What tools do you use when exploring new data?\n- Do you already use any of these tools to explore your data?\n - If you do, why do you use them?\n - If not, where do you think they might be valuable to include?\n- What value--if any--do you see in including these exploratory efforts in your code workflow?\n\n:::\n\n### Quality Control\n\nYou may have encountered the phrase \"QA/QC\" (Quality Assurance / Quality Control) in relation to data cleaning. Technically, quality assurance only encapsulates _preventative_ measures for reducing errors. One example of QA would be using a template for field datasheets because using standard fields reduces the risk that data are recorded inconsistently and/or incompletely. Quality control on the other hand refers to all steps taken to resolve errors _after_ data are collected. Any code that you write to fix typos or remove outliers from a dataset falls under the umbrella of QC.\n\nIn synthesis work, QA is only very rarely an option. You'll be working with datasets that have already been collected and attempting to handle any issues _post hoc_ which means the vast majority of data wrangling operations will be quality control methods. These QC efforts can be **incredibly** time-consuming so using a programming language (like R or Python) is a dramatic improvement over manually looking through the data using Microsoft Excel or other programs like it.\n\n#### QC Considerations\n\nThe datasets you gather for your synthesis project will likely have a multitude of issues you'll need to resolve before the data are ready for visualization or analysis. Some of these issues may be clearly identified in that datasets' metadata or apply to all datasets but it is good practice to make a thorough QC effort as early as is feasible. Keep the following data issues and/or checks in mind as we cover code tools that may be useful in this context later in the module.\n\n- Verify taxonomic classificiations against authorities\n - [ITIS](https://www.itis.gov/), [GBIF](https://www.gbif.org/), and [WoRMS](https://www.marinespecies.org/) are all examples of taxonomic authorities\n - Note that many of these authorities have R or Python libraries that can make this verification step scripted rather than dependent on manual searches\n- Handle missing data\n - Some datasets will use a code to indicate missing values (likely identified in their metadata) while others will just have empty cells\n- Check for unreasonable values / outliers\n - Can use conditionals to create \"flags\" for these values or just filter them out\n- Check geographic coordinates' reasonability\n - E.g., western hemisphere coordinates may lack the minus sign\n- Check date formatting\n - I.e., if all sampling is done in the first week of each month it can be difficult to say whether a given date is formatted as MM/DD/YY or DD/MM/YY\n- Consider spatial and temporal granularity among datasets\n - You may need to aggregate data from separate studies in different ways to ensure that the data are directly comparable across all of the data you gather\n- Handle duplicate data / rows\n\n#### Number Checking\n\nWhen you read in a dataset and a column that _should be_ numeric is instead read in as a character, it can be a sign that there are malformed numbers lurking in the background. Checking for and resolving these non-numbers is preferable to simply coercing the column into being numeric because the latter method typically changes those values to 'NA' where a human might be able to deduce the true number each value 'should be.'\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load the supportR package\nlibrary(supportR)\n```\n\n::: {.cell-output .cell-output-stderr}\n\n```\n\nAttaching package: 'supportR'\n```\n\n\n:::\n\n::: {.cell-output .cell-output-stderr}\n\n```\nThe following object is masked from 'package:dplyr':\n\n count\n```\n\n\n:::\n\n```{.r .cell-code}\n# Create an example dataset with non-numbers in ideally numeric columns\nfish_ct <- data.frame(\"species\" = c(\"salmon\", \"bass\", \"halibut\", \"moray eel\"),\n \"count\" = c(12, \"14x\", \"_23\", 1))\n\n# Check for malformed numbers in column(s) that should be numeric\nbad_nums <- supportR::num_check(data = fish_ct, col = \"count\")\n```\n\n::: {.cell-output .cell-output-stderr}\n\n```\nFor 'count', 2 non-numbers identified: '14x' | '_23'\n```\n\n\n:::\n:::\n\n\nIn the above example, \"14x\" would be coerced to NA if you simply force the column without checking but you could drop the \"x\" with text replacing methods once you use tools like this one to flag it for your attention.\n\n#### Text Replacement\n\nOne of the simpler ways of handling text issues is just to replace a string with another string. Most programming languages support this functionality.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Use pattern match/replace to simplify problem entries\nfish_ct$count <- gsub(pattern = \"x|_\", replacement = \"\", x = fish_ct$count)\n\n# Check that they are fixed\nbad_nums <- supportR::num_check(data = fish_ct, col = \"count\")\n```\n\n::: {.cell-output .cell-output-stderr}\n\n```\nFor 'count', no non-numeric values identified.\n```\n\n\n:::\n:::\n\n\nThe vertical line in the `gsub` example above lets us search for (and replace) multiple patterns. Note however that while you can search for many patterns at once, only a single replacement value can be provided with this function.\n\n#### Regular Expressions\n\nYou may sometimes want to perform more generic string matching where you don't necessarily know--or want to list--all possible strings to find and replace. For instance, you may want remove any letter in a numeric column or find and replace numbers with some sort of text note. \"Regular expressions\" are how programmers specify these generic matches and using them can be a nice way of streamlining code.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Make a test vector\nregex_vec <- c(\"hello\", \"123\", \"goodbye\", \"456\")\n\n# Find all numbers and replace with the letter X\ngsub(pattern = \"[[:digit:]]\", replacement = \"x\", x = regex_vec)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n[1] \"hello\" \"xxx\" \"goodbye\" \"xxx\" \n```\n\n\n:::\n\n```{.r .cell-code}\n# Replace any number of letters with only a single 0\ngsub(pattern = \"[[:alpha:]]+\", replacement = \"0\", x = regex_vec)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n[1] \"0\" \"123\" \"0\" \"456\"\n```\n\n\n:::\n:::\n\n\nThe [`stringr` package cheatsheet](https://github.com/rstudio/cheatsheets/blob/afaa1fec4c5b9aabfa886218b6ba20317446d378/strings.pdf) has a really nice list of regular expression options that you may find valuable if you want to delve deeper on this topic. Scroll to the second page of the PDF to see the most relevant parts.\n\n### Conditionals\n\nRather than finding and replacing content, you may want to create a new column based on the contents of a different column. In plain language you might phrase this as 'if column X has \\[some values\\] then column Y should have \\[other values\\]'. These operations are called conditionals and are an important part of data wrangling.\n\nIf you only want your conditional to support two outcomes (as in an either/or statement) there are useful functions that support this. Let's return to our Plum Island Ecosystems crab dataset for an example.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Make a new colum with an either/or conditional\npie_crab_v2 <- pie_crab %>% \n dplyr::mutate(size_category = ifelse(test = (size >= 15), # <1>\n yes = \"big\",\n no = \"small\"),\n .after = size) \n\n# Count the number of crabs in each category\npie_crab_v2 %>% \n dplyr::group_by(size_category) %>% \n dplyr::summarize(crab_ct = dplyr::n())\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n# A tibble: 2 × 2\n size_category crab_ct\n \n1 big 179\n2 small 213\n```\n\n\n:::\n:::\n\n1. `mutate` makes a new column, `ifelse` is actually doing the conditional\n\nIf you have multiple different conditions you _can_ just stack these either/or conditionals together but this gets cumbersome quickly. It is preferable to instead use a function that supports as many alternates as you want!\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Make a new column with several conditionals\npie_crab_v2 <- pie_crab %>% \n dplyr::mutate(size_category = dplyr::case_when( \n size <= 10 ~ \"tiny\", # <1>\n size > 10 & size <= 15 ~ \"small\",\n size > 15 & size <= 20 ~ \"big\",\n size > 20 ~ \"huge\",\n TRUE ~ \"uncategorized\"), # <2>\n .after = size)\n\n# Count the number of crabs in each category\npie_crab_v2 %>% \n dplyr::group_by(size_category) %>% \n dplyr::summarize(crab_ct = dplyr::n())\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n# A tibble: 4 × 2\n size_category crab_ct\n \n1 big 150\n2 huge 28\n3 small 178\n4 tiny 36\n```\n\n\n:::\n:::\n\n1. Syntax is 'test ~ what to do when true'\n2. This line is a catch-all for any rows that _don't_ meet previous conditions\n\nNote that you can use functions like this one when you do have an either/or conditional if you prefer this format.\n\n:::{.callout-note icon=\"false\"}\n#### Activity: Conditionals\n\nIn a script, attempt the following with the PIE crab data:\n\n- Create a column indicating when air temperature is above or below 13° Fahrenheit\n- Create a column indicating whether water temperature is lower than the first quartile, between the first quartile and the median water temp, between the median and the third quartile or greater than the third quartile\n - _Hint:_ consult the `summary` function output!\n\n:::\n\n### Uniting / Separating Columns\n\nSometimes one column has multiple pieces of information that you'd like to consider separately. A date column is a common example of this because particular months are always in a given season regardless of the specific day or year. So, it can be useful to break a complete date (i.e., year/month/day) into its component bits to be better able to access those pieces of information.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Split date into each piece of temporal info\npie_crab_v3 <- pie_crab_v2 %>% \n tidyr::separate_wider_delim(cols = date, \n delim = \"-\", # <1>\n names = c(\"year\", \"month\", \"day\"),\n cols_remove = TRUE) # <2>\n\n# Check that out\nstr(pie_crab_v3)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\ntibble [392 × 12] (S3: tbl_df/tbl/data.frame)\n $ year : chr [1:392] \"2016\" \"2016\" \"2016\" \"2016\" ...\n $ month : chr [1:392] \"07\" \"07\" \"07\" \"07\" ...\n $ day : chr [1:392] \"24\" \"24\" \"24\" \"24\" ...\n $ latitude : num [1:392] 30 30 30 30 30 30 30 30 30 30 ...\n $ site : chr [1:392] \"GTM\" \"GTM\" \"GTM\" \"GTM\" ...\n $ size : num [1:392] 12.4 14.2 14.5 12.9 12.4 ...\n $ size_category: chr [1:392] \"small\" \"small\" \"small\" \"small\" ...\n $ air_temp : num [1:392] 21.8 21.8 21.8 21.8 21.8 ...\n $ air_temp_sd : num [1:392] 6.39 6.39 6.39 6.39 6.39 ...\n $ water_temp : num [1:392] 24.5 24.5 24.5 24.5 24.5 ...\n $ water_temp_sd: num [1:392] 6.12 6.12 6.12 6.12 6.12 ...\n $ name : chr [1:392] \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" ...\n```\n\n\n:::\n:::\n\n1. 'delim' is short for \"delimiter\" which we covered in the Reproducibility module\n2. This argument specifies whether to remove the original column when making the new columns\n\nWhile breaking apart a column's contents can be useful, it can also be helpful to combine the contents of several columns!\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Re-combine data information back into date\npie_crab_v4 <- pie_crab_v3 %>% \n tidyr::unite(col = \"date\",\n sep = \"/\", # <1>\n year:day, \n remove = FALSE) # <2>\n\n# Structure check\nstr(pie_crab_v4)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\ntibble [392 × 13] (S3: tbl_df/tbl/data.frame)\n $ date : chr [1:392] \"2016/07/24\" \"2016/07/24\" \"2016/07/24\" \"2016/07/24\" ...\n $ year : chr [1:392] \"2016\" \"2016\" \"2016\" \"2016\" ...\n $ month : chr [1:392] \"07\" \"07\" \"07\" \"07\" ...\n $ day : chr [1:392] \"24\" \"24\" \"24\" \"24\" ...\n $ latitude : num [1:392] 30 30 30 30 30 30 30 30 30 30 ...\n $ site : chr [1:392] \"GTM\" \"GTM\" \"GTM\" \"GTM\" ...\n $ size : num [1:392] 12.4 14.2 14.5 12.9 12.4 ...\n $ size_category: chr [1:392] \"small\" \"small\" \"small\" \"small\" ...\n $ air_temp : num [1:392] 21.8 21.8 21.8 21.8 21.8 ...\n $ air_temp_sd : num [1:392] 6.39 6.39 6.39 6.39 6.39 ...\n $ water_temp : num [1:392] 24.5 24.5 24.5 24.5 24.5 ...\n $ water_temp_sd: num [1:392] 6.12 6.12 6.12 6.12 6.12 ...\n $ name : chr [1:392] \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" ...\n```\n\n\n:::\n:::\n\n1. This is equivalent to the 'delim' argument in the previous function\n2. Comparable to the 'cols_remove' argument in the previous function\n\nNote in this output how despite re-combining data information the column is listed as a character column! Simply combining or separating data is not always enough so you need to really lean into frequent data structure checks to be sure that your data are structured in the way that you want.\n\n### Joining Data\n\nOften the early steps of a synthesis project involve combining the data tables horizontally. You might imagine that you have two groups' data on sea star abundance and--once you've synonymized the column names--you can simply 'stack' the tables on top of one another. Slightly trickier but no less common is combining tables by the contents of a shared column (or columns). Cases like this include wanting to combine your sea star table with ocean temperature data from the region of each group's research. You can't simply attach the columns because that assumes that the row order is identical between the two data tables (and indeed, that there are the same number of rows in both to begin with!). In this case, if both data tables shared some columns (perhaps \"site\" and coordinate columns) you can use **joins** to let your computer match these key columns and make sure that only appropriate rows are combined.\n\nBecause joins are completely dependent upon the value in both columns being an _exact_ match, it is a good idea to carefully check the contents of those columns before attempting a join to make sure that the join will be successful.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Create a fish taxonomy dataframe that corresponds with the earlier fish dataframe\nfish_tax <- data.frame(\"species\" = c(\"salmon\", \"bass\", \"halibut\", \"eel\"),\n \"family\" = c(\"Salmonidae\", \"Serranidae\", \"Pleuronectidae\", \"Muraenidae\"))\n\n# Check to make sure that the 'species' column matches between both tables\nsupportR::diff_check(old = fish_ct$species, new = fish_tax$species) \n```\n\n::: {.cell-output .cell-output-stderr}\n\n```\nFollowing element(s) found in old object but not new: \n```\n\n\n:::\n\n::: {.cell-output .cell-output-stdout}\n\n```\n[1] \"moray eel\"\n```\n\n\n:::\n\n::: {.cell-output .cell-output-stderr}\n\n```\nFollowing element(s) found in new object but not old: \n```\n\n\n:::\n\n::: {.cell-output .cell-output-stdout}\n\n```\n[1] \"eel\"\n```\n\n\n:::\n:::\n\n::: {.cell}\n\n```{.r .cell-code}\n# Use text replacement methods to fix that mistake in one table\nfish_tax_v2 <- fish_tax %>% \n dplyr::mutate(species = gsub(pattern = \"^eel$\", # <1>\n replacement = \"moray eel\", \n x = species))\n\n# Re-check to make sure that fixed it\nsupportR::diff_check(old = fish_ct$species, new = fish_tax_v2$species)\n```\n\n::: {.cell-output .cell-output-stderr}\n\n```\nAll elements of old object found in new\n```\n\n\n:::\n\n::: {.cell-output .cell-output-stderr}\n\n```\nAll elements of new object found in old\n```\n\n\n:::\n:::\n\n1. The symbols around \"eel\" mean that we're only finding/replacing _exact_ matches. It doesn't matter in this context but often replacing a partial match would result in more problems. For example, replacing \"eel\" with \"moray eel\" could make \"electric eel\" into \"electric moray eel\".\n\nNow that the shared column matches between the two two dataframes we can use a join to combine them! There are four types of join:\n\n1. left/right join\n2. full join (a.k.a. outer join)\n3. inner join\n4. anti join\n\nYou can learn more about the types of join [here](https://nceas.github.io/scicomp-workshop-tidyverse/join.html) or [here](https://njlyon0.github.io/teach_r-for-biologists/materials/slides_4a.html#/title-slide) but the quick explanation is that the name of the join indicates whether the rows of the \"left\" and/or the \"right\" table are retained in the combined table. In synthesis work a left join or full join is most common (where you have your primary data in the left position and some ancillary/supplementary dataset in the right position).\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Let's combine the fish count and fish taxonomy information\nfish_df <- fish_ct %>% \n # Actual join step\n dplyr::left_join(y = fish_tax_v2, by = \"species\") %>% # <1>\n # Move 'family' column to the left of all other columns\n dplyr::relocate(family, .before = dplyr::everything())\n\n# Look at the result of that\nfish_df\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n family species count\n1 Salmonidae salmon 12\n2 Serranidae bass 14\n3 Pleuronectidae halibut 23\n4 Muraenidae moray eel 1\n```\n\n\n:::\n:::\n\n1. The 'by' argument accepts a vector of column names found in both data tables\n\n:::{.callout-note icon=\"false\"}\n#### Activity: Separating Columns & Joining Data\n\nIn a script, attempt the following with the PIE crab data:\n\n1. Create a data frame where you bin months into seasons (i.e., winter, spring, summer, fall)\n - Use your judgement on which month(s) should fall into each given PIE's latitude/location\n2. Join your season table to the PIE crab data based on month\n - _Hint:_ you may need to modify the PIE dataset to ensure both data tables share at least one column upon which they can be joined\n3. Calculate the average size of crabs in each season in order to identify which season correlates with the largest crabs\n\n:::\n\n### Leveraging Data Shape\n\nYou may already be familiar with data shape but fewer people recognize how playing with the shape of data can make certain operations _dramatically_ more efficient. If you haven't encountered it before, any data table can be said to have one of two 'shapes': either **long** or **wide**. Wide data have all measured variables from a single observation in one row (typically resulting in more columns than rows or \"wider\" data tables). Long data usually have one observation split into many rows (typically resulting in more rows than columns or \"longer\" data tables).\n\nData shape is often important for statistical analysis or visualization but it has an under-appreciated role to play in quality control efforts as well. If many columns have the shared criteria for what constitutes \"tidy\", you can reshape the data to get all of those values into a single column (i.e., reshape longer), perform any needed wrangling, then--when you're finished--reshape back into the original data shape (i.e., reshape wider). As opposed to applying the same operations repeatedly to each column individually.\n\nLet's consider an example to help clarify this. We'll simulate a butterfly dataset where both the number of different species and their sex were recorded in the same column. This makes the column not technically numeric and therefore unusable in analysis or visualization.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Generate a butterfly dataframe\nbfly_v1 <- data.frame(\"pasture\" = c(\"PNW\", \"PNW\", \"RCS\", \"RCS\"),\n \"monarch\" = c(\"14m\", \"10f\", \"7m\", \"16f\"),\n \"melissa_blue\" = c(\"32m\", \"2f\", \"6m\", \"0f\"),\n \"swallowtail\" = c(\"1m\", \"3f\", \"0m\", \"5f\"))\n\n# First we'll reshape this into long format\nbfly_v2 <- bfly_v1 %>% \n tidyr::pivot_longer(cols = -pasture, \n names_to = \"butterfly_sp\", \n values_to = \"count_sex\")\n\n# Check what that leaves us with\nhead(bfly_v2, n = 4)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n# A tibble: 4 × 3\n pasture butterfly_sp count_sex\n \n1 PNW monarch 14m \n2 PNW melissa_blue 32m \n3 PNW swallowtail 1m \n4 PNW monarch 10f \n```\n\n\n:::\n\n```{.r .cell-code}\n# Let's separate count from sex to be more usable later\nbfly_v3 <- bfly_v2 %>% \n tidyr::separate_wider_regex(cols = count_sex, \n c(count = \"[[:digit:]]+\", sex = \"[[:alpha:]]\")) %>% \n # Make the 'count' column a real number now\n dplyr::mutate(count = as.numeric(count))\n\n# Re-check output\nhead(bfly_v3, n = 4)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n# A tibble: 4 × 4\n pasture butterfly_sp count sex \n \n1 PNW monarch 14 m \n2 PNW melissa_blue 32 m \n3 PNW swallowtail 1 m \n4 PNW monarch 10 f \n```\n\n\n:::\n\n```{.r .cell-code}\n# Reshape back into wide-ish format\nbfly_v4 <- bfly_v3 %>% \n tidyr::pivot_wider(names_from = \"butterfly_sp\", values_from = count)\n\n# Re-re-check output\nhead(bfly_v4)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n# A tibble: 4 × 5\n pasture sex monarch melissa_blue swallowtail\n \n1 PNW m 14 32 1\n2 PNW f 10 2 3\n3 RCS m 7 6 0\n4 RCS f 16 0 5\n```\n\n\n:::\n:::\n\n\nWhile we absolutely _could_ have used the same function to break apart count and butterfly sex data it would have involved copy/pasting the same information repeatedly. By pivoting to long format first, we can greatly streamline our code. This can also be advantageous for unit conversions, applying data transformations, or checking text column contents among many other possible applications.\n\n### Loops\n\nAnother way of simplfying repetitive operations is to use a \"for loop\" (often called simply \"loops\"). Loops allow you to iterate across a piece of code for a set number of times. Loops require you to define an \"index\" object that will change itself at the end of each iteration of the loop before beginning the next iteration. This index object's identity will be determined by whatever set of values you define at the top of the loop.\n\nHere's a very bare bones example to demonstrate the fundamentals.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Loop across each number between 2 and 4\nfor(k in 2:4){ # <1>\n \n # Square the number\n result <- k^2\n \n # Message that outside of the loop\n message(k, \" squared is \", result)\n} # <2>\n```\n\n::: {.cell-output .cell-output-stderr}\n\n```\n2 squared is 4\n```\n\n\n:::\n\n::: {.cell-output .cell-output-stderr}\n\n```\n3 squared is 9\n```\n\n\n:::\n\n::: {.cell-output .cell-output-stderr}\n\n```\n4 squared is 16\n```\n\n\n:::\n:::\n\n1. 'k' is our index object in this loop\n2. Note that the operations to iterate across are wrapped in curly braces (`{...}`)\n\nOnce you get the hang of loops, they can be a nice way of simplifying your code in a relatively human-readable way! Let's return to our Plum Island Ecosystems crab dataset for a more nuanced example.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Create an empty list\ncrab_list <- list()\n\n# Let's loop across size categories of crab\nfor(focal_size in unique(pie_crab_v4$size_category)){ # <1>\n \n # Subset the data to just this size category\n focal_df <- pie_crab_v4 %>% \n dplyr::filter(size_category == focal_size)\n \n # Calculate average and standard deviation of size within this category\n size_avg <- mean(focal_df$size, na.rm = T) \n size_dev <- sd(focal_df$size, na.rm = T) \n \n # Assemble this into a data table and add to our list\n crab_list[[focal_size]] <- data.frame(\"size_category\" = focal_size,\n \"size_mean\" = size_avg,\n \"size_sd\" = size_dev)\n} # Close loop\n\n# Unlist the outputs into a dataframe\ncrab_df <- purrr::list_rbind(x = crab_list) # <2>\n\n# Check out the resulting data table\ncrab_df\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n size_category size_mean size_sd\n1 small 12.624270 1.3827471\n2 tiny 8.876944 0.9112686\n3 big 17.238267 1.3650173\n4 huge 21.196786 0.8276744\n```\n\n\n:::\n:::\n\n1. Note that this is not the most efficient way of doing group-wise summarization but is--hopefully--a nice demonstration of loops!\n2. When all elements of your list have the same column names, `list_rbind` efficiently stacks those elements into one longer data table.\n\n### Custom Functions\n\nFinally, writing your own, customized functions can be really useful particularly when doing synthesis work. Custom functions are generally useful for reducing duplication and increasing ease of maintenance (see the note on custom functions in the SSECR [Reproducibility module](https://lter.github.io/ssecr/mod_reproducibility.html#consider-custom-functions)) and also can be useful end products of synthesis work in and of themselves. \n\nIf one of your group's outputs is a new standard data format or analytical workflow, the functions that you develop to aid yourself become valuable to anyone who adopts your synthesis project's findings into their own workflows. If you get enough functions you can even release a package that others can install and use on their own computers. Such packages are a valuable product of synthesis efforts and can be a nice addition to a robust scientific resume/CV.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\n# Define custom function\ncrab_hist <- function(df, size_cat){\n \n # Subset data to the desired category\n data_sub <- dplyr::filter(.data = df, size_category == size_cat)\n \n # Create a histogram\n p <- psych::multi.hist(x = data_sub$size)\n}\n\n# Invoke function\ncrab_hist(df = pie_crab_v4, size_cat = \"tiny\")\n```\n\n::: {.cell-output-display}\n![](mod_wrangle_files/figure-html/custom-fxns-1.png){fig-align='center' width=384}\n:::\n:::\n\n\nWhen writing your own functions it can also be useful to program defensively. This involves anticipating likely errors and writing your own error messages that are more informative to the user than whatever machine-generated error would otherwise get generated\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\n# Define custom function\ncrab_hist <- function(df, size_cat = \"small\"){ # <1>\n \n # Error out if 'df' isn't the right format\n if(is.data.frame(df) != TRUE) # <2>\n stop(\"'df' must be provided as a data frame\")\n \n # Error out if the data doesn't have the right columns\n if(all(c(\"size_category\", \"size\") %in% names(df)) != TRUE) # <3>\n stop(\"'df' must include a 'size' and 'size_category' column\")\n \n # Error out for unsupported size category values\n if(size_cat %in% unique(df$size_category) != TRUE)\n stop(\"Specified 'size_cat' not found in provided data\")\n \n # Subset data to the desired category\n data_sub <- dplyr::filter(.data = df, size_category == size_cat)\n \n # Create a histogram\n p <- psych::multi.hist(x = data_sub$size)\n}\n\n# Invoke new-and-improved function\ncrab_hist(df = pie_crab_v4) # <4>\n```\n\n::: {.cell-output-display}\n![](mod_wrangle_files/figure-html/custom-fxns-improved-1.png){fig-align='center' width=384}\n:::\n:::\n\n1. The default category is now set to \"small\"\n2. We recommend phrasing your error checks with this format (i.e., 'if \\ is _not_ true, then \\)\n3. The `%in%` operator lets you check whether one value matches any element of a set of accepted values. Very useful in contexts like this because the alternative would be a lot of separate \"or\" conditionals\n4. We don't need to specify the 'size_cat' argument because we can rely on the default\n\n:::{.callout-note icon=\"false\"}\n#### Activity: Custom Functions\n\nIn a script, attempt the following on the PIE crab data:\n\n- Write a function that:\n - (A) calculates the median of the user-supplied column\n - (B) determines whether each value is above, equal to, or below the median\n - (C) makes a column indicating the results of step B\n- Use the function on the _standard deviation_ of water temperature\n- Use it again on the standard deviation of air temperature\n- Revisit your function and identify 2-3 likely errors\n- Write custom checks (and error messages) for the set of likely issues you just identified\n\n:::\n\n## Additional Resources\n\n### Papers & Documents\n\n- [Reviews and Syntheses: The Promise of Big Diverse Soil Data, Moving Current Practices Towards Future Potential](https://bg.copernicus.org/articles/19/3505/2022/bg-19-3505-2022.html). Todd-Brown, K.E.O. _et al._, 2022. **Biogeosciences**\n- [The Ultimate Guide to Data Cleaning](https://towardsdatascience.com/the-ultimate-guide-to-data-cleaning-3969843991d4). Elgarby, O. 2019. **Medium**\n- [Some Simple Guidelines for Effective Data Management](https://esajournals.onlinelibrary.wiley.com/doi/full/10.1890/0012-9623-90.2.205). Borer, E. _et al._, 2009. **Ecological Society of America Bulletin**\n\n### Workshops & Courses\n\n- Data Analysis and Visualization in R for Ecologists, [Episode 4: Manipulating, Analyzing, and Exporting Data with `tidyverse`](https://datacarpentry.org/R-ecology-lesson/03-dplyr.html). The Carpentries\n- [Coding in the Tidyverse](https://nceas.github.io/scicomp-workshop-tidyverse/). NCEAS Scientific Computing Team, 2023.\n- NCEAS Learning Hub's coreR Course, [Chapter 8: Cleaning & Wrangling Data](https://learning.nceas.ucsb.edu/2023-10-coreR/session_08.html). NCEAS Learning Hub, 2023.\n- NCEAS Learning Hub's coreR Course, [Chapter 16: Writing Functions & Packages](https://learning.nceas.ucsb.edu/2023-10-coreR/session_16.html). NCEAS Learning Hub, 2023.\n- Open Science Synthesis for the Delta Science Program's [Data Munging / QA / QC / Cleaning](https://nceas.github.io/oss-lessons/data-munging-qa-qc-cleaning/data-munging-qa-qc-cleaning.html)\n\n### Websites\n\n- [Ten Commandments for Good Data Management](https://dynamicecology.wordpress.com/2016/08/22/ten-commandments-for-good-data-management/)\n", + "markdown": "---\ntitle: \"Data Harmonization & Wrangling\"\ncode-annotations: hover\n---\n\n\n## Overview\n\nNow that we have covered how to find data and use data visualization methods to explore it, we can move on to combining separate data files and preparing that combined data file for analysis. For the purposes of this module we're adopting a very narrow view of harmonization and a very broad view of wrangling but this distinction aligns well with two discrete philosophical/practical arenas. To make those definitions explicit:\n\n- \"Harmonization\" = process of combining separate primary data objects into one object. This includes things like synonymizing columns, or changing data format to support combination. This _excludes_ quality control steps--even those that are undertaken before harmonization begins.\n\n- \"Wrangling\" = all modifications to data meant to create an analysis-ready 'tidy' data object. This includes quality control, unit conversions, and data 'shape' changes to name a few. Note that attaching ancillary data to your primary data object (e.g., attaching temperature data to a dataset on plant species composition) _also falls into this category!_\n\n## Learning Objectives\n\nAfter completing this module you will be able to: \n\n- Identify typical steps in data harmonization and wrangling workflows\n- Create a harmonization workflow\n- Define quality control\n- Summarize typical operations in a quality control workflow\n- Use regular expressions to perform flexible text operations\n- Write custom functions to reduce code duplication\n- Identify value of and typical obstacles to data 'joining'\n- Explain benefits and drawbacks of using data shape to streamline code\n- Design a complete data wrangling workflow\n\n## Needed Packages\n\nIf you'd like to follow along with the code chunks included throughout this module, you'll need to install the following packages:\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Note that these lines only need to be run once per computer\n## So you can skip this step if you've installed these before\ninstall.packages(\"ltertools\")\ninstall.packages(\"lterdatasampler\")\ninstall.packages(\"psych\")\ninstall.packages(\"supportR\")\ninstall.packages(\"tidyverse\")\n```\n:::\n\n\nWe'll load the Tidyverse meta-package here to have access to many of its useful tools when we need them later.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load tidyverse\nlibrary(tidyverse)\n```\n:::\n\n\n## Harmonizing Data\n\nData harmonization is an interesting topic in that it is _vital_ for synthesis projects but only very rarely relevant for primary research. Synthesis projects must reckon with the data choices made by each team of original data collectors. These collectors may or may not have recorded their judgement calls (or indeed, any metadata) but before synthesis work can be meaningfully done these independent datasets must be made comparable to one another and combined.\n\nFor tabular data, we recommend using the [`ltertools` R package](https://lter.github.io/ltertools/) to perform any needed harmonization. This package relies on a \"column key\" to translate the original column names into equivalents that apply across all datasets. Users can generate this column key however they would like but Google Sheets is a strong option as it allows multiple synthesis team members to simultaneously work on filling in the needed bits of the key. If you already have a set of files locally, `ltertools` does offer a `begin_key` function that creates the first two required columns in the column key.\n\nThe column key requires three columns:\n\n1. \"source\" -- Name of the raw file\n2. \"raw_name\" -- Name of all raw columns in that file to be synonymized\n3. \"tidy_name\" -- New name for each raw column that should be carried to the harmonized data\n\nNote that any raw names either not included in the column key or that lack a tidy name equivalent will be excluded from the final data object. For more information, consult the `ltertools` [package vignette](https://lter.github.io/ltertools/articles/ltertools.html). For convenience, we're attaching the visual diagram of this method of harmonization from the package vignette.\n\n

\n\"Four\n

\n\n## Wrangling Data\n\nData wrangling is a _huge_ subject that covers a wide range of topics. In this part of the module, we'll attempt to touch on a wide range of tools that may prove valuable to your data wrangling efforts. This is certainly non-exhaustive and you'll likely find new tools that fit your coding style and professional intuition better. However, hopefully the topics covered below provide a nice 'jumping off' point to reproducibly prepare your data for analysis and visualization work later in the lifecycle of the project.\n\nTo begin, we'll load the Plum Island Ecosystems fiddler crab dataset we've used in other modules.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load the lterdatasampler package\nlibrary(lterdatasampler)\n\n# Load the fiddler crab dataset\ndata(pie_crab)\n```\n:::\n\n\n### Exploring the Data\n\nBefore beginning any code operations, it's important to get a sense for the data. Characteristics like the dimensions of the dataset, the column names, and the type of information stored in each column are all crucial pre-requisites to knowing what tools can or should be used on the data.\n\nChecking the data structure is one way of getting a lot of this high-level information.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Check dataset structure\nstr(pie_crab)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\ntibble [392 × 9] (S3: tbl_df/tbl/data.frame)\n $ date : Date[1:392], format: \"2016-07-24\" \"2016-07-24\" ...\n $ latitude : num [1:392] 30 30 30 30 30 30 30 30 30 30 ...\n $ site : chr [1:392] \"GTM\" \"GTM\" \"GTM\" \"GTM\" ...\n $ size : num [1:392] 12.4 14.2 14.5 12.9 12.4 ...\n $ air_temp : num [1:392] 21.8 21.8 21.8 21.8 21.8 ...\n $ air_temp_sd : num [1:392] 6.39 6.39 6.39 6.39 6.39 ...\n $ water_temp : num [1:392] 24.5 24.5 24.5 24.5 24.5 ...\n $ water_temp_sd: num [1:392] 6.12 6.12 6.12 6.12 6.12 ...\n $ name : chr [1:392] \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" ...\n```\n\n\n:::\n:::\n\n\nFor data that are primarily numeric, you may find data summary functions to be valuable. Note that most functions of this type do not provide useful information on text columns so you'll need to find that information elsewhere.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Get a simple summary of the data\nsummary(pie_crab)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n date latitude site size \n Min. :2016-07-24 Min. :30.00 Length:392 Min. : 6.64 \n 1st Qu.:2016-07-28 1st Qu.:34.00 Class :character 1st Qu.:12.02 \n Median :2016-08-01 Median :39.10 Mode :character Median :14.44 \n Mean :2016-08-02 Mean :37.69 Mean :14.66 \n 3rd Qu.:2016-08-09 3rd Qu.:41.60 3rd Qu.:17.34 \n Max. :2016-08-13 Max. :42.70 Max. :23.43 \n air_temp air_temp_sd water_temp water_temp_sd \n Min. :10.29 Min. :6.391 Min. :13.98 Min. :4.838 \n 1st Qu.:12.05 1st Qu.:8.110 1st Qu.:14.33 1st Qu.:6.567 \n Median :13.93 Median :8.410 Median :17.50 Median :6.998 \n Mean :15.20 Mean :8.654 Mean :17.65 Mean :7.252 \n 3rd Qu.:18.63 3rd Qu.:9.483 3rd Qu.:20.54 3rd Qu.:7.865 \n Max. :21.79 Max. :9.965 Max. :24.50 Max. :9.121 \n name \n Length:392 \n Class :character \n Mode :character \n \n \n \n```\n\n\n:::\n:::\n\n\nFor text columns it can sometimes be useful to simply look at the unique entries in a given column and sort them alphabetically for ease of parsing.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Look at the sites included in the data\nsort(unique(pie_crab$site))\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n [1] \"BC\" \"CC\" \"CT\" \"DB\" \"GTM\" \"JC\" \"NB\" \"NIB\" \"PIE\" \"RC\" \"SI\" \"VCR\"\n[13] \"ZI\" \n```\n\n\n:::\n:::\n\n\nFor those of you who think more visually, a histogram can be a nice way of examining numeric data. There are simple histogram functions in the 'base' packages of most programming languages but it can sometimes be worth it to use those from special libraries because they can often convey additional detail.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\n# Load the psych library\nlibrary(psych)\n\n# Get the histogram of crab \"size\" (carapace width in mm)\npsych::multi.hist(pie_crab$size)\n```\n\n::: {.cell-output-display}\n![](mod_wrangle_files/figure-html/multi-hist-1.png){fig-align='center' width=384}\n:::\n:::\n\n\n:::{.callout-warning icon=\"false\"}\n#### Discussion: Data Exploration Tools\n\nWith a group of 4-5 others discuss the following questions:\n\n- What tools do you use when exploring new data?\n- Do you already use any of these tools to explore your data?\n - If you do, why do you use them?\n - If not, where do you think they might be valuable to include?\n- What value--if any--do you see in including these exploratory efforts in your code workflow?\n\n:::\n\n### Quality Control\n\nYou may have encountered the phrase \"QA/QC\" (Quality Assurance / Quality Control) in relation to data cleaning. Technically, quality assurance only encapsulates _preventative_ measures for reducing errors. One example of QA would be using a template for field datasheets because using standard fields reduces the risk that data are recorded inconsistently and/or incompletely. Quality control on the other hand refers to all steps taken to resolve errors _after_ data are collected. Any code that you write to fix typos or remove outliers from a dataset falls under the umbrella of QC.\n\nIn synthesis work, QA is only very rarely an option. You'll be working with datasets that have already been collected and attempting to handle any issues _post hoc_ which means the vast majority of data wrangling operations will be quality control methods. These QC efforts can be **incredibly** time-consuming so using a programming language (like R or Python) is a dramatic improvement over manually looking through the data using Microsoft Excel or other programs like it.\n\n#### QC Considerations\n\nThe datasets you gather for your synthesis project will likely have a multitude of issues you'll need to resolve before the data are ready for visualization or analysis. Some of these issues may be clearly identified in that datasets' metadata or apply to all datasets but it is good practice to make a thorough QC effort as early as is feasible. Keep the following data issues and/or checks in mind as we cover code tools that may be useful in this context later in the module.\n\n- Verify taxonomic classificiations against authorities\n - [ITIS](https://www.itis.gov/), [GBIF](https://www.gbif.org/), and [WoRMS](https://www.marinespecies.org/) are all examples of taxonomic authorities\n - Note that many of these authorities have R or Python libraries that can make this verification step scripted rather than dependent on manual searches\n- Handle missing data\n - Some datasets will use a code to indicate missing values (likely identified in their metadata) while others will just have empty cells\n- Check for unreasonable values / outliers\n - Can use conditionals to create \"flags\" for these values or just filter them out\n- Check geographic coordinates' reasonability\n - E.g., western hemisphere coordinates may lack the minus sign\n- Check date formatting\n - I.e., if all sampling is done in the first week of each month it can be difficult to say whether a given date is formatted as MM/DD/YY or DD/MM/YY\n- Consider spatial and temporal granularity among datasets\n - You may need to aggregate data from separate studies in different ways to ensure that the data are directly comparable across all of the data you gather\n- Handle duplicate data / rows\n\n#### Number Checking\n\nWhen you read in a dataset and a column that _should be_ numeric is instead read in as a character, it can be a sign that there are malformed numbers lurking in the background. Checking for and resolving these non-numbers is preferable to simply coercing the column into being numeric because the latter method typically changes those values to 'NA' where a human might be able to deduce the true number each value 'should be.'\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Load the supportR package\nlibrary(supportR)\n```\n\n::: {.cell-output .cell-output-stderr}\n\n```\n\nAttaching package: 'supportR'\n```\n\n\n:::\n\n::: {.cell-output .cell-output-stderr}\n\n```\nThe following object is masked from 'package:dplyr':\n\n count\n```\n\n\n:::\n\n```{.r .cell-code}\n# Create an example dataset with non-numbers in ideally numeric columns\nfish_ct <- data.frame(\"species\" = c(\"salmon\", \"bass\", \"halibut\", \"moray eel\"),\n \"count\" = c(12, \"14x\", \"_23\", 1))\n\n# Check for malformed numbers in column(s) that should be numeric\nbad_nums <- supportR::num_check(data = fish_ct, col = \"count\")\n```\n\n::: {.cell-output .cell-output-stderr}\n\n```\nFor 'count', 2 non-numbers identified: '14x' | '_23'\n```\n\n\n:::\n:::\n\n\nIn the above example, \"14x\" would be coerced to NA if you simply force the column without checking but you could drop the \"x\" with text replacing methods once you use tools like this one to flag it for your attention.\n\n#### Text Replacement\n\nOne of the simpler ways of handling text issues is just to replace a string with another string. Most programming languages support this functionality.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Use pattern match/replace to simplify problem entries\nfish_ct$count <- gsub(pattern = \"x|_\", replacement = \"\", x = fish_ct$count)\n\n# Check that they are fixed\nbad_nums <- supportR::num_check(data = fish_ct, col = \"count\")\n```\n\n::: {.cell-output .cell-output-stderr}\n\n```\nFor 'count', no non-numeric values identified.\n```\n\n\n:::\n:::\n\n\nThe vertical line in the `gsub` example above lets us search for (and replace) multiple patterns. Note however that while you can search for many patterns at once, only a single replacement value can be provided with this function.\n\n#### Regular Expressions\n\nYou may sometimes want to perform more generic string matching where you don't necessarily know--or want to list--all possible strings to find and replace. For instance, you may want remove any letter in a numeric column or find and replace numbers with some sort of text note. \"Regular expressions\" are how programmers specify these generic matches and using them can be a nice way of streamlining code.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Make a test vector\nregex_vec <- c(\"hello\", \"123\", \"goodbye\", \"456\")\n\n# Find all numbers and replace with the letter X\ngsub(pattern = \"[[:digit:]]\", replacement = \"x\", x = regex_vec)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n[1] \"hello\" \"xxx\" \"goodbye\" \"xxx\" \n```\n\n\n:::\n\n```{.r .cell-code}\n# Replace any number of letters with only a single 0\ngsub(pattern = \"[[:alpha:]]+\", replacement = \"0\", x = regex_vec)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n[1] \"0\" \"123\" \"0\" \"456\"\n```\n\n\n:::\n:::\n\n\nThe [`stringr` package cheatsheet](https://github.com/rstudio/cheatsheets/blob/afaa1fec4c5b9aabfa886218b6ba20317446d378/strings.pdf) has a really nice list of regular expression options that you may find valuable if you want to delve deeper on this topic. Scroll to the second page of the PDF to see the most relevant parts.\n\n### Conditionals\n\nRather than finding and replacing content, you may want to create a new column based on the contents of a different column. In plain language you might phrase this as 'if column X has \\[some values\\] then column Y should have \\[other values\\]'. These operations are called conditionals and are an important part of data wrangling.\n\nIf you only want your conditional to support two outcomes (as in an either/or statement) there are useful functions that support this. Let's return to our Plum Island Ecosystems crab dataset for an example.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Make a new colum with an either/or conditional\npie_crab_v2 <- pie_crab %>% \n dplyr::mutate(size_category = ifelse(test = (size >= 15), # <1>\n yes = \"big\",\n no = \"small\"),\n .after = size) \n\n# Count the number of crabs in each category\npie_crab_v2 %>% \n dplyr::group_by(size_category) %>% \n dplyr::summarize(crab_ct = dplyr::n())\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n# A tibble: 2 × 2\n size_category crab_ct\n \n1 big 179\n2 small 213\n```\n\n\n:::\n:::\n\n1. `mutate` makes a new column, `ifelse` is actually doing the conditional\n\nIf you have multiple different conditions you _can_ just stack these either/or conditionals together but this gets cumbersome quickly. It is preferable to instead use a function that supports as many alternates as you want!\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Make a new column with several conditionals\npie_crab_v2 <- pie_crab %>% \n dplyr::mutate(size_category = dplyr::case_when( \n size <= 10 ~ \"tiny\", # <1>\n size > 10 & size <= 15 ~ \"small\",\n size > 15 & size <= 20 ~ \"big\",\n size > 20 ~ \"huge\",\n TRUE ~ \"uncategorized\"), # <2>\n .after = size)\n\n# Count the number of crabs in each category\npie_crab_v2 %>% \n dplyr::group_by(size_category) %>% \n dplyr::summarize(crab_ct = dplyr::n())\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n# A tibble: 4 × 2\n size_category crab_ct\n \n1 big 150\n2 huge 28\n3 small 178\n4 tiny 36\n```\n\n\n:::\n:::\n\n1. Syntax is 'test ~ what to do when true'\n2. This line is a catch-all for any rows that _don't_ meet previous conditions\n\nNote that you can use functions like this one when you do have an either/or conditional if you prefer this format.\n\n:::{.callout-note icon=\"false\"}\n#### Activity: Conditionals\n\nIn a script, attempt the following with the PIE crab data:\n\n- Create a column indicating when air temperature is above or below 13° Fahrenheit\n- Create a column indicating whether water temperature is lower than the first quartile, between the first quartile and the median water temp, between the median and the third quartile or greater than the third quartile\n - _Hint:_ consult the `summary` function output!\n\n:::\n\n### Uniting / Separating Columns\n\nSometimes one column has multiple pieces of information that you'd like to consider separately. A date column is a common example of this because particular months are always in a given season regardless of the specific day or year. So, it can be useful to break a complete date (i.e., year/month/day) into its component bits to be better able to access those pieces of information.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Split date into each piece of temporal info\npie_crab_v3 <- pie_crab_v2 %>% \n tidyr::separate_wider_delim(cols = date, \n delim = \"-\", # <1>\n names = c(\"year\", \"month\", \"day\"),\n cols_remove = TRUE) # <2>\n\n# Check that out\nstr(pie_crab_v3)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\ntibble [392 × 12] (S3: tbl_df/tbl/data.frame)\n $ year : chr [1:392] \"2016\" \"2016\" \"2016\" \"2016\" ...\n $ month : chr [1:392] \"07\" \"07\" \"07\" \"07\" ...\n $ day : chr [1:392] \"24\" \"24\" \"24\" \"24\" ...\n $ latitude : num [1:392] 30 30 30 30 30 30 30 30 30 30 ...\n $ site : chr [1:392] \"GTM\" \"GTM\" \"GTM\" \"GTM\" ...\n $ size : num [1:392] 12.4 14.2 14.5 12.9 12.4 ...\n $ size_category: chr [1:392] \"small\" \"small\" \"small\" \"small\" ...\n $ air_temp : num [1:392] 21.8 21.8 21.8 21.8 21.8 ...\n $ air_temp_sd : num [1:392] 6.39 6.39 6.39 6.39 6.39 ...\n $ water_temp : num [1:392] 24.5 24.5 24.5 24.5 24.5 ...\n $ water_temp_sd: num [1:392] 6.12 6.12 6.12 6.12 6.12 ...\n $ name : chr [1:392] \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" ...\n```\n\n\n:::\n:::\n\n1. 'delim' is short for \"delimiter\" which we covered in the Reproducibility module\n2. This argument specifies whether to remove the original column when making the new columns\n\nWhile breaking apart a column's contents can be useful, it can also be helpful to combine the contents of several columns!\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Re-combine data information back into date\npie_crab_v4 <- pie_crab_v3 %>% \n tidyr::unite(col = \"date\",\n sep = \"/\", # <1>\n year:day, \n remove = FALSE) # <2>\n\n# Structure check\nstr(pie_crab_v4)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\ntibble [392 × 13] (S3: tbl_df/tbl/data.frame)\n $ date : chr [1:392] \"2016/07/24\" \"2016/07/24\" \"2016/07/24\" \"2016/07/24\" ...\n $ year : chr [1:392] \"2016\" \"2016\" \"2016\" \"2016\" ...\n $ month : chr [1:392] \"07\" \"07\" \"07\" \"07\" ...\n $ day : chr [1:392] \"24\" \"24\" \"24\" \"24\" ...\n $ latitude : num [1:392] 30 30 30 30 30 30 30 30 30 30 ...\n $ site : chr [1:392] \"GTM\" \"GTM\" \"GTM\" \"GTM\" ...\n $ size : num [1:392] 12.4 14.2 14.5 12.9 12.4 ...\n $ size_category: chr [1:392] \"small\" \"small\" \"small\" \"small\" ...\n $ air_temp : num [1:392] 21.8 21.8 21.8 21.8 21.8 ...\n $ air_temp_sd : num [1:392] 6.39 6.39 6.39 6.39 6.39 ...\n $ water_temp : num [1:392] 24.5 24.5 24.5 24.5 24.5 ...\n $ water_temp_sd: num [1:392] 6.12 6.12 6.12 6.12 6.12 ...\n $ name : chr [1:392] \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" \"Guana Tolomoto Matanzas NERR\" ...\n```\n\n\n:::\n:::\n\n1. This is equivalent to the 'delim' argument in the previous function\n2. Comparable to the 'cols_remove' argument in the previous function\n\nNote in this output how despite re-combining data information the column is listed as a character column! Simply combining or separating data is not always enough so you need to really lean into frequent data structure checks to be sure that your data are structured in the way that you want.\n\n### Joining Data\n\nOften the early steps of a synthesis project involve combining the data tables horizontally. You might imagine that you have two groups' data on sea star abundance and--once you've synonymized the column names--you can simply 'stack' the tables on top of one another. Slightly trickier but no less common is combining tables by the contents of a shared column (or columns). Cases like this include wanting to combine your sea star table with ocean temperature data from the region of each group's research. You can't simply attach the columns because that assumes that the row order is identical between the two data tables (and indeed, that there are the same number of rows in both to begin with!). In this case, if both data tables shared some columns (perhaps \"site\" and coordinate columns) you can use **joins** to let your computer match these key columns and make sure that only appropriate rows are combined.\n\nBecause joins are completely dependent upon the value in both columns being an _exact_ match, it is a good idea to carefully check the contents of those columns before attempting a join to make sure that the join will be successful.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Create a fish taxonomy dataframe that corresponds with the earlier fish dataframe\nfish_tax <- data.frame(\"species\" = c(\"salmon\", \"bass\", \"halibut\", \"eel\"),\n \"family\" = c(\"Salmonidae\", \"Serranidae\", \"Pleuronectidae\", \"Muraenidae\"))\n\n# Check to make sure that the 'species' column matches between both tables\nsupportR::diff_check(old = fish_ct$species, new = fish_tax$species) \n```\n\n::: {.cell-output .cell-output-stderr}\n\n```\nFollowing element(s) found in old object but not new: \n```\n\n\n:::\n\n::: {.cell-output .cell-output-stdout}\n\n```\n[1] \"moray eel\"\n```\n\n\n:::\n\n::: {.cell-output .cell-output-stderr}\n\n```\nFollowing element(s) found in new object but not old: \n```\n\n\n:::\n\n::: {.cell-output .cell-output-stdout}\n\n```\n[1] \"eel\"\n```\n\n\n:::\n:::\n\n::: {.cell}\n\n```{.r .cell-code}\n# Use text replacement methods to fix that mistake in one table\nfish_tax_v2 <- fish_tax %>% \n dplyr::mutate(species = gsub(pattern = \"^eel$\", # <1>\n replacement = \"moray eel\", \n x = species))\n\n# Re-check to make sure that fixed it\nsupportR::diff_check(old = fish_ct$species, new = fish_tax_v2$species)\n```\n\n::: {.cell-output .cell-output-stderr}\n\n```\nAll elements of old object found in new\n```\n\n\n:::\n\n::: {.cell-output .cell-output-stderr}\n\n```\nAll elements of new object found in old\n```\n\n\n:::\n:::\n\n1. The symbols around \"eel\" mean that we're only finding/replacing _exact_ matches. It doesn't matter in this context but often replacing a partial match would result in more problems. For example, replacing \"eel\" with \"moray eel\" could make \"electric eel\" into \"electric moray eel\".\n\nNow that the shared column matches between the two two dataframes we can use a join to combine them! There are four types of join:\n\n1. left/right join\n2. full join (a.k.a. outer join)\n3. inner join\n4. anti join\n\nYou can learn more about the types of join [here](https://nceas.github.io/scicomp-workshop-tidyverse/join.html) or [here](https://njlyon0.github.io/teach_r-for-biologists/materials/slides_4a.html#/title-slide) but the quick explanation is that the name of the join indicates whether the rows of the \"left\" and/or the \"right\" table are retained in the combined table. In synthesis work a left join or full join is most common (where you have your primary data in the left position and some ancillary/supplementary dataset in the right position).\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Let's combine the fish count and fish taxonomy information\nfish_df <- fish_ct %>% \n # Actual join step\n dplyr::left_join(y = fish_tax_v2, by = \"species\") %>% # <1>\n # Move 'family' column to the left of all other columns\n dplyr::relocate(family, .before = dplyr::everything())\n\n# Look at the result of that\nfish_df\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n family species count\n1 Salmonidae salmon 12\n2 Serranidae bass 14\n3 Pleuronectidae halibut 23\n4 Muraenidae moray eel 1\n```\n\n\n:::\n:::\n\n1. The 'by' argument accepts a vector of column names found in both data tables\n\n:::{.callout-note icon=\"false\"}\n#### Activity: Separating Columns & Joining Data\n\nIn a script, attempt the following with the PIE crab data:\n\n1. Create a data frame where you bin months into seasons (i.e., winter, spring, summer, fall)\n - Use your judgement on which month(s) should fall into each given PIE's latitude/location\n2. Join your season table to the PIE crab data based on month\n - _Hint:_ you may need to modify the PIE dataset to ensure both data tables share at least one column upon which they can be joined\n3. Calculate the average size of crabs in each season in order to identify which season correlates with the largest crabs\n\n:::\n\n### Leveraging Data Shape\n\nYou may already be familiar with data shape but fewer people recognize how playing with the shape of data can make certain operations _dramatically_ more efficient. If you haven't encountered it before, any data table can be said to have one of two 'shapes': either **long** or **wide**. Wide data have all measured variables from a single observation in one row (typically resulting in more columns than rows or \"wider\" data tables). Long data usually have one observation split into many rows (typically resulting in more rows than columns or \"longer\" data tables).\n\nData shape is often important for statistical analysis or visualization but it has an under-appreciated role to play in quality control efforts as well. If many columns have the shared criteria for what constitutes \"tidy\", you can reshape the data to get all of those values into a single column (i.e., reshape longer), perform any needed wrangling, then--when you're finished--reshape back into the original data shape (i.e., reshape wider). As opposed to applying the same operations repeatedly to each column individually.\n\nLet's consider an example to help clarify this. We'll simulate a butterfly dataset where both the number of different species and their sex were recorded in the same column. This makes the column not technically numeric and therefore unusable in analysis or visualization.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Generate a butterfly dataframe\nbfly_v1 <- data.frame(\"pasture\" = c(\"PNW\", \"PNW\", \"RCS\", \"RCS\"),\n \"monarch\" = c(\"14m\", \"10f\", \"7m\", \"16f\"),\n \"melissa_blue\" = c(\"32m\", \"2f\", \"6m\", \"0f\"),\n \"swallowtail\" = c(\"1m\", \"3f\", \"0m\", \"5f\"))\n\n# First we'll reshape this into long format\nbfly_v2 <- bfly_v1 %>% \n tidyr::pivot_longer(cols = -pasture, \n names_to = \"butterfly_sp\", \n values_to = \"count_sex\")\n\n# Check what that leaves us with\nhead(bfly_v2, n = 4)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n# A tibble: 4 × 3\n pasture butterfly_sp count_sex\n \n1 PNW monarch 14m \n2 PNW melissa_blue 32m \n3 PNW swallowtail 1m \n4 PNW monarch 10f \n```\n\n\n:::\n\n```{.r .cell-code}\n# Let's separate count from sex to be more usable later\nbfly_v3 <- bfly_v2 %>% \n tidyr::separate_wider_regex(cols = count_sex, \n c(count = \"[[:digit:]]+\", sex = \"[[:alpha:]]\")) %>% \n # Make the 'count' column a real number now\n dplyr::mutate(count = as.numeric(count))\n\n# Re-check output\nhead(bfly_v3, n = 4)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n# A tibble: 4 × 4\n pasture butterfly_sp count sex \n \n1 PNW monarch 14 m \n2 PNW melissa_blue 32 m \n3 PNW swallowtail 1 m \n4 PNW monarch 10 f \n```\n\n\n:::\n\n```{.r .cell-code}\n# Reshape back into wide-ish format\nbfly_v4 <- bfly_v3 %>% \n tidyr::pivot_wider(names_from = \"butterfly_sp\", values_from = count)\n\n# Re-re-check output\nhead(bfly_v4)\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n# A tibble: 4 × 5\n pasture sex monarch melissa_blue swallowtail\n \n1 PNW m 14 32 1\n2 PNW f 10 2 3\n3 RCS m 7 6 0\n4 RCS f 16 0 5\n```\n\n\n:::\n:::\n\n\nWhile we absolutely _could_ have used the same function to break apart count and butterfly sex data it would have involved copy/pasting the same information repeatedly. By pivoting to long format first, we can greatly streamline our code. This can also be advantageous for unit conversions, applying data transformations, or checking text column contents among many other possible applications.\n\n### Loops\n\nAnother way of simplfying repetitive operations is to use a \"for loop\" (often called simply \"loops\"). Loops allow you to iterate across a piece of code for a set number of times. Loops require you to define an \"index\" object that will change itself at the end of each iteration of the loop before beginning the next iteration. This index object's identity will be determined by whatever set of values you define at the top of the loop.\n\nHere's a very bare bones example to demonstrate the fundamentals.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Loop across each number between 2 and 4\nfor(k in 2:4){ # <1>\n \n # Square the number\n result <- k^2\n \n # Message that outside of the loop\n message(k, \" squared is \", result)\n} # <2>\n```\n\n::: {.cell-output .cell-output-stderr}\n\n```\n2 squared is 4\n```\n\n\n:::\n\n::: {.cell-output .cell-output-stderr}\n\n```\n3 squared is 9\n```\n\n\n:::\n\n::: {.cell-output .cell-output-stderr}\n\n```\n4 squared is 16\n```\n\n\n:::\n:::\n\n1. 'k' is our index object in this loop\n2. Note that the operations to iterate across are wrapped in curly braces (`{...}`)\n\nOnce you get the hang of loops, they can be a nice way of simplifying your code in a relatively human-readable way! Let's return to our Plum Island Ecosystems crab dataset for a more nuanced example.\n\n\n::: {.cell}\n\n```{.r .cell-code}\n# Create an empty list\ncrab_list <- list()\n\n# Let's loop across size categories of crab\nfor(focal_size in unique(pie_crab_v4$size_category)){ # <1>\n \n # Subset the data to just this size category\n focal_df <- pie_crab_v4 %>% \n dplyr::filter(size_category == focal_size)\n \n # Calculate average and standard deviation of size within this category\n size_avg <- mean(focal_df$size, na.rm = T) \n size_dev <- sd(focal_df$size, na.rm = T) \n \n # Assemble this into a data table and add to our list\n crab_list[[focal_size]] <- data.frame(\"size_category\" = focal_size,\n \"size_mean\" = size_avg,\n \"size_sd\" = size_dev)\n} # Close loop\n\n# Unlist the outputs into a dataframe\ncrab_df <- purrr::list_rbind(x = crab_list) # <2>\n\n# Check out the resulting data table\ncrab_df\n```\n\n::: {.cell-output .cell-output-stdout}\n\n```\n size_category size_mean size_sd\n1 small 12.624270 1.3827471\n2 tiny 8.876944 0.9112686\n3 big 17.238267 1.3650173\n4 huge 21.196786 0.8276744\n```\n\n\n:::\n:::\n\n1. Note that this is not the most efficient way of doing group-wise summarization but is--hopefully--a nice demonstration of loops!\n2. When all elements of your list have the same column names, `list_rbind` efficiently stacks those elements into one longer data table.\n\n### Custom Functions\n\nFinally, writing your own, customized functions can be really useful particularly when doing synthesis work. Custom functions are generally useful for reducing duplication and increasing ease of maintenance (see the note on custom functions in the SSECR [Reproducibility module](https://lter.github.io/ssecr/mod_reproducibility.html#consider-custom-functions)) and also can be useful end products of synthesis work in and of themselves. \n\nIf one of your group's outputs is a new standard data format or analytical workflow, the functions that you develop to aid yourself become valuable to anyone who adopts your synthesis project's findings into their own workflows. If you get enough functions you can even release a package that others can install and use on their own computers. Such packages are a valuable product of synthesis efforts and can be a nice addition to a robust scientific resume/CV.\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\n# Define custom function\ncrab_hist <- function(df, size_cat){\n \n # Subset data to the desired category\n data_sub <- dplyr::filter(.data = df, size_category == size_cat)\n \n # Create a histogram\n p <- psych::multi.hist(x = data_sub$size)\n}\n\n# Invoke function\ncrab_hist(df = pie_crab_v4, size_cat = \"tiny\")\n```\n\n::: {.cell-output-display}\n![](mod_wrangle_files/figure-html/custom-fxns-1.png){fig-align='center' width=384}\n:::\n:::\n\n\nWhen writing your own functions it can also be useful to program defensively. This involves anticipating likely errors and writing your own error messages that are more informative to the user than whatever machine-generated error would otherwise get generated\n\n\n::: {.cell layout-align=\"center\"}\n\n```{.r .cell-code}\n# Define custom function\ncrab_hist <- function(df, size_cat = \"small\"){ # <1>\n \n # Error out if 'df' isn't the right format\n if(is.data.frame(df) != TRUE) # <2>\n stop(\"'df' must be provided as a data frame\")\n \n # Error out if the data doesn't have the right columns\n if(all(c(\"size_category\", \"size\") %in% names(df)) != TRUE) # <3>\n stop(\"'df' must include a 'size' and 'size_category' column\")\n \n # Error out for unsupported size category values\n if(size_cat %in% unique(df$size_category) != TRUE)\n stop(\"Specified 'size_cat' not found in provided data\")\n \n # Subset data to the desired category\n data_sub <- dplyr::filter(.data = df, size_category == size_cat)\n \n # Create a histogram\n p <- psych::multi.hist(x = data_sub$size)\n}\n\n# Invoke new-and-improved function\ncrab_hist(df = pie_crab_v4) # <4>\n```\n\n::: {.cell-output-display}\n![](mod_wrangle_files/figure-html/custom-fxns-improved-1.png){fig-align='center' width=384}\n:::\n:::\n\n1. The default category is now set to \"small\"\n2. We recommend phrasing your error checks with this format (i.e., 'if \\ is _not_ true, then \\)\n3. The `%in%` operator lets you check whether one value matches any element of a set of accepted values. Very useful in contexts like this because the alternative would be a lot of separate \"or\" conditionals\n4. We don't need to specify the 'size_cat' argument because we can rely on the default\n\n:::{.callout-note icon=\"false\"}\n#### Activity: Custom Functions\n\nIn a script, attempt the following on the PIE crab data:\n\n- Write a function that:\n - (A) calculates the median of the user-supplied column\n - (B) determines whether each value is above, equal to, or below the median\n - (C) makes a column indicating the results of step B\n- Use the function on the _standard deviation_ of water temperature\n- Use it again on the standard deviation of air temperature\n- Revisit your function and identify 2-3 likely errors\n- Write custom checks (and error messages) for the set of likely issues you just identified\n\n:::\n\n## Additional Resources\n\n### Papers & Documents\n\n- Todd-Brown, K.E.O. _et al._ [Reviews and Syntheses: The Promise of Big Diverse Soil Data, Moving Current Practices Towards Future Potential](https://bg.copernicus.org/articles/19/3505/2022/bg-19-3505-2022.html). **2022**. _Biogeosciences_\n- Elgarby, O. [The Ultimate Guide to Data Cleaning](https://towardsdatascience.com/the-ultimate-guide-to-data-cleaning-3969843991d4). **2019**. _Medium_\n- Borer, E. _et al._ [Some Simple Guidelines for Effective Data Management](https://esajournals.onlinelibrary.wiley.com/doi/full/10.1890/0012-9623-90.2.205). **2009**. _Ecological Society of America Bulletin_\n\n### Workshops & Courses\n\n\n- The Carpentries. [Data Analysis and Visualization in R for Ecologists: Working with Data](https://datacarpentry.org/R-ecology-lesson/working-with-data.html). **2024**.\n- LTER Scientific Computing Team. [Coding in the Tidyverse](https://lter.github.io/workshop-tidyverse/). **2023**.\n- National Center for Ecological Analysis and Synthesis (NCEAS) Learning Hub [coreR: Cleaning & Wrangling Data](https://learning.nceas.ucsb.edu/2023-10-coreR/session_08.html). **2023**.\n- NCEAS Learning Hub [coreR: Writing Functions & Packages](https://learning.nceas.ucsb.edu/2023-10-coreR/session_16.html). **2023**.\n- Delta Science Program [Data Munging / QA / QC / Cleaning](https://nceas.github.io/oss-lessons/data-munging-qa-qc-cleaning/data-munging-qa-qc-cleaning.html). **2019**.\n\n### Websites\n\n- Fox, J. [Ten Commandments for Good Data Management](https://dynamicecology.wordpress.com/2016/08/22/ten-commandments-for-good-data-management/). **2016**. _Dynamic Ecology_\n", "supporting": [ "mod_wrangle_files" ], diff --git a/mod_credit.qmd b/mod_credit.qmd index 480ad4d..6778fff 100644 --- a/mod_credit.qmd +++ b/mod_credit.qmd @@ -23,9 +23,9 @@ After completing this module you will be able to: ### Papers & Documents -- Puebla _et al._ [Ten simple rules for recognizing data and software contributions in hiring, promotion, and tenure](https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1012296). **2024**. _PLoS Computational Biology_ -- Dahlin _et al._, [Hear Every Voice, Working Groups that Work](https://esajournals.onlinelibrary.wiley.com/doi/10.1002/fee.2115). **2019**. _Frontiers in Ecology and the Environment_ -- Allen _et al._, [How Can We Ensure Visibility and Diversity in Research Contributions? How the Contributor Role Taxonomy (CReDiT) is Helping the Shift from Authorship to Contributorship](https://onlinelibrary.wiley.com/doi/epdf/10.1002/leap.1210). **2018**. _Learned Publishing_ +- Puebla, I. _et al._ [Ten simple rules for recognizing data and software contributions in hiring, promotion, and tenure](https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1012296). **2024**. _PLoS Computational Biology_ +- Dahlin, K.M. _et al._, [Hear Every Voice, Working Groups that Work](https://esajournals.onlinelibrary.wiley.com/doi/10.1002/fee.2115). **2019**. _Frontiers in Ecology and the Environment_ +- Allen, L. _et al._, [How Can We Ensure Visibility and Diversity in Research Contributions? How the Contributor Role Taxonomy (CReDiT) is Helping the Shift from Authorship to Contributorship](https://onlinelibrary.wiley.com/doi/epdf/10.1002/leap.1210). **2018**. _Learned Publishing_ ### Workshops & Courses diff --git a/mod_data-disc.qmd b/mod_data-disc.qmd index 17316b1..f4fa89f 100644 --- a/mod_data-disc.qmd +++ b/mod_data-disc.qmd @@ -25,14 +25,14 @@ After completing this module you will be able to: ### Papers & Documents -- The British Ecological Society's [Better Science Guides](https://www.britishecologicalsociety.org/publications/better-science/) -- Data Management Guide +- British Ecological Society (BES). [Better Science Guides: Data Management Guide ](https://www.britishecologicalsociety.org/publications/better-science/). **2024**. ### Workshops & Courses -- NCEAS coreR [Data Management Essentials](https://learning.nceas.ucsb.edu/2023-10-coreR/session_14.html) lesson -- Open Science Synthesis for the Delta Science Program's [Data Management Essentials and the FAIR & CARE Principles](https://learning.nceas.ucsb.edu/2023-09-ucsb-faculty/session_04.html) -- Open Science Synthesis for the Delta Science Program's [Writing Data Management Plans](https://learning.nceas.ucsb.edu/2023-09-ucsb-faculty/session_05.html) -- NCEAS Scientific Computing team's [Data Acquisition](https://nceas.github.io/scicomp.github.io/data-acquisition.html) instructions +- LTER Scientific Computing Team. [Data Acquisition Guide for TRY and AppEEARS](https://lter.github.io/scicomp/internal_get-data.html). **2024**. +- National Center for Ecological Analysis and Synthesis (NCEAS) Learning Hub. [coreR: Data Management Essentials](https://learning.nceas.ucsb.edu/2023-10-coreR/session_14.html). **2023**. +- NCEAS Learning Hub. [UCSB Faculty Seminar Series: Data Management Essentials and the FAIR & CARE Principles](https://learning.nceas.ucsb.edu/2023-09-ucsb-faculty/session_04.html). **2023**. +- NCEAS Learning Hub. [UCSB Faculty Seminar Series: Writing Data Management Plans](https://learning.nceas.ucsb.edu/2023-09-ucsb-faculty/session_05.html). **2023**. ### Websites diff --git a/mod_data-viz.qmd b/mod_data-viz.qmd index 3e054d4..851d963 100644 --- a/mod_data-viz.qmd +++ b/mod_data-viz.qmd @@ -565,13 +565,13 @@ Check out the [bonus spatial data module](https://lter.github.io/ssecr/mod_spati ### Papers & Documents -- NCEAS [Colorblind Safe Color Schemes](https://www.nceas.ucsb.edu/sites/default/files/2022-06/Colorblind%20Safe%20Color%20Schemes.pdf) reference document +- National Center for Ecological Analysis and Synthesis (NCEAS). [Colorblind Safe Color Schemes](https://www.nceas.ucsb.edu/sites/default/files/2022-06/Colorblind%20Safe%20Color%20Schemes.pdf). **2022**. ### Workshops & Courses -- NCEAS Scientific Computing team's Coding in the Tidyverse workshop [`ggplot2` module](https://nceas.github.io/scicomp-workshop-tidyverse/visualize.html) -- The Carpentries' Data Analysis and Visualization in R for Ecologists [`ggplot2` episode](https://datacarpentry.org/R-ecology-lesson/04-visualization-ggplot2.html) - +- The Carpentries. [Data Analysis and Visualization in R for Ecologists: Data Visualization with `ggplot2`](https://datacarpentry.org/R-ecology-lesson/visualizing-ggplot.html). **2024**. +- The Carpentries. [Data Analysis and Visualization in Python for Ecologists: Making Plots with `plotnine`](https://datacarpentry.org/python-ecology-lesson/07-visualization-ggplot-python.html). **2024**. +- LTER Scientific Computing Team. [Coding in the Tidyverse: 'Visualize' Module](https://lter.github.io/workshop-tidyverse/visualize.html). **2023**. ### Websites diff --git a/mod_facilitation.qmd b/mod_facilitation.qmd index 3d22bb7..9772910 100644 --- a/mod_facilitation.qmd +++ b/mod_facilitation.qmd @@ -23,7 +23,7 @@ After completing this module you will be able to: ### Papers & Documents -- _[Facilitator's Guide to Participatory Decision-Making](https://bookshop.org/p/books/facilitator-s-guide-to-participatory-decision-making-sam-kaner/11179446?ean=9781118404959)_ by Sam Kaner +- Kaner, S. [Facilitator's Guide to Participatory Decision-Making (Revised)](https://bookshop.org/p/books/facilitator-s-guide-to-participatory-decision-making-sam-kaner/11179446?ean=9781118404959). **2014**. ### Workshops & Courses @@ -31,4 +31,4 @@ After completing this module you will be able to: ### Websites -- [Liberating Structures](https://www.liberatingstructures.com/) website +- [Liberating Structures: Including and Uleashing Everyone](https://www.liberatingstructures.com/) diff --git a/mod_findings.qmd b/mod_findings.qmd index 8d1eede..2029b82 100644 --- a/mod_findings.qmd +++ b/mod_findings.qmd @@ -35,8 +35,9 @@ After completing this module you will be able to: ### Workshops & Courses -- Open Science Synthesis for the Delta Science Program's [Communicating Your Science](https://learning.nceas.ucsb.edu/2023-10-delta/session_02.html) + +- NCEAS Learning Hub & Delta Stewardship Council. [Open Science Synthesis: Communicating Your Science ](https://learning.nceas.ucsb.edu/2023-10-delta/session_02.html). **2023**. ### Websites -- Compass' [The Message Box](https://www.compassscicomm.org/leadership-development/the-message-box/) +- COMPASS. [The Message Box](https://www.compassscicomm.org/leadership-development/the-message-box/) diff --git a/mod_interactivity.qmd b/mod_interactivity.qmd index 1cd1b85..740e16c 100644 --- a/mod_interactivity.qmd +++ b/mod_interactivity.qmd @@ -330,9 +330,9 @@ Take a look at [Posit's instructions for deployment](https://shiny.posit.co/r/ar ### Workshops & Courses -- Posit's [Welcome to Shiny](https://shiny.posit.co/r/getstarted/shiny-basics/lesson1/index.html) (for R coders) -- 2022 All Scientists' Meeting [Shiny Apps for Sharing Science](https://njlyon0.github.io/asm-2022_shiny-workshop/) workshop +- Posit. [Welcome to Shiny](https://shiny.posit.co/r/getstarted/shiny-basics/lesson1/index.html). **2024**. +- Lyon, N.J., _et al._ [Shiny Apps for Sharing Science](https://njlyon0.github.io/asm-2022_shiny-workshop/). **2022**. ### Websites -- Posit's [Shiny](https://shiny.posit.co/) website +- diff --git a/mod_next-steps.qmd b/mod_next-steps.qmd index 10debaf..b03577c 100644 --- a/mod_next-steps.qmd +++ b/mod_next-steps.qmd @@ -22,7 +22,7 @@ After completing this module you will be able to: ### Papers & Documents -- Deutsch _et al._, [Leading Inter- and Transdisciplinary Research: Lessons from Applying Theories of Change to a Strategic Research Program](https://www.sciencedirect.com/science/article/pii/S146290112100054X?via%3Dihub). **2021**. _Environmental Science & Policy_ +- Deutsch L., _et al._, [Leading Inter- and Transdisciplinary Research: Lessons from Applying Theories of Change to a Strategic Research Program](https://www.sciencedirect.com/science/article/pii/S146290112100054X?via%3Dihub). **2021**. _Environmental Science & Policy_ ### Workshops & Courses @@ -30,4 +30,4 @@ After completing this module you will be able to: ### Websites -- Shabanov, I. (\@Artifexx) [X post on logic models](https://x.com/artifexx/status/1798936726788051403?s=51&t=TCar5jaRTW3NNuD_rXT8Pg) +- Shabanov, I. (\@Artifexx) [Tweet on logic models](https://x.com/artifexx/status/1798936726788051403?s=51&t=TCar5jaRTW3NNuD_rXT8Pg). **2024**. diff --git a/mod_project-mgmt.qmd b/mod_project-mgmt.qmd index 0163adc..e00da85 100644 --- a/mod_project-mgmt.qmd +++ b/mod_project-mgmt.qmd @@ -27,10 +27,10 @@ After completing this module you will be able to: ### Workshops & Courses -- NCEAS Scientific Computing team's Collaborative Coding with GitHub workshop project management modules - - [GitHub Issues](https://nceas.github.io/scicomp-workshop-collaborative-coding/issues.html) - - [GitHub Projects](https://nceas.github.io/scicomp-workshop-collaborative-coding/projects.html) -- Open Science Synthesis for the Delta Science Program's [Logic Models and Synthesis Development](https://learning.nceas.ucsb.edu/2021-09-delta/session-8-hands-on-logic-models-and-synthesis-development.html) +- LTER Scientific Computing Team. [Collaborative Coding with GitHub: Issues](https://lter.github.io/workshop-github/issues.html). **2024**. +- LTER Scientific Computing Team. [Collaborative Coding with GitHub: Milestones](https://lter.github.io/workshop-github/milestones.html). **2024**. +- LTER Scientific Computing Team. [Collaborative Coding with GitHub: Projects](https://lter.github.io/workshop-github/projects.html). **2024**. +- NCEAS Learning Hub & Delta Stewardship Council. [Open Science Synthesis: Logic Models and Synthesis Development](https://learning.nceas.ucsb.edu/2021-09-delta/session-8-hands-on-logic-models-and-synthesis-development.html). **2021**. ### Websites diff --git a/mod_reports.qmd b/mod_reports.qmd index e6b8b4d..3a7028c 100644 --- a/mod_reports.qmd +++ b/mod_reports.qmd @@ -121,11 +121,10 @@ See [here](https://nceas.github.io/scicomp.github.io/best_practices.html#r-scrip ### Workshops & Courses -- NCEAS coreR [Literate Analysis with Quarto](https://learning.nceas.ucsb.edu/2023-10-coreR/session_04.html) session -- OSS [Reproducible Papers with RMarkdown](https://nceas.github.io/oss-lessons/reproducible-papers-with-rmd/reproducible-papers-with-rmd.html) -- UCSB's Master of Environmental Data Science (MEDS) [Creating your Personal Website using Quarto](https://ucsb-meds.github.io/creating-quarto-websites/) lesson +- Csik, S. [Creating Your Personal Website Using Quarto](https://ucsb-meds.github.io/creating-quarto-websites/). **2024**. +- National Center for Ecological Analysis and Synthesis (NCEAS) Learning Hub. [coreR: Literate Analysis with Quarto](https://learning.nceas.ucsb.edu/2023-10-coreR/session_04.html). **2023**. +- Mecum, B. [Reproducible Papers with RMarkdown](https://nceas.github.io/oss-lessons/reproducible-papers-with-rmd/reproducible-papers-with-rmd.html). **2019**. ### Websites - -- Markdown Guide: [Basic Syntax](https://www.markdownguide.org/basic-syntax/) -- Posit's [Welcome to Quarto](https://quarto.org/) +- [The Markdown Guide: Basic Syntax](https://www.markdownguide.org/basic-syntax/). +- Posit. [Welcome to Quarto](https://quarto.org/) diff --git a/mod_reproducibility.qmd b/mod_reproducibility.qmd index 17f3794..e958720 100644 --- a/mod_reproducibility.qmd +++ b/mod_reproducibility.qmd @@ -578,24 +578,23 @@ Making sure that your project is reproducible requires a handful of steps before ### Papers & Documents -- [A Large-Scale Study on Research Code Quality and Execution](https://www.nature.com/articles/s41597-022-01143-6). Trisovic _et al._, 2022. **Scientific Data** -- [Applying the 'CARE Principles for Indigenous Data Governance' to Ecology and Biodiversity Research](https://www.nature.com/articles/s41559-023-02161-2). Jennings _et al._, 2023. **Nature Ecology & Evolution** -- [Guides to Better Science - Reproducible Code](https://www.britishecologicalsociety.org/publications/better-science/). The British Ecological Society, 2024. -- [FAIR Teaching Handbook](https://fairsfair.gitbook.io/fair-teaching-handbook/). Englehardt _et al._, 2024. -- [R Packages](https://r-pkgs.org/) (2nd ed.). Wickham & Bryan. +- British Ecological Society (BES). [Better Science Guides: Reproducible Code](https://www.britishecologicalsociety.org/publications/better-science/). **2024**. +- Englehardt, C. _et al._ [FAIR Teaching Handbook](https://fairsfair.gitbook.io/fair-teaching-handbook/). **2024**. +- Jennings, L. _et al._ [Applying the 'CARE Principles for Indigenous Data Governance' to Ecology and Biodiversity Research](https://www.nature.com/articles/s41559-023-02161-2). **2023**. _Nature Ecology & Evolution_ +- Wickham, H. & Bryan, J. [R Packages](https://r-pkgs.org/) (2nd ed.). **2023**. +- Trisovic, A. _et al._ [A Large-Scale Study on Research Code Quality and Execution](https://www.nature.com/articles/s41597-022-01143-6). **2022**. _Scientific Data_ ### Workshops & Courses -- Data Analysis and Visualization in R for Ecologists, [Episode 1: Before We Start](https://datacarpentry.org/R-ecology-lesson/00-before-we-start.html). The Carpentries -- Introduction to R for Geospatial Data, [Episode 2: Project Management with RStudio](https://datacarpentry.org/r-intro-geospatial/02-project-intro.html). The Carpentries -- coreR Course, [Chapter 5: FAIR and CARE Principles](https://learning.nceas.ucsb.edu/2023-10-coreR/session_05.html). National Center for Ecological Analysis and Synthesis (NCEAS) Learning Hub, 2023. -- coreR Course, [Chapter 18: Reproducibility & Provenance](https://learning.nceas.ucsb.edu/2023-10-coreR/session_18.html). NCEAS Learning Hub, 2023. +- The Carpentries. [Data Analysis and Visualization in R for Ecologists: Before We Start](https://datacarpentry.org/R-ecology-lesson/00-before-we-start.html). **2024**. +- The Carpentries. [Introduction to R for Geospatial Data: Project Management with RStudio](https://datacarpentry.org/r-intro-geospatial/02-project-intro.html). **2024**. +- National Center for Ecological Analysis and Synthesis (NCEAS) Learning Hub. [coreR: FAIR and CARE Principles](https://learning.nceas.ucsb.edu/2023-10-coreR/session_05.html). **2023**. +- National Center for Ecological Analysis and Synthesis (NCEAS) Learning Hub. [coreR: Reproducibility & Provenance](https://learning.nceas.ucsb.edu/2023-10-coreR/session_18.html). **2023**. ### Websites -- [Coding Tips](https://nceas.github.io/scicomp.github.io/best_practices.html). NCEAS Scientific Computing Team, 2024. -- [Documenting Things: Openly for Future Us](https://openscapes.github.io/documenting-things/#/title-slide). Lowndes _et al.,_ 2023. -- [Team Coding: 5 Essentials](https://nceas.github.io/scicomp.github.io/onboard-scaffold_team_coding.html). NCEAS Scientific Computing Team, 2024. -- Advanced R (1st ed.) [Style Guide](http://adv-r.had.co.nz/Style.html). Wickham -- PEP 8 [Style Guide for Python Code](https://peps.python.org/pep-0008/). van Rossum _et al._ 2013. -- [Google Style Guides](https://google.github.io/styleguide/) +- Google. [Style Guide](http://adv-r.had.co.nz/Style.html). **2024**. +- LTER Scientific Computing Team. [Team Coding: 5 Essentials](https://lter.github.io/scicomp/wg_team-coding.html). **2024**. +- Lowndes, J.S. _et al._ [Documenting Things: Openly for Future Us](https://openscapes.github.io/documenting-things/#/title-slide). **2023**. _posit::conf(2023)_ +- Wickham, H. [Advanced R: Style Guide](http://adv-r.had.co.nz/Style.html). (1st ed.). **2019**. +- van Rossum, G. _et al._ [PEP 8: Style Guide for Python Code](https://peps.python.org/pep-0008/). **2013**. _Python Enhancement Proposals_ diff --git a/mod_spatial.qmd b/mod_spatial.qmd index c246765..f421fa9 100644 --- a/mod_spatial.qmd +++ b/mod_spatial.qmd @@ -297,12 +297,12 @@ In the above output we can see that it has extracted the elevation of _every pix ### Workshops & Courses -- The Carpentries' [Introduction to Geospatial Raster & Vector Data with R](https://datacarpentry.org/r-raster-vector-geospatial/) -- The Carpentries' [Introduction to R for Geospatial Data](https://datacarpentry.org/r-intro-geospatial/index.html) -- Arctic Data Center's [Spatial and Image Data Using GeoPandas](https://learning.nceas.ucsb.edu/2023-03-arctic/sections/geopandas.html) chapter of their Scalable Computing course -- Jason Flower's (UC Santa Barbara) [Introduction to rasters with `terra`](https://jflowernet.github.io/intro-terra-ecodatascience/) -- King, R. [Spatial Data Visualization](https://github.com/king0708/spatial-data-viz) workshop +- The Carpentries. [Introduction to Geospatial Raster and Vector Data with R](https://datacarpentry.org/r-raster-vector-geospatial/). **2024**. +- The Carpentries. [Introduction to R for Geospatial Data](https://datacarpentry.org/r-intro-geospatial/index.html). **2024**. +- King, R. [Spatial Data Visualization](https://github.com/king0708/spatial-data-viz). **2024**. +- Flower, J. [Introduction to Rasters with `terra`](https://jflowernet.github.io/intro-terra-ecodatascience/). **2024**. +- Clark, S.J., _et al._ [Spatial and Image Data Using GeoPandas](https://learning.nceas.ucsb.edu/2023-03-arctic/sections/geopandas.html). **2023**. ### Websites -- NASA's Application for Extracting and Exploring Analysis Ready Samples [(AppEEARS) Portal](https://appeears.earthdatacloud.nasa.gov/) +- NASA. [AppEEARS Portal](https://appeears.earthdatacloud.nasa.gov/) diff --git a/mod_stats.qmd b/mod_stats.qmd index 5a48d59..f5b359f 100644 --- a/mod_stats.qmd +++ b/mod_stats.qmd @@ -337,18 +337,18 @@ After you've calculated all relevant effect sizes--using your chosen flavor of e ### Papers & Documents -- [Understanding ‘It Depends’ in Ecology: A Guide to Hypothesising, Visualising and Interpreting Statistical Interactions](https://onlinelibrary.wiley.com/doi/10.1111/brv.12939). Spake _et al._, 2023. **Biological Reviews** -- [Improving Quantitative Synthesis to Achieve Generality in Ecology](https://www.nature.com/articles/s41559-022-01891-z). Spake _et al._, 2022.**Nature Ecology and Evolution** -- [Doing Meta-Analysis with R: A Hands-On Guide](https://bookdown.org/MathiasHarrer/Doing_Meta_Analysis_in_R/). Harrier _et al._ 2023. -- [Mixed Effects Models and Extensions in Ecology with R](https://link.springer.com/book/10.1007/978-0-387-87458-6). Zuur _et al._, 2009. +- Spake, R. _et al._ [Understanding 'It Depends' in Ecology: A Guide to Hypothesising, Visualising and Interpreting Statistical Interactions](https://onlinelibrary.wiley.com/doi/10.1111/brv.12939). **2023**. _Biological Reviews_ +- Spake, R. _et al._ [Improving Quantitative Synthesis to Achieve Generality in Ecology](https://www.nature.com/articles/s41559-022-01891-z). **2022**. _Nature Ecology and Evolution_ +- Harrier, M. _et al._ [Doing Meta-Analysis with R: A Hands-On Guide](https://bookdown.org/MathiasHarrer/Doing_Meta_Analysis_in_R/). **2021**. +- Zuur, A.F. _et al._ [Mixed Effects Models and Extensions in Ecology with R](https://link.springer.com/book/10.1007/978-0-387-87458-6). **2009**. ### Workshops & Courses -- Matt Vuorre's [Bayesian Meta-Analysis with R, Stan, and brms](https://mvuorre.github.io/posts/2016-09-29-bayesian-meta-analysis/) +- Vuorre, M. [Bayesian Meta-Analysis with R, Stan, and brms](https://mvuorre.github.io/posts/2016-09-29-bayesian-meta-analysis/). **2016**. ### Websites -- [Bayesian Meta-Analysis in brms-II](https://solomonkurz.netlify.app/blog/2020-10-16-bayesian-meta-analysis-in-brms-ii/). Solomon Kurz, A., 2022. -- [Meta-Analysis with R](https://github.com/wviechtb/workshop_2022_ma_esmarconf/blob/master/workshop_meta-analysis.pdf) slide deck. Viechtbauer, W., 2022. -- [The `metafor` Package: A Meta-Analysis Package for R](https://www.metafor-project.org/doku.php/tips:rma_vs_lm_lme_lmer?s%5B%5D=lme4). Viechtbauer, W., 2021. -- [Bayesian Meta-Analysis in brms](https://solomonkurz.netlify.app/blog/bayesian-meta-analysis/). Solomon Kurz, A., 2018. +- Kurz, A. S. [Bayesian Meta-Analysis in brms-II](https://solomonkurz.netlify.app/blog/2020-10-16-bayesian-meta-analysis-in-brms-ii/). **2022**. +- Viechtbauer, W. [Meta-Analysis with R](https://github.com/wviechtb/workshop_2022_ma_esmarconf/blob/master/workshop_meta-analysis.pdf). **2022**. +- Viechtbauer, W. [The `metafor` Package: A Meta-Analysis Package for R](https://www.metafor-project.org/doku.php/tips:rma_vs_lm_lme_lmer?s%5B%5D=lme4). **2021**. +- Kurz, A. S. [Bayesian Meta-Analysis in brms](https://solomonkurz.netlify.app/blog/bayesian-meta-analysis/). **2020**. diff --git a/mod_team-sci.qmd b/mod_team-sci.qmd index c3c8072..f652a8d 100644 --- a/mod_team-sci.qmd +++ b/mod_team-sci.qmd @@ -23,18 +23,18 @@ After completing this module you will be able to: ### Papers & Documents -- Bates _et al._, [Overcome Imposter Syndrome: Contribute to Working Groups and Build Strong Networks](https://www.sciencedirect.com/science/article/pii/S0006320724001289?via%3Dihub). **2024**. _Biological Conservation_ +- Bates, A.E. _et al._, [Overcome Imposter Syndrome: Contribute to Working Groups and Build Strong Networks](https://www.sciencedirect.com/science/article/pii/S0006320724001289?via%3Dihub). **2024**. _Biological Conservation_ - Peterson, D.M., _et al._, [Team Science: A Syllabus for Success on Big Projects](https://onlinelibrary.wiley.com/doi/10.1002/ece3.10343). **2023**. _Ecology and Evolution_ - Gaynor, K.M., _et al._, [Ten Simple Rules to Cultivate Belonging in Collaborative Data Science Research Teams](https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1010567). **2022**. _PLoS Computational Biology_ - Deutsch, L., _et al._, [Leading Inter- and Transdisciplinary Research: Lessons from Applying Theories of Change to a Strategic Research Program](https://www.sciencedirect.com/science/article/pii/S146290112100054X?via%3Dihub). **2021**. _Environmental Science & Policy_ -- Farrell _et al._, [Training Macrosystems Scientists Requires Both Interpersonal and Technical Skills](https://esajournals.onlinelibrary.wiley.com/doi/10.1002/fee.2287). **2021**. _Frontiers in Ecology and the Environment_ -- Hampton & Parker, [Collaboration and Productivity in Scientific Synthesis](https://academic.oup.com/bioscience/article/61/11/900/223655). **2011**. _BioScience_ +- Farrell, K.J. _et al._, [Training Macrosystems Scientists Requires Both Interpersonal and Technical Skills](https://esajournals.onlinelibrary.wiley.com/doi/10.1002/fee.2287). **2021**. _Frontiers in Ecology and the Environment_ +- Hampton, S.E. & Parker, J.N. [Collaboration and Productivity in Scientific Synthesis](https://academic.oup.com/bioscience/article/61/11/900/223655). **2011**. _BioScience_ ### Workshops & Courses -- Amelia Liberatore's [Developing a Successful Team](https://docs.google.com/presentation/d/1xbT59ULxJhV-QF4FAeU7EYkOGGzzrIjTlDEULlNyerw/edit#slide=id.p) workshop slides -- Open Science Synthesis for the Delta Science Program's [Team Science for Synthesis](https://learning.nceas.ucsb.edu/2023-09-ucsb-faculty/session_02.html) -- Open Science Synthesis for the Delta Science Program's [Thinking Preferences](https://learning.nceas.ucsb.edu/2021-09-delta/session-6-collaboration-practices.html#thinking-preferences) +- Liberatore, A. [Developing a Successful Team: Concepts and Strategies to Navigate Change and Conflict Together](https://docs.google.com/presentation/d/1xbT59ULxJhV-QF4FAeU7EYkOGGzzrIjTlDEULlNyerw/edit#slide=id.p). **2024**. +- National Center for Ecological Analysis and Synthesis (NCEAS) Learning Hub & Delta Stewardship Council. [Open Science Synthesis: Team Science for Synthesis](https://learning.nceas.ucsb.edu/2023-09-ucsb-faculty/session_02.html). **2023**. +- Delta Stewardship Council. [Open Science Synthesis: Thinking Preferences](https://learning.nceas.ucsb.edu/2021-09-delta/session-6-collaboration-practices.html#thinking-preferences). **2021**. ### Websites diff --git a/mod_thinking.qmd b/mod_thinking.qmd index 84be7b2..eca4772 100644 --- a/mod_thinking.qmd +++ b/mod_thinking.qmd @@ -20,7 +20,7 @@ After completing this module you will be able to: ### Papers & Documents -- _[Facilitator's Guide to Participatory Decision-Making](https://bookshop.org/p/books/facilitator-s-guide-to-participatory-decision-making-sam-kaner/11179446?ean=9781118404959)_ by Sam Kaner +- Kaner, S. [Facilitator's Guide to Participatory Decision-Making](https://bookshop.org/p/books/facilitator-s-guide-to-participatory-decision-making-sam-kaner/11179446?ean=9781118404959). **2014**. ### Workshops & Courses @@ -28,4 +28,4 @@ After completing this module you will be able to: ### Websites -- [Liberating Structures](https://www.liberatingstructures.com/) website +- [Liberating Structures: Including and Uleashing Everyone](https://www.liberatingstructures.com/) diff --git a/mod_version-control.qmd b/mod_version-control.qmd index fc58ea9..6b3441f 100644 --- a/mod_version-control.qmd +++ b/mod_version-control.qmd @@ -79,10 +79,10 @@ Practically-speaking this also encourages an atmosphere where only one person ca ### Papers & Documents -- Pereira Braga _et al._, [Not Just for Programmers: How GitHub can Accelerate Collaborative and Reproducible Research in Ecology and Evolution](https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.14108). **2023**. _Methods in Ecology and Evolution_ -- GitHub, [Git Cheat Sheet](https://education.github.com/git-cheat-sheet-education.pdf). **2023**. +- Pereira Braga, P.H. _et al._ [Not Just for Programmers: How GitHub can Accelerate Collaborative and Reproducible Research in Ecology and Evolution](https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.14108). **2023**. _Methods in Ecology and Evolution_ +- GitHub. [Git Cheat Sheet](https://education.github.com/git-cheat-sheet-education.pdf). **2023**. ### Workshops & Courses -- [Happy Git and GitHub for the useR](https://happygitwithr.com/). Bryan _et al._, 2024. -- coreR Course, [Chapter 6: Intro to Git and GitHub](https://learning.nceas.ucsb.edu/2023-10-coreR/session_06.html). NCEAS Learning Hub, 2023. +- Bryan J. _et al._ [Happy Git and GitHub for the useR](https://happygitwithr.com/). **2024**. +- National Center for Ecological Analysis and Synthesis (NCEAS) Learning Hub. [coreR: Intro to Git and GitHub](https://learning.nceas.ucsb.edu/2023-10-coreR/session_06.html). **2023**. diff --git a/mod_wrangle.qmd b/mod_wrangle.qmd index d5b4073..b8d57ab 100644 --- a/mod_wrangle.qmd +++ b/mod_wrangle.qmd @@ -543,18 +543,19 @@ In a script, attempt the following on the PIE crab data: ### Papers & Documents -- [Reviews and Syntheses: The Promise of Big Diverse Soil Data, Moving Current Practices Towards Future Potential](https://bg.copernicus.org/articles/19/3505/2022/bg-19-3505-2022.html). Todd-Brown, K.E.O. _et al._, 2022. **Biogeosciences** -- [The Ultimate Guide to Data Cleaning](https://towardsdatascience.com/the-ultimate-guide-to-data-cleaning-3969843991d4). Elgarby, O. 2019. **Medium** -- [Some Simple Guidelines for Effective Data Management](https://esajournals.onlinelibrary.wiley.com/doi/full/10.1890/0012-9623-90.2.205). Borer, E. _et al._, 2009. **Ecological Society of America Bulletin** +- Todd-Brown, K.E.O. _et al._ [Reviews and Syntheses: The Promise of Big Diverse Soil Data, Moving Current Practices Towards Future Potential](https://bg.copernicus.org/articles/19/3505/2022/bg-19-3505-2022.html). **2022**. _Biogeosciences_ +- Elgarby, O. [The Ultimate Guide to Data Cleaning](https://towardsdatascience.com/the-ultimate-guide-to-data-cleaning-3969843991d4). **2019**. _Medium_ +- Borer, E. _et al._ [Some Simple Guidelines for Effective Data Management](https://esajournals.onlinelibrary.wiley.com/doi/full/10.1890/0012-9623-90.2.205). **2009**. _Ecological Society of America Bulletin_ ### Workshops & Courses -- Data Analysis and Visualization in R for Ecologists, [Episode 4: Manipulating, Analyzing, and Exporting Data with `tidyverse`](https://datacarpentry.org/R-ecology-lesson/03-dplyr.html). The Carpentries -- [Coding in the Tidyverse](https://nceas.github.io/scicomp-workshop-tidyverse/). NCEAS Scientific Computing Team, 2023. -- NCEAS Learning Hub's coreR Course, [Chapter 8: Cleaning & Wrangling Data](https://learning.nceas.ucsb.edu/2023-10-coreR/session_08.html). NCEAS Learning Hub, 2023. -- NCEAS Learning Hub's coreR Course, [Chapter 16: Writing Functions & Packages](https://learning.nceas.ucsb.edu/2023-10-coreR/session_16.html). NCEAS Learning Hub, 2023. -- Open Science Synthesis for the Delta Science Program's [Data Munging / QA / QC / Cleaning](https://nceas.github.io/oss-lessons/data-munging-qa-qc-cleaning/data-munging-qa-qc-cleaning.html) + +- The Carpentries. [Data Analysis and Visualization in R for Ecologists: Working with Data](https://datacarpentry.org/R-ecology-lesson/working-with-data.html). **2024**. +- LTER Scientific Computing Team. [Coding in the Tidyverse](https://lter.github.io/workshop-tidyverse/). **2023**. +- National Center for Ecological Analysis and Synthesis (NCEAS) Learning Hub [coreR: Cleaning & Wrangling Data](https://learning.nceas.ucsb.edu/2023-10-coreR/session_08.html). **2023**. +- NCEAS Learning Hub [coreR: Writing Functions & Packages](https://learning.nceas.ucsb.edu/2023-10-coreR/session_16.html). **2023**. +- Delta Science Program [Data Munging / QA / QC / Cleaning](https://nceas.github.io/oss-lessons/data-munging-qa-qc-cleaning/data-munging-qa-qc-cleaning.html). **2019**. ### Websites -- [Ten Commandments for Good Data Management](https://dynamicecology.wordpress.com/2016/08/22/ten-commandments-for-good-data-management/) +- Fox, J. [Ten Commandments for Good Data Management](https://dynamicecology.wordpress.com/2016/08/22/ten-commandments-for-good-data-management/). **2016**. _Dynamic Ecology_