Add several python scripts and tests

Author: MatRouillard

.env

@@ -0,0 +1,3 @@
+TIPG_NAME="OGC Features and Tiles API"
+DATABASE_URL=postgresql://postgres:postgres@127.0.0.1:5432/pgrouting
+TIPG_DB_SCHEMAS='["public","omf","osm","results"]'

.gitignore

@@ -1,4 +1,5 @@
 .venv
 __pycache__/
 cache/
-.streamlit/
+.streamlit/
+tiles

Documentation/tests-visualisation.md

@@ -22,13 +22,17 @@ The draft made for the final web application is provided [here](./OSM_Overture_v
     - [Installation](#installation-2)
     - [Using Dash](#using-dash)
     - [Review](#review-2)
+  - [Shiny for Python](#shiny-for-python)
+    - [Installation](#installation-3)
+    - [Using Shiny](#using-shiny)
+    - [Review](#review-3)
 - [GeoServer](#geoserver)
   - [Why using GeoServer?](#why-using-geoserver)
   - [Installing GeoServer with Docker](#installing-geoserver-with-docker)
   - [Configuring GeoServer](#configuring-geoserver)
     - [Adding layers with the app](#adding-layers-with-the-app)
     - [Adding layers with REST API](#adding-layers-with-rest-api)
-    - [Review](#review-3)
+    - [Review](#review-4)
 
 
 # QGIS
@@ -262,6 +266,36 @@ PyPlot charts (not only maps) can be displayed with Streamlit too, using `st.plo
 
 Here again, it would be interesting to try to use a GeoServer for instance to query the map features in OGC Standards (WMS or WFS, depending on what we want to do), as it would probably be less demanding for the client.
 
+## Shiny for Python
+
+[Shiny for Python](https://shiny.posit.co/py/) or simply Shiny can, according to the website, create "Effortless Python web applications with the power of reactive programming".
+With Shiny, it is possible to create dashboards and also web visualisation for geospatial information.
+The github project can be find [here](https://github.com/posit-dev/py-shiny/).
+Streamlit is under the [MIT License](https://github.com/posit-dev/py-shiny/blob/main/LICENSE).
+
+### Installation
+
+You can find the Shiny for Python documentation [here](https://shiny.posit.co/py/docs/overview.html).
+
+Shiny for python is really easy to install, just with `pip install shiny`.
+
+You can then run `shiny create --template dashboard-tips` to create a first application with a preview of what it is possible to do with Shiny for Python.
+You will have to chose if you want to use Shiny Express or not, and the folder you will want to create the application in.
+Shiny Express is a more compact way for writing python files using Shiny, so you might want to use it.
+To see it, you might need to download the other python packages that you will find in the `requirements.txt` file created in the folder that you have chosen before.
+
+### Using Shiny
+
+
+
+### Review
+
+The Shiny for Python website provides an article for a comparison between frameworks, including Streamlit framework.
+This article can be find [here](https://shiny.posit.co/py/docs/comp-streamlit.html).
+This article is written by Shiny for Python team, so of course the conclusion is that Shiny for Python is better than Streamlit in a way.
+Still, this article explains some core differences between the two framwork, and Shiny for Python seems less intuitive than Streamlit and probably less user-friendly at first sight, but Shiny seems to be more adapted for managing bigger database.
+One of the difference is the reloading of the entire script every time that a change is made with the User Interface (UI) while shiny will only reload the components 
+
 # GeoServer
 
 [GeoServer](https://geoserver.org) is, according to the website, "an open source server for sharing geospatial data".

Python/Web/Shiny/README.md

@@ -0,0 +1 @@
+This template gives you a more "complete" dashboard for exploring the tips dataset. For an overview of what's here, visit [this article](https://shiny.posit.co/py/docs/user-interfaces.html).

Python/Web/Shiny/app.py

@@ -0,0 +1,90 @@
+import geopandas as gpd
+import pandas as pd
+from shiny import reactive
+from shiny.express import input, ui
+from shinywidgets import render_widget, reactive_read
+import sqlalchemy
+import ipywidgets
+from lonboard import Map, ScatterplotLayer, PathLayer
+
+def getEngine(database:str,
+              host:str="127.0.0.1",
+              user:str="postgres",
+              password:str="postgres",
+              port:str="5432") -> sqlalchemy.engine.base.Engine:
+    """Get sqlalchemy engine to connect to the database.
+    
+    Args:
+        database (str): Database to connect to.
+        host (str, optional): Ip address for the database connection. The default is "127.0.0.1"
+        user (str, optional): Username for the database connection. The default is "postgres".
+        password (str, optional): Password for the database connection. The default is "postgres".
+        port (str, optional): Port for the connection. The default is "5432".
+
+    Returns:
+        sqlalchemy.engine.base.Engine: Engine with the database connection.
+    """
+    engine = sqlalchemy.create_engine(f"postgresql://{user}:{password}@{host}:{port}/{database}")
+    return engine
+
+engine = getEngine("pgrouting")
+
+sqlQueryTable = """SELECT table_catalog, table_schema, table_name FROM information_schema.tables AS t
+WHERE t.table_schema != 'pg_catalog' AND t.table_schema != 'information_schema'
+ORDER BY t.table_schema, t.table_name"""
+
+tables = pd.read_sql_query(sqlQueryTable, engine)
+
+print(type(tables))
+
+tables[["table_schema"]]
+
+tables.to_numpy()
+ 
+# schemas = tables[["table_schema"]].unique()
+
+# schemas
+
+# schemas = pd.unique(tables[["table_schema"]])
+
+# schemas
+
+# ui.input_select("select_schema", "Choose a schema", choices=schemas[["table_schema"]])
+
+
+
+@render_widget
+def colorPickerPoint():
+    color_picker_point = ipywidgets.ColorPicker(concise=True, description='Point color', value='#000000')
+    return color_picker_point
+
+@render_widget
+def colorPickerLine():
+    color_picker_line = ipywidgets.ColorPicker(concise=True, description='Line color', value='#000000')
+    return color_picker_line
+
+@render_widget
+def map():
+    gdf = gpd.GeoDataFrame.from_postgis("SELECT * FROM omf.node_tokyo", engine)
+    layer = ScatterplotLayer.from_geopandas(gdf, radius_min_pixels=2)
+    gdf_line = gpd.GeoDataFrame.from_postgis("SELECT * FROM omf.edge_with_cost_tokyo", engine)
+    line = PathLayer.from_geopandas(gdf_line, width_min_pixels=2)
+    return Map([layer, line])
+
+
+@reactive.effect
+def set_fill_color():
+    # Point
+    # map.widget.layers[0].get_fill_color = colors_point[input.color_select_point()]
+    
+    hexPoint = reactive_read(colorPickerPoint.widget, "value")
+    colorPoint = list(int(hexPoint[i:i+2], 16) for i in (1, 3, 5))
+    map.widget.layers[0].get_fill_color = colorPoint
+    
+    
+    # Line
+    # map.widget.layers[1].get_fill_color = colors_line[input.color_select_line()]
+    
+    hexLine = reactive_read(colorPickerLine.widget, "value")
+    colorLine = list(int(hexLine[i:i+2], 16) for i in (1, 3, 5))
+    map.widget.layers[1].get_color = colorLine

Python/Web/Shiny/requirements.txt

@@ -0,0 +1,6 @@
+faicons
+shiny
+shinywidgets
+plotly
+pandas
+ridgeplot

Python/Web/Shiny/shared.py

@@ -0,0 +1,6 @@
+from pathlib import Path
+
+import pandas as pd
+
+app_dir = Path(__file__).parent
+tips = pd.read_csv(app_dir / "tips.csv")

Python/Web/Shiny/styles.css

@@ -0,0 +1,12 @@
+:root {
+  --bslib-sidebar-main-bg: #f8f8f8;
+}
+
+.popover {
+  --bs-popover-header-bg: #222;
+  --bs-popover-header-color: #fff;
+}
+
+.popover .btn-close {
+  filter: var(--bs-btn-close-white-filter);
+}