assignment16

# Load a list of packages. Install them first if they are not available.
# The list of packages to be installed
list.of.packages <- c("sf", "sp", "spatial", "maptools", "rgeos","rgdal",
                      "raster", "grid", "rasterVis",
                      "tidyverse", "magrittr", "ggpubr", "lubridate",
                      "devtools", "htmlwidgets", "mapview",
                      "classInt", "RColorBrewer", "ggmap", "tmap", "leaflet", "mapview", 
                      "ggrepel", "ggsn",
                      "spdep","spatialreg","GWmodel");

# Check out the packages that have not been installed yet.
new.packages <- list.of.packages[!(list.of.packages %in% installed.packages()[,"Package"])]

# Install those missing packages first. It could take a long time for the first time.
if(length(new.packages)>0) install.packages(new.packages)

# Load all packages.
lapply(list.of.packages,function(x) {
  require(x,character.only = TRUE,quietly = TRUE)
})

# Or load specific packages when needed
require(sp); require(sf); library(raster)

install.packages("sf")

library(sf)

#Task 2: Read the NYC postal areas in shapefiles into sf objects 
geojson_file <- file.choose()

postal_areas_sf <- st_read(geojson_file) 

postal_areas_sf

#Task 3: Read the NYC public health services data into a dataframe in R 
#Extract Coordinates 
public_health_services_df <- read.csv(file.choose())

public_health_services_df 

lat_lon <- regmatches(public_health_services_df$Location.1, 
                      gregexpr( "(?<=\\().+?(?=\\))",
                                public_health_services_df$Location.1, perl = T))
lat_lon2 <- data.frame(str_split_fixed(lat_lon, ", ", 2))
#rename columns 
colnames(lat_lon2) <- c("Latitude", "Longitude")

colnames(public_health_services_df) <- c("Facility_Type",
                                         "Borough", "Facility_Name", 
                                         "Cross_Streets", "Phone", "Location", 
                                         "Latitude", "Longitude") 
rm('lat_lon', 'lat_lon2')

# Filter out rows with missing Longitude or Latitude
public_health_services_df <- 
  public_health_services_df[complete.cases(public_health_services_df$Longitude,
                                           public_health_services_df$Latitude),
                            ]

# Convert Longitude and Latitude to numeric
public_health_services_df$Longitude <- 
  as.numeric(public_health_services_df$Longitude)
public_health_services_df$Latitude <- 
  as.numeric(public_health_services_df$Latitude)

#Create sf objects from geographic coordinates 
Public_health_services_sf <- st_as_sf(public_health_services_df , 
                                      coords = c("Longitude", "Latitude"))

Public_health_services_sf

#Task 4: Read and process the NYS retail food stores data. 
#Create sf objects from geographic coordinates for NYS. 

retail_food_stores_df <- read.csv(file.choose())
retail_food_stores_df

# Extract coordinates for Location column
lat_lon <- regmatches(retail_food_stores_df$Location,
                      gregexpr( "(?<=\\().+?(?=\\))",
                                retail_food_stores_df$Location, perl = T))
lat_lon1 <- data.frame(str_split_fixed(lat_lon, ", ", 2))
# Rename the columns
colnames(lat_lon1) <- c("Latitude", "Longitude")
# Convert the latitude and longitude values from character into numeric values
lat_lon1$Latitude <- as.numeric(lat_lon1$Latitude)
lat_lon1$Longitude <- as.numeric(lat_lon1$Longitude)
# There are missing coordinate values, so I will remove those rows from the 
# data frame
retail_food_stores_df <- retail_food_stores_df %>%
  select(c(-Address.Line.2, -Address.Line.3))
retail_food_stores_df <- na.omit(retail_food_stores_df)
rm('lat_lon', 'lat_lon1')
#rename columns
colnames(retail_food_stores_df) <- c("County", "License_Number", 
                                     "Operation_Type", "Establishment_Type", 
                                     "Entity_Name", "DBA_Name", "Street_Number",
                                     "Street_Name", "City", "State", "Zip_Code", 
                                     "Square_Footage","Geoference"
retail_food_stores_sf <- st_as_sf(retail_food_stores_df, coords = c("Longitude", 
                                                                    "Latitude"))

#Task 5: Use simple mapping method, either based on ggmap + ggplot or mapview, 
#with a basemap to verify the above datasets in terms of their geometry 
#locations.


mapview(postal_areas_sf) 
mapview(Public_health_services_sf, zcol = "Facility_Type", layer.name = 
          "Facility Type")
mapview(retail_food_stores_sf, layer.name = "NYS Retail Food Stores")


#Task 6: Save the three sf objects in a RData file or in a single 
#GeoPackage file/database. 

save(postal_areas_sf, public_health_services_sf, retail_food_stores_sf,
     file = "Data9.RData")

# To get the data back into R
# load(file = "Data.RData")

# Save data to a GeoPackage file/database
st_write(postal_areas_sf,
         dsn = "Data.gpkg",
         layer = "postal_areas",
         delete_layer = TRUE)

st_write(public_health_services_sf,
         dsn = "Data.gpkg",
         layer = "public_health_services",
         delete_layer = TRUE)

st_write(retail_food_stores_sf,
         dsn = "Data.gpkg",
         layer = "retail_food_stores",
         delete_layer = TRUE)