data-raw/process_shapefile_info.r

In tweed1e/canmap: Standard Canadian geography shapefiles

library(tidyverse)
library(fs)
library(usethis)
library(httr) # for later.


shapefile_paths <- readr::read_csv("data-raw/shapefile_paths.csv")

base_url_1 <- "http://www12.statcan.gc.ca/census-recensement/"
base_url_2 <- "/geo/bound-limit/files-fichiers/"


shapefile_paths %>% count(geo_code)

# only get shapefiles
shapefile_paths <- shapefile_paths %>%
  filter(format == "ArcGIS (.shp)" & language == "english" & geo_code != "rnf") %>%
  dplyr::mutate(path = stringr::str_c(base_url_1,
                      "2011", # idk why they have 2011 in the url for every year?
                      base_url_2,
                      ref_date,
                      "/",
                      filepath, ".zip")) %>%
    filter(!is.na(file_type) & !is.na(path))

######
# Need to double-check these all exist, they might not.
######
# httr::

si <- shapefile_paths %>%
  distinct() %>%
  rowwise() %>%
  mutate(http_type = path %>%
           httr::HEAD() %>%
           httr::http_type())

si %>% count(http_type)

si <- si %>% filter(http_type == "application/x-zip-compressed")

shapefile_paths <- shapefile_paths %>% semi_join(si, by = "path")

usethis::use_data(shapefile_paths, overwrite = TRUE)


############# and code info

code_book_ <- readr::read_csv("data-raw/code_book.csv")
code_book <- dplyr::select(code_book_, -start_chr, -stop_chr, -long_desc)
usethis::use_data(code_book, internal = FALSE, overwrite = TRUE)

# save this as well.
code_pos <- code_book_ %>%
  dplyr::distinct(code_type, start_chr, stop_chr)
usethis::use_data(code_pos, internal = FALSE, overwrite = TRUE)