Doenload Berlin Shapefiles

download_data.R

#' Download data to draw Berlin template map
#' @examples download_data_berlin()
download_data_berlin <- function() {
  
  ## SETUP ---------------------------------------------------------------------
  ## output dir for raw geo files
  dir <- here::here("data-raw", "geo-raw")
  if (!dir.exists(dir)) dir.create(dir, showWarnings = TRUE, recursive = TRUE)
  
  ## DISTRICTS -----------------------------------------------------------------
  ## Berlin districts (WGS 84)
  ## Source: Technologiestiftung Berlin
  json_file <- here::here("data-raw", "geo-raw", "bezirksgrenzen.geojson")
  if(!file.exists(json_file)) {
    ## Download and unzip Berlin districts (WGS 84)
    ##   - German:  https://daten.odis-berlin.de/de/dataset/bezirksgrenzen
    ##   - English: https://daten.odis-berlin.de/en/dataset/bezirksgrenzen
    link <- "https://tsb-opendata.s3.eu-central-1.amazonaws.com/bezirksgrenzen/bezirksgrenzen.geojson"
    message(paste0("Loading Berlin district data from ", link, "."))
    curl::curl_download(link, destfile = json_file)
    message(paste0("Downloaded to ", json_file, "."))
  } else {
    message(paste0("Berlin district data already exists: ", json_file, "."))
  }
  
  ## OSM DATA ------------------------------------------------------------------
  ## Berlin OSM data (WGS 84)
  ## Source: OpenStreetMaps via Geofabrik
  zip_file <- here::here("data-raw", "geo-raw", "berlin_shapes.zip")
  shp_path <- here::here("data-raw", "geo-raw", "berlin_shapes")
  if(!file.exists(glue::glue("{shp_path}/gis_osm_water_a_free_1.shp"))) {
    ## Download and unzip Berlin shapefiles (WGS 84)
    ## https://download.geofabrik.de/europe/germany/berlin.html
    link <- "https://download.geofabrik.de/europe/germany/berlin-latest-free.shp.zip"
    message(paste0("Loading Berlin OSM data from ", link, "."))
    curl::curl_download(link, destfile = zip_file)
    message(paste0("Unzip Berlin OSM data from ", zip_file, "."))
    unzip(zipfile = zip_file, exdir = shp_path)
    suppressMessages(file.remove(zip_file))
    curl::curl_download(link, destfile = json_file)
    message(paste0("Downloaded to ", shp_path, "/."))
  } else {
    message(paste0("Berlin OSM data already exists: ", shp_path, "/."))
  }
}

## PREPARE DATA --------------------------------------------------------------
## Berlin districts (WGS 84)
## Source: Technologiestiftung Berlin
sf_districts <- sf::read_sf(json_file)

## Derive Berlin Outline
sf_berlin <- suppressMessages(sf::st_union(sf_districts))

## Berlin Waterways (WGS 84)
sf_water <-
  suppressMessages(sf::read_sf(dsn = glue::glue("{shp_path}/gis_osm_water_a_free_1.shp"),
              layer = "gis_osm_water_a_free_1") %>%
  sf::st_intersection(sf_districts))

## Berlin Roads (WGS 84)
if(roads == TRUE) {
  message("Loading road data as well, this might take some time.")
  sf_roads <-
      suppressMessages(sf::read_sf(dsn = glue::glue("{shp_path}/gis_osm_roads_free_1.shp"),
                  layer = "gis_osm_roads_free_1") %>%
      ## keep only roads + links
      dplyr::filter(stringr::str_detect(
        fclass, 
        "motorway|trunk|primary|secondary|tertiary|unclassified|residential|living"
      )) %>%
      sf::st_intersection(sf_districts))
}

if(type == "green_areas") {
  ## Combine relevant natural categories
  sf_type <-
    suppressMessages(
      ## Berlin Landuse Categories (WGS 84)
      sf::read_sf(dsn = glue::glue("{shp_path}/gis_osm_landuse_a_free_1.shp"),
                  layer = "gis_osm_landuse_a_free_1") %>%
      ## Berlin Natural Areas (WGS 84)
      rbind(sf::read_sf(dsn = glue::glue("{shp_path}/gis_osm_natural_a_free_1.shp"),
                        layer = "gis_osm_natural_a_free_1")) %>%
      dplyr::filter(fclass %in% c("forest", "grass", "meadow", "nature_reserve",
                                  "scrub", "heath", "beach", "cliff")) %>%
      sf::st_intersection(sf_districts)
    )
}