Ensure non-intersecting sf-geometries in get_eurostat_geospatial

NAMESPACE

@@ -59,6 +59,7 @@ importFrom(lubridate,year)
 importFrom(methods,as)
 importFrom(readr,col_character)
 importFrom(readr,read_tsv)
+importFrom(sf,st_buffer)
 importFrom(sf,st_read)
 importFrom(sp,merge)
 importFrom(sp,spplot)

R/firstlib.R

@@ -27,6 +27,7 @@
 #' @importFrom readr read_tsv
 #' @importFrom readr col_character
 #' @importFrom sf st_read
+#' @importFrom sf st_buffer
 #' @importFrom sp merge
 #' @importFrom sp spplot
 #' @importFrom stringi stri_match_first_regex

R/get_eurostat_geospatial.R

@@ -25,13 +25,23 @@
 #'        \code{\link{tempdir}}. Directory can also be set with
 #'        \code{option} eurostat_cache_dir.
 #' @param crs projection of the map: 4-digit \href{https://spatialreference.org/ref/epsg/}{EPSG code}. One of:
+#' @param make_valid logical; ensure that valid (multi-)polygon features are returned
+#'        if \code{output_class="sf"}, see Details. Current default \code{FALSE}, will be changed
+#'        in the future.
+#'        
 #' \itemize{
 #' \item "4326" - WGS84
 #' \item "3035" - ETRS89 / ETRS-LAEA
 #' \item "3857" - Pseudo-Mercator
 #' }
 #' @export
 #' @details The data source URL is \url{http://ec.europa.eu/eurostat/web/gisco/geodata/reference-data/administrative-units-statistical-units}.
+#' The source provides feature collections as line strings (GeoJSON format),
+#' not as (multi-)polygons which, in some cases, yields invalid
+#' self-intersecting (multi-)polygon geometries (for some years/resolutions).
+#' This can cause problems, e.g., when using these geometries as input argument
+#' to \code{sf::st_interpolate_aw()}). \code{make_valid = TRUE} makes sure that
+#' only valid (multi-)polygons are returned, example included below.
 #' @author Markus Kainu <markuskainu@gmail.com>
 #' @return a sf, data_frame or SpatialPolygonDataFrame.
 #' @examples
@@ -41,11 +51,36 @@
 #'     spdf <- get_eurostat_geospatial(output_class = "spdf", resolution = "10", nuts_level = "3")
 #'   }
 #' 
-get_eurostat_geospatial <- function(output_class="sf", 
-                                    resolution="60",
+#'   \dontrun{
+#'     # -------------------------------------------------------------------
+#'     # Minimal example to demonstrate reason/effect of 'make_valid = TRUE'
+#'     # Spatial data set; rectangle spanning the entire globe with a constant value of 1L.
+#'     # Requires the R package sf.
+#'     library("sf")
+#'     poly <- st_polygon(list(matrix(c(-180, -90, -180, 90, 180, 90, 180, -90, -180, -90), ncol = 2, byrow = TRUE)))
+#'     data <- st_sf(data.frame(geom = st_sfc(poly), data = 1L), crs = st_crs(4326))
+#'     
+#'     # Causing an error: Self-intersection of some points of the geometry
+#'     NUTS2_A <- get_eurostat_geospatial("sf", 60, nuts_level = 2, year = 2013,
+#'                                        crs = 4326, make_valid = FALSE)
+#'     res <- tryCatch(st_interpolate_aw(data, NUTS2_A, extensive = FALSE),
+#'                     error   = function(e) e)
+#'     print(res)
+#'     
+#'     # Resolving the problem using make_valid = TRUE. 'extensive = FALSE' returns
+#'     # average over each area, thus resulting in a constant value of 1 for each
+#'     # geometry in NUTS2_B.
+#'     NUTS2_B <- get_eurostat_geospatial("sf", 60, nuts_level = 2, year = 2013,
+#'                                        crs = 4326, make_valid = TRUE)
+#'     res <- st_interpolate_aw(data, NUTS2_B, extensive = FALSE)
+#'     print(head(res))
+#'   }
+#' 
+get_eurostat_geospatial <- function(output_class = "sf", 
+                                    resolution = "60",
                                     nuts_level = "all", year = "2016",
                                     cache = TRUE, update_cache = FALSE,
-				    cache_dir = NULL, crs = "4326"){
+                                    cache_dir = NULL, crs = "4326", make_valid = FALSE){
   # Check if you have access to ec.europe.eu. 
   if (!check_access_to_data()){
     message("You have no access to ec.europe.eu. 
@@ -58,35 +93,40 @@ Please check your connection and/or review your proxy settings")
        envir = environment(),
        package = "eurostat")
 
-  # Check resolution is of correct format
-  resolution <- as.character(resolution)
-  resolution <- gsub("^0+", "", resolution)
-  if (!as.numeric(resolution) %in% c(1, 3, 10, 20, 60)) {
-    stop("Resolution should be one of 01, 1, 03, 3, 10, 20, 60")
-  }
-  resolution <- gsub("^1$", "01", resolution)
-  resolution <- gsub("^3$", "03", resolution)
-
   # Check output_class is of correct format
-  if (!output_class %in% c("sf", "df", "spdf")) {
-    stop("output_class should be one of 'sf', 'df' or 'spdf'")
-  }
-
+  stopifnot(length(output_class) == 1L)
+  output_class <- match.arg(as.character(output_class), c("sf", "df", "spdf"))
+
+  # Check resolution is of correct format
+  stopifnot(length(resolution) == 1L)
+  resolution <- as.integer(regmatches(resolution, regexpr("^[0-9]+", resolution)))
+  resolution <- sprintf("%02d", match.arg(as.character(resolution), c(1, 3, 10, 20, 60)))
+
+  # Sanity check for nuts_level
+  stopifnot(length(nuts_level) == 1L)
+  nuts_level <- regmatches(nuts_level, regexpr("^(all|[0-9]+)", nuts_level))
+  nuts_level <- match.arg(nuts_level, c("all", 0:3))
+
   # Check year is of correct format
-  year <- as.character(year)
-  if (!as.numeric(year) %in% c(2003, 2006, 2010, 2013, 2016, 2021)) {
-    stop("Year should be one of 2003, 2006, 2010, 2013, 2016 or 2021")
-  }
+  year <- match.arg(as.character(year), c(2003, 2006, 2010, 2013, 2016, 2021))
+
+  # Sanity check for cache and update_cache
+  stopifnot(is.logical(cache) && length(cache) == 1 && cache %in% c(TRUE, FALSE))
+  stopifnot(is.logical(update_cache) && length(update_cache) == 1 && update_cache %in% c(TRUE, FALSE))
+
+  # Sanity check for cache_dir
+  stopifnot(is.null(cache_dir) || (is.character(cache_dir) && length(cache_dir) == 1L))
+
+  # Check crs is of correct format
+  crs <- match.arg(as.character(crs), c(4326, 3035, 3857))
 
+  # Invalid combination of year and resolution: stop and show hint/error message.
   if (as.numeric(year) == 2003 & as.numeric(resolution) == 60) {
     stop("NUTS 2003 is not provided at 1:60 million resolution. Try 1:1 million, 1:3 million, 1:10 million or 1:20 million")
   }
-
-  # Check crs is of correct format
-  crs <- as.character(crs)
-  if (!as.numeric(crs) %in% c(4326, 3035, 3857)) {
-    stop("crs should be one of 4326, 3035 or 3857")
-  }
+
+  # Sanity check for input `make_valid`
+  stopifnot(is.logical(make_valid) && length(make_valid) == 1L && make_valid %in% c(TRUE, FALSE))
 
 #   message("
 # COPYRIGHT NOTICE
@@ -117,25 +157,20 @@ Please check your connection and/or review your proxy settings")
 
   if (resolution == "60" && year == 2016 && crs == "4326"){
 
-    if (nuts_level %in% c("all")){
+    # nuts_levels are mutually exclusive; either "all"
+    # or one of "0", "1", "2", "3".
+    if (nuts_level == c("all")){
       shp <- eurostat_geodata_60_2016 
+    } else {
+      shp <- filter(eurostat_geodata_60_2016, LEVL_CODE == as.integer(nuts_level))
     }
-    if (nuts_level == "0") shp <- filter(eurostat_geodata_60_2016, 
-                                         LEVL_CODE == 0)
-    if (nuts_level == "1") shp <- filter(eurostat_geodata_60_2016, 
-                                         LEVL_CODE == 1)
-    if (nuts_level == "2") shp <- filter(eurostat_geodata_60_2016, 
-                                         LEVL_CODE == 2)
-    if (nuts_level == "3") shp <- filter(eurostat_geodata_60_2016, 
-                                         LEVL_CODE == 3)
 
     if (output_class == "df"){
       nuts_sp <- as(shp, "Spatial")
       nuts_sp$id <- row.names(nuts_sp)
       nuts_ff <- broom::tidy(nuts_sp)
       shp <- left_join(nuts_ff,nuts_sp@data)
-    }
-    if (output_class == "spdf"){
+    } else if (output_class == "spdf"){
       shp <- as(shp, "Spatial")
     }
 
@@ -168,7 +203,7 @@ Please check your connection and/or review your proxy settings")
   }
 
   # if cache = FALSE or update or new: dowload else read from cache
-  if (!cache || update_cache || !file.exists(cache_file)){
+  if (!cache | update_cache | !file.exists(cache_file)){
 
     if (nuts_level %in% c("0","all")){
       url <- paste0("http://ec.europa.eu/eurostat/cache/GISCO/distribution/v2/nuts/geojson/NUTS_RG_",resolution,"M_",year,"_",crs,"_LEVL_0.geojson")
@@ -209,15 +244,12 @@ Please check your connection and/or review your proxy settings")
         }
     }
 
-    if (nuts_level %in% c("all")){
+    if (nuts_level == "all") {
       shp <- rbind(nuts0, nuts1, nuts2, nuts3)
+    } else {
+      shp <- eval(parse(text = paste0("nuts", nuts_level)))
     }
 
-    if (nuts_level == "0") shp <- nuts0
-    if (nuts_level == "1") shp <- nuts1
-    if (nuts_level == "2") shp <- nuts2
-    if (nuts_level == "3") shp <- nuts3
-
     if (output_class == "df"){
       nuts_sp <- as(shp, "Spatial")
       nuts_sp$id <- row.names(nuts_sp)
@@ -285,6 +317,12 @@ Please check your connection and/or review your proxy settings")
 # 
 # # --------------------------          
 #           ")
+  if (output_class == "sf" & make_valid) {
+      shp <- st_buffer(shp, 0)
+  } else {
+      warning(paste("Default of 'make_valid' for 'output_class=\"sf\"'",
+                    "will be changed in the future (see function details)."))
+  }
 
   # Adding a `geo` column for easier joins with dplyr 
   shp$geo <- shp$NUTS_ID