Add SIDE data, plus several smaller fixes and doc updates

DESCRIPTION

@@ -39,7 +39,8 @@ Suggests:
     terra,
     sf,
     exactextractr,
-    arrow
+    arrow,
+    countrycode
 Config/testthat/edition: 3
 Imports:
     dplyr,

NAMESPACE

@@ -46,6 +46,9 @@ export(gen_naturalearth_cover_share)
 export(gen_ne_disputed_area_share)
 export(gen_ruggedterrain_elevation_mean)
 export(gen_shdi)
+export(gen_side_excluded)
+export(gen_side_included)
+export(gen_side_irrelevant)
 export(gen_speibase6_mean)
 export(gen_traveltime_mean)
 export(gen_traveltime_min)
@@ -92,13 +95,15 @@ export(read_ruggedterrain)
 export(read_shdi)
 export(read_shdi_csv)
 export(read_shdi_shapefile)
+export(read_side)
 export(read_speibase)
 export(read_traveltime)
 export(read_ucdp_ged)
 export(robust_transformation)
 export(ruggedterrain_variable)
 export(save_pgvariable)
 export(shdi)
+export(side)
 export(speibaseN)
 export(ucdp_ged)
 export(ucdpged_distance_within_country)

R/data.R

@@ -69,4 +69,21 @@
 #' }
 "pgchecksum"
 
-
+#' SIDE-LEDA Match Table
+#'
+#' Preprocessed annual links between SIDE ethnic groups and their annual
+#' political relevance categories derived from LEDA/EPR matching.
+#'
+#' This dataset is bundled with the package so the expensive LEDA matching
+#' workflow does not have to run every time SIDE is aggregated to PRIO-GRID.
+#'
+#' @format ## `leda_matches`
+#' A data frame with 5 columns:
+#' \describe{
+#'   \item{iso3c}{Three-letter country code.}
+#'   \item{side_group}{SIDE ethnic group label.}
+#'   \item{year}{Calendar year.}
+#'   \item{link_year_source}{Whether the SIDE-EPR year match was observed or carried by the upstream workflow.}
+#'   \item{status3}{Collapsed EPR status category: included, excluded, or irrelevant.}
+#' }
+"leda_matches"

R/data_ghs_wup.R

@@ -31,6 +31,8 @@
 #'
 #' @return A \code{SpatRaster} object with DEGURBA classification codes
 #'
+#' @param config A \code{pg_config} object. Defaults to \code{\link{pg_current_config}()}.
+#'
 #' @note
 #' \itemize{
 #'   \item Files are automatically extracted from zip archives and cached locally
@@ -116,6 +118,7 @@ read_ghs_wup_degurba <- function(config = pg_current_config()){
 #'
 #' @param urban_definition Numeric vector of DEGURBA codes to classify as urban.
 #'   Valid codes are: 10, 11, 12, 13, 21, 22, 23, 30. See Details for code meanings.
+#' @param config A \code{pg_config} object. Defaults to \code{\link{pg_current_config}()}.
 #'
 #' @details
 #' The DEGURBA classification codes represent:
@@ -183,7 +186,7 @@ ghs_wup_degurba <- function(urban_definition, config = pg_current_config()){
   r <- read_ghs_wup_degurba(config = config)
 
   temporary_directory <- file.path(pg_rawfolder(), "tmp", tempdir() |> basename())
-  dir.create(temporary_directory)
+  dir.create(temporary_directory, recursive = TRUE)
   tmp1 <- tempfile(pattern = "urban_classification", fileext = ".tif", tmpdir = temporary_directory)
 
   r <- terra::classify(r, cl_mat, filename = tmp1)
@@ -427,7 +430,7 @@ gen_ghs_wup_degurba_urban <- function(config = pg_current_config()){
 #' \insertRef{schiavinaGHSWUPDEGURBAR2025AGHSWUP2025}{priogrid}
 #'
 #' \insertRef{europeancommissionApplyingDegreeUrbanisation2021}{priogrid}
-urban_extent <- function(lon, lat, measurement_date, urban_definition = c(21, 22, 23, 30), max_extent = 1000e3){
+urban_extent <- function(lon, lat, measurement_date, urban_definition = c(21, 22, 23, 30), max_extent = 1000e3, config = pg_current_config()){
   # lat <- 59.935320; lon <- 10.763063 Oslo
   # lat <- 28.612738 lon <- 77.231487 New Delhi
 
@@ -452,6 +455,6 @@ urban_extent <- function(lon, lat, measurement_date, urban_definition = c(21, 22
   urban_extent_raster <- terra::ifel(urban_patches == center_patch_id, 1, NA)
   urban_extent_raster <- terra::trim(urban_extent_raster)
   res <- terra::as.polygons(urban_extent_raster, dissolve = TRUE) |> sf::st_as_sf()
-  res <- sf::st_transform(res, pg_current_config()$crs)
+  res <- sf::st_transform(res, sf::st_crs(config$crs))
   res
 }