Merge pull request #347 from ropensci/encoding_UTF8

Set encoding to UTF-8 for tags and user names

Author: jmaspons

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: osmdata
 Title: Import 'OpenStreetMap' Data as Simple Features or Spatial Objects
-Version: 0.2.5.018
+Version: 0.2.5.019
 Authors@R: c(
     person("Mark", "Padgham", , "[email protected]", role = c("aut", "cre")),
     person("Bob", "Rudis", role = "aut"),

NEWS.md

@@ -11,6 +11,7 @@
 - Improved `get_bb(..., format_out = "sf_polygon")` to return full metadata
   along with geometries (#338 thanks to @RegularnaMatrica)
 - Mention key-only feature requests in README (#342 thanks to @joostschouppe)
+- Set encoding to UTF-8 for tags and user names (#347)
 
 
 0.2.5

R/get-osmdata-df.R

@@ -75,7 +75,8 @@ osmdata_data_frame <- function (q,
             colClasses = "character", # osm_id doesn't fit in integer
             check.names = FALSE,
             comment.char = "",
-            stringsAsFactors = stringsAsFactors
+            stringsAsFactors = stringsAsFactors,
+            encoding = "UTF-8"
         )
     } else if (isTRUE (obj$meta$query_type == "adiff")) {
         datetime_from <- obj$meta$datetime_from
@@ -122,13 +123,15 @@ xml_to_df <- function (doc, stringsAsFactors = FALSE) {
 
     tags <- mapply (function (i, k) {
         i <- i [, k, drop = FALSE] # remove osm_id column if exists
+        out <- matrix (
+            NA_character_,
+            nrow = nrow (i), ncol = length (keys),
+            dimnames = list (NULL, keys)
+        )
+        out <- enc2utf8 (out)
         out <- data.frame (
-            matrix (
-                nrow = nrow (i), ncol = length (keys),
-                dimnames = list (NULL, keys)
-            ),
-            stringsAsFactors = stringsAsFactors,
-            check.names = FALSE
+            out,
+            stringsAsFactors = stringsAsFactors, check.names = FALSE
         )
         out [, names (i)] <- i
         return (out)
@@ -214,6 +217,7 @@ xml_adiff_to_df <- function (doc,
     tags_u <- xml2::xml_find_all (osm_actions, xpath = ".//tag")
     col_names <- sort (unique (xml2::xml_attr (tags_u, attr = "k")))
     m <- matrix (
+        NA_character_,
         nrow = length (osm_obj), ncol = length (col_names),
         dimnames = list (NULL, col_names)
     )
@@ -225,6 +229,7 @@ xml_adiff_to_df <- function (doc,
         tagV <- vapply (tag, function (x) x, FUN.VALUE = character (2))
         m [i, tagV [1, ]] <- tagV [2, ]
     }
+    m <- enc2utf8 (m)
 
     osm_type <- xml2::xml_name (osm_obj)
     osm_id <- xml2::xml_attr (osm_obj, "id")
@@ -325,6 +330,7 @@ get_meta_from_xml <- function (osm_obj) {
             osm_uid = xml2::xml_attr (osm_obj, attr = "uid"),
             osm_user = xml2::xml_attr (osm_obj, attr = "user")
         )
+        out$osm_user <- enc2utf8 (out$osm_user)
 
     } else {
         out <- matrix (nrow = length (osm_obj), ncol = 0)

R/get-osmdata-sc.R

@@ -75,6 +75,12 @@ osmdata_sc <- function (q, doc, quiet = TRUE) {
         overpass_version = temp$obj$meta$overpass_version
     )
 
+    has_tags <- c ("nodes", "relation_properties", "object")
+    obj [has_tags] <- lapply(obj [has_tags], function (x) {
+        x [, c ("key", "value")] <- setenc_utf8 (x [, c ("key", "value")])
+        x
+    })
+
     if (!missing (q)) {
         if (!is.character (q)) {
             obj$meta$bbox <- q$bbox

R/get-osmdata-sf.R

@@ -62,8 +62,9 @@ osmdata_sf <- function (q, doc, quiet = TRUE, stringsAsFactors = FALSE) { # noli
     if (!"osm_id" %in% names (res$polygons_kv)[1]) {
         res <- fill_kv (res, "polygons_kv", "polygons", stringsAsFactors)
     }
-    kv_df <- grep ("_kv$", names (res))
-    res[kv_df] <- fix_columns_list (res[kv_df])
+    kv_df <- grep ("_kv$", names (res)) # objects with tags
+    res [kv_df] <- fix_columns_list (res[kv_df])
+    res [kv_df] <- lapply (res [kv_df], setenc_utf8)
 
     if (missing (q)) {
         obj$bbox <- paste (res$bbox, collapse = " ")

R/get-osmdata-sp.R

@@ -71,7 +71,11 @@ osmdata_sp <- function (q, doc, quiet = TRUE) {
     obj$osm_multipolygons <- res$multipolygons
 
     osm_items <- grep ("^osm_", names (obj))
-    obj[osm_items] <- fix_columns_list (obj[osm_items])
+    obj [osm_items] <- fix_columns_list (obj [osm_items])
+    obj [osm_items] <- lapply (obj [osm_items], function (x) {
+        x@data <- setenc_utf8 (x@data)
+        x
+    })
     class (obj) <- c (class (obj), "osmdata_sp")
 
     return (obj)

R/get-osmdata.R

@@ -344,3 +344,19 @@ get_center_from_cpp_output <- function (res, what = "points") {
 
     return (as.data.frame (this))
 }
+
+
+#' Set encoding to UTF-8
+#'
+#' @param x a data.frame or a list.
+#'
+#' @return `x` with all the columns or items of type character with UTF-8 encoding set.
+#' @noRd
+setenc_utf8 <- function (x) {
+    char_cols <- which (vapply (x, is.character, FUN.VALUE = logical (1)))
+    x [char_cols] <- lapply (x [char_cols], function (y) {
+        enc2utf8 (y)
+    })
+
+    return (x)
+}

R/getbb.R

@@ -229,6 +229,8 @@ getbb <- function (place_name,
     )
 
     if (format_out == "data.frame") {
+        utf8cols <- c ("licence", "name", "display_name")
+        obj [, utf8cols] <- setenc_utf8 (obj [, utf8cols])
         return (obj)
     }
 

codemeta.json

@@ -11,13 +11,13 @@
   "codeRepository": "https://github.com/ropensci/osmdata/",
   "issueTracker": "https://github.com/ropensci/osmdata/issues",
   "license": "https://spdx.org/licenses/GPL-3.0",
-  "version": "0.2.5.018",
+  "version": "0.2.5.019",
   "programmingLanguage": {
     "@type": "ComputerLanguage",
     "name": "R",
     "url": "https://r-project.org"
   },
-  "runtimePlatform": "R version 4.3.1 (2023-06-16)",
+  "runtimePlatform": "R version 4.4.1 (2024-06-14)",
   "provider": {
     "@id": "https://cran.r-project.org",
     "@type": "Organization",