climate 1.2.6

* feat: selective download for meteo_imgw_daily

* fix: selective synop download

* fix: styler

* climate 1.2.6rc

* fix: hydro

* fix unit tests

* fix: typo in docs

Author: bczernecki

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: climate
 Title: Interface to Download Meteorological (and Hydrological) Datasets
-Version: 1.2.5
+Version: 1.2.6
 Authors@R: c(person(given = "Bartosz",
            family = "Czernecki",
            role = c("aut", "cre"),

NAMESPACE

@@ -33,6 +33,7 @@ importFrom(archive,archive_read)
 importFrom(curl,curl_download)
 importFrom(curl,has_internet)
 importFrom(data.table,as.data.table)
+importFrom(data.table,data.table)
 importFrom(data.table,fread)
 importFrom(stats,na.omit)
 importFrom(stats,runif)

NEWS.md

@@ -1,3 +1,10 @@
+# climate 1.2.6
+
+* speeding up selective download for given station names in `meteo_imgw_*` and `hydro_imgw_daily()` functions that simultaneously reduce use of memory
+* minor fix for `meteo_imgw_daily` to handle cases with station names containing extra spaces or non-parsable characters
+* verbosing currently processed url for easier detection of problematic or corrupted datasets
+
+
 # climate 1.2.5
 
 * Added 20 sec. pause between subsequent requests to the OGIMET server to avoid server overload. 

R/hydro_imgw_daily.R

@@ -14,7 +14,7 @@
 #' @param ... other parameters that may be passed to the 'shortening' function that shortens column names
 #' @importFrom XML readHTMLTable
 #' @importFrom utils download.file unzip read.csv
-#' @importFrom data.table fread
+#' @importFrom data.table fread data.table
 #' @export
 #' @returns data.frame with historical hydrological data for the daily time interval
 #' @examples \donttest{
@@ -135,10 +135,26 @@ hydro_imgw_daily_bp = function(year,
 
     } #end of loop for (usually monthly) zip files in a given year
 
-    all_data[[length(all_data) + 1]] = merge(codz_data, zjaw_data,
+    all_data[[length(all_data) + 1]] = merge(data.table(codz_data), 
+                                             data.table(zjaw_data),
                                              by = intersect(colnames(codz_data), colnames(zjaw_data)),
                                              all.x = TRUE)
 
+    # station selection and names cleaning:
+    if (!is.null(station)) {
+      if (is.character(station)) {
+        inds = unique(as.numeric(unlist(sapply(station, function(x) grep(pattern = x, x = trimws(all_data[[length(all_data)]]$`Nazwa stacji`))))))
+        if (any(is.na(inds)) || length(inds) == 0) {
+          env$logs = c(
+            env$logs,
+            paste("At least one of selected station(s) is not available in the database. Returning all available stations")
+          )
+        } else {
+          all_data[[length(all_data)]] = all_data[[length(all_data)]][inds, ]
+        }
+      }
+    }
+    
   } # end of loop for years (if more than 1 specified)
 
   all_data = do.call(rbind, all_data)
@@ -171,6 +187,7 @@ hydro_imgw_daily_bp = function(year,
     }
   }
 
+  all_data = as.data.frame(all_data)
   all_data = all_data[do.call(order, all_data[grep(x = colnames(all_data), "Nazwa stacji|Rok hydro|w roku hydro|Dzie")]), ]
   # fix dates and add as seperate column:
   yy_ind = grep(x = colnames(all_data), "Rok hydrologiczny")

R/match_imgw_wmoid_inds.R

@@ -0,0 +1,17 @@
+#' Match WMO station IDs for IMGW SYNOP
+#' @param station vector or station names provided to imgw_meteo_ family of functions
+#' @keywords internal
+#' @noRd
+match_imgw_wmoid_inds = function(station) {
+    if (is.null(station)) {
+        return(NULL)
+    }
+    pattern_combined = paste(station, collapse = "|")
+    ids_idx = grep(
+        pattern = pattern_combined,
+        x = climate::imgw_meteo_stations$station,
+        ignore.case = TRUE
+    )
+    ids = climate::imgw_meteo_stations[ids_idx, "id2"]
+    return(ids[nchar(ids) == 3])
+}