Merge branch 'main' of github.com:ropensci/parzer

AlbanSagouis · AlbanSagouis · commit b703a2360ab2 · 2025-12-02T15:04:12.000+01:00
diff --git a/.github/workflows/check-standard.yml b/.github/workflows/check-standard.yml
@@ -33,7 +33,7 @@ jobs:
       R_KEEP_PKG_SOURCE: yes
 
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v6
 
       - uses: r-lib/actions/setup-pandoc@v2
 
diff --git a/CRAN-SUBMISSION b/CRAN-SUBMISSION
@@ -1,3 +1,3 @@
-Version: 0.4.3
-Date: 2025-05-29 14:46:46 UTC
-SHA: 10a8aa68bea8b18a69cc99326e949c5beb144afd
+Version: 0.4.4
+Date: 2025-07-23 12:38:35 UTC
+SHA: 11f3693d2f3ce60c26146e771d0dd17e7385d318
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -8,7 +8,7 @@ Description: Parse messy geographic coordinates from various character formats
     minutes, or seconds; add and subtract degrees, minutes,
     and seconds. C++ code herein originally inspired from code
     written by Jeffrey D. Bogan, but then completely re-written.
-Version: 0.4.3
+Version: 0.4.4
 Authors@R: c(
     person("Scott", "Chamberlain", role = "aut",
         email = "sckott@protonmail.com",
diff --git a/NEWS.md b/NEWS.md
@@ -1,3 +1,11 @@
+parzer 0.4.4
+============
+
+### BUG FIX
+
+* Fixed a bug in longitude conversion where "74E5423" would be converted to NA while "74W5423" was correctly converted.
+* Improved C++ management of NA values.
+
 parzer 0.4.3
 ============
 
diff --git a/R/dms-fxns.R b/R/dms-fxns.R
@@ -105,33 +105,17 @@ pz_s <- function(x) {
 `+.pz` <- function(e1, e2) {
   e1u <- unclass_strip_atts(e1)
   e2u <- unclass_strip_atts(e2)
-  e1 <- switch(attr(e1, "type"),
-    deg = e1u,
-    min = e1u / 60,
-    sec = e1u / 3600
-  )
-  e2 <- switch(attr(e2, "type"),
-    deg = e2u,
-    min = e2u / 60,
-    sec = e2u / 3600
-  )
+  e1 <- switch(attr(e1, "type"), deg = e1u, min = e1u / 60, sec = e1u / 3600)
+  e2 <- switch(attr(e2, "type"), deg = e2u, min = e2u / 60, sec = e2u / 3600)
   structure(e1 + e2, class = "pz", type = "deg")
 }
 #' @export
 #' @rdname dms
 `-.pz` <- function(e1, e2) {
   e1u <- unclass_strip_atts(e1)
   e2u <- unclass_strip_atts(e2)
-  e1 <- switch(attr(e1, "type"),
-    deg = e1u,
-    min = e1u / 60,
-    sec = e1u / 3600
-  )
-  e2 <- switch(attr(e2, "type"),
-    deg = e2u,
-    min = e2u / 60,
-    sec = e2u / 3600
-  )
+  e1 <- switch(attr(e1, "type"), deg = e1u, min = e1u / 60, sec = e1u / 3600)
+  e2 <- switch(attr(e2, "type"), deg = e2u, min = e2u / 60, sec = e2u / 3600)
   structure(e1 - e2, class = "pz", type = "deg")
 }
 #' @export
diff --git a/R/parzer-package.R b/R/parzer-package.R
@@ -8,4 +8,4 @@ NULL
 #' `r lifecycle::badge("stable")`
 #'
 #' parse geographic coordinates
-"_PACKAGE"
+"_PACKAGE"
diff --git a/R/zzz.R b/R/zzz.R
@@ -1,7 +1,9 @@
 assert <- function(x, y) {
   if (!is.null(x)) {
     if (!inherits(x, y)) {
-      stop(deparse(substitute(x)), " must be of class ",
+      stop(
+        deparse(substitute(x)),
+        " must be of class ",
         paste0(y, collapse = ", "),
         call. = FALSE
       )
diff --git a/codemeta.json b/codemeta.json
@@ -4,23 +4,17 @@
   "identifier": "parzer",
   "description": "Parse messy geographic coordinates from various character formats to decimal degree numeric values. Parse coordinates into their parts (degree, minutes, seconds); calculate hemisphere from coordinates; pull out individually degrees, minutes, or seconds; add and subtract degrees, minutes, and seconds. C++ code herein originally inspired from code written by Jeffrey D. Bogan, but then completely re-written.",
   "name": "parzer: Parse Messy Geographic Coordinates",
-  "relatedLink": ["https://docs.ropensci.org/parzer/", "https://CRAN.R-project.org/package=parzer"],
+  "relatedLink": "https://docs.ropensci.org/parzer/",
   "codeRepository": "https://github.com/ropensci/parzer",
   "issueTracker": "https://github.com/ropensci/parzer/issues",
   "license": "https://spdx.org/licenses/MIT",
-  "version": "0.4.3",
+  "version": "0.4.4",
   "programmingLanguage": {
     "@type": "ComputerLanguage",
     "name": "R",
     "url": "https://r-project.org"
   },
   "runtimePlatform": "R version 4.5.0 (2025-04-11)",
-  "provider": {
-    "@id": "https://cran.r-project.org",
-    "@type": "Organization",
-    "name": "Comprehensive R Archive Network (CRAN)",
-    "url": "https://cran.r-project.org"
-  },
   "author": [
     {
       "@type": "Person",
@@ -198,7 +192,7 @@
   "applicationCategory": "Geospatial",
   "isPartOf": "https://ropensci.org",
   "keywords": ["geospatial", "data", "latitude", "longitude", "parser", "coordinates", "r", "rstats", "geo", "r-package"],
-  "fileSize": "4749.036KB",
+  "fileSize": "6120.123KB",
   "releaseNotes": "https://github.com/ropensci/parzer/blob/master/NEWS.md",
   "readme": "https://github.com/ropensci/parzer/blob/main/README.md",
   "contIntegration": ["https://github.com/ropensci/parzer/actions/", "https://app.codecov.io/github/ropensci/parzer?branch=main"],
diff --git a/cran-comments.md b/cran-comments.md
@@ -4,14 +4,12 @@
 
 * This is a new release.
 * `revdepcheck` passed
-* Local and GHA and win_devel R CMD check passed
-
-## Changes not listed in NEWS
-
-* Removed the README file from the package directory as requested by Professor Ripley.
+* Local and GHA and win_devel and r-project macOS buider R CMD check passed
 
 Dear CRAN reviewer,
 
+The package was silently archived because of a C++ undefined cast. Hopefully this has been fixed.
+
 Thank you for your time reviewing this release.
 
 Dr Alban Sagouis
diff --git a/revdep/README.md b/revdep/README.md
@@ -10,15 +10,16 @@
 |collate  |en_US.UTF-8                                                                  |
 |ctype    |en_US.UTF-8                                                                  |
 |tz       |Europe/Berlin                                                                |
-|date     |2025-05-19                                                                   |
+|date     |2025-07-22                                                                   |
 |pandoc   |3.6.4 @ /opt/homebrew/bin/pandoc                                             |
-|quarto   |1.6.40 @ /Applications/Positron.app/Contents/Resources/app/quarto/bin/quarto |
+|quarto   |1.7.31 @ /Applications/Positron.app/Contents/Resources/app/quarto/bin/quarto |
 
 # Dependencies
 
-|package |old   |new   |Δ  |
-|:-------|:-----|:-----|:--|
-|parzer  |0.4.1 |0.4.2 |*  |
+|package |old |new   |Δ  |
+|:-------|:---|:-----|:--|
+|parzer  |NA  |0.4.4 |*  |
+|Rcpp    |NA  |1.1.0 |*  |
 
 # Revdeps
 
diff --git a/src/RcppExports.cpp b/src/RcppExports.cpp
@@ -67,7 +67,7 @@ BEGIN_RCPP
 END_RCPP
 }
 // split_decimal_degree
-std::vector<double> split_decimal_degree(const double& x, const std::string& fmt);
+Rcpp::List split_decimal_degree(const double& x, const std::string& fmt);
 RcppExport SEXP _parzer_split_decimal_degree(SEXP xSEXP, SEXP fmtSEXP) {
 BEGIN_RCPP
     Rcpp::RObject rcpp_result_gen;
diff --git a/src/latlong.cpp b/src/latlong.cpp
@@ -132,10 +132,11 @@ bool has_non_direction_letters(std::string& s, const std::string& reggex) {
   return z;
 }
 
+// useful to NA on "-45.23232e24"
 bool has_e_with_trailing_numbers(const std::string& s) {
   bool res = false;
   // s = str_tolower(s);
-  std::regex reg("[0-9]+e[0-9]+");
+  std::regex reg("[0-9]+e[0-9]+$");
   std::smatch match;
   if (std::regex_search(s, match, reg)) {
     res = true;
@@ -243,7 +244,7 @@ double convert_lon(std::string& str) {
   if (
       str.size() == 0 ||
         !any_digits(str) ||
-        has_non_direction_letters(str, "abcfghijklmnopqrstuvxyz") ||
+        has_non_direction_letters(str, "abcfghijklmnopqrstuvxyz")  ||
         has_e_with_trailing_numbers(str)
   ) {
     ret = NA_REAL;
@@ -294,4 +295,3 @@ double convert_lon(std::string& str) {
   }
   return ret;
 }
-
diff --git a/src/pz_parse_parts.cpp b/src/pz_parse_parts.cpp
@@ -2,14 +2,15 @@
 #include "latlong.h"
 
 // [[Rcpp::export]]
-std::vector<double> split_decimal_degree(const double& x, const std::string& fmt = "dms") {
+Rcpp::List split_decimal_degree(const double& x, const std::string& fmt = "dms") {
+  //std::vector<double> split_decimal_degree(const double& x, const std::string& fmt = "dms") {
   const double sixty = 60;
   const double thirtysixh = 3600;
   double dir_val = 1.0;
 
-  if ( R_IsNA(x)) {
-    std::vector<double>{0, 0, 0}; // this has to be updated
-    // return Rcpp::List::create(NA_REAL, NA_REAL, NA_REAL);
+  if (std::isnan(x)) {
+    //std::vector<double>{0, 0, 0}; // this has to be updated
+    return Rcpp::List::create(NA_REAL, NA_REAL, NA_REAL);
   }
   // auto x_str = Rcpp::toString(x); // Rcpp should be replaced here
   // if (is_negative(x)) { // does not use Rcpp but uses regex
@@ -23,7 +24,8 @@ std::vector<double> split_decimal_degree(const double& x, const std::string& fmt
   double s = ((x_abs - d) - (m/sixty)) * thirtysixh;
   d = d * dir_val;
 
-  return std::vector<double>{d, m, s};
+  return Rcpp::List::create(d, m, s);
+  // return std::vector<double>{d, m, s};
 }
 
 // [[Rcpp::export]]
@@ -35,7 +37,8 @@ Rcpp::DataFrame pz_parse_parts_lat(std::vector<std::string>& x) {
 
   for (int i=0; i < n; ++i) {
     double out = convert_lat(x[i]); // passed as a reference.
-    std::vector<double> parts = split_decimal_degree(out); // passed as a const reference.
+    //std::vector<double>  = split_decimal_degree(out); // passed as a const reference.
+    Rcpp::List parts = split_decimal_degree(out); // passed as a const reference.
     deg[i] = parts[0];
     min[i] = parts[1];
     sec[i] = parts[2];
@@ -55,7 +58,8 @@ Rcpp::DataFrame pz_parse_parts_lon(std::vector<std::string>& x) {
 
   for (int i=0; i < n; ++i) {
     double out = convert_lon(x[i]); // passed as a reference.
-    std::vector<double> parts = split_decimal_degree(out); // passed as a const reference.
+    //std::vector<double> parts = split_decimal_degree(out); // passed as a const reference.
+    Rcpp::List parts = split_decimal_degree(out); // passed as a const reference.
     deg[i] = parts[0];
     min[i] = parts[1];
     sec[i] = parts[2];
@@ -65,4 +69,3 @@ Rcpp::DataFrame pz_parse_parts_lon(std::vector<std::string>& x) {
                                  Rcpp::_["sec"] = sec,
                                  Rcpp::_["stringsAsFactors"] = false);
 }
-
diff --git a/tests/testthat/setup.R b/tests/testthat/setup.R
@@ -0,0 +1,6 @@
+base::options(digits = 7)
+suppressWarnings({
+  suppressMessages({
+    if (isTRUE(require(paint))) paint::unmask_print()
+  })
+})
diff --git a/tests/testthat/test-dms.R b/tests/testthat/test-dms.R
@@ -82,7 +82,8 @@ invalid_formats <- c(
 
 test_that("dms fxns fail as expected", {
   out <- data.frame(
-    input = invalid_formats, res = NA_real_,
+    input = invalid_formats,
+    res = NA_real_,
     stringsAsFactors = FALSE
   )
   for (i in seq_along(invalid_formats)) {
diff --git a/tests/testthat/test-parse_hemisphere.R b/tests/testthat/test-parse_hemisphere.R
@@ -16,12 +16,18 @@ test_that("parse_hemisphere works", {
   # bad values
   ## one
   expect_equal(
-    suppressWarnings(parse_hemisphere("120", "-240.4183318")), "E")
+    suppressWarnings(parse_hemisphere("120", "-240.4183318")),
+    "E"
+  )
   expect_equal(
-    suppressWarnings(parse_hemisphere("420", "-40.4183318")), "S")
+    suppressWarnings(parse_hemisphere("420", "-40.4183318")),
+    "S"
+  )
   ## both
   expect_equal(
-    suppressWarnings(parse_hemisphere("-200", "-240.4183318")), "")
+    suppressWarnings(parse_hemisphere("-200", "-240.4183318")),
+    ""
+  )
 })
 
 test_that("parse_hemisphere - fails well", {
@@ -32,5 +38,8 @@ test_that("parse_hemisphere - fails well", {
 
   expect_warning(parse_hemisphere("45", "190"), "not within -90")
   expect_warning(parse_hemisphere("45", "190"), "check that you did not invert")
-  expect_warning(parse_hemisphere("190", "45"), "longitude value within 180/360 range, got: 190")
+  expect_warning(
+    parse_hemisphere("190", "45"),
+    "longitude value within 180/360 range, got: 190"
+  )
 })
diff --git a/tests/testthat/test-parse_lat.R b/tests/testthat/test-parse_lat.R
@@ -35,7 +35,8 @@ test_lats <- c(
 
 test_that("parse_lat works: run through test_lats", {
   out <- data.frame(
-    input = test_lats, res = NA_real_,
+    input = test_lats,
+    res = NA_real_,
     stringsAsFactors = FALSE
   )
   for (i in seq_along(test_lats)) {
@@ -77,13 +78,28 @@ invalid_formats <- c(
 # res column should all give NaN
 test_that("parse_lat works: invalid formats fail as expected", {
   out <- data.frame(
-    input = invalid_formats, res = NA_real_,
+    input = invalid_formats,
+    res = NA_real_,
     stringsAsFactors = FALSE
   )
   for (i in seq_along(invalid_formats)) {
-    out[i, "res"] <- suppressWarnings({parse_lat(invalid_formats[i])})
-    expect_warning({aa <- parse_lat(invalid_formats[i])})
+    out[i, "res"] <- suppressWarnings({
+      parse_lat(invalid_formats[i])
+    })
+    expect_warning({
+      aa <- parse_lat(invalid_formats[i])
+    })
     expect_type(aa, "double")
     expect_equal(aa, NaN)
   }
 })
+
+
+test_that("parse_lat correctly processes NA values", {
+  expect_equal(
+    suppressWarnings(
+      parse_lat(c("W60.1", NA, NA_character_, "12' 30'"))
+    ),
+    c(NA_real_, NA_real_, NA_real_, 12.5)
+  )
+})
diff --git a/tests/testthat/test-parse_lat_lon.R b/tests/testthat/test-parse_lat_lon.R
@@ -21,3 +21,25 @@ test_that("parse_lon_lat - fails well", {
   expect_warning(parse_lon_lat("45", "190"), "not within -90")
   expect_warning(parse_lon_lat("45", "190"), "check that you did not invert")
 })
+
+
+test_that("parse_lon_lat correctly processes NA values", {
+  expect_equal(
+    suppressWarnings(
+      parse_lon_lat("S60.1", NA_character_)
+    ),
+    data.frame(lon = NA_real_, lat = NA_real_)
+  )
+  expect_equal(
+    suppressWarnings(
+      parse_lon_lat("12' 30'", NA_character_)
+    ),
+    data.frame(lon = 12.5, lat = NA_real_)
+  )
+  expect_equal(
+    suppressWarnings(
+      parse_lon_lat(NA_character_, "12' 30'")
+    ),
+    data.frame(lon = NA_real_, lat = 12.5)
+  )
+})
diff --git a/tests/testthat/test-parse_llstr.R b/tests/testthat/test-parse_llstr.R
@@ -39,3 +39,12 @@ test_that("parse_llstr - fails well", {
   expect_warning(parse_llstr("190, 45"), "not within -90")
   expect_warning(parse_llstr("190, 45"), "check that you did not invert")
 })
+
+test_that("parse_llstr correctly processes NA values", {
+  expect_equal(
+    suppressWarnings(
+      parse_llstr(c(NA_character_, "12' 30', 12' 30'"))
+    ),
+    data.frame(lat = c(NA_real_, 12.5), lon = c(NA_real_, 12.5))
+  )
+})
diff --git a/tests/testthat/test-parse_lon.R b/tests/testthat/test-parse_lon.R
diff --git a/tests/testthat/test-parse_parts.R b/tests/testthat/test-parse_parts.R
diff --git a/tests/testthat/test-utils.R b/tests/testthat/test-utils.R
