Issue #1938892 by JohnAlbin, Pancho: Fixed Switch from ISO-3166-1 country data to CLDR unicode data.

4 files changed, 143 insertions, 108 deletions

diff --git a/core/lib/Drupal/Core/Locale/CountryManager.php b/core/lib/Drupal/Core/Locale/CountryManager.php
index 33a0a55..6617459 100644
--- a/core/lib/Drupal/Core/Locale/CountryManager.php
+++ b/core/lib/Drupal/Core/Locale/CountryManager.php
@@ -36,13 +36,14 @@ class CountryManager implements CountryManagerInterface {
   }
 
   /**
-   * Get an array of all ISO 3166-1 alpha-2 country code => country name pairs.
+   * Get an array of all two-letter country code => country name pairs.
    *
    * @return array
    *   An array of country code => country name pairs.
    */
   public static function getStandardList() {
     $countries = array(
+      'AC' => t('Ascension Island'),
       'AD' => t('Andorra'),
       'AE' => t('United Arab Emirates'),
       'AF' => t('Afghanistan'),
@@ -71,9 +72,9 @@ class CountryManager implements CountryManagerInterface {
       'BJ' => t('Benin'),
       'BL' => t('Saint Barthélemy'),
       'BM' => t('Bermuda'),
-      'BN' => t('Brunei Darussalam'),
-      'BO' => t('Bolivia, Plurinational State of'),
-      'BQ' => t('Bonaire, Sint Eustatius and Saba'),
+      'BN' => t('Brunei'),
+      'BO' => t('Bolivia'),
+      'BQ' => t('Caribbean Netherlands'),
       'BR' => t('Brazil'),
       'BS' => t('Bahamas'),
       'BT' => t('Bhutan'),
@@ -82,17 +83,18 @@ class CountryManager implements CountryManagerInterface {
       'BY' => t('Belarus'),
       'BZ' => t('Belize'),
       'CA' => t('Canada'),
-      'CC' => t('Cocos (Keeling) Islands'),
-      'CD' => t('Congo, The Democratic Republic of the'),
+      'CC' => t('Cocos [Keeling] Islands'),
+      'CD' => t('Congo - Kinshasa'),
       'CF' => t('Central African Republic'),
-      'CG' => t('Congo'),
+      'CG' => t('Congo - Brazzaville'),
       'CH' => t('Switzerland'),
-      'CI' => t("Côte d'Ivoire"),
+      'CI' => t('Côte d’Ivoire'),
       'CK' => t('Cook Islands'),
       'CL' => t('Chile'),
       'CM' => t('Cameroon'),
       'CN' => t('China'),
       'CO' => t('Colombia'),
+      'CP' => t('Clipperton Island'),
       'CR' => t('Costa Rica'),
       'CU' => t('Cuba'),
       'CV' => t('Cape Verde'),
@@ -101,11 +103,13 @@ class CountryManager implements CountryManagerInterface {
       'CY' => t('Cyprus'),
       'CZ' => t('Czech Republic'),
       'DE' => t('Germany'),
+      'DG' => t('Diego Garcia'),
       'DJ' => t('Djibouti'),
       'DK' => t('Denmark'),
       'DM' => t('Dominica'),
       'DO' => t('Dominican Republic'),
       'DZ' => t('Algeria'),
+      'EA' => t('Ceuta and Melilla'),
       'EC' => t('Ecuador'),
       'EE' => t('Estonia'),
       'EG' => t('Egypt'),
@@ -115,8 +119,8 @@ class CountryManager implements CountryManagerInterface {
       'ET' => t('Ethiopia'),
       'FI' => t('Finland'),
       'FJ' => t('Fiji'),
-      'FK' => t('Falkland Islands (Malvinas)'),
-      'FM' => t('Micronesia, Federated States of'),
+      'FK' => t('Falkland Islands'),
+      'FM' => t('Micronesia'),
       'FO' => t('Faroe Islands'),
       'FR' => t('France'),
       'GA' => t('Gabon'),
@@ -138,12 +142,13 @@ class CountryManager implements CountryManagerInterface {
       'GU' => t('Guam'),
       'GW' => t('Guinea-Bissau'),
       'GY' => t('Guyana'),
-      'HK' => t('Hong Kong'),
+      'HK' => t('Hong Kong SAR China'),
       'HM' => t('Heard Island and McDonald Islands'),
       'HN' => t('Honduras'),
       'HR' => t('Croatia'),
       'HT' => t('Haiti'),
       'HU' => t('Hungary'),
+      'IC' => t('Canary Islands'),
       'ID' => t('Indonesia'),
       'IE' => t('Ireland'),
       'IL' => t('Israel'),
@@ -151,7 +156,7 @@ class CountryManager implements CountryManagerInterface {
       'IN' => t('India'),
       'IO' => t('British Indian Ocean Territory'),
       'IQ' => t('Iraq'),
-      'IR' => t('Iran, Islamic Republic of'),
+      'IR' => t('Iran'),
       'IS' => t('Iceland'),
       'IT' => t('Italy'),
       'JE' => t('Jersey'),
@@ -164,12 +169,12 @@ class CountryManager implements CountryManagerInterface {
       'KI' => t('Kiribati'),
       'KM' => t('Comoros'),
       'KN' => t('Saint Kitts and Nevis'),
-      'KP' => t("Korea, Democratic People's Republic of"),
-      'KR' => t('Korea, Republic of'),
+      'KP' => t('North Korea'),
+      'KR' => t('South Korea'),
       'KW' => t('Kuwait'),
       'KY' => t('Cayman Islands'),
       'KZ' => t('Kazakhstan'),
-      'LA' => t("Lao People's Democratic Republic"),
+      'LA' => t('Laos'),
       'LB' => t('Lebanon'),
       'LC' => t('Saint Lucia'),
       'LI' => t('Liechtenstein'),
@@ -182,16 +187,16 @@ class CountryManager implements CountryManagerInterface {
       'LY' => t('Libya'),
       'MA' => t('Morocco'),
       'MC' => t('Monaco'),
-      'MD' => t('Moldova, Republic of'),
+      'MD' => t('Moldova'),
       'ME' => t('Montenegro'),
-      'MF' => t('Saint Martin (French part)'),
+      'MF' => t('Saint Martin'),
       'MG' => t('Madagascar'),
       'MH' => t('Marshall Islands'),
-      'MK' => t('Macedonia, Republic of'),
+      'MK' => t('Macedonia'),
       'ML' => t('Mali'),
-      'MM' => t('Myanmar'),
+      'MM' => t('Myanmar [Burma]'),
       'MN' => t('Mongolia'),
-      'MO' => t('Macao'),
+      'MO' => t('Macau SAR China'),
       'MP' => t('Northern Mariana Islands'),
       'MQ' => t('Martinique'),
       'MR' => t('Mauritania'),
@@ -224,17 +229,18 @@ class CountryManager implements CountryManagerInterface {
       'PK' => t('Pakistan'),
       'PL' => t('Poland'),
       'PM' => t('Saint Pierre and Miquelon'),
-      'PN' => t('Pitcairn'),
+      'PN' => t('Pitcairn Islands'),
       'PR' => t('Puerto Rico'),
-      'PS' => t('Palestine, State of'),
+      'PS' => t('Palestinian Territories'),
       'PT' => t('Portugal'),
       'PW' => t('Palau'),
       'PY' => t('Paraguay'),
       'QA' => t('Qatar'),
+      'QO' => t('Outlying Oceania'),
       'RE' => t('Réunion'),
       'RO' => t('Romania'),
       'RS' => t('Serbia'),
-      'RU' => t('Russian Federation'),
+      'RU' => t('Russia'),
       'RW' => t('Rwanda'),
       'SA' => t('Saudi Arabia'),
       'SB' => t('Solomon Islands'),
@@ -242,7 +248,7 @@ class CountryManager implements CountryManagerInterface {
       'SD' => t('Sudan'),
       'SE' => t('Sweden'),
       'SG' => t('Singapore'),
-      'SH' => t('Saint Helena, Ascension and Tristan da Cunha'),
+      'SH' => t('Saint Helena'),
       'SI' => t('Slovenia'),
       'SJ' => t('Svalbard and Jan Mayen'),
       'SK' => t('Slovakia'),
@@ -252,11 +258,12 @@ class CountryManager implements CountryManagerInterface {
       'SO' => t('Somalia'),
       'SR' => t('Suriname'),
       'SS' => t('South Sudan'),
-      'ST' => t('Sao Tome and Principe'),
+      'ST' => t('São Tomé and Príncipe'),
       'SV' => t('El Salvador'),
-      'SX' => t('Sint Maarten (Dutch part)'),
-      'SY' => t('Syrian Arab Republic'),
+      'SX' => t('Sint Maarten'),
+      'SY' => t('Syria'),
       'SZ' => t('Swaziland'),
+      'TA' => t('Tristan da Cunha'),
       'TC' => t('Turks and Caicos Islands'),
       'TD' => t('Chad'),
       'TF' => t('French Southern Territories'),
@@ -271,23 +278,24 @@ class CountryManager implements CountryManagerInterface {
       'TR' => t('Turkey'),
       'TT' => t('Trinidad and Tobago'),
       'TV' => t('Tuvalu'),
-      'TW' => t('Taiwan, Province of China'),
-      'TZ' => t('Tanzania, United Republic of'),
+      'TW' => t('Taiwan'),
+      'TZ' => t('Tanzania'),
       'UA' => t('Ukraine'),
       'UG' => t('Uganda'),
-      'UM' => t('United States Minor Outlying Islands'),
+      'UM' => t('U.S. Outlying Islands'),
       'US' => t('United States'),
       'UY' => t('Uruguay'),
       'UZ' => t('Uzbekistan'),
-      'VA' => t('Holy See (Vatican City State)'),
+      'VA' => t('Vatican City'),
       'VC' => t('Saint Vincent and the Grenadines'),
-      'VE' => t('Venezuela, Bolivarian Republic of'),
-      'VG' => t('Virgin Islands, British'),
-      'VI' => t('Virgin Islands, U.S.'),
-      'VN' => t('Viet Nam'),
+      'VE' => t('Venezuela'),
+      'VG' => t('British Virgin Islands'),
+      'VI' => t('U.S. Virgin Islands'),
+      'VN' => t('Vietnam'),
       'VU' => t('Vanuatu'),
       'WF' => t('Wallis and Futuna'),
       'WS' => t('Samoa'),
+      'XK' => t('Kosovo'),
       'YE' => t('Yemen'),
       'YT' => t('Mayotte'),
       'ZA' => t('South Africa'),
diff --git a/core/modules/system/system.api.php b/core/modules/system/system.api.php
index 16bb0de..6321a6c 100644
--- a/core/modules/system/system.api.php
+++ b/core/modules/system/system.api.php
@@ -3468,7 +3468,7 @@ function hook_updater_info_alter(&$updaters) {
  * Alter the default country list.
  *
  * @param $countries
- *   The associative array of countries keyed by ISO 3166-1 country code.
+ *   The associative array of countries keyed by two-letter country code.
  *
  * @see \Drupal\Core\Locale\CountryManager::getList().
  */
diff --git a/core/scripts/update-countries.sh b/core/scripts/update-countries.sh
new file mode 100755
index 0000000..eee5fa2
--- /dev/null
+++ b/core/scripts/update-countries.sh
@@ -0,0 +1,99 @@
+#!/bin/php
+<?php
+
+/**
+ * @file
+ * Updates CLDR codes in CountryManager.php to latest data.
+ *
+ * We rely on the CLDR data set, because it is easily accessible, scriptable,
+ * and in the right human-readable format.
+ */
+
+// Determine DRUPAL_ROOT.
+$dir = dirname(__FILE__);
+while (!defined('DRUPAL_ROOT')) {
+  if (is_dir($dir . '/core')) {
+    define('DRUPAL_ROOT', $dir);
+  }
+  $dir = dirname($dir);
+}
+
+// Determine source data file URI to process.
+$uri = DRUPAL_ROOT . '/territories.json';
+
+if (!file_exists($uri)) {
+  $usage = <<< USAGE
+- Choose the latest release data from http://cldr.unicode.org/index/downloads
+  and download the json.zip file.
+- Unzip the json.zip file and place the main/en/territories.json in the
+  Drupal root directory.
+- Run this script.
+USAGE;
+  exit('CLDR data file not found. (' . $uri . ")\n\n" . $usage . "\n");
+}
+
+// Fake the t() function used in CountryManager.php instead of attempting a full
+// Drupal bootstrap of core/includes/bootstrap.inc (where t() is declared).
+if (!function_exists('t')) {
+  function t($string) {
+    return $string;
+  }
+}
+
+// Read in existing codes.
+// @todo Allow to remove previously existing country codes.
+// @see http://drupal.org/node/1436754
+require_once DRUPAL_ROOT . '/core/lib/Drupal/Core/Locale/CountryManagerInterface.php';
+require_once DRUPAL_ROOT . '/core/lib/Drupal/Core/Locale/CountryManager.php';
+$existing_countries = \Drupal\Core\Locale\CountryManager::getStandardList();
+$countries = $existing_countries;
+
+// Parse the source data into an array.
+$data = json_decode(file_get_contents($uri));
+
+foreach ($data->main->en->localeDisplayNames->territories as $code => $name) {
+  // Use any alternate codes the Drupal community wishes to.
+  $alt_codes = array(
+    // 'CI-alt-variant', // Use CI-alt-variant instead of the CI entry.
+  );
+  if (in_array($code, $alt_codes)) {
+    // Just use the first 2 character part of the alt code.
+    $code = strtok($code, '-');
+  }
+
+  // Skip any codes we wish to exclude from our country list.
+  $exclude_codes = array(
+    'EU', // The European Union is not a country.
+    'ZZ', // Don't allow "Unknown Region".
+  );
+  if (in_array($code, $exclude_codes)) {
+    continue;
+  }
+
+  // Ignore every territory that doesn't have a 2 character code.
+  if (strlen($code) !== 2) {
+    continue;
+  }
+  $countries[(string) $code] = $name;
+}
+if (empty($countries)) {
+  echo 'ERROR: Did not find expected country names.' . PHP_EOL;
+  exit;
+}
+// Sort by country code (to minimize diffs).
+ksort($countries);
+
+// Produce PHP code.
+$out = '';
+foreach ($countries as $code => $name) {
+  // For .po translation file's sake, use double-quotes instead of escaped
+  // single-quotes.
+  $name = (strpos($name, '\'') !== FALSE ? '"' . $name . '"' : "'" . $name . "'");
+  $out .= '      ' . var_export($code, TRUE) . ' => t(' . $name . '),' . "\n";
+}
+
+// Replace the actual PHP code in standard.inc.
+$file = DRUPAL_ROOT . '/core/lib/Drupal/Core/Locale/CountryManager.php';
+$content = file_get_contents($file);
+$content = preg_replace('/(\$countries = array\(\n)(.+?)(^\s+\);)/ms', '$1' . $out . '$3', $content);
+file_put_contents($file, $content);
diff --git a/core/scripts/update-iso-3166.sh b/core/scripts/update-iso-3166.sh
deleted file mode 100644
index e06f059..0000000
--- a/core/scripts/update-iso-3166.sh
+++ /dev/null
@@ -1,72 +0,0 @@
-#!/bin/php
-<?php
-
-/**
- * @file
- * Updates ISO-3166 codes in standard.inc to latest data.
- *
- * We rely on the Debian ISO code repository, because it is easily accessible,
- * scriptable, in the right human-readable format, and all changes went through
- * sufficient FOSS community discussion already.
- */
-
-// Determine DRUPAL_ROOT.
-$cwd = $dir = dirname(__FILE__);
-while (!defined('DRUPAL_ROOT')) {
-  if (is_dir($dir . '/core')) {
-    define('DRUPAL_ROOT', $dir);
-  }
-  $dir = dirname($dir);
-}
-
-// Determine source data file URI to process.
-$uri = $cwd . '/iso_3166.xml';
-// Despite its actual file name, at least Firefox merges and converts the path
-// and filename into a combined filename.
-if (!file_exists($uri)) {
-  $uri = $cwd . '/iso_3166_iso_3166.xml';
-}
-// Fall back and default to original Debian source.
-if (!file_exists($uri)) {
-  $uri = 'http://anonscm.debian.org/gitweb/?p=iso-codes/iso-codes.git;a=blob_plain;f=iso_3166/iso_3166.xml;hb=master';
-}
-
-// Read in existing codes.
-// @todo Allow to remove previously existing country codes.
-// @see http://drupal.org/node/1436754
-require_once DRUPAL_ROOT . '/core/lib/Drupal/Core/Locale/CountryManagerInterface.php';
-require_once DRUPAL_ROOT . '/core/lib/Drupal/Core/Locale/CountryManager.php';
-$existing_countries = \Drupal\Core\Locale\CountryManager::getStandardList();
-$countries = $existing_countries;
-
-// Parse the source data into an array.
-$data = simplexml_load_file($uri);
-foreach ($data->iso_3166_entry as $entry) {
-  // Ignore every territory that doesn't have a alpha-2 code.
-  if (!isset($entry['alpha_2_code'])) {
-    continue;
-  }
-  $name = isset($entry['name']) ? (string) $entry['name'] : (string) $entry['official_name'];
-  $countries[(string) $entry['alpha_2_code']] = $name;
-}
-if (empty($countries)) {
-  echo 'ERROR: Did not find expected alpha_2_code country names.' . PHP_EOL;
-  exit;
-}
-// Sort by country code (to minimize diffs).
-ksort($countries);
-
-// Produce PHP code.
-$out = '';
-foreach ($countries as $code => $name) {
-  // For .po translation file's sake, use double-quotes instead of escaped
-  // single-quotes.
-  $name = (strpos($name, '\'') !== FALSE ? '"' . $name . '"' : "'" . $name . "'");
-  $out .= '    ' . var_export($code, TRUE) . ' => t(' . $name . '),' . "\n";
-}
-
-// Replace the actual PHP code in standard.inc.
-$file = DRUPAL_ROOT . '/core/includes/standard.inc';
-$content = file_get_contents($file);
-$content = preg_replace('/(\$countries = array\(\n)(.+?)(^\s+\);)/ms', '$1' . $out . '$3', $content);
-file_put_contents($file, $content);