Add en-GB-oxendict locale

This commit adds the Oxford English Dictionary variant of the
British English (en-GB) locale to allow for strict spellchecking
when using the corresponding locale (en-GB-oxendict) as distinguished
from the generic British English (en-GB) locale.

Bug: chromium:1064659
Change-Id: Ia625a3566b7d89475c4853562ebbdb3afe7f9ffc
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2140610
Commit-Queue: Guillaume Jenkins <gujen@google.com>
Reviewed-by: edchin <edchin@chromium.org>
Reviewed-by: Guillaume Jenkins <gujen@google.com>
Reviewed-by: anthonyvd <anthonyvd@chromium.org>
Reviewed-by: Avi Drissman <avi@chromium.org>
Reviewed-by: Rouslan Solomakhin <rouslan@chromium.org>
Reviewed-by: Frank Tang <ftang@chromium.org>
Cr-Commit-Position: refs/heads/master@{#784165}

DEPS

@@ -1009,7 +1009,7 @@ deps = {
   },
 
   'src/third_party/hunspell_dictionaries':
-    Var('chromium_git') + '/chromium/deps/hunspell_dictionaries.git' + '@' + '6fdad5388130757e1efb58a14b8e7f9494625691',
+    Var('chromium_git') + '/chromium/deps/hunspell_dictionaries.git' + '@' + '18e09b9197a3b1d771c077c530d1a4ebad04c167',
 
   'src/third_party/icu':
     Var('chromium_git') + '/chromium/deps/icu.git' + '@' + '79326efe26e5440f530963704c3c0ff965b3a4ac',

components/language/core/common/locale_util.cc

@@ -28,13 +28,29 @@ struct LocaleUIFallbackPair {
 // Note that no (Norwegian) is an alias, and should fallback to Norwegian
 // Bokmål (nb)
 const LocaleUIFallbackPair kLocaleUIFallbackTable[] = {
-    {"en", "en-US"},     {"en-AU", "en-GB"},  {"en-CA", "en-GB"},
-    {"en-IN", "en-GB"},  {"en-NZ", "en-GB"},  {"en-ZA", "en-GB"},
-    {"es-AR", "es-419"}, {"es-CL", "es-419"}, {"es-CO", "es-419"},
-    {"es-CR", "es-419"}, {"es-HN", "es-419"}, {"es-MX", "es-419"},
-    {"es-PE", "es-419"}, {"es-US", "es-419"}, {"es-UY", "es-419"},
-    {"es-VE", "es-419"}, {"it-CH", "it"},     {"no", "nb"},
-    {"pt", "pt-PT"}};
+    // clang-format off
+    {"en", "en-US"},
+    {"en-AU", "en-GB"},
+    {"en-CA", "en-GB"},
+    {"en-GB-oxendict", "en-GB"},
+    {"en-IN", "en-GB"},
+    {"en-NZ", "en-GB"},
+    {"en-ZA", "en-GB"},
+    {"es-AR", "es-419"},
+    {"es-CL", "es-419"},
+    {"es-CO", "es-419"},
+    {"es-CR", "es-419"},
+    {"es-HN", "es-419"},
+    {"es-MX", "es-419"},
+    {"es-PE", "es-419"},
+    {"es-US", "es-419"},
+    {"es-UY", "es-419"},
+    {"es-VE", "es-419"},
+    {"it-CH", "it"},
+    {"no", "nb"},
+    {"pt", "pt-PT"}
+    // clang-format on
+};
 
 bool LocaleCompare(const LocaleUIFallbackPair& p1, const std::string& p2) {
   return p1.chosen_locale < p2;

components/language/core/common/locale_util_unittest.cc

@@ -78,7 +78,7 @@ TEST_F(LocaleUtilTest, ContainsSameBaseLanguage) {
   EXPECT_EQ(true, language::ContainsSameBaseLanguage(list, "fr-FR"));
 
   // Multiple elements, no match.
-  list = {"en-US", "es-AR", "en-UK"};
+  list = {"en-US", "es-AR", "en-GB"};
   EXPECT_EQ(false, language::ContainsSameBaseLanguage(list, "fr-FR"));
 
   // Multiple elements, with match.
@@ -187,6 +187,11 @@ TEST_F(LocaleUtilTest, ConvertToActualUILocale) {
   EXPECT_TRUE(is_ui);
   EXPECT_EQ("en-GB", locale);
 
+  locale = "en-GB-oxendict";
+  is_ui = language::ConvertToActualUILocale(&locale);
+  EXPECT_TRUE(is_ui);
+  EXPECT_EQ("en-GB", locale);
+
   locale = "en-IN";
   is_ui = language::ConvertToActualUILocale(&locale);
   EXPECT_TRUE(is_ui);

components/spellcheck/common/spellcheck_common.cc

@@ -29,6 +29,7 @@ struct LanguageVersion {
 static constexpr LanguageRegion kSupportedSpellCheckerLanguages[] = {
     // Several languages are not to be included in the spellchecker list:
     // th-TH, vi-VI.
+    // clang-format off
     {"af", "af-ZA"},
     {"bg", "bg-BG"},
     {"ca", "ca-ES"},
@@ -40,6 +41,7 @@ static constexpr LanguageRegion kSupportedSpellCheckerLanguages[] = {
     {"en-AU", "en-AU"},
     {"en-CA", "en-CA"},
     {"en-GB", "en-GB"},
+    {"en-GB-oxendict", "en-GB-oxendict"},
     {"en-US", "en-US"},
     {"es", "es-ES"},
     {"es-419", "es-ES"},
@@ -79,6 +81,7 @@ static constexpr LanguageRegion kSupportedSpellCheckerLanguages[] = {
     {"tr", "tr-TR"},
     {"uk", "uk-UA"},
     {"vi", "vi-VN"},
+    // clang-format on
 };
 
 bool IsValidRegion(const std::string& region) {
@@ -136,6 +139,9 @@ base::FilePath GetVersionedFileName(base::StringPiece input_language,
 
       // March 2020: Update uk-UA dictionary from upstream.
       {"uk-UA", "-4-0"},
+
+      // June 2020: Add the en-GB-oxendict dictionary.
+      {"en-GB-oxendict", "-9-0"},
   };
 
   // Generate the bdict file name using default version string or special

components/translate/core/browser/translate_prefs_unittest.cc

@@ -11,6 +11,7 @@
 #include <vector>
 
 #include "base/json/json_reader.h"
+#include "base/strings/utf_string_conversions.h"
 #include "base/test/scoped_feature_list.h"
 #include "base/test/test_timeouts.h"
 #include "base/values.h"
@@ -26,6 +27,7 @@
 #include "testing/gmock/include/gmock/gmock.h"
 #include "testing/gtest/include/gtest/gtest.h"
 #include "ui/base/l10n/l10n_util.h"
+#include "ui/base/l10n/l10n_util_collator.h"
 
 namespace {
 
@@ -138,11 +140,11 @@ class TranslatePrefsTest : public testing::Test {
 
   // Returns a vector of display names from the elements of the given
   // |language_list|.
-  std::vector<std::string> ExtractDisplayNames(
+  std::vector<base::string16> ExtractDisplayNames(
       const std::vector<TranslateLanguageInfo>& language_list) const {
-    std::vector<std::string> output;
+    std::vector<base::string16> output;
     for (const auto& item : language_list) {
-      output.push_back(item.display_name);
+      output.push_back(base::UTF8ToUTF16(item.display_name));
     }
     return output;
   }
@@ -362,10 +364,11 @@ TEST_F(TranslatePrefsTest, GetLanguageInfoListOutput) {
   language_list.clear();
   TranslatePrefs::GetLanguageInfoList("en-US", true /* translate_enabled */,
                                       &language_list);
-  const std::vector<std::string> display_names =
+  const std::vector<base::string16> display_names =
       ExtractDisplayNames(language_list);
-  std::vector<std::string> sorted(display_names);
-  std::sort(sorted.begin(), sorted.end());
+  std::vector<base::string16> sorted(display_names);
+  l10n_util::SortVectorWithStringKey("en-US", &sorted, false);
+
   EXPECT_THAT(display_names, ElementsAreArray(sorted));
 }
 

ios/chrome/browser/ui/settings/language/BUILD.gn

@@ -87,7 +87,9 @@ source_set("unit_tests") {
     "//ios/chrome/browser/translate",
     "//ios/chrome/browser/ui/settings/language",
     "//ios/chrome/browser/ui/settings/language/cells",
+    "//testing/gmock",
     "//testing/gtest",
+    "//ui/base:base",
   ]
 }
 

ios/chrome/browser/ui/settings/language/language_settings_mediator_unittest.mm

@@ -23,8 +23,10 @@
 #import "ios/chrome/browser/ui/settings/language/cells/language_item.h"
 #import "ios/chrome/browser/ui/settings/language/language_settings_consumer.h"
 #import "ios/chrome/browser/ui/settings/language/language_settings_mediator.h"
+#include "testing/gmock/include/gmock/gmock.h"
 #include "testing/gtest_mac.h"
 #include "testing/platform_test.h"
+#include "ui/base/l10n/l10n_util_collator.h"
 
 #if !defined(__has_feature) || !__has_feature(objc_arc)
 #error "This file requires ARC support."
@@ -35,18 +37,19 @@ using language::prefs::kAcceptLanguages;
 using sync_preferences::PrefServiceMockFactory;
 using sync_preferences::PrefServiceSyncable;
 using user_prefs::PrefRegistrySyncable;
+using ::testing::ElementsAreArray;
 
 namespace {
 
 // Constant for timeout while waiting for asynchronous sync operations.
 const NSTimeInterval kSyncOperationTimeout = 10.0;
 
-std::vector<std::string> ExtractDisplayNamesFromLanguageItems(
+std::vector<base::string16> ExtractDisplayNamesFromLanguageItems(
     NSArray<LanguageItem*>* language_items) {
-  __block std::vector<std::string> output;
+  __block std::vector<base::string16> output;
   [language_items enumerateObjectsUsingBlock:^(LanguageItem* item,
                                                NSUInteger index, BOOL* stop) {
-    output.push_back(base::SysNSStringToUTF8(item.text));
+    output.push_back(base::SysNSStringToUTF16(item.text));
   }];
   return output;
 }
@@ -182,9 +185,11 @@ TEST_F(LanguageSettingsMediatorTest, TestPrefsChanged) {
 // and excludes languages already in the accept languages list.
 TEST_F(LanguageSettingsMediatorTest, TestSupportedLanguagesItems) {
   NSArray<LanguageItem*>* language_items = [mediator() supportedLanguagesItems];
-  std::vector<std::string> display_names =
+  std::vector<base::string16> display_names =
       ExtractDisplayNamesFromLanguageItems(language_items);
-  EXPECT_TRUE(std::is_sorted(display_names.begin(), display_names.end()));
+  std::vector<base::string16> sorted(display_names);
+  l10n_util::SortVectorWithStringKey("en-US", &sorted, false);
+  EXPECT_THAT(display_names, ElementsAreArray(sorted));
 
   std::vector<std::string> language_codes =
       ExtractLanguageCodesFromLanguageItems(language_items);

ui/base/l10n/l10n_util.cc

@@ -66,7 +66,7 @@ static const char* const kAcceptLanguageList[] = {
     "bs",              // Bosnian
     "ca",              // Catalan
     "ceb",             // Cebuano
-    "ckb",    // Kurdish (Arabci),  Sorani
+    "ckb",             // Kurdish (Arabic),  Sorani
     "co",              // Corsican
     "cs",              // Czech
     "cy",              // Welsh
@@ -81,6 +81,7 @@ static const char* const kAcceptLanguageList[] = {
     "en-AU",           // English (Australia)
     "en-CA",           // English (Canada)
     "en-GB",           // English (UK)
+    "en-GB-oxendict",  // English (UK, OED spelling)
     "en-IN",           // English (India)
     "en-NZ",           // English (New Zealand)
     "en-US",           // English (US)
@@ -383,18 +384,15 @@ bool CheckAndResolveLocale(const std::string& locale,
         tmp_locale.append("-CN");
       }
     } else if (base::LowerCaseEqualsASCII(lang, "en")) {
-      // Map Australian, Canadian, Indian, New Zealand and South African
-      // English to British English for now.
+      // Map Liberian and Filipino English to US English, and everything
+      // else to British English.
       // TODO(jungshik): en-CA may have to change sides once
       // we have OS locale separate from app locale (Chrome's UI language).
-      if (base::LowerCaseEqualsASCII(region, "au") ||
-          base::LowerCaseEqualsASCII(region, "ca") ||
-          base::LowerCaseEqualsASCII(region, "in") ||
-          base::LowerCaseEqualsASCII(region, "nz") ||
-          base::LowerCaseEqualsASCII(region, "za")) {
-        tmp_locale.append("-GB");
-      } else {
+      if (base::LowerCaseEqualsASCII(region, "lr") ||
+          base::LowerCaseEqualsASCII(region, "ph")) {
         tmp_locale.append("-US");
+      } else {
+        tmp_locale.append("-GB");
       }
     }
     if (IsLocaleAvailable(tmp_locale)) {