Revert "Merged: Correct the misuse of uloc_{to,from}LanguageTag"

natorion · Commit Bot · commit 9a132a9e7208 · 2017-10-17T16:10:36.000Z
This reverts commit 593b0d8. Reason for revert: broke some branch builders like https://build.chromium.org/p/client.v8.branches/builders/V8%20arm%20-%20sim%20-%20beta%20branch%20-%20debug Original change's description: > Merged: Correct the misuse of uloc_{to,from}LanguageTag > > Revision: 69bd294 > > Merge to 6.3 branch > > - remove unused Runtime_GetLanguageTagVariants > - add test for another related bug (chromium:770452) as well as for > chromium:770450 . > > BUG=chromium:770450,chromium:770452 > LOG=N > NOTRY=true > NOPRESUBMIT=true > NOTREECHECKS=true > R=​adamk@chromium.org > > Cq-Include-Trybots: master.tryserver.v8:v8_linux_noi18n_rel_ng > Change-Id: Ia2c6dd2156c51995fb18228fc3062a86e78d719c > Reviewed-on: https://chromium-review.googlesource.com/721844 > Reviewed-by: Adam Klein <adamk@chromium.org> > Commit-Queue: Jungshik Shin <jshin@chromium.org> > Cr-Commit-Position: refs/branch-heads/6.3@{#9} > Cr-Branched-From: 094a7c9-refs/heads/6.3.292@{#1} > Cr-Branched-From: 18b8fbb-refs/heads/master@{#48432} TBR=adamk@chromium.org,hablich@chromium.org,jshin@chromium.org Change-Id: I37018f8241efe1431f453ff55cf8216a5daa66de No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: chromium:770450, chromium:770452 Cq-Include-Trybots: master.tryserver.v8:v8_linux_noi18n_rel_ng Reviewed-on: https://chromium-review.googlesource.com/723323 Reviewed-by: Michael Hablich <hablich@chromium.org> Commit-Queue: Michael Hablich <hablich@chromium.org> Cr-Commit-Position: refs/branch-heads/6.3@{#17} Cr-Branched-From: 094a7c9-refs/heads/6.3.292@{#1} Cr-Branched-From: 18b8fbb-refs/heads/master@{#48432}
diff --git a/src/runtime/runtime-intl.cc b/src/runtime/runtime-intl.cc
@@ -68,21 +68,18 @@ RUNTIME_FUNCTION(Runtime_CanonicalizeLanguageTag) {
   v8::String::Utf8Value locale_id(v8_isolate,
                                   v8::Utils::ToLocal(locale_id_str));
 
-  // TODO(jshin): uloc_{for,to}TanguageTag can fail even for a structually valid
-  // language tag if it's too long (much longer than 100 chars). Even if we
-  // allocate a longer buffer, ICU will still fail if it's too long. Either
-  // propose to Ecma 402 to put a limit on the locale length or change ICU to
-  // handle long locale names better. See
-  // https://ssl.icu-project.org/trac/ticket/13417 .
-
   // Return value which denotes invalid language tag.
+  // TODO(jshin): Can uloc_{for,to}TanguageTag fail even for structually valid
+  // language tags? If not, just add CHECK instead of returning 'invalid-tag'.
   const char* const kInvalidTag = "invalid-tag";
 
   UErrorCode error = U_ZERO_ERROR;
   char icu_result[ULOC_FULLNAME_CAPACITY];
-  uloc_forLanguageTag(*locale_id, icu_result, ULOC_FULLNAME_CAPACITY, nullptr,
-                      &error);
-  if (U_FAILURE(error) || error == U_STRING_NOT_TERMINATED_WARNING) {
+  int icu_length = 0;
+
+  uloc_forLanguageTag(*locale_id, icu_result, ULOC_FULLNAME_CAPACITY,
+                      &icu_length, &error);
+  if (U_FAILURE(error) || icu_length == 0) {
     return *factory->NewStringFromAsciiChecked(kInvalidTag);
   }
 
@@ -91,7 +88,7 @@ RUNTIME_FUNCTION(Runtime_CanonicalizeLanguageTag) {
   // Force strict BCP47 rules.
   uloc_toLanguageTag(icu_result, result, ULOC_FULLNAME_CAPACITY, TRUE, &error);
 
-  if (U_FAILURE(error) || error == U_STRING_NOT_TERMINATED_WARNING) {
+  if (U_FAILURE(error)) {
     return *factory->NewStringFromAsciiChecked(kInvalidTag);
   }
 
@@ -137,7 +134,7 @@ RUNTIME_FUNCTION(Runtime_AvailableLocalesOf) {
     error = U_ZERO_ERROR;
     // No need to force strict BCP47 rules.
     uloc_toLanguageTag(icu_name, result, ULOC_FULLNAME_CAPACITY, FALSE, &error);
-    if (U_FAILURE(error) || error == U_STRING_NOT_TERMINATED_WARNING) {
+    if (U_FAILURE(error)) {
       // This shouldn't happen, but lets not break the user.
       continue;
     }
@@ -176,6 +173,91 @@ RUNTIME_FUNCTION(Runtime_GetDefaultICULocale) {
   return *factory->NewStringFromStaticChars("und");
 }
 
+RUNTIME_FUNCTION(Runtime_GetLanguageTagVariants) {
+  HandleScope scope(isolate);
+  v8::Isolate* v8_isolate = reinterpret_cast<v8::Isolate*>(isolate);
+  Factory* factory = isolate->factory();
+
+  DCHECK_EQ(1, args.length());
+
+  CONVERT_ARG_HANDLE_CHECKED(JSArray, input, 0);
+
+  uint32_t length = static_cast<uint32_t>(input->length()->Number());
+  // Set some limit to prevent fuzz tests from going OOM.
+  // Can be bumped when callers' requirements change.
+  if (length >= 100) return isolate->ThrowIllegalOperation();
+  Handle<FixedArray> output = factory->NewFixedArray(length);
+  Handle<Name> maximized = factory->NewStringFromStaticChars("maximized");
+  Handle<Name> base = factory->NewStringFromStaticChars("base");
+  for (unsigned int i = 0; i < length; ++i) {
+    Handle<Object> locale_id;
+    ASSIGN_RETURN_FAILURE_ON_EXCEPTION(
+        isolate, locale_id, JSReceiver::GetElement(isolate, input, i));
+    if (!locale_id->IsString()) {
+      return isolate->Throw(*factory->illegal_argument_string());
+    }
+
+    v8::String::Utf8Value utf8_locale_id(
+        v8_isolate, v8::Utils::ToLocal(Handle<String>::cast(locale_id)));
+
+    UErrorCode error = U_ZERO_ERROR;
+
+    // Convert from BCP47 to ICU format.
+    // de-DE-u-co-phonebk -> de_DE@collation=phonebook
+    char icu_locale[ULOC_FULLNAME_CAPACITY];
+    int icu_locale_length = 0;
+    uloc_forLanguageTag(*utf8_locale_id, icu_locale, ULOC_FULLNAME_CAPACITY,
+                        &icu_locale_length, &error);
+    if (U_FAILURE(error) || icu_locale_length == 0) {
+      return isolate->Throw(*factory->illegal_argument_string());
+    }
+
+    // Maximize the locale.
+    // de_DE@collation=phonebook -> de_Latn_DE@collation=phonebook
+    char icu_max_locale[ULOC_FULLNAME_CAPACITY];
+    uloc_addLikelySubtags(icu_locale, icu_max_locale, ULOC_FULLNAME_CAPACITY,
+                          &error);
+
+    // Remove extensions from maximized locale.
+    // de_Latn_DE@collation=phonebook -> de_Latn_DE
+    char icu_base_max_locale[ULOC_FULLNAME_CAPACITY];
+    uloc_getBaseName(icu_max_locale, icu_base_max_locale,
+                     ULOC_FULLNAME_CAPACITY, &error);
+
+    // Get original name without extensions.
+    // de_DE@collation=phonebook -> de_DE
+    char icu_base_locale[ULOC_FULLNAME_CAPACITY];
+    uloc_getBaseName(icu_locale, icu_base_locale, ULOC_FULLNAME_CAPACITY,
+                     &error);
+
+    // Convert from ICU locale format to BCP47 format.
+    // de_Latn_DE -> de-Latn-DE
+    char base_max_locale[ULOC_FULLNAME_CAPACITY];
+    uloc_toLanguageTag(icu_base_max_locale, base_max_locale,
+                       ULOC_FULLNAME_CAPACITY, FALSE, &error);
+
+    // de_DE -> de-DE
+    char base_locale[ULOC_FULLNAME_CAPACITY];
+    uloc_toLanguageTag(icu_base_locale, base_locale, ULOC_FULLNAME_CAPACITY,
+                       FALSE, &error);
+
+    if (U_FAILURE(error)) {
+      return isolate->Throw(*factory->illegal_argument_string());
+    }
+
+    Handle<JSObject> result = factory->NewJSObject(isolate->object_function());
+    Handle<String> value = factory->NewStringFromAsciiChecked(base_max_locale);
+    JSObject::AddProperty(result, maximized, value, NONE);
+    value = factory->NewStringFromAsciiChecked(base_locale);
+    JSObject::AddProperty(result, base, value, NONE);
+    output->set(i, *result);
+  }
+
+  Handle<JSArray> result = factory->NewJSArrayWithElements(output);
+  result->set_length(Smi::FromInt(length));
+  return *result;
+}
+
 RUNTIME_FUNCTION(Runtime_IsInitializedIntlObject) {
   HandleScope scope(isolate);
 
diff --git a/src/runtime/runtime.h b/src/runtime/runtime.h
@@ -252,6 +252,7 @@ namespace internal {
   F(CanonicalizeLanguageTag, 1, 1)           \
   F(AvailableLocalesOf, 1, 1)                \
   F(GetDefaultICULocale, 0, 1)               \
+  F(GetLanguageTagVariants, 1, 1)            \
   F(IsInitializedIntlObject, 1, 1)           \
   F(IsInitializedIntlObjectOfType, 2, 1)     \
   F(MarkAsInitializedIntlObjectOfType, 2, 1) \
diff --git a/test/intl/general/invalid-locale.js b/test/intl/general/invalid-locale.js
