libv8 3.3.10.2

data/lib/libv8/v8/src/extensions/experimental/language-matcher.cc

@@ -0,0 +1,251 @@
+// Copyright 2011 the V8 project authors. All rights reserved.
+// Redistribution and use in source and binary forms, with or without
+// modification, are permitted provided that the following conditions are
+// met:
+//
+//     * Redistributions of source code must retain the above copyright
+//       notice, this list of conditions and the following disclaimer.
+//     * Redistributions in binary form must reproduce the above
+//       copyright notice, this list of conditions and the following
+//       disclaimer in the documentation and/or other materials provided
+//       with the distribution.
+//     * Neither the name of Google Inc. nor the names of its
+//       contributors may be used to endorse or promote products derived
+//       from this software without specific prior written permission.
+//
+// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+// TODO(cira): Remove LanguageMatcher from v8 when ICU implements
+// language matching API.
+
+#include "language-matcher.h"
+
+#include "i18n-utils.h"
+#include "unicode/datefmt.h"  // For getAvailableLocales
+#include "unicode/locid.h"
+#include "unicode/uloc.h"
+#include "utils.h"
+
+namespace v8 {
+namespace internal {
+
+const unsigned int LanguageMatcher::kLanguageWeight = 75;
+const unsigned int LanguageMatcher::kScriptWeight = 20;
+const unsigned int LanguageMatcher::kRegionWeight = 5;
+const unsigned int LanguageMatcher::kThreshold = 50;
+const unsigned int LanguageMatcher::kPositionBonus = 1;
+const char* const LanguageMatcher::kDefaultLocale = "root";
+
+static const char* GetLanguageException(const char*);
+static bool BCP47ToICUFormat(const char*, char*);
+static int CompareLocaleSubtags(const char*, const char*);
+static bool BuildLocaleName(const char*, const char*, LocaleIDMatch*);
+
+LocaleIDMatch::LocaleIDMatch()
+    : score(-1) {
+  I18NUtils::StrNCopy(
+      bcp47_id, ULOC_FULLNAME_CAPACITY, LanguageMatcher::kDefaultLocale);
+
+  I18NUtils::StrNCopy(
+      icu_id, ULOC_FULLNAME_CAPACITY, LanguageMatcher::kDefaultLocale);
+}
+
+LocaleIDMatch& LocaleIDMatch::operator=(const LocaleIDMatch& rhs) {
+  I18NUtils::StrNCopy(this->bcp47_id, ULOC_FULLNAME_CAPACITY, rhs.bcp47_id);
+  I18NUtils::StrNCopy(this->icu_id, ULOC_FULLNAME_CAPACITY, rhs.icu_id);
+  this->score = rhs.score;
+
+  return *this;
+}
+
+// static
+void LanguageMatcher::GetBestMatchForPriorityList(
+    v8::Handle<v8::Array> locales, LocaleIDMatch* result) {
+  v8::HandleScope handle_scope;
+
+  unsigned int position_bonus = locales->Length() * kPositionBonus;
+
+  int max_score = 0;
+  LocaleIDMatch match;
+  for (unsigned int i = 0; i < locales->Length(); ++i) {
+    position_bonus -= kPositionBonus;
+
+    v8::TryCatch try_catch;
+    v8::Local<v8::Value> locale_id = locales->Get(v8::Integer::New(i));
+
+    // Return default if exception is raised when reading parameter.
+    if (try_catch.HasCaught()) break;
+
+    // JavaScript arrays can be heterogenous so check each item
+    // if it's a string.
+    if (!locale_id->IsString()) continue;
+
+    if (!CompareToSupportedLocaleIDList(locale_id->ToString(), &match)) {
+      continue;
+    }
+
+    // Skip items under threshold.
+    if (match.score < kThreshold) continue;
+
+    match.score += position_bonus;
+    if (match.score > max_score) {
+      *result = match;
+
+      max_score = match.score;
+    }
+  }
+}
+
+// static
+void LanguageMatcher::GetBestMatchForString(
+    v8::Handle<v8::String> locale, LocaleIDMatch* result) {
+  LocaleIDMatch match;
+
+  if (CompareToSupportedLocaleIDList(locale, &match) &&
+      match.score >= kThreshold) {
+    *result = match;
+  }
+}
+
+// static
+bool LanguageMatcher::CompareToSupportedLocaleIDList(
+    v8::Handle<v8::String> locale_id, LocaleIDMatch* result) {
+  static int32_t available_count = 0;
+  // Depending on how ICU data is built, locales returned by
+  // Locale::getAvailableLocale() are not guaranteed to support DateFormat,
+  // Collation and other services.  We can call getAvailableLocale() of all the
+  // services we want to support and take the intersection of them all, but
+  // using DateFormat::getAvailableLocales() should suffice.
+  // TODO(cira): Maybe make this thread-safe?
+  static const icu::Locale* available_locales =
+      icu::DateFormat::getAvailableLocales(available_count);
+
+  // Skip this locale_id if it's not in ASCII.
+  static LocaleIDMatch default_match;
+  v8::String::AsciiValue ascii_value(locale_id);
+  if (*ascii_value == NULL) return false;
+
+  char locale[ULOC_FULLNAME_CAPACITY];
+  if (!BCP47ToICUFormat(*ascii_value, locale)) return false;
+
+  icu::Locale input_locale(locale);
+
+  // Position of the best match locale in list of available locales.
+  int position = -1;
+  const char* language = GetLanguageException(input_locale.getLanguage());
+  const char* script = input_locale.getScript();
+  const char* region = input_locale.getCountry();
+  for (int32_t i = 0; i < available_count; ++i) {
+    int current_score = 0;
+    int sign =
+        CompareLocaleSubtags(language, available_locales[i].getLanguage());
+    current_score += sign * kLanguageWeight;
+
+    sign = CompareLocaleSubtags(script, available_locales[i].getScript());
+    current_score += sign * kScriptWeight;
+
+    sign = CompareLocaleSubtags(region, available_locales[i].getCountry());
+    current_score += sign * kRegionWeight;
+
+    if (current_score >= kThreshold && current_score > result->score) {
+      result->score = current_score;
+      position = i;
+    }
+  }
+
+  // Didn't find any good matches so use defaults.
+  if (position == -1) return false;
+
+  return BuildLocaleName(available_locales[position].getBaseName(),
+                         input_locale.getName(), result);
+}
+
+// For some unsupported language subtags it is better to fallback to related
+// language that is supported than to default.
+static const char* GetLanguageException(const char* language) {
+  // Serbo-croatian to Serbian.
+  if (!strcmp(language, "sh")) return "sr";
+
+  // Norweigan to Norweiaan to Norwegian Bokmal.
+  if (!strcmp(language, "no")) return "nb";
+
+  // Moldavian to Romanian.
+  if (!strcmp(language, "mo")) return "ro";
+
+  // Tagalog to Filipino.
+  if (!strcmp(language, "tl")) return "fil";
+
+  return language;
+}
+
+// Converts user input from BCP47 locale id format to ICU compatible format.
+// Returns false if uloc_forLanguageTag call fails or if extension is too long.
+static bool BCP47ToICUFormat(const char* locale_id, char* result) {
+  UErrorCode status = U_ZERO_ERROR;
+  int32_t locale_size = 0;
+
+  char locale[ULOC_FULLNAME_CAPACITY];
+  I18NUtils::StrNCopy(locale, ULOC_FULLNAME_CAPACITY, locale_id);
+
+  // uloc_forLanguageTag has a bug where long extension can crash the code.
+  // We need to check if extension part of language id conforms to the length.
+  // ICU bug: http://bugs.icu-project.org/trac/ticket/8519
+  const char* extension = strstr(locale_id, "-u-");
+  if (extension != NULL &&
+      strlen(extension) > ULOC_KEYWORD_AND_VALUES_CAPACITY) {
+    // Truncate to get non-crashing string, but still preserve base language.
+    int base_length = strlen(locale_id) - strlen(extension);
+    locale[base_length] = '\0';
+  }
+
+  uloc_forLanguageTag(locale, result, ULOC_FULLNAME_CAPACITY,
+                      &locale_size, &status);
+  return !U_FAILURE(status);
+}
+
+// Compares locale id subtags.
+// Returns 1 for match or -1 for mismatch.
+static int CompareLocaleSubtags(const char* lsubtag, const char* rsubtag) {
+  return strcmp(lsubtag, rsubtag) == 0 ? 1 : -1;
+}
+
+// Builds a BCP47 compliant locale id from base name of matched locale and
+// full user specified locale.
+// Returns false if uloc_toLanguageTag failed to convert locale id.
+// Example:
+//   base_name of matched locale (ICU ID): de_DE
+//   input_locale_name (ICU ID): de_AT@collation=phonebk
+//   result (ICU ID): de_DE@collation=phonebk
+//   result (BCP47 ID): de-DE-u-co-phonebk
+static bool BuildLocaleName(const char* base_name,
+                            const char* input_locale_name,
+                            LocaleIDMatch* result) {
+  I18NUtils::StrNCopy(result->icu_id, ULOC_LANG_CAPACITY, base_name);
+
+  // Get extensions (if any) from the original locale.
+  const char* extension = strchr(input_locale_name, ULOC_KEYWORD_SEPARATOR);
+  if (extension != NULL) {
+    I18NUtils::StrNCopy(result->icu_id + strlen(base_name),
+                        ULOC_KEYWORD_AND_VALUES_CAPACITY, extension);
+  } else {
+    I18NUtils::StrNCopy(result->icu_id, ULOC_LANG_CAPACITY, base_name);
+  }
+
+  // Convert ICU locale name into BCP47 format.
+  UErrorCode status = U_ZERO_ERROR;
+  uloc_toLanguageTag(result->icu_id, result->bcp47_id,
+                     ULOC_FULLNAME_CAPACITY, false, &status);
+  return !U_FAILURE(status);
+}
+
+} }  // namespace v8::internal

data/lib/libv8/v8/src/extensions/experimental/language-matcher.h

@@ -0,0 +1,95 @@
+// Copyright 2011 the V8 project authors. All rights reserved.
+// Redistribution and use in source and binary forms, with or without
+// modification, are permitted provided that the following conditions are
+// met:
+//
+//     * Redistributions of source code must retain the above copyright
+//       notice, this list of conditions and the following disclaimer.
+//     * Redistributions in binary form must reproduce the above
+//       copyright notice, this list of conditions and the following
+//       disclaimer in the documentation and/or other materials provided
+//       with the distribution.
+//     * Neither the name of Google Inc. nor the names of its
+//       contributors may be used to endorse or promote products derived
+//       from this software without specific prior written permission.
+//
+// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+#ifndef V8_EXTENSIONS_EXPERIMENTAL_LANGUAGE_MATCHER_H_
+#define V8_EXTENSIONS_EXPERIMENTAL_LANGUAGE_MATCHER_H_
+
+#include <v8.h>
+
+#include "unicode/uloc.h"
+
+namespace v8 {
+namespace internal {
+
+struct LocaleIDMatch {
+  LocaleIDMatch();
+
+  LocaleIDMatch& operator=(const LocaleIDMatch& rhs);
+
+  // Bcp47 locale id - "de-Latn-DE-u-co-phonebk".
+  char bcp47_id[ULOC_FULLNAME_CAPACITY];
+
+  // ICU locale id - "de_Latn_DE@collation=phonebk".
+  char icu_id[ULOC_FULLNAME_CAPACITY];
+
+  // Score for this locale.
+  int score;
+};
+
+class LanguageMatcher {
+ public:
+  // Default locale.
+  static const char* const kDefaultLocale;
+
+  // Finds best supported locale for a given a list of locale identifiers.
+  // It preserves the extension for the locale id.
+  static void GetBestMatchForPriorityList(
+      v8::Handle<v8::Array> locale_list, LocaleIDMatch* result);
+
+  // Finds best supported locale for a single locale identifier.
+  // It preserves the extension for the locale id.
+  static void GetBestMatchForString(
+      v8::Handle<v8::String> locale_id, LocaleIDMatch* result);
+
+ private:
+  // If langauge subtags match add this amount to the score.
+  static const unsigned int kLanguageWeight;
+
+  // If script subtags match add this amount to the score.
+  static const unsigned int kScriptWeight;
+
+  // If region subtags match add this amount to the score.
+  static const unsigned int kRegionWeight;
+
+  // LocaleID match score has to be over this number to accept the match.
+  static const unsigned int kThreshold;
+
+  // For breaking ties in priority queue.
+  static const unsigned int kPositionBonus;
+
+  LanguageMatcher();
+
+  // Compares locale_id to the supported list of locales and returns best
+  // match.
+  // Returns false if it fails to convert locale id from ICU to BCP47 format.
+  static bool CompareToSupportedLocaleIDList(v8::Handle<v8::String> locale_id,
+                                             LocaleIDMatch* result);
+};
+
+} }  // namespace v8::internal
+
+#endif  // V8_EXTENSIONS_EXPERIMENTAL_LANGUAGE_MATCHER_H_

data/lib/libv8/v8/src/extensions/externalize-string-extension.cc

@@ -0,0 +1,141 @@
+// Copyright 2010 the V8 project authors. All rights reserved.
+// Redistribution and use in source and binary forms, with or without
+// modification, are permitted provided that the following conditions are
+// met:
+//
+//     * Redistributions of source code must retain the above copyright
+//       notice, this list of conditions and the following disclaimer.
+//     * Redistributions in binary form must reproduce the above
+//       copyright notice, this list of conditions and the following
+//       disclaimer in the documentation and/or other materials provided
+//       with the distribution.
+//     * Neither the name of Google Inc. nor the names of its
+//       contributors may be used to endorse or promote products derived
+//       from this software without specific prior written permission.
+//
+// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+#include "externalize-string-extension.h"
+
+namespace v8 {
+namespace internal {
+
+template <typename Char, typename Base>
+class SimpleStringResource : public Base {
+ public:
+  // Takes ownership of |data|.
+  SimpleStringResource(Char* data, size_t length)
+      : data_(data),
+        length_(length) {}
+
+  virtual ~SimpleStringResource() { delete[] data_; }
+
+  virtual const Char* data() const { return data_; }
+
+  virtual size_t length() const { return length_; }
+
+ private:
+  Char* const data_;
+  const size_t length_;
+};
+
+
+typedef SimpleStringResource<char, v8::String::ExternalAsciiStringResource>
+    SimpleAsciiStringResource;
+typedef SimpleStringResource<uc16, v8::String::ExternalStringResource>
+    SimpleTwoByteStringResource;
+
+
+const char* const ExternalizeStringExtension::kSource =
+    "native function externalizeString();"
+    "native function isAsciiString();";
+
+
+v8::Handle<v8::FunctionTemplate> ExternalizeStringExtension::GetNativeFunction(
+    v8::Handle<v8::String> str) {
+  if (strcmp(*v8::String::AsciiValue(str), "externalizeString") == 0) {
+    return v8::FunctionTemplate::New(ExternalizeStringExtension::Externalize);
+  } else {
+    ASSERT(strcmp(*v8::String::AsciiValue(str), "isAsciiString") == 0);
+    return v8::FunctionTemplate::New(ExternalizeStringExtension::IsAscii);
+  }
+}
+
+
+v8::Handle<v8::Value> ExternalizeStringExtension::Externalize(
+    const v8::Arguments& args) {
+  if (args.Length() < 1 || !args[0]->IsString()) {
+    return v8::ThrowException(v8::String::New(
+        "First parameter to externalizeString() must be a string."));
+  }
+  bool force_two_byte = false;
+  if (args.Length() >= 2) {
+    if (args[1]->IsBoolean()) {
+      force_two_byte = args[1]->BooleanValue();
+    } else {
+      return v8::ThrowException(v8::String::New(
+          "Second parameter to externalizeString() must be a boolean."));
+    }
+  }
+  bool result = false;
+  Handle<String> string = Utils::OpenHandle(*args[0].As<v8::String>());
+  if (string->IsExternalString()) {
+    return v8::ThrowException(v8::String::New(
+        "externalizeString() can't externalize twice."));
+  }
+  if (string->IsAsciiRepresentation() && !force_two_byte) {
+    char* data = new char[string->length()];
+    String::WriteToFlat(*string, data, 0, string->length());
+    SimpleAsciiStringResource* resource = new SimpleAsciiStringResource(
+        data, string->length());
+    result = string->MakeExternal(resource);
+    if (result && !string->IsSymbol()) {
+      HEAP->external_string_table()->AddString(*string);
+    }
+    if (!result) delete resource;
+  } else {
+    uc16* data = new uc16[string->length()];
+    String::WriteToFlat(*string, data, 0, string->length());
+    SimpleTwoByteStringResource* resource = new SimpleTwoByteStringResource(
+        data, string->length());
+    result = string->MakeExternal(resource);
+    if (result && !string->IsSymbol()) {
+      HEAP->external_string_table()->AddString(*string);
+    }
+    if (!result) delete resource;
+  }
+  if (!result) {
+    return v8::ThrowException(v8::String::New("externalizeString() failed."));
+  }
+  return v8::Undefined();
+}
+
+
+v8::Handle<v8::Value> ExternalizeStringExtension::IsAscii(
+    const v8::Arguments& args) {
+  if (args.Length() != 1 || !args[0]->IsString()) {
+    return v8::ThrowException(v8::String::New(
+        "isAsciiString() requires a single string argument."));
+  }
+  return Utils::OpenHandle(*args[0].As<v8::String>())->IsAsciiRepresentation() ?
+      v8::True() : v8::False();
+}
+
+
+void ExternalizeStringExtension::Register() {
+  static ExternalizeStringExtension externalize_extension;
+  static v8::DeclareExtension externalize_extension_declaration(
+      &externalize_extension);
+}
+
+} }  // namespace v8::internal