Localized Chromoting Host on Windows.

remoting/tools/localize.py

+#!/usr/bin/env python
+# Copyright 2013 The Chromium Authors. All rights reserved.
+# Use of this source code is governed by a BSD-style license that can be
+# found in the LICENSE file.
+
+"""
+localize.py -- Generates an output file from the given template replacing
+variables and localizing strings.
+
+The script uses Jinja2 template processing library (src/third_party/jinja2).
+Variables available to the templates:
+  - |languages| - the list of languages passed on the command line. ('-l').
+  - Each KEY=VALUE define ('-d') can be accesses as |KEY|.
+  - |official_build| is set to '1' when CHROME_BUILD_TYPE environment variable
+    is set to "_official".
+
+Filters:
+  - GetCodepage - returns the code page for the given language.
+  - GetCodepageDecimal same as GetCodepage, but returns a decimal value.
+  - GetLangId - returns Win32 LANGID.
+  - GetPrimaryLanguage - returns a named Win32 constant specifing the primary
+    language ID.
+  - GetSublanguage -  returns a named Win32 constant specifing the sublanguage
+    ID.
+
+Globals:
+  - SelectLanguage(language) - allows to select the language to the used by
+    {% trans %}{% endtrans %} statements.
+
+"""
+
+import io
+import json
+from optparse import OptionParser
+import os
+import sys
+from string import Template
+
+'''
+See https://code.google.com/p/grit-i18n/source/browse/trunk/grit/format/rc.py
+(revision r79)
+
+This dictionary defines the language lookup table, which is used for replacing
+the GRIT expand variables for language info in Product Version resource. The key
+is the language ISO country code, and the value specifies the corresponding
+locale identifier, code page, primary language and sublanguage.
+
+LCID resource: http://msdn.microsoft.com/en-us/library/ms776294.aspx
+Codepage resource: http://www.science.co.il/language/locale-codes.asp
+Language ID resource: http://msdn.microsoft.com/en-us/library/ms776294.aspx
+
+There is no appropriate sublang for Spanish (Latin America) [es-419], so we
+use Mexico. SUBLANG_DEFAULT would incorrectly map to Spain. Unlike other
+Latin American countries, Mexican Spanish is supported by VERSIONINFO:
+http://msdn.microsoft.com/en-us/library/aa381058.aspx
+
+'''
+
+_LANGUAGE_LANGID = {
+  # Language neutral LCID, unicode(1200) code page.
+  'neutral' : [ '0000', '04b0', 'LANG_NEUTRAL', 'SUBLANG_NEUTRAL' ],
+  # LANG_USER_DEFAULT LCID, unicode(1200) code page.
+  'userdefault' : [ '0400', '04b0', 'LANG_NEUTRAL', 'SUBLANG_DEFAULT' ],
+  'ar' : [ '0401', '04e8', 'LANG_ARABIC', 'SUBLANG_DEFAULT' ],

[garykac, 2013/07/09 00:34:12]

Is there an order to this list?  I expected it to be alphabetized (except for
the few special cases).

[alexeypa (please no reviews), 2013/07/09 01:05:23]

Done.

+  'fi' : [ '040b', '04e4', 'LANG_FINNISH', 'SUBLANG_DEFAULT' ],
+  'ko' : [ '0412', '03b5', 'LANG_KOREAN', 'SUBLANG_KOREAN' ],
+  'es' : [ '040a', '04e4', 'LANG_SPANISH', 'SUBLANG_SPANISH_MODERN' ],
+  'bg' : [ '0402', '04e3', 'LANG_BULGARIAN', 'SUBLANG_DEFAULT' ],
+  # No codepage for filipino, use unicode(1200).
+  'fil' : [ '0464', '04e4', '100', 'SUBLANG_DEFAULT' ],
+  'fr' : [ '040c', '04e4', 'LANG_FRENCH', 'SUBLANG_FRENCH' ],
+  'lv' : [ '0426', '04e9', 'LANG_LATVIAN', 'SUBLANG_DEFAULT' ],
+  'sv' : [ '041d', '04e4', 'LANG_SWEDISH', 'SUBLANG_SWEDISH' ],
+  'ca' : [ '0403', '04e4', 'LANG_CATALAN', 'SUBLANG_DEFAULT' ],
+  'de' : [ '0407', '04e4', 'LANG_GERMAN', 'SUBLANG_GERMAN' ],
+  'lt' : [ '0427', '04e9', 'LANG_LITHUANIAN', 'SUBLANG_LITHUANIAN' ],
+  'zh-CN' : [ '0804', '03a8', 'LANG_CHINESE', 'SUBLANG_CHINESE_SIMPLIFIED' ],
+  'zh-TW' : [ '0404', '03b6', 'LANG_CHINESE', 'SUBLANG_CHINESE_TRADITIONAL' ],
+  'zh-HK' : [ '0c04', '03b6', 'LANG_CHINESE', 'SUBLANG_CHINESE_HONGKONG' ],
+  'el' : [ '0408', '04e5', 'LANG_GREEK', 'SUBLANG_DEFAULT' ],
+  'nb' : [ '0414', '04e4', 'LANG_NORWEGIAN', 'SUBLANG_DEFAULT' ],
+  'no' : [ '0414', '04e4', 'LANG_NORWEGIAN', 'SUBLANG_DEFAULT' ],
+  'th' : [ '041e', '036a', 'LANG_THAI', 'SUBLANG_DEFAULT' ],
+  'he' : [ '040d', '04e7', 'LANG_HEBREW', 'SUBLANG_DEFAULT' ],
+  'iw' : [ '040d', '04e7', 'LANG_HEBREW', 'SUBLANG_DEFAULT' ],
+  'pl' : [ '0415', '04e2', 'LANG_POLISH', 'SUBLANG_DEFAULT' ],
+  'tr' : [ '041f', '04e6', 'LANG_TURKISH', 'SUBLANG_DEFAULT' ],
+  'hr' : [ '041a', '04e4', 'LANG_CROATIAN', 'SUBLANG_DEFAULT' ],
+  # No codepage for Hindi, use unicode(1200).
+  'hi' : [ '0439', '04b0', 'LANG_HINDI', 'SUBLANG_DEFAULT' ],
+  'pt-PT' : [ '0816', '04e4', 'LANG_PORTUGUESE', 'SUBLANG_PORTUGUESE' ],
+  'pt-BR' : [ '0416', '04e4', 'LANG_PORTUGUESE', 'SUBLANG_DEFAULT' ],
+  'uk' : [ '0422', '04e3', 'LANG_UKRAINIAN', 'SUBLANG_DEFAULT' ],
+  'cs' : [ '0405', '04e2', 'LANG_CZECH', 'SUBLANG_DEFAULT' ],
+  'hu' : [ '040e', '04e2', 'LANG_HUNGARIAN', 'SUBLANG_DEFAULT' ],
+  'ro' : [ '0418', '04e2', 'LANG_ROMANIAN', 'SUBLANG_DEFAULT' ],
+  # No codepage for Urdu, use unicode(1200).
+  'ur' : [ '0420', '04b0', 'LANG_URDU', 'SUBLANG_DEFAULT' ],
+  'da' : [ '0406', '04e4', 'LANG_DANISH', 'SUBLANG_DEFAULT' ],
+  'is' : [ '040f', '04e4', 'LANG_ICELANDIC', 'SUBLANG_DEFAULT' ],
+  'ru' : [ '0419', '04e3', 'LANG_RUSSIAN', 'SUBLANG_DEFAULT' ],
+  'vi' : [ '042a', '04ea', 'LANG_VIETNAMESE', 'SUBLANG_DEFAULT' ],
+  'nl' : [ '0413', '04e4', 'LANG_DUTCH', 'SUBLANG_DEFAULT' ],
+  'id' : [ '0421', '04e4', 'LANG_INDONESIAN', 'SUBLANG_DEFAULT' ],
+  'sr' : [ '081a', '04e2', 'LANG_SERBIAN', 'SUBLANG_SERBIAN_CYRILLIC' ],
+  'en-GB' : [ '0809', '040e', 'LANG_ENGLISH', 'SUBLANG_ENGLISH_UK' ],
+  'it' : [ '0410', '04e4', 'LANG_ITALIAN', 'SUBLANG_DEFAULT' ],
+  'sk' : [ '041b', '04e2', 'LANG_SLOVAK', 'SUBLANG_DEFAULT' ],
+  'et' : [ '0425', '04e9', 'LANG_ESTONIAN', 'SUBLANG_DEFAULT' ],
+  'ja' : [ '0411', '03a4', 'LANG_JAPANESE', 'SUBLANG_DEFAULT' ],
+  'sl' : [ '0424', '04e2', 'LANG_SLOVENIAN', 'SUBLANG_DEFAULT' ],
+  'en' : [ '0409', '04b0', 'LANG_ENGLISH', 'SUBLANG_ENGLISH_US' ],
+  # LCID for Mexico; Windows does not support L.A. LCID.
+  'es-419' : [ '080a', '04e4', 'LANG_SPANISH', 'SUBLANG_SPANISH_MEXICAN' ],
+  # No codepage for Bengali, use unicode(1200).
+  'bn' : [ '0445', '04b0', 'LANG_BENGALI', 'SUBLANG_DEFAULT' ],
+  'fa' : [ '0429', '04e8', 'LANG_PERSIAN', 'SUBLANG_DEFAULT' ],
+  # No codepage for Gujarati, use unicode(1200).
+  'gu' : [ '0447', '04b0', 'LANG_GUJARATI', 'SUBLANG_DEFAULT' ],
+  # No codepage for Kannada, use unicode(1200).
+  'kn' : [ '044b', '04b0', 'LANG_KANNADA', 'SUBLANG_DEFAULT' ],
+  # Malay (Malaysia) [ms-MY]
+  'ms' : [ '043e', '04e4', 'LANG_MALAY', 'SUBLANG_DEFAULT' ],
+  # No codepage for Malayalam, use unicode(1200).
+  'ml' : [ '044c', '04b0', 'LANG_MALAYALAM', 'SUBLANG_DEFAULT' ],
+  # No codepage for Marathi, use unicode(1200).
+  'mr' : [ '044e', '04b0', 'LANG_MARATHI', 'SUBLANG_DEFAULT' ],
+  # No codepage for Oriya , use unicode(1200).
+  'or' : [ '0448', '04b0', 'LANG_ORIYA', 'SUBLANG_DEFAULT' ],
+  # No codepage for Tamil, use unicode(1200).
+  'ta' : [ '0449', '04b0', 'LANG_TAMIL', 'SUBLANG_DEFAULT' ],
+  # No codepage for Telugu, use unicode(1200).
+  'te' : [ '044a', '04b0', 'LANG_TELUGU', 'SUBLANG_DEFAULT' ],
+  # No codepage for Amharic, use unicode(1200). >= Vista.
+  'am' : [ '045e', '04b0', 'LANG_AMHARIC', 'SUBLANG_DEFAULT' ],
+  'sw' : [ '0441', '04e4', 'LANG_SWAHILI', 'SUBLANG_DEFAULT' ],
+  'af' : [ '0436', '04e4', 'LANG_AFRIKAANS', 'SUBLANG_DEFAULT' ],
+  'eu' : [ '042d', '04e4', 'LANG_BASQUE', 'SUBLANG_DEFAULT' ],
+  'fr-CA' : [ '0c0c', '04e4', 'LANG_FRENCH', 'SUBLANG_FRENCH_CANADIAN' ],
+  'gl' : [ '0456', '04e4', 'LANG_GALICIAN', 'SUBLANG_DEFAULT' ],
+  # No codepage for Zulu, use unicode(1200).
+  'zu' : [ '0435', '04b0', 'LANG_ZULU', 'SUBLANG_DEFAULT' ],
+  'pa' : [ '0446', '04b0', 'LANG_PUNJABI', 'SUBLANG_PUNJABI_INDIA' ],
+  'sa' : [ '044f', '04b0', 'LANG_SANSKRIT', 'SUBLANG_SANSKRIT_INDIA' ],
+  'si' : [ '045b', '04b0', 'LANG_SINHALESE', 'SUBLANG_SINHALESE_SRI_LANKA' ],
+  'ne' : [ '0461', '04b0', 'LANG_NEPALI', 'SUBLANG_NEPALI_NEPAL' ],
+  'ti' : [ '0873', '04b0', 'LANG_TIGRIGNA', 'SUBLANG_TIGRIGNA_ERITREA' ],
+  'fake-bidi' : [ '040d', '04e7', 'LANG_HEBREW', 'SUBLANG_DEFAULT' ],
+}
+
+# Right-To-Left languages
+_RTL_LANGUAGES = (
+  'ar',  # Arabic
+  'fa',  # Farsi
+  'iw',  # Hebrew
+  'ks',  # Kashmiri
+  'ku',  # Kurdish
+  'ps',  # Pashto
+  'ur',  # Urdu
+  'yi',  # Yiddish
+)
+
+
+def GetCodepage(language):
+  """ Returns the codepage for the given |language|. """
+  langid = _LANGUAGE_LANGID[language]
+  return langid[1]
+
+
+def GetCodepageDecimal(language):
+  """ Returns the codepage for the given |language| as a decimal value. """
+  langid = _LANGUAGE_LANGID[language]
+  return str(int(langid[1], 16))
+
+
+def GetLangId(language):
+  """ Returns the language id for the given |language|. """
+  langid = _LANGUAGE_LANGID[language]
+  return langid[0]
+
+
+def GetPrimaryLanguage(language):
+  """ Returns the primary language ID for the given |language|. """
+  langid = _LANGUAGE_LANGID[language]
+  return langid[2]
+
+
+def GetSublanguage(language):
+  """ Returns the sublanguage ID for the given |language|. """
+  langid = _LANGUAGE_LANGID[language]
+  return langid[3]
+
+
+def IsRtlLanguage(language):
+  return language in _RTL_LANGUAGES;
+
+
+def NormalizeLanguageCode(language):
+  return language.replace('_', '-', 1)
+
+
+def ReadValuesFromFile(values_dict, file_name):
+  """
+  Reads KEYWORD=VALUE settings from the specified file.
+
+  Everything to the left of the first '=' is the keyword,
+  everything to the right is the value.  No stripping of
+  white space, so beware.
+
+  The file must exist, otherwise you get the Python exception from open().
+  """
+  for line in open(file_name, 'r').readlines():
+    key, val = line.rstrip('\r\n').split('=', 1)
+    values_dict[key] = val
+
+
+def ReadMessagesFromFile(file_name):
+  """
+  Reads messages from a 'chrome_messages_json' file.
+
+  The file must exist, otherwise you get the Python exception from open().
+  """
+  messages_file = io.open(file_name, encoding='utf-8-sig')
+  messages = json.load(messages_file)
+  messages_file.close()
+
+  values = {}
+  for key in messages.keys():
+    values[key] = unicode(messages[key]['message']);
+  return values
+
+
+def WriteIfChanged(file_name, contents, encoding='utf-16'):
+  """
+  Writes the specified contents to the specified file_name
+  iff the contents are different than the current contents.
+  """
+  try:
+    target = io.open(file_name, 'r')
+    old_contents = target.read()
+  except EnvironmentError:
+    pass
+  except UnicodeDecodeError:
+    target.close()
+    os.unlink(file_name)
+  else:
+    if contents == old_contents:
+      return
+    target.close()
+    os.unlink(file_name)
+  io.open(file_name, 'w', encoding=encoding).write(contents)
+
+
+class MessageMap:
+  """ Provides a dictionary of localized messages for each language."""
+  def __init__(self, languages, messages_path):
+    self.language = None
+    self.message_map = {}
+
+    # Populate the message map
+    if messages_path:
+      for language in languages:
+        file_name = os.path.join(messages_path,
+                                 language.replace('-', '_', 1),
+                                 'messages.json')
+        self.message_map[language] = ReadMessagesFromFile(file_name)
+
+  def GetText(self, message):
+    """ Returns a localized message for the current language. """
+    return self.message_map[self.language][message]
+
+  def SelectLanguage(self, language):
+    """ Selects the language to be used when retrieving localized messages. """
+    self.language = language
+
+  def MakeSelectLanguage(self):
+    """ Returns a function that can be used to select the current language. """
+    return lambda language: self.SelectLanguage(language)
+
+  def MakeGetText(self):
+    """ Returns a function that can be used to retrieve a localized message. """
+    return lambda message: self.GetText(message)
+
+
+def Localize(source, target, options):
+  # Load jinja2 library.
+  if options.jinja2:
+    jinja2_path = os.path.normpath(options.jinja2)
+  else:
+    jinja2_path = os.path.normpath(os.path.join(os.path.abspath(__file__),
+                                                '../../../third_party/jinja2'))
+  sys.path.append(os.path.split(jinja2_path)[0])
+  from jinja2 import Environment, FileSystemLoader
+
+  # Create jinja2 environment.
+  (template_path, template_name) = os.path.split(source)
+  env = Environment(loader=FileSystemLoader(template_path),
+                    extensions=['jinja2.ext.do', 'jinja2.ext.i18n'])
+
+  # Register custom filters.
+  env.filters['GetCodepage'] = GetCodepage
+  env.filters['GetCodepageDecimal'] = GetCodepageDecimal
+  env.filters['GetLangId'] = GetLangId
+  env.filters['GetPrimaryLanguage'] = GetPrimaryLanguage
+  env.filters['GetSublanguage'] = GetSublanguage
+
+  # Set the list of languages to use
+  languages = map(NormalizeLanguageCode, options.languages)
+  context = { 'languages' : languages }
+  env.globals['IsRtlLanguage'] = IsRtlLanguage
+
+  # Load the localized messages and register the message map with jinja2.i18n
+  # extension.
+  message_map = MessageMap(languages, options.messages_path)
+  env.globals['SelectLanguage'] = message_map.MakeSelectLanguage()
+  env.install_gettext_callables(message_map.MakeGetText(),
+                                message_map.MakeGetText());
+
+  # Add OFFICIAL_BUILD variable the same way chrome/tools/build/version.py
+  # does.
+  if os.environ.get('CHROME_BUILD_TYPE') == '_official':
+    context['official_build'] = '1'
+  else:
+    context['official_build'] = '0'
+
+  # Add all variables defined in the command line.
+  if options.define:
+    for define in options.define:
+      context.update(dict([define.split('=', 1)]));
+
+  # Read KEYWORD=VALUE variables from file.
+  if options.input:
+    for file_name in options.input:
+      ReadValuesFromFile(context, file_name)
+
+  template = env.get_template(template_name)
+  WriteIfChanged(target, template.render(context), options.encoding);
+  return 0;
+
+
+def main():
+  usage = "Usage: localize [options] <input> <output>"
+  parser = OptionParser(usage=usage)
+  parser.add_option(
+      '-d', '--define', dest='define', action='append', type='string',
+      help='define a variable (VAR=VALUE).')
+  parser.add_option(
+      '-i', '--input', dest='input', action='append', type='string',
+      help='read variables from INPUT.')
+  parser.add_option(
+      '-l', '--language', dest='languages', action='append', type='string',
+      help='add LANGUAGE to the list of languages to use.')
+  parser.add_option(
+      '--encoding', dest='encoding', type='string', default='utf-16',
+      help="set the encoding of <output>. 'utf-16' is the default.")
+  parser.add_option(
+      '--jinja2', dest='jinja2', type='string',
+      help="specifies path to the jinja2 library.")
+  parser.add_option(
+      '--messages_path', dest='messages_path', type='string',
+      help="set path to localized messages.")
+
+  options, args = parser.parse_args()
+  if len(args) != 2:
+    parser.error('Two positional arguments (<input> and <output>) are expected')
+  if not options.languages:
+    parser.error('At least one language must be specified')
+  if not options.messages_path:
+    parser.error('--messages_path is required')
+
+  return Localize(args[0], args[1], options)
+
+if __name__ == '__main__':
+  sys.exit(main())
+