Mercurial > babel > old > mirror
changeset 381:6a0e7205790f stable
Ported [407:415/trunk] back to 0.9.x branch.
| author | cmlenz |
|---|---|
| date | Tue, 08 Jul 2008 21:01:28 +0000 |
| parents | 4b63ce84c6ae |
| children | 6ab5550b7945 |
| files | 0.9.x/ChangeLog 0.9.x/babel/core.py 0.9.x/babel/dates.py 0.9.x/babel/localedata.py 0.9.x/babel/messages/catalog.py 0.9.x/babel/messages/frontend.py 0.9.x/babel/messages/plurals.py 0.9.x/babel/messages/pofile.py 0.9.x/babel/messages/tests/__init__.py 0.9.x/babel/messages/tests/checkers.py 0.9.x/babel/messages/tests/frontend.py 0.9.x/babel/messages/tests/plurals.py 0.9.x/babel/messages/tests/pofile.py 0.9.x/babel/numbers.py 0.9.x/babel/tests/dates.py 0.9.x/babel/tests/localedata.py 0.9.x/scripts/dump_data.py 0.9.x/scripts/import_cldr.py |
| diffstat | 18 files changed, 1073 insertions(+), 141 deletions(-) [+] |
line wrap: on
line diff
--- a/0.9.x/ChangeLog +++ b/0.9.x/ChangeLog @@ -11,9 +11,14 @@ forms where the translations were empty (ticket #97). * The stripping of the comment tags in comments is optional now and is done for each line in a comment. - * A JavaScript message extractor was added. - * Updated to CLDR 1.5.1. + * Added a JavaScript message extractor. + * Updated to CLDR 1.6. * Fixed timezone calculations when formatting datetime and time values. + * Added a `get_plural` function into the plurals module that returns the + correct plural forms for a locale as tuple. + * Added support for alias definitions in the CLDR data files, meaning that + the chance for items missing in certain locales should be greatly reduced + (ticket #68). Version 0.9.2
--- a/0.9.x/babel/core.py +++ b/0.9.x/babel/core.py @@ -223,7 +223,7 @@ def _data(self): if self.__data is None: - self.__data = localedata.load(str(self)) + self.__data = localedata.LocaleDataDict(localedata.load(str(self))) return self.__data _data = property(_data) @@ -326,7 +326,7 @@ Mapping of script codes to translated script names. >>> Locale('de', 'DE').variants['1901'] - u'alte deutsche Rechtschreibung' + u'Alte deutsche Rechtschreibung' :type: `dict` """) @@ -481,7 +481,7 @@ >>> Locale('en', 'US').time_zones['Europe/London']['long']['daylight'] u'British Summer Time' >>> Locale('en', 'US').time_zones['America/St_Johns']['city'] - u'St. John\u2019s' + u"St. John's" :type: `dict` """)
--- a/0.9.x/babel/dates.py +++ b/0.9.x/babel/dates.py @@ -185,11 +185,11 @@ >>> get_timezone_gmt(dt, 'short', locale='en') u'-0800' - The long format depends on the locale, for example in France a different - string is used for GMT: + The long format depends on the locale, for example in France the acronym + UTC string is used instead of GMT: >>> get_timezone_gmt(dt, 'long', locale='fr_FR') - u'HMG-08:00' + u'UTC-08:00' :param datetime: the ``datetime`` object; if `None`, the current date and time in UTC is used @@ -488,7 +488,7 @@ if datetime is None: datetime = datetime_.utcnow() elif isinstance(datetime, (int, long)): - datetime = datetime.utcfromtimestamp(datetime) + datetime = datetime_.utcfromtimestamp(datetime) elif isinstance(datetime, time): datetime = datetime_.combine(date.today(), datetime) if datetime.tzinfo is None:
--- a/0.9.x/babel/localedata.py +++ b/0.9.x/babel/localedata.py @@ -23,14 +23,16 @@ import threading except ImportError: import dummy_threading as threading +from UserDict import DictMixin -__all__ = ['exists', 'load'] +__all__ = ['exists', 'list', 'load'] __docformat__ = 'restructuredtext en' _cache = {} _cache_lock = threading.RLock() _dirname = os.path.join(os.path.dirname(__file__), 'localedata') + def exists(name): """Check whether locale data is available for the given locale. @@ -42,6 +44,7 @@ return True return os.path.exists(os.path.join(_dirname, '%s.dat' % name)) + def list(): """Return a list of all locale identifiers for which locale data is available. @@ -54,7 +57,8 @@ os.path.splitext(filename) for filename in os.listdir(_dirname) ] if extension == '.dat' and stem != 'root'] -def load(name): + +def load(name, merge_inherited=True): """Load the locale data for the given locale. The locale data is a dictionary that contains much of the data defined by @@ -74,6 +78,8 @@ True :param name: the locale identifier string (or "root") + :param merge_inherited: whether the inherited data should be merged into + the data of the requested locale :return: the locale data :rtype: `dict` :raise `IOError`: if no locale data file is found for the given locale @@ -84,7 +90,7 @@ data = _cache.get(name) if not data: # Load inherited data - if name == 'root': + if name == 'root' or not merge_inherited: data = {} else: parts = name.split('_') @@ -96,7 +102,7 @@ filename = os.path.join(_dirname, '%s.dat' % name) fileobj = open(filename, 'rb') try: - if name != 'root': + if name != 'root' and merge_inherited: merge(data, pickle.load(fileobj)) else: data = pickle.load(fileobj) @@ -107,17 +113,92 @@ finally: _cache_lock.release() + def merge(dict1, dict2): """Merge the data from `dict2` into the `dict1` dictionary, making copies of nested dictionaries. + >>> d = {1: 'foo', 3: 'baz'} + >>> merge(d, {1: 'Foo', 2: 'Bar'}) + >>> d + {1: 'Foo', 2: 'Bar', 3: 'baz'} + :param dict1: the dictionary to merge into :param dict2: the dictionary containing the data that should be merged """ - for key, value in dict2.items(): - if value is not None: - if type(value) is dict: - dict1[key] = dict1.get(key, {}).copy() - merge(dict1[key], value) + for key, val2 in dict2.items(): + if val2 is not None: + val1 = dict1.get(key) + if isinstance(val2, dict): + if val1 is None: + val1 = {} + if isinstance(val1, Alias): + val1 = (val1, val2) + elif isinstance(val1, tuple): + alias, others = val1 + others = others.copy() + merge(others, val2) + val1 = (alias, others) + else: + val1 = val1.copy() + merge(val1, val2) else: - dict1[key] = value + val1 = val2 + dict1[key] = val1 + + +class Alias(object): + """Representation of an alias in the locale data. + + An alias is a value that refers to some other part of the locale data, + as specified by the `keys`. + """ + + def __init__(self, keys): + self.keys = tuple(keys) + + def __repr__(self): + return '<%s %r>' % (type(self).__name__, self.keys) + + def resolve(self, data): + """Resolve the alias based on the given data. + + This is done recursively, so if one alias resolves to a second alias, + that second alias will also be resolved. + + :param data: the locale data + :type data: `dict` + """ + base = data + for key in self.keys: + data = data[key] + if isinstance(data, Alias): + data = data.resolve(base) + return data + + +class LocaleDataDict(DictMixin, dict): + """Dictionary wrapper that automatically resolves aliases to the actual + values. + """ + + def __init__(self, data, base=None): + dict.__init__(self, data) + if base is None: + base = self + self.base = base + + def __getitem__(self, key): + val = dict.__getitem__(self, key) + if isinstance(val, Alias): # resolve an alias + val = val.resolve(self.base) + if isinstance(val, tuple): # Merge a partial dict with an alias + alias, others = val + val = alias.resolve(self.base).copy() + merge(val, others) + if isinstance(val, dict): # Return a nested alias-resolving dict + val = LocaleDataDict(val, base=self.base) + return val + + def copy(self): + return LocaleDataDict(dict.copy(self), base=self.base)
--- a/0.9.x/babel/messages/catalog.py +++ b/0.9.x/babel/messages/catalog.py @@ -28,7 +28,7 @@ from babel import __version__ as VERSION from babel.core import Locale from babel.dates import format_datetime -from babel.messages.plurals import PLURALS +from babel.messages.plurals import get_plural from babel.util import odict, distinct, LOCALTZ, UTC, FixedOffsetTimezone __all__ = ['Message', 'Catalog', 'TranslationError'] @@ -409,13 +409,10 @@ """) def num_plurals(self): - if not self._num_plurals: + if self._num_plurals is None: num = 2 if self.locale: - if str(self.locale) in PLURALS: - num = PLURALS[str(self.locale)][0] - elif self.locale.language in PLURALS: - num = PLURALS[self.locale.language][0] + num = get_plural(self.locale)[0] self._num_plurals = num return self._num_plurals num_plurals = property(num_plurals, doc="""\ @@ -430,13 +427,10 @@ """) def plural_expr(self): - if not self._plural_expr: + if self._plural_expr is None: expr = '(n != 1)' if self.locale: - if str(self.locale) in PLURALS: - expr = PLURALS[str(self.locale)][1] - elif self.locale.language in PLURALS: - expr = PLURALS[self.locale.language][1] + expr = get_plural(self.locale)[1] self._plural_expr = expr return self._plural_expr plural_expr = property(plural_expr, doc="""\
--- a/0.9.x/babel/messages/frontend.py +++ b/0.9.x/babel/messages/frontend.py @@ -440,7 +440,9 @@ infile = open(self.input_file, 'r') try: - catalog = read_po(infile) + # Although reading from the catalog template, read_po must be fed + # the locale in order to correcly calculate plurals + catalog = read_po(infile, locale=self.locale) finally: infile.close() @@ -956,7 +958,9 @@ infile = open(options.input_file, 'r') try: - catalog = read_po(infile) + # Although reading from the catalog template, read_po must be fed + # the locale in order to correcly calculate plurals + catalog = read_po(infile, locale=options.locale) finally: infile.close()
--- a/0.9.x/babel/messages/plurals.py +++ b/0.9.x/babel/messages/plurals.py @@ -13,6 +13,14 @@ """Plural form definitions.""" + +from operator import itemgetter +from babel.core import default_locale, Locale + + +LC_CTYPE = default_locale('LC_CTYPE') + + PLURALS = { # Afar # 'aa': (), @@ -191,3 +199,58 @@ 'zh_HK': (1, '0'), 'zh_TW': (1, '0'), } + + +DEFAULT_PLURAL = (2, '(n != 1)') + + +class _PluralTuple(tuple): + """A tuple with plural information.""" + + __slots__ = () + num_plurals = property(itemgetter(0), doc=""" + The number of plurals used by the locale.""") + plural_expr = property(itemgetter(1), doc=""" + The plural expression used by the locale.""") + plural_forms = property(lambda x: 'npurals=%s; plural=%s' % x, doc=""" + The plural expression used by the catalog or locale.""") + + def __str__(self): + return self.plural_forms + + +def get_plural(locale=LC_CTYPE): + """A tuple with the information catalogs need to perform proper + pluralization. The first item of the tuple is the number of plural + forms, the second the plural expression. + + >>> get_plural(locale='en') + (2, '(n != 1)') + >>> get_plural(locale='ga') + (3, '(n==1 ? 0 : n==2 ? 1 : 2)') + + The object returned is a special tuple with additional members: + + >>> tup = get_plural("ja") + >>> tup.num_plurals + 1 + >>> tup.plural_expr + '0' + >>> tup.plural_forms + 'npurals=1; plural=0' + + Converting the tuple into a string prints the plural forms for a + gettext catalog: + + >>> str(tup) + 'npurals=1; plural=0' + """ + locale = Locale.parse(locale) + try: + tup = PLURALS[str(locale)] + except KeyError: + try: + tup = PLURALS[locale.language] + except KeyError: + tup = DEFAULT_PLURAL + return _PluralTuple(tup)
--- a/0.9.x/babel/messages/pofile.py +++ b/0.9.x/babel/messages/pofile.py @@ -145,8 +145,14 @@ msgid = tuple([denormalize(m) for m in messages]) else: msgid = denormalize(messages[0]) - if len(translations) > 1: - string = tuple([denormalize(t[1]) for t in translations]) + if isinstance(msgid, (list, tuple)): + string = [] + for idx in range(catalog.num_plurals): + try: + string.append(translations[idx]) + except IndexError: + string.append((idx, '')) + string = tuple([denormalize(t[1]) for t in string]) else: string = denormalize(translations[0][1]) message = Message(msgid, string, list(locations), set(flags), @@ -384,9 +390,14 @@ _write('%smsgid_plural %s\n' % ( prefix, _normalize(message.id[1], prefix) )) - for i, string in enumerate(message.string): + + for idx in range(catalog.num_plurals): + try: + string = message.string[idx] + except IndexError: + string = '' _write('%smsgstr[%d] %s\n' % ( - prefix, i, _normalize(message.string[i], prefix) + prefix, idx, _normalize(string, prefix) )) else: _write('%smsgid %s\n' % (prefix, _normalize(message.id, prefix)))
--- a/0.9.x/babel/messages/tests/__init__.py +++ b/0.9.x/babel/messages/tests/__init__.py @@ -14,13 +14,15 @@ import unittest def suite(): - from babel.messages.tests import catalog, extract, frontend, mofile, pofile + from babel.messages.tests import catalog, extract, frontend, mofile, \ + plurals, pofile, checkers suite = unittest.TestSuite() suite.addTest(catalog.suite()) suite.addTest(extract.suite()) suite.addTest(frontend.suite()) suite.addTest(mofile.suite()) suite.addTest(pofile.suite()) + suite.addTest(checkers.suite()) return suite if __name__ == '__main__':
new file mode 100644 --- /dev/null +++ b/0.9.x/babel/messages/tests/checkers.py @@ -0,0 +1,371 @@ +# -*- coding: utf-8 -*- +# +# Copyright (C) 2008 Edgewall Software +# All rights reserved. +# +# This software is licensed as described in the file COPYING, which +# you should have received as part of this distribution. The terms +# are also available at http://babel.edgewall.org/wiki/License. +# +# This software consists of voluntary contributions made by many +# individuals. For the exact contribution history, see the revision +# history and logs, available at http://babel.edgewall.org/log/. + +from datetime import datetime +import time +import unittest +from StringIO import StringIO + +from babel import __version__ as VERSION +from babel.core import Locale, UnknownLocaleError +from babel.dates import format_datetime +from babel.messages import checkers +from babel.messages.plurals import PLURALS +from babel.messages.pofile import read_po +from babel.util import LOCALTZ + + +class CheckersTestCase(unittest.TestCase): + # the last msgstr[idx] is always missing except for singular plural forms + + def test_1_num_plurals_checkers(self): + for _locale in [p for p in PLURALS if PLURALS[p][0] == 1]: + try: + locale = Locale.parse(_locale) + except UnknownLocaleError: + # Just an alias? Not what we're testing here, let's continue + continue + po_file = (ur"""\ +# %(english_name)s translations for TestProject. +# Copyright (C) 2007 FooBar, Inc. +# This file is distributed under the same license as the TestProject +# project. +# FIRST AUTHOR <EMAIL@ADDRESS>, 2007. +# +msgid "" +msgstr "" +"Project-Id-Version: TestProject 0.1\n" +"Report-Msgid-Bugs-To: bugs.address@email.tld\n" +"POT-Creation-Date: 2007-04-01 15:30+0200\n" +"PO-Revision-Date: %(date)s\n" +"Last-Translator: FULL NAME <EMAIL@ADDRESS>\n" +"Language-Team: %(locale)s <LL@li.org>\n" +"Plural-Forms: nplurals=%(num_plurals)s; plural=%(plural_expr)s\n" +"MIME-Version: 1.0\n" +"Content-Type: text/plain; charset=utf-8\n" +"Content-Transfer-Encoding: 8bit\n" +"Generated-By: Babel %(version)s\n" + +#. This will be a translator comment, +#. that will include several lines +#: project/file1.py:8 +msgid "bar" +msgstr "" + +#: project/file2.py:9 +msgid "foobar" +msgid_plural "foobars" +msgstr[0] "" + +""" % dict(locale = _locale, + english_name = locale.english_name, + version = VERSION, + year = time.strftime('%Y'), + date = format_datetime(datetime.now(LOCALTZ), + 'yyyy-MM-dd HH:mmZ', + tzinfo=LOCALTZ, locale=_locale), + num_plurals = PLURALS[_locale][0], + plural_expr = PLURALS[_locale][0])).encode('utf-8') + + # This test will fail for revisions <= 406 because so far + # catalog.num_plurals was neglected + catalog = read_po(StringIO(po_file), _locale) + message = catalog['foobar'] + checkers.num_plurals(catalog, message) + + def test_2_num_plurals_checkers(self): + # in this testcase we add an extra msgstr[idx], we should be + # disregarding it + for _locale in [p for p in PLURALS if PLURALS[p][0] == 2]: + if _locale in ['nn', 'no']: + _locale = 'nn_NO' + num_plurals = PLURALS[_locale.split('_')[0]][0] + plural_expr = PLURALS[_locale.split('_')[0]][1] + else: + num_plurals = PLURALS[_locale][0] + plural_expr = PLURALS[_locale][1] + try: + locale = Locale(_locale) + date = format_datetime(datetime.now(LOCALTZ), + 'yyyy-MM-dd HH:mmZ', + tzinfo=LOCALTZ, locale=_locale), + except UnknownLocaleError: + # Just an alias? Not what we're testing here, let's continue + continue + po_file = (ur"""\ +# %(english_name)s translations for TestProject. +# Copyright (C) 2007 FooBar, Inc. +# This file is distributed under the same license as the TestProject +# project. +# FIRST AUTHOR <EMAIL@ADDRESS>, 2007. +# +msgid "" +msgstr "" +"Project-Id-Version: TestProject 0.1\n" +"Report-Msgid-Bugs-To: bugs.address@email.tld\n" +"POT-Creation-Date: 2007-04-01 15:30+0200\n" +"PO-Revision-Date: %(date)s\n" +"Last-Translator: FULL NAME <EMAIL@ADDRESS>\n" +"Language-Team: %(locale)s <LL@li.org>\n" +"Plural-Forms: nplurals=%(num_plurals)s; plural=%(plural_expr)s\n" +"MIME-Version: 1.0\n" +"Content-Type: text/plain; charset=utf-8\n" +"Content-Transfer-Encoding: 8bit\n" +"Generated-By: Babel %(version)s\n" + +#. This will be a translator comment, +#. that will include several lines +#: project/file1.py:8 +msgid "bar" +msgstr "" + +#: project/file2.py:9 +msgid "foobar" +msgid_plural "foobars" +msgstr[0] "" +msgstr[1] "" +msgstr[2] "" + +""" % dict(locale = _locale, + english_name = locale.english_name, + version = VERSION, + year = time.strftime('%Y'), + date = date, + num_plurals = num_plurals, + plural_expr = plural_expr)).encode('utf-8') + # we should be adding the missing msgstr[0] + + # This test will fail for revisions <= 406 because so far + # catalog.num_plurals was neglected + catalog = read_po(StringIO(po_file), _locale) + message = catalog['foobar'] + checkers.num_plurals(catalog, message) + + def test_3_num_plurals_checkers(self): + for _locale in [p for p in PLURALS if PLURALS[p][0] == 3]: + po_file = r"""\ +# %(english_name)s translations for TestProject. +# Copyright (C) 2007 FooBar, Inc. +# This file is distributed under the same license as the TestProject +# project. +# FIRST AUTHOR <EMAIL@ADDRESS>, 2007. +# +msgid "" +msgstr "" +"Project-Id-Version: TestProject 0.1\n" +"Report-Msgid-Bugs-To: bugs.address@email.tld\n" +"POT-Creation-Date: 2007-04-01 15:30+0200\n" +"PO-Revision-Date: %(date)s\n" +"Last-Translator: FULL NAME <EMAIL@ADDRESS>\n" +"Language-Team: %(locale)s <LL@li.org>\n" +"Plural-Forms: nplurals=%(num_plurals)s; plural=%(plural_expr)s\n" +"MIME-Version: 1.0\n" +"Content-Type: text/plain; charset=utf-8\n" +"Content-Transfer-Encoding: 8bit\n" +"Generated-By: Babel %(version)s\n" + +#. This will be a translator comment, +#. that will include several lines +#: project/file1.py:8 +msgid "bar" +msgstr "" + +#: project/file2.py:9 +msgid "foobar" +msgid_plural "foobars" +msgstr[0] "" +msgstr[1] "" + +""" % dict(locale = _locale, + english_name = Locale.parse(_locale).english_name, + version = VERSION, + year = time.strftime('%Y'), + date = format_datetime(datetime.now(LOCALTZ), + 'yyyy-MM-dd HH:mmZ', + tzinfo=LOCALTZ, locale=_locale), + num_plurals = PLURALS[_locale][0], + plural_expr = PLURALS[_locale][0]) + + # This test will fail for revisions <= 406 because so far + # catalog.num_plurals was neglected + catalog = read_po(StringIO(po_file), _locale) + message = catalog['foobar'] + checkers.num_plurals(catalog, message) + + def test_4_num_plurals_checkers(self): + for _locale in [p for p in PLURALS if PLURALS[p][0] == 4]: + po_file = r"""\ +# %(english_name)s translations for TestProject. +# Copyright (C) 2007 FooBar, Inc. +# This file is distributed under the same license as the TestProject +# project. +# FIRST AUTHOR <EMAIL@ADDRESS>, 2007. +# +msgid "" +msgstr "" +"Project-Id-Version: TestProject 0.1\n" +"Report-Msgid-Bugs-To: bugs.address@email.tld\n" +"POT-Creation-Date: 2007-04-01 15:30+0200\n" +"PO-Revision-Date: %(date)s\n" +"Last-Translator: FULL NAME <EMAIL@ADDRESS>\n" +"Language-Team: %(locale)s <LL@li.org>\n" +"Plural-Forms: nplurals=%(num_plurals)s; plural=%(plural_expr)s\n" +"MIME-Version: 1.0\n" +"Content-Type: text/plain; charset=utf-8\n" +"Content-Transfer-Encoding: 8bit\n" +"Generated-By: Babel %(version)s\n" + +#. This will be a translator comment, +#. that will include several lines +#: project/file1.py:8 +msgid "bar" +msgstr "" + +#: project/file2.py:9 +msgid "foobar" +msgid_plural "foobars" +msgstr[0] "" +msgstr[1] "" +msgstr[2] "" + +""" % dict(locale = _locale, + english_name = Locale.parse(_locale).english_name, + version = VERSION, + year = time.strftime('%Y'), + date = format_datetime(datetime.now(LOCALTZ), + 'yyyy-MM-dd HH:mmZ', + tzinfo=LOCALTZ, locale=_locale), + num_plurals = PLURALS[_locale][0], + plural_expr = PLURALS[_locale][0]) + + # This test will fail for revisions <= 406 because so far + # catalog.num_plurals was neglected + catalog = read_po(StringIO(po_file), _locale) + message = catalog['foobar'] + checkers.num_plurals(catalog, message) + + def test_5_num_plurals_checkers(self): + for _locale in [p for p in PLURALS if PLURALS[p][0] == 5]: + po_file = r"""\ +# %(english_name)s translations for TestProject. +# Copyright (C) 2007 FooBar, Inc. +# This file is distributed under the same license as the TestProject +# project. +# FIRST AUTHOR <EMAIL@ADDRESS>, 2007. +# +msgid "" +msgstr "" +"Project-Id-Version: TestProject 0.1\n" +"Report-Msgid-Bugs-To: bugs.address@email.tld\n" +"POT-Creation-Date: 2007-04-01 15:30+0200\n" +"PO-Revision-Date: %(date)s\n" +"Last-Translator: FULL NAME <EMAIL@ADDRESS>\n" +"Language-Team: %(locale)s <LL@li.org>\n" +"Plural-Forms: nplurals=%(num_plurals)s; plural=%(plural_expr)s\n" +"MIME-Version: 1.0\n" +"Content-Type: text/plain; charset=utf-8\n" +"Content-Transfer-Encoding: 8bit\n" +"Generated-By: Babel %(version)s\n" + +#. This will be a translator comment, +#. that will include several lines +#: project/file1.py:8 +msgid "bar" +msgstr "" + +#: project/file2.py:9 +msgid "foobar" +msgid_plural "foobars" +msgstr[0] "" +msgstr[1] "" +msgstr[2] "" +msgstr[3] "" + +""" % dict(locale = _locale, + english_name = Locale.parse(_locale).english_name, + version = VERSION, + year = time.strftime('%Y'), + date = format_datetime(datetime.now(LOCALTZ), + 'yyyy-MM-dd HH:mmZ', + tzinfo=LOCALTZ, locale=_locale), + num_plurals = PLURALS[_locale][0], + plural_expr = PLURALS[_locale][0]) + + # This test will fail for revisions <= 406 because so far + # catalog.num_plurals was neglected + catalog = read_po(StringIO(po_file), _locale) + message = catalog['foobar'] + checkers.num_plurals(catalog, message) + + def test_6_num_plurals_checkers(self): + for _locale in [p for p in PLURALS if PLURALS[p][0] == 6]: + po_file = r"""\ +# %(english_name)s translations for TestProject. +# Copyright (C) 2007 FooBar, Inc. +# This file is distributed under the same license as the TestProject +# project. +# FIRST AUTHOR <EMAIL@ADDRESS>, 2007. +# +msgid "" +msgstr "" +"Project-Id-Version: TestProject 0.1\n" +"Report-Msgid-Bugs-To: bugs.address@email.tld\n" +"POT-Creation-Date: 2007-04-01 15:30+0200\n" +"PO-Revision-Date: %(date)s\n" +"Last-Translator: FULL NAME <EMAIL@ADDRESS>\n" +"Language-Team: %(locale)s <LL@li.org>\n" +"Plural-Forms: nplurals=%(num_plurals)s; plural=%(plural_expr)s\n" +"MIME-Version: 1.0\n" +"Content-Type: text/plain; charset=utf-8\n" +"Content-Transfer-Encoding: 8bit\n" +"Generated-By: Babel %(version)s\n" + +#. This will be a translator comment, +#. that will include several lines +#: project/file1.py:8 +msgid "bar" +msgstr "" + +#: project/file2.py:9 +msgid "foobar" +msgid_plural "foobars" +msgstr[0] "" +msgstr[1] "" +msgstr[2] "" +msgstr[3] "" +msgstr[4] "" + +""" % dict(locale = _locale, + english_name = Locale.parse(_locale).english_name, + version = VERSION, + year = time.strftime('%Y'), + date = format_datetime(datetime.now(LOCALTZ), + 'yyyy-MM-dd HH:mmZ', + tzinfo=LOCALTZ, locale=_locale), + num_plurals = PLURALS[_locale][0], + plural_expr = PLURALS[_locale][0]) + + # This test will fail for revisions <= 406 because so far + # catalog.num_plurals was neglected + catalog = read_po(StringIO(po_file), _locale) + message = catalog['foobar'] + checkers.num_plurals(catalog, message) + + +def suite(): + suite = unittest.TestSuite() + suite.addTest(unittest.makeSuite(CheckersTestCase)) + return suite + +if __name__ == '__main__': + unittest.main(defaultTest='suite')
--- a/0.9.x/babel/messages/tests/frontend.py +++ b/0.9.x/babel/messages/tests/frontend.py @@ -13,7 +13,7 @@ from datetime import datetime from distutils.dist import Distribution -from distutils.errors import DistutilsOptionError, DistutilsSetupError +from distutils.errors import DistutilsOptionError from distutils.log import _global_log import doctest import os @@ -280,9 +280,10 @@ self.cmd.initialize_options() def tearDown(self): - locale_dir = os.path.join(self.datadir, 'project', 'i18n', 'en_US') - if os.path.isdir(locale_dir): - shutil.rmtree(locale_dir) + for dirname in ['en_US', 'ja_JP', 'lv_LV']: + locale_dir = os.path.join(self.datadir, 'project', 'i18n', dirname) + if os.path.isdir(locale_dir): + shutil.rmtree(locale_dir) os.chdir(self.olddir) @@ -346,32 +347,7 @@ tzinfo=LOCALTZ, locale='en')}, open(po_file, 'U').read()) - -class InitCatalogNonFuzzyTestCase(unittest.TestCase): - # FIXME: what is this test case about? - - def setUp(self): - self.olddir = os.getcwd() - self.datadir = os.path.join(os.path.dirname(__file__), 'data') - os.chdir(self.datadir) - _global_log.threshold = 5 # shut up distutils logging - - self.dist = Distribution(dict( - name='TestProject', - version='0.1', - packages=['project'] - )) - self.cmd = frontend.init_catalog(self.dist) - self.cmd.initialize_options() - - def tearDown(self): - locale_dir = os.path.join(self.datadir, 'project', 'i18n', 'en_US') - if os.path.isdir(locale_dir): - shutil.rmtree(locale_dir) - - os.chdir(self.olddir) - - def test_with_output_dir(self): + def test_keeps_catalog_non_fuzzy(self): self.cmd.input_file = 'project/i18n/messages_non_fuzzy.pot' self.cmd.locale = 'en_US' self.cmd.output_dir = 'project/i18n' @@ -421,6 +397,107 @@ tzinfo=LOCALTZ, locale='en')}, open(po_file, 'U').read()) + def test_correct_init_more_than_2_plurals(self): + self.cmd.input_file = 'project/i18n/messages.pot' + self.cmd.locale = 'lv_LV' + self.cmd.output_dir = 'project/i18n' + + self.cmd.finalize_options() + self.cmd.run() + + po_file = os.path.join(self.datadir, 'project', 'i18n', 'lv_LV', + 'LC_MESSAGES', 'messages.po') + assert os.path.isfile(po_file) + + self.assertEqual( +r"""# Latvian (Latvia) translations for TestProject. +# Copyright (C) 2007 FooBar, Inc. +# This file is distributed under the same license as the TestProject +# project. +# FIRST AUTHOR <EMAIL@ADDRESS>, 2007. +# +msgid "" +msgstr "" +"Project-Id-Version: TestProject 0.1\n" +"Report-Msgid-Bugs-To: bugs.address@email.tld\n" +"POT-Creation-Date: 2007-04-01 15:30+0200\n" +"PO-Revision-Date: %(date)s\n" +"Last-Translator: FULL NAME <EMAIL@ADDRESS>\n" +"Language-Team: lv_LV <LL@li.org>\n" +"Plural-Forms: nplurals=3; plural=(n%%10==1 && n%%100!=11 ? 0 : n != 0 ? 1 :" +" 2)\n" +"MIME-Version: 1.0\n" +"Content-Type: text/plain; charset=utf-8\n" +"Content-Transfer-Encoding: 8bit\n" +"Generated-By: Babel %(version)s\n" + +#. This will be a translator coment, +#. that will include several lines +#: project/file1.py:8 +msgid "bar" +msgstr "" + +#: project/file2.py:9 +msgid "foobar" +msgid_plural "foobars" +msgstr[0] "" +msgstr[1] "" +msgstr[2] "" + +""" % {'version': VERSION, + 'date': format_datetime(datetime.now(LOCALTZ), 'yyyy-MM-dd HH:mmZ', + tzinfo=LOCALTZ, locale='en')}, + open(po_file, 'U').read()) + + def test_correct_init_singular_plural_forms(self): + self.cmd.input_file = 'project/i18n/messages.pot' + self.cmd.locale = 'ja_JP' + self.cmd.output_dir = 'project/i18n' + + self.cmd.finalize_options() + self.cmd.run() + + po_file = os.path.join(self.datadir, 'project', 'i18n', 'ja_JP', + 'LC_MESSAGES', 'messages.po') + assert os.path.isfile(po_file) + + self.assertEqual( +r"""# Japanese (Japan) translations for TestProject. +# Copyright (C) 2007 FooBar, Inc. +# This file is distributed under the same license as the TestProject +# project. +# FIRST AUTHOR <EMAIL@ADDRESS>, 2007. +# +msgid "" +msgstr "" +"Project-Id-Version: TestProject 0.1\n" +"Report-Msgid-Bugs-To: bugs.address@email.tld\n" +"POT-Creation-Date: 2007-04-01 15:30+0200\n" +"PO-Revision-Date: %(date)s\n" +"Last-Translator: FULL NAME <EMAIL@ADDRESS>\n" +"Language-Team: ja_JP <LL@li.org>\n" +"Plural-Forms: nplurals=1; plural=0\n" +"MIME-Version: 1.0\n" +"Content-Type: text/plain; charset=utf-8\n" +"Content-Transfer-Encoding: 8bit\n" +"Generated-By: Babel %(version)s\n" + +#. This will be a translator coment, +#. that will include several lines +#: project/file1.py:8 +msgid "bar" +msgstr "" + +#: project/file2.py:9 +msgid "foobar" +msgid_plural "foobars" +msgstr[0] "" + +""" % {'version': VERSION, + 'date': format_datetime(datetime.now(LOCALTZ), 'yyyy-MM-dd HH:mmZ', + tzinfo=LOCALTZ, locale='ja_JP')}, + open(po_file, 'U').read()) + class CommandLineInterfaceTestCase(unittest.TestCase): @@ -438,6 +515,10 @@ sys.argv = self.orig_argv sys.stdout = self.orig_stdout sys.stderr = self.orig_stderr + for dirname in ['lv_LV', 'ja_JP']: + locale_dir = os.path.join(self.datadir, 'project', 'i18n', dirname) + if os.path.isdir(locale_dir): + shutil.rmtree(locale_dir) def test_usage(self): try: @@ -631,6 +712,109 @@ 'date': format_datetime(datetime.now(LOCALTZ), 'yyyy-MM-dd HH:mmZ', tzinfo=LOCALTZ, locale='en')}, open(po_file, 'U').read()) + + def test_init_singular_plural_forms(self): + po_file = os.path.join(self.datadir, 'project', 'i18n', 'ja_JP', + 'LC_MESSAGES', 'messages.po') + try: + self.cli.run(sys.argv + ['init', + '--locale', 'ja_JP', + '-d', os.path.join(self.datadir, 'project', 'i18n'), + '-i', os.path.join(self.datadir, 'project', 'i18n', + 'messages.pot')]) + except SystemExit, e: + self.assertEqual(0, e.code) + assert os.path.isfile(po_file) + self.assertEqual( +r"""# Japanese (Japan) translations for TestProject. +# Copyright (C) 2007 FooBar, Inc. +# This file is distributed under the same license as the TestProject +# project. +# FIRST AUTHOR <EMAIL@ADDRESS>, 2007. +# +#, fuzzy +msgid "" +msgstr "" +"Project-Id-Version: TestProject 0.1\n" +"Report-Msgid-Bugs-To: bugs.address@email.tld\n" +"POT-Creation-Date: 2007-04-01 15:30+0200\n" +"PO-Revision-Date: %(date)s\n" +"Last-Translator: FULL NAME <EMAIL@ADDRESS>\n" +"Language-Team: ja_JP <LL@li.org>\n" +"Plural-Forms: nplurals=1; plural=0\n" +"MIME-Version: 1.0\n" +"Content-Type: text/plain; charset=utf-8\n" +"Content-Transfer-Encoding: 8bit\n" +"Generated-By: Babel %(version)s\n" + +#. This will be a translator coment, +#. that will include several lines +#: project/file1.py:8 +msgid "bar" +msgstr "" + +#: project/file2.py:9 +msgid "foobar" +msgid_plural "foobars" +msgstr[0] "" + +""" % {'version': VERSION, + 'date': format_datetime(datetime.now(LOCALTZ), 'yyyy-MM-dd HH:mmZ', + tzinfo=LOCALTZ, locale='en')}, + open(po_file, 'U').read()) + + def test_init_more_than_2_plural_forms(self): + po_file = os.path.join(self.datadir, 'project', 'i18n', 'lv_LV', + 'LC_MESSAGES', 'messages.po') + try: + self.cli.run(sys.argv + ['init', + '--locale', 'lv_LV', + '-d', os.path.join(self.datadir, 'project', 'i18n'), + '-i', os.path.join(self.datadir, 'project', 'i18n', + 'messages.pot')]) + except SystemExit, e: + self.assertEqual(0, e.code) + assert os.path.isfile(po_file) + self.assertEqual( +r"""# Latvian (Latvia) translations for TestProject. +# Copyright (C) 2007 FooBar, Inc. +# This file is distributed under the same license as the TestProject +# project. +# FIRST AUTHOR <EMAIL@ADDRESS>, 2007. +# +#, fuzzy +msgid "" +msgstr "" +"Project-Id-Version: TestProject 0.1\n" +"Report-Msgid-Bugs-To: bugs.address@email.tld\n" +"POT-Creation-Date: 2007-04-01 15:30+0200\n" +"PO-Revision-Date: %(date)s\n" +"Last-Translator: FULL NAME <EMAIL@ADDRESS>\n" +"Language-Team: lv_LV <LL@li.org>\n" +"Plural-Forms: nplurals=3; plural=(n%%10==1 && n%%100!=11 ? 0 : n != 0 ? 1 :" +" 2)\n" +"MIME-Version: 1.0\n" +"Content-Type: text/plain; charset=utf-8\n" +"Content-Transfer-Encoding: 8bit\n" +"Generated-By: Babel %(version)s\n" + +#. This will be a translator coment, +#. that will include several lines +#: project/file1.py:8 +msgid "bar" +msgstr "" + +#: project/file2.py:9 +msgid "foobar" +msgid_plural "foobars" +msgstr[0] "" +msgstr[1] "" +msgstr[2] "" + +""" % {'version': VERSION, + 'date': format_datetime(datetime.now(LOCALTZ), 'yyyy-MM-dd HH:mmZ', + tzinfo=LOCALTZ, locale='en')}, + open(po_file, 'U').read()) def test_compile_catalog(self): po_file = os.path.join(self.datadir, 'project', 'i18n', 'de_DE', @@ -683,7 +867,6 @@ suite.addTest(unittest.makeSuite(CompileCatalogTestCase)) suite.addTest(unittest.makeSuite(ExtractMessagesTestCase)) suite.addTest(unittest.makeSuite(InitCatalogTestCase)) - suite.addTest(unittest.makeSuite(InitCatalogNonFuzzyTestCase)) suite.addTest(unittest.makeSuite(CommandLineInterfaceTestCase)) return suite
new file mode 100644 --- /dev/null +++ b/0.9.x/babel/messages/tests/plurals.py @@ -0,0 +1,25 @@ +# -*- coding: utf-8 -*- +# +# Copyright (C) 2008 Edgewall Software +# All rights reserved. +# +# This software is licensed as described in the file COPYING, which +# you should have received as part of this distribution. The terms +# are also available at http://babel.edgewall.org/wiki/License. +# +# This software consists of voluntary contributions made by many +# individuals. For the exact contribution history, see the revision +# history and logs, available at http://babel.edgewall.org/log/. + +import doctest +import unittest + +from babel.messages import plurals + +def suite(): + suite = unittest.TestSuite() + suite.addTest(doctest.DocTestSuite(plurals)) + return suite + +if __name__ == '__main__': + unittest.main(defaultTest='suite')
--- a/0.9.x/babel/messages/tests/pofile.py +++ b/0.9.x/babel/messages/tests/pofile.py @@ -143,6 +143,30 @@ self.assertEqual(1, len(catalog)) self.assertEqual(0, len(catalog.obsolete)) + def test_singlular_plural_form(self): + buf = StringIO(r'''msgid "foo" +msgid_plural "foo" +msgstr[0] "Voh" +msgstr[1] "Vohs"''') # This is a bad po, ja_JP only uses msgstr[0] + catalog = pofile.read_po(buf, locale='ja_JP') + self.assertEqual(1, len(catalog)) + self.assertEqual(1, catalog.num_plurals) + message = catalog['foo'] + self.assertEqual(1, len(message.string)) + + def test_more_than_two_plural_forms(self): + buf = StringIO(r'''msgid "foo" +msgid_plural "foo" +msgstr[0] "Voh" +msgstr[1] "Vohs"''') # last translation form is missing +#msgstr[2] "Vohss"''') + catalog = pofile.read_po(buf, locale='lv_LV') + self.assertEqual(1, len(catalog)) + self.assertEqual(3, catalog.num_plurals) + message = catalog['foo'] + self.assertEqual(3, len(message.string)) + self.assertEqual('', message.string[2]) + class WritePoTestCase(unittest.TestCase):
--- a/0.9.x/babel/numbers.py +++ b/0.9.x/babel/numbers.py @@ -165,9 +165,9 @@ >>> format_currency(1099.98, 'USD', locale='en_US') u'$1,099.98' >>> format_currency(1099.98, 'USD', locale='es_CO') - u'US$ 1.099,98' + u'US$\\xa01.099,98' >>> format_currency(1099.98, 'EUR', locale='de_DE') - u'1.099,98 \\u20ac' + u'1.099,98\\xa0\\u20ac' The pattern can also be specified explicitly:
--- a/0.9.x/babel/tests/dates.py +++ b/0.9.x/babel/tests/dates.py @@ -29,6 +29,11 @@ fmt = dates.DateTimeFormat(d, locale='cs_CZ') self.assertEqual('1.', fmt['LLL']) + def test_abbreviated_month_alias(self): + d = date(2006, 3, 8) + fmt = dates.DateTimeFormat(d, locale='de_DE') + self.assertEqual(u'Mär', fmt['LLL']) + def test_week_of_year_first(self): d = date(2006, 1, 8) fmt = dates.DateTimeFormat(d, locale='de_DE') @@ -187,7 +192,7 @@ tz = timezone('Europe/Paris') t = time(15, 30, tzinfo=tz) fmt = dates.DateTimeFormat(t, locale='fr_FR') - self.assertEqual(u'Heure de l’Europe centrale', fmt['vvvv']) + self.assertEqual(u'heure d’Europe centrale', fmt['vvvv']) def test_hour_formatting(self): l = 'en_US'
--- a/0.9.x/babel/tests/localedata.py +++ b/0.9.x/babel/tests/localedata.py @@ -16,9 +16,57 @@ from babel import localedata + +class MergeResolveTestCase(unittest.TestCase): + + def test_merge_items(self): + d = {1: 'foo', 3: 'baz'} + localedata.merge(d, {1: 'Foo', 2: 'Bar'}) + self.assertEqual({1: 'Foo', 2: 'Bar', 3: 'baz'}, d) + + def test_merge_nested_dict(self): + d1 = {'x': {'a': 1, 'b': 2, 'c': 3}} + d2 = {'x': {'a': 1, 'b': 12, 'd': 14}} + localedata.merge(d1, d2) + self.assertEqual({ + 'x': {'a': 1, 'b': 12, 'c': 3, 'd': 14} + }, d1) + + def test_merge_nested_dict_no_overlap(self): + d1 = {'x': {'a': 1, 'b': 2}} + d2 = {'y': {'a': 11, 'b': 12}} + localedata.merge(d1, d2) + self.assertEqual({ + 'x': {'a': 1, 'b': 2}, + 'y': {'a': 11, 'b': 12} + }, d1) + + def test_merge_with_alias_and_resolve(self): + alias = localedata.Alias('x') + d1 = { + 'x': {'a': 1, 'b': 2, 'c': 3}, + 'y': alias + } + d2 = { + 'x': {'a': 1, 'b': 12, 'd': 14}, + 'y': {'b': 22, 'e': 25} + } + localedata.merge(d1, d2) + self.assertEqual({ + 'x': {'a': 1, 'b': 12, 'c': 3, 'd': 14}, + 'y': (alias, {'b': 22, 'e': 25}) + }, d1) + d = localedata.LocaleDataDict(d1) + self.assertEqual({ + 'x': {'a': 1, 'b': 12, 'c': 3, 'd': 14}, + 'y': {'a': 1, 'b': 22, 'c': 3, 'd': 14, 'e': 25} + }, dict(d.items())) + + def suite(): suite = unittest.TestSuite() suite.addTest(doctest.DocTestSuite(localedata)) + suite.addTest(unittest.makeSuite(MergeResolveTestCase)) return suite if __name__ == '__main__':
--- a/0.9.x/scripts/dump_data.py +++ b/0.9.x/scripts/dump_data.py @@ -12,12 +12,32 @@ # individuals. For the exact contribution history, see the revision # history and logs, available at http://babel.edgewall.org/log/. +from optparse import OptionParser from pprint import pprint import sys -from babel.localedata import load +from babel.localedata import load, LocaleDataDict -if len(sys.argv) > 2: - pprint(load(sys.argv[1]).get(sys.argv[2])) -else: - pprint(load(sys.argv[1])) + +def main(): + parser = OptionParser(usage='%prog [options] locale [path]') + parser.add_option('--noinherit', action='store_false', dest='inherit', + help='do not merge inherited data into locale data') + parser.add_option('--resolve', action='store_true', dest='resolve', + help='resolve aliases in locale data') + parser.set_defaults(inherit=True, resolve=False) + options, args = parser.parse_args() + if len(args) not in (1, 2): + parser.error('incorrect number of arguments') + + data = load(args[0], merge_inherited=options.inherit) + if options.resolve: + data = LocaleDataDict(data) + if len(args) > 1: + for key in args[1].split('.'): + data = data[key] + pprint(dict(data.items())) + + +if __name__ == '__main__': + main()
--- a/0.9.x/scripts/import_cldr.py +++ b/0.9.x/scripts/import_cldr.py @@ -16,6 +16,7 @@ from optparse import OptionParser import os import pickle +import re import sys try: from xml.etree.ElementTree import parse @@ -26,6 +27,7 @@ sys.path.insert(0, os.path.join(os.path.dirname(sys.argv[0]), '..')) from babel import dates, numbers +from babel.localedata import Alias weekdays = {'mon': 0, 'tue': 1, 'wed': 2, 'thu': 3, 'fri': 4, 'sat': 5, 'sun': 6} @@ -36,6 +38,7 @@ def any(iterable): return filter(None, list(iterable)) + def _text(elem): buf = [elem.text or ''] for child in elem: @@ -43,6 +46,35 @@ buf.append(elem.tail or '') return u''.join(filter(None, buf)).strip() + +NAME_RE = re.compile(r"^\w+$") +TYPE_ATTR_RE = re.compile(r"^\w+\[@type='(.*?)'\]$") + +NAME_MAP = { + 'dateFormats': 'date_formats', + 'dateTimeFormats': 'datetime_formats', + 'eraAbbr': 'abbreviated', + 'eraNames': 'wide', + 'eraNarrow': 'narrow', + 'timeFormats': 'time_formats' +} + +def _translate_alias(ctxt, path): + parts = path.split('/') + keys = ctxt[:] + for part in parts: + if part == '..': + keys.pop() + else: + match = TYPE_ATTR_RE.match(part) + if match: + keys.append(match.group(1)) + else: + assert NAME_RE.match(part) + keys.append(NAME_MAP.get(part, part)) + return keys + + def main(): parser = OptionParser(usage='%prog path/to/cldr') options, args = parser.parse_args() @@ -109,6 +141,8 @@ stem, ext = os.path.splitext(filename) if ext != '.xml': continue + #if stem != 'root': + # break tree = parse(os.path.join(srcdir, 'main', filename)) data = {} @@ -133,25 +167,29 @@ territories = data.setdefault('territories', {}) for elem in tree.findall('//territories/territory'): - if 'draft' in elem.attrib and elem.attrib['type'] in territories: + if ('draft' in elem.attrib or 'alt' in elem.attrib) \ + and elem.attrib['type'] in territories: continue territories[elem.attrib['type']] = _text(elem) languages = data.setdefault('languages', {}) for elem in tree.findall('//languages/language'): - if 'draft' in elem.attrib and elem.attrib['type'] in languages: + if ('draft' in elem.attrib or 'alt' in elem.attrib) \ + and elem.attrib['type'] in languages: continue languages[elem.attrib['type']] = _text(elem) variants = data.setdefault('variants', {}) for elem in tree.findall('//variants/variant'): - if 'draft' in elem.attrib and elem.attrib['type'] in variants: + if ('draft' in elem.attrib or 'alt' in elem.attrib) \ + and elem.attrib['type'] in variants: continue variants[elem.attrib['type']] = _text(elem) scripts = data.setdefault('scripts', {}) for elem in tree.findall('//scripts/script'): - if 'draft' in elem.attrib and elem.attrib['type'] in scripts: + if ('draft' in elem.attrib or 'alt' in elem.attrib) \ + and elem.attrib['type'] in scripts: continue scripts[elem.attrib['type']] = _text(elem) @@ -182,15 +220,15 @@ zone_formats = data.setdefault('zone_formats', {}) for elem in tree.findall('//timeZoneNames/gmtFormat'): - if 'draft' not in elem.attrib: + if 'draft' not in elem.attrib and 'alt' not in elem.attrib: zone_formats['gmt'] = unicode(elem.text).replace('{0}', '%s') break for elem in tree.findall('//timeZoneNames/regionFormat'): - if 'draft' not in elem.attrib: + if 'draft' not in elem.attrib and 'alt' not in elem.attrib: zone_formats['region'] = unicode(elem.text).replace('{0}', '%s') break for elem in tree.findall('//timeZoneNames/fallbackFormat'): - if 'draft' not in elem.attrib: + if 'draft' not in elem.attrib and 'alt' not in elem.attrib: zone_formats['fallback'] = unicode(elem.text) \ .replace('{0}', '%(0)s').replace('{1}', '%(1)s') break @@ -227,88 +265,141 @@ months = data.setdefault('months', {}) for ctxt in calendar.findall('months/monthContext'): - ctxts = months.setdefault(ctxt.attrib['type'], {}) + ctxt_type = ctxt.attrib['type'] + ctxts = months.setdefault(ctxt_type, {}) for width in ctxt.findall('monthWidth'): - widths = ctxts.setdefault(width.attrib['type'], {}) - for elem in width.findall('month'): - if 'draft' in elem.attrib and int(elem.attrib['type']) in widths: - continue - widths[int(elem.attrib.get('type'))] = unicode(elem.text) + width_type = width.attrib['type'] + widths = ctxts.setdefault(width_type, {}) + for elem in width.getiterator(): + if elem.tag == 'month': + if ('draft' in elem.attrib or 'alt' in elem.attrib) \ + and int(elem.attrib['type']) in widths: + continue + widths[int(elem.attrib.get('type'))] = unicode(elem.text) + elif elem.tag == 'alias': + ctxts[width_type] = Alias( + _translate_alias(['months', ctxt_type, width_type], + elem.attrib['path']) + ) days = data.setdefault('days', {}) for ctxt in calendar.findall('days/dayContext'): - ctxts = days.setdefault(ctxt.attrib['type'], {}) + ctxt_type = ctxt.attrib['type'] + ctxts = days.setdefault(ctxt_type, {}) for width in ctxt.findall('dayWidth'): - widths = ctxts.setdefault(width.attrib['type'], {}) - for elem in width.findall('day'): - dtype = weekdays[elem.attrib['type']] - if 'draft' in elem.attrib and dtype in widths: - continue - widths[dtype] = unicode(elem.text) + width_type = width.attrib['type'] + widths = ctxts.setdefault(width_type, {}) + for elem in width.getiterator(): + if elem.tag == 'day': + dtype = weekdays[elem.attrib['type']] + if ('draft' in elem.attrib or 'alt' not in elem.attrib) \ + and dtype in widths: + continue + widths[dtype] = unicode(elem.text) + elif elem.tag == 'alias': + ctxts[width_type] = Alias( + _translate_alias(['days', ctxt_type, width_type], + elem.attrib['path']) + ) quarters = data.setdefault('quarters', {}) for ctxt in calendar.findall('quarters/quarterContext'): + ctxt_type = ctxt.attrib['type'] ctxts = quarters.setdefault(ctxt.attrib['type'], {}) for width in ctxt.findall('quarterWidth'): - widths = ctxts.setdefault(width.attrib['type'], {}) - for elem in width.findall('quarter'): - if 'draft' in elem.attrib and int(elem.attrib['type']) in widths: - continue - widths[int(elem.attrib.get('type'))] = unicode(elem.text) + width_type = width.attrib['type'] + widths = ctxts.setdefault(width_type, {}) + for elem in width.getiterator(): + if elem.tag == 'quarter': + if ('draft' in elem.attrib or 'alt' in elem.attrib) \ + and int(elem.attrib['type']) in widths: + continue + widths[int(elem.attrib['type'])] = unicode(elem.text) + elif elem.tag == 'alias': + ctxts[width_type] = Alias( + _translate_alias(['quarters', ctxt_type, width_type], + elem.attrib['path']) + ) eras = data.setdefault('eras', {}) for width in calendar.findall('eras/*'): - ewidth = { - 'eraAbbr': 'abbreviated', - 'eraNames': 'wide', - 'eraNarrow': 'narrow', - }[width.tag] - widths = eras.setdefault(ewidth, {}) - for elem in width.findall('era'): - if 'draft' in elem.attrib and int(elem.attrib['type']) in widths: - continue - widths[int(elem.attrib.get('type'))] = unicode(elem.text) + width_type = NAME_MAP[width.tag] + widths = eras.setdefault(width_type, {}) + for elem in width.getiterator(): + if elem.tag == 'era': + if ('draft' in elem.attrib or 'alt' in elem.attrib) \ + and int(elem.attrib['type']) in widths: + continue + widths[int(elem.attrib.get('type'))] = unicode(elem.text) + elif elem.tag == 'alias': + eras[width_type] = Alias( + _translate_alias(['eras', width_type], + elem.attrib['path']) + ) # AM/PM periods = data.setdefault('periods', {}) for elem in calendar.findall('am'): - if 'draft' in elem.attrib and elem.tag in periods: + if ('draft' in elem.attrib or 'alt' in elem.attrib) \ + and elem.tag in periods: continue periods[elem.tag] = unicode(elem.text) for elem in calendar.findall('pm'): - if 'draft' in elem.attrib and elem.tag in periods: + if ('draft' in elem.attrib or 'alt' in elem.attrib) \ + and elem.tag in periods: continue periods[elem.tag] = unicode(elem.text) date_formats = data.setdefault('date_formats', {}) - for elem in calendar.findall('dateFormats/dateFormatLength'): - if 'draft' in elem.attrib and elem.attrib.get('type') in date_formats: - continue - try: - date_formats[elem.attrib.get('type')] = \ - dates.parse_pattern(unicode(elem.findtext('dateFormat/pattern'))) - except ValueError, e: - print>>sys.stderr, 'ERROR: %s' % e + for format in calendar.findall('dateFormats'): + for elem in format.getiterator(): + if elem.tag == 'dateFormatLength': + if 'draft' in elem.attrib and \ + elem.attrib.get('type') in date_formats: + continue + try: + date_formats[elem.attrib.get('type')] = \ + dates.parse_pattern(unicode(elem.findtext('dateFormat/pattern'))) + except ValueError, e: + print>>sys.stderr, 'ERROR: %s' % e + elif elem.tag == 'alias': + date_formats = Alias(_translate_alias( + ['date_formats'], elem.attrib['path']) + ) time_formats = data.setdefault('time_formats', {}) - for elem in calendar.findall('timeFormats/timeFormatLength'): - if 'draft' in elem.attrib and elem.attrib.get('type') in time_formats: - continue - try: - time_formats[elem.attrib.get('type')] = \ - dates.parse_pattern(unicode(elem.findtext('timeFormat/pattern'))) - except ValueError, e: - print>>sys.stderr, 'ERROR: %s' % e + for format in calendar.findall('timeFormats'): + for elem in format.getiterator(): + if elem.tag == 'timeFormatLength': + if ('draft' in elem.attrib or 'alt' in elem.attrib) \ + and elem.attrib.get('type') in time_formats: + continue + try: + time_formats[elem.attrib.get('type')] = \ + dates.parse_pattern(unicode(elem.findtext('timeFormat/pattern'))) + except ValueError, e: + print>>sys.stderr, 'ERROR: %s' % e + elif elem.tag == 'alias': + time_formats = Alias(_translate_alias( + ['time_formats'], elem.attrib['path']) + ) datetime_formats = data.setdefault('datetime_formats', {}) - for elem in calendar.findall('dateTimeFormats/dateTimeFormatLength'): - if 'draft' in elem.attrib and elem.attrib.get('type') in datetime_formats: - continue - try: - datetime_formats[elem.attrib.get('type')] = \ - unicode(elem.findtext('dateTimeFormat/pattern')) - except ValueError, e: - print>>sys.stderr, 'ERROR: %s' % e + for format in calendar.findall('dateTimeFormats'): + for elem in format.getiterator(): + if elem.tag == 'dateTimeFormatLength': + if ('draft' in elem.attrib or 'alt' in elem.attrib) \ + and elem.attrib.get('type') in datetime_formats: + continue + try: + datetime_formats[elem.attrib.get('type')] = \ + unicode(elem.findtext('dateTimeFormat/pattern')) + except ValueError, e: + print>>sys.stderr, 'ERROR: %s' % e + elif elem.tag == 'alias': + datetime_formats = Alias(_translate_alias( + ['datetime_formats'], elem.attrib['path']) + ) # <numbers> @@ -318,28 +409,32 @@ decimal_formats = data.setdefault('decimal_formats', {}) for elem in tree.findall('//decimalFormats/decimalFormatLength'): - if 'draft' in elem.attrib and elem.attrib.get('type') in decimal_formats: + if ('draft' in elem.attrib or 'alt' in elem.attrib) \ + and elem.attrib.get('type') in decimal_formats: continue pattern = unicode(elem.findtext('decimalFormat/pattern')) decimal_formats[elem.attrib.get('type')] = numbers.parse_pattern(pattern) scientific_formats = data.setdefault('scientific_formats', {}) for elem in tree.findall('//scientificFormats/scientificFormatLength'): - if 'draft' in elem.attrib and elem.attrib.get('type') in scientific_formats: + if ('draft' in elem.attrib or 'alt' in elem.attrib) \ + and elem.attrib.get('type') in scientific_formats: continue pattern = unicode(elem.findtext('scientificFormat/pattern')) scientific_formats[elem.attrib.get('type')] = numbers.parse_pattern(pattern) currency_formats = data.setdefault('currency_formats', {}) for elem in tree.findall('//currencyFormats/currencyFormatLength'): - if 'draft' in elem.attrib and elem.attrib.get('type') in currency_formats: + if ('draft' in elem.attrib or 'alt' in elem.attrib) \ + and elem.attrib.get('type') in currency_formats: continue pattern = unicode(elem.findtext('currencyFormat/pattern')) currency_formats[elem.attrib.get('type')] = numbers.parse_pattern(pattern) percent_formats = data.setdefault('percent_formats', {}) for elem in tree.findall('//percentFormats/percentFormatLength'): - if 'draft' in elem.attrib and elem.attrib.get('type') in percent_formats: + if ('draft' in elem.attrib or 'alt' in elem.attrib) \ + and elem.attrib.get('type') in percent_formats: continue pattern = unicode(elem.findtext('percentFormat/pattern')) percent_formats[elem.attrib.get('type')] = numbers.parse_pattern(pattern) @@ -360,5 +455,6 @@ finally: outfile.close() + if __name__ == '__main__': main()