def extract_messages(dirs): catalog = Catalog( project='Open Library', copyright_holder='Internet Archive' ) METHODS = [ ("**.py", "python"), ("**.html", "openlibrary.i18n:extract_templetor") ] COMMENT_TAGS = ["NOTE:"] for d in dirs: if '.html' in d: extracted = [(d,) + extract for extract in extract_from_file("openlibrary.i18n:extract_templetor", d)] else: extracted = extract_from_dir(d, METHODS, comment_tags=COMMENT_TAGS, strip_comment_tags=True) for filename, lineno, message, comments, context in extracted: catalog.add(message, None, [(filename, lineno)], auto_comments=comments) path = os.path.join(root, 'messages.pot') f = open(path, 'w') write_po(f, catalog) f.close() print('wrote template to', path)
def convert_xlsm_content(contents): catalog = Catalog() for content in contents: resid = content[0] res_org_content = content[1] rescontent = content[2] if resid == None: continue if rescontent == None: rescontent = u"" if match(r"\((.+)(::(.+))+\)", res_org_content): msg_id = res_org_content[1:-1].split("::") msg_string = rescontent[1:-1].split("::") catalog.add(msg_id, context=resid, string=msg_string) else: if match(r"\".+\"", res_org_content): res_org_content = res_org_content[1:-1] if match(r"\".+\"", rescontent): rescontent = rescontent[1:-1] catalog.add(res_org_content, context=resid, string=rescontent) return catalog
def main(): logging.basicConfig(level=logging.INFO) ap = argparse.ArgumentParser() ap.add_argument("root", nargs="+", help="Directory roots to recurse through") args = ap.parse_args() logging.info("Searching for files...") filenames = list( sanity_utils.find_files(roots=args.root, ignored_dirs=sanity_utils.IGNORED_DIRS + ["migrations"], allowed_extensions=set(m[0] for m in MAPPING))) logging.info("Found %d files. Extracting messages...", len(filenames)) static_catalog = Catalog(charset="UTF-8") js_catalog = Catalog(charset="UTF-8") for filename in filenames: for filename, lineno, message, comments in extract_from_file(filename): catalog = (js_catalog if filename.endswith(".js") else static_catalog) catalog.add(message, None, [(filename, lineno)], auto_comments=comments) with open("shoop-static.pot", "wb") as outf: write_po_wrap(outf, static_catalog) logging.info("Wrote %d static messages to %s.", len(static_catalog), outf.name) with open("shoop-js.pot", "wb") as outf: write_po_wrap(outf, js_catalog) logging.info("Wrote %d JavaScript messages to %s.", len(js_catalog), outf.name)
def _extract(self, app): catalog = Catalog(domain="django", charset="utf8") files = {} for dirpath, dirnames, filenames in filtered_walk(app.path): for filename in filenames: filename = os.path.join(dirpath, filename) if ACCEPTABLE_FILENAMES_RE.match(filename): rel_filename = filename[len(os.path.commonprefix((app.path, filename))) + 1:].replace(os.sep, "/") files[rel_filename] = filename self.log.info("%s: %d translatable files found", app.label, len(files)) extractors = self.get_extractors() for rel_filename, filename in sorted(files.items()): extractor_tup = extractors.get(os.path.splitext(filename)[1]) if not extractor_tup: self.log.warning("Not sure how to extract messages from %s", filename) continue extractor, options = extractor_tup with open(filename, "rb") as fp: for (lineno, message, comments, context) in extract(extractor, fp, options=options): catalog.add(message, locations=[(rel_filename, 0)], auto_comments=comments) if len(catalog): pot_path = self._get_pot_path(app) with open(pot_path, "w") as outf: pofile.write_po(outf, catalog, width=1000, omit_header=True, sort_output=True) self.log.info("%s: %d messages in %s", app.label, len(catalog), pot_path) return catalog
def test_translation_plugins(app, tmpdir): session.lang = 'fr_FR' plugin = MockPlugin(plugin_engine, app) app.extensions['pluginengine'].plugins['dummy'] = plugin plugin.root_path = tmpdir.strpath french_core_str = DICTIONARIES['fr_FR']['This is not a string'] french_plugin_str = "This is not le french string" trans_dir = os.path.join(plugin.root_path, 'translations', 'fr_FR', 'LC_MESSAGES') os.makedirs(trans_dir) # Create proper *.mo file for plugin translation with open(os.path.join(trans_dir, 'messages.mo'), 'wb') as f: catalog = Catalog(locale='fr_FR', domain='plugin') catalog.add("This is not a string", "This is not le french string") write_mo(f, catalog) gettext_plugin = make_bound_gettext('dummy') assert _(u'This is not a string') == french_core_str assert gettext_context(u"This is not a string") == french_core_str assert gettext_plugin(u"This is not a string") == french_plugin_str with plugin.plugin_context(): assert _(u'This is not a string') == french_core_str assert gettext_context(u"This is not a string") == french_plugin_str assert gettext_plugin(u"This is not a string") == french_plugin_str
def create_locale(root, locale): path = os.path.join(root, locale, 'LC_MESSAGES') os.makedirs(path) catalog = Catalog(locale=locale) catalog.add('foo', 'bar {}'.format(locale)) with open(os.path.join(path, 'django.mo'), 'wb') as buf: write_mo(buf, catalog)
def extract_messages(dirs): catalog = Catalog(project='Open Library', copyright_holder='Internet Archive') METHODS = [("**.py", "python"), ("**.html", "openlibrary.i18n:extract_templetor")] COMMENT_TAGS = ["NOTE:"] for d in dirs: if '.html' in d: extracted = [(d, ) + extract for extract in extract_from_file( "openlibrary.i18n:extract_templetor", d)] else: extracted = extract_from_dir(d, METHODS, comment_tags=COMMENT_TAGS, strip_comment_tags=True) for filename, lineno, message, comments, context in extracted: catalog.add(message, None, [(filename, lineno)], auto_comments=comments) path = os.path.join(root, 'messages.pot') f = open(path, 'w') write_po(f, catalog) f.close() print('wrote template to', path)
def assert_convert(cls, po, xml=None, namespaces={}): """Helper that passes the string in ``po`` through our po to xml converter, and checks the resulting xml string value against ``xml``. If ``xml`` is not given, we check against ``po`` instead, i.e. expect the string to remain unchanged. """ key = 'test' catalog = Catalog() catalog.add(po, po, context=key) warnfunc = TestWarnFunc() dom = write_xml(po2xml(catalog, warnfunc=warnfunc), warnfunc=warnfunc) elem = dom.xpath('/resources/string[@name="%s"]' % key)[0] elem_as_text = etree.tostring(elem, encoding=unicode) value = re.match("^[^>]+>(.*)<[^<]+$", elem_as_text).groups(1)[0] match = xml if xml is not None else po print "'%s' == '%s'" % (value, match) print repr(value), '==', repr(match) assert value == match # If ``namespaces`` are set, the test expects those to be defined # in the root of the document for prefix, uri in namespaces.items(): assert dom.nsmap[prefix] == uri # In this case, the reverse (converting back to po) always needs to # give us the original again, so this allows for a nice extra check. if not match: # Skip this if we are doing a special, custom match. TestFromXML.assert_convert(match, po, namespaces) return warnfunc
def extract_messages(dirs: list[str]): catalog = Catalog(project='Open Library', copyright_holder='Internet Archive') METHODS = [("**.py", "python"), ("**.html", "openlibrary.i18n:extract_templetor")] COMMENT_TAGS = ["NOTE:"] for d in dirs: extracted = extract_from_dir(d, METHODS, comment_tags=COMMENT_TAGS, strip_comment_tags=True) counts = {} for filename, lineno, message, comments, context in extracted: counts[filename] = counts.get(filename, 0) + 1 catalog.add(message, None, [(filename, lineno)], auto_comments=comments) for filename, count in counts.items(): path = filename if d == filename else os.path.join(d, filename) print(f"{count}\t{path}", file=sys.stderr) path = os.path.join(root, 'messages.pot') f = open(path, 'wb') write_po(f, catalog) f.close() print('wrote template to', path)
def test_checkbox_null_label(): schema_translation = SchemaTranslation() catalog = Catalog(locale=Locale("cy")) catalog.add( "Rugby", "Rygbi", ) schema_translation.catalog = catalog schema = SurveySchema({ "question": { "answers": [{ "label": None, "options": [{ "label": "Rugby", "value": "Rugby" }] }], } }) translated = schema.translate(schema_translation) expected = { "question": { "answers": [{ "label": None, "options": [{ "label": "Rygbi", "value": "Rugby" }] }], }, "language": "cy", } assert expected == translated.schema
def assert_convert(cls, po, xml=None, namespaces={}): """Helper that passes the string in ``po`` through our po to xml converter, and checks the resulting xml string value against ``xml``. If ``xml`` is not given, we check against ``po`` instead, i.e. expect the string to remain unchanged. """ key = 'test' catalog = Catalog() catalog.add(po, po, context=key) dom = po2xml(catalog) elem = dom.xpath('/resources/string[@name="%s"]' % key)[0] elem_as_text = etree.tostring(elem, encoding=unicode) value = re.match("^[^>]+>(.*)<[^<]+$", elem_as_text).groups(1)[0] match = xml if xml is not None else po print "'%s' == '%s'" % (value, match) print repr(value), '==', repr(match) assert value == match # If ``namespaces`` are set, the test expects those to be defined # in the root of the document for prefix, uri in namespaces.items(): assert dom.nsmap[prefix] == uri # In this case, the reverse (converting back to po) always needs # to give us the original again, so this allows for a nice extra # check. TestFromXML.assert_convert(match, po, namespaces)
def test(self): p = self.setup_project() c = Catalog(locale='de') c.add('en1', 'de1', flags=('fuzzy',), context='foo') c.add('en2', 'de2', context='bar') p.write_po(c, 'de.po') p.program('import', {'--ignore-fuzzy': True}) xml = p.get_xml('de') assert not 'foo' in xml assert 'bar' in xml
def mkcatalog(locale='de'): """Helper that returns a gettext catalog with one message already added. Tests can add a broken message and then ensure that at least the valid message still was processed. """ c = Catalog(locale='de') c.add('valid_message', 'valid_value', context='valid_message') return c
def get_translation_catalog_from_index(index, project_name): # Get all Modulemd.Module object names module_names = index.get_module_names() # Create a list containing highest version streams of a module final_streams = list() for module_name in module_names: module = index.get_module(module_name) stream_names = module.get_stream_names() for stream_name in stream_names: # The first item returned is guaranteed to be the highest version # of that stream in that module. stream = module.search_streams(stream_name, 0)[0] final_streams.append(stream) # A dictionary to store: # key: all translatable strings # value: their respective locations translation_dict = defaultdict(list) for stream in final_streams: # Process description description = stream.get_description("C") if description is not None: location = (("{};{};description").format( stream.props.module_name, stream.props.stream_name), 2) translation_dict[description].append(location) # Process summary summary = stream.get_summary("C") if summary is not None: location = (("{};{};summary").format( stream.props.module_name, stream.props.stream_name), 1) translation_dict[summary].append(location) # Process profile descriptions(sometimes NULL) profile_names = stream.get_profile_names() if(profile_names): for pro_name in profile_names: profile = stream.get_profile(pro_name) profile_desc = profile.get_description("C") if profile_desc is not None: location = (("{};{};profile;{}").format( stream.props.module_name, stream.props.stream_name, pro_name), 3) translation_dict[profile_desc].append(location) catalog = Catalog(project=project_name) for translatable_string, locations in translation_dict.items(): logging.debug("This is a translatable_string: %s" % translatable_string) catalog.add(translatable_string, locations=locations) return catalog
def test_invalid_xhtml(): """Ensure we can deal with broken XML in messages. """ c = Catalog() c.add('Foo', '<i>Tag is not closed', context="foo") # [bug] This caused an exception in 16263b. dom = write_xml(po2xml(c)) # The tag was closed automatically (our loose parser tries to fix errors). assert etree.tostring(dom) == b'<resources><string name="foo"><i>Tag is not closed</i></string></resources>'
def test_translate_sets_language(): catalog = Catalog(locale=Locale("cy")) catalog.add("test") schema_translation = SchemaTranslation(catalog=catalog) schema = SurveySchema({}) schema.translate(schema_translation) translated_schema = schema.translate(schema_translation) assert translated_schema.language == "cy"
def test_invalid_xhtml(): """Ensure we can deal with broken XML in messages. """ c = Catalog() c.add('Foo', '<i>Tag is not closed', context="foo") # [bug] This caused an exception in 16263b. dom = write_xml(po2xml(c)) # The tag was closed automatically (our loose parser tries to fix errors). assert etree.tostring( dom ) == b'<resources><string name="foo"><i>Tag is not closed</i></string></resources>'
def test_load(self): tempdir = tempfile.mkdtemp() try: messages_dir = os.path.join(tempdir, 'fr', 'LC_MESSAGES') os.makedirs(messages_dir) catalog = Catalog(locale='fr', domain='messages') catalog.add('foo', 'bar') write_mo(file(os.path.join(messages_dir, 'messages.mo'), 'wb'), catalog) translations = support.Translations.load(tempdir, locales=('fr',), domain='messages') self.assertEqual('bar', translations.gettext('foo')) finally: shutil.rmtree(tempdir)
def test_multiple_pos(self): """If the language writes to multiple .po files, those are all counted together. Either all of them will be written, or none of them will be. """ p = self.setup_project() p.write_xml(kind='strings') p.write_xml(kind='arrays') # Create two catalogs, one fully translated, the other one not # at all. c = Catalog(locale='de') c.add('translated', 'value', context='translated') p.write_po(c, 'strings-de.po') c = Catalog(locale='de') c.add('untranslated', context='untranslated') p.write_po(c, 'arrays-de.po') # If we require 100% completness on import, both files will # be empty, even though one of them is fully translated. But # the second drags down the total of the group. p.program('import', {'--require-min-complete': '1'}) assert len(p.get_xml('de', kind='strings')) == 0 assert len(p.get_xml('de', kind='arrays')) == 0
def test_update(self): template = Catalog() template.add("1") template.add("2") template.add("3") tmpl_file = os.path.join(self._i18n_dir(), 'temp-template.pot') with open(tmpl_file, "wb") as outfp: write_po(outfp, template) po_file = os.path.join(self._i18n_dir(), 'temp1.po') self.cli.run(sys.argv + ['init', '-l', 'fi', '-o', po_file, '-i', tmpl_file]) with open(po_file, "r") as infp: catalog = read_po(infp) assert len(catalog) == 3 # Add another entry to the template template.add("4") with open(tmpl_file, "wb") as outfp: write_po(outfp, template) self.cli.run(sys.argv + ['update', '-l', 'fi_FI', '-o', po_file, '-i', tmpl_file]) with open(po_file, "r") as infp: catalog = read_po(infp) assert len(catalog) == 4 # Catalog was updated
def setUp(self): # Use a locale which won't fail to run the tests os.environ['LANG'] = 'en_US.UTF-8' messages1 = [ ('foo', {'string': 'Voh'}), ('foo', {'string': 'VohCTX', 'context': 'foo'}), (('foo1', 'foos1'), {'string': ('Voh1', 'Vohs1')}), (('foo1', 'foos1'), {'string': ('VohCTX1', 'VohsCTX1'), 'context': 'foo'}), ] messages2 = [ ('foo', {'string': 'VohD'}), ('foo', {'string': 'VohCTXD', 'context': 'foo'}), (('foo1', 'foos1'), {'string': ('VohD1', 'VohsD1')}), (('foo1', 'foos1'), {'string': ('VohCTXD1', 'VohsCTXD1'), 'context': 'foo'}), ] catalog1 = Catalog(locale='en_GB', domain='messages') catalog2 = Catalog(locale='en_GB', domain='messages1') for ids, kwargs in messages1: catalog1.add(ids, **kwargs) for ids, kwargs in messages2: catalog2.add(ids, **kwargs) catalog1_fp = BytesIO() catalog2_fp = BytesIO() write_mo(catalog1_fp, catalog1) catalog1_fp.seek(0) write_mo(catalog2_fp, catalog2) catalog2_fp.seek(0) translations1 = support.Translations(catalog1_fp) translations2 = support.Translations(catalog2_fp, domain='messages1') self.translations = translations1.add(translations2, merge=False)
def test_load(self): tempdir = tempfile.mkdtemp() try: messages_dir = os.path.join(tempdir, "fr", "LC_MESSAGES") os.makedirs(messages_dir) catalog = Catalog(locale="fr", domain="messages") catalog.add("foo", "bar") with open(os.path.join(messages_dir, "messages.mo"), "wb") as f: write_mo(f, catalog) translations = support.Translations.load(tempdir, locales=("fr",), domain="messages") self.assertEqual("bar", translations.gettext("foo")) finally: shutil.rmtree(tempdir)
def test_load(self): tempdir = tempfile.mkdtemp() try: messages_dir = os.path.join(tempdir, 'fr', 'LC_MESSAGES') os.makedirs(messages_dir) catalog = Catalog(locale='fr', domain='messages') catalog.add('foo', 'bar') with open(os.path.join(messages_dir, 'messages.mo'), 'wb') as f: write_mo(f, catalog) translations = support.Translations.load(tempdir, locales=('fr',), domain='messages') self.assertEqual('bar', translations.gettext('foo')) finally: shutil.rmtree(tempdir)
def _create_catalog(self, path_id, domain=u'foo', locale=u'de', **messages): i18n_dir = os.path.join(self._tempdir, path_id) path = os.path.join(i18n_dir, locale, 'LC_MESSAGES') os.makedirs(path) mo_filename = os.path.join(path, '%s.mo' % domain) assert_false(os.path.exists(mo_filename)) catalog = Catalog(locale=locale, domain=domain, fuzzy=False) for message_id, translation in messages.items(): catalog.add(message_id, translation) mo_fp = file(mo_filename, 'wb') write_mo(mo_fp, catalog) mo_fp.close() return i18n_dir
def test_untranslated(): """Test that by default, untranslated strings are not included in the imported XML. """ catalog = Catalog() catalog.add('green', context='color1') catalog.add('red', 'rot', context='color2') assert po2xml(catalog) == {'color2': 'rot'} # If with_untranslated is passed, then all strings are included. # Note that arrays behave differently (they always include all # strings), and this is tested in test_string_arrays.py). assert po2xml(catalog, with_untranslated=True) ==\ {'color1': 'green', 'color2': 'rot'}
def test_fuzzy_flag_on_catalog_update(): from sphinx_intl import catalog cat = Catalog(locale='ja', domain='domain', fuzzy=False) msg = Message('Hello Internationalized Sphinx World !', u'こんにちは国際化されたSphinxの世界!') cat[msg.id] = msg assert not msg.fuzzy cat_src = Catalog(locale='en', domain='domain', fuzzy=False) msg_src = Message('Hello Internationalized Sphinx World ?') cat_src[msg_src.id] = msg_src catalog.update_with_fuzzy(cat, cat_src) assert msg.id not in cat assert cat[msg_src.id].fuzzy
def read_po(po_path): """ Read a Po file :param po_path: .po path :return: po content """ from babel.messages import pofile logger = logging.getLogger('destral.utils.read_po') try: with open(po_path, 'r') as pot: catalog = pofile.read_po(pot) return catalog except ValueError: # If bad formatted data, replace it with open(po_path, 'r') as potB: data = potB.read() from re import sub, findall from dateutil.parser import parse as date_parse from babel.messages import Catalog replace_string = r'\1 {}\2'.format( date_parse(findall(r"POT-Creation-Date: (.*)\\n", data)[0]).strftime('%Y-%m-%d %H:%M')) data = sub(r"(POT-Creation-Date:).*(\\n)", replace_string, data) replace_string = r'\1 {}\2'.format( date_parse(findall(r"PO-Revision-Date: (.*)\\n", data)[0]).strftime('%Y-%m-%d %H:%M')) data = sub(r"(PO-Revision-Date:).*(\\n)", replace_string, data) catalog = Catalog() parser = pofile.PoFileParser(catalog) parser.parse(data.split('\n')) logger.warning('Data of POfile {} has bad formatted ' 'creation or revision dates'.format(po_path)) return catalog
def test_export_with_empty_master_xml(self): """[Regression] Test that export works fine if the master resource is empty.""" p = self.setup_project(xml_langs=['de']) p.write_xml(data="""<resources></resources>""", lang='de') p.write_po(Catalog('de')) assert not '[failed]' in p.program('export')
def _update(self, app, template_catalog, langs): if template_catalog is None: with open(_get_pot_path(app)) as infp: template_catalog = pofile.read_po(infp, charset="utf8") for lang in langs: po_path = _get_po_path(app, language=lang) if os.path.isfile(po_path): with open(po_path) as infp: lang_catalog = pofile.read_po(infp, charset="utf8") else: lang_catalog = Catalog(locale=lang, charset="utf8") lang_catalog.update(template_catalog) if len(lang_catalog): with open(po_path, "w") as outf: pofile.write_po(outf, lang_catalog, width=1000, omit_header=True, sort_output=True) self.log.info("%s: updated %s", app.label, po_path)
def test_export(self): """Test that the export command maintains the proper plural form, and actually replaces an incorrect one.""" p = self.setup_project() p.write_xml(data="""<resources></resources>""") p.write_xml(data="""<resources></resources>""", lang='ja') # Generate a catalog with different plural rules than we expect catalog = Catalog('ja') catalog._num_plurals, catalog._plural_expr = 2, '(n < 2)' p.write_po(catalog) # Export should override the info assert 'Plural-Forms header' in p.program('export') catalog = p.get_po('ja.po') assert catalog.num_plurals == 1 assert catalog.plural_expr == '(0)'
def test_update(self): template = Catalog() template.add("1") template.add("2") template.add("3") tmpl_file = os.path.join(self._i18n_dir(), "temp-template.pot") with open(tmpl_file, "wb") as outfp: write_po(outfp, template) po_file = os.path.join(self._i18n_dir(), "temp1.po") self.cli.run(sys.argv + ["init", "-l", "fi", "-o", po_file, "-i", tmpl_file]) with open(po_file, "r") as infp: catalog = read_po(infp) assert len(catalog) == 3 # Add another entry to the template template.add("4") with open(tmpl_file, "wb") as outfp: write_po(outfp, template) self.cli.run(sys.argv + ["update", "-l", "fi_FI", "-o", po_file, "-i", tmpl_file]) with open(po_file, "r") as infp: catalog = read_po(infp) assert len(catalog) == 4 # Catalog was updated
def test_update(self): template = Catalog() template.add("1") template.add("2") template.add("3") tmpl_file = os.path.join(self._i18n_dir(), 'temp-template.pot') with open(tmpl_file, "wb") as outfp: write_po(outfp, template) po_file = os.path.join(self._i18n_dir(), 'temp1.po') self.cli.run(sys.argv + ['init', '-l', 'fi', '-o', po_file, '-i', tmpl_file ]) with open(po_file, "r") as infp: catalog = read_po(infp) assert len(catalog) == 3 # Add another entry to the template template.add("4") with open(tmpl_file, "wb") as outfp: write_po(outfp, template) self.cli.run(sys.argv + ['update', '-l', 'fi_FI', '-o', po_file, '-i', tmpl_file]) with open(po_file, "r") as infp: catalog = read_po(infp) assert len(catalog) == 4 # Catalog was updated
def test_clear(self): """Explicitely test that if we ignore a language, the xml file is overwritten with an empty version. Just not processing it is not enough. """ p = self.setup_project() # We start out with one string in the XML p.write_xml({'string1': 'value1'}, lang='de') assert len(p.get_xml('de')) == 1 c = Catalog(locale='de') c.add('string1', context='string1') p.write_po(c, 'de.po') # Now after the import, the resource file is empty. p.program('import', {'--require-min-complete': '1'}) assert len(p.get_xml('de')) == 0
def _update(self, app, template_catalog, langs): if template_catalog is None: with open(self._get_pot_path(app)) as infp: template_catalog = pofile.read_po(infp, charset="utf8") for lang in langs: po_path = self._get_po_path(app, language=lang) if os.path.isfile(po_path): with open(po_path) as infp: lang_catalog = pofile.read_po(infp, charset="utf8") else: # When running on a version of Babel without proper locale data, pretend we don't even want # to create a locale-bound catalog. Baka Babel. lang_catalog = Catalog(locale=(lang if _babel_has_locale_data() else None), charset="utf8") lang_catalog.update(template_catalog) if len(lang_catalog): with open(po_path, "w") as outf: pofile.write_po(outf, lang_catalog, width=1000, omit_header=True, sort_output=True) self.log.info("%s: updated %s", app.label, po_path)
def test_write_and_read_po_file_with_non_ascii_string(temp): from sphinx_intl import catalog cat = Catalog(locale='ja', domain='domain', fuzzy=False) msg = Message('Hello World', u'こんにちは世界') cat[msg.id] = msg po_file = (temp / 'domain.po') catalog.dump_po(po_file, cat) cat2 = catalog.load_po(po_file) assert cat2[msg.id].string == msg.string
def test_fuzzy(self): """This option is affected by the --ignore-fuzzy flag. If it is set, fuzzy strings are not counted towards the total. """ p = self.setup_project() c = Catalog(locale='de') c.add('translated', 'value', context='translated') c.add('fuzzy', 'value', context='fuzzy', flags=('fuzzy',)) p.write_po(c, 'de.po') # When fuzzy strings are counted, the catalog above is 100% # complete. p.program('import', {'--require-min-complete': '1'}) assert len(p.get_xml('de')) == 2 # If they aren't, it won't make the cut and the result should # be no strings at all being written. p.program('import', {'--require-min-complete': '1', '--ignore-fuzzy': True}) assert len(p.get_xml('de')) == 0
def setUp(self): # Use a locale which won't fail to run the tests os.environ["LANG"] = "en_US.UTF-8" messages1 = [ ("foo", {"string": "Voh"}), ("foo", {"string": "VohCTX", "context": "foo"}), (("foo1", "foos1"), {"string": ("Voh1", "Vohs1")}), (("foo1", "foos1"), {"string": ("VohCTX1", "VohsCTX1"), "context": "foo"}), ] messages2 = [ ("foo", {"string": "VohD"}), ("foo", {"string": "VohCTXD", "context": "foo"}), (("foo1", "foos1"), {"string": ("VohD1", "VohsD1")}), (("foo1", "foos1"), {"string": ("VohCTXD1", "VohsCTXD1"), "context": "foo"}), ] catalog1 = Catalog(locale="en_GB", domain="messages") catalog2 = Catalog(locale="en_GB", domain="messages1") for ids, kwargs in messages1: catalog1.add(ids, **kwargs) for ids, kwargs in messages2: catalog2.add(ids, **kwargs) catalog1_fp = BytesIO() catalog2_fp = BytesIO() write_mo(catalog1_fp, catalog1) catalog1_fp.seek(0) write_mo(catalog2_fp, catalog2) catalog2_fp.seek(0) translations1 = support.Translations(catalog1_fp) translations2 = support.Translations(catalog2_fp, domain="messages1") self.translations = translations1.add(translations2, merge=False)
def test_variant_translation(schema_with_question_variants): schema_translation = SchemaTranslation() catalog = Catalog() catalog.add( "First name", "WELSH - First name", context="What is your name?", ) catalog.add( "First name", "WELSH - First name - Proxy", context="What is their name?", ) schema_translation.catalog = catalog variant_schema = SurveySchema(schema_with_question_variants) translated = variant_schema.translate(schema_translation) assert (translated.schema["question_variants"][0]["question"]["answers"][0] ["label"] == "WELSH - First name") assert (translated.schema["question_variants"][1]["question"]["answers"][0] ["label"] == "WELSH - First name - Proxy")
def test_empty_translation_with_fallback(self): catalog1 = Catalog(locale='fr_FR') catalog1.add(u'', '''\ "Content-Type: text/plain; charset=utf-8\n" "Content-Transfer-Encoding: 8bit\n''') catalog1.add(u'Fuzz', '') buf1 = BytesIO() mofile.write_mo(buf1, catalog1) buf1.seek(0) catalog2 = Catalog(locale='fr') catalog2.add(u'', '''\ "Content-Type: text/plain; charset=utf-8\n" "Content-Transfer-Encoding: 8bit\n''') catalog2.add(u'Fuzz', 'Flou') buf2 = BytesIO() mofile.write_mo(buf2, catalog2) buf2.seek(0) translations = Translations(fp=buf1) translations.add_fallback(Translations(fp=buf2)) self.assertEqual(u'Flou', translations.ugettext('Fuzz'))
def _extract(self, app): catalog = Catalog(domain="django", charset="utf8") files = {} for dirpath, dirnames, filenames in filtered_walk(app.path): for filename in filenames: filename = os.path.join(dirpath, filename) if ACCEPTABLE_FILENAMES_RE.match(filename): rel_filename = filename[ len(os.path.commonprefix((app.path, filename))) + 1:].replace(os.sep, "/") files[rel_filename] = filename self.log.info("%s: %d translatable files found", app.label, len(files)) extractors = self.get_extractors() for rel_filename, filename in sorted(files.items()): extractor_tup = extractors.get(os.path.splitext(filename)[1]) if not extractor_tup: self.log.warning("Not sure how to extract messages from %s", filename) continue extractor, options = extractor_tup with open(filename, "rb") as fp: for (lineno, message, comments, context) in extract(extractor, fp, options=options): catalog.add(message, locations=[(rel_filename, 0)], auto_comments=comments) if len(catalog): pot_path = _get_pot_path(app) with open(pot_path, "w") as outf: pofile.write_po(outf, catalog, width=1000, omit_header=True, sort_output=True) self.log.info("%s: %d messages in %s", app.label, len(catalog), pot_path) return catalog
def main(): print "Extracting messages" root = path.abspath(path.join(path.dirname(__file__), '..')) catalog = Catalog(msgid_bugs_address=BUGS_ADDRESS, copyright_holder=COPYRIGHT, charset="utf-8") def callback(filename, method, options): if method != "ignore": print strip_path(filename, root) extracted_py = extract_from_dir(root, PY_METHODS, {}, PY_KEYWORDS, COMMENT_TAGS, callback=callback, strip_comment_tags=True) extracted_js = extract_from_dir(root, [("static/js/**.js", "javascript")], {}, JS_KEYWORDS, COMMENT_TAGS, callback=callback, strip_comment_tags=True) for filename, lineno, message, comments in chain(extracted_py, extracted_js): catalog.add(message, None, [(strip_path(filename, root), lineno)], auto_comments=comments) output_path = path.join(root, "i18n") if not path.isdir(output_path): makedirs(output_path) f = file(path.join(output_path, "messages.pot"), "w") try: write_po(f, catalog, width=79) finally: f.close() print "All done."
def test_catalog_merge_files(): # Refs issues #92, #162 t1 = support.Translations() assert t1.files == [] t1._catalog["foo"] = "bar" fp = BytesIO() write_mo(fp, Catalog()) fp.seek(0) fp.name = "pro.mo" t2 = support.Translations(fp) assert t2.files == ["pro.mo"] t2._catalog["bar"] = "quux" t1.merge(t2) assert t1.files == ["pro.mo"] assert set(t1._catalog.keys()) == {'', 'foo', 'bar'}
def test(self): p = self.setup_project() c = Catalog(locale='de') c.add('en1', 'de1', flags=('fuzzy', ), context='foo') c.add('en2', 'de2', context='bar') p.write_po(c, 'de.po') p.program('import', {'--ignore-fuzzy': True}) xml = p.get_xml('de') assert not 'foo' in xml assert 'bar' in xml
def test(self): p = self.setup_project() c = Catalog(locale='de') c.add('translated', 'value', context='translated') c.add('missing1', context='missing1') c.add('missing2', context='missing2') c.add('missing3', context='missing3') p.write_po(c, 'de.po') # At first, we require half the strings to be available. # This is clearly not the case in the catalog above. p.program('import', {'--require-min-complete': '0.5'}) assert len(p.get_xml('de')) == 0 # Now, only require 25% - this should just make the cut. p.program('import', {'--require-min-complete': '0.25'}) assert len(p.get_xml('de')) == 1
def xml2po(resources, translations=None, filter=None, warnfunc=dummy_warn): """Return ``resources`` as a Babel .po ``Catalog`` instance. If given, ``translations`` will be used for the translated values. In this case, the returned value is a 2-tuple (catalog, unmatched), with the latter being a list of Android string resource names that are in the translated file, but not in the original. Both ``resources`` and ``translations`` must be ``ResourceTree`` objects, as returned by ``read_xml()``. From the application perspective, it will call this function with a ``translations`` object when initializing a new .po file based on an existing resource file (the 'init' command). For 'export', this function is called without translations. It will thus generate what is essentially a POT file (an empty .po file), and this will be merged into the existing .po catalogs, as per how gettext usually """ assert not translations or translations.language catalog = Catalog() if translations is not None: catalog.locale = translations.language.locale # We cannot let Babel determine the plural expr for the locale by # itself. It will use a custom list of plural expressions rather # than generate them based on CLDR. # See http://babel.edgewall.org/ticket/290. set_catalog_plural_forms(catalog, translations.language) for name, org_value in resources.iteritems(): if filter and filter(name): continue trans_value = None if translations: trans_value = translations.pop(name, trans_value) if isinstance(org_value, StringArray): # a string-array, write as "name:index" if len(org_value) == 0: warnfunc("Warning: string-array '%s' is empty" % name, "warning") continue if not isinstance(trans_value, StringArray): if trans_value: warnfunc( ('""%s" is a string-array in the reference ' "file, but not in the translation.") % name, "warning", ) trans_value = StringArray() for index, item in enumerate(org_value): item_trans = trans_value[index].text if index < len(trans_value) else u"" # If the string has formatting markers, indicate it in # the gettext output flags = [] if item.formatted: flags.append("c-format") ctx = "%s:%d" % (name, index) catalog.add(item.text, item_trans, auto_comments=item.comments, flags=flags, context=ctx) elif isinstance(org_value, Plurals): # a plurals, convert to a gettext plurals if len(org_value) == 0: warnfunc("Warning: plurals '%s' is empty" % name, "warning") continue if not isinstance(trans_value, Plurals): if trans_value: warnfunc( ('""%s" is a plurals in the reference ' "file, but not in the translation.") % name, "warning" ) trans_value = Plurals() # Taking the Translation objects for each quantity in ``org_value``, # we build a list of strings, which is how plurals are represented # in Babel. # # Since gettext only allows comments/flags on the whole # thing at once, we merge the comments/flags of all individual # plural strings into one. formatted = False comments = [] for _, translation in org_value.items(): if translation.formatted: formatted = True comments.extend(translation.comments) # For the message id, choose any two plural forms, but prefer # "one" and "other", assuming an English master resource. temp = org_value.copy() singular = ( temp.pop("one") if "one" in temp else temp.pop("other") if "other" in temp else temp.pop(temp.keys()[0]) ) plural = temp.pop("other") if "other" in temp else temp[temp.keys()[0]] if temp else singular msgid = (singular.text, plural.text) del temp, singular, plural # We pick the quantities supported by the language (the rest # would be ignored by Android as well). msgstr = "" if trans_value: allowed_keywords = translations.language.plural_keywords msgstr = ["" for i in range(len(allowed_keywords))] for quantity, translation in trans_value.items(): try: index = translations.language.plural_keywords.index(quantity) except ValueError: warnfunc( ( '"plurals "%s" uses quantity "%s", which ' "is not supported for this language. See " "the README for an explanation. The " "quantity has been ignored" ) % (name, quantity), "warning", ) else: msgstr[index] = translation.text flags = [] if formatted: flags.append("c-format") catalog.add(msgid, tuple(msgstr), flags=flags, auto_comments=comments, context=name) else: # a normal string # If the string has formatting markers, indicate it in # the gettext output # TODO DRY this. flags = [] if org_value.formatted: flags.append("c-format") catalog.add( org_value.text, trans_value.text if trans_value else u"", flags=flags, auto_comments=org_value.comments, context=name, ) if translations is not None: # At this point, trans_strings only contains those for which # no original existed. return catalog, translations.keys() else: return catalog
def do_extract_messages(target=('t', ''), domain=('d', 'messages'), i18n_dir=('i', ''), all=('a', False)): """ Extract messages and create pot file. """ if not domain in ('messages', 'jsmessages'): print_status('invalid domain.') sys.exit(1) if not target and not all: print_status('Please specify target.') sys.exit(1) elif target == 'kay': print_status('Extracting core strings') root = kay.KAY_DIR elif all: targets = get_user_apps() for target in targets: do_extract_messages(target=target, domain=domain, i18n_dir=None, all=False) sys.exit(0) else: root = path.abspath(target) if not path.isdir(root): print_status('source folder missing') sys.exit(1) print_status('Extracting from %s' % root) if domain == 'messages': methods = METHODS else: methods = JSMETHODS catalog = Catalog(msgid_bugs_address=BUGS_ADDRESS, copyright_holder=COPYRIGHT, charset='utf-8') def callback(filename, method, options): if method != 'ignore': print_status(strip_path(filename, root)) option = {} option['extensions'] = ','.join(settings.JINJA2_EXTENSIONS) option.update(settings.JINJA2_ENVIRONMENT_KWARGS) options = { '**/templates/**.*': option, '**.html': option, } extracted = extract_from_dir(root, methods, options, KEYWORDS, COMMENT_TAGS, callback=callback, strip_comment_tags=True) for filename, lineno, message, comments in extracted: catalog.add(message, None, [(strip_path(filename, root), lineno)], auto_comments=comments) if not i18n_dir: i18n_dir = path.join(root, 'i18n') if not path.isdir(i18n_dir): makedirs(i18n_dir) f = file(path.join(i18n_dir, domain+'.pot'), 'w') try: write_po(f, catalog, width=79) finally: f.close() print_status('All done.')
import unittest from datetime import datetime, timedelta from io import BytesIO import babel.dates from babel.messages import Catalog from babel.messages.mofile import write_mo from cms.locale import Translation, DEFAULT_TRANSLATION, filter_language_codes, \ choose_language_code UTC = babel.dates.UTC ROME = babel.dates.get_timezone("Europe/Rome") FRENCH_CATALOG = Catalog(locale="fr") FRENCH_CATALOG.add("%s KiB", "%s Kio") FRENCH_CATALOG.add("%s MiB", "%s Mio") FRENCH_CATALOG.add("%s GiB", "%s Gio") FRENCH_CATALOG.add("%s TiB", "%s Tio") FRENCH_CATALOG_BUF = BytesIO() write_mo(FRENCH_CATALOG_BUF, FRENCH_CATALOG) FRENCH_CATALOG_BUF.seek(0) ENGLISH = DEFAULT_TRANSLATION BRITISH_ENGLISH = Translation("en_GB") FRENCH = Translation("fr", mofile=FRENCH_CATALOG_BUF) HINDI = Translation("hi") ITALIAN = Translation("it") NORWEGIAN = Translation("no") CHINESE = Translation("zh_CN")
def test_more_plural_forms(self): catalog2 = Catalog(locale='ru_RU') catalog2.add(('Fuzz', 'Fuzzes'), ('', '', '')) buf = StringIO() mofile.write_mo(buf, catalog2)
def test_sorting(self): # Ensure the header is sorted to the first entry so that its charset # can be applied to all subsequent messages by GNUTranslations # (ensuring all messages are safely converted to unicode) catalog = Catalog(locale='en_US') catalog.add(u'', '''\ "Content-Type: text/plain; charset=utf-8\n" "Content-Transfer-Encoding: 8bit\n''') catalog.add(u'foo', 'Voh') catalog.add((u'There is', u'There are'), (u'Es gibt', u'Es gibt')) catalog.add(u'Fizz', '') catalog.add(('Fuzz', 'Fuzzes'), ('', '')) buf = StringIO() mofile.write_mo(buf, catalog) buf.seek(0) translations = gettext.GNUTranslations(fp=buf) self.assertEqual(u'Voh', translations.ugettext('foo')) assert isinstance(translations.ugettext('foo'), unicode) self.assertEqual(u'Es gibt', translations.ungettext('There is', 'There are', 1)) assert isinstance(translations.ungettext('There is', 'There are', 1), unicode) self.assertEqual(u'Fizz', translations.ugettext('Fizz')) assert isinstance(translations.ugettext('Fizz'), unicode) self.assertEqual(u'Fuzz', translations.ugettext('Fuzz')) assert isinstance(translations.ugettext('Fuzz'), unicode) self.assertEqual(u'Fuzzes', translations.ugettext('Fuzzes')) assert isinstance(translations.ugettext('Fuzzes'), unicode)