def assert_examples(self, examples, lang=None, logger=None):
cls = type(self)
# store examples
if getattr(cls, '_store_examples', False):
cls._stored_examples.update(examples)
return
errors = []
lang = lang or self.lang
if isinstance(lang, basestring):
lang = hangulize.get_lang(lang)
try:
lang_name = type(lang).__name__
except AttributeError:
lang_name = 'AnonymouseLanguage'
for word, want in examples.items():
try:
got = lang.hangulize(word, logger=logger)
assert want == got
except self.failureException:
errors.append((word, want, got))
if errors:
def form(error):
return " * '%s' should be '%s', but '%s' was given" % (
color(error[0], 'cyan'),
color(error[1], 'green'),
color(error[2], 'red')
)
errors = map(form, errors)
msg = color(lang_name, 'yellow') + '\n' + '\n'.join(errors)
raise HangulizeAssertionError(msg.encode('utf-8'))
def test_singleton(self):
from hangulize import get_lang
from hangulize.langs.ita import Italian
from hangulize.langs.jpn import Japanese
assert Italian() is Italian()
assert get_lang('ita') is Italian()
assert get_lang('ita') is get_lang('ita')
assert Japanese() is Japanese()
assert get_lang('jpn') is Japanese()
assert get_lang('jpn') is get_lang('jpn')
assert Italian() is not Japanese()
assert get_lang('ita') is not Japanese()
assert get_lang('ita') is not get_lang('jpn')
def test_singleton(self):
from hangulize import get_lang
from hangulize.langs.ita import Italian
from hangulize.langs.jpn import Japanese
assert Italian() is Italian()
assert get_lang('ita') is Italian()
assert get_lang('ita') is get_lang('ita')
assert Japanese() is Japanese()
assert get_lang('jpn') is Japanese()
assert get_lang('jpn') is get_lang('jpn')
assert Italian() is not Japanese()
assert get_lang('ita') is not Japanese()
assert get_lang('ita') is not get_lang('jpn')
def _repl():
while True:
lang = self.lang or raw_input(color("Lang: ", "magenta"))
try:
lang = get_lang(lang)
logger.info("** " + color(type(lang).__name__, "green") + " is selected")
break
except Exception, e:
logger.error(color(e, "red"))
self.lang = None
def _repl():
while True:
lang = self.lang or raw_input(color('Lang: ', 'magenta'))
try:
lang = get_lang(lang)
logger.info('** ' + color(type(lang).__name__, 'green') + \
' is selected')
break
except HangulizeError, e:
logger.error(color(e, 'red'))
self.lang = None
def _repl():
while True:
lang = self.lang or raw_input(color('Lang: ', 'magenta'))
try:
lang = get_lang(lang)
logger.info('** ' + color(type(lang).__name__, 'green') +
' is selected')
break
except Exception, e:
logger.error(color(e, 'red'))
self.lang = None
def _repl():
while True:
lang = self.lang or raw_input(color('Lang: ', 'magenta'))
try:
lang = get_lang(lang)
logger.info('** ' + color(type(lang).__name__, 'green') +
' is selected')
break
except Exception as e:
logger.error(color(e, 'red'))
self.lang = None
while True:
string = raw_input(color('==> ', 'cyan'))
if not string:
logger.info('** ' + color('End', 'green'))
break
yield lang.hangulize(string.decode(encoding), logger=logger)
def _repl():
while True:
lang = self.lang or raw_input(color('Lang: ', 'magenta'))
try:
lang = get_lang(lang)
logger.info('** ' + color(type(lang).__name__, 'green') +
' is selected')
break
except Exception as e:
logger.error(color(e, 'red'))
self.lang = None
while True:
string = raw_input(color('==> ', 'cyan'))
if not string:
logger.info('** ' + color('End', 'green'))
break
yield lang.hangulize(string.decode(encoding), logger=logger)
def test_regard_iso639_1(self):
assert type(get_lang('bg', iso639=1)) is type(get_lang('bg'))
assert type(get_lang('ja', iso639=1)) is type(get_lang('ja'))
for i in xrange(len(langs)):
lang = langs.pop(0)
test = lang.replace('.', '_')
test = getattr(__import__('tests.%s' % test), test)
try:
test_case = getattr(test, [x for x in dir(test) \
if x.endswith('TestCase')][0])
test_method = [x for x in dir(test_case) \
if x.startswith('test')][0]
except IndexError:
continue
assert isinstance(test_case.get_examples(test_method), dict)
try:
get_lang('it', iso639=1)
class LanguageCodeTestCase(unittest.TestCase):
table = [('bg', 'bul', 'bul'),
('ca', 'cat', 'cat'),
('cs', 'cze', 'ces'),
('cy', 'wel', 'cym'),
('de', 'ger', 'deu'),
('el', 'gre', 'ell'),
('et', 'est', 'est'),
('fi', 'fin', 'fin'),
(None, 'grc', 'grc'),
(None, None, 'hbs'),
('hu', 'hun', 'hun'),
('ja', 'jpn', 'jpn')]
def test_sub_lang(self):
from hangulize import get_lang
assert get_lang('kat.narrow')
def test_sub_lang(self):
from hangulize import get_lang
assert get_lang('kat.narrow')
for i in range(len(langs)):
lang = langs.pop(0)
test = lang.replace('.', '_')
test = getattr(__import__('tests.%s' % test), test)
try:
test_case = getattr(test, [x for x in dir(test) \
if x.endswith('TestCase')][0])
test_method = [x for x in dir(test_case) \
if x.startswith('test')][0]
except IndexError:
continue
assert isinstance(test_case.get_examples(test_method), dict)
try:
get_lang('it', iso639=1)
class LanguageCodeTestCase(unittest.TestCase):
table = [('bg', 'bul', 'bul'), ('ca', 'cat', 'cat'),
('cs', 'cze', 'ces'), ('cy', 'wel', 'cym'),
('de', 'ger', 'deu'), ('el', 'gre', 'ell'),
('et', 'est', 'est'), ('fi', 'fin', 'fin'),
(None, 'grc', 'grc'), (None, None, 'hbs'),
('hu', 'hun', 'hun'), ('ja', 'jpn', 'jpn')]
def test_regard_iso639_1(self):
assert type(get_lang('bg', iso639=1)) is type(get_lang('bg'))
assert type(get_lang('ja', iso639=1)) is type(get_lang('ja'))
def test_iso639_1(self):
def main(argv):
args = cli.parse_args()
lang = hangulize.get_lang(args.lang)
try:
locale = babel.Locale(lang.iso639_1)
except babel.core.UnknownLocaleError:
locale = None
print('failed to find locale for lang (%s, %s, %s)'
'' % (lang.iso639_1, lang.iso639_2, lang.iso639_3),
file=sys.stderr)
# detect normalize
additional_of_normalize_roman = {}
normalize_roman_called = []
def hacked_normalize_roman(string, additional=None):
if additional:
additional_of_normalize_roman.update(additional)
normalize_roman_called.append(1)
normalize_f = lang.normalize.__func__
normalize_f.__globals__['normalize_roman'] = hacked_normalize_roman
normalize_f(lang, '')
normalize = defaultdict(set)
for src, dst in additional_of_normalize_roman.items():
if src == dst:
continue
normalize[dst].add(src)
# detect script
if normalize_roman_called:
script = 'roman'
else:
script = '???'
print('failed to detect script of lang (%s, %s, %s)'
'' % (lang.iso639_1, lang.iso639_2, lang.iso639_3),
file=sys.stderr)
# find vars
vars_ = []
for attr in dir(lang.__class__):
if attr.startswith('_'):
continue
if hasattr(lang.__class__.__bases__[0], attr):
continue
vars_.append(attr)
if lang.vowels:
vars_.append('vowels')
# group rewrite/transcribe
rewrite = []
transcribe = []
for x, rule in enumerate(lang.notation.rules):
pattern = rule[0]
rpattern = rule[1:]
# ZWSP "/" has been changed with "{}".
pattern = pattern.replace('/', '{}')
# some rpattern is 2d tuple redundantly.
if isinstance(rpattern[0], tuple):
rpattern = rpattern[0]
if isinstance(rpattern[0], hangulize.Phoneme):
transcribe.append((pattern, rpattern))
continue
if rpattern[0] is None:
if transcribe:
transcribe.append((pattern, rpattern))
continue
else:
# "/" -> "{}" here too.
rpattern = rpattern[0].replace('/', '{}')
rewrite.append((pattern, rpattern))
# find test
test_modname = args.lang.replace('.', '_')
test_module = getattr(__import__('tests.%s' % test_modname), test_modname)
for attr, val in vars(test_module).items():
if attr.endswith('TestCase') and not attr.startswith('Hangulize'):
break
test_case = val
examples = test_case.get_examples()
# render
sec = Section('lang')
sec.put('id', args.lang)
sec.put('codes', lang.iso639_1, lang.iso639_3)
if locale is None:
sec.put('english', '???')
sec.put('korean', '???')
else:
sec.put('english', locale.get_language_name('en_US'))
sec.put('korean', locale.get_language_name('ko_KR'))
sec.put('script', script)
print(sec.draw('='), end='')
sec = Section('config')
sec.put('author', args.author)
sec.put('stage', 'draft')
print(sec.draw('='), end='')
sec = Section('macros')
if lang.vowels:
sec.put('@', '<vowels>')
print(sec.draw('=', quote_keys=True), end='')
sec = Section('vars')
for var in vars_:
sec.put(var, *getattr(lang, var))
print(sec.draw('=', quote_keys=True), end='')
sec = Section('normalize')
for to, froms in normalize.items():
sec.put(to, *froms)
print(sec.draw('=', quote_keys=True), end='')
sec = Section('rewrite')
for pattern, rpattern in rewrite:
sec.put(pattern, rpattern)
print(sec.draw('->', quote_keys=True), end='')
sec = Section('transcribe')
for pattern, rpattern in transcribe:
sec.put(pattern, rpattern)
print(sec.draw('->', quote_keys=True), end='')
sec = Section('test')
for loanword, hangul in examples.items():
sec.put(loanword, hangul)
print(sec.draw('->', quote_keys=True), end='')
def test_iso639_2(self):
for iso639_1, iso639_2, iso639_3 in self.table:
if not iso639_2:
continue
assert type(get_lang(iso639_3)) is type(get_lang(iso639_2,
iso639=2))
def test_iso639_3(self):
for iso639_1, iso639_2, iso639_3 in self.table:
assert type(get_lang(iso639_3)) is type(get_lang(iso639_3,
iso639=3))
def test_regard_iso639_1(self):
assert type(get_lang('bg', iso639=1)) is type(get_lang('bg'))
assert type(get_lang('ja', iso639=1)) is type(get_lang('ja'))
def __init__(self, word, code=None, iso639=None, lang=None):
hangulize = self._import_hangulize()
self.lang = lang or hangulize.get_lang(code, iso639)
super(Loanword, self).__init__(word)
def test_iso639_2(self):
for iso639_1, iso639_2, iso639_3 in self.table:
if not iso639_2:
continue
assert type(get_lang(iso639_3)) is type(
get_lang(iso639_2, iso639=2))
def __init__(self, word, code=None, iso639=None, lang=None):
hangulize = self._import_hangulize()
self.lang = lang or hangulize.get_lang(code, iso639)
super(Loanword, self).__init__(word)
def test_iso639_3(self):
for iso639_1, iso639_2, iso639_3 in self.table:
assert type(get_lang(iso639_3)) is type(
get_lang(iso639_3, iso639=3))