def __init__(self, kanjidic_files=None):
dict.__init__(self)
if kanjidic_files is None:
kanjidic_files = [
cjkdata.get_resource('kanjidic'),
cjkdata.get_resource('kanjd212'),
]
line_stream = reduce(chain, [sopen(f) for f in kanjidic_files])
self._parse_kanjidic(line_stream)
def __init__(self, istream=None):
"""
@param dict_file: The radkfile to parse.
"""
if istream is None:
istream = open(cjkdata.get_resource('radkfile'))
self._parse_radkfile(codecs.getreader('utf8')(istream))
def __init__(self):
self._segmenter = get_pinyin_segmenter()
# Load the table mapping hanzi to pinyin.
f = cjkdata.get_resource('tables/gbk_pinyin_table')
with codecs.open(f, 'r', 'utf8') as istream:
for line in istream:
if line.startswith('#'):
continue
entries = line.rstrip().split()
hanzi = entries[0]
numeric_readings = tuple(entries[1:])
self[hanzi] = tuple(numeric_readings)
def get_list(list_name):
"Returns the kanji in the given list."
f = cjkdata.get_resource('lists/char/%s' % list_name)
with codecs.open(f, 'r', 'utf8') as istream:
return scripts.unique_kanji(istream.read())
def get_lists():
"Returns list containing the names of all existing kanji lists."
return os.listdir(cjkdata.get_resource('lists/char'))
def _default_stream():
return open(cjkdata.get_resource('tables/zhuyin_pinyin_conv_table'))