def generate_new_html_for_words(words):
"""
Go through list of words, and generates the HTML + Ruby characters.
*words* a list of chinese words
"""
words_with_pinyin_html = ""
for word in words:
debug("{0} : {1}".format(word, "".join(words)))
debug('CHARACTER: {0}'.format(word))
if not is_entirely_chinese(word):
pi = (' ' * len(word)).split(' ')
else:
zh = hanzi.to_zhuyin(word)
debug('ZHUYIN: {0}'.format(zh))
pi = trans.zhuyin_to_pinyin(
zh, accented=not numbered_pinyin).split(' ')
debug('PINYIN: {0}'.format(pi))
debug("{0}={1}".format(word, "".join(pi)))
words_with_pinyin_html += html.to_html(word, top=pi, minified=True)
return words_with_pinyin_html
def generate_new_html_for_words(words):
"""
Go through list of words, and generates the HTML + Ruby characters.
*words* a list of chinese words
"""
words_with_pinyin_html = ""
for word in words:
debug("{0} : {1}".format(word, "".join(words)))
debug('CHARACTER: {0}'.format(word))
if not is_entirely_chinese(word):
pi = (' '*len(word)).split(' ')
else:
zh = hanzi.to_zhuyin(word)
debug('ZHUYIN: {0}'.format(zh))
pi = trans.zhuyin_to_pinyin(
zh,
accented=not numbered_pinyin).split(' ')
debug('PINYIN: {0}'.format(pi))
debug("{0}={1}".format(word, "".join(pi)))
words_with_pinyin_html += html.to_html(word, top=pi, minified=True)
return words_with_pinyin_html
text = ""
text_type = "unknown"
current_side, side = _return_correct_side(
x, y, top, left, characters, right, bottom)
if side in PUT_TEXT_PLACES:
text_type = _identify(current_side[char_num])
if side in STACKED_SIDES:
text = _stackify(current_side[char_num])
else:
text = current_side[char_num]
if side in NEW_CHARCTER_PLACES:
char_num += 1
_html_add("<td class=\"{0}\">".format(text_type), 3)
_html_add("<span>{0}</span>".format(text), 4)
_html_add("</td>", 3)
_html_add("</tr>", 2)
_html_add("</tbody>", 1)
_html_add("</table>")
return _line_html
if __name__ == '__main__':
zi = '你好，我叫顏毅。我是加拿大人！'
zh = hanzi.to_zhuyin(zi)
pi = trans.zhuyin_to_pinyin(hanzi.to_zhuyin(zi))
print(to_html(zi, bottom=pi, right=zh))
def test_zhuyin_to_pinyin(self):
self.assertEqual(trans.zhuyin_to_pinyin(self.zhuyin), self.accented_pinyin_spaced.lower())
self.assertEqual(trans.zhuyin_to_pinyin(self.zhuyin, accented=False), self.numbered_pinyin_spaced.lower())
def parse_file(filename, words):
with open(filename) as f:
data = json.load(f)
items_parsed = 0
# Each item in the JSON correspond to one or more entries in the dictionary
# Most items map 1:1 to entries, e.g. "物質" is a single entry
# Some items are 多音字, so they map to multiple entries (e.g. 重 -> zhòng and chóng)
#
# In the vocabulary of the the CSLD, each item may correspond to multiple heteronyms,
# and each heteronym maps to a single entry.
for item in data:
# These do not change no matter the heteronym
trad = item["title"]
simp = HanziConv.toSimplified(trad)
jyut = pinyin_jyutping_sentence.jyutping(trad,
tone_numbers=True,
spaces=True)
freq = zipf_frequency(trad, "zh")
# Some items have multiple pronunciations (one for Taiwan, one for Mainland China)
taiwan_pin = mainland_pin = ""
# Build up a list of definitions for each heteronym
taiwan_defs = []
mainland_defs = []
# Distinguish between heteronyms by their pinyin – if the pinyin of the
# current heteronym does not match the old pinyin, then a new heteronym
# must be created
last_heteronym_pin = ""
last_taiwan_pin = last_mainland_pin = ""
# Go through each heteronym, creating Entry objects for each one
for heteronym in item["heteronyms"]:
if "pinyin" not in heteronym:
logging.debug(
f'Could not find pinyin for heteronym of word {trad} with definitions {heteronym["definitions"]}'
)
continue
# Filter out known bad pinyin
if (trad in KNOWN_INVALID_SYLLABLES and heteronym["pinyin"]
in KNOWN_INVALID_SYLLABLES[trad]):
pins = KNOWN_INVALID_SYLLABLES[trad][heteronym["pinyin"]]
else:
pins = heteronym["pinyin"].split("<br>陸⃝")
# Some weird a's cause dragonmapper to break, so replace them with standard a's.
pins = list(map(lambda x: x.replace("ɑ", "a"), pins))
# Remove dashes in pinyin
pins = list(map(lambda x: x.replace("-", " "), pins))
# Remove commas in pinyin
pins = list(map(lambda x: x.replace(",", ""), pins))
# Remove weird characters
pins = list(map(lambda x: x.replace("陸⃟", ""), pins))
# Dragonmapper cannot handle some erhua
pins = list(
map(lambda x: x.replace("diǎr", "diǎn er"), pins))
pins = list(
map(lambda x: x.replace("biār", "biān er"), pins))
try:
# Converting from pinyin -> zhuyin inserts spaces between characters
# Converting from zhuyin -> pinyin conserves these spaces
pins = [
transcriptions.zhuyin_to_pinyin(
transcriptions.pinyin_to_zhuyin(x),
accented=False) for x in pins
]
for x in pins:
if x.count(" ") >= len(trad):
# This means that there was an extra space inserted somewhere; the pinyin is not valid
raise ValueError(
"Too many spaces in parsed Pinyin!")
except Exception as e:
# Try parsing zhuyin as a backup
pins = heteronym["bopomofo"].split("<br>陸⃝")
# Remove weird spaces in zhuyin
pins = list(map(lambda x: x.replace(" ", " "), pins))
try:
pins = [
transcriptions.zhuyin_to_pinyin(x,
accented=False)
for x in pins
]
except Exception as e:
logging.error(
f"Unable to split up Pinyin for word {trad}: {e}, skipping word..."
)
continue
if len(pins) > 1:
taiwan_pin = pins[0]
mainland_pin = pins[1]
else:
taiwan_pin = mainland_pin = pins[0]
if (last_heteronym_pin != ""
and heteronym["pinyin"] != last_heteronym_pin):
# A new different pinyin means that we are now processing a new heteronym.
# We must create an Entry object for the definitions of the old heteronym
# and add it to the list of entries before processing the new one.
entry = objects.Entry(trad,
simp,
last_taiwan_pin,
jyut,
freq=freq,
defs=taiwan_defs)
words.append(entry)
if last_mainland_pin != last_taiwan_pin:
entry = objects.Entry(
trad,
simp,
last_mainland_pin,
jyut,
freq=freq,
defs=mainland_defs,
)
words.append(entry)
# Reset the definitions list
taiwan_defs = []
mainland_defs = []
for definition in heteronym["definitions"]:
taiwan_label = "臺" if taiwan_pin != mainland_pin else ""
mainland_label = "陸" if mainland_pin != taiwan_pin else ""
definition_text = definition["def"]
# Take out parts of definitions that should be in labels
for pattern in LABEL_REGEX_PATTERNS:
if re.match(pattern, definition_text):
definition_label, definition_text = re.match(
pattern, definition_text).group(1, 2)
taiwan_label += ("、" +
definition_label if taiwan_label
else definition_label)
mainland_label += ("、" + definition_label
if mainland_label else
definition_label)
# Remove 臺⃝ and 陸⃝ from definitions, since Qt cannot display them
definition_text = definition_text.replace("臺⃝", "臺：")
definition_text = definition_text.replace("陸⃝", "陸：")
# Insert zero-width spaces so that we can reverse-search the definition
taiwan_def_tuple = objects.DefinitionTuple(
"".join(jieba.cut(definition_text)), taiwan_label, [])
mainland_def_tuple = objects.DefinitionTuple(
"".join(jieba.cut(definition_text)), mainland_label,
[])
# Parse and add examples to this definition
if "example" in definition:
for example in definition["example"]:
if re.match(EXAMPLE_REGEX_PATTERN, example):
# Every example is surrounded by "如：<example>", so only keep the example
example = re.match(EXAMPLE_REGEX_PATTERN,
example).group(1)
# Some examples contain multiple examples, so split them up by enclosing brackets 「」
example_texts = re.findall(
INDIVIDUAL_EXAMPLE_REGEX_PATTERN, example)
else:
logging.warning(
f"Found example that does not fit the normal example regex pattern: {trad}, {example}"
)
# Fall back to splitting on Chinese enumeration comma
example_texts = example.split("、")
for example_text in example_texts:
# Strip out weird whitespace
example_text = re.sub(WHITESPACE_REGEX_PATTERN,
"", example_text)
# Joining and splitting separates series of full-width punctuation marks
# into separate items, which is necessary so that lazy_pinyin() returns
# separate items for each full-width punctuation mark in the list it returns
#
# e.g. "《儒林外史．第四六回》：「成老爹道..." turns into
# "《 儒 林 外 史 ． 第 四 六 回 》 ： 「 成 老 爹 道", which turns into
# ['《', '儒', '林', '外', '史', '．', '第', '四', '六', '回', '》', '：', '「', '成', '老', '爹', '道']
# (Notice how "》：「"" is now split up into three different items)
example_pinyin = lazy_pinyin(
" ".join(example_text).split(),
style=Style.TONE3,
neutral_tone_with_five=True,
)
example_pinyin = " ".join(
example_pinyin).lower()
example_pinyin = example_pinyin.strip(
).replace("v", "u:")
# Since the pinyin returned by lazy_pinyin doesn't always match the pinyin
# given in the heteronym, attempt to replace pinyin corresponding to the
# characters in this heteronym with the pinyin provided by the JSON file.
#
# e.g. example_text = "重新"; example_pinyin = "zhong4 xin1" (returned by lazy_pinyin)
# trad = "重", phrase_pinyin = "chong2"
# means that we should convert "zhong4 xin1" to "chong2 xin1"
# Strip out variant pronunciations for conversion purposes
for index, pin in enumerate(
[taiwan_pin, mainland_pin]):
phrase_pinyin = pin
phrase_pinyin = re.sub(
VARIANT_PRONUNCIATION_REGEX_PATTERN,
"",
phrase_pinyin,
)
phrase_pinyin = re.sub(
COLLOQUIAL_PRONUNCIATION_REGEX_PATTERN,
"",
phrase_pinyin,
)
# Do not try to match entries formatted like "那搭（Namibia)"
if not re.match(
STRANGE_ENTRY_REGEX_PATTERN, trad):
try:
example_pinyin = (
change_pinyin_to_match_phrase(
example_text,
example_pinyin,
trad,
phrase_pinyin,
))
except Exception as e:
logging.warning(
f"Couldn't change pinyin in example for word {trad}: "
f"{''.join(example_text)}, {example_pinyin}, {pin}, "
f"{e}")
traceback.print_exc()
if index == 0:
taiwan_def_tuple.examples.append(
objects.ExampleTuple(
"zho", example_pinyin,
example_text))
elif index == 1:
mainland_def_tuple.examples.append(
objects.ExampleTuple(
"zho", example_pinyin,
example_text))
taiwan_defs.append(taiwan_def_tuple)
mainland_defs.append(mainland_def_tuple)
last_heteronym_pin = heteronym["pinyin"]
last_taiwan_pin = taiwan_pin
last_mainland_pin = mainland_pin
entry = objects.Entry(trad,
simp,
taiwan_pin,
jyut,
freq=freq,
defs=taiwan_defs)
words.append(entry)
if mainland_pin != taiwan_pin:
entry = objects.Entry(trad,
simp,
mainland_pin,
jyut,
freq=freq,
defs=mainland_defs)
words.append(entry)
items_parsed += 1
if not items_parsed % 500:
print(f"Parsed entry #{items_parsed}")
def parse_same_meaning_file(filename, words):
for line in read_csv(filename):
if len(line) != 17 or line[0] == "總分類":
continue
terms = defaultdict(set)
for index in (4, 5, 6):
if line[index]:
terms["臺"].add(line[index])
for index in (7, 8, 9):
if line[index]:
terms["陸"].add(line[index])
for index in (10, 11, 12):
if line[index]:
terms["香"].add(line[index])
for index in (13, 14, 15):
if line[index]:
terms["澳"].add(line[index])
explanation = None
if line[16]:
explanation = objects.DefinitionTuple(
"".join(jieba.cut(line[16])), "差異說明", [])
for location in terms:
for term in terms[location]:
trad = term
simp = HanziConv.toSimplified(trad)
if term == line[4] and line[2]:
# Use the provided pinyin, which always corresponds at least to the first Taiwan term
pin = transcriptions.zhuyin_to_pinyin(line[2].replace(
" ", " "),
accented=False)
else:
pin = lazy_pinyin(
trad,
style=Style.TONE3,
neutral_tone_with_five=True,
)
pin = " ".join(pin).lower()
pin = pin.strip().replace("v", "u:")
jyut = pinyin_jyutping_sentence.jyutping(trad,
tone_numbers=True,
spaces=True)
freq = zipf_frequency(trad, "zh")
defs = terms.keys()
defs = map(
lambda x: objects.DefinitionTuple("、".join(terms[x]), line[
1] + "：" + x, []),
defs,
)
defs = list(defs)
if explanation:
defs.append(explanation)
entry = objects.Entry(trad,
simp,
pin,
jyut,
freq=freq,
defs=defs)
words.add(entry)
hangul_romanize_transliter = Transliter(academic)
romanizers: Mapping[Locale, Callable[[str], Markup]] = {
Locale.parse('ja'):
lambda t: Markup(to_roma(t.replace(' ', ''))),
Locale.parse('ko'):
lambda t: Markup(hangul_romanize_transliter.translit(t.replace(' ', ''))),
Locale.parse('zh_CN'):
lambda t: Markup(to_pinyin(t).replace(' ', '')),
Locale.parse('zh_HK'):
lambda t: Markup(
re.sub(
r'(\d) ?', r'<sup>\1</sup>', t if re.match(r'^[A-Za-z0-9 ]+$', t)
else pinyin_jyutping_sentence.jyutping(t, True, True))),
Locale.parse('zh_TW'):
lambda t: Markup(zhuyin_to_pinyin(to_zhuyin(t)).replace(' ', '')),
}
def romanize(term: str, locale: Locale) -> Markup:
global romanizers
try:
f = romanizers[locale]
except KeyError:
return Markup(term.replace(' ', ''))
return f(term)
class Word(Sequence[Term]):
def __init__(self, id: str, locale: Locale, terms: Iterable[Term]):
self.id = id
else:
print(err)
usage()
htmltext = ""
zh_phons = ""
pi_phons = ""
pi_phons_ac = ""
phonperchar = []
hanzichars = []
phonpairs = []
zh_phons = hanzi.to_zhuyin(hanziin).split(" ")
pi_phons = transcriptions.zhuyin_to_pinyin(hanzi.to_zhuyin(hanziin), accented=False).split(" ")
pi_phons_ac = transcriptions.zhuyin_to_pinyin(hanzi.to_zhuyin(hanziin), accented=True).split(" ")
if remove_chars:
for i in range(0, len(zh_phons)):
try:
remove = known_chars[i]
verbose_print(remove)
if remove == "2":
if zhuyin:
zh_phons[i] = ""
if pinyin:
pi_phons[i] = ""
elif remove == "1":
if zhuyin:
zh_phons[i] = phonetics_to_tone_marks(zh_phons[i], accented)
def test_zhuyin_to_pinyin(self):
self.assertEqual(trans.zhuyin_to_pinyin(self.zhuyin),
self.accented_pinyin_spaced.lower())
self.assertEqual(trans.zhuyin_to_pinyin(self.zhuyin, accented=False),
self.numbered_pinyin_spaced.lower())