def lookup(word, external_cmd=True, cmd=DEFAULT_CMD, dict_name=DEFAULT_DICT_NAME):
if external_cmd:
if os.path.isfile(cmd) and os.access(cmd, os.X_OK):
if dict_name == 'oxford':
dict_path = '/Library/Dictionaries/Simplified Chinese - English.dictionary'
elif dict_name == 'landau':
dict_file = 'langdao-ec-gb.dictionary' if is_english(word) else 'langdao-ce-gb.dictionary'
dict_path = os.path.expanduser('~/Library/Dictionaries/{}'.format(dict_file))
else:
raise DictLookupError('dict name not valid.')
proc = subprocess.Popen([cmd, dict_path, word], stdout=subprocess.PIPE)
definition = proc.stdout.read()
if definition == '(null)\n':
return []
definition = definition.decode('utf-8')
else:
raise DictLookupError('file {} not found or not executable.'.format(cmd))
else:
from DictionaryServices import DCSCopyTextDefinition
unicode_word = word.decode('utf-8')
word_range = (0, len(unicode_word))
definition = DCSCopyTextDefinition(None, unicode_word, word_range)
if definition is None:
return []
result = []
if dict_name == 'oxford':
is_eng = is_english(word)
number = u'①-⑳㉑-㉟㊱-㊿'
chinese = ur'\u4e00-\u9fa5'
pinyin = u'āáǎàēéěèōóǒòīíǐìūúǔùüǘǚǜńň'
phrase = r"a-zA-Z,\. "
sentence = ur"0-9a-zA-Z'‘’«»£\$/\?!,\.\[\]\(\) "
pinyin_all = u"a-zA-Z{}'… ".format(pinyin)
sentence_full = ur'([{1}][{0}]*[{1}]|\([{0}]*[{1}]|[{1}][{0}]*\)) ?[{2}]+'.format(
sentence, sentence.replace(r'\(\) ', ''), chinese)
part_map = {
'noun': 'n.',
'intransitive verb': 'vi.',
'transitive verb': 'vt.',
'adjective': 'adj.',
'adverb': 'adv.',
'determiner': 'det.',
'pronoun': 'pron.',
'preposition': 'prep.',
'conjunction': 'conj.',
'exclamation': 'excl.',
'abbreviation': 'abbr.',
'noun plural': 'pl.',
'modifier': 'mod.'
} if is_eng else {
u'名词': u'n.',
u'动词': u'v.',
u'形容词': u'adj.',
u'副词': u'adv.',
u'数词': u'num.',
u'代词': u'pron.',
u'介词': u'prep.',
u'连词': u'conj.',
u'叹词': u'excl.'
}
ignore_list = [
'Countable and uncountable', 'Uncountable and countable',
'Countable', 'Uncountable', 'British', 'American',
'colloquial', 'euphemistic', 'dated', 'Linguistics'
] if is_eng else [
u'方言', u'客套话', u'委婉语', u'书面语', u'俗语', u'比喻义', u'口语', u'惯用语'
]
phrase_mode = False
if is_eng:
word_escaped = re.escape(word)
if not re.match(word_escaped + '(?= )', definition, re.I):
verb_escaped = re.escape(word.split(' ')[0])
if not re.match(verb_escaped + '(?= )', definition, re.I):
return result
phrase_mode = True
pos = definition.find('PHRASAL VERB')
if phrase_mode:
if pos == -1:
return result
definition = definition[pos:]
match = re.search(r'(({0}:? )([A-Z]\. )?({1}).*?)(?=\b{2} [{3}]*?:? ([A-Z]\. )?({1}))'.format(
word_escaped, '|'.join(part_map.keys()), verb_escaped, phrase), definition)
if match is None:
return result
definition = match.group(1)
start_pos = len(match.group(2))
else:
if pos != -1:
definition = definition[:pos]
if phrase_mode:
result.append(word)
else:
trimmed_len = 0
single_phonetic = True
if is_eng:
phonetics = []
for match in re.finditer(r'[A-Z]\. \|(.*?)\| ?', definition):
phonetic = match.group(1).encode('utf-8').strip()
phonetic = '/{}/'.format(phonetic)
if phonetic not in phonetics:
phonetics.append(phonetic)
start = match.start() + 3 - trimmed_len
end = match.end() - trimmed_len
definition = definition[:start] + definition[end:]
trimmed_len += end - start
if len(phonetics) > 0:
phonetics = ', '.join(phonetics)
result.append('{} {}'.format(word, phonetics))
single_phonetic = False
if single_phonetic:
match = re.search(r'\|(.*?)\| ?'
if is_eng else
ur'([^ ]*[{}][^ ]*) ?'.format(pinyin),
definition)
if match is None:
return result
phonetic = match.group(1).encode('utf-8').strip()
result.append('{}{}'.format(word, ' /{}/'.format(phonetic) if phonetic else ''))
start_pos = match.span()[1]
part_list = []
pattern = (r'({}) ?(\(.*?\))? ?'.format('|'.join(part_map.keys()))
if is_eng else
ur'({}) '.format('|'.join(part_map.keys())))
if 'A. ' not in definition:
match = re.match(pattern, definition[start_pos:])
if match:
part_list.append((start_pos, start_pos + match.span()[1], part_map[match.group(1)]))
else:
for match in re.finditer(ur'[A-Z]\. {}'.format(pattern), definition):
part_list.append((match.start(), match.end(), part_map[match.group(1)]))
last_start_pos = len(definition)
pattern = (ur"([^{4}]*?([{0}][{1}]*? |[{2}]*?(\) |›)))(?=({3}|[{4}]|$))".format(pinyin, pinyin_all, phrase, sentence_full, number)
if is_eng else
ur"(?![a-z] )([^{2}]*?[{0}]* )(?=([→{1}{2}]|$))".format(phrase, chinese, number))
for part in reversed(part_list):
entry_list = []
text = definition[part[1]:last_start_pos]
if u'① ' not in text:
match = re.match(pattern, text)
if match:
entry_list.append(match.group(1))
else:
for match in re.finditer(ur'[{}] {}'.format(number, pattern), text):
entry_list.append(match.group(1))
pos = 1
for entry in entry_list:
entry = re.sub(ur'[{0}]*[{1}][{0}]*'.format(pinyin_all, pinyin)
if is_eng else
r'\[used .*?\]',
'', entry)
entry = re.sub(ur'({})'.format('|'.join(ignore_list)), '', entry)
entry = re.sub(r'\([ /]*\)', '', entry)
entry = re.sub(r' {2,}', ' ', entry).strip()
if is_eng:
entry = entry.replace(u' ;', u';')
entry = (u'{} {}'.format(part[2], entry)).encode('utf-8')
result.insert(pos, entry)
pos += 1
last_start_pos = part[0]
def lookup(word, external_cmd=True, parse_html=True, *args):
if external_cmd:
cmd = '{}/{}'.format(os.path.dirname(os.path.realpath(__file__)), 'systemdict')
if os.path.isfile(cmd) and os.access(cmd, os.X_OK):
dict_name = '牛津英汉汉英词典'
proc = subprocess.Popen([cmd, '-t', 'html' if parse_html else 'text', '-d', dict_name, word],
stdout=subprocess.PIPE, stderr=subprocess.PIPE)
definition = proc.stdout.read()
if definition.strip() == '':
return []
if not parse_html:
definition = definition.decode('utf-8')
else:
raise DictLookupError('file {} not found or not executable.'.format(cmd))
else:
from DictionaryServices import DCSCopyTextDefinition
unicode_word = word.decode('utf-8')
word_range = (0, len(unicode_word))
definition = DCSCopyTextDefinition(None, unicode_word, word_range)
if definition is None:
return []
result = []
is_eng = is_english(word)
mac_ver = StrictVersion(platform.mac_ver()[0])
part_map = {
'noun': 'n.',
'intransitive verb': 'vi.',
'transitive verb': 'vt.',
'reflexive verb': 'vr.',
'adjective': 'adj.',
'adverb': 'adv.',
'determiner': 'det.',
'pronoun': 'pron.',
'preposition': 'prep.',
'conjunction': 'conj.',
'exclamation': 'excl.',
'abbreviation': 'abbr.',
'noun plural': 'pl.',
'modifier': 'mod.'
} if is_eng else {
u'名' if mac_ver >= StrictVersion('10.10') else u'名词': u'n.',
u'动' if mac_ver >= StrictVersion('10.10') else u'动词': u'v.',
u'形' if mac_ver >= StrictVersion('10.10') else u'形容词': u'adj.',
u'副' if mac_ver >= StrictVersion('10.10') else u'副词': u'adv.',
u'数' if mac_ver >= StrictVersion('10.10') else u'数词': u'num.',
u'代' if mac_ver >= StrictVersion('10.10') else u'代词': u'pron.',
u'介' if mac_ver >= StrictVersion('10.10') else u'介词': u'prep.',
u'连' if mac_ver >= StrictVersion('10.10') else u'连词': u'conj.',
u'叹' if mac_ver >= StrictVersion('10.10') else u'叹词': u'excl.'
}
if external_cmd and parse_html:
# use ElementTree to parse html
ns = '{http://www.apple.com/DTDs/DictionaryService-1.0.rng}'
phonetic_spans = []
for xml in definition.split('<?xml')[1:]:
html = ET.fromstring('<?xml' + xml)
entry = html.find('.//{}entry'.format(ns))
word_span = entry.find("./span[@class='hwg']/span[@{}dhw]".format(ns))
if word_span is None:
continue
if word_span.text.encode('utf-8').lower() == word.lower():
root = entry
phonetic_span = entry.find("./span[@class='hwg']/span[@class='pr']".format(ns))
if phonetic_span is not None:
phonetic_spans.append(phonetic_span)
else:
if not is_eng:
continue
phrase_span = entry.find("./span[@class='pvb']//span[@class='pvg'][span='{}']/..".format(word))
if phrase_span is None:
continue
root = phrase_span
for span1 in root.findall("./span[@lexid]"):
phonetic_span = span1.find("./span[@class='pr']")
if phonetic_span is not None:
phonetic_spans.append(phonetic_span)
part_span = span1.find("./span[@class='ps']")
if part_span is None:
continue
part = part_map.get(part_span.text, part_span.text + '.')
for span2 in span1.findall("./span[@lexid]"):
item = ''
for span3 in span2.findall('./span[@class]'):
if span3.attrib['class'] == 'ind':
item += ''.join(span3.itertext())
elif span3.attrib['class'] == 'trg':
for span4 in span3.findall('./span[@class]'):
if span4.attrib['class'] != 'trans ty_pinyin':
item += ''.join(span4.itertext())
item = re.sub(r' {2,}', ' ', item).strip()
if item:
item = u'{} {}'.format(part, item).encode('utf-8')
result.append(item)
phonetics = []
for phonetic_span in phonetic_spans:
phonetic = ''.join(phonetic_span.itertext())
phonetic = re.sub(r' {2,}', ' ', phonetic).encode('utf-8').strip(' |')
phonetic = '/{}/'.format(phonetic)
if phonetic not in phonetics:
phonetics.append(phonetic)
phonetics = ', '.join(phonetics)
if phonetics or len(result) > 0:
result.insert(0, '{} {}'.format(word, phonetics))
else:
# use regular expression to parse text
number = u'①-⑳㉑-㉟㊱-㊿'
chinese = ur'\u4e00-\u9fa5'
pinyin = u'āáǎàēéěèōóǒòīíǐìūúǔùüǘǚǜńň'
phrase = r"a-zA-Z,\. "
sentence = ur"0-9a-zA-Z'‘’«»£\$/\?!,\.\[\]\(\) "
pinyin_all = u"a-zA-Z{}'… ".format(pinyin)
sentence_full = ur'([{1}][{0}]*[{1}]|\([{0}]*[{1}]|[{1}][{0}]*\)) ?[{2}]+'.format(
sentence, sentence.replace(r'\(\) ', ''), chinese)
ignore_list = [
'Countable and uncountable', 'Uncountable and countable',
'Countable', 'Uncountable', 'British', 'American',
'colloquial', 'euphemistic', 'dated', 'Linguistics'
] if is_eng else [
u'方言', u'客套话', u'委婉语', u'书面语', u'俗语', u'比喻义',
u'口语', u'惯用语', u'旧词', u'敬辞'
]
phrase_mode = False
if is_eng:
word_escaped = re.escape(word)
if not re.match(word_escaped + '(?= )', definition, re.I):
verb_escaped = re.escape(word.split(' ')[0])
if not re.match(verb_escaped + '(?= )', definition, re.I):
return result
phrase_mode = True
pos = definition.find('PHRASAL VERB')
if phrase_mode:
if pos == -1:
return result
definition = definition[pos:]
match = re.search(r'(({0}:? )([A-Z]\. )?({1}).*?)(?=\b{2} [{3}]*?:? ([A-Z]\. )?({1}))'.format(
word_escaped, '|'.join(part_map.keys()), verb_escaped, phrase), definition)
if match is None:
return result
definition = match.group(1)
start_pos = len(match.group(2))
else:
if pos != -1:
definition = definition[:pos]
if phrase_mode:
result.append(word)
else:
trimmed_len = 0
single_phonetic = True
if is_eng:
phonetics = []
for match in re.finditer(r'[A-Z]\. \|(.*?)\| ?', definition):
phonetic = match.group(1).encode('utf-8').strip()
phonetic = '/{}/'.format(phonetic)
if phonetic not in phonetics:
phonetics.append(phonetic)
start = match.start() + 3 - trimmed_len
end = match.end() - trimmed_len
definition = definition[:start] + definition[end:]
trimmed_len += end - start
if len(phonetics) > 0:
phonetics = ', '.join(phonetics)
result.append('{} {}'.format(word, phonetics))
single_phonetic = False
if single_phonetic:
match = re.search(r'\|(.*?)\| ?'
if is_eng else
ur'([^ ]*[{}][^ ]*) ?'.format(pinyin),
definition)
if match is None:
return result
phonetic = match.group(1).encode('utf-8').strip()
result.append('{}{}'.format(word, ' /{}/'.format(phonetic) if phonetic else ''))
start_pos = match.span()[1]
part_list = []
pattern = (r'({}) ?(\(.*?\))? ?'.format('|'.join(part_map.keys()))
if is_eng else
ur'({}) '.format('|'.join(part_map.keys())))
if 'A. ' not in definition:
match = re.match(pattern, definition[start_pos:])
if match:
part_list.append((start_pos, start_pos + match.span()[1], part_map[match.group(1)]))
else:
for match in re.finditer(ur'[A-Z]\. {}'.format(pattern), definition):
part_list.append((match.start(), match.end(), part_map[match.group(1)]))
last_start_pos = len(definition)
pattern = (ur"([^{4}]*?([{0}][{1}]*? |[{2}]*?(\) |›)))(?=({3}|[{4}]|$))".format(pinyin, pinyin_all, phrase, sentence_full, number)
if is_eng else
ur"(?![a-z] )([^{2}]*?[{0}]* )(?=([→{1}{2}]|$))".format(phrase, chinese, number))
for part in reversed(part_list):
entry_list = []
text = definition[part[1]:last_start_pos]
if u'① ' not in text:
match = re.match(pattern, text)
if match:
entry_list.append(match.group(1))
else:
for match in re.finditer(ur'[{}] {}'.format(number, pattern), text):
entry_list.append(match.group(1))
pos = 1
for entry in entry_list:
entry = re.sub(ur'[{0}]*[{1}][{0}]*'.format(pinyin_all, pinyin)
if is_eng else
r'\[used .*?\]',
'', entry)
entry = re.sub(ur'({})'.format('|'.join(ignore_list)), '', entry)
entry = re.sub(r'\([ /]*\)', '', entry)
entry = re.sub(r' {2,}', ' ', entry).strip()
if is_eng:
entry = entry.replace(u' ;', u';')
entry = (u'{} {}'.format(part[2], entry)).encode('utf-8')
result.insert(pos, entry)
pos += 1
last_start_pos = part[0]
def lookup(word, external_cmd=True, parse_html=True, *args):
mac_ver = StrictVersion(platform.mac_ver()[0])
if external_cmd:
cmd = '{}/{}'.format(os.path.dirname(os.path.realpath(__file__)),
'systemdict')
if os.path.isfile(cmd) and os.access(cmd, os.X_OK):
dict_name = '牛津英汉汉英词典' if mac_ver >= StrictVersion(
'10.10') else 'Oxford Chinese Dictionary'
proc = subprocess.Popen([
cmd, '-t', 'html' if parse_html else 'text', '-d', dict_name,
word
],
stdout=subprocess.PIPE,
stderr=subprocess.PIPE)
definition = proc.stdout.read()
if definition.strip() == '':
return []
if not parse_html:
definition = definition.decode('utf-8')
else:
raise DictLookupError(
'file {} not found or not executable.'.format(cmd))
else:
from DictionaryServices import DCSCopyTextDefinition
unicode_word = word.decode('utf-8')
word_range = (0, len(unicode_word))
definition = DCSCopyTextDefinition(None, unicode_word, word_range)
if definition is None:
return []
result = []
is_eng = is_english(word)
part_map = {
'noun': 'n.',
'intransitive verb': 'vi.',
'transitive verb': 'vt.',
'reflexive verb': 'vr.',
'adjective': 'adj.',
'adverb': 'adv.',
'determiner': 'det.',
'pronoun': 'pron.',
'preposition': 'prep.',
'conjunction': 'conj.',
'exclamation': 'excl.',
'abbreviation': 'abbr.',
'noun plural': 'pl.',
'modifier': 'mod.'
} if is_eng else {
u'名' if mac_ver >= StrictVersion('10.10') else u'名词': u'n.',
u'动' if mac_ver >= StrictVersion('10.10') else u'动词': u'v.',
u'形' if mac_ver >= StrictVersion('10.10') else u'形容词': u'adj.',
u'副' if mac_ver >= StrictVersion('10.10') else u'副词': u'adv.',
u'数' if mac_ver >= StrictVersion('10.10') else u'数词': u'num.',
u'代' if mac_ver >= StrictVersion('10.10') else u'代词': u'pron.',
u'介' if mac_ver >= StrictVersion('10.10') else u'介词': u'prep.',
u'连' if mac_ver >= StrictVersion('10.10') else u'连词': u'conj.',
u'叹' if mac_ver >= StrictVersion('10.10') else u'叹词': u'excl.'
}
if external_cmd and parse_html:
# use ElementTree to parse html
ns = '{http://www.apple.com/DTDs/DictionaryService-1.0.rng}'
phonetic_spans = []
for xml in definition.split('<?xml')[1:]:
html = ET.fromstring('<?xml' + xml)
entry = html.find('.//{}entry'.format(ns))
word_span = entry.find(
"./span[@class='hwg']/span[@{}dhw]".format(ns))
if word_span is None:
continue
if word_span.text.encode('utf-8').lower() == word.lower():
root = entry
phonetic_span = entry.find(
"./span[@class='hwg']/span[@class='pr']".format(ns))
if phonetic_span is not None:
phonetic_spans.append(phonetic_span)
else:
if not is_eng:
continue
phrase_span = entry.find(
"./span[@class='pvb']//span[@class='pvg'][span='{}']/..".
format(word))
if phrase_span is None:
continue
root = phrase_span
for span1 in root.findall("./span[@lexid]"):
phonetic_span = span1.find("./span[@class='pr']")
if phonetic_span is not None:
phonetic_spans.append(phonetic_span)
part_span = span1.find("./span[@class='ps']")
if part_span is None:
continue
part = part_map.get(part_span.text, part_span.text + '.')
for span2 in span1.findall("./span[@lexid]"):
item = ''
for span3 in span2.findall('./span[@class]'):
if span3.attrib['class'] == 'ind':
item += ''.join(span3.itertext())
elif span3.attrib['class'] == 'trg':
for span4 in span3.findall('./span[@class]'):
if span4.attrib['class'] != 'trans ty_pinyin':
item += ''.join(span4.itertext())
item = re.sub(r' {2,}', ' ', item).strip()
if item:
item = u'{} {}'.format(part, item).encode('utf-8')
result.append(item)
phonetics = []
for phonetic_span in phonetic_spans:
phonetic = ''.join(phonetic_span.itertext())
phonetic = re.sub(r' {2,}', ' ',
phonetic).encode('utf-8').strip(' |')
phonetic = '/{}/'.format(phonetic)
if phonetic not in phonetics:
phonetics.append(phonetic)
phonetics = ', '.join(phonetics)
if phonetics or len(result) > 0:
result.insert(0, '{} {}'.format(word, phonetics))
else:
# use regular expression to parse text
number = u'①-⑳㉑-㉟㊱-㊿'
chinese = ur'\u4e00-\u9fa5'
pinyin = u'āáǎàēéěèōóǒòīíǐìūúǔùüǘǚǜńň'
phrase = r"a-zA-Z,\. "
sentence = ur"0-9a-zA-Z'‘’«»£\$/\?!,\.\[\]\(\) "
pinyin_all = u"a-zA-Z{}'… ".format(pinyin)
sentence_full = ur'([{1}][{0}]*[{1}]|\([{0}]*[{1}]|[{1}][{0}]*\)) ?[{2}]+'.format(
sentence, sentence.replace(r'\(\) ', ''), chinese)
ignore_list = [
'Countable and uncountable', 'Uncountable and countable',
'Countable', 'Uncountable', 'British', 'American', 'colloquial',
'euphemistic', 'dated', 'Linguistics'
] if is_eng else [
u'方言', u'客套话', u'委婉语', u'书面语', u'俗语', u'比喻义', u'口语', u'惯用语', u'旧词',
u'敬辞'
]
phrase_mode = False
if is_eng:
word_escaped = re.escape(word)
if not re.match(word_escaped + '(?= )', definition, re.I):
verb_escaped = re.escape(word.split(' ')[0])
if not re.match(verb_escaped + '(?= )', definition, re.I):
return result
phrase_mode = True
pos = definition.find('PHRASAL VERB')
if phrase_mode:
if pos == -1:
return result
definition = definition[pos:]
match = re.search(
r'(({0}:? )([A-Z]\. )?({1}).*?)(?=\b{2} [{3}]*?:? ([A-Z]\. )?({1}))'
.format(word_escaped, '|'.join(part_map.keys()),
verb_escaped, phrase), definition)
if match is None:
return result
definition = match.group(1)
start_pos = len(match.group(2))
else:
if pos != -1:
definition = definition[:pos]
if phrase_mode:
result.append(word)
else:
trimmed_len = 0
single_phonetic = True
if is_eng:
phonetics = []
for match in re.finditer(r'[A-Z]\. \|(.*?)\| ?', definition):
phonetic = match.group(1).encode('utf-8').strip()
phonetic = '/{}/'.format(phonetic)
if phonetic not in phonetics:
phonetics.append(phonetic)
start = match.start() + 3 - trimmed_len
end = match.end() - trimmed_len
definition = definition[:start] + definition[end:]
trimmed_len += end - start
if len(phonetics) > 0:
phonetics = ', '.join(phonetics)
result.append('{} {}'.format(word, phonetics))
single_phonetic = False
if single_phonetic:
match = re.search(
r'\|(.*?)\| ?' if is_eng else
ur'([^ ]*[{}][^ ]*) ?'.format(pinyin), definition)
if match is None:
return result
phonetic = match.group(1).encode('utf-8').strip()
result.append('{}{}'.format(
word, ' /{}/'.format(phonetic) if phonetic else ''))
start_pos = match.span()[1]
part_list = []
pattern = (r'({}) ?(\(.*?\))? ?'.format('|'.join(part_map.keys()))
if is_eng else ur'({}) '.format('|'.join(part_map.keys())))
if 'A. ' not in definition:
match = re.match(pattern, definition[start_pos:])
if match:
part_list.append((start_pos, start_pos + match.span()[1],
part_map[match.group(1)]))
else:
for match in re.finditer(ur'[A-Z]\. {}'.format(pattern),
definition):
part_list.append(
(match.start(), match.end(), part_map[match.group(1)]))
last_start_pos = len(definition)
pattern = (
ur"([^{4}]*?([{0}][{1}]*? |[{2}]*?(\) |›)))(?=({3}|[{4}]|$))".
format(pinyin, pinyin_all, phrase, sentence_full, number) if is_eng
else ur"(?![a-z] )([^{2}]*?[{0}]* )(?=([→{1}{2}]|$))".format(
phrase, chinese, number))
for part in reversed(part_list):
entry_list = []
text = definition[part[1]:last_start_pos]
if u'① ' not in text:
match = re.match(pattern, text)
if match:
entry_list.append(match.group(1))
else:
for match in re.finditer(ur'[{}] {}'.format(number, pattern),
text):
entry_list.append(match.group(1))
pos = 1
for entry in entry_list:
entry = re.sub(
ur'[{0}]*[{1}][{0}]*'.format(pinyin_all, pinyin)
if is_eng else r'\[used .*?\]', '', entry)
entry = re.sub(ur'({})'.format('|'.join(ignore_list)), '',
entry)
entry = re.sub(r'\([ /]*\)', '', entry)
entry = re.sub(r' {2,}', ' ', entry).strip()
if is_eng:
entry = entry.replace(u' ;', u';')
entry = (u'{} {}'.format(part[2], entry)).encode('utf-8')
result.insert(pos, entry)
pos += 1
last_start_pos = part[0]
def lookup(word, external_cmd=True, cmd=DEFAULT_CMD, dict_name=DEFAULT_DICT_NAME):
if external_cmd:
if os.path.isfile(cmd) and os.access(cmd, os.X_OK):
if dict_name == 'oxford':
dict_path = '/Library/Dictionaries/Simplified Chinese - English.dictionary'
elif dict_name == 'landau':
dict_file = 'langdao-ec-gb.dictionary' if is_english(word) else 'langdao-ce-gb.dictionary'
dict_path = os.path.expanduser('~/Library/Dictionaries/{}'.format(dict_file))
else:
raise DictLookupError('dict name not valid.')
proc = subprocess.Popen([cmd, dict_path, word], stdout=subprocess.PIPE)
definition = proc.stdout.read()
if definition == '(null)\n':
return []
definition = definition.decode('utf-8')
else:
raise DictLookupError('file {} not found or not executable.'.format(cmd))
else:
from DictionaryServices import DCSCopyTextDefinition
unicode_word = word.decode('utf-8')
word_range = (0, len(unicode_word))
definition = DCSCopyTextDefinition(None, unicode_word, word_range)
if definition is None:
return []
result = []
if dict_name == 'oxford':
is_eng = is_english(word)
number = u'①-⑳㉑-㉟㊱-㊿'
chinese = ur'\u4e00-\u9fa5'
pinyin = u'āáǎàēéěèōóǒòīíǐìūúǔùüǘǚǜńň'
phrase = r"a-zA-Z,\. "
sentence = ur"0-9a-zA-Z'‘’«»£\$/\?!,\.\[\]\(\) "
pinyin_all = u"a-z{}'… ".format(pinyin)
sentence_full = ur'([{1}][{0}]*[{1}]|\([{0}]*[{1}]|[{1}][{0}]*\)) ?[{2}]+'.format(
sentence, sentence.replace(r'\(\) ', ''), chinese)
part_map = {
'noun': 'n.',
'intransitive verb': 'vi.',
'transitive verb': 'vt.',
'adjective': 'adj.',
'adverb': 'adv.',
'determiner': 'det.',
'pronoun': 'pron.',
'preposition': 'prep.',
'conjunction': 'conj.',
'exclamation': 'excl.',
'abbreviation': 'abbr.',
'plural noun': 'pl.',
'modifier': 'mod.'
} if is_eng else {
u'名词': u'n.',
u'动词': u'v.',
u'形容词': u'adj.',
u'副词': u'adv.',
u'数词': u'num.',
u'代词': u'pron.',
u'介词': u'prep.',
u'连词': u'conj.',
u'叹词': u'excl.'
}
ignore_list = [
'Uncountable and countable', 'Countable', 'Uncountable', 'British', 'American',
'colloquial', 'euphemistic', 'dated', 'Linguistics'
] if is_eng else [
u'方言', u'客套话', u'委婉语', u'书面语', u'俗语', u'比喻义', u'口语', u'惯用语'
]
phrase_mode = False
if is_eng:
word_escaped = re.escape(word)
if not re.match(word_escaped + '(?= )', definition, re.I):
verb_escaped = re.escape(word.split(' ')[0])
if not re.match(verb_escaped + '(?= )', definition, re.I):
return result
phrase_mode = True
pos = definition.find('PHRASAL VERB')
if phrase_mode:
if pos == -1:
return result
definition = definition[pos:]
match = re.search(r'(({0}:? )([A-Z]\. )?({1}).*?)(?=\b{2} [{3}]*?:? ([A-Z]\. )?({1}))'.format(
word_escaped, '|'.join(part_map.keys()), verb_escaped, phrase), definition)
if match is None:
return result
definition = match.group(1)
start_pos = len(match.group(2))
else:
if pos != -1:
definition = definition[:pos]
if phrase_mode:
result.append(word)
else:
trimmed_len = 0
single_phonetic = True
if is_eng:
phonetics = []
for match in re.finditer(r'[A-Z]\. \|(.*?)\| ?', definition):
phonetic = match.group(1).encode('utf-8').strip()
phonetic = '/{}/'.format(phonetic)
if phonetic not in phonetics:
phonetics.append(phonetic)
start = match.start() + 3 - trimmed_len
end = match.end() - trimmed_len
definition = definition[:start] + definition[end:]
trimmed_len += end - start
if len(phonetics) > 0:
phonetics = ', '.join(phonetics)
result.append('{} {}'.format(word, phonetics))
single_phonetic = False
if single_phonetic:
match = re.search(r'\|(.*?)\| ?'
if is_eng else
ur'([^ ]*[{}][^ ]*) ?'.format(pinyin),
definition)
if match is None:
return result
phonetic = match.group(1).encode('utf-8').strip()
result.append('{}{}'.format(word, ' /{}/'.format(phonetic) if phonetic else ''))
start_pos = match.span()[1]
part_list = []
pattern = (r'({}) ?(\(.*?\))? ?'.format('|'.join(part_map.keys()))
if is_eng else
ur'({}) '.format('|'.join(part_map.keys())))
if 'A. ' not in definition:
match = re.match(pattern, definition[start_pos:])
if match:
part_list.append((start_pos, start_pos + match.span()[1], part_map[match.group(1)]))
else:
for match in re.finditer(ur'[A-Z]\. {}'.format(pattern), definition):
part_list.append((match.start(), match.end(), part_map[match.group(1)]))
last_start_pos = len(definition)
pattern = (ur"([^{4}]*?([{0}][{1}]*? |[{2}]*?(\) |›)))(?=({3}|[{4}]|$))".format(pinyin, pinyin_all, phrase, sentence_full, number)
if is_eng else
ur"(?![a-z] )([^{2}]*?[{0}]* )(?=([→{1}{2}]|$))".format(phrase, chinese, number))
for part in reversed(part_list):
entry_list = []
text = definition[part[1]:last_start_pos]
if u'① ' not in text:
match = re.match(pattern, text)
if match:
entry_list.append(match.group(1))
else:
for match in re.finditer(ur'[{}] {}'.format(number, pattern), text):
entry_list.append(match.group(1))
pos = 1
for entry in entry_list:
entry = re.sub(ur'[{0}]*[{1}][{0}]*'.format(pinyin_all, pinyin)
if is_eng else
r'\[used .*?\]',
'', entry)
entry = re.sub(ur'({})'.format('|'.join(ignore_list)), '', entry)
entry = re.sub(r'\([ /]*\)', '', entry)
entry = re.sub(r' {2,}', ' ', entry).strip()
if is_eng:
entry = entry.replace(u' ;', u';')
entry = (u'{} {}'.format(part[2], entry)).encode('utf-8')
result.insert(pos, entry)
pos += 1
last_start_pos = part[0]
