def wordByWord(self, sentence, hiragana=True):
"""Get reading for every element in provided sentence"""
self.includeSurface().includeReading()
info = self.parse(sentence)
words = []
if info:
# Compile list of {word: reading} excluding okurigana and and so on
for word in info:
# No reading
if not word.get('pronounciation'):
reading = u''
# Word is already in kana
elif (
word.get('pronounciation') == word.get('surface') or
kata2hira(word.get('pronounciation')) == word.get('surface')
):
reading = u''
# Need to convert to hiragana
elif hiragana:
reading = kata2hira(word.get('pronounciation'))
# Otherwise, let it be
else:
reading = word.get('pronounciation')
# Append tuple to word list
words.append((word.get('surface'), reading))
return words
def word_by_word(self, sentence, hiragana=True):
"""Get reading for every element in provided sentence"""
words = []
info = self.include('pronounciation', 'surface').parse(sentence)
if info:
# Compile list of {word: reading} excluding okurigana and and so on
for word in info:
# No reading
if not word.get('pronounciation'):
reading = u''
# Word is already in kana
elif (
word.get('pronounciation') == word.get('surface') or
kata2hira(word.get('pronounciation')) == word.get('surface')
):
reading = u''
# Need to convert to hiragana
elif hiragana:
reading = kata2hira(word.get('pronounciation'))
# Otherwise, let it be
else:
reading = word.get('pronounciation')
# Append tuple to word list
words.append((word.get('surface'), reading))
return words
def parseReadings(self):
items_grouped = {}
for kanji in self.items:
readings = {}
try:
lookup = self.kjd[kanji.character]
for kun in lookup.kun_readings:
kun = kun.replace('.', '').replace('-', '')
for word in kanji.word:
if kun in kata2hira(MecabTool.parseToReadingsKana(word.word)[0]):
if readings.has_key(kun):
readings[kun].append(word.word)
else:
readings[kun] = [word.word]
for on in lookup.on_readings:
on = kata2hira(on.replace('.', '').replace('-', ''))
for word in kanji.word:
if on in kata2hira(MecabTool.parseToReadingsKana(word.word)[0]):
if readings.has_key(on):
readings[on].append(word.word)
else:
readings[on] = [word.word]
except Exception, e:
log.error(e)
# simple solution - difficult implementation
# items_grouped[kanji.character] = readings
# slightly more complicated solution - easier implementation
for reading in readings:
i = 0
if items_grouped.has_key(kanji.character): items_grouped[kanji.character + '_' + str(i)] = (reading, readings[reading]); i += 1
else: items_grouped[kanji.character] = (reading, readings[reading])
def kana_minus_dakuten(char):
if is_katakana(char):
hira = kata2hira(char)
hira = __by_dakuten.get(hira, hira)
return hira2kata(hira)
else:
return __by_dakuten.get(char, char)
def kana_minus_dakuten(char):
if is_katakana(char):
hira = kata2hira(char)
hira = __by_dakuten.get(hira, hira)
return hira2kata(hira)
else:
return __by_dakuten.get(char, char)
def getWordPronunciationFromExample(self, item):
words = MecabTool.parseToWordsFull(self.currentExample.sentence)
answer = self.find(lambda word: item in word['word'] , words)
try:
return kata2hira(answer['pronunciation'])
except Exception:
return u' '
def getCorrectAnswer(self):
words = MecabTool.parseToWordsFull(self.currentExample.sentence)
answer = self.find(lambda word: self.currentItem.character in word['word'] , words)
try:
return kata2hira(answer['pronunciation'])
except Exception:
return u' '
def getWordPronunciationFromText(query, text):
words = MecabTool.parseToWordsFull(text)
answer = MecabTool.findUsingF(lambda word: query in word['word'] , words)
try:
return kata2hira(answer['pronunciation'])
except Exception:
return None
def getWordPronunciationFromText(query, text):
words = MecabTool.parseToWordsFull(text)
answer = MecabTool.findUsingF(lambda word: query in word['word'],
words)
try:
return kata2hira(answer['pronunciation'])
except Exception:
return None
def _reading(node):
surface = node.surface.decode('utf8')
reading = node.feature.decode('utf8').split(',')[-2]
if reading == '*':
return None
reading = jcconv.kata2hira(reading)
return reading
def _reading(node):
surface = node.surface.decode('utf8')
reading = node.feature.decode('utf8').split(',')[-2]
if reading == '*':
return None
reading = jcconv.kata2hira(reading)
return reading
def kana_plus_mini(char):
yield char
is_kata = is_katakana(char)
if is_kata:
char = kata2hira(char)
for char in __to_mini.get(char, ''):
yield hira2kata(char) if is_kata else char
def kana_plus_mini(char):
yield char
is_kata = is_katakana(char)
if is_kata:
char = kata2hira(char)
for char in __to_mini.get(char, ''):
yield hira2kata(char) if is_kata else char
def create_idx_file(datfile, idxfile):
print "Reading in dat file..."
words = OrderedDict()
position = 0
with codecs.open(datfile, 'r', "utf-8") as f:
for line in f:
# TODO create key for each ; separated value
temp = line.split(' ', 1)[0]
temp2 = line.split(';', 1)[0]
if len(temp) < len(temp2):
symbol = temp
else:
symbol = temp2
# vu wouldn't be converted if not reserved, but for clarity purposes ...
symbol = jcconv.kata2hira(symbol, 'ヴ')
if not words.get(symbol):
words.update({symbol: str(position)})
else:
words.update({symbol: words.get(symbol) + "," + str(position)})
symbol_in_brackets = re.search(r'\[(.*?)\]', line.split('/', 1)[0])
if symbol_in_brackets:
symbol_in_brackets = symbol_in_brackets.group(1)
# vu wouldn't be converted if not reserved, but for clarity purposes ...
symbol_in_brackets = jcconv.kata2hira(symbol_in_brackets, 'ヴ')
if words.get(symbol_in_brackets):
words.update({
symbol_in_brackets:
words.get(symbol_in_brackets) + "," + str(position)
})
else:
words.update({symbol_in_brackets: str(position)})
position += len(line)
print "Finished reading in dat file, now sorting index..."
words = OrderedDict(sorted(words.items(), key=lambda t: t[0]))
print "Finished sorting index, now writing idx file..."
with codecs.open(idxfile, 'w+', "utf-8") as f:
for key, value in words.iteritems():
f.write(key + "," + value + "\n")
print "Finished writing idx file"
def create_idx_file(datfile, idxfile):
print "Reading in dat file..."
words = OrderedDict()
position = 0
with codecs.open(datfile, 'r', "utf-8") as f:
for line in f:
# TODO create key for each ; separated value
temp = line.split(' ', 1)[0]
temp2 = line.split(';', 1)[0]
if len(temp) < len(temp2):
symbol = temp
else:
symbol = temp2
# vu wouldn't be converted if not reserved, but for clarity purposes ...
symbol = jcconv.kata2hira(symbol, 'ヴ')
if not words.get(symbol):
words.update({symbol : str(position)})
else:
words.update({symbol : words.get(symbol) + "," + str(position)})
symbol_in_brackets = re.search(r'\[(.*?)\]', line.split('/', 1)[0])
if symbol_in_brackets:
symbol_in_brackets = symbol_in_brackets.group(1)
# vu wouldn't be converted if not reserved, but for clarity purposes ...
symbol_in_brackets = jcconv.kata2hira(symbol_in_brackets, 'ヴ')
if words.get(symbol_in_brackets):
words.update({symbol_in_brackets : words.get(symbol_in_brackets) + "," + str(position)})
else:
words.update({symbol_in_brackets : str(position)})
position += len(line)
print "Finished reading in dat file, now sorting index..."
words = OrderedDict(sorted(words.items(), key=lambda t: t[0]))
print "Finished sorting index, now writing idx file..."
with codecs.open(idxfile, 'w+', "utf-8") as f:
for key, value in words.iteritems():
f.write(key + "," + value + "\n")
print "Finished writing idx file"
def __getattribute__(self, key):
feature_keys = dict.__getattribute__(self, 'feature_keys')
if key in self:
return self[key]
if key in feature_keys:
features = dict.__getattribute__(self, 'feature').split(',')
for name, index in feature_keys.iteritems():
self[name] = features[index]
readings = dict.__getattribute__(self, 'readings')
readings = set([jcconv.kata2hira(e) for e in readings if e not in ('*',)])
self['readings'] = readings
return dict.__getattribute__(self, key)
def all_to_hiragana(string):
out = u''
for index, char in enumerate(string):
if char == u'ー' or char == u'｜':
char = char_to_base_vowel(out[-1])
char = kata2hira(char)
out += char
return out
def reading(self, sentence, hiragana=True):
"""Get reading for provided sentence|word"""
self.includeReading()
info = self.parse(sentence)
if info:
kana = u''.join([
reading.get('pronounciation', '') for reading in info
if reading.get('pronounciation')
])
if hiragana:
return kata2hira(kana)
return kana
def all_to_hiragana(string):
out = u''
for index, char in enumerate(string):
if char == u'ー' or char == u'｜':
char = char_to_base_vowel(out[-1])
char = kata2hira(char)
out += char
return out
def reading(self, sentence, hiragana=True):
"""Get reading for provided sentence|word"""
self.includeReading()
info = self.parse(sentence)
if info:
kana = u''.join([
reading.get('pronounciation', '') for reading in info
if reading.get('pronounciation')
])
if hiragana:
return kata2hira(kana)
return kana
def generate_reading(expression):
expression = expression.encode(settings.MECAB_ENCODING)
proc = subprocess.Popen('mecab', shell=False, stdin=subprocess.PIPE, stdout=subprocess.PIPE)
mecab_output = proc.communicate(expression)[0].decode(settings.MECAB_ENCODING)
lines = mecab_output.split(u'\n')[:-2] #skip the \nEOS\n
ret = u''
for line in lines:
if line[0] == u',':
ret += u','
continue
elif line[:3] == u'EOS':
ret += u'\n'
continue
elif line[0].strip() == '':
ret += line[0]
continue
fields = line.split(u',')
word = fields[0].split()[0]
if len(fields) == 9:
reading = fields[7]
# Has kanji and a reading?
if (jcconv.kata2hira(reading) != word
and reading != word
and any(_code_page(char) != 'hiragana'
and _code_page(char) != 'katakana'
for char in word)):
# The reading comes in as katakana, we want hiragana.
reading = jcconv.kata2hira(reading)
ret += _furiganaize(word, reading, not ret)
else:
ret += word
else:
ret += word
return ret
def get_reading(self, sentence, hiragana=True):
"""
Get reading for provided sentence|word
NB: for some rare words there may be no readings available!
"""
info = self.include('pronounciation').parse(sentence)
if info:
katakana = u''.join([
reading.get('pronounciation', '') for reading in info
if reading.get('pronounciation')
])
if hiragana:
return kata2hira(katakana)
return katakana
def getKanaReading(query):
MECAPI_URL = u'http://mimitako.net/api/mecapi.cgi?sentence='
OPTIONS = u'&response=pronounciation'
XML_TAG = u'word/pronounciation'
url = MECAPI_URL + query + OPTIONS
result = urllib2.urlopen(url)
tree = ElementTree.fromstring(result.read())
reading = []
for node in tree.findall(XML_TAG):
reading.append(node.text)
return kata2hira(''.join(reading))
def reading(self, sentence, hiragana=True):
"""
Get reading for provided sentence|word
NB: for some rare words there may be no readings available!
"""
self.include('pronounciation')
info = self.parse(sentence)
if info:
kana = u''.join([
reading.get('pronounciation', '') for reading in info
if reading.get('pronounciation')
])
if hiragana:
return kata2hira(kana)
return kana
def as_mecab(self):
parser = BakaMeCab(self.front)
parsed_example = []
for word, info in parser.get_info().iteritems():
reading = u''
if(len(info) > 4):
kana = info[6] if len(info) > 6 else info[4]
hiragana = kata2hira(kana)
if kana != word and hiragana != word and word != 'は':
reading = hiragana
parsed_example.append({'front': word, 'reading': reading})
return {
'parsed': parsed_example,
'original': self.front,
'reading': self.reading,
'gloss': self.gloss
}
def getCurrentSentenceReading(self):
return kata2hira(''.join(MecabTool.parseToReadingsKana(self.currentExample.sentence)))
def getWordPronounciation(self, item):
try:
return kata2hira(MecabTool.parseToWordsFull(item)[0]['pronunciation'])
except:
return item
def _can_furigana(self, ma, expression):
return (ma.reading != expression[ma.position:ma.position+ma.word_length] and
kata2hira(expression[ma.position:ma.position+ma.word_length]) != ma.reading and
expression[ma.position:ma.position+ma.word_length] not in u"一二三四五六七八九十０１２３４５６７８９")
total_freq = 0
for file_name in file_list:
print file_name
with codecs.open(file_name, 'r', 'utf-8') as f:
for line in f:
arr = line.split()
base = 0
for i, e in enumerate(arr):
if e[0] <= '9' and e[0] >= '0':
base = i
kanji = arr[base - 1][1:]
freq = arr[base][:-2]
kana = kata2hira(arr[base + 2][:-1])
print kanji, freq, kana
if len(kana) == 0:
continue
elif len(kana) > 1 and kana[0] == '{':
ks = kana[1:kana.index('}')].split('/')
ks = [k + kana[kana.index('}') + 1:] for k in ks if len(k) > 0]
else:
ks = [
kana,
]
freq = int(freq)
for k in ks:
def doGo(self):
#pressing the go button we first see which dict is selected
try:
theWord = str(self.ui.textEdit.toPlainText())
except:
# for JDICT
theWord = self.ui.textEdit.toPlainText()
if self.ui.comboBox.currentText() == 'KATEGLO':
start = time.clock()
kx = kateglo()
data = kx.getData(theWord)
if data == -1:
s = '\nHello : ' + theWord + ' maybe mispelled or not Indonesian or not the correct root'
self.ui.textEdit_2.append(s)
self.ui.textEdit_2.append('Correct and try again')
else:
tr = kx.translator(data)
self.aline()
s = '\nPhrase : ' + theWord
self.ui.textEdit_2.append(s)
rxc = 0
for rx in tr:
if rxc == 0:
s = 'Source : ' + rx
self.ui.textEdit_2.append(s)
else:
self.ui.textEdit_2.append(rx)
rxc += 1
self.aline()
df = kx.definitor(data)
self.ui.textEdit_2.append("\nDefinitions\n")
for rx in df:
self.ui.textEdit_2.append(rx)
pr = kx.proverbor(data)
self.ui.textEdit_2.append("\nProverbs\n")
for rx in pr:
self.ui.textEdit_2.append(rx)
rl = kx.relator(data)
self.ui.textEdit_2.append("\nRelations\n")
for rx in rl:
self.ui.textEdit_2.append(rx)
self.ui.label_3.setText('Finished KATEGLO request ...')
end = time.clock()
s = 'Request duration : ' + str(end - start) + ' secs'
self.ui.label_4.setText(s)
elif self.ui.comboBox.currentText() == 'KBBI':
start = time.clock()
kb = kbbi()
soup = kb.processData(theWord)
sxt = soup.get_text()
atitle = soup.title.string.split('- definisi kata')
s = '\n' + atitle[0]
self.ui.textEdit_2.append(s)
self.aline()
s = '\nKata : ' + atitle[1]
self.ui.textEdit_2.append(s)
sxts = sxt.split('Pranala (link): http://kbbi.web.id/%s' % theWord)
try:
sxts2 = sxts[1]
sxts2 = sxts2.split('Tweet')
res1 = sxts2[0]
res1 = res1.split(';')
s = ''
s = res1[0].strip('-1').strip('-2').strip('-3')
self.ui.textEdit_2.append(s)
s = ''
for rx in range(1, len(res1)):
rc = res1[rx]
s = rc + '\n'
self.ui.textEdit_2.append(s)
except:
self.ui.textEdit_2.append('')
s = theWord + ' ==> Tidak ditemukan - KBBI\n\nMaybe incorrect root word'
self.ui.textEdit_2.append(s)
# we want to get data from the Memuat section , if any
sxtch = sxt.split('Memuat')
try:
sxtchz = sxtch[1].split('Pranala')
self.ui.textEdit_2.append(
'Try with these suggestions provided by kbbi (if any) :'
)
s = sxtchz[0].replace('1', '\n').replace(
'2', '\n').replace(
'3', '\n') # occasionaly there is are subscripts
self.ui.textEdit_2.append(s)
except:
pass
self.ui.label_3.setText('Finished KBBI request ...')
end = time.clock()
s = 'Request duration : ' + str(end - start) + ' secs'
self.ui.label_4.setText(s)
elif self.ui.comboBox.currentText() == 'GLOSBE':
start = time.clock()
lp = str(self.ui.comboBox_2.currentText())
al = lp.split('/')
orglang = al[0]
destlang = al[1]
gb = glosbe()
data = gb.getData(theWord, orglang, destlang)
if data == -1:
s = '\nHello : ' + theWord + ' maybe mispelled or not the correct root word for dicitionary lookup or wrong language code'
self.ui.textEdit_2.append(s)
s = 'Correct and try again'
self.ui.textEdit_2.append(s)
else:
s = 'From : ' + data['from']
self.ui.textEdit_2.append(s)
s = 'Dest : ' + data['dest']
self.ui.textEdit_2.append(s)
s = 'Result : ' + data['result']
self.ui.textEdit_2.append(s)
s = 'Phrases: ' + data['phrase']
self.ui.textEdit_2.append(s)
self.ui.textEdit_2.append('\n')
self.ui.textEdit_2.append('Translations : ')
# translation results
phr = data['tuc']
for item in range(0, len(phr)):
try:
s = data['tuc'][item]['phrase']['text'] + ' , '
self.ui.textEdit_2.append(s)
except:
pass
self.ui.textEdit_2.append(
'\n\n\n Translation + Sample Sentence\n\n')
r2 = requests.get(
'http://glosbe.com/gapi/translate?from=%s&dest=%s&format=json&tm=true&phrase=%s&pretty=true'
% (orglang, destlang, theWord))
try:
data = json.loads(r2.text)
# translation results
phr = data['tuc']
#pprint(phr)
if len(phr) == 0:
self.ui.textEdit_2.append(
' Nothing returned from Glosbe')
else:
# for formatting precalc maxl
maxl = 0
ll = 0
for item in range(0, len(phr)):
try:
if destlang == 'jpn' or destlang == 'zh' or destlang == 'rus':
ll = len(
str(phr[item]['phrase']['text']).rstrip(
' '))
else:
ll = len(
str(phr[item]['phrase']['text']).encode(
'UTF-8').rstrip(' '))
if ll > maxl:
maxl = ll
except:
ll = 10
pass
try:
self.ui.textEdit_2.append('\nPhrase/Meanings :\n')
if len(phr) == 0:
self.ui.textEdit_2.append(
' Nothing returned from Glosbe')
else:
for item in range(0, len(phr)):
try:
if destlang == 'jpn' or destlang == 'zh' or destlang == 'rus':
if phr[item]['phrase']['text'] <> '':
s = 'Phrase : ' + phr[item][
'phrase']['text'].replace(
''', "'").replace(
'’',
"'").replace(
'é', '`')
self.ui.textEdit_2.append(s)
for itx in range(
0, len(phr[item])):
try:
if phr[item]['meanings'][
itx]['text'] <> '':
s = 'Meaning : ' + phr[
item]['meanings'][itx][
'text'].replace(
''',
"'"
).replace(
'’',
"'"
).replace(
'é',
'`')
self.ui.textEdit_2.append(
s)
except:
pass
else:
if phr[item]['phrase']['text'].encode(
'UTF-8') <> '':
s = 'Phrase : ' + phr[item][
'phrase']['text'].encode(
'UTF-8').replace(
''', "'").replace(
'’',
"'").replace(
'é',
'`')
self.ui.textEdit_2.append(s)
for itx in range(
0, len(phr[item])):
try:
if phr[item]['meanings'][
itx]['text'].encode(
'UTF-8') <> '':
s = 'Meaning : ' + phr[
item]['meanings'][
itx]['text'].encode(
'UTF-8'
).replace(
''',
"'"
).replace(
'’',
"'"
).replace(
'é',
'`')
self.ui.textEdit_2.append(
s)
except:
pass
self.aline()
except:
pass
except:
#print 'Error in Phrase/Meanings'
#raise
pass
try:
self.ui.textEdit_2.append('\n')
if data['tuc'][item]['phrase']['text'] <> ' ':
for ite in range(0, len(data['examples'])):
if destlang == 'jpn' or destlang == 'zh' or destlang == 'rus':
ss = data['examples'][ite][
'second'].replace(
'<strong class="keyword">', '')
ss = ss.replace('</strong>', '').replace(
'#', '').replace('|', '')
sf = data['examples'][ite][
'first'].replace(
'<strong class="keyword">', '')
sf = sf.replace('</strong>', '').replace(
'#', '').replace('|', '')
if sf <> '':
s = '\nExamples for : ' + data['tuc'][
item]['phrase']['text']
self.ui.textEdit_2.append(s)
s = sf
self.ui.textEdit_2.append(s)
self.ui.textEdit_2.append(ss)
else:
ss = data['examples'][ite][
'second'].encode('UTF-8').replace(
'<strong class="keyword">', '')
ss = ss.replace('</strong>', '').replace(
'#', '').replace('|', '')
sf = data['examples'][ite][
'first'].replace(
'<strong class="keyword">', '')
sf = sf.replace('</strong>', '').replace(
'#', '').replace('|', '')
if sf.encode('utf-8') <> '':
s = '\nExamples for : ' + data['tuc'][
item]['phrase']['text']
self.ui.textEdit_2.append(s)
s = sf.encode('utf-8')
self.ui.textEdit_2.append(s)
self.ui.textEdit_2.append(ss)
except:
#raise
pass
except:
#raise
self.ui.textEdit_2.append(
'JSon Error, maybe no data retrieved')
self.ui.textEdit_2.append('Re-try')
end = time.clock()
s = 'Request duration : ' + str(end - start) + ' secs'
self.ui.label_4.setText(s)
self.ui.label_3.setText('Finished GLOSBE request ...')
elif self.ui.comboBox.currentText() == 'WEBLIO':
start = time.clock()
try:
self.oldcurrLine = self.lineNumber2()
s = Weblio()
lt = -1
theText = theWord
# we need to limit the length: testing with 80
lt = len(theText)
#firstLine = self.lineNumber2()
#print 'FirstLine :',firstLine
#firstPosition = self.ui.textEdit_3.textCursor().position()
#print 'Firstpos :',firstPosition
if (lt > 0) and (lt < 80):
self.ui.textEdit_2.append('\nWeblio Results')
self.ui.textEdit_2.append('for:')
self.ui.textEdit_2.append(theText)
self.ui.textEdit_2.append('---------------------\n')
#x = self.ui.spinBox_2.value() # how many items to fetch
# here we hardset to 10 examples max
x = 10
res = s.examples(theText, x) # ok
key = 0
for dx in res:
rx = self.remove_comments2(dx[1], '<!--')
rx2 = self.remove_comments2(dx[2], '<!--')
rx3 = self.doMecab(rx2)
key += 1
# now we check if we are ascii and print accordingly
try:
theText.decode('ascii')
except: # we selected a japanese text
rx4 = self.doMecab(rx)
# oks=str(key)+' : '+rx+' '+rx2.encode('utf8')+' '+kata2hira(rx4).decode('utf8')
oks = str(key) + ' : ' + rx
self.ui.textEdit_2.append(oks)
oks = str(key) + ' : ' + rx2.encode('utf8').strip(
'\n')
self.ui.textEdit_2.append(oks)
oks = str(key) + ' : ' + kata2hira(rx4).decode(
'utf8')
self.ui.textEdit_2.append(oks)
else: # we selected a english text
oks = str(key) + ' : ' + rx.encode('utf8')
self.ui.textEdit_2.append(oks)
oks = str(key) + ' : ' + rx2
self.ui.textEdit_2.append(oks)
oks = str(key) + ' : ' + kata2hira(rx3).decode(
'utf8')
self.ui.textEdit_2.append(oks)
self.ui.textEdit_2.append('---------------------\n')
else:
if lt > 1:
oks = 'Weblio line is too long. Length : %i' % lt
self.ui.textEdit_2.append(oks)
self.ui.textEdit_2.append('---------------------\n')
except:
pass
# we try jump to the begining of the latest weblio data
# for time being just jump to bottom
self.ui.textEdit_2.moveCursor(QTextCursor.End)
currLine = self.lineNumber2()
mv = currLine - self.oldcurrLine
# now move
#print '\nOldCurrLine :',self.oldcurrLine
#print 'CurrLine :',currLine
#print 'mv :',mv
#print '\n'
for j in range(0, mv):
self.ui.textEdit_2.moveCursor(
QTextCursor.Up) #,QTextCursor.MoveAnchor)
nowLine = self.lineNumber2()
#print 'NowLine : ',nowLine
self.oldcurrLine = currLine
self.ui.label_3.setText('Finished Weblio request ....')
end = time.clock()
s = 'Request duration : ' + str(end - start) + ' secs'
self.ui.label_4.setText(s)
elif self.ui.comboBox.currentText() == 'JDICT':
start = time.clock()
# try translate from JDic
# while ok it cud be faster
self.ui.textEdit_2.clear()
self.JDictToggleFlag = True
if self.JDictToggleFlag == True:
try:
jdictranslations = JDic().lookup(unicode(theWord))
sl = 0
jdi = 0
# nrset used as divider for linefeed inserts below
nrset = [
'(1)', '(2)', '(3)', '(4)', '(5)', '(6)', '(7)', '(8)',
'(9)', '(10)', '(11)', '(12)', '(13)', '(14)', '(15)',
'(16)', '(17)', '(18)', '(19)', '(20)'
]
for key in jdictranslations.keys():
sl += 1 # if no key than we never come here
if jdi == 0: # only append once per loop
self.ui.textEdit_2.append("\nJDic Info : \n")
jdi = 1
# this gives a wider view but still messy
#self.ui.textEdit_2.append(jdictranslations[key]+"\n")
# this gives a more readable view
tt = ''
for xs in jdictranslations[key]:
if xs <> ";": # only one space or semicolon or we get empty stuff
tt = tt + xs
else:
# below code by trial and error to have a readable representation
tt = tt.replace(
' ',
' ') # note this is a space and a tab
tt = tt.replace('\n',
'') # get rid of linefeeds
tt = tt.replace(
' ', '\n'
) # insert a linefeed if there are 2 spaces
tt = tt + ';' # add the semicolon back
for nx in nrset:
# iterate over our nrset and insert linefeeds for better readability
tt = tt.replace(nx, '\n' + nx + '\n')
self.ui.textEdit_2.append(tt)
tt = ''
if sl == 0:
self.ui.textEdit_2.append('No info from JDict for ' +
theWord)
except:
# occasional non type objects will occure so we just skip it
pass
finally:
# give a notice if run finished
self.ui.textEdit_2.append('JDict-Finished')
#TODO:: try similar wordnet , wordnet has quota so may not work as wanted
end = time.clock()
s = 'Request duration : ' + str(end - start) + ' secs'
self.ui.label_4.setText(s)
total_freq = 0
for file_name in file_list:
print file_name
with codecs.open(file_name, 'r', 'utf-8') as f:
for line in f:
arr = line.split()
base = 0
for i, e in enumerate(arr):
if e[0] <= '9' and e[0] >= '0':
base = i
kanji = arr[base-1][1:]
freq = arr[base][:-2]
kana = kata2hira(arr[base+2][:-1])
print kanji, freq, kana
if len(kana) == 0:
continue
elif len(kana) > 1 and kana[0] == '{':
ks = kana[1:kana.index('}')].split('/')
ks = [k + kana[kana.index('}')+1:] for k in ks if len(k) > 0]
else:
ks = [kana, ]
freq = int(freq)
for k in ks:
if k not in temp_obj:
def getExamplesKana(query):
return JishoClient.getExamples(kata2hira(''.join(MecabTool.parseToReadingsKana(query)))) #it works but slightly incorrect (mecab shenanigans)
#test = JishoClient.getExamplesKana(u'軈て')
#print '\n'.join(test)
