def main():
if len(sys.argv) != 2:
print "Usage: %s inputFile" % sys.argv[0]
sys.exit(1)
ifile = codecs.open(sys.argv[1], 'r', encoding='utf-8')
ofile = codecs.open(sys.argv[1] + '.syllables', 'w', encoding='utf-8')
prev_syllables = u""
prev_count = 0
for line in (l.strip() for l in ifile):
syllables = list(ml.ClusterIter(line)) #ml.ClusterIter(line).next ()
if len(syllables) < 2:
syllables = "".join(syllables)
else:
syllables = "".join(syllables[:2])
if u"\u1039" in syllables or u'\u103F' in syllables:
continue
if syllables == prev_syllables:
prev_count += 1
else:
#if (u"\u1039" in prev_syllables and prev_count > 25) or (prev_count > 50):
if (prev_count > 50):
ofile.write('%s\n' % (prev_syllables))
#else:
#print "skipping ", prev_syllables.encode ('utf8')
prev_count = 0
prev_syllables = syllables
ifile.close()
ofile.close()
def main():
if len(sys.argv) != 2:
print "Usage: %s inputFile" % sys.argv[0]
sys.exit(1)
ifile = codecs.open(sys.argv[1], 'r', encoding='utf-8')
ofile = codecs.open(sys.argv[1] + '.pcol', 'w', encoding='utf-8')
for line in (l.strip() for l in ifile):
for c in ml.ClusterIter(line):
ofile.write("%s " % c)
ofile.write('\n')
ifile.close()
ofile.close()
def main():
if len(sys.argv) != 2:
print "Usage: %s inputFile" % sys.argv[0]
sys.exit(1)
ifile = codecs.open(sys.argv[1], 'r', encoding='utf-8')
ofile = codecs.open(sys.argv[1] + '.prefixes', 'w', encoding='utf-8')
prev_syllables = []
prev_count = 0
match = False
for line in (l.strip() for l in ifile):
syllables = list(ml.ClusterIter(line)) #ml.ClusterIter(line).next ()
if len(syllables) <= 1:
match = False
prev_syllables = []
prev_count = 0
continue
if match:
if startswith(syllables, match):
#print mystr(syllables).encode ('utf-8'), " startswith ", mystr(match).encode ('utf-8')
prev_count += 1
else:
#print mystr(syllables).encode ('utf-8'), " not startswith ", mystr(match).encode ('utf-8')
if prev_count > 5:
ofile.write(mystr(match) + '\n')
#else:
#print "omitting ", mystr(match).encode ('utf-8')
prev_syllables = syllables
prev_count = 0
match = False
else:
match = get_longest_match(syllables, prev_syllables)
#print "syllable = %s, prev_syl =%s , match = %s" %(mystr (syllables).encode ('utf-8'),\
#mystr (prev_syllables).encode ('utf-8'),\
#mystr (match).encode ('utf-8'))
if len(match) < 2:
#if prev_syllables:
#print "omitting ", mystr(prev_syllables).encode ('utf-8')
# ofile.write (mystr (prev_syllables) + '\n')
prev_syllables = syllables
match = False
ifile.close()
ofile.close()
def main():
if len(sys.argv) != 2:
print "Usage: %s inputFile" % sys.argv[0]
sys.exit(1)
ifile = codecs.open(sys.argv[1], 'r', encoding='utf-8')
for k, line in enumerate((l.strip() for l in ifile)):
#print line.encode ('utf8')
try:
syllables = list(ml.ClusterIter(line))
except:
pass
if len(syllables) == 1 and u'\u1039' not in syllables[0]:
print "".join(syllables).encode('utf8')
ifile.close()
def main():
if len(sys.argv) != 2:
print "Usage: %s inputFile" % sys.argv[0]
sys.exit(1)
if (os.path.exists(sys.argv[1] + '.dpkl')):
print "Found pickle."
with open(sys.argv[1] + '.dpkl', 'rb') as fil:
finder4 = cPickle.load(fil)
else:
ifile = codecs.open(sys.argv[1], 'r', encoding='utf-8')
syllables = []
skip = "per()".join(ml.puncts + ml.digits)
for i, line in enumerate(l.strip() for l in ifile):
progress(i)
line = line.replace(" ", "")
for p in skip:
line = line.replace(p, "")
try:
p = [c for c in ml.ClusterIter(line)]
for each in possible_bigram_list(p):
syllables += ["".join(each)]
except Exception, e:
print str(e), line, c
ifile.close()
finder4 = BigramCollocationFinder.from_words(syllables)
finder4.apply_freq_filter(10) # only bigrams that appear 3+ times
# remove stopwords
stopwords = []
with codecs.open('stopwords', 'r', 'utf8') as stopwordFile:
for each in stopwordFile:
stopwords.append(each.strip())
def contains_stopwords(*syllables):
for each in stopwords:
if each in syllables:
return True
return False
finder4.apply_ngram_filter(contains_stopwords)
with open(sys.argv[1] + '.dpkl', 'wb') as fil:
cPickle.dump(finder4, fil)
def main():
if len(sys.argv) != 2:
print "Usage: %s inputFile" % sys.argv[0]
sys.exit(1)
ifile = codecs.open(sys.argv[1], 'r', encoding='utf-8')
for k, line in enumerate((l.strip() for l in ifile)):
print line.encode('utf8')
syllables = list(ml.ClusterIter(line))
bigrams = nltk.bigrams(syllables)
trigrams = nltk.trigrams(syllables)
tetragrams = nltk.bigrams(["".join(x) for x in _bigrams(syllables)
]) if len(syllables) > 3 else []
for each in itertools.chain(bigrams, trigrams, tetragrams):
print "".join(each).encode('utf8')
ifile.close()
def main():
if len(sys.argv) != 3:
print "Usage: %s syllableFile inputFile" % sys.argv[0]
sys.exit(1)
p = re.compile(
u'(\u1004\u103a\u1039)?((\u1000\u103b\u1015\u103a|\u100b\u1039\u100b|\u100b\u1039\u100c|\u100d\u1039\u100d|\u100d\u1039\u100e|\u100f\u1039\u100d)|((\u104e\u1004\u103a\u1038|\u1000|\u1001|\u1002|\u1003|\u1004|\u1005|\u1006|\u1007|\u1008|\u1009|\u100a|\u100b|\u100c|\u100d|\u100e|\u100f|\u1010|\u1011|\u1012|\u1013|\u1014|\u1015|\u1016|\u1017|\u1018|\u1019|\u101a|\u101b|\u101c|\u101d|\u101e|\u101f|\u1020|\u1021|\u1023|\u1024|\u1025|\u1026|\u1027|\u1029|\u102a|\u103f|\u1040|\u1041|\u1042|\u1043|\u1044|\u1045|\u1046|\u1047|\u1048|\u1049|\u104a|\u104b|\u104c|\u104d|\u104e|\u104f)(\u1039\u1010\u103d|\u1039\u1000|\u1039\u1001|\u1039\u1002|\u1039\u1003|\u1039\u1005|\u1039\u1006|\u1039\u1007|\u1039\u1008|\u1039\u100b|\u1039\u100c|\u1039\u100d|\u1039\u100e|\u1039\u100f|\u1039\u1010|\u1039\u1011|\u1039\u1012|\u1039\u1013|\u1039\u1014|\u1039\u1015|\u1039\u1016|\u1039\u1017|\u1039\u1018|\u1039\u1019|\u1039\u101a|\u1039\u101b|\u1039\u101c|\u1039\u101e|\u1039\u1021)?))(\u103a)?(\u103b)?(\u103c)?(\u103d)?(\u103e)?(\u1031)?(\u102d|\u102e)?(\u102f|\u1030)?(\u1032|\u1036)?(\u102b|\u102c)?(\u1037)?(\u103a)?(\u1037)?(\u1038)?',
re.U)
sfile = codecs.open(sys.argv[1], 'r', encoding='utf-8')
ifile = codecs.open(sys.argv[2], 'r', encoding='utf-8')
ofile = codecs.open(sys.argv[2] + '.checked', 'w', encoding='utf-8')
for line in (l.strip() for l in ifile):
for i in ml.ClusterIter(line):
if not p.search(i):
ofile.write(i + '\t' + line + '\n')
break
sfile.close()
ifile.close()
ofile.close()
def main():
if len(sys.argv) != 2:
print "Usage: %s inputFile" % sys.argv[0]
sys.exit(1)
ifile = codecs.open(sys.argv[1], 'r', encoding='utf-8')
ofile = codecs.open(sys.argv[1] + '.pwar2', 'w', encoding='utf-8')
prev_syllables = []
prev_count = 0
match = []
tmp = []
pos = 0
for k, line in enumerate((l.strip() for l in ifile)):
if (k % 1000) == 0:
print k
syllables = list(ml.ClusterIter(line)) #ml.ClusterIter(line).next ()
if (len(syllables) < 1): # < 2
match = []
prev_syllables = []
prev_count = 0
continue
#ofile.write ("\n\ns=%s p=%s:m=%s\n" %(mystr(syllables), mystr(prev_syllables), mystr(match)))
if prev_syllables:
if len(match) > 1 and (len(match) + 1) < len(prev_syllables):
ofile.write("%s\n" % mystr(prev_syllables[len(match):]))
if match:
if startswith(syllables, match):
prev_count += 1
else:
#for u in range (prev_count):
ofile.write("%s\n" % mystr(match))
match = []
prev_syllables = []
prev_count = 0
else:
match = get_longest_match(syllables, prev_syllables)
#ofile.write ("==> prev_count= %d match=%s\n" %(prev_count,mystr(syllables)))
if len(match) < 2:
#if len(match) == 0:
#ofile.write ("==> here= t=%s ps=%s\n" %(mystr(tmp),mystr(prev_syllables)))
if tmp and len(get_longest_match(tmp, prev_syllables)) < 2:
if len(prev_syllables) > 2:
ofile.seek(pos)
ofile.write("%s\n" % mystr(prev_syllables[1:]))
tmp = prev_syllables
match = []
prev_count = 0
else:
if (len(match) + 1) < len(prev_syllables):
ofile.write("%s\n" % mystr(prev_syllables[len(match):]))
prev_count += 1
pos = ofile.tell()
ofile.write("%s\n" % mystr(syllables))
prev_syllables = syllables
ifile.close()
ofile.close()