def chunking (token) :
chunks = []
for chunk in syllabify(token)[0]:
chunks.append(unicodedata.normalize('NFD', chunk))
return chunks
def deaffix(word, affix):
'wordform, affix -> (stem, affixlist)'
stemmed = ()
syls = syllabify(word)
for v in syls:
for allomorph in affix[0]:
if match_affix(v, allomorph, affix[1]):
if affix[1] < 0:
lemma = ''.join(v[:affix[1]])
aff = (), (''.join(v[affix[1]:]), affix[2])
else:
lemma = ''.join(v[affix[1]:])
aff = (''.join(v[:affix[1]]), affix[2]), ()
stemmed = (lemma, aff)
if stemmed:
return stemmed
else:
return (word, ((),()))
def parse_composite(gloss, glossary):
'gloss -> [gloss]'
result = []
stem, afflist, pslist, ge = gloss
# syllabify stem
syls = syllabify(stem)
for scheme in composite_order:
if psmatch(pslist, scheme['pslist']):
for variant in syls:
if len(variant) > 2:
splits = split_composite(variant)
for split in splits:
possible_gloss = product(*[lookup_lemma(strtolemma(stem), glossary) for stem in split])
possible_gloss = [gl for gl in possible_gloss if reduce(and_, map(truth, zip(*gl)[3]))]
for glosses in possible_gloss:
for template in scheme['templates']:
if match_template(glosses, template):
result.append(('.'.join(zip(*glosses)[0]), afflist, psmatch(pslist,scheme['pslist']), '.'.join(zip(*glosses)[3])))
return result
#!/usr/bin/python
# -*- encoding: utf-8 -*-
from morphology import wl
from syllables import syllabify
import unicodedata
for i, lx in wl.items():
for lemma in lx:
if 'n' in lemma[1].split('/'):
t = u''.join(syllabify(lemma[0])[0][1:])
for c in unicodedata.normalize('NFD', t):
if unicodedata.category(c) == 'Mn':
print lemma[0].encode('utf-8'), "\t", "‘" + lemma[2].encode('utf-8') + "’"
break