Created
October 2, 2012 18:27
-
-
Save kmike/3822071 to your computer and use it in GitHub Desktop.
вывод в консоль всех слов из словаря pymorphy 0.5.6 (от Yuri Baburov)
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import pprint | |
import pymorphy | |
import sys | |
sysenc = 'utf-8' | |
def enc(*y): | |
for x in y: | |
if isinstance(x, unicode): | |
yield x.encode(sysenc) | |
else: | |
yield ("%s" % x).encode(sysenc) | |
def re(x): | |
return pprint.pformat(x).decode('unicode_escape') | |
def reprint(*y): | |
for x in enc(*map(re, y)): | |
print x, | |
r = pymorphy.get_morph('/Projects/texts/dicts/pymorphy/ru', backend='shelve') | |
gr = lambda x: reprint(r.get_graminfo(x.upper())) | |
def iter_words(r): | |
listed = set() | |
for lemma in r.data.lemmas.dict.keys(): | |
lemma = lemma.decode('utf-8') | |
forms = r.data.lemmas[lemma] | |
for form in forms: | |
rule = r.data.rules[form] | |
word = lemma + rule[0][0] | |
if not word in listed: | |
yield word | |
listed.add(word) | |
if __name__ == '__main__': | |
o = set() | |
if sys.argv[1] == 'file': | |
f = open(sys.argv[2], 'rt') | |
words = f.read().decode('utf-8').split() | |
for w in words: | |
gi = r.get_graminfo(w.upper()) | |
print 'Total:', len(words), 'words' | |
elif sys.argv[1] == 'all': | |
#print "#class info norm word".replace(' ','\t') | |
for word in iter_words(r): | |
#reprint(norm) | |
for d in r.decline(word): | |
p = (d['class'], d['info'], word, d['word']) | |
key = '\t'.join(p) | |
if not key in o: | |
print key.encode('utf-8') | |
o.add(key) | |
#for w in r.data.: | |
# print w | |
else: | |
gr(sys.argv[1].decode(sysenc)) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment