- 
      
 - 
        
Save pr0n00gler/b7cc763a978b5b2719d0 to your computer and use it in GitHub Desktop.  
  
    
      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
    
  
  
    
  | # -*- coding: utf-8 -*- | |
| # Портирован с Java по мотивам http://www.algorithmist.ru/2010/12/porter-stemmer-russian.html | |
| import re | |
| class Porter: | |
| PERFECTIVEGROUND = re.compile(u"((ив|ивши|ившись|ыв|ывши|ывшись)|((?<=[ая])(в|вши|вшись)))$") | |
| REFLEXIVE = re.compile(u"(с[яь])$") | |
| ADJECTIVE = re.compile(u"(ее|ие|ые|ое|ими|ыми|ей|ий|ый|ой|ем|им|ым|ом|его|ого|ему|ому|их|ых|ую|юю|ая|яя|ою|ею)$") | |
| PARTICIPLE = re.compile(u"((ивш|ывш|ующ)|((?<=[ая])(ем|нн|вш|ющ|щ)))$") | |
| VERB = re.compile(u"((ила|ыла|ена|ейте|уйте|ите|или|ыли|ей|уй|ил|ыл|им|ым|ен|ило|ыло|ено|ят|ует|уют|ит|ыт|ены|ить|ыть|ишь|ую|ю)|((?<=[ая])(ла|на|ете|йте|ли|й|л|ем|н|ло|но|ет|ют|ны|ть|ешь|нно)))$") | |
| NOUN = re.compile(u"(а|ев|ов|ие|ье|е|иями|ями|ами|еи|ии|и|ией|ей|ой|ий|й|иям|ям|ием|ем|ам|ом|о|у|ах|иях|ях|ы|ь|ию|ью|ю|ия|ья|я)$") | |
| RVRE = re.compile(u"^(.*?[аеиоуыэюя])(.*)$") | |
| DERIVATIONAL = re.compile(u".*[^аеиоуыэюя]+[аеиоуыэюя].*ость?$") | |
| DER = re.compile(u"ость?$") | |
| SUPERLATIVE = re.compile(u"(ейше|ейш)$") | |
| I = re.compile(u"и$") | |
| P = re.compile(u"ь$") | |
| NN = re.compile(u"нн$") | |
| def stem(word): | |
| word = word.lower() | |
| word = word.replace(u'ё', u'е') | |
| m = re.match(Porter.RVRE, word) | |
| if m.groups(): | |
| pre = m.group(1) | |
| rv = m.group(2) | |
| temp = Porter.PERFECTIVEGROUND.sub('', rv, 1) | |
| if temp == rv: | |
| rv = Porter.REFLEXIVE.sub('', rv, 1) | |
| temp = Porter.ADJECTIVE.sub('', rv, 1) | |
| if temp != rv: | |
| rv = temp | |
| rv = Porter.PARTICIPLE.sub('', rv, 1) | |
| else: | |
| temp = Porter.VERB.sub('', rv, 1) | |
| if temp == rv: | |
| rv = Porter.NOUN.sub('', rv, 1) | |
| else: | |
| rv = temp | |
| else: | |
| rv = temp | |
| rv = Porter.I.sub('', rv, 1) | |
| if re.match(Porter.DERIVATIONAL, rv): | |
| rv = Porter.DER.sub('', rv, 1) | |
| temp = Porter.P.sub('', rv, 1) | |
| if temp == rv: | |
| rv = Porter.SUPERLATIVE.sub('', rv, 1) | |
| rv = Porter.NN.sub(u'н', rv, 1) | |
| else: | |
| rv = temp | |
| word = pre+rv | |
| return word | |
| stem=staticmethod(stem) | |
| if __name__ == '__main__': | |
| print Porter.stem(u'устойчивость') | 
  
    Sign up for free
    to join this conversation on GitHub.
    Already have an account?
    Sign in to comment