def get_start(self,text): guess = FCUtil.wordcount_to_charcount(self.start_index,text) #matches = s.get_matching_blocks() #match = s.find_longest_match(0,len(self.quote),0,len(text)) matcher = diff_match_patch() m = matcher.match_main(text,self.quote,guess) #print text #print self.quote s = difflib.SequenceMatcher(None,self.quote,text[m:m+len(self.quote)]) r = s.ratio() if r < .9: print "Low Ratio %.2f %d " %(r,self.id) print "Quote: %s" %(self.quote) print "Text: %s" %(text[m:m+len(self.quote)]) if m == -1: print "matching error for: %d" %self.id self.start = m self.end = m+len(self.quote)
