core trie code in new labMTsimple

gistfile1.py

# self.data is a tuple of 2 items
# the first is trie for the non-regex (fixed) words
# and the second for the regex (stem) words
# matching on prefix mimins the [a-z]* match for the stems

# also, this was stolen from an internal function, so
# wordDict is a hash (dict) or "word": count
# for a corpus being scores
totalcount = 0
totalscore = 0.0
for word,count in wordDict.iteritems():
    if word in self.data[0]:
        totalcount += count
        totalscore += count*self.data[0][word][0]
    elif (len(self.data[1].prefixes(word)) > 0):
        totalcount += count
        totalscore += count*self.data[1].prefix_items(word)[0][1][0]
    return totalscore/totalcount