Cleaner way of dealing with punctuation in C1b
[cipher-training.git] / segment.py
1 """Segment a collection of letters into words"""
2
3 import language_models
4 import sys
5 from functools import lru_cache
6 sys.setrecursionlimit(1000000)
7
8 @lru_cache()
9 def segment(text):
10 """Return a list of words that is the best segmentation of text.
11 """
12 if not text: return []
13 candidates = ([first]+segment(rest) for first, rest in splits(text))
14 return max(candidates, key=language_models.Pwords)
15
16 @lru_cache()
17 def segment_wrong(text):
18 """Return a list of words that is the best segmentation of text.
19 """
20 if not text: return []
21 candidates = ([first]+segment(rest) for first, rest in splits(text))
22 return max(candidates, key=language_models.Pwords_wrong)
23
24
25 def splits(text, L=20):
26 """Return a list of all possible (first, rest) pairs, len(first)<=L.
27 """
28 return [(text[:i+1], text[i+1:])
29 for i in range(min(len(text), L))]
30