English word patterns

gistfile1.py

# word pattern => list of corresponding english words
pattern_words = defaultdict(list)

def get_pattern(word):
    """Return the pattern for the given word."""
    map = {}
    letters = iter(string.lowercase)
    pattern = ''
    for char in word:
        if char not in map:
            map[char] = letters.next()
        pattern += map[char]
    return pattern

# read in the dictionary words
with open('../dictionary/wordlist.txt') as f:
    for word in f.read().split():
        words.add(word)

# compute and store the pattern for each dictionary word
for word in words:
    pattern = get_pattern(word)
    pattern_words[pattern] += [word]