replace-elongated-words.py

#!/usr/bin/env python
"""Remove consecutive duplicate characters unless inside a known word.

Usage:
    $ ./replace-elongated-words input_file

http://stackoverflow.com/questions/22471255/replace-mutiple-occurance-of-characters-as-well-as-punctuation-in-strings-of-fil
"""
import fileinput
import re
import sys
from itertools import groupby, product
import enchant # $ pip install pyenchant

def remove_consecutive_dups(s):
    return re.sub(r'(?i)(.)\1+', r'\1', s)

def all_consecutive_duplicates_edits(word, max_repeat=float('inf')):
    chars = [[c*i for i in range(min(len(list(dups)), max_repeat), 0, -1)]
             for c, dups in groupby(word)]
    return map(''.join, product(*chars))

words = enchant.Dict("en")
is_known_word = words.check
for line in fileinput.input(inplace=False):
    #NOTE: unnecessary work, optimize if needed
    output = [next((e for e in all_consecutive_duplicates_edits(s)
                    if e and is_known_word(e)), remove_consecutive_dups(s))
              for s in re.split(r'(\W+)', line)]
    sys.stdout.write(''.join(output))