Primeira refatorado do programa.

sortedcoteiners-refactor.py

import sys

def distrincha():
    words = set()
    plain_text = list()
    codded_text= list()

    # need for speed
    _words_add = words.add
    _codded_text_append = codded_text.append

    # read content
    plain_text = [l.strip().split() for l in sys.stdin]

    # build a word list from content
    for line in plain_text:
        [_words_add(w) for w in line]

    # set() don't have an index method, then use a list
    words = list(words)

    # need for speed
    _words_index = words.index

    # build codded content
    for line in plain_text:
        _codded_text_append([_words_index(w) for w in line])

    # how many words?
    return len(words)

print('Total of words:', distrincha())