131 – Efficiently count words in a string

You can efficiently count how many words a string has by using itertools.groupby:

from itertools import groupby

def count_words(s):
    return sum(is_alpha for is_alpha, _ in groupby(s, key=str.isalpha))

This is so efficient, you could use it to count words on files larger than your computer’s RAM memory. But here’s an example usage with a smaller string:

print(
    count_words("Hey there–how you doin'?")
)  # 5

Note that, depending on the predicate you use as the keyword argument key, you get different definitions of “word”. By using is_alpha, you are implicitly defining a “word” to be a sequence of characters for which the string method str.isalpha returns True.

Under the hood, groupby is grouping characters in runs of characters for which str.isalpha is True and runs for which it is False, and then you’re counting the number of runs associated with the value True.