I wanted to break a string into words, but keeping the index where the word started. E.g., I want to transform 'aaa bbb ccc' into [(0, 'aaa'), (4, 'bbb'), (8, 'ccc')]. This is just the background, not the question.
The problem is that I tried to use itertools.groupby with str.isalpha as key, but it's giving me weird results.
This code shows what I'm talking about (please ignore the list everywhere. I just wanted to be sure I was dealing with iterables, not iterators):
from itertools import groupby
text = 'aaa bbb ccc'
chars = list(groupby(list(enumerate(text)), lambda x: x[1].isalpha()))
result = [list(v) for k, v in chars if k]
print result
assert result == [
[(0, 'a'), (1, 'a'), (2, 'a')],
[(4, 'b'), (5, 'b'), (6, 'b')],
[(8, 'c'), (9, 'c'), (10, 'c')]]
The variable result is ending up as [[(10, 'c')], [], []] and I don't know why. Maybe I'm missing something really simple here, but I just can't see it.
Correct the code:
chars = groupby(l, lambda x: x[1].isalpha())
result = [list(v) for k, v in chars if k]
To figure out the weird output
>>> l = list(enumerate(text))
>>> chars = groupby(l, lambda x: x[1].isalpha())
>>> list(chars.next()[1])
[(0, 'a'), (1, 'a'), (2, 'a')]
>>> for k,v in list(chars): print list(v)
[]
[(10, 'c')]
[]
[]
list would take effect on the sub-iterator in groupby