How to find indexes of string in lists which starts with some substring?

Question

I have some list like this:

lines = [
    "line",
    "subline2",
    "subline4",
    "line",
]

And I want to take list of index of lines which starts with some substring.

I use this approach:

starts = [n for n, l in enumerate(lines) if l.startswith('sub')]

but maybe anybody knows more beautiful approach?

What is wrong with this one? Its a perfectly valid python :) — Marek Kowalski, Feb 22 '13 at 11:03
@MarekKowalski I think every time we have more pythonic approach :) — kharandziuk, Feb 22 '13 at 11:07
I got one! Don't use `l` as a variable - it looks too much like `1`. [PEP8](http://www.python.org/dev/peps/pep-0008/#names-to-avoid) — John La Rooy, Feb 22 '13 at 11:18
I like to use underscore "_" for "non used variables", but we use i18 in our project and _("localize this") is reserved — kharandziuk, Feb 22 '13 at 11:18
There are no unused variables. There are no unnecessary REs, no complicated chains of higher-order functions, no contrived loops -- this code is perfect as it is, at least in terms of clarity and asymptotic complexity. — Fred Foo, Feb 22 '13 at 11:57

score 8 · Answer 1 · answered Feb 15 '18 at 21:21

8

I know it's been a while since this question was active, but here's another solution anyways in case anyone is interested.

Your way seems fine, but here is a similar strategy, using the list.index() method:

starts = [lines.index(l) for l in lines if l.startswith('sub')]

As far as time goes, the two functions clock in at about the same (on average 1.7145156860351563e-06 seconds for your enumerate solution and 1.7133951187133788e-06 seconds for my .index() solution)

answered Feb 15 '18 at 21:21

sacuL

49,704
8
81
106

1

Unfortunately, this solution is not equivalent to the code of the OP: If the list `lines` contains identical entries, your code will always deliver the index of the first matching entry. Moreover, once the list gets more entries, your approach will lose performance wise because more entries will have to be searched. – Dirk Herrmann May 04 '21 at 23:15

score 3 · Answer 2 · answered May 04 '21 at 23:21

While I like your approach, here is another one that handles identical entries in lines correctly (i.e. similar to the way your sample code does), and has comparable performance, also for the case that the length of lines grows:

starts = [i for i in range(len(lines)) if lines[i].startswith('sub')]

How to find indexes of string in lists which starts with some substring?

2 Answers2