Find list index of list items within other list it

问题:

I have a list of long strings and I'd like to get the indexes of the list elements that match a substring of strings in another list. Checking if a list item contains a a single string inside a list is easy to do with list comprehensions, like this question:

my_list = ['abc-123', 'def-456', 'ghi-789', 'abc-456']
thing_to_find = "abc"
matching = [i for i, x in enumerate(my_list) if thing_to_find in x]

However, I'd like to check not only if "abc" is in x, but if any strings in another list are in the list, like so:

my_list = ['abc-123', 'def-456', 'ghi-789', 'abc-456']
things_to_find = ['abc', 'def']

This obviously doesn't work (but it would be really cool if it did):

matching = [i for i, x in enumerate(my_list) if things_to_find in x]

I can find the list indexes if I run commands individually, but it's tedious and horrible:

print([i for i, x in enumerate(my_list) if 'abc' in x])
# [0, 3]
print([i for i, x in enumerate(my_list) if 'def' in x])
# [1]

What's the best way to find the indexes of all instances where elements from one list are found in another list?

回答1:

You are looking for the any() function here:

matching = [i for i, x in enumerate(my_list) if any(thing in x for thing in things_to_find)]

Demo:

>>> my_list = ['abc-123', 'def-456', 'ghi-789', 'abc-456']
>>> things_to_find = ['abc', 'def']
>>> [i for i, x in enumerate(my_list) if any(thing in x for thing in things_to_find)]
[0, 1, 3]

回答2:

Maybe something like?:

my_list = ['abc-123', 'def-456', 'ghi-789', 'abc-456']
things_to_find = ['abc', 'def']
for n, e in enumerate(my_list):
    for m in things_to_find:
        if m in e:
            print '%s is in %s at %s' % (m, e, n)

Output:

abc is in abc-123 at 0
def is in def-456 at 1
abc is in abc-456 at 3

回答3:

You are close:

matching = [i for i, x in enumerate(my_list) for keyword in things_to_find if keyword in x]

which gives [0,1,3].

You need to iterate through the things_to_find list as well, and see if the keyword is in x.

回答4:

Might be a little slow, but why not try:

my_list = ['abc-123', 'def-456', 'ghi-789', 'abc-456']
things_to_find = ['abc', 'def']
for thing_to_find in things_to_find:
    matching = [i for i, x in enumerate(my_list) if thing_to_find in x]

回答5:

Build a regex, then test each list element against that:

import re
#must use search, not match because no wildcards, unless only looking for prefixes
regex = re.compile('|'.join(re.escape(interest) for interest in things_to_find))

Don't rebuild the regex every time you do the search - only rebuild when things_to_find changes.

I suspect you don't want the indices, but the elements:

[x for x in my_list if regex.search(x)]

Or, if you really do want the indices:

[i for i,x in enumerate(my_list) if regex.search(x)]

This will likely perform better than an any(in) solution (which is quadratic) for large things_to_find lists, but will be overkill for short lists. You'll also see more of a gain where the things in things_to_find are similar; and less of a gain if you can sort things_to_find such that more likely matches occur first, and if matches are likely.

回答6:

my_list = ['abc-123', 'def-456', 'ghi-789', 'abc-456']
things_to_find = ['abc', 'def']
matching = [[i for i, x in enumerate(my_list) if y in x]for y in things_to_find]