从list的子列表中提取最长的字符串。 蟒蛇

Question

so i have a list of sublists and within the sublists, there are strings.

the strings are usually at different lengths, but can be the same length as well.

below is an example of the list

sequences = [['aaa'],['aaaa','bb'],[],['aaaaaa','bb','cccccc']]

i want to find a way to extract the LONGEST string from each list and if there are two that are equally long, then take both of those strings

example_output = [['aaa'],['aaaa'],[],['aaaaaa','cccccc']]

usually i would set a threshold in a for-loop where if it was longer than a certain length then append to a list and then after each iteration append that to a list . . . but i don't have a threshold value in this case

if possible i would like try and avoid using lambda and functions since this will be within another function

Answer 1

You can use the length of the longest string seen so far as the threshold ( maxlen in the code below):

def get_longest(seq):
    maxlen = -1
    ret = []
    for el in seq:
        if len(el) > maxlen:
            ret = [el]
            maxlen = len(el)
        elif len(el) == maxlen:
            ret.append(el)
    return ret

sequences = [['aaa'],['aaaa','bb'],[],['aaaaaa','bb','cccccc']]
example_output = list(map(get_longest, sequences))
print(example_output)

This produces:

[['aaa'], ['aaaa'], [], ['aaaaaa', 'cccccc']]

Answer 2

This answer is not the most efficient, but easy to understand.

You can first extract the max lengths (here I'm using a generator expression for that), then extract the strings with those lengths.

lengths = ( max(len(s) for s in sublist) if sublist else 0 for sublist in sequences )
[ [ s for s in sublist if len(s) == l ] for l, sublist in zip(lengths, sequences) ]
-> [['aaa'], ['aaaa'], [], ['aaaaaa', 'cccccc']]

itertools.izip is preferable over zip in this case.

Answer 3

我将使用以下内容（含糊的:)）给我一个镜头：

example_output = [list(filter(lambda x: len(x)==len(max(sub_lst, key=len)), sub_lst)) for sub_lst in sequences]