Python：按大小（反转）对列表列表进行排序并存储原始索引

Question

basically a combination of this:

[x[0] for x in sorted(MYLIST, key=len, reverse=True)]
[x[0] for x in sorted(MYLIST, key=lambda i: i[1])]

Answer 1

If we say that lst_of_lst is the original list,

sorted_lst = [(len(lst), lst) for lst in lst_of_lst]
sorted_lst = sorted(sorted_lst, reverse = True)
sorted_lst = [lst for (length,lst) in sorted_lst]

sorted_lst is the new sorted list.

However, this code may not work depending on how you want to treat cases where two lists have the same length.

Answer 2

Because sorted and enumerate are iterators, it's not really possible to get the effect you want using them together in a single loop. So, instead we gather the info you want seperately.

nested_lists = [
    [1, 2, 3, 4, 5],
    [1, 2],
    [1, 2, 3],
    [1, 2, 3, 4]
]

# Capture original indices
indexed_lists = [(index, sublist) for index, sublist in enumerate(nested_lists)]

# sort them by length of sublist
sorted_lists = sorted(indexed_lists, reverse=True, key=lambda item: len(item[1]))

# Iterate over the sorted object, printing the old and new indexes
for new_index, (old_index, sublist) in enumerate(sorted_lists):
    print('old index:', old_index, 'new_index:', new_index)
    print(sublist, end='\n\n')

The output would be in reverse length order (largest to smallest) printing the old and new indices like this:

old index: 0 new_index: 0
[1, 2, 3, 4, 5]

old index: 3 new_index: 1
[1, 2, 3, 4]

old index: 2 new_index: 2
[1, 2, 3]

old index: 1 new_index: 3
[1, 2]

Answer 3

You can use sorted in conjunction with enumerate and provide the method with the key and the reverse parameters. Below, lst is sorted by length, then reversed.

For example, if you have a list like so:

lst = [[29,53,27], [34,34,76,23,65], [32,39]]

You can use sorted and enumerate in a one liner:

sorted_lst = sorted(enumerate(lst), key=lambda v: len(v[1]), reverse=True)

Which returns a list of tuples with the original index position as the first value and the list as the second value:

[(1, [34, 34, 76, 23, 65]), (0, [29, 53, 27]), (2, [32, 39])]