使用列表中包含的列表删除重复的元组

Question

i have a list with tuples:

managed_list = [ ('a', [1,2]), ('a', [1,2]), ('b', [2,2]), ('b', [2,2])]

need to get:

managed_list = [ ('a', [1,2]), ('b', [2,2]) ]

tried:

seen = set()
[[n for n in x if n not in seen and not seen.add(n)] for x in managed_list]

getting:

TypeError: unhashable type: 'list'

Answer 1

Right, you can't use a list or a structure containing a list (or another unhashable type) in a set . Without changing your input structure, you could use itertools.groupby and then just discard the iterator through the duplicates:

import itertools
uniques = [x[0] for x in itertools.groupby(sorted(managed_list))]

Incidentally if it were not for the unhashable key issue (eg if the lists were tuples instead), your expression could be simplified to:

list(set(managed_list))

You do not need the extra code in the list comprehension for this.

Answer 2

You can also use collections.OrderedDict to remove duplicate keys.

>>> from collections import OrderedDict
>>> OrderedDict([ ('a', [1,2]), ('a', [1,2]), ('b', [2,2]), ('b', [2,2])]).items()
[ ('a', [1,2]), ('b', [2,2]) ]

Keep in mind that in case of duplicate keys, the right-most entry will be the one included in the output.