如何根据不同级别的一组键对Python中的列表进行排序

Question

我在python有一个字典list 。 我想key s 对其进行排序，即首先按chain_id ，然后按start.residue_number ，然后按end.residue_number 。 在下面的list中，第 4 个和第 6 个元素应该是相反的。 我可以用sorted与from operator import itemgetter, attrgetter进行排序chain_id ，但如何做同样的用dictionary里面dictionary ？

            "mappings": [
                {
                    "entity_id": 1,
                    "end": {
                        "residue_number": 63
                    },
                    "start": {
                        "residue_number": 1
                    },
                    "chain_id": "A",
                    "struct_asym_id": "A"
                },
                {
                    "entity_id": 1,
                    "end": {
                        "residue_number": 116
                    },
                    "start": {
                        "residue_number": 1
                    },
                    "chain_id": "A",
                    "struct_asym_id": "A"
                },
                {
                    "entity_id": 1,
                    "end": {
                        "residue_number": 124
                    },
                    "start": {
                        "residue_number": 1
                    },
                    "chain_id": "A",
                    "struct_asym_id": "A"
                },
                {
                    "entity_id": 1,
                    "end": {
                        "residue_number": 116
                    },
                    "start": {
                        "residue_number": 1
                    },
                    "chain_id": "B",
                    "struct_asym_id": "B"
                },
                {
                    "entity_id": 1,
                    "end": {
                        "residue_number": 124
                    },
                    "start": {
                        "residue_number": 1
                    },
                    "chain_id": "B",
                    "struct_asym_id": "B"
                },
                {
                    "entity_id": 1,
                    "end": {
                        "residue_number": 63
                    },
                    "start": {
                        "residue_number": 1
                    },
                    "chain_id": "B",
                    "struct_asym_id": "B"
                }
            ]

Answer 1

我假设d作为你的字典，用sorted和lambda解决

# Sort with chain_id
In [93]: sort_1 = sorted(d['mappings'],key=lambda x:x['chain_id'])
# Sort with start residue number
In [94]: sort_2 = sorted(sort_1,key=lambda x:x['start']['residue_number'])
# Sort with end residue number
In [95]: sort_3 = sorted(sort_2,key=lambda x:x['end']['residue_number'])

或者你可以在一条线上做。

sorted(d['mappings'],key=lambda x:(x['chain_id'],x['start']['residue_number'],x['end']['residue_number']))

灵感来自 leotrubach 的回答。

Answer 2

如果所有键都按降序排序，您可以使用list.sort()或sorted()函数并创建一个返回键函数的元组：

print(
    sorted(
        mappings, 
        key=lambda x: (
            x['chain_id'], 
            x['start']['residue_number'], 
            x['end']['residue_number']
        )
    )
)