For example, in programming, there is the following python list
l
l=[
{"a":1,"b":2},
{"a":4,"b":1},
{"a":2,"b":5},
{"a":1,"b":4},
{"a":1,"b":1},
{"a":1,"b":3}
]
Sorts the list
l
, sorting only by the
a
field
l.sort(key=lambda item:item['a'])
Results are as follows:
[
{'a': 1, 'b': 2},
{'a': 1, 'b': 4},
{'a': 1, 'b': 1},
{'a': 1, 'b': 3},
{'a': 2, 'b': 5},
{'a': 4, 'b': 1}
]
You can see that
a
is sorted from smallest to largest, but when
a
is
1
, these elements are not sorted by
b
:
{'a': 1, 'b': 2},
{'a': 1, 'b': 4},
{'a': 1, 'b': 1},
{'a': 1, 'b': 3},
If we use multi-field sort
l.sort(key=lambda item:(item['a'],item['b']))
, the sorted result is sorted by the
a
field first, and then sorted by the
b
field if the
a
field is the same:
[
{'a': 1, 'b': 1},
{'a': 1, 'b': 2},
{'a': 1, 'b': 3},
{'a': 1, 'b': 4},
{'a': 2, 'b': 5},
{'a': 4, 'b': 1}
]
You can see that when a is all 1, it is sorted by element b from smallest to largest:
{'a': 1, 'b': 1},
{'a': 1, 'b': 2},
{'a': 1, 'b': 3},
{'a': 1, 'b': 4}
partial order plays an important role in programming. When comparing two elements in sorting, whether the comparison relationship between elements meets partial order will lead to different sorting results. More applications, in the database query,
l
=[
{
"a"
:
1
,
"b"
:
2
},
{
"a"
:
4
,
"b"
:
1
},
{
"a"
:
2
,
"b"
:
5
},
{
"a"
:
1
,
"b"
:
4
},
{
"a"
:
1
,
"b"
:
1
},
{
"a"
:
1
,
"b"
:
3
}
]
7 keyword, you can specify the database query according to 1 to more fields to do sort.
