Python-Pyspark对Column使用UDF.md

发表于 | 更新于 | 分类于 ， | 阅读次数：

Python代码如下：

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
# -*- coding: utf-8 -*-
# @Time    : 2022-06-17

from pyspark.sql import Column
from pyspark.sql.types import IntegerType
from pyspark.sql import SparkSession
from pyspark.sql import functions as F


def count_add_val(col, val):
    def add(cell):
        return cell + val

    count_col = F.count(col)
    return F.udf(add, IntegerType())(count_col)


def count_add(col: Column) -> Column:
    def add(cell):
        return cell + 1

    count_col = F.count(col)
    return F.udf(add, IntegerType())(count_col)


def count_base(col: Column) -> Column:
    return F.count(col)


if __name__ == "__main__":
    spark = SparkSession.builder.getOrCreate()
    df = spark.createDataFrame([("A", 1),
                                ("B", 2),
                                ("B", 4)], ("name", "age"))
    df.show()

    df.select(count_base(df.age)).show()
    df.groupby(df.name).agg(count_base(df.name)).show()

    df.select(count_add(df.age)).show()
    df.groupby(df.name).agg(count_add(df.name)).show()

    df.select(count_add_val(df.age, val=3)).show()
    df.groupby(df.name).agg(count_add_val(df.name, val=3)).show()

运行的结果为：

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
>>> df.show()
+----+---+
|name|age|
+----+---+
|   A|  1|
|   B|  2|
|   B|  4|
+----+---+

>>> df.select(count_base(df.age)).show()
+----------+
|count(age)|
+----------+
|         3|
+----------+

>>> df.groupby(df.name).agg(count_base(df.name)).show()  
+----+-----------+
|name|count(name)|
+----+-----------+
|   B|          2|
|   A|          1|
+----+-----------+

>>> df.select(count_add(df.age)).show()
+---------------+
|add(count(age))|
+---------------+
|              4|
+---------------+

>>> df.groupby(df.name).agg(count_add(df.name)).show()
+----+----------------+
|name|add(count(name))|
+----+----------------+
|   B|               3|
|   A|               2|
+----+----------------+

>>> df.select(count_add_val(df.age, val=3)).show()
+---------------+
|add(count(age))|
+---------------+
|              6|
+---------------+

>>> df.groupby(df.name).agg(count_add_val(df.name, val=3)).show()
+----+----------------+
|name|add(count(name))|
+----+----------------+
|   B|               5|
|   A|               4|
+----+----------------+
Thanks for rewarding