UDFRegistration

Wrapper for user-defined function registration. This instance can be accessed by spark.udf.

Syntax

Python

# Access through SparkSession
spark.udf

Properties

Property	Description
logs	Returns a UDFLogs instance for UDF logging. This feature is experimental and unstable.

Methods

Method	Description
register(name, f, returnType)	Registers a Python function (including lambda functions) or a user-defined function as a SQL function. Supports Spark Connect.
registerJavaFunction(name, javaClassName, returnType)	Registers a Java user-defined function as a SQL function. When returnType is not specified, it is inferred via reflection. Supports Spark Connect.
registerJavaUDAF(name, javaClassName)	Registers a Java user-defined aggregate function as a SQL function. Supports Spark Connect.

Examples

Python

strlen = spark.udf.register("stringLengthString", lambda x: len(x))
spark.sql("SELECT stringLengthString('test')").collect()

Output

[Row(stringLengthString(test)='4')]

Python

from pyspark.sql.types import IntegerType
from pyspark.sql.functions import udf

slen = udf(lambda s: len(s), IntegerType())
_ = spark.udf.register("slen", slen)
spark.sql("SELECT slen('test')").collect()

Output

[Row(slen(test)=4)]

Python

import pandas as pd
from pyspark.sql.functions import pandas_udf

@pandas_udf("integer")
def add_one(s: pd.Series) -> pd.Series:
    return s + 1

_ = spark.udf.register("add_one", add_one)
spark.sql("SELECT add_one(id) FROM range(3)").collect()

Output

[Row(add_one(id)=1), Row(add_one(id)=2), Row(add_one(id)=3)]