spark教程(18)-sparkSQL 自定义函数

sparkSQL 也允许用户自定义函数，包括 UDF、UDAF，但没有 UDTF

官方 API

class pyspark.sql.UDFRegistration(sparkSession)[source]

　　　　register(name, f, returnType=None)[source]

　　　　　　Register a Python function (including lambda function) or a user-defined function as a SQL function.

　　　　　　Parameters

　　　　　　name – name of the user-defined function in SQL statements.

　　　　　　f – a Python function, or a user-defined function. The user-defined function can be either row-at-a-time or vectorized. See pyspark.sql.functions.udf() andpyspark.sql.functions.pandas_udf().

　　　　　　returnType – the return type of the registered user-defined function. The value can be either a pyspark.sql.types.DataType object or a DDL-formatted type string.

　　　　　　Returns

　　　　　　a user-defined function.

  registerJavaFunction(name, javaClassName, returnType=None)[source]

  registerJavaUDAF(name, javaClassName)

示例代码

strlen = spark.udf.register("stringLengthString", lambda x: len(x))
spark.sql("SELECT stringLengthString('test')").collect()        # test 只是个字符
# [Row(stringLengthString(test)=u'4')]
spark.sql("SELECT stringLengthString(name) from hive1101.person limit 3").collect()     # read  hive table
# [Row(stringLengthString(name)=u'4'), Row(stringLengthString(name)=u'4'), Row(stringLengthString(name)=u'4')]

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()
# [Row(slen(test)=4)]

参考资料：

https://spark.apache.org/docs/latest/api/python/pyspark.sql.html#pyspark.sql.UDFRegistration　　官网，也更多例子