cloud-fan commented on a change in pull request #28079: [SPARK-31312][SQL] Cache Class instance for the UDF instance in HiveFunctionWrapper URL: https://github.com/apache/spark/pull/28079#discussion_r400756377
########## File path: sql/hive/src/test/scala/org/apache/spark/sql/hive/HiveUDFDynamicLoadSuite.scala ########## @@ -0,0 +1,254 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.spark.sql.hive + +import org.apache.spark.sql.{QueryTest, Row} +import org.apache.spark.sql.catalyst.expressions.{AttributeReference, Expression} +import org.apache.spark.sql.hive.HiveShim.HiveFunctionWrapper +import org.apache.spark.sql.hive.test.TestHiveSingleton +import org.apache.spark.sql.test.SQLTestUtils +import org.apache.spark.sql.types.{IntegerType, StringType} +import org.apache.spark.util.Utils + +class HiveUDFDynamicLoadSuite extends QueryTest with SQLTestUtils with TestHiveSingleton { + + case class UDFTestInformation( + identifier: String, + funcName: String, + className: String, + fnVerifyQuery: () => Unit, + fnCreateHiveUDFExpression: () => Expression) + + private val udfTestInfos: Seq[UDFTestInformation] = Array( + // UDF + // UDFExampleAdd2 is slightly modified version of UDFExampleAdd in hive/contrib, + // which adds two integers or doubles. + UDFTestInformation( + "UDF", + "udf_add2", + "org.apache.hadoop.hive.contrib.udf.example.UDFExampleAdd2", + () => { + checkAnswer(sql("SELECT udf_add2(1, 2)"), Row(3) :: Nil) + }, + () => { + HiveSimpleUDF( + "default.udf_add2", + HiveFunctionWrapper("org.apache.hadoop.hive.contrib.udf.example.UDFExampleAdd2"), + Array( + AttributeReference("a", IntegerType, nullable = false)(), + AttributeReference("b", IntegerType, nullable = false)())) + }), + + // GenericUDF + // GenericUDFTrim2 is cloned version of GenericUDFTrim in hive/contrib. + UDFTestInformation( + "GENERIC_UDF", + "generic_udf_trim2", + "org.apache.hadoop.hive.contrib.udf.example.GenericUDFTrim2", + () => { + checkAnswer(sql("SELECT generic_udf_trim2(' hello ')"), Row("hello") :: Nil) + }, + () => { + HiveGenericUDF( + "default.generic_udf_trim2", + HiveFunctionWrapper("org.apache.hadoop.hive.contrib.udf.example.GenericUDFTrim2"), + Array(AttributeReference("a", StringType, nullable = false)()) + ) + } + ), + + // AbstractGenericUDAFResolver + // GenericUDAFSum2 is cloned version of GenericUDAFSum in hive/exec. + UDFTestInformation( + "GENERIC_UDAF", + "generic_udaf_sum2", + "org.apache.hadoop.hive.ql.udf.generic.GenericUDAFSum2", + () => { + import spark.implicits._ + val df = Seq((0: Integer) -> 0, (1: Integer) -> 1, (2: Integer) -> 2, (3: Integer) -> 3) + .toDF("key", "value").createOrReplaceTempView("t") + checkAnswer(sql("SELECT generic_udaf_sum2(value) FROM t GROUP BY key % 2"), + Row(2) :: Row(4) :: Nil) + }, + () => { + HiveUDAFFunction( + "default.generic_udaf_sum2", + HiveFunctionWrapper("org.apache.hadoop.hive.ql.udf.generic.GenericUDAFSum2"), + Array(AttributeReference("a", IntegerType, nullable = false)()) + ) + } + ), + + // UDAF + // UDAFExampleMax2 is cloned version of UDAFExampleMax in hive/contrib. + UDFTestInformation( + "UDAF", + "udaf_max2", + "org.apache.hadoop.hive.contrib.udaf.example.UDAFExampleMax2", + () => { + import spark.implicits._ + val df = Seq((0: Integer) -> 0, (1: Integer) -> 1, (2: Integer) -> 2, (3: Integer) -> 3) + .toDF("key", "value").createOrReplaceTempView("t") + checkAnswer(sql("SELECT udaf_max2(value) FROM t GROUP BY key % 2"), + Row(2) :: Row(3) :: Nil) + }, + () => { + HiveUDAFFunction( + "default.udaf_max2", + HiveFunctionWrapper("org.apache.hadoop.hive.contrib.udaf.example.UDAFExampleMax2"), + Array(AttributeReference("a", IntegerType, nullable = false)()), + isUDAFBridgeRequired = true + ) + } + ), + + // GenericUDTF + // GenericUDTFCount3 is slightly modified version of GenericUDTFCount2 in hive/contrib, + // which emits the count for three times. + UDFTestInformation( + "GENERIC_UDTF", + "udtf_count3", + "org.apache.hadoop.hive.contrib.udtf.example.GenericUDTFCount3", + () => { + checkAnswer( + sql("SELECT udtf_count3(a) FROM (SELECT 1 AS a FROM src LIMIT 3) t"), + Row(3) :: Row(3) :: Row(3) :: Nil) + }, + () => { + HiveGenericUDTF( + "default.udtf_count3", + HiveFunctionWrapper("org.apache.hadoop.hive.contrib.udtf.example.GenericUDTFCount3"), + Array.empty[Expression] + ) + } + ) + ) + + udfTestInfos.foreach { udfInfo => + val jarUrl = getHiveUDFTestJarUrl + test("SPARK-26560 Spark should be able to run Hive UDF using jar regardless of " + + s"current thread context classloader (${udfInfo.identifier}") { + testHiveUDFUsingJarWithChangingClassloader( + udfInfo.funcName, + jarUrl, + udfInfo.className, + udfInfo.fnVerifyQuery + ) + } + + test("SPARK-31312 Transformed Hive UDF using jar expression should not be failed to run " + + s"regardless of current thread context classloader (${udfInfo.identifier})") { + testHiveUDFUsingJarWithChangingClassloaderWithCopyUDFExpression( + udfInfo.funcName, + jarUrl, + udfInfo.className, + udfInfo.fnVerifyQuery, + udfInfo.fnCreateHiveUDFExpression) + } + } + + private def testHiveUDFUsingJarWithChangingClassloader( + funcName: String, + jarUrl: String, + className: String, + // this function must call Hive UDF to load JAR + fnVerifyQuery: () => Unit): Unit = { + // force to use Spark classloader as other test (even in other test suites) may change the + // current thread's context classloader to jar classloader + Utils.withContextClassLoader(Utils.getSparkClassLoader) { + withUserDefinedFunction(funcName -> false) { + val sparkClassLoader = Thread.currentThread().getContextClassLoader + + sql(s"CREATE FUNCTION $funcName AS '$className' USING JAR '$jarUrl'") + + assert(Thread.currentThread().getContextClassLoader eq sparkClassLoader) + + // JAR will be loaded at first usage, and it will change the current thread's + // context classloader to jar classloader in sharedState. + // See SessionState.addJar for details. + fnVerifyQuery() + + assert(Thread.currentThread().getContextClassLoader ne sparkClassLoader) + assert(Thread.currentThread().getContextClassLoader eq + spark.sqlContext.sharedState.jarClassLoader) + + // Roll back to the original classloader and run query again. Without this line, the test + // would pass, as thread's context classloader is changed to jar classloader. But thread + // context classloader can be changed from others as well which would fail the query; one + // example is spark-shell, which thread context classloader rolls back automatically. This + // mimics the behavior of spark-shell. + Thread.currentThread().setContextClassLoader(sparkClassLoader) + + fnVerifyQuery() + } + } + } + + private def testHiveUDFUsingJarWithChangingClassloaderWithCopyUDFExpression( + funcName: String, + jarUrl: String, + className: String, + // this function must call Hive UDF to load JAR + fnVerifyQuery: () => Unit, + fnCreateHiveUDFExpression: () => Expression): Unit = { + // force to use Spark classloader as other test (even in other test suites) may change the + // current thread's context classloader to jar classloader + Utils.withContextClassLoader(Utils.getSparkClassLoader) { + withUserDefinedFunction(funcName -> false) { + val sparkClassLoader = Thread.currentThread().getContextClassLoader + + sql(s"CREATE FUNCTION $funcName AS '$className' USING JAR '$jarUrl'") + + assert(Thread.currentThread().getContextClassLoader eq sparkClassLoader) + + // JAR will be loaded at first usage, and it will change the current thread's + // context classloader to jar classloader in sharedState. + // See SessionState.addJar for details. + fnVerifyQuery() + + assert(Thread.currentThread().getContextClassLoader ne sparkClassLoader) + assert(Thread.currentThread().getContextClassLoader eq + spark.sqlContext.sharedState.jarClassLoader) + + val udfExpr = fnCreateHiveUDFExpression() Review comment: I think this test should start here. The above test is the same as `testHiveUDFUsingJarWithChangingClassloader`. ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
