cloud-fan commented on a change in pull request #29359: URL: https://github.com/apache/spark/pull/29359#discussion_r466240736
########## File path: sql/core/src/test/scala/org/apache/spark/sql/CTEHintSuite.scala ########## @@ -0,0 +1,158 @@ +/* + * 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 + +import org.apache.log4j.Level + +import org.apache.spark.sql.catalyst.plans.PlanTest +import org.apache.spark.sql.catalyst.plans.logical.{BROADCAST, HintInfo, Join, JoinHint, Repartition, RepartitionByExpression, ResolvedHint, SHUFFLE_HASH, SHUFFLE_MERGE, SHUFFLE_REPLICATE_NL} +import org.apache.spark.sql.execution.adaptive.AdaptiveSparkPlanHelper +import org.apache.spark.sql.test.SharedSparkSession + +class CTEHintSuite extends PlanTest with SharedSparkSession with AdaptiveSparkPlanHelper { + + def verifyCoalesceHint(df: DataFrame): Unit = { + val optimized = df.queryExecution.optimizedPlan + val repartitions = optimized collect { + case r: Repartition => r + case r: RepartitionByExpression => r + case _: ResolvedHint => fail("ResolvedHint should not appear after optimize.") + } + assert(repartitions.nonEmpty) + } + + def verifyJoinHint(df: DataFrame, expectedHints: Seq[JoinHint]): Unit = { + val optimized = df.queryExecution.optimizedPlan + val joinHints = optimized collect { + case Join(_, _, _, _, hint) => hint + case _: ResolvedHint => fail("ResolvedHint should not appear after optimize.") + } + assert(joinHints == expectedHints) + } + + def verifyJoinHintWithWarnings( + df: => DataFrame, + expectedHints: Seq[JoinHint], + warnings: Seq[String]): Unit = { + val logAppender = new LogAppender("join hints") + withLogAppender(logAppender) { + verifyJoinHint(df, expectedHints) + } + val warningMessages = logAppender.loggingEvents + .filter(_.getLevel == Level.WARN) + .map(_.getRenderedMessage) + .filter(_.contains("hint")) + assert(warningMessages.size == warnings.size) + warnings.foreach { w => + assert(warningMessages.contains(w)) + } + } + + def msgNoJoinForJoinHint(strategy: String): String = + s"A join hint (strategy=$strategy) is specified but it is not part of a join relation." + + def msgJoinHintOverridden(strategy: String): String = + s"Hint (strategy=$strategy) is overridden by another hint and will not take effect." + + test("Resolve coalesce hint in CTE") { + // COALESCE, + // REPARTITION, + // REPARTITION_BY_RANGE + withTable("t") { + sql("CREATE TABLE t USING PARQUET AS SELECT 1 AS id") + verifyCoalesceHint( + sql("WITH cte AS (SELECT /*+ COALESCE(1) */ * FROM t) SELECT * FROM cte")) + verifyCoalesceHint( + sql("WITH cte AS (SELECT /*+ REPARTITION(3) */ * FROM t) SELECT * FROM cte")) + verifyCoalesceHint( + sql("WITH cte AS (SELECT /*+ REPARTITION(id) */ * FROM t) SELECT * FROM cte")) + verifyCoalesceHint( + sql("WITH cte AS (SELECT /*+ REPARTITION(3, id) */ * FROM t) SELECT * FROM cte")) + verifyCoalesceHint( + sql("WITH cte AS (SELECT /*+ REPARTITION_BY_RANGE(id) */ * FROM t) SELECT * FROM cte")) + verifyCoalesceHint( + sql("WITH cte AS (SELECT /*+ REPARTITION_BY_RANGE(3, id) */ * FROM t) SELECT * FROM cte")) + } + } + + test("Resolve join hint in CTE") { + // BROADCAST, + // SHUFFLE_MERGE, + // SHUFFLE_HASH, + // SHUFFLE_REPLICATE_NL + withTable("t", "s") { + sql("CREATE TABLE a USING PARQUET AS SELECT 1 AS a1") + sql("CREATE TABLE b USING PARQUET AS SELECT 1 AS b1") + sql("CREATE TABLE c USING PARQUET AS SELECT 1 AS c1") + verifyJoinHint( + sql( + """ + |WITH cte AS ( + | SELECT /*+ BROADCAST(a) */ * FROM a JOIN b ON a.a1 = b.b1 + |) + |SELECT * FROM cte + |""".stripMargin), + JoinHint( + Some(HintInfo(strategy = Some(BROADCAST))), + None) :: Nil + ) + verifyJoinHint( + sql( + """ + |WITH cte AS ( + | SELECT /*+ SHUFFLE_HASH(a) */ * FROM a JOIN b ON a.a1 = b.b1 + |) + |SELECT * FROM cte + |""".stripMargin), + JoinHint( + Some(HintInfo(strategy = Some(SHUFFLE_HASH))), + None) :: Nil + ) + verifyJoinHintWithWarnings( + sql( + """ + |WITH cte AS ( + | SELECT /*+ SHUFFLE_HASH MERGE(a, c) BROADCAST(a, b)*/ * FROM a, b, c + | WHERE a.a1 = b.b1 AND b.b1 = c.c1 + |) + |SELECT * FROM cte + |""".stripMargin), + JoinHint( + None, + Some(HintInfo(strategy = Some(SHUFFLE_MERGE)))) :: + JoinHint( + Some(HintInfo(strategy = Some(SHUFFLE_MERGE))), + Some(HintInfo(strategy = Some(BROADCAST)))) :: Nil, + msgNoJoinForJoinHint("shuffle_hash") :: + msgJoinHintOverridden("broadcast") :: Nil + ) + verifyJoinHint( + sql( + """ + |WITH cte AS ( + | SELECT /*+ SHUFFLE_REPLICATE_NL(a) SHUFFLE_HASH(b) */ * FROM a JOIN b ON a.a1 = b.b1 + |) + |SELECT * FROM cte + |""".stripMargin), + JoinHint( + Some(HintInfo(strategy = Some(SHUFFLE_REPLICATE_NL))), + Some(HintInfo(strategy = Some(SHUFFLE_HASH)))) :: Nil + ) + } + } +} Review comment: can we move https://github.com/apache/spark/pull/29062/files#diff-5a2e7f03d14856c8769fd3ddea8742bdR3564 to this new suite? ---------------------------------------------------------------- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
