minni31 commented on code in PR #12077: URL: https://github.com/apache/gluten/pull/12077#discussion_r3227452263
########## backends-velox/src/test/scala/org/apache/gluten/execution/VeloxRDDScanSuite.scala: ########## @@ -0,0 +1,189 @@ +/* + * 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.execution + +import org.apache.gluten.execution._ + +import org.apache.spark.SparkConf +import org.apache.spark.sql.Row +import org.apache.spark.sql.classic.ClassicDataset +import org.apache.spark.sql.execution.adaptive.AdaptiveSparkPlanHelper +import org.apache.spark.sql.types._ + +class VeloxRDDScanSuite extends VeloxWholeStageTransformerSuite with AdaptiveSparkPlanHelper { + + override protected val resourcePath: String = "/tpch-data-parquet" + override protected val fileFormat: String = "parquet" + + override protected def sparkConf: SparkConf = { + super.sparkConf + .set("spark.sql.ansi.enabled", "false") + } + + override def beforeAll(): Unit = { + super.beforeAll() + createTPCHNotNullTables() + } + + test("basic RDDScanExec is replaced by VeloxRDDScanTransformer") { + val data = spark.sql("SELECT l_orderkey, l_partkey FROM lineitem LIMIT 10") + val expectedAnswer = data.collect() + + val node = LogicalRDD.fromDataset( + rdd = data.queryExecution.toRdd, + originDataset = data, + isStreaming = false) + val df = ClassicDataset.ofRows(spark, node).toDF() + + checkAnswer(df, expectedAnswer) + val cnt = collect(df.queryExecution.executedPlan) { case _: VeloxRDDScanTransformer => true } + assert(cnt.nonEmpty, "Expected VeloxRDDScanTransformer in plan") + } + + test("RDDScan with string and numeric types") { + val data = spark.sql("""SELECT l_returnflag, l_linestatus, l_quantity, l_extendedprice + |FROM lineitem LIMIT 20""".stripMargin) + val expectedAnswer = data.collect() + + val node = LogicalRDD.fromDataset( + rdd = data.queryExecution.toRdd, + originDataset = data, + isStreaming = false) + val df = ClassicDataset.ofRows(spark, node).toDF() + + checkAnswer(df, expectedAnswer) + val cnt = collect(df.queryExecution.executedPlan) { case _: VeloxRDDScanTransformer => true } + assert(cnt.nonEmpty, "Expected VeloxRDDScanTransformer in plan") + } + + test("RDDScan with aggregation downstream") { + val query = + """SELECT l_returnflag, sum(l_quantity) AS sum_qty + |FROM lineitem + |WHERE l_shipdate <= date'1998-09-02' + |GROUP BY l_returnflag""".stripMargin + val data = spark.sql(query) + val expectedAnswer = data.collect() + + val node = LogicalRDD.fromDataset( + rdd = data.queryExecution.toRdd, + originDataset = data, + isStreaming = false) + val df = ClassicDataset.ofRows(spark, node).toDF() + + checkAnswer(df, expectedAnswer) + } + + test("RDDScan with empty RDD") { + val data = spark.sql("SELECT l_orderkey FROM lineitem WHERE 1 = 0") + val expectedAnswer = data.collect() + + val node = LogicalRDD.fromDataset( + rdd = data.queryExecution.toRdd, + originDataset = data, + isStreaming = false) + val df = ClassicDataset.ofRows(spark, node).toDF() + + checkAnswer(df, expectedAnswer) + assert(df.count() == 0) + } + + test("RDDScan preserves data correctness with multiple re-reads") { + val data = spark.sql("SELECT l_orderkey, l_partkey FROM lineitem LIMIT 50") + val expectedAnswer = data.collect() + + val node = LogicalRDD.fromDataset( + rdd = data.queryExecution.toRdd, + originDataset = data, + isStreaming = false) + val df = ClassicDataset.ofRows(spark, node).toDF() + + // Read twice to verify idempotency + checkAnswer(df, expectedAnswer) + checkAnswer(df, expectedAnswer) + } + + test("RDDScan with null values") { + val rdd = spark.sparkContext.parallelize( + Seq( + Row(1, "a", null), + Row(null, "b", 2.0), + Row(3, null, 3.0) + )) + val schema = StructType( + Seq( + StructField("id", IntegerType, nullable = true), + StructField("name", StringType, nullable = true), + StructField("value", DoubleType, nullable = true) + )) + val data = spark.createDataFrame(rdd, schema) + val expectedAnswer = data.collect() + + val node = LogicalRDD.fromDataset( + rdd = data.queryExecution.toRdd, + originDataset = data, + isStreaming = false) + val df = ClassicDataset.ofRows(spark, node).toDF() + + checkAnswer(df, expectedAnswer) + } + + test("RDDScan with all supported primitive types") { + val rdd = spark.sparkContext.parallelize( + Seq( + Row( + true, + 1.toByte, + 2.toShort, + 3, + 4L, + 5.0f, + 6.0, + "hello", + java.sql.Date.valueOf("2024-01-01"), + java.sql.Timestamp.valueOf("2024-01-01 12:00:00"), + Array[Byte](1, 2, 3), + BigDecimal("123.45").underlying() + ) + )) + val schema = StructType( + Seq( + StructField("bool", BooleanType), + StructField("byte", ByteType), + StructField("short", ShortType), + StructField("int", IntegerType), + StructField("long", LongType), + StructField("float", FloatType), + StructField("double", DoubleType), + StructField("string", StringType), + StructField("date", DateType), + StructField("timestamp", TimestampType), + StructField("binary", BinaryType), + StructField("decimal", DecimalType(10, 2)) + )) + val data = spark.createDataFrame(rdd, schema) + val expectedAnswer = data.collect() + + val node = LogicalRDD.fromDataset( + rdd = data.queryExecution.toRdd, + originDataset = data, + isStreaming = false) + val df = ClassicDataset.ofRows(spark, node).toDF() + + checkAnswer(df, expectedAnswer) + } +} Review Comment: Added 4 new tests: array type, map type, struct type, and unsupported-type fallback (DayTimeIntervalType → verifies VeloxRDDScanTransformer is absent from plan). Total coverage is now 11 tests. -- 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. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
