sunchao commented on code in PR #39687: URL: https://github.com/apache/spark/pull/39687#discussion_r1092296833
########## sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/util/InternalRowComparableWrapper.scala: ########## @@ -0,0 +1,84 @@ +/* + * 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.catalyst.util + +import scala.collection.mutable + +import org.apache.spark.SparkException +import org.apache.spark.sql.catalyst.InternalRow +import org.apache.spark.sql.catalyst.expressions.{Expression, Murmur3HashFunction, RowOrdering} +import org.apache.spark.sql.catalyst.plans.physical.KeyGroupedPartitioning +import org.apache.spark.sql.connector.read.{HasPartitionKey, InputPartition} +import org.apache.spark.sql.types.{DataType, StructField, StructType} + +/** + * Wraps the [[InternalRow]] with the corresponding [[DataType]] to make it can be compared with Review Comment: "to make it can be compared with" -> "to make it comparable with"? ########## sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/DataSourceV2ScanExecBase.scala: ########## @@ -124,28 +124,24 @@ trait DataSourceV2ScanExecBase extends LeafExecNode { inputPartitions: Seq[InputPartition]): Option[Seq[(InternalRow, Seq[InputPartition])]] = { if (!SQLConf.get.v2BucketingEnabled) return None keyGroupedPartitioning.flatMap { expressions => - val results = inputPartitions.takeWhile { - case _: HasPartitionKey => true - case _ => false - }.map(p => (p.asInstanceOf[HasPartitionKey].partitionKey(), p)) - - if (results.length != inputPartitions.length || inputPartitions.isEmpty) { + if (inputPartitions.isEmpty || inputPartitions.count(!_.isInstanceOf[HasPartitionKey]) > 0) { Review Comment: I think originally I chose `takeWhile` because it can quickly exit the loop if any input partition doesn't implement `HasPartitionKey`, but with `count` it now has to loop through all input partitions. ########## sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/util/InternalRowComparableWrapper.scala: ########## @@ -0,0 +1,84 @@ +/* + * 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.catalyst.util + +import scala.collection.mutable + +import org.apache.spark.SparkException +import org.apache.spark.sql.catalyst.InternalRow +import org.apache.spark.sql.catalyst.expressions.{Expression, Murmur3HashFunction, RowOrdering} +import org.apache.spark.sql.catalyst.plans.physical.KeyGroupedPartitioning +import org.apache.spark.sql.connector.read.{HasPartitionKey, InputPartition} +import org.apache.spark.sql.types.{DataType, StructField, StructType} + +/** + * Wraps the [[InternalRow]] with the corresponding [[DataType]] to make it can be compared with + * the values in [[InternalRow]]. + * It uses Spark's internal murmur hash to compute hash code from an row, and uses [[RowOrdering]] + * to perform equality checks. + * + * @param dataTypes the data types for the row + */ +class InternalRowComparableWrapper(val row: InternalRow, val dataTypes: Seq[DataType]) { + + private val structType = StructType(dataTypes.map(t => StructField("f", t))) + private val ordering = RowOrdering.createNaturalAscendingOrdering(dataTypes) + + override def hashCode(): Int = Murmur3HashFunction.hash(row, structType, 42L).toInt + + override def equals(other: Any): Boolean = { + if (!other.isInstanceOf[InternalRowComparableWrapper]) { + return false + } + val otherWrapper = other.asInstanceOf[InternalRowComparableWrapper] + if (!otherWrapper.dataTypes.equals(this.dataTypes)) { + return false + } + ordering.compare(row, otherWrapper.row) == 0 + } +} + +object InternalRowComparableWrapper { + + def apply(partition: InputPartition, + partitionExpression: Seq[Expression]): InternalRowComparableWrapper = { + if (!partition.isInstanceOf[HasPartitionKey]) { + throw new SparkException("partition row should implement `HasPartitionKey`") + } + new InternalRowComparableWrapper( + partition.asInstanceOf[HasPartitionKey].partitionKey(), partitionExpression.map(_.dataType)) + } + + def apply(partitionRow: InternalRow, + partitionExpression: Seq[Expression]): InternalRowComparableWrapper = { + new InternalRowComparableWrapper(partitionRow, partitionExpression.map(_.dataType)) + } + + def mergePartitions(leftPartition: KeyGroupedPartitioning, rightPartition: KeyGroupedPartitioning, Review Comment: ditto maybe rename parameters to `leftPartitioning` and `rightPartitioning` ########## sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/BatchScanExec.scala: ########## @@ -80,22 +79,21 @@ case class BatchScanExec( "during runtime filtering: not all partitions implement HasPartitionKey after " + "filtering") } - - val newRows = new InternalRowSet(p.expressions.map(_.dataType)) - newRows ++= newPartitions.map(_.asInstanceOf[HasPartitionKey].partitionKey()) - - val oldRows = p.partitionValues.toSet + val newPartitionKeys = newPartitions Review Comment: I think these should be called "partition values" instead of "partition keys". I made a mistake previously and it's better to correct them now. ########## sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/util/InternalRowComparableWrapper.scala: ########## @@ -0,0 +1,84 @@ +/* + * 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.catalyst.util + +import scala.collection.mutable + +import org.apache.spark.SparkException +import org.apache.spark.sql.catalyst.InternalRow +import org.apache.spark.sql.catalyst.expressions.{Expression, Murmur3HashFunction, RowOrdering} +import org.apache.spark.sql.catalyst.plans.physical.KeyGroupedPartitioning +import org.apache.spark.sql.connector.read.{HasPartitionKey, InputPartition} +import org.apache.spark.sql.types.{DataType, StructField, StructType} + +/** + * Wraps the [[InternalRow]] with the corresponding [[DataType]] to make it can be compared with + * the values in [[InternalRow]]. + * It uses Spark's internal murmur hash to compute hash code from an row, and uses [[RowOrdering]] + * to perform equality checks. + * + * @param dataTypes the data types for the row + */ +class InternalRowComparableWrapper(val row: InternalRow, val dataTypes: Seq[DataType]) { + + private val structType = StructType(dataTypes.map(t => StructField("f", t))) + private val ordering = RowOrdering.createNaturalAscendingOrdering(dataTypes) + + override def hashCode(): Int = Murmur3HashFunction.hash(row, structType, 42L).toInt + + override def equals(other: Any): Boolean = { + if (!other.isInstanceOf[InternalRowComparableWrapper]) { + return false + } + val otherWrapper = other.asInstanceOf[InternalRowComparableWrapper] + if (!otherWrapper.dataTypes.equals(this.dataTypes)) { + return false + } + ordering.compare(row, otherWrapper.row) == 0 + } +} + +object InternalRowComparableWrapper { + + def apply(partition: InputPartition, + partitionExpression: Seq[Expression]): InternalRowComparableWrapper = { + if (!partition.isInstanceOf[HasPartitionKey]) { + throw new SparkException("partition row should implement `HasPartitionKey`") + } + new InternalRowComparableWrapper( + partition.asInstanceOf[HasPartitionKey].partitionKey(), partitionExpression.map(_.dataType)) + } + + def apply(partitionRow: InternalRow, + partitionExpression: Seq[Expression]): InternalRowComparableWrapper = { + new InternalRowComparableWrapper(partitionRow, partitionExpression.map(_.dataType)) + } + + def mergePartitions(leftPartition: KeyGroupedPartitioning, rightPartition: KeyGroupedPartitioning, + partitionExpression: Seq[Expression]): Seq[InternalRow] = { Review Comment: It's a bit strange that we have `mergePartitions`, which is only related to `KeyGroupedPartitioning`, in the `InternalRowComparableWrapper` class. Could we move this method to some place more relevant, like `KeyGroupedShuffleSpec`? ########## sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/util/InternalRowComparableWrapper.scala: ########## @@ -0,0 +1,84 @@ +/* + * 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.catalyst.util + +import scala.collection.mutable + +import org.apache.spark.SparkException +import org.apache.spark.sql.catalyst.InternalRow +import org.apache.spark.sql.catalyst.expressions.{Expression, Murmur3HashFunction, RowOrdering} +import org.apache.spark.sql.catalyst.plans.physical.KeyGroupedPartitioning +import org.apache.spark.sql.connector.read.{HasPartitionKey, InputPartition} +import org.apache.spark.sql.types.{DataType, StructField, StructType} + +/** + * Wraps the [[InternalRow]] with the corresponding [[DataType]] to make it can be compared with + * the values in [[InternalRow]]. + * It uses Spark's internal murmur hash to compute hash code from an row, and uses [[RowOrdering]] + * to perform equality checks. + * + * @param dataTypes the data types for the row + */ +class InternalRowComparableWrapper(val row: InternalRow, val dataTypes: Seq[DataType]) { + + private val structType = StructType(dataTypes.map(t => StructField("f", t))) + private val ordering = RowOrdering.createNaturalAscendingOrdering(dataTypes) + + override def hashCode(): Int = Murmur3HashFunction.hash(row, structType, 42L).toInt + + override def equals(other: Any): Boolean = { + if (!other.isInstanceOf[InternalRowComparableWrapper]) { + return false + } + val otherWrapper = other.asInstanceOf[InternalRowComparableWrapper] + if (!otherWrapper.dataTypes.equals(this.dataTypes)) { + return false + } + ordering.compare(row, otherWrapper.row) == 0 + } +} + +object InternalRowComparableWrapper { + + def apply(partition: InputPartition, + partitionExpression: Seq[Expression]): InternalRowComparableWrapper = { + if (!partition.isInstanceOf[HasPartitionKey]) { + throw new SparkException("partition row should implement `HasPartitionKey`") + } + new InternalRowComparableWrapper( + partition.asInstanceOf[HasPartitionKey].partitionKey(), partitionExpression.map(_.dataType)) + } + + def apply(partitionRow: InternalRow, Review Comment: ditto ########## sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/util/InternalRowComparableWrapper.scala: ########## @@ -0,0 +1,84 @@ +/* + * 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.catalyst.util + +import scala.collection.mutable + +import org.apache.spark.SparkException +import org.apache.spark.sql.catalyst.InternalRow +import org.apache.spark.sql.catalyst.expressions.{Expression, Murmur3HashFunction, RowOrdering} +import org.apache.spark.sql.catalyst.plans.physical.KeyGroupedPartitioning +import org.apache.spark.sql.connector.read.{HasPartitionKey, InputPartition} +import org.apache.spark.sql.types.{DataType, StructField, StructType} + +/** + * Wraps the [[InternalRow]] with the corresponding [[DataType]] to make it can be compared with + * the values in [[InternalRow]]. + * It uses Spark's internal murmur hash to compute hash code from an row, and uses [[RowOrdering]] + * to perform equality checks. + * + * @param dataTypes the data types for the row + */ +class InternalRowComparableWrapper(val row: InternalRow, val dataTypes: Seq[DataType]) { + + private val structType = StructType(dataTypes.map(t => StructField("f", t))) + private val ordering = RowOrdering.createNaturalAscendingOrdering(dataTypes) + + override def hashCode(): Int = Murmur3HashFunction.hash(row, structType, 42L).toInt + + override def equals(other: Any): Boolean = { + if (!other.isInstanceOf[InternalRowComparableWrapper]) { + return false + } + val otherWrapper = other.asInstanceOf[InternalRowComparableWrapper] + if (!otherWrapper.dataTypes.equals(this.dataTypes)) { + return false + } + ordering.compare(row, otherWrapper.row) == 0 + } +} + +object InternalRowComparableWrapper { + + def apply(partition: InputPartition, Review Comment: nit: put each parameter in a separate line following the style guide: https://github.com/databricks/scala-style-guide#spacing-and-indentation, e.g.: ```scala def apply( partition: InputPartition, partitionExpression: Seq[Expression]): InternalRowComparableWrapper = { .. } ``` ########## sql/catalyst/src/main/scala-2.12/org/apache/spark/sql/catalyst/util/InternalRowSet.scala: ########## @@ -1,65 +0,0 @@ -/* Review Comment: Should we also remove the same file under `scala-2.13`? -- 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]
