huaxingao commented on a change in pull request #35395: URL: https://github.com/apache/spark/pull/35395#discussion_r805292241
########## File path: sql/catalyst/src/main/java/org/apache/spark/sql/connector/write/RowLevelOperation.java ########## @@ -0,0 +1,92 @@ +/* + * 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.connector.write; + +import org.apache.spark.annotation.Experimental; +import org.apache.spark.sql.connector.expressions.NamedReference; +import org.apache.spark.sql.connector.read.Scan; +import org.apache.spark.sql.connector.read.ScanBuilder; +import org.apache.spark.sql.util.CaseInsensitiveStringMap; + +/** + * A logical representation of a data source DELETE, UPDATE, or MERGE operation that requires + * rewriting data. + * + * @since 3.3.0 + */ +@Experimental +public interface RowLevelOperation { + + /** + * The SQL row-level operation. + */ + enum Command { + DELETE, UPDATE, MERGE + } + + /** + * Returns the description associated with this row-level operation. + */ + default String description() { + return this.getClass().toString(); + } + + /** + * Returns the actual SQL operation being performed. + */ + Command command(); + + /** + * Returns a {@link ScanBuilder} to configure a {@link Scan} for this row-level operation. + * <p> + * Sources fall into two categories: those that can handle a delta of rows and those that need + * to replace groups (e.g. partitions, files). Sources that handle deltas allow Spark to quickly + * discard unchanged rows and have no requirements for input scans. Sources that replace groups + * of rows can discard deleted rows but need to keep unchanged rows to be passed back into + * the source. This means that scans for such data data sources must produce all rows in a group Review comment: nit: `data data` typo? ########## File path: sql/catalyst/src/main/java/org/apache/spark/sql/connector/write/RowLevelOperationInfo.java ########## @@ -0,0 +1,41 @@ +/* + * 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.connector.write; + +import org.apache.spark.annotation.Experimental; +import org.apache.spark.sql.connector.write.RowLevelOperation.Command; +import org.apache.spark.sql.util.CaseInsensitiveStringMap; + +/** + * An interface with logical information for a row-level operation such as DELETE or MERGE. Review comment: nit: shall we also put `UPDATE` here? ########## File path: sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/DataSourceV2Strategy.scala ########## @@ -255,7 +255,13 @@ class DataSourceV2Strategy(session: SparkSession) extends Strategy with Predicat case OverwritePartitionsDynamic(r: DataSourceV2Relation, query, _, _, Some(write)) => OverwritePartitionsDynamicExec(planLater(query), refreshCache(r), write) :: Nil - case DeleteFromTable(relation, condition) => + case DeleteFromTableWithFilters(r: DataSourceV2Relation, filters) => + DeleteFromTableExec(r.table.asDeletable, filters.toArray, refreshCache(r)) :: Nil + + case DeleteFromTable(_, _, Some(rewritePlan)) => + planLater(rewritePlan) :: Nil + + case DeleteFromTable(relation, condition, None) => Review comment: Is this the same as `DeleteFromTableWithFilters`? Is it possible to combine these two? ########## File path: sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/OptimizeMetadataOnlyDeleteFromTable.scala ########## @@ -0,0 +1,63 @@ +/* + * 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.datasources.v2 + +import org.apache.spark.sql.catalyst.expressions.{Expression, Literal, PredicateHelper, SubqueryExpression} +import org.apache.spark.sql.catalyst.plans.logical.{DeleteFromTable, DeleteFromTableWithFilters, LogicalPlan} +import org.apache.spark.sql.catalyst.rules.Rule +import org.apache.spark.sql.connector.catalog.SupportsDelete +import org.apache.spark.sql.execution.datasources.DataSourceStrategy +import org.apache.spark.sql.sources + +/** + * Checks whether a delete using filters is possible and nullifies the rewrite plan + * if the source can handle this delete without executing the rewrite plan. + * + * Note this rule must be run after expression optimization. + */ +object OptimizeMetadataOnlyDeleteFromTable extends Rule[LogicalPlan] with PredicateHelper { Review comment: Do we have a test for this new rule? ########## File path: sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/RowLevelCommandScanRelationPushDown.scala ########## @@ -0,0 +1,81 @@ +/* + * 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.datasources.v2 + +import org.apache.spark.sql.catalyst.expressions.{AttributeReference, AttributeSet, Expression, PredicateHelper, SubqueryExpression} +import org.apache.spark.sql.catalyst.planning.RewrittenRowLevelCommand +import org.apache.spark.sql.catalyst.plans.logical.LogicalPlan +import org.apache.spark.sql.catalyst.rules.Rule +import org.apache.spark.sql.connector.expressions.filter.{Filter => V2Filter} +import org.apache.spark.sql.connector.read.ScanBuilder +import org.apache.spark.sql.execution.datasources.DataSourceStrategy +import org.apache.spark.sql.sources.Filter + +object RowLevelCommandScanRelationPushDown extends Rule[LogicalPlan] with PredicateHelper { + import DataSourceV2Implicits._ + + override def apply(plan: LogicalPlan): LogicalPlan = plan transformDown { + // push down the filter from the command condition instead of the filter in the rewrite plan, + // which may be negated for sources that only support replacing groups of data (e.g. files) Review comment: is it possible to push down the negated filter in the rewrite plan? -- 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]
