huaxingao commented on a change in pull request #29396: URL: https://github.com/apache/spark/pull/29396#discussion_r470898996
########## File path: sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/jdbc/JDBCScan.scala ########## @@ -0,0 +1,50 @@ +/* + * 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.jdbc + +import org.apache.spark.rdd.RDD +import org.apache.spark.sql.{Row, SQLContext} +import org.apache.spark.sql.connector.read.V1Scan +import org.apache.spark.sql.execution.datasources.jdbc.JDBCRelation +import org.apache.spark.sql.sources.{BaseRelation, Filter, TableScan} +import org.apache.spark.sql.types.StructType + +case class JDBCScan( + relation: JDBCRelation, + prunedSchema: StructType, + pushedFilters: Array[Filter]) extends V1Scan { Review comment: Using ```V1Scan``` here for now because we are still calling V1 ```JDBCRDD.scanTable``` underneath. Migrating step by step. ########## File path: sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/jdbc/JDBCScanBuilder.scala ########## @@ -0,0 +1,70 @@ +/* + * 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.jdbc + +import org.apache.spark.sql.SparkSession +import org.apache.spark.sql.connector.read.{Scan, ScanBuilder, SupportsPushDownFilters, SupportsPushDownRequiredColumns} +import org.apache.spark.sql.execution.datasources.PartitioningUtils +import org.apache.spark.sql.execution.datasources.jdbc.{JDBCOptions, JDBCRDD, JDBCRelation} +import org.apache.spark.sql.jdbc.JdbcDialects +import org.apache.spark.sql.sources.Filter +import org.apache.spark.sql.types.StructType + +case class JDBCScanBuilder( + session: SparkSession, + schema: StructType, + jdbcOptions: JDBCOptions) + extends ScanBuilder with SupportsPushDownFilters with SupportsPushDownRequiredColumns { + + private val isCaseSensitive = session.sessionState.conf.caseSensitiveAnalysis + + private var pushedFilter = Array.empty[Filter] + + private var prunedSchema = schema + + override def pushFilters(filters: Array[Filter]): Array[Filter] = { + if (jdbcOptions.pushDownPredicate) { + val dialect = JdbcDialects.get(jdbcOptions.url) + val (pushed, unSupported) = filters.partition(JDBCRDD.compileFilter(_, dialect).isDefined) + this.pushedFilter = pushed + unSupported Review comment: Thanks for your comment. I agree that it's safer to return the original filters, but it seems to me that we want to push down filters to the underlying datasource for better performance, so I guess we don't want to return the original filter to re-evaluate the filters that have already evaluated in the datasources. ########## File path: sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/jdbc/JDBCScanBuilder.scala ########## @@ -0,0 +1,70 @@ +/* + * 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.jdbc + +import org.apache.spark.sql.SparkSession +import org.apache.spark.sql.connector.read.{Scan, ScanBuilder, SupportsPushDownFilters, SupportsPushDownRequiredColumns} +import org.apache.spark.sql.execution.datasources.PartitioningUtils +import org.apache.spark.sql.execution.datasources.jdbc.{JDBCOptions, JDBCRDD, JDBCRelation} +import org.apache.spark.sql.jdbc.JdbcDialects +import org.apache.spark.sql.sources.Filter +import org.apache.spark.sql.types.StructType + +case class JDBCScanBuilder( + session: SparkSession, + schema: StructType, + jdbcOptions: JDBCOptions) + extends ScanBuilder with SupportsPushDownFilters with SupportsPushDownRequiredColumns { + + private val isCaseSensitive = session.sessionState.conf.caseSensitiveAnalysis + + private var pushedFilter = Array.empty[Filter] + + private var prunedSchema = schema + + override def pushFilters(filters: Array[Filter]): Array[Filter] = { Review comment: Seems JDBC doesn't support nested columns yet. In JdbcUtils.getCatalystType, we map `java.sql.Types.ARRAY` to null and `java.sql.Types.STRUCT` to `StringType`. In `JdbcUtils.getJDBCType`, we don't have a mapping for `ArrayType` and `StructType`. I think we don't need to consider nested column pruning and nested column predicates for now. I have open jira SPARK-32593 for nested column support. ########## File path: sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/jdbc/JDBCWriteBuilder.scala ########## @@ -0,0 +1,46 @@ +/* + * 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.jdbc + +import org.apache.spark.sql._ +import org.apache.spark.sql.connector.write._ +import org.apache.spark.sql.execution.datasources.jdbc.{JdbcOptionsInWrite, JdbcUtils} +import org.apache.spark.sql.internal.SQLConf +import org.apache.spark.sql.sources.InsertableRelation +import org.apache.spark.sql.types.StructType + +case class JDBCWriteBuilder(schema: StructType, options: JdbcOptionsInWrite) extends V1WriteBuilder + with SupportsTruncate { + + private var isTruncate = false + + override def truncate(): WriteBuilder = { + isTruncate = true + this + } + + override def buildForV1Write(): InsertableRelation = new InsertableRelation { + override def insert(data: DataFrame, overwrite: Boolean): Unit = { Review comment: Seems `overwrite` can't be true according to the `V1WriteBuilder` doc. ---------------------------------------------------------------- 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]
