sunchao commented on a change in pull request #32931: URL: https://github.com/apache/spark/pull/32931#discussion_r654595568
########## File path: sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/ShowCreateTableExec.scala ########## @@ -0,0 +1,172 @@ +/* + * 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 scala.collection.mutable + +import org.apache.spark.sql.AnalysisException +import org.apache.spark.sql.catalyst.InternalRow +import org.apache.spark.sql.catalyst.catalog.BucketSpec +import org.apache.spark.sql.catalyst.expressions.Attribute +import org.apache.spark.sql.catalyst.util.escapeSingleQuotedString +import org.apache.spark.sql.connector.catalog.{CatalogV2Util, Table, TableCatalog} +import org.apache.spark.sql.connector.expressions.{BucketTransform, FieldReference, IdentityTransform} +import org.apache.spark.sql.errors.QueryExecutionErrors +import org.apache.spark.sql.execution.LeafExecNode +import org.apache.spark.sql.execution.command.DDLUtils +import org.apache.spark.unsafe.types.UTF8String + +/** + * Physical plan node for show create table. + */ +case class ShowCreateTableExec( + output: Seq[Attribute], + table: Table, + asSerde: Boolean = false) extends V2CommandExec with LeafExecNode { + override protected def run(): Seq[InternalRow] = { + val builder = StringBuilder.newBuilder + if (isView(table)) { + // It is better to use `ShowCreateViewExec` independent from `ShowCreateTableExec` + throw new AnalysisException("SHOW CREATE TABLE is not supported for V2 Views.") + } else { + if (asSerde) { + throw new AnalysisException("SHOW CREATE TABLE is not supported for v2 tables of" + + " Hive Serde table to generate the Hive DDL.") + } else { + // it is used to generate Spark DDL for given table. include Hive Serde table + showCreateTable(table, builder) + } + } + + Seq(InternalRow(UTF8String.fromString(builder.toString))) + } + + private def showCreateTable(table: Table, builder: StringBuilder): Unit = { + builder ++= s"CREATE TABLE ${table.name()}" + val isHive = DDLUtils.isHiveTable(Option(table.properties.get(TableCatalog.PROP_PROVIDER))) + if (isHive) { + throw new AnalysisException("SHOW CREATE TABLE is not supported for Hive serde table" + + " to generate the Spark DDL.") + } + showTableDataColumns(table, builder) + showTableUsing(table, builder) + showTableOptions(table, builder) + showTableNonDataColumns(table, builder) + showTableComment(table, builder) + showTableLocation(table, builder) + showTableProperties(table, builder) + } + + private def showTableDataColumns(table: Table, builder: StringBuilder): Unit = { + val columns = table.schema().fields.map(_.toDDL) + builder ++= concatByMultiLines(columns) + } + + private def showTableUsing(table: Table, builder: StringBuilder): Unit = { + if (table.properties.containsKey(TableCatalog.PROP_PROVIDER)) { + builder ++= + s"USING ${escapeSingleQuotedString(table.properties().get(TableCatalog.PROP_PROVIDER))}\n" + } + } + + private def showTableOptions(table: Table, builder: StringBuilder): Unit = { + import scala.collection.JavaConverters._ + val dataSourceOptions = table.properties.asScala + .filterKeys(_.startsWith(TableCatalog.OPTION_PREFIX)) + if (dataSourceOptions.nonEmpty) { + val props = dataSourceOptions.map { case (key, value) => + s"'${escapeSingleQuotedString(key)}' = '${escapeSingleQuotedString(value)}'" + } + + builder ++= "OPTIONS " + builder ++= concatByMultiLines(props) + } + } + + private def showTableNonDataColumns(table: Table, builder: StringBuilder): Unit = { + if (!table.partitioning.isEmpty) { + val identityCols = new mutable.ArrayBuffer[String] + var bucketSpec = Option.empty[BucketSpec] + + table.partitioning.map { + case IdentityTransform(FieldReference(Seq(col))) => + identityCols += col + case t : BucketTransform => + bucketSpec = Some(BucketSpec(t.numBuckets.value, t.columns.map(col => col.toString), Nil)) + case transform => + throw QueryExecutionErrors.unsupportedPartitionTransformError(transform) + } + if (identityCols.nonEmpty) { + builder ++= s"PARTITIONED BY ${identityCols.mkString("(", ", ", ")")}\n" Review comment: @Peng-Lei yes in DSv2 bucket is just a special type of partitioning scheme, so `bucket` in either `CLUSTERED BY` or `PARTITIONED BY` will yield to the same result. Because of this, IMO I think we should convert it to `PARTITIONED BY` for the purpose of consistency. -- 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]
