AnishMahto commented on code in PR #55836: URL: https://github.com/apache/spark/pull/55836#discussion_r3244449137
########## sql/pipelines/src/main/scala/org/apache/spark/sql/pipelines/autocdc/ChangeArgs.scala: ########## @@ -0,0 +1,124 @@ +/* + * 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.pipelines.autocdc + +import org.apache.spark.sql.{AnalysisException, Column} +import org.apache.spark.sql.catalyst.parser.CatalystSqlParser +import org.apache.spark.sql.types.StructType + +/** + * A column reference that must be a single, unqualified identifier (no nested field path and + * no table/alias qualifier). The constructor parses [[name]] with the Spark SQL parser and + * throws an [[AnalysisException]] if it does not resolve to exactly one name part. + */ +case class UnqualifiedColumnName(name: String) { + UnqualifiedColumnName.validate(name) +} + +object UnqualifiedColumnName { + private def validate(columnName: String): Unit = { + val nameParts = CatalystSqlParser.parseMultipartIdentifier(columnName) + if (nameParts.length != 1) { + throw multipartColumnIdentifierError(columnName, nameParts) + } + } + + private def multipartColumnIdentifierError( + columnName: String, + nameParts: Seq[String] + ): AnalysisException = + new AnalysisException( + errorClass = "AUTOCDC_INVALID_COLUMN_SELECTION.MULTIPART_COLUMN_IDENTIFIER", + messageParameters = Map( + "columnName" -> columnName, + "nameParts" -> nameParts.mkString(", ") + ) + ) +} + +sealed trait ColumnSelection +object ColumnSelection { + type ColumnList = Seq[UnqualifiedColumnName] + + case class IncludeColumns(columns: ColumnList) extends ColumnSelection + case class ExcludeColumns(columns: ColumnList) extends ColumnSelection + + /** + * Applies [[ColumnSelection]] to a [[StructType]] and returns the filtered schema. + * Field names are matched exactly. Field order follows the original schema (filtered in place). + */ + def applyToSchema(schema: StructType, columnSelection: Option[ColumnSelection]): StructType = + columnSelection match { + case None => + // A none column selection is interpreted as a no-op. + schema + case Some(IncludeColumns(includeColumns)) => + validateColumnsExistInSchema(includeColumns, schema) + + val includeColumnSet = includeColumns.map(_.name).toSet + StructType(schema.fields.filter(f => includeColumnSet.contains(f.name))) + case Some(ExcludeColumns(excludeColumns)) => + validateColumnsExistInSchema(excludeColumns, schema) + + val excludeColumnSet = excludeColumns.map(_.name).toSet + StructType(schema.fields.filterNot(f => excludeColumnSet.contains(f.name))) + } + + private def validateColumnsExistInSchema(columns: ColumnList, schema: StructType): Unit = { + val schemaColumns = schema.fieldNames.toSet + val missingColumns = columns.map(_.name).filterNot(schemaColumns.contains).distinct Review Comment: Choosing not to do this, feels a bit like premature optimization. I don't want to conflate responsibility of `validateColumnsExistInSchema` by making it additionally return the set. ########## sql/pipelines/src/test/scala/org/apache/spark/sql/pipelines/autocdc/ChangeArgsSuite.scala: ########## @@ -0,0 +1,155 @@ +/* + * 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.pipelines.autocdc + +import org.apache.spark.SparkFunSuite +import org.apache.spark.sql.AnalysisException +import org.apache.spark.sql.catalyst.parser.ParseException +import org.apache.spark.sql.types.{IntegerType, StringType, StructType} + +class ChangeArgsSuite extends SparkFunSuite { + + private val sourceSchema = new StructType() + .add("id", IntegerType, nullable = false) + .add("Name", StringType) + .add("age", IntegerType) + + test("ColumnSelection None leaves schema unchanged") { + assert(ColumnSelection.applyToSchema(sourceSchema, None) == sourceSchema) + } + + test("ColumnSelection IncludeColumns filters by exact name in schema order") { + val filteredSchema = ColumnSelection.applyToSchema( + sourceSchema, + Some(ColumnSelection.IncludeColumns( + Seq(UnqualifiedColumnName("age"), UnqualifiedColumnName("Name"))))) + + assert(filteredSchema == new StructType() + .add("Name", StringType) + .add("age", IntegerType)) + } + + test("ColumnSelection ExcludeColumns filters by exact name") { + val filteredSchema = ColumnSelection.applyToSchema( + sourceSchema, + Some(ColumnSelection.ExcludeColumns(Seq(UnqualifiedColumnName("id"))))) + + assert(filteredSchema == new StructType() + .add("Name", StringType) + .add("age", IntegerType)) + } + + test("ColumnSelection IncludeColumns fails for columns not present in schema") { + checkError( + exception = intercept[AnalysisException] { + ColumnSelection.applyToSchema( + sourceSchema, + // Column inclusion is case-sensitive; "name" will not match against "Name". + Some(ColumnSelection.IncludeColumns( + Seq(UnqualifiedColumnName("name"), UnqualifiedColumnName("missing")))) + ) + }, + condition = "AUTOCDC_INVALID_COLUMN_SELECTION.COLUMNS_NOT_FOUND", + sqlState = "42703", + parameters = Map( + "missingColumns" -> "name, missing", + "availableColumns" -> "id, Name, age" + ) + ) + } + + test("ColumnSelection ExcludeColumns fails for columns not present in schema") { + checkError( + exception = intercept[AnalysisException] { + ColumnSelection.applyToSchema( + sourceSchema, + // Column exclusion is case-sensitive; "NAME" will not match against "Name". + Some(ColumnSelection.ExcludeColumns( + Seq(UnqualifiedColumnName("NAME"), UnqualifiedColumnName("missing")))) + ) + }, + condition = "AUTOCDC_INVALID_COLUMN_SELECTION.COLUMNS_NOT_FOUND", + sqlState = "42703", + parameters = Map( + "missingColumns" -> "NAME, missing", + "availableColumns" -> "id, Name, age" + ) + ) + } + + test("UnqualifiedColumnName accepts a simple single-part identifier") { + assert(UnqualifiedColumnName("col").name == "col") + } + + test("UnqualifiedColumnName accepts a backtick-quoted name containing a literal dot") { + // Backticks make the dot part of a single name part, so this passes validation. + assert(UnqualifiedColumnName("`a.b`").name == "`a.b`") Review Comment: Done -- 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]
