aokolnychyi commented on code in PR #40308: URL: https://github.com/apache/spark/pull/40308#discussion_r1167348702
########## sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/AssignmentUtils.scala: ########## @@ -0,0 +1,167 @@ +/* + * 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.analysis + +import scala.collection.mutable + +import org.apache.spark.sql.catalyst.SQLConfHelper +import org.apache.spark.sql.catalyst.expressions.{Attribute, CreateNamedStruct, Expression, GetStructField, Literal} +import org.apache.spark.sql.catalyst.plans.logical.Assignment +import org.apache.spark.sql.catalyst.util.CharVarcharUtils +import org.apache.spark.sql.connector.catalog.CatalogV2Implicits._ +import org.apache.spark.sql.errors.QueryCompilationErrors +import org.apache.spark.sql.types.{DataType, StructType} + +object AssignmentUtils extends SQLConfHelper with CastSupport { + + /** + * Aligns assignments to match table columns. + * <p> + * This method processes and reorders given assignments so that each target column gets + * an expression it should be set to. If a column does not have a matching assignment, + * it will be set to its current value. For example, if one passes table attributes c1, c2 + * and an assignment c2 = 1, this method will return c1 = c1, c2 = 1. This allows Spark to + * construct an updated version of a row. + * <p> + * This method also handles updates to nested columns. If there is an assignment to a particular + * nested field, this method will construct a new struct with one field updated preserving other + * fields that have not been modified. For example, if one passes table attributes c1, c2 + * where c2 is a struct with fields n1 and n2 and an assignment c2.n2 = 1, this method will + * return c1 = c1, c2 = struct(c2.n1, 1). + * + * @param attrs table attributes + * @param assignments assignments to align + * @return aligned assignments that match table attributes + */ + def alignAssignments( + attrs: Seq[Attribute], + assignments: Seq[Assignment]): Seq[Assignment] = { + + val errors = new mutable.ArrayBuffer[String]() + + val output = attrs.map { attr => + applyAssignments( + col = restoreActualType(attr), + colExpr = attr, + assignments, + addError = err => errors += err, + colPath = Seq(attr.name)) + } + + if (errors.nonEmpty) { + throw QueryCompilationErrors.invalidRowLevelOperationAssignments(assignments, errors.toSeq) + } + + attrs.zip(output).map { case (attr, expr) => Assignment(attr, expr) } + } + + private def restoreActualType(attr: Attribute): Attribute = { + attr.withDataType(CharVarcharUtils.getRawType(attr.metadata).getOrElse(attr.dataType)) + } + + private def applyAssignments( + col: Attribute, + colExpr: Expression, + assignments: Seq[Assignment], + addError: String => Unit, + colPath: Seq[String]): Expression = { + + val (exactAssignments, otherAssignments) = assignments.partition { assignment => + assignment.key.semanticEquals(colExpr) + } + + val fieldAssignments = otherAssignments.filter { assignment => + assignment.key.exists(_.semanticEquals(colExpr)) + } + + if (exactAssignments.size > 1) { Review Comment: @cloud-fan, I've changed the approach to avoid deconstructing references. However, I decided to keep the validation while recursing vs doing this in a separate step as we discussed. When I tried to implement that idea, it turned out to be pretty involved with lots of edge cases. For instance, we can't have multiple assignments per top-level key but keys can reference top-level fields many times, `a.b.c` and `a.b.d` are allowed but not `a.b` and `a.b.c`. It felt easier to validate while recursing, just like `TableOutputResolver`. Let me know what you think. -- 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]
