spark UpdateFields 源码

spark UpdateFields 代码

文件路径：/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/UpdateFields.scala

/*
 * 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.optimizer

import java.util.Locale

import scala.collection.mutable

import org.apache.spark.sql.catalyst.expressions.{Expression, UpdateFields, WithField}
import org.apache.spark.sql.catalyst.plans.logical.LogicalPlan
import org.apache.spark.sql.catalyst.rules.Rule
import org.apache.spark.sql.catalyst.trees.TreePattern.UPDATE_FIELDS


/**
 * Optimizes [[UpdateFields]] expression chains.
 */
object OptimizeUpdateFields extends Rule[LogicalPlan] {
  private def canOptimize(names: Seq[String]): Boolean = {
    if (conf.caseSensitiveAnalysis) {
      names.distinct.length != names.length
    } else {
      names.map(_.toLowerCase(Locale.ROOT)).distinct.length != names.length
    }
  }

  val optimizeUpdateFields: PartialFunction[Expression, Expression] = {
    case UpdateFields(structExpr, fieldOps)
      if fieldOps.forall(_.isInstanceOf[WithField]) &&
        canOptimize(fieldOps.map(_.asInstanceOf[WithField].name)) =>
      val caseSensitive = conf.caseSensitiveAnalysis

      val withFields = fieldOps.map(_.asInstanceOf[WithField])
      val names = withFields.map(_.name)
      val values = withFields.map(_.valExpr)

      val newNames = mutable.ArrayBuffer.empty[String]
      val newValues = mutable.HashMap.empty[String, Expression]
      // Used to remember the casing of the last instance
      val nameMap = mutable.HashMap.empty[String, String]

      names.zip(values).foreach { case (name, value) =>
        val normalizedName = if (caseSensitive) name else name.toLowerCase(Locale.ROOT)
        if (nameMap.contains(normalizedName)) {
          newValues += normalizedName -> value
        } else {
          newNames += normalizedName
          newValues += normalizedName -> value
        }
        nameMap += normalizedName -> name
      }

      val newWithFields = newNames.map(n => WithField(nameMap(n), newValues(n)))
      UpdateFields(structExpr, newWithFields.toSeq)

    case UpdateFields(UpdateFields(struct, fieldOps1), fieldOps2) =>
      UpdateFields(struct, fieldOps1 ++ fieldOps2)
  }

  def apply(plan: LogicalPlan): LogicalPlan = plan.resolveExpressionsWithPruning(
    _.containsPattern(UPDATE_FIELDS), ruleId)(optimizeUpdateFields)
}

/**
 * Replaces [[UpdateFields]] expression with an evaluable expression.
 */
object ReplaceUpdateFieldsExpression extends Rule[LogicalPlan] {
  def apply(plan: LogicalPlan): LogicalPlan = plan.transformAllExpressionsWithPruning(
    _.containsPattern(UPDATE_FIELDS)) {
    case u: UpdateFields => u.evalExpr
  }
}

相关信息

spark 源码目录

相关文章

spark ComplexTypes 源码

spark CostBasedJoinReorder 源码

spark DecorrelateInnerQuery 源码

spark EliminateResolvedHint 源码

spark InjectRuntimeFilter 源码

spark InlineCTE 源码

spark LimitPushDownThroughWindow 源码

spark MergeScalarSubqueries 源码

spark NestedColumnAliasing 源码

spark NormalizeFloatingNumbers 源码