-
Notifications
You must be signed in to change notification settings - Fork 28.4k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[SPARK-22197][SQL] push down operators to data source before planning #19424
Changes from 2 commits
File filter
Filter by extension
Conversations
Jump to
Diff view
Diff view
There are no files selected for viewing
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,64 @@ | ||
/* | ||
* 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 java.util.Objects | ||
|
||
import org.apache.spark.sql.catalyst.expressions.{Attribute, AttributeReference} | ||
import org.apache.spark.sql.sources.v2.reader._ | ||
|
||
/** | ||
* A base class for data source reader holder and defines equals/hashCode methods. | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more.
|
||
*/ | ||
trait DataSourceReaderHolder { | ||
def fullOutput: Seq[AttributeReference] | ||
def reader: DataSourceV2Reader | ||
|
||
override def equals(other: Any): Boolean = other match { | ||
case other: DataSourceV2Relation => | ||
val basicEquals = this.fullOutput == other.fullOutput && | ||
this.reader.getClass == other.reader.getClass && | ||
this.reader.readSchema() == other.reader.readSchema() | ||
|
||
val samePushedFilters = (this.reader, other.reader) match { | ||
case (l: SupportsPushDownCatalystFilters, r: SupportsPushDownCatalystFilters) => | ||
l.pushedCatalystFilters().toSeq == r.pushedCatalystFilters().toSeq | ||
case (l: SupportsPushDownFilters, r: SupportsPushDownFilters) => | ||
l.pushedFilters().toSeq == r.pushedFilters().toSeq | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. The evaluation order of these filters must be the same? If the orders are different, they are still the same, right? There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. good catch! |
||
case _ => true | ||
} | ||
|
||
basicEquals && samePushedFilters | ||
|
||
case _ => false | ||
} | ||
|
||
override def hashCode(): Int = { | ||
val state = Seq(fullOutput, reader.getClass, reader.readSchema()) | ||
val filters: Any = reader match { | ||
case s: SupportsPushDownCatalystFilters => s.pushedCatalystFilters().toSeq | ||
case s: SupportsPushDownFilters => s.pushedFilters().toSeq | ||
case _ => Nil | ||
} | ||
(state :+ filters).map(Objects.hashCode).foldLeft(0)((a, b) => 31 * a + b) | ||
} | ||
|
||
lazy val output: Seq[Attribute] = reader.readSchema().map(_.name).map { name => | ||
fullOutput.find(_.name == name).get | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Shall we use resolver instead of string comparison? There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. These names should already be normalized before reaching here. |
||
} | ||
} |
Original file line number | Diff line number | Diff line change |
---|---|---|
|
@@ -29,20 +29,12 @@ import org.apache.spark.sql.execution.metric.SQLMetrics | |
import org.apache.spark.sql.sources.v2.reader._ | ||
import org.apache.spark.sql.types.StructType | ||
|
||
/** | ||
* Physical plan node for scanning data from a data source. | ||
*/ | ||
case class DataSourceV2ScanExec( | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more.
|
||
fullOutput: Array[AttributeReference], | ||
@transient reader: DataSourceV2Reader, | ||
// TODO: these 3 parameters are only used to determine the equality of the scan node, however, | ||
// the reader also have this information, and ideally we can just rely on the equality of the | ||
// reader. The only concern is, the reader implementation is outside of Spark and we have no | ||
// control. | ||
readSchema: StructType, | ||
@transient filters: ExpressionSet, | ||
hashPartitionKeys: Seq[String]) extends LeafExecNode { | ||
|
||
def output: Seq[Attribute] = readSchema.map(_.name).map { name => | ||
fullOutput.find(_.name == name).get | ||
} | ||
fullOutput: Seq[AttributeReference], | ||
@transient reader: DataSourceV2Reader) extends LeafExecNode with DataSourceReaderHolder { | ||
|
||
override def references: AttributeSet = AttributeSet.empty | ||
|
||
|
@@ -74,7 +66,7 @@ class RowToUnsafeRowReadTask(rowReadTask: ReadTask[Row], schema: StructType) | |
override def preferredLocations: Array[String] = rowReadTask.preferredLocations | ||
|
||
override def createReader: DataReader[UnsafeRow] = { | ||
new RowToUnsafeDataReader(rowReadTask.createReader, RowEncoder.apply(schema)) | ||
new RowToUnsafeDataReader(rowReadTask.createReader, RowEncoder.apply(schema).resolveAndBind()) | ||
} | ||
} | ||
|
||
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Seems we don't use
Expression
here?There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
good catch. I'll remove it in my following PRs.