Skip to content

[SPARK-49246][SQL][FOLLOW-UP] The behavior of SaveAsTable should not be changed by falling back to v1 command #48019

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.

Sign up for GitHub

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

Closed
amaliujia wants to merge 3 commits into from
Closed

[SPARK-49246][SQL][FOLLOW-UP] The behavior of SaveAsTable should not be changed by falling back to v1 command #48019

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
eaf6cf6
[SPARK-49246][SQL][FOLLOW-UP] The behavior of SaveAsTable should not …
amaliujia Sep 6, 2024
61b69a8
fix test
cloud-fan Sep 7, 2024
d912e90
update test
cloud-fan Sep 7, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 4 additions & 2 deletions sql/core/src/main/scala/org/apache/spark/sql/internal/DataFrameWriterImpl.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -426,8 +426,10 @@ final class DataFrameWriterImpl[T] private[sql](ds: Dataset[T]) extends DataFram
import org.apache.spark.sql.connector.catalog.CatalogV2Implicits._

val session = df.sparkSession
val canUseV2 = lookupV2Provider().isDefined ||
df.sparkSession.sessionState.conf.getConf(SQLConf.V2_SESSION_CATALOG_IMPLEMENTATION).isDefined
val canUseV2 = lookupV2Provider().isDefined || (df.sparkSession.sessionState.conf.getConf(
SQLConf.V2_SESSION_CATALOG_IMPLEMENTATION).isDefined &&
!df.sparkSession.sessionState.catalogManager.catalog(CatalogManager.SESSION_CATALOG_NAME)
.isInstanceOf[DelegatingCatalogExtension])

session.sessionState.sqlParser.parseMultipartIdentifier(tableName) match {
case nameParts @ NonSessionCatalogAndIdentifier(catalog, ident) =>
Expand Down
9 changes: 9 additions & 0 deletions .../test/scala/org/apache/spark/sql/connector/DataSourceV2DataFrameSessionCatalogSuite.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -88,6 +88,15 @@ class DataSourceV2DataFrameSessionCatalogSuite
assert(tableInfo.properties().get("provider") === v2Format)
}
}

test("SPARK-49246: saveAsTable with v1 format") {
withTable("t") {
sql("CREATE TABLE t(c INT) USING csv")
val df = spark.range(10).toDF()
df.write.mode(SaveMode.Overwrite).format("csv").saveAsTable("t")
verifyTable("t", df)
}
}
}

class InMemoryTableSessionCatalog extends TestV2SessionCatalogBase[InMemoryTable] {
Expand Down
15 changes: 5 additions & 10 deletions sql/core/src/test/scala/org/apache/spark/sql/connector/TestV2SessionCatalogBase.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,8 +22,7 @@ import java.util.concurrent.atomic.AtomicBoolean

import scala.jdk.CollectionConverters._

import org.apache.spark.sql.catalyst.catalog.CatalogTableType
import org.apache.spark.sql.connector.catalog.{CatalogV2Util, Column, DelegatingCatalogExtension, Identifier, Table, TableCatalog, V1Table}
import org.apache.spark.sql.connector.catalog.{CatalogV2Util, Column, DelegatingCatalogExtension, Identifier, Table, TableCatalog}
import org.apache.spark.sql.connector.expressions.Transform
import org.apache.spark.sql.types.StructType

Expand Down Expand Up @@ -53,14 +52,10 @@ private[connector] trait TestV2SessionCatalogBase[T <: Table] extends Delegating
if (tables.containsKey(ident)) {
tables.get(ident)
} else {
// Table was created through the built-in catalog
super.loadTable(ident) match {
case v1Table: V1Table if v1Table.v1Table.tableType == CatalogTableType.VIEW => v1Table
case t =>
val table = newTable(t.name(), t.schema(), t.partitioning(), t.properties())
addTable(ident, table)
table
}
// Table was created through the built-in catalog via v1 command, this is OK as the
// `loadTable` should always be invoked, and we set the `tableCreated` to pass validation.
tableCreated.set(true)
super.loadTable(ident)
}
}

Expand Down