[GitHub] [spark] sunchao commented on a change in pull request #30211: [SPARK-33305][SQL] DSv2: DROP TABLE command should also invalidate cache

[GitBox <gi...@apache.org>]

sunchao commented on a change in pull request #30211:
URL: https://github.com/apache/spark/pull/30211#discussion_r520028366



##########
File path: sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/DropTableExec.scala
##########
@@ -26,13 +27,16 @@ import org.apache.spark.sql.connector.catalog.{Identifier, TableCatalog}
  * Physical plan node for dropping a table.
  */
 case class DropTableExec(
+    session: SparkSession,
     catalog: TableCatalog,
     ident: Identifier,
     ifExists: Boolean,
     purge: Boolean) extends V2CommandExec {
 
   override def run(): Seq[InternalRow] = {
     if (catalog.tableExists(ident)) {
+      val table = catalog.loadTable(ident)
+      session.sharedState.cacheManager.uncacheV2Table(table)

Review comment:
       Thanks @cloud-fan . Originally I used the first option but since this is called irrespective whether the target table is cached or not, it means we need to handle streaming tables as well. And since `SparkSession.table` always create `UnresolvedRelation` with `isStreaming = false`, the call will fail later on with those.
   
   I think the second option is promising. Will try that. Thanks.
   
   




----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org