maropu commented on a change in pull request #24200: [SPARK-27266][SQL] Support 
ANALYZE TABLE to collect tables stats for cached catalog views
URL: https://github.com/apache/spark/pull/24200#discussion_r268959862
 
 

 ##########
 File path: 
sql/core/src/main/scala/org/apache/spark/sql/execution/command/AnalyzeTableCommand.scala
 ##########
 @@ -35,19 +35,27 @@ case class AnalyzeTableCommand(
     val tableIdentWithDB = TableIdentifier(tableIdent.table, Some(db))
     val tableMeta = sessionState.catalog.getTableMetadata(tableIdentWithDB)
     if (tableMeta.tableType == CatalogTableType.VIEW) {
-      throw new AnalysisException("ANALYZE TABLE is not supported on views.")
-    }
-
-    // Compute stats for the whole table
-    val newTotalSize = CommandUtils.calculateTotalSize(sparkSession, tableMeta)
-    val newRowCount =
-      if (noscan) None else 
Some(BigInt(sparkSession.table(tableIdentWithDB).count()))
-
-    // Update the metastore if the above statistics of the table are different 
from those
-    // recorded in the metastore.
-    val newStats = CommandUtils.compareAndGetNewStats(tableMeta.stats, 
newTotalSize, newRowCount)
-    if (newStats.isDefined) {
-      sessionState.catalog.alterTableStats(tableIdentWithDB, newStats)
+      // Analyzes a catalog view if the view is cached
+      val table = sparkSession.table(tableIdent.quotedString)
+      val cacheManager = sparkSession.sharedState.cacheManager
+      if (cacheManager.lookupCachedData(table.logicalPlan).isDefined) {
+        // To collect table stats, materializes an underlying columnar RDD
+        table.collect()
 
 Review comment:
   I wrote this code to do the same thing with the normal case:
   
https://github.com/apache/spark/blob/90b72512f46ac08008d574577c65263d9b13a33c/sql/core/src/main/scala/org/apache/spark/sql/execution/command/AnalyzeTableCommand.scala#L44

----------------------------------------------------------------
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:
us...@infra.apache.org


With regards,
Apache Git Services

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

Reply via email to