[GitHub] [spark] HyukjinKwon commented on a change in pull request #32411: [SPARK-28551][SQL]In CTAS with LOCATION , should not allow to a non-empty directory.

[GitBox]

HyukjinKwon commented on a change in pull request #32411:
URL: https://github.com/apache/spark/pull/32411#discussion_r624870695



##########
File path: 
sql/core/src/main/scala/org/apache/spark/sql/execution/command/DataWritingCommand.scala
##########
@@ -96,4 +98,20 @@ object DataWritingCommand {
       sparkContext.getLocalProperty(SQLExecution.EXECUTION_ID_KEY),
       metrics.values.toSeq)
   }
+
+  def assertEmptyRootPath(locationUri: Option[URI], configuration: 
Configuration) {
+    locationUri.fold() { tablePathUri: URI =>
+      val tablePath = tablePathUri.toString
+      val filePath = new org.apache.hadoop.fs.Path(tablePath)
+      val fs = filePath.getFileSystem(configuration)
+      if(fs.exists(filePath)) {
+        val cs = fs.getContentSummary(filePath)

Review comment:
       This kind of listing can potentially be expensive, and looks 
`getContentSummary` is recursive too. I think we should at least guard by a 
legacy configuration with updating the migration guide at 
https://github.com/apache/spark/blob/master/docs/sql-migration-guide.md




-- 
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



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