cloud-fan commented on a change in pull request #31286:
URL: https://github.com/apache/spark/pull/31286#discussion_r567794818
##########
File path: docs/sql-migration-guide.md
##########
@@ -49,6 +49,8 @@ license: |
* and the method `spark.catalog.refreshTable`
In Spark 3.1 and earlier, table refreshing leaves dependents uncached.
+ - In Spark 3.2, the usage of `count(table.*)` is blocked to avoid producing
ambiguous results. Because `count(*)` and `count(table.*)` will output
differently if there is any null values. To restore the behavior before Spark
3.2, you can set `spark.sql.legacy.allowStarWithSingleTableIdentifierInCount`
to `true`.
Review comment:
`table` is a keyword, so `count(table.*)` is not parse-able under ansi
mode. How about `count(tblName.*)`?
----------------------------------------------------------------
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:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
