Github user dongjoon-hyun commented on the issue: https://github.com/apache/spark/pull/22208 Thank you for updating and adding a test case, @icexelloss . - First of all, my previous comment about using `resolver` means the following. Instead of `queryExecution.analyzed.resolveQuoted(xxx, resolver).isDefined`, the following will be enough and fast. ```scala - if (schema.fieldNames.contains(colName)) { + if (schema.fieldNames.exists(resolver(_, colName))) { ``` - Given that this is about appending additional note at the end of error message, the third commit looks too aggressive change. Could you rollback that in order to minimize the touched line?
--- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org