Github user gatorsmile commented on the issue: https://github.com/apache/spark/pull/15690 Found another side effect caused by implicit casting. When we having `CAST` in the filter conditions, we are unable to push down these filters into the data sources. Performance could be very bad if users do not properly specify the data types of these literals. For example, when the data type of `a` is integer, the following filter can be pushed down to the parquet: ``` spark.read.parquet(path).filter("a < 100") ``` However, users will not see the filter push down if the literal is not an integer. ``` spark.read.parquet(path).filter("a < 100.0") ```
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. --- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org