maryannxue commented on issue #26813: [SPARK-30188][SQL] Resolve the failed unit tests when enable AQE URL: https://github.com/apache/spark/pull/26813#issuecomment-580336242 I did not go through the entire list of changes. I've noticed these categories of changes so far: 1. Fix: https://github.com/apache/spark/pull/26813/files/8b5e7442c63fe326db7c7f46f7a194fbae8f0d46#diff-f0ed5bcd435a874d4eb2faf0c2a1b091R33 2. Things I'm not sure if they are correctness issue workaround or performance optimization: https://github.com/apache/spark/pull/26813#discussion_r371920340 3. Use of AdaptiveSparkPlanHelper to resolve leaf node issues and hard-code plan checks. 4. Ordering results using "sort" 5. Disabling AQE flags: e.g., SQLMetricsSuite, https://github.com/apache/spark/pull/26813#discussion_r373044512 A big +1 on (3). Let's pull (1) into different PRs for easy tracking. Let's verify (2) and see what to do. Look into (4) and at least understand why. I'm sure some of these changes are legitimate, but users may find results to be different when turning on AQE. Pull them out into different PRs, and open a JIRA ticket for each of these tests in (5).
---------------------------------------------------------------- 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] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
