xiangfu0 opened a new pull request, #19569: URL: https://github.com/apache/pinot/pull/19569
## Summary The binary compatibility job can attribute changes already on master to an unrelated PR because its synthetic merge and `pull_request.base.sha` use different master snapshots. In [the failure on #19530](https://github.com/apache/pinot/actions/runs/34904461122/job/104229846477), the candidate contained master `47a5f867`, while the baseline was `b65be1d5`. - Fetch two commits and compare PR merges against their actual first parent. On master pushes, compare against `github.event.before` so the entire pushed range is checked; reject a missing baseline explicitly. - Restore the public `TableConfig` constructor ending in `boolean materializedView`, removed when #16033 added page-cache warmup configuration. The deprecated overload delegates to the canonical constructor with a null warmup configuration. - Add a regression covering the restored signature, preserved materialized-view/config settings, and null warmup default. ## Validation - 26 focused SPI tests passed: `TableConfigTest`, `TableConfigBuilderTest`, and `TableConfigsSerializationTest`, built with JDK 25 and compiler warnings enabled. No warnings on added lines. - Compiled the historical, broken, and fixed `TableConfig` versions: japicmp 0.23.1 reproduces `CONSTRUCTOR_REMOVED` before the fix and exits successfully after it. - `actionlint` passed. Executed the exact workflow selection script in shallow Git checkouts for an advanced-base PR merge and a multi-commit push, plus empty/all-zero baseline failures. - Affected-module Spotless, Checkstyle, and license checks passed; workflow license check passed. -- 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. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
