SEZ9 commented on PR #12269: URL: https://github.com/apache/seatunnel/pull/12269#issuecomment-6964384270
@Dev-next-gen thanks for the follow-up. Keeping `getSinkPluginJarPathsUnderTurkishLocale` in `SeaTunnelSourcePluginDiscoveryTest` is fine with me. Reusing the `duplicate` fixture and the jars from `before()` is a reasonable justification, and the comment you added in 528f5e3a1 settles the test-placement point. From the latest review on this thread, I understand the earlier points now stand as follows: - **CLI converters:** the two converters in seatunnel-starter that used `PluginType.valueOf(value.toUpperCase())` now use `Locale.ROOT`, so `-pt sink` resolves under tr-TR. - **Multi-jar branch:** the Turkish-locale test now covers the multiple-jar selection as well as the single-jar case. - **Core fix:** the `Locale.ROOT` change in `AbstractPluginDiscovery` remains correct and minimal, and no docs or changelog action is needed. Two small things I'd still like you to confirm: 1. **Resource lock on the Turkish-locale test:** the lock was flagged earlier as inert in this module, and a sibling test uses a different lock key. Does the lock actually take effect here, and are the keys aligned so the locale-mutating tests can't run concurrently? If it has no effect in this module, please say so in the PR. 2. **Remaining `toLowerCase()` calls:** have the bare calls on the identifier in the same method been switched to `Locale.ROOT` too? This one is style-level and not a blocker. Thanks also for the local JDK 17 run (`spotless:check` passing, `SeaTunnelSourcePluginDiscoveryTest` with 6 tests, 0 failures and 0 errors). <!-- streview-comment:2424 --> -- 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]
