alhudz commented on PR #1725: URL: https://github.com/apache/commons-lang/pull/1725#issuecomment-4776671131
@sarankumarbaskar has it right. `String.regionMatches(true, ...)` only folds a supplementary code point at the code point level from Java 9 on; on Java 8 it stays char-by-char and returns `false`, so the Deseret rows in the shared `TEST_DATA` failed on the `String`/`CSString` runners there. This PR only touches the green non-String path, which now folds the whole code point on every JDK (matching Java 9+ `String`). On Java 8 that leaves the green path a touch ahead of the JDK's own `String`, which seems like the right call for a `CharSequence` helper rather than reproducing the old surrogate-by-surrogate behaviour. I've moved those cases out of the String-parity data into a dedicated `testRegionMatchesSupplementaryCaseFold` that drives only the non-String path, and kept `TEST_DATA` on BMP inputs that agree with `String` everywhere. Build is green on 8 through 25 now. -- 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]
