On Wed, 16 Apr 2025 16:29:23 GMT, Jaikiran Pai <j...@openjdk.org> wrote:
> > Nevertheless, the tests are there, so no _new_ ones are needed. We can > > duplicate them, if a majority thinks it is beneficial. > > As noted in the contribution guide > (https://openjdk.org/guide/#testing-the-jdk), regression tests are expected > for a majority of changes that are done in the JDK. There are situations > where regressions tests cannot be added, but this isn't one. I believe a new > method with a default specified implementation in a public exported interface > deserves a regression test of its own. The docs do not say that the regression test must be *separate*. Again: We *do have* testing covered. The question is *not* whether we need a test -we *do have one*-, but whether to *duplicate* it. ------------- PR Comment: https://git.openjdk.org/jdk/pull/21730#issuecomment-2810119199
