On Wed, Mar 02 2022, Junio C Hamano wrote: > Ævar Arnfjörð Bjarmason <avarab@xxxxxxxxx> writes: > >> Change a couple of uses of "test_expect_failure" to use a >> "test_expect_success" to positively assert the current behavior, and >> replace the intent of "test_expect_failure" with a "TODO" comment int >> the description. >> >> As noted in [1] the "test_expect_failure" feature is overly eager to > > And noted in [2], it is not a good idea to abuse "test_expect_success" > for this purpose, either, though. > > [2] https://lore.kernel.org/git/xmqq4k9kj15p.fsf@gitster.g/ As noted I do have a "test_todo" (or "test_expect_todo") replacement for "test_expect_failure" which I think I think will address your concern there. But do you mind if this is left like this for now? Due to the semantics of "test_expect_failure" we can't use it in conjunction with "test_must_fail" currently and not hide segfaults or abort(). So having it marked as "ok ... # TODO" v.s. "not ok ... # TODO" isn't ideal, but certainly better than silently hiding abort() and segfaults.
