On Wed, Nov 16, 2022 at 05:04:10PM -0500, Jeff King wrote: > > I haven't looked too deeply at what is going on here, but let's make > > sure to resolve this before graduating the topic down (which I would > > otherwise like to do in the next push-out, probably tomorrow or the next > > day). > > The issue is that some of the tests assume that hiding "refs/" should > produce no output from "--exclude-hidden=receive --all". But it will > also show HEAD, even if it points to a hidden ref (which I think is OK, > and matches what receive-pack would do). > > But because the setup uses "main" as one of the sample refs, HEAD may or > may not be valid, depending on what it points to (without setting > GIT_TEST_DEFAULT_INITIAL_BRANCH_NAME it points to master, which is > unborn). > > So the fix is just: > > [...] Makes perfect sense, and thanks for looking into it. Patrick: it sounds like there was one typo in the earlier round which you may want to pick up also, assuming you reroll this. Thanks, Taylor
