On Mon, Jun 21 2021, Roland Hieber wrote: > On Sun, Jun 20, 2021 at 10:10:10PM -0400, Eric Sunshine wrote: >> On Sun, Jun 20, 2021 at 10:00 PM brian m. carlson >> <sandals@xxxxxxxxxxxxxxxxxxxx> wrote: >> > On 2021-06-21 at 00:35:49, Ævar Arnfjörð Bjarmason wrote: >> > > On Sun, Jun 20 2021, Roland Hieber wrote: >> > > > Currently, calling 'git bisect' from a directory other than the top >> > > > level of a repository only comes up with an error message: >> > > > >> > > > You need to run this command from the toplevel of the working tree. >> > > >> > > How does this affect out-of-tree scripts that will be run with "git >> > > bisect run", is the cwd set to the root as they now might expect git to >> > > check, or whatever subdirectory you ran the "run" from? >> > >> > As for the idea itself, I think it's a good one assuming everything >> > continues to work. It will certainly be more convenient for a lot of >> > people. >> >> There have been multiple patches sent to the project over the years >> with the same purpose. One problem, I believe, which has never been >> fully addressed is what happens when the subdirectory from which >> git-bisect is run gets deleted as part of the bisection. >> >> Here are a couple recent threads triggered by previous such patches >> (but there are probably several more): >> >> https://lore.kernel.org/git/pull.765.git.1603271344522.gitgitgadget@xxxxxxxxx/ >> https://lore.kernel.org/git/pull.736.git.git.1584868547682.gitgitgadget@xxxxxxxxx/ > > Ah, thanks for explaining the problem. Would a patch that adds a short > explanatory comment in git-bisect.sh on the matter help to prevent > people sending such patches? Having skimmed the linked discussions I don't think the consensus is that this shouldn't exist, but that someone who wants it should do some research on the relevant edge cases, come up with test cases for them, discuss the trade-offs in a commit message etc. I for one would welcome such a feature, it's often annoyed me, it should just work like "rebase exec" in that a "run" script should cd to the root, but (as discussed in the linked threads) I don't see why we'd prevent it any more than several other commands that already have this edge case, but don't explicitly prevent this.
