Hi Elijah,
> Le 14 oct. 2020 à 18:02, Elijah Newren <newren@xxxxxxxxx> a écrit :
>
> On Wed, Oct 14, 2020 at 8:27 AM Philippe Blain
> <levraiphilippeblain@xxxxxxxxx> wrote:
>>
>> Hello all,
>>
>> Up to recently I had an alias 'st' for `git status`, i.e. `git config alias.st status`.
>> This works well in the main working tree as well as secondary worktrees (`git worktree add ...`),
>> and also shows paths relative to the current directory if I'm not at the root of the repo,
>> just like `git status`.
>>
>> Now I wanted to change my alias so that it ran `git status` and then an additional command, i.e.
>> `git config alias.st '!git status && date'` (for demonstration purposes).
>> This works correctly in the main worktree and secondary worktrees, but since aliases
>> run from the root of the repo, the paths in the output of 'git st' are not relative
>> to the current directory.
>>
>> So my next attempt was `git config alias.st '!git -C "${GIT_PREFIX}" status && date'`,
>> so that paths would be relative in the output. However, this works in the main worktree
>> but fails badly in the secondary worktree:
>>
>> ```
>> rm -rf test &&
>> rm -rf test2 &&
>> git init test &&
>> cd test &&
>> mkdir folder &&
>> date>>folder/file &&
>> git add folder/file &&
>> git ci -m "commit" &&
>> git config alias.st '!git -C "${GIT_PREFIX}" status && date' &&
>> date >>folder/file &&
>> echo '=== `git st` in main worktree at root ===' &&
>> git st &&
>> cd folder &&
>> echo '=== `git st` in main worktree in folder ===' &&
>> git st &&
>> git worktree add ../test2 &&
>> cd ../test2 &&
>> date >>folder/file &&
>> echo '=== `git st` in secondary worktree at root ===' &&
>> git st &&
>> cd folder &&
>> echo '=== `git st` in secondary worktree in folder ===' &&
>> git st
>> ```
>>
>> The last commands ouputs:
>>
>> ```
>> === `git st` in secondary worktree in folder ===
>> On branch test2
>> Changes not staged for commit:
>> (use "git add/rm <file>..." to update what will be committed)
>> (use "git restore <file>..." to discard changes in working directory)
>> deleted: folder/file
>>
>> Untracked files:
>> (use "git add <file>..." to include in what will be committed)
>> file
>>
>> no changes added to commit (use "git add" and/or "git commit -a")
>> ```
>>
>> So something is wrong in the automatic worktree path detection...
>> If I replace the alias with
>>
>> git config alias.st '!git -C "${GIT_PREFIX}" --work-tree=$PWD status && date'
>>
>> then it works correctly, but I have a feeling this should not be necessary...
>>
>> I've CC-ed Eric (because of his work on `git worktree`) and Elijah (because
>> I remember he worked on `dir.c` so maybe this is related to that code, but I'm
>> not sure).
>>
>> Cheers,
>> Philippe
>
> I don't think there's anything dir.c-specific here (thank
> goodness...). I extended your aliases slightly to echo the command
> they would run (so I could see the value of $GIT_PREFIX) as well as to
> print all the GIT_* environment variables, and found that GIT_DIR was
> set as well as GIT_PREFIX, so you can duplicate the funny results with
>
> GIT_DIR=$WHEREVER git -C folder status
>
> or even
>
> git --git-dir=$WHEREVER -C folder status
>
> In fact, you don't need any special worktrees setup; using these
> commands will trigger the funny status output. Looking at the
> documentation for core.worktree, I see
>
> "If --git-dir or GIT_DIR is specified but none of --work-tree,
> GIT_WORK_TREE and core.worktree is specified, the current working
> directory is regarded as the top level of your working tree."
Yes, this is in line with git(1) [1] :
"Specifying the location of the ".git" directory using this option (or GIT_DIR environment variable)
turns off the repository discovery that tries to find a directory with ".git" subdirectory
(which is how the repository and the top-level of the working tree are discovered),
and tells Git that you are at the top level of the working tree.
If you are not at the top-level directory of the working tree,
you should tell Git where the top-level of the working tree is,
with the --work-tree=<path> option (or GIT_WORK_TREE environment variable)"
> and indeed, I find that if I run from the toplevel:
>
> git --work-tree=folder/ status
>
> then I see the same output that you are complaining about. So, it
> seems that "the current working directory is regarded as the top level
> of your working tree" is implemented to take effect AFTER the "-C
> folder" first changes the current working directory.
>
> I haven't tracked down where in the code this happens, but I suspect
> that this is what is happening and is the culprit behind the behavior
> you are seeing. If I am right, it doesn't answer whether this is a
> bug, of course -- it could be that this ordering is intentional and
> desired, or it could be that this ordering is not wanted. Anyway,
> does this help you track it down?
I found two threads [2] and [3] which are manifestations of similar (or the same ?) problems...
Maybe git-config(1) could document which variables will
be set in aliases that shell out, but even that might be tricky because it
seems the answer is "it depends" (according to [2])...
I also found 57ea7123c8 (git.c: make sure we do not
leak GIT_* to alias scripts, 2015-12-20), in topic 'nd/clear-gitenv-upon-use-of-alias',
which tries to clear out GIT_* variables from aliases environment,
which I guess is defeated by the fact that I run `git -C` in my alias...
Anyway thanks for your answer.
[1] https://git-scm.com/docs/git#Documentation/git.txt---git-dirltpathgt
[2] https://lore.kernel.org/git/20200130102933.GE840531@xxxxxxxxxxxxxxxxxxxxxxx/
[3] https://lore.kernel.org/git/20200228190218.GC1408759@xxxxxxxxxxxxxxxxxxxxxxx/
Philippe.
