From: Phillip Wood <phillip.wood123@gmail.com>
Date: Wed, 18 Feb 2026 14:19:56 GMT
Subject: Re: [PATCH 1/2] wt-status: avoid passing NULL worktree
Message-ID: <d881edeb-d73b-43ba-bdb3-1b664e1cb882@gmail.com>
In-Reply-To: <CAOLa=ZQKLqFn4w3s7PD87FZ_120gohoqKX5c3uLKo2vASsbxfA@mail.gmail.com>

On 17/02/2026 17:46, Karthik Nayak wrote:
> Phillip Wood <phillip.wood123@gmail.com> writes:
>> From: Phillip Wood <phillip.wood@dunelm.org.uk>
>>
>> This updates the only callers that pass a NULL worktree to
>> worktree_git_path().
>>
> 
> I was thinking surely there must be other places where we also pass NULL
> for worktree, but doesn't seem like there are any such instances.

Yes, I was pleasantly surprised there weren't more sites to convert.

>> +struct worktree *get_worktree_from_repository(struct repository *repo)
>> +{
>> +	struct worktree *wt = xcalloc(1, sizeof(*wt));
>> +	char *gitdir = absolute_pathdup(repo->gitdir);
>> +	char *commondir = absolute_pathdup(repo->commondir);
>> +
>> +	wt->repo = repo;
>> +	if (repo->worktree)
>> +		wt->path = absolute_pathdup(repo->worktree);
> 
> Shouldn't this always be set? I guess my question is, will
> `repo->worktree` ever be NULL?

Oh, wt->path should never be NULL. repo->worktree is NULL in bare 
repositories but then we should use repo->gitdir as the worktree path.

>> +	wt->is_bare = !!repo->worktree;
>> +	if (fspathcmp(gitdir, commondir))
>> +		wt->id = xstrdup(find_last_dir_sep(commondir) + 1);
> 
> So here we continue to treat NULL as the main worktree. Okay.
> 
>> +	wt->is_current = is_current_worktree(wt);
> 
> Since we're getting the worktree from the repo, shouldn't this be
> 'true'?

That's what I thought initially. However is_current_worktree() compares 
"repo->gitdir" to "the_repository->gitdir" so the "current" worktree is 
the one that the process was started in, which is not necessarily the 
same as the one matching "wt->repo->gitdir". It's possible that we will 
want to change that definition in the future but I opted to make this 
function consistent with the status quo.

>> -/*
>> +/* Construct a struct worktree from a struct repository */
>> +struct worktree *get_worktree_from_repository(struct repository *repo);
>> +
>> + /*
> 
> Nit: extra space?

Good spot I'll fix it

Thanks for the review

Phillip