From: Bello Olamide <belkid98@gmail.com>
Date: Wed, 07 Jan 2026 15:33:03 GMT
Subject: Re: [Outreachy PATCH v2] environment: move "core.attributesFile" into repo-setting
Message-ID: <CAD=f0L9in5tjLNUcM86uzd_caTaZYre9RRygmizO4G=S_DBFRQ@mail.gmail.com>
In-Reply-To: <8899016f-eeef-404b-8da6-ff3a90e81cea@gmail.com>

On Wed, 7 Jan 2026 at 15:19, Phillip Wood <phillip.wood123@gmail.com> wrote:
>
> On 07/01/2026 10:26, Phillip Wood wrote:
> > On 06/01/2026 13:44, Bello Olamide wrote:
> >>
> >> But won't this be a temporary solution since the goal is to prevent
> >> the use of
> >> `the_repository`?
> >
> > Yes but it would be a good start as passing a repository down to
> > git_default_config() will be quite invasive.
>
> To expand on this the first steps could be
>    (i) create a new struct to hold the config settings from
>        git_default_config()
>   (ii) add that struct as a member of `struct repository`
> (iii) one-by-one, for each setting parsed by git_default_config() add a
>        new member to the config struct, store the parsed value in
>        `the_repository` and adjust any code that uses the variable.
>
> Then later we can tackle the intrusive change to pass a `struct
> repository` down to git_default_config() and store the settings in that
> rather than `the_repository`. If we add a local variable to
> git_default_config() in step (iii) above then getting it to use the
> repository passed down the call chain will simply be a matter of doing
> something like
>
> -       struct repository *r = the_repository;
> +       struct repository *r = cb ? cb : the_repository;
>
> Thanks
>
> Phillip
>

Thanks for the detailed proposal, which makes a lot of sense.
I’ll align my changes with this staged approach.

Bello