Re: [PATCH v2 1/2] config: allow giving separate author and committer idents

From: Junio C Hamano <gitster@pobox.com>
Date: Jan 28, 2019, 21:33 UTC
Message-ID: <xmqqmunka2nb.fsf@gitster-ct.c.googlers.com>
In-Reply-To: <871s4w4khs.fsf@evledraar.gmail.com>

Ævar Arnfjörð Bjarmason <avarab@gmail.com> writes:

Show 21 quoted lines

> I'm happy to help if you'd like. I had a thinko with "table", and I
> think our asciidoc dialect doesn't support it (maybe I'm wrong), but
> thinking about it again we could just describe these variables all in
> the same documentation. As in this hunk (which you could squash in):
>
> BEGIN QUOTE
> diff --git a/Documentation/config/user.txt b/Documentation/config/user.txt
> index 18e1ec3c1b..ad3c43cf47 100644
> --- a/Documentation/config/user.txt
> +++ b/Documentation/config/user.txt
> @@ -1,39 +1,20 @@
> +user.name::
> +user.email::
>  author.name::
> +author.email::
>  committer.name::
> +committer.email::
> +	The `user.name` and `user.email` variables determine what ends
> +	up in the `author` and `committer` field of commit
> +	objects. These config variables will be overridden by
> +	`GIT_COMMITTER_NAME` and `GIT_COMMITTER_EMAIL`,

You forgot to list two more obvious ones here.

Show 7 quoted lines

> ++
> +Most users should have no reason to set the `author.*` and
> +`committer.*` variables, but can do so to e.g. set different a
> +different E-Mail for the `committer` field. Like the `user.name` and
> +`user.email` variables, these can be overridden in the environment
> +with `GIT_AUTHOR_NAME`, `GIT_AUTHOR_EMAIL`, `GIT_COMMITTER_NAME` and
> +`GIT_COMMITTER_EMAIL`.

I do not see a strong reason to say "most users should have no reason", especially without arguing why (and I do not think this is a place to make such an argument, either).

    The `user.*` variables can be used to set both author and committer
    names and e-mail addresses to the same value; users who want to set
    the committer and author identities differently can use the
    `author.*` and `committer.*` variables.

or something along that line, perhaps?