Re: How to undo previously set configuration? (again)

From: Jeff King <peff@peff.net>
Date: May 1, 2019, 21:15 UTC
Message-ID: <20190501211525.GB15902@sigill.intra.peff.net>
In-Reply-To: <87r29iqsf9.fsf@evledraar.gmail.com>

On Wed, May 01, 2019 at 02:18:34PM +0200, Ævar Arnfjörð Bjarmason wrote:

> We can make it fancier, but we already deal with this, e.g. if you do
> "git config -l" we'll show "include{,if}" directives at the same "level"
> as other "normal" keys.

We show them, but we _do_ interpret them if the caller asks for it with --includes (which defaults to on when doing the usual "look in all files").

I think we'd have something similar here, where the caller can ask to apply excludes or not.

Show 20 quoted lines

> We also provide no way in "git config" to properly interpret a
> value. E.g. does a "user.email" showing up twice for me mean I have two
> E-Mails at the same time, or does the last one win? We both know the
> answer, but git-config itself doesn't, and that information lives in
> docs/code outside of it.
> 
> Similarly we'd just print a sequence of:
> 
>     user.name=foo
>     user.email=bar
>     exclude.key=user.*
>     user.name=baz
> 
> And it would be up to some "smarter" reader of the config data to
> realize that the end result is one where we have no "user.email" set,
> and "user.name=baz".
> 
> But yeah, optionally having some new --list-normalized or
> --list-after-excludes or whatever would be great, and presumably not
> hard if we had some central "excludes" mechanism...

I think that is all because "--list" really is just about dumping all values, not about interpreting. If we had an exclude mechanism, then I'd expect:

  git config user.name

to apply it just like git_config_get_string() would.

Because of its lack of interpretation, I don't think --list is actually good for much besides debugging. Some scripts do use it to avoid making a bunch of individual git-config calls, but they'd be much better served by a --stdin mode which let you feed in a sequence of operations ("I want x.y.z, as a single last-one-wins value, and interpreted as a bool") and get a sequence of outputs.

-Peff