Re: [PATCH] builtin/pack-objects.c: introduce `pack.extraCruftTips`

From: Taylor Blau <me@ttaylorr.com>
Date: Apr 21, 2023, 02:14 UTC
Message-ID: <ZEHxctCveqt06g4a@nand.local>
In-Reply-To: <CAPx1GvdmzDVbiZGeguVVjWe+eAhj6=yG18Rd6wLhZVrnd4jiBg@mail.gmail.com>

On Thu, Apr 20, 2023 at 05:10:31PM -0700, Chris Torek wrote:

Show 7 quoted lines

> On Thu, Apr 20, 2023 at 10:35 AM Taylor Blau <me@ttaylorr.com> wrote:
> > +Output must contain exactly one hex object ID per line, and nothing
> > +else. Objects which cannot be found in the repository are ignored.
>
> The first part sounds good, as a rigid output format can always be
> relaxed later if necessary. The second I'm less sure about: should
> there perhaps be a --{no-,}missing-objects option to control it?

Maybe, although specifying it might be a little tricky.

I think you'd have to tie that setting into the configuration instead of passing it through as a command-line argument. So things get interesting when you have multiple `pack.extraCruftTips` commands: which of them are allowed (or not allowed) to specify missing objects?

I think you could do something like:

    [packExtraCruft "foo"]
        exec = /path/to/script
        allowMissing = true

and specify multiple pack.extraCruftTips-esque programs, that each allow (or not) passing missing objects.

I dunno. It might be overkill, but I could also see a compelling argument towards doing something like that. It may equally be OK to set `pack.allowMissingExtraCruftTips` once and have it apply to all values of `pack.extraCruftTips`.

That might be a reasonable compromise, and it doesn't feel like overkill to me. I can't easily imagine a circumstance where you'd want to allow some programs to specify missing objects and not others.

Thanks, Taylor