Re: [PATCH 00/12] builtin/rerere: convert to use subcommands
- From
Junio C Hamano <gitster@pobox.com>
- Date
- Oct 27, 2026, 19:52 UTC
- Message-ID
- <xmqqqzhaj4zv.fsf@gitster.g>
- In-Reply-To
- <20261027-pks-rerere-subcommands-v1-0-d836c36ce0c5@pks.im>
Patrick Steinhardt <ps@pks.im> writes:
Show 9 quoted lines
> while reviewing some patch series around git-rerere(1) recently I was a > bit appalled by our custom subcommand parsing that we do here. This > patch series here thus cleans up and modernizes that part a bit so that > git-rerere(1) behaves like every other Git command that happens to have > subcommands. > > This series is based on 37b3475fb53 (The 9th batch, 2026-10-25) with > ps/rerere-remove-global-state at b948c10c8b3 (rerere: get rid of global > rerere cache state, 2026-10-23) merged into it.
As the inventor of the mechanism, I took special interest in the topic, and overall found them well done. I wasn't even aware that we were so close to get rid of USE_THE_REPOSITORY_VARIABLE in builtin/rerere.c with such a small change (even though as I've said I do not think it is high priority to make cmd_foo() not rely on the_repository). Very nicely done.
Thanks.