On Thu, Jan 24, 2019 at 08:11:05AM -0500, Jeff King wrote:
> I've mentioned before that I have a series which compiles cleanly with
> -Wunused-parameters. I split this work roughly into three groups:
> 
>   1. Patches that fix bugs (i.e., we should have been using the
>      parameter but didn't).
> 
>   2. Patches that drop unused parameters (i.e., code cleanup).
> 
>   3. Patches that annotate undroppable parameters (e.g., ones that are
>      present due to a callback interface).
> 
> All of the patches from group 1 have been merged already. So this series
> starts us off on group 2. There are about 50 patches in that group.
> Given that none of them are urgent, I plan to feed them in batches to
> avoid overwhelming reviewers. I'm also ordering them to avoid conflicts
> with other topics in flight (this batch has no conflicts with 'next',
> and only one minor textual conflict with 'pu').

All of these patches seemed straightforward. I did see a few where there
may have originally been some consistency benefit to keeping parameters
(always passing file name to *_to_worktree, for example), but I'm fine
dropping them if it means we can get better development help from the
compiler.

I am also, for the record, in favor of ignoring the effects of
relativity for the purposes of Git. :)
-- 
brian m. carlson: Houston, Texas, US
OpenPGP: https://keybase.io/bk2204

Attachment: signature.asc
Description: PGP signature

Reply via email to