Re: [RFC PATCH 0/3] refspec: centralize refspec-related logic

> On Wed, Jan 22, 2025 at 01:21:51PM +0530, Meet Soni wrote:
>> As Patrick pointed out in [1], the logic related to refspec is currently
>> split across multiple headers. This patch series addresses that by
>> relocating refspec-related logic from remote to refspec for improved
>> cohesion.
>> 
>> There are a few functions that seem to be on the borderline. They seem
>> to have refspec-related logic, but also contain some elements related
>> to remote logic, so I'm unsure whether they should be moved.
>> Specifically:
>> 
>>     * get_fetch_map()
>>     * count_refspec_match()
>>     * check_push_refs()
>>     * match_push_refs()
>> 
>> [1]: https://lore.kernel.org/git/ZysQvUyxgdRqjvj2@pks.im/
>
> Thanks for this series! A couple comments from my side, but I think that
> the overall goal of this series is sensible (well, I seemingly proposed
> it at one point, so no surprise there).