Re: [PATCH 11/17] refs: replace `refs_for_each_ref_in()`

> On Mon, Feb 23, 2026 at 04:11:52AM -0500, Karthik Nayak wrote:
>> Patrick Steinhardt <ps@pks.im> writes:
>> > diff --git a/builtin/rev-parse.c b/builtin/rev-parse.c
>> > index 9032cc6327..02703f2fb8 100644
>> > --- a/builtin/rev-parse.c
>> > +++ b/builtin/rev-parse.c
>> > @@ -613,13 +613,18 @@ static int opt_with_value(const char *arg, const char *opt, const char **value)
>> >
>> >  static void handle_ref_opt(const char *pattern, const char *prefix)
>> >  {
>> > -	if (pattern)
>> > +	if (pattern) {
>> >  		refs_for_each_glob_ref_in(get_main_ref_store(the_repository),
>> >  					  show_reference, pattern, prefix,
>> >  					  NULL);
>> > -	else
>> > -		refs_for_each_ref_in(get_main_ref_store(the_repository),
>> > -				     prefix, show_reference, NULL);
>> > +	} else {
>> > +		struct refs_for_each_ref_options opts = {
>> > +			.prefix = prefix,
>> > +			.trim_prefix = strlen(prefix),
>>
>> Tangent: I wonder if it makes sense to make `trim_prefix` a bool and
>> then internally trim strlen(prefix). Is there a usecase where
>> `.trim_prefix != strlen(prefix)`?
>
> I don't think there is right now, and I cannot think about any myself.
> How about we leave this as a #leftoverbit though?
>
> Patrick