Eli Zaretskii <eliz@gnu.org> writes:
>> From: Spencer Baugh <sbaugh@janestreet.com>
>> Cc: Mekeor Melire <mekeor@posteo.de>,  Author <sbaugh@catern.com>,
>>    66283@debbugs.gnu.org,  Committer <eliz@gnu.org>
>> Date: Wed, 04 Oct 2023 11:13:06 -0400
>> 
>> Two fixes to the previous patch:
>> 
>> - There was always a bug with which-function-mode where if you enabled
>>   it when there were already existing buffers, it would behave
>>   differently than if you enabled it and then created those buffers.
>>   Namely it would enable which-func-mode for every buffer, even if the
>>   buffer didn't support imenu.  This is especially noticeable when
>>   toggling which-function-mode and using the header line, so I fixed it.
>> 
>> - Also, I accidentally dropped a paren before submitting.  Now that's
>>   fixed :)
>
> Thanks, but how about some tests for these fixes?  It's high time
> which-mode had some test suite, I think.

Sure, here's a patch including a test.