Re: removing mutt patches

On Thu, May 04, 2023 at 06:33:05PM -0400, Derek Schrock wrote:
> tl;dr Removing feature patches in mail/mutt!  Do you care?
> 
> A recent bug [1] has been created with a claim that one of the remote
> patches QUOTE_PATCH [2] causes issues even without using the feature.
> This patch was originally a vvv patch [3] left unmaintained then later 
> taken over by some capacity by va [4].
> 
> While it was taken over (mainly for context updates?) I don't think it's
> fully maintained since there's still parts of it that fail at build
> time.  However, this got me thinking that maybe it's time to just drop
> the quote patch all together... Thinking some more why not all remote
> patches... Or maybe all patches all together.  I'm not really looking to
> maintain feature patches.  That's best left to fight upstream to
> include.
> 
> I feel these patches were added maybe by request or maybe they were used
> by the previous mail/mutt maintainers.  Trying to track down some
> patches removed for reasons/requests normally just lead to a
> non-descriptive commit of the inclusion so this might be all
> speculation.
> 
> However, given that some of the patches are part of neomutt I think that
> if you need these features you can either install that or build locally
> with local patches or better yet work with upstream to include to
> directly to mutt.
> 
> I've always wanted to drop all patches since I don't use any of the
> features so in turn don't have a good means to test however I've kept
> them in for legacy reasons.
> 
> Even though the mailing list will be a limited set of users I think it
> might be a good (or at least the best) way to test user feedback.  Maybe
> poking the mutt-user mailing list too could be a good idea.
> 
> However, with them removed it only means faster port releases and better
> well tested packages.
> 
> So do any of the ports/pkg mutt users deeply depend on any feature
> patches from the port and would the world end if you didn't have them?
> 
> [1] https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=271190
> [2] http://vc.org.ua/mutt/mutt-2.2.10.vvv.quote
> [3] https://www.mutt.org.ua/download/stable/
> [4] http://vc.org.ua/mutt/

I use %I specifier from vvv.initials (QUOTE_PATCH option), but have vvv.quote
commented out in the port's Makefile. I don't remember exactly why, it seems
caused some formatting issues.
I've no opinion about vc.greeting, patch-ifdef, patch-reverse_reply.
The patch-date-conditional plus patch-dgc-deepif is a more flexible variant of
patch-smartdate.
The patch-maildir-mailtime is occasionally useful.
The smime-sender can be safely dropped.

tl;dr: I would like vvv.initials to stay. Will be nice to have maildir-mailtime,
but I can live without it. I've no opinion about other patches.


-- 
Alex