Re: git: 082542e6c694 - 2024Q1 - */*: Chase editors/emacs update

From: Joseph Mingrone <>
Date: Sun, 07 Apr 2024 03:32:28 UTC
On Sat, 2024-04-06 at 22:00, Joseph Mingrone <> wrote:

> On Sat, 2024-04-06 at 19:53, "Jason E. Hale" <> wrote:

>> On Fri, Apr 5, 2024 at 10:32 PM Joseph Mingrone <> wrote:

>>> On Fri, 2024-04-05 at 19:13, "Jason E. Hale" <> wrote:

>>> > On Tue, Mar 26, 2024 at 7:57 PM Joseph Mingrone <> wrote:

>>> >> The branch 2024Q1 has been updated by jrm:

>>> >> URL:

>>> >> commit 082542e6c694d58c24d1c425c9f06441c6a16db7
>>> >> Author:     Joseph Mingrone <>
>>> >> AuthorDate: 2024-01-26 14:13:34 +0000
>>> >> Commit:     Joseph Mingrone <>
>>> >> CommitDate: 2024-03-26 23:41:02 +0000

>>> >>     */*: Chase editors/emacs update

>>> >>     - Bump Emacs version in Mk/Uses/ to update version-specific
>>> >>       paths

>>> >>     - Bump PORTREVISION of ports with USES=emacs.  This is required for two
>>> >>       reasons.  Emacs lisp files need to be byte compiled for the new Emacs
>>> >>       version, and files installed under, e.g., EMACS_VERSION_SITE_LISPDIR
>>> >>       need to be relocated.

>>> >>     Reviewed by:    ashish
>>> >>     Sponsored by:   The FreeBSD Foundation
>>> >>     Differential Revision:

>>> >>     (cherry picked from commit ab463bdca9c29ec22be0c7e6d7aa27bb2a980b48)
>>> >> ---

>>> > Hi,

>>> > You bumped more than PORTREVISION on several ports with this cherry
>>> > pick, leading to breakage in the 2024Q1 branch since the
>>> > (POR|DIS)TVERSIONs no longer match distinfo.

>>> > - Jason

>>> My apologies if I messed up the resolution of some of the many merge
>>> conflicts.  I'll take a closer look this weekend, but since we're now in
>>> Q2, from a user perspective, isn't this now moot?

>>> J.

>> Considering the delay in cutting the 2024Q2 branch, I don't think it's
>> moot to fix errors in the current quarterly branch which is still
>> 2024Q1 at time of writing. Had the Q2 branch already been cut, I
>> wouldn't have received the numerous pkg-fallout emails that caused me
>> to investigate this in the first place, so I'm not pointing this out
>> just to be petty. :) I'm not sure how much longer we'll be on Q1, so I
>> think the responsible thing to do would be to revert the
>> PORTVERSION/DISTVERSION changes for audio/emms, devel/clojure-cider,
>> and mail/anubis. For mail/anubis, the PORTREVISION should then be 18,
>> because it was 17 before this change.

> I only had a few minutes to write the last email and didn't have a
> chance to check the quarterly branch status and just assumed Q2 had been
> cut.  I'm working on fixing things now and will loop you in soon.

I reverted the changes for the messed-up ports and just gave them a

As an aside, I think this is a good example of why another discussion on
the costs versus benefits of having a quarterly branch is warranted.
Ashish rightfully merged a security update for editors/emacs, but many
ports would have been broken without a PORTREVISION bump.  Merging the
commits for the PORTREVSION bumps was a pain because of the divergence
between the branches.

I believe the results of the recent community survey will be released
soon.  If I recall correctly, something in the range of 15% of
respondents prefer infrequent package updates.  I think that begs the
question whether maintaining two port branches is worth the effort.