On Wed, 6 Aug 2025 09:05:31 +0200
Ralf Hildebrandt via Postfix-users <postfix-users@postfix.org> wrote:
> > thanks for the continued or revived maintenance of this venerable
> > and useful software.
>
> Same here.
You're likewise welcome, Ralf :)
>
> > I came across one change which broke my crontab, and that is that
> > 1.1.11 renamed --verp_mung to --verp-mung.
[snip]
>
> And yes, this happeneded to me too, but then I just read the man
> page :) (afterwards, that is :) )
Hah, yeah—I may have to find a better way to handle versioning.
On Wed, 6 Aug 2025 10:53:56 +0200
Matthias Andree via Postfix-users <postfix-users@postfix.org> wrote:
> Am 06.08.25 um 05:35 schrieb Jim Seymour via Postfix-users:
[snip]
> > I did. It's both in the ChangeLog and was posted here:
[snip]
> >
> > That and the other major UI change were right at the top.
>
> Yes, right. I missed that because I was looking specifically for
> "verp". I also only cursorily read this because ...
[snip]
>
> Also, in my software, I routinely have a section "Breaking Changes"
> or similar that documents changed interfaces to make those changes
> stand out, ...
That's a good suggestion. In fact, here’s how the 1.1.12 (Beta)
ChangeLog is currently shaping up:
rel-1.1.12 202508??
*** Breaking Changes ***
Date::Calc now Required due to date(-range) enhancements.
UI Changes
Renamed option --unprocd to --unprocd-file for better
clarity. (--unprocd was introduced in v1.1.11.)
Changed --pscrn-detail option to behave like all other
report-limiting options: Full detail unless this option
is specified to limit or suppress it. (--pscrn-detail was
introduced in v1.1.11.)
Do you think something like that would better flag changes in the
future?
> ... I would have bumped the major revision to 2.0.0 ...
Once the current set of changes settles, I plan to bump to v1.2.0 for
the Production release. I reserve full major version bumps (e.g., 1.x
--> 2.x) for very substantial architectural or behavioral changes.
That said, you're absolutely right: Dropping underscore-style options
should’ve had explicit notice and a deprecation period. In hindsight,
my focus was more on tidying up than on minimizing disruption. Lesson
learned.
And, in my defense: These *were* Beta releases. Personally, I tend to
expect Beta releases to possibly be a mite unpredictable ;)
> ... in that situation unless that
> old form with --verp_mung would have been announced deprecated for
> a long time.
Yeah, that one's definitely on me. I'd *thought* it had been, when,
in fact, the only "deprecated" notifications were for switches that
were being entirely discontinued.
> So maybe I need to not assume everyone does that
> semantic versioning, too... <https://semver.org> :-)
Yeah—I aim to follow SemVer loosely. That is, I try not to bump the
major version unless the changes really reshape the tool. But you've
got me thinking that I might need to be more communicative about
compatibility changes either way.
There may eventually be a pflogsumm 2.x (I've been noodling ideas),
but if that happens it’ll be a significant redesign. For example: The
current interface inconsistency—some flags disable features by
default, others enable them—that kind of thing would get reworked.
In any case: I’m sorry for the breakage. I really do appreciate the
feedback, and I’ll be more cautious and deliberate going forward.
Regards,
Jim
--
Note: My mail server employs *very* aggressive anti-spam
filtering. If you reply to this email and your email is
rejected, please accept my apologies and let me know via my
web form at <http://jimsun.LinxNet.com/contact/scform.php>.
_______________________________________________
Postfix-users mailing list -- postfix-users@postfix.org
To unsubscribe send an email to postfix-users-le...@postfix.org
