Re: isfinite-no-c++: Make more C++ safe

"G. Branden Robinson" <[email protected]> Tue, 3 Feb 2026 11:33:52 -0600

Newsgroups	gmane.comp.printing.groff.general,gmane.comp.lib.gnulib.bugs
Message-ID	<20260203173352.pbti4ijsk4mj5365@illithid>

[looping in groff list; release plans are discussed]

At 2026-02-03T18:11:41+0100, Bruno Haible wrote:
> G. Branden Robinson wrote:
> > > This patch now fixes it, by ensuring REPLACE_ISFINITE = 0 instead
> > > of = 1.
> > 
> > Thanks for the heads-up.  Are you planning to cherry-pick this fix
> > over to the stable-202601 branch?
> 
> Yes.
> 
> The next update of the stable branches is due on 2026-02-15. Does this
> fit with your planned groff release schedule?

Probably.  I don't have a "schedule"; the scheme is:

1.  Cut a release candidate (RC).
2.  See what feedback comes back.
3.  If something demands fixing before we can release with a clear
    conscience, fix it, collect other items like this, and go to step 1.
4.  Otherwise, nothing alarming has been revealed about the most recent
    RC.  Make a final release.

It appears that: we exhaust our testers' supply of feedback within 8
days from a release candidate; we're due for an rc3; and I have
accumulated sufficient changes to make one (about 64 commits).  But I
also still have a C++ code regression to fix before we can go final.[1]

Even if I put out rc3 today, it will be about the 11th before I close
the evaluation window.  I'm not averse to waiting another few days after
that to cut rc4.  Also, I haven't heard from Nelson Beebe since rc1, and
I'd kind of like to before going final.

Regards,
Branden

[1] https://savannah.gnu.org/bugs/?67992

signature.asc (application/pgp-signature, 833 B)