Re: master bfa951cdfa: Fix invalid current buffer after print-unreadable-function signals

Re: master bfa951cdfa: Fix invalid current buffer after print-unreadable-function signals

[Lars Ingebrigtsen]

Po Lu via Mailing list for Emacs changes <emacs-diffs@gnu.org> writes:

>     Fix invalid current buffer after print-unreadable-function signals
>
>     * src/print.c (PRINTPREPARE): Remove `old' and
>     record_unwind_current_buffer instead.
>     (PRINTFINISH): Stop restoring `old'.  (bug#56773)

This apparently leads to:

1 unexpected results:
   FAILED  terpri

in print-tests.el.

Re: master bfa951cdfa: Fix invalid current buffer after print-unreadable-function signals

[Stefan Kangas]

Lars Ingebrigtsen <larsi@gnus.org> writes:

> Po Lu via Mailing list for Emacs changes <emacs-diffs@gnu.org> writes:
>
>>     Fix invalid current buffer after print-unreadable-function signals
>>
>>     * src/print.c (PRINTPREPARE): Remove `old' and
>>     record_unwind_current_buffer instead.
>>     (PRINTFINISH): Stop restoring `old'.  (bug#56773)
>
> This apparently leads to:
>
> 1 unexpected results:
>    FAILED  terpri
>
> in print-tests.el.

If there is no quick fix, could we please revert the change?

It's currently stopping me from merging emacs-28 to master.

Re: master bfa951cdfa: Fix invalid current buffer after print-unreadable-function signals

[Eli Zaretskii]

> From: Stefan Kangas <stefankangas@gmail.com>
> Date: Thu, 28 Jul 2022 01:14:30 -0700
> Cc: Po Lu <luangruo@yahoo.com>
> 
> Lars Ingebrigtsen <larsi@gnus.org> writes:
> 
> > Po Lu via Mailing list for Emacs changes <emacs-diffs@gnu.org> writes:
> >
> >>     Fix invalid current buffer after print-unreadable-function signals
> >>
> >>     * src/print.c (PRINTPREPARE): Remove `old' and
> >>     record_unwind_current_buffer instead.
> >>     (PRINTFINISH): Stop restoring `old'.  (bug#56773)
> >
> > This apparently leads to:
> >
> > 1 unexpected results:
> >    FAILED  terpri
> >
> > in print-tests.el.
> 
> If there is no quick fix, could we please revert the change?
> 
> It's currently stopping me from merging emacs-28 to master.

Why does it prevent a merge?

Re: master bfa951cdfa: Fix invalid current buffer after print-unreadable-function signals

[Stefan Kangas]

Eli Zaretskii <eliz@gnu.org> writes:

> Why does it prevent a merge?

Because the tests are failing, and admin/automerge doesn't like that.

Re: master bfa951cdfa: Fix invalid current buffer after print-unreadable-function signals

[Eli Zaretskii]

> From: Stefan Kangas <stefankangas@gmail.com>
> Date: Thu, 28 Jul 2022 01:47:47 -0700
> Cc: larsi@gnus.org, emacs-devel@gnu.org, luangruo@yahoo.com
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> > Why does it prevent a merge?
> 
> Because the tests are failing, and admin/automerge doesn't like that.

I think you are taking this too seriously.  A known failure on master
shouldn't hold off the merge, IMO.  Can't you ignore the test failure?

Re: master bfa951cdfa: Fix invalid current buffer after print-unreadable-function signals

[Stefan Kangas]

Eli Zaretskii <eliz@gnu.org> writes:

> I think you are taking this too seriously.  A known failure on master
> shouldn't hold off the merge, IMO.  Can't you ignore the test failure?

Sure, it's a bit more work is all.  It's not the end of the world, but
I'd prefer if we could avoid breaking the test suite on master
instead.

Re: master bfa951cdfa: Fix invalid current buffer after print-unreadable-function signals

[Eli Zaretskii]

> From: Stefan Kangas <stefankangas@gmail.com>
> Date: Thu, 28 Jul 2022 11:14:57 +0200
> Cc: Lars Ingebrigtsen <larsi@gnus.org>, Emacs developers <emacs-devel@gnu.org>, Po Lu <luangruo@yahoo.com>
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> > I think you are taking this too seriously.  A known failure on master
> > shouldn't hold off the merge, IMO.  Can't you ignore the test failure?
> 
> Sure, it's a bit more work is all.  It's not the end of the world, but
> I'd prefer if we could avoid breaking the test suite on master
> instead.

But you aren't breaking the test suite, it's already broken.

The part of automerge that runs the test suite is meant to ensure that
the merge doesn't introduce new regressions.  This is not that case.

Re: master bfa951cdfa: Fix invalid current buffer after print-unreadable-function signals

[Stefan Kangas]

Eli Zaretskii <eliz@gnu.org> writes:

> The part of automerge that runs the test suite is meant to ensure that
> the merge doesn't introduce new regressions.  This is not that case.

Of course, but it the script is too primitive to decide if the
regression is new or not.  So every time it happens, I need to
intervene manually.