GNU bug report logs - #31974
If a phase returns #f, a warning is issued, but the build continues

Previous Next

Full log

View this message in rfc822 format

From: help-debbugs <at> gnu.org (GNU bug Tracking System)
To: Mark H Weaver <mhw <at> netris.org>
Subject: bug#31974: closed (Re: bug#31974: If a phase returns #f, a
 warning is issued, but the build continues)
Date: Wed, 09 Jan 2019 20:39:03 +0000

[Message part 1 (text/plain, inline)]

Your bug report

#31974: If a phase returns #f, a warning is issued, but the build continues

which was filed against the guix package, has been closed.

The explanation is attached below, along with your original report.
If you require more details, please reply to 31974 <at> debbugs.gnu.org.

-- 
31974: http://debbugs.gnu.org/cgi/bugreport.cgi?bug=31974
GNU Bug Tracking System
Contact help-debbugs <at> gnu.org with problems

[Message part 2 (message/rfc822, inline)]

From: Ludovic Courtès <ludo <at> gnu.org>
To: Mark H Weaver <mhw <at> netris.org>
Cc: 31974-done <at> debbugs.gnu.org
Subject: Re: bug#31974: If a phase returns #f, a warning is issued,
 but the build continues
Date: Wed, 09 Jan 2019 21:38:25 +0100

Mark H Weaver <mhw <at> netris.org> skribis:

> I just noticed that I made a mistake in commit
> d8a3b1b9e847d4a44d2695f95af77170d4d2788f, which changed 'gnu-build' in
> (guix build gnu-build-system) to issue a warning if a phase returns a
> value other than #t.
>
> The result is that if a phase returns a value other than #t, a warning
> is issued, but the build nonetheless continues to the next phase, and
> the build could ultimately "succeed" even some phases returned #f.

This was fixed in commit 82230603ce06de7aa3e4aef2fa093a6dbf0ef8df.
Closing!

Ludo'.

[Message part 3 (message/rfc822, inline)]

From: Mark H Weaver <mhw <at> netris.org>
To: bug-guix <at> gnu.org
Subject: If a phase returns #f, a warning is issued, but the build continues
Date: Mon, 25 Jun 2018 23:13:03 -0400

I just noticed that I made a mistake in commit
d8a3b1b9e847d4a44d2695f95af77170d4d2788f, which changed 'gnu-build' in
(guix build gnu-build-system) to issue a warning if a phase returns a
value other than #t.

The result is that if a phase returns a value other than #t, a warning
is issued, but the build nonetheless continues to the next phase, and
the build could ultimately "succeed" even some phases returned #f.

In other words, the return values of phases are effectively ignored,
except for the warning, so the only way to cause a build with phases to
fail is to raise an exception.

We can't fix this problem without a full rebuild.  Sorry about that :-(

The only good news here is that I've already fixed all of the core
packages to return #t from all phases and to report errors using
exceptions.  Ditto for all of the packages that I use on my GNOME
desktop.  The remaining packages that might return #f from a phase could
be fixed in 'master'.

      Mark

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.