Bug#390888: [Pbuilder-maint] Bug#390888: Full support for experimental

Junichi Uekawa dancer at netfort.gr.jp
Tue Oct 24 15:30:28 UTC 2006


Hi,

> > One thing that I'm weary about applying this change is that, by
> > applying this change, and potentially introducing breakage to Debian
> > sid, people will start filing serious FTBFS bugs. At this time of
> > imminent freeze, I don't think it's a good idea to generally upload
> > such change into unstable.
> 
>  This is of course what I would expect from a good maintainer.
> 
>  I don't think there's the risk of serious bugs as, IIRC, the patched
>  pbuilder will only run new code if the normal path fails.
> 
>  Perhaps you can try running with the patch and see for yourself?

One thing I'm missing from pbuilder testsuite is a testsuite for Build-Depends parsing.

I would like to do one of the following 

1. A script to ensure all Build-Depends*: field in Sources.gz can be
   parsed and installed in clean chroot. Might be more feasible now
   with cowdancer. This would give me confidence that
   pbuilder-satisfydepends is working.

2. A collection of corner-case samples of Build-Depends. This probably
   needs to be worked on in a longer-term to find regressions.


The main problem being that I don't have too much resource at hand.

regards,
	junichi
-- 
dancer@{debian.org,netfort.gr.jp}   Debian Project




More information about the Pbuilder-maint mailing list