[Python-modules-team] Backporting matplotlib and its dependencies (Was: Sagemath Build-Depends backported - unfortunately build error (Was: Help needed to backport ipywidgets ... (Was: Backporting sagemath))]

[Andreas Tille]

Hi,

On Sun, Apr 23, 2017 at 12:08:57AM +0200, Andreas Tille wrote:
> On Sat, Apr 22, 2017 at 12:09:00PM +0000, Ximin Luo wrote:
> > >> So, you could try to build the version currently in sid, against sid. If it FTBFS you can file an RC bug and force the maintainer to fix it. :p
> > > 
> > > I admit I do not understand this hint, sorry.
> > 
> > If you look at the "Logs" column of the table linked above, it says "no logs" for arch:all and arch:amd64. This means the DD built these locally and uploaded the binaries, the buildds did not build for these architectures.
> > 
> > Since your error is in generating the documentation, there might be an inherent problem with building that documentation, that was obscured on the DD's own machine. (Maybe missing B-D). I was suggesting that you try the build against sid, to check that if this is the case. If so (i.e. it FTBFS), then it needs a RC bug filed against it. If not, then my hint here isn't useful.
> 
> I can confirm that matplotlib builds fine in an unstable chroot
> and I can not see any reason for a bug report.

Is there any other suggestion how to get matplotlib build in Jessie
backports? 

Kind regards

      Andreas.

-- 
http://fam-tille.de