Re: packages newer in Ubuntu than in Debian (reduced false positives)

[Eric Lavarde <deb@zorglub.s.bawue.de>]

Hello Bart,

is there some kind of agreement between Debian and Ubuntu concerning the 
distribution part of the version?

I ask this because you seem to assume that:
	X.Y.Z-K (Debian) << X.Y.Z-L (Ubuntu)
	X.Y.Z-K (Debian) << X.Y.Z-KubuntuA (Ubuntu)
	(also dfsg stuff doesn't seem to be completely right)

I would suggest that you split your report between:
	1) A.B.C << X.Y.Z
and
	2) X.Y.Z-KKK << X.Y.Z-LLL

The 2nd part would be a watchout (yellow) whereas the 1st part would 
really mean a different version of the software (red).
Well... Not always true but good enough :-) (check liblog4net-cil with 
1.2.8+1.2.9beta-1 vs. 1.2.9beta-0ubuntu2)

Cheers, Eric

Bart Martens wrote:
> On Tue, 2007-05-01 at 11:49 +0200, Martin Michlmayr wrote:
> > * Bart Martens <bartm@debian.org> [2007-05-01 11:21]:
> > > Another approach for identifying packages to be updated in Debian to
> > > newer upstream releases is by comparing Debian with Ubuntu.  Here is a
> > > list of packages that are newer in Ubuntu than in Debian, grouped by
> > > maintainer:
> > > http://people.debian.org/~bartm/borg/outdated.html
> > The first package in your list shows that you're not handling epochs
> > properly.  1.0~rc1-13 vs 2:1.0~rc1-0ubuntu9.  Debian isn't out of date
> > here.  You need to ignore epochs.
>
> The handling of epochs is correct, but some packages have different
> epochs in Debian and Ubuntu for the same upstream version.
>
> I have updated the list to hide packages with identical upstream version
> numbers but with different epochs.  This might hide some real positives
> but most likely hides more false positives.
>
> So thanks for the feedback; this is useful.
>
> Regards,
>
> Bart Martens