[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: New perl packages uploaded to the SVN



Vincent Danjean dijo [Sat, Dec 08, 2007 at 10:29:18AM +0100]:
> > I agree, a policy "forbidding" CDBS is not what I want; but stating
> > the current situation as a matter of fact -- that most members don't
> > like CDBS (for its black magic and lack of documentation) and that
> > the probability for sponsorship is lowder therefore -- seems ok to
> > me.
> 
> Just for the record (as all people that talked here express concerns with
> cdbs): I like CDBS and I use it for most of my packages (except packages I
> created with dh-make-perl...)
> I agree with the lack of documentation but I welcome the fact that the
> packaging of lots of packages can be improved by modifying only one package
> (CDBS) without the need to edit all packages. For example, automake-ed
> packages automatically gain support for cross-compiling, management of
> noopt, nostrip ... options, ...
> I'm always a bit disapointed when I see all perl package with the same
> debian/rules file. Or, more exactly, debian/rules of a perl package can
> tell us the date of the packaging. They are so similar that a include of
> a common file that can be improved over time seems a better technical
> solution to me.
> Bt the way, I does not have enought free time to document CDBS or to work
> on the packaging of all perl packages, so let's continue with plain debhelpers
> debian/rules.

FWIW, I'm on the camp that prefers dh_* but does not really oppose
CDBS - I use CDBS for my Ruby modules (because it's the group policy,
go figure)... So, Vincent: Is there a good CDBS rules file that acts
quite like our standard behaviour? (or going a couple of steps
further: Like the four dh-make-perl variants? MakeMaker arch:all,
MakeMaker arch:any, Module::Build arch:all and Module::Build arch:any)
If so, well... We could properly study and document them... And see if
migration is possible/desirable :)

It would _clearly_ reduce duplication. But yes, it would increase
obscurity. 

-- 
Gunnar Wolf - gwolf@gwolf.org - (+52-55)5623-0154 / 1451-2244
PGP key 1024D/8BB527AF 2001-10-23
Fingerprint: 0C79 D2D1 2C4E 9CE4 5973  F800 D80E F35A 8BB5 27AF



Reply to: