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

Re: Misc developer news (#21)

Mike Hommey schrieb am Saturday, den 20. February 2010:

> On Sat, Feb 20, 2010 at 09:03:10AM +0100, Marc Haber wrote:
> > On Fri, 19 Feb 2010 18:51:11 +0100, Raphael Hertzog
> > <hertzog@debian.org> wrote:
> > >On Fri, 19 Feb 2010, Marc Haber wrote:
> > >> On Fri, Feb 19, 2010 at 12:08:16PM +0100, Raphael Hertzog wrote:
> > >> >   * More than 1000 source packages[4] are already using the new source
> > >> >     formats "3.0 (quilt)" and "3.0 (native)". Have you updated your own
> > >> >     packages already?
> > >> 
> > >> Why should I?
> > >
> > >http://wiki.debian.org/Projects/DebSrc3.0#WhyshouldIconvertmypackageto3.0.28quilt.29format.3F
> > 
> > I don't see any valid reason to convert packages which already use a
> > patch system such as dpatch to the new thing. Debian is in dire need
> > of manpower taking care of our core infrastructure, converting
> > dpatch-based packages to quilt is a total waste of manpower.
> If your dpatches are simply patches and not scripts, converting to 3.0
> (quilt) is a few minutes job: rename the files, rename 00list to series,
> adapt its content, remove your patch system from debian/rules, change
> the build-deps accordingly, write a debian/source/format file containing
> 3.0 (quilt). Done.
I think dpatch is much easier to use and simpler to debug than quilt (yes I
use both systems). 

So I still don't see a reason to convert. 

For many packages (one upstream, already working patch system) converting to
quilt just means more work. 

source 3.0 (quilt) is still missing tools to make it working out of the box.
If you quilt not only in debian packages its fucking annoying that you need
to change to patch of your patches from patches/ to debian/patches depending
on what you are working on. 

IMHO 3.0 would have been much more successfull if it wouldn't have called
quilt and dpkg would have all tools needed to work with the system (push,
pop, create patch and so on). 

The way it is now, I don't see a reason to invest the extra work.

Just my 2 cent


Reply to: