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

Re: debconf

On Tue, Oct 17, 2000 at 06:19:24PM +0200, Robert Bihlmeyer wrote:
> Marcus Brinkmann <Marcus.Brinkmann@ruhr-uni-bochum.de> writes:
> > >     export PATH=$PATH:`pwd`/Client
> [...]
> > Uh, this would be a bug.
> debconf build-depends on debconf-utils which depends on debconf,
> again. I'll ask on debian-devel if this is legal.

Well, they have the same source, but debconf build-depends on
debconf-utils, so it's really a self-dependency.  It's not really
"illegal" AFAIK, but the developer's reference says not to do this (8.1
- this is the section on being kind to porters).  There's nothing about
this that I can find in the policy or packaging manuals.

> > > and dpkg warnings are architecture
> > > 'i386' not in remapping table,
> > 
> > the latter means something calls dpkg --print-*-architecture, which
> > would be a bug, too (most likely).
> Why would that be a bug? Because the envvar-override is not heeded?
> Most debhelper scripts call this in their initialization phase - see
> /usr/lib/perl5/Debian/Debhelper/Dh_Lib.pm line 347

Marcus can answer this better than I, but I think the gist is that
the dpkg --print-*-architecture stuff is broken on hurd so he wrote
dpkg-architecture to do things "the right way".  The old way is marked as
deprecated in the dpkg-architecture man page and and it's been in dpkg-dev
for awhile now so hopefully maintainers are thinking of fixing this up.
Maybe with some bug reports they will :-).


Steve Bowman  <sbowman@frostwork.net> (preferred)
Buckeye, AZ   <sbowman@goodnet.com> <bowmanc@acm.org>

Powered by Debian GNU/Linux and GNU/Hurd <http://www.debian.org>

Reply to: