Re: binNMUs for arch:all packages too

2009-07-07 Thread Matthias Julius
Frank Küster writes: > The ${source:Version} thing is a point. However, I also see a need for > such binNMUs, or rather a case where it would be helpful. > > That's the case when a package Build-Depends on some package because it > needs to incorporate code (or configuration settings or data or

Re: binNMUs for arch:all packages too

2009-07-07 Thread Frank Küster
Steve Langasek wrote: > On Mon, Jul 06, 2009 at 01:50:14PM +0300, Kari Pahula wrote: >> only builds foo_1.0-1+b1 for arch:any packages. No +b1 is built for >> any possible -doc packages. Often, this is what's expected, but not >> always. > >> I don't know offhand if this has been discussed befo

Re: binNMUs for arch:all packages too

2009-07-06 Thread Steve Langasek
Also: On Mon, Jul 06, 2009 at 01:50:14PM +0300, Kari Pahula wrote: > Haskell library -doc packages include .haddock files, which are > derived from the ABI at build time. If this is true, then it is an *error* to include these files in an architecture: all package. There is no guarantee that th

Re: binNMUs for arch:all packages too

2009-07-06 Thread Philipp Kern
On 2009-07-06, Kari Pahula wrote: >> > What's needed to get this working? >> I don't think it should be made to work. > So you think that sourceful uploads for Haskell libraries is the > expected thing to do? We're talking about tens of packages. You can do no-change sourceful uploads, yes. And

Re: binNMUs for arch:all packages too

2009-07-06 Thread Kari Pahula
On Mon, Jul 06, 2009 at 10:33:56AM -0700, Steve Langasek wrote: > Because: > > - there are no autobuilders configured to build arch: all packages in debian Because there was no need for them before. > - allowing arch: all packages to be binNMUed breaks the invariant that > packages may use ${b

Re: binNMUs for arch:all packages too

2009-07-06 Thread Steve Langasek
On Mon, Jul 06, 2009 at 01:50:14PM +0300, Kari Pahula wrote: > only builds foo_1.0-1+b1 for arch:any packages. No +b1 is built for > any possible -doc packages. Often, this is what's expected, but not > always. > I don't know offhand if this has been discussed before. Any reasons > why this sho

binNMUs for arch:all packages too

2009-07-06 Thread Kari Pahula
Currently, asking for nmu foo_1.0-1 . ALL . -m 'rebuild against bar 2.3' only builds foo_1.0-1+b1 for arch:any packages. No +b1 is built for any possible -doc packages. Often, this is what's expected, but not always. The specific scenario I have in mind is Haskell libraries. They change ABIs