Re: binNMUing on some architectures breaks Multi-arch: same

2012-04-13 Thread Adam D. Barratt
On 13.04.2012 08:58, Goswin von Brederlow wrote: "Adam D. Barratt" writes: On 11.04.2012 10:30, Goswin von Brederlow wrote: The current state is that M-A: same packages must have the same version and identical Changelog files across all architectures. That means binNMU on all architectures or

Re: binNMUing on some architectures breaks Multi-arch: same

2012-04-13 Thread Goswin von Brederlow
"Adam D. Barratt" writes: > On 11.04.2012 10:30, Goswin von Brederlow wrote: >> The current state is that M-A: same packages must have the same >> version >> and identical Changelog files across all architectures. That means >> binNMU on all architectures or a sourcefull upload. > > A source uplo

Re: binNMUing on some architectures breaks Multi-arch: same

2012-04-11 Thread Joachim Breitner
Hi, Am Mittwoch, den 11.04.2012, 12:05 +0100 schrieb Adam D. Barratt: > >> A source upload isn't "just a rebuild" in Debian. It's an NMU and > >> requires building and testing the package with at least as much care > >> as > >> any other NMU would. It doesn't scale, it's a waste of resources >

Re: binNMUing on some architectures breaks Multi-arch: same

2012-04-11 Thread Adam D. Barratt
On 11.04.2012 11:50, Joachim Breitner wrote: Am Mittwoch, den 11.04.2012, 11:01 +0100 schrieb Adam D. Barratt: On 11.04.2012 10:30, Goswin von Brederlow wrote: > The current state is that M-A: same packages must have the same > version > and identical Changelog files across all architectures. Th

Re: binNMUing on some architectures breaks Multi-arch: same

2012-04-11 Thread Joachim Breitner
Hi, Am Mittwoch, den 11.04.2012, 11:01 +0100 schrieb Adam D. Barratt: > On 11.04.2012 10:30, Goswin von Brederlow wrote: > > The current state is that M-A: same packages must have the same > > version > > and identical Changelog files across all architectures. That means > > binNMU on all archite

Re: binNMUing on some architectures breaks Multi-arch: same

2012-04-11 Thread Julian Andres Klode
On Wed, Apr 11, 2012 at 11:47:35AM +0200, Julien Cristau wrote: > On Wed, Apr 11, 2012 at 11:30:08 +0200, Goswin von Brederlow wrote: > > > The current state is that M-A: same packages must have the same version > > and identical Changelog files across all architectures. That means > > binNMU on a

Re: binNMUing on some architectures breaks Multi-arch: same

2012-04-11 Thread Adam D. Barratt
On 11.04.2012 10:30, Goswin von Brederlow wrote: The current state is that M-A: same packages must have the same version and identical Changelog files across all architectures. That means binNMU on all architectures or a sourcefull upload. A source upload isn't "just a rebuild" in Debian. It'

Re: binNMUing on some architectures breaks Multi-arch: same

2012-04-11 Thread Julien Cristau
On Wed, Apr 11, 2012 at 11:30:08 +0200, Goswin von Brederlow wrote: > The current state is that M-A: same packages must have the same version > and identical Changelog files across all architectures. That means > binNMU on all architectures or a sourcefull upload. > Or accept that such packages a

Re: binNMUing on some architectures breaks Multi-arch: same

2012-04-11 Thread Goswin von Brederlow
Julien Cristau writes: > On Sat, Apr 7, 2012 at 12:15:58 +0200, Julian Andres Klode wrote: > >> Discussion in Bug#595139 led to the conclusion that packages which are >> Multi-arch: same must not be binNMUed (or in fact, binNMUed on all >> architectures). >> > I very much disagree with that "co

Re: binNMUing on some architectures breaks Multi-arch: same

2012-04-07 Thread Julien Cristau
On Sat, Apr 7, 2012 at 12:15:58 +0200, Julian Andres Klode wrote: > Discussion in Bug#595139 led to the conclusion that packages which are > Multi-arch: same must not be binNMUed (or in fact, binNMUed on all > architectures). > I very much disagree with that "conclusion". Cheers, Julien signa

Re: binNMUing on some architectures breaks Multi-arch: same

2012-04-07 Thread Adam D. Barratt
On Sat, 2012-04-07 at 12:15 +0200, Julian Andres Klode wrote: > So, a solution would be to add code to your tools to prevent Multi-arch: > same packages from being binNMUed on a subset of the architectures (or at > all) and upload sourceful NMUs for the affected packages. Assuming by "your" here,

Re: binNMUing on some architectures breaks Multi-arch: same

2012-04-07 Thread Julian Andres Klode
On Sat, Apr 07, 2012 at 10:56:27AM +0100, Adam D. Barratt wrote: > On Sat, 2012-04-07 at 11:52 +0200, Julian Andres Klode wrote: > > it seems that libffi5 received a binNMU on armhf. As libffi5 is > > marked Multi-arch: same, it is now not possible to have the armhf > > version installed on an amd6

Re: binNMUing on some architectures breaks Multi-arch: same

2012-04-07 Thread Adam D. Barratt
On Sat, 2012-04-07 at 11:52 +0200, Julian Andres Klode wrote: > it seems that libffi5 received a binNMU on armhf. As libffi5 is > marked Multi-arch: same, it is now not possible to have the armhf > version installed on an amd64 system, as the armhf version has a > binNMU and the amd64 one does not

binNMUing on some architectures breaks Multi-arch: same

2012-04-07 Thread Julian Andres Klode
Hi everyone, it seems that libffi5 received a binNMU on armhf. As libffi5 is marked Multi-arch: same, it is now not possible to have the armhf version installed on an amd64 system, as the armhf version has a binNMU and the amd64 one does not have one. The following packages have unmet dependencie