Re: [gentoo-dev] Packages without source code

2014-04-23 Thread Ulrich Mueller
On Thu, 3 Jan 2013, Rich Freeman wrote: On Thu, Jan 3, 2013 at 9:39 AM, Ulrich Mueller u...@gentoo.org wrote: We could easily solve this by adding a binary-only or no-source-code tag to such packages. It would be included in the @BINARY-REDISTRIBUTABLE license group, but not in @FREE. So

[gentoo-dev] multilib RFC: improving wrapping compatibility with non-multilib and multilib-portage systems

2014-04-23 Thread Michał Górny
Hi, all. Following the issue brought up in bug #508278 (${CHOST}-pango-querymodules not available on non-multilib systems), I was thinking how to improve the wrapping conditionals so that it would work best for both multilib and non-multilib systems, while keeping it compatible with

[gentoo-dev] [PATCH 1/3] Move wrapping conditionals to the wrapping functions.

2014-04-23 Thread Michał Górny
This way, we don't have to repeat them in each eclass involved, and we make use of the low-level functions easier. --- eclass/cmake-multilib.eclass | 9 +++-- eclass/multilib-build.eclass | 4 eclass/multilib-minimal.eclass | 8 +++- 3 files changed, 10 insertions(+), 11

[gentoo-dev] [PATCH 2/3] Disable wrapping only if multilib-portage is used.

2014-04-23 Thread Michał Górny
In other words, enable wrapping for non-multilib systems as well. While this isn't necessary, it shouldn't hurt and it improves consistency between systems. As a result, ebuilds can now safely use '${CHOST}-executable' calls without checking whether the system is multilib. ---

[gentoo-dev] [PATCH 3/3] Create '${CHOST}-executable' symlinks in multilib-portage.

2014-04-23 Thread Michał Górny
To support multilib-portage abi-wrapper properly, we need to install executables non-prefixed on all ABIs. However, we can safely create prefixed symlinks to the executable to improve compatibility with non-multilib-portage systems. All of the symlinks point to the same executable but since it is

Re: [gentoo-dev] [PATCH 2/3] Disable wrapping only if multilib-portage is used.

2014-04-23 Thread Michał Górny
Dnia 2014-04-23, o godz. 14:57:00 Michał Górny mgo...@gentoo.org napisał(a): In other words, enable wrapping for non-multilib systems as well. While this isn't necessary, it shouldn't hurt and it improves consistency between systems. As a result, ebuilds can now safely use

[gentoo-dev] Last rites: dev-python/manifestdestiny

2014-04-23 Thread Mike Gilbert
# Mike Gilbert flop...@gentoo.org (23 Apr 2014) # Unused python library. Should have been removed with the moz* packages. # Removal in 30 days. dev-python/manifestdestiny

Re: [gentoo-dev] ARM64 stable keyword

2014-04-23 Thread William Hubbs
On Tue, Apr 22, 2014 at 09:43:24PM +0200, Tom Wijsman wrote: On Tue, 22 Apr 2014 22:13:16 +0400 Mikle Kolyada zlog...@gentoo.org wrote: 22.04.2014 21:59, Mike Gilbert пишет: Ok, then the stable keyword is going to get lost when I drop old versions. Vapier can restore stable

Re: [gentoo-dev] Re: ARM64 stable keyword

2014-04-23 Thread Rich Freeman
On Wed, Apr 23, 2014 at 12:26 AM, Duncan 1i5t5.dun...@cox.net wrote: Yes, but... I think stable keywords on such archs must be used differently, and by virtue of necessity, mean something else than they mean on more mainstream archs. This was basically the gist of the Council meeting that

Re: [gentoo-dev] ARM64 stable keyword

2014-04-23 Thread Tom Wijsman
On Wed, 23 Apr 2014 15:50:03 -0500 William Hubbs willi...@gentoo.org wrote: The arm64 profiles are marked exp in profiles.desc, so we don't have to care about the keyword status. Ah okay; nevermind then, my concern didn't take that into account. -- With kind regards, Tom Wijsman (TomWij)

Re: [gentoo-dev] ARM64 stable keyword

2014-04-23 Thread Anthony G. Basile
On 04/23/2014 05:04 PM, Tom Wijsman wrote: On Wed, 23 Apr 2014 15:50:03 -0500 William Hubbs willi...@gentoo.org wrote: The arm64 profiles are marked exp in profiles.desc, so we don't have to care about the keyword status. Ah okay; nevermind then, my concern didn't take that into account. I