Re: [gentoo-dev] Re: ffmpeg vs libav choice of default
On Sun, 08 Feb 2015 23:34:57 +0100 Luca Barbato lu_z...@gentoo.org wrote: Sadly the time I can spend doing opensource stuff can be compressed from time to time and maybe is nicer develop interesting stuff such as useful API and features than write tons of s:CODEC_ID:AV_CODEC_ID: over countless packages. can't agree more; by people I really meant fresh blood; I had been doing such boring tasks for the past years and am also busy with other things these days; even ffmpeg integration is not in the shape it used to be. (nb: you can see that this precise one is mostly fixed already.) Alexis.
Re: [gentoo-portage-dev] [PATCH 0/2] Implement soname dependencies
-BEGIN PGP SIGNED MESSAGE- Hash: SHA256 On 09/02/15 06:41, Zac Medico wrote: 49 files changed, 2766 insertions(+), 149 deletions(-) So... is anyone anywhere near qualified to (N)ACK this? - -- Alexander berna...@gentoo.org https://secure.plaimi.net/~alexander -BEGIN PGP SIGNATURE- Version: GnuPG v2 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iF4EAREIAAYFAlTY4wAACgkQRtClrXBQc7WMAAEArw5snyAG2oHnlTMiPj8N6O3Z gabXWdk5w+HCBTINlicA/iYbi9QR95trQIFj4VkVxtEbMffJ+EqqxPO4szi9taZR =LRfF -END PGP SIGNATURE-
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in eclass: toolchain.eclass
On 02/09/2015 07:19 PM, Anthony G. Basile wrote: On 02/09/15 06:40, Michał Górny wrote: This breaks a few packages [1,2]. Please fix the breakage and run repoman next time. [1]:https://travis-ci.org/gentoo/gentoo-portage-rsync-mirror/jobs/50118225#L4304 [2]:https://travis-ci.org/gentoo/gentoo-portage-rsync-mirror/jobs/50118228#L923 Someone has a shiny new toy :) Seriously this is great, but asking someone to run repoman when making a change to an eclass is extreme. You'd have to run it against the whole tree and while I haven't timed it, that's going to be a heavy task. Travis times it: 18m 10s ;-) @mgorny could you steal AutoRepomanWarning parser from Patrick and apply it to your travis job? Cheers, Kacper BTW, when I saw this I went out and added travis-ci to all my github repos. What a wonderful service. If only I can figure out how to make it work with clang static analysis. signature.asc Description: OpenPGP digital signature
Re: [gentoo-dev] Git mirror news: travis-ci running repoman, git-cvs merge revert tools
On 8 February 2015 at 18:38, Michał Górny mgo...@gentoo.org wrote: Hello, everyone. I would like to announce that our little rsync-git band-aid mirror [1] is doing fine and we're actively working towards improving Gentoo development experience. First of all, we have enabled tree-wide repoman scans using travis-ci [2]. Besides providing regularly updated repository state report, it can be used to scan Pull Requests for tree-wide damage :). Asides from the benefit to external contributors, Gentoo developers can use it to avoid having to run repoman locally. For example, if you are doing a big old version cleanup, do it in git and submit a Pull Request, and travis will figure out if you don't break any revdeps. Secondly, I have committed app-portage/lightweight-cvs-toolkit for your committing pleasure. It consists of three tools: a. lcvs-init -- that can be used to quickly create partial CVS checkout, having only pure categories checked out. The repository is set to use 'gentoo' as master, so you can easily commit into CVS while keeping the dependencies, eclasses and profiles synced to your regular rsync/git checkout (with working cache!). The idea is explained more thoroughly on the wiki [3] and in the script output. b. lcvs-merge-pr -- a convenient tool to merge github PR (or any git patch) into your CVS checkout. It 'cvs up -dP' directories as necessary, git-applies the patch omitting ChangeLogs and Manifests, calls cvs add/cvs rm as appropriate. All you have to do is update the ChangeLog and commit :). More about merging on the wiki [4]. c. lcvs-revert -- a convenient tool to revert commits. Pretty much the idea is: someone breaks something e.g. by removing ebuilds, and you want to revert that. Instead of playing all the fancy 'cvs add' magic, you just find the matching git commit and lcvs-revert it. Using logic similar to lcvs-merge-pr, it fetches the diff and reverse-applies it. Then you check if everything went fine, ChangeLog and commit :). More info on the wiki [5] as well. Thanks to all the people that helped me get this running, and have fun :). [1]:https://github.com/gentoo/gentoo-portage-rsync-mirror [2]:https://travis-ci.org/gentoo/gentoo-portage-rsync-mirror [3]:https://wiki.gentoo.org/wiki/Lightweight_CVS_Checkout [4]:https://wiki.gentoo.org/wiki/Project:Git_mirror/Merging_Pull_Requests [5]:https://wiki.gentoo.org/wiki/Project:Git_mirror/Reverting_Gentoo_commits Thank you! This looks really useful. -- Cheers, Ben | yngwin Gentoo developer
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in eclass: toolchain.eclass
On 02/09/15 06:40, Michał Górny wrote: This breaks a few packages [1,2]. Please fix the breakage and run repoman next time. [1]:https://travis-ci.org/gentoo/gentoo-portage-rsync-mirror/jobs/50118225#L4304 [2]:https://travis-ci.org/gentoo/gentoo-portage-rsync-mirror/jobs/50118228#L923 Someone has a shiny new toy :) Seriously this is great, but asking someone to run repoman when making a change to an eclass is extreme. You'd have to run it against the whole tree and while I haven't timed it, that's going to be a heavy task. BTW, when I saw this I went out and added travis-ci to all my github repos. What a wonderful service. If only I can figure out how to make it work with clang static analysis. -- Anthony G. Basile, Ph.D. Gentoo Linux Developer [Hardened] E-Mail: bluen...@gentoo.org GnuPG FP : 1FED FAD9 D82C 52A5 3BAB DC79 9384 FA6E F52D 4BBA GnuPG ID : F52D4BBA
Re: [gentoo-portage-dev] [PATCH] repoman: make SRC_URI.mirror test result non-fatal
-BEGIN PGP SIGNED MESSAGE- Hash: SHA256 LGTM. Go ahead push. - -- Alexander berna...@gentoo.org https://secure.plaimi.net/~alexander -BEGIN PGP SIGNATURE- Version: GnuPG v2 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iF4EAREIAAYFAlTY1QEACgkQRtClrXBQc7WERgEAtISEPd0KrDKwlQ39vSeZDogD eqWGIqsbfbX61XYm5QMA+wRFVz27t6LjKmOWxbGLP3UvkJMEy/KhtlMDEzLmP4ej =Z1Bf -END PGP SIGNATURE-
[gentoo-dev] Lastrite: app-crypt/pinentry-qt
-BEGIN PGP SIGNED MESSAGE- Hash: SHA512 # Kristian Fiskerstrand k...@gentoo.org (09 Feb 2015) # Dead upstream, will be removed in 30 days. # Qt support is provided by app-crypt/pinentry[qt4] app-crypt/pinentry-qt - -- Kristian Fiskerstrand Public PGP key 0xE3EDFAE3 at hkp://pool.sks-keyservers.net fpr:94CB AFDD 3034 5109 5618 35AA 0B7F 8B60 E3ED FAE3 -BEGIN PGP SIGNATURE- iQEcBAEBCgAGBQJU2R5iAAoJEP7VAChXwav6Ba0H/ifZZLKHLwspr8Brz4C2u9aB h4J7ZL2zs/ZKXGAiHn0pQhoPV7XE+O2sB+XiUcyy6jjVRvKn3rizmmztBUWibbWW MaDzz1Nuy+rzfRgZ/oi/8kknXQFkcV3m9qNs/0bX3D/9VN5KDktPn3qViX7SxR/c eGYcrlDW65d8ZlCRUcGiaMtBWWjuwXfYvr7Bh1tx6AROK3hJuWxqixdrZO5dU4Zc OYGhHuElW9XO3mSRoMC3mATWeivEf9z/SBn9hxu+WeSc4SA6T1087qMNcusHpY5a NRU2sVIzxYPIbhNXVcs9bH2jSwuqaecjMLKDRQ+6rTaKoODsBmuLHUffbN+pwlM= =l2a2 -END PGP SIGNATURE-
Re: [gentoo-dev] Re: ffmpeg vs libav choice of default
On 09/02/15 10:17, Alexis Ballier wrote: (nb: you can see that this precise one is mostly fixed already.) I hope it is completely fixed already =\ lu
Re: [gentoo-portage-dev] [PATCH 0/2] Implement soname dependencies
On 02/09/2015 08:40 AM, Alexander Berntsen wrote: On 09/02/15 06:41, Zac Medico wrote: 49 files changed, 2766 insertions(+), 149 deletions(-) So... is anyone anywhere near qualified to (N)ACK this? Sebastian Luther it the only other person who is really familiar with portage's dependency resolver. @Sebastian: Would you like to review my patch for soname dependency resolution? https://github.com/zmedico/portage/tree/binpkg-soname-deps -- Thanks, Zac
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in eclass: toolchain.eclass
Dnia 2015-02-09, o godz. 20:05:07
Mike Frysinger (vapier) vap...@gentoo.org napisał(a):
vapier 15/02/09 20:05:07
Modified: toolchain.eclass
Log:
use multislot for all cross-compilers and versions older than gcc-4.6
Revision ChangesPath
1.650eclass/toolchain.eclass
file :
http://sources.gentoo.org/viewvc.cgi/gentoo-x86/eclass/toolchain.eclass?rev=1.650view=markup
plain:
http://sources.gentoo.org/viewvc.cgi/gentoo-x86/eclass/toolchain.eclass?rev=1.650content-type=text/plain
diff :
http://sources.gentoo.org/viewvc.cgi/gentoo-x86/eclass/toolchain.eclass?r1=1.649r2=1.650
Index: toolchain.eclass
===
RCS file: /var/cvsroot/gentoo-x86/eclass/toolchain.eclass,v
retrieving revision 1.649
retrieving revision 1.650
diff -u -r1.649 -r1.650
--- toolchain.eclass 5 Feb 2015 23:28:17 - 1.649
+++ toolchain.eclass 9 Feb 2015 20:05:07 - 1.650
@@ -1,6 +1,6 @@
# Copyright 1999-2015 Gentoo Foundation
# Distributed under the terms of the GNU General Public License v2
-# $Header: /var/cvsroot/gentoo-x86/eclass/toolchain.eclass,v 1.649
2015/02/05 23:28:17 blueness Exp $
+# $Header: /var/cvsroot/gentoo-x86/eclass/toolchain.eclass,v 1.650
2015/02/09 20:05:07 vapier Exp $
# Maintainer: Toolchain Ninjas toolch...@gentoo.org
@@ -155,7 +155,7 @@
IUSE+= ${IUSE_DEF[*]}
# Support upgrade paths here or people get pissed
-if use multislot ; then
+if ! tc_version_is_at_least 4.6 || is_crosscompile || use multislot ; then
SLOT=${GCC_CONFIG_VER}
else
SLOT=${GCC_BRANCH_VER}
This breaks a few packages [1,2]. Please fix the breakage and run
repoman next time.
[1]:https://travis-ci.org/gentoo/gentoo-portage-rsync-mirror/jobs/50118225#L4304
[2]:https://travis-ci.org/gentoo/gentoo-portage-rsync-mirror/jobs/50118228#L923
--
Best regards,
Michał Górny
pgpAcF_UXR8db.pgp
Description: OpenPGP digital signature
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in eclass: toolchain.eclass
On Mon, Feb 9, 2015 at 6:40 AM, Michał Górny mgo...@gentoo.org wrote:
Dnia 2015-02-09, o godz. 20:05:07
Mike Frysinger (vapier) vap...@gentoo.org napisał(a):
vapier 15/02/09 20:05:07
Modified: toolchain.eclass
Log:
use multislot for all cross-compilers and versions older than gcc-4.6
Revision ChangesPath
1.650eclass/toolchain.eclass
file :
http://sources.gentoo.org/viewvc.cgi/gentoo-x86/eclass/toolchain.eclass?rev=1.650view=markup
plain:
http://sources.gentoo.org/viewvc.cgi/gentoo-x86/eclass/toolchain.eclass?rev=1.650content-type=text/plain
diff :
http://sources.gentoo.org/viewvc.cgi/gentoo-x86/eclass/toolchain.eclass?r1=1.649r2=1.650
Index: toolchain.eclass
===
RCS file: /var/cvsroot/gentoo-x86/eclass/toolchain.eclass,v
retrieving revision 1.649
retrieving revision 1.650
diff -u -r1.649 -r1.650
--- toolchain.eclass 5 Feb 2015 23:28:17 - 1.649
+++ toolchain.eclass 9 Feb 2015 20:05:07 - 1.650
@@ -1,6 +1,6 @@
# Copyright 1999-2015 Gentoo Foundation
# Distributed under the terms of the GNU General Public License v2
-# $Header: /var/cvsroot/gentoo-x86/eclass/toolchain.eclass,v 1.649
2015/02/05 23:28:17 blueness Exp $
+# $Header: /var/cvsroot/gentoo-x86/eclass/toolchain.eclass,v 1.650
2015/02/09 20:05:07 vapier Exp $
# Maintainer: Toolchain Ninjas toolch...@gentoo.org
@@ -155,7 +155,7 @@
IUSE+= ${IUSE_DEF[*]}
# Support upgrade paths here or people get pissed
-if use multislot ; then
+if ! tc_version_is_at_least 4.6 || is_crosscompile || use multislot ; then
SLOT=${GCC_CONFIG_VER}
else
SLOT=${GCC_BRANCH_VER}
This breaks a few packages [1,2]. Please fix the breakage and run
repoman next time.
[1]:https://travis-ci.org/gentoo/gentoo-portage-rsync-mirror/jobs/50118225#L4304
[2]:https://travis-ci.org/gentoo/gentoo-portage-rsync-mirror/jobs/50118228#L923
Why is this change being made in the first place? I see this slotting
change gcc and previously in binutils, with no explanation given.
