#archlinux-ports | Logs for 2017-07-09

Back
[00:31:08] -!- isacdaavid has quit [Ping timeout: 260 seconds]
[00:47:20] -!- Faalagorn has joined #archlinux-ports
[00:59:44] -!- Faalagorn has quit [Remote host closed the connection]
[01:22:09] -!- isacdaavid has joined #archlinux-ports
[02:01:56] -!- fhdrin has quit [Quit: Leaving]
[04:20:27] eschwartzd is now known as eschwartz
[05:02:03] -!- fhdrin has joined #archlinux-ports
[06:55:29] -!- isacdaavid has quit [Read error: Connection reset by peer]
[07:30:35] -!- fhdrin has quit [Quit: Leaving]
[09:24:11] -!- eschwartz has quit [Remote host closed the connection]
[09:26:07] -!- eschwartz has joined #archlinux-ports
[10:56:56] -!- City-busz has quit [Remote host closed the connection]
[10:57:03] -!- City-busz has joined #archlinux-ports
[10:57:03] -!- City-busz has quit [Changing host]
[10:57:03] -!- City-busz has joined #archlinux-ports
[10:57:16] -!- phrik has quit [Read error: Connection reset by peer]
[10:57:16] -!- City-busz has quit [Remote host closed the connection]
[11:00:07] -!- City-busz has joined #archlinux-ports
[11:00:08] -!- City-busz has quit [Changing host]
[11:00:08] -!- City-busz has joined #archlinux-ports
[11:00:21] -!- phrik has joined #archlinux-ports
[18:38:05] -!- eschwartz has quit [Remote host closed the connection]
[18:39:01] -!- eschwartz has joined #archlinux-ports
[20:06:05] -!- deep42thought has joined #archlinux-ports
[20:36:44] <deep42thought> woffs: thanks for the info about botan
[20:39:08] <deep42thought> rewbycraft: I had the idea, too, to skip the "-nocheck" ending, however I wanted to check first if the build scripts really work if two different packages have the same name
[20:52:25] <deep42thought> btw: What do you think about following "trunk" instead of the stable repos?
[20:53:25] <deep42thought> Some package compilation errors only get fixed in trunk and it takes some time until the changes propagate to the stable repositories
[21:37:55] <City-busz> deep42thought: if a package in a stable repository does not build, then it should be reported on bug tracker. It's a valid report even if the trunk version builds fine.
[21:38:19] <City-busz> deep42thought: why do you want to name a package with '-nocheck'?
[21:38:37] <deep42thought> because I have a cyclic dependency otherwise:
[21:39:02] <deep42thought> perl-pango and gtk2-perl depend on each other and thus the exact perl version they were built with
[21:39:32] <deep42thought> however, one of the dependencies is only a checkdep, so I want to make this package w/o check, name it -nocheck
[21:39:41] <deep42thought> and use it to "bootstrap" the perl version
[21:41:02] <City-busz> I think it would be easier to simply skip the check function in pango-perl, it's not so important.
[21:41:17] <deep42thought> hmm
[21:41:41] <deep42thought> I had no idea about the importance of those checks
[21:44:35] <City-busz> When it causes troubles, then I would say simply ignore them.
[21:44:48] <rewbycraft> That... is also an option
[21:46:16] <jelle> or bug upstream ;-)
[21:46:27] <deep42thought> City-busz: regarding the bug-thing: I don't think, that each modification necessary to keep a rebuild possible, needs a pkgrel-bump and thus a rebuild
[21:47:17] <deep42thought> jelle: What exactly would the upstream bug be?
[21:48:09] <jelle> deep42thought: well pango-perl?
[21:49:20] <jelle> deep42thought: test passes for me, on 64 bit
[21:49:29] <deep42thought> the test does not fail
[21:49:38] <deep42thought> the problem is the dependencies of the packages
[21:49:44] <jelle> hmmmmm
[21:49:52] <jelle> I missed something then :P
[21:50:33] <City-busz> deep42thought: regarding the bug-thing: I don't understand what you said :)
[21:50:40] <deep42thought> sry
[21:51:25] <deep42thought> City-busz: if some other change breaks a package build but not the package, then this does not call for a pkgrel-increase
[21:52:42] <deep42thought> jelle: pango-perl and gtk2-perl depend on each other and on the exact perl version they were built against - so if I want to increase the perl version, I can't build either, because the other still depends on the old perl version
[21:52:54] <jelle> aha
[21:53:31] -!- Faalagorn has joined #archlinux-ports
[21:55:41] <City-busz> I would open a bug report for pango-perl to disable the check function. We should avoid these circular build dependencies in Arch Linux.
[21:58:00] <deep42thought> I always thought of those circular (check)dependencies as a sort of regression-test
[21:58:26] <deep42thought> x depends on y, and y checkdepends on x, because each version bump of y tests if x is still fine
[21:58:49] <City-busz> Yes, these are mostly regression test. But this is really bad when it's circular.
[22:00:58] <City-busz> deep42thought: I agree that a build fix does not necessarily need a pkgrel increase. But the build fix should be applied for the stable packages too even if rebuild does not happened. We always have to ensure that stable packages are buildable in Arch Linux.
[22:02:06] <City-busz> So if a package maintainer fix the trunk, and leave the repos broken, it's an incomplete fix.
[22:03:58] <deep42thought> Is this some sort of an official statement?
[22:04:09] <deep42thought> Because I got a lot responses like "trunk works"
[22:06:37] <City-busz> If it's not an official statement, then I'll write to the arch-dev-public mailing list. :)
[22:06:46] <deep42thought> :-)
[22:06:49] <City-busz> Could you show me some of these responses?
[22:07:37] <deep42thought> yeah, w8
[22:08:27] <deep42thought> FS54735 got fixed in trunk, but not in the repo
[22:10:05] <deep42thought> my personal favourite: FS#54413
[22:11:29] <deep42thought> FS#54381
[22:11:41] <jelle> !bug 54381
[22:11:42] <phrik> http://bugs.archlinux.org
[22:12:17] <deep42thought> jelle: thanks for the hint
[22:12:43] <deep42thought> !bug 54735
[22:12:43] <phrik> http://bugs.archlinux.org
[22:12:51] <deep42thought> !bug 54413
[22:12:51] <phrik> http://bugs.archlinux.org
[22:26:36] <deep42thought> ok, maybe "3" is not "a lot" in an absolute sense, but it feels like many :-/
[22:34:00] -!- fhdrin has joined #archlinux-ports
[22:52:48] <City-busz> I fixed 54381 and 54735
[22:53:02] -!- eschwartz has quit [Remote host closed the connection]
[22:53:03] <deep42thought> City-busz: I noticed, thanks!
[22:54:22] <City-busz> If more happens (fixed in trunk but not in repos), please notify me, and I'll ensure that the things gets fixed (at least for the [community] repository).
[22:54:40] <deep42thought> City-busz: thanks for your help :-)
[22:55:23] -!- eschwartz has joined #archlinux-ports
[23:10:55] -!- eschwartz has quit [Remote host closed the connection]
[23:11:32] -!- eschwartz has joined #archlinux-ports
[23:19:12] -!- eschwartz has quit [Remote host closed the connection]
[23:23:05] -!- Polichronucci has quit [Ping timeout: 240 seconds]
[23:24:13] -!- Polichronucci has joined #archlinux-ports
[23:26:50] -!- eschwartz has joined #archlinux-ports
[23:51:02] -!- deep42thought has quit [Quit: Leaving.]