From FedoraProject

Jump to: navigation, search
[18:05] *** Kevin_Kofler nastavil téma kanálu na "KDE SIG Meeting - - automoc: What's the right way of packaging it? Upstream is confusing about that".
[18:05] --> JSchmitt has joined this channel (
[18:05] <Kevin_Kofler> First topic: automoc: What's the right way of packaging it? Upstream is confusing about that
[18:05] <Kevin_Kofler> lvillani put this on the agenda.
[18:05] --> kital has joined this channel (n=Joerg_Si@fedora/kital).
[18:05] <lvillani> yep
[18:06] <lvillani> I sent an email to upstream developer regarding
[18:06] <lvillani> if it's needed or not
[18:06] <-- drago01 has left this server (Read error: 104 (Connection reset by peer)).
[18:07] --> drago01 has joined this channel (
[18:07] <lvillani> but it seems (after reading Automoc4Config.cmake) that is not really needed
[18:07] <lvillani> or at least, we don't need it to be packaged in %{_libdir}
[18:07] <than> lvillani: don't include it if it's not needed
[18:07] --> rnorwood has joined this channel (n=rnorwood@nat/redhat/x-abac00903191ff15).
[18:08] <Kevin_Kofler> But is it really not needed?
[18:08] <lvillani> but upstream says that it's a must-have and that it must be in it's installed location (%{_libdir})
[18:08] <lvillani> I forwarded his email to rdieter, he agrees that this is confusing
[18:09] <lvillani> anyone willing to do a test build with this automoc package?
[18:10] <Kevin_Kofler> Just build it in dist-f10 and I'll try to build akonadi against it.
[18:10] <Kevin_Kofler> Is the multilib issue fixed yet?
[18:11] <than> lvillani: we just remove the file in rawhide. If it causess build problem we always can add it back
[18:11] <lvillani> it should be
[18:12] <-- kital has left this server (Remote closed the connection).
[18:12] <Kevin_Kofler> OK, then as than said, go ahead and build it in dist-f10. As I said, I'll build akonadi against it, we'll see if it builds.
[18:12] <lvillani> than, sounds good
[18:12] <Kevin_Kofler> Anything else on this or can we move on?
[18:13] <lvillani> fine, I think we can move on
[18:13] *** Kevin_Kofler nastavil téma kanálu na "KDE SIG Meeting - - KDE 4.0.5 update - status update".
[18:13] <Kevin_Kofler> Next topic: KDE 4.0.5 update - status update
[18:13] <Kevin_Kofler> ltinkl issues a kdelibs build with some more fixes, I'm waiting for the build to complete to update the update.
[18:13] <Kevin_Kofler> Then the plan is to move it to stable, unless there are any objections?
[18:14] <Kevin_Kofler> Just one thing: the crystalsvg changes: should I add a versioned kdelibs3 Requires or a versioned crystalsvg-icon-them Conflicts to prevent updating only part of these and running into file conflicts or is it not worth it?
[18:15] <than> Kevin_Kofler: it's not worth it
[18:15] <Kevin_Kofler> OK.
[18:15] <Kevin_Kofler> Then I think that's all about 4.0.5, move on?
[18:15] *** Kevin_Kofler nastavil téma kanálu na "KDE SIG Meeting - -".
[18:16] <Kevin_Kofler> OK, now we're in the recent bugs discussion.
[18:16] <Kevin_Kofler> First one:
[18:16] <bugbot_> Bug 446093: low, low, ---, Ngo Than, ASSIGNED , split kdeedu RFE
[18:16] <than> it's bad idea to seperate package for each app.
[18:16] <than> -math should be fine with me
[18:16] <Kevin_Kofler> But then how do we accomodate the SIGs which want to have only parts of kdeedu?
[18:17] <Kevin_Kofler> But what is -math? For example, is Step Math?
[18:17] <Kevin_Kofler> It's a Physics tool, but of course there's lots of (applied) Mathematics in it.
[18:18] <than> calgebra, kalgebra, kbruch, kmplot, kpercentage
[18:18] <Kevin_Kofler> Maybe we should rediscuss this next week when rdieter is back from vacation?
[18:18] <than> it could be packages in -math
[18:18] <Kevin_Kofler> I'm OK with a -math subpackage, though I anticipate there will be requests for some more subpackages for other topics.
[18:19] <SMParrish> I think you can justify going in either direction.  Personally I think -math is the solution.  Can address other SIG requests as we get them
[18:19] <than> is it really worth splitting it because there's no space on live CD?
[18:20] <than> how about removing other packages ?
[18:22] <Kevin_Kofler> I'd say we go with kdeedu-math, but we should discuss this with rdieter when he comes back before implementing it.
[18:22] <MathStuf> what else is taking up the space on the CD that kdeedu is looked at for trimming "fat"?
[18:23] <Kevin_Kofler> There are tons of Mathematics-related tools in Fedora.
[18:23] <than> it's fine with me with kdeedu-math, but no go with seperate package for each app!
[18:23] <MathStuf> ah, kdeedu is 79MB
[18:23] <Kevin_Kofler> *maxima, octave, several plotting tools, just to mention some.
[18:24] <Kevin_Kofler> And soon there might be even a huge mastodont of a program called "SAGE", but that's a PITA to package.
[18:24] <Kevin_Kofler> Mainly because it tries to be its own distribution.
[18:24] <Kevin_Kofler> But that's not really a topic for KDE SIG. ;-)
[18:25] <Kevin_Kofler> In any case, I can definitely see how a Mathematics spin is heavily size-constrained.
[18:28] <than> ok, we should move on
[18:29] <Kevin_Kofler> Right. I'll ask rdieter when he comes back whether he's OK with this, but right now the conclusion is to go for kdeedu-math.
[18:29] *** Kevin_Kofler nastavil téma kanálu na "KDE SIG Meeting - -".
[18:29] <Kevin_Kofler>
[18:29] <bugbot_> Bug 447852: low, low, ---, Ngo Than, CLOSED UPSTREAM, system:/ kio broken
[18:29] --> rnorwood_ has joined this channel (n=rnorwood@nat/redhat/x-60fd88bb99c1a48e).
[18:29] <-- rnorwood has left this server (Read error: 110 (Connection timed out)).
[18:29] <Kevin_Kofler> MathStuf found out that the offending dialog actually comes from KDE 3.
[18:29] <Kevin_Kofler> That's why it tries to use ioslaves which don't exist anymore in KDE 4.
[18:30] <Kevin_Kofler> We have to disable that KDE 3 dialog.
[18:30] <than> +1
[18:31] <than> Matstuff: is there a testcase how to reproduce the problem?
[18:32] <Kevin_Kofler> I guess installing kdebase3 is part of it?
[18:32] <Kevin_Kofler> AFAIK that code is in kdebase3.
[18:33] <MathStuf> yeah, just put in a removable media
[18:33] <MathStuf> the KDE3 dialog and the plasmoid show up
[18:34] <than> ah i can reproduce it
[18:35] <than> the KDE3 dialog should be disable in thi scase
[18:35] <ltinkl> is the system ioslave part of kdebase3? then just remove it from there
[18:35] <than> ltinkl: yes
[18:36] <ltinkl> I don't think it's really needed outside of old kdebase, KDE 4 doesn't use it
[18:36] <than> ltinkl: could you please take care of it?
[18:36] <ltinkl> sure
[18:36] <than> ltinkl: thanks
[18:36] * pusling takes notes
[18:36] <Kevin_Kofler> The system:/ ioslave is harmless though.
[18:36] <Kevin_Kofler> What really needs to be disabled is the offending dialog.
[18:37] <ltinkl> hmm, right, that would be better
[18:37] <ltinkl> I'll try to come up with something that just disables the dialog
[18:37] <ltinkl> it's more probably kded_mediamanager or sth like that 
[18:37] [Chyba] checks: neznámý příkaz.
[18:38] *** stickster_afk je nyní znám jako stickster.
[18:39] <ltinkl> oh, and the previously mentioned bug is related to that
[18:39] <bugbot_> Bug 163717: medium, medium, ---, Petr Machata, CLOSED CURRENTRELEASE, Need ltrace 0.3.36 to debug x32 binaries on x64
[18:39] <than> ltinkl:  yes, it's in kded_mediamanager
[18:39] <ltinkl> stupid bugbot
[18:39] <Kevin_Kofler> OK. ltinkl: please take care of this.
[18:40] <Kevin_Kofler> And let's move on.
[18:40] *** Kevin_Kofler nastavil téma kanálu na "KDE SIG Meeting - -".
[18:40] <Kevin_Kofler>
[18:40] <bugbot_> Bug 439587: medium, low, ---, Kevin Kofler, ASSIGNED , Unable to move Icons in panel
[18:40] <ltinkl> the KDE bug crashed in 
[18:40] <ltinkl> #11 0x00002aaab063bd58 in libhal_device_get_property_QString () from /usr/lib64/kde3/
[18:40] <ltinkl> same stuff probably
[18:40] <Kevin_Kofler> Right.
[18:41] <Kevin_Kofler> #439587 (unable to move icons in panel) has been reopened because the patch doesn't apply as is for KDE 4.1.
[18:41] <than> the feature "moving icons on panel" is not included in beta1?
[18:41] <Kevin_Kofler> We had hoped that it would get fixed upstream, but unfortunately they don't want to implement this before 4.2 because they consider it a "new feature".
[18:41] <than> i thougt it's iin beta1
[18:42] <Kevin_Kofler> No, users confirmed it isn't.
[18:42] <-- rnorwood_ has left this server (Remote closed the connection).
[18:42] <than> it seems we have to adapt the patch for 4.1
[18:43] <Kevin_Kofler> Right. Now the question is who does it.
[18:43] <pusling> unless we screwed up in debian packaging, I can also confirm - whats in panel is impossible to move from where you drag+dropped it
[18:43] <Kevin_Kofler> I could have some fun with it, but I don't have Rawhide installed anywhere (I have F8 here and F9 on my new laptop).
[18:44] <Kevin_Kofler> It's probably easier to work on for somebody who actually has Rawhide or some other KDE 4.1 installed somewhere.
[18:44] --> rnorwood has joined this channel (n=rnorwood@nat/redhat/x-18ccf6fd7663d216).
[18:45] <Kevin_Kofler> Patch against 4.0.3 here:
[18:45] <Kevin_Kofler> But that uses the KDE 4.0 libplasma API.
[18:45] <Kevin_Kofler> So it won't work as is in 4.1. (It doesn't even apply, because the relevant files have also changed.)
[18:45] --> huff has joined this channel (n=dhuff@nat/redhat/x-59858684a982d493).
[18:46] <Kevin_Kofler> (The patch comes from Stefan Binner from OpenSUSE.)
[18:47] * spstarr_work is sitting out today's meeting busy @ work
[18:48] <Kevin_Kofler> So, do we have any volunteer to work on porting that patch?
[18:50] <than> it seems noone has interest to work on this.
[18:51] <Kevin_Kofler> We can't really count on the OpenSUSE folks to do it for us, because they'll ship 11.0 with KDE 4.0.
[18:51] <ltinkl> I can at least try having trunk installed locally from sources
[18:51] <ltinkl> no promise though, the API is completely different
[18:52] <than> yes. there's a lot of change there
[18:52] <than> i will take a look at this
[18:52] <ltinkl> some stuff got better in trunk (wrt plasma and desktop) but I still considers this very unfinished
[18:53] <ltinkl> -s
[18:53] <Kevin_Kofler> Yeah, a lot of things got postponed to 4.2. :-(
[18:53] <ltinkl> cf my systray adventures in
[18:53] <bugbot_> Bug 155381: medium, medium, ---, Dave Jones, CLOSED RAWHIDE, drivers/block/cfq-iosched.c:1070: spin_is_locked on uninitialized spinlock d9d133b8.
[18:53] <Kevin_Kofler> IMHO they're way too strict with their freezes.
[18:53] <Kevin_Kofler> Plasma should have been exempted for 4.1 like it was for 4.0.
[18:54] <Kevin_Kofler> And some other stuff should get freeze exemptions too.
[18:56] <ltinkl> k, anything else on the agenda?
[18:56] <Kevin_Kofler> No.
[18:56] *** Kevin_Kofler nastavil téma kanálu na "KDE SIG Meeting - - Open discussion".
[18:57] <Kevin_Kofler> We have 3 minutes left for anything else.
[18:57] <ltinkl> since svahl is not here, I'll post the minutes if noone objects
[18:57] <-- danishka has left this server ("Leaving").
[18:57] <Kevin_Kofler> ltinkl: OK
[18:58] <Kevin_Kofler> There's also the minutes of the previous 2 weeks, I wanted to take care of those and didn't get around to it. :-(
[18:58] <lvillani> it seems that Automoc4Config.cmake does not require but Akonadi still tries to include it, no objections about that but...
[18:59] <lvillani> the point is that FindAutomoc4.cmake looks for Automoc4Config.cmake in %{_libdir}/automoc4 which, imho, is wrong
[18:59] <lvillani> suggestions?
[19:00] <lvillani> i think that %{_datadir}/cmake/Modules is a better place for Automoc4Config.cmake, but this will probably involves patching FindAutomoc4.cmake for packages shipping it
[19:00] <lvillani> (ie: akonadi)
[19:00] <lvillani> (I just did a couple of local test builds, so I'm not 100% sure about that)
[19:02] --> iWolf_ has joined this channel (
[19:02] <Kevin_Kofler> Well, our time is up.
[19:02] <Kevin_Kofler> lvillani: We can discuss this on #fedora-kde later.
[19:03] <lvillani> ok

There are no pages that link to this file.