15:07:46 #startmeeting kde-sig -- http://fedoraproject.org/wiki/SIGs/KDE/Meetings/2011-07-19 15:07:46 Meeting started Tue Jul 19 15:07:46 2011 UTC. The chair is rdieter. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:07:46 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:07:53 #topic roll call 15:07:57 hi, who's present today? 15:08:22 * jreznik is here 15:08:32 * rnovacek here 15:09:05 Kevin_Kofler said he might not get it on time or at all 15:09:52 #chair jreznik rnovacek 15:09:52 Current chairs: jreznik rdieter rnovacek 15:09:53 * than is present 15:09:56 #chair than 15:09:56 Current chairs: jreznik rdieter rnovacek than 15:10:07 #info jreznik rnovacek than rdieter present 15:10:15 #topic agenda 15:10:20 what to discuss today? 15:10:33 brb, afk for a couple min... 15:10:43 4.7rc1 status 15:13:47 i have appointment at 5:30 and, leave the meeting early 15:13:59 looks like we'll have a short meeting then 15:14:09 #topic 4.7rc1 status 15:15:04 see also our tracking bug, https://bugzilla.redhat.com/showdependencytree.cgi?id=kde-4.7&hide_resolved=1 15:15:32 * jreznik gave up building monolithic kdebingings - some parts are already splitted - so it has to be finished... 15:15:44 jreznik: agreed 15:16:00 jreznik: it doesn't make sense to build monolithic kdebingings 15:16:09 I think we have the minimal required pieces of kdebindings done already 15:16:13 it just makes more problem 15:16:35 is there anything else we ship that uses any of the missing kdebindings stuff? 15:17:36 * rdieter also submitted 2 more reviews for newer split tarballs 15:17:49 .bug 722993 15:17:51 rdieter: Bug 722993 Review Request: kate - Advanced Text Editor - https://bugzilla.redhat.com/show_bug.cgi?id=722993 15:18:04 .bug 723064 15:18:09 rdieter: Bug 723064 Review Request: konsole - KDE Terminal Emulator - https://bugzilla.redhat.com/show_bug.cgi?id=723064 15:18:19 * jreznik already took konsole 15:18:36 so kdelibs, kdebase at least no longer needs hacks 15:18:42 but I was finishing herqq review for the changed qtsoap 15:18:43 maybe kdesdk too, need another closer look 15:19:20 kate being packaged separately now, apps using/needing katepart will need to explicitly add some dependencies though 15:20:05 do we already have review request for smokegen, smokekde, smokeqt ? 15:20:24 alternatively, kdelibs can make a subpkg for libktexteditor, and *that* can have a dep on kate-part 15:20:34 than: no 15:21:01 ok, i will do Review Request for those 15:21:21 we have kross-interpreters 15:21:36 https://bugzilla.redhat.com/show_bug.cgi?id=656997 15:21:41 PyKDE4, kross-interpreters are the only ones done so far 15:21:54 than: ping me and I'll do review 15:22:14 jreznik: i will do, thanks :) 15:22:16 wrt kdelibs and kate, which approach seems better? opinions? 15:22:51 1. explictly require apps to add Requires: kate-part, or 2. make kdelibs-ktexteditor subpkg with Requires: kate-part ? 15:23:20 Does it make any sense to have kdelibs without kate-part? 15:23:42 I don't think so, so +1 to 1. 15:24:23 rnovacek: 2 implies apps that already link against libktexteditor will continue to "just work" without modification 15:24:32 1. requires packaging changes 15:25:43 but it's not a big packaging change... 15:26:15 * rdieter likes option 1 better, less work. :) 15:26:23 I think so 15:26:23 err... oops, make that 2 15:26:31 2 is less work, less error-prone 15:26:53 ok, +1 to 2. too :) 15:26:58 1 is not so much work 15:27:14 2 is better because of ktexteditor withou kate-part is nonsense 15:28:10 ok, once kate review is done, I'll implement option 2? 15:29:08 #action than will submit a few more kdebindings (smoke) related reviews 15:29:37 #action rdieter will implement kdelibs-ktexteditor subpkg once kate review is done 15:29:50 any bikeshedding for subpkg name, or is kdelibs-ktexteditor subpkg name ok? 15:30:08 :) 15:30:54 ok 15:31:05 anything else 4.7-related? 15:31:40 I think kdelibs-ktexteditor is OK. 15:31:47 I'm a bit worried because the dep is still circular to some extent. 15:32:11 i.e. a subpackage of kdelibs depending on something that needs kdelibs-devel to build 15:32:29 Kevin_Kofler: good poing, I'll have to think about that a bit 15:32:38 good point even 15:33:32 maybe option 1 is better after all 15:34:09 * rdieter curses again for not having soft dependencies 15:34:52 I'm not sure what the best solution is, both suck to some extent. :-( 15:35:15 Kevin_Kofler: could be a little more evil, and have libktexteditor *not* get pulled in via kdelibs-devel 15:35:34 and apps needing it will have to add another BR 15:36:11 but now we're getting back to having to modify all apps that need it, but that's a *little* safer than the original option 1, imo 15:36:50 I'll think about it some more, and make a proposal later. 15:37:00 #topic open discussion 15:37:04 anything else to discuss today? 15:37:52 rdieter: That wouldn't help. 15:38:05 The Kate stuff would need kdelibs-ktexteditor-devel anyway. 15:38:10 So the dep would still be circular. 15:38:54 ok, but at least the breakage is limited to just kate (and anything that BR's it), instead of breaking everything that BR: kdelibs-devel 15:39:13 TBH, I think the best solution is to just keep building that stuff as part of kdelibs. 15:39:34 kate/libktexteditor is best built together, true 15:39:57 I'll ping upstream and suggest something like that, this situation sucks 15:40:16 I think it was a very bad idea from upstream to move the KatePart out of kdelibs. :-( 15:40:30 no argument from me 15:41:16 more fallout from the arbitrariness of producing split tarballs based on git repos, without forethought on packaging implications 15:42:02 I think the repo organization is wrong to begin with. 15:42:09 Plasma is doing just fine with libplasma in the kdelibs git. 15:42:24 calling it 'organization' is too kind. :) 15:43:54 to be clear, katepart is the only ktexteditor implementation ? 15:44:41 Not sure. It's the only one actually used, in any case. :-) 15:46:03 * rdieter will close meeting if there's nothing else to discuss 15:47:08 thanks everyone! 15:47:09 #endmeeting