15:11:48 <rdieter> #startmeeting kde-sig
15:11:48 <zodbot> Meeting started Tue Oct  9 15:11:48 2012 UTC.  The chair is rdieter. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:11:48 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:11:55 <rdieter> #meetingname kde-sig
15:11:55 <zodbot> The meeting name has been set to 'kde-sig'
15:12:00 <rdieter> #topic roll call
15:12:04 <Kevin_Kofler> Present.
15:12:11 <dvratil> me!
15:12:16 <jgrulich> present
15:12:45 <rdieter> than, ltinkl, rnovacek : ping
15:12:51 * ltinkl is here
15:12:59 * rnovacek is here
15:13:02 <than> present
15:14:45 <rdieter> #chair Kevin_Kofler dvratil jgrulich ltinkl rnovacek than
15:14:45 <zodbot> Current chairs: Kevin_Kofler dvratil jgrulich ltinkl rdieter rnovacek than
15:14:54 <rdieter> #info rdieter Kevin_Kofler dvratil jgrulich ltinkl rnovacek than present
15:14:58 <rdieter> #topic agenda
15:15:12 <rdieter> moving on to agenda, nothing published, so what should we discuss today?
15:15:52 <than> rdieter:  any blocker bugs for f18
15:16:14 * rdieter checks https://bugzilla.redhat.com/show_bug.cgi?id=F18Blocker-kde
15:16:28 <rdieter> .bug 829881
15:16:32 <zodbot> rdieter: Bug 829881 Use polkit .rules files instead of .pkla files - https://bugzilla.redhat.com/show_bug.cgi?id=829881
15:17:06 <Kevin_Kofler> Yeah well, that one isn't really a blocker, I doubt QA would let it pass as a real blocker, NTH at best.
15:17:17 <rdieter> true
15:17:22 * rdieter checks https://bugzilla.redhat.com/show_bug.cgi?id=F18Target-kde
15:17:26 <Kevin_Kofler> The .rules file will just have no effect, which means setting the clock requires a password by default.
15:17:28 * tdfischer is here, late
15:17:36 <Kevin_Kofler> Uh, I mean the old .pkla file.
15:17:48 <rdieter> a few on target too
15:17:51 <rdieter> .bug 688956
15:17:54 <zodbot> rdieter: Bug 688956 kcm_touchpad: "Enable Tapping" doesn't work - https://bugzilla.redhat.com/show_bug.cgi?id=688956
15:17:58 <Kevin_Kofler> If I get some time, I'll see if I can come up with a .rules file for that.
15:18:06 <rdieter> qwt6...
15:18:23 <rdieter> blocking on someone doing qwt5 review
15:18:23 <Kevin_Kofler> Ah, that kcm_touchpad one is fun, default tapping actions being empty is lame.
15:18:34 <Kevin_Kofler> You shouldn't have to enable tapping twice to enable it.
15:18:41 <rdieter> yeah
15:18:50 <Kevin_Kofler> That's also something I had wanted to look into, sigh……
15:18:55 <rdieter> .bug 845591
15:18:58 <zodbot> rdieter: Bug 845591 Review Request: qwt5 - Qt Widgets for Technical Applications - https://bugzilla.redhat.com/show_bug.cgi?id=845591
15:19:12 <rdieter> so... any help on these would be appreciated
15:19:33 <rdieter> (I just realized, we're still in agenda items, sorry)
15:19:46 <rdieter> anything else to discuss today, before moving on?
15:20:02 <tdfischer> phonone-gstreamer is being ported to gst1.0
15:20:13 <Kevin_Kofler> tdfischer: Yay!
15:20:14 <tdfischer> that's my only contribution
15:20:18 <rdieter> tdfischer: thanks!
15:20:25 <tdfischer> np
15:20:26 <Kevin_Kofler> tdfischer: But it's a great one. :-)
15:20:46 <rdieter> moving on officially then...
15:20:52 <rdieter> #topic F18 blocker/target bugs
15:21:02 <rdieter> https://bugzilla.redhat.com/show_bug.cgi?id=F18Blocker-kde
15:21:08 <rdieter> https://bugzilla.redhat.com/show_bug.cgi?id=F18Target-kde
15:21:50 <Kevin_Kofler> So for both #829881 and #688956, I have a pretty clear-cut idea of what needs to be done to fix those, but I just didn't get around to looking into them.
15:22:10 <Kevin_Kofler> I'll have a look this week.
15:22:38 <rdieter> Kevin_Kofler: could you maybe document your ideas in those bugs... given that, maybe someone else could jump in and help finish it up.
15:22:42 <Kevin_Kofler> (That's the PolicyKit rules and the kcm_touchpad stupidity.)
15:23:02 <rdieter> the pk thing should be fairly straight-forward tho
15:23:12 <rdieter> just adapting to the newer syntax
15:23:14 <Kevin_Kofler> Well, for the PolicyKit rules, look at some existing .rules files and then make one which matches what the .pkla file did.
15:23:33 <Kevin_Kofler> Yeah, just look at examples to see how the new syntax should look like.
15:24:20 <Kevin_Kofler> For kcm_touchpad, I think that we need something like "if tapping is disabled, uncheck 'Enable tapping' and set tapping actions to some default actions rather than reading the settings which are all disabled".
15:24:56 <Kevin_Kofler> I guess the time where it loads the driver settings and sets the UI based on that is probably the best time to set those.
15:25:34 <Kevin_Kofler> Or alternatively, one could set some default actions when the user checks the box, but I think that's less nice, checking the box should only affect the grayed-out state of the tapping actions, not their values.
15:26:26 <rdieter> that sounds fun.
15:26:51 <Kevin_Kofler> Oh, there's an issue which SHOULD be a blocker, but I don't think it's filed yet: the Anaconda rendering issues with oxygen-gtk!
15:27:14 <rdieter> that's news to me, what's up with that?
15:27:15 <Kevin_Kofler> DaemonFC showed a screenshot where almost the whole UI was empty and as a result one couldn't make the required input to proceed to the next step.
15:27:22 <Kevin_Kofler> So IMHO it's a blocker!
15:27:41 <rdieter> guess time to try a newer snapshot, that was definitely not the case with f18-alpha
15:27:43 <Kevin_Kofler> There was some discussion about graphical glitches on the chan before, but it seems to be really bad.
15:27:54 <Kevin_Kofler> rdieter: Yes, it's a post-alpha regression (probably in GTK+).
15:27:57 <dvratil> Kevin_Kofler, could #863276 be related?
15:29:05 <Kevin_Kofler> dvratil: I'm having a look…
15:29:14 <dvratil> you can't see the hub (just a gray area), but it reacts on clicks
15:29:31 <rdieter> fyi, f18beta-tc2-kde x86_64 image is currently ~816mb
15:29:39 <Kevin_Kofler> No, it's not the same.
15:29:51 <Kevin_Kofler> The whole VM is gray in the screenshot in #863276.
15:30:19 <Kevin_Kofler> In DaemonFC's screenshot, parts of the liveinst dialog are visible.
15:30:26 <Kevin_Kofler> And it only happens with oxygen-gtk.
15:30:44 <Kevin_Kofler> #863276 is standalone Anaconda, which doesn't use oxygen-gtk.
15:31:34 <dvratil> ok...I was able to reliably reproduce it in KDE spin, but not in Gnome
15:31:39 <Kevin_Kofler> Let me see if I can find his screenshot (he posted it to the chan last night, it must be in tigcc_bot's logs).
15:32:26 <Kevin_Kofler> dvratil: I think what you're seeing is DaemonFC's bug, not the OP of #863276's bug.
15:33:23 <dvratil> hmm, that could explain why I can't see it in Gnome :))
15:33:47 <Kevin_Kofler> https://imageshack.us/a/img43/760/snapshot1dv.png
15:33:53 <Kevin_Kofler> This is DaemonFC's screenshot.
15:33:57 <dvratil> yeah, that exactly what I can get
15:34:02 <dvratil> *-can
15:34:10 <Kevin_Kofler> Compare with https://bugzilla.redhat.com/attachment.cgi?id=621834 from #863276.
15:34:10 <rdieter> wow,nasty
15:34:24 <dvratil> the content is still clickable though, you just can't see the buttons
15:34:53 <Kevin_Kofler> We need a separate bug filed (against GTK+ maybe?) and proposed as F18Beta blocker.
15:35:44 <dvratil> there are some runtime Gtk warnings, so maybe oxygen-gtk can't cope with it
15:37:04 <rdieter> dvratil: can you check which version of gtk is in use?
15:37:08 <Kevin_Kofler> In any case, please file a separate bug and propose it as a blocker.
15:37:29 * rdieter will test too after meeting, as time allows
15:37:44 <dvratil> ok, I'll deal with it
15:38:04 <rdieter> .bug 856367
15:38:07 <zodbot> rdieter: Bug 856367 Impossible to rename a pdf file after choosing the directory - https://bugzilla.redhat.com/show_bug.cgi?id=856367
15:38:14 <rdieter> i was almost going to suspect ^^, but the symptoms are very different
15:38:48 <rdieter> and I guess that's largely limited to gtk2 these days (anaconda is gtk3 now or not?)
15:39:20 <dvratil> gtk3-3.6.0-1.fc18
15:39:43 <dvratil> whoa, we ship gtk2 too?
15:40:21 <rdieter> of course, lot's of things still use it
15:40:35 <dvratil> even on the live cd?
15:41:06 <rdieter> pretty sure yes, i'll double check when I test the f18-beta-tc2 image here in a bit
15:41:15 <rdieter> and take note what pulls it in
15:41:39 <rdieter> may be worth trying to omit it if it's not too much pain
15:41:45 <Kevin_Kofler> Anaconda and all the system-config-* used to be PyGTK2.
15:43:10 <Kevin_Kofler> These days, I don't know. But I think at least firstboot is still gtk2.
15:43:37 <dvratil> ok, I uninstalled oxygen-gtk3, now I can see the hub again
15:44:26 <dvratil> so either it's oxygen-gtk3 issue, or it's caused by Anaconda misusing Gtk (thus runtime warnings) which confuses oxygen-gtk
15:44:35 <Kevin_Kofler> (Last I checked, the firstboot developers said they can't port to gtk3 because they want to keep supporting all the vendor-provided custom firstboot screens written for older RHEL versions. :-/ )
15:45:18 <Kevin_Kofler> dvratil: Or it's a gtk3 bug which breaks oxygen-gtk3.
15:45:55 <dvratil> indeed, that's an option too
15:46:06 <Kevin_Kofler> (and FWIW, oxygen-gtk3 definitely does misuse GTK+ to some extent ;-) but normally it's supposed to work)
15:46:48 <rdieter> f18-Alpha had gtk3-3.5.10-1.fc18 oxygen-gtk3-1.1.0-2.fc18
15:47:05 <Kevin_Kofler> Was Anaconda in Alpha already using gtk3?
15:47:14 <rdieter> now we're at (f18 tag): gtk3-3.6.0-1.fc18 oxygen-gtk3-1.1.1-1.fc18
15:47:33 <Kevin_Kofler> We need to try downgrading things to see what changed.
15:48:07 <rdieter> ok, we have a plan on how to test that (after meeting).
15:48:19 <rdieter> #topic open discussion
15:48:26 <rdieter> anything else for today, before we close our meeting?
15:48:46 * dvratil is curious what will we do about the Nepomuk plugins
15:49:14 <rdieter> dvratil: ship what we can (ffmpeg is out of course)
15:49:21 <Kevin_Kofler> Wait for how upstream tars them up, then package them where we can, get the rest into RPM Fusion.
15:49:41 <dvratil> nice
15:50:48 <Kevin_Kofler> I don't think there's much more we can decide before we have the tarballs.
15:51:03 <Kevin_Kofler> And I guess then it'll be straightforward, one specfile per tarball.
15:51:37 <Kevin_Kofler> And we can decide whether and what to Require (obviously we can't Require the ffmpeg stuff, but we're probably expected to drag the rest in by default somehow).
15:51:56 <Kevin_Kofler> But we need to see what the pieces are before we can make any reasonable decision there.
15:52:22 <dvratil> sure, this won't happen before KDE 4.10 anyway
15:55:53 <rdieter> looks like discussion tapered off, let's wrap up.  thanks everyone!
15:55:56 <rdieter> #endmeeting