[00:04] * yofel wonders when the hell we were talking about making stable daily builds of kde === jackyalcine is now known as zz_jackyalcine === zz_jackyalcine is now known as jackyalcine === jackyalcine is now known as jackyalcine_ === jackyalcine_ is now known as zz_jackyalcine [01:11] Cool. [01:13] !testers | new images 20120213 has arrived [01:13] new images 20120213 has arrived: Help is needed in #kubuntu-devel. Please ping Riddell, yofel, soee, Tm_T, shadeslayer, BluesKaj, James147, smartboyhw, Quintasan, lordievader for information. === zz_jackyalcine is now known as jackyalcine [05:38] ::runtime-bugs:: [1123763] Veromix volume control broken by update. @ https://bugs.launchpad.net/bugs/1123763 (by Daniel Clem) === Adityab_ is now known as Adityab [07:48] QStringList scripts = KGlobal::dirs()->findAllResources("data", appName + "/init/*.js"); [07:48] scripts.sort(); [07:48] Riddell: that's why the plasma init script is not run [07:50] didn't we fix that already [07:50] s/we/I/ [07:50] shadeslayer meant: "didn't I fix that already" [07:50] not upstream apparently? [07:50] lolwot [07:50] untrue [07:50] dunno [07:50] looking at the sourcery right now [07:50] I upstreamed it man [07:50] I upstream all the things [07:51] shadeslayer: not fixed in the 4.8 branch [07:51] bah [07:52] or [07:52] hm [07:52] patch it patch it patch it [07:52] 4.8 was EOL [07:52] right? [07:52] there also was no patch in 4.8.2 [07:53] wtf [07:53] @_@ [07:54] * shadeslayer sighs [07:55] "I asked that to the kubuntu guys so i could use it myself, but it seems [07:55] packaging is too hard in ubuntu land and they never got back to me." [07:55] when was this? [07:55] from kde-packagers [07:56] someone should tell Albert about Project Neon :P [07:56] ehm [07:57] someone should tell Rohan about quoting of private lists [07:58] eh .. durrr [07:58] don't see no such mail [07:58] so [07:58] it was [07:58] QStringList scripts = KGlobal::dirs()->findAllResources("data", appName + "/init/*.js"); [07:58] QStringList scriptPaths; [07:59] then it got a sort [07:59] right [07:59] both shit because master looks completely different [07:59] we could actually get rid of all the weird logic [07:59] well, not completely different [07:59] just vastly different [08:00] ah [08:00] we should patch master into .5 [08:00] dude, just look at my commit or sth [08:00] much nicer code [08:00] heh [08:00] this is in workspace right? [08:00] it uses findDirs rather than findAllResources and then builds the script paths manually [08:01] http://quickgit.kde.org/?p=kde-workspace.git&a=blob&h=620915cc319bae50830a3b521f0f981a72c0d772&hb=e656398cefa2167aee1913244f633579676db985&f=libs%2Fplasmagenericshell%2Fscripting%2Fscriptengine.cpp [08:01] last function [08:01] yus, that's me [08:01] omg apachelogger said my code is nice [08:01] my life is complete [08:02] :P [08:02] well [08:02] IIRC I came up with the algorithm :P [08:02] did not realize it was such a well thought through one [08:02] :) [08:03] eh, no one here wanted to know that :P [08:03] shadeslayer: so you did not patch .8 for which you should get a beating really [08:03] and that .sort there is what breaks it obviously [08:03] no, I was under the assumption that .8 was EOL so it was pointless [08:03] what with /usr/share/kde4 < /usr/share/kubuntu-* [08:03] but turns out we ship EOL software for precise [08:04] version+2 needs to be released for a release to go EOL upstream [08:04] shadeslayer: it was not EOL at the time [08:04] and we discovered this issue in the first 4.10 release IIRC [08:04] it only went EOL like 2 weeks ago :S [08:04] shadeslayer: nope [08:04] that should have been .9 [08:04] oh [08:05] hm [08:05] fine, patch it [08:05] otherwise they'd not have it backported to .8.5 [08:05] shadeslayer: u break it, you fix it :P [08:05] I didn't break it [08:05] Marco broke it :P [08:05] I need to deal with all the other shit that I got told is broken yesterday [08:06] will have a look after lunch [08:12] shadeslayer: assigned it to you for the time being [08:13] kk [08:26] ::workspace-bugs:: [1123126] 12.04 plasma init script order wrong @ https://bugs.launchpad.net/bugs/1123126 (by lordievader) [08:35] yofel_: btw, I have been thinking... since we track all of KDE SC branches under kubuntu-packaging for stable stuff a branch set could be done under a different project (kubuntu-packaging-precise-4.8.5 for example) that would IMO make excessive release branching a lot easier to parse and understand by humans [08:36] i.e. instead of having the branches differ by name we make them differ by their containing project [08:43] apachelogger: and when will we have the new vlc backend packages? I still wait for 0.6.1... [08:44] ::runtime-bugs:: [973205] nepomukservicestub crashed with SIGSEGV @ https://bugs.launchpad.net/bugs/973205 (by Søren Holm) [08:45] 13.04 [08:45] oh come on, we have backports PPAs [08:48] *shrug* [08:48] shadeslayer: bug 1123763 [08:48] bug 1123763 in kde-runtime (Ubuntu) "Veromix volume control broken by update." [Undecided,New] https://launchpad.net/bugs/1123763 [08:48] apachelogger: sawn [08:49] do somethign with it then plz [08:49] I propose you reply to that [08:49] certainly not kde-runtime I say [08:49] shadeslayer: I did not look into it, you did... [08:49] No, I did not look into it either, i merely said that it was broken on raring [08:49] you had some choice words about it though :P [08:50] that will not help the fact that I do not have raring and someone reported a bug against -runtime so someone needs to check and find out whether -runtime is in fact affected, and if not move the bug somewhere else [08:51] apachelogger: shrug? Hello? Shouldn't you actually want packages to be distributed? [08:51] actually no [08:51] somebody else: could we have those new backend packages ASAP? We already skipped 0,6.1 that were important bugfixes [08:52] apachelogger: you suck [08:52] yeah, not so much [08:52] I have a 2 person QA policy [08:52] both releases were QA'd by one person [08:52] .1 had a regression [08:52] so if that is any indication [08:52] .2 will have a regression [08:52] well, why do you release, then? Don't you tst your stuff? [08:53] test* [08:53] ehm [08:53] yeah whatever [08:57] shadeslayer: me, Harald and Albert talked about that a while ago here, I just can't find the discussion [08:58] yofel__: oh okay [08:58] (what you quoted was the release-team ML btw.) [08:58] yofel__: true, I don't have a kde-packagers label, and since it was originally sent to kde-packagers, gmail applied the kde-packagers label to all the replies [08:58] apachelogger: would work I guess. It leaves the question what to do with our master branches around ~4.X.0 or even around RC's [08:59] o.O [09:00] * shadeslayer makes a release-team label [09:00] duplicate those and merge on release? [09:00] that sounds like a shitload of work [09:01] yofel_: yeah well [09:01] we could also switch to git :P [09:01] still a shitload of work to merge 150 or so branches even with git [09:02] why would they need merging though? [09:03] my raring builder is broken -.- [09:03] let's take the current nepomuk-core 4.10.0 branch [09:03] today is also a very nice day [09:03] that has an upstream patch [09:03] so the master branch can't build git [09:04] one would need a branch with the patch removed to build git, and around 4.10.1 merge that into master [09:04] ohm [09:04] OR [09:04] that's the wrong way [09:04] you apply all changes again [09:05] master would always build git [09:05] and what tracks the archive? [09:05] another branch [09:05] so we have 2 after all [09:06] you don't get around a shitload of branches [09:06] it's just a question of whether the branches need manual merging which ought not be the case [09:08] ah hm, yeah. I think I get what you mean [09:08] I guess that would need a test run on a couple of packages [09:10] right now you'd probably have: master (building git master), 4.10 (building git KDE/4.10), 4.10-raring (branched from 4.10 some time ago carrying raring patches), 4.10-quantal (branched from 4.10 some time ago carrying quantal patches), 4.10-precise (branched from 4.10 some time ago carrying precise patches) [09:10] on 4.10.1 release the 4.10 branch is merged into all series branches (which in fact would be a fast-forward merge as there should be no conflicts) [09:11] i.e. the tricky part is to ensure that packaging is done in such a way that all supported/targetted series can be supported by the same upstream-tracking branch [09:12] I'm against having the 4.10-quantal and 4.10-precise branches, our backports work well enough there, but otherwise +1 [09:12] if that cannot be the case a major headache appears [09:12] yofel_: architecturally it is better to the have the branches anyway, even if they carry no changes [09:12] IMO [09:12] ok, if you have some use for them [09:12] well, you never know [09:14] for all I care we can also just have 4.10-raring, but the requirement of one packaging to be able to cater all targets remains implicitly there [09:15] which is rather impossible. Sure it works in neon, but even there I have some odd looking dependency workarounds to support multiple releases at once [09:16] for quantal it might work. Precise is just nonsense [09:16] yeah, so you'd want branches [09:17] when a delta needs to be carried there is still a good chance an auto-merge will work [09:19] thinkin about it,sort-and-wrap may be a requirement at that point [09:19] it would help at least [09:19] AFAIK the chances of auto-merging are way better to succeed on line-based changes [09:20] in fact I am reasonable certain git will be able to do that in just about all cases as long as one does not exchange half the deps in one commit ^^ [09:21] yofel_: what I am unsure in all cases though is the changelog, it causes conflicts all and everywhere [09:24] well. You could simply make it policy to never edit the stable branch changelog and instead fill it later during merge from the bzr diff [09:25] would need changelog conformant log entries [09:25] (which we currently do anyway) [09:25] yeah, I guess [09:31] stupid ISP, I have 2 seconds of lag [09:37] apachelogger: should I also backport the fix to the 4.8 branch [09:37] or is it useless [09:37] technically useless but if it's no hassle I'd do it anyway [09:48] wtf [09:48] cannot build precise package [09:50] ? [09:52] pbuilder doesn't have -updates [09:52] apachelogger: ooh [09:53] apachelogger: got a fix? (again) [09:53] Riddell: needs this : http://quickgit.kde.org/?p=kde-workspace.git&a=commitdiff&h=a651fff01cfcef8874c5ddcf7a080467edc49d16&hp=e44b48c6fa1f9875f1189f495b9981ac92a1eb36 [09:53] backported to 4.8.5 [09:53] I'm working on it === jackyalcine is now known as zz_jackyalcine === zz_jackyalcine is now known as jackyalcine [09:59] this is going to take some time :| [10:00] thanks shadeslayer [10:00] yeah I recognise that one [10:00] SteveRiley: ping, fancy doing alpha 2 notes? [10:01] I haven't seen SteveRiley in forever [10:01] no, he did e-mail last month but hasn't followed up as far as I can tell :( [10:03] hm [10:06] rofl [10:06] http://www.neowin.net/news/its-cheaper-to-fly-to-the-usa-than-buy-adobe-cs6-in-australia [10:07] and whaaaa : http://www.opera.com/press/releases/2013/02/13/ [10:08] um [10:08] "Opera will make a gradual transition to the WebKit engine, as well as Chromium, for most of its upcoming versions of browsers" [10:08] so..what's the point of Opera? [10:09] I must be looking at it from a very technical view, Opera must be doing something significant other than making an html engine [10:13] opera makes a browser suite thing [10:17] Riddell: yeah, I didn't quite understand that as well [10:17] * apachelogger should make nu splash and stuff [10:17] -.- [10:18] nu splash! :D [10:18] dunno what tho [10:18] ask nuno [10:18] the new wall is totally not plymouthable [10:18] nu-no :D [10:18] way too much colory [10:18] hehe [10:19] Could not find a configuration file for package "QJSON" that is compatible [10:19] afiestas: qjson for libkscreen? [10:19] * apachelogger notes that qjson is really really really bad with performance [10:20] what you expect monitors to use XML? they keep with the times for fashionable data exchange. [10:20] well if you ignore this ugly VGA connector thing [10:20] I expect kde software to use sensible libraries :P [10:21] and why 0.8 for crying out loud -.- [10:21] ah well, screw that [10:21] my raring vm is broken [10:21] no new plymouth theme for 13.04 [10:21] aww :( [10:22] just because you can't get virtualbox to work? [10:23] also because I have no idea what to do anyway [10:24] make it all purble and blue like the wallpaper? [10:24] yeah, but pattern-wise [10:24] honestly, if you can't make a new one, please revert to the pre-grey one and put a neon-shiny gear into it [10:24] as long as it looks plasma-glossy people will be happy [10:24] yeah [10:24] * apachelogger smacks yofel_ [10:24] Mirv: qtbase and qtdeclarative look good to go, I'll upload them as soon as I get on a connection with decent upload speed (in a cafe now) [10:25] actually [10:25] shadeslayer: you could try to convince nuno to send me the svg of the wallpaper :P [10:25] Mirv: however patches with "Forward: no" scare and annoy me, I'd really like a comment somewhere on why they shouldn't be forwarded [10:25] apachelogger: isn't that opensource somewhere? [10:25] he refuses to give it away, but if I had the SVG I could perhaps try to replicate the entire thing [10:26] wat [10:26] Mirv: the patches have comments which is great but I'm still scared of ending up like qt4 with dozens of patches nobody is quite sure what should be done with [10:26] i.e. the problem is that while the grey splash thingy could be shipped at 800x* and it would nicely scale up the new one does not [10:27] so what can be done is reuse the patterns and manually align them on a fixed color background such that it looks like the wallpaper without really being the wallpaper [10:27] all in all still shitty tho [10:27] * apachelogger reads plymouth codez again to find new exciting ways to make splashy [10:28] ::workspace-bugs:: [1123126] 12.04 plasma init script order wrong @ https://bugs.launchpad.net/bugs/1123126 (by lordievader) [10:28] Riddell: in dep3 meaning "no" means: not yet forwarded. Dist-specific patches are "not-needed" [10:29] apachelogger: don't worry an dupdate qjson to 0.8.1 [10:29] dude [10:29] quantal has 0.7.1 [10:29] * apachelogger did not realize that json changed the last 6 months.... :P [10:31] Riddell: ok, thank you! they are internally being pushed forward, but should have had some more comment like "being evaluated for upstreaming" [10:32] Riddell: upstreaming those are also on a To-Do list, I'll continue to make efforts to push people into working on those [10:33] * apachelogger falls off chair [10:33] Mirv: add a comment saying that at the top of the patch if you can [10:33] oh now it is gone [10:33] phew [10:33] yofel__: I know but it would be good to know why [10:33] there was the most terrible wikipedia please support us message ever [10:34] was like half my screen in yello and black text, a lot of text [10:34] * Riddell out to neuro doctor [10:34] Riddell: ok, adding [10:36] where's sheytan when you need him -.- === jackyalcine is now known as zz_jackyalcine [10:43] who has a amd64 precise VM? [10:53] shadeslayer: what evern happend with the webcam support in ubiquity? [10:59] webcam in ubiquity? why would we need that? o.O [11:09] jussi: because it's fin [11:09] s/fin/fun [11:09] apachelogger: didn't get the time to work on that over the last couple of days [11:10] apachelogger: plus, needs a new pgst [11:10] with snapshot() support [11:12] or we could just patch the current pgst with snapshot support [11:15] jussi: http://i.imgur.com/RfPAXZW.png [11:15] shadeslayer: I suppose... [11:15] :P [11:16] hehe [11:18] apachelogger: I still have to fix that show() stuff that you PM'd me [11:18] * apachelogger show()s [11:18] noo .. not publically :P [11:18] actually [11:19] :( [11:19] shadeslayer: http://4.bp.blogspot.com/-GF1yMm5Y3BY/URbUZSfDbOI/AAAAAAAACMI/jA2UzYtE4_s/s1600/installer-wifi.png [11:19] wait what huh [11:19] zomg [11:19] I totally don't get the choose from file button, but otherwise I think that would be nice [11:19] why the default is a spooky snake is beyond me tho [11:19] apachelogger: where is that from [11:19] looks like a QML port :D [11:20] that polish artist guy :P [11:20] ah [11:20] we should get a qml port at some point tho [11:20] okay, a) Choose from file is pointless [11:20] we also should get a plymouth qml port ;) [11:20] b) take webcam picture should have a live feed from webcam :P [11:20] shadeslayer: I reckon that button is meant to be ... choose from nice default pictures [11:21] shadeslayer: well yeah, but you get to the webcam thing via that page [11:21] apachelogger: you mean like from the left side bar that I have here : http://i.imgur.com/RfPAXZW.png [11:21] i.e. do not make it an own page IMO [11:21] oh [11:21] hm [11:21] IMO the text fields on the user creation page are too wide, and putting the user icon there resolves that nicely plus it really belongs to user creation [11:22] true that [11:22] * apachelogger wonders how to get qml into plymouth :D === zz_jackyalcine is now known as jackyalcine [11:32] shadeslayer: extra points if you embed the webcam snapshot thing into that picture preview when you click the button [11:32] that way you don't get any new page at all === jackyalcine is now known as zz_jackyalcine === zz_jackyalcine is now known as zz_zz_jackyalcin === zz_zz_jackyalcin is now known as jackyalcine === jackyalcine is now known as zz_jackyalcine === zz_jackyalcine is now known as jackyalcine === jackyalcine is now known as zz_jackyalcine === zz_jackyalcine is now known as jackyalcine === blaze` is now known as aske [11:55] I really do not know what to do for the splash [11:55] not even sure on the colors to use === jackyalcine is now known as zz_jackyalcine === zz_jackyalcine is now known as jackyalcine [12:06] aubergine background with blue gear? [12:06] that would at least kinda match ubuntu [12:06] Hi guys [12:07] apachelogger: how flexible are plymouth gradients actually? Is anything more than the previous blue to black gradient even doable? [12:07] hi smartboyhw [12:15] yofel__: one can write a plymouth plugin ;) [12:15] there you can pull all the shit you desire [12:16] yofel__: what gradients do you want? [12:17] yofel__: that was the plan [12:17] shadeslayer++ [12:18] apachelogger: I just copied kde-default.png to /lib/plymouth/kubuntu-logo/background.png and I get all the colors [12:18] yofel__: that is... a plymouth window supports a top-bottom gradient only, everything else needs to be done from scratch [12:18] apachelogger: not sure yet, I kind of come up blank yet trying to think of something that blends nicely into lightdm :/ [12:18] no need for gradients or stuff [12:18] shadeslayer: yes, that's an image, not a gradient :P [12:18] also [12:18] how big was that file again? [12:18] yeah, so, why not use that? [12:19] also it fails if the driver only supports 4bpp [12:19] shadeslayer: it scales [12:19] it scales up and it scales down [12:19] oh heh [12:19] and the more it scales the shittier it looks [12:19] :D [12:19] with the previous one it was nto noticable [12:20] well, it was on HD 56inch [12:20] hehe [12:20] but still not as if you wanted to throw up because of it [12:20] but that was due to the very nature of the previous pic, so that will not work now [12:20] hello everyone === jackyalcine is now known as zz_jackyalcine === zz_jackyalcine is now known as jackyalcine [12:21] hm [12:21] yofel_: go investigate lightdm-kde rendering later [12:21] clones? [12:22] spooky [12:22] so the intartubes also have no good ideas for splash screens [12:22] not really, this here is webchat from work. The place I'm at right now blocks almost everything [12:22] * apachelogger sad :( [12:22] oh === murthy_ is now known as murthy [12:24] http://www.youtube.com/watch?v=S9KQyC3Po40 opensuse has a nice one [12:24] though quite frankly it reminds me of the windows7 one [12:24] smartboyhw: cantata packaged? [12:24] * yofel__ likes the win7 one [12:25] murthy, shoud have been [12:25] smartboyhw: ?? [12:25] murthy, not in the archive. Ask Riddell [12:27] Riddell: why bug 1103383 not closed? [12:27] bug 1103383 in Ubuntu "[needs-packaging] cantata" [Wishlist,Confirmed] https://launchpad.net/bugs/1103383 [12:27] it got rejected from NEW twice, Riddell might know more [12:27] yofel_, oh no [12:27] * smartboyhw goes for a reboot [12:28] yofel__: there is a new version of wine available, does it needs packaging or the wine team will take care of it? === jackyalcine is now known as zz_jackyalcine === zz_jackyalcine is now known as jackyalcine [12:29] yofel__, shadeslayer: what about something simple like osx? [12:30] plane background + logo + indicator [12:30] murthy: they are usually fast on that I believe [12:30] haven't used wine in ages though [12:31] yofel__: so we have to just copy from their ppa? [12:31] apachelogger: I don't mind. But can we add a kind of transition animation for lightdm then? A hard change from plain blue or so to the background would be urgh [12:31] murthy: dunno, I never looked at the packaging [12:32] shadeslayer: does osx have a transition there? [12:32] yofel__: no clue [12:32] why do you people take a clue from other OS ? [12:32] I really dunno what the animation would do [12:33] like there is nothing that would make the blue go away in any more reasonable fashion [12:33] murthy: we're always open for other suggestions [12:34] governing dynamics gentleman [12:34] * yofel__ actually likes the debian-kde ksplash. That looks like KDE and has simplicity [12:34] pics or it didnt happen [12:35] I personally would have a black background with with circular progress indicator [12:35] apachelogger: how about going with a black background with some sort of kubuntu logo? [12:35] :D [12:35] have you seen a latest animated one, made in theme of a game? [12:35] yofel__: plane background + logo + indicator [12:35] I did not say the background should be blue :P [12:36] in fact it is hard to find a sensible blue :P [12:36] yeah, that was just stuck in my mind [12:36] so it would probably a black with a hint of blue [12:36] murthy: huh? [12:36] apachelogger: searching the video [12:37] what would probably look neat from an artistic POV is some sort of 3dish sin vs. cos thing [12:39] apachelogger: we need some more eye candy [12:39] *shrug* [12:39] * yofel__ wonders why he had to think of those win95 screensaver bezier curves when reading 3dish [12:39] those kinds of thingare reeeeeeeeeeeeeaaaaaaaaaaaaaaaaaaaallllllllllllllllyyyyyyyyy expensive to create [12:39] ya, but thats what people demand [12:40] actually create + run ... i.e. since plymouth works in two dimensions you'd have loads of sprites moving loads of images in all possible fashions [12:40] apachelogger, LOL [12:40] murthy: not really [12:40] google search on planet.kde.org with "site:" doesn't work? [12:40] people demand stuff that works :P [12:41] apachelogger: also shiny stuff too :) [12:41] having 2 people spend half a year making a splash screen that on a SSD system you see for exactly 1 second seems not worthwhile [12:41] like, absolutely not worthwhile [12:41] apachelogger: you should think different [12:42] this is why mint scores [12:43] shadeslayer: reminder: ktp-desktop-applets debian/copyright needs verification [12:43] yofel__: http://www.flickr.com/photos/54316230@N06/5425492224/ that looks nice [12:44] murthy: perhaps I should join mint [12:44] wow [12:44] apachelogger: no what i mean is ubuntu+gift warpping=mint [12:44] yofel__: though I'd have dropped the dots and instead had the bubbles animate [12:45] which is actually pretty easy [12:45] murthy: don't get your point [12:45] well, the dots are classic, which does have it's style [12:45] so I should not go to mint, but mint does what the people want, so why would I not go to mint? [12:45] yofel__: nah [12:45] bubbles > dots [12:45] way more organic [12:45] apachelogger: we need girl models for promoting high tech cars :) [12:45] i.e. you see something goes on [12:46] murthy, srsly?:P [12:46] apachelogger: like what? "Only for testing - the OS is still cooking"? [12:46]