[01:22] JontheEchidna: trying to install muon in maverick and receive the following error: /var/lib/dpkg/info/muon.postinst: 11: update-apt-xapian-index: not found [01:47] jjesse-netbook: ah, my mistake [01:48] jjesse-netbook: out of curiosity, search in muon wouldn't happen to be working for you, would it? [01:48] JontheEchidna: let me check [01:48] I don't think it can without apt-xapian-index [01:48] no i cant find any packages with the search [01:48] right, thanks [01:49] installing apt-xapian-index should fix things [01:50] installing that also installs the package python-xapian [01:52] yeah, update-apt-xapian-index is a python script [05:11] [libqapt] jmthomas * 1168579 * trunk/extragear/sysadmin/libqapt/src/backend.cpp Versions of libapt-pkg < 0.8.x need apt-pkg/error.h explicitly included here, or else they FTBFS === _LibertyZero is now known as LibertyZero === _LibertyZero is now known as LibertyZero [09:13] anyone planning to make a Qt Creator 2.0.1 package? [09:22] markey: I'm hoping someone is :-) [09:24] :) [09:25] It's easy enough to install from the downloads on the Qt site, but I'd prefer a package [09:26] huh? [09:26] qtcreator: [09:26] Installed: 2.0.1-0ubuntu1 === yofel_ is now known as yofel [09:26] it is already in maverick [09:27] ossum [09:27] * superfly is not running maverick yet [09:28] me neither [09:28] would be nice to have it in backports [09:29] that said, 2.0.1 doesn't fix the editor sluggishness I'm seeing (just tried it from Nokia site) [10:34] hi. I'm getting patch failures, but I can't understand why they're failing [10:34] * apache2logger is wondering if UPS really manages to get a box from CZ to upper austria in less than a day [10:35] TheKro: Source might have changed perhaps? [10:35] So, I *think* http://paste.ubuntu.com/484410/ fixes koffice/armel FTBFS : what's the best procedure to get it into the archives (I like debdiffs in bugs, but have heard there's a process used by Kubuntu folk) [10:36] nigelb: perhaps - but how do I debug the patch. It seems like it should make the change if I look at the source, but I'm not very familiar with the patch format [10:36] persia: the best way is to find a sponsor and then follow his requests ;) [10:36] nigelb: does @@ -945,14 +945,11 @@ mean change the 14 lines ending at line 945 with the 11 lines? [10:36] apache2logger, OK. Let me know if you see a sponsor around :) [10:37] nigelb: because that is still what should happen [10:37] * apache2logger notes that Kubuntu usually also uses debdiffs in bugreports, or simply debdiffs, except for large stuff which is preferred to go to revu [10:37] TheKro: that neans 14 lines removed from line 945 and 11 added. [10:37] s/neans/means [10:37] * persia files a bug [10:37] nigelb: starting at line 945? [10:37] TheKro: yep [10:37] generally it will all start to make sense when you look at line 945 [10:39] is mcafee security stuff any good for windows? [10:40] nigelb: so then the code has changed - but if I change the 945 to 929 (where that code is), that hunk of the patch still fails [10:40] * apache2logger just received the new laptop for his mum and is wondering whether that default stuff is to be favored over microsofts security essential stuff things [10:40] I think most folks run both MS's stuff and someone else's. [10:40] At least all the shops seem to normall upsell with a secondary scanner of some sort. [10:41] apache2logger: I don't know, my family doesn't use Windows [10:41] well, there is this malware protection thingy from microsoft that is on by default [10:41] TheKro: you need to compare what the patch does in that lines to what is already there [10:41] but apparently they also have a proper virus protection now, just not installed for competition reasons [10:41] TheKro: whats happening is, what the patch thing is there, isn't there [10:41] superfly: lucky you [10:42] apache2logger: Just install fluffy ;) [10:42] fluffy does not suite office needs :P [10:42] Depends on the office, really. [10:42] nigelb: i'm looking in the source file, and what the patch is being told is there, is there [10:43] TheKro: ok, do you have a vcs link to what you're trying to patch? [10:43] persia: +1 [10:44] nigelb: what is vcs? [10:44] http://www.microsoft.com/windows/antivirus-partners/windows-7.aspx silly microsoft does not help its users make decisions [10:44] they only add more things to not know things about [10:44] -.- [10:44] persia: the office where you need an office app :P [10:44] You know, kubuntu is remarkably virus free these days... [10:44] TheKro: er, forget I said that. What are you trying to patch/ [10:44] * apache2logger notes that fluffy does not include a office suite by default, neither is it supported to use one [10:44] (theming wise that is ^^) [10:44] apache2logger: Then Kubuntu ;) [10:44] nigelb: I'm trying to patch ktimetracker, using the patch from comment 7 of http://bugs.kde.org/show_bug.cgi?id=226915 [10:45] KDE bug 226915 in general "manually edting task times is a pain in 4 4 (regression ?)" [Normal,New] [10:45] nigelb: I've downloaded the source of kdepim [10:45] apache2logger: kde uses svn? [10:45] nigelb: that for sure [10:45] but not primary [10:45] :P [10:45] well [10:45] primary [10:45] but not only [10:45] nigelb: yes, but not for long [10:45] nigelb: and i'm sitting in the ktimetracker dir running patch < timetracker.patch (cut and pasted from the comment) [10:45] git migration efforts are under away [10:46] apache2logger, Wouldn't the theme work with webservice-office-zoho (essentially launches a browser)? [10:46] TheKro: doh, you might be going wrong with patch levels. I'm not good at explaining that. [10:46] persia: I dunno [10:46] apache2logger: I hate svn browsing :x [10:47] I wonder if we have a patch level documentation somewhere... [10:47] Microsoft’s competitors will argue that it’s less secure and doesn’t have as many features, but they should be worried. Overall, Security Essentials is excellent and I highly recommend it. [10:47] that is all bogus [10:47] well, first output is: [10:47] brrrr [10:47] patching file taskview.cpp [10:47] Hunk #1 FAILED at 929. [10:47] 1 out of 1 hunk FAILED -- saving rejects to file taskview.cpp.rej [10:47] TheKro: can you bin the code? [10:47] !pastebin [10:47] For posting multi-line texts into the channel, please use http://paste.ubuntu.com | To post !screenshots use http://tinyurl.com/imagebin | !pastebinit to paste directly from command line | Make sure you give us the URL for your paste - see also the channel topic. [10:48] http://paste.ubuntu.com/484414/ [10:48] nigelb: ^ [10:49] looking [10:49] TheKro: um, no. THe taskview.cpp file. [10:49] Paste that file [10:51] nigelb: sorry for the delay - this problem came with a lucid upgrade, and now my vim yank to clipboard isn't working either [10:51] TheKro: heh, no probs [10:52] nigelb: http://paste.ubuntu.com/484417/ [10:53] TheKro: which revision of kdepim did you download? form what I see the patch is already applied to svn [10:54] yofel: I did an apt-get src [10:54] ah, 4.4.5 then? [10:55] yofel: kdepim-4.4.2 [10:55] yofel: I'm not sure all of it is applied yet [10:55] TheKro: why not 4.4.5 ? [10:56] yofel: erm, 4.4.2 are what's in my repo? [10:56] TheKro: we always fix the development version, not current one [10:56] can you install 4.4.5 from the kubuntu updates ppa and see if that helps? [10:56] so first in svn, then backport it [10:56] nigelb: if you tell me how... [10:57] or point me somewhere [10:57] where are minions when you want them :p [10:57] Ah, shadeslayer is away for a week [10:57] as I see it's fixed in svn [10:58] TheKro: if its already fixed in svn as yofel assures me, you don't have to do anything [10:58] nigelb: i need to do something - the program isn't working for me now [10:58] yofel: side note, how do track history in svn from the web interface/ [10:58] TheKro: yofel did tell you how. Install the ppa for the newer one [10:59] nigelb: i don't know how to do that [10:59] meh, doesn't seem to be fixed in 4.4.5 :/ let me try to apply the patch [10:59] * nigelb kicks yofel [10:59] :p [10:59] hey [11:00] as for web interface, no idea, I rarely use it [11:00] yofel: you clone the repo and then cmd-line? [11:01] right, note that kde svn is HUGE (I build it from svn sometimes, that's why I have it here) [11:02] how do I install from the kubuntu updates ppa? I have lucid-updates in my sources.list but don't know about the ppa [11:02] or is it pointless to bother if 4.4.5 is broken as well [11:05] yofel: any luck with the patch? [11:06] yes, give me a moment [11:13] nigelb: (by the way, yank to clipboard wasn't broken, I was just yanking to root's clipboard instead of the user with the browser open) [11:19] hmm. what are the rules for stringfreeze? is it ok to mark a string as translatable that wasn't translatable before (i.e. got overlooked in the source)? this technically introduces a new string, but it's not worsening the situation [11:20] * persia would recommend checking with the translators to ensure it won't make them scream in frustration before doing so [11:20] dpm: ^^ [11:39] yofel: still busy? [11:39] test building in pbuilder, should finished in less than 5 min [11:39] *should be [11:40] cool [11:42] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827104216-0v4eq7hn38ga309g * src/ (Dialog.cpp Error.h) change marshalling order [11:49] yofel: where will I be able to get it when you're done? I've got to leave in a minute [11:49] TheKro: for now I'll put the binaries to http://yofel.dyndns.org/ext/kdepim-4.4.2_patched/ I'll upload it to one of my ppas later maybe [11:51] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827105105-m7endpqqngf7jxs1 * src/ (4 files in 2 dirs) type all dbus errors to Error and add registerMetaType() func [11:51] yofel: thanks [12:11] Sput, I'd recommend checking with translators first by sending an e-mail to ubuntu-tranlators (at) lists (dot) ubuntu (dot) com. If the change is absolutely necessary, you should also notify translators to give them a heads up to translate the string. There is nothing more frustrating than completing a translation and then noticing on release that it is in fact not translated (thanks for the heads up jussi) [12:12] dpm: well, currently the string isn't marked as translateable (missing tr()) [12:12] so it won't be translated without the fix either... [12:20] Sput, ah, right, then I'd just suggest going ahead with the changes and send a brief heads up to translators. They don't mind these fixes, but it's good to keep everyone up to date to make sure they don't miss them and the string gets translated [12:20] thx [12:21] al: ^^^ you'd like to organize that? :D [12:21] Sput, if you or al need help with the announcement or need advice, just give me a shout [12:28] Sput: just leave it out :> [13:15] txwikinger: Sput has a testing request you could take up for your global jam list of stuff: [13:15] [05:10:05] ScottK: if you could have people test if quassel-qt4 and -kde both integrate properly into kubuntu and ubuntu, with all the new-fangled stuff like dbusmenu, appindicators, message indicator, unified menubar etc, that would be great [13:15] [05:10:11] ScottK: we should be fully supporting all that now [14:09] Riddell: to run a script when a new version of a package is installed we need a trigger in the new package right? [14:10] New grantlee release. [14:10] erm [14:11] The following NEW packages will be installed: hplip kubuntu-konqueror-shortcuts openoffice.org-calc rekonq << rekonq?!!! [14:11] aha! kubuntu-konqueror-shortcuts pulls that in ^_^ [14:27] ok.. ScottK got the message.. will put it on my list... Thanks!!! [14:28] Sput: ^^^ testers. [14:44] qt4-x11 4.7 rc1 is in Debian Experimental. We can't upload to the archive until after beta, but it would be worth merging and putting in a PPA probably. [14:45] yofel: sorry, do I need to install all the .deb's in the kdepim-4.4.2_patched directory? [14:47] TheKro: probably not, the ktimetracker one should be enough [14:51] yofel: nope - needs exact version of libraries, and it seems to be cascading [14:52] yofel: ok, only needed 3 libraries [14:53] yofel: Thank you!!! [14:55] Riddell: ping [14:55] Riddell: I messed up with my MPRIS patch for Amarok, there is a new one on Launchpad, can you update the package? [14:58] agateau: Is this critical for Maverick Beta? Riddell is away until Monday. [14:59] ScottK: oh [14:59] ScottK: it would make Amarok work with the Sound Menu on gnome [14:59] Sounds worthwhile. [14:59] apache2logger, JontheEchidna, some other kubuntu-dev ^^^ [14:59] ScottK: it should be quite safe: just an update of an existing patch, not a new patch [14:59] Please have a look at agateau's patch. [15:00] agateau: If I upload it, then I can't be the reviewer for the release team. [15:00] With Riddell away, we don't want that. [15:00] I can take a look [15:00] Thanks. [15:00] ScottK: JontheEchidna: ok, thanks a lot [15:00] bug 617367? [15:00] Launchpad bug 617367 in VLC media player "mpris2 feature exception" [Unknown,Confirmed] https://launchpad.net/bugs/617367 [15:06] JontheEchidna: yes, that's it [15:06] ScottK, txwikinger: thanks :) [15:22] hm, so a DVD only has more language packs, nothing from universe? [15:22] how disappointing [15:40] It also has some additional stuff from Main too. [15:40] It should have koffice, for example, but I will confess I didn't look. [15:51] it should have most of KDE on there too [15:51] http://pastebin.com/QyJwF50e [16:08] Did anyone try the updated Ubiquity yet? [16:15] ScottK: is it possible to download packages from Debian NEW? [16:15] debfx: No. [16:15] It's not public. [16:20] * debfx downloads a snapshot from git [16:21] who messed up my kopete-gcall package! :( [16:21] i demand you fix it now :S [16:21] debfx: If it's qt4-x11 you can get the packaging from qt-kde git/svn (don't recall which) [16:22] apparently someone/something removed the binary [16:22] ScottK: they switched to git [16:22] Thanks [16:23] -usr/bin/googletalk-call [16:23] GAH!!!! [16:24] that reminds me, I wanted to move rekonq to git [16:24] debfx: do give me the link as well... ( mail it to me :P ) [16:25] Riddell: no kopete gcall, for some reason the binary got removed [16:25] dpm: are translation changes made using launchpad accesible via bzr? [16:26] debfx: care to fix kopete gcall and get it uploaded? [16:27] oh wow... crappy timing to break kdenetwork.. just read the topic :( [16:29] guess i'd at least need to change the branch to "Hosted" [16:30] al, it depends. When we set up quassel for translation in Kubuntu, we only set it up for the source packages. In that case, translation changes are not available via bzr. However, if you've got a bzr branch already, you can set up a project, enable translations there and enable automatic translation commits, which will put all translation changes in a branch of your choice for you [16:30] It's actually quite neat [16:30] shadeslayer: why don't you fix it? :P [16:30] if it would just support git ;) [16:31] al: I understand git-bzr (or bzr-git, I forget which) works well. [16:32] debfx: exams!! [16:32] i dont have the time.. itll have to wait for a week ... [16:32] ScottK: yea, i was planning on using that to merge launchpad changes back to the git repo. but as it seems now, i'd also need to do the forward merging, which is currently done by launchpad [16:33] don't know if i really want that [16:35] al, Scott will be able to answer much better than me on that, but why wolud you need to do merging from git to bzr? Why would a mirrored branch from git not work? [16:36] dpm: i have a mirrored branch right now and i don't see any changes made in launchpad translations in that bzr branch [16:36] al, let me see, have you enabled automatic translation exports? [16:36] where is the branch? [16:36] lp:quassel/i18n-master - maybe i'm doing it wrong [16:37] mh, "Currently not exporting translations to a branch. Choose a target branch." [16:37] that sounds wrong [16:37] :) [16:39] shadeslayer: problem is, that I have no idea which files should be in the kopete-gcall package [16:39] debfx: just usr/bin/googletalk-call [16:39] "There is 1 error. - Invalid value" [16:40] it got removed with the last upload .. [16:40] * shadeslayer points to kdenetwork diff [16:41] maybe i can only choose lp hosted branches as export target branch? [16:41] meh, back to square 1 :> [16:42] al, let me ask an LP developer that, but first let's take it by parts, I've got a question: what's the difference between the 'trunk' and 'i18n-master' series? [16:42] dpm: i18n-master is where i do the translation stuff on launchpad and transifex [16:43] dpm: trunk is the official upstream repository [16:43] well, a mirror of [16:43] not actually in use for anything on launchpad [16:44] yeah, basically theres a plugin called bzr-git which imports git repos to bzr [16:44] jelmer @ #launchpad can help you understand :P [16:44] al, so you are using both LP and Transifex? Obviously I'm a bit biased for which translation system to use, but I'll just comment on the fact that using two different sources for translations will make your life more difficult and possibly create conflicts between translators [16:45] dpm: just solve me the launchpad problem ;) [16:45] hahaha [16:45] al, so i18n-master is mirrored or not? [16:45] dpm: it is [16:47] I'm not sure you can do your exports on the mirrored branch then. Let me check that again with an LP developer if that's possible... [16:48] im pretty sure you cant [16:50] hello [16:50] is the KMix.desktop file special? [16:51] al, yeah, I just got confirmation that they are read-only. You'll need to create another branch for translation exports [16:51] dpm: ok, thanks :> [16:51] two-way git-bzr gateway it is then [16:53] al, why two way? Can't you just have git -> bzr as mirror, then a Launchpad-only branch to fetch automatic translation exports, and then manually do bzr->git to merge that branch upstream? [16:54] I try to figure out why all .desktop files in ~/.config/autostart are autostarting except kmix.desktop [16:54] apache2logger, thanks a lot for your answers, even though WP doesn't love me :) [16:55] agateau, ScottK: ack, I apt-get source'd amarok and then forgot about it for a while and then $WORK called. I've just uploaded it now, though [16:55] JontheEchidna: thanks! [16:55] dpm: hmm, yea, possibly [16:56] JontheEchidna: Thanks. I'll have a look. [16:59] small question: how should I understand this version: lbmtp8 1.0.2-1ubuntu1 [17:00] it reads 1.0.2, but the current upstream is 1.2.1, so this is 0.2, right? oh my... [17:01] dpm: doesnt love me either, the third example I had in my original reply was supposed to present the filename tag ^^ [17:02] "upstream is 1.0.12, typo [17:02] apache2logger, yeah, anyway I understood how the markup works now [17:02] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827160215-9ugz19dbl0oc3ixl * src/ (BaseWidget.cpp BaseWidget.h CMakeLists.txt) basewidget deriving from qwidget with a pure virtual setErrorMessage func [17:04] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827160423-ouo7ds4wrih6rdeh * src/ (LoginWidget.cpp LoginWidget.h) port loginwidget to basewidget, setErrorMessage is a dummy currently [17:05] Mamarok: quite simple : 1.0.2 is upstream version [17:05] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827160457-t581wh0t23j16ho3 * src/Error.h make error marshalling a bit more verbose [17:05] then 1ubuntu1 says that there is 1 package in debian [17:05] ( the first 1 ) [17:05] more like 1 rev. upload of package in debian [17:06] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827160509-e2jkwburq7m1r0ap * src/ (Dialog.cpp Dialog.h) implement generic gotError func in Dialog, qobject casts currenPage's widget to BaseWidget and calls setErrorMessage [17:06] and ubuntu1 says that we took debian package > modified it for ubuntu > uploaded it [17:06] so packages syncd from debian will have 1.0.2-1 whereas ones with ubuntu specific modifications are called 1.0.2-1ubuntuX [17:07] shadeslayer: kopete-gcall is fixed in bzr [17:07] and when debian goes to rev. 2, we will have 1.0.2-2ubuntu1 [17:07] * shadeslayer hugs debfx [17:07] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827160706-dbc623qkzse4rivk * src/Dialog.cpp do no check for busyoverlay since it is a QPointer anyway [17:08] \o [17:08] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827160815-ci9an8gylbnw4fl3 * src/Dialog.cpp ensure busyOverlay is deleted in dtor [17:08] Quintasan: hey [17:08] shadeslayer: how is kdesupport import? [17:08] look at my discussions with yofel in #project-neon and decide what is to be done [17:09] not good... still failing [17:09] JontheEchidna: Looking at http://launchpadlibrarian.net/54502404/amarok_2:2.3.1-1ubuntu6_2:2.3.1-1ubuntu7.diff.gz I'm reasonably sure ebian/patches/debian-changes-2:2.3.1-1ubuntu7 is not what you want. [17:09] or last time i checked it was at that [17:09] I can reject it so you can reupload. Let me know. [17:09] kmix.desktop des not run by the %i argument [17:09] bah [17:09] kmix: Unexpected argument '%i'. [17:10] kmix: Use --help to get a list of available command line options. [17:10] I dont know what is this %i argument for [17:11] and I dont know why it seems not to affect KDE users [17:11] but someone mat want to user kmix in a non-KDE desktop [17:12] %i expands as per fdo spec [17:12] I am the operator with my pocket calculator! [17:12] and the %i is as unknown in the command shell and in the ~/.config/autostart and braeaks the kmix autostart in a non-KDE desktop [17:13] EagleScreen: that is nonesense [17:13] * shadeslayer wonders if Mamarok got scared by ubuntu versioning [17:13] environments that implement .config/autostart must implement the desktop entry spec since .config/autostart is dependent on desktop files [17:13] shadeslayer: who didnt? ;) [17:14] :> [17:14] why kmix dont know the &i argument in the terminal? [17:14] is it a KDE specific thing? [17:14] EagleScreen: read what I wrote [17:14] if you do not understand it ask google [17:15] there is no issue [17:15] there is issue but it may not be your [17:16] THERE IS NO ISSUE YOU ARE JUST USING IT WRONG! [17:16] http://standards.freedesktop.org/desktop-entry-spec/desktop-entry-spec-latest.html [17:16] http://standards.freedesktop.org/autostart-spec/autostart-spec-latest.html [17:17] bash does not implement either spec [17:17] bash does also not do autostart [17:17] bash hence does not have a problem with %i [17:17] Mamarok: id suggest #ubuntu-packaging for starting out in packaging [17:17] then who do autostart in .config/autostart ? [17:17] apache2logger: ^ hang out in that channel and help new people :P [17:18] in case your not already on there [17:18] we might want to add "abort-on-upstream-changes" to debian/source/format as I have seen too many uploads with unwanted automatically created patches [17:18] shadeslayer: how can upstream be 1.0.2 when this website says 1.0.1? -> http://libmtp.sourceforge.net/download.php [17:19] debfx: alternatively people could just use bzr-buildpackage :/ [17:19] shadeslayer: I don't want to package anything, I want to know the version number, and apparently there is no 1.0.2 upstream [17:20] Mamarok: in fact upstream is at 1.0.3 [17:20] where? [17:20] Mamarok: http://packages.debian.org/source/experimental/libmtp [17:20] apache2logger++ [17:20] http://sourceforge.net/projects/libmtp/files/libmtp/ [17:20] sorry, my reference is the upstream website, which says 1.0.1 [17:20] * apache2logger uses the upstream tarball pool as reference ^^ [17:20] just pointing out that they have a 1.0.3 release [17:21] people always forget updating websites ^^ [17:21] yeah, seen that now, they really should upgrade their website [17:22] Mamarok: yep, actually if they were smart they would just embedd that sf page into theirs, since they have release notes on that sf page anyway [17:22] since the page looks like crap anyway that is ;) [17:27] * shadeslayer installs playonlinux to checkout steam [17:28] shadeslayer: Don't we have Linux steam packages? [17:28] i dont think so [17:28] * ScottK thought so. [17:28] not yet, I don't think they've been released yet [17:28] theres a article on phoronix [17:28] http://www.phoronix.com/scan.php?page=article&item=source_linux&num=1 [17:29] OK. Having to click on a link exceeds the amount I care. [17:29] which is pretty much i could find for steam on ubuntu [17:29] hehe :P [17:30] okay, then some .desktop files that complain with freedesktop specs can not be autostarted in ~/.config/autostart, is not this a bug? where must I report this? [17:32] shadeslayer: http://www.engadget.com/2010/08/23/valve-denies-having-a-linux-version-of-steam-in-the-works/ [17:33] debfx: saw that too [17:33] im being optimistic :) [17:33] no steam for u :P [17:33] * shadeslayer shoots neon lights in face of Quintasan [17:34] when i create a bzr branch on launchpad, do i have to wait for it to become usable? [17:34] Quintasan: anyways we have wine [17:34] I'm neon immune shadeslayer [17:34] ill survive [17:34] Quintasan: this is the KDE neon.. no one survives [17:34] al: A minute or two, IIRC. [17:34] ScottK: hm, ok [17:35] Quintasan: btw ask sheytan to make a new icon for us :P [17:35] fingers crosse [17:35] d [17:35] kk [17:35] will spam him [17:35] ^_^ [17:39] cya laer [17:39] *later [17:41] JontheEchidna: Did you see my ping about amarok? [18:01] does this branch work for anyone? lp:~quassel-dev/quassel/quassel-i18n-lpexp [18:01] i created it half an hour ago but maybe i need to wait longer? [18:09] ScottK: no, I must have confused quilt [18:09] (was out for lunch) [18:09] JontheEchidna: I'll reject it then and you can reupload with the save version number. [18:10] ScottK: sorry, was out to lunch [18:10] ScottK: yeah, looks like I confused quilt pretty good === jtechidna is now known as JontheEchidna [18:26] http://aplg.kollide.net/images/osiris/013.png [18:26] is it just me or does that hello widget there look completely stupid [18:27] * apache2logger was thinking about using it for notifiying the user about errors [18:28] http://aplg.kollide.net/images/osiris/014.png [18:28] in oxygen it seems even worse [18:28] :S [18:29] al: branches should be published almost instantly [18:30] al: does not work here though [18:47] apache2logger: are you writing a web app or a desktop app :-P [19:04] k [19:04] wrong channel... [19:07] superfly: a desktop app to replace a web app [19:07] ... [20:28] jtechidna: Did you already reupload amarok and someone else accepted it? [20:28] * ScottK was out for awhile. [20:28] ScottK: no, the patch doesn't apply [20:28] Urgh. [20:29] and no frenchman in sight === jtechidna is now known as JontheEchidna [20:29] As long as we get it on Monday, it'll be fine. [20:38] apache2logger: I just ask in jest, because that looks very much like what a web app would do [21:12] superfly: I do not even quite understand why we cannot have the existing web interface do it [21:13] apache2logger: as far as I can make out (from the Ubuntu documentation, anyways), that is what Ubuntu does [21:14] yes, I am replicating their UI for the KDE platform [21:14] I think it's cool, but I completely see your point [21:15] * apache2logger still does not know where to put error messages [21:15] maybe I should just abuse the overlay [21:16] "Our webserver exploded, please try agian [Ok]" [21:16] hahaha [21:16] actually that might not be the worst idea [21:16] KRatingPainter is inefficient: http://imgur.com/q4tzF [21:17] It creates a new KIconEffect class every paint to make a disabled pixmap [21:17] renders my earlier changes to a BaseWidget useless, but... [21:18] It's about twice as expensive as the already somewhat-expensive KIcon(QString) lookup :s [21:18] JontheEchidna: I have to ask... how did you get the menu bar in the panel at the top of your screen? [21:18] I've played, and I can't figure it out :-/ [21:18] * apache2logger notes that kicon(qstring) lookups are not that expensiv because they usually have high cache hit rates [21:19] http://imgur.com/05u52 [21:19] superfly: I'm using plasma-widget-menubar in maverick [21:19] ah [21:19] JontheEchidna: how would you fix the ratingpainter waste though? [21:19] static global instance of iconeffect? [21:20] apache2logger: create the pixmap once in the constructor maybe? [21:20] scaling? [21:20] oh, hold on [21:21] * apache2logger needs to look what that class actually does ^^ [21:21] hmm, no good. the pixamp it applies the affect to is dynamic [21:21] JontheEchidna: you cannot avoid the painting [21:22] you could reduce the amount of painting by storing previous state information (parent size constraint, visibility etc..) and make painting dependent on that [21:22] but that is bound to fail [21:23] for this snippet here: http://paste.ubuntu.com/484643/ [21:23] couldn't we just create one KIconEffect object and let hoverPix and disabledratingpix share it? [21:24] JontheEchidna: I do not think so [21:25] well [21:25] maybe [21:25] if we could that'd cut KIconEffect constructions in half [21:25] JontheEchidna: it however makes me wonder why apply is not some static thing [21:26] In Soviet Russia, KIconEffect constructions cut you. [21:28] hm [21:29] JontheEchidna: now I am a little tired but from my POV it looks to me that the usage there is wrong [21:29] instead the static functions of KIconEffect should be used [21:30] e.g. KIconEffect::disabled, etc? [21:30] *deSaturate [21:30] http://api.kde.org/4.5-api/kdelibs-apidocs/kdeui/html/classKIconEffect.html [21:30] JontheEchidna: yeah [21:31] from the looks of it they aren [21:32] 't using the static ones because the mess with the effect strength [21:32] then it calls for global static instance of KIconEffect [21:32] oh, but that's not really an excuse because deSaturate takes a strength argument [21:33] since it essentially calls upon static stuff just that it requires effect strength help [21:33] er *toGrey() [21:33] JontheEchidna: well then... [21:33] those calls should be replaced by toGray(pixmap, strength) [21:34] resulting in not having to construct thousands of KIconEffects and all their configy needs [21:34] and probably resulting in some nice painting optimization for dolphin and gwenview when using the rating widget [21:34] aye [21:34] good catch [21:44] This could have been done also: KIconEffect *effect = KIconLoader::global()->iconEffect(); [21:50] I just might have to blog about this one. [21:53] ^^ [21:53] JontheEchidna: what do you think about http://aplg.kollide.net/images/osiris/015.png for errors? [21:54] fancy [21:54] * apache2logger thinks that the overlay somehow must be made overlap the buttonbox [21:54] yeah [21:54] I do not realy feel like manually setting and unsetting enables there ^^ [21:55] getting the overlay down there is gonna be tricky though, since it is outside the baseWidget and only a part of the windowing widget [21:55] * apache2logger is scared [21:56] holy code duplication batman: http://lxr.kde.org/source/KDE/kdelibs/knewstuff/knewstuff3/ui/ratingpainter.cpp [21:57] *pow* [21:57] JontheEchidna: must be from the days when nepomuk did not live in libs I suppose [21:57] yeah [21:57] good thing whoever put it there did not bother to get rid of it again :P [21:59] * apache2logger scratches head [22:00] gaze at the expensiveness: http://imgur.com/0E1Js [22:01] * apache2logger is wondering if one can get the relative position of a widget inside another widget [22:02] JontheEchidna: needs to go on a diet clearly [22:02] bbiab [22:16] oh [22:16] uh [22:16] ha! [22:16] QSize baseSize = m_baseWidget->size(); [22:16] int heightBelowWidget = window()->size().height() - baseSize.height() - parentPos.y(); [22:16] baseSize.setHeight( baseSize.height() + heightBelowWidget); [22:16] ^^ [22:20] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827211950-89neb97ppcrpinh4 * src/ (5 files) get rid of stinky old basewidget - we shall favor overlays muhahahah [22:27] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827212710-v8olhqddkwa6jrpk * src/ (5 files) overlay expands over button box, overlay is now a stacked widget, overlay can show error with ok button, that foo get used for errors now [22:28] * apache2logger notes that now there is more designer ui file xml than cpp in this project :P [22:29] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827212905-dxj3oe6t45bajxx1 * src/BusyOverlay.cpp reorder error showing [22:34] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827213334-nhip5ocktu0c52cc * src/Error.h use first line of error reply as message, everything else is pyth0rn junk [22:35] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827213535-cd9e6lpf9i2pzdvu * src/RegisterWidget.ui default register checkboxes to checked [22:36] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827213622-b02sgv6n73m6i0h0 * src/RegisterWidget.cpp make checked terms checkbox condition to register [22:37] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827213722-3ml2alazw2kn6ygt * src/Dialog.cpp remove ktitlewidget [22:43] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827214229-d6qk8igkrrzbva7d * src/RequestPasswordWidget.ui make request description string more useful [22:43] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827214117-1cqd7enrpfe1ckw8 * src/RegisterWidget.cpp upstream says the helps&tips stuff does currently not have any function -> hide [22:48] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827214809-d5edfhdl29hx7ak7 * src/LoginWidget.cpp run login email address through emailhelper [22:51] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827215057-im99w2m011zf0wwy * src/main.cpp set organization domain to you bun too dot com [22:52] apache2logger: even on arm qt4-x11 uses lzma for compression, is that intentional? [22:52] *shurg* [22:53] why is it disabled for kde packages anyway [22:53] debfx: if it got hardcoded lzma foo in the rules ... port it to --with-lzma [22:53] debfx: time outs [22:53] IIRC [22:53] if qt works fine I imagine kde packages do as well [22:53] dunno if that is still valid, even then I would argue that it should be done per-package [22:54] debfx: bindings is quite the beast, kdepim too... [22:54] Installer fixes in http://cdimage.ubuntu.com/kubuntu/daily-live/current/ need testing. [22:54] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827215438-382zmkx5hpjgl58v * src/main.cpp remove client suffix from name [22:56] Anyone around that can test the new ISOs? [22:56] I thought we enabled lzma for all kde, and only disabled it for Qt on ARM [22:57] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827215651-14rsyguru8znw312 * src/Dialog.cpp when on success page -> only show close button [22:57] [muon] jmthomas * 1168921 * trunk/extragear/sysadmin/muon/src/DetailsTabs/MainTab.cpp Hide the purge button when installed, since we already have the purge action as a menu item of the remove item when the package is installed [22:58] JontheEchidna: well currently it's the opposite ^^ [22:58] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827215756-7tseiqsy61h9xruc * src/SuccessWidget.ui mark placeholder labels in success ui not translatable [22:58] :s [23:01] are you sure? [23:01] lzma.pm should enable it for everything kde, except when on arm and when DEB_NO_LZMA is true [23:02] [ubuntu-sso-client-kde] Harald Sitter * apachelogger@ubuntu.com-20100827220142-a4jd5an22rrnczkd * src/ (RegisterWidget.cpp RegisterWidget.ui) move busywidet into a widget with layout rather than a layout alone, so it can be hidden properly [23:05] meh [23:05] my new laptop is still in frankfurt -.- [23:05] * apache2logger will not be able to sleep properly until monday [23:06] JontheEchidna: lzma is disabled if on arm OR DEB_NO_LZMA=1 [23:06] yeah, that's what I meant [23:06] the latest muon build from yesterday is lzma'd: http://launchpadlibrarian.net/54471545/buildlog_ubuntu-maverick-amd64.muon_1.0.1-0ubuntu2_FULLYBUILT.txt.gz [23:06] as is kde4libs: http://launchpadlibrarian.net/53747391/buildlog_ubuntu-maverick-amd64.kde4libs_4:4.5.0b-0ubuntu2_FULLYBUILT.txt.gz [23:07] yes, but not on arm [23:07] oh [23:08] I thought you meant that all KDE on all archs was disabled when you said [17:53:16] why is it disabled for kde packages anyway [23:10] ah no, I was only talking about arm [23:11] yeah, just a misunderstanding on my part [23:11] hm I should have uploaded qt from a server ^^ [23:12] 200MB with 1 MBit/s takes a while ... [23:12] Got spare download bandwidth to grab the latest ISO for testing while you wait? [23:13] * yofel just uploads 230MiB of kdesupport to neon ppa over 1Mbit [23:13] I feel with you... [23:14] dput needs a bandwidth limitation option [23:22] meh, KIconEffect::toGray won't operate on QPixmaps, only QImages :( [23:22] oh well, at least I can use the global KIconEffect instance to avoid two constructions on every paint [23:25] ScottK: current maverick amd64 iso gives me the violet ubuntu splash after selecting start kubuntu, is that known? (KVM) [23:26] yofel: There is a bug on that, yes. [23:26] yofel: Can you install? [23:26] just trying, installer looks nice :) [23:28] btw, does anyone else get 'Exception AttributeError: "'NoneType' object has no attribute 'close'" in ignored' from dput in maverick? [23:28] (using sftp) [23:29] yofel: Excellent. If it installs that will be a huge piece of good news. [23:29] well, it seems to do something while I'm currently at the keyboard configuration [23:30] JontheEchidna: QPixmap::toImage()? [23:30] It's supposed to start installing right away, so that's expected. [23:30] apache2logger: might be just cheaper to get a pointer to the global KIconEffect instance [23:30] The idea is to minimize total time required by doing stuff that doesn't need your input while getting input from you on the stuff it does. [23:31] JontheEchidna: I doubt apply() does it otherwise [23:31] right, seems to work fine so far, where did the advanced tab go btw? or did I just miss that somewhere? [23:31] yofel: dput works fine for me [23:31] else if ( effect != NoEffect ) [23:31] { [23:31] QImage tmpImg = pixmap.toImage(); [23:31] JontheEchidna: ^ [23:32] what if effect == ToGray? [23:32] debfx: it works fine too, but after it finishes uploading it prints that error. I guess I should suspect that bzr sftp plugin it's using [23:32] JontheEchidna: it does that and then passes the image to a QImage version of apply [23:32] and so forth [23:33] so we'd just be doing the convert 1 step earlier [23:33] from what I see the apply qpixmap foo is even more expensive than you would think [23:33] if you call the qpixmap apply version with least arguments you will need at least 3 function stacks to get anywhere [23:34] only to end up with [23:34] case ToGray: [23:34] toGray(image, value); [23:34] yofel: There's some hidden way to get to it. IDK what. [23:34] where image is the image created as per above [23:34] yofel: Let me know when it finishes please. [23:34] and once a QImage gets out, what do you think gets used to convert back to QPixmap? [23:34] result = QPixmap::fromImage(tmpImg); :P [23:34] ScottK: I'll look for it later, expected it to be in the 'Install' tab, but that seems to only show the slideshow now [23:35] sure (currently 51%) [23:35] I'll accept a basic install as working as big progress for the moment. [23:37] JontheEchidna: so the gain is little but you save a couple of function calls which with a lot of calls to the ratingpainter will probably make more a difference [23:40] * apache2logger pays his bed a visit [23:40] nite [23:40] gn8 [23:43] considering that the progress bar resets for the different installation phases an overall progress bar would be nice :/ [23:57] oh great, ubiquity was killed by OOM [23:59] most of the memory was used up by 147 instances of 'plugininstall.p' where each uses 10MiB [23:59] ScottK: ^ [23:59] Ouch.