[00:00] Darkwing: ah === Quintasan_ is now known as Quintasan [09:28] Good morning [09:47] yofel_: Ah, I figured it out, Apper fails to update when there is a conflict [09:47] I had some multiarch libs to upgrade and there was a conflict [09:47] But, still, with multiarch on by default it makes me a bit uneasy about uploading it === yofel_ is now known as yofel [09:55] moin [09:56] well, it would be as broken as aptitude, so I agree there [10:23] apachelogger: pingy [10:26] morning [12:10] uh.... does rekonq do something for anyone right now? [12:10] I enter an url, press enter and get an white page [12:11] oh fun, konqueror: [12:11] Additional Information: Unable to create io-slave: klauncher said: Error loading 'kio_http'. [12:16] hm, reboot fixed it -.- [12:26] I hope we can fit firefox onto cd at some point [12:28] well, firefox doesn't give me access to www-admin.kubuntu.org so it wouldn't have helped there ^^ [12:28] it would be interesting to know which browser and email client kubuntu users prefer [12:29] heh, is that still broken on everything except (re)konq? [12:30] needs rekonq or konqueror [12:30] wasn't that a security feature at some point? :P [12:34] heh, I hope not [13:38] yofel, shadeslayer already seen the nepomuk diff in kde-packager? === bulldog98_ is now known as bulldog98 [13:43] hey guys [13:44] is rc of kubuntu ready too? [13:56] * bulldog98 is testbuilding with that patch. Then I’ll upload that to ninjas === neversfelde_ is now known as neversfelde [15:18] Quintasan: hi! can you update your system using pkcon instead of Apper? [15:18] dantti: ping [15:19] hey :) [15:19] It might be aptcc that needs fixing but we can only be sure if pkcon or gnome-packagekit have the same behavior [15:20] also it would be nice if you know which package is causing that.. [15:21] the error information is not really good enough [15:22] maybe we need to poke hughsie again to allow terminal output to be passed to the backend for debugging [15:24] ximion: if I use gdebug in the code and packagekitd --verbose is run I could use that but aptcc code doesn't debug this part.. [15:25] my biggest concern is about + and ~ being used on the pkg version, maybe packagekit removes it 3+git20111005+r160-8~oneiric1 [15:25] improving aptcc is already on my TODO list, as well as making AppStream support available for Apper [15:26] I don't think it does... [15:26] only * is forbidden, afaik [15:27] ~ and + are bothe allowed [15:29] weird.. [15:29] I have no clue ... [15:34] yofel: hi! have you been able to reproduce this "fails to update" bug with apper? [15:34] dantti: could this be related to multiarch? [15:34] ximion: that would be my second bet, but I dunno never used it. [15:35] It only tell that the package is not found if it can't find a package at that version [15:35] it doen't look at the arch part [15:35] but that might be the issue [15:35] dantti: yes, I guess it is... [15:35] I dunno if some new code is needed to know the arch.. [15:36] but: I use multiarch and it works good [15:37] so, hmm [15:40] dantti: it is a bit difficult to convince someone to upload Apper if it does not do what it should do... [15:41] sure, but it's also hard to fix this issue without a test case, also Apper probably does what it should do, I believe aptcc is the one to blame [15:41] dantti: in this case, it should be safe to upload Apper, because then it works for 90% of all users [15:42] but I wonder why I don't get bug reports against gpk for this... [15:42] ximion: probably safe to upload, and also upload a fix for packagekit-aptcc [15:42] usually, if something's broken, they recognize it in a few hours [15:42] I believe few people use it ... [15:43] or don't you think so? [15:43] There's nothing special in the update ui, it simple pass the pkg id it got from pk, so it should be broken on gpk and pkcon too [15:43] there are many users, according to popcon [15:43] hmm [15:44] of course, not as much as USC users :P [15:44] hehe [15:47] dantti: btw, what was the reason not to leave apt's cache open? [15:48] you once told me that there were "errors" when you did this... [15:49] ximion: the main problem is that if you run apt-get update on the cmd line the gui will not be aware of that.. [15:49] (oh, and we also need to take a look at the gui-filter: for some users, it doesn't work :P => http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=638643 ) [15:49] Debian bug 638643 in packagekit "packagekit results in no results with graphical package filter." [Normal,Open] [15:49] tho this could be changed as we have an /etc/apt.d script that tell us it was changed [15:50] but it also has the problem with the /etc/sources.list which could be fixed too with a watcher... [15:50] dantti: So, the problem is that if the cache stays open, you can't run apt-get update? [15:50] or just that the cache needs to be reopened again? [15:50] you can, but you have a in memory outdated cache.. [15:51] yes you need to reopen it [15:51] afaik... [15:51] on openSUSE, they just tell PK to update [15:51] (zYpper has a plugin for this) [15:51] the same should be possible for apt, I guess [15:52] the problem there is that the gui filter filters by the package section... [15:52] together with the reworked parallel transactions hughsie is working on, PK-aptcc should become really fast [15:52] (fast enough for the USC) [15:52] ximion: it should be, tho on openSUSE things are very bad on that backend... it actualy refreshes the cache before each operation [15:53] ximion: sure, it just needs testing but would be great to only open the cache on the first run.. [15:53] I let the way it is now as it's safer.. [15:54] true [15:55] but PK needs to become effectively better than the other solutions ;-) [15:55] at time we have native tools, aptd, qapt and packagekit [15:55] so we have three different implementations of the same thing [15:55] having just one would be great [15:56] and having PK as the solution would be excellent, so code is shared across distributions. [15:56] and there are already KDE apps making use of PK [15:59] ximion: I did my last commit for apper 0.7, now descriptions are back to normal. I'll release a tar ball as soon as possible... [15:59] ximion: I'm giving linux classes right now and the students have just came back from lunch [16:00] if you have more details or can build a test case we fix this issue but I believe 100% that it is a backend bug... [16:01] okay [16:02] cya [17:49] Quintasan: still need that testbuild and testing stuff? [18:02] bulldog98: seen now, is your package ready? [18:03] oh my [18:03] yofel: in about 1min [18:03] ASSERT: "s->parsed == false" in file /build/buildd/project-neon-kdelibs-2+git20110924+r92127/kdecore/kernel/kcmdlineargs.cpp, line 526 [18:05] ximion: as I don't usually use mulitarch, nope - for me it worked. The only issue I had was refreshing the cache being far too slow. [18:05] huh, runs with kcm [18:05] shadeslayer: where do you have that ancient version from? [18:07] yofel: neon, haven't updated in forever [18:07] current is 2+git20111002+snap4.7-59 [18:07] yup i know [18:07] wait what [18:07] snap ? [18:08] snapshot, as that's the only way to get something from the 4.7 branch [18:08] the import is junk [18:08] ah ok [18:08] right, now i remember [18:10] yofel: uploaded to ninjas [18:10] thanks [18:10] * yofel resyncs kdesdk bzr with archive [18:11] yofel: btw the 4.7.2 stuff is already uploaded? [18:11] it'll get into -proposed soon, I'll push it to the public PPA first in an hour or so [18:12] (make that after your patch is built and tested) [18:12] Quintasan: on neon, compiling apper from git and trying to open something from "Categories" gives me : QSqlDatabasePrivate::database: unable to open database: "unable to open database file Error opening database" [18:13] so thats broken [18:13] shadeslayer: yep, same what he got [18:13] well, i don't get a error dialog [18:13] and me I think, although I didn't look at xsession-errors [18:13] thats the debug output from konsole [18:13] * yofel reads backlog [18:15] well from the backlog and from the error, it looks as if the db is busted [18:16] * Quintasan would like the Policy KCM back for Precise [18:16] shadeslayer: Well, yeah, thanks [18:16] np [18:16] Quintasan: congrats on the visa btw [18:17] * Quintasan wonders why his panel is not transparent when on the right side of the screen [18:17] Don't forget yer swimming shorts, there's a hot tub there ^_^ [18:17] I have a new freakin gfx card and effects are not snappy and smooth [18:17] Why the hell deleting .kde is solution to this [18:17] Quintasan: AMD? [18:17] s/AMD/ATi/ [18:17] shadeslayer meant: "Quintasan: ATi?" [18:18] shadeslayer: You kiddin? GeForce GTX560 [18:18] ah ok [18:18] maybe it's too new? :P [18:20] ... [18:20] shouldn't be... my 250GTS worked fine from the day I got it [18:21] Currently I have three problems with my Linux experience [18:21] 1. Why the hell Nepomuk sucks so much [18:22] 2. Why the hell the desktop effects sometime work and sometimes work like shit and sometimes don't work [18:22] 3. Why on Earth deleting your config should resolve problems after updates [18:22] Hmm [18:22] four actually [18:22] 4. Why the hell "sync" isn't getting called when copying stuff onto usb thumb drives [18:23] 1) sebas is trying to fix that, 2) ask martin 3) agreed 4) at the end or while copying? [18:24] at the end [18:24] which reminds me I forgot to file a bug for my most annoying plasma issue... [18:24] what's wrong with sync running at the end? [18:24] (should be fsync() though I guess) [18:24] NOT running [18:24] Quintasan: I've seen that as well [18:24] I mean [18:24] files don't get copied properly [18:24] oooops [18:24] I copy files to a thumb drive [18:24] I'm tired [18:24] It says it's done [18:25] yup [18:25] well, that's why you need to unmount it... [18:25] I try unmounting and it says it can't unmount since something is still being done [18:25] or "eject" [18:25] yofel: hangs forever when you do that [18:25] Exactly [18:25] * shadeslayer waves fist at rygel [18:25] ah... I think I saw that once [18:26] I somehow have a habit of running sync by hand lately... [18:26] i just use rsync now [18:26] or cp for smaller files [18:26] for some reason I'm typing in the dark even tho my keyboard has a backlight feature -.- [18:29] any plans on packaging plasma active 1 ? [18:31] [lp:~kubuntu-packagers/kubuntu-packaging/kdesdk] Philip Muškovac * 146 * debian/ (patches/no_structures_tool.diff changelog patches/series) Guard includes which are not available if the structures tool is not built. LP: #852430. [18:35] yofel: okay, thanks. [18:35] Quintasan: have you read what dantti said? [18:35] ximion: Nope [18:36] do you think it is okay to upload Apper if it works for most users? [18:36] the problem you are axperiencing seems to be a corner case... [18:36] I asked some other people and noone else has this error [18:36] also, this is a PK error I guess and has nothing to do with Apper [18:37] Well, apart from that I had no problems whatsover with Apper [18:38] ximion: Please recheck if REVU has the right version and let me know so I can take last look at it [18:56] Quintasan: the version on revu is okay and working [18:56] Splendid. [18:56] and it's the same as my local version this time :D [19:03] ScottK, ximion: Apper uploaded [19:12] Quintasan: thank you! [19:50] Back on KDE after a month of trying Unity. Boy am I glad to be back === yofel_ is now known as yofel [20:03] anyone about? [20:12] no [20:17] jjesse: lol [20:18] ANyways, Im in the middle of reporting a bug, but jockey-kde is not being started with kdesudo, therefore not allowing you to change anything. [20:27] bug 871469 for all interested peoples [20:27] Launchpad bug 871469 in jockey (Ubuntu) "jockey-kde is not started with root privelidges" [Undecided,New] https://launchpad.net/bugs/871469 === Quintasan_ is now known as Quintasan [20:52] jussi: Why would you even try this heinous abomination of an GUI? [20:59] Quintasan: everyone of us has to be open minded [21:05] bulldog98_: your new kde4libs upload adds a new symbol, please update the symbols file for libplasma3 [21:07] please push to bzr after doing that and fixing the changelog (and mention that the patch is from upstream please and/or use a dep-3 header in the patch) [21:16] [lp:~kubuntu-packagers/kubuntu-packaging/kde-l10n-common] Philip Muškovac * 108 * debian/ (changelog config) New upstream release (svn: 1256933, type: stable) (PPA upload) === BarkingFish is now known as Guest79173 === claydoh_ is now known as claydoh [22:15] ScottK: Apper is in the Oneiric queue now :) === micahg_ is now known as micahg