[00:05] I see marble stuff [03:05] ovidiu-florin: ping [06:16] ahoneybun: pong [06:18] I'll test it tonight [06:19] if all goes good, can/will it be pushed to official backports? [06:19] Thank you ahoneybun, AWESOME work!!! [06:19] * ovidiu-florin hugs ahoneybun [06:22] good morning [07:10] meh, still this problem in Wily: libstreamanalyzer0 : Depends: libclucene-core1 (>= 2.3.3.4) but it is not installable [07:17] yofel: I hear marble is driving you mad? [07:19] Riddell: needs double build for kf5 AND qt4, and so far dhmk isn't liking me [07:19] feel free to give it a shot, all I did is in git [07:19] yofel: dropping qt4 seems acceptable to me [07:19] Riddell: then you'll have to cripple digikam [07:19] Good morning. [07:20] moin [07:20] is ScottK traveling or something? :/ [07:20] Riddell, yofel: need review for https://code.launchpad.net/~apachelogger/kubuntu-packaging/pykde4-no-kdepimlibs/+merge/268193 [07:20] can't transition kdepim without that change [07:22] yofel: curiously rdepends on libmarblewidget21 doesn't show digikam or libkgeomap [07:23] CMakeLists.txt:find_package(KdepimLibs REQUIRED) [07:23] kopete......... [07:24] Riddell: because something in digikam went wrong :/ [07:24] CMake Warning at extra/kipi-plugins/CMakeLists.txt:131 (message): [07:24] libkgeomap: Version information not found, your version is probably too [07:24] old. [07:25] -- Please install the libkgeomap (version >= 3.0.0) development package. [07:25] what's with that versioning.. [07:25] yofel: yeah cos digikam can't sort out their release, they depend on a newer version of libkgeomap than is released I think [07:25] yofel: so let's just drop it [07:25] SIGH [07:26] Riddell: ok, feel free to revert the last couple commit [07:26] s [07:26] Riddell: konsole CI still broken apparently [07:28] what do we do with kopete? it requires kdepimlibs for bonjour protocol, so we either patch kopete and rip out that protocol or make a kde4pimlibs source working around the minor installation incompats or annoy dvratil until he resvoles the incompats [07:28] or remove kopete which I guess won't sit well with its new maintainer :P [07:28] sitter: no .install files changed with that pykde change? [07:29] sitter: I'm for ripping it out [07:29] simplest option [07:29] apparently it didn't install an artifact [07:29] Riddell: any thoughts https://plus.google.com/u/0/103317747728601767381/posts/PwYjWRqCo8M ? [07:30] soee: tell him to talk to kde upstream [07:30] sudo apt install dmz-cursor-theme [07:34] what brings my blood to a boil is when upstream regulars do not merge their applications/* branch upstream after making a change [07:34] leaving it to me to figure out how to sort out conflicts [07:35] which is a right chore knowing nothing of the source base or recent changes [07:36] Riddell: didn't pali the kopete maintainer idle here? [07:36] or was he just on the list [07:36] * sitter seems to recall chatting on irc [07:36] sitter: he does sometimes yet [07:36] yes [07:36] what's his nick? [07:36] pali [07:37] ah well [07:38] I think I'll start packaging up applications for upload to wily [07:40] Riddell: would be good to have few confilcts in Wily fixed [07:41] soee: what's conflicting? [07:41] Riddell: yofel knows more as we talked about it yesterday. Here @ work where i upgraded yetserday to Wily from Vivid, i have atm: libstreamanalyzer0 : Depends: libclucene-core1 (>= 2.3.3.4) but it is not installable [07:43] that sort of thing will get solved on upload of more stuff, or at least it'll show up in excuses so we know where to fix it [07:47] Riddell: kopete isn't blocking anyway [07:47] I'll take it upstream [07:47] Riddell: the sooner we upload this stuff the better [07:47] we have 15.08 final to get on [07:47] right [07:47] any improvements and things gone wrong can be sorted for that [07:47] if it builds its good enough IMO [07:49] sitter: I'll pause CI [07:50] although I need to sort out a reminder for unpausing it cos I keep failing you there and I don't want you to smite me [07:51] "okgoogle set alarm in 2 hours" :P [07:52] ^^ [07:57] yofel: marble patchery upstreamed === 6A4AAE3QZ is now known as TuxMario [08:05] folks, I'm about to head out for the rest of the week -- my husband has his last section of the trail to do and I'll spend those days up north [08:05] I'll have my travel computer with me but probably won't be in irc much if at all [08:05] ciao! [08:35] lordievader: btw are you sure you entered the right kwallet password? [08:35] note that kwallet4 and kwallet5 have different wallets and thus potentially different passwords [08:36] Yes, first it was nothing because of this I opened it and changed it. [08:36] I opened the kwallet* [08:38] most peculiar [08:38] That is what I thought. [09:02] eek [09:03] Riddell: crippling kopete seems like a lot of work now. kdepimlibs is also used in actual libkopete classes (many of them actually) and used in the contactlist -.- [09:07] sitter: hmm fooey, do you know how hard it would be to make a kdepimlibs4 package? [09:24] Riddell: just needs dan renaming a bunch of files [09:25] Riddell: the bigger problem is that I think parts of it won't be functinal [09:25] since there'd be no akonadi runtime [09:25] sitter: they could just port it to frameworks... [09:25] so I have no clue how well this would work [09:25] also apparently the new stuff apparently doesn't talk to the old stuff [09:26] so whatever happens via kdepimlibs4 is essentially lost data anyway [09:26] Riddell: apparently happening see thread on release-team list [09:27] kdepim feels a lot like an as poor transition as kwallet [09:29] sitter: no epich in kimap is deliberate? [09:29] epoch [09:30] you put the epochs there [09:30] there is 0 need for epochs on any of th enew libs [09:31] simplicity of scripting mostly [09:31] should proof to be fun if those libs become frameworks [09:31] mm good point [10:10] yofel: hows marble coming along? [10:10] Riddell will know [10:10] * clivejo runs and hides [10:13] yofel: fancy transitioning kopete? [10:13] sitter is looking at kopete [10:14] :) [10:14] Im running 15.07.90 apps [10:14] Riddell: custom epochs are also fun if debian decides not to take them which makes our packages unsyncable [10:14] please don't add any where they're not needed [10:14] gotcha [10:14] kopete build: https://www.youtube.com/watch?v=npjOSLCR2hE [10:15] and the kwallet popups seem to have settled down :) [10:15] lol [10:16] parrots FTW [10:16] sitter: lol... [10:18] Riddell: instead of unpausing please tell me when the upload is done [10:18] kopete <3 [10:18] Riddell: unless I left for the day already [10:18] sitter: ok, still ongoing.. [10:18] * sitter tries to fiddle together an ISO [10:24] so kopete is dead as a dead parrot? [10:27] its build is [10:56] Hi all [11:04] hi BluesKaj [11:04] hi clivejo [11:22] mh [11:22] actually [11:22] Riddell: why does the applications upload cause builds anyway? [11:22] isn't that NOCI? [11:23] ah yes [11:23] Riddell: why is that not marked NOCI? [11:23] sitter: I seem to remember last time I used the script with NOCI it didn't make any diference [11:23] sitter: how come there are build failuures on KCI but not in the Staging PPA? [11:23] clivejo: KCI is ahead, also quality constraints are higher in KCI [11:23] clivejo: It's not dead, it's pining for the fjords [11:23] Riddell: seems like something that should be fixed? [11:24] Riddell: done pushing yet? [11:24] DCI full of builds too :( [11:26] Riddell: http://kci.pangea.pub/job/merger_okteta/202/ [11:26] stuff like that is probably why NOCI didn't work. if you push more than a NOCI commit the entire thing will need integration [11:27] also that merge failed [11:27] FYI [11:28] how does one get an account at Debian, Im going round and round in circles on the website! [11:30] clivejo: what kind of account? [11:32] to access Alioth [11:37] get alioth account, set ssh key, join on irc and ask for access https://pkg-kde.alioth.debian.org/join.html [11:38] latter will probably need someone to vouche for you not going to break the repos [11:40] someone like Harold Sitter? [11:42] Harald even! [11:43] badumtss === jussi is now known as Guest55514 [11:44] clivejo: yes, I am not comfortable enough to do that already though [11:45] :( [11:45] Riddell generally is more confident with this sort of thing ;) [11:45] sure happy to [11:49] shadeslayer_, Riddell: new script in pangea-tooling/jenkins_unqueue.rb to remove jobs matching a regex from a jenkins queue [11:49] sitter: thanks Harold [11:50] Riddell: also, I've just wiped the CI queue, might as well do the pointless CI at night when nobody's watching [11:50] shadeslayer_: ... === shadeslayer_ is now known as shadeslayer [11:50] sitter: :P [11:51] my qemu doesn't like my arguments anymore :( [11:52] uff, somehow that binary is no qemu-user [11:53] sitter: I see your problem, you are arguing with it, try asking nicely ;) [11:53] lol [12:01] uploading applications 15.07 [12:03] http://kci.pangea.pub/ lots of things can do with improvements btw [12:08] * shadeslayer is busy fixing DCI [12:25] lots of new packages, which should I remove the epoch from? https://launchpad.net/ubuntu/wily/+queue [12:25] stuff like dolphin need to keep it [12:25] Riddell: you also need ot remove them from the CI repos now... [12:25] akonadi needs it [12:26] baloo-widgets needs it [12:26] ktnef doesn't [12:26] kpimtextedit doesn't [12:26] kontactinterface doesn't [12:26] kaccounts-integration technically doesn't but isn't all new I think [12:27] kmime, kmbox, kmailtransport, kldap, kidentitymanagement, kholidays, kcontacts, kdebugsettings, kcalutils, kcalcore, kblog, kalarmcal, gpgmepp don't [12:28] so, everything but akonadi, baloo-widgets, and dolphin can have their epoch dropped [12:28] err.... so, everything but akonadi, baloo-widgets, dolphin, and kaccounts-integration can have their epoch dropped [12:29] thanks sitter! [12:29] sitter: how to remove them from the CI repos? just delete in launchpad? [12:30] o/ [12:30] Riddell: guess so, removing from CI is fiddly nonesense that I am entirely not happy about [12:30] perhaps I should introduce a save guard against epoch bumps where another person needs to ack the bump for it to integrate [12:30] it's not the first time we had arbitrary bumps that then got undone again [13:03] Riddell: computer disagrees with ark packaging apparently http://kci.pangea.pub/job/wily_unstable_ark/2/parsed_console/ [13:20] ovidiu-florin: lots of it was yofel [13:22] you should *test* it before putting it in backports. From what I saw, a couple things were missing during configure. So you'll have to decide whether 4.12 with some things missing is better than 4.9 [13:23] (e.g. thanks to their libkgeomap integration sucking, you won't have location resolving for geolocations in exif data) [13:24] hm, maybe we should re-enable the builtin libkgeomap for wily [13:27] at least the kde4 libs in the backports are pretty much stable these days, so I don't think you have to do another rebuild to put it there [13:27] all I know is that ovidiu-florin is happy [13:27] no clue on what works [13:28] sitter: no traveling, just busy. The pykde thing is probably fine. [13:28] one person being happy is not sufficent QA to put something into our public PPA's ;) [13:28] oh I know [13:28] just saying that I have not tested it [13:28] you don't have to test every little feature, but at least compare it in general [13:29] plus I'm on wily anyway [13:29] k [13:29] thanks for giving it a try ;) [13:29] what packaging it? [13:30] https://twitter.com/OvidiuB13/status/633524406158012416 [13:52] sure, otherwise it wouldn't be there now [13:52] and I hope I wasn't too fast ^^ [14:03] FYI: yesterday & today a fresh wily installation in virtualbox 4.3.30 and 5.0.2 is not usable as soon as the guest OS extenstion installed: Only plasma background and only Popups work are shown. All plasma panel & widget are not shown. Terminal, dolphin etc are invisible and never the keyboard/mouse input. Final touch is 'Fallback session' does not work (xterm is installed) and login prompt reappears after a sec. [14:04] Something disaterous happend between blog '.. 5.4rc ready for testing in wily' and yesterday ;-) [14:16] xsession-error mostly consists of of lines like: OpenGL Warning: {gl*NV not found in mesa table|XGetVisual ...}' [14:18] you were a bit fast yofel lol [14:22] all KCI wily buids are being redone to adopt gcc5 [14:22] starting later today [14:22] half an hour or so [14:46] ahoneybun: yeah, sorry. I hope you saw a couple useful things anyway [14:47] ovidiu-florin: ahoneybun: you guys still needing this ec2? [15:00] <[Relic]> has session saving been fixed in 15.04 yet? does it properly save where programs are running on each different virtual desktop when you shut down a restart yet? === mgraesslin_ is now known as mgraesslin [15:04] Riddell: yofel and I got the package backported just need some Q&A [15:04] also Riddell we need to get kubuntu-settings out! [15:13] Riddell: http://kci.pangea.pub/job/wily_unstable_qt-gstreamer/8/ [15:21] Hello all. I use the master archive server and I seem to have a weird dependency issue that can be replicated in a virtual machine. apt dist-upgrade does not work. I ran apt with the pgkProblemSolver debug option: https://paste.kde.org/p0yysht3g [16:01] kci flooded with builds [16:02] * sitter out [16:04] The issue seens to be related to libqca2 and libqca2v5 which can not be installed at the same time - and some applications depend on one and some on the other. This seems to be especially important for libplasma3. [16:04] Sitter: Thank you for your reply. I am kind of new to this; who is kci? [16:05] This kci: https://community.kde.org/Kubuntu/CI ? === phoenix__ is now known as phoenix_firebrd === phoenix_firebrd is now known as summa [20:15] would someone have time to help me package plasma-widget-awesome-widgets? [20:41] hi -I've noticed the firefox package in wily is older than in the stable releases - i.e wily = 0.38 vivid=0.40 [20:42] Is this the version with that seriously javascript pdf viewer flaw? [20:44] not sure, you can try to install firefox 40 from wily-proposed [20:45] yofel: ok will do.. [20:50] is the best way just to add a wily-proposed line to sources.list then install via apt? [20:51] yossarianuk: yes, but try to just apt install firefox, and disable proposed after that [20:52] (it will probably pull a couple other things though) [20:52] yofel: cheers I saw apt-get upgrade pulled in a few things after adding the list (I didn;t ok that though..) [20:59] yofel: where do I find the version of the license for GPL-3.0 that lintian is happy with, its saying the one I took from http://spdx.org/licenses/GPL-3.0 is too long! [20:59] can you show me the passage it complains about? [21:00] but usually, you would take the one that's shipped with the source you're packaging [21:01] yofel: I tried this - https://github.com/arcan1s/awesome-widgets/blob/master/COPYING [21:01] but its too long aparently [21:01] well, you don't put all that in debian/copyright [21:02] and when I just use the short form I get this - E: plasma-widget-awesome-widgets: copyright-should-refer-to-common-license-file-for-gpl [21:02] where is the common licence file ! [21:02] did you read the documentation for that tag? [21:03] read the first line of it ^^ [21:20] There must be somewhere with the short licences [21:21] clivejo: the short forms are embedded in the full ones [21:21] see END OF TERMS AND CONDITIONS in GPL-3 [21:21] Ive copied and pasted out of a previous copyright file [21:21] that's fine [21:22] but there must be an easier way [21:22] there must be a list of just the text needed in the copyright file [21:24] not that I know of, there are examples in the DEP-5 documentation. From that you're supposed to figure it out yourself [21:25] for the GPL, you take the short form for insertion in the code files, strip the copyright holder, and append a section pointing to common-licenses [21:27] but usually people just C&P from other files I think [21:27] after all the copyright file is shipped wit the installation, so you have hundreds of examples in /usr/share/doc/ [21:28] I just want to know what to put in the copyright file [21:30] that's actually a very hard to answer question. You need to put all licenses and copyright holders of the whole source code in there. [21:30] For licenses that have a copy in /usr/share/common-licenses, you may include the short form and point to the files on the system. For other licenses you need to include the full copy [21:30] there is 'licensecheck' which attempts to tell you which licenses are used in the source files [21:32] yofel: would you have time to check plasma-widget-awesome-widgets - 2.4.0-0ubuntu1~ubuntu15.10~ppa1 in my PPA? [21:33] maybe in a bit [21:33] The rules file Im not convinced is right [21:38] grrrr why is it wanting to install mailutils! [21:41] smartmontools recommends mailx | mailutils [21:42] does the package not do something useful without hddtemp, mpd, smartmontools? [21:42] he says they are optional deps here - https://github.com/arcan1s/awesome-widgets [21:43] maybe I should just ommit them and let the user figure it out? [21:43] no, we have soft-dependencies for that [21:43] just demote them to "Recommends:" [21:43] see https://www.debian.org/doc/debian-policy/ch-relationships.html#s-binarydeps [21:44] will that install them? [21:44] in the default desktop system configuration, yes [21:44] quoted from above page: "The Recommends field should list packages that would be found together with this one in all but unusual installations." [21:46] for dep, the rule is "The Depends field should be used if the depended-on package is required for the depending package to provide a significant amount of functionality." [21:46] it's your job to decide where to draw the line [21:47] well, worst case someone will file a bug ^^ [21:47] well I thought having SMART info, HDD temp and access to media player would be the main function of the widget :/ [21:48] but didnt know it was gonna pull in half the archive to provide that functionality! [21:48] ok, then leaving it as a dep might be reasonable [21:48] Ive added a recommends [21:49] what would you do in this case? [21:49] where would you draw the line? [21:52] I would say recommends, it seems to do a couple other things as well. But I would really need to try it for the final decision [21:53] Im only practicing [21:53] I found it on KDE apps and couldnt see it packaged for kubuntu anywhere [21:54] dont even know it it is any use! [21:55] btw. if you want to get deeper into debian packaging, you should read the entire debian policy at some point. The PDF version has 105 pages, and it's somewhat dry, so try to find some boring moments to read through it until ~christmas [21:55] it explains how most of the files in debian/ work and what rules you should be following [21:55] yofel: I could read books til the cows come home, but unless I actually do it nothing sinks in [21:56] yeah sure, you shoud just roughly know what's inside, so you know where to look when you need it [21:56] have you a link handy [21:57] would be a good reference point [21:57] either read the online version https://www.debian.org/doc/debian-policy/ or install debian-policy and read /usr/share/doc/debian-policy/policy.pdf.gz [21:57] the pdf version is handy for searching [21:58] https://www.debian.org/doc/debian-policy/policy.pdf [21:58] or that ^^ [21:59] :) [21:59] bedtime reading me thinks! [22:01] * clivejo sends it to kdeconnect [22:04] rules looks fine from what I see. In case the plasmoid isn't properly installed you might need --with=kde or --with=kf5 (those are debhelper plugins) [22:05] have I passed -DKDE_INSTALL_USE_QT_SYS_PATHS=ON properly [22:06] or should I use --with=kf5 instead? [22:06] hm... [22:07] might not work like that. If that's a cmake define, you might need: [22:07] override_dh_auto_configure: [22:07] dh_auto_configure -- -DKDE_INSTALL_USE_QT_SYS_PATHS=ON [22:07] yeah, that looks better to me [22:09] should I - include /usr/share/pkg-kde-tools/qt-kde-team/2/debian-qt-kde.mk? [22:09] as it's kde software, --with=kf5 might still be required [22:10] or use that, but for kf5, you need v3, not v2 [22:10] oh yeah, 3 is kf5 [22:10] so confusing! [22:12] anyway, I'm off to bed [22:12] gn8 [22:13] thanks and night night [23:47] <[Relic]> has session saving been fixed in 15.04 yet? does it properly save where programs are running on each different virtual desktop when you shut down a restart yet?