[06:27] good morning [07:59] Good morning. === ghostcube_ is now known as ghostcube [09:13] ouch, my dentist had a break down half way through drilling out my tooth :( [09:14] shadeslayer: they did one a couple of years ago too, I've got the files if you want them [09:16] Riddell, LOL [09:27] Riddell: I see, well, I still have to watch the rest of that one [09:27] I watched 30 mins of it, then I started to have a headache [09:37] shadeslayer, oh:( === 20WACYJOB is now known as toscalix [09:58] Riddell, you are a Debian developer right? [10:03] smartboyhw: not I [10:04] Riddell, :O [10:04] ScottK: has rights I believe [10:04] * smartboyhw wants somebody to sponsor his first package into Debian.... [10:05] smartboyhw: what's the lucky package? [10:05] Riddell, zbackup [10:06] It's a new one, so it's a good opportunity to train it out my packaging skills [10:06] (Albeit it's quite an easy one) [10:08] okay I think login is faster if cache is populated from the install [10:10] for kde it will be, cache generation takes quite a while [10:12] shadeslayer: what's your current method? [10:13] Riddell: http://paste.kde.org/p171e1662/ [10:13] Riddell: though I can remove the dir creation stuff by running kbuildsycoca4 twice , once before I copy the cache, and once I copy it [10:14] * smartboyhw forgotten the command to check copyright again, yofel [10:14] IDIOT [10:14] which is what I'm trying right now [10:14] smartboyhw: licensecheck [10:14] ^ [10:15] heh, also missing a ] in the subprocess.call [10:19] smartboyhw: also run suspicious-source [10:20] Riddell, yeah? [10:20] Riddell: sebas agreed that copying the cache should work [10:24] smartboyhw: so you don't let in any non-preferred modifiable form files [10:40] yofel: does python have a method to copy all files matching a regex? [10:40] then I can just copy over *.kcache [10:41] I don't think so, you can use a shell call or just match the files in python easy enough [10:42] subprocess.call("cp * foo/", shell=True) [10:42] I think [10:42] ah yes, shell call is probably easier [10:43] python has a glob module I think that you could use, but shell would be easier === ghostcube_ is now known as ghostcube [12:07] hi manchicken [12:07] Hiya [12:10] 'Morning all [12:17] xnox: ping [12:17] xnox: is there a way I can extract the name of the live session user? [12:18] $USER, no? [12:19] ubiquity will be running as root [12:19] so won't $USER be root? [12:19] problem is : cache_path = subprocess.getoutput('kde4-config --path cache') : gives me the cache of the root user [12:19] which does not have the cache files [12:20] but the live user does [12:20] shadeslayer: class DM has self.username, there is also get_casper('USERNAME', default_username) [12:20] shadeslayer: SUDO_USER should be defined in the environment as well. [12:21] (SUDO_USER used in a couple of places in the installer itself) [12:21] shadeslayer: else fallback to root, as e.g. it is valid to run ubiquity as root with no "normal" accounts. [12:22] xnox: I don't wnat to fallback to root because it doesn't have the KDE cache files populated :) [12:22] shadeslayer: or simply use the drop privileges decorator, to executed something as a mere mortal. [12:22] aha [12:22] that sounds better [12:22] shadeslayer: preexec_fn=misc.drop_all_privileges [12:22] xnox: can drop_privileges be used in scripts/plugininstall.py ? [12:23] no idea =)( [12:23] no idea =) [12:23] shadeslayer: try and see what breaks =) [12:23] lol [12:23] xnox: once dropped, can I elevate again? [12:23] ( doesn't seem like it ) [12:27] xnox: v [12:27] erm [12:27] xnox: http://paste.kde.org/p699b323c/ [12:27] xnox: only problematic line is line 10 [12:28] need to figure out how to run that as the live user [12:28] shadeslayer: drop_privileges is for running stuff as $user in the live session, not in target. [12:30] aha, there's a regain priviliges [12:30] xnox: right, I understand [13:02] xnox: btw, do you have an idea if populating the xapian cache on the target system has been investigated? [13:05] shadeslayer: which xapian cache? i thought on ubuntu cds it does happen at the end when software-centre is configured. [13:05] - the apt-xapian cache that is. [13:05] yes, that's the one I'm talking about [13:05] xnox: is that generic or specific to the ubuntu CD> [13:05] or even the package [13:05] no idea, I just know that it happens =) [13:06] heh =) [13:09] shadeslayer: so apt-xapian-index is purged from target with a comment that it will be reindexed later. [13:09] yes I saw that [13:09] something along the lines of "Once the target systems apt is fully configured we will re index" [13:15] shadeslayer: are you just pondering or you did live-session install, then mounted /target and saw there is no apt-xapian inde? [13:15] xnox: installing to check if my guess is correct [13:15] xnox: atleast from the logs I don't think it's doing anything related to the apt xapian index [13:22] Anyone here have any ideas as to which team in Debian should I join? [13:26] the KDE one? [13:26] shadeslayer, except the KDE one? [13:27] not the GNOME one? [13:27] shadeslayer, I know nothing about GNOME:P [13:27] I bet apachelogger can enlighten you about it [13:28] shadeslayer, oh [13:28] http://wstaw.org/m/2013/08/27/plasma-desktopyY2304.png phonon5 playing video [13:28] apachelogger, shadeslayer say you can enlighten me on above topic^ [13:29] GNOME is awesome. [13:29] apachelogger, ...... [13:29] all the Enlightenment you need [13:29] apachelogger, ................. [13:29] apachelogger, bah :) [13:29] or perhaps you should have a look at your Windows [13:29] I heard they tried to remove middle click paste [13:30] to see an Apple falling from a tree, landing on a Bee [13:30] apachelogger, you are obviously answering the wrong question... [13:30] and then it was reverted [13:30] smartboyhw: yeah, because there was no question [13:30] apachelogger, um the question is Anyone here have any ideas as to which team in Debian should I join? [13:30] or there was one, and that one I cannot answer as I am not in any debian team [13:30] apachelogger, oh [13:30] https://git.gnome.org/browse/gnome-settings-daemon/commit/plugins/xsettings/gsd-xsettings-manager.c?id=696e04d41a485f84870a98c85b819979928b69e9 [13:30] shadeslayer, can you get somebody who can ACTUALLY answer?: [13:30] :P [13:30] so shadeslayer #fail [13:31] how about the debiam kde team , if there is one :) [13:31] what [13:31] BluesKaj: ofcourse there is one :P [13:32] I heared they all switched to fedora [13:32] yeah , no kidding ' [13:32] OK, so you want me all to switch to Debian and do packaging work there instead of here?:P [13:32] I see alot of fewrora users in #kde [13:32] oops fedora [13:33] * BluesKaj needs more coffee [13:33] so fedora must have problems on kde [13:34] BluesKaj, :O [13:35] or they have users [13:35] trolololo [13:36] ^^ [13:36] apachelogger, shadeslayer so, the Debian Qt/KDE team? [13:37] ¯\_(ツ)_/¯ [13:37] shadeslayer, OK, what the hell is that? [13:38] I don't know [13:38] shadeslayer, huh [13:38] THen why are you typing that? [13:38] It's a Japanese word + arms [13:39] yes, I know, I just used it as a substitute for IDK since it better conveys what I wanted to say [13:39] shadeslayer, .............................................................................................................................................................................................................................................................................. [13:44] it's an emoticon that's what it is [13:45] look s like , "I have no idea" :) [13:46] It looks like to me as \o/ [13:49] it looks like *shrug* [13:57] halp http://paste.kde.org/~shadeslayer/p3fda168e/ [13:57] causes crash : http://paste.ubuntu.com/6032727/ [13:57] line 958 ^^ [13:58] * Riddell tunes into http://summit.ubuntu.com/uds-1308/meeting/21887/intro-and-keynote/ [13:58] * smartboyhw too [13:58] Aug 27 13:50:17 kubuntu ubiquity: File "/usr/share/ubiquity/plugininstall.py", line 1674 [13:58] Aug 27 13:50:17 kubuntu ubiquity: universal_newlines=True).communicate()[0].strip('\n') [13:58] yes [13:58] shadeslayer: invalid syntax [13:58] why? [13:59] shadeslayer: missing comma [13:59] preexec_fn=misc.drop_privileges [13:59] there ^^ [13:59] %$^%*& [14:00] shadeslayer, CALM DOWN [14:00] my brain hurts [14:00] python is driving me crazy [14:01] * smartboyhw drives shadeslayer even more crazy [14:03] /o\ [14:04] * smartboyhw thinks shadeslayer should be sent to "Tsing Shan" (a.k.a. hospital for crazy people) [14:13] -.- [14:13] the cp command still doesn't work :/ [14:41] hmm [14:41] xnox: I don't think drop_priviliges is working [14:42] shadeslayer: define not work =) on kubuntu cd, it will correctly drop priviliges to the live-cd kubuntu user. [14:43] xnox: http://paste.kde.org/~shadeslayer/p5bae65c4/ [14:43] that's the code [14:44] in syslog I see "Building KDE cache from" [14:44] no path [14:44] check /var/log/installer/dm anything there? [14:45] not really no [14:46] http://paste.ubuntu.com/6032895 [14:47] xnox: ^^ [14:49] shadeslayer: at least on ubuntu, ubiquity bootted with debug-ubiquity parameter allows ctrl-alt-t to bring up the terminal with the same environment as the installer, for you to test what (if anything) is wrong with that call. [14:50] hmm [14:50] nope [14:50] nothing [14:50] xnox: you mean ubiquity --debug right? [14:51] ctrl+alt+t hasn't worked here for ages [14:51] * xnox has no idea about kde frontend. but it does work in ubuntu. [14:51] shadeslayer: no, debug-ubiquity as kernel boot arg, cause you want it in ubiquity-dm, not in the live session. [14:54] just a Q, are you folks aware that apport-kde is seg-faulting? [14:54] aha [14:55] xnox: but my code is only meant to be run in the live session [14:57] shadeslayer: what do you mean by live session? installer should do everything regardless if it was run from live session or from ubiquity-dm session. [14:58] shadeslayer: probably environment variables are missing, as we use pkexec and it clears environment, and when you drop priviliges back the environment will not be the same. [14:58] xnox: no no, this code should only be run when on the live session [14:58] shadeslayer: why does it need to be run anyway then? [14:58] * xnox doesn't understand, if you are generating caches, you should chroot into /target and regenerate caches there. [14:58] shadeslayer: do note that your live session != the target installed environment.... [14:59] xnox: yes I know, what I'm doing is copying caches from the live-session to the target [14:59] don't do that. [15:00] chroot into target and regenerate them. [15:00] shadeslayer: what does it gain you? [15:00] hmm, kind of difficult since there's no command to generate the cache on the target via command line that I know of [15:00] xnox: login time [15:00] or atleast it seems to me that login is faster if I copy the cache [15:00] *first login [15:00] shadeslayer: well the caches are generated somehow, if it's not exposed yet, make it exposed. [15:01] xnox: they're most likely generated programatically [15:01] xnox: for eg the plasma theme cache [15:01] or the icon cache [15:01] combined, the compromise about 90 MB's [15:01] benchmark it. I don't like something that only works via one method vs the other. [15:02] hm, I see [15:02] talk to kde upstream, maybe they can expose cache generation. [15:02] *nod* [15:04] shadeslayer: for example icon caches should be all up to date globally, as we preserve time stamps during livefs build. [15:05] not really, if I understand this correctly, KDE caches icons in /var/tmp/kdecache-USER/ [15:05] since you don't know the value of USER, the file is not there on the target system [15:18] DOH, Google doesn't allow me to join the vUDS hangouts since I'm < 18:( [15:19] lol? [15:20] shadeslayer, true [15:20] It is actually Google's policy [15:23] have patience smartboyhw . you'll be 18 someday :) [15:23] BluesKaj, what? That's 3 YEARS, for holy's sake [15:24] And that's 36 vUDS... [15:24] I sure hope not [15:24] * shadeslayer is still rooting for in person UDS to come back [15:25] like the saying goes , patience is a virtue [15:25] * smartboyhw wrote http://smartboyhwubuntu.wordpress.com/2013/08/27/google-please-let-me-join-vuds/ in fury [15:26] write vUDS-use-something-other-than-google instead [15:26] then Scott could join too [15:27] ^^ [15:27] yofel, wait, why Scott can't? [15:27] It's proprietary [15:27] same reason I try and actively avoid it [15:27] smartboyhw: he doesn't agree with google's TOS afaik [15:27] yofel, oh:) [15:27] and that as well ^^ [15:28] shadeslayer, yofel updated [15:28] At least pleading Ubuntu to change [15:28] I am unsure why they did not go with mumble [15:28] I don't think video is *that* important [15:30] well, video *is* nice [15:30] it does give it a bit more personal feeling [15:30] But, this TOS is :( [15:30] * smartboyhw has to re-consider whether he will be trying to join Google in the future [15:38] shadeslayer, yofel their answer: "NO SOLUTION TO THIS" [15:38] And they seem to stick to Video more [15:38] * yofel expected that [15:41] yofel, next thing you know from popey and IdleOne: Why not lie?:P [15:42] Next thing you know from slangasek and rbasak: Go and write one yourself -.- [15:58] xnox: fwiw the xapian-apt index is not created on kubuntu [15:58] there is no /target/var/lib/apt-xapian-index [15:58] but daily cron will create it. [15:59] xnox: do you think we should prepopulate it from the installer ? [16:00] shadeslayer: not sure. [16:00] so, where is this db used? [16:00] muon? [16:00] JontheEchidna: ^ ? [16:00] apol: ^^ [16:01] libqapt? [16:01] well, it's probably implemented in libqapt, but the question is whether muon uses it [16:01] it does [16:01] I think the issue was our language selector [16:01] for searching [16:01] that doesn't work without the index [16:02] ah hmm, the trello card gives no reasons why each one of those caches should be populated [16:02] just says that they should be [18:05] hiho [18:35] Hey Riddell, can you spare a moment? [18:38] When you have the chance Riddell, please ping infinity & slangasek over -release with the amount of alphas and betas Kubuntu want to participate in during the T cycle. It would help them with a bit of their discussion from a short session about schedule planning. [18:38] * skellat wanders off === claydoh_ is now known as claydoh [19:52] skellat: all of them I think [20:41] Riddell: Just please make sure to ping infinity & slangasek directly in -release with that. knome did so for us and I think I need to track down gilir perhaps yet too. === dantti_laptop|2 is now known as dantti_laptop