[00:09] TheMuso: hi [00:09] robert_ancell: Hey there. [00:11] TheMuso: I have a BZR question... I've made a branch lp:~robert-ancell/gdl/ubuntu which I was trying some packaging ideas. Now I don't want these changes anymore. How do I a) mege with lp:~ubuntu-desktop/gdl/ubuntu or b) delete this branch? Which is more correct? [00:11] robert_ancell: If you want to delete the branch, you want to go to http://launchpad.net/gdl/+branches and delete the branch from there. [00:13] TheMuso: thanks, I assume I can do that from the command-line too? [00:14] robert_ancell: Not that I know of. [00:14] Sure you can delete the branch locally, but I am not sure about deleting it from LP on the command line. [00:14] hi TheMuso [00:15] thanks for your awesome interview feedback [00:15] Hey rickspencer3. [00:15] rickspencer3: np [00:15] I would like to put it on the wiki as an example of good quality interview feedback [00:15] would that be ok with you? [00:15] rickspencer3: BTW, I have a spec that I need approved. Its in the review stage, but hasn't yet been looked at. Do I need to do anything special? [00:15] rickspencer3: Sure, go ahead./ [00:15] in terms of the spec, who is the approver? [00:16] rickspencer3: It was foundations/Robbie, but the approver was changed to you. let me double check that. [00:16] TheMuso: send me a link [00:16] the approver should probably be pitti [00:16] https://blueprints.edge.launchpad.net/ubuntu/+spec/desktop-karmic-gnome-speech-replacement [00:18] TheMuso: I set pitti as approver, and set it to "pending approval" which will trigger pitti to review it [00:18] rickspencer3: Thanks a lot. [00:18] he'll set it to approved, or make comments and send it back to "drafting" [00:18] n/p [00:18] Gotcha, I know the process. [00:38] good night ! === asac_ is now known as asac === rickspencer3 is now known as rickspencer3-afk [02:17] hallo, i can't log in after installing the latest ati driver [02:20] Rammler1983: What version of Ubuntu are you using? [02:21] 9.04 [02:21] Rammler1983: Are you talking about the fglrx proprietary driver, or the free driver? [02:22] after updating with update manager in ubuntu and i restart the laptop the system freezes [02:23] How does it freeze? DO you get a login screen? [02:23] no [02:23] after choosing [02:23] ubuntu 9.04 [02:24] i get the logo [02:24] Rammler1983: Ok. Do you still have the 2.6.28-7 kernel installed, and if so, can you boot that ok? [02:24] but it not sharp [02:24] no [02:24] i cant [02:25] i tried the alle the recovery options [02:25] and still not [02:26] Hmm. If you boot with recovery mode, do you get any error messages at the time of the freeze? [02:26] no [02:26] it just freezes [02:26] with out updating everything is fine [02:26] but i can't use the effects [02:29] Rammler1983: ok, so you can't boot at all. What kernel are you trying to boot? [02:30] the latest one [02:30] just a sec. [02:31] http://wiki.cchtml.com/index.php/Ubuntu_Gutsy_Installation_Guide [02:32] i used method 1 [02:33] 2.6.28-11-generic kernel [02:35] hrm ok [02:35] Sorry I'm not sure whats going on then. === Amaranth_ is now known as Amaranth === calc` is now known as calc === Amaranth_ is now known as Amaranth [07:28] good morning === tkamppeter_ is now known as tkamppeter [08:26] * robert_ancell has made lots of enemies with his bug triaging in compiz. Apparently LP stupidly is happily spamming everyone who is subscribed with duplicate reports (there are >200 of them) [09:03] Good morning [09:16] pitti: guten tag! [10:00] seb128: good morning. I am unsure about your comment on bug 345727. I can't find a change done outside debian/* and apparently the patch still gets applied in the latest ubuntu-package. If I look at the diff introducing the fix (http://launchpadlibrarian.net/24141917/seahorse-plugins_2.26.0-0ubuntu1_2.26.0-0ubuntu2.diff.gz) I see that it is properly made as a patch and in revision 6 of the bzr branch this change was applied. [10:00] Launchpad bug 345727 in seahorse-plugins "Seahorse-agent writes an empty ~/.gnupg/gpg.conf on first run, breaking email signing in KDE" [Medium,Triaged] https://launchpad.net/bugs/345727 [10:01] Ampelbein: the changelog entry is misleading then [10:01] " * libseahorse/seahorse-gpg-options.c: [10:02] - Copy gpg.conf skel instead of creating blank file [10:02] " [10:02] I've not looked into the source [10:02] seb128: right, got confused by that at first too. [10:02] not sure why they claim the change is not in the current version then [10:02] anyway that should be sent upstream in any case [10:05] seb128: I will do that. [10:08] seb128: and i can confirm that this works correctly in karmic: "mv ~/.gnupg/gpg.conf .." ; seahorse-agent produces a correctly copied skeleton gpg.conf [10:25] grrrr at karmic [10:25] so running away from my laptop for half an hour means I've a system I need to powerdown using the power button when coming back now [10:26] why is that? [10:26] because the box does nothing [10:26] blank screen doesn't react to vt changes, etc [10:27] hm, i have no problems at all, nvidia graphics. can leave for hours, still working after. [10:39] ugh, I hope Robert used a script to fix the duplicates and didn't do this by hand [10:39] which ones? [10:40] all the compiz ones [10:40] got some 100 mails [10:40] we discussed duplicates of duplicates at uds and listed a tool to do the magic for that if that's what you are talking about [11:28] what's the point of all this de- and re-duping of that compiz bug? [11:28] it will be flooding lots of peoples inboxes [11:35] Laney: not keeping a duplicate open I guess [11:36] launchpad doesn't let you mark duplicate a bug which has duplicate [11:36] I just wonder if the gain is worth the pain [11:37] it didn't cost a lot either [12:09] hey guys [12:09] dunno where to bing this topic up [12:10] but at UDS i talked to Bono [12:10] i mean jono [12:10] he had the idea of writing an application that teaches people who cant read and write to do so [12:38] Haha, Bono :) [12:41] sorry [12:41] :P === kalon33 is now known as Kalon33 === Kalon33 is now known as loup_74 === loup_74 is now known as Loup_74 === Loup_74 is now known as kalon33 [12:47] ^^ [12:53] * seb128_ kicks karmic [12:54] if somebody talked to me recently please say whatever you said again, karmic just crash my laptop when I'm away for a while === seb128_ is now known as seb128 [12:55] it's karmic, just logical it is not yet quite stable ^^ [12:55] but it doesn't work so bad here :) [13:13] hello crevette [13:13] hello [13:14] hey kalon33 [13:38] kenvandine: confirmed that there will be a gajim 0.12.3 soon. [13:38] Nafallo: cool [13:38] Nafallo: did you get a chance to see why notifications aren't called? [13:38] kenvandine: nope [13:38] kenvandine: I did however get a chance to package 0.12.2 :-D [13:38] :) [13:39] still broken though? [13:39] kenvandine: yes. [13:39] kenvandine: are you sure it ever worked? ;-) [13:39] yes... positive... i have screenshots somewhere :) [13:40] in it's current state, at least last i tried, it doesn't even show the libnotify notifications [13:40] ehrm. does for me. [13:40] if you mean notify-osd stuff. [13:40] yes [13:41] well then it should work with the indicator [13:41] that had stopped working for me [13:41] the method that does that wasn't even getting called [13:41] the only part that works with the indicator is that I can see gajim in the list :-) [13:41] i even tried in a guest session, so no local config [13:41] hmm. weird. [13:46] Nafallo: can you please upload latest gajim crack now? [13:47] i am still suffering bugs i haveing filed a report for ;) [13:47] havent [13:47] asac: I have it in my PPA. what's the bug? [13:47] asac: also... I have it in the wrong ppa. let me fix that :-) [13:48] Nafallo: error dialogs jumping in my face whenever you (in particular) send a message [13:48] Nafallo: why not push to karmic? [13:48] i dont want to add yet another PPA ;) [13:48] unless you have snapshots in there or something similar sexy [13:49] asac: oki. fair enough. [13:51] Nafallo: so why didnt you upload to karmic ;)? [13:52] asac: because the publisher haven't ran yet? [13:52] Nafallo: ah so you already uploaded to real archive? nice. [13:53] and ppa [14:09] pitti: what do we do if our specs don't fit well with having a list of tasks? [14:10] hey seb128 [14:10] seb128: as I wrote, just split them up as sensible [14:10] lut crevette_ [14:11] pitti: hum, I'm not sure how to split the GNOME3 one, that's virtually an hundred small tasks ... should I spend a day listing all those? [14:11] seb128: e. g. the banshee one could be "package new version", "measure memory requirements", "write MIRs", "change seeds" [14:11] seb128: no, of course not [14:11] ie gnome-vfs has 37 rdepends in main [14:12] libglade around 60 [14:12] etc [14:12] seb128: packaging gnome/zeitgeist are obvious ones [14:12] seb128: I guess you can't work on all 97 anyway? :-) [14:13] no, but I'm not sure what are realistic or not so I'm having difficulties putting a metric on that [14:13] that's why I'm asking ;-) [14:13] I was going to go on the best effort metric [14:13] seb128: yeah, this spec is very blurry [14:14] seb128: you could just split it up by "drop gnome-vfs" and "drop libglade", together with the two packaging ones [14:14] I can add "reduce gnome-vfs use" [14:14] but that will not be DONE 100% in karmic [14:14] seb128: just ignore them for now, I think, and just add zeitgeist/gshell [14:14] ok [14:14] thanks! [14:15] I will add those as goal with a comment saying having a precise metric will not be easy [14:15] well, since we don't have a precise goal, you can't have a precise metric :) [14:15] but let's not worry about this too much [14:16] having the "must have" facts in work items will suffice for the purpose [14:16] WI don't cover all of our work anyway [14:16] (bugs, mail, SRU, and whatnot) [14:50] rickspencer3-afk: ping [14:50] anyone familiar with the nautilus internals? [14:51] i need to know which part is responsible for the smtp dialogue [14:51] i guess seb128 ^^ [14:52] -nautilus+evolution [14:52] nautilus and smtp? [14:52] oops [14:52] ha [14:52] what dialog? [14:52] SMTP [14:52] smtp is what you use to send emails [14:52] between the client and the server [14:52] do you speak about the sent bar? the preference dialog? [14:52] the account details? [14:52] or the protocol impl? [14:53] rickspencer3-afk: I got my work items parsing/CSV generation script ready [14:53] anyway better to ask on #evolution on irc.gnome.org [14:53] rickspencer3-afk: but it seems that the trend line in burndown.py is utterly hardcoded? [14:53] they are responsive and probably know the code better, I'm hanging there too [14:53] seb128, asac: the protocol implementation, the headers sent by evolution are not good enough to use in my corporate environment === rickspencer3-afk is now known as rickspencer3 [14:54] pitti: yes, totally hard coded [14:54] rickspencer3: good morning [14:54] I couldn't figure out how to access x,y coords of elements in the chart [14:54] seems pychart just burns off the image, and then forgets about it [14:54] fta: the communication is in evolution-data-server [14:54] fta: yeah the word dialog is then confusing because this usually refers to UI. [14:54] rickspencer3: this is line.draw([(0,380),(725, 0)]) [14:54] ? [14:55] pitti: right [14:55] and line is actually an arrow :) [14:56] seb128, thanks. looking into that now. [15:05] ok, found it.. camel/providers/smtp/camel-smtp-transport.c 1520 lines, yeah! oh, joy! [15:14] Riddell, mvo: hello. I would like to push packagekit 0.4.8 to karmic. [15:15] Riddell, mvo: Unfortunately it contains an api break in the qt libs without a so name change [15:17] Riddell, so kpackagekit 0.4.1 has to be uploaded too === crevette_ is now known as crevette [15:56] is there a graphical frontend for x redirection, already? [15:57] i'm looking for something that allows to explore the gnome menu of a remote machine and select and start programs for execution with display on my local screen [15:58] another cool feature would be to start a remote desktop session with display in a xnest on the local machine [16:01] if nobody knows such a tool i create a blueprint [16:05] pitti: putting a blueprint into "review" sends it to the approver for review? [16:13] kenvandine: yes [16:17] what is the recommended python way to read a txt file on a http location? [16:17] urllib2?? [16:18] urllib2.urlopen('http://www.google.com').read() [16:19] I would say that, it is included in the default python install, but perhaps there is better? [16:19] that's pretty good [16:20] you need more if you want specific things with redirects, etags etc., but that'll work fine [16:20] I don't need to read a web page, just a txt file on a people location [16:21] urllib2.urlopen('http://www.google.com/anyfile.txt').read() would work too I guess seb128 [16:21] I was pondering copying it locally and checking the timestamp before downloading again [16:25] pitti: do you happen to know why bip seems to prepend + to every message? [16:26] calc: I'm not using bip, I don't know [16:26] pitti: ok [16:26] seb128: I was typing that 5 mins ago, but I got disconnected [16:27] for line-by-line this is most convenient: [16:27] for line in urllib.urlopen('http://foo'): [16:27] print line [16:27] pitti: thanks [16:29] pitti: new release of telepathy-glib do out and uploaded to debian within the hour [16:29] :) [16:29] \o/ [16:29] with this bug fixed of course [16:29] pitti: tp upstream is great :) [16:31] seb128: any word on tomboy 14.2 for jaunty? upstream fixed jaunty bugs and they are keen on getting it out to people [16:33] jcastro: go for it? [16:33] jcastro: I'm not working on tomboy but I'm fine having a new version sru [16:34] seb128: is anyone available to work on it? it kind of sucks when an upstream fixes our bugs and they don't end up back in the distro. (understand that everyone is busy) [16:34] dunno but I'm sure we can find somebody easily next week [16:34] k [16:34] it's just almost 6pm on a friday there [16:35] * jcastro nods [16:35] I didn't mean right now. :D [16:35] I will find somebody to do the update and do the sponsoring next week [16:35] so we already have an updated and a reviewed in the loop and we just need sru verification [16:35] which pedro can easily do ;-) [16:37] yeah it can just get frustrating for an upstream when they ship fixes and there's a lag there. [16:37] and then the users are bugging them about bugs they fixed already [16:39] well we don't do stable version updates in stable for non lts versions [16:40] if nobody comes to bug us about the update that's it [16:40] jcastro: fine if someone from community wants to do the packaging and put it into jaunty-backports or so [16:45] seb128: gnome-python-destkop had a bugbuddy binding without depending on bugbuddy package (as it's in universe). With the package split (there is one package for gnome-python-bugbuddy), do I achieve the same thing? Having this package with the binding, but not depending to the universe bugbuddy package? [16:46] didrocks: I think it'd make sense for gnome-python-bugbuddy to depend on bugbuddy [16:46] didrocks: yes [16:46] seb128: don't we cherry pick fixes from stable gnome point releases? [16:46] jcastro: only if they match our SRU criteria and someone requests them [16:46] pitti: but bugbuddy is in universe, gnome-python-desktop in main [16:46] ok [16:46] jcastro: we do when we have issues which seem worth doing that, not sure if there is one on tomboy right now [16:46] didrocks: right, but if there's a split out g-p-bugbuddy, this particular package could be in universe; I suppose we don't need it in main? [16:47] pitti: packages that "import bugbuddy" will likely need patching need ... [16:47] pitti: it's easy to install the binding and make it handle the case where bug-buddy is not installed [16:47] seb128: for the g-p-d -> g-p-b dependency change? [16:47] if you want to move the binary to universe [16:48] g-p-bugbuddy that is [16:48] pitti: is it possible in Ubuntu policy? Having a source package providing bin packages in both main and universe? [16:48] I think app importing it should depend on it [16:48] didrocks: yes [16:48] ok, so, I can remove the patch, adding the b-d and put the splitted package in universe [16:48] well, will pychess start if pybugbuddy is not installed? [16:49] or will it bail on the import bugbuddy which is not in a try statement? [16:49] seb128: I will try this === rickspencer3 is now known as rickspencer3-afk === sabdfl1 is now known as sabdfl [17:26] pitti, ping [17:26] hi hggdh === rickspencer3-afk is now known as rickspencer3 [17:27] hi pitti. I am writing an apport hook for Evolution; the main objective is to sanitise the BTs so that private data is masked. So... is there a way of reconstructing the crash file from a bug? [17:27] this will allow me to locally run the hook under apport-retrace, and check it [17:33] hggdh: there's no CLI for it, but it's not particularly hard [17:33] hang on [17:34] hggdh: but if you just want to test a hook, why bother with uploading it first at all? [17:34] you could just use the details expander to see what it would send, and then cancel? [17:35] hggdh: problem is, there is a function to download a report, but it doesn't grab all attachments; just the ones needed for retracing and duplication [17:35] but for your purpose you probably want all attachments [17:40] rickspencer3: fixed the "work items" syntax on https://blueprints.edge.launchpad.net/ubuntu/+spec/desktop-karmic-bug-workflow FYI [17:40] pitti, yes. The problem is that I have to test it against bugs that do have the fields I need to mask out; on the other hand, I do not really need all attachments, just the BT and StackBT, so the function you have might be a good start point [17:40] if, of course, you do not mind sharing it [17:41] pitti: oops, I just stomped on your changes, I think [17:41] hggdh: right, the curent function does fetch those [17:41] hggdh: look at sr/lib/python2.6/dist-packages/apport/crashdb_impl/launchpad.py, APPORT_FILES [17:42] hggdh: are those enough for you? [17:42] looking [17:43] http://people.ubuntu.com/~seb128/versions.html [17:43] ^ first version, I've to make that clear and nicer now [17:43] seb128: nice! [17:44] but that list debian, ubuntu, upstream versions [17:44] pitti, yes, thank you [17:44] and color lines for merges and updates required [17:44] hggdh: python -c 'from apport.crashdb import get_crashdb; c=get_crashdb(None); r=c.download(338478); r.write(open("/tmp/report.crash", "w"))' [17:44] hggdh: sorry, I know it's not nice, but perhaps throw it in a small shell script and call it apport-download? [17:44] I guess I should add that to the apport package itself [17:44] :-) [17:45] If what I am trying to do succeeds, we might consider adding it to the hooks-utils [17:45] seb128: what's the color key right now? [17:45] red = new upstream version [17:45] orange = newer upstream and in debian [17:46] yellow = current upstream version but newer revision in debian [17:46] great [17:46] so yellow = to merge [17:46] red = to upgrade [17:46] orange is to merge and upgrade [17:46] rickspencer3: np [17:47] rickspencer3: fixed again [17:47] ugh, today I got 7 new specs to review [17:48] I see, doesn't like my moin :) [17:48] pitti: I'm not much with the regex, but doesn;t the i in "items" need to be lower? [17:49] rickspencer3: //i -> case independent [17:49] rickspencer3: wOrk iTEMs will work as well [17:49] rickspencer3: but *shrug* I can just extend it to digest = .. = [17:50] don't bother [17:51] thanks though [17:52] pitti: in terms of quickly, .quickly file is bad? no hidden files? [17:53] rickspencer3: why would you want hidden files? that's confusing [17:53] that's fair feedback [17:53] and I'm not even sure whether Debian Policy allows shipping hidden files in packages [17:53] what should w call it? [17:53] this would *create* a hidden file, not ship one [17:53] quickly.conf [17:53] ? [17:53] it would make "magic" [17:54] rickspencer3: what does that file do? [17:54] this would be a per *instantiated* file [17:54] it is a container for information about a particular project [17:54] so if I do: [17:54] $quickly new ubuntu-project birds [17:54] this would create a project called "birds" [17:55] in that directory called "birds" is a file that contains information that quickly needs, most importantly the template that created it [17:55] so it would have: [17:55] template = ubuntu-project [17:55] but you could imagine extended it [17:55] currently this file is .quickly [17:56] but perhaps naming it something more descriptive and making it visible makes sense [17:56] didrocks: ^^^^ [17:56] proj [17:56] proj.conf [17:56] quickly.proj [17:56] birds.proj [17:56] ??? [17:56] birds.quickly [17:58] rickspencer3: I wouldn't make the name project specific [17:58] rickspencer3: yes ? :) [17:58] quickly.conf is easy to find, and similarly static than "Makefile" [17:58] why not (out of curiousity?) [17:58] (IMHO) [17:58] (backlogging) [17:58] didrocks: discussing the naming of ".quickly" [17:59] rickspencer3: well, you could parse *.conf and decide which ones are for quickly, but what happens if you have several? [17:59] does quickly.conf not suggest that you are configuring quickly itself, rather than the app you are creating? [17:59] well, "Makefile" doesn't configure make itself, just how to use make [17:59] if it's birds.quickly, it would be quite easy to find, and the meaning of the file may be quite clear to users [17:59] is '.quickly' a desktop file? [17:59] pitti: the new spec revision is more verbose about this .quickly file [17:59] right, but this is a different user (not that I'm disagreeing) [18:00] rickspencer3: does it make sense to have several? [18:00] pitti: not really [18:00] I guess RoR names all their files the same, not per project [18:00] rickspencer3: that's what I thought, too [18:01] so, .quickly file seems good, no? :) [18:01] so why not call it "quickly.conf" then? then you can't have two [18:01] didrocks: I think making it hidden is not optimal [18:01] pitti: I think this is an internal quickly function. The user doesn't have to change it [18:01] rickspencer3: see above :) [18:02] my purpose is to manipulate it by quickly commands [18:02] so, no user direct interaction. (like .bzr folder in bzr branch) [18:02] well, it's your baby, but personally I find hidden files very confusing [18:02] it's way too easy to miss them when producing tarballs, backups, packages, etc. [18:02] pitti: may we leave it .quickly until we get some user feedback? [18:02] rickspencer3: i agree with pitti [18:02] I think didrocks gets to decide [18:02] :) [18:03] didrocks: you don't tar up .bzr/ and mail it to a friend [18:03] you should rather quickly deb [18:03] :) [18:03] no, I just wait for "professional feedback", so we can make it quickly.conf [18:03] pitti: will you approve the spec if I leave this an open issue? [18:03] I'm juste afraid quickly user won't understand it and remove it [18:04] just* [18:04] rickspencer3: sure, but I wanted to mention it and point out the potential traps [18:04] yes [18:04] and be aware that it's hard to change afterwards [18:04] I don't think we need to block on it now though [18:04] and it'd be the first build system which relies on hidden files [18:05] I'm tending to agree with pitti at this point, and am wondering whether having a simple to edit configuration file might be a good thing [18:05] though I still like birds.quickly [18:05] pitti: I don't see really any difference with .quickly and .bzr-builddeb === Amaranth_ is now known as Amaranth [18:07] didrocks: *shrug* okay; when I wrote that, I didn't know what .quickly was (that was another point that I mentioned which needed clarification) [18:07] if the user should never touch it, fine for me [18:07] pitti: is the new spec version more clear? [18:07] didrocks: didn't have a look at the newer version (-EOVERLOAD) [18:08] pitti: ok, that's why I didn't understand :) [18:08] didrocks: I am extracting work items and putting them on the whiteboard so we can track during Karmic [18:08] could you review those when I am done? [18:08] so, having an hidden file makes sense? [18:09] rickspencer3: sure, but I hadn't the time to follow the meeting about work items handling, are they just opened bug? [18:09] didrocks: no, we're tracking on the whitebaord [18:09] it will be rather obvious I think [18:09] check it out in like 1 minute [18:09] rickspencer3: ok, just drop me the link and I will review them :) [18:10] (still fighting with gnome-python-desktop stuff ^^) [18:11] didrocks: https://blueprints.edge.launchpad.net/ubuntu/+spec/desktop-karmic-quickly [18:11]