[04:29] * RAOF heads off to the physio. [05:07] jasoncwarner_, when do you leave for desktop summit? [05:19] good morning [05:20] didrocks: Good morning, I just sent you an email about what the problem was [05:26] hey bschaefer! didn't receive it yet [05:26] where did you send it to?? [05:26] ah gmail :) [05:26] to your gmail account i think [05:26] yeah to many to choose from on launchpad haha [05:27] heh ;) [05:27] bschaefer: so ok, I'm not sure about it's generated from this .lemony to .cc [05:27] (again another preprocessor…) [05:28] so in the trunk version it compiles the lemon.c then uses that to generate the queryparser_internal.cc [05:28] bschaefer: a patch patching everything needed without any regeneration (as the tarball which generates everything) would be nice! [05:28] bschaefer: patch for our package should be in the -p1 format btw [05:29] for applying the patch or making it with diff? [05:30] (new to making patchs) but I thought mine was in the -p1 for applying the patch since you would go into xapian and type: [05:30] bschaefer: the patch (which is in debian/patches) for you cjk patch in ubuntu/oneiric branch should be in -p1 format, (does it make sense?) [05:30] patch -p1 < ../cjk.patch [05:31] didrocks: yeah that makes sense ill make sure to do that for the new one [05:34] bschaefer: thanks a lot! :-) [05:34] kenvandine: I believe he flew out this morning. [05:34] bschaefer: so do not hesitate to ping me once you have the one patching the generated files as well [05:34] didrocks: will do [05:35] bschaefer: interesting for this .lemony files, I'll add that to my doc of "ok, I patch this file, but take care to generate those" [05:35] bschaefer: idealy, the build system will pick it and regenerate the needed file, but oh well :-) [05:35] TheMuso, ah... i figured it was soon :) [05:36] didrocks: yeah that is what it should do, and why I built it like that [05:37] didrocks: I had it in the queryparser_internal.cc to start out with but if you look at my branch there is no queryparser_internal.cc like in oneiric [05:38] didrocks: (if you want to look) http://bazaar.launchpad.net/~brandontschaefer/xapian/cjk-support-patch/files/head:/xapian-core/queryparser/ [05:39] bschaefer: indeed, however, I prefer not remove what's in the tarball already. The preferred way is to patch both file (the "master" one and the generated one) [05:40] didrocks: yes! Sounds good, I should have looked at the one in oneiric to base my patch off, but it shouldn't be much work [05:41] bschaefer: no worry, thanks for looking at it! :) [05:43] didrocks: no problem, was a lot more worried when you said the testing failed haha [05:44] bschaefer: so did I! ;) [05:44] was* [05:44] didrocks: but I should have it to you by tomorrow, I need head off for the night, but will finish as soon as i can [05:45] bschaefer: sounds excellent, thanks! :) [05:45] didrocks: Yep, have a good day :) [05:46] bschaefer: have a good night :) [06:27] didrocks: Good morning! [06:29] hey RAOF, how are you? [06:30] I'm pretty good. My back's feeling even better after a little physio. [06:30] RAOF: nice to hear. Try to get some quiet time on week-end to not hurt your back too much! [06:31] And some boxing on Saturday morning! [06:32] Oh, and I've thought of a reason why wayland-scanner would want to be shipped in libwayland-dev: compositors are expected to extend the protocol and will probably want to do so via the same XML specification mechanism. [06:32] heh, you should try some wreslting, a bad move can maybe fix you! [06:32] RAOF: oh, that makes totally sense then ;) [06:33] wrestling* [06:35] I'll see about addressing those MIR concerns now, pull a new upstream snapshot with some copyright fixes, and multiarch the sucker. I'd like to ensure that all the mesa libs and their transitive dependencies are installable on i386 ;) [06:39] RAOF: excellent, do you think that can be done for your tomorrow? I'm travelling for desktop summit but will surely be online for few hours before [06:39] RAOF: not sure how urgent this is ;) [06:39] RAOF: as long as you comment on the MIR bug, I can process that during the week-end with the promotion as well [06:40] Ok. I'd like wayland to be ready once we're unfrozen from A3, so I can upload mesa 7.11 without futzing around. [06:40] Other than that, it's not particularly urgent. === tkamppeter_ is now known as tkamppeter [06:47] RAOF: ok, keep me updated please :) [06:49] tkamppeter: Good day! colord's now in Debian new, and should shortly be syncable. Alternatively, I could just upload it directly. [06:54] RAOF, so colord arrived in time. Great. Thank you. [06:55] RAOF, only problem is that there is near no time left to rebuild all dependent packages, to MIR all the CM stuff in Universe, ... What needs to be done before FF and what can be done after? [06:58] tkamppeter: How many rdepends will there be for colord? cups, gnome-color-manager, … ? [06:59] I guess that strictly speaking, everything should be done before FF, although the uploads could be done before the MIR and just block on it. [07:00] It shouldn't be too hard to get a FFe for the colour stuff. [07:02] didrocks, hi [07:02] hey tkamppeter [07:02] RAOF, so I will try ... [07:03] RAOF: not gnome-color-manager, colord can be in maybe, but it would need a MIR report, which isn't the case [07:03] didrocks, it seema that we will succeed with color management in Oneiric in the last minute. Can you process the MIR in bug 788145? Thanks. [07:03] Launchpad bug 788145 in gnome-color-manager "[MIR] gnome-color-manager" [Medium,Incomplete] https://launchpad.net/bugs/788145 [07:03] tkamppeter: not sure about gnome-color-manger, isn't that deprecated for GNOME3 and in g-c-c? [07:03] tkamppeter: and colord needs a separate MIR [07:04] and I'm not found of adding 2MB on an already oversized CD [07:04] good morning [07:04] didrocks, did not now that gnome-color-manager is deprecated. If so, its successor must be taken care of to get in. [07:04] apart if you find the room for it :) [07:04] hey chrisccoulson [07:04] hi didrocks [07:04] tkamppeter: as you are looking at this, can you please ensure that? [07:05] tkamppeter: and I think, there is no room for g-c-m anyway, but colord can maybe be in if we have a MIR for it [07:06] tkamppeter: I thought gnome-color-manager now depended on colord? How has it been building without colord? [07:06] g-c-m is still depending on gconf as well [07:07] and have some recommends in universe [07:07] RAOF, I did not know that gnome-color-manager needs colord. It seems not, as Debian/Ubuntu had g-c-m but never colord. [07:07] didrocks, https://code.launchpad.net/~chrisccoulson/messagingmenu-extension/support-for-inbox-only/+merge/70388 ;) [07:07] 2.32.0-0ubuntu1 [07:07] -> I guess it's definitively not GNOME 3 and in g-c-c now :) [07:08] * didrocks would like some research to be done before spending time on MIRs ;) [07:08] chrisccoulson: \o/ you rock! thanks ;) [07:23] chrisccoulson: btw, no startup cache issue for now :) [07:23] didrocks, that's good. hopefully it's fixed now :) [07:24] well, "fixed". the fix is actually to work around a bug in the JS engine ;) [07:24] argh ;) [07:24] well, as long as it works! [07:26] RAOF: You should be able to get FFe you need, but otherwise you could upload with an Ubuntu upstream version 0.1.11~bz2-0ubuntu1 and then sync when LP catches up [07:26] otherwise/if you want to get it in earlier [07:27] Laney: I'm still under the impression that Launchpad will handle .orig.tar.xz fine. In fact, I might just upload colord to check this! [07:27] lifeless could be misinformed I suppose, check with wgrant if you like — he knows all :-) [07:27] or try a PPA [07:28] Well, wgrant filed and marked as fixed the “support .orig.tar.xz” bug, so… [07:28] as far as I read the comments, that means the code has landed but it still needs testingn to be deployed [07:28] not that I know how they work [07:30] It looks like it got tagged qa-ok, and then marked as fix released. [07:34] Yeah. The question is 'is that the same as "in production"?' [07:37] * Laney is uploading to a PPA [07:41] Too slow, already done it. [07:42] you beaut [07:43] Hah. Except that I suck, because PPAs don't build for “unstable”, surprisingly enough. [07:44] RAOF: accepted! [07:44] hmm, I may be thinking of cjwatsons other patch :( [07:45] Some launchpad technical architect *you* turn out to be :P [07:45] Doesn't even know every edge case that gets fixed! [07:45] :) [07:46] there's .data.tar.xz which is something a bit different :-) [07:46] Oh, right. Subtly different to lzma, which mesa's used forever.\ [07:47] yus [08:03] RAOF: I think you're confusing me with someone that cares :P [08:05] hey seb128! it appears that gnome-settings-daemon crashes during the upgrade from natty->oneiric. at least that is what the auto-upgrade-tester says :) is that a known issue? [08:05] hey mvo, hard to say without seeing a stacktrace [08:06] mvo, but I wouldn't be surprised if lot of things were going weird with the gconf to gsettings upgrades under their feet [08:06] mvo, like the gconf key and schemas they use are cleaned but the running instance still use gconf [08:07] seb128: ok, I will see that I can get it uploaded to LP so that the retracer can run [08:07] mvo, thanks [08:08] seb128: its a bit difficult as this was happening on a different machine and my workstations nvidia card died today. so may take until tomorrow, I was mostly wondering if its known or not [08:08] (actually the nvidia card maybe fine and its just compiz/nouveau thats gone insane ;) [08:08] mvo, well, we get a collection of segfault so it might be one we already received [08:08] mvo, but from the description I can't say it's known [08:08] like the upgrade is not know to break g-s-d [08:14] mvo: try to put it in the oven! :-) [08:15] didrocks: right, I read about that, do you still have the link? [08:15] its dead anyway, I could as well try that [08:16] mvo: it was a video on youtube for my specific hardware (to show how to remove it from my laptop) not sure it will apply to you [08:17] last time I tried, I put it in the oven at 200°C, for 20 minutes (without pre-heating) and opening the door at then to cool it and not have some thermal shock [08:17] you just have to think about removing the thermal paste [08:19] ok [08:20] ah, DIY reflow :) [08:35] mvo, btw is that a design decision that update-manager stopped showing download and upgrade time estimation [08:52] ok, thunderbird is weird [08:52] it didn't know about ubuntu.com but got all the canonical.com server settings right it seems [08:52] that's nice [08:53] but now it lists only half my boxes [08:53] it seems to be still downloading though but it doesn't make it obvious it does [08:53] it's a bit weird, evolution get at list the boxes listed from start [08:53] then it fills the content and counts, etc [08:54] seems like tb does it the other way [08:59] hey seb128, how are you? [08:59] hey chrisccoulson, I'm fine thanks [08:59] chrisccoulson, how are you? [08:59] seb128, tired ;) [08:59] trying to get tb to work [08:59] my daughter didn't sleep too well again last night [08:59] did you work all night again? or did you daugher refused to sleep? or both? [08:59] :-( [09:00] i did some work last night too ;) [09:00] i implemented inbox-only mode in the message menu for kenvandine and didrocks ;) [09:01] nice [09:01] didrocks, how is the unitydialog thing called in ccsm? [09:01] I need to turn it off, it breaks the thunderbird multiple identity dialog [09:01] seb128: unitydialogs, but it shouldn't be starting with latest unity [09:02] like you only have part of the dialog and no way to access the buttons [09:02] seb128: the xml file isn't shipped [09:02] and I don't have them there [09:02] it's in use for me :-( [09:02] hum, how come? :/ [09:02] but not showing anywhere in ccsm [09:02] waow [09:02] is the xml used at runtime? or only for the config? [09:02] normally, removing the xml removed the detection of it [09:02] but the .so is still installed [09:02] like if my gconf says to load it will it load the .so? [09:02] I bet that's the case [09:02] let me tweak gconf [09:02] yeah, but when I tried that (and tested there), removing the xml was enough [09:03] and I don't have them anymore [09:03] not for me [09:03] so yeah, we can remove safly the iso [09:03] safely* [09:03] so* [09:03] /usr/lib/compiz/libunitydialog.so [09:04] didrocks, that's fine, I tweaked the gconf config [09:04] seb128: well, I'll upload after alpha3 with removing it [09:04] thanks [09:04] yw [09:05] chrisccoulson, tb sucks [09:05] lol [09:05] :( [09:05] it doesn't list half my imap folders! [09:05] seb128: I had to add them one by one [09:05] including the launchpad box and the ubuntu-desktop and ubuntu-devel ones [09:06] (and ask for checking all updates one by one by right-clicking as it didn't check them there) [09:06] yeah, I'm not going to do that [09:06] yeah, you can edit folder subscriptions, but i'm not sure how it decides which folders to subscribe to by default [09:06] normal users wouldn't do that either [09:06] chrisccoulson, what about "all"? ;-) [09:07] like if I bothered filtering emails in boxes that's because I use those [09:07] boxes->folders [09:07] seb128, it seems to get it right for gmail (ie, it hides the Bin, All Mail and Starred virtual folders by default) [09:08] it decided to not list desktop-devel-list either [09:08] i'm not sure how it decides though ;) [09:08] like it's listing all the old boxes I don't care about [09:08] like dapper-changes [09:08] lucid-changes [09:08] seb128, can you check to make sure you're not subscribed to those boxes? [09:08] but none of the natty, oneiric, desktop lists, launchpad etc [09:09] chrisccoulson, is "subscribed" something stored on the server side? [09:09] how do I check that? [09:09] seb128, if you right click on the account name in the folder pane (on the left), and click "Subscribe" in the context menu, you will get a dialog [09:09] not sure how that translates in french ;) [09:10] ok [09:10] it's not in french [09:10] I've apt-get install thunderbird and that doesn't bring any translation with it ;-) [09:10] do you not have the french language pack installed? (thunderbird-locale-fr) [09:10] no [09:10] will do that in a bit ;-) [09:10] it's not, I poked pitti about it and he told me he would fix that later [09:10] seb128, yeah, we discussed that problem at UDS (automatically pulling in new language packs when an app requires them) [09:11] didrocks, not sure how we can "fix" that [09:11] we can do depends on binary-$(locale) [09:11] i.e dynamic depends [09:11] seb128: as th is in the default, shouldn't it be recommended by the langpack? [09:11] can't [09:11] (isn't the same issue than firefox?) [09:11] didrocks, yeah, same issue for firefox [09:11] and same issue for installing KDE apps on GNOME [09:12] didrocks, well I was using evolution and we don't change the email client on upgrade I think [09:12] chrisccoulson, so ok, I'm only subscribed to those folders I don't care about according to thunderbird [09:12] seb128, subscribe to the ones you care about for now, and i'll try and find out how tbird decides which ones to subscribe too ;) [09:12] which account is this btw? [09:13] chrisccoulson, my canonical one [09:13] imap [09:13] well evolution seems to list the same subscription [09:13] so maybe it's a server side issues and I checked those by then [09:13] but evo default to "show everything" and not "show only subscribed" [09:13] so I never noticed ;-) [09:14] can I tell tb to show everything? [09:14] hum [09:14] seb128, not sure there's a way to do that. i guess that could be easy to add though [09:15] seb128, ok, it's something on the server [09:15] not sure what yet though ;) [09:15] chrisccoulson, yeah, as said evo has the same ones checked [09:15] so they agree on that [09:16] chrisccoulson, let's say I had a broken config and not blame it on tb [09:16] having a "subscribe to all" button on the subscribe dialog would be nice though [09:16] seb128, it missed some of my folders too. i wonder if we need to tweak something on our server [09:16] in any case, gmail seemed to get it right [09:16] ok, now I've all my box [09:17] but no email count indicated [09:17] clicking on "get mail" does nothing [09:17] but I've unread emails in evolution [09:17] seb128, chrisccoulson can I interrupt real quick to ask something weird about desktop summit? [09:17] rickspencer3, sure [09:17] i'm not going to the desktop summit :( [09:18] my wife had arranged for an apartment in Berlin next week [09:18] but, she's not going [09:18] she can cancel it, but I thought maybe someone might want to use it [09:18] so, just throwing it out there [09:18] she'll cancel it tonight unless someone expresses interest [09:18] rickspencer3, thanks for the info [09:19] well I think those going from Canonical got room booked in an hotel close from the venue and together [09:19] yeah [09:19] I like better staying at the same place as everybody [09:19] seb128, to tell tb to show everything go to "account settings / server setting / advanced" and untick "show only subscribed folder" [09:19] the offer is open to more than just Canonical folks [09:19] I was just putting it out there [09:19] rickspencer3, but thanks for the offer ;-) [09:19] maybe that should be the default, if that's the default in evo [09:20] mvo, ^ you were pondering maybe commit to summit, in case that's something that could interest you [09:20] jibel, thanks [09:20] chrisccoulson, ok, so tb insist I've no email when I've some [09:20] chrisccoulson, clicking on get mail does nothing [09:20] like doesn't update the counts or seems to check for emails [09:21] do I need to click on each box to see if they got emails or something? [09:21] seb128: in folders? [09:21] didrocks, yes [09:21] ahah, I'm not the only one \o/ [09:21] remembers my discussion some weeks ago? [09:21] seb128, yeah, it lazy loads some folders by default [09:21] didrocks, not really, I didn't follow the details of your tb issues ;-) [09:22] chrisccoulson, it means I need to click manually 25 folders every time I want to check which one got new messages? [09:22] that seems suboptimal :-( [09:22] seb128, do you do server-side filtering too? [09:22] basically, you have to right click on each folder, property, and check the option "check in this folder" or something like that [09:22] doesn't tb handle imap idle? [09:22] chrisccoulson, yes [09:23] seb128: remember the email I missed on ubuntu-devel? that was the caused :p [09:23] seb128, yeah, that's why i don't see this, as i do all my filtering in tbird ;) [09:23] hum it has "use IDLE if the server support it" checked [09:23] seb128, to force checking, try clicking on the arrow next to 'Get Mail' to open the menu and select 'Get All new messages' [09:23] and the server supports it, it works in evo [09:23] jibel, that does nothing as well [09:24] seb128, didrocks: Oops, looks like when fixing rhythmbox's window not closing on gnome-shell, now it won't show when clicking notification :( [09:24] like no visible actions, no spinner, no update [09:24] dunno if that's broken on unit too [09:24] xclaesse, if you want to fix it updated patches are welcome ;-) [09:25] no even a message in the status bar saying 'Connect', 'Check for new messages', ... ? [09:25] jibel, no, nothing [09:25] seb128, I'll let you know if I find why :) [09:25] xclaesse, thanks [09:26] now I understand why chrisccoulson never see bug emails [09:26] tb just doesn't show new emails :p [09:26] seb128, i see all mine because i do local filtering ;) [09:26] seb128, by default, checking is disabled for !Inbox folders until you view them [09:26] i wonder why that's the default though [09:26] hum [09:27] but "get all new messages" should work at list [09:27] even if "get mail" doesn't [09:27] yeah, that sounds like a bug [09:27] do you want me to keep it the broken state? [09:27] seb128, no, feel free to set it up how you want :) [09:28] well, do you need debug infos to fix that issue? [09:28] I can keep using evolution for a bit [09:28] seb128, i don't think so [09:28] ok [09:30] do users really prefer having emails not sorted by discussions by default? [09:31] ok, I've got my tb mostly set up, let's see how it goes at use ;-) [09:31] I'm sure chrisccoulson will soon with I stayed on evo :p [09:32] with->wih [09:32] wish [09:32] can't type!!! ;-) [09:33] bah tb is not better than evo as displaying an email preview while updating indexes [09:33] it's over a minute and it's still not previewing it [09:34] it's ok once it's done the initial sync and index ;) [09:34] the tooltip on boxes with unread messages is nice [09:34] it tells you what you got there without having to switch ;-) [09:35] oh, i've never noticed that before [09:37] when do we unfreeze? i've got a firefox beta to upload [09:37] we're out of date ;) [09:38] chrisccoulson, it's a soft freeze ;-) [09:38] seb128, yeah, but i don't want to annoy everyone in #ubuntu-release ;) [09:39] well skaet is still sleeping, just go for it ;-) [09:39] joke aside small uploads are probably fine, maybe wait later today for firefox [09:39] chrisccoulson, you can fix g-s-d though if you want ;-) [09:39] yeah, firefox definitely isn't a small upload ;) [09:39] i could :) [09:40] you should! [09:40] you own me a favor now that I'm using your email client :p [09:40] (wonder if that will work :p) [09:41] lol [09:41] seb128, i think the deal was that if you didn't find any bugs, i would owe you a beer ;) [09:42] doh, I knew I should just have tweaked things locally without saying anything :p [09:42] chrisccoulson, well I think the only real bug there is the "get all new messages" doing nothing by default [09:43] heh, DOM inspector is great. i can find exactly where in the code that button is, so i can view exactly what it does when you press it ;) [09:44] I know what it does [09:44] "nothing" [09:44] I tried it ;-) [09:45] lol [09:46] this is fun! when selecting "try ubuntu" I get a desktop but without reboot/shutdown option [09:46] mvo, why would you ever want to stop running Ubuntu? ;-) [09:47] espcially when it runs from a speedy medium like a CDROM … [09:47] mvo, "known bug" I think ;-) [09:47] ok, fair enough [09:47] not a big deal [09:51] chrisccoulson, now the indicator and launcher counts disagree with the tb ui [09:51] like the count is 3 but the boxes the indicator list with unread messages have no unread messages [09:53] hmmm :/ [09:53] i'll need to think about why that could happen ;) [09:54] well, I used evolution to read my emails while tb is indexing [09:54] so maybe 2 clients accessing the server confused it [09:55] morning everyone btw :) [09:57] lut huats [10:02] hello seb128 [10:05] stupid question [10:05] how do I "clean up deleted messages" in tb? [10:06] seb128: empty the trash [10:06] tjaalton, doesn't work [10:07] yeah wait.. I'm using both tb and alpine, I'll check how it's done.. [10:07] but I don't move deleted messages to it, I selected 'just mark as deleted" option in the account options [10:07] i.e I like keeping those crossed in the box when they arrived until I clean up the list [10:08] it makes easier to undo the "ups, I didn't want to delete this one" [10:10] right, I use the default [10:11] ok, google help [10:11] "compact folder" does it [10:11] very intuitive ;-) [10:12] yeah, very :) [10:12] I'm wondering if that did other things as well [10:15] bah that's doing other things [10:15] it's like going over my folders and compating those now, I hope it will not screw things [10:15] ouch [10:16] it just displayed an error saying it couldn't compact a folder because it was being indexed [10:17] i guess it goes through all the folders of the account looking for deleted messages [10:18] well I would feel better if the label was "cleaning deleted messages" [10:18] if that's what it's doing [10:18] because I don't know what is hiding behind the current wording ;-) [10:18] indexing is something else, I've always disabled it since tb3 came out [10:18] http://kb.mozillazine.org/Thunderbird_:_Tips_:_Compacting_Folders [10:18] that would suggest it only purges deleted messages [10:23] is it just me or is it odd that "keyboard" in gnome has no option to change the keyboard options (like make caps-lock a control). its under "regions and languages" now [10:24] mvo, they plan to drop it in 3.2 [10:26] mvo, gnome bug #654617 [10:26] Gnome bug 654617 in Region & Language "Remove the "Options" dialogue" [Normal,New] http://bugzilla.gnome.org/show_bug.cgi?id=654617 [10:26] eh [10:26] remove it entirely? [10:26] yes [10:26] sigh [10:26] they said it's rather a tweak tool thing [10:26] like no normal user who use those options... [10:27] see bug #648919 [10:27] Launchpad bug 648919 in virtkey "package python-virtkey 0.60.0-0ubuntu1 failed to install/upgrade: Unterprozess installiertes post-installation-Skript gab den Fehlerwert 1 zurück (dup-of: 648695)" [Undecided,New] https://launchpad.net/bugs/648919 [10:27] Launchpad bug 648695 in virtkey "package python-virtkey (not installed) failed to install/upgrade: file does not exist: /usr/lib/python2.6/dist-packages/python_virtkey-0.60.0.egg-info" [Critical,Fix released] https://launchpad.net/bugs/648695 [10:27] ups [10:27] gnome bug #654617 [10:27] Gnome bug 654617 in Region & Language "Remove the "Options" dialogue" [Normal,New] http://bugzilla.gnome.org/show_bug.cgi?id=654617 [10:27] so its removed and then someone needs to go and add it to the tweaktool? [10:27] mvo, indeed [10:27] * mvo tries to control his temper [10:30] are you fucking kidding me? [10:30] * desrt doesn't do as well as mvo [10:31] i see this is a very old discussion.... [10:32] I'm fine having it in the tweak tool, but I want to have this option, I really really do, its the first thing I configure on every machine I work on [10:32] it's kinda funny [10:32] my girlfriend couldn't find how to turn on the compose key yesterday [10:32] maximize with ctrl-0 and make caps a additional ctrl [10:32] and we had almost exactly this conversation: https://bugzilla.gnome.org/show_bug.cgi?id=648919 [10:32] Gnome bug 648919 in Region & Language "Region and Language is wrong place for ctrl key position preference" [Normal,Resolved: duplicate] [10:33] mvo: my use case is capslock for compose [10:34] desrt, very old like 2 weeks? [10:34] hey desrt btw ;-) [10:34] seb128: since 2005 [10:34] Opened the keyboard preferences dialog and checked out "Layout Options". Wohoha! [10:34] Expandorama! [10:34] that comment was made in 2005 [10:34] desrt, what discussion? [10:34] you commented on it back then, even [10:34] https://bugzilla.gnome.org/show_bug.cgi?id=311672 [10:34] Gnome bug 311672 in Keyboard "Clean up of Layout Options" [Normal,Resolved: duplicate] [10:34] desrt, well the bug you pointed and the bug it's duplicate of are 2 weeks [10:35] seb128: there's another duplicate [10:35] old [10:36] it's cool [10:36] just ship tweaktool on the CD [10:37] and use the now-public control centre API to include it in the control centre :) [10:43] desrt, ;-) [10:44] bah, the thunderbird column resizing is buggy [11:20] weeeeee, somebody known with modemmanager code ? :D [11:38] dupondje, cyphermox does [11:38] or I think he does, check with him still ;-) [11:40] playing around to enable my gps device in modemmanager ^^ === MacSlow is now known as MacSlow|lunch [11:49] chrisccoulson, ok, tb is done indexing, I visited most of my folder and restarted it [11:49] it still fails to update any folder count :-( [11:49] oh the "get all new messages" does it [11:50] it seems "get mail" doesn't do though and it doesn't do it on start either [12:25] hmm, there is a lack of PPA builders again [12:26] probably wasn't a good idea for me to upload 10 firefox builds and 8 thunderbird builds [12:26] chrisccoulson: that's why my PPA builds are stuck!!! [12:26] lol] [12:27] didrocks, what are you trying to build? [12:27] chrisccoulson: some unity stuff ;) [12:27] ah, ok [12:27] that's not important ;) [12:28] * didrocks adds some if chriscoulson ;) [12:28] you guys stop locking the buildds! [12:28] the archive builders are idling though [12:28] upload to the archive ;-) [12:29] seb128, you want me to upload firefox 8.0 to the archive? ;) [12:29] i'm happy to do that [12:29] yes! [12:29] chrisccoulson: add to the changelog "because seb128 told me" [12:30] lol [12:30] can you upload a working tb as well? [12:30] ;-) [12:30] heh [12:30] seb128: isn't it a one line patch? :) [12:30] like one which actually notice new emails [12:30] why does ubuntu-desktop depend on xterm, does anyone know? [12:30] hum, I do :) [12:30] fallback [12:31] nevermind [12:31] fallback? xterm is my main session! :) [12:31] hehe [12:31] * didrocks prepares his upload for getting ready before chrisccoulson uploads firefox :) [12:32] well there's a bug requesting that it included 'NoDisplay' to not show in the classic menu by default, but I'm leaning towards closing it as Opinion [12:50] seb128, good morning... i fixed that flickering of the bottom tile on scrolling :) [12:52] kenvandine: is today you're day for fixing my bugs or something :) [12:53] last night was :) [12:53] Yay! [12:53] davmor2, i think i fixed some of the ones you filed and didn't mark them as resolved [12:53] it was 3am... was a little tired :) [12:54] kenvandine: I'll go through them after and look [12:54] please do :) [12:54] davmor2, so working better now? [12:57] kenvandine: Yeap, home show all messages in sequence, messages is in sync with everything else, opens from the menu, isn't using more than one core to display stuff, it sends messages and reply works so on the whole much better :) [12:58] :-D [12:58] awesome [12:58] maybe i can sleep tonight :) [12:58] njpatel, would it be insane to make StreamViewTile just an HBox which swaps out other tile types? [12:59] so i could have some tiles that are laid out nicer depending on the content? [12:59] kenvandine, no, not insane [12:59] we only create around ten, so it's fine, go crazy :D [13:00] well, I guess 20-30 [13:00] right now set_details is nuts [13:00] njpatel, some of the tweaks i made this week has made scrolling much faster again [13:01] sweet! [13:01] and memory seems under control [13:01] njpatel, i really want the async image loading [13:01] and actions :) [13:02] njpatel, and i think the sizing issues have resolved themselves somehow [13:02] kenvandine, weekend dude, still trying to make all this lens stuff work for FF [13:02] kenvandine, magic! [13:02] i never get the half streams anymore :) [13:02] kenvandine, if I leave my code alone long enough, it fixes itself [13:02] hehe [13:03] i think we were trigger "refresh" too often in stream-view [13:03] so it was happening too often, and sometimes allocation size was wrong [13:04] seb128, i fixed that flickering bottom tile [13:04] hey kenvandine [13:04] kenvandine, yeah, I read that in the changelog, nice! [13:04] it should also always fill the screen now with tiles on startup [13:05] sweet [13:05] njpatel, i would like to hide all the empty tiles for that 1-2s it takes to load the data on startup [13:05] but doing that breaks all the sizing stuff [13:06] kenvandine, don't hide the tiles, update their draw function to just return if they don't have useful data [13:06] you get the same effect :) [13:06] humm [13:06] kenvandine, ok, I can confirm that the flickering is fixed, it does take like 3 seconds to react to the scroll click though [13:06] 3? [13:06] ugh [13:07] seb128, indeed... [13:07] * kenvandine wonders why... [13:07] but that's only an i5 laptop not a modern config according to dx criterious ;-) [13:07] hahaha! [13:08] seb128, i can probably clean up more on setting the position [13:08] didrocks, do you plan to do the bamf update today? [13:08] we are probably too agressive there too [13:08] seb128: you guys know that gnome 3.2 will only run on sandybridge, right? [13:08] seb128: hard dependency in gnome-settings-daemon [13:08] seb128: waiting for the archive to be unfrozen, yeah [13:08] desrt, ;-) [13:08] desrt, haha [13:08] seb128: I did the upstream release on purpose if you didn't note :-) [13:08] seb128: i see that you're happy. i'm glad you agree! [13:08] notice* [13:09] made* [13:09] desrt, why is the gsettings manpage replaced by a stupid message when you build without regenerating it? [13:09] seb128: damned if i know [13:09] seb128: --enable-man [13:09] desrt, like the tarball has a valid version why not shipping that rather than an error message? [13:09] * desrt isn't even sure what framework we use for man [13:09] desrt, no thanks, I don't want to install that stack only for a manpage [13:10] desrt, I will just hack the rules to copy the tarball one over the make install version after build [13:10] davmor2, also, fyi the lens still isn't getting new messages... known issue [13:10] i need to look at why [13:10] $(man_MANS): echo Man generation disabled. Creating dummy $@. Configure with --enable-man to enable it. [13:10] uh.... [13:10] you're right. that's stupid. [13:10] ;-) [13:10] seb128: bug and cc: me please [13:11] seb128: patches welcome :) [13:11] desrt, ok [13:11] kenvandine: oh yeah I forgot about that one :) [13:12] seb128: i guess the logic should be something like: if the manpage exists, install it [13:12] and if we have the tools to update it and it needs to be updated, do that [13:12] desrt, right [13:13] i wonder what we do if we don't have the tools, and it's out of date [13:13] maybe delete it [13:13] (and not install it, obviously) [13:13] why would that happen? [13:13] the make dist should create an updated version [13:13] hypothetical [13:13] if the .xml gets touched, or something [13:15] seb128: just so you know -- you're on the bottom of my todo list :) [13:15] although ted is on vacation this week and next [13:15] i'm sure he won't mind if i hack glib's makefiles instead of doing his silly stuff === m_conley_away is now known as m_conley [13:16] desrt, that's fine, I've nothing worth being high on your todolist ;-) [13:16] desrt, when do you fly btw? [13:16] in a few hours [13:16] you are still in Canada? [13:16] ok [13:16] i think i arrive around noon tomorrow [13:16] where are you guys booked? [13:17] desrt, Motel One Berlin-Alexanderplatz [13:17] Dircksenstraße 36 [13:17] we're at some weird out of the way hotel [13:17] fjord [13:18] ours is just next to Alexanderplatz [13:18] i'm south of potsdamerplatz [13:19] got a bit of a walk [13:19] and now i find out it's going to rain the whole time :) [13:19] fortunately there are worse places to walk around than berlin :-) [13:27] desrt: enjoy the (current) german summer :) [13:37] dupondje: modemmanager might have some idea of GPS devices, but I'm not sure how far you can go, IIRC checking in d-feet or API docs would tell you what parameters to pass; and then you'd likely need to include a ttyUSB* device that points to the GPS chip [13:40] cyphermox: It has indeed mm-modem-location.c :) [13:40] where the basic api is in, but seems nothing uses it yet [13:40] nope [13:42] http://ubuntu.dupondje.be/modemmanager.diff [13:42] made this now :) but guess its not totally right ^^ === MacSlow|lunch is now known as MacSlow [13:45] cyphermox, about gnome-keyring caps that made ubiquity hang, does it make a difference if the installation is run on HW or VM ? [13:45] jibel: no [13:45] but stgraber fixed that [13:46] cyphermox, I know he fixed that, but charlie-tca still has the problem on xubuntu amd64 on HW. [13:46] jibel: charlie-tca: do you have the manifest file on that iso/cd/etc to check what version of gnome-keyring is in? [13:46] jibel: I also remember charlie-tca saying that the lightdm fix doesn't work for him... are we 100% sure the xubuntu image is fully up to date? [13:47] how do we verify it is 100 [13:47] % [13:47] ? [13:47] stgraber, cyphermox that's what I'm checking [13:47] charlie-tca: what version are you testing? [13:47] 2011-08-03.2 [13:48] charlie-tca: i386 or amd64? [13:48] The latest image I have available [13:48] amd64 [13:48] looking [13:49] I did file a new bug 820731 for it [13:49] Launchpad bug 820731 in ubiquity "Oneiric Ocelot Xubuntu Desktop images fail to install" [High,Incomplete] https://launchpad.net/bugs/820731 [13:49] according to the manifest, both casper and gnome-keyring are up to date [13:50] I'm downloading it here now to have a look [13:51] stgraber, right, and I can't reproduce the problem in a VM. [13:51] hence my question, would bare-metal testing make a difference? [13:51] no [13:52] k [13:54] charlie-tca: just triple checking, you have: [13:54] 19815fe1a15674db6fa51b8960629bc3 oneiric-desktop-amd64.iso [13:55] oh nevermind, I see jibel asked already in another chan :) [13:55] yes, let's take this back to #ubuntu-testing :) [13:58] dupondje: that patch looks okay-ish, but do you really need to initialize it that way? isn't that done in mm-modem-location.c? my guess is that this should be done by calling the right method when needed, and the location itself probably might be just one async call [14:00] dupondje: but anyway, best is to bring this up on the networkmanager mailing list: http://mail.gnome.org/mailman/listinfo/networkmanager-list (or on #nm) [14:01] cyphermox: you need to enable that card with special *GPS codes :) [14:01] yes, but the enabling does belong in mm-modem-location.c, not mbm, I think [14:03] mm-modem-location.c is the abstract class no ? [14:04] the enabling of the modem for example is also done in mbm-modem.c btw [14:04] I don't know I'm not overly familiar with that part of the code. otherwise you should have some other file per device or something [14:04] mbm is IIRC meant for actuall communications [14:04] the plugins folder is full of device specific code :) [14:05] all the commands to send PIN code etc are in the mbm file also [14:08] anyway I play some more with it :D [14:08] anyway, if it works, nobody will complain ;) [14:10] héhé true :) but its always good to do it the good way :) [14:10] but guess there is nothing in the gui yet to see LOCATION support ? :) [14:10] of networkmanager then [14:17] seb128, i found the slowness in stepping with the scrollbar... it is only on down, and only with overlay scrollbars [14:17] the value gets set 21 times with overlay scrollbars :) [14:18] weird ;) [14:22] kenvandine: latest gwibber seems to repeat Facebook posts, I'm going to restart to ensure it still happens from a fresh start [14:23] davmor2, ugh... [14:23] yeah... i bet i know why [14:24] kenvandine: only facebook though [14:24] yeah [14:24] because facebook sends me posts back if it changed [14:24] like when someone posts a comment [14:24] or likes it [14:24] and i bet my dupe detection isn't matching them for some reason [14:26] kenvandine: Yeap 55 likes on first post 56 likes on the second [14:28] kenvandine: also https://twitter.com/#!/matthewrevell/statuses/99122602014015488 on old gwibber shows in new gwibber you get <and> instead [14:30] davmor2, yeah... i need to fix message parsing a bit so it doesn't break the Gtk.Label [14:30] right now i do some escaping so it doesn't break the label [14:30] but it over does it :/ [14:31] althought that is interesting that online it has the work "and" [14:31] oh... you mean ? [14:32] not [14:32] davmor2, ^^ [14:32] kenvandine: yeah on the word primary [14:32] that is getting escaped as a tag :/ [14:33] if anyone has ideas how to sanitize the data to be pango friendly... [14:33] i would love to hear them [14:33] we gotta figure that out [14:34] kenvandine: two last things how do you search for people? if I type in kenvandine it is displayed as #kenvandine only, the search box is really close to the edge of the app it's quite easy to miss the box and click on the app behind if you have a fullscreen app behind it :) [14:35] not implemented yet [14:35] the entry is a place holder until njpatel makes it not ugly [14:35] :) [14:35] that is just doing twitter/identica searches [14:35] not user searches [14:35] hahaha :) [14:35] so like hash tags [14:36] i so add a # as the first character [14:36] user streams are in the works [14:36] cool I think thats it from this one I'll keep looking though :) [14:37] cool [14:37] davmor2, things like image previews in the tiles and such are coming this weekend [14:37] and i am going to make the content presentation much better in the tiles [14:38] kenvandine: nice I'll keep my eyes out for the changes :) [14:47] interesting [14:47] ross opened a bug about letting g-c-c allow external panels again or adding a connman one for meego ;-) [14:50] is it WONTFIX yet? :) [14:50] not yet ;-) [15:20] how do I configure auto-login in lightdm? [15:21] and why does my datetime indicator only display the text "Time" instead of displaying it? [15:21] geser, translator error maybe? [15:21] geser, the autologin, in a config file for now I think [15:22] seb128: is /etc/lightdm/lightdm.conf the right one? [15:22] yes [15:24] I've set default-user=michael and default-user-timeout=0 below the [seat-0] section and still had the greeter instead of auto-login [15:24] wrong keys? [15:26] mvo, https://bugs.launchpad.net/ubuntu/+source/software-properties/+bug/815480 [15:26] Ubuntu bug 815480 in software-properties "software-properties-gtk crashed with UnicodeDecodeError in init_distro(): 'ascii' codec can't decode byte 0xd0 in position 0: ordinal not in range(128)" [Undecided,Confirmed] [15:26] mvo, have you seen that? I've something similar on an a3 live session (using a chinese local if that makes a difference) [15:27] geser, the livecd has [15:27] [SeatDefaults] [15:27] then "autologin-user=ubuntu" [15:27] with "autologin-user-timeout=0" [15:28] geser, I will check with robert_ancell why lightdm stopped install a config file [15:28] it was easier to edit [15:28] if you have one on disk it might be a leftover from the previous version [15:28] the format changed in 0.9 [15:29] yes, it might be a leftover file from the previous version [15:30] seb128: I haven't but I check it now [15:30] mvo, I've the system with the issue running if you need infos [15:30] seb128, i noticed some locale issues on a clean install [15:30] perhaps it is related [15:30] the output of the locale command is right [15:30] kenvandine, well that's the live session and it seems to work fine otherwise [15:30] seb128: oh? what does "lsb-release -a" gives you? [15:30] distro.id should really be just plain ascii [15:30] seb128, yeah... gwibber was having a problem collating [15:30] mvo, "command not found" [15:30] which used LC_COLLATE [15:31] mvo: it's not installed [15:31] lsb_release -a ? [15:31] which was set... and looked correct [15:31] but it failed using the collator in glib [15:31] worked fine on a dist-upgraded box from natty [15:31] seb128: not even with the underscore? I think it should be [15:31] haven't been able to nail why though [15:31] mvo, ups sorry [15:31] seb128: my mistake [15:31] seb128: I pasted the wrong one earlier [15:32] pango in gir is crahing, but pi'fixit'ti is on vac, right? [15:32] mvo, [15:32] $ lsb_release -a [15:32] No LSB modules are available. [15:32] Distributor ID: Ubuntu [15:32] Description: Ubuntu oneiric (development branch) [15:33] Release: 11.10 [15:33] Codename: oneiric [15:33] hm, that looks all like ascii, no? [15:34] indeed [15:34] mvo, pitti will be back on monday I think [15:34] he seems to be back online though, he did reply to some emails and did a debian upload [15:34] could you put a "print self.distro.id" before "/usr/lib/python2.7/dist-packages/softwareproperties/gtk/SoftwarePropertiesGtk.py" line 269 please? [15:34] just cowboy it in via sudo ;) [15:35] yeah [15:36] mvo, "Ubuntu" it prints [15:37] * mvo is confused [15:38] and after the print it give the error? [15:38] ohh, its probably the translation for software, but why is it giving a error there [15:38] not if I remove the _() around the "%s Software" [15:39] this is in .fr, right? [15:39] no, it's chinese ;-) [15:39] zh_CN.UTF-8 [15:41] woah [15:42] maybe a incorrect (non-utf8) translation for this string? [15:43] mvo, https://translations.launchpad.net/ubuntu/oneiric/+source/software-properties/+pots/software-properties/zh_CN/42/+translate [15:43] is seems [15:43] I wonder if rosetta could catch those [15:44] not sure if that's an incorrect encoding one [15:46] let me grep my logs [15:46] we had several similar bugs with pygtk softwares before [15:46] or pygobject ones [15:46] * desrt takes oneiric for a spin [15:47] * desrt senses that he may regret this choice [15:47] desrt, of course not! [15:48] mvo, using "*% Software".decode('utf-8') works [15:48] *%->%s [15:51] so _("%s Software").decode('utf-8') % self.distro.id ? [15:51] yes [15:51] like addind .decode('utf-8') to the string [15:51] what does it print if you do "locale" ? [15:51] seb128: I don't get why its needed for this particular string and not all of them [15:52] zh_CN [15:52] it's similar to https://bugs.launchpad.net/ubuntu/+source/apport/+bug/738632 [15:52] Ubuntu bug 738632 in apport "apport-kde crashed with UnicodeDecodeError in run_crash(): 'ascii' codec can't decode byte 0xc3 in position 5: ordinal not in range(128)" [Low,Fix released] [15:53] mvo, I wish I did understand python encoding handling enough to say [15:54] mvo, but maybe you are right and the translation is broken [15:59] mvo, so [15:59] $ LANGUAGE=zh_CN.UTF-8 gettext software-properties "%s Software" [15:59] %s 软件 [15:59] would you mind adding the info you have to the report? [15:59] so that it does not get lost? [15:59] will do [15:59] I assume the "isutf8" tool is happy with the gettext output? [16:00] tremolux: I think I found the pnago issue [16:00] mvo: oh yeah? [16:01] it looks like a incorrect annoation in get_line_readonly() I'm testing building now to see if that is true or not [16:01] mvo, yeah, other tools are happy with it [16:01] why do we ship xterm on the Desktop CD? [16:02] seb128: I'm confused, I will give it the full force of my attention when this pango issue is solved [16:03] mvo: ah no kidding, thanks for checking into that! === ara_ is now known as ara [16:04] I ask because of bug 129041, showing the .desktop in the default install is ugly but if people intentionally install it [16:04] Launchpad bug 129041 in xterm "xterm icon available by default" [Low,Opinion] https://launchpad.net/bugs/129041 [16:04] then they should have a launcher [16:18] jbicha, somebody mentioned it was for the failsafe session earlier [16:19] read the irclog for details ;-) [16:20] mvo, ok, I dumped in the bug the infos I have and subscribed pitti [16:21] * desrt clicks 'lightdm' [16:21] mvo, but he said that a while ago [16:21] now I run into tons of trouble with assertions/crashes, having to call encode('UTF-8') everywhere (because Gtk expects UTF-8 strings, not unicode), and into all other sorts of type errors [16:23] mvo, he did that in jockey http://bazaar.launchpad.net/~jockey-hackers/jockey/trunk/revision/711 [16:23] well though in this case the string is coming from gettext and should be utf8 [16:23] yeah [16:24] I know about the stdout issue [16:24] * mvo scratches head [16:24] mvo, debug your pango issue first, no hurry [16:24] mvo, I'm done with what I have as debug ideas, I dumped that in the bug [16:27] thanks seb128, much appreciated [16:28] one pango crash fixed, but its crashing still :/ [16:32] have a good evening everyone! [16:35] seb128: I put bug 820773 in the sponsorship queue for after alpha3 to allow the libglew transition to actually happen, it just removes the provides for the unversioned dev packages so that anything that build-deps on the versioned ones won't be touched [16:35] Launchpad bug 820773 in glew1.5 "glew1.5 shouldn't provide unversioned dev packages" [Medium,Confirmed] https://launchpad.net/bugs/820773 [16:36] micahg, ok thanks [16:46] tremolux, hey [16:46] seb128: hi! [16:47] tremolux, could you add a small summary of what is new in software-center in a3 to https://wiki.ubuntu.com/OneiricOcelot/TechnicalOverview [16:47] if there is anything worth mentioning [16:47] ? [16:47] seb128: sure, I will [16:47] tremolux, thanks ;-) [16:48] seb128: sure thing [16:49] wow. that sucked hard. [16:49] desrt, what? [16:49] during the upgrade grub was trying to probe my partitions or something [16:49] and it opened the mmc card reader [16:49] and i guess the driver has a bug or something because it locked up grub and went into an extremely fast infinite loop of spewing messages to the console [16:49] filling the console in 1 second or so, say [16:50] so even if you press ^L your screen is full again in less than a second [16:50] so i login, fuser -k mmcblk0 [16:50] and it opens it again with a new PID to retry [16:50] i had to rename the device, then fuser -k on the renamed one (so it wouldn't find it again) to get out of that mess [16:51] urg [16:51] "fun" [16:51] ;-) [16:51] i renamed it to a certain 4 letter word [16:51] * desrt was getting pissed off [16:53] anyway.. other than the usual flood of "this thing you don't care about crashed" dialogs, it seems pretty usable [16:53] \o/ [17:14] seb128: I have dinner now, pango seems to be ok now, but tremolux needs to double check first if it wasn't fun flares or anything [17:21] mvo, enjoy [17:24] hello, not sure if someone mentioned it here already but ISO testing found bug 820372 [17:24] Launchpad bug 820372 in ltsp "default session set to GNOME and fails. Should be Unity" [High,Confirmed] https://launchpad.net/bugs/820372 [17:24] which isn't an LTSP bug but a gnome-session one [17:24] LTSP calls /etc/X11/Xsession to start whatever is the default session (no parameter) [17:24] Xsession then starts gnome-session but without a --session= parameter which makes it fail [17:24] I think we had a fix in Natty to make it use a the right "default" session but that apparently got lost somewhere [17:25] was wondering what source package I should move that bug to? [17:25] stgraber: what I would see was aside from the Ubuntu and Ubuntu 2D entries there was an extra GNOME one that shouldn't be there [17:26] (outside of ltsp) [17:27] with Edubuntu we have gnome-session-fallback installed so apparently gnome-session starts the fallback session then (instead of unity-3d => unity-2d => gnome-session-fallback) but on regular Ubuntu, it just fails [17:28] mmkay, so which case is failing now? [17:29] calling "/etc/X11/Xsession" with no parameter doesn't start the default session on regular Ubuntu [17:29] the same thing on Edubuntu works but starts the session fallback instead of unity (which is wrong too) [17:31] simply calling "/etc/X11/Xsession" from a terminal on your system should reproduce the issue [17:32] stgraber, can you check with didrocks when he's around? [17:32] stgraber, or better assigned the bug to canonical desktop team [17:32] we will sort it [17:41] seb128: ok, will do [18:46] -Werror=unused-but-set-variable is supposed to cause a build-failure if -Werror is used, right? [18:47] I just want to confirm whether I should open a FTBFS bug upstream for it [19:03] jbicha: i would think that option would always cause that warning to be an error, but would not cause other warnings to also be errors. but i don't claim to know what gcc actually wants to do with it :) [19:05] ok, well upstream was building was -Werror so I'll go ahead and open the bug there, thanks! [19:09] well, it's a change in the defaults somewhere; we had to fix a couple of those in ubuntuone stuff. so yeah, upstream should fix it, but it's really annoying when compile flags change out from under us :) [19:11] dobey: so you're saying that that warning didn't used to cause a build failure? [19:11] I don't understand gcc which was why I asked because it doesn't seem to me to be a critical warning at all [19:14] jbicha: that warning didn't used to appear at all [19:15] oh, hmm [19:15] jbicha: but in oneiric it does now [19:15] yeah i don't know why it got changed to an error by default via flags [19:17] well I believe -Werror just treats all warnings as errors which will stop the build [19:17] yes [19:18] but -Werror=foo was added as a default in CFLAGS [19:18] and not by upstream devs [19:27] Can someone help me with building a package? I have problems on the linking step. [19:30] It is argyll from this PPA: https://launchpad.net/~pmjdebruijn/+archive/ppa [19:30] It builds on Natty but not on Oneiric. [19:31] I need it to build on Oneiric to include it as part of the new Color Management support. === hggdh_ is now known as hggdh [20:05] tkamppeter: i don't see such a thing in that PPA, and everything in it has successful builds already [20:08] dobey, sorry, wrong link, he has tons of PPAs. Here we go: https://launchpad.net/~pmjdebruijn/+archive/argyll-release [20:09] He built it only on Natty and older. The problem is when one takes the Natty version and tries to rebuild it on Oneiric. [20:09] oh [20:09] then pastebin the error - http://paste.ubuntu.com/ [20:11] dobey, http://paste.ubuntu.com/658851/ [20:12] that's a lot of compiler warnings, but it looks like you're missing X development libs [20:13] dobey, I have already checked that. They are installed. [20:14] tkamppeter: probably an --as-needed issue then, the order of the libs matter [20:14] Also adding -L/usr/lib/x86_64-linux-gnu to the linker command line (for the multiarch packages) does not help. [20:16] micahg, is there an "--as-needed" command line option to place somewhere? [20:17] tkamppeter: well, the solution is to reorder the libs on the linker line [20:17] dobey, micahg, I can run the linker with the "ld" or "ld.bfd.real" command and I get only the warning [20:18] ld.bfd.real: warning: cannot find entry symbol _start; defaulting to 0000000000401280 [20:18] and 0 as exit value. [20:18] But if I try to run the linked program I get [20:18] ./plot/plot [20:19] -bash: ./plot/plot: No such file or directory [20:19] dobey, micahg, ^^ [20:23] ok, i'm not sure. i'm not on your machine trying to build it, and can't easily try myself. but you're a developer, so i'm sure you'll figure it out :) [20:24] tkamppeter: this explains it somewhat: https://wiki.ubuntu.com/NattyNarwhal/ToolchainTransition#How_to_Fix_a_Problem [20:38] micahg, thanks. I have tried a lot of such things. No I have replaced the linker by ld, which reduces the problem to the mentioned _start warning and only a few missing symbols. [20:40] tkamppeter: I'm sorry I don't have more time to look at it right now, but maybe people in #ubuntu-motu could help if no one is around here [20:42] wassup tkamppeter [20:44] tkamppeter, "cc -o plot/plot plot/plot_plot.o spectro/libconv.a numlib/libnum.a -L/usr/lib -ldl -lrt -lX11 -lXext -lXxf86vm -lXinerama -lXrandr -lXau -lXdmcp -lXss -ltiff -lm -lpthread" should fix your problem [20:44] (well, that would be the first thing i'd try. the ordering is wrong in any case) [20:50] chrisccoulson, now "plot" actually works. [20:50] What is the difference of your command line to the one in the package? How did you obtain this command line? [20:51] chrisccoulson, ^^ [20:51] \o/ [20:51] tkamppeter, when linking an object, the linker only looks for symbols in libraries that are specified after it [20:51] i think ;) [20:52] IIRC, it was explained to me as with --as-needed and symbols not needed when encountering a library on the linker line are dropped [20:52] s/and/any/ [20:55] chrisccoulson, this recipe works! === m_conley is now known as m_conley_away [22:45] is there any way to store the passphrase for an ssh key in the keyring non-transiently? [23:11] broder: What do you mean? You in such a way as it's automatically available when your session is unlocked? [23:12] RAOF: i don't want to store my ssh key unencrypted, but i don't want to have to remember an extra password since i already have the keyring which is encrypted to my login password [23:12] Either gnome-keyring-daemon or seahorse does that. I forget which. [23:13] the gnome-keyring-daemon will prompt me for the password to the ssh key and then store it for the duration of my session [23:13] One of the options there is “unlock whenever my keyring is unlocked”, isn't it? [23:14] oh, huh. yeah, that sounds like what i'm looking for [23:14] of course...i won't be able to test it until i get this new key whose passphrase i know onto my server... [23:14] Heh. [23:15] an unfortunate side effect of having locked down this laptop and removing all interesting key material before going to DEFCON... === tkamppeter_ is now known as tkamppeter [23:39] kenvandine: did that bug I logged recently make any sense, crash on start up of Gwibber on natty :s [23:56] broder: have fun!