[00:17] morning all! robert_ancell TheMuso RAOF last few days in AU :) [00:18] jasoncwarner: Good morning! [00:21] 01:21:11 up 7 days, 14:50, 9 users, load average: 39.61, 16.86, 11.65 [00:21] 01:21:11 up 7 days, 14:50, 9 users, load average: 39.61, 16.86, 11.65 [00:21] 01:21:11 up 7 days, 14:50, 9 users, load average: 39.61, 16.86, 11.65 [00:21] 01:21:11 up 7 days, 14:50, 9 users, load average: 39.61, 16.86, 11.65 [00:21] whops [00:22] as you can see, my machine actually is doing something against the cool winter. [00:23] jasoncwarner: good morning btw. note to self: This really mean I should go to bed. [00:23] :) [00:23] get to bed! [00:23] jasoncwarner: is that an order? [00:25] * bryce waves to jasoncwarner [00:26] hey bryce ! [00:26] Sweetshark: yes ;) [00:27] alrighty ;) [00:27] night Sweetshark [00:28] jasoncwarner: I hope your last days in Au aren't too hot. [00:28] TheMuso shouldn't be. was 41 on monday, but been cooler since. [00:28] Twas still 34 in Sydney at midnight on Wednesday morning, I hope that never happens again. [00:28] TheMuso 47 last friday, though! [00:28] yikes [00:28] jasoncwarner: Ouch. [00:30] ouch. my sympathies. [00:32] jasoncwarner, coincidentally it was 47 here in Oregon the other night too ;-) [00:32] bryce lol. I'd take that 47, though! [00:41] bryce: it got that warm? :) [00:51] jasoncwarner, hello! [00:52] oh, I've missed him ;-) [00:54] well, I'm heading to the gym. back later [01:58] heh: http://www.smh.com.au/environment/weather/temperatures-off-the-charts-as-australia-turns-deep-purple-20130108-2ce33.html [01:58] so hot in australia that the weather guys had to add new colors to their thermometers [01:59] wow, those temperatures are crazy [02:00] I think I'd be a rapidly-evaporating puddle [02:03] bryce: Not the whole country. It's now cool over here :) [02:04] And down here, too. [02:04] Although we're still on fire. [02:04] RAOF: You ok where you are? [02:05] Oh, absolutely. [02:05] Aside from the whole state being on fire [02:05] It's nowhere near Hobart, and even if it were, we're not in a particularly likely area to get incinerated. [02:05] RAOF: Suburbia? [02:05] West Hobart, yeah. [02:06] We don't have suburbia like Perth has suburbia ☺ [02:07] Incidentally, we're heading to Perth at the beginning of Feb; we should catch up. [02:07] I've found it interesting that the eastern Australia finds 40 degrees hot or unusual. And high-40s unprecedented. [02:07] -the [02:08] RAOF: Yes, I am here, still. Sounds good [02:08] duflu: That's because Perth is adjacent to the Vents of Hell :P [02:10] No wonder I keep getting burnt [02:12] duflu RAOF I saw these and laughed. http://www.jokeoverflow.com/wp-content/uploads/2011/01/australia-500x487.jpg and http://www.australianblogs.com.au/blog/wp-content/uploads/2011/09/Australia-Danger-map.jpg [02:14] jasoncwarner: Heh, yes so accurate too [02:30] Once that hot air mass over central Aus is discipated, things should improve overall. [02:30] But I believe that will be a while, since it aint going anywhere. [02:59] desrt: hmm, the only thing I'm doing is defining an interface in a shared library and that's it [02:59] I can quickly double check if I missed something [03:01] duflu: RAOF I assume you have all seen this: http://i.imgur.com/4GO07.jpg [03:04] smspillaz: Don't know what you're talking about. The window is open and I have a cool breeze [03:08] its nicer now [03:09] I was working in a kitchen a few days ago and it was fun going from the freezer back into the boiler rooms [03:09] hair on fire, hair frozen [03:26] duflu: BTW, kinda annoying news, the "solution" that khronos gave me (after mesa said NOTOURBUG) was that they were going to break the header file anyways and that we should include gl2.h locally in our source tree [03:27] which I think is /insane/, but there's not a whole lot I can really do [03:27] I guess the cast really is the best option at this point :( [03:28] * smspillaz thinks its a little stupid when someone tells you that their header files never change, so they won't add versioning macros, yet they later make changes to their header files [03:30] smspillaz: Well, it almost makes sense. GL(ES) is a well-defined API for which you should be able to use a single header forever unless you need new features. But yeah the cast is better [03:30] duflu: right, it just seems as though their logic contradicts itself [03:31] they changed their header in a non backwards-compatiable way, and then said that they wont add versioning info to their header because they don't change their header [03:31] smspillaz: Look at it this way: ALL driver vendors have to share the same header file. So all drivers have to work with that header at a binary level (at which point constness of parameters is irrelevant) [03:32] of course, the binary level is fine [03:32] smspillaz: And yes you're right. They messed up [03:32] I'm talking about the fact that they've gone and broken client code [03:32] They messed up badly considering that existing GL definitions should never change [03:32] what annoys me is that they won't admit they messed up -.- [03:33] * duflu wonders if there's another calling syntax compatible with the single and double-const versions. [03:34] there's very little reason why they had to make that change. All adding the double-const does is ensures that the driver gets a nonwritable char *, but the driver can just cast it away anyways .. [03:35] perhaps const char * const * is not compatible with const char ** [03:35] but again ... I don't think that's a good reason to change it post-facto [03:35] *sigh* oh well [03:40] smspillaz: Yes "const char **" means the first level of indirection is writable (like when you want to return a "const char *" by parameter). The more correct syntax for an input-only parameter representing a sparse 2D array of chars is the double-const version. But I don't think there's a syntax compatible with both so they should never change it [03:41] Hmm, this is basic C I should be able to give a definite answer on :S [03:46] duflu: well, I figure that its difficult to give answers on stuff you don't come across much in practise [03:47] I don't come across the const char * const * variant much either [03:47] I mean, passing a char* to a "const char *" is normal and compatible but it looks like you lose that flexibility at a second level [03:47] char* const* is a "pointer to a const pointer to char", const char** "a pointer to a pointer to a const char" [03:47] pointers themselves can be const too [03:47] * smspillaz likes the go syntax for this kind of thing [03:48] a const * Int [03:49] "a is a const pointer to an Int" [03:49] annoying to read if you're used to C though [03:50] const binds to the left if there is anything to the left of it, else to the right [03:50] tsimpson: Thanks, yes we know [03:51] then it's easy enough to read [03:52] char * const * const :) [03:52] tsimpson: The problem is: How to define a variable that can be passed (without casting) to both a "const char **" and "const char * const *" [03:56] duflu: const char** can be converted to const char* const* automatically [03:57] tsimpson: but int (const char **) can't be converted to int (const char * const *) :( [03:59] tsimpson: which is what we're complaining about at the moment. Khronos (the vendor of the OpenGL|ES API) changed the header file that they ship to all of the graphics driver vendors to change a function signature from void (uint, size_t, int, const char **) to void (uint, size_t, int, const char * const *) [04:00] its not that big a deal but annoying nonetheless :) [04:03] smspillaz: is that one of those ones where you need to "request" a function pointer by name? (I don't play with OpenGL much) [04:04] smspillaz: Yeah actually it works (just tested). A parameter compatible with both function versions can be defined as either "const char **x" or "const char *x[n]" (same thing) [04:05] So it's likely a relatively safe API change. Not sure how compiz broke it [04:05] duflu: function pointers [04:05] the parameter itself is compatible [04:05] the pointer to the function is not [04:05] smspillaz: Ah yep [04:05] Definitely [04:05] tsimpson: not necessarily, although what we were doing was taking a pointer to glShaderSource as defined in the header file as a convenience [04:06] because in desktop mode its not core functionality so you have to use GetProcAddress [04:06] and having #ifdefs everywhere is annoying [04:06] desrt: so this: http://paste.ubuntu.com/1515714/ hangs [04:20] smspillaz: you're using G_DEFINE_INTERFACE incorrectly [04:20] the last argument is the pre-requisite type [04:20] probably you want G_TYPE_OBJECT there [04:20] in effect, you are declaring an interface type dependent on itself -- that's why you crash on re-entering the get_type() function... the lock is already taken [04:22] desrt: ah, I knew it [04:23] desrt: thanks [04:24] (this always happens when I try and brave it out and not copy-paste the boilerplate code :)) [05:22] Good morning [05:47] Laney: FYI, I committed doko's Ubuntu glib changes to the Debian exp svn, so if you update to the new version we can just sync again === Amaranthus is now known as Amaranth === mechanobot is now known as amithkk [06:51] good morning [07:14] bonjour didrocks, comment vas-tu? [07:15] pitti: je vais bien, et toi, tu te sens un peu plus en forme qu'hier? [07:16] oui, je me sens beaucoup mieux [07:16] ah, super :) [07:16] woke up at 6 again [07:20] still have issues to woke up at 7 here :) between 7h20 and 7h40 to start working :) [07:20] but finished at 8h50PM yesterday evening, so I'm not shy about starting 40 minutes afterwards :p [07:22] didrocks: as-tu domier bien la nuit derniére? [07:22] pitti: oui oui, encore juste réveillé à 5h, mais j'ai pu me rendormir de suite. Par contre, ça fait 3 nuits de suite que je me réveille d'un coup, je sais pas pourquoi… [07:23] ERROR: connecting to Launchpad failed: unclosed token: line 13661, column 6 [07:23] erk [07:23] ^ current retracer failure [07:23] what the hell is that supposed to mean? [07:24] even not a file associated with the line? :) [07:24] * pitti removes the .launchpadlib caches [07:24] or is it in the data that launchpad is sending to you? :) [07:24] pitti: I finally had to workaround this multi-process doesn't work on launchpadlib [07:24] looks like a LP errormesasge [07:25] like setting a launchpad cache dir per process [07:25] urgh, ~/.launchpadlib is 20 GB [07:25] hah, that did it [07:27] I wonder if one could disable caching altogether [07:27] the retracer won't ever look at the same bug twice anyway [07:29] well, if you set the cache dir to none, it will use ~/.launchpadlib, so I didn't see that option [07:30] one could probably use mkdtemp and set the cache dir to that [07:31] yeah, but you are still creating them [07:31] that's fine for one run [07:31] I just don't want them to pile up over time [07:32] yeah, that should do it [07:45] sigh [07:45] where's ancell? [07:48] hey guys, we are working on a unity configuration tool. is there any procedure to get it into universe section? [07:48] i asked a MOTU a week ago and he said it has to be approved by desktop devs. [07:49] so, here am i. [07:57] jokerdino: hum, it doens't need to be improve by desktop devs, any MOTU can sponsor it :) [07:57] didrocks: i was told since it is a unity tool, desktop devs need to verify if it breaks unity or something. [07:58] also, is there any branding issues if we name it Unity settings hub? [07:58] jokerdino: ah, I would prefer something that doesn't make it official [07:59] like myunity was clearly not official [07:59] unity tweak tool? :) [07:59] not sure if it was taken [07:59] so, we can use unity tweak tool? [07:59] i was of the opinion only ubuntu brand cant be used. [07:59] but i wasnt sure about using unity in the name [08:00] ubuntu tweak tool was never official [08:00] unity tweak tool sounds less "official" to me than "unity settings" TBH :) [08:00] do you base this on an existing tool such as myunity? in that case, you could just reintroduce "myunity" [08:00] pitti: it wasn't. we wrote it from scratch using python and gsettings [08:01] myunity was basic and shell commands [08:01] *basic = Gambas [08:01] didrocks: the name is actually "unity settings hub". [08:01] but i hear your point. [08:02] yeah, I'm afraid will type "unity settings" and think this is the official supported options [08:02] people* [08:02] okay :) [08:02] unity tweak tool, (if not already taken) sounds good to me, but that's just IMHO :) [08:02] we can't work on making it official, can we? [08:03] I thought not having too many settings was a design decision [08:03] jokerdino: well, we prefer to add the settings we officially support in gnome-control-center [08:03] the settings we exposed through gsettings that are not in g-c-c are not officially supported [08:03] also, each new setting leads to combinatorial explosion in testing and bug fixing [08:04] ah, that makes sense. [08:04] yeah, look at ccsm and how you can break everything in less than 20s :) [08:04] haha, that is a fair point. [08:05] but it's good that for geeks, there are those kinds of tools, just try to hilight that those combinations may lead to issue [08:05] (and a revert to default would even be awesome btw ;)) [08:05] ok, once we finalize on a new name, i'll get in touch again. [08:05] btw, why does unity reset (some of) the settings on package update? like focus-follows-mouse seems to revert to the default everytime here [08:05] sure, do not hesitate! [08:05] tjaalton: this is a bug, smspillaz knows more than I about it [08:06] didrocks: heh, we have a restore defaults button on every tab. [08:06] didrocks: ah, good [08:06] jokerdino: excellent! [08:06] tjaalton: no, we don't hate you *that* much :) [08:06] didrocks: *phew*.. :) [08:06] heh [08:07] ok, i'll disturb you guys later on ;-) [08:07] no worry, thanks jokerdino! :) [08:07] btw, there was an interesting suggestion to improve it on the "global menu bug", to add what gnome-shell apparently has, which is a 20ms(?) timeout to change the focus if the cursor is not moved [08:08] tjaalton: maybe talk with JohnLea about it? [08:08] "dwell timeout" [08:08] didrocks: ok I will [08:08] he will be back this Thursday or next Monday, I don't remember :) [08:09] but one of the two [08:16] didrocks: one last tiny thing. assuming unity-tweak-tool name is fine, can we just update the display name while sticking to the current package name? [08:16] the package name is currently ush. [08:17] jokerdino: not sure a lot of people will find "ush", but as long as the description of them is unity tweak tool (in debian/control), I don't see any issue :) [08:17] jokerdino: but TBH, I would personnaly recommend renaming the package before entering the archive [08:18] hey desktopers [08:19] didrocks: that's a fair point. i think we'll proceed to renaming it. [08:19] salut seb128! [08:19] * jokerdino researches on unity branding. [08:19] lut didrocks ;-) [08:43] bonjour seb128 [08:43] pitti, salut, ca va bien ? [08:43] seb128: oui, merci! et toi? [08:44] pitti, je vais bien merci ! [08:45] seb128: packgin for next week already? [08:55] good morning everyone [09:04] chrisccoulson, hi [09:05] heya [09:05] pitti: great! [09:05] :foo build-deps work on Debian? [09:06] I thought Debian would have all the multi-arch stuff now? [09:07] hmm, no Build-Depends.*:any in current sid main _Sources [09:07] pitti, not yet, will do the packing tomorrow [09:07] chrisccoulson, hey, how are you? [09:07] I remember that they were in the specification but not implemented yet - not sure where that stands ATM [09:08] Laney: meh, seems you are right; sid's dpkg doesn't parse that yet [09:09] * pitti reverts that part [09:09] dang, there is our delta [09:09] chrisccoulson, my firefox is back in french \o/ [09:10] better to check with cjw or infinity [09:10] en effet! mon firefox parle allemagne à nouveau [09:10] pitti, ;-) [09:15] hi ricotz [09:15] hi seb128, pitti, how are you? [09:15] seb128, oh, that's good about firefox :) [09:15] hey chrisccoulson [09:20] pitti, oh, you were affected by bug 1094376 too? i'm glad it works again :) [09:20] Launchpad bug 1094376 in firefox (Ubuntu Raring) "Language pack isn't used" [Critical,Fix released] https://launchpad.net/bugs/1094376 [09:20] chrisccoulson: apparently, yes; the menu was in English === Quintasan_ is now known as Quintasan [11:17] chrisccoulson: hm, firefox now hangs itself when I try to click an "Attach file.." button [11:17] oh, it just takes effing long [11:20] pitti, any particular site? [11:21] bugzilla.gnome.org [11:29] pitti, works fine here ... or I'm not clicking on the same buttons, I tried the "Add an attachment" on the new bug forms and on an existing bug [11:29] seb128: thanks for trying; probably something local then [11:30] nautilus is also dog slow for me, presumably something in between ubuntuone, gvfs, and my local fs [11:30] could be yeah === vrruiz_ is now known as rvr_ [12:04] seb128 afternoon, https://bugs.launchpad.net/ubuntu/+source/gnome-settings-daemon/+bug/1031034 :) [12:04] Launchpad bug 1031034 in gnome-settings-daemon (Ubuntu Precise) "Breaks with smart card auth with pkcs11" [High,Triaged] [12:04] Do I need to do anything else ? [12:04] sru , and debdiff [12:05] ritz, hey, no need to do anything else, that looks good, thanks for the work on it [12:06] seb128, thank you :) [12:26] https://bugs.launchpad.net/ubuntu/+source/unity/+bug/1096954 [12:26] Launchpad bug 1096954 in unity (Ubuntu Precise) " Enabling Xinerama causes Unity Panel/Dash to become all black" [Medium,Confirmed] [12:26] anyone good with unity-2d here ? [12:35] ritz, try #ubuntu-unity rather [12:36] seb128 thank you :) === MacSlow is now known as MacSlow|lunch === Ursinha is now known as Ursinha-afk === MacSlow|lunch is now known as MacSlow === jjardon__ is now known as jjardon === m_conley_away is now known as m_conley [14:55] Has anyone seen "quilt refresh" create invalid patches? [14:55] Something's screwy. [14:56] yes, if your original tarball isn't actually original, or files have been modified before quilt add was done on them. [14:56] Hmm, neither of those here, xnox. [14:57] qengho: in that case define "invalid" =) [14:57] Will do. ... [15:06] xnox, ah, it was me. Nevermind. Updating the path with s! src/!a/src/! (note the missing space) made valid patches for updating, but on emitting refreshed, it doubled-up all blocks. [15:06] I pre-processed wrongly. === rickspencer3_ is now known as rickspencer3 === Ursinha-afk is now known as Ursinha [15:41] is anyone here running precise on real hardware which has multimedia keys on the keyboard? [15:41] https://bugs.launchpad.net/ubuntu/+source/gnome-settings-daemon/+bug/1034090 needs verification [15:41] Launchpad bug 1034090 in gnome-settings-daemon (Ubuntu Precise) "Hotkeys not functional after upgrade to quantal's xorg (new xinput version)" [High,Fix committed] [15:44] seb128, speaking about verifying bugs, how can I test your devhelp upload? I don't see how to enable proposed archive, software-properties-gtk crash at startup. and lp is missing a link to -common .deb [15:45] xclaesse, mkdir test; cd test; wget https://launchpad.net/ubuntu/+source/devhelp/3.6.1-0ubuntu0.1/+build/4024122/+files/devhelp_3.6.1-0ubuntu0.1_amd64.deb https://launchpad.net/ubuntu/+source/devhelp/3.6.1-0ubuntu0.1/+build/4024122/+files/libdevhelp-3-1_3.6.1-0ubuntu0.1_amd64.deb https://launchpad.net/ubuntu/+source/devhelp/3.6.1-0ubuntu0.1/+build/4024125/+files/devhelp-common_3.6.1-0ubuntu0.1_all.deb; sudo dpkg -i *.d [15:45] eb [15:46] xclaesse, the arch-independant binaries (like -common) are built on i386, which is a bit confusing ... you need to go to the i386 build page on launchpad to find those [15:46] seb128, ah ok good to know [15:46] thx [15:46] yw [15:46] xclaesse, what's the issue with software-properties-gtk? can you pastebin the python stacktrace? [15:48] I've told to report the bug when the crash window appear, but it does not seems to do anything... anyway: http://pastebin.com/mLGbejSs [15:48] python3... [15:49] urg, yet another UnicodeDecodeError... [15:49] seb128, FYI devhelp works perfectly :) [15:49] thanks [15:50] xclaesse, excellents, thanks for testing ... can you comment on https://bugs.launchpad.net/ubuntu/+source/devhelp/+bug/1079694 ? [15:50] Launchpad bug 1079694 in devhelp (Ubuntu Quantal) "3.6.1 stable update, fix ctrl-C copy" [Low,Fix committed] [15:51] seb128: I won't be the only one to get it today \O/ [15:52] didrocks, the UnicodeDecodeError? [15:52] yep, but I got one on UTAH [15:52] xclaesse, do you have davmail installed? [15:53] N: Impossible de trouver le paquet davmail [15:54] xclaesse, can you run [15:54] "for F in /var/lib/apt/lists/*Packages; do iconv -f utf-8 -t ucs-2le $F > /dev/null || echo $F; done" [15:55] seb128, done, software-properties-gtk still crash [15:55] xclaesse, did it print any output? [15:55] no [15:55] :-( [15:56] the issue is likely similar to https://bugs.launchpad.net/ubuntu/+source/dpkg/+bug/1053749 [15:56] Launchpad bug 1053749 in ubuntu-drivers-common (Ubuntu Raring) "UnicodeDecodeError from broken package descriptions" [High,Fix released] [15:56] e.g one package has datas in invalid encoding [15:56] xclaesse, does that command prints anything? "grep-available -r . | iconv -f utf-8 -t ucs-2le 1> /dev/null" [15:57] $ grep-available -r . | iconv -f utf-8 -t ucs-2le 1> /dev/null [15:57] iconv: séquence d'échappement non permise à la position 1677550 [16:03] xclaesse, sorry I was talking to somebody [16:04] xclaesse, grep-available -r . | head -c 1677550| tail -n 6 [16:04] Package: eid-mw [16:04] Priority: optional [16:04] Section: libs [16:04] Installed-Size: 2676 [16:04] oh, eid... that's the belgian electronic id card app [16:05] that package is not from archive, I've installed it manually from gov website [16:05] xclaesse, well, in any case that's the issue ... you can probably edit /var/lib/dpkg/status look for eid-mw and clean the nonutf chars [16:06] xclaesse, ah, https://bugs.launchpad.net/ubuntu/+source/dpkg/+bug/1053749/comments/21 [16:06] Launchpad bug 1053749 in ubuntu-drivers-common (Ubuntu Raring) "UnicodeDecodeError from broken package descriptions" [High,Fix released] [16:06] Maintainer: Frank Mari\EBn [16:07] there is the problem :) [16:07] yeah [16:07] xclaesse, edit /var/lib/dpkg/status and maybe /var/lib/dpkg/available as well [16:07] somebody should contact whoever is doing the packaging to tell them that the encoding is broken [16:07] seb128, ok file edited, and now it works [16:07] still a bug in software-properties for not handling it though [16:08] xclaesse, great [16:08] seb128, thanks! [16:08] yw ;-) [16:46] smspillaz: thank you thank you thank you for fixing my pet-peeve compiz bug! (LP: #1037164) [16:46] Launchpad bug 1037164 in compiz (Ubuntu Quantal) "Clicking on semi-maximized windows in a different workspace fails to switch to the correct workspace" [Medium,Triaged] https://launchpad.net/bugs/1037164 [16:49] * mdeslaur hugs smspillaz [16:51] np [16:51] * smspillaz hugs mdeslaur [16:54] smspillaz: remind me, didn't I promise you a bottle of scotch or something if you did? [16:56] I'm having problems with a Marvell 88E8001 GbE NIC on desktop 12.10 amd64 [16:56] can anyone assist me in troubleshooting it? [16:58] oops, and I was typing === mspencer is now known as iBelieve [18:04] * didrocks waves good evening [19:03] anyone else running into issues with archive.ubuntu.com? [19:04] libcupsimage2 and libcups2-dev in particular [19:22] attente, not me [19:27] huh.. thanks kenvandine [19:28] attente, what kind of issues? I haven't had any issues but I also haven't had a printer in 2 months. [19:29] kenvandine, robru, ah sorry, i just forgot to update my pbuilder environment [22:19] robert_ancell: could you take a look at bug 1095092 since I believe you were the last to touch it? [22:19] Launchpad bug 1095092 in libgsf (Ubuntu) "Sync libgsf 1.14.25-2 (main) from Debian unstable (main)" [Wishlist,New] https://launchpad.net/bugs/1095092 [22:19] jbicha, sure === m_conley is now known as m_conley_away [23:28] robert_ancell: Any reason why the accessible description is not translatable? [23:29] Oh sorry, it is. [23:29] DIdn' [23:29] DIdn't see that both source files deal with it. [23:29] Don't mind me.