[00:45] darkxst: You need to set a commit message for https://code.launchpad.net/~darkxst/unity-settings-daemon/lp1224217/+merge/210913 as unity-settings-daemon is managed by the CI landing system. [00:47] darkxst: Also, you don't add a changelog entry, since the CI landing bot creates the changelog when the release is made. The commit message serves as your changelog entry. === stgraber_ is now known as stgraber [01:25] TheMuso, hi [01:25] I had a commit message?, so I just need to remove the changelog entry? [01:27] darkxst: Yes but the commit message needs to be set on the merge page. [01:32] TheMuso, ok, should be fixed now [01:34] darkxst: Right, xnox has also commented. [01:38] darkxst: can you explain the complete set of changes? [01:38] darkxst: when Laney and I were designing this, we assumed _everything_ will want gnome-settings-daemon, but Ubuntu/unity7 session. [01:38] xnox, only GNOME3 sessions will use g-s-d. GNOME2-ish sessions are much closer to Unity, so makes more sense for them to be using u-s-d. [01:39] darkxst: if tht's the case, just flipping the conditions should be fine. But then only u-s-d and g-s-d are affected. [01:39] darkxst: did you already upload anything else? i'd like to review all of proposed changes. [01:40] TheMuso may have uploaded the other bits? [01:40] darkxst: e.g. seeds will need to change as well. [01:40] darkxst: Right, xnox has also commented.I already uploaded gnome-session and gnome-settings-daemon as sponsor. THings made sense to me when I looked things over. [01:40] TheMuso: let me look at the diffs, as that may complete break next image generation =/ [01:40] Fair enough. [01:42] darkxst: TheMuso: right the changelog entires now make sense to me =) [01:42] TheMuso: i'd rather unity-settings-daemon go in as a direct upload into the archive, together with the rest of the uploads. [01:44] TheMuso: darkxst: desktop environments / seeds that use gnome-settings-daemon should probably change to unity-settings-daemon now. [01:44] darkxst: this sort of change should not have been done really, without consulting them =/ [01:44] mitya57 is taking care of the gnome-flashback side [01:44] darkxst: cool. [01:44] and I assume edubuntu seed [01:44] darkxst: i'm not sure who else sould be required to adapt. [01:45] edubuntu: dvd [01:45] ubuntu-gnome: daily-live [01:45] ubuntukylin: daily-live [01:45] darkxst: right edubuntu & ubuntu kylin. [01:45] darkxst: i'll take care of kylin. [01:47] xnox, ok thanks [01:50] I'm not sure what happens if one is supposed to directly override a CI controlled package... [01:51] Or more to the point, what I am supposed to do. [02:27] desrt, what sets XDG_RUNTIME_DIR? [02:27] I'm thinking a PAM module somewhere [02:28] logind [02:30] see pam_systemd(8) [02:34] ta === qengho is now known as CardinalFang === CardinalFang is now known as qengho === Zachary_DuBois is now known as Tokyo === Tokyo is now known as Zachary_DuBois === Zachary_DuBois is now known as Zachary_DuBois|A [05:20] Good morning [06:50] Hi, all, I'm confused about the 'DesktopName' field in gnome session's .session file, for ubuntu defines it as DesktopName=Unity, is there any docs about how it comes or how can i define a new DesktopName, hope for help, thanks! [07:36] penghuan, just don't! the whole desktop thing is a slight mess atm [07:37] there is an XDG list of approved desktops for OnlyShowIn [07:37] which is loosed used for XDG_CURRENT_DESKTOP [07:37] loosely [07:38] beyond that you are guaranteed to break things by using undefined values [07:38] pitti: Good morning! [07:39] hey RAOF, how are you? [07:39] pitti: I don't suppose you've ever tried to record a touchscreen device with umockdev, have you? [07:39] pitti: I'm pretty good. [07:40] How about your fine self? [07:40] RAOF: I thought I did [07:40] RAOF: quite well, thanks! learning lots of cloud stuff technology recently :) [07:41] RAOF: I certainly recorded touchpad events (that's in iumockdev's test suite); shouldn't really be that different? [07:41] Hah! [07:41] Yes, I thought so. [07:41] But my touchpad playback works fine, but when I try to record my mako's touchscreen it all goes doolally. [07:42] RAOF: back when I wrote these I didn't have a linux touch screen at hand; now I have the nexus 4, so I guess I could do some touch recording there [07:42] RAOF: interesting [07:42] As in - I get a lot of touch events, but they're all for the same point. [07:42] RAOF: i. e. evtest itself works, but when you run it on the umockdev playback it fails? [07:42] I'm not sure if this is a bug in my test harness or not. [07:43] RAOF: ah, so you get the right number of events at the right time, but with wrong data? [07:43] Maybe? [07:43] I get roughly the right number of events, but they all seem to have the same coordinates. [07:43] Until I change direction, or something. [07:43] RAOF: maybe some endianess confusion (unlikely), or 32 vs. 64 bit truncation somwhere? [07:44] Yeah, maybe. [07:50] Hm. Running evtest on the umockdev recording makes evtest segfault... [07:59] RAOF: re; trying that now on the G4 [08:02] RAOF: so evtest on the real touch device on mako seems quite fine; I guess that's how you started, too? [08:02] I actually didn't start like that; I just started by recording Mir. [08:03] RAOF: hm, udev's input-id doesn't even seem to recognize teh touch screen as such [08:03] I had expected some ID_INPUT_TOUCHSCREEN=1 [08:05] Yeah, I expected the same. [08:05] But we probably just don't have the appropriate udev rules set up. [08:06] It's not like anyone has ever actually cared that the N4's touchscreen doesn't have the appropriate udev tags before. [08:06] yeah; it just caught my eye [08:06] anyway, mako-touch.script is a wonderful pile of binary garbage; the time offsets look right, though [08:07] phablet@ubuntu-phablet:~$ umockdev-run -d mako-touch.umockdev -s /dev/input/event2=mako-touch.script evtest /dev/input/event2 [08:07] evtest: can't get version: Invalid argument [08:07] RAOF: is that what you saw? [08:08] RAOF: ah, I suspect I need to record/replay the ioctls too, as evdev needs those for the MT extensions [08:08] pitti: Right. I recorded the ioctls too, and then evtest segfaults :) [08:09] $ sudo umockdev-record -i /dev/input/event2=mako-touch.ioctl -s /dev/input/event2=mako-touch.script evtest /dev/input/event2 [08:10] $ umockdev-run -d mako-touch.umockdev -i /dev/input/event2=mako-touch.ioctl -s /dev/input/event2=mako-touch.script evtest /dev/input/event2 [08:10] RAOF: that works just perfectly here [08:10] and of course the initial umockdev-record /dev/input/event2 > mako-touch.umockdev [08:10] pitti: Hm. I'll try recording again... [08:12] RAOF: note that .script files are appended, so better remove them first [08:12] (.ioctl too) [08:14] pitti: I get http://paste.ubuntu.com/7088883/ [08:15] (It segfaults at about the time the first event would be sent. [08:15] ) [08:15] all these errors at the top shouldn't be there really [08:16] Input device name: "touch_dev" [08:16] RAOF: it shouldn't be "unknown", and the sysfs bits exist; I wonder what went wrong there [08:16] RAOF: how does your .umockdev look like? [08:16] phone, brb [08:17] pitti: http://paste.ubuntu.com/7088898/ [08:20] RAOF: that looks identical to mine [08:25] pitti: So, um...? What's going on? :( [08:26] RAOF: just to confirm, are you running umockdev-run on the N4 as well, or on your laptop? [08:26] Ah. [08:26] Running umockdev-run on my laptop. [08:26] Are you running it on the N4? [08:26] ah [08:26] yes [08:27] So, I'd very much like this to work on my laptop, because I'd very much like this to work on our all our CI infrastructure :) [08:27] RAOF: right, I get the same error here [08:27] Right. We've got some architecture dependencies somewhere. [08:27] HUZZAH! [08:31] good morning [08:31] hey chrisccoulson [08:31] i'm finally upgrading to trusty! https://twitter.com/chrisccoulson/status/444389398470086656 [08:32] this must be the latest i've ever upgraded :/ [08:32] oh, wow [08:32] what's saucy? [08:32] heh [08:33] pitti, the upgrader is inaccurate - it took just under 7 minutes to download the update, not 10 https://twitter.com/chrisccoulson/status/444388773149687808 ;) [08:34] chrisccoulson: OMG - critical bug! [08:34] heh [08:34] chrisccoulson: so if that's the biggest problem that you've found → WIN! :) [08:40] $ src/umockdev-run -d /tmp/mako-touch.umockdev -- udevadm info --export-db [08:40] ERROR: uevent_sender_send: No such device /sys/devices/virtual/input/input2/event2 [08:40] ERROR: uevent_sender_send: No such device /sys/devices/virtual/input/input2 [08:40] RAOF: ^ so this reproduces the first errors, and is a smaller problem; looking into that now [08:41] RAOF: it looks like some fallout from the recent "send add uevent for device additions" [08:41] Probably so. [08:46] RAOF: ok, I'll have a look at this today; this is excellent offline hacking material for my train ride this afternoon :) [08:46] pitti: Superb! [08:57] good morning desktopers! [08:58] grrrr [08:58] TheMuso, darkxst: no no no, doing manual landing of unity-settings-daemon which is under CI landing is not cool, it creates extra work for us. Also those changes seem like they should have been discussed on the mailing list or including somebody from desktop firts [09:00] seb128, I did chat with pitti about it [09:00] and I have no idea about the CI stuff [09:02] seems like TheMuso told you it was under CI landing (reading night logs) [09:02] yeh, then xnox told him to push it manually [09:02] TheMuso, "I'm not sure what happens if one is supposed to directly override a CI controlled package..." [09:02] don't [09:02] right [09:02] because the other ones got uploaded in an uncoordinated fashion [09:03] which forced you to make extra wrong to fix it [09:03] shrug, more work for me today to fix the mess I guess [09:04] seb128, I fail to see how that is my fault [09:05] it's not, it's TheMuso's I guess [09:08] seb128, Laney, I know its getting pretty late but are you willing to ack bug 1228765 [09:08] Launchpad bug 1228765 in unity-control-center (Ubuntu) "[FFe] Implement DisplayConfig dbus interface and transition to gnome-desktop 3.10" [Low,In progress] https://launchpad.net/bugs/1228765 [09:10] I didn't change my position since last time, I think it's late and it's wrong to do for a LTS quality wise for Ubuntu [09:10] but I would like to accomodate GNOME Remix as well [09:10] otherwise I would already have nacked it [09:10] so I'm a bit unsure what to do [09:11] do you have a ppa current that I can optin for to test the update? [09:13] * seb128 tries to under the gnome-session change from DESKTOP_SESSION to XDG_CURRENT_DESKTOP [09:27] seb128, no current ppa, it was too much work keeping it up to date with all the churn relating to u-s-d/u-c-c etc [09:28] seb128, DESKTOP_SESSION is to granular, ultimately atleast currently it comes down to run u-s-d or run g-s-d [09:29] once the legacy code is dropped other gnome2 sessions will be using u-s-d [09:29] I hope the DESKTOP_SESSION change doesn't create issue for e.g Kylin or such, who use unity but maybe with a different .session because they have different required components [09:29] kylin should not be using g-s-d [09:29] you changed the semantic of the job [09:29] DESKTOP_SESSION is not used only by g-s-d/u-s-d [09:29] that can be used by any job [09:29] some that might not even be in the archive [09:29] that's not of an ABI change [09:29] seb128, the upstart instance is *only* used by g-s-d/u-s-d [09:30] how do you know? [09:30] I checked [09:30] " some that might not even be in the archive" [09:30] you can't check local systems for all our users [09:30] some might have written jobs relying on the old semantic [09:32] seb128, you guys break our stuff, that is in the archive from time-to-time, yet you expect me to grep users local machines or something? [09:33] no [09:33] I'm just saying that it's a behaviour change, it's like changing an API in an incompatible way [09:34] that should ideally not happen after ff [09:34] I understand why you did it [09:34] that doesn't mean it's risk free or right, or that we couldn't have find another way [09:34] well, let's see if anyone complains [09:36] ok [09:37] yay, everything still works \o/ [09:38] chrisccoulson, upgraded to trusty? ;-) [09:38] chrisccoulson, hey, how are you? [09:38] seb128, yeah, i did, finally :) [09:38] i'm not too bad thanks. how are you? [09:38] hm, who can i talk to about unity's new window-decorator? [09:40] ochosi, Trevinho but he's off skiing until eow [09:40] ochosi, but just ask your question, he might read backlog in between or others might be able to reply [09:40] chrisccoulson, I'm good thanks, it's sunny friday ;-) [09:40] chrisccoulson, I saw you on video in a vUDS session! [09:40] lol [09:40] ok, thanks seb128 :) [09:41] seb128, that was the only session i participated in all week ;) [09:41] Trevinho: hey, i wanted to ask whether there's a way to have bright window-deco and dark panel with unity's new window-decorator? seems like it's not behaving as i'd have expected: https://github-camo.global.ssl.fastly.net/8b0e0720c71a10a947fdcddb0ab0932aed4a2757/687474703a2f2f692e696d6775722e636f6d2f79723774544a342e706e67 === sil2100_ is now known as sil2100 [10:01] https://bugs.launchpad.net/ubuntu/+source/unity/+bug/1292442 [10:01] Launchpad bug 1292442 in unity (Ubuntu) "lockscreen: menus interactable but render behind the lockscreen" [Undecided,New] [10:04] shrug [10:04] bregma, Trevinho: ^ [10:04] must be a kernel issue :) [10:05] seb128: the event semantics were only introduced recently at the sprint, just a few weeks back. [10:05] seb128: and indeed specifically for gsd/usd [10:06] xnox, there is no written rules that they can't be used for custom jobs [10:06] seb128: in-development releases are not supported. [10:07] xnox, sure, let me change a glib api that was added this cycle the week before release, and you tell me how that goes :p [10:08] seb128: the event stayed, and parameters as well. [10:09] https://bugs.launchpad.net/ubuntu/+source/unity/+bug/1292446 [10:09] Launchpad bug 1292446 in unity (Ubuntu) "lockscreen: on autolock on suspend the fade does not complete before suspend leaving desktop partially visible for a short period on resume" [Undecided,New] [10:10] apw, just using launchpad is fine you know ;-) [10:11] xnox, what are you arguing for? that changing semantics of our jobs, apis, etc after ff is fine? [10:12] seb128, heh ... i tend to lose bugs in launchpad so i like people bringing them :) [10:12] seb128: apis -> no, events -> yes. we only guarantee a small stable subset of the job events both on the system and session init, and to my knowledge we never broke that. The "Well-known events" in man 7 upstart-events. [10:13] e.g. dbus, dconf, desktop-start, desktop-end, file, session-end, startup, :sys:*, xsession. [10:14] apw, yeah, fair enough, thanks for testing reporting those btw! (I can't believe that bregma and co didn't even try opening an indicator before landing the update :/) [10:14] seb128: and re:apis, we are in process landing qt5.0->5.2 merely weeks before release ;-) [10:14] xnox, but it got prepared and got a ffe [10:15] xnox, also qt releases are supposed to be api/binary compatible (if you omi the qreal change on arm) [10:15] seb128, i am thorough, it is in my nature, else i couldn't look after a kernel :) [10:15] seb128: here it's a bugfix, since wrong s-d got used under some environments. [10:15] apw, seb128 would always beat the LP bug DB in a contest, he knows all desktop bugs from the top of his head ;) [10:16] ogra_, yeah i have somone on my team like that, i forget the ones i am activly working on and have to look them up all the time, numbers are nightmare things [10:16] * ogra_ learned long time ago that asking him is faster than the LP search page ;) [10:17] apw, same here :) [10:17] xnox, let's not argue about it, we obviously have different view of what it takes to have a solid LTS out (I might be too conservative, or might not be enough ;-) [10:19] heh, well it _is_ your product =) and i've never released an LTS yet. [10:19] seb128: let me file an FFe about the plymouth-upstart-bridge rewrite and plymouth ubuntu-logo theme rewrite. [10:19] lol [10:19] seb128, oh, i like the new lock screen ;) [10:20] chrisccoulson, thanks, kudos to the unity guys ;-) [10:20] i should have upgraded earlier, i've been missing out on all this stuff [10:24] seb128, is it known that the timer setting in settings (for the lockscreen) is not working? [10:24] ie. it says "lock after 30s" and after several minutes it has not yet done so [10:25] apw, what settings? [10:25] what panel rather [10:26] the privacy one or the lock one? (that's a known duplication/issue, I think privacy is buggy) [10:30] Anybody here has HiDPI machine and wants to test a scaled up plymouth theme? [10:31] * seb128 doesn't have an Hi-DPI machine [10:32] expense one ! [10:40] xnox: any changes coming up in plymouth that flavors should adapt their themes to? [10:40] (just asking cause i recently got this bug reported: https://bugs.launchpad.net/bugs/1220076) [10:40] Launchpad bug 572155 in plymouth (Ubuntu) "duplicate for #1220076 Splash image wrong size when two displays connected" [Undecided,Confirmed] [10:41] oh, that one has meanwhile been marked... [10:44] ochosi: no, not really. I only tweaked ubuntu-logo theme, to do Image.Scale () on images to get them the right size on HiDPI screens and that's it. [10:45] seb128, " [10:45] ochosi: there is no automatic scaling factor detection, yet. thus one could copy/mimic the static upscale, or wait for scaling factor detection. [10:45] seb128, "Brightness and Lock" has 30s on it on mine, and its not 30s for sure [10:50] seb128: https://bugs.launchpad.net/ubuntu/+source/plymouth/+bug/1292458 [10:50] Launchpad bug 1292458 in plymouth (Ubuntu) "[FFe] a bunch of updates to plymouth" [Undecided,New] [10:50] apw, that's an ages old bug [10:50] apw, lock time is added to the "turn off screen after" one [10:50] apw, so "30s" is "30s after screen is off" [10:50] (I know it's confusing) [10:51] so we so should add "a further" to the text for that [10:52] apw, yeah, we should probably do that [10:52] I think we discussed that with mpt some cycles ago, need to find the bug back/his recommendations [10:54] apw, https://bugs.launchpad.net/ubuntu/+source/gnome-control-center/+bug/1063682 [10:54] Launchpad bug 1063682 in gnome-control-center (Ubuntu) "Screensaver settings GUI is confusing" [Low,Triaged] [10:59] seb128, the other commenters on there have a point that if they were independant times, things would work a lot better, and a lot more like people expect. as i want my screen to lock after 5m and only power off after 30m to save my screen being turned on and off alllll the time [11:00] apw, right, I'm unsure why GNOME made it this way [11:00] we should look at fixing it, we just never found the free slots to do that :/ === jjardon__ is now known as jjardon === Sweetsha1k is now known as Sweetshark [12:01] seb128, do you know if it is possible to add mouse bindings to window titles any more, i used to have a "lower this window" binding on the title bar, and it went away with the recent updates [12:03] apw, I don't know sorry, Trevinho would know (but it's on vac this week) === MacSlow is now known as MacSlow|lunch === alan_g is now known as alan_g|lunch [12:42] xnox: ah ok, well good to know, i might apply the same method on xubuntu's plymouth theme then. thanks! === alan_g|lunch is now known as alan_g === m_conley_away is now known as m_conley === MacSlow|lunch is now known as MacSlow [14:01] seb128: I wonder how exactly I would add the caladea/carlito fonts to libreoffice. my instinct would tell me to add them as a 'Suggests:' to libreoffice-writer or libreoffice-core. But then they would need to explicitly be added to the default install. Also we currently have no font suggests on either package currently, so this feel a bit odd. Suggestions? [14:10] Sweetshark, well, those are used by writer no? why would writer recommends the frequently used fonts? [14:22] seb128: the issue is not that they are intended to be user for user because they are pretty or something. Its that these fonts are metric compatible fallbacks for fonts distributed by default with new MS Windows versions. [14:23] Sweetshark, right, so it seems like you want those coming with writer to get a good experience, so writer should recommends them [14:23] seb128: k, will do so. ;) [14:24] although .. [14:25] the same applies for e.g. impress -- no fallback font: suddenly text is bigger/smaller on the slides. (thus I considered to add it to libreoffice-core, not -writer). [14:28] well, you know better than me where things are useful [14:29] if normal users would benefit to have those fonts coming with any libreoffice component, make code recommends it [14:29] it->those [14:33] seb128: k [14:34] desrt, hey, happy friday! [14:34] desrt, new glib is in trusty since yesterday evening, no complain so far, so looks good [14:35] desrt, I can't tell yet if the reports stopped, but we should have a better view of that next week [14:35] reports of the issue it's supposed to fix [14:35] (we didn't get any today, but it might be just that nobody got the update & hit the bug, we didn't get reports from that issue that often) [14:39] seb128: i noticed the landing from the bug report [14:39] so i pushed the patches upstream already [14:39] great === sil2100_ is now known as sil2100 === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha [16:39] pitti: hey. around? [16:41] pitti: i'm wondering what the state of jhbuild-on-ubuntu-via-jenkins is... it doesn't look great [16:55] didrocks: I have my ex boss pinging me about Ubuntu Touch ... [16:55] Sweetshark: yelling at you? :p [16:57] didrocks: "Tried the image, all looking quite nice. But of course, one wants to have a closer look then: I found the build instructions kinda hard to dig through thogh ..." [16:57] ah, this is not me :p [16:59] didrocks: so essentially he is asking for help, but at also seems to want to invest some time (and possibly contribute). is there any ML list or other channel for getting started there? [16:59] didrocks: or to put it different: if its not you, who is it? [16:59] :D [16:59] Sweetshark: yeah, the best way is the ubuntu phone ML [16:59] Sweetshark: tell him to join https://lists.launchpad.net/ubuntu-phone/maillist.html [17:02] didrocks: alright, thanks! [17:02] yw ;) [17:02] ah, he seemed to have joined lp and the phone team already: https://launchpad.net/~mathias-bauer [17:03] Sweetshark, yeah, he made some noise already :) [17:04] i think sergio even tried to help him [17:04] (not sure why anyone would want to build from source though ... it isnt an easy task) === mjohnson15_2 is now known as mjohnson15 [17:25] * Sweetshark tests the build from the ppa with autopkgtests in a VM. *excited* [17:31] "after this operation 5137MB of additional discspace will be used" <- libreoffice autopkgtest installing its deps. === mhr3__ is now known as mhr3 [17:59] *grumble* no space left on device *grumble* === alan_g is now known as alan_g|EOW === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha === Ursinha is now known as Ursinha-afk === gatox_ultra is now known as gatox === Ursinha-afk is now known as Ursinha === gatox_ultra is now known as gatox === gatox_ultra is now known as gatox === m_conley is now known as m_conley_away === tkamppeter_ is now known as tkamppeter === e11bits_ is now known as e11bits === qengho is now known as CardinalFang === CardinalFang is now known as qengho === jasoncwarner___ is now known as jasoncwarner === jjardon__ is now known as jjardon