[06:40] morning o/ [07:09] good morning desktoppers [07:15] good morning! [07:18] salut didrocks [08:00] tjaalton, patch attached. Use it however you want... https://bugs.launchpad.net/ubuntu/+source/mesa/+bug/1753776 [08:00] Ubuntu bug 1753776 in mesa (Ubuntu Bionic) "Graphics corruption just before login animation to Xorg sessions (Intel gen9 GPUs only)" [Medium,In progress] [08:01] Morning tsimonq2, oSoMoN, didrocks [08:02] duflu: thanks, I was about to just revert the commit, but this is smaller [08:02] 18.0.0 should be released RSN [08:02] Yeah, smaller patches are nicer to maintain === pstolowski|afk is now known as pstolowski [08:40] duflu: https://bugs.launchpad.net/ubuntu/+source/gdm3/+bug/1754218 => ping me if you want me to debug this a bit more [08:40] Ubuntu bug 1754218 in gdm "[regression] gdm 3.27 shows blank login screen on primary monitor when two monitors are connected on a hybrid intel/nvidia system" [Medium,Confirmed] [08:40] quite annoying bug :D [08:41] dupondje, "want" is a strong word. I was only trying to triage/sort it, rather than fix it [08:41] :) [08:45] hehe, I,m trying to get it fixed [08:48] Mar 21 09:48:05 lt-jeanlouis gnome-shell[1598]: Failed to set CRTC mode 1920x1080: Invalid argument [08:48] meh :( [08:59] dupondje, sounds like the eglnative backend. Try using a Xorg session instead of Wayland [08:59] * duflu hops into a meeting [08:59] yep Xorg work but missing fractional scaling there :) [09:00] dupondje, interesting, maybe ask Trevinho why [09:01] I only have trouble with the cursor failing to scale in Xorg sessions, and only on shell elements [09:02] ey [09:09] ahasenack, morning! Could you test 1.10.3 of libinput (should be in bionic today - might need to wait for mirrors to update)? (re: swipe of death) [09:10] morning Laney [09:11] morning Laney, willcooke, duflu [09:11] Morning didrocks [09:15] Hello all [09:16] hey alexarnaud didrocks [09:16] duflu: https://gitlab.gnome.org/GNOME/mutter/issues/18 seems like this one [09:17] dupondje, I think that fix is already in mutter 3.28.0 so you might want a different bug [09:18] Yes, bionic already has that fix [09:19] After discussing accessibility with didrocks yesterday, I would like to tell you the following things: the installer of Ubuntu 18.04 is no longer usable with a screen reader due to those two bugs: [09:19] https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/1741690 [09:19] https://bugs.launchpad.net/ubuntu-mate/+bug/1741507 [09:19] Ubuntu bug 1741690 in ubiquity (Ubuntu) "You can't enable the Orca screen reader until after you click "Try Ubuntu" on Ubuntu bionic" [Undecided,Confirmed] [09:19] Ubuntu bug 1741507 in ubuntu-mate "Can't start screen reader with Ctrl+S and Alt+Super+S" [Undecided,New] [09:19] usable with a screen reader means a visual-impaired will not be able to install the next LTS [09:20] For a visual-impaired perspective it is as grave as if the screen doesn't work. [09:20] duflu: Inded :( [09:21] xnox: are you able to have a look at what changed in the ubiquity installer itself ^ [09:21] alexarnaud: have yo utried on 17.10 as suggested? [09:21] I'll try, it's downloading. [09:21] that will help to know wheb that started [09:21] tanks! [09:21] thanks* [09:21] let's see with xnox [09:22] dupondje, I think we might have steered into a dead end asking for a fix in gdm. Mutter is likely to blame so please also log it here: https://gitlab.gnome.org/GNOME/mutter/issues [09:22] yea i'll do :) [09:23] I assume Jonas or someone there has the hardware, since they wrote the hybrid support [09:26] almost every laptop has hybrid these days ... :) [09:26] not that exceptional [09:28] dupondje, actually most laptops don't :) [09:29] Hybrid GPUs makes them more expensive, heavier, more power hungry [09:29] At least a little. So for the sake of cost and margins you will find most are not hybrid [09:30] i'm glad lenovo t4x0s dropped the dgpu ages ago [09:40] how can I make mutter to print some debugging ? :) [09:41] cause damn its the hell :( connecting second screen while gdm3 is running just crashes it, and unable to recover (so hard reboot needed) [09:47] argh, cannot unlock my encrypted usb drive :((( [09:47] bug 1757321 [09:47] bug 1757321 in udisks2 (Ubuntu) "Can't mount encrypted USB drive after upgrade to bionic" [Undecided,New] https://launchpad.net/bugs/1757321 === ShriHari is now known as IDK === IDK is now known as Guest88476 === Guest88476 is now known as IDKIDKIDK [09:56] looks like we need https://bugs.launchpad.net/ubuntu/+source/volume-key/+bug/1754422 and then promote libblockdev-crypto2 and make udisks2 pull it in [09:56] Ubuntu bug 1754422 in volume-key (Ubuntu) "[MIR] volume-key" [Undecided,Incomplete] [09:56] dupondje, you have to rebuild it with --enable-debug AND run it with various environment settings :P [10:00] didrocks, sorry, it was never me that had to deal with these things. [10:01] didrocks, it sounds to me like e.g. gnome-settings-daemon (or equivalent of thereof) is not started correctly and/or not responding to those shortcuts correctly or some such =/ [10:01] xnox: could be, who did work on the ubiquity installer session? can be a missing g-s-d, I guess the last one to look at it was seb [10:02] alexarnaud: anyway, I'll keep an eye on it, seb is on holidays and we'll discuss it next week, mind tagging it as rls-bb-incoming? [10:02] didrocks, well, i think mate has different set of "settings-daemons" [10:02] hum, true, unsure for mate, flexiondotorg ^ [10:02] so, quite a coincidence that both breaks [10:02] xnox: the shortcut is broken on Mate and GNOME installer [10:03] didrocks: I assume it's not a coincidence. [10:03] hence my sentence… [10:10] didrocks, xnox: same bug on Ubuntu 17.10.1 [10:11] alexarnaud: thanks for confirming! Sad that nobody reported it by then though :( [10:11] *sigh* [10:13] right but the bug 1741690 exists since the beginning of the year [10:13] bug 1741690 in ubiquity (Ubuntu) "You can't enable the Orca screen reader until after you click "Try Ubuntu" on Ubuntu bionic" [Undecided,Confirmed] https://launchpad.net/bugs/1741690 [10:15] didrocks: Maybe it could be great to have a test to prevent such regression. Do you know if tests exists in ubiqity ? [10:16] alexarnaud: there are some autopilot tests, jibel would know more. Interested in contributing? [10:16] alexarnaud: basically python code [10:17] didrocks: I'm interesting indeed but we're so few people involved in a11y QA and I'm not sure to have enough time to take to this in my spare time (Ubuntu helps is only in my spare time) [10:18] didrocks, alexarnaud the shortcut isseu might be bug 1750846 [10:18] bug 1750846 in gnome-settings-daemon (Ubuntu Bionic) "several g-s-d services are timing out when booting a live session" [High,New] https://launchpad.net/bugs/1750846 [10:18] jibel: sounds to impact as well 17.10 though (and we didn't have that timeout issue, did we?) [10:19] didrocks, I don't know, we had several issues with a11y in 17.10. [10:22] didrocks, alexarnaud in 17.10 shortcuts worked but then there was bug 1719995 [10:22] bug 1719995 in ubiquity (Ubuntu) "Screen reader does not read Ubiquity windows" [High,Confirmed] https://launchpad.net/bugs/1719995 [10:24] jibel: On my test VM, I'm unable to launch the screen reader on the first welcome screen. [10:25] he bug you describes is about the live session [10:25] *the [10:25] didrocks, sounds like unit ordering, and/or bug in ubiquity-dm, as the Try Ubuntu runs before the full session is started.... [10:25] hence things might be missing.... [10:27] I don't have the time to look at it right now, let's wait for seb to be back === ShriHari is now known as Ubuntu_admin [11:04] didrocks: what you mean for "maybe with a "sendMessage = True" for the 2 first cases" [11:07] andyrock: you have 3 conditions here [11:08] you only send the message on the first 2 ones if I read it correctly [11:08] so, set a flag to tell later on "do the dbus call" [11:10] dupondje: https://bugs.launchpad.net/ubuntu/+source/mutter/+bug/1757401 [11:10] Ubuntu bug 1757401 in mutter (Ubuntu) "hybrid GPU: one screen stays black in mutter 3.28" [Undecided,New] [11:10] and https://gitlab.gnome.org/GNOME/mutter/issues/81 [11:10] cause hell, thats an annoying bug :) [11:10] mmm didrocks as you prefer [11:11] andyrock: it's just to have complex requests one [11:11] andyrock: but nothing mandatory for that one, it's a personal preference [11:11] as long as you fix the timeout issue, I'm fine :) [11:19] Laney, first refresh happens 5 min after initialization of the core snap [11:23] dupondje: try latest mainline kernel [11:24] didrocks: pushed! [11:25] Laney, er, refresh happened 2s after I set the hold value [11:25] https://paste.ubuntu.com/p/7M48WMzZRn/ [11:25] bad luck? [11:25] tjaalton: i'm running the latest (with drm-tip) :) [11:25] ok [11:26] cause without some recent kernel my system locks up when I connct a second screen :) so its a requirement :D [11:30] blacklist nouveau? [11:30] andyrock: excellent! Did you look at all templates content for translations? [11:31] for what I can see are all there [11:31] tjaalton: guess then I can't even use my external screen :) [11:31] tjaalton: and it worked fine on 17.10 ... so why would it be a kernel issue? [11:33] because of a regression? [11:33] does 4.13 work then? [11:34] 4.14 you mean? [11:34] no, 4.13 from artful [11:35] Laney, mvo there is something weird. On a live session, it seems that setting refresh.hold triggers a refresh [11:35] jibel: hmm, how did you set that? [11:35] mvo, Laney https://paste.ubuntu.com/p/CKmgFZ8vHV/ [11:35] just manually? [11:35] yes [11:35] andyrock: on line 1553, enabled is false by default? [11:35] (I only see that you added token='') [11:35] that is not good [11:36] I tried twice and there is a refresh event immediately after setting the config option [11:36] tjaalton: I did already run daily kernel on artful, and there it worked :) [11:36] hold... but in a minute... [11:36] might want to tell mwhudson this [11:36] andyrock: apart from that one, all looks good to me (need some manual testing on my side) [11:36] so guess its some other component (mutter?) that broke it [11:36] yup moving to #snappy [11:38] * Laney lurks [11:38] kk livepatch seems broken right now... some problem with "execl" [11:38] not sure what's going on [11:38] but it's on their side [11:39] andyrock: ok, and on my remark for enabled=false? [11:39] fixed [11:39] I was not really able to test that change because I cannot enable livepatch anymore :D [11:40] andyrock: tssssss ;) at least, peer reviews are for this! :) [11:41] andyrock: do you have an ETA for them fixing it? [11:41] I wonder if I should wait for testing or not [11:41] nope maybe on the command line "canonical-livepatch status" [11:42] and check if it works [11:42] maybe it's just here [11:42] i mean it's broken just here [11:42] andyrock: ah, so the dbus call should work, and status would change? [11:42] will give it a try anyway (not right now, but later) [11:42] you'll get the error dialog [11:43] ok, at least, testing the error path this way :) [11:43] I guess this will need a UIFe [11:43] andyrock: nice work, and thanks for addressing the issue! [11:43] andyrock: well, depends if the FFe is acked, I can upload it [11:43] as UIFe is tomorrow [11:43] andyrock: maybe chase someone from the release team? [12:00] I'm the only one that cannot dismiss low-battery notification in bionic? [12:13] didrocks: ffe approved [12:18] andyrock: nice! :) [12:20] andyrock: I'm changing 0.96.24.25 by 0.96.24.24 [12:20] kk [12:25] andyrock: hum, layout is weird [12:25] screenshot? [12:25] * didrocks reruns with LANG=C [12:26] * didrocks grr at recursive search blocking everything in nautilus [12:26] andyrock: https://imgur.com/a/qWpRD [12:27] mmm that's the wanted layout [12:27] looking for the mockup [12:28] https://wiki.ubuntu.com/SoftwareUpdates [12:29] didrocks: ^^^ [12:30] andyrock: mind sharing my screenshot with mpt? [12:31] andyrock: I think he wanted that be aligned with leftmost content [12:31] from this mockup [12:31] mpt? [12:31] let's see if he answers here [12:31] andyrock: there is also another couple of question, I just commented on the MP [12:32] popey, did you ever get an answer to https://forum.snapcraft.io/t/docs-examples-for-build-snaps/2433 ? [12:32] andyrock: https://code.launchpad.net/~azzar1/software-properties/add-canonical-livepatch/+merge/341427/comments/893290 [12:32] andyrock: I think #2 will be with mpt [12:32] I would appreciate an answer for #3 [12:32] quite lost [12:32] didrocks: there never was a design for the login dialog [12:33] i tried to get one but never a respose [12:33] ok, let's see what he will say thus [12:33] oSoMoN: hah, no [12:33] andyrock: and in g-c-c? I have 2 accounts setup, but unsure what to do [12:33] also with the theme fixes it gets better [12:33] just pick one [12:33] pick, like "click"? [12:33] you should get both of them in the list [12:34] clicking just say "use this account for mail/agenda/…" [12:34] it's not really a pickler [12:34] do you mind sharing a screenshot? [12:34] of what you get? [12:35] are you running on bionic? [12:35] yes [12:35] sure, let me see if I'm just not up to date and there is a g-c-c I'm missing [12:35] you've to pick directly from the list [12:35] willcooke: updating libinput got rid of my X crash, I updated the upstream bug [12:36] I have the default u-o-a panel to configure/add/remove accounts [12:36] without opening g-c-c [12:36] https://bugs.launchpad.net/ubuntu/+source/libinput/+bug/1751086 can be closed then? [12:36] Ubuntu bug 1751086 in libinput (Ubuntu) "Xorg assert failure: Xorg: ../src/evdev-mt-touchpad-tap.c:1002: tp_tap_handle_state: Assertion `tp->tap.nfingers_down > 0' failed." [High,Confirmed] [12:36] there is a new g-c-c [12:36] maybe that's what I'm missing [12:36] * didrocks upgrades [12:38] https://usercontent.irccloud-cdn.com/file/oIb1Q86T/image.png [12:38] this is what I get [12:38] you just need to select one of the accounts [12:39] andyrock: ah, you saw what I have in the picker (see my second link) [12:39] which is what looked weird "Sign in" [12:39] I think it's a missing g-c-c, in that case, it means you need a dep [12:39] yeah because you've no ubuntu sso configured in gnome-online-acocunts [12:39] so you need to sign-in before [12:40] hum, I have my canonical account, isn't that an ubuntu sso account? [12:40] do you have it configured in gnome-online-acocunts? [12:40] yes [12:40] as said in my comments [12:41] - once I end up in UAO, I don't really know what to do, I have 2 accounts (my personal Google one and my canonical one) which are already subscribed… Clicking on them doesn't mention livepatch or anything at all, what should I do? [12:41] do you mind sharing a screenshot of the online-accounts panel in g-c-c? [12:42] andyrock: there are my accounts, let me blur it thus first [12:42] didrocks: kk [12:47] andyrock: https://imgur.com/a/y3JdU [12:47] you've a google account not an ubuntu sso account [12:47] you should have something like that [12:48] https://usercontent.irccloud-cdn.com/file/PyRjGM01/image.png [12:49] Laney: aisleriot & almanah need to have asgen rerun against them in bionic because their screenshots were accidentally deleted from the web earlier (but they're back now) [12:49] it has been uploaded one month ago [12:49] Laney: can you handle that or do you want me to just upload no-change rebuilds for those 2 packages [12:49] didrocks: ^^^ [12:50] andyrock: what has been uploaded? So, I need to remove my google account and add an SSO one? [12:50] jbicha: ok [12:50] didrocks: no need to remove the google account, that for gmail (email, etc) [12:50] *that's [12:50] google calendar integration in gnome-shell ftw [12:50] didrocks: the new gnome-online-accounts should be already in bionic [12:51] andyrock: ok, so I think something should tell "select " [12:51] it's quite puzzling [12:51] (in the previous sign in dialog) [12:51] we can remove the "settings" button [12:52] you don't select from g-c-c [12:52] but from the list [12:52] yeah, removing the settings button I would say and in the "sign in", tell to create/select an ubuntu SSO account [12:52] or you use sign-in button to login [12:52] yeah, but I need to create the association in g-c-c first, correct? [12:52] but you still got a problem [12:52] didrocks: nope [12:52] hum, puzzled [12:53] didrocks: you don't have an ubuntu-sso-provider [12:53] can you check the gnome-online-acocunts' version? [12:54] ah it's still in proposed? :O [12:54] 3.27.92-1 [12:54] ah [12:54] weird [12:54] so, updated bionic wasn't enough [12:54] yeah sorry [12:54] and you miss a versioned dep [12:55] not sure why [12:55] I'll fix this [12:55] let me get it from proposed thus [12:55] and retry :) [12:55] I'll remote the settings icon [12:55] yeah ;) [12:58] andyrock: ah, way better! [12:59] ok let me have lunch and I'll push the fixes [13:00] the problem with all this is that multiple passwords are required [13:00] one to login inside snapd, the other to enable livepatch [13:00] andyrock: ah, this is why it asked for 2 passcodes [13:00] ok, but then [13:00] the window closed [13:00] and I have again the "Sign in" dialog [13:01] didrocks: are you using a yubikey? [13:01] andyrock: a manual counter, but similar [13:01] ok, so signing in didn't change the "select account dialog" [13:01] and it asks me again to sign in… [13:01] didrocks: mmm I guess you need to restart gnome-online-accounts daemon [13:02] I am getting this stacktrace: [13:02] Traceback (most recent call last): [13:02] File "/usr/lib/python3/dist-packages/softwareproperties/gtk/DialogAuth.py", line 199, in _listbox_accounts_row_activated_cb [13:02] self._spawn_goa_with_args('add', 'ubuntusso') [13:02] File "/usr/lib/python3/dist-packages/softwareproperties/gtk/DialogAuth.py", line 138, in _spawn_goa_with_args [13:02] proxy.call_sync('Activate', param, Gio.DBusCallFlags.NONE, -1, None) [13:02] GLib.Error: g-io-error-quark: Le délai d’attente est dépassé (24) [13:02] ohhh [13:02] you're probably right [13:02] ok done, let's retry [13:04] ok, now checkbox is fine [13:04] and then failure [13:04] execl fails [13:04] I guess that's what you meant? [13:04] yeah not our fault [13:05] not sure it's a problem in snap [13:05] the stacktrace is linked to that, I guess? ^ [13:05] or in livepatch [13:05] you are seeing it as well, correct? [13:05] the "execl fails" things right? [13:05] you should get an ugly error dialog [13:05] yes, but also a stacktrace ^ [13:05] the one I pasted 3 minutes ago [13:06] yeah but it was linked to the fact gnome-online-accounts deamon was not restarted [13:06] that was after [13:06] something to fix in goa post-init [13:06] after a fresh restart, I still have this [13:07] mmm not sure [13:07] I'm not getting this [13:07] I'll check after lunch [13:07] sure :) [13:08] writing the summary on the MR [13:13] ok, should be summarized [13:14] time for a break soon for me as well :) [13:22] mpt: if you're here please can you take a look at this: https://code.launchpad.net/~azzar1/software-properties/add-canonical-livepatch/+merge/341427 [13:30] Laney, have you had any problems with suspend/resume on your XPS13? [13:31] willcooke: no not really [13:31] I had some resume failures a few months ago but not lately [13:33] cool, thanks Laney [13:46] GunnarHj: I think I'll just let robert_ancell handle your gnome-software translations merge proposal [13:49] jbicha: I called his attention to it here yesterday, response "makes sense". Thought he would merge, but he didn't, and UI freeze is tomorrow... [13:49] hmm [13:51] could you email him then and point out the deadline. If he doesn't get to it, we can upload anyway tomorrow [13:56] jbicha: Sure, I could do that. OTOH why does it matter who merge it? Do you have any doubts about the proposal? [13:56] it's just that it's two parts, the d/patch part needs to be pushed to our GNOME gitlab branch also [13:57] he's been handling most of the gnome-software uploads lately [13:58] jbicha: Ok, I'll send him a note mentioning UI freeze. [14:01] jbicha: Sounds like there is one repository too much, btw. ;) [14:02] we might drop the extra one when we move our work from bzr to git next cycle [14:07] Laney, did you reproduce bug 1750465 ? [14:07] bug 1750465 in plymouth (Ubuntu Artful) "upgrade attempting to process triggers out of order (package plymouth-theme-ubuntu-text 0.9.2-3ubuntu17 failed to install/upgrade: dependency problems - leaving triggers unprocessed)" [Undecided,In progress] https://launchpad.net/bugs/1750465 [14:28] tjaalton: found already the commits that broke it :) so just need to get this fixed now :) [14:28] dupondje: gnome? [14:29] mutter [14:29] https://gitlab.gnome.org/GNOME/mutter/issues/81 [14:31] great [14:31] tseliot: maybe you hit the same bug? ^ [14:32] https://bugs.launchpad.net/ubuntu/+source/gdm3/+bug/1754218 seems to be related [14:32] Ubuntu bug 1754218 in gdm "[regression] gdm 3.27 shows blank login screen on primary monitor when two monitors are connected on a hybrid intel/nvidia system" [Medium,Confirmed] [14:51] Laney: hey... can you please publish https://bileto.ubuntu.com/#/ticket/3201 [14:53] Trevinho: ok [14:53] Trevinho, Laney: thanks! [14:53] ta [14:53] you should ask for better changelogs though :-) === pstolowski is now known as pstolowski|lunch [14:54] Yeah, right.. I guess bileto didn't take both lines [14:55] Laney, it's concise, it says what it does, and it links to a bug report, what would you suggest to improve it? [14:56] (indeed bileto only kept the first line of the commit message, and the comma looks out of place) [14:59] oSoMoN: I would prefer to see a short explanation of the changes and why they fix the problem [14:59] the actual individual commits have better messages, they just didn't end up in the debian changelog [15:00] thanks for the fix though :P [15:03] Trevinho, Laney: so bileto automatically generates a changelog entry based on the merge request commit message ; if I had added a new changelog entry in my merge request would it have used it instead? [15:03] yes [15:03] it won't touch d/changelog if the MP touches it [15:04] iirc anyway [15:04] cool, I'll do that next time then [15:05] Trevinho, I submitted a similar fix for libindicator: https://code.launchpad.net/~osomon/libindicator/bionic-ftbfs-fix/+merge/341772 [15:05] I'll add a changelog entry [15:05] oSoMoN: yeah, better use debchange if you want to show things better on debian side [15:06] Trevinho, shall I leave the changelog entry UNRELEASED ? [15:07] oSoMoN: yep [15:07] bileto will fix it [15:07] or add bionic, as you wish [15:12] done [15:12] Trevinho, ready for review and bileto [15:23] didrocks: pushed! [15:24] didrocks: mpt is off today [15:24] not sure blocking on design is good at this point [15:25] if it's about paddings/etc we can always fix it later without UIFe [15:25] tjaalton: it could be that === pstolowski|lunch is now known as pstolowski [15:59] kenvandine, hey, I spent some time this morning looking into https://forum.snapcraft.io/t/evince-snap-does-not-open-comic-book-file-cbr-while-deb-does-so/4523/11 (see my comments), but I kinda stalled and didn't want to spend the day on it, any ideas? [15:59] willcoke: for reference this is how it looks like now https://usercontent.irccloud-cdn.com/file/bkCXxMnB/sp1.png https://usercontent.irccloud-cdn.com/file/8kivUljT/sp2.png https://usercontent.irccloud-cdn.com/file/2w6o2WjG/sp3.png [16:01] oSoMoN, i looked at that too, no ideas [16:01] jbicha: There is an issue with install links in gnome-user-docs. Upstream uses 'action="install:"' (which AFAIK doesn't work on Ubuntu) while the format we use is 'href="apt:"'. This message on gnome-doc-list illustrates the conflict: [16:01] https://mail.gnome.org/archives/gnome-doc-list/2018-March/msg00079.html [16:01] Do you know if it's possible to make the upstream variant work on Ubuntu? Or should we propose Ubuntu conditionals? [16:01] I noticed the existence of the gnome-packagekit-session package (not installed by default). Does that package have anything to do with it? [16:08] gnome-packagekit-session is only a transitional package [16:09] jbicha: Then we can forget about that. :) [16:09] I think install: used to work… [16:11] jbicha: There were no such links in 16.04. [16:14] jbicha: apt: is nice, since it works also when browsing the HTML pages (at least with FF). [16:15] try running xdg-open install:gnome-clocks in 14.04 (maybe even 16.04) [16:15] I don't have 14.04 installed here since I'm no longer SRUing to it (except very rarely) [16:16] doesn't work in 16.04 [16:18] jbicha: I have trusty in a separate partition. Will test, and get back in a while. [16:19] it doesn't work in Fedora either [16:20] GunnarHj: come back! [16:31] jbicha: [16:31] $ lsb_release -cr [16:31] Release: 14.04 [16:31] Codename: trusty [16:31] $ xdg-open install:gnome-clocks [16:31] gvfs-open: install:gnome-clocks: error opening location: The specified location is not supported [16:31] it doesn't work on Fedora either [16:33] jbicha: See Mario's comment here: [16:33] https://mail.gnome.org/archives/gnome-doc-list/2018-March/msg00076.html [16:34] GunnarHj: there is an alternative appstream:org.gnome.clocks.desktop that works on Debian but not Ubuntu! [16:34] !! [16:35] apt:gnome-sushi is valid though since that paragraph is only shown on Ubuntu, right? [16:35] jbicha: Right. [16:35] what Andre said :) [16:36] jbicha: The problem is the other links which are shown everywhere. [16:36] if you want install: , I would file a gnome-software bug upstream [16:37] jbicha: How can it be an upstream thing if it works on Fedora? (Which Mario claims it does.) [16:38] it didn't work for me, but I'm updating my F28 now¸ give it 20 minutes [16:39] jbicha: Ok. It's the behavior when clicking such a link from yelp we are talking about, right? [16:53] andyrock, sorry missed that ping - yeah, looks good! [17:47] GunnarHj: hmm, xdg-open doesn't work but the install: links in yelp half-way work. Not sure why launching from yelp is different [17:47] by half-way, I mean it opens GNOME Software but can't find any results for gnome-tweaks :( [17:49] also, Fedora manages to figure out how to provide an install link in GNOME Software for gnome-user-share, rygel and vino but Ubuntu can't do that since there isn't appstream metadata [17:50] so I guess file an Ubuntu bug against gnome-software for that feature not working [17:51] yelp isn't working in my Debian Buster install so I can't tell whether they have the same bug [17:55] Have a nice evening. See you tomorrow. [18:11] dinner time, night all [18:19] jbicha: Thanks for your help to investigate! [18:19] I think I'll file a gnome-software (Ubuntu) bug, even if I'm not sure that the install: link type is the best approach for us considering that apt: works smoothly both in yelp and with FF. OTOH, I can't reasonably propose Ubuntu specific conditionals in gnome-user-docs before we have sorted out if we can make it work. [18:20] you can just (carefully) run sed in debian/rules to switch from install: to apt: [18:21] jbicha: Yeah, I thought of that... Maybe I'll give it a try. [18:43] jbicha: One last thing about docs links: [18:43] When building the HTML, we have for many cycles converted apt: to https://apps.ubuntu.com/cat/applications/. See for instance the link to install Synaptic here: [18:43] https://help.ubuntu.com/16.04/ubuntu-help/addremove-install-synaptic.html [18:43] However, apps.ubuntu.com now redirects to https://snapcraft.io/store, so those links are broken. :( My plan is to drop that conversion, and use href="apt:" also on the HTML pages. [18:43] Now the question: Are you aware of any other web interface to the package archive that could be used instead of apps.ubuntu.com? (packages.ubuntu.com isn't good enough for several reaons.) [18:43] no [18:44] jbicha: Ok, that's what I thought. Thanks! === amano_ is now known as amano === siel_ is now known as siel