| KGB-0 | gnome-shell ubuntu/master cb58172 Marco Trevisan (Treviño) debian/changelog * Update changelog * https://deb.li/32OqV | 00:28 |
|---|---|---|
| KGB-0 | gnome-shell ubuntu/master 8846fe0 Marco Trevisan (Treviño) debian/changelog * Upload to kinetic * https://deb.li/zXZ1 | 00:28 |
| KGB-0 | gnome-initial-setup ubuntu/master 8665a6d Jeremy Bicha * pushed 90 commits (first 5 follow) * https://deb.li/igM6M | 00:52 |
| KGB-0 | gnome-initial-setup ubuntu/master c9df3b5 Aleksandr Melman po/ru.po * Update Russian translation * https://deb.li/2ceI | 00:52 |
| KGB-0 | gnome-initial-setup ubuntu/master b773d6f Will Thompson gnome-initial-setup/pages/account/ (7 files) * Merge commit '131b2f5' * https://deb.li/3hX2A | 00:52 |
| KGB-0 | gnome-initial-setup ubuntu/master 055b800 Luming Zh po/zh_CN.po * Update Chinese (China) translation * https://deb.li/8V0e | 00:52 |
| KGB-0 | gnome-initial-setup ubuntu/master e1f6bea Fran Dieguez po/gl.po * Update Galician translation * https://deb.li/3NOC0 | 00:52 |
| KGB-0 | gnome-initial-setup ubuntu/master 1e93957 Jeremy Bicha debian/changelog * Update debian/changelog * https://deb.li/ROAz | 00:52 |
| oSoMoN | good morning desktoppers | 06:15 |
| seb128 | goood morning desktopers! | 06:19 |
| seb128 | lut oSoMoN, en forme ? | 06:19 |
| jibel | Good morning | 06:21 |
| seb128 | lut jibel, en forme ? | 06:21 |
| jibel | Salut seb128, oui, et toi? | 06:22 |
| seb128 | un peu fatigué, il fait trop chaud pour bien dormir, mais sinon ça va! | 06:22 |
| jibel | ici la chaleur est pas mal retombée et le baignades en soirée font du bien | 06:26 |
| lissyx | que des français finalement :) | 06:31 |
| seb128 | :) | 06:32 |
| seb128 | lissyx, re the backlog from yesterday, yes I will try to see what we can do to get those portal issues moving | 06:35 |
| duflu | Morning oSoMoN, seb128, jibel | 06:41 |
| oSoMoN | salut seb128, jibel, lissyx | 06:43 |
| oSoMoN | hey duflu | 06:43 |
| lissyx | https://bugzilla.mozilla.org/show_bug.cgi?id=1773316#c9 | 06:43 |
| ubottu | Mozilla bug 1773316 in Core "Can't see my printer with firefox 101 on Ubuntu 22.04 (snap)" [--, Unconfirmed] | 06:43 |
| lissyx | oSoMoN, that's starting to be several people with supposed-to-work auto-connect that did fail | 06:44 |
| lissyx | between the opengl one and the printers one | 06:44 |
| seb128 | hey duflu, how are you? | 06:47 |
| duflu | seb128, not bad, you? | 06:47 |
| seb128 | I'm alright, just a bit tired due to the heat | 06:48 |
| lissyx | oSoMoN, https://bugs.launchpad.net/ubuntu/+source/lightdm/+bug/1593407 do you know what is broken precisely? | 07:03 |
| ubottu | Launchpad bug 1593407 in snapd (Ubuntu) "Guest session cannot run snaps" [High, Confirmed] | 07:03 |
| seb128 | lissyx, the guest session thing use apparmor to restrict permissions and seems to block things snapd needs to work correctly | 07:05 |
| seb128 | note that guest sessions isn't something we had been providing or supporting for 5 years | 07:05 |
| seb128 | lissyx, my guess is that some access to /snap and maybe other locations would need to be get added to https://github.com/canonical/lightdm/blob/main/data/apparmor/lightdm-guest-session.in | 07:06 |
| seb128 | or https://github.com/canonical/lightdm/blob/main/data/apparmor/abstractions/lightdm | 07:07 |
| seb128 | 'cannot open snapd info file "/snap/snapd/current/usr/lib/snapd/info": open /snap/snapd/current/usr/lib/snapd/info: permission denied' | 07:08 |
| didrocks | good morning | 07:12 |
| luna__ | morning | 07:12 |
| seb128 | lut didrocks, hey luna__, how are you? | 07:14 |
| luna__ | i am okay | 07:14 |
| didrocks | salut seb128, ça va, et toi ? | 07:15 |
| seb128 | didrocks, ça va, un peu fatigué, fait trop chaud pour bien dormir... | 07:16 |
| duflu | Hi didrocks and luna__ | 07:18 |
| KGB-2 | gnome-control-center ubuntu/jammy 54a193d Sebastien Bacher debian/ changelog patches/series patches/panels-wwan-Add-WWAN-5G-connection-support.patch * debian/patches: Add 5G connection support (LP: #1983645) * https://deb.li/i2bv4 | 07:18 |
| ubottu | Launchpad bug 1983645 in OEM Priority Project "Can't find 5G option in Network Mode under Mobile Network" [Critical, Confirmed] https://launchpad.net/bugs/1983645 | 07:18 |
| KGB-2 | gnome-control-center signed tags 8a1383c Sebastien Bacher ubuntu/1%41.7-0ubuntu0.22.04.5 * gnome-control-center Debian release 1:41.7-0ubuntu0.22.04.5 * https://deb.li/U63U | 07:18 |
| lissyx | seb128, so it's not supported but people are loudly complaining | 07:21 |
| lissyx | seb128, especially the bug on our side was opened against MATE that seems to still support guest session | 07:22 |
| lissyx | while I get that it's their problem, it is also somehow ours since people report it to us .. | 07:22 |
| seb128 | lissyx, right, well if someone (you? ;-) want to fix the guest session profile that would be welcome, I'm happy to drive the landing in Ubuntu | 07:23 |
| seb128 | should be easy enough to test, basically it's switch the login manager to lightdm and set the config Olivier shared | 07:24 |
| lissyx | hence why asking if you knew precisely what is broken :) | 07:24 |
| seb128 | ack | 07:24 |
| didrocks | hey duflu | 07:24 |
| seb128 | well I would start by adding read permission to /snap to the apparmor profiles I pointed out | 07:24 |
| seb128 | it will probably also require other things that are currently blocked, it's an iteration game :) | 07:25 |
| lissyx | also, on the fonts problem | 07:25 |
| lissyx | what fix is going to be done? | 07:26 |
| lissyx | a snap providing fonts? | 07:26 |
| seb128 | a don't know what's the proper solution going forward | 07:27 |
| seb128 | but I think the first easy step is to do what James suggested and add the font (or at least its fontconfig config) to the gnome sdk snap | 07:27 |
| seb128 | they are probably some other fonts in the same case so i'm unsure how we build a list, or if we just go one by one as issues get reported | 07:28 |
| lissyx | like on gnome-3-38-2004-sdk branch | 07:32 |
| lissyx | (I cant find in gnome sdk snap where fonts are) | 07:32 |
| lissyx | seb128, I worry that just adding this font will lead to people hitting the same issue later | 07:35 |
| lissyx | oSoMoN, just runnning 'speechSynthesis.speak(new SpeechSynthesisUtterance("Testing 1 2 3"));' on a nightly from us does not work | 07:37 |
| lissyx | I can't find anything related to speech synthesis exposed in GNOME Settings | 07:37 |
| lissyx | is this supposed to work out of the box? | 07:37 |
| lissyx | speech-dispatcher and its friends seems installed | 07:37 |
| lissyx | oSoMoN, seb128, also, for mozregressionn support, it might be easier for everyone if we update our side of snap production to re-sue your snapcraft.yaml | 07:39 |
| lissyx | oSoMoN, seb128 however the current one is unable to get a source tree from anything else than https://github.com/canonical/firefox-snap/blob/nightly/snapcraft.yaml#L197-L201 | 07:40 |
| seb128 | lissyx, is speechSynthesis the screen reader? if so the service we use for that is orca and I think it only gets activated if you turn on the screen reader in the accessiblity settings | 07:41 |
| lissyx | if I want that to run on autoland and m-c, I will have to provide a way to specify accessing sources differently | 07:41 |
| lissyx | would that be okay to you? | 07:41 |
| lissyx | seb128, ah let me chack that | 07:41 |
| lissyx | lol | 07:43 |
| lissyx | nope | 07:43 |
| lissyx | the screen reader reads me the code I'm pasting in the devtool console, but speech synthesis interface itself does nothing | 07:43 |
| seb128 | lissyx, re. fonts, yes adding one will probably mean we will get other similar reports for other fonts later. I wonder if we should just bundle the fonts config directory from a standard desktop installation. Though I don't know how fontconfig handles the case where there is a config for a font which isn't available? | 07:43 |
| lissyx | I dont know much about fonts either | 07:44 |
| seb128 | lissyx, https://sources.debian.org/src/firefox/104.0-1/dom/media/webspeech/synth/speechd/SpeechDispatcherService.cpp/?hl=31#L321 ... we probably need to stage libspeechd2 for that to work | 08:01 |
| seb128 | lissyx, I don't really know about speech-to-text but spd-say works and I can confirm that firefox doesn't speak | 08:01 |
| lissyx | well libspeechd2 is installed | 08:06 |
| seb128 | in the snap? | 08:07 |
| lissyx | no | 08:07 |
| lissyx | our tar | 08:07 |
| seb128 | I assumed you were trying that in the snap | 08:08 |
| seb128 | it's not working from the tar version either? | 08:08 |
| lissyx | yes it's not working from the tar either | 08:08 |
| lissyx | we have dom/media/webspeech/synth/test/test_speech_simple.html | 08:09 |
| lissyx | let me check if it passes tests ... | 08:09 |
| seb128 | lissyx, is 'spd-say hey' working for you? | 08:10 |
| lissyx | yes | 08:10 |
| seb128 | k, so at least the service is working | 08:10 |
| lissyx | seb128, unrelated, but a long time ago I was on a bug on launchpad (and you were too as much as I know) related to LTE support for Quectel EM120GL-R | 08:11 |
| lissyx | firmware crashing | 08:11 |
| lissyx | I can't find it anymore | 08:11 |
| seb128 | lissyx, there was https://bugs.launchpad.net/ubuntu/+source/modemmanager/+bug/1949621 according to my firefox memory | 08:13 |
| ubottu | Launchpad bug 1949621 in modemmanager (Ubuntu) "modemmanager/libmbim cannot make use of Quectel EM120" [Low, Fix Released] | 08:13 |
| lissyx | nice | 08:14 |
| lissyx | wooo | 08:19 |
| lissyx | seb128, looks like the test is passing but tryign to get something spoken does not work | 08:19 |
| seb128 | :( | 08:20 |
| oSoMoN | for once, it's not a snap-specific issue… | 08:23 |
| lissyx | oSoMoN, it might just be some other pieces are missinng | 08:26 |
| lissyx | I'm testing from devtools, maybe it is not supposed to work | 08:29 |
| lissyx | https://codepen.io/matt-west/pen/DpmMgE I have nothing in the "Voices" drop down | 08:30 |
| lissyx | working on windows | 08:35 |
| lissyx | did we regressed on linux ? | 08:35 |
| oSoMoN | looks like it… for a comparison point, TTS works for me in chrome, but not in the chromium snap, and neither in the firefox upstream tarball nor in the snap | 08:37 |
| lissyx | no point in investigating the snap if it's broken outside of it :) | 08:43 |
| seb128 | lissyx, we might still want to add libspeechd2 to the stage for the day it gets fixed | 08:44 |
| seb128 | https://bugs.launchpad.net/ubuntu/+source/chromium-browser/+bug/1792300 suggests it was working in firefox in 22.04 | 08:47 |
| ubottu | Launchpad bug 1792300 in chromium-browser (Ubuntu) "Speech Synthesis Not Working" [Medium, Confirmed] | 08:47 |
| seb128 | so probably something that regressed | 08:47 |
| KGB-2 | switcheroo-control upstream/latest 52d4fe9 Jeremy Bicha * pushed 28 commits (first 5 follow) * https://deb.li/3fNLh | 08:48 |
| KGB-2 | switcheroo-control upstream/latest fdae539 Bastien Nocera tests/ meson.build unittest_inspector.py * tests: Split the integration test into individual tests * https://deb.li/MGW2 | 08:48 |
| KGB-2 | switcheroo-control upstream/latest 085091f Bastien Nocera data/meson.build meson.build * build: Install D-Bus config file in /usr * https://deb.li/ik48C | 08:49 |
| KGB-2 | switcheroo-control upstream/latest 0eba9b4 Bastien Nocera tests/meson.build * build: Fix run_command() warning in meson * https://deb.li/7GEg | 08:49 |
| KGB-2 | switcheroo-control upstream/latest de36074 Bastien Nocera src/switcheroo-control.c * main: Add Vulkan support * https://deb.li/3r40l | 08:49 |
| KGB-2 | switcheroo-control upstream/latest d3d8399 Bastien Nocera .gitlab-ci.yml tests/integration-test.py * tests: Add test for Vulkan support * https://deb.li/aNex | 08:49 |
| KGB-2 | switcheroo-control pristine-tar 240e1d9 Jeremy Bicha switcheroo-control_2.6.orig.tar.bz2.delta switcheroo-control_2.6.orig.tar.bz2.id * pristine-tar data for switcheroo-control_2.6.orig.tar.bz2 * https://deb.li/3MK2E | 08:49 |
| lissyx | mozregression, firefox 91.0a1, broken as well | 08:50 |
| lissyx | either it is broken for years, or it is something else ... | 08:50 |
| lissyx | https://bugzilla.mozilla.org/show_bug.cgi?id=1666703 | 08:55 |
| ubottu | Mozilla bug 1666703 in Core "SpeechSynthesis.getVoices() returns empty array instead of SpeechSynthesisVoice objects" [S3, New] | 08:55 |
| oSoMoN | it must be something else, because people mentioned in the chromium bug report that it was working in firefox, and that was not so long ago | 08:55 |
| lissyx | missing dep or something like that | 08:58 |
| oSoMoN | or incompatible version of a lib? | 09:03 |
| lissyx | but I'm pretty sure empty getVoices is related to the issue | 09:05 |
| oSoMoN | yes | 09:05 |
| oSoMoN | very likely | 09:05 |
| lissyx | interesting | 09:09 |
| lissyx | https://searchfox.org/mozilla-central/rev/30ac44262374be08ccd4262ee36f0716152868d3/dom/media/webspeech/synth/speechd/SpeechDispatcherService.cpp#367-368 | 09:10 |
| lissyx | list is not null but it's empty | 09:10 |
| jbicha | ricotz: btw, we'll be rebuilding libreoffice for the evolution-data-server transition probably later today | 09:19 |
| lissyx | oSoMoN, seb128 do you know how voices are to be installed? | 09:24 |
| lissyx | "spd-say -L" returns nothing | 09:24 |
| oSoMoN | hrm, I can confirm this in jammy, but in focal it returns a list of voices as expected | 09:27 |
| seb128 | lissyx, I got it working in the tar, no idea what I'm doing but I found https://stackoverflow.com/questions/46617366/speechsynthesis-getvoices-not-listing-voices-in-firefox which is quite informative | 09:27 |
| seb128 | installing speech-dispatcher-pico made it work | 09:27 |
| seb128 | $ spd-say -L | grep fr | 09:27 |
| seb128 | virginie fr-FR none | 09:27 |
| seb128 | it was empty before | 09:28 |
| seb128 | now firefox list language and speaks | 09:28 |
| lissyx | seb128, looks like it's this | 09:28 |
| seb128 | not the snap though, but that's expected since the libspeech library is missing | 09:28 |
| lissyx | seb128, espeak-ng-mbrola | 09:28 |
| seb128 | I wonder if we should install that by default? | 09:30 |
| seb128 | we have espeak-ng-data installed by default which has voices in /usr/lib/x86_64-linux-gnu/espeak-ng-data/voices/ , unsure why that's not working | 09:32 |
| lissyx | seb128, I think it's just that default setup has two espeak modules and the first default one has no voice | 09:35 |
| seb128 | sounds like a bad default then? | 09:35 |
| lissyx | that's your default | 09:36 |
| lissyx | maybe we should be more resilient to it as well | 09:36 |
| lissyx | https://searchfox.org/mozilla-central/rev/30ac44262374be08ccd4262ee36f0716152868d3/dom/media/webspeech/synth/test/file_setup.html#82-85 | 09:36 |
| lissyx | our test is not correct here | 09:36 |
| lissyx | we never verify we have voices ... | 09:36 |
| seb128 | yes, I was mentioning that as a distro problem | 09:37 |
| seb128 | but firefox could handle things better | 09:37 |
| seb128 | + the snap needs fixing | 09:37 |
| duflu | tjaalton, are we really not allowed to patch xwayland? https://salsa.debian.org/xorg-team/wayland/xwayland/-/blob/upstream-unstable/.gitignore#L68 | 09:40 |
| tjaalton | duflu: use .diff | 09:40 |
| tjaalton | or git add -f | 09:40 |
| duflu | Umm OK | 09:41 |
| tjaalton | it's pretty common for all xorg projects to have that :) | 09:41 |
| duflu | Yeah 8 years old. Looks historical like it's better to have a file that's wrong with no diff against it | 09:42 |
| oSoMoN | seb128, lissyx : on the distro level, it looks like a regression between focal and jammy | 09:47 |
| lissyx | i'm wondeing if it's not a bug on speech-dispatcher side | 09:47 |
| lissyx | https://github.com/brailcom/speechd/blob/6712d5265192bcbd6a1498656f11be83a5b772e0/src/api/c/libspeechd.c#L1469-L1506 | 09:47 |
| seb128 | oSoMoN, what was the difference in focal? different default? | 09:47 |
| lissyx | there's nothing in the API allowing us to select modules | 09:47 |
| lissyx | seb128, I think they introduced mbrola | 09:47 |
| seb128 | lissyx, oSoMoN, I reported the bug against Debian now, the maintainer is usually quite responsive and involved upstream in the accessibiliy stacks | 09:49 |
| seb128 | will share the reference once the Debian email interface handle the report | 09:49 |
| oSoMoN | thx | 09:50 |
| lissyx | https://github.com/brailcom/speechd/pull/652 | 09:50 |
| ubottu | Pull 652 in brailcom/speechd "Fix listing voices of default module" [Merged] | 09:50 |
| lissyx | HM. | 09:50 |
| seb128 | https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1018075 | 09:51 |
| seb128 | ah, the reporter of that github is the Debian maintainer :) | 09:51 |
| oSoMoN | seb128, in a focal VM, "spd-say -L" lists a bunch of voices, whereas in a jammy VM it doesn | 09:51 |
| oSoMoN | doesn't | 09:52 |
| seb128 | ack | 09:52 |
| lissyx | look at this PR | 09:52 |
| seb128 | well I would let Samuel reply at this point | 09:52 |
| seb128 | lissyx, it seems right if that was listing voices which were not used before? | 09:53 |
| seb128 | it still feels like that the issue is that the default is a module which has no voice associated | 09:53 |
| lissyx | but maybe with this patch it's not a problem anymore | 09:53 |
| duflu | tjaalton, thanks. Please forgive me for any breach of convention in advance :) https://salsa.debian.org/xorg-team/wayland/xwayland/-/merge_requests/2 | 09:53 |
| ubottu | Merge 2 in xorg-team/wayland/xwayland "Add xwayland-Detect-gbm_bo_get_fd_for_plane-at-runtime.patch" [Opened] | 09:53 |
| tjaalton | the patch description is too verbose ;) | 09:55 |
| duflu | Well, it's that or require the reader to find the original web page | 09:55 |
| duflu | This solution works without the internet | 09:56 |
| tjaalton | heh, no worries, the patches usually just have context in the filename :) | 09:56 |
| tjaalton | unless already upstream, in which case git format-patch is fine | 09:56 |
| tjaalton | I mean, it's great now | 09:56 |
| tjaalton | as is | 09:56 |
| seb128 | lissyx, this patch is included in the 22.04 speech-dispatcher | 09:57 |
| lissyx | is it? | 09:58 |
| lissyx | I dont see it here. | 09:58 |
| lissyx | I just apt-get sourced it | 09:58 |
| tjaalton | duflu: I'll merge 22.1.3 too | 10:02 |
| tjaalton | oh, looks like the local branch on this machine was just behind | 10:03 |
| duflu | Thanks | 10:03 |
| seb128 | lissyx, oSoMoN, I wonder if the 'no voice' is a bug in https://salsa.debian.org/tts-team/speech-dispatcher/-/blob/master/debian/patches/mbrola-paths | 10:04 |
| lissyx | I dont see patches applied from debian/patches/series | 10:04 |
| seb128 | lissyx, which version did you get? | 10:05 |
| lissyx | and every time I read https://wiki.debian.org/debian/patches I never understand | 10:05 |
| lissyx | speech-dispatcher_0.11.1-1ubuntu1.dsc | 10:05 |
| lissyx | isn't "debuild -b -us -uc" supposed to apply the patches ? | 10:06 |
| seb128 | it does | 10:07 |
| seb128 | when dpkg unpacks the source it applies the patches | 10:07 |
| lissyx | there's 0 mention of it in the build log | 10:07 |
| seb128 | export QUILT_PATCHES=debian/patches | 10:07 |
| seb128 | quilt applied | 10:08 |
| seb128 | to see what is applied | 10:08 |
| seb128 | lissyx, https://salsa.debian.org/tts-team/speech-dispatcher/-/blob/debian/0.11.1-1/src/server/output.c#L134 | 10:08 |
| seb128 | to me it looks like the new variant of the code past that PR? | 10:08 |
| seb128 | or I'm reading it wrong? | 10:08 |
| lissyx | the PR is said to be merged only on master | 10:08 |
| lissyx | others are shown as 0.11.1 tag | 10:08 |
| seb128 | lissyx, https://github.com/brailcom/speechd/commits/speech-dispatcher-0-11 | 10:09 |
| seb128 | it's the 3 item right? | 10:09 |
| lissyx | https://salsa.debian.org/tts-team/speech-dispatcher/-/commits/debian/0.11.1-1/src/server/output.c | 10:09 |
| seb128 | dunno why the PR isn't showing the tag | 10:09 |
| seb128 | but it is included unless I'm really confused | 10:09 |
| lissyx | yeah | 10:10 |
| lissyx | looks like github UI is painful now | 10:10 |
| seb128 | lissyx, I would suggest to let Samuel some hours to reply and to move to something else meanwhile, he knows the stack better and that will probably be more productive than us trying to figure out details of a project we are not familiar with | 10:10 |
| lissyx | I get your point | 10:10 |
| ricotz | jbicha, hi, ah, I could have waited with the last upload then :( | 10:11 |
| lissyx | repro'd on CI: https://treeherder.mozilla.org/jobs?repo=try&revision=c44518128399cc014b8b7da184a6bc19aa32ed1b | 10:11 |
| ricotz | jbicha, would be good to let the libreoffice arm* builds finish first | 10:12 |
| lissyx | seb128, still i'm blocked even to fix the snap ... | 10:26 |
| lissyx | I have no idea what I am doing but removing /etc/speech-dispatcher/modules/espeak-ng-mbrola.conf, spd-say -O still lists espeak-ng-mbrola, but -L shows a correct list | 10:27 |
| KGB-0 | adwaita-icon-theme ubuntu/master fd312df Marco Trevisan * pushed 264 commits (first 5 follow) * https://deb.li/O8fC | 10:27 |
| KGB-0 | adwaita-icon-theme ubuntu/master d0ecd9c Sam Hewitt (6 files in 6 dirs) * fullcolor: few improvements to folder-remote; fix it being off-baseplate * https://deb.li/3hx3G | 10:27 |
| KGB-0 | adwaita-icon-theme ubuntu/master 3f96f1d Jakub Steiner Adwaita/scalable/actions/selection-mode-symbolic.svg src/symbolic/core.svg * symbolic: selection-mode * https://deb.li/vuBc | 10:28 |
| lissyx | and firefox from us now supports :) | 10:28 |
| KGB-0 | adwaita-icon-theme ubuntu/master 6b85bbe Jakub Steiner Adwaita/scalable/actions/selection-mode-symbolic.svg src/symbolic/core.svg * Merge branch 'wip/jimmac/selection-mode' into 'master' * https://deb.li/3oDKy | 10:28 |
| lissyx | I guess it's good enough for me so far ... | 10:28 |
| KGB-0 | adwaita-icon-theme ubuntu/master 5d4e335 Jakub Steiner NEWS configure.ac * prepare for 43.beta release * https://deb.li/SDYd | 10:28 |
| KGB-0 | adwaita-icon-theme ubuntu/master 98b4569 Jakub Steiner configure.ac * build: adjust dist tarball creation * https://deb.li/3b8OV | 10:28 |
| seb128 | lissyx, try to install speech-dispatcher-pico maybe? that was enough to make firefox tar work for me | 10:29 |
| seb128 | without config change | 10:29 |
| seb128 | then you should be able to poke at the snap | 10:29 |
| seb128 | I expect we need the lib and maybe to give access to the speech-dispatcher socket in XDG_RUNTIME_DIR/speech-dispatcher/speechd.sock | 10:30 |
| lissyx | I dont really care at that point | 10:31 |
| lissyx | if breaking a bit my system helps | 10:31 |
| lissyx | I need to rebuild a new 106, previous builds where 105 | 10:32 |
| lissyx | guess I'll launch it while doing some biking then eating and come back at this later | 10:32 |
| lissyx | and as you said, hopefully it will just be about lib + socket access | 10:32 |
| seb128 | enjoy the biking! | 10:33 |
| seb128 | though with the heat I'm unsure if that's not 'good luck' instead | 10:33 |
| lissyx | home trainer indeed | 10:34 |
| lissyx | when I dont commute to the office, 1h of it, 40km.h of speed | 10:35 |
| lissyx | with some hand-crafted bike | 10:35 |
| seb128 | :) | 10:35 |
| lissyx | Novo Force, speed 3, 52 teeth on the front plateau, 11 on the derailleur | 10:36 |
| lissyx | if it's a distro issue we can't workaround on our side, it's going to be tricky to fix on CI I fear | 10:36 |
| seb128 | right :/ | 10:37 |
| seb128 | well, let's see first what Samuel has to say | 10:37 |
| seb128 | bbiab, relocating to somewhere less warm for lunch | 10:40 |
| KGB-2 | orca ubuntu/master 16ea0b4 Marco Trevisan * pushed 8 commits (first 5 follow) * https://deb.li/3W8t2 | 10:52 |
| KGB-2 | orca ubuntu/master 61f07b4 Joanmarie Diggs README.md configure.ac * Post-release version bump * https://deb.li/i47Eg | 10:52 |
| KGB-2 | orca ubuntu/master 1a37355 Joanmarie Diggs src/orca/speech_generator.py * Web: Make cell ancestor presentation optional * https://deb.li/Wsiz | 10:52 |
| KGB-2 | orca ubuntu/master ce74a5a Joanmarie Diggs NEWS * Release prep * https://deb.li/3F7cs | 10:52 |
| KGB-2 | orca ubuntu/master 35aa38c Samuel Thibault (10 files in 3 dirs) * New upstream version 42.3 * https://deb.li/aihv | 10:52 |
| KGB-2 | orca ubuntu/master bfc2193 Samuel Thibault (10 files in 3 dirs) * Update upstream source from tag 'upstream/42.3' * https://deb.li/3uCyn | 10:52 |
| jbicha | good morning | 12:36 |
| lissyx | oSoMoN, build is failing for me | 12:45 |
| lissyx | + cp /root/stage/debug-symbols/firefox-106.0a1.en-US.linux-x86_64.crashreporter-symbols.zip /root/stage/debug-symbols/firefox-105.0a1.en-US.linux-x86_64.crashreporter-symbols.zip /root/project/firefox_106.0a1_amd64.debug | 12:45 |
| lissyx | cp: target '/root/project/firefox_106.0a1_amd64.debug' is not a directory | 12:45 |
| lissyx | oh probably my fault :) | 12:46 |
| lissyx | wish me luck | 13:14 |
| lissyx | I'm going to install a new wifi card :) | 13:14 |
| lissyx | [ 24.634079] iwlwifi 0000:03:00.0: Detected Intel(R) Wi-Fi 6 AX210 160MHz, REV=0x420 | 13:30 |
| lissyx | :) | 13:30 |
| lissyx | seb128, ^ moved from rtw8859ae (really crashing driver, said to be due to bogus UEFI) to ac9260 and now ax210 | 13:33 |
| KGB-2 | babl tags 1b9c4a4 Jeremy Bicha upstream/0.1.96 * Upstream version 0.1.96 * https://deb.li/47IA | 13:54 |
| KGB-2 | babl upstream/latest f51c9b6 Jeremy Bicha * pushed 29 commits (first 5 follow) * https://deb.li/ipCS7 | 13:54 |
| KGB-2 | babl upstream/latest 2f9d5e7 Jehan bin/babl.c * bin: add -r/--intent for setting the rendering intent on the CLI tool. * https://deb.li/Wiky | 13:54 |
| KGB-2 | babl upstream/latest d72b5dd Jehan babl/babl-icc.c * babl: fix typos in error strings. * https://deb.li/3ZC14 | 13:54 |
| KGB-2 | babl upstream/latest 035c7b7 Jehan bin/babl.c * bin: profile-aware babl format defaults. * https://deb.li/A7ZM | 13:54 |
| KGB-2 | babl upstream/latest 0cfa2ed Axel Viala babl/base/formats.c * https://deb.li/301Nj | 13:54 |
| KGB-2 | Fix warning: comparison of integer expressions of different signedness: ‘int’ and ‘long unsigned int’ [-Wsign-compare] | 13:54 |
| KGB-2 | babl upstream/latest 933b3bb Axel Viala tools/babl_fish_path_fitness.c * babl_fish_path: comparison of integer expressions of different signedness * https://deb.li/dYWN | 13:55 |
| lissyx | not sure how I am supposed to grant access to the socket :) | 14:00 |
| lissyx | maybe just a plug? | 14:02 |
| lissyx | looks like we already have wayland-0 and this is history on that: https://forum.snapcraft.io/t/wayland-dconf-and-xdg-runtime-dir/186 | 14:07 |
| seb128 | lissyx, you can't grant random access from the snap, that needs to be in snapd | 14:14 |
| seb128 | lissyx, something similar to https://github.com/snapcore/snapd/pull/11208 for example | 14:15 |
| ubottu | Pull 11208 in snapcore/snapd "interfaces: allow access to new at-spi socket location in desktop-legacy" [Merged] | 14:15 |
| lissyx | seb128, I was just bugging mardy about that on #snappy | 14:16 |
| KGB-0 | gnome-online-accounts ubuntu/master 084d3e2 Jeremy Bicha debian/patches/0001-ubuntu-sso-provider.patch * Update ubuntu-sso-provider.patch * https://deb.li/Ft7E | 15:03 |
| lissyx | seb128, so it's not working, and I dont see anything in journalctl -xe | 15:08 |
| seb128 | lissyx, :( | 15:17 |
| seb128 | lissyx, I can give it a try tomorrow. Did you rebuild a snap with the library bundled? I guess the dlopen equivalent finds it where it's intalled? | 15:18 |
| lissyx | seb128, the lib is visible in the snap | 15:18 |
| lissyx | as to whether it loads properly, it's hard to tell | 15:18 |
| lissyx | there is no error reported at all | 15:18 |
| lissyx | the few logging avaible does not show anything | 15:18 |
| lissyx | there seems to be some kind of auto spawn of speech-dispatcher | 15:19 |
| seb128 | was it showing an error before you added it? | 15:19 |
| oSoMoN | lissyx, you can use "snap run --strace …" to strace the execution of a snapped app, if this helps | 15:19 |
| lissyx | i.e., it's not running, do "spd-say -L", a pgrep -af speech shows a process with --spawn | 15:19 |
| lissyx | seb128, no | 15:19 |
| seb128 | weird, the code has | 15:20 |
| seb128 | if (!speechdLib) { | 15:20 |
| seb128 | NS_WARNING("Failed to load speechd library"); | 15:20 |
| seb128 | where should the NS_WARNING be displayed? | 15:20 |
| lissyx | but NS_WARNING is only shown on debug build as much as I remember | 15:20 |
| lissyx | it's not a debug build | 15:20 |
| seb128 | ah :( | 15:20 |
| lissyx | (and because the build process will re-download the whole hg history ...) | 15:21 |
| lissyx | we really need to fix that also | 15:21 |
| KGB-2 | gnome-online-accounts ubuntu/master c6248c8 Jeremy Bicha debian/patches/0002-livepatch-auth.patch * Update livepatch patch * https://deb.li/3RSFE | 15:21 |
| lissyx | it makes it super super painful to work on the snap package | 15:21 |
| seb128 | one trick you could try I used before is to set LD_DEBUG=libs | 15:22 |
| lissyx | bingo | 15:22 |
| lissyx | it's the kind of auto spawn that seems to be broken | 15:22 |
| seb128 | ah :/ | 15:22 |
| lissyx | I started my firefox, speechSynthesis.getVoices() | 15:22 |
| lissyx | then in parallel a spd-say -L | 15:23 |
| lissyx | all using SPEECHD_ADDRESS=inet_socket | 15:23 |
| lissyx | it spawn a process | 15:23 |
| lissyx | and now I have voices returned | 15:23 |
| lissyx | I see lang in codepen example | 15:23 |
| lissyx | ok so I have no sound | 15:25 |
| lissyx | but it is on my whole VM | 15:25 |
| jbicha | seb128: gnome-online-accounts has switched to meson. Do you know if I can drop the dh_translations override or do I need to port it to ninja? | 15:25 |
| jbicha | https://salsa.debian.org/gnome-team/gnome-online-accounts/-/blob/ubuntu/master/debian/rules#L30-38 | 15:25 |
| seb128 | jbicha, you should be able to drop it if it has the standard meson target working | 15:26 |
| jbicha | ok, I'll drop and we can check later if things still work as expected | 15:26 |
| seb128 | 👍 | 15:26 |
| lissyx | seb128, oSoMoN BONJOUR ICI C'EST SNAP | 15:26 |
| lissyx | too bad you can't see my nice screenshot | 15:26 |
| lissyx | and hear it | 15:27 |
| lissyx | but with all those hacks, it is working | 15:27 |
| oSoMoN | \o/ | 15:27 |
| oSoMoN | you rock | 15:27 |
| lissyx | I'm going to brain dump because it's full of hacks ... | 15:27 |
| lissyx | https://bugzilla.mozilla.org/show_bug.cgi?id=1729750#c8 | 15:36 |
| ubottu | Mozilla bug 1729750 in Firefox Build System "Speech synthesis doesn't work in the snap" [S3, Unconfirmed] | 15:36 |
| seb128 | thanks! | 15:42 |
| lissyx | as expected | 16:05 |
| lissyx | 22.04.1, more people hit snap bugs: https://bugzilla.mozilla.org/show_bug.cgi?id=1786392 | 16:05 |
| ubottu | Mozilla bug 1786392 in Core "Firefox snap Ubuntu 22.04.1 LTS handles /tmp reference incorrectly." [--, Unconfirmed] | 16:05 |
| jbicha | seb128: did you try demoting the mutter test binaries to -proposed for the 43~beta-3ubuntu2 upload? I think they went back to main because of the 2nd upload to -proposed | 16:10 |
| ricotz | jbicha, can we get libreoffice transition first before the e-d-s rebuild? | 16:56 |
| jbicha | ricotz: probably, if no one else schedules the rebuild. armhf autopkgtest queue is a little slow but maybe it can catch up over the weekend | 16:58 |
| ricotz | jbicha, ok :), not that I expect the e-d-s transition won't be smooth | 17:02 |
| jbicha | it's a bigger than usual transition entangled with some other libraries switching to libsoup3 | 17:02 |
| ricotz | yeah, I noticed, that is why I would like to have last LO move out of -proposed first | 17:03 |
| ricotz | before it gets entangled in there :) | 17:04 |
| jbicha | I understand. That's why I made sure gnome-shell 43 is getting through first too :) | 17:04 |
| KGB-2 | gnome-online-accounts ubuntu/master 9d5d0d0 Jeremy Bicha debian/patches/0001-ubuntu-sso-provider.patch * Update ubuntu-sso patch again * https://deb.li/GzsA | 18:28 |
| seb128 | lissyx, I got a reply from the Debian maintainer who wrote | 18:39 |
| seb128 | 'But when there is no voice available there, it | 18:39 |
| seb128 | won't be working so in that case speech-dispatcher shouldn't even enable | 18:39 |
| seb128 | it, and thus fallback to espeak-ng. I'll have a look at fixing that,' | 18:39 |
| KGB-0 | gnome-online-accounts signed tags c0108b8 Jeremy Bicha ubuntu/3.45.2-0ubuntu1 * gnome-online-accounts Debian release 3.45.2-0ubuntu1 * https://deb.li/ljOG | 18:57 |
| KGB-2 | gnome-maps upstream/latest 7b82f2c Jeremy Bicha * pushed 163 commits (first 5 follow) * https://deb.li/3rdLy | 20:20 |
| KGB-2 | gnome-maps upstream/latest 0a00add Jiri Grönroos po/fi.po * Update Finnish translation * https://deb.li/yhWG | 20:20 |
| KGB-2 | gnome-maps upstream/latest 2212dc9 Marcus Lundblad src/ (5 files) * Fix broken route printing after ES6 modules migration * https://deb.li/3MG8M | 20:20 |
| KGB-0 | gnome-maps pristine-tar 7ed0977 Jeremy Bicha gnome-maps_43~beta.orig.tar.xz.delta gnome-maps_43~beta.orig.tar.xz.id * pristine-tar data for gnome-maps_43~beta.orig.tar.xz * https://deb.li/i7KNg | 20:20 |
| KGB-2 | gnome-maps upstream/latest aa06d84 Goran Vidović po/hr.po * Update Croatian translation * https://deb.li/I9k1 | 20:20 |
| KGB-2 | gnome-maps upstream/latest 3d19e57 Goran Vidović po/hr.po * Update Croatian translation * https://deb.li/3mkqN | 20:20 |
| KGB-2 | gnome-maps upstream/latest 3864bae Goran Vidović po/hr.po * Update Croatian translation * https://deb.li/3kTZV | 20:20 |
| ricotz | jbicha, the needed autopkgtests for libreoffice 1:7.4.1~rc1-0ubuntu1 are running, so I would expect them to finish within the next 8 hours | 20:46 |
| jbicha | cool, I remembered that we're going to have to wait for https://launchpad.net/ubuntu/+source/webkit2gtk/2.37.90-1 to build on riscv64 to finish the evolution transition | 20:47 |
| ricotz | jbicha, oh, this riscv build can take quite some time | 20:49 |
| jbicha | yeah, brand new is that we are really building webkit 3 times (jammy had gtk3+libsoup2 and gtk3+libsoup3; kinetic now adds gtk4+libsoup3) | 20:50 |
| ricotz | yeah, I know | 20:50 |
| KGB-2 | gnome-shell ubuntu/master 2a44daa Jeremy Bicha debian/ control control.in rules * Build with libsoup3 * https://deb.li/Ak5C | 20:51 |
| KGB-2 | gnome-shell ubuntu/master 6b30471 Jeremy Bicha debian/changelog * releasing package gnome-shell version 43~beta-1ubuntu2 * https://deb.li/30dZG | 20:51 |
| KGB-2 | gnome-shell signed tags fe5a942 Jeremy Bicha ubuntu/43_beta-1ubuntu2 * gnome-shell Debian release 43~beta-1ubuntu2 * https://deb.li/dd9F | 20:51 |
| jbicha | gtk4 is needed by gnome-builder and gnome-initial-setup 43 😞 | 20:52 |
| ricotz | jbicha, might be worth to try to reuse the same builddir | 20:58 |
| ricotz | which works locally while meson seems smart enough to do that | 20:59 |
| jbicha | do you want to work on a merge proposal for that idea? | 21:00 |
| jbicha | https://salsa.debian.org/webkit-team/webkit/-/blob/wk2/experimental/debian/rules | 21:00 |
| jbicha | it's going to be a problem for Security if we can't routinely build webkit in a reasonable amount of time | 21:01 |
| ricotz | those webkit build times are quite comparable to the libreoffice ones now :\ | 21:03 |
| ricotz | EOD here, I hoping this works out :) | 21:10 |
| === bdrung_ is now known as bdrung | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!