[01:44] <duflu> jbicha, I saw a comment/rumor that gnome-shell only saves some settings on a clean shutdown (when it hasn't crashed). Do you know anything about that?
[01:44] <duflu> Because I *think* I might have just fixed a logout crash that was our top crasher
[01:49] <jbicha> I don't know anything about that
[01:50] <jbicha> GNOME's doing 3.26.2 releases this week so your fix might make it in 🙂
[01:54] <duflu> jbicha, yes that fix is in 3.26.2. Slightly sad the font rendering fix is only in 3.27. Bionic will get that though right?
[01:56] <jbicha> if you're asking whether 18.04 will get GNOME Shell 3.28, that's still undecided
[01:57] <duflu> Oh dear. OK, I'll need to work on either distro patching or pushing for it in 3.26
[01:58] <duflu> I was thinking 3.27 with hybrid support should theoretically solve many people's display issues too
[01:59] <jbicha> you mean fractional scaling, or something different?
[01:59] <jbicha> multi-monitor?
[02:18] <duflu> jbicha, I mean various bug reports where users with multiple GPUs have no picture on screen, or on the wrong screen. I thought the hybrid work would start to solve those
[02:19] <duflu> Even with a single GPU... if I install a radeon card and intel is disabled when Wayland doesn't work. It feels like that stuff could be related to the 3.27 enhancements
[02:20] <duflu> *then* Wayland doesn't work
[02:23] <duflu> 3.26 doesn't feel like a great choice for LTS to me. But that's based purely on the assumption that 3.28 will be ready early enough and will be significantly better
[02:23] <duflu> otherwise I'm just talking about bugs whose fixes can always be backported
[07:05] <oSoMoN> good morning desktoppers
[08:43] <willcooke> hoy
[08:45] <duflu> ahoy willcooke
[08:45] <duflu> Morning oSoMoN
[08:52] <willcooke> taking a surprise day off since I don't have a window in my study
[08:53] <willcooke> I'll be on telegram etc etc
[08:53] <willcooke> have a good day all
[08:53] <willcooke> seb128 will run the meeting
[08:53] <duflu> OK then. Bye willcooke
[09:02] <Laney> PHEW
[09:02] <Laney> what's up
[09:03] <duflu> <insert pun>
[09:03] <duflu> Hi Laney
[09:09] <seb128> hey Laney & desktopers
[09:09] <seb128> hey duflu
[09:09] <duflu> Hello seb128
[09:10] <seb128> duflu, trying to see if you can DoS the meeting with your weekly summaries? ;)
[09:10] <duflu> seb128, verbosity is an inherited problem
[09:11] <duflu> Plus it seems if I ever miss something out, someone will ask about it
[09:11] <seb128> hehe
[09:21] <Laney> hey seb128
[09:21] <Laney> how's it going?
[09:21] <Laney> & duflu!
[09:22] <seb128> Laney, good! you?
[09:23] <seb128> I slept well this night, 11pm to almost 6pm without being waken up in between :)
[09:23] <Laney> 6pm!!!!
[09:23] <Laney> :P
[09:23] <Laney> well done
[09:23] <seb128> haha, thanks
[09:26] <seb128> Laney, seems like you have IRC closed earlier, just as a fyi willcooke decided to take the day off since he has workers redoing his office and it was not possible to use it today
[09:26] <Laney> alright
[09:26]  * Laney is inside 9999 emails
[09:26] <seb128> one more and it might overflow and go to 0
[09:26] <seb128> who knows!
[09:30]  * Laney overflowed
[13:15] <kenvandine> oSoMoN, i think i know the cause of the theme issue
[13:15] <kenvandine> oSoMoN, glib in the backports PPA doesn't have the per desktop overrides support
[13:16] <kenvandine> oSoMoN, fixing that now
[14:17] <seb128> oSoMoN, seems like this morning guess was right there :)
[14:26] <andyrock> seb128: can you schedule me at the end of the meeting
[14:26] <seb128> andyrock, sure
[14:26] <andyrock> I'm preparing the notes
[14:29] <seb128> :-)
[14:30] <seb128> ok, it's meeting time
[14:30] <seb128> let's see if anyone gets surprised due to DST :p
[14:30] <seb128> #startmeeting Desktop Team Weekly meeting - 2017-10-31
[14:30] <meetingology> Meeting started Tue Oct 31 14:30:49 2017 UTC.  The chair is seb128. Information about MeetBot at http://wiki.ubuntu.com/meetingology.
[14:30] <meetingology> Available commands: action commands idea info link nick
[14:30] <seb128> Roll call: andyrock, dgadomski, didrocks (out), duflu (out), jbicha, jamesh (out), jibel/heber, kenvandine, laney, oSoMoN, seb128, tkamppeter, trevinho (out), robert_ancell (out)
[14:31] <kenvandine> o/
[14:31] <jbicha> o/
[14:31] <oSoMoN> 🐵/
[14:31] <seb128> lol
[14:31] <seb128> ok, let's get started
[14:31] <seb128> #topic dgadomski
[14:31] <seb128> dgadomski, hey
[14:32] <dgadomski> hey
[14:32] <Laney> meow
[14:32] <dgadomski> * attended OpenSourceSummit Europe
[14:32] <dgadomski> * found a workaround for bug #1718688, waiting for confirmation from the user who reported that to me
[14:32] <dgadomski> * added some changes and unittest to my fix for bug #1699179
[14:32] <dgadomski> * monitoring bug #1638695 in case any decision made there will be applicable to Xenial.
[14:32] <dgadomski> eof
[14:32] <seb128> dgadomski, thanks
[14:33] <seb128> #topic duflu
[14:33] <seb128> (be ready, long summary coming)
[14:33] <seb128> * Mutter/Xwayland crash handling: https://bugzilla.gnome.org/show_bug.cgi?id=789086
[14:33] <seb128>   - Completed several patch iterations during the week.
[14:33] <seb128>   - Landed upstream: https://git.gnome.org/browse/mutter/commit/?id=054c25f693a
[14:33] <seb128> * Mutter: blurry shell fonts: https://bugzilla.gnome.org/show_bug.cgi?id=645433
[14:33] <seb128>   - Completed several patch iterations during the week.
[14:33] <seb128>   - Landed upstream: https://git.gnome.org/browse/mutter/commit/?id=a37956c9
[14:33] <seb128> * Totem's big stutter bug: https://bugzilla.gnome.org/show_bug.cgi?id=733780
[14:33] <seb128>   - Finally, almost got a code review of my one line fix. Although that was only a Rejection without any reason given. The maintainer does however request much more complicated API changes (affecting multiple projects) instead.
[14:33] <seb128>   - I think I've lost patience with totem for now and will just distro patch the one-line fix soon. Let upstream change the API to their own liking in their own time.
[14:33] <seb128> * Totem's smaller (bug more frequent) stutter bug in GTK also has fixes pending review (links below).
[14:33] <seb128> * Unresponsive touchpads: https://bugs.freedesktop.org/show_bug.cgi?id=98839
[14:33] <seb128>   - Yes, already. I thought this was a task for next year but upstream suddenly woke up this week and has been asking me to test patches. Looks like a fix is close to landing!
[14:34] <seb128>   - Testing patches, testing hardware
[14:34] <seb128>   - Reading kernel code and finding more things worth fixing (next year-ish).
[14:34] <seb128> * Incorrect %CPU in gnome-system-monitor: https://bugzilla.gnome.org/show_bug.cgi?id=788922
[14:34] <seb128>   - Just got frustrated and fixed it myself as a quick Friday evening job.
[14:34] <seb128>   - Patch proposed upstream.
[14:34] <seb128> * Daily bug management across gnome-shell, mutter, gdm3, ubuntu-themes, bluez, pulseaudio, dkms, mir, wayland, totem, mpv.
[14:34] <seb128>   - Mountains and mountains of bug traffic this week, but we mostly kept up.
[14:34] <seb128> * Patches awaiting Ubuntu sponsorship (or needing refreshing for 18.04):
[14:34] <seb128>   - https://code.launchpad.net/~vanvugt/gtk/fix-1698270/+merge/331846
[14:34] <seb128> * Patches awaiting upstream review:
[14:34] <seb128>   - https://bugzilla.gnome.org/show_bug.cgi?id=787001
[14:34] <seb128>   - https://bugzilla.gnome.org/show_bug.cgi?id=773453
[14:34] <seb128>   - https://bugzilla.gnome.org/show_bug.cgi?id=787665
[14:34] <seb128>   ^^^ These have either been ignored for months, or received spurious reviews which I answered and then get ignored. But I would rather they stayed quiet till after my vacation.
[14:34] <seb128> * PulseAudio 11:
[14:34] <seb128>   - Been working on a TODO list for this for a long time. I hope to start on it ahead of schedule (ie. this week instead of December/January as planned).
[14:34] <seb128>  
[14:34] <seb128> (giving us a minute to read/digest that ;)
[14:35] <seb128> ok, next
[14:35] <seb128> #topic jbicha
[14:35] <seb128> jbicha, hey
[14:35] <jbicha> • Synced cairo and merged fontconfig (LP: #1702544) for color emoji support.
[14:35] <jbicha> • Packaged nototools and fonts-noto-color-emoji, uploading to Debian NEW soon
[14:35] <jbicha> • LocutusOfBorg merged remmina 1.2~ (uses freerdp2) & I built vinagre with freerdp2 support too.
[14:35] <jbicha> • Uploaded gtk 3.22.25 to artful/unapproved LP: #1728421
[14:35] <jbicha> • Uploaded fix to bionic for LP: #1722809, robert_ancell will SRU
[14:35] <jbicha> • libgit2 transition
[14:35] <jbicha> • Helped remove mozjs24. IMO mozjs (1.8.5, from Firefox 4) is removable, leaving just mozjs38 and 52 (LP: #1714821)
[14:36] <jbicha> • mozjs52 / gjs and its rdeps are back on s390x. I assume we don't want to start building ubuntu-desktop there again?
[14:36] <jbicha> • Help Needed for troubleshooting third-party scanner driver issues, see LP: #1707352
[14:36] <jbicha> eof
[14:37] <seb128> jbicha, the gnome-user-share bug referenced there was a copy error?
[14:37] <seb128> or is mozjs having to do with it?
[14:38] <jbicha> oops, should have been LP: #1728038
[14:38] <seb128> jbicha, did you try to talk to Robert about the sane-backends issue? it's sort of his domain
[14:38] <seb128> that bug reference makes more sense :p
[14:38] <jbicha> yes, but he didn't have answers yet
[14:38] <seb128> k
[14:38] <seb128> thanks jbicha
[14:38] <seb128> #topic jamesh
[14:38] <seb128> jamesh, hey, unsure if you are around?
[14:39] <seb128> willcooke didn't fwd me your update
[14:39] <seb128> k, seems he's not there
[14:39] <seb128> next
[14:39] <seb128> #topic kenvandine
[14:39] <seb128> kenvandine, hey
[14:39] <kenvandine> * oSoMon found an issue with gtk theme settings not being picked up in a fresh user session.  This only affects 17.10, works as expected on 16.04, zesty, etc.  I suspect this has to do with per desktop overrides.  I'm refreshing our backport of glib now to get that change to confirm it fixes it.
[14:39] <kenvandine> * Did some digging into yelp support for our snaps.  I think we might be able to solve the issue of finding help files with the use of the OpenURI support in th
[14:39] <kenvandine> e freedesktop portal and a patch to snapd to support help uris.
[14:39] <kenvandine> * Did some work on the desktop helpers, creating symlinks to the xdg dirs so they properly appear in the file picker.  Need to do a little more work before subm
[14:40] <kenvandine> itting a PR.
[14:40] <kenvandine> * Backporting some artful fixes to the gnome-3-26 PPA to get them in the platform snap
[14:40] <kenvandine> eof
[14:41] <seb128> kenvandine, the glib/per desktop override is basically what we suspected when we discussed that with Olivier this morning, let's see how the refresh is working
[14:41] <seb128> thanks kenvandine
[14:41] <oSoMoN> thanks kenvandine !
[14:41] <kenvandine> ah, you guys talked about it :)
[14:42] <seb128> yeah, I read the backlog and I mentioned that the dconf file existing on start was probably not the issue
[14:42] <seb128> if the file exists it's probably something buggy that does a write on loggin
[14:42] <seb128> which shouldn't be done, desrt chassed those buggy cases in the past
[14:42] <seb128> anyway let's see
[14:42] <seb128> next
[14:43] <seb128> #topic Laney
[14:43] <seb128> Laney, hey
[14:43] <Laney> hi
[14:43] <Laney> • v. short week 💀
[14:43] <Laney> • finally figured out a way to filter shell extensions in gnome-software so you see the usable ones, going to submit a MR later 👻
[14:43] <Laney> • bos02 cloud region stability work got finished, worked today on enabling arm64/s390x cloud autopkgtesting - should be good to go in proposed-migration once some more hardware is enabled in openstack 🧛
[14:43] <Laney> • helped a bit with bionic archive opening (NewReleaseCycleProcess checklist stuff) 🕷️
[14:43] <Laney> 🎃
[14:43] <oSoMoN> happy halloween Laney :)
[14:43] <jbicha> which ones are usable?
[14:44] <kenvandine> :)
[14:44] <seb128> thanks Laney
[14:45] <seb128> #topic oSoMoN
[14:45] <seb128> oSoMoN, hey
[14:45] <Laney> ones that are compatible with your shell version
[14:45] <oSoMoN> hey there
[14:45] <oSoMoN> • chromium 62.0.3202.62 published to -security and -updates for all supported series, and 62.0.3202.75 ready for publication
[14:45] <oSoMoN> • managed to build chromium 63 (beta) on trusty, using gcc-mozilla (gcc 4.9) and statically linking its libstdc++
[14:45] <oSoMoN> • first chromium 64 (dev) build ongoing in PPA
[14:45] <oSoMoN> • updated chromium snap after a call for testing, useful feedback and a few fixes: 62.0.3202.75 in stable channel, 63.0.3239.18 in beta channel
[14:45] <oSoMoN> • prepared libreoffice 5.4.2 SRU for artful (bug #1728072), now in unapproved queue
[14:45] <oSoMoN> • preparing libreoffice 5.4.2 upload for bionic with a couple of fixes (debhelper and icu)
[14:45] <oSoMoN> • libreoffice snap 5.4.2: enabled spellchecking (bug #1680928), hyphenation and thesaurus, and issued call for testing, got good feedback so far, planning on promoting to stable in a couple of days if no blocking issue arises
[14:45] <oSoMoN> EOF
[14:45] <seb128> thanks oSoMoN
[14:46] <seb128> #topic seb128
[14:46] <seb128> • spent most of the week triaging&debugging&sending upstream launchpad incoming bugs
[14:46] <seb128> • sponsoring (libreoffice artful SRU, u-c-c bugfix + xenial SRU)
[14:46] <seb128> • some trello board reviews and administrative work

[14:47] <seb128> #topic tkamppeter
[14:47] <seb128> tkamppeter, hey
[14:47] <tkamppeter> - OpenPrinting: Studied GitHub Pages as possibility to host the site and manage the content, including news blog in GIT repos.
[14:47] <tkamppeter> - Google Summer of Code 2017: Planned the distribution of the funds received from Google.
[14:47] <tkamppeter> - hplip: Uploaded missing proprietary plugins to OpenPrinting. HP failed to send them to me earlier.
[14:47] <tkamppeter> - cups, cups-filters, hplip, ...: First syncs into Bionic.
[14:47] <tkamppeter> - CPDB (Common Print Dialog Backends): Posted on Debian Printing mailing list to ask them to start packaging.
[14:47] <tkamppeter> - Bugs.
[14:48] <seb128> thanks tkamppeter
[14:48] <seb128> #topic robert_ancell
[14:48] <seb128> - gnome-software/snapd-glib bug fixing
[14:48] <seb128> - Uploaded updated gnome-software/snapd-glib packages to ppa:ubuntu-desktop/gnome-software
[14:48] <seb128> - Prepared artful gnome-software/snapd-glib SRU (in QA testing now).
[14:48] <seb128> - snapd-glib 1.24 released
[14:49] <seb128> #topic andyrock
[14:49] <seb128> andyrock, your turn :)
[14:49] <andyrock> 1. Livepatch desktop integration:
[14:49] <andyrock>      1.1 working on software-properties to integrate with the new macaron-based API to retrieve lp token
[14:49] <andyrock>      1.2 talking with the right guys to get pymacaroonbakery in Debian and artful
[14:49] <andyrock> 2. OEM bug (https://bugs.launchpad.net/bugs/1716359)
[14:49] <andyrock>      2.1 mp for GCC master (already committed)
[14:49] <andyrock>      2.2 mp for ucc trunk
[14:49] <andyrock>      2.3 sru for ucc xenial
[14:49] <andyrock> 3. Updated the branch to fix osd progress bars in artful
[14:49] <andyrock> 4. Sru for the gtk crash when adding new online accounts
[14:49] <andyrock> eow
[14:50] <seb128> thanks andyrock
[14:50] <seb128> #topic aob
[14:50] <seb128> k, jbicha has some of those at least
[14:50] <seb128> jbicha, I guess you can start :)
[14:50] <jbicha> https://lists.ubuntu.com/archives/ubuntu-desktop/2017-October/005272.html
[14:51] <jbicha> #1 fontconfig: should we keep the templates in /etc indefinitely or ok to follow upstream?
[14:52] <seb128> it's probably ok to follow upstream/debian if you are confident it's not going to create issues for users upgrading
[14:53] <seb128> I think I'm the one who rolled back on those changes at the time
[14:53] <jbicha> I am not confident, it will break users who symlinked one of the extra available configs
[14:53] <seb128> but that's because I didn't want to have to deal with the migration and we were updating ahead of Debian
[14:53] <jbicha> yes, I think that was like 2012 (!)
[14:53] <seb128> do we have any issue with the current situation?
[14:54] <jbicha> no
[14:54] <seb128> well, my position would be to stay with what we have unless somebody is wanting to deal with the change properly, which means addressing the upgrade problems
[14:55] <jbicha> ok, that's fine
[14:55] <jbicha> moving on to #2: dealing with headerbar patches
[14:55] <jbicha> Khurshid also replied to my email: https://lists.ubuntu.com/archives/ubuntu-desktop/2017-October/005273.html
[14:56] <seb128> tricky
[14:57] <jbicha> the headerbar patches are useful for Unity becuase otherwise LIM is broken there
[14:57] <seb128> we said we would drop them to lower delta/maintainance cost
[14:57] <seb128> but if we have contributors doing the work it's a bit less easy to say "no" rather than including them for this cycle
[14:58] <seb128> are those patches known to create any problem?
[14:58] <jbicha> they look bad on other desktops (that bug was filed by elementary OS ) and I saw similar badness in Xfce
[14:58] <jbicha> so, changing the patches to only apply against Unity are an improvement
[14:58] <seb128> right, but the updates patch change the behaviour to be in unity only
[14:58] <seb128> right
[14:58] <seb128> wdyt?
[14:59] <seb128> I think I'm fine either way
[14:59] <jbicha> actually, some of the patches were set to apply anywhere that wasn't GNOME which was why the bug was filed
[14:59] <seb128> I guess Laney would be in favor of dropping them to low delta with upstream
[14:59] <seb128> right
[14:59] <seb128> that was probably the wrong choice
[15:00] <seb128> I think I would be slightly in favor of removing those changes
[15:00] <seb128> we can investigate this gtk3-nocsd mentioned on the list
[15:00] <jbicha> it sort of made sense at the time but I think CSD are reasonably well supported by desktops now
[15:00] <seb128> I don't know about it
[15:00] <jbicha> how about we think about it more then?
[15:00] <seb128> yes
[15:00] <seb128> let's follow up on the list
[15:01] <jbicha> ok, #3 ok to sync gnome-user-share?
[15:02] <seb128> what's the user experience today with that one?
[15:02] <seb128> you can enable it from an ui and it prompts you to install packages?
[15:02] <jbicha> there is no user experience in the latest version of gnome-user-share itself
[15:02] <seb128> well, with the one we currently have
[15:03] <seb128> like what do we give up on?
[15:03] <jbicha> with the old version, it has a file sharing thing used by Unity
[15:04] <jbicha> but we have been patching it to break the file sharing feature for years becuase we didn't want that apache bin in the desktop install
[15:05] <seb128> ideally we would have g-c-c installing packages on demand when you try to enable it there I guess?
[15:05] <sil2100> kenvandine: hey!
[15:05] <sil2100> kenvandine: you around?
[15:05] <seb128> we are in a meeting
[15:05] <sil2100> Ah, oops, excuse me then!
[15:05] <seb128> but almost over
[15:05] <seb128> no worry
[15:05] <kenvandine> hey sil2100
[15:05]  * sil2100 hides
[15:05] <seb128> jbicha, let's continue on the list
[15:05] <jbicha> ok
[15:06] <seb128> sorry I feel like we don't have a good quorum today
[15:06] <seb128> willcooke is off and didrocks as well (who usually has input on such topics)
[15:06] <jbicha> g-c-c hides the File Sharing subpanel if gnome-user-share is not installed
[15:06] <xnox> Laney, whoop whoop re:bos02 =)
[15:06] <seb128> and others don't seem to have much opinion
[15:06] <Laney> hi
[15:07] <seb128> jbicha, on a similar topic we should perhaps revisit what we do with nautilus-share, it's quite buggy atm
[15:07] <seb128> jbicha, the pam integration package it tries to install has been remove in xenial
[15:07] <Laney> might be worth thinking about 3b in jbicha's email
[15:07] <seb128> so installation fails and auth isn't working unless you go add users manually
[15:07] <seb128> hey Laney :)
[15:08] <Laney> it'd be cool to have those features
[15:08] <seb128> right
[15:08] <seb128> I think I would like to have it
[15:08] <seb128> either by installation things
[15:08] <Laney> guessing security would want to remove it
[15:08] <Laney> review*
[15:08] <Laney> or remove :P
[15:08] <seb128> or by changing g-c-c to do "install on demand" rather than hidding
[15:08] <Laney> mmm
[15:09] <jbicha> Laney: you were right the first time ;)
[15:10] <seb128> jbicha, ok, let's continue on the list, it seems like we would like to have the feature user visible but we need to define the best way to get there
[15:11] <jbicha> sure, thanks
[15:11] <seb128> thanks for bringing those topics
[15:11] <seb128> sorry that there was not much traction to discuss them
[15:11] <seb128> as said we are missing some people
[15:11] <seb128> let's continue on the list and maybe try to get a conclusion next week
[15:11] <jbicha> it's fine, some are complicated and it's good to think things through
[15:11] <seb128> right
[15:11] <seb128> any other topic?
[15:13] <seb128> seems not, it's a wrap then
[15:13] <Laney> nein
[15:13] <Laney> thanks!
[15:13] <seb128> thanks everyone!
[15:13] <seb128> #endmeeting
[15:13] <meetingology> Meeting ended Tue Oct 31 15:13:12 2017 UTC.
[15:13] <meetingology> Minutes:        http://ubottu.com/meetingology/logs/ubuntu-desktop/2017/ubuntu-desktop.2017-10-31-14.30.moin.txt
[15:13] <oSoMoN> thanks everyone
[15:13]  * kenvandine waves
[15:14] <oSoMoN> kenvandine, please let me know when the glib package in the backports PPA is refreshed so I can trigger a LO snap rebuild
[15:15] <kenvandine> oSoMoN, i will
[15:15] <kenvandine> waiting for an LP build that's queued
[15:17] <Laney> jbicha: on fontconfig, would a symlink /etc/fonts/conf.avail/ -> the new directory work?
[15:17] <Laney> that or a maintainer script fixing up any symlinks in conf.d
[15:18] <Laney> you'd have to fix all font pkgs that install into conf.avail too
[15:19] <Laney> or make fontconfig read both?
[15:20] <jbicha> fontconfig itself reads from /etc/fonts/conf.d/ these are just templates that can be symlinked
[15:21] <jbicha> maybe you're familiar with web servers using sites-available and sites-enabled ?
[15:21] <Laney> I remember the seup
[15:21] <Laney> setup
[15:21] <jbicha> I didn't know anything about fontconfig until I started working on this emoji thing
[15:21] <Laney> so the only problem is that someone might have manually symlinked something that moves into /usr/share/
[15:22] <jbicha> right
[15:22] <jbicha> I'm not sure how making that directory a symlink would work since there are lots of other Debian packages that try to install files to that /etc directory
[15:22] <Laney> shouldn't be required, I don't think emptying it should be a goal
[15:23] <Laney> at least not now
[15:23] <jbicha> apt-file search /etc/fonts/conf.d
[15:24] <Laney> most packages install into .d and .avail
[15:24] <Laney> and then when they move to /usr they should fix old symlinks
[15:25] <oSoMoN> chrisccoulson, did you see my ping yesterday that chromium 62.0.3202.75 is ready for publication in the stage PPA?
[15:25] <jbicha> we can symlink files, I don't have any experience with symlinking directories in Debian packages if other Debian packages might try using those directories
[15:25] <Laney> nah don't think that is required
[15:26] <jbicha> I think I'm confused by what you are suggesting then
[15:26] <Laney> I think fontconfig itself should only need to care about packages that it is now moving to /usr that weren't symlinked from conf.d before
[15:26] <jbicha> ok, we can add symlinks for all of those files
[15:26] <Laney> s/packages/files/
[15:27] <Laney> either symlink those from conf.avail or a maintainer script to fix up any now dangling symlinks in conf.d
[15:28] <Laney> don't actually see any on my system though
[15:28] <Laney> oh no I do
[15:29] <Laney> e.g. 10-no-sub-pixel.conf
[15:32] <Laney> an advantage of fixing them up is that it's a droppable patch post-LTS but symlinking less so
[15:34] <jbicha> Laney: so sort of like https://anonscm.debian.org/git/pkg-freedesktop/fontconfig-debian.git/tree/debian/fontconfig-config.maintscript
[15:35] <jbicha> but with mv_conffile instead of rm_conffile
[15:36] <Laney> that's the removing half, to fix up it's something like for file in /etc/conf.d/*; if symlink pointing to one of these removed files, remove it and make symlink to /usr/share/...; done
[15:37] <Laney> did Debian bother doing anything like that?
[15:38] <jbicha> it doesn't look like it based on https://anonscm.debian.org/git/pkg-freedesktop/fontconfig-debian.git/log/debian/rules
[15:39] <jbicha> let me check…
[15:39] <Laney> not that I can see
[15:44] <jbicha> well this doesn't work: https://bugs.debian.org/714157
[15:44] <Laney> yeah don't do that if other packages are going to try to put files there
[15:46] <jbicha> yeah my instinct said that was a bad idea
[15:46] <Laney> I thought the goal was to stop putting anything in that /etc directory
[15:46] <Laney> but it's not, so no need to go through that pain
[15:48] <jbicha> yeah, it seems likely there are 3rd party deb's that we couldn't fix
[15:50] <jbicha> I don't have much experience with maintainer scripts but I think I have enough to start with now
[15:50] <jbicha> thanks for the ideas
[15:51] <Laney> should be fun
[15:51] <Laney> happy to review it if you want to show me something
[15:53] <Laney> laney@nightingale> readlink -e /etc/fonts/conf.d/10-antialias.conf                                                                                                                                                                ~
[15:53] <Laney> /etc/fonts/conf.avail/10-antialias.conf
[15:53] <Laney> that's probably a thing to use
[15:54] <Laney> lies, it doesn't work for broken symlnks
[15:54] <Laney> readlink -m
[16:37] <oSoMoN> ricotz, have you verified that packages build in bionic with https://git.launchpad.net/~libreoffice/ubuntu/+source/libreoffice/commit/?id=91b98436ca90dad3bb11a5944be70b39c057e8ec ?
[16:39] <ricotz> oSoMoN, yes
[16:40] <oSoMoN> ricotz, excellent, thanks!
[17:04]  * andyrock misses c++ 💔
[17:14] <kenvandine> oSoMoN, new glib didn't fix it
[17:14] <oSoMoN> :/
[17:15] <kenvandine> oSoMoN, it must still be related though
[17:15] <kenvandine> i'm sure this worked relatively recently
[17:15] <kenvandine> but i probably haven't tested this scenario since the per desktop overrides work landed
[17:34] <xnox> https://code.launchpad.net/~indicator-applet-developers/+activereviews looks ungardened
[18:18]  * oSoMoN EOD
[18:18] <oSoMoN> good night all
[22:54] <jbicha> I opened a trello card for the sane-backends issues since I don't think I'll be able to do much more troubleshooting myself on it