=== RAOF is now known as Guest56578
[09:19] <cjwatson> sil2100 (CC seb128, wgrant): LP's language pack exports for disco look sensible.  I suggest going ahead and doing a langpack packages build from them.  Any objection to me unhiding translations for disco?
[09:19] <seb128> cjwatson, I though I did that last week?
[09:20] <cjwatson> seb128: Oh.  You disregarded the order on the process wiki page :)
[09:20] <cjwatson> (https://wiki.canonical.com/Launchpad/Translations/UbuntuOpenings)
[09:20] <seb128> ah, sorry
[09:21] <cjwatson> Should be no harm done; that part of the process is much less delicate than the start
[09:26] <seb128> k
[09:27] <seb128> cjwatson, but do you know what team has access to the +admin from translations? I think I used to have access to that page (is that the one where there is the box telling if the next export is a full one or a delta?)
[09:31] <sil2100> Seeing that we have the first full language export already, let me kick some base langpacks in some moments
[09:32] <seb128> speaking of langpacks, I got an email of someone wondering why 18.10 isn't getting weejkly ppa updates
[09:32] <seb128> is that something we should have set up and didn't?
[09:34] <cjwatson> seb128: ~ubuntu-translations-coordinators and ~rosetta-admins, I think
[09:47] <Laney> mwhudson: hah
[09:48] <cjwatson> mvo: just a post-sprint reminder to cut down the set of Commands files on cnf-extractor.internal and to tell me what the client actually uses :)
[09:48] <Laney> cjwatson: thanks for that merge
[09:48]  * Laney notices that partman-base is quite far behind in Ubuntu and screams
[09:49] <Laney> WOH someone did actually just merge it
[09:49] <cjwatson> NP.  Good luck with that merge ...
[09:49] <cjwatson> Oh, hah
[09:49] <mvo> cjwatson: thanks! sorry for letting you wait on this, I was on vac but will get to it today :)
[09:49]  * Laney hugs xnox 
[09:49] <cjwatson> mvo: It's OK, so was I
[09:50] <Laney> oh right, it was more or less just translations
[09:51] <xnox> and i was away on leave =) back today =)
=== cpaelzer_ is now known as cpaelzer
[10:13] <Laney> :>
=== yiko is now known as yikoru
=== ricab is now known as ricab|lunch
[13:04] <cpaelzer> jbicha: juliank; mitya57: vorlon: I have seen that all of you hit retest for tryton-server blocking various things. I haven't found a bug about it at the tryton-server package nor does the issue trigger when run in a local VM
[13:04] <juliank> It's an odd thing
[13:04] <cpaelzer> is there any insight on what is going on there already - after all it is a cross-arch segfault which is odd
[13:04] <mitya57> cpaelzer: it succeeds if you retry it with all-proposed=1, that pulls sqlite3 from -proposed
[13:05] <juliank> it passes in Debian, but used to crash there too
[13:05] <mitya57> see Debian #923038
[13:05] <ubottu> Debian bug 923038 in libsqlite3-0 "tryton-server tests segfault with sqlite 3.27.1" [Important,Fixed] http://bugs.debian.org/923038
[13:05] <cpaelzer> mitya57: that might be why it passes for me atm, thanks mitya57
[13:05] <juliank> well, that's helpful
[13:06] <cpaelzer> mitya57: but I see you had triggered sphinx with sqlite3/3.27.2-1 and it still failed a few times
[13:07] <cpaelzer> mitya57: what made the difference eventually?
[13:07] <mitya57> cpaelzer: sqlite3 is part of base chroot so adding it to triggers does not actually upgrade it. Only all-proposed=1 makes it really use the proposed sqlite3.
[13:09] <cpaelzer> thanks
[13:11] <jbicha> ooh, thanks :)
[13:16] <dja> ddstreet: sorry about the chaos for LP: #1818340 ! Was not my intention to leave you with that sort of time bomb!
[13:16] <ubottu> Launchpad bug 1818340 in systemd (Ubuntu Cosmic) "systemd-networkd core dumps in bionic-proposed" [Critical,In progress] https://launchpad.net/bugs/1818340
[13:18] <dja> ddstreet: by the way, is there anything in git-ubuntu to make dealing with quilt less painful? Like to turn a git patch into a debian/patches/ patch? I ended up doing it with a bunch of awful intermediate steps when I was working on the original bug and have lost all the command history, so I sort of gave up for this.
[13:20] <ddstreet> dja no prob, it happens :)
[13:20] <ddstreet> re: git ubuntu, i dont think there is anything yet that easily will turn normal git patches that you've applied already into d/p/ patches
[13:21] <ddstreet> you can just quilt import a git patch, but of course if it needs adjustment, the quilt push will fail and you'll have to hand-edit it, which can be quite annoying
[13:22] <dja> Ah I wonder why I never thought of that
[13:22] <dja> or maybe I did and they just never worked
[13:22] <dja> oh well
[13:22] <dja> It's late here so I will be interested to read your comments on the bug tomorrow!
[13:22] <ddstreet> yep, i'll try to dig into it and see what i can find :)
[13:22] <ddstreet> thnx!
[13:33] <cjwatson> git-dpm and git-buildpackage both have various approaches to let you work on a separate patched branch and have that be automatically serialised into debian/patches/.  If you use those systematically then you never have to touch quilt.  Do read the documentation in depth first though.
[13:40]  * juliank loves gbp pq
[13:41] <juliank> One thing I don't like is that it reexports unmodified patches, so you can't just import the patches, and add your own when you want to make a change on a package not using it
[13:41] <juliank> because you get unrelated changes that way
=== ricab|lunch is now known as ricab
[14:43] <rbasak> tjaalton: just glanced at your libxkbcommon upload to Bionic. While you're fixing up the changelog, I'm of the opinion that the changelog should explain what is being fixed in the SRU itself, since users actually read those.
[14:43] <tjaalton> rbasak: alright
[14:44] <rbasak> Like "sync packaging to allow coinstallation of i386 and amd64 (LP: #XXXXXX)" or similar would be better for example.
[14:44] <rbasak> I'm interested to know if others on the SRU team would agree.
[14:44] <tjaalton> I'll just copy the entry from 0.8.0-2
[14:44] <rbasak> (so this isn't a hard reject if someone else thinks that's OK, etc)
[14:44] <tjaalton> and upload again
[14:44] <tjaalton> no big deal
[14:44] <tjaalton> it should show on the diff though, but I know users don't read that far
[14:45] <rbasak> Agreed. I'd prefer users to want to take SRUs, rather than react with "yet another SRU? why?!"
[14:46] <rbasak> Thanks :)
[14:46] <tjaalton> dropped the old one too
[15:30] <rbalint> i would like to create the ubuntu-wsl metapackage (to help us adding utilities later) but i can't find any process for that
[15:31] <rbalint> should i just go ahead and prepare and upload it and let the archive admins decide or i should update ubuntu-meta?
[15:33] <rbalint> btw is there a vcs for ubuntu-meta?
[15:34] <rbasak> I'm not very familiar with ubuntu-meta. I have only used it to regenerate seed-generated metapackages.
[15:34] <rbasak> I don't believe there's a VCS.
[15:34] <rbasak> Does ubuntu-meta do non-seed-based metapackages? I' m not aware of that.
[15:36] <cjwatson> It's of course technically possible but IMO it shouldn't.  If you want something in ubuntu-meta, add a seed for it
[15:37] <rbalint> cjwatson, i'm ready to do that if that's the process
[15:37] <cjwatson> There at least never used to be a VCS for ubuntu-meta because so much of it was autogenerated that there didn't seem a great deal of point
[15:38] <rbalint> rbasak, cjwatson, since i need to add ubuntu-wsl to ubuntu-meta's i guess i need to update the seed first, then upload an updated ubuntu-meta
[15:39] <juliank> I was just wondering: How would I i18n a string like "%d updates are available, %d are security updates"
[15:39] <juliank> There are variants for each %d
[15:39] <rbalint> cjwatson, my imagined point would be to agree on a change first before it hits the archive
[15:39] <juliank> like this one needs 4 variants in english
[15:49] <rbasak> juliank: some discussion here: https://www.gnu.org/software/gettext/manual/html_node/Plural-forms.html
[18:06] <sil2100> cjwatson, seb128: base langpack importer is running
[18:26] <seb128> sil2100, great
[21:55] <rbalint> cjwatson, rbasak i'm adding the wsl seed, i hope it is ok this way https://code.launchpad.net/~rbalint/ubuntu-seeds/+git/platform/+merge/364003
[21:59] <infinity> rbalint: There seems to be a lot of duplication there with minimal.
[22:00] <infinity> rbalint: Is this all to avoid having init?
[22:22] <rbalint> infinity, yes there is a lot of duplication with minimal, but wsl does not need initramfs-tools
[22:22] <rbalint> infinity, no, it is about pulling in wslu and pulling in packages later when they show up in the archive
[22:23] <rbalint> infinity,  minimal is too much for wsl