[01:57] <tsimonq2> rbasak: It seems you had a discussion about this back in March, but ~ubuntu-sponsors has been subscribed: bug 1758196
[01:57] <tsimonq2> It would be great if you could take a look when you have a second.
[10:51] <ricotz> acheronuk, the firefox 62 beta7 build will fix the problem
[10:51] <acheronuk> ricotz: great. thanks. :)
[11:28] <rbasak> tsimonq2: I commented on the bug
[11:55] <mdeslaur> juliank: hi! do you have a libgcrypt merge planned soon?
[11:58] <juliank> mdeslaur: no, but I can do that now
[11:58] <mdeslaur> juliank: awesome, thanks!
[15:57] <bdmurray> juliank: Thank for opening bug 1780996, to work on it I should assign a package task to myself and then see what release it has an await trigger in correct?
[15:58] <juliank> bdmurray: I think that works
[15:58] <juliank> JFTR for everyone else: This is a huge meta bug tracking the conversion of triggers to noawait in a ton of packages
[15:59] <juliank> In the hope to make xenial -> bionic upgrades more successful
[15:59] <juliank> People are welcome to join
[15:59] <juliank> :D
[15:59] <bdmurray> Its important to look and see if there are other changes needed in addition to just flipping the trigger though.
[15:59] <juliank> yes
[16:00] <juliank> We either need to flip it, or if it really needs to be await we need to add dependencies to all packages triggering
[16:00] <juliank> it
[16:00]  * juliank is bad at line breaks, apparently
[16:00] <juliank> File triggers should generally be the most safe to convert, except for gsettings schema and some friends.
[16:01] <juliank> Basically ask yourself the question: If this trigger does not get run, is the package unusable?
[16:02] <juliank> or well, does not get run before the package is being configured.
[16:02] <juliank> Because you do want the package to be usable when configured.
[16:04] <juliank> The rule for triggering packages is: Either depend on what you're triggering or do not wait for it.
[16:04] <juliank> We might not catch all of the packages triggering file triggers
[16:05] <juliank> um
[16:05] <Laney> Would appreciate you forwarding any changes you find back to Debian.
[16:05] <Laney> at least for desktop stuff
[16:05] <bdmurray> Laney: we searched for things that had already converted from await to noawait since xenial
[16:06] <bdmurray> so they are presumably already fixed in Debian
[16:06] <juliank> Hmm
[16:06] <juliank> I think I also added some packages that are not fixed in cosmic yet
[16:07] <juliank> best to check cosmic, bionic, and xenial for each package
[16:07] <juliank> and mark the task as fix released or triaged accordingly
[16:07] <juliank> like I did for appstream
[16:07] <juliank> and libreoffice
[16:08] <juliank> I'm _not_ going to do libreoffice, my bandwidth is not high enough for that
[16:08] <bdmurray> Right, so looking at xpdf its the same version in bionic and cosmic and has changed to noawait so I'll set those to Fix Released.
[16:13] <juliank> bdmurray: I'm using  dch -i "Convert triggers to noawait (LP: #1780996)"  for the changelog
[16:13] <juliank> and then dch -r and distro fixing
[16:14] <ricotz> juliank, bdmurray, you can forward the noawait problem of libreoffice to debian too
[16:17] <juliank> bdmurray: Hmm, we could also mark the already fixed ones as invalid; and then have a way to distinguish "really fixed" vs "already fixed" for forwarding purposes? Not sure.
[16:39] <juliank> ricotz: it's fixed in cosmic/bionic already, so I don't think we're going to check if that's a local change or not
[16:39] <juliank> especially as libreoffice is a -0ubuntu
[16:40] <juliank> I think the rule is: If it's not fixed in cosmic yet, we check and forward it
[16:40] <juliank> otherwise it gets annoying
[16:56] <LocutusOfBorg> [16:55:59] <LocutusOfBorg> can anybody please give me an hint about libunistring and libidn build failures? they seems to be a regression in release, not happening in debian, glib related, but I can't see toolchain changes there
[16:58] <smoser> bdmurray: your review of https://code.launchpad.net/~smoser/software-properties/trunk.lp-1779302-retry-recv-keys/+merge/349213 would be appreciated or *someone* to review that.
[16:58] <smoser> bug 1779302
[17:06] <bdmurray> smoser: I'll have a look in a bit
[17:27] <bdmurray> mitya57: Do you plan on fixing sphinx for xenial?
[17:28] <ricotz> juliank, you mean it is not an issue in libreoffice 6.0.3-0ubuntu1 ?
[17:28] <juliank> ricotz: yes
[17:28] <ricotz> so "interest-noawait" is fine?
[17:28] <juliank> yes
[17:28] <ricotz> ah I see
[17:28] <ricotz> good
[17:29] <juliank> bad are interest and interest-await, and activate, and activate-await
[17:29] <juliank> well, the -await ones are explicit, so they are likely really needed.
[17:29] <juliank> the ones without a suffix are await for legacy reasons
[17:36] <mitya57> bdmurray: I don’t have much time so if you want to fix it yourself, please do
[17:37] <smoser> bdmurray: i certainlyi would not be opposed to review of https://code.launchpad.net/~smoser/ubuntu-archive-tools/package-sets/+merge/348334 also.
[17:37] <mitya57> bdmurray: the relevant commit is https://salsa.debian.org/python-team/modules/sphinx/commit/e65ad395b01708fda2367030c54ef4e702d28a3b
[17:38] <bdmurray> mitya57: Cool, I'll take it. There were no other changes needed?
[17:39] <mitya57> bdmurray: no other changes, just changing the type worked fine
[21:45] <tsimonq2> rbasak: Thank you.