[00:31] hi there [00:32] RikMills: do you mind if I build frameworks 5.79 on the ninjas PPA to add it to the experimental PPA [00:32] mostly because we will end up having fw 5.79 + plasma 5.21 and fw is going to take a while to be actually available [00:36] santa_: that sounds ok. obviously please make sure you change the settings so not building against proposed. and test the 2 ppas together, before copying them together into one [00:37] sure, that's the idead [00:37] in fact I'm testing with VMs, but the staging packages depend on -proposed things [00:38] I'm rebuilding everything of fw/plasma/apps in tritemio btw [00:38] http://tritemio-groomlake.duckdns.org/build-status/buildstatus_ubuntu-exp/ [00:38] ↑ doing fw atm [00:39] [16:29] discuss unfreezing the archive tomorrow. [00:39] [16:29] figure out if we can allow glibc to migrate. [00:39] that from #u-release sounds hopeful [00:40] but must assume glibc will be stuck for a while just in case [00:45] yes, I have been reading the channel, before the upload queue "spam" XD [00:45] I am probably being over hopeful :/ [00:48] as long as there are no more issues, I will probably put 5.21.0 into the hirsute queue in the morning [00:49] as do the syntaxhighlighting hotfix [00:49] *and do [01:31] btw, no issues so far with ppa:kubuntu-ppa/experimental. But I have only run it for few minutes.... [01:47] But did it look nearly as good as the youtube video? [02:23] santa_: why are you building amd64 only on the wrong ppa? [02:23] ppa:kubuntu-ninjas/frameworks only has 2GB as well [02:24] would be better IMO in ppa:kubuntu-ninjas/plasma even though it is frameworks? [02:24] especially as you plan to copy to a ppa which has all arches enabled [02:28] hmmm I guess no huge deal [02:31] I would have just tried to match the eventual target ppa [02:44] syntax-highlighting and kdnssd 5.79.1 hotfixes staged [12:22] Howdy folks [15:14] hi everyone [15:14] afternoon [15:15] RikMills: yes, I realized after uploading, but not a big deal imho, since this is just a temporary thing for testing [15:15] ok [15:15] also the reason why we have all archs activated in staging is mostly to update the symbols files, which we already did [15:16] so yeah, if someone is testing plasma in an IBM mainframe, he/she won't have the new fw, only plasma 5.21 [15:16] lol [15:18] * RikMills puts his mainframe away in disappointment [15:18] needless to say if I were rich I would have a mansion with a small cpd one machine per ubuntu/debian arch to build and test [15:18] and an assitant named Alfred [15:20] hopefully the 'fun-with-fakeroot' is almost over [15:21] yeah, I have seen the bug report, my test rebuilds in the night were fine btw [15:21] anyway, I have to leave you know. I will have time again in the night to finish fw, TODO: [15:21] yeah, today has just been, 'why is riscv64 now borked?' [15:21] - provide that hotfixes you do [15:21] - test [15:22] - copy packages to experimental PPA [15:22] feel free to do those if you have time before I do [15:22] see you later, 100000000000000 thanks for your work :) [15:22] thank you as well! [17:37] santa_: there was a reason I dropped kconfig acc test :/ [17:37] test deps are not satisfiable on i386 [17:37] so FAIL [17:37] I forgot [17:40] I am just going to nuke them again. There is no way to skip [21:15] RikMills: ack, and I'm find with the workaround. just FYI I think we could still void them just for i386 [21:16] in debian/tests/control: [21:16] set dh-acc depends as "dh-acc [!i386]" [21:17] in debian/tests/acc: [21:17] add an if to detected if the arch is i386; if so, exit 0 [21:17] I think that could work [21:18] oh, for some reason I thought that wouldn't work in tests/control [21:18] FTR: https://people.debian.org/~mpitt/autopkgtest/README.package-tests.html [21:18] RikMills: have you disabled other tests for the same reason? if the strategy above works maybe we could re-enable them [21:19] no, I think that was the only one [21:20] another possibility would be disabling them like you were attempting to do some time ago; the strategy would be: [21:20] rename debian/tests/control -> debian/tests/control.disable [21:21] I would add support in gbp-tritemio to undo the rename on the fly without touching the git local clone [21:22] this way we would get them built with acc tests in tritemio, no acc tests in the archive, saving some cycles for the ubuntu infra [21:22] let me ponder on that. I am not messing with the version I put in the upload queue for now [21:22] in principle that sounds useful though [21:23] sure, of course, just think about it and we could assess it better one of these weekends [21:24] btw I think I'm going to have time today to finish the fw 5.79 + plasma 5.21 case if possibru [21:25] but I need to do some personal stuff before that [21:25] cool. I got distracted