[09:29] hi clivejo [09:29] morning [09:41] clivejo: moin, the removal in that last log you posted was a package install failure [09:43] uhm, that's very wrong actually o.O [10:52] hm... I'm kind of starting to like ruby... [10:52] * yofel feels like he just said something crazy [10:59] * lordievader stares at yofel wide-eyed [11:05] * clivejo checks yofel's temperature [11:06] I think you're ill [11:06] yofel: http://paste.ubuntu.com/14025920/ [11:06] https://speakerdeck.com/erniemiller/thats-not-very-ruby-of-you?slide=26 *cough* [11:06] XD [11:07] full sample for general enjoyment http://paste.ubuntu.com/14025930/ [11:09] alias print puts, wth? [11:10] this is just.. [11:11] mad? [11:11] crazy? [11:11] lets say, misdirected mindset [11:11] it's for when you want your print to behave incorrectly [11:11] a print that appends newlines is no real print IMO :P [11:13] is that how you have to define an object in ruby? [11:13] clivejo: no, that's how you define a ruby object to look and work as a python one [11:14] just in case you ever want to write a language to behave like a different language [11:14] O_O [11:15] sitter: does this look reasonable or did I do something very wrong? https://git.launchpad.net/~kubuntu-packagers/+git/kubuntu-automation/tree/git-push-all [11:15] (haven't actually tested the push part yet) [11:20] * sgclark is very scared [11:26] * yofel originally wanted to rewrite that 500-line spaghetti code called ppa-build-status and was looking at ruby so he might be able to use the CI parsers. [11:26] as usual I ended up writing something else [11:31] hehe [11:36] yofel: still using the stuff in automation-ng? [11:37] santa_: no, I gave up on that for now after git-clone-all didn't do anything useful to me due to too much hardcoding of os.getcwd() [11:38] right now I'm trying to patch the old stuff into shape so we can get stuff done while we deprecate it [11:39] yofel: ok, so parallelizing git-clone-all sounds like an interesting thing to do? [11:40] possibly, at least with moszumanska being as slow as it is [11:40] k [11:40] requires git stdout/stderr redirection though to properly show errros [11:41] 'Morning all [11:41] hey [11:43] Hey yofel [11:48] yofel: doesn't look too bad [11:49] kate isn't launching after unstalling the packages from the frameworks/build_status ppa yesterday ..installed kwrite and it works ok [11:50] the build status that is still red? [11:50] dbus-launch opens kate , but that's a pita [11:51] well if it is kf5 in staging-frameworks, it is not done, [11:51] and I am about to hop on a plane so it won't be done soon unless someone else steps up to help [11:52] It's building actually, except nmqt [11:53] well yes of course, I have retry running. but have to pack up soon [11:54] sgclark, np, it works fine otherwise and yes it's the Staging Frameworks ppa [11:57] I'm feeling adventurous these days, maybe it's the winter boredom setting in and a I need a little excitement :-) [11:58] lol [12:20] sitter: "Build step 'Console output (build log) parsing' changed build result to UNSTABLE" gets done if there's any KCI-W detected by kci? [12:21] Riddell: yup [12:29] kdelibs4support - 5.17.0-0ubuntu1~ubuntu16.04~ppa1 i386 logfile Dependency wait: kdesignerplugin (>= 5.16.0~) [12:30] wait, why does that say >= 5.16, should be >= 5.17 [12:32] oh great, another script with now busted epoch handling [12:32] eh what? [12:32] looks like I did a bad job at looking for those :( [12:32] oh [12:35] oh wait, that has a different array o.O [12:36] O.o [12:36] static package list files overload -.- [12:37] I always did love those static lists [12:38] haha [12:38] anyway, can someone make a tester call when the last bits finish please, I am heading back home now. [12:39] sgclark: you uploaded the new network-manager somewhere? [12:39] ahh crud. nevermind. no. [12:40] ok, our epoch handling sucks [12:41] that needs redesign on paper before I touch it any more [12:42] yofel: couldn't you simply switch all version mangling with the python debian lib thingy's version class? [12:43] not if that lib still deletes comments from files, otherwise that's probably good idea [12:44] well. no more or less than wrap-and-sort already does, seeing as that also uses the lib IIRC [12:44] good point [12:50] OTOH, the actual control file editing isn't the problem really. The tracking of what package has what epoch is a PITA [12:50] yofel: you don't need to track that though [12:51] it already shoudl be there [12:51] so all you need to do is not touch it [12:51] Doing a git-clone-all before starting to work and extracting the information from there is more reasonable than using 2 JSON arrays and a dozen text files [12:51] if there is an epoch already you preserve it and if there isn't you won't add one [12:51] also yeah, using life data is of course the [12:51] yes, I think that's what my old code did. The current one does.. magic [12:51] best way :) [12:51] very static magic [13:33] does this mean that the wily backport will go straight to plasma5.5.1 (due to bugs in 5.5)? [13:33] https://www.kde.org/announcements/plasma-5.5.1.php [13:35] well yeah, considering we didn't even start yet we'll start with what's available when we do [13:37] good stuff - 5.4.3 is serving me well anyway... [13:38] I assume that 16.04 will have plasma 5.5.x ? [13:43] yes [13:45] sgclark, clivejo: As a warning: I just removed the git push logic from staging-upload and ubuntu-archive-upload, that's now an extra script called git-push-all so remember to run that [13:45] that a) finally allows you to mess up while generating the packages, b) shortens the time that CI has to be paused a lot [14:59] Plasma 5.6 meeting in #plasma in a minute! [15:01] * yofel watches [15:02] has to go out soon, but Ill read it later :) [15:05] * clivejo going to dry the puppy [15:15] * yofel has the punch line of the meeting now: "Terminology is important" [15:16] :-D === mck182 is now known as mck182|afk === mck182|afk is now known as mck182_ === mck182_ is now known as mck182