Friday, 2017-12-15

*** hashar has joined #openstack-jjb08:54
waynrhmmm i was thinking about this a few weeks ago and i've changed my mind about the parser09:41
waynrelectrofelix is right that the original point of the refactor was to allow an approach to convert from straight up job data structure straight to xml without necessitating yaml09:42
waynrbut that doesn't mean that we can't optionally pass a parser reference through to the xml expansionmodules for the primary use case of expanding on yaml parser generated data structures09:42
*** electrofelix has joined #openstack-jjb09:53
openstackgerritDarragh Bailey proposed openstack-infra/jenkins-job-builder master: No warn on old plugin conf format not in use  https://review.openstack.org/52828114:22
waynri kind of want to try writing a CI DSL that's not a mutated form of yaml14:27
waynrdefine the core features any CI system should have call everything else an extension and use JJB xml generator as the first backend implementation that connects the DSL to jenkins14:28
electrofelixwaynr: just catching up with responses early this morning that I missed via eavesdrop14:29
openstackgerritDarragh Bailey proposed openstack-infra/jenkins-job-builder master: Prevent deprecation warning using test with stdin  https://review.openstack.org/52828414:30
electrofelixwaynr: so your thinking that we pass something down that when it hits the dispatch call to go to the builders can perform a lazy expansion of any macros at that point in time, for the few modules that are using others?14:31
electrofelixas, in make it a hidden expansion triggered by the yaml -> xml code trying to access it?14:32
electrofelixDoes that mean needing to include all the template params to fill in for any params not explicitly set in macros?14:33
electrofelixI'm assuming we support that now for most normal templates, that in expanding out the macros early, they have access to other params from the project to use in the macro, not just those set directly?14:35
electrofelixI think in any case can expect that I can have something working by the end of next week, time during the last 2 weeks was spent relearning how the code works in that area and hooking up pycharms for some detailed debugging/inspection14:36
electrofelixwaynr zxiiro: were we supposed to have a quick meeting about 30 mins ago?14:38
zxiiroIt's in 20 minutes according to my calender14:38
*** openstackgerrit has quit IRC14:48
zxiirowell I'm at my computer now14:52
zxiiroI don't have any updates. Unfortunately migrating clouds took up all my time the past 2 weeks. I think the only topic this week is the YAML parser again.14:53
electrofelixzxiiro: listed as being 'Fridays at 1400 UTC' ;-)14:54
zxiiroah...14:55
zxiirooh daylight savings...14:55
zxiirothat's why14:55
zxiiroi entered it into my google calendar as EST i guess14:56
zxiirofixed in my calendar. I set it to iceland GMT+014:58
zxiiroWe just need waynr I guess.14:58
zxiirohe probably had the same problem as me14:58
electrofelixzxiiro: I used the ical to load it into google calendar, probably why it came up without an issue15:01
waynryeah my calendar just says 9am CST15:01
zxiiroalright we're all here15:01
zxiiro#startmeeting jjb15:02
openstackMeeting started Fri Dec 15 15:02:00 2017 UTC and is due to finish in 60 minutes.  The chair is zxiiro. Information about MeetBot at http://wiki.debian.org/MeetBot.15:02
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.15:02
*** openstack changes topic to " (Meeting topic: jjb)"15:02
openstackThe meeting name has been set to 'jjb'15:02
zxiiro#topic YAML Parser15:02
*** openstack changes topic to "YAML Parser (Meeting topic: jjb)"15:02
electrofelixI'm guessing I should attempt to get a POC of the delayed expansion based suggestion from Waynr, assuming that is a reasonable approach?15:09
electrofelixDo we lose any abilities that we thought were supported in going that way?15:09
electrofelixthe two changes above https://review.openstack.org/528281 & https://review.openstack.org/528284 both deal with some warning logs emitted unexpectedly, one from use of the trigger-parameterized code and the other from reading a yaml file in via stdin15:17
*** electrofelix has quit IRC15:19
waynrisn't delayed expansion the way we were doing it before?15:31
waynrmaybe it's been too long since i've been familiar with the code15:31
zxiiroI guess he left15:49
zxiiro#endmeeting15:50
*** openstack changes topic to " (Meeting topic: testing?)"15:50
openstackMeeting ended Fri Dec 15 15:50:22 2017 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)15:50
openstackMinutes:        http://eavesdrop01.openstack.org/meetings/jjb/2017/jjb.2017-12-15-15.02.html15:50
openstackMinutes (text): http://eavesdrop01.openstack.org/meetings/jjb/2017/jjb.2017-12-15-15.02.txt15:50
openstackLog:            http://eavesdrop01.openstack.org/meetings/jjb/2017/jjb.2017-12-15-15.02.log.html15:50
zxiirowaynr: I'll pass on what you said next time I see him15:50
waynrcool15:53
*** hashar has quit IRC16:35
*** smyers has joined #openstack-jjb18:57
*** smyers has quit IRC19:38
*** hashar has joined #openstack-jjb19:53
*** smyers has joined #openstack-jjb21:06
*** smyers has quit IRC21:09
*** smyers has joined #openstack-jjb21:12
*** smyers has quit IRC21:29
*** smyers has joined #openstack-jjb21:33
*** caphrim007 has joined #openstack-jjb21:57
caphrim007anyone know what version of jenkins-job-builder has the !include-jinja2 syntax?21:57
caphrim007in 2.0.0.0b2 it raises an error about "could not determine a constructor for the tag !include-jinja2"21:57
zxiirocaphrim007: no versions released have it21:57
zxiirocaphrim007: we have a jjb 2.0 b locker preventing us from releasing 2.021:57
caphrim007ahh ok21:58
caphrim007i'll just grab it directly from scm in the meantime21:58
*** hashar has quit IRC22:43

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!