16:32:40 <djmitche> #startmeeting weekly
16:32:40 <bb-supy> Meeting started Tue Mar  8 16:32:40 2016 UTC and is due to finish in 60 minutes.  The chair is djmitche. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:32:40 <bb-supy> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:32:40 <bb-supy> The meeting name has been set to 'weekly'
16:32:45 <djmitche> sorry, was delayed fixing a broken puppet server :/
16:32:52 <djmitche> #topic Introduction
16:32:56 <djmitche> https://titanpad.com/buildbot-agenda
16:32:58 <infobob> http://paste.pound-python.org/show/zmTpujQoBBFepYOIb6nZ/ (repasted for djmitche)
16:33:27 <djmitche> #topic Bug 2340
16:33:41 <rutsky> First part of buildbot-worker package transition ready and need review: https://github.com/buildbot/buildbot/pull/2033
16:33:49 <rutsky> djmitche and tardyp already reviewed
16:34:17 <djmitche> is there anyone else you're interested in hearing from?
16:34:22 <rutsky> at this moment there is no comments that requires some fixing
16:34:23 <djmitche> it's not a wide field :)
16:34:29 <rutsky> djmitche: not really
16:34:35 <djmitche> ok, ship it then! :)
16:35:10 <rutsky> done!
16:35:11 <tardyp> +1
16:35:16 <bb-github> [13buildbot] 15rutsky closed pull request #2033: [bug 2340] first part of "buildslave" to "buildbot-worker" transition (06bug2340...06bug2340-worker-copy) 02https://git.io/v2A8Q
16:35:30 <rutsky> I'll create similar branches for next stages of buildbot-worker transition
16:35:34 <djmitche> #info first step for buildbot-worker package shipped - copy package to a new name
16:35:37 <djmitche> ok
16:35:50 <rutsky> but we will not yet merge it master, yes?
16:35:56 <djmitche> #info next step - rename classes/interfaces/etc with tools for refactoring
16:36:02 <rutsky> currently I merged PR into bug2340 branch
16:36:02 <tardyp> I think we should merge it to master
16:36:17 <djmitche> I tend to think so too - why not?
16:36:26 <rutsky> ok, lets merge!
16:36:31 <tardyp> and there are some conflicts to resolve
16:36:37 <rutsky> I'll take a look later about current conflicts
16:36:55 <tardyp> please make a new PR to master, with the current branch, and we'll merge it after travis check
16:36:59 <tardyp> w/o review
16:37:19 <rutsky> ok, travis checks is good idea
16:37:28 <rutsky> djmitche: action for me then?
16:38:11 <djmitche> #action rutsky to merge bug2340 to master, and continue merging buidlbot-worker PRs to master
16:38:16 <djmitche> cool!
16:38:22 <djmitche> any other discussion on this project?
16:38:43 <rutsky> I think no
16:39:11 <djmitche> ok
16:39:22 <djmitche> #topic development week in review
16:39:56 <djmitche> #info tardyp did a good bit of pull request grooming - thanks!
16:40:28 <djmitche> looks like some PR's around docs -- anything to highlight there?
16:41:33 <tardyp> we need some love for eight branch
16:42:12 <tardyp> there are a bunch of PRs for eight that are mergeable
16:42:19 <djmitche> ok
16:42:27 <djmitche> sa2ajj: ^^ that's still on you -- what do you think?
16:42:32 <tardyp> do we just refuse them, because we won't do anymore eight release?
16:42:43 <djmitche> #info a number of eight-targeted PRs are mergeable
16:42:46 <tardyp> or do we merge them, but we will need to handle another eight release
16:42:57 <djmitche> it's up to misha really
16:43:05 <tardyp> agreed
16:43:18 * djmitche sends devel@ email
16:43:56 <rutsky_> hi again
16:44:27 * djmitche waves
16:44:46 <djmitche> http://trac.buildbot.net/ticket/3479 worries me
16:45:08 <djmitche> but I don't think it blocks 0.9.0
16:45:10 <djmitche> agree?
16:45:59 <bb-trac_> [trac] #3478/enhancement (new) updated by dustin (empty comment) http://trac.buildbot.net/ticket/3478
16:46:26 <djmitche> http://trac.buildbot.net/ticket/3474 - rutsky_ this was one of yours, regarding PR templates
16:46:40 <rutsky_> yes
16:46:50 <rutsky_> what do you think about PR templates?
16:46:53 <djmitche> I don't think this is especially useful for us -- I feel like the PRs we get from contributors are already in pretty good shape (or not, because they aren't paying attention)
16:47:01 <bb-trac_> [trac] #3479/defect (new) updated by dustin (empty comment) http://trac.buildbot.net/ticket/3479
16:47:40 <rutsky_> I agree, posted to get feedback from others --- maybe someone have ideas how they might be useful for us
16:47:57 <bb-trac_> [trac] #3475/defect (new) updated by dustin (empty comment) http://trac.buildbot.net/ticket/3475
16:48:22 <bb-trac_> [trac] #3476/defect (new) updated by dustin (empty comment) http://trac.buildbot.net/ticket/3476
16:48:27 <djmitche> ok
16:48:30 <djmitche> I'll comment
16:48:33 <tardyp> we may add some checkboxes like: [x] I ran validate.sh  / [x] I did a make hooks to install the git hooks
16:48:57 <rutsky_> checkboxes from list that we already have on trac
16:49:07 <tardyp> [x] I added unit tests  [x] I added relnote
16:49:17 <rutsky_> from here: http://trac.buildbot.net/wiki/PatchReview
16:49:20 <djmitche> ok
16:49:35 <djmitche> sounds good -- more explicit, for sure
16:50:03 <bb-trac_> [trac] #3477/defect (new) updated by dustin (empty comment) http://trac.buildbot.net/ticket/3477
16:50:09 <djmitche> want to comment on the bug?
16:50:25 <rutsky_> I'll do
16:51:22 <djmitche> ok
16:51:28 <djmitche> anything else to bring up today?
16:52:18 <rutsky_> http://trac.buildbot.net/ticket/3477
16:52:23 <rutsky_> tardyp: can you comment on this one?
16:52:37 <rutsky_> I think it's you, who changes this implementation
16:53:07 <rutsky_> (which doesn't mean that you need to resolve this issue, but probably you know better than other what should go into docs here)
16:53:08 <djmitche> #agree will set up a pull-request template (bug 3474) to encourage contributors to check the boxes on the contribution guide
16:54:02 <bb-trac_> [trac] #3474/undecided (new) updated by rutsky (From discussion in IRC: we can add template with checkboxes of items from ...) http://trac.buildbot.net/ticket/3474
16:54:20 <tardyp> rutsky_: I agree this part is not documented a lot
16:55:05 <tardyp> there is a bit of reverse engineering to find out the list of layout
16:55:17 <tardyp> and to look at baseParameter.type
16:55:39 <tardyp> the eight version was not very much documented as well, so I may have missed this last parameter
16:56:21 <djmitche> is that an easy thing? should we move it to 0.9.0?
16:58:10 <tardyp> yes, its just a matter of documentation
16:58:28 <djmitche> ok
16:58:44 <bb-trac_> [trac] #3477/defect (assigned) updated by dustin (empty comment) http://trac.buildbot.net/ticket/3477
16:58:53 <rutsky_> I think layout types and field types are defined here: https://github.com/buildbot/buildbot/blob/master/www/md_base/src/app/common/directives/forcebuildform/forcebuildform.directive.coffee#L40
16:59:29 <tardyp> indeed
16:59:34 <tardyp> no
16:59:40 <tardyp> you are looking at md_base
16:59:42 <rutsky_> do we want to somehow sync docs with code there?
16:59:46 <rutsky_> oh, crap :)
17:00:19 <djmitche> backup to 18 0.9.0 bugs now..  :)
17:00:19 <tardyp> md_base is much cleaner code on that part
17:00:34 <rutsky_> anyway, supported types should be same in md_base and base
17:00:55 <rutsky_> can we somehow make more explicit that list of supported types/layouts?
17:01:52 <tardyp> yes
17:02:14 <rutsky_> I'm worried that when someone will add new type he will forgot to implement is support in md_base or base
17:02:29 <tardyp> its actually implicit in base code
17:02:39 <tardyp> given the list of directives available
17:02:50 <tardyp> this allows  plugins to define new layouts or new types
17:03:08 <tardyp> this is why the list is not hardcoded with a nice switch
17:03:31 <tardyp> www/base/src/app/common/directives/forcefields
17:03:40 <rutsky_> and UI just must be able properly to handle unknown to it type? Like output warning and don't display that property?
17:04:19 <tardyp> www/base/src/app/common/directives/forcefields/forcefields.directive.coffee
17:04:43 <tardyp> https://www.irccloud.com/pastebin/optGFu0S/
17:04:46 <rutsky_> having open set of types sound good to me. But we should make sure that default UI will gracefully handle new types
17:04:50 <djmitche> #info ..discussion of matching force-scheduler field types to UI implementations..
17:07:28 <rutsky_> tardyp: what will "base" UI do if master will return field type that is unknown to UI?
17:07:53 <djmitche> I need to focus on other stuff folks -- I'll make rutsky_ chair, and you can close the meeting when you're done, ok?
17:07:57 <djmitche> #chair rutsky_
17:07:57 <bb-supy> Current chairs: djmitche rutsky_
17:08:33 <rutsky_> oh, too much responsibility :)
17:08:44 <rutsky_> do we have other topics to discuss today?
17:09:10 <tardyp> that's fine for me
17:10:49 <rutsky_> tardyp: what will "base" UI do if master will return field type that is unknown to UI?
17:11:41 <rutsky_> I see that base UI define Angular directives for each known type
17:12:10 <rutsky_> so if unknown type will arise, probably we will get some stragely looking half-raw HTML?
17:12:22 <tardyp> it will crash badely :-(
17:12:31 <rutsky_> oh...
17:12:33 <tardyp> actually not.
17:12:46 <tardyp> it will create a <mytype></mytype> html code
17:12:55 <tardyp> which angular wont recognise as a directive
17:13:04 <tardyp> so you will get an empty field
17:13:35 <tardyp> https://www.irccloud.com/pastebin/NloKeGYK/
17:13:40 <rutsky_> sometimes no error is worse than badly looking error...
17:13:41 <tardyp> this is the code responsible of this
17:14:00 <rutsky_> I see
17:14:18 <tardyp> we can actually verify in that code if there is a directive define
17:14:55 <rutsky_> we can store somewhere global list of supported types?
17:15:03 <rutsky_> and check is type in it there?
17:15:13 <tardyp> no, angular has the list of directive available
17:15:34 <tardyp> so we can check if it is available instead of just running $compile
17:15:40 <tardyp> and throw a nice error message
17:15:47 <rutsky_> sounds good!
17:16:13 <rutsky_> will you create ticket/implement this?
17:17:42 <tardyp> we can do in the same ticket
17:17:57 <tardyp> I'm googling this, and I am not sure yet how to get the list of defined directives
17:19:09 <rutsky_> #action tardyp to handle unknown field types in base UI graceful manner
17:19:45 <tardyp> I know... we can just make sure that the result of $compile did actually something, and not leave the same html code
17:20:44 <tardyp> did you encounter unsupported type?
17:20:49 <rutsky_> it's an option
17:20:53 <rutsky_> no
17:21:07 <rutsky_> so this is a low priority actually
17:22:06 <rutsky_> do we have anything else to discuss?
17:25:25 <tardyp> no
17:25:37 <rutsky_> #endmeeting