mozilla :: #conduit

20 Apr 2017
06:08zalunDo we have a document with links to our knowledge base?
08:48zalunis our README in conduitdemo still valid?
09:23zalunOK port is wrong - it should just be removed
12:51imaduemedavidwalsh: i set you as the reviewer for (bug 1357852) and added some comments in the diff for explanation.
12:51imaduemeAlso I tried, but if anyone knows how to mock autolanding end to end in rb development that would be great to know! I could only get as far as bypassing the permissions.
12:51firebot NEW, Regression: Remember try syntax from last run on review
12:52imadueme(I also meant to say feel free to pass the review to someone else if you're busy)
13:46imaduemeDo the rb tests normally take 30+ minutes to run? holy cow
13:58dkl|afkDr appt running long this morning so will be late for planning meeting.
13:58dkl|afksmacleod: I added some more notes to the etherpad about phab privacy.
14:21smacleoddkl|afk: k, thanks
14:25dkl|afksmacleod: you may have better ideas
14:25smacleoddkl|afk: writing up coments right now
14:33globdkl|afk: meeting time
14:33dkl|afksmacleod: sorry for confusion before. I mistakenly thought we would leave the revision in the private space. Please add a trello card about implementing the bot and assign me
14:39janxsmacleod or mcote: Can I help bring back the "mozreview.commits.published" pulse message? Is there a bug filed about what's blocking it? If not, where can I file one?
14:41firebotBug 1332697 NEW, Pulse publishing timeout
15:32* janx heads off to another meeting, but hopes to resolve this meeting conflict one day, because sprint planning seems even more interesting for me
15:52dklis the meeting still going?
15:52smacleoddkl: yup, join up
15:54imaduemesmacleod: do you think it is worth editing bugzilla itself to call out to the service rather than a cron? Considering this is so important
15:54smacleodimadueme: doing that sort of stuff is built-in, bugzilla has daemon's for things like this. But our initial MVP doesn't need to go through that hassle
15:57dklsmacleod: there is a project.edit.{add,remove}.members conduit API call
15:58smacleoddkl: awesome :)
16:00imaduemejust a heads up, I'm voiping a lot more than usual, sry if i miss something
16:02imaduemethat's about right :)
16:02imaduemewoah smacleod: that pic just brought my power back :D
16:08mcoteheh that would make a great user pic
16:25imaduemeHave we thought about storing the revision id in bugzilla (i.e. the other way around)? Then it seems like we just need a smaller extension to link to the bug. That makes bugzilla the source truth
16:26imaduemezalun, smacleod
16:26mcoteimadueme: yes, we're going to do that too, but we still need a way to get from revision -> bug
16:26mcotewithout search bugzilla for the revision id
16:27smacleodI did consider another service to do this
16:27smacleodthe easiest way to display this though in phabricator, is zalun's plan
16:27imaduemegot it
16:29smacleodFor anyone who was wondering about the extensions folder, here is my docker-compose mapping:
16:29smacleod- ./extensions:/srv/phabricator/phabricator/src/extensions
16:29smacleodzalun / mars ^
16:48mcoteimadueme: do you have a new mic/headset? it's really clear :)
16:48dklimadueme: yeah makes you sound older, more mature :)
16:48mcoteradio voice :)
16:49janxsmacleod: do you have an idea how easy it would be to publish static-analysis feedback as actual reviews? (using ReviewBoard or Phabricator API), or do you think it's more reasonable to publish a simple comment, e.g. on the bug?
16:49smacleodjanx: I have a library for Review Board in python that makes it pretty easy
16:49imaduememcote, dkl: yup $100 :). Did it because during onboarding they really recommened getting a good setup :P
16:50smacleodnot sure about Phabricator yet tbh
16:50mcoteimadueme: wow you look like a radio host now too :)
16:50smacleodjanx: probably best to start with simple comment and then make it better later?
16:51janxsmacleod: the python library sounds useful, could you please share the link? (we thought about starting with a simple bz comment, but maybe it's even easier to just read stdout on the taskcluster task, and directly work on the next step which is posting an actual review, if it doesn't require an unreasonable amount of work)
17:02* janx recently discovered that Phabricator's API is called Conduit
17:04imaduememcote: btw i ran the tests (which take forever) for that bug you passed on to me. I did get errors but they do not seem related to the change, does mozreview currently have tests that are supposed to be failing
17:06* imadueme moves to #mozreview
18:02smacleodckolos: flow for login:
18:54dklckolos: dklawren
18:54marsckolos: mars-f
19:33davidwalshdkl: Is there a scenario where an Auth Delection Request "Accept" -> POST would provide 0 data in the post? Because that's what I'm getting
19:34davidwalshI thought maybe Phabricator was cleansing $_POST but when I use cURL the $_POST data gets there
19:34davidwalshSo I can only believe that BMO isn't sending any POST data when I hit that "Accept" button
19:43dkldavidwalsh: do not recall. dylan might know
19:54imaduememars whenever you have time to discuss those autoland cards I should be ready. In the meantime I'm running tests (that take forever) for mark's bug and I was thinking about making a video about git commands
20:22davidwalshWho's ready to throw up?
20:23imaduemeanother one?
20:26* imadueme
20:27* imadueme also not sure why I shouldn't like that line
21 Apr 2017
No messages
Last message: 9 days and 11 hours ago