=== petevg_ is now known as petevg
=== dooferlad_ is now known as dooferlad
=== frankban_ is now known as frankban
=== rogpeppe is now known as rogpeppe1
=== rogpeppe1 is now known as rogpeppe
[12:09] <mhilton> anyone else having problems with canonical irc?
[12:10] <frankban> uiteam are you able to connect to canonical irc?
[12:11] <rogpeppe> frankban, mhilton: no, it's down
[12:11] <mhilton> frankban: I can't
[12:11] <fabrice> down
[12:11] <frankban> ok
[12:11] <rogpeppe> frankban, mhilton: see discussion on #juju-dev
[12:15] <rogpeppe> uiteam: sorry, i don't think I made it clear - I'll need this to be approved and landed before the testing changes: https://github.com/juju/utils/pull/242
[14:21] <hatch_> kadams54: you could take another look at that revision branch you were working on and switch it from using the setup.py to not
[14:22] <rick_h_> and this channel comes alive :P
[14:23] <kadams54> hatch_: Yeah, that's what I'm doing at the moment. Just didn't know if there was anything that should take precedent.
[14:23] <kadams54> rick_h_: I know! I saw the notification for hatch's message and then couldn't find it ;-)
[14:24] <hatch_> kadams54: here is how you can get the version in the Makefile
[14:24] <hatch_> $(eval GUI_VERSION=$(shell sed -n -e '/current_version =/ s/.*\= *//p' .bumpversion.cfg))
[14:24] <hatch_> kadams54: I'd also like it to grab the current git sha and include it in the file
=== hatch_ is now known as hatch
[14:24] <hatch> probably in a json format
[14:24] <hatch> or js
=== hatch is now known as Guest72453
[14:25] <Guest72453> { "version": "2.1.13", "sha": "asdfasdfasdf" }
[14:25] <Guest72453> bah
[14:30] <kadams54> hatch: Yeah, so... when we talked last, we wanted to use the same version that goes on the tarball. That version is pulled from setup.py (`python setup.py sdist --formats=bztar` in the fast-dist target), which is why I took the approach I did. .bumpversion.cfg updates setup.py when you run `make bumpversion`, so I'd like to keep both the dist and version
[14:30] <kadams54> targets using the same approach.
=== BradCrittenden is now known as bac
=== hatch_ is now known as hatch
[14:39] <hatch> there we go
[14:39] <hatch> kadams54: sorry did you see my messages there? I was having issues :)
[14:40] <kadams54> ha ha, no, I did not.
[14:40] <hatch> last one was about the version in json?
[14:40] <hatch> or in js
[14:48] <kadams54> sha in json
[14:49] <kadams54> hatch: The message you sent as Guest72453
[14:50] <kadams54> ;-)
[14:50] <frankban> uiteam call in 10
[15:00] <bac> kadams54, antdillon: standup
[15:14] <frankban> hatch: I am now available for any GUI tasks that you need for 2.2. I am near my EOD but please fire me an email before your EOD with anything I can help doing tomorrow
[15:15] <hatch> frankban: sure thing - honestly the big thing right now I think is that subordinate issue
[15:15] <hatch> frankban: the DF was working quite well
[15:15] <hatch> so I haven't had enough time to really try and break it
[15:15] <hatch> :D
[15:15] <hatch> kadams54: 
[15:15] <hatch> kadams54: did you need any help?
[15:15] <hatch> on your charm?
[15:17] <hatch> frankban: so if you remember that subordinat issue lazypower had actually brought it up again
[15:17] <hatch> and then ther ewas also the juju-info issue with subordinates
[15:17] <hatch> not being able to create them
[15:17] <hatch> frankban: does this sound like something you'd be interested in?
[15:18] <kadams54> hatch: on my charm? You mean on the version stuff?
[15:18] <hatch> yeah the version stuff
[15:18] <hatch> did that sed stuff work for you?
[15:19] <hatch> kadams54: I was working on it for when I was generating a custom tarball not using setup.py
[15:20] <kadams54> hatch: probably if I'd put it in place yet :-) working on the route issue first.
[15:20] <frankban> hatch: sure, is this https://github.com/juju/juju-gui/issues/1483 ?
[15:20] <hatch> frankban: yeah that's the juju-info one thanks
[15:20] <kadams54> hatch: still not sold on that approach. If we change that I feel like we ought to change the fast-dist target
[15:20] <frankban> hatch: ok I'll take a look at that one tomorrow
[15:21] <hatch> kadams54: yeah totally it's all going to change - but it's going to be a slack task
[15:21] <hatch> thanks frankban 
[15:21] <hatch> I'm sure lazypower would be very happy :D
[15:21] <hatch> frankban: plus there are some bundles that can't be made in the GUI, and this will fix that
[15:24] <hatch> bac: let me know when you have a moment and I can fill you in on the charm release stuff
[15:24] <hatch> for whatever reason it wasn't working for me (even after jrwren got me settled on the proper charm version)
[15:25] <hatch> jrwren: you had mentioned something about azure and python from js? But I couldn't find a js version of those tools
[15:26] <hatch> or maybe that was jcsackett ?
[15:30] <hatch> frankban: maybe while you're in there you could take a look at https://github.com/juju/juju-gui/issues/1975
[15:30] <hatch> assuming it is indeed in that area
[15:32] <frankban> hatch: sure
[15:34] <hatch> frankban: we're down under 90 open issues now :)
[15:34] <hatch> yay
[15:39] <hatch> Makyo: have you experienced this in a while https://github.com/juju/juju-gui/issues/1767
[15:39] <hatch> since you're now back on the gui stuff?
[15:41] <Makyo> hatch: uh, hmm.  I think so, but I'll keep an eye on it.
[15:41] <hatch> Makyo: sure thing, it might have been another one of those vagrant only issues
[15:41] <hatch> since i haven't actually ever experienced it :)
[15:49] <kadams54> Taking a quick lunch break.
=== kadams54 is now known as kadams54-lunch
[15:50] <hatch> mm
[15:51] <hatch> oh I can qa these using local charms :)
[15:51] <hatch> great
[15:55] <hatch> frankban: https://jujucharms.com/docs/stable/authors-subordinate-services#declaring-subordinate-charms here is some information about the 'container vs scope' it sort of helps haha
[15:56] <frankban> :-) sort of
[16:00] <jrwren> hatch: https://www.npmjs.com/package/azure has lots of cli tools and used to be the official cli tools for azure.
[16:01] <hatch> jrwren: oh I always thought this was a library for use with node tools
[16:01] <hatch> not a cli tool
[16:02] <jrwren> hatch: https://azure.microsoft.com/en-us/documentation/articles/xplat-cli-install/
[16:02] <hatch> interesting
[16:02] <jrwren> hatch: nice how juju hides all that cmdline stuff for you, eh?
[16:03] <hatch> haha oh yes, yes Juju is pretty awesome there
[16:04] <hatch> jrwren: so they aren't going to support this any longer and switch exclusively to the Python implementation?
[16:04] <jrwren> hatch: i don't know. I expect they will tell us, eventually. Maybe they will support both?
[16:06] <jrwren> hatch: the github calls the python one "2.0 - preview" so maybe they are abandoning the node cli tools and moving it all to the new python ones for 2.0. That was my assumption.
[16:06] <hatch> that's a good guess
[16:06] <hatch> I could see the advantage
[16:06] <hatch> everyone has python ;)
[16:08] <jrwren> hatch: that isn't true. no one on windows has python OOTB. so its just as much a support nightmare for their 99% use case.
[16:08] <hatch> ohh, that's a good point!
=== kadams54-lunch is now known as kadams54
=== frankban is now known as frankban|afk