[00:47] <rick_h_> hatch: sorry for the late, did qa and noted issues. 
[00:47] <hatch> rick_h_ np, looking at comment
[00:50] <hatch> replied
[00:51] <hatch> ugh I hate trying to balance such a complicated switch between the feature flags
[00:51] <hatch> so many wasted hours on this one
[00:52] <hatch> oh now I see why
[00:52] <rick_h_> yea, I'll bet these are small ones
[00:54] <rick_h_> ok, time to run away for the night. 
[00:54] <rick_h_> have a good night all
[00:54] <hatch> it's using the new code for the curated but the old code for the search so it doesn't notice the change
[00:54] <hatch> I may just have to hack this to work 
[00:54] <hatch> cyaz
[01:01] <hatch> nice I found a non hacky solution
[01:02] <hatch> yay finally landing this
[01:11] <hatch> huwshimi I'm going to head out - any q's before I do?
[01:12] <huwshimi> hatch: All good. I'll leave any questions on the pr if need be.
[01:12] <hatch> sounds like a plan, have a good one
[07:48] <rogpeppe> mornin' all
[07:55] <huwshimi> rogpeppe: Morning
[07:55] <rogpeppe> huwshimi: hiya
[09:37] <frankban> hi fwereade, thanks for the review! I am available for a live chat when you want
[09:38] <frankban> morning rogpeppe, how is it going?
[09:38] <rogpeppe> frankban: hiya
[09:38] <rogpeppe> frankban: not bad, thanks
[11:09] <rick_h_> morning everyone
[11:16] <rick_h_> rogpeppe: morning, did you get the invite to the interview today? Are you able to participate?
[11:17] <rogpeppe> rick_h_: hi rick
[11:17] <rogpeppe> rick_h_: yes, i saw the invite on my phone and forgot to respond this morning. i can come.
[11:18] <rick_h_> rogpeppe: awesome, appreciate it.
[11:33] <frankban> rogpeppe: FYI after lunch I'll start working on creating a BaseSuite in github's testing. The next step is to make utils/* stuff use the new generic suite.
[11:38] <rogpeppe> frankban: that might be ok. i'm not entirely sure about BaseSuite though as a name - it doesn't really imply any particular semantics and it kind of implies that it's the centre of the universe :-) what does it mean when we've got two BaseSuites, one in juju-core and one in github/juju/testing?
[11:43] <frankban> rogpeppe: from a pre-emp with William the idea is 1) BaseSuite in github includes logging, osenv reset/restore (all vars) and cleanup, and it's not juju-core related: this is a generic base for having base isolation for all the tests. and 2) BaseSuite in core will be eventually replaced by 1, after potentially fixing the juju-core tests which depends on the env vars not currently removed
[12:17] <rogpeppe> frankban: perhaps IsolationSuite might be a better name
[12:31] <bac> rick_h_: did we get advance copy of homework from candidate?
[12:32] <rick_h_> bac: yes, I thought he copied everyone /me looks
[12:33] <rick_h_> bac: sorry, that was the afternoon interview. I'm working on it. 
[12:41] <frankban> rogpeppe: both BaseSuite and IsolationSuite sgtm, fwereade? ^^^
[12:41] <luca> rick_h_: is there another charm that someone can use instead of NodeJS?
[12:42] <luca> rick_h_: could someone use Ruby on Rails for doing similar stuff?
[12:42] <luca> rick_h_: broadly speaking of course :D
[12:42] <rick_h_> luca: yea, there's a rails charm, a django charm
[12:43] <luca> rick_h_: brilliant, thanks
[12:52] <fwereade> rogpeppe, frankban: I'm easy, IsolationSuite sounds fine to me
[12:52] <frankban> fwereade: cool thanks
[12:53] <fwereade> frankban, rogpeppe: except, ha
[12:53] <fwereade> frankban, rogpeppe: no forget it, even the Cleanup stuff is all about isolation really
[12:55] <rick_h_> jcsackett: ping
[12:56]  * rogpeppe wishes that the setup and teardown stuff wasn't so reflection-driven. i'd like to be able to do gc.RegisterSuites(c, &mySuite.LoggingSuite, &mySuite.CleanupSuite)
[12:56] <rogpeppe> or something like that
[12:56] <rogpeppe> without needing to do the painful SetUpSuite/TearDownSuite calling of underlying suites
[12:57] <rogpeppe> 'cos that pain is the only real reason we combine suites into super-suites
[12:57] <rogpeppe> IMO
[12:58] <rick_h_> frankban: rogpeppe reminder call in 3. 
[12:58] <rogpeppe> rick_h_: yup
[12:58] <rogpeppe> rick_h_: ta
[12:59] <fwereade> rogpeppe, I feel your pain, but I don't think it's quite the only reason -- it's quite useful to have them all batched up together *anyway*, imagine the hassle of adding network isolation to every single suite -- even without manual setup/teardown, it'd be way too easy to have them drift out of sync
[13:00] <rogpeppe> fwereade: i definitely think that isolation is useful to have in its own suite
[13:00] <rogpeppe> fwereade: but logging is a bit of an outlier there, as is cleanup, really.
[13:03] <jcsackett> Morning, all. 
[13:07] <frankban> rick_h_: I do not have access to the google doc
[13:07] <frankban> (Presentation of homework)
[13:07] <rick_h_> frankban: his doc? yea, I'll ask him to open it up. I think it's just his bullet notes to go through
[13:08] <frankban> rick_h_: OIC, ok  so nevermond
[13:08] <frankban> nevermind even
[13:08] <rick_h_> I think he's doing this more a 'conference presentation' model or lightning talk
[13:08] <rick_h_> it's not really specified, but first time we've had this route 
[13:57] <rick_h_> rogpeppe: stay on the call please
[14:12] <hatch> afternoon luca
[14:22] <hatch> rick_h_ when you get a chance I'd like to chat about moving the search over to the consolidated view - it's going to be a considerable amount of work to keep both the unflagged and flagged functionality
[14:23] <rick_h_> hatch: rgr, on a call atm but will ping when I'm free
[14:23] <rick_h_> hatch: I'm basically going to ask you about the other remaining cards up there for IL
[14:23] <rick_h_> hatch: so maybe check how many of those apply and they need to happen first?
[14:24] <hatch> yeah - I'm doing a qa on it right now
[14:39] <hatch> rick_h_ i've created cards for all the bugs ive found and placed them in the order I think they should be completed behind the green il card
[14:43] <rick_h_> hatch: cool thanks
[14:43] <rick_h_> hatch: we can chat after the standup
[14:43] <hatch> sounds good
[14:48] <hatch> rick_h_ for il can we go back to having the inspector render by default on ghost?
[14:49] <rick_h_> hatch: sounds good to me
[14:50] <Makyo> jujugui call in 10
[14:59] <Makyo> jujugui call in 1
[15:00] <rick_h_> jujugui I'm going to be a couple of min late, hatch please run with it
[15:00] <hatch> on it
[15:01] <hatch> jujugui call now
[15:01] <kadams54> Fighting with Google, will be in in a moment
[15:11] <frankban> rogpeppe: quick call?
[15:11] <rogpeppe> frankban: sure
[15:11] <rogpeppe> frankban: wanna start a hangout?
[15:12] <frankban> rogpeppe: https://plus.google.com/hangouts/_/gz42gpplaq7jndtftpitkl3oeaa?authuser=0&hl=en
[15:16] <rick_h_> Makyo: after your card can you coordinate with hatch on helping move towards getting rid of the il flag please? He's got notes and such to help coordinate which can be done in parallel vs serial
[15:16] <rick_h_> jcsackett: standup hangout?
[15:17] <hatch> Makyo here is the diff for the card I'm working on right now https://gist.github.com/8bc67fa011aeecd7aaa3 I just need to write tests for it then it'll be up for review 
[15:17] <Makyo> rick_h_, sure thing.  Thanks hatch 
[15:23] <rick_h_> jcsackett: ok leaving that hangout. Shoot me a message when you're back
[15:25] <luca> hatch: heya, I was in an interview
[15:25] <hatch> did you get the job?
[15:25] <hatch> kehehe
[15:25] <luca> hatch: rofl
[15:26] <luca> hatch: how goes it?
[15:26] <hatch> it's going, it's going, summer finally so getting some time outside :)
[15:26] <hatch> well, time outside without 3" of insulation 
[15:26] <luca> hehe nice
[15:45] <luca> hatch__: do you need anything from me? :)
[15:45] <hatch__> luca I don't think so
[15:45] <hatch__> have anything I want?
[15:45] <luca> hatch: not at the moment :)
[15:46] <luca> hatch: looking forward to QA’ing MV and IotL
[15:46] <hatch> are you sure? Maybe we disregarded everything you said :P
[15:46] <luca> hatch: ha, that could be an improvement
[15:46] <hatch> lol
[15:47] <hatch> nah it's good :)
[15:47] <luca> :D
[15:50] <hatch> luca I've been working on getting il ready to go and it looks awesome on the left - I've been wondering if we maybe shouldn't also have the charmbrowser black to match too
[15:51] <luca> hatch: hehe yeah
[15:53] <hatch> huw is going to kill me for suggesting that he redo all the css :P
[15:53] <rick_h_> hah, I think it's nice to have some clear seperation of things, and that has nothing to do with the fact that I'm not a fan of the black :P
[15:54] <hatch> haha - I was also thinking of that
[15:55] <hatch> I however am a fan of black
[15:55] <hatch> :P
[15:55] <rick_h_> but it's so mushy, the scroll boundries are so unclear and soft.
[15:56] <hatch> with the black?
[15:56] <rick_h_> hatch: yes
[15:56] <hatch> whaaat, it's a very high contrast to the grey compared to grey on grey, how is that MORE squishy? 
[15:57] <hatch> me thinks you've gone cra'cra'
[15:57]  * hatch is learning the kids lingo
[15:59] <rick_h_> hatch: but the scrollbar is darker and stands out. The boundry from charm to charm is done with lines. and studies show reading black on white > white on black for eye fatigue and such
[16:01] <hatch> well the dividers would also need to be adjusted accordingly. I thought that the studies showed the opposite? I much prefer reading white on black for coding
[16:01] <hatch> jujugui lf a review/qa for a small diff https://github.com/juju/juju-gui/pull/345
[16:07] <Makyo> On it.
[16:08] <hatch> thanks
[16:20] <hatch> remember that storm I was posting about yesterday.....yeah it rained a bit https://twitter.com/J_Dubs83/status/471083954754572288
[16:29] <rick_h_> heh, that's a few oz of water
[16:29]  * rick_h_ runs away for lunch 
[16:30] <hatch> :-)
[16:42]  * Makyo runs to grab prescription.
[17:02] <hatch> oh look we use YUI's DataSource, how did this get in here
[17:03] <rick_h_> hatch: charmworld api stuff in store
[17:03] <hatch> yeah :-) I've never actually found a use for it before
[17:04] <hatch> I'm not sure we actually use it for anything beyond an IO call though right?
[17:05] <hatch> this might actually make my current task much easier
[17:10] <hatch> I was wrong *sadface*
[17:15] <rick_h_> hazmat: ping, trying to run the deployer on a mac for this demo purpose and getting a bzrlib. 
[17:15] <rick_h_> hazmat: does this run on a mac? Is there a bzrlib way to do that off the top of your head? 
[17:15] <hazmat> rick_h_, pip install bzrlib
[17:16] <hazmat> er.. pip install bzr
[17:16]  * rick_h_ doesn't see it in pypi
[17:16] <hazmat> rick_h_, traceback
[17:16] <rick_h_> ah, bzr
[17:16]  * rick_h_ smacks head for searching for wrong lib
[17:16] <hazmat> rick_h_, alternatively we could fix bzr cli to support what's needed  ;-)
[17:18] <hazmat> might be worth checking if its supported.. i'm looking at the comment and its a little unclear.. its about detecting working copy changes when the branch/checkout is pinned at a rev
[17:20] <rick_h_> hazmat: cool yea working on trying to get it to work. we'll see
[17:20] <hazmat> rick_h_, so bzr status on a branch that's not on head rev goes.. 'working tree is out of date, run 'bzr update'
[17:21] <hazmat> rick_h_, but yeah osx is supported and has worked  b4
[17:21] <rick_h_> hazmat: yea working around pip atm 
[17:21] <rick_h_> sudo pip install juju-deployer bzr --allow-all-external --allow-insecure=bzr
[17:21] <rick_h_> wheeee
[17:21] <hazmat> rick_h_, oh yeah.. pip got sane about security
[17:21] <rick_h_> for not upload builds to pypi
[17:21] <hazmat> rick_h_, we just need to upload a bzr src tarball to pypi
[17:25] <rick_h_> hazmat: running juju-deployer -c xxx.yaml and getting a tsl error? "alert protocol version"? ring any bells?
[17:25] <rick_h_> ssk
[17:25] <rick_h_> bah
[17:26] <rick_h_> ssl error that is
[17:27] <hazmat> rick_h_, likely your using an old version of jujuclient
[17:27] <rick_h_> hazmat: k
[17:27] <hazmat> rick_h_, pip install -U jujuclient
[17:28] <rick_h_> yea, just intsalled, already up to date
[17:30] <hatch> of course, when I need charmworld to be slow it's blazing fast lol
[17:34] <rick_h_> bah, man this guy is finding every broken pita thing we've got. 
[17:34] <rick_h_> and now my lunch is cold boooo
[17:43] <hatch> internet throttled....oh boy this is slow
[17:43] <rick_h_> whooops
[17:43] <hatch> on purpose
[17:43] <hatch> i mean
[17:43] <rick_h_> lol
[17:43] <rick_h_> what are you trying to replicate?
[17:43] <hatch> I'm working on the charm list headers rendering over the inspector
[17:44] <hatch> but the results were coming back faster than I could click on the service icon
[17:46] <hatch> the silver lining is that the GUI works well on a throttled network connection
[17:46] <hatch> :)
[17:46] <hatch> it's slow, no doubt about that, but functional
[17:48] <hatch> there are definitely some improvements to make on the slow loading side of things
[17:51] <hatch> nice, it works
[17:54] <hatch> hey rick_h_ , CI merge hung so I killed it but now I think you'll need to go in and killall node processes
[17:57] <hatch> ahhh back to full speed again
[18:36] <hatch> rick_h_ yeah EADDRINUSE
[18:36] <rick_h_> hatch: sorry, missed the ping looking
[18:37] <hatch> rick_h_ maybe before you go on holidays you add a killall at the start of the CI script? :)
[18:37] <hatch> ya know....cuz I'm sure you have nothing else to do right? haha
[18:37] <rick_h_> hatch: killed
[18:37] <hatch> thanky
[18:37] <rick_h_> hatch: well I put the ssh key in the wiki for others to be able to ssh in and kill :P
[18:37] <hatch> oh ok cool
[18:38] <kadams54> guihelp: Need review and QA on https://github.com/juju/juju-gui/pull/346
[18:38] <hatch> sure i'll take it
[18:39] <hatch> kadams54 you commented on a loc but it didn't expand in the conversation or show in the code.....odd
[18:39] <kadams54> Hmm, that is odd
[18:40] <kadams54> I've never seen that happen.
[18:40] <kadams54> https://github.com/kadams54/juju-gui/commit/a9744286283696056dcae11eaec92cd3213a126a#commitcomment-6462503
[18:40] <hatch> ahhh you commented on the commit not on the PR
[18:40] <hatch> interesting
[18:41] <kadams54> Oops, yeah, you're right
[18:41] <kadams54> I clicked on the commit hash in the PR "Conversation" rather than using the "Files changed" tab.
[18:41] <kadams54> Moving it to the PR.
[18:44] <hatch> I think the failure is a lint issue
[18:46] <kadams54> Yeah, forgot to "document" initializer. It's fixed and pushed.
[18:50] <rick_h_> man I can't wait until quickstart works on osx, it'll make this crap so much nicer
[18:51] <rick_h_> don't appreciate how nice something is until you have to work without
[18:52] <kadams54> afk to pick the kids up from school. Should be back around 3:30-ish.
[18:52] <kadams54> (Eastern)
[19:20] <rick_h_> hazmat: filing some deployer bugs. We will probably check them out as part of quickstart on osx fyi
[19:31] <rick_h_> Makyo: call
[20:33] <hatch> kadams54 I made a comment in your PR, lemme know if you want to chat about it in more detal
[20:36] <kadams54> hatch: yeah, that was essentially my thought in that comment about splitting out into a handlebars template
[20:37] <kadams54> I was also tempted to make the header label a widget with a render method
[20:40] <hatch> that's probably overkill
[20:40] <kadams54> hatch: hah, OK, when you think it's overkill… ;-)
[20:40] <hatch> because if the label is a template all you need to do is ` var label =  this.labelTemplate(e.newVal); this.get('container').one('.label-span').setHTML(label); `
[20:41] <hatch> haha
[20:41] <kadams54> yeah ill go that route
[20:42] <hatch> your current approach is just 'too good'
[20:42] <hatch> haha
[20:44] <hatch> rick_h_ with your kinesis how do you hit the ctrl + alt buttons? Did you remap them?
[20:45] <rick_h_> yes, ctrl is caps lock, alt is a thumb based key
[20:46] <hatch> ahh ok - I was looking at the key layout and I didn't think it would work at all for coding with the defaults
[20:47] <rick_h_> hatch: it works pretty well, the only thing is that the curved layout makes ; a little harder for my little pinky
[20:47] <rick_h_> but you get used to it
[20:47] <hatch> I spent too much time last night researching it and the TECK keyboard
[20:47] <hatch> heh
[20:48] <rick_h_> heh, been there done that
[20:48] <rick_h_> ok, I'm out for now. Dog to the boarding place so I can go on holiday!
[20:48] <rick_h_> have a good night all
[20:48] <hatch> lataz
[21:11] <kadams54> hatch: switched to using a template. Much simpler: https://github.com/kadams54/juju-gui/commit/dd60b784bb7802a7149a5e7df89da7083eab12bd#diff-30cfb18f4179247092a96597a449670dR55
[21:11] <hatch> looking
[21:12] <hatch> :)
[21:12] <hatch> gota love it
[21:21] <hatch> kadams54 lint error again :)
[21:22] <hatch> you should have a pre-commit hook that runs the linter lol
[21:22] <kadams54> I really should
[21:53] <hatch> handlebars helpers in the template sure look odd
[21:53] <hatch> `pluralize label count` heh
[21:57] <hatch> kadams54 I have another suggestion for simplification, sorry I should have seen this before
[22:12] <kadams54> Go for it
[22:12] <kadams54> hatch: ^
[22:12] <hatch> I made it in the PR :)
[22:12] <kadams54> Ah yes, I see that now :-)
[22:13] <kadams54> I like that.
[22:13] <hatch> don't worry, we'll have removed 90% of the code you wrote by the end of the day!
[22:13] <hatch> lol
[22:13] <hatch> sorry haha
[22:59] <huwshimi> Morning
[23:01] <hatch> jujugui looking for a review for bug #1322255 https://github.com/juju/juju-gui/pull/348
[23:01] <_mup_> Bug #1322255: Sticky headers render over inspector <juju-gui:In Progress by hatch> <https://launchpad.net/bugs/1322255>
[23:01] <hatch> morning huwshimi 
[23:04] <huwshimi> hatch: Morning