2014/12/05/#juju-gui.txt

[00:07] <rick_h_> hatch: sure
[00:07] <rick_h_> hatch: is this one done/ready for review?
[00:07] <rick_h_> ah, I see it is in there now
[00:07] <rick_h_> hatch: maybe try to get one review from Makyo or huwshimi while they're around (not sure how much longer that's true)
[00:12] <hatch> rick_h_: yep it's all ready now
[00:13] <hatch> ok will do
[00:28] <Makyo> Can't get selenium tests to even run, much less pass >:/
[00:28] <Makyo> Seems selenium can't find firefox or something.
[00:28] <hatch__> that's odd...any errors?
[00:29]  * hatch__ is trying out this terminal irc thingy (weechat) :)
[00:29] <Makyo> Yeah, it can't load the firefox profile anymore.
[00:29] <hatch__> hmm I've never seen that error before
[00:29] <Makyo> http://paste.ubuntu.com/9374600/
[00:30] <hatch__> apparently I have to hold ctrl to click links in this thing :)
[00:31] <rick_h_> hatch: middle-click ftw
[00:31] <rick_h_> :)
[00:31] <rick_h_> Makyo: is this locally?
[00:31] <rick_h_> Makyo: or on CI?
[00:31] <hatch__> apparently I don't have a middle click....clicking the scroll wheel does not appear to work
[00:32] <Makyo> rick_h_, locally.  The one on CI I got around by `make clean-all && make build`
[00:32] <rick_h_> hatch: well then you need a better mouse
[00:32] <Makyo> Or...that was two of them ago.
[00:32] <rick_h_> hatch: so sucks if the one I got you doesn't have middle-click
[00:32] <Makyo> Either way, it's an environment problem, not a test problem (so far)
[00:33] <hatch__> rick_h_: haha well we'll see - I just got weechat running so I may be doing something wrong
[00:34] <hatch__> initially however I'm really liking it
[00:34] <hatch__> can bomb around the controls pretty fast
[00:34] <hatch__> need to see if I can see all the channels/servers in a list somwhere though..
[00:35] <hatch__> oh they call them 'buffers'
[00:35] <hatch__> at least now I know what to search for :D
[00:36] <hatch__>  /whois hatch_
[00:37] <hatch__> hmm
[00:38] <hatch__> niiice now I have that working
[00:38] <teslanick>  /ghost hatch
[00:38] <hatch__> haha
[00:38] <hatch__> says hatch_ is taken, but whois returns nothing
[00:38] <teslanick> No, srsly, that's how you boot your dead selfs
[00:38] <teslanick> Huh
[00:39] <hatch__> oh there it is
[00:39] <hatch__> must have been a typo or something
[00:39] <hatch__> yup someone else is hatch_ :/
[00:39] <Makyo> You put a space in front of it.
[00:39] <hatch__> lame sauces
[00:40]  * hatch__ <--- real slim shady
[00:40] <Makyo> Best stand up, then
[00:40] <teslanick> that's funnier to me because you're like... super-canadian.
[00:40] <hatch__> hahahaha
[00:40] <hatch__> wha my name is, who my name is, wha my name is, chicka chicka slim shady!
[00:41] <Makyo> Oy gevalt
[00:41] <hatch__> well so far I'm digging weechat
[00:41] <hatch__> now if only I could make vim be sublime then I'd only need chrome and terminal :)
[00:42] <teslanick> And your journey to the nerd side will be complete.
[00:42] <teslanick> Actually, that journey's end is when you can make vim into chrome.
[00:43] <hatch__> lol does links count?
[00:43] <teslanick> I used to use links.
[00:44] <hatch__> I just can't figure out why sublime and terminal use the identical font but render it totally different
[00:44] <hatch__> it's irritating me because I want terminal to look like sublime :)
[00:44] <teslanick> In OSX? Good luck.
[00:45] <hatch__> no I'm all Ubuntu now
[00:45] <hatch__> Ubumini IS ALIVE!
[00:45] <teslanick> Oh, well then you just need to alter your /etc/cfg/nobody-gaf.conf
[00:45] <teslanick> (trollface)
[00:45] <hatch__> rofl!
[00:46] <teslanick> I gotta ressurect my gaming machine. It was running ubuntu, but the graphics card *hates* the nouveau drivers.
[00:46] <teslanick> And ununtu hates my mouse. It's a lot of hate in that family of silicon.
[00:46] <hatch__> haha 
[00:48] <hatch__> it takes something like 25s longer to run lint on ubumini than on my MBP
[00:48] <hatch__> *sadface*
[00:49] <hatch__> they put some fast stuff in there
[00:49] <teslanick> What's an ubumini?
[00:49] <hatch__> macmini with ubuntu on it :P
[00:49] <teslanick> Sounds like a kind of dog.
[00:49] <hatch__> it's actually an old mini, probably 2011 era
[00:51] <teslanick> Reminder: Mac Minis are basically Macbook Air's with a fatter hard drive.
[00:51] <hatch__> this true
[00:53] <hatch__> it's actually a pretty smart piece of hardware
[00:53] <hatch__> when I put more ram and a SSD in it, I was impressed by how it all fits together
[00:53] <teslanick> It is. I really like the Mac Mini.
[00:53] <hatch__> unfortunately now they are all soldered/glued 
[00:53] <teslanick> The new ones are? That's too bad.
[00:53] <hatch__> yeah :/
[00:54] <teslanick> Seems like a double-helping of "missing the point"
[01:04] <hatch> huwshimi: think you'll have time today to do that review?
[01:16] <huwshimi> hatch: Yep, probably won't get to it till after lunch though
[01:16] <huwshimi> If that's OK
[01:17] <huwshimi> hatch: It's not a small review...
[01:19] <hatch> haha no, no it's not
[01:19] <hatch> anytime is fine - I'm not going to need it again until the morning :)
[01:20] <huwshimi> Great :)
[01:49] <arosales_> is it correct to state there is no 1.5 quickstart love for OSX?
[01:49] <arosales_> it is looking like brew is only finding 1.4 :-/
[01:49] <arosales_> https://code.launchpad.net/~bac/juju-quickstart/brew-commands looks last updated back in may
[01:52] <rick_h_> arosales_: that's an oversight on our part if there's no 1.5.
[01:52] <arosales_> rick_h_: I am checking  . . .
[01:53] <arosales_> I may have a cached .rb on this OSX machine
[01:53] <arosales_> looks like brew search finds it
[01:54] <rick_h_> arosales_: yea, the formula is 1.5 https://github.com/Homebrew/homebrew/blob/9b63c2864d890a137e0e841ea043a0aad58cc20a/Library/Formula/juju-quickstart.rb
[01:54] <rick_h_> in brew trunk
[01:54] <arosales_> rick_h_: ok, thanks for verifying
[01:54] <rick_h_> np, thanks for mentioning it. I know I forget about brew sometimes
[02:00] <arosales_> rick_h_: ok 1.5 looks to be downloaded
[02:00] <arosales_> rick_h_: sorry for the false alarm
[02:01]  * arosales_ grabs some dinner
[02:16] <hatch> teslanick: you're back - did ya rage quit? :D
[02:17] <teslanick> No, I decided to eat something
=== kadams54 is now known as kadams54-away
[02:56] <hatch> hmm being connected to two servers is confusing - I can easily see myself accidentally typing in the wrong channel...wonder if there is a way to better separate them
=== kadams54 is now known as kadams54-away
=== kadams54-away is now known as kadams54
[04:28] <hatch> huwshimi: I'm looking at your branch right now
[04:29] <huwshimi> hatch: Ah thanks. It's horrible, I wouldn't look at it if I were you
[04:30] <hatch> where are these remove/addClass methods defined?
[04:31] <huwshimi> hatch: Just to remove the duplication of having to do that same thing five times
[04:31] <huwshimi> hatch: YUI is not available.
[04:31] <huwshimi> hatch: As I say, it's horrible.
[04:32] <hatch> ohh ok
[04:32] <hatch> I'm guessing you didn't know about element.classList.add() and .remove() :)
[04:32] <huwshimi> ...
[04:32] <huwshimi> Can I just finish for the day then?
[04:32] <hatch> lol
[04:32] <hatch> I'll comment in the PR 
[04:33] <hatch> hey, at least I stopped in to take a peek haha
[04:33] <huwshimi> hatch: Thank you :)
[04:37] <hatch> huwshimi: ok done reviewing
[04:37] <huwshimi> hatch: Thanks!
[04:39] <hatch> :) np
[04:49] <hatch> huwshimi: mind sending me a PM? I'd like to see how weechat handles it
=== kadams54 is now known as kadams54-away
[05:10] <huwshimi> yes
[05:10] <hatch> lol
[05:10] <hatch> I thought for a second there I screwed something up 
[05:10]  * hatch is trying to figure out weechat :)
[05:10] <hatch> ok so why do you have to mock it out?
[05:11] <huwshimi> hatch: My irc client defaults to #juju-dev when it opens up
[05:11] <hatch> I thought that you were passing the element in ?
[05:11] <huwshimi> for some reason
[05:11] <hatch> oh hhah
[05:11] <huwshimi> hatch: I am, only so that in the test I can pass in a mock (I assume that's why it has been set up that way)
[05:11] <hatch> oh, why don't you pass in an element?
[05:12] <huwshimi> hatch: I assume because the dom is not ready when we execute this js
[05:12] <hatch> it doesn't need to be https://developer.mozilla.org/en-US/docs/Web/API/document.createElement
[05:13] <hatch> element.classList.... is a JS representation of the DOM element (JSDOM) so as long as you pass in something similar it won't care if the dom is ready or not
[05:13] <huwshimi> Or I could do that...
[05:13] <hatch> :)
[05:14] <hatch> with that said however....the DOM should be available in that test
[05:15] <hatch> but if you're just unit testing it then yeah might as well just use the above method
[05:15] <hatch> method above*
[05:16] <huwshimi> hatch: Oh, I might have been having issues as I was trying to use our getDocument() method
[05:16] <huwshimi> which obviously wouldn't work there
[05:17] <huwshimi> But neither is this :(
[05:17] <hatch> is it throwing an error?
[05:18] <huwshimi> Error: INVALID_CHARACTER_ERR: DOM Exception 5
[05:18] <hatch> oo fun
[05:18] <huwshimi> Might be that we're overwriting it var document = getDocument();
[05:19] <huwshimi> and then getDocument = function() {
[05:19] <hatch> what do you have for your createElement line?
[05:19] <hatch> that error is usually because of that
[05:19] <hatch> the arguments passed to that method
[05:19] <huwshimi> var logoutMock = document.createElement('<div></div>');
[05:20] <hatch> document.createElement('div')
[05:20] <hatch> createElement takes a 'tagName' as it's first argument 
[05:21] <huwshimi> Ah
[05:21] <hatch> yeah the YUI wrapper hides all that stuff away in the background
[05:21] <hatch> it essentially takes a string and creates the respective commands
[05:28] <hatch> huwshimi: got it all working?
[05:28] <huwshimi> hatch: Yep, all good, thanks for that
[05:28] <hatch> np I'm off, have a good weekend!
[05:29] <huwshimi> hatch: Thanks have a good one!
[09:17] <mhilton> uiteam: morning
[09:17] <urulama> morning
[09:17] <urulama> a change of servers? :)
[09:20] <rogpeppe> mhilton: hiya
[09:21] <mhilton> uiteam: Could I get a second review of https://github.com/juju/charmstore/pull/248 please? and a QA
[09:42] <rogpeppe> mhilton: looking
[12:49] <rick_h_> hatch: external links please :P
=== kadams54 is now known as kadams54-away
=== kadams54-away is now known as kadams54
[15:13] <hatch> rick_h_: oops sorry 
[15:13] <hatch> I just figured out scrollback :)
[15:13] <rick_h_> huh?
[15:13] <hatch> in weechat
[15:13] <hatch> I didn't see your msg
[15:13] <rick_h_> which one?
[15:13]  * rick_h_ is lost/confused
[15:13] <hatch> external links
[15:13] <rick_h_> still confused
[15:14] <hatch> about 3h ago you said "external links please :P"
[15:15] <rick_h_> hatch: in another channel
[15:15] <rick_h_> ?
[15:15]  * rick_h_ is so confused 
[15:15] <hatch> hmm
[15:16] <hatch> 06:49:15   rick_h_ | hatch: external links please :P
[15:17] <hatch> maybe mentions from all channels go into this window? Now I'm wondering if I broke something setting this up lol
[15:32] <hatch> rick_h_: I would like to land that branch today just in case the api release gets delayed :)
[15:32] <hatch> unless you're pretty confident that's not going to happen
[15:32] <rick_h_> hatch: understand, I've ok'd it as is
[15:32] <hatch> thnks
[15:50] <rick_h_> uiteam call in 10 kanban please
[15:57] <hatch> sheesh I ordered some boots which had overnight shipping 6 days ago - they are being delivered today
[15:58] <hatch> so much for under promise over deliver
[17:39] <hatch> I think I'm going to run for an early lunch 
=== kadams54 is now known as kadams54-away
=== kadams54-away is now known as kadams54
[19:14] <rick_h_> Makyo: any luck with your functional tests?
[19:15] <Makyo> rick_h_, no luck yet. Trying to get sauce connect up and running so I can run them locally.
=== kadams54 is now known as kadams54-away
[20:03] <Makyo> uiteam running up against a wall with sauceconnect, keep getting 502 Bad Gateway.  Anyone have luck running selenium tests locally?
[20:04] <hatch> Makyo: can sauce reach your machine from outside?
[20:05] <hatch> Makyo: I usually do 'local' sauce tests on ec2 or the like
[20:05] <hatch> because I've had issues with incoming requests on my network
[20:05] <Makyo> oh, that makes sense.  Probably easier than screwing with my netowkr. 
[20:05] <Makyo> Or network, for that matter.
[20:05] <hatch> :) hope it helps!
[20:05] <hatch> can someone mention me? I want to see if dings are working now
[20:06] <rick_h_> Makyo: just claim CI
[20:06] <rick_h_> Makyo: ssh to it, start the server, run it from there and you know sauce can talk to it
[20:06] <Makyo> hatch ping
[20:06] <Makyo> rick_h_, ah, alright, will do
[20:07] <hatch> it is not working :/ well I'll have to investigate after eod
=== kadams54 is now known as kadams54-away
[20:58] <hatch> Hmm, I'm wondering if fetching charms and bundles should always return a promise...
[20:58] <hatch> or remove the promise bit entirely
=== kadams54-away is now known as kadams54
[20:58] <hatch> ^ Makyo you probably have some input in this :)
[20:58] <Makyo> :T
[20:59] <Makyo> I really like the idea of promises, just not the implementation.
[20:59] <Makyo> I'd say move away from promises, hatch.
[20:59] <Makyo> If we're looking at moving away from YUI, I suggest as few little bits of it in place as possible.
[21:00] <hatch> I'd be ok with that - I just don't want to have to implement both like the current charmworld api
[21:00] <hatch> makes it confusing to work with
[21:00] <Makyo> Ahh, yeah
[21:00] <hatch> promiseBundle isn't actually ever used 
[21:00] <hatch> lol
[21:01] <Makyo> Oy
[21:01] <hatch> actually tbh I don't see why we need a bundle and charm charmworld api endpoint now that they are on equal footing
[21:02] <Makyo> If there
[21:03] <Makyo> 's a way to simplify, great
[21:03] <hatch> yeah this saves probably 50+ loc's from the charmstore api
[22:18] <rick_h_> Makyo: better progress on CI?
[22:19] <rick_h_> hatch: as for the two endpoings, we can simplify but we have to note the diff in charm vs bundle metadata
[22:20] <hatch> rick_h_: well the client will request data from the endpoint and then the supplied callbacks will handle what's returned
[22:20] <rick_h_> hatch: sure, just saying there's still a little diff in the two things that needs some <3
[22:21] <hatch> oh yeah for sure
[22:21] <hatch> maybe I'll run into an issue when I pass the details into the tesseract
=== kadams54 is now known as kadams54-away