[10:59] <rick_h_> morning
[12:51] <hatch> morning
[13:00] <hatch> rick_h_, thoughts on changing the browser flag to just be 'browser' ?
[13:00] <rick_h_> hatch: yes :)
[13:01] <rick_h_> hatch: so it's in flux at the moment. I'm finding some issues following the old LP FF conventions because it used a . notation which in JS looks like nested objects. So went to -, then _. Trying to work out the best way still
[13:01] <rick_h_> hatch: but we could easily want to have a series of flags in the browser subapp so I kind of prefer the explicit browser_enabled/browser_newux and such since it's a subapp
[13:01] <rick_h_> hatch: be glad I dropped the original plan of subapps_browser_enabled :P
[13:02] <rick_h_> which I like but don't want to type
[13:02] <hatch> haha - I am just using :flags:/inspector as a truthy check
[13:02] <rick_h_> yea, everything so far has been a single term, but I think anything in a subapp at least should be NS
[13:03] <rick_h_> as the start of some useful conventions
[13:24] <hatch> maybe we can use https://github.com/adambom/parallel.js to speed up our tests :)
[13:25] <rick_h_> hatch: not until we rewrite everything :P
[13:25]  * rick_h_ has been doing too much window. in the code to feel that we're anywhere near parallel tests. 
[13:26] <hatch> haha hey it's a good dream
[13:27] <rick_h_> +1
[13:31] <hatch> I was looking at these new designs from luca on my laptop and am kind of concerned that the top 'chrome' takes up a good 1/5th of my screen on my laptop
[13:32] <hatch> so I can only actually see 6 charms in the search results page and can't even see their full descriptions
[13:32] <hatch> it feels very 'zoomed in' 
[13:32] <hatch> this of course likely isn't an issue on my desktop :)
[13:32]  * gary_poster hasn't seen yet...
[13:33] <rick_h_> jcsackett: standup ping
[13:33] <jcsackett> rick_h_: fighting with g+
[13:33] <rick_h_> hatch: up for some review love? https://codereview.appspot.com/9731047 jcsackett as well please ^^
[13:34] <benji> has anyone seen this while trying to do an lbox propose? http://paste.ubuntu.com/5710198/
[13:34] <hatch> rick_h_, sure
[13:34] <hatch> benji, never seen that
[13:34] <gary_poster> me either...
[13:34] <rick_h_> benji: think I saw that in my raring install. Sec, checking my history
[13:35]  * hatch is glad he isn't on raring yet ;)
[13:35] <benji> I recently upgraded to raring.
[13:35] <rick_h_> benji: sec, on call. Will dbl check.
[13:36] <benji> even though I've had some small issues (this and the fact that lbox isn't available for it) it has been worth it because my networking (wifi) has been so much better
[13:36] <gary_poster> yay!
[13:36] <hatch> benji, I have read others complaining about wifi issues haha
[13:36] <hatch> maybe you took their good luck :)
[13:36] <benji> heh
[13:37] <hatch> ubuntu needs a better driver story
[13:38] <hatch> I don't know how many hours i've fought with installing graphics drivers in the correct order to get things to work properly
[13:38] <hatch> I know it's not ubuntu's fault but...ya know
[13:39] <gary_poster> lol, I go to uistage.jujucharms.com:8080, and Chrome says "This page is in Galician.  Would you like to translate it?"  um?
[13:39] <gary_poster> I reported an incorrect language detection
[13:40] <hatch> gary_poster, haha I've ran into some odd languages there too!
[13:40] <gary_poster> :-)
[13:40] <gary_poster> hatch I compared https://launchpadlibrarian.net/140916463/Juju_Full_Home.png against http://uistage.jujucharms.com:8080/bws/fullscreen/
[13:41] <gary_poster> new version is definitely taller at top
[13:41] <gary_poster> but only in "Charm browser" section, for one thing
[13:41] <gary_poster> and not *wildly* taller
[13:41] <gary_poster> 20 or 30 px maybe?
[13:42] <hatch> ahh yes you're right - I guess the design itself has always been like that
[13:42] <gary_poster> luca, hi.  fwiw, hatch reported that the charmbrowser header took up 1/5 of his laptop screen
[13:43] <gary_poster> hatch, what's that screen res?
[13:43] <hatch> sorry I meant the whole gui chrome
[13:43] <hatch> 1366x768
[13:43] <gary_poster> hatch, yeah sorry I meant that when you are using the charmbrowser in full screen, 1/5 of the screen is in the header
[13:43] <luca> gary_poster: hatch I'm working on minimising the headers in the GUI to something more "refined" a lot of it currently is throw back from previous designers
[13:44] <gary_poster> luca cool, thanks.
[13:44] <luca> gary_poster: hatch mainly because it does the same thing on my laptop hehe :)
[13:44] <gary_poster> lol
[13:44] <hatch> haha 
[13:44] <hatch> oh right you're on a air
[13:44] <hatch> an*
[13:46] <luca> hatch: yup
[13:46] <hatch> alright then - so you can easily see the low res issues
[13:46] <hatch> I'm not even sure this is an issue for us, but I kind of assumed it would be used by ppl on their laptops
[13:47] <gary_poster> frankban, did/does the NOT LGTM review on the API address juju core branch mean a big change?
[13:47] <gary_poster> hatch, oh no, people definitely don't use laptops ;-)
[13:47] <hatch> lol
[13:47] <hatch> maybe they are all using retina MBP's so it's not an issue :)
[13:47] <gary_poster> :-)
[13:47] <frankban> gary_poster: quick hangout?
[13:48] <gary_poster> sure frankban 
[13:48] <frankban> gary_poster: I am in the gui chat
[13:50] <rick_h_> benji: ok, this was part of my virtualenv hell 
[13:50] <rick_h_> benji: so I had to for install an updated virtualenv 1.9 to get that working. 
[13:51] <benji> please say there is a ppa, please say there is a ppa
[13:51] <rick_h_> benji: wget "https://pypi.python.org/packages/source/v/virtualenv/virtualenv-1.9.1.tar.gz" && sudo pip install virtualenv-1.9.1.tar.gz
[13:51] <rick_h_> lol
[13:51] <benji> darn
[13:51] <rick_h_> well I didn't look for a PPA
[13:51] <rick_h_> I'm anti .deb for my .py so it was a 'whatever works' in this lxc
[13:51] <benji> I hate installing junk in my OS.  I guess I need to go back to heavy lxc use
[13:52] <rick_h_> benji: but yea, there's some issue in how python got multi-arch'd and virtualenv in 13.04 that causes issues and the python-virtualenv isn't new enough to handle it in 13.04
[13:52] <rick_h_> jcsackett: picking up a review on your caching branch
[13:53] <benji> you'd think that would be on the did-we-break-python checklist
[13:53] <rick_h_> yea, it's how I found out I was the only one using a raring lxc on the squad :/
[13:55] <jcsackett> rick_h_: wait, it's up?
[13:55] <rick_h_> jcsackett: just heads up I'm doing on of the reivews for your cache branch.
[13:56] <jcsackett> rick_h_: right, i'm just surprised--from my end it looks like it kept dying without completing the proposal.
[13:56] <jcsackett> "it" being lbox.
[13:56] <rick_h_> jcsackett: oh, got the emails
[13:56] <jcsackett> rick_h_: well, fantastic.
[14:05] <bac> hi rick_h_, i see you added 'bin/*' to .bzrignore for juju-gui.  those files are versioned so that seems ungood to me.
[14:05] <rick_h_> bac: oh hmm, /me dbl checks what was up with that
[14:05] <hatch> rick_h_, I don't see anything setting noop in your tests...isn't that what it was for?
[14:05] <bac> rick_h_: ok, let me know if i shouldn't revert it.
[14:06] <rick_h_> bac: this is from a while ago?
[14:08] <rick_h_> bac: yes, please revert if you need it now or I can drive-by in my current branch. This was from my own putting the gui in a virtualenv and didn't realize I was colliding with existing stuff
[14:08] <rick_h_> hatch: looking
[14:10] <rick_h_> hatch: it's set and used in the app/subapps/browser/browser.js
[14:10] <bac> rick_h_: ok, i'll do it in my branch
[14:11] <rick_h_> bac: thanks
[14:12] <hatch> rick_h_, ohh ok now I see the intended use
[14:12] <rick_h_> hatch: cool, please suggest any docs updates/etc to help make it clearer. Been a while since I looked at it now myself :/
[14:12] <rick_h_> stupid long hanging branches of doom
[14:13] <hatch> yeah I have one of those myself
[14:15] <hatch> rick_h_, LGTM'd now to qa
[14:28] <hatch> rick_h_, QA looks OK
[14:28] <rick_h_> hatch: cool thanks
[14:31] <benji> rick_h_: have you seen this one? http://paste.ubuntu.com/5710371/
[14:33] <rick_h_> benji: yep, got that one as well. I've not gotten past that one. I only get it on lbox submit and have been doing submit from my laptop which isn't in an lxc. I thought I got that around getting the gslint package from the local path or something
[14:33] <rick_h_> -cr work fine :/
[14:34] <rick_h_> benji: but i had assumed that was more to do with me running a local pypi mirror and it's not in the lxc yet so ugh that you get it as well
[14:34] <benji> hrm
[14:38] <rick_h_> benji: https://pastebin.canonical.com/91696/ is how my version looks
[14:39] <rick_h_> benji: not sure why it's looking for mypi since I commented that out of my pip config, but seems like it's still somewhere. Assumed that's why I was getting this error.
[14:40] <benji> grr, it is using zipped eggs, so debugging is harder
[14:46] <hatch> benji, are zipped eggs some kind of southern delicacy? 
[14:46] <benji> heh
[14:46] <hatch> :)
[14:52] <rick_h_> jcsackett: notes inbound, let me know if you want to hangout on it
[14:56] <jcsackett> rick_h_: may want to chat, reading notes/considering.
[14:57] <rick_h_> jcsackett: rgr
[14:57] <hatch> luca, I just noticed in one of your mockups you use a comma in 10,234 and the other two don't - not sure if it matters but thought I'd point it out :)
[14:58] <rick_h_> jcsackett: just invite me to a hangout or we might even be able to co-opt guichat. 
[14:59] <luca> hatch: there isn't meant to be a comma, the visual designer is sending me a masthead file for you guys to reference. Cheers for the heads up :)
[15:01] <hatch> :)
[15:05] <benji> score one for tests: I just made a "small" change that would have broken my branch if I hadn't had tests
[15:05] <hatch> go tests....go tests.....go tests.......
[15:09] <hatch> there is TDD, BDD, and I am going to coin the term RDD
[15:09] <hatch> Regression Driven Development :D
[15:12] <gary_poster> yay, thank you rick_h_ .  have not checked out browser default flag yet, but saw the landing :-)
[15:12] <rick_h_> gary_poster: cool, hatch qa'd and should be good. next up...icon svg out of the api updates
[15:12] <gary_poster> yay! :-)
[15:12] <benji> perky yet refined review available: https://codereview.appspot.com/9657046
[15:15] <teknico> how classy :-)
[15:16] <teknico> Makyo: got time for a quick chat? d3 is confusing me
[15:16] <Makyo> teknico, Sure.
[15:16] <teknico> Makyo: https://plus.google.com/hangouts/_/02bb45411739e441fe107c9f66e2a8cc36ba4ba7?authuser=0&hl=en
[15:26] <abentley> rick_h_: Since you created "Update charmworld docs/ to match new life with injestor.", could you please review https://code.launchpad.net/~abentley/charmworld/ingest-docs/+merge/166080 to see if it addresses it?
[15:27] <rick_h_> abentley: sure, give me a couple and I'll peek at it.
[15:27] <hatch> hmm I'm getting the Uncaught SyntaxError: Unexpected end of input  errors from jujucharms again
[15:44] <rick_h_> hatch: more details req'd. DNS issue? actually getting a bad response? 
[15:45] <rick_h_> url that you're hitting it's failing on, etc
[15:45] <hatch> rick_h_, it will just cut off the transfer
[15:45] <hatch> it happened at home on my desktop too
[15:45] <hatch> so I know it's not me
[15:45] <hatch> but that's all I have - if you make the request directly it works 100% of the time
[15:45] <gary_poster> could be following you around hatch.  We have our ways.
[15:45] <hatch> but through the app it appears to be cutting off the transfer
[15:45] <hatch> lol
[15:46] <rick_h_> hatch: so compare headers? something diff between app req vs manual
[15:47] <jcsackett> rick_h_: i've published my replies. you want to read them and then see if we can have a chat before the jujugui standup? or do you want me to ping you after that?
[15:47] <jcsackett> oops, forgot the hyphen. sorry for the needless alerts, all. :-P
[15:47] <rick_h_> jcsackett: sure, reading now.
[15:47] <hatch> rick_h_, forget that - I just got it to cut off when going direct
[15:50] <hatch> rick_h_, yeah from time to time - apparently without reason https://manage.jujucharms.com/api/0/charms/interesting will just stop sending data between 8-12s 
[15:51] <hatch> it always appears to stop during one of the svg's though
[15:51] <hatch> not sure if that helps with anything though :)
[15:54] <abentley> jcastro: I think https://bugs.launchpad.net/charmworld/+bug/1015767 has already been done.  Can you confirm?
[15:54] <_mup_> Bug #1015767: Queue time needs to be measured by last modified, not creation time. <charmers> <charmworld:Triaged> <https://launchpad.net/bugs/1015767>
[15:54] <gary_poster> jujugui call in 6. kanban now?
[15:54] <jcastro> abentley: still open afaict
[15:55] <abentley> jcastro: The age column is based on last modified.  You can see that in "Merge for apache2 from precise/apache2/additional-modules" for example.
[15:56] <jcastro> oh, I misunderstood!
[15:56] <abentley> jcastro: Maybe we need a text change?
[15:57] <jcastro> yeah, there was something we wanted tweaked
[15:57] <jcastro> sec, searching mails for it
[15:58] <jcastro> abentley: oh, we wanted non-employee submissions prioritized
[15:58] <jcastro> but I guess that's a new bug right?
[15:58] <abentley> jcastro: I would say.
[15:58] <gary_poster> jujugui call in 2
[15:58] <jcastro> ok so should I resolve this one?
[15:59] <abentley> jcastro: The only think I see outstanding is hazmat wanted date modified to be displayed as a column.
[15:59] <jcastro> yeah that works for me
[15:59] <abentley> jcastro: And if you can think of a clearer column name for "age", that would be nice too.
[15:59] <jcastro> Time in Queue
[16:00] <abentley> jcastro: last-modified doesn't measure time in queue.  Is that going to be okay?
[16:00] <jcastro> abentley: we had a bunch of ideas on the queue, if you want me to like outline them all etc I can do that
[16:00] <jcastro> ok, how about "Last Modified" then?
[16:00] <jcastro> I don't care too much how long something is in there, I care that when someone posts a new revision it gets responded to quickly.
[16:00] <abentley> jcastro: It would be confusing to have two columns called Last Modified.
[16:01] <jcastro> like if a charm takes a month to make it through that's fine, as long as we're responsive.
[16:01] <jcastro> I guess leave it as age then?
[16:02] <abentley> Okay.
[16:03] <jcastro> https://bugs.launchpad.net/charmworld/+bug/1185087
[16:03] <_mup_> Bug #1185087: Prioritize new charms in the review queue <charmworld:New> <https://launchpad.net/bugs/1185087>
[16:03] <jcastro> I think that does what we want instead of prioritizing based on employer.
[16:04] <jcsackett> welp, that was the uber crash, jujugui. be back in a moment.
[16:07] <hatch> https://manage.jujucharms.com/api/0/charms/interesting
[16:07] <hatch> ^ jcsackett 
[16:08] <hatch> it appears to frequently cut off around 8-12s inside one of the svg's 
[16:10] <abentley> hatch: rick_h_ is updating to API 1, which will remove the svgs from the output.
[16:12] <hatch> abentley, ok great - I hope that helps fix the issue as a side effect
[16:13] <abentley> hatch: It seems to.  https://manage.jujucharms.com/api/1/charms/interesting loads fine for me.
[16:15] <hatch> abentley, ahh I didn't know that the api version was up yet - that's so much faster and loads without issue
[16:15] <hatch> excellent
[16:15] <rick_h_> hatch: yea, work in progress. Backend is updated and much nicer. Front end in progress
[16:16] <hatch> right on
[16:18] <hatch> thx
[16:27] <Makyo> ...typo D:
[16:27] <Makyo> All tests pass now.
[16:33] <gary_poster> yay Makyo :-)
[16:36] <teknico> bac: thanks, anyone up for an easy one? https://codereview.appspot.com/9731048/
[16:36] <bac> teknico: benji already did and i'm looking at it
[16:37] <teknico> bac: right, sorry, it was benji
[16:37] <teknico> and Makyo knows about it already, so it sort of makes three :-)
[16:38] <bac> teknico: done
[16:38] <teknico> bac: thanks (for real this time :-) )
[16:45] <benji> bac: review of https://codereview.appspot.com/9836043/ done
[16:46] <bac> benji: thanks for the productive and entertaining review
[16:47] <benji> I can think of no higher praise.
[17:10] <benji> gary_poster: I'll contact you after lunch about how I can help.
[17:11] <gary_poster> sorry benji, thanks.  pulled off to lunch
[17:15] <gary_poster> hatch https://docs.google.com/a/canonical.com/file/d/0B1IM--9A1RkTVGR1N05CdmVfN3c/edit (slides 18, 20, etc.) includes sketches on future directions for inspector.  if you want to incorporate now, great.  if not, we can iterate towards this soon
[17:18] <gary_poster> hazmat, how do I get on uistage again?  ubuntu@ doesn't work any more
[17:19] <hatch> gary_poster, so does this mean we aren't going to the 'floating' panels anymore?
[17:19] <gary_poster> hatch, uh? :-) guichat really quick?
[17:19] <hazmat> gary_poster, root@
[17:20] <gary_poster> ah thanks hazmat 
[17:20] <hazmat> gary_poster, working dir is in /opt
[17:20] <gary_poster> ack thanks again hazmat 
[17:20] <hatch> gary_poster, I can't seem to log in, lemme try to reconnect - the internet here is horrible
[17:20] <gary_poster> ok
[17:22] <gary_poster> bcsaller, I see wordpress on uistage.  I think maybe we don't restart any more?
[17:22] <gary_poster> restart improv I mean
[17:22] <bcsaller> ahh, maybe
[17:39] <gary_poster> thanks hatch. :-) ttyl
[17:49] <hatch> gary_poster, sorry - someone called, apparently that causes the internet to go down
[17:49] <jcastro> heya gary_poster 
[17:50] <gary_poster> hatch, everybody knows you are not allowed to talk on the phone and surf at the same time
[17:50] <gary_poster> jcastro, hey
[17:50] <jcastro> we'd like to add "framework" as a category to show in the gui, do I file a bug or bring it up with design or what?
[17:50] <hatch> I feel like it's 1996
[17:50] <hatch> lol
[17:50] <gary_poster> jcastro, file a bug, gimme the number and I'll run with it
[17:50] <rick_h_> jcastro: file a bug please and if you can bring it up with design add that as well. 
[17:50] <jcastro> ack
[17:51] <rick_h_> jcastro: I think we have to add support to the charm linter, add it to the backend, and then we need a new icon for it
[17:51] <gary_poster> jcastro, I have now run with it.  whew!  that was hard.  thanks rick_h_ :-)
[17:51] <rick_h_> jcastro: so it'll be a few steps to get it through. 
[17:51] <rick_h_> jcastro: and we'll probably ask for another to keep a nice balanced even number of categories for design :P
[17:51] <rick_h_> need two teams to join the league at the same time
[17:52] <jcastro> https://bugs.launchpad.net/juju-gui/+bug/1185125
[17:52] <_mup_> Bug #1185125: Add "framework" as a category  <juju-gui:New> <https://launchpad.net/bugs/1185125>
[17:52] <rick_h_> jcastro: cool thanks. Will bring it up and see if we can't get the ball rolling. 
[17:53] <abentley> orangesquad: I'm linking of using soupmatchers in our charmworld view tests.  Is there a better option?
[17:53] <jcastro> rick_h_: yeah, I'm sure we'll need to come up with ways to remove/add categories, so I figure might as well figure it out now
[17:53] <rick_h_> abentley: lxml?
[17:53] <rick_h_> jcastro: yea, I'm writing out the list, since it's across several code bases and requires design work I'm not sure there's going to be a good answer forit
[17:54] <jcsackett> rick_h_, abentley: i have more familiarity with soupmatchers than lxml.
[17:54] <abentley> rick_h_: How do you test with lxml?
[17:54] <hatch> rick_h_, curious as to the ETA on the api changeover? (planning my day)
[17:54] <rick_h_> abentley: just parse the doc and assert using normal assertions?
[17:54] <rick_h_> hatch: depends, working on it now. In an ideal world I'll have it tomorrow but I've not run all the tests through yet. 
[17:54] <rick_h_> hatch: tomorrow or wed I'd say
[17:55] <rick_h_> abentley: guess I'm not sure what you're using it to check. 
[17:55] <hatch> ahh ok thanks
[17:55] <abentley> rick_h_: the HTML output of a view.
[17:55] <hatch> gary_poster, guichat to resume where we left off?
[17:55] <rick_h_> abentley: right, so you can lxml parse the doc and using the find/etc functions to locate notes/etc. It's C-backed so fast. 
[17:56] <gary_poster> sure hatch
[18:06] <gary_poster> mm, bcsaller, no, uistage is hosed :-( it is blank for me now.  I tried to add mediawiki and it told me that a service of that name already existed.  going to look at ws messages...
[18:07] <jcastro> rick_h_: welcome back man, I missed bothering you
[18:07] <bcsaller> gary_poster: yeah, when I glaced at it it looked like it just wasn't rendering anymore. You can follow the notifications to actual mentioned services and units for example
[18:07] <rick_h_> jcastro: :P
[18:07] <rick_h_> jcastro: we still have to chat blog post. I want to get that out one day
[18:09] <jcastro> rick_h_: I'm on the SE podcast soon until EOD, I can go now though if you want?
[18:10] <rick_h_> jcastro: how about we setup something later int he week. I need to re-read it myself and trying to get this change asap
[18:10] <jcastro> sure
[18:10] <rick_h_> jcastro: long 3hr CHC wed. :P
[18:10] <jcastro> the biggest thing I saw, is that you need to intro it to someone who does not know what you work on
[18:11] <rick_h_> jcastro: yea, agreed
[18:11] <jcastro> right now to understand it you basically have to work @ canonical
[18:11] <jcastro> actually, I wouldn't even mention juju-gui itself, that will just confuse people
[18:11] <jcastro> "I am working on a project and I want to show people a prototype"
[18:11] <rick_h_> jcastro: cool
[18:16] <gary_poster> jujugui trunk appears to be broken.  (1) notifications are hidden when you click on them (2) there's a white bar that is always present on the left, even on inner pages
[18:16] <gary_poster> to demo first one, join to rapi-rollup or go to uistage
[18:17] <gary_poster> to demo second one, should be obvious
[18:17] <gary_poster> There's a third problem that is on uistage: the services do not show up
[18:17] <gary_poster> I  can't dupe that locally yet
[18:17] <gary_poster> could I have someone look at #1 and #2?
[18:18] <bac> other than that mrs. lincoln...
[18:18] <gary_poster> :-)
[18:18] <bac> o/
[18:18] <bac> hatch, rick_h_: either of you care to do an easy review?  linked off my cardd
[18:19] <gary_poster> thank you bac.  I suspect that the merge with browser default might be a part of #2
[18:19] <rick_h_> gary_poster: white bar is my fault. Since we don't load the subapp now it doesn't auto hide it's html bits. 
[18:19] <rick_h_> gary_poster: FF doesn't change the index.html DOM content :/ 
[18:20] <gary_poster> rick_h_, gotcha.  not the end of the world.  #1 is much worse, and yet-to-be-reported #3...
[18:21] <gary_poster> large.json works fine locally...
[18:21] <gary_poster> as does sample.json
[18:25] <bac> so chrome on os x thinks uistage is written in galacian and offers to translate it.
[18:25] <gary_poster> yeah, linux too
[18:25] <gary_poster> I reported it to google <shrug>
[18:25] <gary_poster> not sure what we would have done to cause that
[18:31] <benji> gary_poster: ready to talk when you are
[18:35] <gary_poster> thanks benji.  bcsaller, downloading on uistage and locally (shift d) is not working for me.  it causes a relogin for me.  that's realtively minor compared to the fact that uistage shows no services.  could you please figure out what is wrong with uistage and fix it, other than the #1 and #2 I mentioned above? :-)  I am pretty sure that simply restarting the improv on uistage will fix this but I would rather address
[18:35] <gary_poster>  the problem
[18:36] <gary_poster> bcsaller, will try out benji's tool in anger to see if I can dupe locally that way :-) will report back.  benji, wanna join on guichat and you can talk me through it, and I'll review while I'm at it, and then we'll talk about the rest?
[18:36] <benji> gary_poster: sounds good
[18:36] <bcsaller> gary_poster: I can look at it now
[18:36] <gary_poster> thank you very much bcsaller
[18:37] <bcsaller> app.db.services.size() [18:39] <rick_h_> abentley: ping, in checking out the api1 changes I notice we lost the count and such in search results? http://manage.jujucharms.com/api/1/charms?text=cassandra vs https://pastebin.canonical.com/91727/
[18:40] <abentley> rick_h_: looking.
[18:42] <abentley> rick_h_: No, I don't think API 0 ever supported that, just the pre-0 API.
[18:43] <rick_h_> abentley: ah ok then wasn't used/won't be missed. 
[18:43] <rick_h_> abentley: just checknig around if it was used or not. thanks for peeking
[18:44] <abentley> rick_h_: np.
[18:44] <bcsaller> gary_poster: export currently only works with sandbox as it defers to the backend of the impl. Improv gets the 'exportEnvironment' call and resets the connection on error. We could easily make the fakeback impl the real clientside impl, but that wasn't the intention. We can also limit the binding of the hotkey based on mode, but that also isn't the case 
[18:45] <abentley> rick_h_: You can still get that output at http://manage.jujucharms.com/search/json?search_text=cassandra for now.
[18:46] <gary_poster> bcsaller, ah ok cool, makes sense thanks
[18:46] <gary_poster> bcsaller, you looking at crazy service stuff now?
[18:48] <bcsaller> gary_poster: I suspect someone added a ton of them, there are only something like 140 or 150 units so its mostly services of 1 unit
[18:49] <bcsaller> ahh, but now size is zero
[18:56] <bac> gary_poster: r 689 seems responsible for the left bar but not the notifications
[18:56] <rick_h_> bac: white bar fix coming in a sec. Pushing through lbox now
[18:57] <bac> rick_h_: oh, cool.  what was it?
[18:57] <gary_poster> bac, 689 is doc branch from nicola?
[18:57] <bac> sorry, 690
[18:57] <bac> i confused myself with the reverse cherry pick
[18:57] <rick_h_> bac: putting the browser behind a FF means the code that it ran on startup to hide the minified version wasn't run. :)
[18:57] <gary_poster> ah ok thanks bac
[18:58] <gary_poster> bac, error notifications bug is older?
[18:58] <bac> gary_poster: yes.  will look for it now
[18:58] <gary_poster> cool thanks
[19:06] <rick_h_> bac: gary_poster k, white bar fix landed in trunk. 
[19:06] <gary_poster> awesome thank you rick_h_ 
[19:14] <bac> gary_poster: the notifications occlusion is due to r688
[19:15] <gary_poster> bac, ok, thank you.  fix obvious, or should you ask ben for direction?
[19:16] <bac> gary_poster: haven't found it yet.  will bug ben if needed.
[19:16] <gary_poster> thanks
[19:25] <bac> hi bcsaller, in r688 (Hotkey extensions) you added 'overflow:hidden' for .navbar.  that causes the notification panel to be cut off below the navbar.  i can revert that css change but what were you trying to accomplish?  will doing that revert break something else you were fixing?
[19:26] <bcsaller> bac: ahh, I didn't think that should have that side-effect. When you shrink the width of the viewport a white bar cuts the screen in half w/o something like that fix. This happens alot with the chrome dev tools open on the right side 
[19:26] <bac> ok, let me try to duplicate
[19:27] <bcsaller> bac: different css/clipping overflow on a child element might work, other options might be to position:absolute the notification popup and give it a z-index
[19:28] <bac> ok
[19:28] <bcsaller> the positioning would need to listen on the viewport size changes events so a pure CSS solution might be better if its quick to find
[19:42] <bac> bcsaller, gary_poster: i'm going to land a fix to the notications panel now, though i have reproduced the problem ben talks about.  i'll make a card for fixing it.
[19:42] <gary_poster> bac cool thank you
[20:10] <gary_poster> bac, http server has LGTM with trivial
[20:11] <bac> gary_poster: thanks
[20:11] <gary_poster> bcsaller, should I just restart improv do you think?  or are you trying to figure out why the gui is exploding with 80+ services?
[20:12] <bcsaller> app.db.services.get('displayName')
[20:13] <bcsaller> it looks like its back up to 123 again, not sure if someone is adding those or if they are stale in the system
[20:13] <bcsaller> I'd save the improv log or some of it if possible 
[20:13] <bcsaller> gary_poster: ^^
[20:27] <gary_poster> bcsaller, there is no improv log afaik
[20:29] <bcsaller> gary_poster: it would be in the console, for example if it was being run in tmux, I don't know for sure, if its detached then there is little to be gained that way 
[20:29] <gary_poster> detached
[20:29] <gary_poster> killed and restarted
[20:32] <gary_poster> uistage is AOK for now, thanks all
[20:33] <hatch> sorry if someone was trying to msg me - I just noticed that the internet was down again, not sure how long I was off for
[20:34] <gary_poster> uistage was not AOK for long :-P
[20:34] <hatch> poor uistage
[20:35] <gary_poster> bcsaller, uistage has only five services now and does not render :-/
[20:40] <gary_poster> bcsaller, I can dupe locally with benji's branch and collected websocket output http://pastebin.ubuntu.com/5711447/
[20:40] <gary_poster> so it's something to do with those messages
[20:41] <benji> yay for websocket logging
[20:41] <gary_poster> :-)
[20:41] <gary_poster> thought you might like that
[20:41] <benji> yep
[20:42] <bcsaller> Yeah, I like the dev tools, network websocket,  frames thing as well for this but recording it makes it simpler to talk about 
[20:45] <gary_poster> huh
[20:45] <gary_poster> it's just the delta
[20:48] <bcsaller> gary_poster: what do you mean?
[20:48] <bcsaller> I'm able to run large.json on current improv and trunk locally fine
[20:50] <gary_poster> yes, can only dupe with benji's tool atm.  with that tool discovered that after only the delta, you still do not see services.  when I restart improv on uistage it works for a bit.  I have that delta and am comparing.
[20:50] <bcsaller> gary_poster: ahh, good
[20:51] <gary_poster> yes, the delta diff is the issue.  new pristine delta alone renders fine.  
[20:52] <gary_poster> good version
[20:52] <gary_poster> http://pastebin.ubuntu.com/5711485/
[20:52] <gary_poster> bad version 
[20:52] <gary_poster> http://pastebin.ubuntu.com/5711488/
[20:53] <gary_poster> bad version has annotations...
[20:54] <gary_poster> mere presence of annotations does not trigger
[20:57] <bac> gary_poster: can you ssh to uistage?
[20:57] <gary_poster> bac, yes
[20:57] <bac> hmm, i can't.  must not like my key
[20:57] <bac> had no problem last week
[20:59] <gary_poster> bcsaller, Makyo I see problem with uistage.  trivial but annoying.  guichat, and one of you can take the solution.
[21:00] <bcsaller> joining
[21:00] <bcsaller> server errors
[21:00] <Makyo> Yeah, weird.
[21:00] <gary_poster> bac please remind me tomorrow and I will set up
[21:00] <gary_poster> yeah same here
[21:01] <bcsaller> actual server error
[21:01] <bcsaller> heh
[21:06] <Makyo> gary_poster, bcsaller, Sorry, dryer guy showed up (and he's inexplicably super angry).  Let me know if I can do anything to help
[21:06] <gary_poster> Makyo, !! :-( sorry,  ok thanks,
[21:13] <gary_poster> bac trying jenkins rerun; looks crazy
[21:43] <Makyo> Alright, have a dryer.  Doing laundry, because I'm down to T-shirts *horror*
[21:44] <hatch> so what makes a dryer guy inexplicably angry?
[21:46] <Makyo> He left pre-explication. :)
[21:48] <hatch> haha
[21:48]  * Makyo dogwalks, now that angryguy is gone.
[21:52] <hatch> I know a retired guy who fixes appliances for fun
[23:14] <hatch> gary_poster, still around?