[00:24] <Makyo> jujugui lf quickstart review/qa https://codereview.appspot.com/39610049
[13:00] <bac> hi benji
[13:14] <gary_poster> I'm taking Makyo's quickstart review
[13:16] <rick_h__> gary_poster: cool thanks. Let me know if you want to reschedule our chat today sometime. 
[13:17] <gary_poster> cool rick_h__ .  definitely.  I'll drag something over on the calendar.  you're feeling much better, then?
[13:18] <rick_h__> gary_poster: yea, working on it. Letting them medicate me to hopefully avoid rotten holidays
[13:18] <gary_poster> heh, cool
[14:08] <bac> hey benji
[14:09] <bac> yoo whoo
[14:09] <benji> bac: it's been a while since I've done that
[14:09] <bac> benji: but its always amusing
[14:09] <benji> heh
[14:09] <bac> benji: i'm about to ask webops for an update to production for charmworld.  any reason i should wait, i.e. new code from you for the app?
[14:09] <benji> I guess I need to write an IRC client over Christmas that will keep me from doing that.
[14:10] <bac> s/i.e./e.g./
[14:10] <benji> nope, but if you know where they keep the production .ini file, I would like to know
[14:10] <benji> or we can ash whomever you get ahold of
[14:11] <bac> benji: is it not in the charm?
[14:12] <benji> that sounds like a question I should have asked myself
[14:13] <benji> bac: any hints as to where the charm lives?
[14:14] <benji> (the way charms are store in LP makes them hard for me to find)
[14:14] <bac> yes, they are a mess.  but you can search for charmworld on manage.jujucharms.com and click on 'repository'
[14:15] <bac> benji: otherwise you have to know who owns the given charm.
[14:15] <benji> how meta
[14:15] <benji> I suppose the ~juju-jitsu version is the official one.
[14:15] <bac> benji: i just happened to have it opne: https://code.launchpad.net/~juju-jitsu/charms/precise/charmworld/trunk
[14:15] <benji> thanks
[14:24] <hatch> so hows everyone doing this fine fine morning
[14:26] <bac> benji: charmworld upgrade deferred to monday.  i forgot the friday rule.
[14:26] <bac> and nothing is on fire
[14:28] <hatch> last night I decided to start learning lisp
[14:28] <hatch> not sure why...
[15:03] <frankban> hey benji: morning, any problems with my review?
[15:04] <benji> frankban: the only problem is that I forgot to do it, very sorry.  Can you send me a link?
[15:04] <frankban> benji: np, and thanks: https://codereview.appspot.com/41350043
[15:04] <benji> I'll look now.
[15:04] <frankban> cool
[15:08] <benji> rick_h__: I found what appears to generate the INI but it didn't answer my question.  I'm trying to figure out why a script I had run in production generated this error: https://pastebin.canonical.com/101921/
[15:09] <rick_h__> benji: so you have to set an INI= env var with the path to the INI file to use
[15:09] <rick_h__> benji: so I'd expect your command to be INI=/path/to/production.ini bin/python /tmp/xxxx
[15:10] <benji> could be, I'll try that next
[15:10] <rick_h__> benji: in the charmworld utils is a get_ini() function that tries to find a default or ENV specified ini file to work
[15:10] <benji> I'll observe that requiring carefully set environment variables for forware to run isn't such a great thing.
[15:11] <rick_h__> well, I think it defaults to a charmworld.ini if none is specified
[15:12] <rick_h__> and production uses production.ini vs charmword.ini which is why it needs the override
[15:12] <rick_h__> patches/ideas on improvement welcome. 
[15:12] <rick_h__> personally I wish it just overwrote charmworld.ini
[15:13] <rick_h__> but it makes it more difficult to test production vs dev locally
[15:22] <gary_poster> bac, thanks for looking at bug 1259704.  After you do that, or as a break if it is easy, could you look at updating comingsoon to use git?  There's a high priority card for that down in the CI swim lane.  +1 on doing it as cheaply as we've done the bzr integration, for now.
[15:22] <_mup_> Bug #1259704: Bundle branch command incorrect. <juju-gui:Triaged by bac> <https://launchpad.net/bugs/1259704>
[15:23] <bac> gary_poster: ok, i missed that one.  i'll be glad to do it now
[15:23] <gary_poster> thanks bac, either way
[15:37] <hatch> so there is an issue with us switching to git
[15:37] <hatch> we have to push our branches up to lp to test in a real environment
[15:37] <hatch> haha
[15:43] <rick_h__> hatch: just until the charm is updated
[15:43] <rick_h__> hatch: which is a card to be done soon
[15:43] <rick_h__> hatch: so hit me back next week
[15:45] <hatch> psssshhhhht
[15:45] <hatch> but I need it nooooooow
[15:45] <rick_h__> then go fix the charm nooooooow :P
[15:45] <rick_h__> you know how to hacky hack
[15:46] <hatch> but....but....but....
[15:46] <rick_h__> pssssssshhhht
[15:46] <hatch> http://www.digitalblasphemy.com/preview.shtml?i=shadedpathwinternight1
[15:47] <gary_poster> hatch, not true: http://jujugui.wordpress.com/2013/10/15/if-you-want-to-run-a-custom-gui/
[15:47] <rick_h__> oh right! gary_poster ftw
[15:47] <gary_poster> :-)
[15:48] <hatch> oooo hacky hacky
[15:48] <rick_h__> except then you might want my current branch that git-ifies that stuff. 
[15:48] <gary_poster> now we have to say NO_BZR BRANCH_IS_GOOD make distfile I bet hatch
[15:48]  * rick_h__ isn't sure
[15:48] <gary_poster> sorry "NO_BZR=1 BRANCH_IS_GOOD=1 make distfile"
[15:49] <gary_poster> yeah or get branch from rick_h__ 
[15:50] <gary_poster> jujugui call in 10
[15:50] <hatch> http://updates.html5rocks.com/2013/12/300ms-tap-delay-gone-away
[15:51] <rick_h__> hatch: yea, coolio
[15:51] <rick_h__> but watch out broken sites!
[15:52] <gary_poster> looks like it only affects double tap to zoom
[15:52] <gary_poster> negatively I mean
[15:54] <hatch> yeah the whole reasoning for it was to enable double tap support in the first place
[15:58] <gary_poster> jujugui call in 2
[16:55]  * gary_poster heads to lunch.  back later
[16:55] <gary_poster> fwiw http://jujugui.wordpress.com/2013/12/13/weekly-retrospective-notes-2/
[16:55] <rick_h__> bac: http://alblue.bandlem.com/2011/07/git-tip-of-week-tracking-branches.html looks like it might be decent for the tracking in git
[16:56] <rick_h__> jujugui ^ if you're interested in shortcutting the push/pull commands to not be as explicit and setup tracking branches kind of like Makyo was mentioning today
[16:57] <rick_h__> jujugui and I setup http://paste.ubuntu.com/6567622/ as a quick way to get around Makyo's question. Use that to sync with upstream and it'll keep your fork's develop in sync as well
[16:57] <rick_h__> I'll update the docs with that alias in a bit. 
[16:57] <bac> jujugui: comingsoon now using the git repo
[16:57] <gary_poster> yay! thanks
[16:57] <rick_h__> woot! thanks bac 
[16:58] <gary_poster> inspector is so much freaking faster.  yay
[16:58]  * gary_poster runs for real
[16:58] <rick_h__> I feel the need...for speed!
[16:59] <rick_h__> http://stackoverflow.com/questions/520650/how-do-you-make-an-existing-git-branch-track-a-remote-branch is the copy paste tracking post
[17:15] <rick_h__> hatch: k, free here. Let me know when you want to chat
[17:16] <benji> frankban: I finished your long-awaited review.  Your branch looks good.
[17:16] <benji> Lunch looks good too, so I'll do that next.
[17:17] <frankban> benji: cool, thanks
[17:20] <hatch> rick_h__ just grabbing something to eat, didn't have breakfast again hah - 15mins?
[17:20] <rick_h__> hatch: no problem
[17:45] <hatch> rick_h__ rdy?
[17:45] <hatch> https://plus.google.com/hangouts/_/72cpj5h9np735l8cvrlk97jsfc?hl=en
[18:04] <hatch> jujugui looking for a review/qa on the login system changes https://github.com/juju/juju-gui/pull/19
[18:10] <rick_h__> hatch: will look in a sec, lunch is nuking
[18:10] <hatch> rick_h__ ^ this branch also fixes the odd minimize issues
[18:10] <hatch> so plz test that too
[18:10] <rick_h__> hatch: ok coolio
[18:10] <hatch> oh wait nm
[18:10] <hatch> cache 
[18:10] <hatch> it doesn't fix it
[18:10] <hatch> lol
[18:10] <rick_h__> lol
[18:11] <rick_h__> ok, keep that card on the board then
[18:11] <hatch> I'm pretty confident it's separate from the url one
[18:12] <hatch> I'll have to do another bisect to find the offending diff
[18:12] <hatch> I'll do that now
[18:14] <rick_h__> cool, thanks
[18:21] <rick_h__> hatch: close, question and a tweak to the logic please
[18:22] <hatch> rick_h__ the featured issue has been here for -well- over a month and I still haven't found one which works properly so I'm going to create a real bug for it
[18:22] <hatch> looking
[18:22] <rick_h__> hatch: ?
[18:22] <rick_h__> which featured issue?
[18:23] <hatch> the minimize apache2 details one
[18:23] <rick_h__> oh, that interesting/reload thing?
[18:23] <rick_h__> wow, :/
[18:23] <hatch> yeah
[18:23]  * rick_h__ promises to do better at QA day
[18:23] <rick_h__> sorry to assume/stick that on you
[18:24] <hatch> no problem :)
[18:24] <rick_h__> yay git bisect
[18:24] <hatch> I'm not sure I agree with your comments on my branch
[18:25] <hatch> but I'm sure I can be easily convinced 
[18:25] <hatch> why would the credentials cookies be removed if the user hasn't logged out?
[18:25] <rick_h__> well, it's not trivial for things to get out of sync, but isLogged in should respect your credentials right?
[18:25] <rick_h__> hatch: well, my use case is manual dev testing. 
[18:26] <hatch> ok you convinced me
[18:26] <hatch> :)
[18:26] <rick_h__> but browser goes boom, something goes wrong, I'm not sure. It's more a 'correctness' issue that's easy to fix. In the check credentials stuff, if they're not set, then also make sure isLogged in is set to false
[18:26] <rick_h__> yay 
[18:30] <hatch> running the tests again before pushing that 1line fix
[18:30] <rick_h__> hatch: ty
[18:32] <hatch> rick_h__ pushed
[18:36] <rick_h__> hatch: so I don't get a login window
[18:36] <rick_h__> hatch: if I remove the session cookie, or I go logout, and go back to /
[18:37] <rick_h__> I get the onboarding and not a login screen
[18:38] <hatch> hmm
[18:38] <hatch> it's working fine here, lemme try clearing the cache again
[18:39] <rick_h__> yea, if I clear the cache, go to / I get the environment. I hit logout, get redirected to /login. If I manually change the url to /
[18:39] <rick_h__> I'm right back in
[18:40] <hatch> well wth make prod is broken now?
[18:40] <rick_h__> huh? I'm just using make devel
[18:41] <hatch> ohh well you're reloading the page then if you're in make devel
[18:41] <hatch> which will make you log back in
[18:41] <rick_h__> I expect a login box but I can't get one unless I go to /login
[18:42] <rick_h__> hatch: hangout/screenshare?
[18:42] <hatch> https://plus.google.com/hangouts/_/72cpj5h9np735l8cvrlk97jsfc?hl=en
[18:42] <hatch> bah
[18:42] <hatch> https://plus.google.com/hangouts/_/7ecpi6darhr73v0qj8gn8hhi9g?hl=en
[18:43] <gary_poster> rick_h__, replied to git update email.  Lemme know if you want me to send.
[18:44] <rick_h__> gary_poster: rgr
[18:44] <rick_h__> otp, sec
[18:44] <gary_poster> np
[18:55] <rick_h__> gary_poster: did we want to send it to anyone besides john?
[18:56] <gary_poster> rick_h__, minimally John, Curtis, and mramm.  If you feeling excited, cc canonical-juju, but that might be more expensive than we want. :-P
[18:56] <rick_h__> gary_poster: rgr, sounds good. 
[18:56] <gary_poster> thank you
[18:57] <hatch> rick_h__ https://bugs.launchpad.net/juju-gui/+bug/1260831
[18:57] <_mup_> Bug #1260831: Featured browser results shown instead of search results <juju-gui:New> <https://launchpad.net/bugs/1260831>
[18:59] <hatch> rick_h__ did you want that one? You're probably better suited to it
[19:00] <rick_h__> hatch: I can, not sure when I'll get to it atm as I'm trying to get CI finished up. 
[19:00] <rick_h__> hatch: but feel free to stick it on me for now
[19:00] <hatch> cool, card assigned
[19:00] <rick_h__> thanks
[19:00] <hatch> it's been there for a long time just want to get as many bugs fixed before 1.0 :)
[19:01] <rick_h__> hatch: rgr, I'll get to it next week for sure. It might even be an easy fix once it's traced
[19:01] <rick_h__> the tracing is the pita part
[19:01] <hatch> imagine how simple all that will be once double dispatch is gone
[19:01] <rick_h__> shush, you want to talk about a lot of code to work around something
[19:02]  * Makyo sneaks to car+shop thing
[19:02] <rick_h__> you've already kidded hours of my worklife :P
[19:02] <hatch> haha 
[19:02] <rick_h__> "I don't get why rick complained about this stuff...it's super easy now"
[19:02] <rick_h__> - fullscreen, - double dispatch
[19:03] <gary_poster> "kidded" -> "killed" ?
[19:03] <rick_h__> umm yea
[19:03] <rick_h__> drugs!
[19:03] <rick_h__> :)
[19:03] <gary_poster> heh
[19:05] <gary_poster> My 100th jujugui blog post celebrate the speed improvements from Makyo: http://jujugui.wordpress.com/2013/12/13/great-speed-improvements/
[19:05] <rick_h__> woot! nice
[19:05] <benji> cool
[19:05] <hatch> yay!!
[19:05] <gary_poster> I'll make a git post after the release stuff has landed
[19:06] <gary_poster> And looking forward to a "feature complete" post for quickstart next week :-)
[19:06] <hatch> I tried to go buy a MBP yesterday, noone had the model I wanted in stock :(
[19:06] <hatch> a sign?
[19:07] <gary_poster> heh
[19:07] <gary_poster> yeah, a sign to order online
[19:07] <hatch> I want to buy local so if there is an issue I can return local :D
[19:07] <hatch> it's expensive enough I want to be able to.....talk.....to someone face to face if there is an issue
[19:07] <hatch> :P
[19:23] <rick_h__> running for coffee to try to go a bit long today. brb
[19:31] <hatch> if it exists, it will be written in javascript http://coolwanglu.github.io/vim.js/web/vim.html
[19:32] <hatch> ^ rick_h__  benji 
[19:36] <benji> hatch: heh, yeah I saw that earlier, strange days
[19:36] <hatch> it opens up some cool opportunities though
[19:37] <hatch> not sure how they are worth the work over just ssh'ing into the machine...buuuut
[19:37] <hatch> haha
[19:38] <bac> jujugui: what's the equiv of 'bzr info'?  i'm looking for the name of my feature branch.
[19:38] <bac> (which i found in my bash history.  but how does one find it via git?)
[19:38] <gary_poster> git branch
[19:39] <bac> well, isn't that easy
[19:39] <hatch> bac PS1 config to show it in your prompt https://gist.github.com/hatched/7826871
[19:40] <hatch> hatch@beep:~/canonical/juju/juju-gui (fix-login-redirect) :
[19:40] <hatch> for example
[19:40] <bac> hatch: cool.  too many \ though
[19:42] <hatch> bac yeah? I know almost thing about sed - I made it by merging a number of PS1 examples :)
[19:42] <rick_h__> bac: git branch will show you the branch you're on with a *
[19:43] <rick_h__> and you already got told that...finish reading the scrollback rick
[19:45] <hatch> yeah rick!
[19:45] <gary_poster> uh oh, he's getting cranky with himself now!
[19:45] <rick_h__> lol
[19:45] <bac> so, what's the diff between 'git checkout' vs 'git branch' if they both create branches?
[19:45] <marcoceppi> hey party people, can I get some help in #juju?
[19:45] <hatch> bac oh so you need to 'make check' before pushing
[19:45] <rick_h__> bac: heads up, lint failure on your pull request. 
[19:46] <hatch> that's why your ci failed
[19:46] <rick_h__> marcoceppi: maybe
[19:46] <marcoceppi> rick_h__: it's about charmworld, and ingesting, and fun stuff like that
[19:46] <gary_poster> benji, #juju looks like it is related to what you are looking at
[19:46] <gary_poster> marcoceppi, rick_h__ ^^^
[19:46] <rick_h__> bac: http://stackoverflow.com/questions/7987687/what-is-the-difference-between-git-branch-and-git-checkout-b
[19:46]  * benji looks
[19:47] <rick_h__> oh, have we ever added trusty to the gui or api for support?
[19:48] <marcoceppi> rick_h__: well it seems to be in the API
[19:48] <rick_h__> marcoceppi: k, it might not be all the way through the UX. There's a few places where it's filtered in charmworld and in the gui I believe. 
[19:49] <gary_poster> app/models/browser.js does not have saucy or trusty but that's the only places where those strings are & I haven't seen how we use it yet...
[19:50] <gary_poster> app/widgets/filter.js ...
[19:50] <rick_h__> gary_poster: yea, but charmworld might need some <3 there as well. It doesn't show trust charms by default. Though they're indexed for search as a couple do pull up via search. http://manage.jujucharms.com/~xnox/trusty
[19:50] <hatch> bac I can do the review/qa once you get ci passing
[19:50] <gary_poster> which is in a widget we don't use AFAIK
[19:51] <gary_poster> rick_h__, yeah, my first guess would be in charmworld
[19:51] <rick_h__> gary_poster: but yea, we should make cards to audit trusty all the way through charmworld -> gui
[19:51] <gary_poster> ack.  on it
[19:51] <rick_h__> there's a filter object in the browser.js models that we use for doing api calls
[19:51] <rick_h__> so it might need a small one liner in there to add trusty 
[19:52] <gary_poster> rick_h__, see the scrollback, luke ;-)
[19:52] <rick_h__> gary_poster: yea, I'm just agreeing with it :)
[19:52] <gary_poster> ah ok
[19:52] <gary_poster> ]but that is only used in that filter widget which is hidden at best
[19:53] <rick_h__> gary_poster: no, the filter is used in the main browser I believe and passed to the store for making calls
[19:53] <gary_poster> rick_h__, btw good progress with Antonio and team account
[19:53] <rick_h__> it's how categories work and such
[19:53] <gary_poster> oh ok
[19:53] <rick_h__> woot, oh boy, I get to recreate jenkins! 
[19:53] <gary_poster> :-P
[19:53] <gary_poster> we can make someone else to do it to try out your instructions!
[19:53] <gary_poster> valuable
[19:53] <rick_h__> hah, they're not that step by step tbh
[19:53] <gary_poster> :-)
[19:54] <bac> hatch: ok, i fixed lint and pushed a new revision.  did that update the pull request?  i don't see the diff anymore on the pull request page.
[19:54] <rick_h__> I'm trying to avoid getting others to hate me remember 
[19:54] <rick_h__> bac: yep updated and tests are marked as passing now
[19:54] <bac> thx
[19:55] <gary_poster> heh
[19:55] <rick_h__> bac: just make sure before you submit for $$merge$$ you git rebase --autosqaush and squash the lint commit
[19:55] <rick_h__> bac: let me know if you need a walk through on it for the first time
[19:55] <rick_h__> it's the one doc I've not added yet. 
[19:55] <bac> i will
[19:56] <gary_poster> oh, didn't know that was necessary
[19:56] <hatch> bac on it
[19:56] <hatch> might take me a bit, lunch just arrived
[19:58] <gary_poster> https://jujucharms.com/fullscreen/search/~xnox/trusty/sbuild-0/?text=sbuild but not in search
[19:58] <gary_poster> https://jujucharms.com/fullscreen/search/?text=sbuild
[20:00] <rick_h__> gary_poster: yea, sounds like charmworld needs something added to the search api to pick up the extra series
[20:00] <gary_poster> trying the filter thing really quick beofre I write email #500076889758
[20:00] <rick_h__> https://manage.jujucharms.com/api/3/search?text=trusty returns them, so it's clearly not checking the series field
[20:01] <rick_h__> oh wait, they are in the sbuild search. Yep, we're filtering them heh, sucky me. https://manage.jujucharms.com/api/3/search?text=sbuild
[20:02] <bac> rick_h__: so if i do a review i can just say "looks nice $$merge$$" and bob's your uncle?
[20:02] <rick_h__> bac: rgr
[20:03] <rick_h__> then 4min later it should show up in trunk/develop
[20:03] <bac> well that sure seems nice
[20:03] <rick_h__> hopefully :)
[20:04] <gary_poster> rick_h__, no I think this was more ingestion time
[20:04] <gary_poster> http://comingsoon.jujucharms.com/sidebar/search/?text=sbuild
[20:04] <gary_poster> it is there now
[20:04] <rick_h__> gary_poster: oh, cool then
[20:04] <gary_poster> elasticsearch was the last to hear about it, I suspect
[20:04] <rick_h__> even better
[20:04] <gary_poster> yeah
[20:11] <rick_h__> gary_poster: got a sec?
[20:12] <gary_poster> sure rick_h__ .  hangout?
[20:12] <rick_h__> gary_poster: sure
[20:12] <gary_poster> https://plus.google.com/hangouts/_/72cpis84sht2ofk1388nj78i4k?hl=en
[20:19] <hatch> bac all done lgtm'd
[20:20] <bac> hatch:
[20:20] <bac> hatch: ok, so i need to rebase and then do my own $$merge$$ comment?
[20:21] <hatch> well first I'd like to see some changes made :P
[20:21] <hatch> but yes you would then rebase your local branch, push the changes up
[20:21] <bac> wait, does $$merge$$ go in a pull request comment or a commit message?  latter, right?
[20:21] <hatch> then add the $$merge$$ comment
[20:21] <hatch> it goes as a comment in the pull request
[20:22] <hatch> see https://github.com/juju/juju-gui/pull/18
[20:22] <hatch> for example
[20:25] <bac> rt.  thanks.
[20:29] <hatch> rick_h__ have you had a chance to qa my branch on a real env yet?
[20:30] <rick_h__> hatch: no, I was actually working on make distfile in my own branch and figured I'd see if I could get it to work from there but hit an issue
[20:30] <hatch> ohh ok cool cool
[20:30] <rick_h__> honestly, this is kind of complex and I want to stick through this before going off on the live qa. I'll try to get to it in a few
[20:31] <rick_h__> if you can find someone else that's cool, but I don't want to lose track of this makefile stuff over the weekend if I can
[20:31] <bac> wow, i like how github lets you delete reviewer comments you find annoying.  nice!
[20:31] <rick_h__> lol
[20:31] <hatch> hahaha
[20:31] <hatch> bac has removed all of my comments
[20:32] <bac> justSomeOfThem
[20:32]  * rick_h__ chuckles
[20:32] <gary_poster> lol
[20:36] <hatch> rick_h__ if you git rebase --autosquash and then push up to the PR does it remove all of the previous commits? or do you need to force push?
[20:37] <rick_h__> hatch: you have to force push or it won't push
[20:37] <rick_h__> git push --force origin {branchname}
[20:37] <hatch> is that a good idea to do? because then it gets rid of all of the PR history?
[20:37] <hatch> shoudl we maybe say only PR after you have rebased, then after that we just deal with it?
[20:37] <rick_h__> hatch: well, it's still there. If you look at the comments you can view the original context they were applied against
[20:37] <rick_h__> hatch: meh, I don't see why. The branch goes away anyway
[20:38] <rick_h__> and normally those changes are "fix lint"
[20:38] <rick_h__> it's not that much of a history to lose imo
[20:38] <rick_h__> and worth it for a cleaner reading develop line
[20:38] <hatch> well if you force push the comments will have nothing to reference because you rewrote history
[20:38] <rick_h__> hatch: right, but github gives you the option to look at the original history
[20:38] <hatch> oh does it? cool
[20:38]  * rick_h__ goes to find one, knows he saw it
[20:38] <rick_h__> https://github.com/juju/juju-gui/pull/20
[20:39] <rick_h__> hatch: go down to your comment with the 'show outdated diff' links
[20:39] <hatch> right but the previous commits are still there
[20:39] <hatch> they are just outdated
[20:40] <rick_h__> I thought it did the same thing with force/squashed, I'll look for another example
[20:40] <rick_h__> https://github.com/juju/juju-gui/pull/10
[20:40] <rick_h__> hatch: ^^
[20:40] <rick_h__> hatch: notice one commit, but there's an outdated diff reference
[20:40] <rick_h__> hatch: I mean rebasing doesn't get rid of the commit. reflog ftw
[20:40]  * gary_poster just sent a super, super, way important message to peeps.
[20:41] <rick_h__> it just moves it out of the tree of commits that make up the history path
[20:41] <hatch> rick_h__ ohh I see what's going on now
[20:41] <hatch> I thought rebase in this context was actually destroying the commits
[20:42] <hatch> gary_poster reading
[20:42] <rick_h__> gary_poster: +1 on the :shipit:. I'll change the jenkins box if someone will change the docs
[20:42] <gary_poster> rick_h__, awesome. :-) I'll do docs, since I'm the perpetrator
[20:42] <gary_poster> rick_h__, we can wait to see if anyone throws things at me first though
[20:42] <rick_h__> gary_poster: changed
[20:42] <rick_h__> :)
[20:42] <gary_poster> rick_h__, lol
[20:42] <rick_h__> bac: ^^
[20:42] <gary_poster> oh ok
[20:43]  * gary_poster makes branch for docs
[20:43] <hatch> lol!
[20:43] <hatch> most superest importantest email eva
[20:43] <gary_poster> :-)
[20:49] <bac> hatch: did you do qa?  am i read to grey squirrel?
[20:50] <gary_poster> lol
[20:50] <hatch> haha gary_poster https://www.evernote.com/shard/s219/sh/1f0d13b7-e5bf-408a-8e41-f0b10aacd61f/9689e4b586a251a1b245ddf6105bb32a
[20:50] <hatch> bac yep I said QA OK!
[20:50] <gary_poster> hatch, oh yeah :-)
[20:50] <bac> oh didn't see.  ty.
[20:50] <hatch> bac did you autosquash?
[20:50] <hatch> I'm curious to see that in action
[20:51] <bac> i tried
[20:51] <hatch> it doesn't work because you already pushed right?
[20:51] <bac> hatch: sorry, i was looking on the files tab not the conversation tab so i didn't see it.
[20:51] <bac> why does it not work?
[20:52] <hatch> ohh nm I think we are just confusing eachother
[20:52] <hatch> :)
[20:52] <bac> hatch: i did this:
[20:52] <bac> git rebase -i --autosquash
[20:52] <bac> git push origin fix-branch-reference
[20:52] <bac> should be right, no?
[20:53] <rick_h__> bac: when you did the rebase you changed the lint/fixer-upper commits to s ?
[20:53] <bac> s?
[20:53] <rick_h__> yea, hangout?
[20:53] <bac> ffs
[20:53] <bac> sure
[20:53] <gary_poster> fast review of https://github.com/juju/juju-gui/pull/21 ?
[20:53]  * bac no steroids
[20:53] <hatch> can I join too?
[20:54] <gary_poster> no qa
[20:54] <rick_h__> we can not hangout :/
[20:54] <hatch> link me up!
[20:54] <hatch> :/
[20:54] <gary_poster> :\
[20:54] <rick_h__> https://plus.google.com/hangouts/_/72cpjheea3948s8nfq85l9gh40?hl=en
[20:54] <hatch> (╯°□°）╯︵ ┻━┻
[20:54] <gary_poster> lol
[20:54] <hatch> haha I love that
[20:55]  * gary_poster inclined to shipit without review. trivial
[20:55] <gary_poster> doing so.
[20:56] <hatch> lgtm'd
[20:56] <gary_poster> ty :-)
[21:00] <gary_poster> jujugui, if you don't see email or HACKING doc change, use :shipit: now instead of $$merge$$
[21:02] <rick_h__> jujugui release process branch for review please. Note the *disclaimers* in the notes. https://github.com/juju/juju-gui/pull/22
[21:02] <hatch> oh it
[21:02] <hatch> n
[21:02] <gary_poster> I'll take it
[21:03] <gary_poster> ok, two reviews probably wise
[21:04] <rick_h__> yea, appreciate all the eyeballs on it please. Though I still ack that I'll probably want a tweak/two when we do the real release. I didn't want to go all the way atm
[21:07] <gary_poster> rick_h__, do I understand correctly that the tag will be on master but not develop branch?
[21:08] <rick_h__> gary_poster: it will be made from the sha of the commit on master after the merge from develop
[21:08] <rick_h__> gary_poster: but it will be visible from anywhere via git tag
[21:08] <gary_poster> ok, that sounds good, and hopefully I'll see how it works when I try it myself
[21:09] <rick_h__> yep
[21:09] <rick_h__> you must experience the joy...wheeeee
[21:12] <gary_poster> heh
[21:14] <Makyo> jujugui python question re: __future__.print_function
[21:15]  * bac eod.  have a nice weekend.
[21:15] <hatch> cya bac enjoy!
[21:16] <rick_h__> bac: you need to publicize your juju membership before you can land
[21:16] <rick_h__> bac: https://github.com/juju?tab=members and publicize 
[21:18] <hatch> flyer person is driving door to door to deliver the flyers....
[21:18] <bac> rick_h__: so now i just :shipit: again?
[21:19] <bac> hatch: are the flyers for some environmental org?
[21:19] <gary_poster> have a nice weekend bac
[21:19] <bac> gary_poster: oh, hell, no.  not until this thing lands.
[21:19]  * benji waits with bated breath for Makyo's question.
[21:19] <hatch> bac haha no, they are for all of the big box stores in the city....very anti environment
[21:19] <gary_poster> Makyo, you never said the question :-)
[21:20] <hatch> lol
[21:20] <Makyo> Sorry, was getting the gist ready: https://gist.github.com/makyo/7951503
[21:20] <Makyo> gary_poster, benji ^^^
[21:21] <Makyo> With the end kwarg, nothing gets printed until the loop breaks, then all the dots get printed at once.  Without, it prints one per second, but on new lines.
[21:21] <benji> Makyo: flush
[21:21] <Makyo> Aha!
[21:21] <benji> (or run your python unbuffered (python -u))
[21:22] <rick_h__> bac: it should pick it up on the next cron run
[21:22] <rick_h__> we just don't allow anyone not in the org to merge
[21:23] <Makyo> benji, cool, thanks.
[21:23] <benji> (or reopen stdout as unbuffered, or use the PYTHONUNBUFFERED environment variable)
[21:23] <rick_h__> bac: landed, thanks!
[21:29] <rick_h__> no one's ever allowed to edit the hacking doc again. It'll never have a better line https://github.com/juju/juju-gui
[21:33] <gary_poster> :-)
[21:33] <benji> heh
[21:33] <benji> this gives me so many ideas for commit messages
[21:33] <gary_poster> rick_h__, changes lgtm with comments.  trying qa now.
[21:34] <rick_h__> gary_poster: thanks
[21:37] <hatch> rick_h__ you aren't allowed to EOD until you finish qa'ing my branch
[21:37] <hatch> new rule
[21:37] <hatch> house rules if you will
[21:37] <hatch> :P
[21:37] <rick_h__> hatch: :P drugs!
[21:37] <hatch> lol
[21:37] <rick_h__> sorry man, been crazy today. 
[21:38] <hatch> yeah it's fine, I'd just like to land it before my EOD as well :)
[21:38] <rick_h__> k, working on it
[21:38] <hatch> wel tbh it could be any time, I'm not going anywhere
[21:38] <rick_h__> well I want to get out of here. Running low today
[21:40] <hatch> yeah that's understandable 
[21:43] <rick_h__> running make deploy now
[21:46] <gary_poster> so nice to not see the bzr complaints any more
[21:47] <Makyo> jujugui quickstart re-review/qa: https://codereview.appspot.com/39610049 Have to use ssh-agent for juju-core to recognize the keys.
[21:47] <Makyo> ^^^ includes changes to autogenerating keys.
[21:47] <gary_poster> will look asap if no-one else does.  tryingto finish rick review
[21:47] <Makyo> No worries.
[21:47] <Makyo> I need to pick up a prescription now that I finally have my car back. Shouldn't be too long
[21:52] <rick_h__> bah, error state...
[21:54] <gary_poster> rick_h__, one ignorable qa bit so far: I expected "IS_TRUNK_BRANCH=1 make distfile" to work in dev branch.  Can't figure out why yet, because the other var is fine.  However, BRANCH_IS_GOOD=1 make distfile works great.  Oh!
[21:54] <gary_poster> you changed the logic to check 0
[21:54] <gary_poster> rather than definition
[21:54] <rick_h__> gary_poster: yes, I couldn't get the length stuff to work right
[21:54] <gary_poster> ok, that's alright
[21:54] <rick_h__> and I hit that and it was confusing to me as well, I tried the same thing
[21:55] <rick_h__> but this seemed a simpler check. did it return 0 or 1 where 0 is a good code and 1 is a bad return code
[21:56] <gary_poster> rick_h__, IS_TRUNK_BRANCH=0 make distfile works :-) so if we cared we could say ifdef IS_TRUNK_BRANCH; IS_TRUNK_BRANCH=0; endif
[21:56] <gary_poster> and same for BRANCH_IS_CLEAN
[21:56] <gary_poster> but eh
[21:56] <gary_poster> although
[21:56] <gary_poster> if we don't do that
[21:56] <gary_poster> we should fix up the HACKING doc
[21:57] <gary_poster> rick_h__, I'd say the ifdef hack is the easier of the two
[21:57] <rick_h__> gary_poster: oic, so map the 1 to a 0 for the sake of getting the UX consistent with the actual check
[21:57] <gary_poster> right
[21:58] <rick_h__> ok, I'm game
[21:58] <gary_poster> rick_h__, or actually changing hacking might be better
[21:58] <gary_poster> rick_h__, <shrug> you choose :-)
[21:58] <rick_h__> hacking? or process.rst?
[21:58] <gary_poster> hacking
[21:58] <rick_h__> oh, missed it was in hacking
[21:59] <rick_h__> k, yea that'll be easier I think and less confusing to parse
[21:59] <gary_poster> cool +1
[21:59] <gary_poster> rick_h__, I thnk given that your disclaimers, QA good.  the rubber will hit the road next week, but looks good so far
[22:00] <rick_h__> and I'll change BRANCH_IS_GOOD to 0 as well
[22:00] <gary_poster> heh
[22:00] <gary_poster> ok
[22:00] <rick_h__> just for the sake of it looking the same :)
[22:00] <rick_h__> gary_poster: cool, thanks for walking through it. I think it's in the ballpark 
[22:01] <gary_poster> definitely
[22:01] <gary_poster> :+1: done
[22:02] <rick_h__> hatch: did you get this release dir thing to work?
[22:03] <hatch> having issues?
[22:03] <rick_h__> hatch: yea, it keeps failing on install saying it can't find a release
[22:03] <hatch> I used NO_BZR=1 or something
[22:03] <hatch> I can't remember exactly
[22:03] <hatch> umm
[22:04] <rick_h__> ah, I missed that when I did the distfile
[22:08] <gary_poster> oof, switch between bzr and git is going to make me dizzy
[22:08] <rick_h__> +1
[22:08] <rick_h__> I forsee us moving more stuff just to keep sanity
[22:09] <gary_poster> the thing is that quickstart and charm both have to be in LP because of machinery (PPA and charm store, respectively) :-/
[22:09] <gary_poster> we could do import
[22:09] <gary_poster> if that's even working still
[22:10] <gary_poster> so git is source of truth and LP regularly imports it
[22:10] <rick_h__> I hear people do it, but not sure who is tbh. 
[22:11] <rick_h__> but yea, another level of stuff to deal with
[22:11] <rick_h__> hatch: I'm working on getting it to deploy, but I've got to get our of here and get the boy from day care. I'll keep working on this qa when I get back/tonight. 
[22:12] <hatch> rick_h__ sure np just update the PR and I'll get back to it whenever
[22:12] <rick_h__> hatch: if it's good I'll just shipit for you. I think you've addressed anything I had in code
[22:14] <rick_h__> ah crap, it deployed to ec2
[22:14] <rick_h__> heh, oh well
[22:14] <hatch> ok cool that works too
[22:14] <rick_h__> looks good, will land
[22:17] <gary_poster> Makyo, got same error as before ("juju-quickstart: error: ERROR error parsing environment "ec2": no public ssh keys found").  I need to go soon.  do you want to do a quick hangout with me to diagnose, or get someone else to look at it?
[22:19] <gary_poster> Mm, I think I know why.
[22:21] <gary_poster> nope. :-/
[22:26] <gary_poster> still think I might know
[22:35] <hatch> oh ricks branch got landed
[22:35] <hatch> lol
[22:35] <hatch> WELL THEN
[22:35] <hatch> lgtm
[22:35] <hatch> :P
[22:53] <gary_poster> Makyo, gave some hints.  Must run
[22:53] <gary_poster> have a good weekend all!
[23:13] <rick_h__> hatch: heh yea sorry. gary_poster|away went through it pretty well and I needed to get out
[23:34] <Makyo> Damn.
[23:41] <rick_h__> Makyo: hey, there has been a lot of agent/key mgt commits to juju-core lately I've seen go by
[23:42] <rick_h__> any of those help you with how they're managing some of this stuff?
[23:42] <Makyo> rick_h__, yeah, that's the next step - seeing how core does it.
[23:42] <rick_h__> cool, yea there's been a bunch of commits over this week that seemed like it might be useful. 
[23:42] <Makyo> rick_h__, The code I have is still valid, just for passphrase-less keys.
[23:42] <rick_h__> gotcha
[23:42] <Makyo> rick_h__, will check it out, thanks.
[23:42] <rick_h__> good luck!
[23:48] <Makyo> Damn again.  Compounding pharmacy didn't actually compound anything - same stupid prescription.  Waste :/