=== rogpeppe is now known as rogpeppe1 [13:12] morning [13:12] party [13:13] hatch: does that mean I have a +1 for the docs update? [13:13] or more notes inbound? [13:13] oh....woops I meant to add the +1 to the initial comment :) [13:14] I don't use the aliases anyways so I'm a horrible person to hold up changes to them :) [13:15] heh that's ok [13:15] I want to get all opinions and such [13:15] but hope this will help the immediate issues for everyone [13:20] yep it's a good thing to have in there for sure [14:12] Can charms have multiple maintainers? and if so, what is the definition style in metadata.yaml for said multiple maintainership declaration? [14:14] lazyPower I'm pretty sure, no [14:14] although that's definitely something that we should have [14:14] lazyPower we need to propose to the list, get concensous on how to do it (shouldn't be hard) then have that made known to the teams and have them implement it as needed [14:14] filing a bug [14:15] would that go against -gui or -core? assuming -gui [14:16] well... [14:16] lazyPower: charmworld and gui probably [14:16] charmworld needs to be able to parse it [14:16] too [14:16] yeah [14:16] ack. [14:16] I'll tack that on to both and reference [14:26] ugh osx is such a steaming pile [14:30] yay local charm upgrade works on a real env [14:31] rick_h_ this is a much better UI than dragging on the inspector [14:31] +1 [14:33] hey hatch, can you give me some git help? [14:34] bac I can try :) [14:34] shoot [14:34] hatch: i have a branch 'constraint-export'. i pushed it to github and the commit shows up in my 'develop' branch up there [14:34] hatch: i want to create a pull request [14:34] can you link me your fork? [14:34] just so I can see the repo structure [14:35] hatch: https://github.com/bac/juju-gui/tree/develop [14:36] bac so you need to do a PR from a branch [14:36] oh that's not entirely true [14:36] you can specify a commit....but... [14:36] when you do `git branch` does it show your branch locally? [14:36] hatch: if i try to do a pull request it shows changes from other contributors going back to dec-12 [14:36] so is it not updated to the project tip correctly? [14:37] hatch: yes, git branch shows this branch, develop, and two others [14:37] lets have a hangout [14:37] k [14:37] https://plus.google.com/hangouts/_/7acpi3t6o8bq2dmt0gsebvqj9s?hl=en [14:42] hatch: cool on the upgrade. I sent luca__ a screenshot of the deploy/upgrade inspector in our call today and they'll look at it in the next week or so. [14:46] lazyPower: hatch yea, we'll have to update charmworld/gui to handle it but nothing crazy to support. [14:47] +1 i've got an open thread on the mailing list for discussion. [14:47] lazyPower: yep, tracking [14:47] with a glorious typo in the subject line [14:47] * lazyPower facepalms [14:47] frakkin mondays [14:50] frankban: what version of ubuntu are you using for your GUI development? [14:50] benji: saucy [14:51] trusty here with a saucy lxc for charmworld [14:51] thanks; I'm still on raring and the latest nodejs package isn't available for it [14:51] rick_h_ have a second to help bac with some git issues? [14:51] hatch: yep [14:51] hangout above? [14:51] https://plus.google.com/hangouts/_/7acpi3t6o8bq2dmt0gsebvqj9s?hl=en [14:51] yeah, saucy host + saucy LXC for the GUI [14:51] rick_h_: yeah, I think I'm going to go straight to trusty [14:52] I really didn't want to do an OS upgrade today [15:15] benji: heh, welcome back...here's a ton of packages! [15:15] wb benji [15:15] :) [15:23] hey rick_h_, when i pushed it went to my develop branch on github again. http://paste.ubuntu.com/7028047/ [15:24] rick_h_: should i have had a colon in that command, as shown further down in the hacking doc? [15:28] bac: can you paste your .git/conf? [15:28] .git/config [15:29] rick_h_: http://paste.ubuntu.com/7028077/ [15:29] bac: ok, so in there your merge = refs/heads/develop is setup to push to develop [15:29] so now the question is why is that set that way. [15:29] how did you create hte feature branch? [15:29] guihelp: I need two review + QA for https://github.com/juju/juju-gui/pull/152 anyone available? thanks! [15:30] frankban I'll take one once I finish my tests [15:30] ~20mins [15:30] rick_h_: here is all that i did http://paste.ubuntu.com/7028092/ [15:31] hatch: thank you [15:31] bac: k doing a local test in a new directory here as well [15:34] bac: I dupe the issue here, looking into it [15:45] oops [15:45] here we are thinking everyone else is running the commands wrong [15:45] when the commands were wrong :) [15:46] hatch: this is different [15:46] hatch: and at one time worked. I'm wondering if something changed in git in trusty or something [15:46] ohh that might be it [15:47] git version 1.8.1.2 [15:47] ^ that's what I'm using fyi [15:47] hatch: I've been doig git push origin $newbranch for years now without issue [15:47] suddently that's going to the remote develop vs tracking a new branch on the remote [15:47] hmm i've been always doing it long handed [15:47] yeah [15:48] long hand being git push origin branch:branc [15:48] h ? [15:48] git push origin upgrade-from-local-charm:refs/heads/branch_to_create [15:48] yeah [15:48] ugh [15:48] but now I use grb so I don't need to do that anymore [15:48] just `grb create foo` [15:48] until git changes something that breaks grb :P [15:49] rick_h_: as i said, trusty has git 1.9.0 [15:49] well it does everything -very- explicitly.....but now that it's no longer maintained - that is a very real possibility haha [15:49] bac: yes looking. I can dupe, will try to get an update once I figure out what's changed/wrong [15:50] jujugui call in 10 [15:51] git push origin master or more generally any git push $where $what has been enhanced to use the same ref mapping git push origin would use to determine which ref at the origin is to be updated to the local master. For example, with this configuration: [15:51] so git push origin xxx got some changes [15:51] http://blogs.atlassian.com/2014/02/whats-new-git-1-9/ [15:53] bah, so the 'enhancement' brreaks how we set it up to be used lovely [15:53] bac: sorry, you're the first one to hit 1.9 and find the change [15:53] red shirt [15:54] rick_h_: nice find, though. now we know what's happened [15:54] yea, I'll update the docs and mention it on the call [15:54] you have to be more verbose [15:54] git push origin branchname:branchname [15:54] should tab complete at least [15:54] jujugui lf a review/qa on real env https://github.com/juju/juju-gui/pull/153 [15:54] at least it does in zsh [15:55] hurrah for DRY [15:55] hurrah for DRY [15:55] no kidding [15:55] love how they call it 'enhanced' and all I can think is "more explicit" [15:55] rick_h_: so my develop on github is screwed again, so i need to start over? [15:56] rick_h_ just fyi - tab complete does not work on osx unless you have installed an extra package [15:56] el stupido limitationo [15:57] hatch: do you use the github tool on os x? [15:57] hatch: sucks to be osc [15:57] osx [15:57] bac I don't... tbh I didn't even know there was one :) [15:58] hmm, I don't see the daily meeting on my calendar any more [15:58] there is and it looks nice. i haven't played with it much since i don't develop on os x [15:58] oh, there it is [15:58] jujugui call in 2 [15:58] I wont' be dev'ing in osx either once rick_h_ confirms they fixed the kernel issue for osx [15:59] er apple [15:59] my ethernet adapter is on the way [15:59] yay [16:13] rick_h_: is there any easier way for me to get github untangled than killing everything and starting over? [16:14] bac: 'easier' if you have a better feel what you're doing. If you don't I worry it'll be messier [16:14] er, ok [16:15] bac: if you want to chase the other way I can walk you through it :) [16:15] nah [16:28] benji did you do a qa on frankban's branch? [16:28] hatch: nope; shall I? [16:28] I can, I just didn't want both of us to do it :) [16:29] k [16:29] frankban the qa can be done on lxc right? [16:29] hatch: no, it requires ec2 and juju-core trunk [16:29] hatch: core does not support lxc inside lxc [16:29] ohh ok [16:30] frankban do you know if 1.17.4 will work? [16:31] hatch: I don't know for sure [16:31] ok np thanks, I'll get on the qa now [16:31] do you want me to shipit, if everything is good? [16:32] hatch: sure thanks [16:33] Currently Playing: Good Vibrations (feat. Loleatta Holloway) - Marky Mark and the Funky Bunch [16:33] aww yeah [16:33] lol [16:34] 90s ftw [16:34] never been so glad to work from home [16:34] rofl [16:36] :-) [16:36] I can see frankban rockin out to some Marky Mark [16:37] rofl [16:37] oh man the 90s [16:38] So. If ever you need to move your VMs, make sure to try and figure out vboxmanage (or at the very least kill VBoxSVC before you edit VirtualBox.xml) [16:41] uh oh [16:41] did you break something? [16:49] just in case you want to take your laptop on your river rapid tour http://www.stormtech.ca/wlc-1-stormtech-waterproof-laptop-carrier.html lol [16:50] Yeah, but I fixed it :P Just note that if you have the VBoxSVC running, all of your changes to your config file get overwritten as soon as you start virtualbox again. [16:50] that's interesting [16:50] jujugui: simple pull request: https://github.com/juju/juju-gui/pull/155 for one review / QA [16:54] bac: on it [17:01] bac: when I export the bundle I see comma separated constraints [17:36] Makyo in my branch review rick_h_ asked about a bug with upgrade charm ui listing 'old' charms which don't exist....do we have a bug for that? I thought so but I can't seem to find it [17:37] hatch, yes. Someone made a non-linear version thingy. [17:37] One sec. [17:37] hatch: if you find that can you make a card in the maint. backlog for it as well please? [17:38] you bet [17:38] Makyo: hatch luca is going to look at upgrade UX in the next week or two as well :) [17:40] awesome [17:40] hatch, https://bugs.launchpad.net/juju-gui/+bug/1246928 [17:40] <_mup_> Bug #1246928: Do not offer to upgrade/downgrade local charms [17:42] frankban you still around? [17:42] Makyo thanks! [17:42] hatch: yes [17:43] frankban when running app.env.deploy('cs:precise/wordpress-21', 'wordpress', null, null, 1, null, '1/lxc/0'); I get an error in the console which is.... [17:43] Relation added without matching service [17:43] models.js:818 [17:43] hatch: is the service deployed? [17:44] it is deploying..... [17:44] sec checking status [17:44] https://gist.github.com/hatched/9330381 [17:44] for the wordpress machine [17:45] hatch: if so, that's an error in the stream handlers, maybe we'll have to sort the messages by entity. [17:45] ok so is that a qa issue for this branch? [17:45] hatch: this is not related to the branch, could you please file a bug with the stacktrace and steps to dupe? [17:46] sure [17:46] thanks [17:53] frankban https://bugs.launchpad.net/juju-gui/+bug/1287283 just fyi [17:53] <_mup_> Bug #1287283: deploying service to lxc container throws relation error [17:55] frankban +1'd [17:56] hatch: thanks [17:58] sorry frankban, a set of changes must've got lost in my github shenanigans. [17:58] rick_h_ should this go in the Pool or On deck? [17:58] bac: np, I only see constraints parsing related code indeed [17:58] I'm guessing pool? [17:58] hatch: stick it in the on deck atm and I'll look otp [17:58] not entirely sure what the difference is [17:59] on deck is the limited stuff I pull out of the pool to poker plan on [17:59] ok cool, added [17:59] hatch: so that we don't go thorugh everything ever to do [18:00] definitely a good idea with our backlog haha [18:00] :) [18:07] hey hatch, during my git branch recovery, the patching dropped one set of changes. i just re-applied them but when i try to push to github it says the branch there is newer and needs to be merged. that isn't true. [18:08] looking [18:08] hatch: i don't want to get messed up again [18:08] bac did you rebase your fix in? [18:08] yes [18:08] ok then you need to use force [18:08] git push -f [18:08] but probably best to be explicit [18:09] ok [18:09] git push origin constraint-export -f [18:09] assuming your origin is your fork [18:09] there you go [18:09] :) [18:09] https://github.com/juju/juju-gui/pull/155/files [18:09] all good? [18:10] yeah, thanks [18:11] frankban: the lost changes are reapplied in https://github.com/juju/juju-gui/pull/155 : model.js and test_model.js are the only two that changed [18:12] bac, np, just be VERY careful with -f because we have the ability to force update develop which will blow out everyone elses stuff causing us to fix it :) [18:12] bac: redoing QA [18:24] bac: done [18:25] thanks frankban. [18:25] yw [18:47] hey bac [18:47] you had a launchpad link to show all the bundles, have that handy? [18:48] jcastro: https://code.launchpad.net/charms/bundles ? [18:48] ta [18:52] rick_h_ I'm going to be writing the integration tests then onto 'real' work again :) is there a Project you would like me on? 1 vs A? [18:52] I'm thinking A [19:01] hatch: yep, souds good. I think it's something you wanted to get involved with [19:02] hatch: let me know if we need to pre-imp or anything [19:02] sure [19:12] jujugui, need another review for https://github.com/juju/juju-gui/pull/155 since frankban has EODed. rick_h_ if you have just a sec to weigh in that would be appreciated [19:13] bac: rgr, looking [19:20] bac: commented [19:22] rick_h_: ok, so you get it a +1, but with a request to do something. is it going to start merging? [19:22] bac: no, :shipit: is the merge command [19:23] bac: :+1: is the LGTM [19:23] rick_h_: so you want me to un-promise the three tests i touched, right? [19:24] bac: yes please. [19:36] bac: heads up on test failure due to make lint [19:36] rick_h_: thx. [19:49] bac heh yeah, you'll get used to running `make lint && make test-debug` before your final commit :D [19:49] meh, lander catches it quickly and emails you. not the worst thing [19:50] hatch: oh, i'm used to it [19:50] no but then you have to rebase commit bleh bleh bleh [19:52] man sometimes I wish our yui modules would just fall over if their local containers didn't include the proper deps [19:52] so if you make incremental changes during code review, do you rebase before pushing each up? [19:52] bac I do, so I don't have a bunch of 'lint' commits in trunk [19:52] isn't that what messes up github causing you to 'push -f'? [19:52] yes [19:53] it seems you should only rebase before the first push and landing [19:53] that's not a "mess up" it's ok to push -f [19:53] it's only effecting your local branch [19:54] rick_h_: oh, i thought it was like '--overwrite' [19:54] in bzr [19:54] well it is - if you use -f on anything other than your current branch [20:05] bac: well in a way it is, but since it's just your local space it's safe. It's one of the reasons we all work in our local forks [20:21] aaaand I'm not back [20:22] install crashed and took my OS partition AND MY HOME PARTITION [20:23] benji: doh! [20:24] benji what upgrade was this? [20:24] see hatch, this here is just a lie: https://github.com/juju/juju-gui/pull/155/commits [20:25] bac: but it's all the lie we care about [20:25] :) git is really just a database [20:25] since proposing the branch i've pushed two subsequent revisions. now the history, even in the review, is collapsed [20:25] hatch: upgrading my laptop to trusty [20:25] benji from? [20:25] raring [20:26] benji: fresh install time? [20:26] oh, well that's no good, were you able to file a bug? [20:26] rick_h_: I'm speaking to you from a fresh install now. But I'll have to re-install because it installed to my HOME PARTITION... stupid stupid thing [20:26] benji: oh [20:26] rick_h_: would you look to see if i unpromised that mess correctly and gimme a shipit if so? [20:27] bac: sure thing looking [20:27] this will take so long to recover [20:27] ok, off to install again (or maybe munge paritions since I have nothing left to loose) [20:27] if I'm not back by the morning send a search party [20:28] hah [20:33] thanks rick_h_. god that was harder than it should've been. [20:34] bac: thanks for bearing throughit. Hopefully it'll go smoother since we found the new git issue, you've gone throgh the process, etc [20:44] have a good day all /me runs away [20:45] jujugui lf a quick review of my integration tests [20:45] https://github.com/juju/juju-gui/pull/156 [20:45] so far that card was considerably less work than we estimated [20:45] lol [20:45] helps that I wrote it once before ;) [22:01] hatch: will look in a few min [22:01] cool thanks [22:02] and hey, card can take less time :) we'll need it for the longer ones [22:06] haha [22:06] I was hoping to just take the extra time off? [22:06] lol [22:12] hatch: hah, that's why we have slack cards [22:23] hatch: so ignore the 'thinking out loud' on the helper stuff. Let me know if I'm nuts on any of the rest [22:24] ok cool looking [22:26] rick_h_ replied [22:26] I'll add some periods and the like then land it [22:27] thanks [22:29] hatch: k, more meant commend on the argument objects you're building vs the "calling callback" which is ok [22:30] ohh ok [22:30] yeah I can add that [22:31] it reads like "ok, yep you stub that, set the 4th arg to something, whatever that is. I guess it works." [22:31] I found out our elliptical takes D batteries....I'm not even sure where to get those anymore lol [22:31] harder to give real review feedback without going back/forth a ton [22:31] haha - yeah ok [22:31] lol [22:31] I've got enoloops with a case around to fit C and D [22:32] oh thats a nice idea, I have some Duracell 'Staycharged' but no cases [22:32] I should look for some of those [22:33] so get http://www.amazon.com/eneloop-Capacity-Pre-Charged-Rechargeable-Batteries/dp/B0058GZWXQ/ref=sr_1_1?s=electronics&ie=UTF8&qid=1393885972&sr=1-1&keywords=eneloop+max [22:33] and http://www.amazon.com/eneloop-SEC-DSPACER4PK-Spacers-battery-cells/dp/B002V8C3W2/ref=pd_sim_e_16?ie=UTF8&refRID=0Y7QENAYX6Y1ERTSG2GT [22:33] they don't last as long since they're AA in the core, but they're rechargable so you charge your thing once a month or whatever [22:33] yeah that's cool thanks [22:34] I'll see if amazon will sell to the great white north [22:34] hah [22:34] the last few things I tried to buy from .com they woudln't ship here [22:34] well just let me know [22:34] they let ship to you last time [22:34] let me * [22:34] yeah, will do thanks [23:00] oo boy 20 test failures removing the multiple inspector support [23:13] wheee [23:13] then hopefully some of those tests go away as well