/srv/irclogs.ubuntu.com/2013/06/18/#juju-gui.txt

=== rogpeppe2 is now known as rogpeppe
benjiHi all, I have a branch up for review that fixes the problem with running the charm tests locally (the "TypeError: __init__() got an unexpected keyword argument 'desired_capabilities'" excaption): https://codereview.appspot.com/10387043/12:20
bacbenji: on it12:22
benjithanks12:22
bacbenji: no card?12:22
benjibac: oh, it apparently didn't move; it should by in the review lane now: "packaged version of selenium doesn't work with local test-charm"12:23
bacbenji: oh, i didn't recognize that card as the same issue12:24
benjiI should have been more specific.12:24
bacnp12:24
gary_posterbenji, LGTM with doc suggestions12:26
benjigary_poster: look out, we might be asked to use this in the GUI: http://mintchaos.github.io/flippant.js/12:26
benjiand thanks for the review 12:26
gary_posterbenji, huh, not bad12:27
gary_posterluca__, fwiw see benji12:27
gary_poster's link above12:27
gary_postertry the "Card" buttons12:28
bacbenji: land it!12:29
gary_posterbenji, something fascinating--at least in chrome, look at the "js" when it flips at the top of the page.  it leaves a little part of the j around12:29
benjiheh, that is funny12:29
gary_posterI assume that's an artifact of it using opengl bits12:29
benjisome sort of bounding box mis-calculation I guess12:30
gary_postermaybe.  I think the bounding box would either take a bit of the period or leave a bit of the j12:30
gary_posterassuming it has to be a rectangle12:31
gary_posterHey luca__ , for future reference, do you know if Canonical has some kind of a shared SurveyMonkey account?12:40
luca__gary_poster: hmm, I'm not sure, I can ask Peter when he gets back from Lunch12:42
gary_posterthanks luca__ 12:54
teknicoguihelp: need Javascript and jshint help with a paradoxical branch I cannot propose: lp:~teknico/juju-gui/loadfixture-no-hashes13:07
rick_hteknico: what's up?13:08
teknicoi.e. the branch is ready, tests pass, but the JS spec and jshint say they should not :-)13:08
teknicorick_h: thanks, guichat? it's a bit long winded13:08
rick_hteknico: sure thing13:08
abentleyorangesquad: Could you please review https://code.launchpad.net/~abentley/charmworld/related-updates/+merge/170072 ?13:17
sinzuiI can take it13:18
teknicorick_h: no joy :-(13:27
rick_hteknico: really? ouch13:27
teknicorick_h: another weird thing I forgot to mention is that breakpoints on that "delete" line do not break :-o13:27
rick_hteknico: push the updated diff please. /me wants to pull it down and peek13:27
rick_hteknico: just make the line before `degugger;`13:27
rick_hthat should force it13:28
rick_herr, debugger; that is13:28
teknicorick_h: pushed. I don't get what you mean13:28
rick_hteknico: if you put debugger; as a line in your code it'll force a breakpoint13:29
teknicorick_h: oh, nice and useful13:29
rick_hteknico: yea, I've got a short macro for it in vim. `bug` will enter debugger; in js, import pdb in python, etc13:30
teknicoyeah, like "import pdb: pdb.set_trace()" :-)13:30
rick_hexactly13:30
teknicorick_h: still does not break :-o how can it happen that afterEach is not called in tests?13:32
rick_hteknico: oh hmm, that's strange13:33
gary_posterbenji, your branch broke CI...13:39
gary_posterRunning Tests13:39
gary_posterbin/test-charm: 41: bin/test-charm: virtualenv/bin/python: not found13:39
gary_posterUnable to deploy Charm.13:39
benjiit is unfortunate that the highlighted download on https://launchpad.net/juju-gui/ is a developer release.  I suppose we would have to change the Development focus to "stable"13:39
gary_postercould you look into it, or find someone else to do so?13:39
benjigary_poster: darn; I saw the message but the end said that it just needed to be re-run so I figured it was a CI error not a failure.  I'll look at it now.13:40
gary_posterbenji, agree with highlighted download.  had exact same thought as you, followed by "but "stable" is not our development focus!13:40
gary_poster"13:40
benjiyep13:40
gary_posterbut maybe from a LP perspective t is...13:40
gary_posterbecause for LP they are thinking of "saucy" as the development focus...13:41
gary_posterbecause stuff is going to be released in saucy13:41
teknicorick_h: could it be a problem with afterEach not using the done() function?13:41
gary_posterand we are going to release all of our stuff to trunk...13:41
rick_hteknico: otp atm13:42
teknicok13:42
rick_hteknico: done I think only applies if you declare done as a argument to the function13:42
teknicoyeah13:42
benjigary_poster: how about we switch it and see if it works for us13:42
gary_posterbenji, sure. it is just semantic AFAIK anyway.13:43
gary_posterthanks for raising benji.  you doing it, or shall I?13:43
benjigary_poster: done13:43
gary_posterthanks :-)13:43
benjithe download link is right now13:43
gary_posteryay :-)13:44
gary_posternot that anyone should actually be using it, but still :-)13:44
benjigary_poster: where is that Jenkins script stored?13:44
gary_posterbenji, it is in jenkins itself, and documented in https://wiki.canonical.com/CDO/Juju/GUI/CI13:45
gary_posterbenji, thank you for taking bug 1178462.  hatch has some advice for this I think, fwiw (maybe simply to use YUI DD?)13:51
_mup_Bug #1178462: Can't drag a charm from the charm browser sidebar to the main canvas <juju-gui:Triaged> <https://launchpad.net/bugs/1178462>13:51
benjigary_poster: cool; I'll do a pre-imp with him when I get Jenkins sorted13:52
gary_postercool13:52
rick_hgary_poster: benji let me know if you anything drives questions. Might be cool to add that to the widgets for charm-container or charm-token if container won't work for any reason13:52
rick_h /cool/easy13:53
gary_posterrick_h, :-) great.  maybe pre-imp from benji with you in addition to/instead of hatch?13:53
benjirick_h: first sentence: thanks; second sentence: I don't follow. :)13:53
hatchmorning13:58
gary_postermorning13:58
rick_hbenji: so the things you want to make draggable for that bug are a widget called charm-token13:58
gary_posterI *think* bcsaller is back today, albeit jet-lagged13:58
rick_hbenji: those are built and container in charm-container13:59
rick_hbenji: man I can't type the words in my head today. charm-tokens are built and wrapped in charm-container13:59
hatchbenji: http://yuilibrary.com/yui/docs/dd/proxy-drag.html13:59
rick_hhatch: can you just include a dd-node plugin on a widget and set it up during render?14:02
* rick_h hasn't done dd tbh14:03
teknicohatch: hi, interested in a couple more javascript enigmas? :-)14:03
hatchrick_h: can't remember, if it's written properly then yes14:04
rick_hhatch: benji http://yuilibrary.com/yui/docs/resize/constrain-resize-plugin.html is like what I'm thinking in my head14:05
rick_hso it'd be simple to add the plugin to widgets/charm-token.js and setup the config for it. Then catch the drop event on the env14:05
benjicool14:05
hatchheh...nothing is simple wrt the env :P14:06
rick_hhatch: ok, well I'm hoping the widget work into the charm-token will make the drag part easy then :P14:06
hatchyeah it should be *crosses fingers*14:06
rick_hat laest it was part of the master plan back when we started it lol14:07
hatchsuuuuuure it was ;)14:07
gary_posterrick_h, hatch, benji, something both clean and expedient should be ideal.  If there's some way of taking advantage of hatch's DD knowledge/experience, +1.  If tying this into the widget code slows things down significantly, -1.14:08
benjisounds good14:09
gary_posterthanks14:09
hatchI don't really foresee any issues with it pending that there isn't any obscure issues with D314:09
teknicohatch: I see my offer does not thrill you. I can understand that. :-)14:12
bacbenji: are you still looking at the jenkins failure?14:13
benjibac: yep14:13
hatchteknico:  hah sorry I didn't see that14:13
bacbenji: i see that lp:juju-gui has no development branch set.14:13
bacbenji: so i cannot 'bzr branch lp:juju-gui'14:13
hatchteknico: what was the path again?14:13
baci think that may be significatnt14:13
teknicolp:~teknico/juju-gui/loadfixture-no-hashes14:13
hatchok looking14:13
teknicohatch: ^^14:13
benjibac: that may be a side-effect of me setting the "development focus" to stable14:13
benjiif so, we may have to set it back14:14
teknicohatch: it needs explanation, though14:14
bacbenji: ok.14:14
benjiwe did that so the latest download would be the right thing14:14
bacbenji: but it appears to have been unset14:14
bacbenji: witnesseth https://code.launchpad.net/juju-gui14:14
hatchteknico: ok it's just checking out and making now14:15
teknicohatch: I'm in guichat14:15
hatchgary_poster: just fyi I finished the conflict prototype last night so I can demo it after I'm done with teknico14:15
gary_posterhatch, great!14:16
benjibac: oh, I know what's wrong.  Try it now.14:16
gary_posterFWIW, there's a newly available high priority card in story 1, jujugui14:22
benjigary_poster: do you mean "Story A"?14:23
gary_posterbenji, oh sorry yes, story A14:23
gary_posterthanks14:24
* benji adds "Story Alpha", "Story I", and "Story 1st" sections.14:24
gary_poster:-P14:24
jcsackettluca__: do you have a moment to chat?14:25
luca__jcsackett: via a hangout?14:25
jcsackettluca__: indeed.14:25
luca__jcsackett: sure, send me a link :)14:25
jcsackettluca__: open google+, an invite should arrive in a moment.14:26
frankbangary_poster: thanks for the review. re deploy(), perhaps I could add a flag (check_deployed) to ensure the service is ready after the call?14:26
gary_posterfrankban, would be fine, or just add a comment in the deploy method about the limitation and we fix it later if someone needs to?14:27
frankbangary_poster: +1 on fixing it later if required14:27
gary_postercool14:27
abentleyorangesquad: Can you please review https://code.launchpad.net/~abentley/charmworld/remove-api-1/+merge/170097 ?14:28
* sinzui takes it14:29
bacbenji: yes, that works.  thanks14:31
benjicool14:31
sinzuiabentley, r=me14:34
abentleysinzui: Thanks.14:34
hatchgary_poster: done with teknico, did you want me to demo now?14:37
gary_posterhatch, yeah!  guichat?14:37
hatchyup14:37
sinzuiabentley, about that short_url issue on the missing page. While ingest did not provide the attr, I can make every listed charm because I know the owner, series, and name14:38
sinzuiabentley, Maybe the model should make the short_url if it has those three parts. Maybe ingest never needs to make it since the pages are the only users of the attr14:39
abentleysinzui: +114:39
benjiarg, once more unto the (Jenkins) breach14:45
hatchwb bcsaller14:54
bcsallerThank you14:55
gary_posterbcsaller, hi15:01
gary_posterglad you are back15:01
gary_posteralejandraobregon, luca__ where are we meeting?15:02
bcsalleryeah, I'll need to get caught up15:02
luca__gary_poster: just added a hangout to the meeting15:02
gary_posterbcsaller, everything is in ClojureScript now15:02
bcsallersounds about right15:02
hatchbcsaller: actually we told gary_poster it was Clogure but really it's C#15:04
gary_posterlol15:04
hatchhaha15:04
rick_hMonooooooooooooooooooooooo15:04
hatchhaha you have the right idea15:05
hatchI don't think there is anything more confusing as to the difference between c# and Mono15:05
hatchI don't even think people who code C# know15:05
hatch:P15:05
rick_hheh, got a buddy that did C# for a long time and loved Mono since it was often ahead of C# implementing features in the languages spec15:05
hatchyeah then you throw .NET in there15:06
hatchand you have no idea what class is from where15:06
hatch:)15:06
hatchmy biggest gripe is that they have utility methods that start with caps15:06
hatchand that they have two different methods to loop through multidimensional arrays15:06
=== rogpeppe2 is now known as rogpeppe
abentleysinzui: Could you please review https://code.launchpad.net/~abentley/charmworld/fix-commits-in-past-30-days/+merge/170108 ?15:12
* sinzui looks15:12
sinzuiabentley, r=me15:13
benjiI suck.  Forgotten "bzr add".  Trying Jenkins again.15:14
abentleysinzui: Thanks.15:14
abentleysinzui: Any idea what this means? https://jenkins.qa.ubuntu.com/job/charmworld-merger-trunk/335/console15:18
sinzuiabentley, I think the instance restarted15:19
abentleysinzui: Oh.15:20
hatchDoes this mean that I'm the idiot? http://dilbert.com/strips/comic/2013-06-18 lol15:20
hatch^ gary_poster15:20
gary_posteron call will look15:20
sinzuihatch, possibly. I decided not to compete with all the brain power in the room. I choose to out dress them.15:21
hatchhaha15:21
sinzuiMy wife asked my why I was wearing a polka-dotted tie with a flowered shirt a few months ago. I replied, "Baby, I'm going into a meeting of alpha males. My only hope show dominance by power clashing."15:23
hatchhaha - I wonder if this phenomenon will catch on15:24
hatchjujugui call in 7, kanban now15:53
bacteknico: hey can you do a quick review of https://codereview.appspot.com/10395043 before you EOD?  it is exposing use-analytics in the charm.15:57
teknicobac: will do15:57
hatchjujugui call now15:59
gary_posterjujugui almost there16:00
bacthx teknico16:01
abentleysinzui: In terms of my changes, r273 is deployable.16:01
sinzuifab, I will make the formal request16:01
rick_hthanks abentley!16:02
bacteknico: branch looks good16:24
teknicobac: thanks, your one too :-)16:24
bacexcellent news.  i think i'll celebrate with lunch.16:25
gary_posterhatch, you available to show off your prototype soon to luca__ if he is available?16:26
teknicofrankban: branch approved, how about a look at mine? :-)16:27
hatchjust in a call with Ben right now so we can stop whenever16:27
gary_posterok thanks hatch16:27
gary_posterluca__, you available for a call about the inspector?16:27
luca__gary_poster: sure16:27
luca__gary_poster: guichat?16:27
gary_posterI think it is full luca__ , one sec16:28
benjirick_h and/or hatch: do you have a minute for a pre-imp on the drag-and-drop?16:28
gary_posterhatch, bcsaller do you mind if luca__ and I crash your party in guichat for a few minutes?16:28
rick_hbenji: I can, but I think hatch would be good since I can only speak to our widget side and not full DD16:28
bcsallerthats fine16:28
rick_hbenji: sounds like hatch is tied up atm16:28
gary_postercome on by luca__ 16:28
rick_hbenji: but I can background the widgets if that's helpful 16:29
benjirick_h: verbing weirds language16:29
benji:P16:29
rick_hbenji: "I can provide background on the widgets if that's helpful"?16:30
benjirick_h: the regular hangout is busy, I'll make a new one16:30
rick_hbenji: rgr16:30
frankbanteknico: thanks, I'll do16:30
benjirick_h: https://plus.google.com/hangouts/_/1a34cdb70051a661112a75e12dd50779a5e2a0a4?hl=en16:32
sinzuiabentley, r273 is deployed16:50
hatchbenji: rick_h ok I'm available again, am I sitll needed or are you guys O K?17:17
hatchguessing ok :)17:20
hatchwhat do we do if we forget our password to the canonical admin ?17:32
rick_hhatch: we get lastpass17:35
hatchlol this was pre-lastpass17:36
rick_hhatch: not sure, check the wiki and then I'd check with #is I guess. 17:38
hatchalright will try is17:39
=== matsubara is now known as matsubara-lunch
sinzuiorangesquad, any available to review https://code.launchpad.net/~sinzui/charmworld/charm-model-cruft/+merge/17015117:44
rick_habentley: I'm always getting has_icon = false in the api atm. the mysql charm in here I believe should have one? 17:50
rick_hsinzui: I can look17:50
bacjujugui: anyone have time for a second review of https://codereview.appspot.com/10395043 ?17:52
Makyobac, sure17:53
bacty17:53
rick_hsinzui: conflict in the file fyi17:56
sinzuioh dear17:56
rick_hsinzui: heh, nothing major. 17:56
rick_hsinzui: r=me otherwise17:59
abentleyrick_h: looking...18:00
sinzuithanks rick. I thought I had already merged abentley's api1 removal...but the MP clear shows I did not18:01
abentleyrick_h: Okay, I know what I did wrong.  Will be a trivial fix.18:07
sinzuiorangesquad, one of you just tested staging. You got a 500, but that is because we have a 4040 logic18:07
rick_habentley: cool thanks. Did you want me to file a bug/card?18:07
abentleyrick_h: Nah, I will take care of it.18:07
rick_habentley: thanks18:07
abentleysinzui: That was me.  I'm surprised, because I thought the API handled that case.18:08
sinzuiabentley, no, I used it as a test case for something a few weeks ago. I reported a bug, but set it aside until charm model was done18:08
sinzuiabentley, this is bug roughly matches the TB I just got in my email: https://bugs.launchpad.net/charmworld/+bug/117446618:10
_mup_Bug #1174466: bad URLs cause 500 errors where 404s would be expected <api> <oops> <charmworld:Triaged> <https://launchpad.net/bugs/1174466>18:10
abentleysinzui: Could you please review https://code.launchpad.net/~abentley/charms/precise/charmworld/worker-interval/+merge/170127 ?18:21
sinzuisure18:22
sinzuiabentley, r=me18:23
=== BradCrittenden is now known as bac
sinzuiabentley, I think we need an rt to get production to update the charm. They can do that when we do the next rollout of m.jc.com18:24
abentleysinzui: Sounds good.18:24
abentleyorangesquad: could you please review https://code.launchpad.net/~abentley/charmworld/fix-has-icon/+merge/17015918:24
jcsackettabentley: looking.18:25
jcsackettabentley: r=me.18:25
abentleyjcsackett: Thanks!18:35
hatchbcsaller: when you were investiaging canvas for the env did you see http://paperjs.org/ ? It's for doing vector in canvas18:37
* hatch isn't saying we should switch :)18:38
bcsallerhatch: I'd seen that before, but a long time ago, not in the recent search18:38
hatchhttp://paperjs.org/examples/hit-testing/ see this for an env like demo18:38
=== matsubara-lunch is now known as matsubara
hatchI was doing some canvas research this weekend (for fun, cuz apparently I'm a nerd) and came across that18:39
bcsallerits pretty cool18:39
hatchyeah, I am still pretty sure svg is the right thing for us...but with this it removes the argument about vector sizing18:40
abentleysinzui: ever seen anything like this? https://pastebin.canonical.com/93040/18:45
sinzuiabentley, never18:46
sinzuiabentley, but I just issues the command and got it.18:46
sinzuiabentley, I issued the same command a few hours ago and saw staging was lovely18:46
abentleysinzui: Oh, maybe I'm somehow misconfigured.18:46
sinzuiabentley, This is the timestamp from my last successful run:18:47
sinzui2013-06-18 08:49:10,23818:47
sinzuiI think that is EDT18:47
abentleysinzui: So you're saying you get the same error as me, right?18:48
sinzuiI do now18:48
abentleysinzui: Okay, I'm going to try restarting zk18:49
abentleysinzui: Oh, I can't ssh to machine 0, because I get the same error.  Do you have the ip address of machine 0?18:50
sinzuiI do, let me past the status from 6 hours18:51
sinzuiabentley, https://pastebin.canonical.com/93041/18:51
abentleysinzui: Thanks.18:52
abentleysinzui: restarting ZK didn't help.  Rebooting the machine didn't help.18:59
sinzui:?18:59
sinzui:/18:59
sinzuiabentley, do you think webops would know about this? canonistack might throwing a wobbly?19:01
* hazmat backlogs19:02
abentleysinzui: I don't think it's likely.  It looks like a juju/ZK problem.19:02
hazmathatch, bcsaller paper.js is fun  rn19:03
hazmatabentley, is this an ops setup or staging?19:03
abentleyhazmat: staging.19:04
hazmatabentley, not being able to connect to status isn't zk/juju but object storage and local config19:04
hazmatfind zk is before it even connects to zk19:04
hazmatits trying to query object storage for the control bucket and parse where the zk/instance is19:05
abentleyhazmat: Okay, so object storage then, because neither sinzui nor I had this before today, now both do.19:05
hazmatabentley, the empty string  at the end of the error msg is suspect19:05
hazmatabentley, if you have swift tools can you do an ls on the bucket?19:06
hazmatclient tools that is19:06
hazmatabentley, or alternatively pastebin the bucket name19:06
hazmater. priv msg19:06
hatchhazmat: I'm not really sure what to make of that statement :)19:11
hazmatabentley, sinzui i get internal server error on swift..19:12
hazmatdon't think i can help19:12
sinzuithank you for playing hazmat19:12
sinzuiabentley, the nuclear option?19:12
abentleyhazmat: Can you give us a pastebin we can give to is?19:13
hazmatabentley,  could just be an account issue for me .. http://pastebin.ubuntu.com/5778053/19:14
hazmati've had other issues with swift on canonistack in the last two months19:14
abentleyhazmat: okay, thanks.19:14
hazmathatch, paper.js seems really nice at interactivity and modeling a scene graph. i've seen it before but  after i think after d3 got sold for gui, i had hoped we could use more d3 builtins at the time and when it was basically d3 vs raphael.19:18
hatchahh19:19
hazmatthe degree of interactivity reminds of a much improved processing.js.. ie. twould be nice for games as well19:19
abentleysinzui: hrm.  The bucket seems to be missing for me: http://pastebin.ubuntu.com/5778076/19:21
sinzuioh.19:21
* sinzui tries19:21
sinzuiabentley, I agree swift has issues.19:26
hatchhazmat: yeah I was thinking about how hard it would be to put a gameloop with paperjs19:27
abentleysinzui: swift issues are now reported in #is.19:28
hatchhazmat: there are of course a number of other canvas game frameworks including impact.js19:32
sinzuiorangesquad, I have a short branch to fix an oops seen on staging today: https://code.launchpad.net/~sinzui/charmworld/no-handler-is-404/+merge/17017619:34
abentleysinzui: looking.19:34
abentleysinzui: r=me.19:35
sinzuithank you abentley19:35
sinzuiabentley, I think staging was up for 11 days this time, and it wasn't elastic search dying.19:37
abentleysinzui: Right.  We may hope that elasticsearch no-startiness is a thing of the past.19:38
sinzuiabentley, and we haven't see the ES restart issue since we switched to 0.90.0.release. I ponder changing the bug to incomplete19:38
abentleysinzui: +119:39
hatchthis is my third MS Ergo 4000 keyboard, and on every single one the alt key stops working lol20:10
hatchawesome new Android humble bundle http://www.humblebundle.com/20:53
bacgary_poster: huw should be working in a bit, no?20:55
gary_posterbac, he starts in about 1.5 to 2 hours20:55
bacgary_poster: ok, cool, just checking to make sure he was not out.  in that case, i think i'll go dog walk and check back later to talk to him about css tweaks20:56
gary_postersounds good bac, thanks.  have a nice evening20:56
MakyoCute. https://twitter.com/codinghorror/status/347070841059692545/photo/121:01
hatchlol what's it an add for?21:03
MakyoNo idea, just saw it pop up earlier :)21:04
* Makyo dogwalk21:13
hatchdatabinding conflict resolution prototype review https://codereview.appspot.com/10397045/ bcsaller gary_poster21:16
hatchbcsaller: I kept the code pretty close to pre our discussion so feel free to put those notes on the review as we discussed21:18
gary_posterhatch, taking a glance, but must run very soon21:19
hatchnp, thanks21:20
gary_posterhatch, why deleted?21:23
gary_poster if (this.model) {this.unbind();}21:23
hatchguichat real quick?21:23
gary_posterin BindingEngine.prototype._bind21:23
gary_posterheh, *really* quick, ok21:23
hatchpssst anyone here?22:08
hatchmorning huwshimi22:41
huwshimihatch: Morning22:41
huwshimi(evening)22:41
hatch:) how are things?22:42
huwshimihatch: Good, lots to do!22:45
bachi huwshimi22:58

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!