/srv/irclogs.ubuntu.com/2012/11/21/#juju-gui.txt

=== Guest84948 is now known as rogpeppe
=== rogpeppe is now known as rogpeppe1
=== rogpeppe1 is now known as rogpeppe2
=== rogpeppe2 is now known as rogpeppe3
=== rogpeppe3 is now known as rogpeppe
=== mattuk1972_ is now known as mattuk1972
hazmatfrankban, looking over the notifications.. why the Y.timer (gallery-timer) instead of just Y.later?12:19
hazmatthis is for the pause functionality when hovering?12:20
frankbanhazmat: for the doc, Y.later does not seem to have pause functionality12:21
frankbanhazmat: heh12:21
hazmatfrankban, no worries, just trying to keep track of our gallery modules12:21
frankbanhazmat: cool12:21
* frankban lunches12:31
=== jovan2_ is now known as jovan2
benjigary_poster: I just remembered to put in my sick day for Monday.13:34
gary_posterack13:35
bacgary_poster: hey i'm now seeing the favicon in chromium but only on the tab, not in the address bar.  is that what you see?  #notreallyobsessed13:37
gary_posteryes bac13:37
bacoh13:38
bacwell itfm then13:38
bacer, iwfm13:39
gary_posterbac, that's the way other sites work for me on chrome too.13:45
gary_posterbenji, approved fwiw. :-)13:45
benjinow I can begin the psycological healing13:46
gary_posterlol13:54
* tveronezi brb13:58
* tveronezi back14:17
teknicobenji, weird, I got the email from rietveld with your answers to reviews, but I don't see your changes on launchpad: https://code.launchpad.net/~benji/charms/precise/juju-gui/second/+merge/13437014:30
benjiteknico: hmm, that is weird; let me look at the lbox log, maybe it pushed it to the wrong place14:30
benjiteknico: lp:~juju-gui/charms/precise/juju-gui/trunk looks fine to me14:33
teknicobenji, indeed, so it's just some miscommunication between rietveld and launchpad14:35
benjiteknico: oh, I know what it is: I did not push the review changes to the branch.  I never do in this new, lboxy world.  Since you are interested in seeing them, I will push them now.14:36
benjipushed14:37
teknicobenji, oh, ok, thanks, I was just wondering whether something had gone haywire (sp?)14:37
teknicobenji, uhm, I understand pushing the latest revision from bzr, but what do you mean exactly by "pushing the review changes to the branch"? doesn't rietveld do that automatically?14:40
benjiteknico: not according to the log (I did an "lbox submit", "propose" probably does push to the branch)14:44
teknicoah, lbox, right14:45
* hazmat runs an errand, bbiab14:54
gary_posterhery tveronezi .  Why does the old line 38/39 approach in https://codereview.appspot.com/6856070/diff/9001/test/test_app.js?column_width=80 not work anymore again?  Also, one test is failing reliably and others seem more fragile in your branch15:10
gary_posterAnd in an urelated note, the tests are mutating the URL again 15:11
gary_posterso that you end up on localhost:8084 instead of localhost:8084/test15:11
gary_poster(And "Environmenton excellent provider" is at the top of the page)15:11
gary_posterbut all those last notes are from trunk as well15:11
gary_postertveronezi, failing test in your branch is in relation notifications: "should generate messages about one-party relations"15:13
gary_posterLooks like it is a bad space removal: 15:13
gary_posterexpected 'Relation with endpoint1 (relation type "relation1") was created' to equal 'Relation with endpoint1 (relation type"relation1") was created'15:13
tveronezigary_poster: We cannot know the order of the tests. YUI loads the dependencies asynchronously.15:13
tveronezigary_poster: It wasnt failing yesterday. I will check it again.15:14
gary_postertveronezi, that's why we were using "done"15:14
tveronezi"done" gives timeout sometimes.15:14
benjiit sounds like we need some URL assertions in whatever test that is15:15
gary_posterso you mean if I revert your branch to the previous approach in test_app.js it will intermittently fail tveronezi ?15:15
gary_poster(as an example)15:15
tveronezigary_poster: yeap.15:16
gary_postertveronezi, why is that different in your branch?  I have not seen that fragility in the trunk15:16
gary_posterat least I don't think I have :-P15:16
tveronezigary_poster: In fact, we should change all the tests. We should wrap the tests with the "YUI().use" call in order to have the Y object in order to follow the YUI application principle discussed in previous branches.15:19
gary_postertveronezi, that's a reasonable argument.  I would expect that to go in a separate branch (and possibly after team discussion, since it is a pretty big change) rather than changing things piecemeal here. I guess you are saying that these particular tests were the most fragile, and they need to be changed in order for your branch to be successful.  I wish I understood why that these changes are necessary for this bran15:24
gary_posterch.  Do you know?15:24
tveronezigary_poster: I've figured out why the tests are broken. It is due to a last minute attempt to make "prep" and "lint" happy. I've removed a white space by mistake.15:25
tveronezigary_poster: I have a meeting now... I will be back in one hour.15:26
gary_posterok15:30
gary_postertveronezi, approved, with reservations noted and a small suggestion.15:42
gary_posterbenji, teknico there is a "setup tests for Juju GUI charm" in Active review on kanban.  We can move that to done-done, right?15:48
benjigary_poster: oh, yep; thanks15:49
gary_postercool, race you15:49
teknicogary_poster, oh, right15:49
gary_posterI dunno who won, but I did it :-)15:49
teknicothis seems useful as far as YUI optimizations are concerned, title notwithstanding: http://speakerdeck.com/yaypie/when-not-to-use-yui15:53
benjiI won: I didn't do anything and you moved it for me. ;)15:53
gary_poster:-)15:54
teknicothen it's a tie!15:54
benjiteknico: I'm only part way through, but we should all look at that slide deck15:56
teknicobenji, yeah, it's quite startling15:56
benjiThe part about YUI.Base using 2k per instance may be very important for us if we are representing thousands of units.15:57
hazmattveronezi, gary_poster the yui loader takes an option on the async loading15:57
gary_posterhazmat, ah yes, that rings a bell15:58
hazmathttp://yuilibrary.com/yui/docs/api/classes/Loader.html 15:58
hazmatasync defaults to true15:58
gary_posterthe change he made is still a reasonable one though, I think.  But it would be better to talk about itbevore we go down that road15:58
gary_posterbac benji frankban hazmat jovan2 Makyo teknico tveronezi call in 215:58
teknicobenji, and it partly explains YUI's bad placement in speed comparisons like http://jsperf.com/js-mvc-frameworks-models/315:58
benjihazmat: I muted your keyboard. ;)16:02
hazmatbenji, danke ;-)16:03
hazmatbenji, re 2k per instance, that's why we're using the lazy their16:03
hazmatthere16:03
* benji buys hazmat some cherry brown switches for his keyboard for Christmas.16:03
benjimmm, makes sense16:04
hazmatbenji, cherry red atm16:06
hazmathmm.. actually cherry blue.. the brown's would sound nice though.. stealth mode16:07
bacbut i just refactored the picker!  :(16:12
gary_posterben is on a bus with spotty wifi16:17
gary_posterhis review email disappeared for some reason16:18
gary_posterbut it is here:16:18
gary_posterhttps://codereview.appspot.com/6858045/16:18
gary_posterhazmat, ^^^16:18
hazmatgary_poster, aha..16:18
hazmatgary_poster, cool, thanks16:18
hazmatgary_poster, odd about the lack of lp email16:18
hazmatgary_poster, oh. its the wrong merge target16:19
gary_posterbac benji frankban hazmat Makyo teknico tveronezi if you have some time, bcsaller would like reviews of https://codereview.appspot.com/6858045/ .  Two people can do the more thorough review, but he'd be interested in broader, bigger picture reviews too16:20
gary_posterThis is the environment view refactoring16:20
hazmatactually pls hold off on that for a few..16:20
hazmatthe diff is bad16:20
gary_posterit is supposed to show a pattern for writing the environment view and for writing environment view tests16:20
gary_posterok cool hazmat.  you contacting him or fixing it?16:21
hazmatgary_poster, yeah.. on the phone16:21
gary_postercool16:21
benjik16:21
hazmatgary_poster, so he's go no network access atm.. i'm going to just resubmit with the correct parent16:22
gary_postercool thanks hazmat16:23
benjiThe UI is very brittle regarding font sizes; upsizing the font by one unit (by pressing control +) breaks the layout.16:54
hazmatbenji, all these absolute sizes are badness..16:56
benjiyep16:57
hazmatgotta think about ems and relatives16:57
gary_posterhey Makyo, do bac and bcsaller know that you are waiting on their reply to move lp:~makyo/juju-gui/adding-subordinates ?17:01
Makyogary_poster, I just wanted bcsaller to check on the test, but I suppose anyone can okay that.  I tried pinging him yesterday, but flaky internet got the best of us.17:03
MakyoThat's https://codereview.appspot.com/6856067/diff/5001/test/test_environment_view.js if anyone would like to take a quick peek.17:04
MakyoDog's being a brat, going to take him for a quick walk and hope that helps.17:08
bacMakyo: +1 on the test17:27
Makyobac, thanks.  If it's good with you, I'll land.17:27
bacsure17:27
* Makyo -> doctor19:01
hazmatthe env refactor is at https://codereview.appspot.com/6842084/19:27
hazmathad some troubles on the initial propose19:27
benjihazmat: that is the branch that Ben wants many people to peruse, right?19:34
hazmatbenji, it is19:34
benjicool, I'll do that19:34
hazmatbenji, awesome thanks..19:34
hazmati'm hoping he'll be around in the next hr to correct the mp himself.. but that at least gives a nice review diff19:35
hazmatman tty.js is pretty sweet19:48
bacubuntu + mbp + bulging battery = very low untethered work time20:30
* gary_poster here with less tooth than before20:37
bacyikes20:40
* bac dog walk. bbiab.20:40
* benji recommits to flossing more often.20:46
benjigary_poster: is bug 1078978 up for grabs?  It looks fun.20:56
_mup_Bug #1078978: CSS is not minified in static deployment <juju-gui:Triaged> < https://launchpad.net/bugs/1078978 >20:56
benjihmm, it may be blocked on bug 1078910 though20:59
_mup_Bug #1078910: development server is unnecessarily complicated and different from production delivery <juju-gui:Triaged> < https://launchpad.net/bugs/1078910 >20:59
gary_posterbenji, I don't think one has to be solved before the other, but could be wrong21:04
benjigary_poster: I would at least need to base my branch on the branch for 1078910, right?  Wait, maybe not... enough minimization is in the trunk that I can base it on that, right? 21:05
gary_posterbut yes, they are up for grabs.  benji, https://code.launchpad.net/~tveronezi/juju-gui/change-requires-param ought to be landed, or massaged and landed, before that.  If you could take that it would be helpful, and would set you up for the rest21:06
gary_posterdo you want to have a call?21:06
benjigary_poster: sure21:06
gary_posterI didn't have laughing guess, but I am numb, so my speech might be at least mildly amusing21:06
gary_postergas21:06
benjiI'll take what I can get.21:07
gary_poster:-)21:07
gary_posterok juju-ui looks empty21:07
MakyoUgh, sorry, awful traffic.  Back.22:19
gary_posterhave a great rest of week, weekend, and Thanksgiving where appropriate, everyone!22:23
=== gary_poster is now known as gary_poster|away

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