[00:03] figured as much but haven't looked yet [00:08] hatch: https://codereview.appspot.com/12494043/ if you want to review [00:10] done [00:11] hatch: Thanks! [00:13] no probs [00:14] huwshimi: being a css guy I figured you might be interested in my latest blog post http://fromanegg.com/post/57464448123/css-selectors-the-next-level [00:14] lemme know what you think (or if I have any errors ;) ) [00:16] hatch: Nice. [00:17] hatch: I think the second code example should be ".dog::first-letter" [00:18] checking [00:18] yup [00:18] :) thanks [00:18] :) [00:18] Other than that it's a handy guide [00:19] thanks, glad you like it :) === schwuk_away is now known as schwuk [11:26] gary_poster, Morning just landed a branch with updated zoom slider styles: lp:~ya-bo-ng/juju-gui/zoom-slider-update [11:30] antdillon: do you need help getting this into review and such then? === schwuk is now known as schwuk_away [11:42] rick_h, Not sure Gary wanted me sending stuff for review [11:42] rick_h, But I just do lbox propose -cr right? === antdillon is now known as ant-lunch [11:48] ant-lunch: https://codereview.appspot.com/12523043 [11:49] ant-lunch: added a card to the board. Will get it reviewed and such. [11:55] ant-lunch: comment/question on the images in the code review for you https://codereview.appspot.com/12523043/ === ant-lunch is now known as antdillon [12:33] rick_h, Will sprite them now [12:33] antdillon: thanks [13:08] rick_h, Done, want me to do a mp? [13:08] antdillon: do just push an update to your branch and I'll pull it down and resync it with the code review in progress [13:09] rick_h, Thanks [13:10] orangesquad, is there a way to check for errors on a charm? https://jujucharms.com/sidebar/~pavel-pachkovskij/precise/rack-4/ shows no readme but there is one in the source. [13:10] rick_h: You can look at the ingest logs. [13:27] orangesqad, bac or benji: Could you please review https://code.launchpad.net/~abentley/charmworld/restore-questions/+merge/178753 ? [13:27] abentley: sure [13:27] abentley: I'll take a look. === schwuk_away is now known as schwuk [13:27] or not if bac is dying to do it (abently: do you need one or two reviews?) [13:28] bac: Thanks. benji, I don't need two reviews. [13:28] benji: But thanks anyway. [13:28] * benji and bac thumb wrestle; bac wins. [13:28] \o/ [13:29] * bac hops up and down, rocky style [13:29] abentley, benji: how did gary's demo go yesterday? [13:30] bac: The server was down, but he said it went well anyway. [13:30] non-esistent; the server was down so he did the "demo" with hand waves [13:30] darn [13:33] antdillon: sorry, think we had a miscommunication there [13:33] antdillon: any png in the assets/images directory is turned into a sprite automatically. we can use that with the css "sprite filename". [13:34] antdillon: if you look around you can see it used in a bunch of places [13:34] antdillon: so I was more just looking for the css styles to be onthe divs you used vs the image links themselves. [13:34] rick_h, Ah ok, on it [13:34] antdillon: sorry, didn't mean to make you sprite up the images you added on their own. [13:34] rick_h, I've put the images as backgrounds of the divs instead of images [13:35] antdillon: it's why we've got that 'non-sprites' directory. For images that don't need ot go into the auto-sprite [13:35] antdillon: right, and that's what the sprite tooling in place does [13:37] abentley: the text in iter_categories() is user-visible i assume. did you grab that from somewhere else or make it up? i ask because there are some grammar issues but i don' t want to spend a lot of time on them if they are not user visible nor under our control. [13:37] bac: I grabbed it from the original migration that produced the questions. It is user-visible. [13:39] bac: The original: http://bazaar.launchpad.net/~juju-jitsu/charmworld/trunk/view/243/migrations/versions/001_load_qa_questions.py [13:40] abentley: ok, some just look clunky but perhaps read correctly in context. [13:40] abentley: and 'user' is only members of ~charmers, right? [13:41] as in 'user visible' [13:41] bac: The user is anyone who looks at the QA ratings. [13:41] oh, ok [13:44] bac: There are definitely grammar problems. Some questions refer to the subject as singular "passes tests...", while others refer to it as plural "Contain a suite..." [13:44] abentley: done [13:44] abentley: if you could make a pass at those it would be nice. [13:45] rick_h, Ok to keep the button and hover images together? Or should I separate them all and change the background image [13:45] antdillon: no, split them up back like they were. Just don't link directly to them. Use the sprite css classes to load them into the backgroud of your divs [13:45] rick_h, Got ya [13:46] bac: The content of those questions is the ~charmers' choice AIUI. We can certainly point out the issues. [13:47] antdillon: run `bzr grep "sprite "` to get an idea of uses there [13:49] rick_h, Cheers [13:54] morning [14:07] frankban: looks like you fixed CI :) [14:07] yay! [14:07] guihelp: we ignored that poor jenkins guy for so long... but (for) now he's back to sanity! [14:07] thanks for doing that :) [14:07] yay, frankban [14:07] :-) [14:08] frankban: he's such a jenk, you know ;-) [14:08] yay, frankban, thanks! [14:09] jujugui, fwiw, I'm going to work on Huw's proposed branch for a break. :-) [14:09] gary_poster: this one https://codereview.appspot.com/12506044/ ? [14:09] hatch yeah [14:09] ok I'm also in the process of reviewing [14:09] teknico: doing your 2nd review [14:10] hatch oh ok. you do it then. did you already fix the destroy service duplication and the expose issues? [14:10] hatch, if not I'll do it :-) [14:10] negative [14:10] bac: thanks, you may want to hold off qa though, I'll fix the problem frankban pointed out first [14:10] ok [14:11] gary_poster: so far the main issue I've found is that he moves the viewlet-manager-footer into the service-configuration.partial [14:11] hatch, that's per request [14:11] which isn't really semantically correct [14:11] hatch oh [14:11] I know the button should move...just that we should keep the footer class in the wrapper :) [14:12] abentley: got a sec to help me debug what's up with this charm please? [14:12] got it. hatch, may I work on the expose issues and the destroy issue duplication? :-) I need a little fun [14:12] haha you bet [14:12] rick_h: Sure. Is it the one you mentioned earlier? [14:12] :-) thanks [14:13] how was the flight? [14:13] very long hatch, three legs. :-/ [14:13] but fine otherwise [14:13] I suppose that's the best you can hope for :) [14:13] abentley: yea, I finally setup my ssh to get into staging. On staging I'm seeing http://paste.mitechie.com/show/994/ in the logs a ton in app.log [14:14] abentley: but having a hard time seeing where this error comes from. The log line says "update_charm" but update_charm doesn't do a warning, and the error format has "error" in the name so it's from elsewhere and trying to chase it. [14:14] rick_h: The exception is being raised by Bazaar. [14:14] abentley: greeping for part of that message comes up with no results so guessing it's coming out of something else? [14:18] rick_h: It looks like the user uncommitted that revision, because I can't see it on the Launchpad branch. [14:18] abentley: ah, ok. /me didn't realize you could uncommit a revision [14:19] abentley: what's the way I can suggest he moves forward? Can he get back the uncommitted revision and move forward from there? [14:21] rick_h: So, the uncommitted revision is actually on Launchpad, but because it's not part of the current history, it's not being fetched. He's just committed a new revision, so we should stop caring about the old one soon if we haven't already. [14:21] abentley: yea, I thought maybe he was hitting the proof error of no revision file so he pushed an updated version with a revision file and we waited for ingest to pick it up [14:22] rick_h: The new revision is from 48 minutes ago. Do you have any errors more recently than 30 minutes ago? [14:22] abentley: m.j.c. pulled the revision enough that the changelog is updated, but the config/readme and other files seem stuck pre this issue http://manage.jujucharms.com/~pavel-pachkovskij/precise/rack [14:23] abentley: yes, latest is from 3min ago [14:23] abentley: looks like this error is hit every ingest cycle [14:23] rick_h: m.j.c is on revno 5. [14:23] log files is spammed to death [14:24] abentley: well app.log [14:24] rick_h: Okay, so I guess the charm store is slow. [14:24] abentley: yea, that's the latest rev in bzr. he went from r4 to r5 today as part of working this out [14:24] bac: bug fixed, qa may proceed, thanks again :-) [14:25] rick_h, Do you need to use js to add the hovered class to display the hovered image? [14:25] antdillon: I'd assume we could use a :hover css selector? [14:25] antdillon: oh wait, so there's a hover on the sprite image? [14:26] antdillon: if so then yes, we'd need to change the css class on mouse in/out [14:26] rick_h, I think I will because changing the background pos isnt working [14:26] rick_h: So, the charm store still is using pavel.pachkovskij@altoros.com-20130301074156-vkto1nnhg3pujiyq. Once it updates, we'll update: https://store.juju.ubuntu.com/charm-info?charms=cs:~pavel-pachkovskij/precise/rack-4 [14:27] abentley: ok, will tell him to hold on and keep an eye on it. Thanks for the help [14:27] abentley: and I'll check with him about the uncommit usage to verify that's what caused it to become a mess [14:27] rick_h: Technically, that URL should be https://store.juju.ubuntu.com/charm-info?charms=cs:~pavel-pachkovskij/precise/rack so it tracks head. [14:28] abentley: ok [14:28] rick_h, Unless I can plus or minus from current bk pos :) [14:28] antdillon: huh? I'm not following. [14:28] antdillon: what are we hovering? [14:30] rick_h, So the sprite system is putting the sprites in the middle sprite sheet so I can't use hover to move the bk to the corrent hover pos. I'll add classes with js [14:30] rick_h, Hovering the plus and minus buttons on the zoom [14:31] antdillon: ok, right. We'd have to use js to change the css class to the hover image title [14:31] antdillon: it's a good idea anyway to avoid flicker while the hover image is loaded on the first runs [14:37] gary_poster: there is one other issue in https://codereview.appspot.com/12506044/ that you could also fix ;) The inspector height is always the full allowable height regardless of the content [14:37] ack hatch, thanks, will discuss with luca et al. Wondering about interaction with charm panel. [14:38] gary_poster: I *thought* we set both the height of the left breakout along with the inspector height in the 'recalculateHeight' method [14:38] gary_poster: so they should be kept in sync [14:38] * rick_h reads link to catch up [14:39] rick_h, right. the question is whether the desire to have a nice full charm side panel overrides the desire to have a short charm panel when needed. [14:39] gary_poster: ahh right I forgot about that change [14:48] frankban: teknico do either of you live here? https://plus.google.com/u/0/103804991882843552936/posts/5SANwPcSYrG :D [14:52] gary_poster: how do we respond to the jujugui documentation doc? [14:54] hatch: no we don't. that's near Naples, there are beautiful places in that area. [14:54] yeah - I would imagine expensive to live there too? [14:56] anyone have a few minutes to help me diagnose some test failures, it looks like I have a crossed wire. lp:~bcsaller/juju-gui/conflict-ux/ [14:56] bcsaller: you bet [14:56] hatch: well, I guess buying an house there is more expensive than the average in Itay, and the average here is ... expensive. [14:56] thank you, this thing is driving me crazy, hopefully shallow with another set of eyes [14:57] bcsaller: it's because you're up to early [14:57] you should sleep in for another hour [14:57] :) [14:57] oh, this is from yesterday too though [14:57] frankban: haha yeah? what's the avg per square foot? [14:58] hatch, jujugui documentation doc: could comment on it? [14:58] gary_poster: ok can do - I just wasn't sure if there was a discussion thread somewhere or not [14:58] hatch: you are asking too many unit conversions here ;-) [14:59] frankban: oh right...sorry per sq/m :) [14:59] hatch, nope. some verbal discussions here, but nick is out this week [14:59] ahh [15:00] bcsaller: first issue.... :8084/test/test_conflict_ux.js 404 (Not Found) [15:00] :) [15:00] hatch: added, sorry and pushed [15:00] :D [15:00] those tests are not the one with the issue though [15:01] wait till you see this one... :-/ [15:01] waiting......... [15:01] * hatch needs faster computer [15:01] * hatch starts indigogo campaign for new MBP [15:01] http://0.0.0.0:8084/test/index.html?grep=Inspector%20Constraints is the short cut in this case [15:02] 5 failures so far [15:02] oh, I only get 4 in that module [15:02] allllmost doneeee [15:02] alllllmmmoooossssttttt [15:02] and done [15:02] 75s [15:02] oh man haha [15:02] it's 50s slower in the console [15:03] er faster [15:03] ok investigating now [15:04] I get the 4 errors you speak of, and an error in Inspector Conflict UX [15:04] should indicate conflict and allow resolution of constraints [15:04] if you were wondering what the other error was [15:04] hatch: http://www.immobiliare.it/Salerno/casa-Positano.html it's in italian but you can easily find the important bits of info, e.g. € 680.000 | 85 m² ;-) [15:05] hatch: oh, that last test is new, I'll add a skip, was trying to figure out what happened with the rest [15:05] bcsaller: oh ok cool [15:05] frankban: haha yep that's expensive :D [15:07] bcsaller: your asserts are backwards [15:07] just fyi [15:07] hatch: I didn't write those, but I agree [15:07] expected is the second param [15:07] confusing...I know [15:07] ohh [15:08] existing tests that now break [15:08] but in such a strange way [15:08] yeah I see that [15:08] subscribers seem to be getting crossed [15:08] ok grabbing my coffee then I'll dig in deeper [15:08] oh me too, me too [15:08] bcsaller: so I got failures on that section when I had left over dom elements from an earlier test [15:09] bcsaller: when I did proper html cleanup it went away [15:09] bcsaller: so I'd start out with debugger; in there before those tests run and peek at Y.one('body').get('innerHTML') [15:09] rick_h: thats good to know, but even running them in isolation seems to cause it [15:10] rick_h: container seems to populate alright and I made selection of the nodes in the callbacks goto the viewlet container rather than the viewlet-manager container (its ancestor) [15:10] bcsaller: by isolation you mean with a .only? or by commenting out the rest of the test files? [15:10] rick_h: with only/grep [15:10] bcsaller: I only figured this out after commenting out the actual index.html references and going through them debugging. [15:11] bcsaller: ah ok [15:11] hatch: I've been there, it's as steep as it looks like :-) [15:11] rick_h: maybe I'll give that a quick try [15:20] jujugui call in 10, kanban now [15:20] oh yeah it's Tuesday thanks Makyo [15:20] :) [15:20] * hatch goes to review board [15:21] * Makyo on autopilot [15:21] haha === teknico1 is now known as teknico [15:29] jujugui call in 1 [15:31] teknico: guichat now [15:49] bcsaller: so somehow button.save-constraints was hitting button.confirm ....I'm a little concerned here :) [15:51] ugh, bug trackers, where idea go to die. [15:52] rick_h: heh not ALL bug trackers [15:52] hatch: yeah, me too, we should have that talk at some point, the distance is definition and invocation of callbacks is all messed up [15:53] sure - can we bench it until after the chat with Gary? [15:53] absolutely, thats why I said at some point [15:56] bcsaller: do you need another review? [15:56] hatch: you have looked at this branch before, but if you have bandwidth another look would be good [15:56] antdillon: how are we doing? Need a hand with it? [15:57] rick_h, Almost there, just trying to workout how to get the sprites.css to have the hover styles after the normal style [15:58] antdillon: k, going to do this user testing call and afterwards I'll ping you and we can do a hangout and get it straightened out if that's cool? [15:58] rick_h, Cool, let me know [15:58] abentley: could you please have a look here https://code.launchpad.net/~adeuring/charmworld/etag-for-charm-api-view/+merge/178792 ? [15:59] adeuring: Sure thing. [15:59] gary_poster: call now? [16:00] rick_h, Done it, just updated my branch [16:02] antdillon: ok, will pull down after this call and check it out. Thanks for working through it [16:02] rick_h, Thanks [16:04] gary_poster: meeting? [16:06] adeuring: This approach to hashing doesn't seem to distinguish between a list with ['a', 'b'] and a dict with {'a': 'b'}. [16:07] abentley: oh, good catch. [16:07] adeuring: Maybe just serialize to json and hash that? [16:07] abentley: nice idea, simple and reliable. I'll do this [16:08] adeuring: Cool. You'll want to specify sort_keys=True. Other than that, I think it's a stable representation. [16:08] abentley: right [16:10] adeuring: Also, you can simplfy "if x in y: del[x]" as "y.pop(x, None)" [16:10] abentley: ah, right [16:19] jujugui, sorry I was pulled away. can we start the call in 5 minutes, at 25 after the hour? [16:19] we are already in it haha [16:20] :-) ok cool. will join then [16:22] adeuring: I think it's reasonable to call _charm_result(Charm(charm_data)) rather than _charm_results([charm_data])[0] [16:23] abentley: yeah, I thought too that thisa bit more complicated than necessary, but did not want to change existing stuuf... [16:23] adeuring: Cool. [17:09] gary_poster: so it looks like your laptop has a very directional mic - when luca was talking we could barely hear him if it wasn't pointed at him [17:09] that's pretty cool [17:09] hatch: we still need to have that other event selector call but I could use a break from video for a bit [17:09] a-greed [17:09] 20mins? [17:11] hatch: sounds good [17:11] abentley: any hints as to why "bin/migrations upgrade --init" would generate this exception? migrations.migrate.MissingExodusIndex: Exodus index "charms_pending_010" does not exist. [17:12] benji: pyc file left around? [17:13] rick_h: It's a fresh branch (I don't colocate dev branches, they are all unique snowflakes that deserve their own HD space) [17:14] hatch, ack will try to remember [17:14] gary_poster: oh and one more thing - are you going to be landing huw's branch or am I taking over at some point (just because it's getting late there already) [17:21] jujugui can I help get a second review for ant's branch? https://codereview.appspot.com/12523043 [17:21] sure [17:21] you forgot a G in your lgtm :) [17:22] hatch: hah [17:23] ugh, and he's gone [17:23] rick_h: so now that we are pulling the CSS into our own stylesheet we can drop the yui3-skin-sam and just go with the .yui3-slider-y [17:23] hatch: but I thought the sam skin was put on a container of the slider [17:23] so we need to match the specific-ness to override the default [17:24] hatch: oh hmm, not on the slider then. [17:25] well yeah we could now stop loading the yui css and then just rely on our own styling [17:29] rick_h: done - w/QA [17:30] hatch: cool [17:36] hatch: whenever you want to have that chat === schwuk is now known as schwuk_away [17:37] rick_h: with all of those styles being overridden it probably isn't much of a stretch to copy over the remaining required styles and then stop loading the yui css for the slider [17:37] bcsaller: ok guichat? [17:44] hatch: yes, we've had this discussion. However, I've lost my day to helping land this branch and chasing a charm bug. At this point, I'm letting this go, not making anything worse in the current app, and letting there remain the todo to clean up skin-sam to another day. [17:46] benji: Sounds like you didn't run prepare-upgrade first. [17:47] ahh; let me try that [17:48] abentley: that worked. Thanks! [17:50] benji: np [18:09] rick_h: alright - I'd like to create a task/card to go through the app and remove all of the yui css entirely [18:10] rick_h: you landed the branch without the user select classes :/ [18:10] rules* [18:10] hatch: ? /me goes to look again [18:11] unless I'm on the wrong diff [18:11] hatch: it's here, /me checks the diff [18:11] oh woops [18:11] sorry [18:11] I was on the wrong diff [18:11] my bad [18:12] hatch: ah ok. cool. Had me nervous I was editing and submitting two different things :/ [18:12] hatch: card added [18:15] thanks [18:59] * benji takes a really late lunch. [19:37] rick_h: bcsaller jcsackett thought you guys would find this interesting https://github.com/yui/yui3/pull/1063 [19:39] hatch: that is fantastic. [19:39] yeah could really help simplify some of our routing [19:47] yeah, not bad, function or regex makes sense to me there, but regex is a good start [19:48] bcsaller: yeah - his argument against supporting functions was that at that point you should use middleware [19:48] which I suppose is valid [19:50] hmm, I guess it can help clean up our 3 different view route points, but at the expense of complicating up the routes themselves [19:50] so 1/2 one and 1/2 another imo [19:53] yeah I actually think that the subapp routes can be distilled into a cascading series of routes instead of how it's currently done [19:53] (which is how they are supposed to work) [19:53] even a bunch of our primary routes could do the same [19:53] well I suppose we are removing like 90% of our primary routes haha [19:58] yea, if we could verify we hit once and only hit the first matching route we'd be in good shape, but those aren't the rules [19:58] well how routes are 'supposed' to work is that they are supposed to cascade [19:59] so you 'build up' your data set as you step through routes and then show the UI on the final one [20:00] I've never successfully developed that system from the start of an app [20:00] but always ends up being where they logically progress to in the end [20:10] abentley: on friday you had a bundle ingested to staging from launchpad, correct? i'm trying to view the json but get no_such_bundle [20:11] abentley: is it still there? shouldn't this be the url: [20:11] http://staging.jujucharms.com/api/2/bundle/wiki-bundle/wiki [20:11] bac: Yes. You are missing the revision. http://staging.jujucharms.com/api/2/bundle/~abentley/wiki-bundle-1/wiki [20:12] worse, i was missing the fact it wasn't promulgated! [20:21] bcsaller: so it appears that nested describes execute the code in their parents even when run in .only() mode.....very interesting [20:21] * hatch is trying to organize tests like you do [20:26] orangesquad, bac or benji: Could you please review https://code.launchpad.net/~abentley/charmworld/separate-qa/+merge/178841 ? [20:27] abentley: I'll take a look [20:27] benji: Thanks. [20:27] hatch: I tend to do that to reuse test code where I think it makes sense, but I like the simple hierarchy by itself [20:27] yeah I like how it can say these are the ghosts tests and these are the sections to which it's testing [20:28] make them easier to find/organize [20:56] abentley: your branch looks good [20:57] benji: Thanks. [20:57] benji or abentley: Branch for review for bug 1208035 -- should be quick. https://code.launchpad.net/~bac/charmworld/bug-1208035/+merge/178849 [20:57] <_mup_> Bug #1208035: TypeError Bundle takes exactly 2 arguments [20:57] bac: I'll get it. [20:58] abentley: thanks for the help earlier. i am able to access staging now. [20:58] Woot. [20:59] bac: Why are there two test_no_paths? [20:59] one charms one bundles [20:59] bac: r=me. [21:00] thanks abentley [21:02] I get nervous when a tests doesn't fail [21:06] rick_h: somehow the slider sprites got broked http://comingsoon.jujucharms.com/ [21:07] just realized it's past your EOD too :) [21:36] jujugui anyone available for a review? https://codereview.appspot.com/12567043/ [21:36] hatch: I'll look [21:36] thanks! [21:36] wb gary_poster [21:37] isn't it reeeaaaal late there? :) [21:37] hey hatch :-) [21:37] 10:37 :-) [21:37] oh ok so not too bad [21:39] bcsaller: the ghost_inspector file changes only add a describe wrapper [21:40] but bzr doesn't handle indentation changes well [21:53] bcsaller: thanks for the review - re-sending the commands at scale - did you mean env should throttle these or core should accept a list? [21:54] hatch: it could take a list, which at scale is still a race, or it could support things like 'retry units in state x' as a server side operation [21:54] I think that style API makes more sense when you have 1k units that need replacing for example [21:55] yeah I'd like to offload the list to the server, but we can batch the requests to the server right now to get around that in the mean time [21:56] if you think it's necessary [21:56] I think I'll make a ticket regardless [22:00] bcsaller, hatch, +1 on fixing the bug you emailed me about. However, I'd like it to happen after the conflict resolution and scale up/down work. We could then carve out a few days with a strong expectation that tests and docs for these things would be significantly improved during that time. Then bcsaller and/or Makyo could work on the charmbrowser bundle code (or prototype) and then bcsaller could help luca prototype the dragging thing. [22:01] that sounds like a plan - today ended up getting bogged down with some reviews/discussions and whatnot so I'm behind anyways :/ [22:01] ack hatch [22:02] gary_poster: you may also have input on this https://bugs.launchpad.net/juju-gui/+bug/1209015 [22:02] <_mup_> Bug #1209015: Unit action buttons could lock up the UI at scale [22:04] gary_poster: that sounds like what we expected, thanks [22:04] hatch, thanks. I don't think it is a burning issue, but will be one to chip away at in the coming months. Thanks for highlighting it. [22:04] bcsaller: actually caught it :) [22:05] nick complete made that read funny :) [22:05] haha woops [22:07] I have filed 26 bugs since I started....look at me go [22:07] about to be 27 [22:09] https://bugs.launchpad.net/juju-gui/+bug/1209016 [22:09] <_mup_> Bug #1209016: Right hand zoom slider handles sprited improperly [22:09] boom [22:09] jujugui need one more review on https://codereview.appspot.com/12567043/ [22:14] hey hatch, what should I do to fix viewlet-manager-footer in huwshimi's branch? Can you give me a quick patch? [22:14] umm [22:14] one sec [22:15] hatch should I just dupe the style and give it a new class name? that looks like the heart of your issue IIUC [22:16] yeah the footer style should be reverted and it moved back to where it was in the wrapper [22:16] then a new styles on that element [22:16] s/a// [22:16] cool, on it hatch, thanks [22:16] later all [22:16] np [22:16] asd [22:17] a [22:17] sorry I broke my client [22:18] did you get my last msg about reverting the style? [22:41] oh valuechange event why won't you fire for me [22:43] hatch I had to deal with that as well [22:44] hatch: there is a work-around on the tests of my branch [22:44] that took me a while to figure out as well [22:45] well it works fine in chrome but not in phantom for some reason [22:45] looking at your branch [22:45] ahhhh simulate focus [22:45] niiice [22:45] I was .focus() [22:46] that has to be a bug with YUI [22:48] thanks for figuring that out [22:51] hatch, https://codereview.appspot.com/12506044/ has my comments to Huw. review if you get a chance, otherwise np. I want to review your branch, but it needs qa, and I really should get to bed. [22:51] so will leave it for someone tomorrow. Sorry :-/ [22:51] sure np :) have a good night [22:51] ttyl, all [22:51] thanks :-) [22:54] bcsaller: so setting the id of the ghost automatically updates the element in the DOM in the app, but it's not updating the DOM in the tests [22:54]