[09:03:11] not sure where jzaefferer is [09:03:35] rworth is out of town [09:04:14] danheberden: any infrastructure news? [09:04:24] haven't heard anything from clarkbox in weeks [09:04:53] scott_gonzalez was looking at the .it stuff, but no - haven't heard from clarkbox either [09:05:05] he's been offline too [09:05:23] i'll be looking at the bugs server this afternoon [09:05:37] ok [09:06:41] hey guys [09:06:53] daylight saving switch got me miss meetings, but I'm here now! [09:07:48] Jörn can haz meetings now? [09:08:01] yeah [09:08:17] I'm going to assume there are no docs updates right now [09:08:46] well we got somewhere with the learning site deployment stuff, so that might prove useful for docs later? [09:09:00] so a baby update [09:09:43] yay [09:10:09] on to testing [09:10:18] I implemented .simulate( "focus" ) and .simulate( "blur" ) [09:10:33] based on the discoveries gnarf and I made last week [09:10:49] that fixed the autocomplete and tooltip tests [09:11:23] I've also been working on getting TestSwarm to be completely green [09:11:26] getting very close [09:11:39] just some effects failures left in IE [09:11:53] that's pretty cool stuff [09:11:55] which seem to not have existed until I fixed every other failing test :-/ [09:12:04] except that one [09:12:51] im here, runnin a lil late [09:12:53] Dave is still working on draggable and droppable [09:13:06] good work too it seems :) [09:13:31] yeah, and I was even online at the same time as him last night [09:14:37] if you haven't heard, Everett Zufelt will be doing an accessibility review next week [09:14:37] zufelt.ca/blog/jquery-ui-accessibility-review [09:15:38] hoping to takes Hans' work into account that we didn't land yet [09:16:11] what's outstanding besides tabs? [09:17:31] high contrast mode, which affects a lot of widgets [09:17:37] slider maybe? [09:18:16] well Everett is blind, so he won't notice that :-P [09:20:15] BrowserStack is finalizing their API [09:20:20] so we should have something soon [09:20:35] any other updates? [09:21:11] also autocomplete, Hans had built something that got screenreaders to announce the menu opening [09:21:24] something like "x suggestions available" [09:21:51] by adding a (hidden) element with that text and aria live region attributes [09:22:12] yeah, that's what Drupal does [09:22:57] ah, okay [09:23:02] maybe they took his code :p [09:25:39] i'm still working on Menu, Menubar and Popup ... a little sidetracked with some client work lately but hoping to dig back in soon [09:25:56] still having trouble getting Menu to work with tables [09:27:15] events get screwy causing some weird visual stuff to happen (adding/removing styles at the wrong times) [09:29:01] it works with arbitrary markup though, right? [09:29:13] and non-direct-child menu items? [09:30:19] arbitrary markup, yes ... non-direct-child it should but i need to test [09:31:49] ok, if those work, then let's just move on without table support for now [09:31:54] we can come back to it later [09:33:18] ok ... looks like non-direct-child is a little off ... but should be an easy fix. i'll get that working and move on [09:33:26] great [09:33:32] anything else? [09:35:56] oh, danheberden: jquery-ui-git.js is still stuck on a build from Oct 7 [09:36:09] ok, i'll give it a kick [09:36:49] thanks [09:38:03] I guess that's it [09:39:10] no table menus sounds very reasonable [09:39:19] we're messing with tables enough already on grid... [09:40:58] yeah, i think it just gets messy because to get them to look like non-table menus you have to make tbody(if it exists) and tr's display:block; which messes with the flow and event bubbling gets crazy [09:41:15] specifically in IE [09:42:20] yeah, let's don't do that [09:45:57] grid meeting in 15 minutes then, this is probably the longest agenda we had for grid in a long time [09:46:10] most is progress updates though [10:00:45] hey [10:04:13] anyone around? [10:04:23] yah [10:05:47] everyoe else was just here... ;) [10:05:50] hey brado23 [10:06:02] boris not around? [10:06:17] dunno...let me check... [10:06:40] He doesn't seem to be around. [10:07:49] well, not much so say about template then [10:08:06] we (briefly) discussed our wiki updates on the call last week already [10:08:32] sorry - got held up... [10:08:37] hey borismoore [10:08:46] haven't actually started yet [10:09:07] ok, cool :) [10:09:56] so yeah, we should talk about the Tmpl proposal on the wiki, both API and syntax [10:10:12] Richard and me couldn't quite agree on nested tags, so those are still missing [10:10:23] I'm afraid I haven't been able to get to it [10:10:37] should be schedule another meeting to review it? [10:10:46] *should we? [10:11:00] I will be doing a commit today or tomorrow, with some changes we could look at [10:11:18] yes, but let me ping you when I have had cycles to look at that aspect [10:11:32] so, not actually schedule yet...:) [10:11:52] sure, but from my perspective, the major blocker in saying what it is the project wants to have as template was the lack of a spec that worked for us [10:12:03] sure [10:12:04] Mike's work was way too academic to be actually maintainable for us [10:12:24] I'm getting close to something I'll document on the wiki [10:12:34] okay, then check it out, and we can review together afterwards [10:12:34] that you can then review per your requirements [10:13:03] okay [10:13:08] sure [10:13:16] sounds good [10:13:17] we kicked out a lot of implementation specific detail from that page [10:13:26] I saw [10:13:29] makes sense [10:13:31] so depending on what it is you wanna document, repo readme or separate wiki page might be a better spot [10:13:54] yeah, i'll choose 'the best place' [10:13:59] alright [10:14:13] Dataview, brado23 still hasn't delivered his nitpick review :p [10:14:24] I did some updates, promoting dataviewlocal to the ui/ folder [10:14:40] yes, I'll do that today. [10:14:57] getting some unit tests in place for preloader, which didn't help so far, didn't yet get my finder on where the race conditions happen [10:15:09] once I figure that out, tests should help fix it [10:15:40] (I'll go on with the rest of the agenda, just stop me if there's something you wanna discuss) [10:15:58] Menugrid doesn't exist anymore, instead we now have two separate extensions for sorting and filtering [10:16:10] sorting is simple enough that we can consider merging it into grid itself [10:16:19] though that would go against the idea of the "zero feature grid" [10:16:51] I've chatted with Wes Bos, who we met around the ARIA Hackathon in Toronto, first task would be to give the slideshow demos some actual visual design [10:17:02] he may start on that in two weeks, hopefully sticks around [10:17:08] we can use some prettier demos [10:17:32] the Pager got an update, it can now be added without having to have existing markup [10:17:41] the useless prevStep/nextStep methods are gone [10:18:08] there's now a prototype for scroll-based paging, exercising the dataview in an interesting way: http://view.jqueryui.com/grid/grid-spf/products-scrolling.html [10:18:31] so far its just a custom dataview generating infinite amount of random data, but should work just as well with actual remote data [10:18:53] editing/selectable now uses a dialog to edit rows: http://view.jqueryui.com/grid/grid-editing/grid-selectable.html [10:19:27] we've got some nice progress on Editable, it looks a lot better now: http://view.jqueryui.com/editable/demos/editable/buttons.html (click 'Edit all') [10:19:46] Mask and Timepicker are pretty close to land in master, working on some more demos [10:20:02] Selectmenu is not quite as close, not real unit test coverage yet [10:20:11] need to replace various visual tests with unit tests [10:20:16] that's all I got [10:21:14] you've been busy! [10:21:55] I can update on "observer"...if now's the right time. [10:21:58] done a lot of reviewing end of last week, implementations on the train on sunday... [10:22:22] also Felix, Rafael and Corey did a lot of work on custom inputs [10:22:32] so yeah, forgot about that on the agenda [10:22:35] gr8 [10:22:42] please go ahead with observer [10:23:02] Yes, so I updated jquery.ui.grid.js to use "observer". I'm pretty happy with the result. [10:23:25] Fixed the remaining grid-editing demos too to remove custom binding (in favor of $observer). [10:23:44] Here's a link to an observer demo...to give a sense of how it works: [10:23:46] http://brado23.github.com/jquery-ui/grid-editing/observer.html [10:24:34] It probably makes sense to take a look at the grid diff....I'll grab an URL here... [10:25:42] k [10:25:50] https://github.com/brado23/jquery-ui/commit/c82ab8cd85adfb67ae8fb80fd8b86a49bfbcb240#ui/jquery.ui.grid.js [10:25:57] also an explanation of that demo would be helpful [10:26:15] sure [10:26:59] If you view source on the demo, you'll see a JS literal for a list of customers. The customers are heavily nested (array nesting objects nesting arrays...). [10:27:21] right [10:27:26] gnarf, you around? [10:27:27] When you run the demo, you'll see a sequence of observable edits at various levels of the model. [10:27:32] yea [10:27:39] reading over this diff :) [10:27:43] cool [10:28:08] ...along with the as-spec'd edit event....along with a path down to the array/object that was edited. [10:28:12] brado23, before we go on: [10:28:18] sure [10:28:30] how is this different from or relates to data binding concepts like jsviews implements, or sproutcore or knockoutjs [10:28:31] ? [10:28:49] seems like it addresses same or at least similar issue [10:28:54] good question [10:28:54] or does it? [10:29:09] would help me understand the concept better [10:30:50] $.observer is similar to JsViews, et al, in the sense that it is a helper that gets folks out of the business of dynamic binding to the various atoms in a model.. [10:31:02] okay [10:31:52] let me poke around a bit at some details: [10:31:58] It's different from JsViews, etc, in the sense that it doesn't dictate a model for recomputing UI....that's up to the client of $.observer. [10:32:06] looking at: $.observable( customersObserver.getAt(0).data ).property( "age", 35 ); [10:32:22] could that be this? customersObserver.getAt(0).property( "age", 35 ); [10:33:08] JsViews is to do with declarative binding of UI to data, with live binding for changes. So it listens to deep changes, when required, and provides the UI binding [10:33:18] observer is a programmatic layer, [10:33:36] which could be used by JsViews to replace some of its current code [10:33:45] so observer is not coupled to the UI, but abstracts the bindings below? [10:33:52] - the deep binding aspect [10:33:56] yes, that's right. [10:34:04] that's what I am seeing [10:34:12] but Brad can confirm [10:34:35] yes. It's just a way to demux editing events from the N atoms in a model up to some single listener. [10:34:41] I haven't seen this before, so I am not looking at the observer implementation [10:35:15] As conceived today....observer and observable are decoupled. [10:35:28] they only share editing events as a common dependency [10:35:37] and it does the bind/unbind when you add remove from arrays... [10:35:40] ? [10:35:46] yes [10:35:53] wanna resume on Skype? [10:35:55] or add object members [10:35:59] yup [10:36:05] lemme find a quite spot [10:36:14] it maintains a parallel tree of observers [10:36:27] ...parallel to the model... [10:36:44] but does it not use observable internally ? [10:36:49] ( i can't do skype today, but go ahead w/o me ;) ) [10:36:59] or is that only the events that you really need? [10:37:01] no, it only uses editing events internally [10:37:13] that's all that's needed [10:37:36] observer is only for reasoning over events...not making edits. [10:37:44] observable is for making edits [10:38:29] brado23: wanna join on Skype? [10:38:45] yeah...one sec [10:39:14] sigh....slow Skype... [11:26:19] we'll likely cancel the grid meeting next week [11:44:17] i wont be here for meetings next week [11:45:06] out of town in austin [11:47:44] Boris likely isn't around, I'll be at a conference [11:48:05] dunno Richard's status for next week, but there's probably not much going to happen on Monday/Tuesday [19:23:15] anybody there [19:31:16] hello , i wanna help , how i can change values of get ex: example.com/?p=home i wanna way way to change p=index