[09:00:05] <scott_gonzalez> jzaefferer: arschmitz fnagel1 gnarf kborchers mikesherov rxaviers tj_vantoll
[09:00:19] <rxaviers> Hello
[09:00:19] <jzaefferer> aloha!
[09:00:21] <arschmitz> Hi
[09:01:00] <tj_vantoll> Hey
[09:01:06] <fnagel1> hey
[09:01:29] <kborchers> hi
[09:01:59] <gnarf> !
[09:02:54] <scott_gonzalez> rxaviers: Wanna start with an AMD update?
[09:03:00] <mikesherov1> here
[09:03:24] <rxaviers> sure, it's not much updates actually.
[09:03:42] <rxaviers> I've ran Jörn's rename script, and
[09:04:09] <rxaviers> I've tested that the require.js build is working just fine (so, Download Builder supports it), and also that the resulting bundle has no jshint issue: https://gist.github.com/rxaviers/7757969
[09:04:22] <rxaviers> It needs more eyeballs to make sure the rename did not mess up demos, or tests...
[09:04:54] <rxaviers> I have also updated the UMD return exports of all files.
[09:05:32] <scott_gonzalez> Thanks.
[09:05:40] <rxaviers> ah! right, it needs AMD tests
[09:05:43] <jzaefferer> rxaviers: I mentioned in the PR that we need some kind of test for the UMD support
[09:05:48] <jzaefferer> exactly :-)
[09:06:03] <rxaviers> :) it's on my TODO
[09:06:28] <scott_gonzalez> No update on 1.10.4 yet. I've been getting pulled in a million directions again. But at least this time most of it is for other parts of UI :-)
[09:06:45] <scott_gonzalez> I reviewed the tooltip PR again, but found some more issues.
[09:06:50] <scott_gonzalez> https://github.com/jquery/jquery-ui/pull/1118
[09:06:58] <jzaefferer> maybe have some setup that runs an AMD loader in phantomjs, loading each widget once and trying to instantiate it using the constructor (the return value)? That way we could be sure that the returns are in place for new widgets as well, and that all dependencies are met
[09:07:51] <scott_gonzalez> jzaefferer: Why not just have an AMD loader that gets used in the normal unit tests?
[09:08:35] <rxaviers> Should we have all tests run twice? One using normal script tags, another one using AMD loader?
[09:09:13] <tj_vantoll> Seems like overkill.
[09:09:23] <scott_gonzalez> It does and it doesn't.
[09:09:34] <scott_gonzalez> It does because we assume that if we return the constructor, everything works.
[09:09:42] <scott_gonzalez> Actually, nevermind.
[09:09:48] <jzaefferer> so far we haven't discussed if and how we're going to change how we load resources in our tests and and demos
[09:09:54] <scott_gonzalez> I was going to say that the return value might break part of the AMD module.
[09:10:07] <scott_gonzalez> But if it did, it'd break the <script> version too, since the return would happen in both.
[09:10:14] <scott_gonzalez> We won't change our demos at all.
[09:10:32] <jzaefferer> tests?
[09:11:09] <tj_vantoll> If we changed the tests to load the files using AMD that provides an automatic test for their functionality.
[09:11:12] <scott_gonzalez> I think we probably just need one additional test per widget.
[09:11:53] <tj_vantoll> Presumably. that test would load the file using an AMD loader and ensure it gets the constructor?
[09:11:56] <scott_gonzalez> If we change all the tests to use AMD, that would work too.
[09:12:09] <scott_gonzalez> If the demos are using <script>, we'll quickly notice if something breaks (which is doubtful that it would).
[09:12:22] <scott_gonzalez> tj_vantoll: Yeah, that'd be the test.
[09:12:30] <jzaefferer> We already have the resource loader for unit tests, we can probably update or replace that with an AMD loader
[09:12:50] <tj_vantoll> jzaefferer: That's more or less what I was thinking.
[09:12:54] <scott_gonzalez> That would work for me.
[09:13:24] <arschmitz> not that we do it right but thats what mobile does
[09:13:45] <jzaefferer> So update or replace resource loader with an AMD loader. If that turns out to be a bad idea, add a test for each widget that loads it via AMD.
[09:14:04] <rxaviers> roger
[09:14:05] <jzaefferer> arschmitz: you're using requirejs in unit tests?
[09:14:23] <jzaefferer> also, what are you doing wrong?
[09:14:26] <arschmitz> yes
[09:14:55] <arschmitz> jzaefferer: as far as what we are doing wrong i dont recall you told us we were doing it wrong in austin
[09:15:21] <tj_vantoll> fyi here's the current resource loader UI's tests use: https://github.com/jquery/jquery-ui/blob/master/tests/unit/testsuite.js#L40
[09:15:27] <jzaefferer> uh, okay
[09:15:32] <jzaefferer> can't remember that /o\
[09:15:42] <tj_vantoll> Which essentially just document.write()'s <script> and <link> tags.
[09:15:43] <arschmitz> jzaefferer: it was about how we were using qunit
[09:15:49] <jzaefferer> oh, that
[09:17:18] <jzaefferer> have you changed it since then?
[09:17:43] <arschmitz> no
[09:17:47] <jzaefferer> this should be just <div id="qunit">: https://github.com/jquery/jquery-mobile/blob/master/tests/unit/button/index.html#L52-L56
[09:18:22] <jzaefferer> I can't remember the details
[09:18:32] <arschmitz> oh is that it i thought it had something to do with the require.js setup we were using
[09:18:57] <jzaefferer> maybe that was it, not sure
[09:19:16] <arschmitz> anyway
[09:19:25] <jzaefferer> arschmitz: got something on button?
[09:19:30] <arschmitz> yup
[09:19:56] <arschmitz> http://view.jqueryui.com/button/demos/button/checkbox.html
[09:20:27] <arschmitz> the checkbox radio widget is working and i switched to not use an icon for radio button at uGoMobi's suggestion
[09:20:41] <arschmitz> need to work out some theme / color issues with scott_gonzalez still though
[09:20:51] <rxaviers> scott_gonzalez, do you plan to release 1.11 this year or in 2014?
[09:21:06] <scott_gonzalez> Probably 2014.
[09:21:08] <rxaviers> ok
[09:21:11] <arschmitz> also started working on controlgroup / buttonset
[09:21:15] <scott_gonzalez> Though possibly a pre-release this year.
[09:21:17] <jzaefferer> What does "not using an icon" mean?
[09:21:17] <scott_gonzalez> Who knows.
[09:21:35] <arschmitz> last week when i showed it it used the bullet icon
[09:21:46] <mikesherov1> in re: to 1.11 I'd like to get effects rewrite actually done this month
[09:22:06] <arschmitz> now it uses border and and border radius to make the radio "icon" instead
[09:22:08] <scott_gonzalez> mikesherov1: That'd be great.
[09:22:22] <mikesherov1> now that draggable is in a decent place and there is DaveStein and another guy at Behance interested in working on interactions (both current and future)
[09:22:36] <mikesherov1> recruiting fresh blood
[09:22:55] <DaveStein> the irony is i also recruited you so it'd be easier to work for you here :D
[09:22:56] <jzaefferer> arschmitz: ah, okay; that's how you got rid of the subpixel offset issues?
[09:23:02] <DaveStein> so we're just blatantly using each other
[09:24:01] <jzaefferer> DaveStein: hey! You wanna get rid of these failures right here this red row? http://swarm.jquery.org/job/2052
[09:24:15] <kborchers> scott_gonzalez: wrt interactions, we probably want to consider merging interactions-polymer into interactions some time soon too, right?
[09:24:27] <jzaefferer> uh, not that job, more like http://swarm.jquery.org/job/2047
[09:25:02] <arschmitz> jzaefferer: yes
[09:25:05] <scott_gonzalez> kborchers: Is https://github.com/jquery/jquery-pointer-events ready to be used?
[09:25:59] <kborchers> if not, it's close. i kind of left it hanging when i had to get my talk ready for brazil and haven't gotten back to it
[09:26:08] <DaveStein> jzaefferer: I'll be able to fix those tests soonish. I was working on tests monday with mikesherov, but there were a few fun things we found that kept me from tackling new ones
[09:26:21] <arschmitz> jzaefferer: also on button stuff for controlgroup / buttonset i added proposed api to doc https://docs.google.com/document/d/1yFD_Y9QuuMTI4NOc-JQYRjPTrcVV52h7ZeLKLPbSj40/edit
[09:26:23] <kborchers> scott_gonzalez: i think we need to find some way to add tests to that repo
[09:26:25] <scott_gonzalez> kborchers: Isn't that a prerequisite for interactions-polymer?
[09:26:48] <kborchers> yeah, the main thing is testing ... i was testing it via interactions-polymer which is not ideal
[09:27:49] <jzaefferer> DaveStein: okay. The timeframe for this overall is pretty open, but I'd like to get us to green TestSwarm builds asap. It would be better to fix those properly instead of disabling the failing tests.
[09:28:26] <jzaefferer> arschmitz: as much as our pbworks wiki sucks: could you port that gdoc to the wiki anyway?
[09:28:30] <DaveStein> i know i didn't disable anything
[09:28:41] <DaveStein> mike just went to help someone with something quickly but i wouldn't thin he did
[09:28:48] <arschmitz> jzaefferer: yeah no problem
[09:29:09] <kborchers> scott_gonzalez: let me try to find some time this week to get things in a good place on jquery-pointer-events and we can revisit
[09:29:18] <jzaefferer> DaveStein: I mean, I'd rather disable stuff then leave it failing for another few months...
[09:29:48] <DaveStein> well in regards to 2.0 tests, i am working on that... the swarm 1.* stars mikesherov was working on with a new hire here
[09:30:04] <DaveStein> mikesherov what do you think of timeframe working with mark?
[09:30:12] <mikesherov1> yeah, I can disable the tests in IE
[09:30:51] <mikesherov1> as discussed over email, jzaefferer, we can disable draggable for the moment
[09:31:06] <DaveStein> im shedding a tear on the inside
[09:31:06] <mikesherov1> or similar
[09:31:38] <mikesherov1> I'll resolve the reds soon. I've done a lot of work to get green in the past too jzaefferer so I'm with you
[09:31:45] <mikesherov1> whatever gets us green for now
[09:32:09] <jzaefferer> we can leave it a few more weeks; still need to get rid of a non-reproducable dialog failure in Opera 15 :/
[09:32:21] <jzaefferer> or stop testing Opera 15, hoping that the issue is gone in 16 or 17
[09:32:31] <jzaefferer> stupid focus tests :/
[09:33:11] <mikesherov1> I made a focus thing that works in all browser
[09:33:21] <jzaefferer> oh yeah?
[09:33:22] <mikesherov1> it's in draggable testhelpers actually
[09:35:14] <jzaefferer> I don't see any focus-related methods in there
[09:36:31] <jzaefferer> fnagel1: did you see my TODO/comment about the width option in selectmenu?
[09:36:45] <mikesherov1> jzaefferer:let me find it
[09:36:46] <fnagel1> jzaefferer: no, not yet
[09:37:26] <fnagel1> What to we want to cover?
[09:37:29] <jzaefferer> not using hangouts lately? anyway, I couldn't find a single demo, unit or visual test that uses the width option; we should fix that along with writing the API docs
[09:37:37] <mikesherov1> https://github.com/jquery/jquery-ui/blob/master/tests/unit/testsuite.js#L184
[09:37:44] <mikesherov1> jzaefferer: ^
[09:38:03] <mikesherov1> that works all the way back to jquery 1.6
[09:38:11] <mikesherov1> 'all the way back to IE7
[09:38:12] <fnagel1> jzaefferer: no, still sick.
[09:38:15] <jzaefferer> mikesherov1: thanks. Scott and me were debugging flaky non-finishing datepicker tests the other day, this method didn't help either
[09:38:47] <jzaefferer> fnagel1: oh, okay. Hope you get better soon
[09:38:48] <fnagel1> I will add some width related tests asap.
[09:39:03] <mikesherov1> jzaefferer: yeah, it's being polluted by other tests
[09:39:17] <mikesherov1> I was trying to debug those flakes months ago :-)
[09:39:26] <jzaefferer> so much fun /o\
[09:39:27] <jzaefferer> rxaviers: update on cldr?
[09:39:45] <mikesherov1> I just never got around to finding the offending tests that felt the state unclean
[09:39:48] <mikesherov1> left*
[09:40:27] <rxaviers> Yeap. The two big topics are: Globalize-CLDR implementation vs. new datepicker; and also the cooperation work with moment.js
[09:40:51] <fnagel1> How to proceed with selectmenu's native menu option?
[09:41:13] <rxaviers> I have been answering some Iskren's questions about porting moment.js to CLDR.
[09:41:46] <rxaviers> By the way, jzaefferer, can you help to schedule a new meeting tomorrow with us from the 1st meeting that are interested in participating, and also Timo
[09:41:52] <jzaefferer> fnagel1: its not a blocker for 1.11, though eventually we should solve that, so that jQuery Mobile can adopt our selectmenu. I don't have answers for specific issues though
[09:42:38] <jzaefferer> rxaviers: I won't be able to attend anything tomorrow; maybe you invite Iskren and Krinkle for a time that works for you?
[09:42:45] <rxaviers> Ok
[09:42:57] <rxaviers> About Globalize-CLDR (aka 1.x.x), I have finished the first port/prototype work of date parse.
[09:43:13] <rxaviers> I have updated new datepicker (and $.date) to use the new Globalize 1.0.0 with success.
[09:43:26] <fnagel1> jzaefferer: arschmitz wanted to take a look at the test branch soon, perhaps its best to wait until then
[09:43:48] <Krinkle> hi
[09:43:55] <rxaviers> I have a question for you guys about Globalize-CLDR milestone.
[09:44:11] <rxaviers> We all now Globalize 1.x.x (major 1) is going to be the new CLDR-Globalize.
[09:44:19] <jzaefferer> fnagel1: yeah, seems good; feel free to ping him every once in a while, he's a busy guy ;)
[09:44:31] <arschmitz> fnagel1: yes please bug me lol
[09:44:45] <arschmitz> fnagel1: im working on 100 things and easily distracted
[09:45:06] <jzaefferer> hey Krinkle. Can you get together with Rafael and Iskren of Moment to discuss Globalize and Moment and CLDR and what role Wikimedia can play in that?
[09:45:15] <rxaviers> Considering we currently accept no more 0.x.x features, and considering the full CLDR port is a long term work. I would like to know what you think about breaking the "eternal CLDR-complete-and-perfect" (Im exaggerating) milestone into smaller (more feasible) ones? Eg. land on master as soon as we have a stable date format and parse (but, yet to be done number format and parse), etc.
[09:45:31] <Krinkle> jzaefferer: Yep!
[09:45:37] <jzaefferer> good
[09:45:42] <Krinkle> jzaefferer: Are they both on IRC?
[09:45:47] <rxaviers> Krinkle, does this time work for you tomorrow?
[09:45:50] <Krinkle> (or GitHub handles)
[09:45:58] <fnagel1> arschmitz: I know that feeling :-)
[09:46:04] <rxaviers> this time - 45min
[09:46:05] <scott_gonzalez> rxaviers: We can't break our existing integrations like that.
[09:46:05] <jzaefferer> rxaviers: how would you deal with .net and cldr stuff at the same time?
[09:46:19] <Krinkle> jzaefferer: Yes
[09:46:36] <rxaviers> Does master need to be always stable?
[09:46:48] <rxaviers> Can we land alpha versions?
[09:47:03] <scott_gonzalez> As far as I'm concerned, everything can go into master.
[09:47:22] <scott_gonzalez> It's 0.x software.
[09:47:35] <scott_gonzalez> We just can't do a release that would break the existing integration with UI.
[09:47:56] <scott_gonzalez> So spinner needs to continue to work with whatever the latest version of Globalize is.
[09:48:23] <rxaviers> Alright, yeap.
[09:48:30] <tj_vantoll> fnagel1: fyi selectmenu's docs have been merged into the 1-11 branch of api.jqueryui.com. I made a pass over the docs and I think we're good other than adding the width option.
[09:48:35] <jzaefferer> we could do a 0-x-stable branch in Globalize?
[09:48:53] <fnagel1> tj
[09:48:57] <jzaefferer> Though since we have no intention to update the old stuff, that seems pointless
[09:49:15] <scott_gonzalez> 0.x never needs back compat.
[09:49:21] <scott_gonzalez> So 0-x-stable wouldn't make sense anyway.
[09:49:57] <scott_gonzalez> Globalize can contain as many breaking changes as want.
[09:50:02] <jzaefferer> rxaviers Krinkle: I've revived the moment/cldr email thread, suggesting 17:00 GMT+1 tomorrow
[09:50:04] <rxaviers> Are we going to support Globalize-.NET along with Globalize-CLDR?
[09:50:07] <scott_gonzalez> We just can't break jQuery UI.
[09:50:14] <fnagel1> tj_vantoll: ok, I just thought about work a little on that. Good to know I need to pull another branch. Thanks for the hint
[09:50:16] <scott_gonzalez> No, we will never support .NET again.
[09:50:31] <rxaviers> hehe
[09:50:39] <rxaviers> ok
[09:50:55] <rxaviers> cool, excellent, ok then.
[09:51:15] <jzaefferer> so merge CLDR branch to master, but don't release anything until it works with spinner (and datepicker)?
[09:51:24] <rxaviers> jzaefferer, and it's really revived. I hope things go well on the cooperation.
[09:51:32] <jzaefferer> that way it should be much more obvious that we have no intention to change the current implementation
[09:51:33] <Krinkle> jzaefferer: OK
[09:51:57] <jzaefferer> actually by 17:00 I might be able to attend as well, not sure yet
[09:52:22] <tj_vantoll> fnagel1: No problem. One more thing, the demos on the docs site won't work unless you update these references to the git versions: https://github.com/jquery/api.jqueryui.com/blob/master/entries2html.xsl#L12-15
[09:52:43] <rxaviers> guys that's all from my side. We're almost running over (it's 8mins left). So, feel free to jump to other sections if you need to.
[09:53:19] <rxaviers> Krinkle, I will ping you pvt to know your availability tomorrow
[09:53:48] <fnagel1> tj_vantoll: Ahh ok, I will try to remember :-)
[09:54:26] <scott_gonzalez> Anyone have anything else to discuss right now?
[09:55:24] <scott_gonzalez> Ok, guess that's it for today. See everyone back in -dev.
[09:55:26] <scott_gonzalez> Thanks everyone.