[12:01:20] ?pepteam [12:01:20] arschmitz bethge jdalton jrossi jrossi1 jrossi2 jrossi3 jrossi9billion M4rius scott_gonzalez snover [12:01:28] hi [12:01:43] Hi [12:01:50] hi [12:02:22] I have to admit, I got wrapped up in "real work" and did not prepare at all. I just remembered the meeting four minutes ago :-/ [12:03:20] bethge: I see that you were doing more work on nested pointerenter/leave events. [12:03:20] https://github.com/jquery/PEP/pull/274 [12:04:40] Yeah, indeed, sorry going back and forth between "wip" and "ready" [12:05:57] I hope I understand correctly, that whenever a finger touches an element, all its ancestors get a `pointerenter` [12:07:36] I noticed more and more things to consider. I feel like having functional tests for these kinds of issues would be great, since they are not covered by w3c tests, afaik. [12:09:03] bethge seems like a good idea [12:09:29] If the W3C tests don't cover this, we can expand them. [12:12:53] Should we write our own tests and run them in parallel or expand the w3c tests and ask the PE wg if they want to add them? [12:13:30] I'd prefer expanding the W3C tests. [12:14:04] If it tests spec conformance, it makes sense to have it in the W3C test suite. [12:14:06] im sure the WG will be more then happy to add tests [12:15:43] Ok cool, I will cook something up for PR-274 [12:15:50] great [12:16:24] arschmitz: What's the status of https://github.com/jquery/PEP/pull/245? [12:16:32] I still get errors when I run `grunt test` with that. [12:17:29] weird i thought it was good [12:17:37] no one ever commented that they got errors [12:19:03] i see yours about node 5.3.0 but that does not look at all related to the selenium server thats an intern error [12:19:16] https://github.com/jquery/PEP/pull/245#issuecomment-181905262 [12:19:49] wasnt it 5.3.0 that broke grunt basicly [12:19:58] and fixed with 5.3.1 [12:20:15] oh yeah? [12:20:21] one of the 5.x did [12:20:27] cant remember which [12:20:39] Testing on 0.12 right now. [12:20:56] they broke path.relative in one of the 5.x releases [12:21:02] Well that worked. [12:21:19] Install 5.9.1 now. [12:21:38] 5.x also has the change to node_modules [12:21:44] which might break something in the chain [12:22:01] WHY DO YOU HATE INNOVATION?!?!?!?!?!? [12:22:03] :-P [12:22:42] Looks like selenium didn't stop even on 0.12. [12:22:53] hmm [12:23:04] i have a manual "oh shit" hook i use on a different project [12:23:20] to stop it if its running no matter what when grunt errors or exits [12:23:23] i can add that [12:23:39] i use a different package here and didnt think i was going to need it [12:23:51] i didnt get the issues with not stopping [12:24:04] bethge pretty sure you had tested this too right? [12:24:19] Seems to be working properly with 5.9.1 though. [12:24:51] Successfully ran it three times in a row on 5.9.1 just now. [12:25:36] hmm [12:25:43] i honestly never tried with 12 [12:25:45] just ran it on 5.7 with an error: Failed to load module intern/node_modules/dojo/node from ... [12:25:46] i tried with 10 and 4 [12:25:52] 5.7.0? [12:26:00] that the broken version i just found it lol [12:26:06] https://github.com/nodejs/node/issues/5383 [12:26:29] yeah, downloading 5.9.1 now [12:26:52] yeah nothing is likely to work on 5.7.0 [12:28:36] it works! [12:29:00] With 5.9.1 and deleting the node_modules folder [12:32:45] arschmitz: Do you want to check if we still need the modified shrinkwrap before we land this? [12:32:57] yeah ill double check [12:33:03] Hopefully they fixed that while we were waiting :- [12:33:13] can hope [12:40:34] Anyone have anything else to discuss? [12:54:24] oh i did [12:54:28] but got distracted lol [12:55:43] i was looking at any potential breakage from ios 9.3 [12:56:02] and realized we make an assumption that has not been correct for a long time [12:56:19] we assume that pointer event support === touch-action support [12:58:26] which did i guess save us from breaking on iOS 9.3 [12:58:31] but is still not correct [13:07:13] arschmitz: you mean pointer-event-support imples touch-action-support or touch-action-support implies pointer-event-support or both ways? [13:07:36] pointer event support implies touch-action support [13:07:40] is what we have now [13:07:54] but chrome has just touch-action for a long time [13:08:04] and now ios 9.3 has partial support [13:08:37] bethge we set touch action props [13:08:42] https://github.com/jquery/PEP/blob/master/src/touch-action.js#L26 [13:08:54] bethge exactly [13:09:09] uuh, yeah, that is dangerous [13:09:39] and just plain wrong [13:10:51] looks like we never added the manipulation value either [13:13:19] Hmm ... where do we use touch-action as a css property? [13:15:04] I thought it is mostly for handling touch input, but for touch we also read the attribute of the html node, iiuc: https://github.com/jquery/PEP/blob/master/src/touch.js#L48 [13:18:15] bethge we copy the attribute [13:18:22] to be a css rule [13:18:29] for browsers that support it native [13:18:35] is the theory there [13:20:48] stupid question: why do we do that? [13:20:48] so you dont have to write it twice in 2 different places [13:20:48] other wise you would need to do the attibute and the style [13:20:49] to cover all browsers [13:21:05] ah ok [13:21:16] but we should use native where we can instead of faking it [13:21:17] like in chrome [13:22:27] kinda comes around to what we talked about with just using the style attribute too i guess [13:23:20] and at this point i have no idea if apple has any intention of supporting values besides manipulation [14:00:43] hello! [14:01:26] Hey! It's not moderated ... well, I guess it was left open from the last meeting. [14:01:58] freedom! [14:02:02] howdy [14:02:11] anarchy! [14:02:29] Hey [14:05:21] hey sorry [14:05:33] yeah we never officially ended last meeting lol [14:05:45] sfrisk jasperdegroot agcolom [14:05:46] hi all [14:06:59] so lots and lots of activity [14:07:01] this week [14:07:11] we are down from 34 open pr's to 4 [14:07:20] YAY! [14:07:25] great work everyone! [14:07:43] Wow! [14:08:06] sweet [14:08:27] so tables are landed [14:08:36] all test are green including demos [14:09:24] nice [14:09:32] any idea what was going wrong? [14:09:41] with the demos spider I mean [14:10:36] (hey, I'm in a meeting so limited availability) [14:12:13] yes [14:12:23] so it was the url re-writing on the root page [14:12:37] we re-write all the panel and slide bar links [14:12:43] and it was not working [14:12:54] on the search panel [14:13:47] ah I see [14:14:01] on the other tests [14:14:03] so the links were only broken when using search [14:14:08] yes [14:14:10] ok [14:14:13] and only from root page [14:14:19] if you were on a demo page it worked fine [14:14:36] there were about a dozen other issues too [14:14:37] good thing we have that spider [14:14:41] broken links and what not [14:14:48] yeah I can imagine [14:14:56] so thats all fixed [14:15:16] tests we had a breakage in transitions [14:15:30] on core because they change the whole api for then / when in deffereds [14:15:34] thats fixed too [14:15:48] we got the whole test bootstrap landed [14:16:02] with a little docs page to explain how it works [14:16:14] the demos updates are all landed [14:16:24] know we need more but goot to get what he had landed [14:16:45] all the updates for page visibility and the pagecontainer classes are landed [14:16:54] nice work! [14:16:54] all apsdehal test updates are landed [14:17:30] so the only thing blocking master is still selectmenu [14:17:40] gabrielschulhof any updates there? [14:17:47] Yeah. [14:17:58] I found another pagehide handler hiding in selectmenu. [14:18:08] I removed it and I've just found the tests passing. [14:18:15] great [14:18:22] My next step is to rebase onto 1.5-dev. [14:18:28] so one more thing now is to update to use the bootstrap and rebase with 1.5-dev [14:18:34] Yeah. [14:18:45] Looking forward to the conflicts :) [14:18:55] we didnt rebase 1.5 [14:19:00] Oh, no? [14:19:02] no [14:19:09] So, wait, onto what should I rebase? [14:19:15] going to wait at this point till we go to master [14:19:17] 1.5-dev [14:19:42] we landed stuff there but didnt rebase so it should not conflict everything [14:19:46] just legit things [14:20:00] Yeah, I was planning on rebasing onto 1.5-dev [14:20:19] Didn't even think about master. [14:20:20] i dont think apsdehal updates the slectmenu tests anywhere either [14:20:25] so you shoudl be good [14:20:34] we will go to master [14:20:40] with 1.5-dev once selectmenu lands [14:20:43] OK. [14:21:01] we still have a lot of broken demos until that lands [14:21:13] Yes, I haven't touched selectmenu [14:21:35] Except the index.html file I think [14:21:38] OK, then I'll look at the other tests as a template, and I'll apply bootstrap to selectmenu. [14:21:45] i have not seen any timeouts or flaky tests since this update either [14:21:54] Good stuff! [14:22:03] gabrielschulhof https://github.com/jquery/jquery-mobile/tree/1.5-dev/tests [14:22:13] we added a readme to tests [14:22:26] Very nice! [14:22:44] oh we also now have assert.domEqual [14:22:50] gabrielschulhof: Ping me, if you have any questions related to bootstrap [14:22:53] to replace testhelper version [14:22:56] and it gives a diff [14:23:21] but im actually working on seperating that and maybe changing the api [14:23:25] arschmitz: Is that somewhere in external/? [14:23:27] so i wouldnt jump on it just yet [14:23:36] gabrielschulhof no but thats where it will be [14:23:45] OK. [14:23:46] its just a copyed file in the test/lib [14:23:48] from ui [14:24:01] but im making it a stand alone asserting extention [14:24:07] so we can bot hjust depend on it [14:24:35] gabrielschulhof ill actually be looking for some feedback similar to qunit-assert-classes [14:24:43] it will be a suite of dom assertions [14:25:10] arschmitz: Between testHelper and UI's version, I think it covers most of the common use cases. [14:25:13] so would like to figure out what scenerios we would like to cover [14:25:23] gabrielschulhof i want to remove the testhelper [14:25:44] and add things for checking different types of dom equality [14:25:46] arschmitz: I know, but as you create the extension, you can look at what testhelper does and what UI's version does and do the union. [14:26:01] Mostly testHelper is used atm for navigations [14:26:04] yeah i was also thinking things like sameNode [14:26:14] differentNode [14:26:35] arschmitz: What are the semantics of that? [14:26:54] Anyway, we can talk about this on -dev. [14:26:58] well in general we do $( "thing" )[ 0 ] === $( otherthing )[ 0 ] [14:27:01] But it's a good idea. [14:27:01] yeah [14:27:40] anyway so if anyone has ideas for useful dom assertions [14:27:43] let me know [14:28:40] so next thing [14:28:44] cogwurx [14:28:46] jasperdegroot [14:28:50] demos are looking great [14:28:54] once we are happy there [14:29:00] i would love a good website refresh [14:29:06] to call attention to new version [14:29:21] website refresh? [14:29:26] jquerymobile.com [14:29:33] fix it up [14:29:39] make it all shiny and new [14:30:02] you mean within the possibilities of the WordPress theme that we are using? [14:30:05] yes [14:30:19] would that change the other sites that use that theme as well? [14:30:30] no [14:30:36] we wouldn't change the theme [14:30:39] just the content [14:30:50] we also have theme override stuff [14:30:50] https://github.com/jquery/jquery-wp-content/tree/master/themes/jquerymobile.com [14:31:03] okay cool [14:31:04] it has its own theme seperate that inherits from others [14:31:17] yeah I did that for this PR https://github.com/jquery/jquerymobile.com/pull/120 [14:31:19] so we have a lot of flexibility within that [14:31:52] so i would really like to see that get spruced p [14:32:00] it will bring more attention to new release [14:32:18] about that PR ... maybe the team members that already submitted a text want to review it since the PR is more than a year old [14:32:21] cogwurx jasperdegroot does that sound like something the 2 of you would be interest in [14:32:30] yes lets get new members added [14:32:32] and land that [14:32:49] ?? redesign/updated theme for jquerymobile.com? [14:32:49] apsdehal sfrisk for sure [14:32:53] cogwurx yes [14:33:03] not really redesign [14:33:20] "refresh" [14:33:37] I think we are not supposed to make big changes, only like small customizations [14:33:38] yeah [14:33:39] as far as im concerned go as crazy as you want within the confines of what we can do without effecting other sites [14:33:54] and breaking layouts for common stuff [14:33:58] like the header and footer [14:33:59] I'm game... [14:34:10] but breaking things is half the fun! [14:34:30] I think we should focus on the content first [14:34:33] like we of course need to keep the ad banners and what not [14:34:43] if some content needs changes in layout/style then go for it [14:34:46] cogwurx: I totally read that as "but breaking things in half is fun" [14:34:54] lol [14:35:00] i think both apply [14:35:07] breaking things in half can be lots of fun [14:35:15] there is still enough to do to get demos in really good shape [14:35:19] espically if you get ot use like a big sword [14:35:26] jasperdegroot yes demos first [14:35:35] I like your thinking @arschmitz [14:35:35] i would want the site for the release of 1.5 stable [14:37:01] I think users prefer good/working demos and docs over a site redesign [14:37:21] jasperdegroot yes demos are first for sure [14:37:27] and docs are not optional [14:37:34] since we don't have that much time, let's just focus on those things before 1.5-stable [14:37:45] jasperdegroot there is a while for stable [14:37:50] we have not even done alpha yet [14:37:59] I know [14:38:23] and docs should be done by the people who made the changes so kinda unrelated [14:38:48] i think you and cogwurx should focus on demos then site [14:39:06] apsdehal gabrielschulhof cgack we should all be focusing on docs / bugs [14:39:13] right on... [14:39:18] sure [14:39:45] im going to focus on finishing the changelog and upgrade guide [14:39:47] then docs [14:39:48] we also have a themeroller that needs to be updated [14:39:56] what's left on demos for the design? I know I have one minor fix to do, if I can get git to cooperate [14:40:01] and I don't know about the download builder [14:40:02] arschmitz: will do. I'm afk all next week, but I'll dig into the docs/bugs for things when I return [14:40:13] jasperdegroot download build wont be happening most likely [14:40:24] which brings up next topic [14:40:40] gabrielschulhof you have an old PR open to make the build work from command line [14:40:44] we are going to need this [14:40:55] since download builder is currently a bust [14:41:30] https://github.com/jquery/jquery-mobile/pull/8067 [14:41:51] Boy! [14:42:37] That's an old one. [14:42:43] if we can at least have a command line build [14:42:55] that relieves a lot of the need for the download builder site [14:43:09] we still want it to get fixed but thats sadly a lot of work [14:43:14] in other repos [14:43:27] Well, my intention was to make a command line build work via some magical download builder so we only ever have one order of inclusion. [14:43:47] gabrielschulhof right but thats not going to happen as planned in time [14:43:53] so i think that PR could address it [14:44:10] Yeah, so we have what we have and we need that PR. [14:44:20] yup [14:44:31] OK. I can have another look at it. [14:44:49] gabrielschulhof thanks might be easyest to just cherry-pick your commits that matter at this point [14:45:08] Yeah, probably. [14:45:16] Though we haven't touched the build system much. [14:45:24] no but that because of rebase [14:45:30] has lots of unrelated commits in it [14:45:56] it should have 5 [14:46:02] but has 27 [14:46:17] just because 1.5-dev was rebased [14:46:26] at some point since last year lol [14:46:30] Oh, I see that. [14:46:49] OK, I'll see what the common base is and rebase. [14:46:59] gabrielschulhof i would just cherry pick those 5 commits [14:47:02] to 1.5-dev [14:47:12] OK, so I should just rebase/land that PR? [14:47:17] do a range cherry-pick [14:47:27] gabrielschulhof maybe do another PR [14:47:30] or update that one [14:47:34] OK. I'll do that. [14:47:52] im not sure you want the add DB to package.json either [14:48:03] i think that was you going to try to start switch to new builder [14:48:52] gabrielschulhof but lets use that as a starting point for anything else we need to do to support a decent command line build [14:50:01] Oh, dang! That's right. [14:50:04] OK. [14:50:19] ... and it adds the boilerplate which we've already added. [14:50:30] OK, so not gonna happen in five minutes. [14:52:30] yeah [14:52:43] id say that should be priority after selectmenu [14:53:14] /o [14:57:57] Alright. I need to take off. L8R all! [14:59:51] later! [15:00:00] me too... later! [15:07:18] later [15:07:20] lats wrap [15:07:24] see everyone back on dev