**** BEGIN LOGGING AT Mon Dec 15 02:59:59 2014
Dec 15 03:53:14 <DougReeder> Hey, enyos, I’ve defined a kind that looks normal to me: https://gist.github.com/DougReeder/af13fb82ff352adac792
Dec 15 03:53:59 <DougReeder> But when it’s instantiated by another kind, I get errors: https://gist.github.com/DougReeder/af13fb82ff352adac792
Dec 15 03:54:36 <DougReeder> (ack, the kind is at https://gist.github.com/DougReeder/85b7b5c2518c6629c973
Dec 15 03:55:25 <DougReeder> Any idea why I get the warning “enyo.kind.inherited: unable to find requested super-method from -> undefined” ?
Dec 15 05:31:30 <Roy__> EricBlade: I'll be there tomorrow, too.  Come see me.  How are you packaging it?
Dec 15 05:32:17 <Roy__> DougReeder: Still on 2.4?
Dec 15 05:33:26 <Roy__> DougReeder: Also, it's probably not a good idea to override the create method of InputDecorator that way.
Dec 15 05:33:54 <DougReeder> Yes, ATM we’re still on 2.4
Dec 15 05:34:32 <Roy__> I believe the issue is you've overridden the create method and it's bound to your other kind.
Dec 15 05:34:40 <DougReeder> I get the same error if I wrap the InputDecorator in a generic Control and try to add a create method to that.
Dec 15 05:34:40 <Roy__> Not to inputdecorator.
Dec 15 05:35:05 <DougReeder> Bound to which kind?
Dec 15 05:35:25 <EricBlade> Roy__: I'll be in sometime between noon and 1 probably.  It's going to be quite difficult for me to mess with as my travel PC runs Win 7 though
Dec 15 05:35:40 <Roy__> that's when I get in, too.
Dec 15 05:35:59 <Roy__> EricBlade: Get a better travel computer. :D
Dec 15 05:36:20 <Roy__> DougReeder: can you make a fiddle?
Dec 15 05:36:22 <EricBlade> well, when the company eliminates dependencies on win7/ie8 ...
Dec 15 05:36:49 <DougReeder> I tried, but I couldn’t get it to render.
Dec 15 05:38:24 <Roy__> I use VMWare on my Mac.
Dec 15 05:38:28 <Roy__> EricBlade: ^
Dec 15 05:40:10 <EricBlade> theoretically, i could travel with the Linux machine but I've already filled it's entire disk capacity without having any VMware on it
Dec 15 05:41:54 <EricBlade> so, anyway, what i do is i just do a "cordova build windows" and then i open the cordova built project in visual studio, and then run the debugger on it
Dec 15 05:43:12 <EricBlade> i'm not too worried about debugging that part of it, though, i can do that when i get back from the office. .. if you want a really tough task ;) i can show you that i'm getting a totally screwed up app from the minification process ..
Dec 15 05:44:28 <EricBlade> no errors reported, it's just not displaying one of my panels
Dec 15 05:46:57 <Roy__> Hmm, I didn't know Cordova did a target Windows.
Dec 15 05:47:01 <Roy__> Is that Win 8?
Dec 15 05:47:09 <DougReeder> I’ve done a fiddle at http://jsfiddle.net/dougreeder/2L24bLkg/
Dec 15 05:47:13 <Roy__> If so, then yes, there are some special considerations.
Dec 15 05:47:26 <Roy__> I think we have some write-up on it on the wiki.
Dec 15 05:47:31 <Roy__> On github.
Dec 15 05:47:34 <Roy__> Or else in the docs.
Dec 15 05:47:43 <Roy__> http://enyojs.com/docs/latest/developer-guide/deploying-apps/enyo-apps-on-windows-8.html
Dec 15 05:47:45 <EricBlade> yes, that is a win 8 target ..
Dec 15 05:47:51 <Roy__> Did you read that?
Dec 15 05:48:44 <Roy__> If it's not helpful/up-to-date, please let me know.
Dec 15 05:49:35 <Roy__> DougReeder: Have to add Enyo to it.
Dec 15 05:49:44 <EricBlade> well, basically, what i think cordova did was built that project for me, then i used it to launch the app from visual studio :-)  so, what happened, was that it would blow up almost instantly on startup, complaining that we were feeding insecure data to innerHtml
Dec 15 05:49:48 <DougReeder> Just did; reload
Dec 15 05:50:27 <Roy__> DougReeder: Still loading the old version, if forcing refresh.
Dec 15 05:50:32 <Roy__> Did you update base?
Dec 15 05:50:45 <Roy__> rather, set as base
Dec 15 05:51:13 <DougReeder> I have now.
Dec 15 05:52:10 <Roy__> Don't see any exception now.
Dec 15 05:52:30 <EricBlade> Now, that was running in 8.1, despite having a target as 8.0, I don't know if that might've made a difference.  Or if my app is causing some "insecure" html to go through the innerHtml that it didn't like
Dec 15 05:52:34 <Roy__> contacts.textTypeRepeater.create(): email New email address Object {container: enyoConstructor} undefined
Dec 15 05:53:26 <Roy__> EricBlade: As I recall, you have to set something so it'll allow the insecure javascript.
Dec 15 05:53:27 <DougReeder> that’s the normal log message
Dec 15 05:53:30 <Roy__> Not positive.
Dec 15 05:53:50 <DougReeder> Looks like this reduced example doesn’t evoke the error.
Dec 15 05:54:03 <Roy__> DougReeder: You probably need to set some rows in the repeaster.
Dec 15 05:54:13 <Roy__> s/repeaster/repeater/
Dec 15 05:54:13 <synergybot> Roy__ meant: DougReeder: You probably need to set some rows in the repeater.
Dec 15 05:54:19 <DougReeder> Oh, right no data in the collection
Dec 15 05:55:00 <EricBlade> Roy__: the microsoft docs on it didn't provide any obvious way of getting around it, other than making sure to run any html through window.toStaticHTML() before giving it to innerHtml
Dec 15 05:55:22 <Roy__> EricBlade: Could be a platform detection problem.
Dec 15 05:56:32 <Roy__> EricBlade: This should prevent the problem:
Dec 15 05:56:33 <Roy__> 	// override setInnerHtml for Windows 8 HTML applications
Dec 15 05:56:33 <Roy__> 	if (typeof window.MSApp !== 'undefined') {
Dec 15 05:56:33 <Roy__> 		enyo.dom.setInnerHtml = function(node, html) {
Dec 15 05:56:33 <Roy__> 			window.MSApp.execUnsafeLocalFunction(function() {
Dec 15 05:56:33 <Roy__> 				node.innerHTML = html;
Dec 15 05:56:34 <Roy__> 			});
Dec 15 05:56:34 <Roy__> 		};
Dec 15 05:56:35 <Roy__> 	}
Dec 15 05:56:44 <EricBlade> so, what i did to get it to run, was i stuck in some if(window.toStaticHTML) { return window.toStaticHTML(...) }'s in HTMLStringDelegate's generate*Html functions
Dec 15 05:56:47 <Roy__> Can you see if window.MSApp is defined.
Dec 15 05:59:41 <EricBlade> yes, that exists.
Dec 15 06:01:40 <DougReeder> Ok, now the error manifest in the fiddle
Dec 15 06:09:06 <EricBlade> so.. does enyo.dom.setInnerHtml somehow override all the places in enyo that do "element.innerHTML = ..." ?
Dec 15 06:10:01 <EricBlade> (if it's supposed to, it doesn't seem to)
Dec 15 06:10:47 <EricBlade> i have it working now, so long as i don't click inside an input. clicking inside an input causes a crash in lang.js .. but since i'm using minified source, i can't really tell where it is
Dec 15 06:18:24 <EricBlade> something that minifies down to for(var l in n)e=n[l],a[l]!==e&&(r.exists&&!e||r.ignore&&t[l]||(r.filter?!r.filter(l,e,n,t,r):0)||(t[l]=e)); ... it's blowing up in the "e = n[l]" section with "Invalid Argument".
Dec 15 06:21:13 <EricBlade> looks like in the non-minified source, that's probably occurring in enyo.mixin, for(var key in src) { val = src[key]; } ... which doesn't look like something capable of blowing up
Dec 15 06:23:41 <EricBlade> the non-minified source is not throwing any kind of an error.. in IE.  i'm going to have to spend some time that i don't have right now to figure out how to get the non-minified source to package as a Win app
Dec 15 06:25:43 <EricBlade> minified source doesn't throw an error in IE, either.. so.. some kind of an error in WinJS
Dec 15 06:27:13 <EricBlade> Aha! Windows debugger.
Dec 15 06:27:43 <EricBlade> Accessing "currentPoint" of a PointerEvent is not allowed.
Dec 15 06:29:00 <EricBlade> which means simply iterating over a PointerEvent will cause an application crash.
Dec 15 06:44:59 <Roy__> Is Enyo doing that?  Nothing should really do that.
Dec 15 06:45:25 <EricBlade> yes, it is
Dec 15 06:45:36 <EricBlade> enyo.mixin !
Dec 15 06:50:20 <EricBlade> ok, so here we go.. when the input receives a "click" event, enyo clones the event to make a "tap" event.  enyo.clone() does a "return enyo.mixin({}, base)" where base is the event being cloned.  This causes enyo.mixin() to iterate the event, which causes a crash.
Dec 15 07:03:20 <EricBlade> So, unless following the Object.create() path in enyo.clone() gets around that failure, I don't see an obvious way through that without explicitly checking windows && object is a PointerEvent && key === "currentPoint" and then bypassing it.  i wonder if there are any other properties of other things, and a list of them
**** ENDING LOGGING AT Tue Dec 16 02:59:58 2014