Forums

I'm not sure I agree (or maybe I don't understand the context). You mean in reference to your follow up point about getter/setter support being rolled into ES5? I don't think that warrants the "too heavy because it was too ambitious" classification.

Widget's goal is to define a pattern which separates State from UI, and binds them through state change events, because this was considered to be useful for designing easily customizable/maintainable/extendable widgets (based on the YUI 2 experience, and evolution of YUI 2 Widgets). Is it this goal which you find "too ambitious"?

If we're just talking about how the implementation could be optimized, that's a different story. For example, now that most browsers have caught up with getter/setters as a potential way to centralize all the code around the setting/getting particular piece of object state, we may want to look at how attribute could delegate to the native impl where possible.



I'm not sure how that crept onto stage, but it's not supposed to be there. Sorry about the confusion.

I imagine it got left in, by "forgot to search/replace" mistake, with the TreeView documentation push.

Since we haven't gotten enough feedback/coverage on the WidgetHTMLRenderer approach it's not broken out as in independent module as part of 3.6.0 GA. The intention was to push it out as treeview-htmlrenderer along with TreeView as the first use case (as mentioned in the tickets/design docs : http://yuilibrary.com/projects/yui3/ticket/2531369, https://gist.github.com/1386836).