- From: Dimitri Glazkov <dglazkov@google.com>
- Date: Tue, 6 Jan 2015 10:54:38 -0800
- To: Anne van Kesteren <annevk@annevk.nl>
- Cc: WebApps WG <public-webapps@w3.org>
Received on Tuesday, 6 January 2015 18:55:05 UTC
On Tue, Jan 6, 2015 at 10:38 AM, Anne van Kesteren <annevk@annevk.nl> wrote: > On Tue, Jan 6, 2015 at 7:06 PM, Dimitri Glazkov <dglazkov@google.com> > wrote: > > Yes to the first question. I wasn't planning on doing anything different > > there. > > It seems simple prototype munging but not actually changing identity > will fail for all but the basic cases of subclassing. E.g. if I > subclass an HTMLInputElement I would expect it to have an internal > [[value]] slot of sorts, but since the element created is a plain > HTMLElement whose prototype is munged that will not be the case. > Right, that's why to create a valid custom element that subclasses HTMLInputElement, you should use type extensions. With type extensions, the HTMLInputElement is what's instantiated. :DG<
Received on Tuesday, 6 January 2015 18:55:05 UTC