W3C home > Mailing lists > Public > public-script-coord@w3.org > October to December 2009

Re: [WebIDL] Trying to understand IndexGetter/NameGetter

From: Jason Orendorff <jason.orendorff@gmail.com>
Date: Wed, 21 Oct 2009 17:38:39 -0500
Message-ID: <bb8868b90910211538h7455f848kf98e534bc35a8108@mail.gmail.com>
To: Boris Zbarsky <bzbarsky@mit.edu>
Cc: public-script-coord@w3.org
On Wed, Oct 21, 2009 at 1:19 PM, Boris Zbarsky <bzbarsky@mit.edu> wrote:
> Proposal:  Trying to create an indexed property outside the range supported
> by the IndexSetter throws.  If there is an IndexGetter but no IndexSetter,
> the relevant range is empty.
>
> This has an obvious drawback: throwing in situations where UAs currently do
> not.  Said UAs don't interoperate very well on such cases, though (see
> http://lists.w3.org/Archives/Public/public-script-coord/2009OctDec/0042.html)
> so it might be that the compat impact is low.
>
> Obvious benefit: not having to define how user-defined properties have to
> interact with IndexGetter/Setter stuff, since there is no interaction
> between the two anymore.

Well, there is if the range changes, right? Maybe I don't understand
the proposal. What range would be supported for a DOMNodeList?

I think we have to cope with "real" ES5 properties interacting with
"virtual" properties, but I don't think it has to be complicated.
Suppose all real properties simply shadow all virtual properties, and
virtual properties shadow the prototype chain. That leaves a bunch of
little issues, but are there any really troubling problems?

-j
Received on Wednesday, 21 October 2009 22:39:07 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 8 May 2013 19:30:02 UTC