Cameron McCormack wrote:
Cameron McCormack:
* In section 6, I don’t think it’s necessary to explicitly mention
undefined, since it’s already handled by the annotation in the
IDL. If you do want to include this in the prose, I think it
needs to be qualified to say that this applies to an ECMAScript
language binding of the interface. (null’s OK, since you can
talk about null at the level of IDL values so it’s applicable to
any language.)
Lachlan Hunt:
I don't see why this is a problem. Technically both null and undefined
are handled by the IDL, but stating it implicitly in the prose makes it
clearer.
I don’t mind it being repeated.
I got convinced on IRC that the redundancy isn't good and so I've now
removed the statement entirely.
I've also updated the IDL to include the [ImplementedOn=] extended
attribute and replaced the preceding prose about it being implemented on
those interfaces.
Finally, I updated the Terminology and Conventions section to define
that the interfaces referred to in the spec come from DOM3Core.
http://dev.w3.org/2006/webapi/selectors-api/
--
Lachlan Hunt - Opera Software
http://lachy.id.au/
http://www.opera.com/
