- From: Cameron McCormack <cam@mcc.id.au>
- Date: Mon, 11 Oct 2010 18:52:55 +1300
- To: Garrett Smith <dhtmlkitchen@gmail.com>
- Cc: public-script-coord@w3.org
-various people Garrett Smith: > | A named constructor object that exists due to one or more > | [NamedConstructor] extended attributes with a given identifier > | is a function object. It MUST have a [[Call]] internal property, > | allows construction of objects that implement the interface on > | which the [NamedConstructor] extended attributes appear. > > Again, it is redundant to state that a function object has a [[Call]] > property. I agree that it is redundant to state that a function object has a [[Call]] property. I don’t think it is redundant to state that it has a [[Call]] property with particular behaviour, though. > It is also redundant to say that the function is an object. > Functions are objects. That is true. Are you asking for me to change instances of “function object” to just “function”? > Change every occurrence of "named constructor object" to "named > constructor". I named that term “named constructor object” analogously to “interface object”, “interface prototype object”, etc. But I can see that it is not consistent with the name of the [NamedConstructor] extended attribute, either. I’m fine with changing it to “named constructor”. > I suggest rewording that. > > [NamedConstructor] function > When a named constructor is called as a function, it creates and > initializes a new object of the interface type. Yeah, it’s probably fine to avoid talking about [[Call]] explicitly and talk rather about the behaviour of the function when called. > Question: Do you have any real examples of a NamedConstructor? And if > so, please provide one. TIA. There is one in the spec: http://dev.w3.org/2006/webapi/WebIDL/#NamedConstructor Search for “Audio”. That’s taken from the HTML5 spec. -- Cameron McCormack ≝ http://mcc.id.au/
Received on Monday, 11 October 2010 05:53:38 UTC