On Thu, 7 Dec 2006, Cyril Concolato wrote: > > Is the ECMAScript syntax in the 'implementation' element free like: > > <implementation> > foo = 1; > </implementation> > > or shall/should it obey some specific ECMAScript syntax (as all the > examples in the specification suggest), i.e. start with ({, contain only > fonction declarations, and end with }). The spec requires that whatever is executed evaluate to an object. So I guess you could do something like: <implementation> var x = new Object(); x.foo = 1; x; </implementation> > If the syntax is free, please say so in the specification. If not, > please give link to specifics constructs allowed from the ECMAScript > specification. This would satisfy me. This is already covered by section 5.4 as far as I can tell. (Specifically, 5.4.1.) On Fri, 8 Dec 2006, Marcos Caceres wrote: > > Cyril raises an interesting point there regarding implementations and > how they should be handled by scripting langauges within the context of > XBL. For ECMAScript, it might be nice to write the constructor without > using the old ({ ... }). Can I do the following?: > > <implemenation> > //where "this" is the scope of this implementation > this.x = 1234; > this._y = 4321; > this.funky = function(){...} > this._helper = {( functionName: function() { return true; })} > </implemenation> "this" in this context would be "window", so that wouldn't work. But things like the example above would, as far as I can tell. (I'm no JS expert.) -- Ian Hickson U+1047E )\._.,--....,'``. fL http://ln.hixie.ch/ U+263A /, _.. \ _\ ;`._ ,. Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'Received on Friday, 5 January 2007 23:50:56 GMT
This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:10:21 GMT