Yes this wasn't just an attempt. This made a measurable speed difference
when callInherited was written in AS. Seems safe to remove it now --
especially if we can run a test that shows that the speed is comparable.
A
On May 17, P T Withington wrote:
> This should work, but personally I'd like to eliminate these special
> cases. I think that these were done in an attempt to speed up start
> up (by avoiding our slow runtime callInherited mechanism), but this
> should not be necessary any more because the runtime mechanism is
> quite a bit more efficient (it does not have to loop through the
> class heirarrchy).
>
> On 2006-05-16, at 22:21 EDT, Henry Minsky wrote:
>
> > The code at the start of LaszloView.as does a manual call to
> > "super" as
> > follows below.
> > Should this just be expected to work with the new class substrate?
> > Seems
> > like it ought to
> > work OK... I'm just doing testing now with a doppelganger LzView
> > class.
> >
> >
> >
> >
> >
> >
> > var LzView = Class( "LzView" , LzNode );
> >
> > LzView.prototype._nodeconstruct = LzNode.prototype.construct;
> > //--------------------------------------------------------------------
> > ----------
> > //Base level constructor for views. See <method
> > //classname="LzNode">construct</method> for more on this.
> > //
> > // @keywords protected
> > //--------------------------------------------------------------------
> > ----------
> > LzView.prototype.construct = function ( parent , args ) {
> > //this.callInherited( "construct" , arguments.callee ,
> > parent ,args );
> > this._nodeconstruct( parent ? parent : canvas, args );
> >
> >
> >
> > --
> > Henry Minsky
> > Software Architect
> > [EMAIL PROTECTED]
>
> _______________________________________________
> Laszlo-dev mailing list
> [email protected]
> http://www.openlaszlo.org/mailman/listinfo/laszlo-dev
>
_______________________________________________
Laszlo-dev mailing list
[email protected]
http://www.openlaszlo.org/mailman/listinfo/laszlo-dev
