"xxx : this.yyy" in prototype

I did not find any related subject in discussions or specs so here it is :

var f=function(name) {
     this._name=name;
};

f.prototype={
     get nodeName() {return this._name},
     get tagName() {return this._name},
     get anotherNameInvention() {return this._name},
     ...
     set nodeName() {},
     set tagName() {},
     set anotherNameInvention() {},
     ...
};

So in that case we have to define several getters and setters for 
something that is refering to the same object, slowing down the overall 
script execution, it would probably be better to have something like :

f.prototype={
     nodeName <operator> this._name,
     tagName <operator> this._name,
     anotherNameInvention <operator> this._name
     ...
}

var g=new f('test') --> g.nodeName is equal to g._name

While calling new the <operator> would indicate that g.nodeName must be 
equal to g._name whether _name is a property or a method.

nodeName, tagName, etc could be defined in f but this does not solve the 
case if we want them to be inherited properties.

Maybe no <operator> is required and the default behavior should be the 
one described above (what could be the use of declaring something like 
xxx : this.yyy in prototype ??? I have never seen it).

Example of use : js w3c dom implementation where objects own a lot of 
redundant properties.

Regards

A. Vitte

-- 
jCore
Email :  avitte at jcore.fr
Web :    www.jcore.fr
Webble : www.webble.it
Extract Widget Mobile : www.extractwidget.com
BlimpMe! : www.blimpme.com

Declaring something like a:this.b in prototype is of no use (because in 
all cases it will be equivalent at the end to declare something like a:c 
where c=("this" bounded object).b).

Intuitively, if we forget the specs and current behaviour, we could 
think that doing such, "this" could have a chance to refer to the new 
created object after the "new" statement, but of course it is not the case.

Instead we have to define getters and setters as mentionned below.

This can be compared to the var that/self=this trick in terms of 
repeatability.

Something that is useless could be transformed to something usefull, but 
right now I don't see a simple way to do it that creates more benefit 
than harm.

Regards

A. Vitte

Le 15/02/2012 19:21, Aymeric Vitte a écrit :
> I did not find any related subject in discussions or specs so here it 
> is :
>
> var f=function(name) {
>     this._name=name;
> };
>
> f.prototype={
>     get nodeName() {return this._name},
>     get tagName() {return this._name},
>     get anotherNameInvention() {return this._name},
>     ...
>     set nodeName() {},
>     set tagName() {},
>     set anotherNameInvention() {},
>     ...
> };
>
> So in that case we have to define several getters and setters for 
> something that is refering to the same object, slowing down the 
> overall script execution, it would probably be better to have 
> something like :
>
> f.prototype={
>     nodeName <operator> this._name,
>     tagName <operator> this._name,
>     anotherNameInvention <operator> this._name
>     ...
> }
>
> var g=new f('test') --> g.nodeName is equal to g._name
>
> While calling new the <operator> would indicate that g.nodeName must 
> be equal to g._name whether _name is a property or a method.
>
> nodeName, tagName, etc could be defined in f but this does not solve 
> the case if we want them to be inherited properties.
>
> Maybe no <operator> is required and the default behavior should be the 
> one described above (what could be the use of declaring something like 
> xxx : this.yyy in prototype ??? I have never seen it).
>
> Example of use : js w3c dom implementation where objects own a lot of 
> redundant properties.
>
> Regards
>
> A. Vitte
>

-- 
jCore
Email :  avitte at jcore.fr
Web :    www.jcore.fr
Webble : www.webble.it
Extract Widget Mobile : www.extractwidget.com
BlimpMe! : www.blimpme.com