Subj : Strange Rhino NativeString behavior
To   : netscape.public.mozilla.jseng
From : Matt McShane
Date : Sun Oct 24 2004 12:31 am

Hello,

I am having trouble understanding some Rhino behavior, hopefully someone 
here can explain for me.

I have added a function to Object.protoype, let's call it myFunc. Here 
is the Java implementation:

public static Object myFunc(Context cx, Scriptable thisObj,
                               Object[] args, Function funOb)
{
Object someAttributeValue = thisObj.get("someAttribute",thisObj);
System.out.println(someAttributeValue.toString());
return Context.getUndefinedValue();
}

I am simply assuming that the object has a property called 
"someAttribute" and printing it out (I took out null checks, etc. for 
clarity)

This works for the first two of the following tests but not the third.

var obj = new Object()
obj.someAttribute = "xyz"
obj.myFunc() //prints xyz

var date = new Date()
date.someAttribute = "xyz"
date.myFunc() //prints xyz


var msg = "Hello, world."
msg.someAttribute = "xyz"
msg.myFunc() //prints o.m.j.UniqueTag@1027b4d: NOT_FOUND

I understand why the someAttribute property in the NativeString test 
isn't found - it's because the instance of NativeString on which the 
property is "put" is different than the instance passed as thisObj to my 
myFunc. In other words, Rhino is creating a different instance of 
NativeString with new NativeString("Hello, world.") before passing it to 
myFunc. But I don't know why this is. There may be a good reason to use 
different instances - but why aren't properties copied over?

Thanks,
Matt.

.