[jira] Commented: (XAP-80) setupClassAsSubclassOf() has some unexpected behavior

["Turyn, Michael (JIRA)" <xa...@incubator.apache.org>]

    [ http://issues.apache.org/jira/browse/XAP-80?page=comments#action_12448648 ] 
            
Turyn, Michael commented on XAP-80:
-----------------------------------

Agreed; this is a problem caused by the fact that  "this.superclass" is _not_ context-sensitive when called with call(ths,...) or apply(this,...).

If a way could be found to determine the object a function call  actually came from, we could create a superclass() method that would do the right thing.  I will give this a little time, but my guess is that  we can't, since JS goes through a lot to obscure this.  The member "arguments.callee" in every function knows which function is actually being called (it knows it's in me.prototype.prototype.foo(), not  me.foo() if  the latter doesn't exist) but I don't think there's a way of going from the member to the object that owns it.

Otherwise, I'll just go through the code and replace the this.superclass with  the explicit object references.

> setupClassAsSubclassOf() has some unexpected behavior
> -----------------------------------------------------
>
>                 Key: XAP-80
>                 URL: http://issues.apache.org/jira/browse/XAP-80
>             Project: XAP
>          Issue Type: Bug
>            Reporter: James Margaris
>
> In TableBridge I have the following:
> xap.bridges.dojo.TableBridge.prototype.init = function() {
> 	this.superclass.init.call(this);
> 	//add listener last so we don't fire for the initial set
> 	dojo.event.connect(this.getPeer(), "onActivateRow",this,"onActivateRow");	
> 	dojo.event.connect(this.getPeer(), "onSelectRow",this,"onSelectRow");	
> 	dojo.event.connect(this.getPeer(), "onDeselectRow",this,"onDeselectRow");
> 	dojo.event.connect(this.getPeer(), "onExpandRow",this,"onExpandRow");	
> }
> Now I set up a subclass of table bridge that does NOT override init.:
>  /**
>  * @fileoverview
>  * 
>  * A bridge class with dojo toolkit box panel peer.
>  */
>  
> xap.bridges.dojo.TreeBridge= function() {
> 	xap.bridges.dojo.TableBridge.call(this);
> }
> Xap.setupClassAsSubclassOf(
> 				"xap.bridges.dojo.TreeBridge",
> 				"xap.bridges.dojo.TableBridge"						
> );
> When you map a tag to treeBridge and run, you get an infinite recursion.
> If I change the init call in TableBridge to do:
> xap.bridges.dojo.DojoWidgetBridge.prototype.init.call(this);
> It works fine.
> It seems to me that what is happening here:
> We look for an init() and find one in TreeBridge that is actually the version from TableBridge
> We run that init
> The this.superclass call, since we are technically in TreeBridge, maps to the TableBridge (?) method
> Some recursion occurs ??
> We should look at how dojo works, there instead of:
> this.superclass
> You do:
> xap.bridges.TreeBridge.superclass
> Avoiding the use of "this" which has some context issues..

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: http://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira