[jira] Commented: (TAP5-128) Render phase short-circuiting is not implemented for methods of the same phase.

["Howard M. Lewis Ship (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/TAP5-128?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12643346#action_12643346 ] 

Howard M. Lewis Ship commented on TAP5-128:
-------------------------------------------

I can see a problem in this code:

 private void invoke(boolean reverse, ComponentCallback callback)
    {
        try
        { // Optimization: In the most general case (just the one component, no mixins)
            // invoke the callback on the component and be done ... no iterators, no nothing.

            if (components == null)
            {
                callback.run(coreComponent);
                return;
            }

            Iterator<Component> i = reverse ? InternalUtils.reverseIterator(components) : components.iterator();

            while (i.hasNext()) callback.run(i.next());
        }
        catch (RuntimeException ex)
        {
            throw new TapestryException(ex.getMessage(), getLocation(), ex);
        }
    }

Looks like we work our way through the full components list (mixins plus the component itself) even once the event is aborted (because a non-null/non-void value was returned).

> Render phase short-circuiting is not implemented for methods of the same phase.
> -------------------------------------------------------------------------------
>
>                 Key: TAP5-128
>                 URL: https://issues.apache.org/jira/browse/TAP5-128
>             Project: Tapestry 5
>          Issue Type: Bug
>            Reporter: Nick Westgate
>            Priority: Critical
>         Attachments: jira1662sample.tgz
>
>
> According to the docs on render phase short-circuiting for methods of the same phase:
> http://tapestry.apache.org/tapestry5/tapestry-core/guide/rendering.html
> "If a method returns a true or false value, this will short circuit processing.
> Other methods within the phase that would ordinarily be invoked will not be invoked."
> This is currently unimplemented. (Only other "inner" phases are short-circuited & not invoked.)
> Changes would primarily affect:
> org.apache.tapestry.internal.structure.ComponentCallback
> org.apache.tapestry.internal.structure.ComponentPageElementImpl
> In ComponentPageElementImpl.invoke(...), the loop:
>             while (i.hasNext())
>                 callback.run(i.next());
> needs to implement the short circuiting, for example:
>             while (i.hasNext())
>                 if (callback.run(i.next()))
>                     break;
> This change would cascade up to ComponentCallback and all its usage in ComponentPageElementImpl.
> Cheers,
> Nick.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@tapestry.apache.org
For additional commands, e-mail: dev-help@tapestry.apache.org