Mailing-List: contact dev-help@myfaces.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "MyFaces Development" <dev@myfaces.apache.org>
Message-ID: <33011797.57791283880513389.JavaMail.jira@thor>
Date: Tue, 7 Sep 2010 13:28:33 -0400 (EDT)
From: "Leonardo Uribe (JIRA)" <dev@myfaces.apache.org>
To: dev@myfaces.apache.org
Subject: [jira] Reopened: (MYFACES-2908) UIViewRoot.addComponentResource()
 adding multiple components with same ID instead of replacing
In-Reply-To: <12996088.145941283454713095.JavaMail.jira@thor>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


     [ https://issues.apache.org/jira/browse/MYFACES-2908?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Leonardo Uribe reopened MYFACES-2908:
-------------------------------------


The code that check for a duplicate component is on the right place.

It seems the previous algorithm takes into consideration facelets added components but not programatically added ones. So the fix is correct if a component is added programatically but makes facelets added component (h:outputScript, h:outputStylesheet). Right now, we have tag handlers for h:outputScript and h:outputStylesheet that prevents double addition, but if some user create a tag with a similar behavior it will fail.

> UIViewRoot.addComponentResource() adding multiple components with same ID instead of replacing
> ----------------------------------------------------------------------------------------------
>
>                 Key: MYFACES-2908
>                 URL: https://issues.apache.org/jira/browse/MYFACES-2908
>             Project: MyFaces Core
>          Issue Type: Bug
>          Components: JSR-314
>    Affects Versions: 2.0.2-SNAPSHOT
>            Reporter: Michael Concini
>            Assignee: Michael Concini
>             Fix For: 2.0.2-SNAPSHOT
>
>         Attachments: MYFACES-2908-patch.txt
>
>
> Looks like when MYFACES-2854 was committed, there were two issues it caused where we break spec compliance.  Both are around behavior that "If the component ID of componentResource matches the the ID of a resource that has allready been added, remove the old resource."
> The first is that it looks like the "else if (componentId != null)" statement is in the wrong place.  
> If a UIComponent is added with the same ID as an existing component, we'll never get to this check since we'll  be false on the isInView() check.  
> This will cause duplicate ID exceptions to be thrown if multiple objects with the same ID are added.  
> This is easily resolved by moving the else if to kick in whenever isInView() is false.
> The second issue, and the more important since this is breaking a TCK test, is that since we were only comparing to the parent's ID to the location prefix, we weren't handling the case where the same object gets added a second time after updating the target.  
> This can be resolved by changing 
>             if (componentResource.getParent() != null &&
>                 componentResource.getParent().getId() != null &&
>                 componentResource.getParent().getId().startsWith(JAVAX_FACES_LOCATION_PREFIX)) ...
> to
>             if (componentResource.getParent() != null &&
>                 componentResource.getParent().getId() != null &&
>                 componentResource.getParent().getId().equals(JAVAX_FACES_LOCATION_PREFIX + target))
> I'll attach a patch for review while I wait for our CTS team to try the change out.  I probably wont' be able to commit until after the US Labor Day holiday since I'm away Friday. 

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