[
https://issues.apache.org/jira/browse/TRINIDAD-2110?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Scott O'Bryan resolved TRINIDAD-2110.
-------------------------------------
Resolution: Fixed
Fix Version/s: 1.2.15-core
Added a check for the Bridge's hack for Mojarra 1.2
> workaround for jsf portlet bridge 2.0 nullviewstatetoken
> --------------------------------------------------------
>
> Key: TRINIDAD-2110
> URL: https://issues.apache.org/jira/browse/TRINIDAD-2110
> Project: MyFaces Trinidad
> Issue Type: Bug
> Components: Portlet
> Affects Versions: 1.2.14-core
> Reporter: Scott O'Bryan
> Assignee: Scott O'Bryan
> Fix For: 1.2.15-core
>
>
> There is a bug related to a workaround in the MyFaces Portlet Bridge 2.0 R.I.
> After a goLink renders a new portlet view, the bridge changes a null
> VIEWSTATE parameter value to the string
> "org.apache.myfaces.portlet.faces.nullViewState". The reason this is done is
> that apparently there is a bug in Mojarra where, during a post, if the
> VIEWSTATE is null, the render phases are skipped. Unfortunately this
> workaround in the bridge cannot be removed in the bridge.
> As a result of this, however, our check for the format of the viewStateToken
> incorrectly returns the VIEWSTATE encoding which rely's on a null VIEWSTATE
> to make part of it's decision. We need to check for this
> "org.apache.myfaces.portlet.faces.nullViewState" token as well.
> Also, this workaround in the bridge is ONLY in bridges which operate on JSF
> 1.2. The MyFaces Portlet Bridge 3.0 (which is based off JSF 2.0) has this
> hack removed. As such, this bug fix should only apply to JSF 1.2 branches
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
