[
https://issues.apache.org/jira/browse/TRINIDAD-2000?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Blake Sullivan updated TRINIDAD-2000:
-------------------------------------
Resolution: Fixed
Fix Version/s: 2.0.0-alpha-2
Status: Resolved (was: Patch Available)
Resolved in revision 1058396
Trinidad-2000 partial response writers should handle nesting of CDATA blocks.
Patch to fix cases where nested calls are made to startCDATA in the
ResponseWriter (which will cause the CDATA block to be closed prematurely when
the nested endCDATA is called) by ignoring all but the outermost CDATA block.
Also fixes places where Trinidad was outputting the start and end CDATA tokens
in text to instead output them using startCDATA/endCDATA on the ResponseWriter
> partial response writers should handle nesting of CDATA blocks.
> ---------------------------------------------------------------
>
> Key: TRINIDAD-2000
> URL: https://issues.apache.org/jira/browse/TRINIDAD-2000
> Project: MyFaces Trinidad
> Issue Type: Bug
> Components: Components
> Affects Versions: 2.0.0.4-core
> Reporter: Pavitra Subramaniam
> Fix For: 2.0.0-alpha-2
>
> Attachments: jira-2000-II.patch, uiDebugInJSF.zip, uiDebugInTrin.zip
>
>
> The issue is with the use of <ui:debug> tag especially when it's placed
> inside a PPR-ed component (in Trinidad 2.x). The JSF UIDebug component
> encodes the script for this feature as follows [1] and this causes issues
> especially when rendering a partial page response as it's already inside
> another CDATA section.
> This used to a problem in Mojarra as well, but I could not reproduce this
> with a vanilla JSF app. Turns out this issue had been resolved recently in
> JIRA issue 1463 [2]. The fix was made to the RI's response writer
> (HtmlResponseWriter) so when they write out a script or a style element they
> make a note of whether they are already inside a CDATA section and if so
> remove the (nested) CDATA block from [1].
> But the above fix does nothing for a Trinidad application because we
> implement our own partial response writers. So this bug is to implement a
> similar fix.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
