Regards,
Eelco
p.s. you should have CVS access allready. Just log in using ssh (extssh in Eclipse) with you SF account.
Gwyn Evans wrote:
Should I set up CVS access, or does anyone want to review patches?
Here's this one in the meantime...
Index: src/java/com/voicetribe/wicket/RequestCycle.java =================================================================== RCS file: /cvsroot/wicket/Wicket/src/java/com/voicetribe/wicket/RequestCycle.java,v retrieving revision 1.10 diff -u -r1.10 RequestCycle.java --- src/java/com/voicetribe/wicket/RequestCycle.java 29 Nov 2004 23:58:31 -0000 1.10 +++ src/java/com/voicetribe/wicket/RequestCycle.java 1 Dec 2004 11:00:43 -0000 @@ -357,8 +357,10 @@ // it means that the handling of the total request is not done yet { // detach all model objects from the page - detachModels(page); - // clear the ui messages and reset the original component models + if (page != null) { + detachModels(page); + } + // clear the ui messages and reset the original component models UIMessages.release(); } current.set(null); // reset ThreadLocal reference
------ One thing is that some of the code's using real tabs set to show as 4 spaces, while my setup was using all spaces. What's the project standard, as it's easy enough to set Intellij to whichever?
Gwyn
On Wed, 1 Dec 2004 11:53:34 +0100, Juergen Donnerstag
<[EMAIL PROTECTED]> wrote:
If I then choose the DisplayTags exmaples, then the ExampleExport, then Export to CSS, I get a Null Pointer Exception showing up in HttpRequestCycle.
What seems to be happeing is that the rendering for 0.exportCSV does a setPage(), then invokes the 'linkClicked" method on the ExportLink component, which does a setPage(null), but when the rendering gets to the 'finally' block in RequestCycle, there's no check in either the call to or the implementaton of 'detatchModels(page)', which gives the NPE.
A check, either in the call or the body, fixes the problem, but I'm not sure if it's the real problem or showing something else up - comments?
ExportLink.linkClicked() sets the page == null in order to prevent any page rendering (the response contains the export data not the page rendered). This worked fine so far. RequestCycle.render(), especially the finally block however depends since recently on a valid page. Currently I don't see other means to return the export data and prevent rendering the page. Thus, the finally block should not rely on a valid page variable.
regards
Juergen
-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now. http://productguide.itmanagersjournal.com/
_______________________________________________
Wicket-develop mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/wicket-develop
