[ 
https://jira.jboss.org/jira/browse/RF-2393?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12441292#action_12441292
 ] 

Nick Belaevski commented on RF-2393:
------------------------------------

Mike,

Can I ask you for a simpler test case? The one attached is somewhat complicated 
for investigation and debugging and will take a lot of time.



> Tree nested inside node of an outer tree has strange behaviour on ajax submit 
> when there are validation errors
> --------------------------------------------------------------------------------------------------------------
>
>                 Key: RF-2393
>                 URL: https://jira.jboss.org/jira/browse/RF-2393
>             Project: RichFaces
>          Issue Type: Bug
>    Affects Versions: 3.1.4
>         Environment: Glassfish v2, JSF RI 1.2
>            Reporter: Mike Hanafey
>            Assignee: Nick Belaevski
>         Attachments: NestedTree.tar.gz, NestedTree.war
>
>
> I have situation where there is an outer tree whose nodes have an inner tree. 
> Specifically, the outer tree is over DNA sequences, and each DNA sequence may 
> have a tree of comments attached to it.
> Both the inner and outer trees have a couple of faces, one is used when a 
> node is in display mode, and the other when it is in edit mode.
> If the inner tree is put into edit mode at a node (via an ajax request) and a 
> validation error occurs on a second ajax request to submit the edits, the 
> whole inner tree for that node disappears. It can be recovered by doing a 
> browser redisplay. If there is no validation error, the edit works as expected
> Looking at the XmlHttpRequest response when there is a validation error shows 
> that the response does not contain the expected tree nodes.
> When this same comment tree code (both the backing beans and the facelets 
> views) is applied on an unnested tree this problem does not occur. 
> A test war is attached (the war is big only because it includes the library 
> dependencies -- the test code itself is small).
> To demonstrate, deploy the war and look at the top page.
> Click on the + next to "Comments" (this is the un-nested tree). Click "Ajax 
> Save". Observe the "Value is required" messages.
> Enter something into the first textfield, and do the "Ajax Save". Now a 
> custom validation message appears.
> Now for the first strangeness. Enter something in both the textfield and the 
> textarea. The custom validation message appears on both inputs, but now the 
> inputs just made are erased. When only one of these fields have data, the 
> data entered in the other field is preserved, and I believe the clearing of 
> input data is another error, but not the one in the title of this this issue.
> To see the main issue, open the tree labeled "Sequences". This has a comments 
> tree just like the one above nested inside the sequence node.
> Click on the + next the "Comments" under "Sequences".
> Click on "Ajax Save". Now instead of seeing the "Value is required" error, 
> the inner tree node collapses. Click on the "Submit" button, an reopen the 
> "Sequences" tree to see that the inner tree is in fact still intact.
> Click on "Ajax Immediate". Now the tree does not collapse. The only 
> difference is this button is immediate, so the validation steps are not done.
> Finally, clicking on "Plain Save" shows the expected validation errors (now 
> the Sequence tree collapses after the submit so it must be manually opened, 
> but I guess this is expected behavior because the open / close state is not 
> maintained across the request). The problem where the inputs that fail 
> validation are erased remains even with this non-ajax update mode).

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: 
https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira

        
_______________________________________________
richfaces-issues mailing list
[email protected]
https://lists.jboss.org/mailman/listinfo/richfaces-issues

Reply via email to