[ https://issues.apache.org/jira/browse/WICKET-6604?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16665647#comment-16665647 ]
ASF subversion and git services commented on WICKET-6604: --------------------------------------------------------- Commit ec1549118b3c583afdfb3ef45b0e62d870f318b6 in wicket's branch refs/heads/wicket-8.x from [~ivaynberg] [ https://git-wip-us.apache.org/repos/asf?p=wicket.git;h=ec15491 ] WICKET-6604 fixed bug where controllers of enclosures were not correctly repainted in ajax if they were outside enclosures > Ajax repaint is not correctly handled when component being repainted has an > enclosure associated with it and is not a child of the enclosure > -------------------------------------------------------------------------------------------------------------------------------------------- > > Key: WICKET-6604 > URL: https://issues.apache.org/jira/browse/WICKET-6604 > Project: Wicket > Issue Type: Bug > Affects Versions: 7.10.0, 8.1.0 > Reporter: Igor Vaynberg > Assignee: Igor Vaynberg > Priority: Minor > Labels: ajax > Fix For: 7.11.0, 8.2.0 > > > When a component is repainted with ajax we first check if that component is a > controlling component of the enclosure and if it is we repaint the enclosure > instead of the component. However, we make the assumption that the > controlling component of the enclosure is always contained within the > enclosure, which may not always be true with inline enclosures. > For example: > {code:java} > <label wicket:child="name" wicket:for="name">Name</label> > <input type="text" wicket:id="name"/>{code} > In this case the inline enclosure will only encompass the label. So if we > repaint the textfield by adding it to the ajax request handler we will get a > strange result where any header contributions associated with the name > component will get rendered, but the markup for the name component will not - > because we make the assumption that since it is controlling an enclosure it > is also inside it so there is no point in repainting it directly. > > One may argue that controlling components have to be inside the enclosure and > we should throw an error if they are not, but I prefer this more flexible > approach which allows the controller to be outside. -- This message was sent by Atlassian JIRA (v7.6.3#76005)