Doesn't it make more sense to override WebPage's setHeaders(WebResponse) method? You get the response as a param instead of pulling it from the request cycle and the method name is just right. _____
From: Richard Nichols [mailto:[email protected]] To: [email protected] Sent: Thu, 23 Sep 2010 17:47:24 -0400 Subject: Re: chrome + wicket ajax + back button = problem? Take a look at http://www.richardnichols.net/2010/03/apache-wicket-force-page-reload-to-fix-ajax-back/ On Wed, Sep 22, 2010 at 12:58 PM, Ryan Crumley <[email protected]> wrote: > All, > > I ran into strange behavior involving wicket ajax updates + chrome + back > button. I am pretty sure the issue is not caused by wicket but I am hoping > the problem might sound familiar to someone and they can point me in the > right direction. The scenario goes like this: > > - An ajax request is made that updates the DOM. > - The user navigates to another page by clicking on a link. > - The user uses the browser back button to go back to the original page. > > After clicking back Google Chrome shows the page as it was originally > rendered (before the DOM update). All other browsers show the page as it > was > when the user left the page (with the DOM update). Besides the possibility > of showing stale data the other problem is that the page state has been > altered by the ajax request and links that were originally rendered may no > longer be valid. > > I found a few links where people are having similar issues: > > > http://www.maintaino.com/nuts-and-bolts/2010/04/08/teaching-chrome-not-to-cache-your-rails-pages/ > > Possibly related to using jQuery and/or window.onUnload() and "bfcache": > > > http://stackoverflow.com/questions/158319/cross-browser-onload-event-and-the-back-button > https://developer.mozilla.org/En/Using_Firefox_1.5_caching > > In addition to using Wicket 1.4.12 I am also using jQuery 1.4.2 and google > analytics on these pages. I don't think any unload handlers have been added > however my only check was looking at all the attributes on the "window" > object via the inspector. > > Has anyone else seen similar behavior? Even better, anyone else have a > solution? You may be seeing errors like this one in your logs: > > org.apache.wicket.protocol.http.request.InvalidUrlException: > org.apache.wicket.WicketRuntimeException: component wid1:wid2:1:wid3 not > found on page com.foo.xxx > > > Thanks! > > Ryan > -- Richard Nichols http://www.richardnichols.net/ :: http://onmydoorstep.com.au/
