Thanks Patrick for that info. How does this work if there are no forms on a page?
Will Tapestry ever change this behavior. This seems like a really important feature to support in the framework. I know HLS was influenced by WebObjects and I know it does support this bahavior. Thanks E --- Patrick Casey <[EMAIL PROTECTED]> wrote: > > What I ended up doing was rolling my own persist > layer. The tapestry > persist layer persists on the session/page level, > which falls apart if you > have multiple instances of the same page up e.g. > > List of Dogs > Right Click -> Open in new Window -> Poodle > Right Click -> Open in new Window -> Rotwilder > Click on poolde form. > Name it muffy. > Push save. > > Congrats, your rotwilder is now named muffy. > > I ended up having to put a hidden keyfield atop > each and every one > of my edit forms and then bootstrapping the form > back based on that key. > Intitially I was just stuffing the ID value of > whatever was getting edited, > but it turns out in practice you usually need to > persist more than that e.g. > what page do I redirect to *after* a save? The user > expects to go back to > the list, etc. > > So what I do is generate a "pesistencecontext" > object for each form > and then save it in a LRU cache in the session (with > a unique key). This key > to this context goes out with each edit form on > *top* of the form so it > rewinds first, and then I rebuild form state in the > setPersistKey() > function. > > Sort of awkward to describe, but once you get it in > place it works > pretty well. > > --- Pat > > > -----Original Message----- > > From: Edward Scanzano [mailto:[EMAIL PROTECTED] > > Sent: Wednesday, August 17, 2005 11:20 AM > > To: tapestry-user; [EMAIL PROTECTED] > > Subject: Please expand on this > > > > Hi All (and Dario), > > > > How am I supposed to access the value of a field > on > > the form from within the java code if it is not in > a > > form post? > > > > I find it hard to believe that when Tapestry > persists > > the state of a page it does not attach it to an > > instance of a page. This makes supporting backing > a > > real pain and basically my responsibility. > > > > Do I have this right. > > > > > > > > > > > > This is expected behavior. If your parameters are > > persisted, then they > > will have the last value they were set to, > regarding > > of how many > > "backs" you perform. > > > > You need to use Hidden fields to avoid this. > > > > > > On 8/10/05, Edward Scanzano <[EMAIL PROTECTED]> > > wrote: > > > Hi all, > > > > > > I have a page when have a number of components > that > > > are passed parameters which are persisted in the > > page. > > > This page has an initialization method this is > > called > > > as follows. > > > > > > CWorkspace page = (CWorkspace) > > > cycle.getPage("CWorkspace"); > > > page.setup((Container)item,null); > > > > > > So far, no problem. > > > > > > The user then performs a click with in one of > these > > > components and the same page is called on this > time > > > with a new object passed to the setup method. > > > > > > So far, no problem. > > > > > > Now, I back on the browser to the first page and > > when > > > I click on another link, the components are > seeing > > the > > > data passed to them from the second page. It is > as > > if > > > the page did not have its persistant variables > reset > > > for page 1. They remain for page 2. > > > > > > What is causing this and how should I fix it. > > > > > > Thanks > > > E > > > > > > > > > > > > > > > > > --------------------------------------------------------------------- > > To unsubscribe, e-mail: > [EMAIL PROTECTED] > > For additional commands, e-mail: > [EMAIL PROTECTED] > > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: > [EMAIL PROTECTED] > For additional commands, e-mail: > [EMAIL PROTECTED] > > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
