You can manually set an id value on all of your components. Or you can try to reverse-engineer the patch Martin put in to apply it to 1.1.1.
I doubt there's an easy solution if you are not willing to use a nightly at this point. You might also be able to downgrade your version of facelets -- certain versions of facelets worked better with certain versions of Myfaces. I don't know if there's one that works perfectly with Myfaces 1.1.1, though. On 3/3/06, Andrew Robinson <[EMAIL PROTECTED]> wrote: > Okay, I brought up this issue before, but regarding my project I am > working on at home. The problem is that when I post back to the JSF > page using the same view (no navitation rules matched - I am just > changing UI state on the current page, not going to a new page), > myfaces generates new IDs but not remembering the old IDs. So if the > new IDs are the same as the old IDs, there is an error. > > Unlike my home project, I cannot afford to be on the development > release at work. I need a stable release. Since this is a major bug, > aka show stopper, can there be a service pack, hot fix or at least a > source patch for this bug for the 1.1.1 released source code? (1.1.1b > or something?) > > This is stopping me dead in my tracks and there doesn't seem to be a > work around except for always creating a new view by making navigation > rules like: > from-view-id myview.xhtml > from-outcome * > to-view-id myview.xhtml > > This is not elegant and is not good for the server either right (no > point in having to create a new view for every post back to the same > page)? > > The issue seems centered on new components being added to the tree on > a postback to the same page. > > My page has collapsable panels that are saving their state. Instead of > rendering the children and hiding them with CSS, I am not rendering > them from the server at all if the panel is collapsed. I am doing it > this way so my backing beans on the server can remember all of the > panels' expanded state (right now there are 3 panels on this page). If > I do the CSS solution and don't post back, the user will lose the > state if they navigate away without posting my form. > > Thanks, > Andrew > > FYI - I need a solution ASAP if at all possible (meaning in the next > few hours not days), so even temporary workarounds that are not as > ugly as the navigation case one I just mentioned are appreciated as > well. >
