Hi Werner Ok, now I understand. Right now I'm working on enhance myfaces test, but I'll dedicate some time to this one as well.
Leonardo 2010/7/13 Werner Punz <[email protected]> > Am 13.07.10 07:49, schrieb Leonardo Uribe: > > Hi Werner >> >> This seems to be an undocumented feature. I didn't knew @form exists. It >> doesn't appear on the draft of jsf 2.0 rev A. Anyway, we have to >> implement it and tell to EG that is not documented. >> >> regards, >> >> Leonardo >> > Hi Leonardo, sorry for the misunderstanding, I will describe the issue more > clearly here, it is not @form (which is part of the spec btw, see below) but > a nav case fails, if done within a ppr context. The issue is simply that if > you navigate from page1 to page 2 within a nav case it works but as soon as > you want to navigate back you are stuck on the on page2 if you have ppr > enabled for the navigation, (a render viewroot is issued however correctly, > but in the second page just for second page again) > > the second navigation case within ppr simply fails, the @form stuff is > something which is resolved on the javascript side and is unrelated to the > issue. > > > Since navigation is vital ppr and outside of ppr I marked this issue as > stopper, I simply think 2.0.1 should not have this bug anymore, because the > overall quality of the build is so much better. > > The @form attribute is btw. defined in the secion 14.2.2 of the jsf2 spec > and is javascript only, so no worries that you never encountered it. It just > means the element defined is the enclosing form, in case of execute it means > the entire form is encoded and sent to the server, on the server side it > never shows up. > > > So here is the exact description of the failure > > page1 navigate to page2 via an action and implicit navigation -> page2 > shows -> page2 should navigate back to page1 via an action and implicit > navigation -> page2 is rerendered again. > > This only happens if you use f:ajax on the commandLinks and in myfaces. > Mojarra correctly switches back to page1. > > I will debug into it I assume it is a bug in the nav handler which has to > treat the ppr case slightly differently, maybe I will find it. > > > > > > > Werner > > >
