If you're using JSF 1.2, it is more stable. -- Adam
On 5/29/07, Luka Surija <[EMAIL PROTECTED]> wrote:
How stable is Trinidad 1.2 comparing with 1.0.1? Luka Adam Winer wrote: > The Trinidad 1.2 code sends javax.faces.ViewState (the 1.2 spec > requires this), but the trunk doesn't, instead sending > org.apache.myfaces.trinidad.faces.STATE. If you're using JSF 1.2, > you could (and should) use the Trinidad 1.2 code. > > -- Adam > > > > On 5/29/07, D. Cardon <[EMAIL PROTECTED]> wrote: >> Thanks for your response, I looked into it and found out that when >> G4JSF builds its request, it >> send the javax.faces.ViewState variable in the request. Do I need to >> have it send something >> different for Trinidad? >> >> --David >> >> --- Adam Winer <[EMAIL PROTECTED]> wrote: >> >> > If there's no tree, I suspect that the state parameter is >> > not getting submitted. Check the G4JSF code to see if >> > it has hardcoded a list of request parameters to submit. >> > >> > -- Adam >> > >> > >> > On 5/29/07, D. Cardon <[EMAIL PROTECTED]> wrote: >> > > Hi all, >> > > >> > > I'm using MyFaces Trinidad and am integrating the Google Web >> Toolkit (GWT) to work along with >> > it. >> > > I am using the G4JSF project along with some of my own >> modifications. So far, everything on >> > the >> > > client side is working just fine, but I am having difficulty >> communicating with the server >> > when >> > > the GWT fires an event. Since the G4JSF project works just fine >> without Trinidad, I figured a >> > > little enlightenment on Trinidad's workings would be very helpful. >> > > >> > > Here's the part of the process that is having problems: >> > > >> > > 1. The G4JSF project latches correctly onto the GWT event, which >> is correctly parsed from the >> > > client's request. >> > > >> > > 2. G4JSF implements a phase listener, which propagates the GWT >> event after the Restore View >> > phase >> > > has completed. >> > > >> > > 3. After the restore view phase completes, the phase listener's >> code correctly runs, but the >> > > component tree is not "restored". That is, there IS a view root >> available, but it has no >> > > children. So, when "invokeOnComponent" is called on the >> facesContext.getViewRoot(), the GWT >> > event >> > > never makes it to the GWT component. Also, I noticed that the >> > > >> org.apache.myfaces.trinidadinternal.context.AdfFacesPhaseListener.POSTBACK >> variable in the >> > request >> > > is set to 'false'. >> > > >> > > Now my questions: >> > > >> > > - Shouldn't the component tree be completely constructed by the >> end of the Restore View phase? >> > > - If not, is there a better way to propagate the GWT event to a >> function in the backing bean? >> > > - Since the G4JSF client-server communication works fine in JSF >> without Trinidad, how does >> > > Trinidad modify the process? >> > > >> > > Thanks for any help! >> > > >> > > --David Cardon >> > > >> > > >> > > >> > > >> ____________________________________________________________________________________Choose >> the >> > right car based on your needs. Check out Yahoo! Autos new Car >> Finder tool. >> > > http://autos.yahoo.com/carfinder/ >> > > >> > >> >> >> >> >> ____________________________________________________________________________________Need >> a vacation? Get great deals >> to amazing places on Yahoo! Travel. >> http://travel.yahoo.com/ >> >
