myfaces-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Daniel Young" <dyo...@synyati.com.au>
Subject RE: Jsf_tree_64
Date Wed, 31 Jan 2007 05:57:09 GMT

Cool, that makes sense - thank u so much :)

-----Original Message-----
From: Mario Ivankovits [mailto:mario@ops.co.at] 
Sent: Wednesday, 31 January 2007 3:55 PM
To: MyFaces Discussion
Subject: Re: Jsf_tree_64

Hi Daniel!
> As Simon said, it's basically server-side state saving with a single
> client-side token.  But normal server-side state saving already has a
token
> (the SESSION ID!!) that maps to the HttpSession anyway... so what's the
> difference??
>   
The thing is to not only find the correct session, but also the correct
request.
It's required to support the browser-back button where, due to the page
flow, the system awaits to find a JSF state for the page. If you do not
maintain a request id you'll always could have the last JSF state (for
the last page), any browser-back will fail.

MyFaces will store a configureable number of JSF states, say for the
last 20 requests. After then, the browser-back issue is back.


Ciao,
Mario


Mime
View raw message