Walden,

See below, please.

On Wed, Nov 19, 2008 at 2:29 PM, walden <[EMAIL PROTECTED]> wrote:
>
> Olivier,
>
> I'm still a little perplexed, see below.
>
>> >>  * session expiration, because the GWT RPC will fail soon (401).
>> >>  * forbiden because the GWT RPC will fail soon (403).
>>
>> When session is expired, the RPC will fail soon with a 401 (Auth
>> required status), before GWT 1.5 it was not (easily ) possible to
>> detect such failure. But session expiration is not an issue for HTTP
>> basic.>>  * activation of widget when authority is granted.
>
> Originally, I thought your points were against HTTP auth, but now it
> looks like they were for it?
>

I'm not talking of HTTP Basic Scheme where AFAIK there is no
expiration. I'm talking of Session Base mecanism like Acegi or Form
Based authentication.

What I was (trying) to explain is that when relying on a previous
authentication, then the GWT application is in fact unaware of being
under a restricted access. That might be a good (as it simple). But
when an error (security errors Auth Required (401) when session has
expired , a forbidden access (403)) occurs on a GWT-RPC call  the GWT
application has to handle this error (much simpler under GWT >= 1.5).

So the GWT application has to handle some security concern (Auth
required && Forbidden).

>
>>
>> About widget activation && authorization, I my proposal the widget are
>> aware of the authentication events so they can activate/desactivate
>> when login/logout occurs.
>
> This doesn't come up for me.  I secure my site in such a way that you
> don't get any widgets until you're authenticated and authorized.  I
> thought you were referring to a more fine grained authorization scheme
> where certain widgets appear only for certain users.

I do ! Some GWT element may be notified for the authentication event
(granted authorities) and then they can do what they want ...

>That sort of
> entitlement management goes beyond authorization, and the point I was
> making was that it seems somewhat orthogonal to what protocol you use
> for auth.
Definitively !
>
> Walden

Regards
Olivier.

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"Google Web Toolkit" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to