Rapha�l Luta wrote:
<snip>
> > Yeah.. that is a good point.  Maybe in user PSML we should have the
> > equivalent of a name in the registry PSML.  Maybe an id="" but the XPATH
> > id() would be the default.  I wonder if this is provided in Castor?
> >
> > This would work.
> >
> 
> That's what I was thinking, with the additionnal requirement that the id should
> be unique across all PSML files since from this ID the PortletFactory should be
> able to locate exactly the corresponding portlet description, so what we need is
> something like:
> 
>         pid = concat( <PSML-handle>, id() )
> 
> BUT, this kind of id() is stale as soon as there's a change in the PSML, this may
> cause issue in shared PSML scenarios such as:
> 
>         User 1 views User 2 PSML file
>         User 2 adds a portlet to its PSML file
>         User 1 clicks Maximize on a portlet of User 2 PSML
> 
> With the proposed pid, there may be an error since the portlet may have changed
> its XPATH id() and we don't have a built-in recovery method in the id().

Yes... I know.  This is a problem.  I don't know if there is an adequate
way around this.  We should probably just reset everything if the user
PSML changes.  There are no clean ways around this :(

Kevin

-- 
Kevin A Burton (e-mail: [EMAIL PROTECTED], UIN: 73488596, ZKey:
burtonator)
http://relativity.yi.org
Message to SUN:  "Please Open Source Java!"
To fight and conquer in all your battles is not supreme excellence;
supreme 
excellence consists in breaking the enemy's resistance without fighting.
    - Sun Tzu, 300 B.C.


--
--------------------------------------------------------------
Please read the FAQ! <http://java.apache.org/faq/>
To subscribe:        [EMAIL PROTECTED]
To unsubscribe:      [EMAIL PROTECTED]
Archives and Other:  <http://java.apache.org/main/mail.html>
Problems?:           [EMAIL PROTECTED]

Reply via email to