Function use in the IdP

[Brent Putman]

On 11/12/14 1:07 PM, Rod Widdowson wrote:
>> We don't have the latter, because it's not possible. The message contexts
>> aren't children of the PRC, they're stored in the PRC itself. Crosswalking
>> that isn't possible as far as I can see.
> Damn.  So basically if all I have is a MessageContext I cannot get the PRC.
> That would kinda indicate that we could usefully have a ScriptedFunction
> which takes a MessageContext.
>

Hmm, I always thought we could do that, via MessageContext (really
BaseContext) #getParent().  It was my intention that you could.  Just
looked. Looks like in reality the MessageContext parent isn't getting
automatically set in the InOutOperationContext ctor, like it does when
you a add context child in the usual way. I think that's just an
oversight, that constructor should do that.  It would be trivial to add
*there*.

Unfortunately, however, the PRC doesn't expose the 2 argument ctor of
the superclass InOutOperationContext.  I think that's also a (bigger)
oversight, maybe a legacy of that code starting in the IdP.  So that's
obviously a problem, since we aren't doing that.  I imagine it may be
too late to change every place where we instantiate a PRC to use a (new)
2-arg constructor - and that's assuming the message context instances
are even available at that point.  But you guys are more familiar.  If
it's just changing a few places, etc, then maybe worth doing. 

Can't believe we haven't hit upon this til now.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://shibboleth.net/pipermail/dev/attachments/20141112/7480d035/attachment.html