Shibboleth Developers

[Chad La Joie <>]

I'm not sure I understand what you mean.  For a look up by entity ID
you'd have something like

<baseUrl>/<urlEcodedEntityID>

for a look up by artifact hash you'd have

<baseUrl>/{sha}<hash>

In either case you may have additional tags, but those would just be
added, so for example looking for entity ID has in the 'test'
collection might be:
<baseUrl>/{sha}<hash>+test

or, if you knew you were always going to query the test collection you
would just put the "test+" part first as part of the base URL.  What
the plugin wouldn't be able to do is figure out what additional tags
to add on the fly, at least not as I was planning.  I don't know how
it would be able to.

On Thu, Jul 8, 2010 at 21:13, Scott Cantor 
<>
 wrote:
>> Well there is a standard tag for the transform at least and I would
>> assume the client would know which it needed to do.  For the metadata
>> provider I was planning on implement I was thinking I would have just
>> have the base URL, everything up until tag, and then the code would
>> know how to construct the tag.
>
> I get that part, but I had thought that the tag syntax might come to include
> multiple tags of different types and the tag part might be more than just
> the data, but also some kind of qualifier/label for the data.
>
> So I was thinking that just a fixed base URL wouldn't cut it for tags that
> represent different data (an entityID vs. a source ID). Maybe I'm
> misunderstanding the intent?
>
> -- Scott
>
>
>



-- 
Chad La Joie
www.itumi.biz
trusted identities, delivered