<html>
<head>
<meta content="text/html; charset=windows-1252"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<p><br>
</p>
<br>
<div class="moz-cite-prefix">On 6/9/16 3:10 PM, Cantor, Scott wrote:<br>
</div>
<blockquote
cite="mid:9846A6064BD102419D06814DD0D78DE11294163B@CIO-KRC-D2MBX08.osuad.osu.edu"
type="cite">
<blockquote type="cite">
<pre wrap="">Apparently the schema is different for the v3 port.
</pre>
</blockquote>
<pre wrap="">
Hmm, you're right, I'm wrong. I guess the schema was changed. Or more to the point, the old type in the extension schema was added to the SAML relying-party schema that the IdP comes with for all the other profiles. Strange choice, I don't remember us deciding to do that, it seems odd to me.</pre>
</blockquote>
<br>
It's complicated. IIRC the support for the legacy config schema and
parsing was in 3.0 from the beginning. Chad did that when he
originally ported things for v2. I had to fix up some stuff when we
actually added runtime support for all that. <br>
<br>
<blockquote
cite="mid:9846A6064BD102419D06814DD0D78DE11294163B@CIO-KRC-D2MBX08.osuad.osu.edu"
type="cite">
<pre wrap="">
Anyway, I do see a SAML2SSOSProfile type in the "urn:mace:shibboleth:2.0:relying-party" namespace. But sure enough, I don't see a mapping in the code to handle that type. I think it's a bug. But it may be the bug was actually that we added it to the schema at all, Brent would have to comment there.</pre>
</blockquote>
<br>
You do? I'm not seeing that... I see it defined for
"urn:mace:shibboleth:2.0:relying-party:saml" in
shibboleth-relying-party-saml.xsd. <br>
<br>
<blockquote
cite="mid:9846A6064BD102419D06814DD0D78DE11294163B@CIO-KRC-D2MBX08.osuad.osu.edu"
type="cite">
<pre wrap="">
The general line we took is that anything that wasn't part of the V2 code base wouldn't be accounted for in the legacy relying-party support, and would require migrating to the native version. This is straddling the fence, but is to my mind something I would have argued was on the extension side and would not have been supported.</pre>
</blockquote>
<br>
Well, Chad had already done it before we even released 3.0. I think
it was just sitting in there without any of us realizing it. So I
had to make it work.<br>
<blockquote
cite="mid:9846A6064BD102419D06814DD0D78DE11294163B@CIO-KRC-D2MBX08.osuad.osu.edu"
type="cite">
</blockquote>
<br>
</body>
</html>