CAS attribute naming

[Cantor, Scott]

On 6/12/17, 4:54 PM, "users on behalf of Marvin Addison" <users-bounces at shibboleth.net on behalf of marvin.addison at gmail.com> wrote:

> To that end, please speak up if you need additional controls on what gets returned. The attribute ID seemed like a reasonable
> choice absent other requirements.

I was hesitant to suggest the obvious, because nobody really wants to have to add a whole other set of AttributeEncoders to the resolver, we need to come up with a separate and easier-to-configure strategy for that. I think that's better than adding encoders inside out for every protocol. It's conceptually ok and it's clean enough, but it's just a hassle.

-- Scott