NameID as a Function of RP

Cantor, Scott cantor.2 at osu.edu
Tue Jul 12 09:44:26 EDT 2016


On 7/12/16, 9:38 AM, "users on behalf of Marvin Addison" <users-bounces at shibboleth.net on behalf of marvin.addison at gmail.com> wrote:

> Thanks Scott, that was very helpful. Since both since the source attribute and name ID
> generator apply exclusively to Google, I used a common activation condition reference in
> both places. The following wiki page had all the examples I needed:

It's a slight optimization to do it in both. If you have a conditional attribute, running the generator won't take long, it will just bail out on the missing attribute, but I guess if it logs anything ugly that would be another reason to bypass it.

My own personal style is that if it's a piece of data I would want out of the resolver anyway, I'd just generate it as a standard attribute I use for other things and then conditionally run the NameID generator.

Or I would if I had migrated my NameID generation out of the resolver anyway.

-- Scott




More information about the users mailing list