Multiple instances of the same authentication flow on Shibboleth IdP

[Cantor, Scott]

> Am I correct assuming each flow auto-register itself with an
> hardcoded identifier, and won't work if configured with
> anything else ?

Yes. The changes to the Password flow were to get around having to do it in that case, being the most common one.

I did just redo the c14n layer (mostly so it doesn't need flows at all), to allow for this possibility by indirecting the key of the flow from the flow ID while defaulting it in.

I believe it may (emphasis on may) be possible to change the authn flow handling in a similar way so I can take a look at it, it would be a desirable thing to do but it would require defining theose flow descriptors.

As for RemoteUser and RemoteUserInternal, the documentation explains the differences and no, they're not implemented with the same exact code, they can't be because of what makes them distinct.

-- Scott