Re: [keycloak-dev] UMA 2.0
Depends on amount of work. If it's not to much extra work I'd prefer to
have backwards compatibility for a while to allow users to migrate then
remove in the last 3.x release. If that is a lot of work then we should
just remove it.
On 14 June 2017 at 14:50, Pedro Igor Silva <psilva(a)redhat.com> wrote:
Hi,
I would like to review our UMA implementation (which is based on v1), and
get it aligned with the new version, v2.
One of the main changes we need is that now UMA has a specific grant type
that should be used by clients to obtain RPTs. The Authorization API no
longer exists.
Other changes are basically related with parts of the specs we are missing
that don't really bring issues for people already using UMA in Keycloak.
But new features and better UMA support.
My question is if it is reasonable to have those changes in 3.2.0.CR1 and
how ? For instance, if we decide to have those changes in, specially the
new UMA grant type, should we keep/deprecate the legacy Authorization API
for backward compatibility or just remove it from AuthZ REST API ?
Regards.
Pedro Igor
_______________________________________________
keycloak-dev mailing list
keycloak-dev(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/keycloak-dev