Michael Paquier <mich...@paquier.xyz> wrote: > Please note that the CF entry has been marked as committed. We should > really do something about having a cleaner separation between SASL, > the mechanisms and the AUTH_REQ_* codes, in the long term, though > honestly I don't know yet what would be the most elegant and the least > error-prone approach. And for anything that touches authentication, > simpler means better.
I've taken another shot at this over on the OAuth thread [1], for those who are still interested; see v40-0002. It's more code than my previous attempt, but I think it does a clearer job of separating the two concerns. Thanks, --Jacob [1] https://postgr.es/m/CAOYmi+=FzVg+C-pQHCwjW0qU-POHmzZaD2z3CdsACj==14h...@mail.gmail.com
