Henry B. Hotz wrote:
I have no interest in long-term maintenance; that's not how *I* make
my living. Whatever I do should be stable enough to survive without
me. That means its dependencies need to be on things that are stable,
and actively supported elsewhere. "Support" for me, means someone
else will support it if I build it and walk away.
Sure. Likewise, my interest is in creating something stable enough that
requires as little of my time supporting it as possible. Putting
together a build that was actually sanely configured is a big part of
achieving that.
I'm not sure if any of this directly addresses the question of how to
provide SASL support on Windows. I can't see PostgreSQL being happy
adding a capability that doesn't work on Windows. If someone were
willing to build against KfW, then they could build the existing Krb5
support on Windows. PostgreSQL already has a rich selection of
authentication methods, so their incremental gain isn't what you might
expect from adoption of SASL.
My specific motivation is to provide an auth method that supports
Kerberos in the Java client. I would still get that, but it's looking
like I don't get as much beyond that as I hoped for.
Why doesn't that already work? Sorry if I missed something obvious, but
doesn't Java's GSS already give you Kerberos support using the native
Windows ticket cache?
--
-- Howard Chu
Chief Architect, Symas Corp. http://www.symas.com
Director, Highland Sun http://highlandsun.com/hyc
OpenLDAP Core Team http://www.openldap.org/project/
