[1834] in Kerberos_V5_Development
Re: Krb5 Ccache DLL proposal
daemon@ATHENA.MIT.EDU (Theodore Y. Ts'o)
Tue Oct 1 12:50:20 1996
Date: Tue, 1 Oct 1996 12:49:20 -0400
From: "Theodore Y. Ts'o" <tytso@MIT.EDU>
To: Marc Horowitz <marc@MIT.EDU>
Cc: "Theodore Y. Ts'o" <tytso@MIT.EDU>, mikec@geek.grf.ov.com,
harryc@cam.ov.com, pbh@MIT.EDU, krbdev@MIT.EDU, dosdev@MIT.EDU
In-Reply-To: Marc Horowitz's message of Mon, 30 Sep 1996 13:16:10 EDT,
<199609301716.NAA25752@beeblebrox.MIT.EDU>
Date: Mon, 30 Sep 1996 13:16:10 EDT
From: Marc Horowitz <marc@MIT.EDU>
Also, wrt these promises, exactly what parts of the API did you
promise you wouldn't change? I wouldn't call anything named krb5_*
part of the API?
I'm talking about the krb5_* part of the API. My original plan was to
have a glue layer that sat below the existing krb5_cache_* functions,
and called out to the DLL functions defined in the proposal I outlined.
Yes, there would be a few "impedence mismatches" that the glue layer
would have to deal with, but they would be relatively small.
The advantage is that the DLL writer could implement a relatively clean
interface, as opposed to the moderately gross one that we currently have
inside our library.
- Ted
