>An example hash is:
4d8d39a8283a53aeb51061457edc32c6a8e5404864b7571bf7b33d2c3bd5c580869ac1635be0d8ee57581f28b8aa9799729244852eb7aa010e7ac7b1cd69638f
>
> And an example salt is:
> 91f3c641110b2ded640c418046b3cb5af257da06ca98f192085cb10c7443e5e9

I think hex encoding is common and safe to assume generally. Just to
clarify, these values are stored in separate columns in the same
table? I would hope that's common and a safe assumption. Supporting
values across separate tables is possible, but it's hopefully an edge
case we could ignore.

> I also have to be able to specify the number of iterations, of course, which 
> is another issue!

That's a common input for salted hashing, so that would be an
anticipated configuration point.

The only concern you didn't address is whether the salt is the first
input to the digest or the last; I have seen both approaches used in
practice. I find it strange that there is no standard or guidance on
the matter despite having researched it fairly extensively. It looks
like it's simply an arbitrary choice and the CAS component would have
to support both approaches.

M

-- 
You are currently subscribed to [email protected] as: 
[email protected]
To unsubscribe, change settings or access archives, see 
http://www.ja-sig.org/wiki/display/JSG/cas-user

Reply via email to