Re: [boost] Any interest in hashing algorithms SHA and/or FNV1a?

On Wed, Nov 13, 2013 at 8:31 AM, Marshall Clow wrote:

On Nov 12, 2013, at 10:52 AM, foster brereton wrote:

[ snip initial proposal / rfc ]

I believe so, yes.

There's been interest off and on over the years in a Boost.Crypto library. So far, there's been either (a) a lack of people willing to put in the work, or (b) disagreement on the interface.

While it is probably too early to start religious wars about interfaces, traditionally I have seen hash algorithms with APIs in triples: init -> update -> finish. I would be in favor of replicating that interface, as well as providing a single digest(...) call that wraps them all together, for more straightforward cases.

A set of hash implementations would be a nice start for this. I'll try to take a look at your code this week.

Great - if you have any questions/comments/concerns just ask. As I mentioned to Jeffrey Flinn I am willing to loop in ASL's md5 implementation, too, if it is worthwhile to do so. -foster