Actually, I meant to remove Sapphire and forgot to do it. It's done now and checked into CVS. Thanks for the report.
On Fri, Oct 18, 2002 at 03:08:39PM -0400, J Smith wrote: > > I've noticed the Sapphire cipher classes seems to be incomplete in Crypto++ > 5.0. Is this on purpose, or am I missing something here? > > Specifically, a bunch of virtual methods from StreamTransformation aren't > implemented in Sapphire: > > void StreamTransformation::ProcessData(byte*, const byte*, unsigned int) > bool StreamTransformation::IsRandomAccess() const > bool StreamTransformation::IsSelfInverting() const > bool StreamTransformation::IsForwardInverting() const > > I assume this was an oversight, as there are no tests for the Sapphire cipher > in any of the benchmarks. > > Any suggestion for a fix? Would using something like taking ProcessData() from > ARC4_Base, modifying it a bit and using it in SapphireBase work? As for the > Is*() methods, what should they be returning for Sapphire, 'cause I don't > have a clue. (I don't use Sapphire at all, I'm just using it for > completeness.) > > J
