Re: [boost] Message Hashing Interface (SHA-1/256/384/512, MD4/5)

Subject: Re: [boost] Message Hashing Interface (SHA-1/256/384/512, MD4/5)
From: Rob Riggs (rob_at_[hidden])
Date: 2010-04-06 23:32:42

On 04/06/2010 10:55 AM, Scott McMurray wrote:
> On 6 April 2010 10:55, Rob Riggs<rob_at_[hidden]>  wrote:
>    
>> This is also the scheme used by crypto++ (C++), the Perl API, PyCrypto
>> (Python) and openssl (C).  I think the Boost crypto hash API should more
>> closely mimic the existing de facto standard interfaces as much as possible.
>>
>>      
> Thanks for the references.  They're not completely consistent, though.
>
> I do like the idea that digest() be const, since conceptually it's
> just inspecting the result of the previous updates.
>
> Here's a hybrid idea, with definitions to show equivalences:
>
> class hash {
>    public:
>      void update(byte);
>      void reset() { *this = hash(); }
>      digest_type digest() const {
>          return hash(*this).end_of_message();
>      }
>      digest_type end_message() {
>          digest_type h = digest();
>          reset();
>          return h;
>      }
> };
>
> That way it provides for rolling or message-oriented digests, and
> adapts for the other as best it can if needed.
>
> Is that a reasonable compromise?
>    

I like it!

Rob