SHA-based subclass for random module

[Trevor Perrin]

Paul Rubin wrote:

> trevp at trevp.net (Trevor Perrin) writes:
> 
>>
>>What about a subclass of Random that provides an interface to
>>platform-specific secure RNGs:
[...]
> 
> I think it's extremely desirable and important to add such an
> interface, and maybe it should go in the random module, but it
> shouldn't be a subclass of Random.  Random is intended to be a
> deterministic generator.


I see that seed(), getstate() and setstate() imply determinism.

Aside from those, the nice thing about subclassing Random is you'd only 
have to implement random() and getrandbits() and you'd get randrange(), 
randint(), choice(), shuffle(), and sample() for free.  So it might be 
worth it, even if you have to raise NotImplementedError for a few methods.

Actually, Random could be made even easier to subclass.  getrandbits 
(which is new in 2.4) returns a long.  It would be better for us if 
there was a getrandstring() function that an underlying generator could 
implement.  This would also be a helpful addition to the API, at least 
for cryptographic uses.


Trevor


http://www.python.org/dev/doc/devel/lib/module-random.html