On Tue, Sep 15, 2015 at 8:31 PM, John Cowan <xxxxxx@mercury.ccil.org> wrote:

Alex Shinn scripsit:

> However providing the salt directly in the API has several
> advantages:

Do you mean allowing a salt argument to the constructors? I thought the
whole point of the salt was that it's supposed to be produced from
a high-quality random source.

No, the hash table implementation provides the salt.

My proposal is for the hash function API to be:

(hash <object> <salt>)

The salt could be global or per table, at the implementations

discretion. Using different values for the salt effectively

generates multiple hash functions, allowing double hashing

and cuckoo hashing, etc.

Alex