[issue34751] Hash collisions for tuples

[Tim Peters]

Tim Peters <tim at python.org> added the comment:

>> I've noted before, e.g., that sticking to a prime
>> eliminates a world of regular bit patterns in the
>> multiplier.

> Why do you think this? 0x1fffffffffffffff is prime :-)

Point taken ;-)  But "a world of" is not the same as "the universe".  For example, sticking to a prime you'll never get 8 bytes the same.  Etc - "a world of" extremely regular patterns are eliminated.


> Having regular bit patterns and being prime are independent
> properties.
>
> To be clear: I don't have anything against picking a prime
> but we just shouldn't pretend that primes are important
> when they are not. That's all...

I don't like arguments from ignorance.  As I've said, I don't know why SeaHash uses a prime.  Neither do you.  Our collective ignorance doesn't imply the designer didn't have a good reason.

I can't think of a "good reason" for it either, but that's where we differ:  I think "so don't make gratuitous changes" while you think "therefore it can't possibly matter" ;-)

----------

_______________________________________
Python tracker <report at bugs.python.org>
<https://bugs.python.org/issue34751>
_______________________________________