Re: hash_value( complex<T> )

From: Peter Dimov (pdimov_at_[hidden])
Date: 2021-10-19 01:47:23

> Andrey Semashev wrote:
> > Not that I have a particular opinion on one way or another, but having the
> > hash changing the value between releases could be a breaking change, if
> > someone is using it e.g. in a memory mapped file.
> 
> Sure. As far as I can see, however, boost::hash has never promised hash
> stability, and the hash values have already changed a number of times.

On second thought, you are right. While it's true that boost::hash does not
promise hash stability in principle, and that hash values have indeed changed
a number of times, the library has been relatively stable so it's possible that
people do depend on hash stability, and this should be kept in mind.

Thanks.

I'm about to add a test with reference values that will tell us if some of them
change, whether inadvertently or by design. The commit history of this test
will also be usable as a log of which release changed which hash values.

https://github.com/boostorg/container_hash/blob/feature/reference/test/hash_reference_values.cpp