From: "Shot (Piotr Szotkowski)" <shot@...>
Date: 2009-09-11T00:56:14+09:00
Subject: What’s the best general implementation of custom #hash	methods?

--Mjqg7Yu+0hL22rav
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

Hello, ruby-talk.

What=E2=80=99s the best (fast, yet with relatively few cases when equal has=
hes
are generated for non-eql objects) way to implement a hash method?

Is XOR-ing the hashes of an object=E2=80=99s instance variables (assuming t=
hat
two objects are eql if their ivars are eql) a good general approach?

I=E2=80=99m asking because I use this pattern in my code, and it seems quite
fast; I needed to work around a bug in MRI=E2=80=99s Set#hash and this solu=
tion=C2=B9
seems about as fast as the original MRI=E2=80=99s Set#hash. I=E2=80=99m jus=
t not sure
whether it=E2=80=99s hashy enough (so to speak) in that two un-eql objects
hashed in this way produce different hashes often enough.

=C2=B9 http://github.com/Chastell/art-decomp/commit/b465660b9dcc8eada86f50b=
bcb8f951aa8145a19

=E2=80=94 Shot
--=20
Ill-informed qmail-bashing is better than
no qmail-bashing at all.      [Don Marti]

--Mjqg7Yu+0hL22rav
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: Digital signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)

iEYEARECAAYFAkqpIZoACgkQi/mCfdEo8UpDpQCeP3ttOJXQLioMVXL0GSFlJBKU
e24AoJmbRhF+IRLg9sah1otjWw3rV+3/
=jGk5
-----END PGP SIGNATURE-----

--Mjqg7Yu+0hL22rav--