From: "Shot (Piotr Szotkowski)" <shot@...>
Date: 2008-10-20T08:08:02+09:00
Subject: Re: newbie q: stripping duplicates

--i9LlY+UWpKt15+FH
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

Marcin Miel=C5=BCy=C5=84ski:

> 1.8.7 and 1.9.x use deep hashing for hashes, to achieve that in
> 1.8.6 you need to monkey patch Hash: http://pastie.org/pastes/272194

Right, and I use =E2=80=98alias eql? =3D=3D=E2=80=99 and hand-crafted hash =
methods
in most of my classes that need to be sane Set elements.

I understand the idea behind the hash method (let=E2=80=99s have a quick way
to check whether two objects are different =E2=80=93 and look closely with =
eql?
only if their hashes are the same), but I wonder whether there are any
rules-of-thumb for finding the sweet spot between making it fast and
making it return different results for different objects often.

For example, is 137 in the above pastie snippet a =E2=80=98magic number=E2=
=80=99 that
it=E2=80=99s good to multiply by? I understand how the bitwise XORs make ev=
ery
key and value impact the hash, but why not also multiply by 137 between
the key+value iterations?

-- Shot
--=20
Man will never be free until the last king is strangled with
the entrails of the last priest.            -- Denis Diderot

--i9LlY+UWpKt15+FH
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)

iD8DBQFI+7+Pi/mCfdEo8UoRAtL+AKCaHPWYIqu7ekKKg9Slrturz/lMFACeOfuJ
D1IXivMBx6pS4KpZcLtO7vQ=
=YfvK
-----END PGP SIGNATURE-----

--i9LlY+UWpKt15+FH--