-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Nathaniel Smith wrote: > If that's really what you want, you need something like > > x1 ^ x2 ^ x3 ^ reinterpret_cast<int>(my_pointer); > > (Or perhaps multiplication would be better, or something, hash > functions are not that easy to design.) In fact if x1 and x2 are likely to change by 1 it would be quite easy to have the case (x1+1 ^ x2+1 = x1 ^ x2) when the hash is a collision (if x1 and x2 had a 0 lower bit) or similar cases. If speed is not a real concern, it really would be better something like: x1*p1 + x2*p2 + x3*p3 where p1,p2,p3 are some "big" different primes.
(and me not being a cryptographer I guess it would have "other problems" anyway, but at least it wouldn't be "weak" to such frequently-happening use cases ^_^) - -- Lapo Luchini [EMAIL PROTECTED] (OpenPGP & X.509) www.lapo.it (Jabber, ICQ, MSN) -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2.1 (Cygwin) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iQIcBAEBAgAGBQJFDa+aAAoJELBiMTth2oCDERkP/jN3t+U0r5kIPlqp0QIp+jKk ZOnzhqbBgB8w7zHVFNLXvbUx5w5sqq4dNn4t9VINan09QZXsqGBvLpotgFKTNnKE g82Dtm0Qn4aok+yht4eFoGgI/8OWAfvT79ToI3RkIf0XGH6SuQBg277voSoP0uGd dPLU4oKhau7aNpRxQvBObeZkpbO6ka2EU1bgs8j7PfdCXIdlZg9ANVbPXgv9x74A 9vAWCEn/fwcrBqlAB4qcrBVA5zv8s4mWwFDYPODlrDKdb3yEICMZnrtsjNyR8fhk FXRFq6UW82iJC3/qBg8brX6fn2K5zzaRqJz3foV8zieasgPPFY3MTe3dCHBADb/R AfQ1U4kR7n4euJFv9SggdjIhiuoTFaOdYR3cDwtMspBx+VWU7og3ONn4SzuAnRJv 2Elun5cOygJYxOyGJ8lJlG9lrSTjP80blwt5dE51+SyCFAerEP+4442V5PiLIvXC o6BX/mmQjiyg+3k54sqNNy7kupjaDIwmtED0TLfu+b7xCFzUYsFA84erVQpYKACx jL/FI6i9Fo9U0wNFjciJYqGd6gPMSZB8Gcroc/e4SOhpv0qyiV3Q2mNlnCvDrGcc r5YHUH/oFXqbJ91WxEwdBPjGFbJnEg36MuqVFTftFvDMkfSFRq0169V3zGJ6LCzW 75+u4AEQ4ZbxWuagD4Wm =LsRQ -----END PGP SIGNATURE----- _______________________________________________ Monotone-devel mailing list Monotone-devel@nongnu.org http://lists.nongnu.org/mailman/listinfo/monotone-devel