On 4/16/02 12:27 PM, "Larry Wall" <[EMAIL PROTECTED]> claimed:
> You guys are thinking in terms of a single $obj.hash method. I think > there will be more than one hashish (er...) method available, and each > hash will be able to choose at least whether it wants to hash by $obj._ > (the default), by $obj.hash (mutable), or by $obj.id (immutable). The > hash function is not sufficient in cases of hash collision, so you also > need to think in terms of multiple comparison methods. Wow, that will be pretty cool, I think. > Since these are just properties of the hash, the hash could even have > properties containing closures that define the hash and/or comparison > functions. These properties can force the issue, regardless of whether > the object in question cooperates. So you could hash objects on any > attribute or combination of attributes, for instance. Yes, this sounds ideal to me. > But the default is gonna look a lot like Perl 5. That's certainly good enough for me! Regards, David -- David Wheeler AIM: dwTheory [EMAIL PROTECTED] ICQ: 15726394 http://david.wheeler.net/ Yahoo!: dew7e Jabber: [EMAIL PROTECTED]