I had assumed Perl6 will have copy-on-write references, so that
$x = [1,2,3];
@y = *$x;
would not require a copy. However, on thinking about it further, I
realized that that would get you in trouble here:
$x = [1,2,3];
%h{$x} = 1;
@y = *x;
$x[0]++;
print %h{$x};
Thi
On Wed, 2002-07-24 at 12:34, Fisher Mark wrote:
> > But then sometimes you'd *want* hashing to be based on the
> > content.
>
> OK, I'll bite -- when would you want this behavior? This behavior means
> that once you change the contents, the hash value would become irretrievable
> unless you rest
> But then sometimes you'd *want* hashing to be based on the
> content.
OK, I'll bite -- when would you want this behavior? This behavior means
that once you change the contents, the hash value would become irretrievable
unless you restored the contents of the key. (Is this useful in functional
As the last person to change the key hash algorithm, I'd like to chime in
here with a request that each PMC provide a string that the key hashing
algorithm can operate on. To some degree this is just selfish on my part --
I've got plans for upgrading the key hash algorithm in Perl 5 and Perl 6
wh
