> > I like this very much and I'm assuming it's faster than
> > putting them in hash and returning the keys as an array
> > since you don't have to build a hash.
>
> Actually you *do* build a hash. :-)
Oh yeah, duh what am I smoking?
>
> > @unique = do {
> > my %seen;
> > grep !$seen{$_}++, @arr
> > };
>
> The advantages are that (a) it doesn't change the order
> of the input array and (b) it's a common idiom that the
> next programmer in line will recognize instantly.
>
> The hash-keys version would probably be a bit faster:
>
> @unique = keys %{
> my %set;
> @set{ @arr } = (); # fast!
> \%set;
> };
I mihgt play around with that a bit too. Thanks!
Dan
>
> But karma is more important than (a smidgen of) speed.
>
> --
> Steve
>
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
