Randal L. Schwartz wrote:
> I prefer a pattern that looks more like this:
>
> sub expensive_to_calculate {
>
> my $self = shift;
> my $input = shift; # cache on this scalar
>
> our %cache_for_expensive_to_calculate;
> return $cache_for_expensive_to_calculate{$input} ||= do {
> expensive calculation here;
> more things based on $input;
> last expression is value;
> };
>
> }
>
> Yes, in 5.10, this can be simplified, but I find 5.8 to still be
> the dominant Perl.
>
> This also presumes that a "false" value is never needed to be cached.
Out of curiosity, if it's gone this far, is it fair for me to recommend
reading Mr. Dominus' Memoizing chapter in Higher-Order Perl
(http://hop.perl.plover.com/)?
Fantastic reading regarding caching. Fantastic book overall. Note also
that the author (Mark Jason Dominus) now has the entire book on the
website as a free PDF download.
It is a mind-twisting read, from front to end.
I will never, ever forget a particular footnote statement (it makes me
smile every time I think about it):
"I sometimes enjoy the mind-bending exercise of imagining the result of
memoizing the Unix fork() function." (Chapter 3, page 80).
Steve
--
To unsubscribe, e-mail: beginners-unsubscr...@perl.org
For additional commands, e-mail: beginners-h...@perl.org
http://learn.perl.org/
