Rafaqat Ali Chaudhry wrote:
> Dear all,
>
> I've a function which searches a given number from an array and
> returns the result.
>
> Function:
>
> 1. sub checkNumber($)
> 2. {
> 3. my ($l_number) = @_;
> 4. my $l_status;
> 5. my @matches= grep { /$l_number/ } @numbers;
Others have mentioned why grep() is a bad idea here.
Why are you using a regex here? Assuming your looking at numbers, dont' you
want a numeric equality test like $x == $y
> 6. if (@matches) {
> 7. $l_status="TRUE";
> 8. }
> 9. else {
> 10. $l_status="FALSE";
> 11. }
> 12. return $l_status;
Using "TRUE" and "FALSE" as return codes is unconventional. You can't use
these in Perl boolean expressions. Do they have some other use in your
program?
> 13. }
>
> The Way I call it:
>
> 1. $check=checkNumber($num1);
>
> The array has at least 20K elements. This piece of code is taking up
> almost 100% of the CUP cycles. Would somebody please suggest some
> alternative solution.
If the array needs to be searched repeatedly, consider constructing a hash.
Then you can do direct lookups. Or, look into using mjd's Memoize module to
cache results so repeated tests are faster.
A more efficient iterative approach using a numeric equality test would be:
sub checkNumber {
$_[0] == $_ && return 1 for @numbers;
return;
}
This returns 1 for a match and undef for no match.
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]