On 8/17/05, Paul Johnson <[EMAIL PROTECTED]> wrote:
> On Wed, Aug 17, 2005 at 06:26:16PM +0100, marcos rebelo wrote:
>
> > If I'm not wrong, Changing this lines:
> >
> > my @col = grep(!/\t/, split(/(\t)/, $line));
> > push(@col, "") if $line =~ /\t$/;
> >
> >
> > by
> >
> >
> > $line .= "\t";
> > my @col;
> > my $lastIndex = 0;
> > foreach my $actualIndex (0..length($line)) {
> > if (substr($line, $actualIndex, 1) eq "\t") {
> > push(@col, substr($line, $lastIndex,
> > $actualIndex - $lastIndex));
> > $lastIndex = $actualIndex+1;
> > }
> > }
> >
> > give me 10% faster code.
> >
> >
> > Is it possible to do faster?
>
> Probably.
>
> my @col = split /\t/, $line, -1;
>
> is much clearer anyway.
>
> perldoc -f split
>
> --
> Paul Johnson - [EMAIL PROTECTED]
> http://www.pjcj.net
>
really
my @col = split /\t/, $line, -1;
is clearer and even faster.
I'm passing one subroutine from 50 sec to 7 sec, really any small
inprovement is nice.
thanks all
Marcos
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
<http://learn.perl.org/> <http://learn.perl.org/first-response>
