On Tue, Jan 22, 2008 at 05:52:35PM -0500, Chas. Owens wrote:
> On Jan 22, 2008 2:58 PM, lerameur <[EMAIL PROTECTED]> wrote:
> > Hello,
> >
> > I wrote a short perl script (65 lines), simply to count some log file
> > time, and take the average. The script takes 4 minutes to run and go
> > through about 8 millions lines.
> > I would like to know if I can make it run faster. Why?, if I use the
> > command 'wc -l filename' , I get the number of lines in about a
> > minute, that is three less then the small script. I am right by
> > thinking the script can be reprogrammed so it can be process the file
> > faster ???
> snip
>
> Since we can't see your code, we can't tell if it can be done faster.
> Please note that wc -l is doing something that in Perl could be
> accomplished in 1 line:
>
> perl -nle 'END { print $c } $c++' file
>
> or the slightly more efficient
>
> perl -ne 'BEGIN { $/ = \4196 } END { print "$c\n" } $c += tr/\n//' file
>
> so 65 lines is a significantly more complex program and trying to
> compare them is not a good idea. You should also make sure that you
> are not seeing the effects of caching when comparing the two programs.
Yes, give wc something to do and it will probably take a little longer.
For example, on my machine here
wc -w big_file
takes about twice as long as
perl -nle '$w++ while /\S+/g; END { print $w }' big_file
which just goes to prove ... er ... hmmm ... well, not very much, really.
--
Paul Johnson - [EMAIL PROTECTED]
http://www.pjcj.net
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
http://learn.perl.org/
