Anonymous Monk has asked for the wisdom of the Perl Monks concerning the following question:

Hi, I have this file of stats: 
"\\TEST\Memory\Pages/sec"    "\\TEST\PhysicalDisk"
"06/20/2002 08:27:19.087"    "0.87367121907974665"    
"06/20/2002 08:27:34.088"    "3.3339663084377777"
"06/20/2002 08:27:49.090"    "3.9996207699833071"
"06/20/2002 08:28:04.091"    "3.198195381472539"

[download]
Now I have this code to grab the data, and I now want to calculate the averages on each column. Is there an easy way to do that dynamically counting the rows then obviously I would add them and calculate the average: 
#!C:\Perl\bin\perl

use Getopt::Std;

$USAGE = <<USAGE;
 Usage: Absolute_Path\\Logfile

    A script to retrieve perf numbers and calculate averages
    
USAGE
;
 
# process command line
getopts('x') || die $USAGE;
 
# get parameters
$Perf    = shift || die $USAGE;

chomp ($Perf);

open (LOG, "$Perf") ||
        die "Error: Couldn't open $Perf : $!\n";

@read = <LOG>;

print "Contents of read are: @read\n";

[download]

Replies are listed 'Best First'.
Re: Formating Text Files
by hotshot (Prior) on Jun 20, 2002 at 15:35 UTC
    If I undestood you right, then you can try something like that: 
    $i = 0,
$total = 0;
while ($line = <FILE>) {
  chomp $line;
  if ($line =~ /"([\d\/]+)\s([\d:.]+)"\s+"([\d.]+)"/) {
    total += $3;
    $i++;
  }
}
$avarage = $total/$i;

    [download]
    This will give you only the avarage of the number in the right column, but you can easily do that for the time and date (write a timeAvarage function), I'll leave that for you.

    Thanks.

    Hotshot
[reply]
[d/l]
Re: Formating Text Files
by robobunny (Friar) on Jun 20, 2002 at 15:48 UTC
    if your data file is large, you might see some performance benefit from using a split instead of a more complicated regex: 
    # skip the header line, assuming it's the first line
# otherwise, you'll have to check in the loop to see
# if you have grabbed a number or a string
my $header = <FILE>;
my $total = 0;
my $count = 0;
while(<FILE>) {
  $total += (split('"', $_))[3];
  $count++;
}
print "Average: ",$total/$count, "\n";

    [download]
    if you have multiple header lines inside the file, you will need to check in the loop to make sure that you skip the header lines.
[reply]
[d/l]
      Thanks, good ideas! I was taking the 'scenic route' as I usually do. and making things to complicated.
[reply]
Re: Formating Text Files
by DamnDirtyApe (Curate) on Jun 20, 2002 at 15:38 UTC

    Assuming that $Perf in the data file in your example, you can just read @read in scalar context to find the number of elements.

    $lines_in_file = @read;

    Update: Forgot about counting the header row (thanks bronto). Read below.


    _______________
    D a m n D i r t y A p e
    Home Node | Email
[reply]
[d/l]
[select]
      ...minus one (the heading row). So, actually, he could use $#read directly, or --$lines_in_file ;
[reply]
[d/l]
[select]

Back to Seekers of Perl Wisdom