Parsing a file into an array

Question

I'm really new to Perl, but I want to basically read in a file to get some data out of it. I want to parse this data into an array. Why an array? Because, I want to use the data to generate an graph (bar or pie of the data).

Here's my code for Perl so far:

#!/usr/bin/perl -w
use warnings;

#Creating an array 
my @cpu_util;

#Creating a test file to read data from
my $file = "test.txt";      
#Opening the while      
open(DATA, $file) || die "Can't open $file: $!\n";

#Looping through the end of the file
while (<DATA>) 
{
if (/(\w+)\s+\d+\s+(\d+)\.(\d+)\%/){ #Getting only the "processes"
chomp;
push @cpu_util, split /\t/; #I was hoping this would split the data at the tabs
}
}

close ($file);

foreach $value (@cpu_util) {
print $value . "\n";
}

Here's the file that I am reading in (test.txt):

=========================================

CPU Utilization 

=========================================

Name      CPU Time     CPU Usage    

-----------------------------------------

System                7962         3.00% 

Adobe Photoshop       6783         0.19% 

MSN Messenger         4490         0.01% 

Google Chrome         8783         0.02% 

Idle                   120        94.00% 

=========================================

However, what I notice is that I successfully populate the array, but it doesn't split the tabs and give me a multidimensional array. I don't really care about the CPU time field, but I do want the CPU Usage, So I want to print an XY chart with Y-axis having the CPU-usage and x-axis, the process name.

I was hoping to have an array "cpu_util" and have cpu_util[0][0] = System and cpu_util[0][1] = 3.00. Is this even possible? I thought the split /\/t\ would take care of it, but apparently I was mistaken ...

Answer 1

I think you want a hash there. The key would be the CPU name and the value would be the percent use. You're almost there. After a successful match, use the captures ($1 and $2) to populate the hash:

my %cpu_util;   
while (<DATA>) 
    {
    if (/(.*?)\s+\d+\s+(\d+\.\d+)\%/){ #Getting only the "processes"
        $cpu_util{ $1 } = $2;
        }
    }

use Data::Dumper;   
print Dumper( \%cpu_util );

Your data structure should be much easier to use then:

$VAR1 = {
          'Google Chrome' => '0.02',
          'System' => '3.00',
          'Adobe Photoshop' => '0.19',
          'Idle' => '94.00',
          'MSN Messenger' => '0.01'
        };

Your split probably doesn't work because there aren't tabs there. It's not really they way to go anyway.

Answer 2

[wrong answer redacted]

Answer 3

Here is your code fixed ihateme:

use Data::Dumper;
#use strict;

#Creating an array
my @cpu_util;

#Creating a test file to read data from
my $file = "bar.txt";
#Opening the while
open(DATA, $file) || die "Can't open $file: $!\n";

#Looping through the end of the file
while (<DATA>)
{
  if (/(\w+)\s+\d+\s+(\d+)\.(\d+)\%/)
  {
    chomp;
    my ( $name, $cpuusage, @rest  ) = split ( /\s+\d+\s+(\d+\.\d+)\%/);
    push @cpu_util, [$name, $cpuusage ];
  }

}
close $file;

print Dumper(\@cpu_util);
$ref = $cpu_util[0];  # this will set the reference to the first array
$ref -> [2]; # this will return the 3rd value
print Dumper ($ref -> [1]);


 $VAR1 = [
      [
        'System',
        '3.00'
      ],
      [
        'Adobe Photoshop',
        '0.19'
      ],
      [
        'MSN Messenger',
        '0.01'
      ],
      [
        'Google Chrome',
        '0.02'
      ],
      [
        'Idle',
        '94.00'
      ]
    ];

$VAR1 = '3.00';