Well there are quite a few ways. One method would be to store the log in a hash (or an array):

my %log_file_hash = ();
my @log_file_array;
my $linenum = 0;
my $line;

open FILE, "logfile.txt";

while ($line = <FILE>)
{
  chomp $line;
  $log_file_hash{$linenum} = $line;
  push @log_file_array, $line;
  $linenum++;
}

#now the whole file is in the hash and array.
print "The first line is " . $log_file_hash{"0"} . "\n";
print "From the array... " . $log_file_array[0] . "\n";
[download]

Another way to do this is to just store the last 4 lines and sort of move a frame over your data:

$line_0 = "";
$line_1 = "";
$line_2 = "";
$line_3 = "";

while ($line = <FILE>)
{
  chomp $line;
  $line_3 = $line_2;
  $line_2 = $line_1;
  $line_1 = $line_0;
  $line_0 = $line;
}
[download]

Hopefully this is useful. If I misunderstood your question just let me know.

Update: minor inplace grammar edits.

Based on your described algorithm, I'd suggest using Tie::File. If you're using large log files, be sure to read the section on memory.

Basically, this module allows you to "tie" a file to an array where each line in the file is an element in the array. This allows you to use array control functions (such as push and pop) to manipulate the file (such as adding a new line of text and deleting a line.)

The (untested) code below should help you get started with using Tie::File with what you're trying to do.

use strict;
use Tie::File;

my $filename = "some.log";
my @file;

tie @file, 'Tie::File', $filename || die "Unable to open file\n";

my $index = 0;
foreach my $line (@file) {
  if ($line =~ m/xyz/i) {
    my $var1 = $file[$index-4];
    my $var2 = $file[$index+4];
    print "Found values '$var1' and '$var2'\n";
  }
  $index++;
}

untie @file;
[download]

a not so perfect solution , but might work

#!/usr/bin/perl
use strict;
use warnings;

my @array;
my $PreviousLines = 4;
while (<DATA>) {
    chomp;    
    if (/xyz/) {
        print join "\n", @array;
        print "\n\n";        
    }else{
        push @array, $_;
        if (@array > $PreviousLines ) {
            # delete first element 
            splice @array, 0,1;
        }
        
    }    
}
[download]

Once you have it in English, it's easy to translate to Perl, C++or anything else.

Note:English, in the above, is a placeholder for human language you're most comfortable writing in. 8-)

#!perl 
use tie::file


my $count = 0;

tie @array, 'Tie::File', 'log.txt' or die;

my $index = 0;
foreach my $line (@array) {

  if ($line =~ m/Age: 3/i) {
  
    my $age = @array[$index];
    my $var1 = @array[$index-4];
    my $var2 = @array[$index+6];
    my $var3 = @array[$index+7];
    
    

    if ($var2 =~ m/Added to non/i) {
    print "$var1\n"; 
    print "$age\n";
    print "$var2\n";
    print "-----------------\n";
    }
    
    if ($var3 =~ m/Added to non/i) {
    print "$var1\n"; 
    print "$age\n";
    print "$var3\n";
    print "-----------------\n";
    }    
    

    }
    
  $index++;
}


untie @array;
[download]