access log parser

stats.pl

#!/usr/bin/env perl

# use warnings;
# use strict;

die('File not specified') if not $ARGV[0];
open FILE, $ARGV[0] or die $!;

my %months = ();
my ($firstRecord, $lastRecord);
my ($records, $total);
my @excludeIp = ('127.1.1.1', '127.2.2.2', '127.3.3.3', '127.4.4.4');
my $apacheRegex = '
	^(\S+)\s                      # IP
	\S+\s+                        # remote logname
	(?:\S+\s+)+                   # remote user
	\[([^]]+)\]\s                 # date
	"(\S*)\s?                     # method
	(?:((?:[^"]*(?:\\")?)*)\s     # URL
	([^"]*)"\s|                   # protocol
	((?:[^"]*(?:\\")?)*)"\s)      # or, possibly URL with no protocol
	(\S+)\s                       # status code
	(\S+)\s                       # bytes
	"((?:[^"]*(?:\\")?)*)"\s      # referrer
	"(.*)"$                       # user agent';
my $dateRegex = '^\S+\/(\S+)\/\S+\s\S+';

# Loop records
while (<FILE>) {
	$total++;
	my ($ip, $date, $method, $url, $protocol, $alt_url, $code, $bytes, $referrer, $ua) = (m/$apacheRegex/xs);
	$firstRecord = $date if not $firstRecord;
	$lastRecord = $date;
	next if &isExcludeIp($ip);
	&setDownloadCount(&getMonthFromDate($date));
}

# Print month stats
printf("First record: %s\n", $firstRecord);
printf("Last record: %s\n\n", $lastRecord);


foreach my $month (keys %months) {
	printf("%s: %d\n", $month, $months{$month});
	$records += $months{$month};
}

printf("\nTotal records: %d (%d excluded)\n\n", $records, ($total - $records));

#######
## SUBS
sub isExcludeIp($) {
	my ($ip) = @_;
	if (grep(/^$ip/, @excludeIp)) { return 1 } else { return 0 };
}
sub getMonthFromDate($) {
	my ($date) = @_;
	my $month = $date =~ m/$dateRegex/;
	return $1;
}
sub setDownloadCount($) {
	my ($date) = @_;
	$months{$date}++;
}