Preprocessing Apache Log

simplify_apache_log.rb

#!/usr/bin/env ruby

require 'date'
require 'apachelogregex'

# apache log format
parser = ApacheLogRegex.new('%h %l %u %t \"%r\" %>s %O \"%{Referer}i\" \"%{User-Agent}i\" %I %b %D')

File.open(ARGV[0]).readlines.each do | line |
	p = parser.parse(line)
	next if p.nil?

	# prepare needed columns
	path = p['%r'].split(' ')[1]

	# from zend log default to YYYY-MM-DD H:i:S
	date = DateTime.strptime(p['%t'], '[%d/%b/%Y:%H:%M:%S %Z]').strftime('%Y-%m-%d %H:%M:%S')
	puts "#{date}\t#{p['%h']}\t#{path}"
end