#!/usr/local/bin/ruby
#
#### logana.rb 1.0.0
#
# Copyright (C) 2005 phonondrive
# You can redistribute it and/or modify it under GPL2.
#
#### How to use
#
#    http://example.net/logana.rb?file=/usr/foo/log/expamle.com.log&regexp=POST
#
# You may use regular expression.
#    http://example.net/logana.rb?file=/usr/foo/log/expamle.com.log&regexp=.jpg|.gif
#    http://example.net/logana.rb?file=/usr/foo/log/expamle.com.log&regexp=GET|404
#
# *NOTE* Use '%3B' instead of ';'.
#    http://example.net/logana.rb?file=/usr/foo/log/expamle.com.log&regexp=Mozilla/4.76%20[en]%20(Win98%3B%20U)
#

BEGIN { $defout.binmode }
$KCODE = 'n'

# Set a path to your Apache's log file here if you don't want to use CGI parameters.
FILE_PATH = "example.com.log"

begin

	require 'cgi'
	@cgi = CGI::new

	print "Content-type: text/html\n\n"

	if @cgi.params['regexp'][0]
		file = @cgi.params['file'][0] || FILE_PATH
		regexp = Regexp.new( "#{ Regexp.quote( CGI::escapeHTML( @cgi.params['regexp'][0] ) ) }", Regexp::IGNORECASE )

		output = ""
		url_h = {}
		url_r = {}

		open( CGI::escapeHTML( file ) ) {|line|
			while l = line.gets
				if regexp =~ l
					url = l.split(" ").first
					if url_h[url]
						url_h[url] += 1
						url_r[url] += "#{l}<br>"
					else
						url_h[url] = 1
						url_r[url] = "#{l}<br>"
					end
					output += l
				end if regexp
			end
		}

		print "<h1>Logs matching to '#{regexp.source}'</h1>\n"

		url_h.sort {|i, j|
			( j[1] <=> i[1] ).nonzero? or i[0] <=> j[0]
		}.each {|key, num|
			print "<h2>#{num} : #{key}</h2>\n<p style=\"font-size:x-small\">#{url_r[key]}</p>\n"
		}
	end

	print "complete!"

rescue Exception
	puts 'Content-Type: text/plain'
	puts
	puts "#$! (#{$!.class})"
	puts
	puts $@.join("\n")
end