#!/usr/bin/env ruby

# Copyright (C) 2008 Metasploit LLC

#
# This script extracts the forms from the main page of each
# web site in a list. The output of this can be used with
# Metasploit (and other tools) to obtain the saved form data
# of these domains.
#

require 'rubygems'  # install rubygems
require 'hpricot'   # gem install hpricot
require 'open-uri'
require 'timeout'

def usage
	$stderr.puts "#{$0} [site list] [output-dir]"
	exit(0)
end

sitelist = ARGV.shift() || usage()
output   = ARGV.shift() || usage()

File.readlines(sitelist).each do |site|
	site.strip!
	next if site.length == 0
	next if site =~ /^#/
	
	out = File.join(output, site + ".txt")
	File.unlink(out) if File.exists?(out)
	
	fd  = File.open(out, "a")
	

	["", "www."].each do |prefix|
		begin
			Timeout.timeout(10) do 
				doc = Hpricot(open("http://#{prefix}#{site}/"))
				doc.search("//form").each do |form|

					# Extract the form
					res = "<form"
					form.attributes.each do |attr|
						res << " #{attr[0]}='#{attr[1].gsub("'", "")}'"
					end
					res << "> "

					# Strip out the value
					form.search("//input") do |inp|

						inp.attributes.keys.each do |ikey|
							if (ikey.downcase == "value")
								inp[ikey] = ""
								next
							end

							if(inp.attributes[ikey] =~ /^http/i)
								inp[ikey] = ""
								next
							end

						end

						res << inp.to_html
					end
					res << "</form>"

					fd.write(res)
				end
			end
			break
		rescue ::Timeout::Error
			$stderr.puts "#{prefix}#{site} timed out"
		rescue ::Interrupt
			raise $!
		rescue ::Exception => e
			$stderr.puts "#{prefix}#{site} #{e.class} #{e}"
		end
	end
	
	fd.close
	
	File.unlink(out) if (File.size(out) == 0)

end