parsing HTML in Ruby

url-and-http.rb

#require 'socket'
require 'net/http'
require 'open-uri'
require 'nokogiri'

client = TCPSocket.open('indoexchanger.co.id', 'www')
client.send("OPTIONS /~dave/ HTTP/1.0\n\n", 0)
puts client.readlines
client.close

#scrap image in page

page = gets.chomp.to_s

http = Net::HTTP.new(url, 80)
response = http.get(page)

if response.message == "OK"
  puts response.body.scan(/<img alt=".*?" src="(.*?)"/m).uniq[0,3]
else
  puts "Failed"
end


puts response.message
if response.message == "OK"
  puts response.body.scan(/<img src="(.*?)"/m)
else
  puts "Failed"
end

#another way get image link
open('http://muhammadyana.web.id') do
  |f|
  puts f.read.scan(/<img src="(.*?)"/m).uniq[0,3]
end

#parsing HTMl
page = open('http://muhammadyana.web.id').read
if page =~ %r{<title>(.*?)</title>}m
  puts "Title is #{$1.inspect}"
end
doc = Nokogiri::HTML(open("http://pragprog.com/"))
puts "Page title is " + doc.xpath("//title").inner_html

puts doc.css('div#copyright p')

puts "\nSecond hyperlink is"
puts doc.xpath('id("site-links")//a[2]')
puts doc.css('#site-links a:nth-of-type(2)')