validate html of a login protected page, gems used :: mechanize and w3c_validator

w3c_validator_with_mechanize_authentication.rb

#dependencies :
#gem install mechanize 
# ^ if this doesnt work, and you are on ubuntu, you'll have to install nokogiri 
# gem install __name__ is now by default equivalent of gem install __name__ -y (-y for including all dependencies)
#gem install w3c_validtors


# I had to write this in order to validate pages against html5 compliance for our in house product
# my first repo.


require "rubygems"
require "w3c_validators"
require 'mechanize'
include W3CValidators

def validate(text_frag)

  @validator = MarkupValidator.new
  
  @validator.set_doctype!(:html5)
  # for example, change :html5 to :html32 and so on to validate against diff versions
  
  html = text_frag
  
  results = @validator.validate_text(html)
  
  #puts "\n\n\n-------here goes validation by using a url-------\n\n\n\n"
  
   
  k = [ ]
  if results.errors.length > 0
      results.errors.each do |err|
        puts err.to_s
        k << err.to_s
      end
    else
      puts 'Valid!'
    end
  
  num = 1
  
  File.open('errorlog_html'.concat(num.to_s), 'a') do |f|
    num = num +1 
    f.puts "generating error log for url "
    f.puts "\nat ".concat(Time.now.to_s), " "
    f.puts k
    f.puts "\n"
    
  end
  #puts "exiting..."
  #exit()

end


def get_list_of_html_fragments(links)
  #this function collects html source of all the links you give it
  #links must be in ruby array format
  
  bodies = [ ]
  # ^this list will have html soruces
  
  agent = Mechanize.new
  page = agent.get 'www.twitter.com'
  form = page.forms.first
  
  form.userName = ' '
  form.pwd = ' '
  
  page = agent.submit form
  #puts page.body
  
  bodies << page.body
  
  links.each do |link|
    page = agent.get link
    bodies << page.body
  end
  
  return bodies
  
end



# you can add your own implementation here, import links from a file etc


links = [ ]

links << 'www.twitter.com/railstroll'

links << 'www.twitter.com/someone_else'

bodies =  get_list_of_html_fragments(links)


bodies.each do |frag|
  puts "validating this frag ", frag
  validate(frag)
  puts "\n\n\n successfully validated.....\n\n\n"
end


puts "done done done"