Mechanize例子

This list based on my (german) article: "Web scraping mit Ruby/Mechanize" http://sixserv.org/2009/05/27/webscripting-mit-ruby-und-mechanize/ #00 Initialization

require 'rubygems'require 'mechanize'agent = WWW::Mechanize.newagent.set_proxy('localhost', '8000')agent.user_agent = 'Individueller User-Agent'agent.user_agent_alias = 'Linux Mozilla'agent.open_timeout = 3agent.read_timeout = 4agent.keep_alive = falseagent.max_history = 0 # reduce memory if you make lots of requests
#01 manual get requests
url = 'http://apoc.sixserv.org/requestinfo/'page = agent.get url# or ...page = agent.get(url, {"name" => "value", "key" => "val"})
#02 manual post submits

url = 'http://apoc.sixserv.org/requestinfo/'page = agent.post(url, {"name" => "value", "key" => "val"})

#03 form post submits
page = agent.get 'https://twitter.com/login'login_form = page.form_with(:action => 'https://twitter.com/sessions')login_form['session[username_or_email]'] = '[Username]'login_form['session[password]'] = '[Password]'page = agent.submit login_form
#04 link and history navigation
page = agent.get 'http://www.heise.de/'page = agent.click(page.link_with(:text => /Telepolis/))page = agent.click(page.link_with(:href => /artikel/))agent.backagent.backputs page.body
#05 exceptions
begin  page = agent.get 'http://apoc.sixserv.org/diese/seite/gibt/es/nicht/'rescue WWW::Mechanize::ResponseCodeError  puts "ResponseCodeError - Code: #{$!}"end
#06 referer
page = agent.get(:url => 'http://apoc.sixserv.org/requestinfo/',:referer => 'http://google.com/this/is/a/custom/referer')puts page.body
#07 request header manipulation
agent.pre_connect_hooks << lambda do |params|  params[:request]['X-Requested-With'] = 'XMLHttpRequest'end
#08 response header
page = agent.head 'http://sixserv.org'server_version = page.header['server']puts "Server: #{server_version}"if page.header.key? 'x-powered-by'  php_version = page.header['x-powered-by']  puts "X-Powered-By: #{php_version}"end# redirection urls:agent.redirect_ok = falsepage = agent.get 'http://www.sixserv.org/'puts page.header['location']
#09 content parsing
# X Path / CSS-Selector:page = agent.get 'http://xkcd.com/'img = page.search '/html/body/div/div[2]/div/div[2]/div/div/img'puts img# Regular Expression:page = agent.get 'http://example.com/'page.body.match /< h3>([^<]+)< \/h3>/puts "Heading 3: #{$1}"
#10 "with" method examples
# *_with: form, link, base, frame or iframe# get the first link including "foo" inside url:page.link_with(:href => /foo/)# all links with text 'more'page.links_with(:text => 'more')# get the form with the name 'foo'page.form_with('foo') # or form_with(:name => 'foo')