mechanize 2.5.1 → 2.6.0

data/lib/mechanize/http/auth_challenge.rb

@@ -1,6 +1,6 @@
 class Mechanize::HTTP
 
-  AuthChallenge = Struct.new :scheme, :params
+  AuthChallenge = Struct.new :scheme, :params, :raw
 
   ##
   # A parsed WWW-Authenticate header
@@ -18,7 +18,10 @@ class Mechanize::HTTP
     # The authentication parameters
 
     ##
-    # :method: initialize(scheme = nil, params = nil)
+    # :method: initialize
+    #
+    # :call-seq:
+    #   initialize(scheme = nil, params = nil)
     #
     # Creates a new AuthChallenge header with the given scheme and parameters
 
@@ -52,13 +55,9 @@ class Mechanize::HTTP
     end
 
     ##
-    # The reconstructed, normalized challenge
-
-    def to_s
-      auth_params = params.map { |name, value| "#{name}=\"#{value}\"" }
+    # The raw authentication challenge
 
-      "#{scheme} #{auth_params.join ', '}"
-    end
+    alias to_s raw
 
   end
 

data/lib/mechanize/http/content_disposition_parser.rb

@@ -86,7 +86,7 @@ class Mechanize::HTTP::ContentDispositionParser
 
     while true do
       return nil unless param = rfc_2045_token
-      param.downcase
+      param.downcase!
       return nil unless @scanner.scan(/=/)
 
       value = case param

data/lib/mechanize/http/www_authenticate_parser.rb

@@ -25,6 +25,7 @@ class Mechanize::HTTP::WWWAuthenticateParser
 
     while true do
       break if @scanner.eos?
+      start = @scanner.pos
       challenge = Mechanize::HTTP::AuthChallenge.new
 
       scheme = auth_scheme
@@ -43,6 +44,7 @@ class Mechanize::HTTP::WWWAuthenticateParser
           challenge.params = @scanner.scan(/.*/)
         end
 
+        challenge.raw = www_authenticate[start, @scanner.pos]
         challenges << challenge
         next
       else
@@ -62,10 +64,15 @@ class Mechanize::HTTP::WWWAuthenticateParser
         unless name then
           challenge.params = params
           challenges << challenge
-          break if @scanner.eos?
+
+          if @scanner.eos? then
+            challenge.raw = www_authenticate[start, @scanner.pos]
+            break
+          end
 
           @scanner.pos = pos # rewind
-          challenge = '' # a token should be next, new challenge
+          challenge.raw = www_authenticate[start, @scanner.pos].sub(/(,+)? *$/, '')
+          challenge = nil # a token should be next, new challenge
           break
         else
           params[name] = value
@@ -73,8 +80,6 @@ class Mechanize::HTTP::WWWAuthenticateParser
 
         spaces
 
-        return nil unless ',' == @scanner.peek(1) or @scanner.eos?
-
         @scanner.scan(/(, *)+/)
       end
     end

data/lib/mechanize/page.rb

@@ -212,7 +212,10 @@ class Mechanize::Page < Mechanize::File
   def_delegator :parser, :at, :at
 
   ##
-  # :method: form_with(criteria)
+  # :method: form_with
+  #
+  # :call-seq:
+  #   form_with(criteria)
   #
   # Find a single form matching +criteria+.
   # Example:
@@ -221,7 +224,9 @@ class Mechanize::Page < Mechanize::File
   #   end
 
   ##
-  # :method: forms_with(criteria)
+  # :method: forms_with
+  #
+  # :call-seq: forms_with(criteria)
   #
   # Find all forms form matching +criteria+.
   # Example:
@@ -232,14 +237,19 @@ class Mechanize::Page < Mechanize::File
   elements_with :form
 
   ##
-  # :method: link_with(criteria)
+  # :method: link_with
+  #
+  # :call-seq: link_with(criteria)
   #
   # Find a single link matching +criteria+.
   # Example:
   #   page.link_with(:href => /foo/).click
 
   ##
-  # :method: links_with(criteria)
+  # :method: links_with
+  #
+  # :call-seq:
+  #   links_with(criteria)
   #
   # Find all links matching +criteria+.
   # Example:
@@ -250,14 +260,18 @@ class Mechanize::Page < Mechanize::File
   elements_with :link
 
   ##
-  # :method: base_with(criteria)
+  # :method: base_with
+  #
+  # :call-seq: base_with(criteria)
   #
   # Find a single base tag matching +criteria+.
   # Example:
   #   page.base_with(:href => /foo/).click
 
   ##
-  # :method: bases_with(criteria)
+  # :method: bases_with
+  #
+  # :call-seq: bases_with(criteria)
   #
   # Find all base tags matching +criteria+.
   # Example:
@@ -268,14 +282,18 @@ class Mechanize::Page < Mechanize::File
   elements_with :base
 
   ##
-  # :method: frame_with(criteria)
+  # :method: frame_with
+  #
+  # :call-seq: frame_with(criteria)
   #
   # Find a single frame tag matching +criteria+.
   # Example:
   #   page.frame_with(:src => /foo/).click
 
   ##
-  # :method: frames_with(criteria)
+  # :method: frames_with
+  #
+  # :call-seq: frames_with(criteria)
   #
   # Find all frame tags matching +criteria+.
   # Example:
@@ -286,14 +304,18 @@ class Mechanize::Page < Mechanize::File
   elements_with :frame
 
   ##
-  # :method: iframe_with(criteria)
+  # :method: iframe_with
+  #
+  # :call-seq: iframe_with(criteria)
   #
   # Find a single iframe tag matching +criteria+.
   # Example:
   #   page.iframe_with(:src => /foo/).click
 
   ##
-  # :method: iframes_with(criteria)
+  # :method: iframes_with
+  #
+  # :call-seq: iframes_with(criteria)
   #
   # Find all iframe tags matching +criteria+.
   # Example:
@@ -304,14 +326,18 @@ class Mechanize::Page < Mechanize::File
   elements_with :iframe
 
   ##
-  # :method: image_with(criteria)
+  # :method: image_with
+  #
+  # :call-seq: image_with(criteria)
   #
   # Find a single image matching +criteria+.
   # Example:
   #   page.image_with(:alt => /main/).fetch.save
 
   ##
-  # :method: images_with(criteria)
+  # :method: images_with
+  #
+  # :call-seq: images_with(criteria)
   #
   # Find all images matching +criteria+.
   # Example:

data/lib/mechanize/page/image.rb

@@ -165,7 +165,7 @@ class Mechanize::Page::Image
         page.uri + src
       end
     else
-      src
+      URI src
     end
   end
 

data/lib/mechanize/parser.rb

@@ -54,21 +54,30 @@ module Mechanize::Parser
   attr_accessor :code
 
   ##
-  # :method: [](header)
+  # :method: []
+  #
+  # :call-seq:
+  #   [](header)
   #
   # Access HTTP +header+ by name
 
   def_delegator :header, :[], :[]
 
   ##
-  # :method: []=(header, value)
+  # :method: []=
+  #
+  # :call-seq:
+  #   []=(header, value)
   #
   # Set HTTP +header+ to +value+
 
   def_delegator :header, :[]=, :[]=
 
   ##
-  # :method: key?(header)
+  # :method: key?
+  #
+  # :call-seq:
+  #   key?(header)
   #
   # Is the named +header+ present?
 

data/lib/mechanize/pluggable_parsers.rb

@@ -1,6 +1,7 @@
 require 'mechanize/file'
 require 'mechanize/file_saver'
 require 'mechanize/page'
+require 'mechanize/xml_file'
 
 ##
 # Mechanize allows different parsers for different content types.  Mechanize
@@ -64,7 +65,7 @@ class Mechanize::PluggableParser
     :xhtml => 'application/xhtml+xml',
     :pdf   => 'application/pdf',
     :csv   => 'text/csv',
-    :xml   => 'text/xml',
+    :xml   => ['text/xml', 'application/xml'],
   }
 
   attr_accessor :default
@@ -74,7 +75,9 @@ class Mechanize::PluggableParser
       CONTENT_TYPES[:html]  => Mechanize::Page,
       CONTENT_TYPES[:xhtml] => Mechanize::Page,
       CONTENT_TYPES[:wap]   => Mechanize::Page,
-      'image'               => Mechanize::Image
+      'image'               => Mechanize::Image,
+      'text/xml'            => Mechanize::XmlFile,
+      'application/xml'     => Mechanize::XmlFile,
     }
 
     @default = Mechanize::File
@@ -138,7 +141,9 @@ class Mechanize::PluggableParser
   # Registers +klass+ as the parser for text/xml content
 
   def xml=(klass)
-    register_parser(CONTENT_TYPES[:xml], klass)
+    CONTENT_TYPES[:xml].each do |content_type|
+      register_parser content_type, klass
+    end
   end
 
   ##

data/lib/mechanize/test_case.rb

@@ -78,6 +78,19 @@ class Mechanize::TestCase < MiniTest::Unit::TestCase
     Mechanize::Page.new uri, nil, body, 200, @mech
   end
 
+  ##
+  # Creates a Mechanize::CookieJar by parsing the given +str+
+
+  def cookie_jar str, uri = URI('http://example')
+    jar = Mechanize::CookieJar.new
+
+    Mechanize::Cookie.parse uri, str do |cookie|
+      jar.add uri, cookie
+    end
+
+    jar
+  end
+
   ##
   # Runs the block inside a temporary directory
 

data/lib/mechanize/test_case/digest_auth_servlet.rb

@@ -15,19 +15,19 @@ class DigestAuthServlet < WEBrick::HTTPServlet::AbstractServlet
     :Logger => Logger.new(nil)
   })
 
-  def do_GET(req,res)
+  def do_GET req, res
     def req.request_time; Time.now; end
     def req.request_uri; '/digest_auth'; end
-    def req.request_method; "GET"; end
+    def req.request_method; 'GET'; end
 
     begin
-      @@authenticator.authenticate(req,res)
+      @@authenticator.authenticate req, res
       res.body = 'You are authenticated'
     rescue WEBrick::HTTPStatus::Unauthorized
       res.status = 401
     end
-    FileUtils.rm('digest.htpasswd') if File.exists?('digest.htpasswd')
   end
+
   alias :do_POST :do_GET
 end
 

data/lib/mechanize/util.rb

@@ -27,7 +27,7 @@ class Mechanize::Util
   #   Mechanize::Util::DefaultMimeTypes.replace(mimetab)
   DefaultMimeTypes = WEBrick::HTTPUtils::DefaultMimeTypes
 
-  def self.build_query_string(parameters, enc=nil)
+  def self.build_query_string(parameters, enc = nil)
     parameters.map { |k,v|
       # WEBrick::HTTP.escape* has some problems about m17n on ruby-1.9.*.
       [CGI.escape(k.to_s), CGI.escape(v.to_s)].join("=") if k
@@ -35,7 +35,7 @@ class Mechanize::Util
   end
 
   # Converts string +s+ from +code+ to UTF-8.
-  def self.from_native_charset(s, code, ignore_encoding_error=false, log=nil)
+  def self.from_native_charset(s, code, ignore_encoding_error = false, log = nil)
     return s unless s && code
     return s unless Mechanize.html_parser == Nokogiri::HTML
 

data/lib/mechanize/xml_file.rb

@@ -0,0 +1,47 @@
+##
+# This class encapsulates an XML file. If Mechanize finds a content-type
+# of 'text/xml' or 'application/xml' this class will be instantiated and
+# returned. This class also opens up the +search+ and +at+ methods available
+# on the underlying Nokogiri::XML::Document object.
+#
+# Example:
+#
+#   require 'mechanize'
+#
+#   agent = Mechanize.new
+#   xml = agent.get('http://example.org/some-xml-file.xml')
+#   xml.class #=> Mechanize::XmlFile
+#   xml.search('//foo[@attr="bar"]/etc')
+
+class Mechanize::XmlFile < Mechanize::File
+  extend Forwardable
+
+  # The underlying Nokogiri::XML::Document object
+
+  attr_reader :xml
+
+  def initialize(uri = nil, response = nil, body = nil, code = nil)
+    super uri, response, body, code
+    @xml = Nokogiri.XML body
+  end
+
+  ##
+  # :method: search
+  #
+  # Search for +paths+ in the page using Nokogiri's #search.  The +paths+ can
+  # be XPath or CSS and an optional Hash of namespaces may be appended.
+  #
+  # See Nokogiri::XML::Node#search for further details.
+
+  def_delegator :xml, :search, :search
+
+  ##
+  # :method: at
+  #
+  # Search through the page for +path+ under +namespace+ using Nokogiri's #at.
+  # The +path+ may be either a CSS or XPath expression.
+  #
+  # See also Nokogiri::XML::Node#at
+
+  def_delegator :xml, :at, :at
+end

data/test/htdocs/tc_follow_meta_loop_1.html

@@ -0,0 +1,8 @@
+<html>
+  <head>
+    <meta http-equiv="refresh" content="0; url=http://localhost/tc_follow_meta_loop_2.html">
+  </head>
+  <body>
+    This page has a meta refresh.
+  </body>
+</html>

data/test/htdocs/tc_follow_meta_loop_2.html

@@ -0,0 +1,8 @@
+<html>
+  <head>
+    <meta http-equiv="refresh" content="0; url=http://localhost/tc_follow_meta_loop_1.html">
+  </head>
+  <body>
+    This page has a meta refresh.
+  </body>
+</html>

data/test/test_mechanize.rb

@@ -76,6 +76,8 @@ class TestMechanize < Mechanize::TestCase
     end
 
     assert_equal ssl_certificate, mech.certificate
+    assert_equal ssl_certificate, mech.cert
+    assert_equal ssl_private_key, mech.key
   end
 
   def test_cert_store
@@ -98,6 +100,18 @@ class TestMechanize < Mechanize::TestCase
                  @mech.history.last.uri.to_s)
   end
 
+  def test_click_bogus_link_with_cookies
+    @mech.cookie_jar = cookie_jar("a=b")
+
+    page = html_page <<-BODY
+<a href="http:///index.html">yes really</a>
+    BODY
+
+    page.links[0].click
+
+    assert_equal '/index.html', requests.first.path
+  end
+
   def test_click_frame
     frame = node 'frame', 'src' => '/index.html'
     frame = Mechanize::Page::Frame.new frame, @mech, fake_page
@@ -384,20 +398,11 @@ but not <a href="/" rel="me nofollow">this</a>!
   end
 
   def test_get_digest_auth
-    block_called = false
-
-    @mech.add_auth(@uri, 'user', 'pass')
+    @mech.add_auth @uri, 'user', 'pass'
 
-    @mech.pre_connect_hooks << lambda { |_, request|
-      block_called = true
-      request.to_hash.each do |k,v|
-        assert_equal(1, v.length)
-      end
-    }
+    page = @mech.get @uri + '/digest_auth'
 
-    page = @mech.get(@uri + '/digest_auth')
-    assert_equal('You are authenticated', page.body)
-    assert block_called
+    assert_equal 'You are authenticated', page.body
   end
 
   def test_get_follow_meta_refresh
@@ -411,6 +416,21 @@ but not <a href="/" rel="me nofollow">this</a>!
                  @mech.history.first.uri.to_s)
     assert_equal('http://localhost/index.html', page.uri.to_s)
     assert_equal('http://localhost/index.html', @mech.history.last.uri.to_s)
+
+    [5, 6].each { |limit|
+      @mech.redirection_limit = limit
+      begin
+        @mech.get('http://localhost/tc_follow_meta_loop_1.html')
+      rescue => e
+        assert_instance_of Mechanize::RedirectLimitReachedError, e
+        assert_equal limit, e.redirects
+        if limit % 2 == 0
+          assert_equal '/tc_follow_meta_loop_1.html', e.page.uri.path
+        else
+          assert_equal '/tc_follow_meta_loop_2.html', e.page.uri.path
+        end
+      end
+    }
   end
 
   def test_get_follow_meta_refresh_anywhere
@@ -1009,6 +1029,40 @@ but not <a href="/" rel="me nofollow">this</a>!
     assert_equal URI('http://user:pass@localhost:8080'), http.proxy_uri
   end
 
+  def test_shutdown
+    uri = URI 'http://localhost'
+    jar = Mechanize::CookieJar.new
+    Mechanize::Cookie.parse uri, 'a=b' do |cookie|
+      jar.add uri, cookie
+    end
+
+    @mech.cookie_jar = jar
+
+    @mech.get("http://localhost/")
+
+    assert_match(/Hello World/, @mech.current_page.body)
+    refute_empty @mech.cookies
+    refute_empty Thread.current[@mech.agent.http.request_key]
+
+    @mech.shutdown
+
+    assert_nil Thread.current[@mech.agent.http.request_key]
+    assert_empty @mech.history
+    assert_empty @mech.cookies
+  end
+
+  def test_start
+    body, id = nil
+
+    Mechanize.start do |m|
+      body = m.get("http://localhost/").body
+      id = m.agent.http.request_key
+    end
+
+    assert_match(/Hello World/, body)
+    assert_nil Thread.current[id]
+  end
+
   def test_submit_bad_form_method
     page = @mech.get("http://localhost/bad_form_test.html")
     assert_raises ArgumentError do