mechanize 2.5.1 → 2.6.0

data/lib/mechanize/download.rb

@@ -65,5 +65,7 @@ class Mechanize::Download
     end
   end
 
+  alias save_as save
+
 end
 

data/lib/mechanize/element_matcher.rb

@@ -6,14 +6,14 @@ module Mechanize::ElementMatcher
         criteria = if String === criteria then
                      {:name => criteria}
                    else
-                     criteria.map do |k, v|
-                       k = :dom_id if k.to_sym == :id
-                       k = :dom_class if k.to_sym == :class
-                       [k, v]
-                     end
+                     Hash[criteria.map do |k, v|
+                            k = :dom_id if k.to_sym == :id
+                            k = :dom_class if k.to_sym == :class
+                            [k, v]
+                          end]
                    end
 
-        f = #{plural}.find_all do |thing|
+        f = select_#{plural}(criteria.delete(:search)).find_all do |thing|
           criteria.all? do |k,v|
             v === thing.send(k)
           end
@@ -28,6 +28,17 @@ module Mechanize::ElementMatcher
         f
       end
 
+      def select_#{plural} selector
+        if selector.nil? then
+          #{plural}
+        else
+          nodes = search(selector)
+          #{plural}.find_all do |element|
+            nodes.include?(element.node)
+          end
+        end
+      end
+
       alias :#{singular} :#{singular}_with
     CODE
   end

data/lib/mechanize/file.rb

@@ -49,11 +49,18 @@ class Mechanize::File
   end
 
   ##
-  # Use this method to save the content of this object to +filename+
+  # Use this method to save the content of this object to +filename+.
+  #
+  #   file.save 'index.html'
+  #   file.save 'index.html' # saves index.html.1
+  #   file.save 'index.html'
 
   def save filename = nil
     filename = find_free_name filename
 
+		dirname = File.dirname filename
+		FileUtils.mkdir_p dirname
+
     open filename, 'wb' do |f|
       f.write body
     end
@@ -61,5 +68,22 @@ class Mechanize::File
 
   alias save_as save
 
+  ##
+  # Use this method to save the content of this object to +filename+.
+  # This method will overwrite any existing filename that exists with the
+  # same name.
+  #
+  #   file.save 'index.html'
+  #   file.save! 'index.html' # overwrite original file
+
+  def save! filename = nil
+    dirname = File.dirname filename
+    FileUtils.mkdir_p dirname
+
+    open filename, 'wb' do |f|
+      f.write body
+    end
+  end
+
 end
 

data/lib/mechanize/form.rb

@@ -1,7 +1,7 @@
 require 'mechanize/element_matcher'
 
 # This class encapsulates a form parsed out of an HTML page.  Each type of
-# input fields available in a form can be accessed through this object.
+# input field available in a form can be accessed through this object.
 #
 # == Examples
 #
@@ -38,7 +38,7 @@ class Mechanize::Form
   attr_reader :form_node
   attr_reader :page
 
-  def initialize(node, mech=nil, page=nil)
+  def initialize(node, mech = nil, page = nil)
     @enctype = node['enctype'] || 'application/x-www-form-urlencoded'
     @form_node        = node
     @action           = Mechanize::Util.html_unescape(node['action'])
@@ -60,26 +60,75 @@ class Mechanize::Form
 
   alias :has_key? :has_field?
 
+  # Returns whether or not the form contains a field with +value+
   def has_value?(value)
     fields.find { |f| f.value == value }
   end
 
-  def keys; fields.map { |f| f.name }; end
+  # Returns all field names (keys) for this form
+  def keys
+    fields.map { |f| f.name }
+  end
+
+  # Returns all field values for this form
+  def values
+    fields.map { |f| f.value }
+  end
+
+  # Returns all buttons of type Submit
+  def submits
+    @submits ||= buttons.select { |f| f.class == Submit }
+  end
+
+  # Returns all buttons of type Reset
+  def resets
+    @resets ||= buttons.select { |f| f.class == Reset }
+  end
+
+  # Returns all fields of type Text
+  def texts
+    @texts ||= fields.select { |f| f.class == Text }
+  end
+
+  # Returns all fields of type Hidden
+  def hiddens
+    @hiddens ||= fields.select { |f| f.class == Hidden }
+  end
+
+  # Returns all fields of type Textarea
+  def textareas
+    @textareas ||= fields.select { |f| f.class == Textarea }
+  end
 
-  def values; fields.map { |f| f.value }; end
+  # Returns all fields of type Keygen
+  def keygens
+    @keygens ||= fields.select { |f| f.class == Keygen }
+  end
+
+  # Returns whether or not the form contains a Submit button named +button_name+
+  def submit_button?(button_name)
+    submits.find { |f| f.name == button_name }
+  end
 
-  def submits  ; @submits   ||= buttons.select { |f| f.class == Submit   }; end
-  def resets   ; @resets    ||= buttons.select { |f| f.class == Reset    }; end
-  def texts    ; @texts     ||=  fields.select { |f| f.class == Text     }; end
-  def hiddens  ; @hiddens   ||=  fields.select { |f| f.class == Hidden   }; end
-  def textareas; @textareas ||=  fields.select { |f| f.class == Textarea }; end
-  def keygens  ; @keygens   ||=  fields.select { |f| f.class == Keygen   }; end
+  # Returns whether or not the form contains a Reset button named +button_name+
+  def reset_button?(button_name)
+    resets.find { |f| f.name == button_name }
+  end
 
-  def submit_button?(button_name)   submits.find{|f| f.name == button_name}; end
-  def reset_button?(button_name)     resets.find{|f| f.name == button_name}; end
-  def text_field?(field_name)         texts.find{|f| f.name == field_name}; end
-  def hidden_field?(field_name)     hiddens.find{|f| f.name == field_name}; end
-  def textarea_field?(field_name) textareas.find{|f| f.name == field_name}; end
+  # Returns whether or not the form contains a Text field named +field_name+
+  def text_field?(field_name)
+    texts.find { |f| f.name == field_name }
+  end
+
+  # Returns whether or not the form contains a Hidden field named +field_name+
+  def hidden_field?(field_name)
+    hiddens.find { |f| f.name == field_name }
+  end
+
+  # Returns whether or not the form contains a Textarea named +field_name+
+  def textarea_field?(field_name)
+    textareas.find { |f| f.name == field_name }
+  end
 
   # This method is a shortcut to get form's DOM id.
   # Common usage:
@@ -117,7 +166,6 @@ class Mechanize::Form
   # following:
   #
   #   form.set_fields :foo => { 1 => 'bar' }
-
   def set_fields fields = {}
     fields.each do |name, v|
       case v
@@ -139,18 +187,14 @@ class Mechanize::Form
     end
   end
 
-  # Fetch the value of the first input field with the name passed in
-  # ==Example
-  # Fetch the value set in the input field 'name'
+  # Fetch the value of the first input field with the name passed in. Example:
   #  puts form['name']
   def [](field_name)
     f = field(field_name)
     f && f.value
   end
 
-  # Set the value of the first input field with the name passed in
-  # ==Example
-  # Set the value in the input field 'name' to "Aaron"
+  # Set the value of the first input field with the name passed in. Example:
   #  form['name'] = 'Aaron'
   def []=(field_name, value)
     f = field(field_name)
@@ -173,8 +217,9 @@ class Mechanize::Form
     super
   end
 
-  # Submit this form with the button passed in
-  def submit button=nil, headers = {}
+  # Submit the form. Does not include the +button+ as a form parameter.
+  # Use +click_button+ or provide button as a parameter.
+  def submit button = nil, headers = {}
     @mech.submit(self, button, headers)
   end
 
@@ -209,7 +254,9 @@ class Mechanize::Form
 
     successful_controls = []
 
-    (fields + checkboxes).sort.each do |f|
+    (fields + checkboxes).reject do |f|
+      f.node["disabled"]
+    end.sort.each do |f|
       case f
       when Mechanize::Form::CheckBox
         if f.checked
@@ -481,7 +528,7 @@ class Mechanize::Form
       next if type == 'reset'
       @buttons << Button.new(node)
     end
-    
+
     # Find all keygen tags
     form_node.search('keygen').each do |node|
       @fields << Keygen.new(node, node['value'] || '')
@@ -532,7 +579,6 @@ class Mechanize::Form
 
     body
   end
-
 end
 
 require 'mechanize/form/field'
@@ -550,4 +596,3 @@ require 'mechanize/form/option'
 require 'mechanize/form/radio_button'
 require 'mechanize/form/check_box'
 require 'mechanize/form/select_list'
-

data/lib/mechanize/form/field.rb

@@ -16,9 +16,13 @@
 class Mechanize::Form::Field
   attr_accessor :name, :value, :node, :type
 
+  # This fields value before it's sent through Util.html_unescape.
+  attr_reader :raw_value
+
   def initialize node, value = node['value']
     @node = node
     @name = Mechanize::Util.html_unescape(node['name'])
+    @raw_value = value
     @value = if value.is_a? String
                Mechanize::Util.html_unescape(value)
              else
@@ -45,7 +49,7 @@ class Mechanize::Form::Field
     node['id']
   end
 
-  # This method is a shortcut to get field's DOM id.
+  # This method is a shortcut to get field's DOM class.
   # Common usage: form.field_with(:dom_class => "foo")
   def dom_class
     node['class']
@@ -54,7 +58,7 @@ class Mechanize::Form::Field
   def inspect # :nodoc:
     "[%s:0x%x type: %s name: %s value: %s]" % [
       self.class.name.sub(/Mechanize::Form::/, '').downcase,
-      object_id, @type, @name, @value
+      object_id, type, name, value
     ]
   end
 

data/lib/mechanize/form/select_list.rb

@@ -28,11 +28,11 @@ class Mechanize::Form::SelectList < Mechanize::Form::MultiSelectList
     end
   end
 
-  def value=(new)
-    if new != new.to_s and new.respond_to? :first
-      super([new.first])
+  def value=(new_value)
+    if new_value != new_value.to_s and new_value.respond_to? :first
+      super([new_value.first])
     else
-      super([new.to_s])
+      super([new_value.to_s])
     end
   end
 

data/lib/mechanize/http/agent.rb

@@ -69,6 +69,12 @@ class Mechanize::HTTP::Agent
   # Maximum number of redirects to follow
   attr_accessor :redirection_limit
 
+  # :section: Allowed error codes
+
+  # List of error codes to handle without raising an exception.
+
+  attr_accessor :allowed_error_codes
+
   # :section: Robots
 
   # When true, this agent will consult the site's robots.txt for each access.
@@ -120,6 +126,7 @@ class Mechanize::HTTP::Agent
   # implementation detail of mechanize and its API may change at any time.
 
   def initialize
+    @allowed_error_codes      = []
     @conditional_requests     = true
     @context                  = nil
     @content_encoding_hooks   = []
@@ -239,11 +246,9 @@ class Mechanize::HTTP::Agent
     end
 
     # Add If-Modified-Since if page is in history
-    page = visited_page(uri)
-
-    if (page = visited_page(uri)) and page.response['Last-Modified']
-      request['If-Modified-Since'] = page.response['Last-Modified']
-    end if(@conditional_requests)
+    if page = visited_page(uri) and last_modified = page.response['Last-Modified']
+      request['If-Modified-Since'] = last_modified
+    end if @conditional_requests
 
     # Specify timeouts if given
     connection.open_timeout = @open_timeout if @open_timeout
@@ -283,12 +288,12 @@ class Mechanize::HTTP::Agent
     meta = response_follow_meta_refresh response, uri, page, redirects
     return meta if meta
 
+    if robots && page.is_a?(Mechanize::Page)
+      page.parser.noindex? and raise Mechanize::RobotsDisallowedError.new(uri)
+    end
+
     case response
     when Net::HTTPSuccess
-      if robots && page.is_a?(Mechanize::Page)
-        page.parser.noindex? and raise Mechanize::RobotsDisallowedError.new(uri)
-      end
-
       page
     when Mechanize::FileResponse
       page
@@ -301,7 +306,11 @@ class Mechanize::HTTP::Agent
       response_authenticate(response, page, uri, request, headers, params,
                             referer)
     else
-      raise Mechanize::ResponseCodeError.new(page, 'unhandled response')
+      if @allowed_error_codes.any? {|code| code.to_s == page.code} then
+        page
+      else
+        raise Mechanize::ResponseCodeError.new(page, 'unhandled response')
+      end
     end
   end
 
@@ -403,6 +412,11 @@ class Mechanize::HTTP::Agent
     end
   end
 
+  # Closes all open connections for this agent.
+  def shutdown
+    http.shutdown
+  end
+
   ##
   # Decodes a gzip-encoded +body_io+.  If it cannot be decoded, inflate is
   # tried followed by raising an error.
@@ -431,7 +445,7 @@ class Mechanize::HTTP::Agent
     end
   ensure
     # do not close a second time if we failed the first time
-    zio.close if zio and not (zio.closed? or gz_error)
+    zio.close if zio and !(zio.closed? or gz_error)
     body_io.close unless body_io.closed?
   end
 
@@ -622,6 +636,8 @@ class Mechanize::HTTP::Agent
       end
     end
 
+    uri.host = referer_uri.host if referer_uri && URI::HTTP === uri && uri.host.nil?
+
     scheme = uri.relative? ? 'relative' : uri.scheme.downcase
     uri = @scheme_handlers[scheme].call(uri, referer)
 
@@ -696,7 +712,7 @@ class Mechanize::HTTP::Agent
       message = 'WWW-Authenticate header missing in response'
       raise Mechanize::UnauthorizedError.new(page, nil, message)
     end
-                                               
+
     challenges = @authenticate_parser.parse www_authenticate
 
     unless @auth_store.credentials? uri, challenges then
@@ -800,7 +816,7 @@ class Mechanize::HTTP::Agent
     begin
       if Tempfile === body_io and
          (StringIO === out_io or out_io.path != body_io.path) then
-        body_io.close! 
+        body_io.close!
       end
     rescue IOError
       # HACK ruby 1.8 raises IOError when closing the stream
@@ -845,7 +861,7 @@ class Mechanize::HTTP::Agent
     sleep delay
     @history.push(page, page.uri)
     fetch new_url, :get, {}, [],
-          Mechanize::Page.new, redirects
+          Mechanize::Page.new, redirects + 1
   end
 
   def response_log response
@@ -1147,7 +1163,7 @@ class Mechanize::HTTP::Agent
 
     out_io
   ensure
-    inflate.close
+    inflate.close if inflate.finished?
   end
 
   def log
@@ -1200,6 +1216,11 @@ class Mechanize::HTTP::Agent
     size >= @max_file_buffer
   end
 
+  def reset
+    @cookie_jar.clear!
+    @history.clear
+  end
+
 end
 
 require 'mechanize/http/auth_store'