mechanize 0.5.4 → 0.6.0

data/lib/mechanize/form_elements.rb

@@ -71,12 +71,12 @@ module WWW
       super(name, value)
     end
 
-    def tick
+    def check
       uncheck_peers
       @checked = true
     end
 
-    def untick
+    def uncheck
       @checked = false
     end
 
@@ -88,7 +88,7 @@ module WWW
     def uncheck_peers
       @form.radiobuttons.name(name).each do |b|
         next if b.value == value
-        b.untick
+        b.uncheck
       end
     end
   end
@@ -117,12 +117,10 @@ module WWW
       @options = WWW::Mechanize::List.new
   
       # parse
-      node.each_recursive {|n|
-        if n.name.downcase == 'option'
-          option = Option.new(n, self)
-          @options << option
-        end
-      }
+      (node/'option').each do |n|
+        option = Option.new(n, self)
+        @options << option
+      end
       super(name, value)
     end
 

data/lib/mechanize/hpricot.rb

@@ -0,0 +1,12 @@
+require 'hpricot'
+class Hpricot::Elem
+  def all_text
+    text = ''
+    children.each do |child|
+      if child.respond_to? :content
+        text << child.content
+      end
+    end
+    text
+  end
+end

data/lib/mechanize/inspect.rb

@@ -1,5 +1,3 @@
-require 'pp'
-
 module WWW
   # :stopdoc:
   class Mechanize
@@ -11,7 +9,6 @@ module WWW
         q.pp current_page
       }
     end
-    alias :inspect :pretty_print_inspect
 
     class Page
       def pretty_print(q)
@@ -43,7 +40,6 @@ module WWW
           }
         }
       end
-      alias :inspect :pretty_print_inspect
     end
 
     class Link
@@ -53,7 +49,6 @@ module WWW
           q.breakable; q.pp href
         }
       end
-      alias :inspect :pretty_print_inspect
     end
 
     class Form
@@ -82,7 +77,6 @@ module WWW
           }
         }
       end
-      alias :inspect :pretty_print_inspect
     end
 
     class RadioButton

data/lib/mechanize/mech_version.rb

@@ -1,7 +1,5 @@
-# DO NOT EDIT
-# This file is auto-generated by build scripts
 module WWW
   class Mechanize
-    Version = '0.5.4'
+    Version = '0.6.0'
   end
 end

data/lib/mechanize/page.rb

@@ -1,4 +1,5 @@
 require 'fileutils'
+require 'hpricot'
 
 module WWW
   class Mechanize
@@ -14,27 +15,18 @@ module WWW
     #  agent.get('http://google.com/').class  #=> WWW::Mechanize::Page
     #
     class Page < File
-      attr_accessor :watch_for_set
-      attr_finder :frames, :iframes, :links, :forms, :meta, :watches
+      attr_reader :root, :title, :watch_for_set
+      attr_reader :frames, :iframes, :links, :forms, :meta, :watches
 
-      # Alias our finders so that we can lazily parse the html
-      alias :find_frames   :frames
-      alias :find_iframes  :iframes
-      alias :find_links    :links
-      alias :find_forms    :forms
-      alias :find_meta     :meta
-      alias :find_watches  :watches
-    
       def initialize(uri=nil, response=nil, body=nil, code=nil)
         super(uri, response, body, code)
-        @frames       = nil
-        @iframes      = nil
-        @links        = nil
-        @forms        = nil
-        @meta         = nil
-        @watches      = nil
-        @root         = nil
-        @title        = nil
+        @watch_for_set = {}
+
+        yield self if block_given?
+
+        raise Mechanize::ContentTypeError.new(response['content-type']) unless
+            content_type() =~ /^text\/html/ 
+        parse_html if body && response
       end
     
       # Get the response header
@@ -44,57 +36,23 @@ module WWW
     
       # Get the content type
       def content_type
-        @response['Content-Type']
-      end
-    
-      # Get a list of Form associated with this page.
-      def forms(*args)
-        parse_html() unless @forms
-        find_forms(*args)
-      end
-    
-      # Get a list of Link associated with this page.
-      def links(*args)
-        parse_html() unless @links
-        find_links(*args)
-      end
-    
-      # Get the root XML parse tree for this page.
-      def root
-        parse_html() unless @root
-        @root
-      end
-    
-      # This method watches out for a particular tag, and will call back to the
-      # class specified for the tag in the watch_for_set method.  See the example
-      # in this class.
-      def watches(*args)
-        parse_html() unless @watches 
-        find_watches(*args)
-      end
-    
-      # Get a list of Meta links, usually used for refreshing the page.
-      def meta(*args)
-        parse_html() unless @meta 
-        find_meta(*args)
+        @response['content-type']
       end
 
-      # Get a list of Frame from the page
-      def frames(*args)
-        parse_html() unless @frames
-        find_frames(*args)
+      # Search through the page like HPricot
+      def search(*args)
+        @root.search(*args)
       end
 
-      # Get a list of IFrame from the page
-      def iframes(*args)
-        parse_html() unless @iframes
-        find_iframes(*args)
+      def at(*args)
+        @root.at(*args)
       end
 
-      # Fetch the title of the page
-      def title
-        parse_html() unless @title
-        @title
+      alias :/ :search
+    
+      def watch_for_set=(obj)
+        @watch_for_set = obj
+        parse_html if @body
       end
 
       def form(name)
@@ -104,29 +62,8 @@ module WWW
       private
     
       def parse_html
-        raise Mechanize::ContentTypeError.new(content_type()) unless
-          content_type() =~ /^text\/html/ 
-    
         # construct parser and feed with HTML
-        parser = HTMLTree::XMLParser.new
-        begin
-          parser.feed(@body)
-        rescue => ex
-          if ex.message =~ /attempted adding second root element to document/ and
-            # Put the whole document inside a single root element, which I
-            # simply name <root>, just to make the parser happy. It's no
-            #longer valid HTML, but without a single root element, it's not
-            # valid HTML as well.
-  
-            # TODO: leave a possible doctype definition outside this element.
-            parser = HTMLTree::XMLParser.new
-            parser.feed("<root>" + @body + "</root>")
-          else
-            raise
-          end
-        end
-    
-        @root = parser.document
+        @root = Hpricot.parse(@body)
     
         @forms    = WWW::Mechanize::List.new
         @links    = WWW::Mechanize::List.new
@@ -135,39 +72,57 @@ module WWW
         @iframes  = WWW::Mechanize::List.new
         @watches  = {}
     
-        @root.each_recursive {|node|
-          name = node.name.downcase
-    
-          case name
-          when 'form'
-            form = Form.new(node)
-            form.action ||= @uri
-            @forms << form
-          when 'title'
-            @title = node.text
-          when 'a'
-            @links << Link.new(node)
-          when 'meta'
-            equiv   = node.attributes['http-equiv']
-            content = node.attributes['content']
-            if equiv != nil && equiv.downcase == 'refresh'
-              if content != nil && content =~ /^\d+\s*;\s*url\s*=\s*(\S+)/i
-                node.attributes['href'] = $1
-                @meta << Meta.new(node)
-              end
+        # Set the title
+        @title = if (@root/'title').text.length > 0
+          (@root/'title').text
+        end
+
+        # Find all the form tags
+        (@root/'form').each do |html_form|
+          form = Form.new(html_form)
+          form.action ||= @uri
+          @forms << form
+        end
+
+        # Find all the 'a' tags
+        (@root/'a').each do |node|
+          @links << Link.new(node)
+        end
+
+        # Find all 'meta' tags
+        (@root/'meta').each do |node|
+          next if node.attributes.nil?
+          next unless node.attributes.has_key? 'http-equiv'
+          next unless node.attributes.has_key? 'content'
+          equiv   = node.attributes['http-equiv']
+          content = node.attributes['content']
+          if equiv != nil && equiv.downcase == 'refresh'
+            if content != nil && content =~ /^\d+\s*;\s*url\s*=\s*(\S+)/i
+              node.attributes['href'] = $1
+              @meta << Meta.new(node)
             end
-          when 'frame'
-            @frames << Frame.new(node)
-          when 'iframe'
-            @iframes << Frame.new(node)
-          else
-            if @watch_for_set and @watch_for_set.keys.include?( name )
-              @watches[name] = [] unless @watches[name]
-              klass = @watch_for_set[name]
-              @watches[name] << (klass ? klass.new(node) : node)
+          end
+        end
+
+        # Find all 'frame' tags
+        (@root/'frame').each do |node|
+          @frames << Frame.new(node)
+        end
+
+        # Find all 'iframe' tags
+        (@root/'iframe').each do |node|
+          @iframes << Frame.new(node)
+        end
+
+        # Find all watch tags
+        unless @watch_for_set.nil?
+          @watch_for_set.each do |key, klass|
+            (@root/key).each do |node|
+              @watches[key] ||= []
+              @watches[key] << (klass ? klass.new(node) : node)
             end
           end
-        }
+        end
       end
     end
   end

data/lib/mechanize/page_elements.rb

@@ -12,22 +12,25 @@ module WWW
       attr_reader :node
       attr_reader :href
       attr_reader :text
+      attr_reader :attributes
       alias :to_s :text
     
       def initialize(node)
+        node.attributes ||= {}
         @node = node
         @href = node.attributes['href'] 
         @text = node.all_text
+        @attributes = node.attributes
 
         # If there is no text, try to find an image and use it's alt text
-        if (@text.nil? || @text.length == 0) && @node.has_elements?
+        if (@text.nil? || @text.length == 0) && (node/'img').length > 0
           @text = ''
-          @node.each_element { |e|
-            if e.name == 'img'
-              @text << (e.has_attributes? ? e.attributes['alt'] || '' : '')
-            end
-          }
+          (node/'img').each do |e|
+            e.attributes ||= {}
+            @text << (e.attributes.has_key?('alt') ? e.attributes['alt'] : '')
+          end
         end
+
       end
 
       def uri
@@ -51,6 +54,7 @@ module WWW
       alias :name :text
 
       def initialize(node)
+        node.attributes ||= {}
         @node = node
         @text = node.attributes['name']
         @href = node.attributes['src']

data/test/htdocs/frame_test.html

@@ -21,7 +21,7 @@
       <P>This frameset document contains:
       <UL>
          <LI><A href="/google.html">Some neat contents</A>
-         <LI><A href="/form_test.html">Form Test</A>
+         <LI><A href="/form_test.html" class="bar">Form Test</A>
          <LI><A href="/file_upload.html">Some other neat contents</A>
       </UL>
   </NOFRAMES>

data/test/htdocs/tc_no_attributes.html

@@ -0,0 +1,16 @@
+<html>
+  <meta>
+  <head><title></title>
+  <body>
+    <a>Hello</a>
+    <a><img /></a>
+    <form>
+      <input />
+      <select>
+        <option />
+      </select>
+      <textarea></textarea>
+    </form>
+    <frame></frame>
+  </body>
+</html>

data/test/tc_checkboxes.rb

@@ -15,7 +15,7 @@ class TestCheckBoxes < Test::Unit::TestCase
 
   def test_select_one
     form = @page.forms.first
-    form.checkboxes.name('green').tick
+    form.checkboxes.name('green').check
     assert_equal(true,  form.checkboxes.name('green').checked)
     assert_equal(false, form.checkboxes.name('red').checked)
     assert_equal(false, form.checkboxes.name('blue').checked)
@@ -26,7 +26,7 @@ class TestCheckBoxes < Test::Unit::TestCase
   def test_select_all
     form = @page.forms.first
     form.checkboxes.each do |b|
-      b.tick
+      b.check
     end
     form.checkboxes.each do |b|
       assert_equal(true, b.checked)
@@ -36,17 +36,17 @@ class TestCheckBoxes < Test::Unit::TestCase
   def test_select_none
     form = @page.forms.first
     form.checkboxes.each do |b|
-      b.untick
+      b.uncheck
     end
     form.checkboxes.each do |b|
       assert_equal(false, b.checked)
     end
   end
 
-  def test_tick_one
+  def test_check_one
     form = @page.forms.first
     assert_equal(2, form.checkboxes.name('green').length)
-    form.checkboxes.name('green')[1].tick
+    form.checkboxes.name('green')[1].check
     assert_equal(false,  form.checkboxes.name('green')[0].checked)
     assert_equal(true,  form.checkboxes.name('green')[1].checked)
     page = @agent.submit(form)
@@ -54,11 +54,11 @@ class TestCheckBoxes < Test::Unit::TestCase
     assert_equal('green:on', page.links.first.text)
   end
 
-  def test_tick_two
+  def test_check_two
     form = @page.forms.first
     assert_equal(2, form.checkboxes.name('green').length)
-    form.checkboxes.name('green')[0].tick
-    form.checkboxes.name('green')[1].tick
+    form.checkboxes.name('green')[0].check
+    form.checkboxes.name('green')[1].check
     assert_equal(true,  form.checkboxes.name('green')[0].checked)
     assert_equal(true,  form.checkboxes.name('green')[1].checked)
     page = @agent.submit(form)