* lib/hpricot/elements.rb: added block syntax to attr, new methods…

… `remove_attr` and `remove_class`. Got rid of `set`, making it an alias for `attr` which does much more.

git-svn-id: https://code.whytheluckystiff.net/svn/hpricot/trunk@150 9e82565c-fa0a-db11-9fa0-00132028b6dc

README

@@ -257,7 +257,7 @@ So, let's go beyond just trying to fix the hierarchy.  The
 
 What measures does <tt>:xhtml_strict</tt> take?
 
- 1. Shift elements into their proper containers just like <tt>:fixup_tags</tt>.
+ 1. Shift elements into their proper containers just like :fixup_tags.
  2. Remove unknown elements.
  3. Remove unknown attributes.
  4. Remove illegal content.
@@ -270,6 +270,9 @@ on the standard mode.  The main difference is that :xml mode won't try to output
 tags which are friendlier for browsers.  For example, if an opening and closing
 <tt>br</tt> tag is found, XML mode won't try to turn that into an empty element.
 
+XML mode also doesn't downcase the tags and attributes for you.  So pay attention
+to case, friends.
+
 The primary way to use Hpricot's XML mode is to call the Hpricot.XML method:
 
  doc = open("http://redhanded.hobix.com/index.xml") do |f|

lib/hpricot/elements.rb

@@ -172,10 +172,36 @@ def wrap(str = nil, &blk)
       end
     end
 
-    def attr key, value = nil
-      if value
+    # Gets and sets attributes on all matched elements.
+    #
+    # Pass in a +key+ on its own and this method will return the string value
+    # assigned to that attribute for the first elements.  Or +nil+ if the 
+    # attribute isn't found.
+    #
+    #   doc.search("a").attr("href")
+    #     #=> "http://hacketyhack.net/"
+    #
+    # Or, pass in a +key+ and +value+.  This will set an attribute for all
+    # matched elements.
+    #
+    #   doc.search("p").attr("class", "basic")
+    # 
+    # You may also use a Hash to set a series of attributes:
+    #
+    #   (doc/"a").attr(:class => "basic", :href => "http://hackety.org/")
+    # 
+    # Lastly, a block can be used to rewrite an attribute based on the element
+    # it belongs to.  The block will pass in an element.  Return from the block
+    # the new value of the attribute.
+    #
+    #   records.attr("href") { |e| e['href'] + "#top" }
+    #
+    # This example adds a <tt>#top</tt> anchor to each link.
+    #
+    def attr key, value = nil, &blk
+      if value or blk
         each do |el|
-          el.set_attribute(key, value)
+          el.set_attribute(key, value || blk[el])
         end
         return self      
       end    
@@ -186,7 +212,13 @@ def attr key, value = nil
         return self[0].get_attribute(key)
       end
     end
+    alias_method :set, :attr
 
+    # Adds the class to all matched elements.
+    #
+    #   (doc/"p").add_class("bacon")
+    #
+    # Now all paragraphs will have class="bacon".
     def add_class class_name
       each do |el|
         next unless el.respond_to? :get_attribute
@@ -196,26 +228,37 @@ def add_class class_name
       self
     end
 
-    # Sets an attribute for all elements in this list.  You may use
-    # a simple pair (<em>attribute name</em>, <em>attribute value</em>):
+    # Remove an attribute from each of the matched elements.
     #
-    #   doc.search('p').set(:class, 'outline')
+    #   (doc/"input").remove_attr("disabled")
     #
-    # Or, use a hash of pairs:
+    def remove_attr name
+      each do |el|
+        next unless el.respond_to? :remove_attribute
+        el.remove_attribute(name)
+      end
+      self      
+    end
+
+    # Removes a class from all matched elements.
     #
-    #   doc.search('div#sidebar').set(:class => 'outline', :id => 'topbar')
+    #   (doc/"span").remove_class("lightgrey")
     #
-    def set(k, v = nil)
-      case k
-      when Hash
-        each do |node|
-          k.each { |a,b| node.set_attribute(a, b) }
-        end
-      else
-        each do |node|
-          node.set_attribute(k, v)
+    # Or, to remove all classes:
+    #
+    #   (doc/"span").remove_class
+    # 
+    def remove_class name = nil
+      each do |el|
+        next unless el.respond_to? :get_attribute
+        if name
+          classes = el.get_attribute('class').to_s.split(" ")
+          el.set_attribute('class', (classes - [name]).uniq.join(" "))
+        else
+          el.remove_attribute("class")
         end
       end
+      self      
     end
 
     ATTR_RE = %r!\[ *(?:(@)([\w\(\)-]+)|([\w\(\)-]+\(\))) *([~\!\|\*$\^=]*) *'?"?([^\]'"]*)'?"? *\]!i

test/test_alter.rb

@@ -37,10 +37,27 @@ def test_add_class
   def test_change_attributes
     all_ps = (@basic/"p").attr("title", "Some Title")
     all_as = (@basic/"a").attr("href", "http://my_new_href.com")
+    all_lb = (@basic/"link").attr("href") { |e| e.name }
     assert_changed(@basic, "p", all_ps) {|p| p.attributes["title"] == "Some Title"}
     assert_changed(@basic, "a", all_as) {|a| a.attributes["href"] == "http://my_new_href.com"}
+    assert_changed(@basic, "link", all_lb) {|a| a.attributes["href"] == "link" }
   end
 
+  def test_remove_attr
+    all_rl = (@basic/"link").remove_attr("href")
+    assert_changed(@basic, "link", all_rl) { |link| link['href'].nil? }
+  end
+
+  def test_remove_class
+    all_c1 = (@basic/"p[@class*='last']").remove_class("last")
+    assert_changed(@basic, "p[@class*='last']", all_c1) { |p| p['class'] == 'final' }
+  end
+
+  def test_remove_all_classes
+    all_c2 = (@basic/"p[@class]").remove_class
+    assert_changed(@basic, "p[@class]", all_c2) { |p| p['class'].nil? }
+  end
+
   def assert_changed original, selector, set, &block
     assert set.all?(&block)
     assert Hpricot(original.to_html).search(selector).all?(&block)