oga 1.0.3-java → 1.1.0-java

data/ext/ragel/base_lexer.rl

@@ -419,18 +419,24 @@
         any $count_newlines;
     *|;
 
-    # Characters that can be used for unquoted HTML attribute values.
-    # See https://html.spec.whatwg.org/multipage/introduction.html#intro-early-example
-    # for more info.
-    html_unquoted_value =
-        ^(squote | dquote | whitespace_or_newline)
-        ^('`' | '=' | '<' | '>' | whitespace_or_newline)+;
-
     # Machine used after matching the "=" of an attribute and just before moving
     # into the actual attribute value.
     attribute_pre := |*
         whitespace_or_newline $count_newlines;
 
+        squote | dquote => {
+            fhold;
+
+            if ( lines > 0 )
+            {
+                advance_line(lines);
+
+                lines = 0;
+            }
+
+            fnext quoted_attribute_value;
+        };
+
         any => {
             fhold;
 
@@ -443,25 +449,33 @@
 
             if ( html_p )
             {
-                fnext html_attribute_value;
+                fnext unquoted_attribute_value;
             }
+            /* XML doesn't support unquoted attribute values */
             else
             {
-                fnext xml_attribute_value;
+                fret;
             }
         };
     *|;
 
-    # Machine used for processing HTML attribute values.
-    html_attribute_value := |*
-        squote | dquote => {
-            fhold;
-            fnext xml_attribute_value;
-        };
-
-        # Unquoted attribute values are lexed as if they were single quoted
-        # strings.
-        html_unquoted_value => {
+    # Machine for processing unquoted HTML attribute values.
+    #
+    # The HTML specification describes a set of characters that can be allowed
+    # in an unquoted value at https://html.spec.whatwg.org/multipage/introduction.html#intro-early-example.
+    #
+    # As is always the case with HTML everybody completely ignores this
+    # specification and thus every library and browser out these is expected to
+    # support input such as `<a href=lol("javascript","is","great")></a>.
+    #
+    # Oga too has to support this, thus the only characters it disallows in
+    # unquoted attribute values are:
+    #
+    # * > (used for terminating open tags)
+    # * whitespace
+    #
+    unquoted_attribute_value := |*
+        ^('>' | whitespace_or_newline)+ => {
             callback_simple(id_on_string_squote);
 
             callback(id_on_string_body, data, encoding, ts, te);
@@ -472,8 +486,8 @@
         any => hold_and_return;
     *|;
 
-    # Machine used for processing XML attribute values.
-    xml_attribute_value := |*
+    # Machine used for processing quoted XML/HTML attribute values.
+    quoted_attribute_value := |*
         # The following two actions use "fnext" instead of "fcall". Combined
         # with "element_head" using "fcall" to jump to this machine this means
         # we can return back to "element_head" after processing a single string.

data/lib/oga/version.rb

@@ -1,3 +1,3 @@
 module Oga
-  VERSION = '1.0.3'
+  VERSION = '1.1.0'
 end # Oga

data/lib/oga/xml/attribute.rb

@@ -101,7 +101,7 @@ module Oga
           full_name = name
         end
 
-        enc_value = value ? Entities.encode(value) : nil
+        enc_value = value ? Entities.encode_attribute(value) : nil
 
         %Q(#{full_name}="#{enc_value}")
       end

data/lib/oga/xml/entities.rb

@@ -32,6 +32,20 @@ module Oga
         '<' => '&lt;',
       }
 
+      ##
+      # Hash containing characters and the corresponding XML entities to use
+      # when encoding XML/HTML attribute values.
+      #
+      # @return [Hash]
+      #
+      ENCODE_ATTRIBUTE_MAPPING = {
+        '&' => '&amp;',
+        '>' => '&gt;',
+        '<' => '&lt;',
+        "'" => '&apos;',
+        '"' => '&quot;'
+      }
+
       ##
       # @return [String]
       #
@@ -56,6 +70,12 @@ module Oga
       #
       ENCODE_REGEXP = Regexp.new(ENCODE_MAPPING.keys.join('|'))
 
+      ##
+      # @return [Regexp]
+      #
+      ENCODE_ATTRIBUTE_REGEXP =
+        Regexp.new(ENCODE_ATTRIBUTE_MAPPING.keys.join('|'))
+
       ##
       # Decodes XML entities.
       #
@@ -87,6 +107,16 @@ module Oga
       def self.encode(input, mapping = ENCODE_MAPPING)
         input.gsub(ENCODE_REGEXP, mapping)
       end
+
+      ##
+      # Encodes special characters in an XML attribute value.
+      #
+      # @param [String] input
+      # @return [String]
+      #
+      def self.encode_attribute(input)
+        input.gsub(ENCODE_ATTRIBUTE_REGEXP, ENCODE_ATTRIBUTE_MAPPING)
+      end
     end # Entities
   end # XML
 end # Oga

data/lib/oga/xml/lexer.rb

@@ -499,7 +499,7 @@ module Oga
       ##
       # Called on the closing tag of an element.
       #
-      # @param [String] ns_name The name of the element (minus namespace
+      # @param [String] name The name of the element (minus namespace
       #  prefix). This is not set for self closing tags.
       #
       def on_element_end(name = nil)

data/lib/oga/xml/node.rb

@@ -152,6 +152,27 @@ module Oga
         return node_set.delete(self) if node_set
       end
 
+      ##
+      # Replaces the current node with another.
+      #
+      # @example Replacing with an element
+      #  element = Oga::XML::Element.new(:name => 'div')
+      #  some_node.replace(element)
+      #
+      # @example Replacing with a String
+      #  some_node.replace('this will replace the current node with a text node')
+      #
+      # @param [String|Oga::XML::Node] other
+      #
+      def replace(other)
+        if other.is_a?(String)
+          other = Text.new(:text => other)
+        end
+
+        before(other)
+        remove
+      end
+
       ##
       # Inserts the given node before the current node.
       #

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: oga
 version: !ruby/object:Gem::Version
-  version: 1.0.3
+  version: 1.1.0
 platform: java
 authors:
 - Yorick Peterse
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2015-06-16 00:00:00.000000000 Z
+date: 2015-06-29 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: ast