oga 0.2.2 → 0.2.3

data/ext/java/org/liboga/xml/Lexer.rl

@@ -187,14 +187,13 @@ public class Lexer extends RubyObject
     }
 
     /**
-     * Returns true if we're in an HTML script tag. See
-     * Oga::XML::Lexer#inside_html_script? for more information.
+     * See * Oga::XML::Lexer#literal_html_element? for more information.
      */
-    public Boolean inside_html_script_p()
+    public Boolean literal_html_element_p()
     {
         ThreadContext context = this.runtime.getCurrentContext();
 
-        return this.callMethod(context, "inside_html_script?").isTrue();
+        return this.callMethod(context, "literal_html_element?").isTrue();
     }
 }
 

data/ext/ragel/base_lexer.rl

@@ -328,11 +328,9 @@
         '>' => {
             callback_simple(id_on_element_open_end);
 
-            if ( inside_html_script_p() )
+            if ( literal_html_element_p() )
             {
-                mark = ts + 1;
-
-                fnext script_text;
+                fnext literal_html_element;
             }
             else
             {
@@ -401,14 +399,15 @@
         };
     *|;
 
-    # <script> tags in HTML can contain basically anything except for the
-    # literal "</script>". As a result of this we can't use the regular text
-    # machine.
-    script_text := |*
-        '</script>' => {
-            callback(id_on_text, data, encoding, mark, ts);
+    # Certain tags in HTML can contain basically anything except for the literal
+    # closing tag. Two examples are script and style tags.  As a result of this
+    # we can't use the regular text machine.
+    literal_html_closing_tags = '</script>' | '</style>';
+    literal_html_allowed = (any* -- literal_html_closing_tags) $count_newlines;
 
-            mark = 0;
+    literal_html_element := |*
+        literal_html_allowed => {
+            callback(id_on_text, data, encoding, ts, te);
 
             if ( lines > 0 )
             {
@@ -416,13 +415,23 @@
 
                 lines = 0;
             }
+        };
 
-            callback_simple(id_on_element_end);
+        literal_html_allowed %{ mark = p; } literal_html_closing_tags => {
+            callback(id_on_text, data, encoding, ts, mark);
+
+            p    = mark - 1;
+            mark = 0;
+
+            if ( lines > 0 )
+            {
+                advance_line(lines);
+
+                lines = 0;
+            }
 
             fnext main;
         };
-
-        any $count_newlines;
     *|;
 
     # The main machine aka the entry point of Ragel.

data/lib/oga/version.rb

@@ -1,3 +1,3 @@
 module Oga
-  VERSION = '0.2.2'
+  VERSION = '0.2.3'
 end # Oga

data/lib/oga/xml/lexer.rb

@@ -41,12 +41,11 @@ module Oga
       attr_reader :html
 
       ##
-      # Element name used to determine if a tag being processed is a Javascript
-      # tag.
+      # Names of HTML tags of which the content should be lexed as-is.
       #
-      # @return [String]
+      # @return [Array]
       #
-      SCRIPT_TAG = 'script'.freeze
+      LITERAL_HTML_ELEMENTS = %w{script style}
 
       ##
       # @param [String|IO] data The data to lex. This can either be a String or
@@ -190,12 +189,12 @@ module Oga
       end
 
       ##
-      # Returns true if the current element is the HTML `<script>` element.
+      # Returns true if the current element's content should be lexed as-is.
       #
       # @return [TrueClass|FalseClass]
       #
-      def inside_html_script?
-        return html? && current_element == SCRIPT_TAG
+      def literal_html_element?
+        return html? && LITERAL_HTML_ELEMENTS.include?(current_element)
       end
 
       ##

data/lib/oga/xml/text.rb

@@ -12,8 +12,8 @@ module Oga
         node = parent
         root = root_node
 
-        if root.is_a?(Document) and node.is_a?(Element) \
-        and node.name == Lexer::SCRIPT_TAG and root.html?
+        if root.is_a?(Document) and node.is_a?(Element) and root.html? \
+        and Lexer::LITERAL_HTML_ELEMENTS.include?(node.name)
           return super
         else
           return Entities.encode(super)

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: oga
 version: !ruby/object:Gem::Version
-  version: 0.2.2
+  version: 0.2.3
 platform: ruby
 authors:
 - Yorick Peterse
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-03-03 00:00:00.000000000 Z
+date: 2015-03-04 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: racc