Use separate Ragel machines for script/style tags

Previously a single Ragel machine was used for processing HTML
script and style tags. This had the unfortunate side-effect that the
following was not parsed correctly (while being valid HTML):

    <script>
    var foo = "</style>";
    </script>

The same applied to style tags:

    <style>
    /* </script> */
    </style>

By using separate machines we can work around the above issue. The
downside is that this can produce multiple T_TEXT nodes, which have to
be stitched back together in the parser.

lib/oga/xml/lexer.rb

@@ -40,12 +40,18 @@ module Oga
     class Lexer
       attr_reader :html
 
+      # @return [String]
+      HTML_SCRIPT = 'script'
+
+      # @return [String]
+      HTML_STYLE = 'style'
+
       ##
       # Names of HTML tags of which the content should be lexed as-is.
       #
       # @return [Array]
       #
-      LITERAL_HTML_ELEMENTS = %w{script style}
+      LITERAL_HTML_ELEMENTS = [HTML_SCRIPT, HTML_STYLE]
 
       ##
       # @param [String|IO] data The data to lex. This can either be a String or
@@ -189,12 +195,17 @@ module Oga
       end
 
       ##
-      # Returns true if the current element's content should be lexed as-is.
-      #
       # @return [TrueClass|FalseClass]
       #
-      def literal_html_element?
-        return html? && LITERAL_HTML_ELEMENTS.include?(current_element)
+      def html_script?
+        return html? && current_element == HTML_SCRIPT
+      end
+
+      ##
+      # @return [TrueClass|FalseClass]
+      #
+      def html_style?
+        return html? && current_element == HTML_STYLE
       end
 
       ##

lib/oga/xml/parser.rll

@@ -176,7 +176,17 @@ xml_decl
 # Plain text
 
 text
-  = T_TEXT { on_text(val[0]) }
+  = T_TEXT text_follow
+    {
+      text = val[1] ? val[0] + val[1] : val[0]
+
+      on_text(text)
+    }
+  ;
+
+text_follow
+  = T_TEXT text_follow { val[1] ? val[0] + val[1] : val[0] }
+  | _                  { nil }
   ;
 
 # Strings