kramdown 0.7.0 → 0.8.0

data/lib/kramdown/parser/kramdown.rb

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 #
 #--
-# Copyright (C) 2009 Thomas Leitner <t_leitner@gmx.at>
+# Copyright (C) 2009-2010 Thomas Leitner <t_leitner@gmx.at>
 #
 # This file is part of kramdown.
 #
@@ -69,7 +69,7 @@ module Kramdown
     #
     #   Kramdown::Document.new(input_text, :input => 'ERBKramdown').to_html
     #
-    class Kramdown
+    class Kramdown < Base
 
       include ::Kramdown
 
@@ -79,13 +79,12 @@ module Kramdown
 
       # Create a new Kramdown parser object for the Kramdown::Document +doc+.
       def initialize(doc)
-        @doc = doc
-        @extension = @doc.options[:extension] || Kramdown::Parser::Kramdown::Extension.new
+        super(doc)
 
         @src = nil
         @tree = nil
         @stack = []
-        @text_type = :text
+        @text_type = :raw_text
         @block_ial = nil
 
         @doc.parse_infos[:ald] = {}
@@ -95,7 +94,7 @@ module Kramdown
 
         @block_parsers = [:blank_line, :codeblock, :codeblock_fenced, :blockquote, :table, :atx_header,
                           :setext_header, :horizontal_rule, :list, :definition_list, :link_definition, :block_html,
-                          :footnote_definition, :abbrev_definition, :ald, :block_math, :extension_block_depr,
+                          :footnote_definition, :abbrev_definition, :ald, :block_math,
                           :block_extension, :block_ial, :eob_marker, :paragraph]
         @span_parsers =  [:emphasis, :codespan, :autolink, :span_html, :footnote_marker, :link, :smart_quotes, :inline_math,
                          :span_extension, :span_ial, :html_entity, :typographic_syms, :line_break, :escaped_chars]
@@ -104,11 +103,6 @@ module Kramdown
       private_class_method(:new, :allocate)
 
 
-      # Parse the string +source+ using the Kramdown::Document +doc+ and return the parse tree.
-      def self.parse(source, doc)
-        new(doc).parse(source)
-      end
-
       # The source string provided on initialization is parsed and the created +tree+ is returned.
       def parse(source)
         configure_parser
@@ -122,12 +116,6 @@ module Kramdown
         tree
       end
 
-      # Add the given warning +text+ to the warning array of the Kramdown document.
-      def warning(text)
-        @doc.warnings << text
-        #TODO: add position information
-      end
-
       #######
       protected
       #######
@@ -177,12 +165,12 @@ module Kramdown
         status
       end
 
-      # Update the tree by parsing all <tt>:text</tt> elements with the span level parser (resets
-      # +@tree+, +@src+ and the +@stack+) and by updating the attributes from the IALs.
+      # Update the tree by parsing all <tt>:raw_text</tt> elements with the span level parser
+      # (resets +@tree+, +@src+ and the +@stack+) and by updating the attributes from the IALs.
       def update_tree(element)
         element.children.map! do |child|
-          if child.type == :text
-            @stack, @tree = [], nil
+          if child.type == :raw_text
+            @stack, @tree, @text_type = [], nil, :text
             @src = StringScanner.new(child.value)
             parse_spans(child)
             child.children
@@ -195,7 +183,7 @@ module Kramdown
       end
 
       # Parse all span level elements in the source string.
-      def parse_spans(el, stop_re = nil, parsers = nil, text_type = :text)
+      def parse_spans(el, stop_re = nil, parsers = nil, text_type = @text_type)
         @stack.push([@tree, @text_type]) unless @tree.nil?
         @tree, @text_type = el, text_type
 
@@ -232,21 +220,6 @@ module Kramdown
         stop_re_found
       end
 
-      # Modify the string +source+ to be usable by the parser.
-      def adapt_source(source)
-        source.gsub(/\r\n?/, "\n").chomp + "\n"
-      end
-
-      # This helper method adds the given +text+ either to the last element in the +tree+ if it is a
-      # +type+ element or creates a new text element with the given +type+.
-      def add_text(text, tree = @tree, type = @text_type)
-        if tree.children.last && tree.children.last.type == type
-          tree.children.last.value << text
-        elsif !text.empty?
-          tree.children << Element.new(type, text)
-        end
-      end
-
       # Update the attributes with the information from the inline attribute list and all referenced ALDs.
       def update_attr_with_ial(attr, ial)
         ial[:refs].each do |ref|
@@ -259,29 +232,11 @@ module Kramdown
       # Create a new block level element, taking care of applying a preceding block IAL if it exists.
       def new_block_el(*args)
         el = Element.new(*args)
+        el.options[:category] ||= :block
         el.options[:ial] = @block_ial if @block_ial && el.type != :blank && el.type != :eob
         el
       end
 
-      # Extract the part of the StringScanner backed string specified by the +range+. This method
-      # also works correctly under Ruby 1.9.
-      def extract_string(range, strscan = @src)
-        result = nil
-        if RUBY_VERSION >= '1.9'
-          begin
-            enc = strscan.string.encoding
-            strscan.string.force_encoding('ASCII-8BIT')
-            result = strscan.string[range].force_encoding(enc)
-          ensure
-            strscan.string.force_encoding(enc)
-          end
-        else
-          result = strscan.string[range]
-        end
-        result
-      end
-
-
       @@parsers = {}
 
       # Holds all the needed data for one block/span level parser.

data/lib/kramdown/parser/kramdown/attribute_list.rb

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 #
 #--
-# Copyright (C) 2009 Thomas Leitner <t_leitner@gmx.at>
+# Copyright (C) 2009-2010 Thomas Leitner <t_leitner@gmx.at>
 #
 # This file is part of kramdown.
 #

data/lib/kramdown/parser/kramdown/autolink.rb

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 #
 #--
-# Copyright (C) 2009 Thomas Leitner <t_leitner@gmx.at>
+# Copyright (C) 2009-2010 Thomas Leitner <t_leitner@gmx.at>
 #
 # This file is part of kramdown.
 #

data/lib/kramdown/parser/kramdown/blank_line.rb

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 #
 #--
-# Copyright (C) 2009 Thomas Leitner <t_leitner@gmx.at>
+# Copyright (C) 2009-2010 Thomas Leitner <t_leitner@gmx.at>
 #
 # This file is part of kramdown.
 #

data/lib/kramdown/parser/kramdown/blockquote.rb

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 #
 #--
-# Copyright (C) 2009 Thomas Leitner <t_leitner@gmx.at>
+# Copyright (C) 2009-2010 Thomas Leitner <t_leitner@gmx.at>
 #
 # This file is part of kramdown.
 #

data/lib/kramdown/parser/kramdown/codeblock.rb

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 #
 #--
-# Copyright (C) 2009 Thomas Leitner <t_leitner@gmx.at>
+# Copyright (C) 2009-2010 Thomas Leitner <t_leitner@gmx.at>
 #
 # This file is part of kramdown.
 #

data/lib/kramdown/parser/kramdown/codespan.rb

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 #
 #--
-# Copyright (C) 2009 Thomas Leitner <t_leitner@gmx.at>
+# Copyright (C) 2009-2010 Thomas Leitner <t_leitner@gmx.at>
 #
 # This file is part of kramdown.
 #

data/lib/kramdown/parser/kramdown/emphasis.rb

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 #
 #--
-# Copyright (C) 2009 Thomas Leitner <t_leitner@gmx.at>
+# Copyright (C) 2009-2010 Thomas Leitner <t_leitner@gmx.at>
 #
 # This file is part of kramdown.
 #

data/lib/kramdown/parser/kramdown/eob.rb

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 #
 #--
-# Copyright (C) 2009 Thomas Leitner <t_leitner@gmx.at>
+# Copyright (C) 2009-2010 Thomas Leitner <t_leitner@gmx.at>
 #
 # This file is part of kramdown.
 #

data/lib/kramdown/parser/kramdown/escaped_chars.rb

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 #
 #--
-# Copyright (C) 2009 Thomas Leitner <t_leitner@gmx.at>
+# Copyright (C) 2009-2010 Thomas Leitner <t_leitner@gmx.at>
 #
 # This file is part of kramdown.
 #

data/lib/kramdown/parser/kramdown/extension.rb

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 #
 #--
-# Copyright (C) 2009 Thomas Leitner <t_leitner@gmx.at>
+# Copyright (C) 2009-2010 Thomas Leitner <t_leitner@gmx.at>
 #
 # This file is part of kramdown.
 #
@@ -26,94 +26,6 @@ module Kramdown
   module Parser
     class Kramdown
 
-      # The base extension class.
-      #
-      # This class provides implementations for the default extensions defined in the kramdown
-      # specification.
-      #
-      # An extension is a method called <tt>parse_EXTNAME</tt> where +EXTNAME+ is the extension name.
-      # These methods are called with three parameters:
-      #
-      # [+parser+]
-      #    The parser instance from which the extension method is called.
-      # [+opts+]
-      #    A hash containing the options set in the extension.
-      # [+body+]
-      #    A string containing the body of the extension. If no body is available, this is +nil+.
-      class Extension
-
-        # Just ignore everything and do nothing.
-        def parse_comment(parser, opts, body)
-          nil
-        end
-
-        # Add the body (if available) as <tt>:raw</tt> Element to the +parser.tree+.
-        def parse_nomarkdown(parser, opts, body)
-          parser.tree.children << Element.new(:raw, body) if body.kind_of?(String)
-        end
-
-        # Update the document and parser options with the options set in +opts+.
-        def parse_options(parser, opts, body)
-          opts.select do |k,v|
-            k = k.to_sym
-            if Kramdown::Options.defined?(k)
-              parser.doc.options[k] = Kramdown::Options.parse(k, v) rescue parser.doc.options[k]
-              false
-            else
-              true
-            end
-          end.each do |k,v|
-            parser.warning("Unknown kramdown option '#{k}'")
-          end
-        end
-
-      end
-
-
-      EXT_BLOCK_START_STR_DEPR = "^#{OPT_SPACE}\\{::(%s):(:)?(#{ALD_ANY_CHARS}*)\\}\s*?\n"
-      EXT_BLOCK_START_DEPR = /#{EXT_BLOCK_START_STR_DEPR % ALD_ID_NAME}/
-
-      # Parse the block extension at the current location.
-      def parse_extension_block_depr
-        @src.pos += @src.matched_size
-
-        ext = @src[1]
-        opts = {}
-        body = nil
-        parse_attribute_list(@src[3], opts)
-
-        warn('DEPRECATION warning: This syntax is deprecated, use the new extension syntax')
-        if !%w{comment nomarkdown options}.include?(ext)
-          warn('DEPRECATION warning: Custom extensions will be removed in a future version - use a template processor like ERB instead')
-        end
-
-        if !@extension.public_methods.map {|m| m.to_s}.include?("parse_#{ext}")
-          warning("No extension named '#{ext}' found - ignoring extension block")
-          body = :invalid
-        end
-
-        if !@src[2]
-          stop_re = /#{EXT_BLOCK_START_STR_DEPR % ext}/
-          if result = @src.scan_until(stop_re)
-            parse_attribute_list(@src[3], opts)
-            body = result.sub!(stop_re, '') if body != :invalid
-          else
-            body = :invalid
-            warning("No ending line for extension block '#{ext}' found - ignoring extension block")
-          end
-        end
-
-        @extension.send("parse_#{ext}", self, opts, body) if body != :invalid
-
-        true
-      end
-      define_parser(:extension_block_depr, EXT_BLOCK_START_DEPR)
-
-
-      ##########################################
-      ### Code for handling new extension syntax
-      ##########################################
-
       def parse_extension_start_tag(type)
         @src.pos += @src.matched_size
 
@@ -145,7 +57,7 @@ module Kramdown
         when 'comment'
           # nothing to do
         when 'nomarkdown'
-          @tree.children << Element.new(:raw, body, :type => type) if body.kind_of?(String)
+          @tree.children << Element.new(:raw, body, :category => type) if body.kind_of?(String)
         when 'options'
           opts.select do |k,v|
             k = k.to_sym

data/lib/kramdown/parser/kramdown/footnote.rb

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 #
 #--
-# Copyright (C) 2009 Thomas Leitner <t_leitner@gmx.at>
+# Copyright (C) 2009-2010 Thomas Leitner <t_leitner@gmx.at>
 #
 # This file is part of kramdown.
 #

data/lib/kramdown/parser/kramdown/header.rb

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 #
 #--
-# Copyright (C) 2009 Thomas Leitner <t_leitner@gmx.at>
+# Copyright (C) 2009-2010 Thomas Leitner <t_leitner@gmx.at>
 #
 # This file is part of kramdown.
 #

data/lib/kramdown/parser/kramdown/horizontal_rule.rb

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 #
 #--
-# Copyright (C) 2009 Thomas Leitner <t_leitner@gmx.at>
+# Copyright (C) 2009-2010 Thomas Leitner <t_leitner@gmx.at>
 #
 # This file is part of kramdown.
 #

data/lib/kramdown/parser/kramdown/html.rb

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 #
 #--
-# Copyright (C) 2009 Thomas Leitner <t_leitner@gmx.at>
+# Copyright (C) 2009-2010 Thomas Leitner <t_leitner@gmx.at>
 #
 # This file is part of kramdown.
 #
@@ -20,179 +20,102 @@
 #++
 #
 
-require 'rexml/parsers/baseparser'
+require 'kramdown/parser/html'
 
 module Kramdown
   module Parser
     class Kramdown
 
-      #:stopdoc:
-      # The following regexps are based on the ones used by REXML, with some slight modifications.
-      #:startdoc:
-      HTML_COMMENT_RE = /<!--(.*?)-->/m
-      HTML_INSTRUCTION_RE = /<\?(.*?)\?>/m
-      HTML_ATTRIBUTE_RE = /\s*(#{REXML::Parsers::BaseParser::UNAME_STR})\s*=\s*(["'])(.*?)\2/m
-      HTML_TAG_RE = /<((?>#{REXML::Parsers::BaseParser::UNAME_STR}))\s*((?>\s+#{REXML::Parsers::BaseParser::UNAME_STR}\s*=\s*(["']).*?\3)*)\s*(\/)?>/m
-      HTML_TAG_CLOSE_RE = /<\/(#{REXML::Parsers::BaseParser::NAME_STR})\s*>/m
-
-
-      HTML_PARSE_AS_BLOCK = %w{applet button blockquote colgroup dd div dl fieldset form iframe li
-                               map noscript object ol table tbody td th thead tfoot tr ul}
-      HTML_PARSE_AS_SPAN  = %w{a abbr acronym address b bdo big cite caption code del dfn dt em
-                               h1 h2 h3 h4 h5 h6 i ins kbd label legend optgroup p pre q rb rbc
-                               rp rt rtc ruby samp select small span strong sub sup tt var}
-      HTML_PARSE_AS_RAW   = %w{script math option textarea}
-
-      HTML_PARSE_AS = Hash.new {|h,k| h[k] = :raw}
-      HTML_PARSE_AS_BLOCK.each {|i| HTML_PARSE_AS[i] = :block}
-      HTML_PARSE_AS_SPAN.each {|i| HTML_PARSE_AS[i] = :span}
-      HTML_PARSE_AS_RAW.each {|i| HTML_PARSE_AS[i] = :raw}
-
-      #:stopdoc:
-      # Some HTML elements like script belong to both categories (i.e. are valid in block and
-      # span HTML) and don't appear therefore!
-      #:startdoc:
-      HTML_SPAN_ELEMENTS = %w{a abbr acronym b big bdo br button cite code del dfn em i img input
-                              ins kbd label option q rb rbc rp rt rtc ruby samp select small span
-                              strong sub sup textarea tt var}
-      HTML_BLOCK_ELEMENTS = %w{address applet button blockquote caption col colgroup dd div dl dt fieldset
-                               form h1 h2 h3 h4 h5 h6 hr iframe legend li map ol optgroup p pre table tbody
-                               td th thead tfoot tr ul}
-      HTML_ELEMENTS_WITHOUT_BODY = %w{area br col hr img input}
-
-      HTML_BLOCK_START = /^#{OPT_SPACE}<(#{REXML::Parsers::BaseParser::UNAME_STR}|\?|!--|\/)/
-      HTML_RAW_START = /(?=<(#{REXML::Parsers::BaseParser::UNAME_STR}|\/))/
-
-      # Parse the HTML at the current position as block level HTML.
-      def parse_block_html
-        if result = @src.scan(HTML_COMMENT_RE)
-          @tree.children << Element.new(:xml_comment, result, :type => :block)
-          @src.scan(/[ \t]*\n/)
-          true
-        elsif result = @src.scan(HTML_INSTRUCTION_RE)
-          @tree.children << Element.new(:xml_pi, result, :type => :block)
-          @src.scan(/[ \t]*\n/)
-          true
-        else
-          if result = @src.check(/^#{OPT_SPACE}#{HTML_TAG_RE}/) && !HTML_SPAN_ELEMENTS.include?(@src[1])
-            @src.pos += @src.matched_size
-            handle_html_start_tag
-            true
-          elsif result = @src.check(/^#{OPT_SPACE}#{HTML_TAG_CLOSE_RE}/) && !HTML_SPAN_ELEMENTS.include?(@src[1])
-            @src.pos += @src.matched_size
-            name = @src[1]
-
-            if @tree.type ==:html_element && @tree.value == name
-              throw :stop_block_parsing, :found
-            else
-              warning("Found invalidly used HTML closing tag for '#{name}' - ignoring it")
-              true
-            end
-          else
-            false
-          end
-        end
-      end
-      define_parser(:block_html, HTML_BLOCK_START)
-
-
-      # Return the HTML parse type defined by the string +val+, i.e. raw when "0", default parsing
-      # (return value +nil+) when "1", span parsing when "span" and block parsing when "block". If
-      # +val+ is nil, then the default parsing mode is used.
-      def get_parse_type(val)
-        case val
-        when "0" then :raw
-        when "1" then :default
-        when "span" then :span
-        when "block" then :block
-        when NilClass then nil
-        else
-          warning("Invalid markdown attribute val '#{val}', using default")
-          nil
-        end
-      end
-
-      # Process the HTML start tag that has already be scanned/checked.
-      def handle_html_start_tag
-        curpos = @src.pos
-        name = @src[1]
-        closed = !@src[4].nil?
-        attrs = {}
-        @src[2].scan(HTML_ATTRIBUTE_RE).each {|attr,sep,val| attrs[attr] = val}
+      include Kramdown::Parser::Html::Parser
 
+      def handle_kramdown_html_tag(el, closed)
         parse_type = if @tree.type != :html_element || @tree.options[:parse_type] != :raw
-                       (@doc.options[:parse_block_html] ? HTML_PARSE_AS[name] : :raw)
+                       (@doc.options[:parse_block_html] ? HTML_PARSE_AS[el.value] : :raw)
                      else
                        :raw
                      end
-        if val = get_parse_type(attrs.delete('markdown'))
-          parse_type = (val == :default ? HTML_PARSE_AS[name] : val)
+        if val = html_parse_type(el.options[:attr].delete('markdown'))
+          parse_type = (val == :default ? HTML_PARSE_AS[el.value] : val)
         end
 
         @src.scan(/[ \t]*\n/) if parse_type == :block
-
-        el = Element.new(:html_element, name, :attr => attrs, :type => :block, :parse_type => parse_type)
         el.options[:outer_element] = true if @tree.type != :html_element
         el.options[:parent_is_raw] = true if @tree.type == :html_element && @tree.options[:parse_type] == :raw
-        @tree.children << el
+        el.options[:parse_type] = parse_type
 
-        if !closed && HTML_ELEMENTS_WITHOUT_BODY.include?(el.value)
-          warning("The HTML tag '#{el.value}' cannot have any content - auto-closing it")
-        elsif !closed
+        if !closed
           if parse_type == :block
             end_tag_found = parse_blocks(el)
             if !end_tag_found
               warning("Found no end tag for '#{el.value}' - auto-closing it")
             end
           elsif parse_type == :span
+            curpos = @src.pos
             if result = @src.scan_until(/(?=<\/#{el.value}\s*>)/m)
-              add_text(extract_string(curpos...@src.pos), el)
+              add_text(extract_string(curpos...@src.pos, @src), el)
               @src.scan(HTML_TAG_CLOSE_RE)
             else
               add_text(@src.scan(/.*/m), el)
               warning("Found no end tag for '#{el.value}' - auto-closing it")
             end
           else
-            parse_raw_html(el)
+            parse_raw_html(el, &method(:handle_kramdown_html_tag))
           end
           @src.scan(/[ \t]*\n/) unless (@tree.type == :html_element && @tree.options[:parse_type] == :raw)
         end
       end
 
-      # Parse raw HTML until the matching end tag for +el+ is found or until the end of the
-      # document.
-      def parse_raw_html(el)
-        @stack.push(@tree)
-        @tree = el
-
-        done = false
-        endpos = nil
-        while !@src.eos? && !done
-          if result = @src.scan_until(HTML_RAW_START)
-            endpos = @src.pos
-            add_text(result, @tree, :html_text)
-            if @src.scan(HTML_TAG_RE)
-              handle_html_start_tag
-            elsif @src.scan(HTML_TAG_CLOSE_RE)
-              if @tree.value == @src[1]
-                done = true
-              else
-                warning("Found invalidly used HTML closing tag for '#{@src[1]}' - ignoring it")
-              end
+      # Return the HTML parse type defined by the string +val+, i.e. raw when "0", default parsing
+      # (return value +nil+) when "1", span parsing when "span" and block parsing when "block". If
+      # +val+ is nil, then the default parsing mode is used.
+      def html_parse_type(val)
+        case val
+        when "0" then :raw
+        when "1" then :default
+        when "span" then :span
+        when "block" then :block
+        when NilClass then nil
+        else
+          warning("Invalid markdown attribute val '#{val}', using default")
+          nil
+        end
+      end
+
+
+      HTML_BLOCK_START = /^#{OPT_SPACE}<(#{REXML::Parsers::BaseParser::UNAME_STR}|\?|!--|\/)/
+
+      # Parse the HTML at the current position as block level HTML.
+      def parse_block_html
+        if result = @src.scan(HTML_COMMENT_RE)
+          @tree.children << Element.new(:xml_comment, result, :category => :block)
+          @src.scan(/[ \t]*\n/)
+          true
+        elsif result = @src.scan(HTML_INSTRUCTION_RE)
+          @tree.children << Element.new(:xml_pi, result, :category => :block)
+          @src.scan(/[ \t]*\n/)
+          true
+        else
+          if result = @src.check(/^#{OPT_SPACE}#{HTML_TAG_RE}/) && !HTML_SPAN_ELEMENTS.include?(@src[1])
+            @src.pos += @src.matched_size
+            handle_html_start_tag(&method(:handle_kramdown_html_tag))
+            Kramdown::Parser::Html::ElementConverter.new.process(@tree.children.last) if @doc.options[:html_to_native]
+            true
+          elsif result = @src.check(/^#{OPT_SPACE}#{HTML_TAG_CLOSE_RE}/) && !HTML_SPAN_ELEMENTS.include?(@src[1])
+            @src.pos += @src.matched_size
+            name = @src[1]
+
+            if @tree.type == :html_element && @tree.value == name
+              throw :stop_block_parsing, :found
             else
-              add_text(@src.scan(/./), @tree, :html_text)
+              warning("Found invalidly used HTML closing tag for '#{name}' - ignoring it")
+              true
             end
           else
-            result = @src.scan(/.*/m)
-            add_text(result, @tree, :html_text)
-            warning("Found no end tag for '#{@tree.value}' - auto-closing it")
-            done = true
+            false
           end
         end
-
-        @tree = @stack.pop
-        endpos
       end
+      define_parser(:block_html, HTML_BLOCK_START)
 
 
       HTML_SPAN_START = /<(#{REXML::Parsers::BaseParser::UNAME_STR}|\?|!--|\/)/
@@ -200,9 +123,9 @@ module Kramdown
       # Parse the HTML at the current position as span level HTML.
       def parse_span_html
         if result = @src.scan(HTML_COMMENT_RE)
-          @tree.children << Element.new(:xml_comment, result, :type => :span)
+          @tree.children << Element.new(:xml_comment, result, :category => :span)
         elsif result = @src.scan(HTML_INSTRUCTION_RE)
-          @tree.children << Element.new(:xml_pi, result, :type => :span)
+          @tree.children << Element.new(:xml_pi, result, :category => :span)
         elsif result = @src.scan(HTML_TAG_CLOSE_RE)
           warning("Found invalidly used HTML closing tag for '#{@src[1]}' - ignoring it")
         elsif result = @src.scan(HTML_TAG_RE)
@@ -212,36 +135,33 @@ module Kramdown
           attrs = {}
           @src[2].scan(HTML_ATTRIBUTE_RE).each {|name,sep,val| attrs[name] = val.gsub(/\n+/, ' ')}
 
-          do_parsing = (HTML_PARSE_AS_RAW.include?(@src[1]) ? false : @doc.options[:parse_span_html])
-          if val = get_parse_type(attrs.delete('markdown'))
+          do_parsing = (HTML_PARSE_AS_RAW.include?(@src[1]) || @tree.options[:parse_type] == :raw ? false : @doc.options[:parse_span_html])
+          if val = html_parse_type(attrs.delete('markdown'))
             if val == :block
               warning("Cannot use block level parsing in span level HTML tag - using default mode")
             elsif val == :span
               do_parsing = true
             elsif val == :default
-              (HTML_PARSE_AS_RAW.include?(@src[1]) ? false : true)
+              do_parsing = !HTML_PARSE_AS_RAW.include?(@src[1])
             elsif val == :raw
               do_parsing = false
             end
           end
 
-          el = Element.new(:html_element, @src[1], :attr => attrs, :type => :span)
+          el = Element.new(:html_element, @src[1], :attr => attrs, :category => :span, :parse_type => (do_parsing ? :span : :raw))
+          @tree.children << el
           stop_re = /<\/#{Regexp.escape(@src[1])}\s*>/
-          if @src[4]
-            @tree.children << el
-          elsif HTML_ELEMENTS_WITHOUT_BODY.include?(el.value)
+          if !@src[4] && HTML_ELEMENTS_WITHOUT_BODY.include?(el.value)
             warning("The HTML tag '#{el.value}' cannot have any content - auto-closing it")
-            @tree.children << el
-          else
-            if parse_spans(el, stop_re, (do_parsing ? nil : [:span_html]), (do_parsing ? :text : :html_text))
-              end_pos = @src.pos
+          elsif !@src[4]
+            if parse_spans(el, stop_re, (do_parsing ? nil : [:span_html]))
               @src.scan(stop_re)
             else
               warning("Found no end tag for '#{el.value}' - auto-closing it")
-              add_text(@src.scan(/.*/m))
+              add_text(@src.scan(/.*/m), el)
             end
-            @tree.children << el
           end
+          Kramdown::Parser::Html::ElementConverter.new.process(el) if @doc.options[:html_to_native]
         else
           add_text(@src.scan(/./))
         end