Fix HTML/XML parser to correctly, case sensitively parse XML

The previous version of the parser converted all tags and attribute names
to lowercase. However, this is only valid for HTML since HTML tags are
case insensitive, XML tags and attribute namesare case sensitive.

Fixes #310

lib/kramdown/parser/html.rb

@@ -59,6 +59,12 @@ module Constants
                                figcaption footer form h1 h2 h3 h4 h5 h6 header hgroup hr html head iframe legend menu
                                li map nav ol optgroup p pre section summary table tbody td th thead tfoot tr ul}
         HTML_ELEMENTS_WITHOUT_BODY = %w{area base br col command embed hr img input keygen link meta param source track wbr}
+
+        HTML_ELEMENT = Hash.new(false)
+        (HTML_SPAN_ELEMENTS + HTML_BLOCK_ELEMENTS + HTML_ELEMENTS_WITHOUT_BODY +
+         HTML_CONTENT_MODEL.keys).each do |a|
+          HTML_ELEMENT[a] = true
+        end
       end
 
 
@@ -77,9 +83,10 @@ module Parser
         # element is already closed, ie. contains no body; the third parameter specifies whether the
         # body - and the end tag - need to be handled in case closed=false).
         def handle_html_start_tag(line = nil) # :yields: el, closed, handle_body
-          name = @src[1].downcase
+          name = @src[1]
+          name.downcase! if HTML_ELEMENT[name.downcase]
           closed = !@src[4].nil?
-          attrs = parse_html_attributes(@src[2], line)
+          attrs = parse_html_attributes(@src[2], line, HTML_ELEMENT[name])
 
           el = Element.new(:html_element, name, attrs, :category => :block)
           el.options[:location] = line if line
@@ -99,10 +106,13 @@ def handle_html_start_tag(line = nil) # :yields: el, closed, handle_body
         # Parses the given string for HTML attributes and returns the resulting hash.
         #
         # If the optional +line+ parameter is supplied, it is used in warning messages.
-        def parse_html_attributes(str, line = nil)
+        #
+        # If the optional +in_html_tag+ parameter is set to +false+, attributes are not modified to
+        # contain only lowercase letters.
+        def parse_html_attributes(str, line = nil, in_html_tag = true)
           attrs = Utils::OrderedHash.new
           str.scan(HTML_ATTRIBUTE_RE).each do |attr, sep, val|
-            attr.downcase!
+            attr.downcase! if in_html_tag
             if attrs.has_key?(attr)
               warning("Duplicate HTML attribute '#{attr}' on line #{line || '?'} - overwriting previous one")
             end
@@ -155,10 +165,11 @@ def parse_raw_html(el, &block)
                   handle_html_start_tag(&block) # DEPRECATED: method needs to accept line number in 2.0
                 end
               elsif @src.scan(HTML_TAG_CLOSE_RE)
-                if @tree.value == @src[1].downcase
+                if @tree.value == (HTML_ELEMENT[@tree.value] ? @src[1].downcase : @src[1])
                   done = true
                 else
-                  warning("Found invalidly used HTML closing tag for '#{@src[1].downcase}' on line #{line} - ignoring it")
+                  add_text(@src.matched, @tree, :text)
+                  warning("Found invalidly used HTML closing tag for '#{@src[1]}' on line #{line} - ignoring it")
                 end
               else
                 add_text(@src.getch, @tree, :text)

lib/kramdown/parser/kramdown/html.rb

@@ -113,14 +113,15 @@ def parse_span_html
           warning("Found invalidly used HTML closing tag for '#{@src[1]}' on line #{line}")
           add_text(result)
         elsif result = @src.scan(HTML_TAG_RE)
-          tag_name = @src[1].downcase
+          tag_name = @src[1]
+          tag_name.downcase! if HTML_ELEMENT[tag_name.downcase]
           if HTML_BLOCK_ELEMENTS.include?(tag_name)
             warning("Found block HTML tag '#{tag_name}' in span-level text on line #{line}")
             add_text(result)
             return
           end
 
-          attrs = parse_html_attributes(@src[2], line)
+          attrs = parse_html_attributes(@src[2], line, HTML_ELEMENT[tag_name])
           attrs.each {|name, value| value.gsub!(/\n+/, ' ')}
 
           do_parsing = (HTML_CONTENT_MODEL[tag_name] == :raw || @tree.options[:content_model] == :raw ? false : @options[:parse_span_html])
@@ -139,7 +140,8 @@ def parse_span_html
           el = Element.new(:html_element, tag_name, attrs, :category => :span, :location => line,
                            :content_model => (do_parsing ? :span : :raw), :is_closed => !!@src[4])
           @tree.children << el
-          stop_re = /<\/#{Regexp.escape(tag_name)}\s*>/i
+          stop_re = /<\/#{Regexp.escape(tag_name)}\s*>/
+          stop_re = Regexp.new(stop_re.source, Regexp::IGNORECASE) if HTML_ELEMENT[tag_name]
           if !@src[4] && !HTML_ELEMENTS_WITHOUT_BODY.include?(el.value)
             if parse_spans(el, stop_re, (do_parsing ? nil : [:span_html]))
               @src.scan(stop_re)

test/test_files.rb

@@ -65,6 +65,8 @@ class TestFiles < Minitest::Test
                           'test/testcases/block/15_math/mathjax_preview.html', # bc of mathjax preview
                           'test/testcases/block/15_math/mathjax_preview_simple.html', # bc of mathjax preview
                           'test/testcases/span/05_html/mark_element.html', # bc of tidy
+                          'test/testcases/block/09_html/xml.html', # bc of tidy
+                          'test/testcases/span/05_html/xml.html', # bc of tidy
                          ].compact
     EXCLUDE_HTML_TEXT_FILES = ['test/testcases/block/09_html/parse_as_span.htmlinput',
                                'test/testcases/block/09_html/parse_as_raw.htmlinput',
@@ -169,6 +171,8 @@ def tidy_output(out)
                           'test/testcases/span/math/mathjaxnode.text', # bc of tidy
                           'test/testcases/span/01_link/link_defs_with_ial.text', # bc of attribute ordering
                           'test/testcases/span/05_html/mark_element.text', # bc of tidy
+                          'test/testcases/block/09_html/xml.text', # bc of tidy
+                          'test/testcases/span/05_html/xml.text', # bc of tidy
                          ].compact
     Dir[File.dirname(__FILE__) + '/testcases/**/*.text'].each do |text_file|
       next if EXCLUDE_TEXT_FILES.any? {|f| text_file =~ /#{f}$/}
@@ -219,6 +223,8 @@ def tidy_output(out)
                              'test/testcases/block/15_math/mathjax_preview_simple.html', # bc of mathjax preview
                              'test/testcases/span/01_link/link_defs_with_ial.html', # bc of attribute ordering
                              'test/testcases/span/05_html/mark_element.html', # bc of tidy
+                             'test/testcases/block/09_html/xml.html', # bc of tidy
+                             'test/testcases/span/05_html/xml.html', # bc of tidy
                             ].compact
     Dir[File.dirname(__FILE__) + '/testcases/**/*.{html,html.19}'].each do |html_file|
       next if EXCLUDE_HTML_KD_FILES.any? {|f| html_file =~ /#{f}(\.19)?$/}

test/testcases/block/09_html/not_parsed.html

@@ -7,7 +7,7 @@
 </div></div>
 
 <div>
-
+&lt;/p&gt;
 </div>
 
 <div>

test/testcases/block/09_html/simple.html

@@ -49,10 +49,6 @@
   <p>Another para.</p>
 </div>
 
-<webgen:block name="test" />
-
-<some:url name:spac="hallo">doit</some:url>
-
 <p><em>Test</em></p>
 
 <p><em>Test</em></p>

test/testcases/block/09_html/simple.html.19

@@ -49,10 +49,6 @@ weiter
   <p>Another para.</p>
 </div>
 
-<webgen:block name="test" />
-
-<some:url name:spac="hallo">doit</some:url>
-
 <p><em>Test</em></p>
 
 <p><em>Test</em></p>

test/testcases/block/09_html/simple.text

@@ -40,14 +40,10 @@ hallo
 
 para6
 
-<div><div cLASs="clear"></div>
+<div><DiV cLASs="clear"></dIv>
 Another para.
 </div>
 
-<webgen:block name="test" />
-
-<some:url name:spac='hallo'>doit</some:url>
-
 <em>Test</em>
 
 <p><em>Test</em></p>

test/testcases/block/09_html/xml.html

@@ -0,0 +1,8 @@
+<webgen:block name="test" />
+
+<some:url name:spac="hallo">doit</some:url>
+
+<SoMe:UrL NamE:SpAC="test">doit</SoMe:UrL>
+
+<SoMe>doit&lt;/some&gt;
+</SoMe>

test/testcases/block/09_html/xml.text

@@ -0,0 +1,7 @@
+<webgen:block name="test" />
+
+<some:url name:spac='hallo'>doit</some:url>
+
+<SoMe:UrL NamE:SpAC='test'>doit</SoMe:UrL>
+
+<SoMe>doit</some>

test/testcases/span/05_html/xml.html

@@ -0,0 +1,5 @@
+<p>This <some:url name:spac="hallo">doit</some:url> test</p>
+
+<p>This <SoMe:UrL NamE:SpAC="test">doit</SoMe:UrL> test</p>
+
+<p>This <SoMe>doit&lt;/some&gt; test</SoMe></p>

test/testcases/span/05_html/xml.text

@@ -0,0 +1,5 @@
+This <some:url name:spac='hallo'>doit</some:url> test
+
+This <SoMe:UrL NamE:SpAC='test'>doit</SoMe:UrL> test
+
+This <SoMe>doit</some> test