Fix XSS vulnerabilities in chunk-handling

lib/chunks/category.rb

@@ -16,7 +16,8 @@ class Category < Chunk::Abstract
 def initialize(match_data, content)
     super(match_data, content)
     @hidden = match_data[1]
-    @list = match_data[2].split(',').map { |c| c.strip }
+    @list = match_data[2].split(',').map { |c| c.to_s.is_utf8? ? html_escape(c. strip) : nil }
+    @list.compact!
     @unmask_text = ''
     if @hidden
       @unmask_text = ''
@@ -28,6 +29,6 @@ def initialize(match_data, content)
 
   # TODO move presentation of page metadata to controller/view
   def url(category)
-    %{<a class="category_link" href="../list/?category=#{category}">#{category}</a>}
+    %{<a class="category_link" href="../list/#{category}">#{category}</a>}
   end
 end

lib/chunks/chunk.rb

@@ -74,6 +74,14 @@ module Chunk
       @content.delete_chunk(self)
     end
 
+    def  html_escape(string)
+      string.gsub( /&/, "&" ).
+             gsub( /</, "&lt;" ).
+             gsub( />/, "&gt;" ).
+             gsub( /'/, "&#39;" ).
+             gsub( /"/, "&quot;" )
+    end
+
   end
 
 end

lib/chunks/nowiki.rb

@@ -13,6 +13,10 @@ require 'chunks/chunk'
 #
 # Author: Mark Reid <mark at threewordslong dot com>
 # Created: 8th June 2004
+
+require 'sanitize'
+include Sanitize
+
 class NoWiki < Chunk::Abstract
 
   NOWIKI_PATTERN = Regexp.new('<nowiki>(.*?)</nowiki>', Regexp::MULTILINE)
@@ -22,7 +26,7 @@ class NoWiki < Chunk::Abstract
 
   def initialize(match_data, content)
     super
-    @plain_text = @unmask_text = match_data[1]
+    @plain_text = @unmask_text = sanitize_html(match_data[1])
   end
 
 end

lib/sanitize.rb

@@ -203,3 +203,28 @@ module Sanitize
           style = clean.join(' ')
       end
 end      
+
+# Some useful additions to the String class
+
+class String
+
+# Check whether a string is valid utf-8
+#
+# :call-seq:
+#    string.is_utf8?    -> boolean
+#
+# returns true if the sequence of bytes in string is valid utf-8
+   def is_utf8?
+     self =~  /^(
+         [\x09\x0A\x0D\x20-\x7E]            # ASCII
+       | [\xC2-\xDF][\x80-\xBF]             # non-overlong 2-byte
+       |  \xE0[\xA0-\xBF][\x80-\xBF]        # excluding overlongs
+       | [\xE1-\xEC\xEE\xEF][\x80-\xBF]{2}  # straight 3-byte
+       |  \xED[\x80-\x9F][\x80-\xBF]        # excluding surrogates
+       |  \xF0[\x90-\xBF][\x80-\xBF]{2}     # planes 1-3
+       | [\xF1-\xF3][\x80-\xBF]{3}          # planes 4-15
+       |  \xF4[\x80-\x8F][\x80-\xBF]{2}     # plane 16
+     )*$/x;
+   end
+
+end