#!/usr/bin/env ruby require File.join(File.dirname(__FILE__), 'preamble') require 'html5lib/html5parser' require 'html5lib/liberalxmlparser' require 'html5lib/treewalkers' require 'html5lib/serializer' require 'html5lib/sanitizer' class SanitizeTest < Test::Unit::TestCase include HTML5lib def sanitize_xhtml stream XHTMLParser.parseFragment(stream, {:tokenizer => HTMLSanitizer, :encoding => 'utf-8'}).to_s end def sanitize_html stream HTMLParser.parseFragment(stream, {:tokenizer => HTMLSanitizer, :encoding => 'utf-8'}).to_s end def sanitize_rexml stream require 'rexml/document' doc = REXML::Document.new("
foo
foo <bad>bar</bad> baz
" htmloutput = "foo <bad>bar</bad> baz
" check_sanitization(input, htmloutput, output, output) end end HTMLSanitizer::ALLOWED_ATTRIBUTES.each do |attribute_name| define_method "test_should_forbid_#{attribute_name.upcase}_attribute" do input = "foo
foo <bad>bar</bad> baz
" check_sanitization(input, output, output, output) end end HTMLSanitizer::ALLOWED_PROTOCOLS.each do |protocol| define_method "test_should_allow_#{protocol}_uris" do input = %(foo) output = "foo" check_sanitization(input, output, output, output) end end HTMLSanitizer::ALLOWED_PROTOCOLS.each do |protocol| define_method "test_should_allow_uppercase_#{protocol}_uris" do input = %(foo) output = "foo" check_sanitization(input, output, output, output) end end def test_should_allow_anchors input = "" output = "<script>baz</script>" check_sanitization(input, output, output, output) end # RFC 3986, sec 4.2 def test_allow_colons_in_path_component input = "foo" output = "foo" check_sanitization(input, output, output, output) end %w(src width height alt).each do |img_attr| define_method "test_should_allow_image_#{img_attr}_attribute" do input = "" output = "" rexmloutput = "" check_sanitization(input, output, output, rexmloutput) end end def test_should_handle_non_html input = 'abc' output = 'abc' check_sanitization(input, output, output, output) end def test_should_handle_blank_text input = '' output = '' check_sanitization(input, output, output, output) end [%w(img src), %w(a href)].each do |(tag, attr)| close = VOID_ELEMENTS.include?(tag) ? "/>boo" : ">boo#{tag}>" xclose = VOID_ELEMENTS.include?(tag) ? " />" : ">boo#{tag}>" input = %(<#{tag} #{attr}="javascript:XSS" title="1">boo#{tag}>) output = %(<#{tag} title='1'#{close}) rexmloutput = %(<#{tag} title='1'#{xclose}) define_method "test_should_strip_#{attr}_attribute_in_#{tag}_with_bad_protocols" do check_sanitization(input, output, output, rexmloutput) end define_method "test_should_strip_#{attr}_attribute_in_#{tag}_with_bad_protocols_and_whitespace" do input = %(<#{tag} #{attr}=" javascript:XSS" title="1">boo#{tag}>) output = %(<#{tag} title='1'#{close}) rexmloutput = %(<#{tag} title='1'#{xclose}) check_sanitization(input, output, output, rexmloutput) end end [%(), %(), %(), %(), %(), %(), %(), %(), %(), %(), %(), %(), %(), %(), %()].each_with_index do |img_hack, i| define_method "test_should_not_fall_for_xss_image_hack_#{i}" do output = "" rexmloutput = "" rexmloutput = "Ill-formed XHTML!" if i == 1 check_sanitization(img_hack, output, output, rexmloutput) end end def test_should_sanitize_tag_broken_up_by_null input = %(