vendor/rails/actionpack/test/controller/html-scanner/sanitizer_test.rb

   1 require 'abstract_unit'
   2
   3 class SanitizerTest < Test::Unit::TestCase
   4   def setup
   5     @sanitizer = nil # used by assert_sanitizer
   6   end
   7
   8   def test_strip_tags
   9     sanitizer = HTML::FullSanitizer.new
  10     assert_equal("<<<bad html", sanitizer.sanitize("<<<bad html"))
  11     assert_equal("<<", sanitizer.sanitize("<<<bad html>"))
  12     assert_equal("Dont touch me", sanitizer.sanitize("Dont touch me"))
  13     assert_equal("This is a test.", sanitizer.sanitize("<p>This <u>is<u> a <a href='test.html'><strong>test</strong></a>.</p>"))
  14     assert_equal("Weirdos", sanitizer.sanitize("Wei<<a>a onclick='alert(document.cookie);'</a>/>rdos"))
  15     assert_equal("This is a test.", sanitizer.sanitize("This is a test."))
  16     assert_equal(
  17     %{This is a test.\n\n\nIt no longer contains any HTML.\n}, sanitizer.sanitize(
  18     %{<title>This is <b>a <a href="" target="_blank">test</a></b>.</title>\n\n<!-- it has a comment -->\n\n<p>It no <b>longer <strong>contains <em>any <strike>HTML</strike></em>.</strong></b></p>\n}))
  19     assert_equal "This has a  here.", sanitizer.sanitize("This has a <!-- comment --> here.")
  20     assert_equal "This has a  here.", sanitizer.sanitize("This has a <![CDATA[<section>]]> here.")
  21     assert_equal "This has an unclosed ", sanitizer.sanitize("This has an unclosed <![CDATA[<section>]] here...")
  22     [nil, '', '   '].each { |blank| assert_equal blank, sanitizer.sanitize(blank) }
  23   end
  24
  25   def test_strip_links
  26     sanitizer = HTML::LinkSanitizer.new
  27     assert_equal "Dont touch me", sanitizer.sanitize("Dont touch me")
  28     assert_equal "on my mind\nall day long", sanitizer.sanitize("<a href='almost'>on my mind</a>\n<A href='almost'>all day long</A>")
  29     assert_equal "0wn3d", sanitizer.sanitize("<a href='http://www.rubyonrails.com/'><a href='http://www.rubyonrails.com/' onlclick='steal()'>0wn3d</a></a>")
  30     assert_equal "Magic", sanitizer.sanitize("<a href='http://www.rubyonrails.com/'>Mag<a href='http://www.ruby-lang.org/'>ic")
  31     assert_equal "FrrFox", sanitizer.sanitize("<href onlclick='steal()'>FrrFox</a></href>")
  32     assert_equal "My mind\nall <b>day</b> long", sanitizer.sanitize("<a href='almost'>My mind</a>\n<A href='almost'>all <b>day</b> long</A>")
  33     assert_equal "all <b>day</b> long", sanitizer.sanitize("<<a>a href='hello'>all <b>day</b> long<</A>/a>")
  34
  35     assert_equal "<a<a", sanitizer.sanitize("<a<a")
  36   end
  37
  38   def test_sanitize_form
  39     assert_sanitized "<form action=\"/foo/bar\" method=\"post\"><input></form>", ''
  40   end
  41
  42   def test_sanitize_plaintext
  43     raw = "<plaintext><span>foo</span></plaintext>"
  44     assert_sanitized raw, "<span>foo</span>"
  45   end
  46
  47   def test_sanitize_script
  48     assert_sanitized "a b c<script language=\"Javascript\">blah blah blah</script>d e f", "a b cd e f"
  49   end
  50
  51   # fucked
  52   def test_sanitize_js_handlers
  53     raw = %{onthis="do that" <a href="#" onclick="hello" name="foo" onbogus="remove me">hello</a>}
  54     assert_sanitized raw, %{onthis="do that" <a name="foo" href="#">hello</a>}
  55   end
  56
  57   def test_sanitize_javascript_href
  58     raw = %{href="javascript:bang" <a href="javascript:bang" name="hello">foo</a>, <span href="javascript:bang">bar</span>}
  59     assert_sanitized raw, %{href="javascript:bang" <a name="hello">foo</a>, <span>bar</span>}
  60   end
  61
  62   def test_sanitize_image_src
  63     raw = %{src="javascript:bang" <img src="javascript:bang" width="5">foo</img>, <span src="javascript:bang">bar</span>}
  64     assert_sanitized raw, %{src="javascript:bang" <img width="5">foo</img>, <span>bar</span>}
  65   end
  66
  67   HTML::WhiteListSanitizer.allowed_tags.each do |tag_name|
  68     define_method "test_should_allow_#{tag_name}_tag" do
  69       assert_sanitized "start <#{tag_name} title=\"1\" onclick=\"foo\">foo <bad>bar</bad> baz</#{tag_name}> end", %(start <#{tag_name} title="1">foo bar baz</#{tag_name}> end)
  70     end
  71   end
  72
  73   def test_should_allow_anchors
  74     assert_sanitized %(<a href="foo" onclick="bar"><script>baz</script></a>), %(<a href="foo"></a>)
  75   end
  76
  77   # RFC 3986, sec 4.2
  78   def test_allow_colons_in_path_component
  79     assert_sanitized("<a href=\"./this:that\">foo</a>")
  80   end
  81
  82   %w(src width height alt).each do |img_attr|
  83     define_method "test_should_allow_image_#{img_attr}_attribute" do
  84       assert_sanitized %(<img #{img_attr}="foo" onclick="bar" />), %(<img #{img_attr}="foo" />)
  85     end
  86   end
  87
  88   def test_should_handle_non_html
  89     assert_sanitized 'abc'
  90   end
  91
  92   def test_should_handle_blank_text
  93     assert_sanitized nil
  94     assert_sanitized ''
  95   end
  96
  97   def test_should_allow_custom_tags
  98     text = "<u>foo</u>"
  99     sanitizer = HTML::WhiteListSanitizer.new
 100     assert_equal(text, sanitizer.sanitize(text, :tags => %w(u)))
 101   end
 102
 103   def test_should_allow_only_custom_tags
 104     text = "<u>foo</u> with <i>bar</i>"
 105     sanitizer = HTML::WhiteListSanitizer.new
 106     assert_equal("<u>foo</u> with bar", sanitizer.sanitize(text, :tags => %w(u)))
 107   end
 108
 109   def test_should_allow_custom_tags_with_attributes
 110     text = %(<blockquote cite="http://example.com/">foo</blockquote>)
 111     sanitizer = HTML::WhiteListSanitizer.new
 112     assert_equal(text, sanitizer.sanitize(text))
 113   end
 114
 115   def test_should_allow_custom_tags_with_custom_attributes
 116     text = %(<blockquote foo="bar">Lorem ipsum</blockquote>)
 117     sanitizer = HTML::WhiteListSanitizer.new
 118     assert_equal(text, sanitizer.sanitize(text, :attributes => ['foo']))
 119   end
 120
 121   [%w(img src), %w(a href)].each do |(tag, attr)|
 122     define_method "test_should_strip_#{attr}_attribute_in_#{tag}_with_bad_protocols" do
 123       assert_sanitized %(<#{tag} #{attr}="javascript:bang" title="1">boo</#{tag}>), %(<#{tag} title="1">boo</#{tag}>)
 124     end
 125   end
 126
 127   def test_should_flag_bad_protocols
 128     sanitizer = HTML::WhiteListSanitizer.new
 129     %w(about chrome data disk hcp help javascript livescript lynxcgi lynxexec ms-help ms-its mhtml mocha opera res resource shell vbscript view-source vnd.ms.radio wysiwyg).each do |proto|
 130       assert sanitizer.send(:contains_bad_protocols?, 'src', "#{proto}://bad")
 131     end
 132   end
 133
 134   def test_should_accept_good_protocols
 135     sanitizer = HTML::WhiteListSanitizer.new
 136     HTML::WhiteListSanitizer.allowed_protocols.each do |proto|
 137       assert !sanitizer.send(:contains_bad_protocols?, 'src', "#{proto}://good")
 138     end
 139   end
 140
 141   def test_should_reject_hex_codes_in_protocol
 142     assert_sanitized %(<a href="&#37;6A&#37;61&#37;76&#37;61&#37;73&#37;63&#37;72&#37;69&#37;70&#37;74&#37;3A&#37;61&#37;6C&#37;65&#37;72&#37;74&#37;28&#37;22&#37;58&#37;53&#37;53&#37;22&#37;29">1</a>), "<a>1</a>"
 143     assert @sanitizer.send(:contains_bad_protocols?, 'src', "%6A%61%76%61%73%63%72%69%70%74%3A%61%6C%65%72%74%28%22%58%53%53%22%29")
 144   end
 145
 146   def test_should_block_script_tag
 147     assert_sanitized %(<SCRIPT\nSRC=http://ha.ckers.org/xss.js></SCRIPT>), ""
 148   end
 149
 150   [%(<IMG SRC="javascript:alert('XSS');">),
 151    %(<IMG SRC=javascript:alert('XSS')>),
 152    %(<IMG SRC=JaVaScRiPt:alert('XSS')>),
 153    %(<IMG """><SCRIPT>alert("XSS")</SCRIPT>">),
 154    %(<IMG SRC=javascript:alert(&quot;XSS&quot;)>),
 155    %(<IMG SRC=javascript:alert(String.fromCharCode(88,83,83))>),
 156    %(<IMG SRC=&#106;&#97;&#118;&#97;&#115;&#99;&#114;&#105;&#112;&#116;&#58;&#97;&#108;&#101;&#114;&#116;&#40;&#39;&#88;&#83;&#83;&#39;&#41;>),
 157    %(<IMG SRC=&#0000106&#0000097&#0000118&#0000097&#0000115&#0000099&#0000114&#0000105&#0000112&#0000116&#0000058&#0000097&#0000108&#0000101&#0000114&#0000116&#0000040&#0000039&#0000088&#0000083&#0000083&#0000039&#0000041>),
 158    %(<IMG SRC=&#x6A&#x61&#x76&#x61&#x73&#x63&#x72&#x69&#x70&#x74&#x3A&#x61&#x6C&#x65&#x72&#x74&#x28&#x27&#x58&#x53&#x53&#x27&#x29>),
 159    %(<IMG SRC="jav\tascript:alert('XSS');">),
 160    %(<IMG SRC="jav&#x09;ascript:alert('XSS');">),
 161    %(<IMG SRC="jav&#x0A;ascript:alert('XSS');">),
 162    %(<IMG SRC="jav&#x0D;ascript:alert('XSS');">),
 163    %(<IMG SRC=" &#14;  javascript:alert('XSS');">),
 164    %(<IMG SRC=`javascript:alert("RSnake says, 'XSS'")`>)].each_with_index do |img_hack, i|
 165     define_method "test_should_not_fall_for_xss_image_hack_#{i+1}" do
 166       assert_sanitized img_hack, "<img>"
 167     end
 168   end
 169
 170   def test_should_sanitize_tag_broken_up_by_null
 171     assert_sanitized %(<SCR\0IPT>alert(\"XSS\")</SCR\0IPT>), "alert(\"XSS\")"
 172   end
 173
 174   def test_should_sanitize_invalid_script_tag
 175     assert_sanitized %(<SCRIPT/XSS SRC="http://ha.ckers.org/xss.js"></SCRIPT>), ""
 176   end
 177
 178   def test_should_sanitize_script_tag_with_multiple_open_brackets
 179     assert_sanitized %(<<SCRIPT>alert("XSS");//<</SCRIPT>), "&lt;"
 180     assert_sanitized %(<iframe src=http://ha.ckers.org/scriptlet.html\n<a), %(&lt;a)
 181   end
 182
 183   def test_should_sanitize_unclosed_script
 184     assert_sanitized %(<SCRIPT SRC=http://ha.ckers.org/xss.js?<B>), "<b>"
 185   end
 186
 187   def test_should_sanitize_half_open_scripts
 188     assert_sanitized %(<IMG SRC="javascript:alert('XSS')"), "<img>"
 189   end
 190
 191   def test_should_not_fall_for_ridiculous_hack
 192     img_hack = %(<IMG\nSRC\n=\n"\nj\na\nv\na\ns\nc\nr\ni\np\nt\n:\na\nl\ne\nr\nt\n(\n'\nX\nS\nS\n'\n)\n"\n>)
 193     assert_sanitized img_hack, "<img>"
 194   end
 195
 196   # fucked
 197   def test_should_sanitize_attributes
 198     assert_sanitized %(<SPAN title="'><script>alert()</script>">blah</SPAN>), %(<span title="'&gt;&lt;script&gt;alert()&lt;/script&gt;">blah</span>)
 199   end
 200
 201   def test_should_sanitize_illegal_style_properties
 202     raw      = %(display:block; position:absolute; left:0; top:0; width:100%; height:100%; z-index:1; background-color:black; background-image:url(http://www.ragingplatypus.com/i/cam-full.jpg); background-x:center; background-y:center; background-repeat:repeat;)
 203     expected = %(display: block; width: 100%; height: 100%; background-color: black; background-image: ; background-x: center; background-y: center;)
 204     assert_equal expected, sanitize_css(raw)
 205   end
 206
 207   def test_should_sanitize_with_trailing_space
 208     raw = "display:block; "
 209     expected = "display: block;"
 210     assert_equal expected, sanitize_css(raw)
 211   end
 212
 213   def test_should_sanitize_xul_style_attributes
 214     raw = %(-moz-binding:url('http://ha.ckers.org/xssmoz.xml#xss'))
 215     assert_equal '', sanitize_css(raw)
 216   end
 217
 218   def test_should_sanitize_invalid_tag_names
 219     assert_sanitized(%(a b c<script/XSS src="http://ha.ckers.org/xss.js"></script>d e f), "a b cd e f")
 220   end
 221
 222   def test_should_sanitize_non_alpha_and_non_digit_characters_in_tags
 223     assert_sanitized('<a onclick!#$%&()*~+-_.,:;?@[/|\]^`=alert("XSS")>foo</a>', "<a>foo</a>")
 224   end
 225
 226   def test_should_sanitize_invalid_tag_names_in_single_tags
 227     assert_sanitized('<img/src="http://ha.ckers.org/xss.js"/>', "<img />")
 228   end
 229
 230   def test_should_sanitize_img_dynsrc_lowsrc
 231     assert_sanitized(%(<img lowsrc="javascript:alert('XSS')" />), "<img />")
 232   end
 233
 234   def test_should_sanitize_div_background_image_unicode_encoded
 235     raw = %(background-image:\0075\0072\006C\0028'\006a\0061\0076\0061\0073\0063\0072\0069\0070\0074\003a\0061\006c\0065\0072\0074\0028.1027\0058.1053\0053\0027\0029'\0029)
 236     assert_equal '', sanitize_css(raw)
 237   end
 238
 239   def test_should_sanitize_div_style_expression
 240     raw = %(width: expression(alert('XSS'));)
 241     assert_equal '', sanitize_css(raw)
 242   end
 243
 244   def test_should_sanitize_img_vbscript
 245     assert_sanitized %(<img src='vbscript:msgbox("XSS")' />), '<img />'
 246   end
 247
 248   def test_should_sanitize_cdata_section
 249     assert_sanitized "<![CDATA[<span>section</span>]]>", "&lt;![CDATA[&lt;span>section&lt;/span>]]>"
 250   end
 251
 252   def test_should_sanitize_unterminated_cdata_section
 253     assert_sanitized "<![CDATA[<span>neverending...", "&lt;![CDATA[&lt;span>neverending...]]>"
 254   end
 255
 256   def test_should_not_mangle_urls_with_ampersand
 257      assert_sanitized %{<a href=\"http://www.domain.com?var1=1&amp;var2=2\">my link</a>}
 258   end
 259
 260 protected
 261   def assert_sanitized(input, expected = nil)
 262     @sanitizer ||= HTML::WhiteListSanitizer.new
 263     if input
 264       assert_dom_equal expected || input, @sanitizer.sanitize(input)
 265     else
 266       assert_nil @sanitizer.sanitize(input)
 267     end
 268   end
 269
 270   def sanitize_css(input)
 271     (@sanitizer ||= HTML::WhiteListSanitizer.new).sanitize_css(input)
 272   end
 273 end