html-sanitizer 0.0.1

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: f2aab4300a74e8223593489a108c42ccce6fc4a9
+  data.tar.gz: cc97dc16365fec2e38ab4ed4ac7cf8c250572236
+SHA512:
+  metadata.gz: a1918cd81328e7051456b8e3f73cded0195fb78d3a76ccbb3cc7f8fcc8bb0af95e307699440830db5fb08363f2b5a3c463532c3528616c6e6478ee2c0905a6ff
+  data.tar.gz: 2df201d4cddd0df59b2d7240c818debf3d5d5b3efc93459e8648d0af54211a224b0e7f999e55caf787572d9c4ec0ad30533dbf5a02bac90cadd4c0b711a73a72

data/LICENSE.txt

@@ -0,0 +1,22 @@
+Copyright (c) 2013 Rafael Mendonça França, Kasper Timm Hansen
+
+MIT License
+
+Permission is hereby granted, free of charge, to any person obtaining
+a copy of this software and associated documentation files (the
+"Software"), to deal in the Software without restriction, including
+without limitation the rights to use, copy, modify, merge, publish,
+distribute, sublicense, and/or sell copies of the Software, and to
+permit persons to whom the Software is furnished to do so, subject to
+the following conditions:
+
+The above copyright notice and this permission notice shall be
+included in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

data/README.md

@@ -0,0 +1,124 @@
+This gem is a fork from [Rails Html Sanitizers](https://github.com/rails/rails-html-sanitizer) without rails dependency.
+
+## Installation
+
+Add this line to your application's Gemfile:
+
+    gem 'html-sanitizer', github: 'dpisarewski/html-sanitizer'
+
+And then execute:
+
+    $ bundle
+
+## Usage
+
+### Sanitizers
+
+All sanitizers respond to `sanitize`.
+
+#### FullSanitizer
+
+```ruby
+full_sanitizer = Html::FullSanitizer.new
+full_sanitizer.sanitize("<b>Bold</b> no more!  <a href='more.html'>See more here</a>...")
+# => Bold no more!  See more here...
+```
+
+#### LinkSanitizer
+
+```ruby
+link_sanitizer = Html::LinkSanitizer.new
+link_sanitizer.sanitize('<a href="example.com">Only the link text will be kept.</a>')
+# => Only the link text will be kept.
+```
+
+#### WhiteListSanitizer
+
+```ruby
+white_list_sanitizer = Html::WhiteListSanitizer.new
+
+# sanitize via an extensive white list of allowed elements
+white_list_sanitizer.sanitize(@article.body)
+
+# white list only the supplied tags and attributes
+white_list_sanitizer.sanitize(@article.body, tags: %w(table tr td), attributes: %w(id class style))
+
+# white list via a custom scrubber
+white_list_sanitizer.sanitize(@article.body, scrubber: ArticleScrubber.new)
+
+# white list sanitizer can also sanitize css
+white_list_sanitizer.sanitize_css('background-color: #000;')
+```
+
+### Scrubbers
+
+Scrubbers are objects responsible for removing nodes or attributes you don't want in your HTML document.
+
+This gem includes two scrubbers `Html::PermitScrubber` and `Html::TargetScrubber`.
+
+#### `Html::PermitScrubber`
+
+This scrubber allows you to permit only the tags and attributes you want.
+
+```ruby
+scrubber = Html::PermitScrubber.new
+scrubber.tags = ['a']
+
+html_fragment = Loofah.fragment('<a><img/ ></a>')
+html_fragment.scrub!(scrubber)
+html_fragment.to_s # => "<a></a>"
+```
+
+#### `Html::TargetScrubber`
+
+Where `PermitScrubber` picks out tags and attributes to permit in sanitization,
+`Html::TargetScrubber` targets them for removal.
+
+
+```ruby
+scrubber = Html::TargetScrubber.new
+scrubber.tags = ['img']
+
+html_fragment = Loofah.fragment('<a><img/ ></a>')
+html_fragment.scrub!(scrubber)
+html_fragment.to_s # => "<a></a>"
+```
+
+#### Custom Scrubbers
+
+You can also create custom scrubbers in your application if you want to.
+
+```ruby
+class CommentScrubber < Html::PermitScrubber
+  def allowed_node?(node)
+    !%w(form script comment blockquote).include?(node.name)
+  end
+
+  def skip_node?(node)
+    node.text?
+  end
+
+  def scrub_attribute?(name)
+    name == "style"
+  end
+end
+```
+
+See `Html::PermitScrubber` documentation to learn more about which methods can be overridden.
+
+## Read more
+
+Loofah is what underlies the sanitizers and scrubbers of rails-html-sanitizer.
+- [Loofah and Loofah Scrubbers](https://github.com/flavorjones/loofah)
+
+The `node` argument passed to some methods in a custom scrubber is an instance of `Nokogiri::XML::Node`.
+- [`Nokogiri::XML::Node`](http://nokogiri.org/Nokogiri/XML/Node.html)
+- [Nokogiri](http://nokogiri.org)
+
+## Contributing
+
+1. Fork it
+2. Create your feature branch (`git checkout -b my-new-feature`)
+3. Commit your changes (`git commit -am 'Add some feature'`)
+4. Push to the branch (`git push origin my-new-feature`)
+5. Create new Pull Request

data/lib/html/sanitizer.rb

@@ -0,0 +1,22 @@
+require "html/sanitizer/version"
+require "loofah"
+require "html/sanitizer/scrubbers"
+require "html/sanitizer/sanitizer"
+
+module Html
+  class Sanitizer
+    class << self
+      def full_sanitizer
+        Html::FullSanitizer
+      end
+
+      def link_sanitizer
+        Html::LinkSanitizer
+      end
+
+      def white_list_sanitizer
+        Html::WhiteListSanitizer
+      end
+    end
+  end
+end

data/lib/html/sanitizer/sanitizer.rb

@@ -0,0 +1,139 @@
+module Html
+  XPATHS_TO_REMOVE = %w{.//script .//form comment()}
+
+  class Sanitizer # :nodoc:
+    def sanitize(html, options = {})
+      raise NotImplementedError, "subclasses must implement sanitize method."
+    end
+
+    private
+
+    def remove_xpaths(node, xpaths)
+      node.xpath(*xpaths).remove
+      node
+    end
+  end
+
+  # === Html::FullSanitizer
+  # Removes all tags but strips out scripts, forms and comments.
+  #
+  # full_sanitizer = Html::FullSanitizer.new
+  # full_sanitizer.sanitize("<b>Bold</b> no more!  <a href='more.html'>See more here</a>...")
+  # # => Bold no more!  See more here...
+  class FullSanitizer < Sanitizer
+    def sanitize(html, options = {})
+      return unless html
+      return html if html.empty?
+
+      Loofah.fragment(html).tap do |fragment|
+        remove_xpaths(fragment, XPATHS_TO_REMOVE)
+      end.text
+    end
+  end
+
+  # === Html::LinkSanitizer
+  # Removes a tags and href attributes leaving only the link text
+  #
+  # link_sanitizer = Html::LinkSanitizer.new
+  # link_sanitizer.sanitize('<a href="example.com">Only the link text will be kept.</a>')
+  # # => Only the link text will be kept.
+  class LinkSanitizer < Sanitizer
+    def initialize
+      @link_scrubber = TargetScrubber.new
+      @link_scrubber.tags = %w(a href)
+      @link_scrubber.attributes = %w(href)
+    end
+
+    def sanitize(html, options = {})
+      Loofah.scrub_fragment(html, @link_scrubber).to_s
+    end
+  end
+
+  # === Html::WhiteListSanitizer
+  # Sanitizes html and css from an extensive white list (see link further down).
+  #
+  # === Whitespace
+  # We can't make any guarentees about whitespace being kept or stripped.
+  # Loofah uses Nokogiri, which wraps either a C or Java parser for the
+  # respective Ruby implementation.
+  # Those two parsers determine how whitespace is ultimately handled.
+  #
+  # When the stripped markup will be rendered the users browser won't take
+  # whitespace into account anyway. It might be better to suggest your users
+  # wrap their whitespace sensitive content in pre tags or that you do
+  # so automatically.
+  #
+  # === Options
+  # Sanitizes both html and css via the white lists found here:
+  # https://github.com/flavorjones/loofah/blob/master/lib/loofah/html5/whitelist.rb
+  #
+  # WhiteListSanitizer also accepts options to configure
+  # the white list used when sanitizing html.
+  # There's a class level option:
+  # Html::WhiteListSanitizer.allowed_tags = %w(table tr td)
+  # Html::WhiteListSanitizer.allowed_attributes = %w(id class style)
+  #
+  # Tags and attributes can also be passed to +sanitize+.
+  # Passed options take precedence over the class level options.
+  #
+  # === Examples
+  # white_list_sanitizer = Html::WhiteListSanitizer.new
+  #
+  # Sanitize css doesn't take options
+  # white_list_sanitizer.sanitize_css('background-color: #000;')
+  #
+  # Default: sanitize via a extensive white list of allowed elements
+  # white_list_sanitizer.sanitize(@article.body)
+  #
+  # White list via the supplied tags and attributes
+  # white_list_sanitizer.sanitize(@article.body, tags: %w(table tr td),
+  # attributes: %w(id class style))
+  #
+  # White list via a custom scrubber
+  # white_list_sanitizer.sanitize(@article.body, scrubber: ArticleScrubber.new)
+  class WhiteListSanitizer < Sanitizer
+    class << self
+      attr_accessor :allowed_tags
+      attr_accessor :allowed_attributes
+    end
+
+    def initialize
+      @permit_scrubber = PermitScrubber.new
+    end
+
+    def sanitize(html, options = {})
+      return unless html
+      return html if html.empty?
+
+      loofah_fragment = Loofah.fragment(html)
+
+      if scrubber = options[:scrubber]
+        # No duck typing, Loofah ensures subclass of Loofah::Scrubber
+        loofah_fragment.scrub!(scrubber)
+      elsif allowed_tags(options) || allowed_attributes(options)
+        @permit_scrubber.tags = allowed_tags(options)
+        @permit_scrubber.attributes = allowed_attributes(options)
+        loofah_fragment.scrub!(@permit_scrubber)
+      else
+        remove_xpaths(loofah_fragment, XPATHS_TO_REMOVE)
+        loofah_fragment.scrub!(:strip)
+      end
+
+      loofah_fragment.to_s
+    end
+
+    def sanitize_css(style_string)
+      Loofah::HTML5::Scrub.scrub_css(style_string)
+    end
+
+    private
+
+    def allowed_tags(options)
+      options[:tags] || self.class.allowed_tags
+    end
+
+    def allowed_attributes(options)
+      options[:attributes] || self.class.allowed_attributes
+    end
+  end
+end

data/lib/html/sanitizer/scrubbers.rb

@@ -0,0 +1,146 @@
+module Html
+  # === Html::PermitScrubber
+  #
+  # Html::PermitScrubber allows you to permit only your own tags and/or attributes.
+  #
+  # Html::PermitScrubber can be subclassed to determine:
+  # - When a node should be skipped via +skip_node?+.
+  # - When a node is allowed via +allowed_node?+.
+  # - When an attribute should be scrubbed via +scrub_attribute?+.
+  #
+  # Subclasses don't need to worry if tags or attributes are set or not.
+  # If tags or attributes are not set, Loofah's behavior will be used.
+  # If you override +allowed_node?+ and no tags are set, it will not be called.
+  # Instead Loofahs behavior will be used.
+  # Likewise for +scrub_attribute?+ and attributes respectively.
+  #
+  # Text and CDATA nodes are skipped by default.
+  # Unallowed elements will be stripped, i.e. element is removed but its subtree kept.
+  # Supplied tags and attributes should be Enumerables.
+  #
+  # +tags=+
+  # If set, elements excluded will be stripped.
+  # If not, elements are stripped based on Loofahs +HTML5::Scrub.allowed_element?+.
+  #
+  # +attributes=+
+  # If set, attributes excluded will be removed.
+  # If not, attributes are removed based on Loofahs +HTML5::Scrub.scrub_attributes+.
+  #
+  # class CommentScrubber < Html::PermitScrubber
+  #   def allowed_node?(node)
+  #     !%w(form script comment blockquote).include?(node.name)
+  #   end
+  #
+  #   def skip_node?(node)
+  #     node.text?
+  #   end
+  #
+  #   def scrub_attribute?(name)
+  #     name == "style"
+  #   end
+  # end
+  #
+  # See the documentation for Nokogiri::XML::Node to understand what's possible
+  # with nodes: http://nokogiri.org/Nokogiri/XML/Node.html
+  class PermitScrubber < Loofah::Scrubber
+    attr_reader :tags, :attributes
+
+    def initialize
+      @direction = :bottom_up
+      @tags, @attributes = nil, nil
+    end
+
+    def tags=(tags)
+      @tags = validate!(tags, :tags)
+    end
+
+    def attributes=(attributes)
+      @attributes = validate!(attributes, :attributes)
+    end
+
+    def scrub(node)
+      return CONTINUE if skip_node?(node)
+
+      unless keep_node?(node)
+        return STOP if scrub_node(node) == STOP
+      end
+
+      scrub_attributes(node)
+    end
+
+    protected
+
+    def allowed_node?(node)
+      @tags.include?(node.name)
+    end
+
+    def skip_node?(node)
+      node.text? || node.cdata?
+    end
+
+    def scrub_attribute?(name)
+      !@attributes.include?(name)
+    end
+
+    def keep_node?(node)
+      if @tags
+        allowed_node?(node)
+      else
+        Loofah::HTML5::Scrub.allowed_element?(node.name)
+      end
+    end
+
+    def scrub_node(node)
+      node.before(node.children) # strip
+      node.remove
+    end
+
+    def scrub_attributes(node)
+      if @attributes
+        node.attribute_nodes.each do |attr|
+          attr.remove if scrub_attribute?(attr.name)
+        end
+
+        scrub_css_attribute(node)
+      else
+        Loofah::HTML5::Scrub.scrub_attributes(node)
+      end
+    end
+
+    def scrub_css_attribute(node)
+      if Loofah::HTML5::Scrub.respond_to?(:scrub_css_attribute)
+        Loofah::HTML5::Scrub.scrub_css_attribute(node)
+      else
+        style = node.attributes['style']
+        style.value = Loofah::HTML5::Scrub.scrub_css(style.value) if style
+      end
+    end
+
+    def validate!(var, name)
+      if var && !var.is_a?(Enumerable)
+        raise ArgumentError, "You should pass :#{name} as an Enumerable"
+      end
+      var
+    end
+  end
+
+  # === Html::TargetScrubber
+  #
+  # Where Html::PermitScrubber picks out tags and attributes to permit in
+  # sanitization, Html::TargetScrubber targets them for removal.
+  #
+  # +tags=+
+  # If set, elements included will be stripped.
+  #
+  # +attributes=+
+  # If set, attributes included will be removed.
+  class TargetScrubber < PermitScrubber
+    def allowed_node?(node)
+      !@tags.include?(node.name)
+    end
+
+    def scrub_attribute?(name)
+      @attributes.include?(name)
+    end
+  end
+end

data/lib/html/sanitizer/version.rb

@@ -0,0 +1,5 @@
+module Html
+  class Sanitizer
+    VERSION = "0.0.1"
+  end
+end

data/test/sanitizer_test.rb

@@ -0,0 +1,489 @@
+require "minitest/autorun"
+require "sanitizer"
+require "rails/dom/testing/assertions/dom_assertions"
+
+class SanitizersTest < Minitest::Test
+  include Rails::Dom::Testing::Assertions::DomAssertions
+
+  def test_sanitizer_sanitize_raises_not_implemented_error
+    assert_raises NotImplementedError do
+      Html::Sanitizer.new.sanitize('')
+    end
+  end
+
+  class XpathRemovalTestSanitizer < Html::Sanitizer
+    def sanitize(html, options = {})
+      fragment = Loofah.fragment(html)
+      remove_xpaths(fragment, options[:xpaths]).to_s
+    end
+  end
+
+  def test_remove_xpaths_removes_an_xpath
+    html = %(<h1>hello <script>code!</script></h1>)
+    assert_equal %(<h1>hello </h1>), xpath_sanitize(html, xpaths: %w(.//script))
+  end
+
+  def test_remove_xpaths_removes_all_occurences_of_xpath
+    html = %(<section><header><script>code!</script></header><p>hello <script>code!</script></p></section>)
+    assert_equal %(<section><header></header><p>hello </p></section>), xpath_sanitize(html, xpaths: %w(.//script))
+  end
+
+  def test_remove_xpaths_called_with_faulty_xpath
+    assert_raises Nokogiri::XML::XPath::SyntaxError do
+      xpath_sanitize('<h1>hello<h1>', xpaths: %w(..faulty_xpath))
+    end
+  end
+
+  def test_remove_xpaths_called_with_xpath_string
+    assert_equal '', xpath_sanitize('<a></a>', xpaths: './/a')
+  end
+
+  def test_remove_xpaths_called_with_enumerable_xpaths
+    assert_equal '', xpath_sanitize('<a><span></span></a>', xpaths: %w(.//a .//span))
+  end
+
+  def test_strip_tags_with_quote
+    input = '<" <img src="trollface.gif" onload="alert(1)"> hi'
+    assert_equal ' hi', full_sanitize(input)
+  end
+
+  def test_strip_invalid_html
+    assert_equal "", full_sanitize("<<<bad html")
+  end
+
+  def test_strip_nested_tags
+    expected = "Weia onclick='alert(document.cookie);'/&gt;rdos"
+    input = "Wei<<a>a onclick='alert(document.cookie);'</a>/>rdos"
+    assert_equal expected, full_sanitize(input)
+  end
+
+  def test_strip_tags_multiline
+    expected = %{This is a test.\n\n\n\nIt no longer contains any HTML.\n}
+    input = %{<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}
+
+    assert_equal expected, full_sanitize(input)
+  end
+
+  def test_strip_comments
+    assert_equal "This is ", full_sanitize("This is <-- not\n a comment here.")
+  end
+
+  def test_strip_cdata
+    assert_equal "This has a ]]&gt; here.", full_sanitize("This has a <![CDATA[<section>]]> here.")
+  end
+
+  def test_strip_unclosed_cdata
+    assert_equal "This has an unclosed ]] here...", full_sanitize("This has an unclosed <![CDATA[<section>]] here...")
+  end
+
+  def test_strip_blank_string
+    [nil, '', '   '].each { |blank| assert_equal blank, full_sanitize(blank) }
+  end
+
+  def test_strip_tags_with_plaintext
+    assert_equal "Dont touch me", full_sanitize("Dont touch me")
+  end
+
+  def test_strip_tags_with_tags
+    assert_equal "This is a test.", full_sanitize("<p>This <u>is<u> a <a href='test.html'><strong>test</strong></a>.</p>")
+  end
+
+  def test_strip_tags_with_many_open_quotes
+    assert_equal "", full_sanitize("<<<bad html>")
+  end
+
+  def test_strip_tags_with_sentence
+    assert_equal "This is a test.", full_sanitize("This is a test.")
+  end
+
+  def test_strip_tags_with_comment
+    assert_equal "This has a  here.", full_sanitize("This has a <!-- comment --> here.")
+  end
+
+  def test_strip_tags_with_frozen_string
+    assert_equal "Frozen string with no tags", full_sanitize("Frozen string with no tags".freeze)
+  end
+
+  def test_strip_links_with_tags_in_tags
+    expected = "a href='hello'&gt;all <b>day</b> long/a&gt;"
+    input = "<<a>a href='hello'>all <b>day</b> long<</A>/a>"
+    assert_equal expected, link_sanitize(input)
+  end
+
+  def test_strip_links_with_unclosed_tags
+    assert_equal "", link_sanitize("<a<a")
+  end
+
+  def test_strip_links_with_plaintext
+    assert_equal "Dont touch me", link_sanitize("Dont touch me")
+  end
+
+  def test_strip_links_with_line_feed_and_uppercase_tag
+    assert_equal "on my mind\nall day long", link_sanitize("<a href='almost'>on my mind</a>\n<A href='almost'>all day long</A>")
+  end
+
+  def test_strip_links_leaves_nonlink_tags
+    assert_equal "My mind\nall <b>day</b> long", link_sanitize("<a href='almost'>My mind</a>\n<A href='almost'>all <b>day</b> long</A>")
+  end
+
+  def test_strip_links_with_links
+    assert_equal "0wn3d", link_sanitize("<a href='http://www.rubyonrails.com/'><a href='http://www.rubyonrails.com/' onlclick='steal()'>0wn3d</a></a>")
+  end
+
+  def test_strip_links_with_linkception
+    assert_equal "Magic", link_sanitize("<a href='http://www.rubyonrails.com/'>Mag<a href='http://www.ruby-lang.org/'>ic")
+  end
+
+  def test_strip_links_with_a_tag_in_href
+    assert_equal "FrrFox", link_sanitize("<href onlclick='steal()'>FrrFox</a></href>")
+  end
+
+  def test_sanitize_form
+    assert_sanitized "<form action=\"/foo/bar\" method=\"post\"><input></form>", ''
+  end
+
+  def test_sanitize_plaintext
+    assert_sanitized "<plaintext><span>foo</span></plaintext>", "<span>foo</span>"
+  end
+
+  def test_sanitize_script
+    assert_sanitized "a b c<script language=\"Javascript\">blah blah blah</script>d e f", "a b cd e f"
+  end
+
+  def test_sanitize_js_handlers
+    raw = %{onthis="do that" <a href="#" onclick="hello" name="foo" onbogus="remove me">hello</a>}
+    assert_sanitized raw, %{onthis="do that" <a href="#" name="foo">hello</a>}
+  end
+
+  def test_sanitize_javascript_href
+    raw = %{href="javascript:bang" <a href="javascript:bang" name="hello">foo</a>, <span href="javascript:bang">bar</span>}
+    assert_sanitized raw, %{href="javascript:bang" <a name="hello">foo</a>, <span>bar</span>}
+  end
+
+  def test_sanitize_image_src
+    raw = %{src="javascript:bang" <img src="javascript:bang" width="5">foo</img>, <span src="javascript:bang">bar</span>}
+    assert_sanitized raw, %{src="javascript:bang" <img width="5">foo</img>, <span>bar</span>}
+  end
+
+  tags = Loofah::HTML5::WhiteList::ALLOWED_ELEMENTS - %w(script form)
+  tags.each do |tag_name|
+    define_method "test_should_allow_#{tag_name}_tag" do
+      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)
+    end
+  end
+
+  def test_should_allow_anchors
+    assert_sanitized %(<a href="foo" onclick="bar"><script>baz</script></a>), %(<a href=\"foo\"></a>)
+  end
+
+  def test_video_poster_sanitization
+    assert_sanitized %(<video src="videofile.ogg" autoplay  poster="posterimage.jpg"></video>), %(<video src="videofile.ogg" poster="posterimage.jpg"></video>)
+    assert_sanitized %(<video src="videofile.ogg" poster=javascript:alert(1)></video>), %(<video src="videofile.ogg"></video>)
+  end
+
+  # RFC 3986, sec 4.2
+  def test_allow_colons_in_path_component
+    assert_sanitized "<a href=\"./this:that\">foo</a>"
+  end
+
+  %w(src width height alt).each do |img_attr|
+    define_method "test_should_allow_image_#{img_attr}_attribute" do
+      assert_sanitized %(<img #{img_attr}="foo" onclick="bar" />), %(<img #{img_attr}="foo" />)
+    end
+  end
+
+  def test_should_handle_non_html
+    assert_sanitized 'abc'
+  end
+
+  def test_should_handle_blank_text
+    [nil, '', '   '].each { |blank| assert_sanitized blank }
+  end
+
+  def test_setting_allowed_tags_affects_sanitization
+    scope_allowed_tags %w(u) do |sanitizer|
+      assert_equal '<u></u>', sanitizer.sanitize('<a><u></u></a>')
+    end
+  end
+
+  def test_setting_allowed_attributes_affects_sanitization
+    scope_allowed_attributes %w(foo) do |sanitizer|
+      input = '<a foo="hello" bar="world"></a>'
+      assert_equal '<a foo="hello"></a>', sanitizer.sanitize(input)
+    end
+  end
+
+  def test_custom_tags_overrides_allowed_tags
+    scope_allowed_tags %(u) do |sanitizer|
+      input = '<a><u></u></a>'
+      assert_equal '<a></a>', sanitizer.sanitize(input, tags: %w(a))
+    end
+  end
+
+  def test_custom_attributes_overrides_allowed_attributes
+    scope_allowed_attributes %(foo) do |sanitizer|
+      input = '<a foo="hello" bar="world"></a>'
+      assert_equal '<a bar="world"></a>', sanitizer.sanitize(input, attributes: %w(bar))
+    end
+  end
+
+  def test_should_allow_custom_tags
+    text = "<u>foo</u>"
+    assert_equal text, white_list_sanitize(text, tags: %w(u))
+  end
+
+  def test_should_allow_only_custom_tags
+    text = "<u>foo</u> with <i>bar</i>"
+    assert_equal "<u>foo</u> with bar", white_list_sanitize(text, tags: %w(u))
+  end
+
+  def test_should_allow_custom_tags_with_attributes
+    text = %(<blockquote cite="http://example.com/">foo</blockquote>)
+    assert_equal text, white_list_sanitize(text)
+  end
+
+  def test_should_allow_custom_tags_with_custom_attributes
+    text = %(<blockquote foo="bar">Lorem ipsum</blockquote>)
+    assert_equal text, white_list_sanitize(text, attributes: ['foo'])
+  end
+
+  def test_scrub_style_if_style_attribute_option_is_passed
+    input = '<p style="color: #000; background-image: url(http://www.ragingplatypus.com/i/cam-full.jpg);"></p>'
+    assert_equal '<p style="color: #000;"></p>', white_list_sanitize(input, attributes: %w(style))
+  end
+
+  def test_should_raise_argument_error_if_tags_is_not_enumerable
+    assert_raises ArgumentError do
+      white_list_sanitize('<a>some html</a>', tags: 'foo')
+    end
+  end
+
+  def test_should_raise_argument_error_if_attributes_is_not_enumerable
+    assert_raises ArgumentError do
+      white_list_sanitize('<a>some html</a>', attributes: 'foo')
+    end
+  end
+
+  def test_should_not_accept_non_loofah_inheriting_scrubber
+    scrubber = Object.new
+    def scrubber.scrub(node); node.name = 'h1'; end
+
+    assert_raises Loofah::ScrubberNotFound do
+      white_list_sanitize('<a>some html</a>', scrubber: scrubber)
+    end
+  end
+
+  def test_should_accept_loofah_inheriting_scrubber
+    scrubber = Loofah::Scrubber.new
+    def scrubber.scrub(node); node.name = 'h1'; end
+
+    html = "<script>hello!</script>"
+    assert_equal "<h1>hello!</h1>", white_list_sanitize(html, scrubber: scrubber)
+  end
+
+  def test_should_accept_loofah_scrubber_that_wraps_a_block
+    scrubber = Loofah::Scrubber.new { |node| node.name = 'h1' }
+    html = "<script>hello!</script>"
+    assert_equal "<h1>hello!</h1>", white_list_sanitize(html, scrubber: scrubber)
+  end
+
+  def test_custom_scrubber_takes_precedence_over_other_options
+    scrubber = Loofah::Scrubber.new { |node| node.name = 'h1' }
+    html = "<script>hello!</script>"
+    assert_equal "<h1>hello!</h1>", white_list_sanitize(html, scrubber: scrubber, tags: ['foo'])
+  end
+
+  [%w(img src), %w(a href)].each do |(tag, attr)|
+    define_method "test_should_strip_#{attr}_attribute_in_#{tag}_with_bad_protocols" do
+      assert_sanitized %(<#{tag} #{attr}="javascript:bang" title="1">boo</#{tag}>), %(<#{tag} title="1">boo</#{tag}>)
+    end
+  end
+
+  def test_should_block_script_tag
+    assert_sanitized %(<SCRIPT\nSRC=http://ha.ckers.org/xss.js></SCRIPT>), ""
+  end
+
+  def test_should_not_fall_for_xss_image_hack_with_uppercase_tags
+    assert_sanitized %(<IMG """><SCRIPT>alert("XSS")</SCRIPT>">), "<img>\"&gt;"
+  end
+
+  [%(<IMG SRC="javascript:alert('XSS');">),
+   %(<IMG SRC=javascript:alert('XSS')>),
+   %(<IMG SRC=JaVaScRiPt:alert('XSS')>),
+   %(<IMG SRC=javascript:alert(&quot;XSS&quot;)>),
+   %(<IMG SRC=javascript:alert(String.fromCharCode(88,83,83))>),
+   %(<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;>),
+   %(<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>),
+   %(<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>),
+   %(<IMG SRC="jav\tascript:alert('XSS');">),
+   %(<IMG SRC="jav&#x09;ascript:alert('XSS');">),
+   %(<IMG SRC="jav&#x0A;ascript:alert('XSS');">),
+   %(<IMG SRC="jav&#x0D;ascript:alert('XSS');">),
+   %(<IMG SRC=" &#14;  javascript:alert('XSS');">),
+   %(<IMG SRC="javascript&#x3a;alert('XSS');">),
+   %(<IMG SRC=`javascript:alert("RSnake says, 'XSS'")`>)].each_with_index do |img_hack, i|
+    define_method "test_should_not_fall_for_xss_image_hack_#{i+1}" do
+      assert_sanitized img_hack, "<img>"
+    end
+  end
+
+  def test_should_sanitize_tag_broken_up_by_null
+    assert_sanitized %(<SCR\0IPT>alert(\"XSS\")</SCR\0IPT>), ""
+  end
+
+  def test_should_sanitize_invalid_script_tag
+    assert_sanitized %(<SCRIPT/XSS SRC="http://ha.ckers.org/xss.js"></SCRIPT>), ""
+  end
+
+  def test_should_sanitize_script_tag_with_multiple_open_brackets
+    assert_sanitized %(<<SCRIPT>alert("XSS");//<</SCRIPT>), "alert(\"XSS\");//"
+    assert_sanitized %(<iframe src=http://ha.ckers.org/scriptlet.html\n<a), ""
+  end
+
+  def test_should_sanitize_unclosed_script
+    assert_sanitized %(<SCRIPT SRC=http://ha.ckers.org/xss.js?<B>), ""
+  end
+
+  def test_should_sanitize_half_open_scripts
+    assert_sanitized %(<IMG SRC="javascript:alert('XSS')"), "<img>"
+  end
+
+  def test_should_not_fall_for_ridiculous_hack
+    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>)
+    assert_sanitized img_hack, "<img>"
+  end
+
+  def test_should_sanitize_attributes
+    assert_sanitized %(<SPAN title="'><script>alert()</script>">blah</SPAN>), %(<span title="#{CGI.escapeHTML "'><script>alert()</script>"}">blah</span>)
+  end
+
+  def test_should_sanitize_illegal_style_properties
+    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;)
+    expected = %(display: block; width: 100%; height: 100%; background-color: black; background-x: center; background-y: center;)
+    assert_equal expected, sanitize_css(raw)
+  end
+
+  def test_should_sanitize_with_trailing_space
+    raw = "display:block; "
+    expected = "display: block;"
+    assert_equal expected, sanitize_css(raw)
+  end
+
+  def test_should_sanitize_xul_style_attributes
+    raw = %(-moz-binding:url('http://ha.ckers.org/xssmoz.xml#xss'))
+    assert_equal '', sanitize_css(raw)
+  end
+
+  def test_should_sanitize_invalid_tag_names
+    assert_sanitized(%(a b c<script/XSS src="http://ha.ckers.org/xss.js"></script>d e f), "a b cd e f")
+  end
+
+  def test_should_sanitize_non_alpha_and_non_digit_characters_in_tags
+    assert_sanitized('<a onclick!#$%&()*~+-_.,:;?@[/|\]^`=alert("XSS")>foo</a>', "<a>foo</a>")
+  end
+
+  def test_should_sanitize_invalid_tag_names_in_single_tags
+    assert_sanitized('<img/src="http://ha.ckers.org/xss.js"/>', "<img />")
+  end
+
+  def test_should_sanitize_img_dynsrc_lowsrc
+    assert_sanitized(%(<img lowsrc="javascript:alert('XSS')" />), "<img />")
+  end
+
+  def test_should_sanitize_div_background_image_unicode_encoded
+    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)
+    assert_equal '', sanitize_css(raw)
+  end
+
+  def test_should_sanitize_div_style_expression
+    raw = %(width: expression(alert('XSS'));)
+    assert_equal '', sanitize_css(raw)
+  end
+
+  def test_should_sanitize_across_newlines
+    raw = %(\nwidth:\nexpression(alert('XSS'));\n)
+    assert_equal '', sanitize_css(raw)
+  end
+
+  def test_should_sanitize_img_vbscript
+    assert_sanitized %(<img src='vbscript:msgbox("XSS")' />), '<img />'
+  end
+
+  def test_should_sanitize_cdata_section
+    assert_sanitized "<![CDATA[<span>section</span>]]>", "section]]&gt;"
+  end
+
+  def test_should_sanitize_unterminated_cdata_section
+    assert_sanitized "<![CDATA[<span>neverending...", "neverending..."
+  end
+
+  def test_should_not_mangle_urls_with_ampersand
+     assert_sanitized %{<a href=\"http://www.domain.com?var1=1&amp;var2=2\">my link</a>}
+  end
+
+  def test_should_sanitize_neverending_attribute
+    assert_sanitized "<span class=\"\\", "<span class=\"\\\">"
+  end
+
+  [
+    %(<a href="javascript&#x3a;alert('XSS');">),
+    %(<a href="javascript&#x003a;alert('XSS');">),
+    %(<a href="javascript&#x3A;alert('XSS');">),
+    %(<a href="javascript&#x003A;alert('XSS');">)
+  ].each_with_index do |enc_hack, i|
+    define_method "test_x03a_handling_#{i+1}" do
+      assert_sanitized enc_hack, "<a>"
+    end
+  end
+
+  def test_x03a_legitimate
+    assert_sanitized %(<a href="http&#x3a;//legit">), %(<a href="http://legit">)
+    assert_sanitized %(<a href="http&#x3A;//legit">), %(<a href="http://legit">)
+  end
+
+protected
+
+  def xpath_sanitize(input, options = {})
+    XpathRemovalTestSanitizer.new.sanitize(input, options)
+  end
+
+  def full_sanitize(input, options = {})
+    Html::FullSanitizer.new.sanitize(input, options)
+  end
+
+  def link_sanitize(input, options = {})
+    Html::LinkSanitizer.new.sanitize(input, options)
+  end
+
+  def white_list_sanitize(input, options = {})
+    Html::WhiteListSanitizer.new.sanitize(input, options)
+  end
+
+  def assert_sanitized(input, expected = nil)
+    if input
+      assert_dom_equal expected || input, white_list_sanitize(input)
+    else
+      assert_nil white_list_sanitize(input)
+    end
+  end
+
+  def sanitize_css(input)
+    Html::WhiteListSanitizer.new.sanitize_css(input)
+  end
+
+  def scope_allowed_tags(tags)
+    Html::WhiteListSanitizer.allowed_tags = %w(u)
+    yield Html::WhiteListSanitizer.new
+
+  ensure
+    Html::WhiteListSanitizer.allowed_tags = nil
+  end
+
+  def scope_allowed_attributes(attributes)
+    Html::WhiteListSanitizer.allowed_attributes = attributes
+    yield Html::WhiteListSanitizer.new
+
+  ensure
+    Html::WhiteListSanitizer.allowed_attributes = nil
+  end
+end

data/test/scrubbers_test.rb

@@ -0,0 +1,160 @@
+require "minitest/autorun"
+require "sanitizer"
+
+class ScrubberTest < Minitest::Test
+  protected
+
+    def assert_scrubbed(html, expected = html)
+      output = Loofah.scrub_fragment(html, @scrubber).to_s
+      assert_equal expected, output
+    end
+
+    def to_node(text)
+      Loofah.fragment(text).children.first
+    end
+
+    def assert_node_skipped(text)
+      assert_scrub_returns(Loofah::Scrubber::CONTINUE, text)
+    end
+
+    def assert_scrub_stopped(text)
+      assert_scrub_returns(Loofah::Scrubber::STOP, text)
+    end
+
+    def assert_scrub_returns(return_value, text)
+      node = to_node(text)
+      assert_equal return_value, @scrubber.scrub(node)
+    end
+end
+
+class PermitScrubberTest < ScrubberTest
+
+  def setup
+    @scrubber = Html::PermitScrubber.new
+  end
+
+  def test_responds_to_scrub
+    assert @scrubber.respond_to?(:scrub)
+  end
+
+  def test_default_scrub_behavior
+    assert_scrubbed '<tag>hello</tag>', 'hello'
+  end
+
+  def test_default_attributes_removal_behavior
+    assert_scrubbed '<p cooler="hello">hello</p>', '<p>hello</p>'
+  end
+
+  def test_leaves_supplied_tags
+    @scrubber.tags = %w(a)
+    assert_scrubbed '<a>hello</a>'
+  end
+
+  def test_leaves_only_supplied_tags
+    html = '<tag>leave me <span>now</span></tag>'
+    @scrubber.tags = %w(tag)
+    assert_scrubbed html, '<tag>leave me now</tag>'
+  end
+
+  def test_leaves_only_supplied_tags_nested
+    html = '<tag>leave <em>me <span>now</span></em></tag>'
+    @scrubber.tags = %w(tag)
+    assert_scrubbed html, '<tag>leave me now</tag>'
+  end
+
+  def test_leaves_supplied_attributes
+    @scrubber.attributes = %w(cooler)
+    assert_scrubbed '<a cooler="hello"></a>'
+  end
+
+  def test_leaves_only_supplied_attributes
+    @scrubber.attributes = %w(cooler)
+    assert_scrubbed '<a cooler="hello" b="c" d="e"></a>', '<a cooler="hello"></a>'
+  end
+
+  def test_leaves_supplied_tags_and_attributes
+    @scrubber.tags = %w(tag)
+    @scrubber.attributes = %w(cooler)
+    assert_scrubbed '<tag cooler="hello"></tag>'
+  end
+
+  def test_leaves_only_supplied_tags_and_attributes
+    @scrubber.tags = %w(tag)
+    @scrubber.attributes = %w(cooler)
+    html = '<a></a><tag href=""></tag><tag cooler=""></tag>'
+    assert_scrubbed html, '<tag></tag><tag cooler=""></tag>'
+  end
+
+  def test_leaves_text
+    assert_scrubbed('some text')
+  end
+
+  def test_skips_text_nodes
+    assert_node_skipped('some text')
+  end
+
+  def test_tags_accessor_validation
+    e = assert_raises(ArgumentError) do
+      @scrubber.tags = 'tag'
+    end
+
+    assert_equal "You should pass :tags as an Enumerable", e.message
+    assert_nil @scrubber.tags, "Tags should be nil when validation fails"
+  end
+
+  def test_attributes_accessor_validation
+    e = assert_raises(ArgumentError) do
+      @scrubber.attributes = 'cooler'
+    end
+
+    assert_equal "You should pass :attributes as an Enumerable", e.message
+    assert_nil @scrubber.attributes, "Attributes should be nil when validation fails"
+  end
+end
+
+class TargetScrubberTest < ScrubberTest
+  def setup
+    @scrubber = Html::TargetScrubber.new
+  end
+
+  def test_targeting_tags_removes_only_them
+    @scrubber.tags = %w(a h1)
+    html = '<script></script><a></a><h1></h1>'
+    assert_scrubbed html, '<script></script>'
+  end
+
+  def test_targeting_tags_removes_only_them_nested
+    @scrubber.tags = %w(a)
+    html = '<tag><a><tag><a></a></tag></a></tag>'
+    assert_scrubbed html, '<tag><tag></tag></tag>'
+  end
+
+  def test_targeting_attributes_removes_only_them
+    @scrubber.attributes = %w(class id)
+    html = '<a class="a" id="b" onclick="c"></a>'
+    assert_scrubbed html, '<a onclick="c"></a>'
+  end
+
+  def test_targeting_tags_and_attributes_removes_only_them
+    @scrubber.tags = %w(tag)
+    @scrubber.attributes = %w(remove)
+    html = '<tag remove="" other=""></tag><a remove="" other=""></a>'
+    assert_scrubbed html, '<a other=""></a>'
+  end
+end
+
+class ReturningStopFromScrubNodeTest < ScrubberTest
+  class ScrubStopper < Html::PermitScrubber
+    def scrub_node(node)
+      Loofah::Scrubber::STOP
+    end
+  end
+
+  def setup
+    @scrubber = ScrubStopper.new
+  end
+
+  def test_returns_stop_from_scrub_if_scrub_node_does
+    assert_scrub_stopped '<script>remove me</script>'
+  end
+end

metadata

@@ -0,0 +1,127 @@
+--- !ruby/object:Gem::Specification
+name: html-sanitizer
+version: !ruby/object:Gem::Version
+  version: 0.0.1
+platform: ruby
+authors:
+- Rafael Mendonça França
+- Kasper Timm Hansen
+autorequire:
+bindir: bin
+cert_chain: []
+date: 2014-12-19 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: '2.0'
+  name: loofah
+  prerelease: false
+  type: :runtime
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: '2.0'
+- !ruby/object:Gem::Dependency
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: '1.3'
+  name: bundler
+  prerelease: false
+  type: :development
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: '1.3'
+- !ruby/object:Gem::Dependency
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  name: rake
+  prerelease: false
+  type: :development
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  name: minitest
+  prerelease: false
+  type: :development
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  name: rails-dom-testing
+  prerelease: false
+  type: :development
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+description: HTML sanitization
+email:
+- dieter.pisarewski@gmail.com
+- rafaelmfranca@gmail.com
+- kaspth@gmail.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- LICENSE.txt
+- README.md
+- lib/html/sanitizer.rb
+- lib/html/sanitizer/sanitizer.rb
+- lib/html/sanitizer/scrubbers.rb
+- lib/html/sanitizer/version.rb
+- test/sanitizer_test.rb
+- test/scrubbers_test.rb
+homepage: https://github.com/dpisarewski/html-sanitizer
+licenses:
+- MIT
+metadata: {}
+post_install_message:
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project:
+rubygems_version: 2.2.2
+signing_key:
+specification_version: 4
+summary: This gem is responsible to sanitize HTML fragments.
+test_files:
+- test/sanitizer_test.rb
+- test/scrubbers_test.rb