Extract SanitizeNodeLink and apply to WikiLinkFilter

The SanitizationFilter was running before the WikiFilter. Since
WikiFilter can modify links, we could see links that _should_ be stopped
by SanatizationFilter being rendered on the page. I (kerrizor) had
previously addressed the bug in: https://gitlab.com/gitlab-org/gitlab-ee/commit/7bc971915bbeadb950bb0e1f13510bf3038229a4
However, an additional exploit was discovered after that was merged.
Working through the issue, we couldn't simply shuffle the order of
filters, due to some implicit assumptions about the order of filters, so
instead we've extracted the logic that sanitizes a Nokogiri-generated
Node object, and applied it to the WikiLinkFilter as well.

On moving filters around:
Once we start moving around filters, we get cascading failures; fix one,
another one crops up. Many of the existing filters in the WikiPipeline
chain seem to assume that other filters have already done their work,
and thus operate on a "transform anything that's left" basis;
WikiFilter, for instance, assumes any link it finds in the markdown
should be prepended with the wiki_base_path.. but if it does that, it
also turns `href="@user"` into `href="/path/to/wiki/@user"`, which the
UserReferenceFilter doesn't see as a user reference it needs to
transform into a user profile link. This is true for all the reference
filters in the WikiPipeline.

changelogs/unreleased/security-60143-patch-additional-xss-vector-in-wikis.yml

+---
+title: Patch XSS issue in wiki links
+merge_request:
+author:
+type: security

lib/banzai/filter/autolink_filter.rb

@@ -18,6 +18,7 @@ module Banzai
     #
     class AutolinkFilter < HTML::Pipeline::Filter
       include ActionView::Helpers::TagHelper
+      include Gitlab::Utils::SanitizeNodeLink
 
       # Pattern to match text that should be autolinked.
       #
@@ -72,19 +73,11 @@ module Banzai
 
       private
 
-      # Return true if any of the UNSAFE_PROTOCOLS strings are included in the URI scheme
-      def contains_unsafe?(scheme)
-        return false unless scheme
-
-        scheme = scheme.strip.downcase
-        Banzai::Filter::SanitizationFilter::UNSAFE_PROTOCOLS.any? { |protocol| scheme.include?(protocol) }
-      end
-
       def autolink_match(match)
         # start by stripping out dangerous links
         begin
           uri = Addressable::URI.parse(match)
-          return match if contains_unsafe?(uri.scheme)
+          return match unless safe_protocol?(uri.scheme)
         rescue Addressable::URI::InvalidURIError
           return match
         end

lib/banzai/filter/sanitization_filter.rb

@@ -7,8 +7,8 @@ module Banzai
     # Extends HTML::Pipeline::SanitizationFilter with a custom whitelist.
     class SanitizationFilter < HTML::Pipeline::SanitizationFilter
       include Gitlab::Utils::StrongMemoize
+      extend Gitlab::Utils::SanitizeNodeLink
 
-      UNSAFE_PROTOCOLS        = %w(data javascript vbscript).freeze
       TABLE_ALIGNMENT_PATTERN = /text-align: (?<alignment>center|left|right)/.freeze
 
       def whitelist
@@ -51,7 +51,7 @@ module Banzai
         # Allow any protocol in `a` elements
         # and then remove links with unsafe protocols
         whitelist[:protocols].delete('a')
-        whitelist[:transformers].push(self.class.remove_unsafe_links)
+        whitelist[:transformers].push(self.class.method(:remove_unsafe_links))
 
         # Remove `rel` attribute from `a` elements
         whitelist[:transformers].push(self.class.remove_rel)
@@ -69,35 +69,6 @@ module Banzai
       end
 
       class << self
-        def remove_unsafe_links
-          lambda do |env|
-            node = env[:node]
-
-            return unless node.name == 'a'
-            return unless node.has_attribute?('href')
-
-            begin
-              node['href'] = node['href'].strip
-              uri = Addressable::URI.parse(node['href'])
-
-              return unless uri.scheme
-
-              # Remove all invalid scheme characters before checking against the
-              # list of unsafe protocols.
-              #
-              # See https://tools.ietf.org/html/rfc3986#section-3.1
-              scheme = uri.scheme
-                .strip
-                .downcase
-                .gsub(/[^A-Za-z0-9\+\.\-]+/, '')
-
-              node.remove_attribute('href') if UNSAFE_PROTOCOLS.include?(scheme)
-            rescue Addressable::URI::InvalidURIError
-              node.remove_attribute('href')
-            end
-          end
-        end
-
         def remove_rel
           lambda do |env|
             if env[:node_name] == 'a'

lib/banzai/filter/wiki_link_filter.rb

@@ -8,15 +8,19 @@ module Banzai
     # Context options:
     #   :project_wiki
     class WikiLinkFilter < HTML::Pipeline::Filter
+      include Gitlab::Utils::SanitizeNodeLink
+
       def call
         return doc unless project_wiki?
 
-        doc.search('a:not(.gfm)').each { |el| process_link_attr(el.attribute('href')) }
-        doc.search('video').each { |el| process_link_attr(el.attribute('src')) }
+        doc.search('a:not(.gfm)').each { |el| process_link(el.attribute('href'), el) }
+
+        doc.search('video').each { |el| process_link(el.attribute('src'), el) }
+
         doc.search('img').each do |el|
           attr = el.attribute('data-src') || el.attribute('src')
 
-          process_link_attr(attr)
+          process_link(attr, el)
         end
 
         doc
@@ -24,6 +28,11 @@ module Banzai
 
       protected
 
+      def process_link(link_attr, node)
+        process_link_attr(link_attr)
+        remove_unsafe_links({ node: node }, remove_invalid_links: false)
+      end
+
       def project_wiki?
         !context[:project_wiki].nil?
       end

lib/banzai/filter/wiki_link_filter/rewriter.rb

@@ -4,8 +4,6 @@ module Banzai
   module Filter
     class WikiLinkFilter < HTML::Pipeline::Filter
       class Rewriter
-        UNSAFE_SLUG_REGEXES = [/\Ajavascript:/i].freeze
-
         def initialize(link_string, wiki:, slug:)
           @uri = Addressable::URI.parse(link_string)
           @wiki_base_path = wiki && wiki.wiki_base_path
@@ -37,8 +35,6 @@ module Banzai
 
         # Of the form `./link`, `../link`, or similar
         def apply_hierarchical_link_rules!
-          return if slug_considered_unsafe?
-
           @uri = Addressable::URI.join(@slug, @uri) if @uri.to_s[0] == '.'
         end
 
@@ -58,10 +54,6 @@ module Banzai
         def repository_upload?
           @uri.relative? && @uri.path.starts_with?(Wikis::CreateAttachmentService::ATTACHMENT_PATH)
         end
-
-        def slug_considered_unsafe?
-          UNSAFE_SLUG_REGEXES.any? { |r| r.match?(@slug) }
-        end
       end
     end
   end

lib/gitlab/utils/sanitize_node_link.rb

+# frozen_string_literal: true
+
+require_dependency 'gitlab/utils'
+
+module Gitlab
+  module Utils
+    module SanitizeNodeLink
+      UNSAFE_PROTOCOLS  = %w(data javascript vbscript).freeze
+      ATTRS_TO_SANITIZE = %w(href src data-src).freeze
+
+      def remove_unsafe_links(env, remove_invalid_links: true)
+        node = env[:node]
+
+        sanitize_node(node: node, remove_invalid_links: remove_invalid_links)
+
+        # HTML entities such as <video></video> have scannable attrs in
+        #   children elements, which also need to be sanitized.
+        #
+        node.children.each do |child_node|
+          sanitize_node(node: child_node, remove_invalid_links: remove_invalid_links)
+        end
+      end
+
+      # Remove all invalid scheme characters before checking against the
+      # list of unsafe protocols.
+      #
+      # See https://tools.ietf.org/html/rfc3986#section-3.1
+      #
+      def safe_protocol?(scheme)
+        return false unless scheme
+
+        scheme = scheme
+          .strip
+          .downcase
+          .gsub(/[^A-Za-z\+\.\-]+/, '')
+
+        UNSAFE_PROTOCOLS.none?(scheme)
+      end
+
+      private
+
+      def sanitize_node(node:, remove_invalid_links: true)
+        ATTRS_TO_SANITIZE.each do |attr|
+          next unless node.has_attribute?(attr)
+
+          begin
+            node[attr] = node[attr].strip
+            uri = Addressable::URI.parse(node[attr])
+
+            next unless uri.scheme
+            next if safe_protocol?(uri.scheme)
+
+            node.remove_attribute(attr)
+          rescue Addressable::URI::InvalidURIError
+            node.remove_attribute(attr) if remove_invalid_links
+          end
+        end
+      end
+    end
+  end
+end

spec/lib/banzai/filter/wiki_link_filter_spec.rb

@@ -70,47 +70,5 @@ describe Banzai::Filter::WikiLinkFilter do
         expect(filtered_link.attribute('href').value).to eq(invalid_link)
       end
     end
-
-    context "when the slug is deemed unsafe or invalid" do
-      let(:link) { "alert(1);" }
-
-      invalid_slugs = [
-        "javascript:",
-        "JaVaScRiPt:",
-        "\u0001java\u0003script:",
-        "javascript    :",
-        "javascript:    ",
-        "javascript    :   ",
-        ":javascript:",
-        "javascript:",
-        "javascript:",
-        "javascript:",
-        "javascript:",
-        "java\0script:",
-        "   javascript:"
-        ]
-
-      invalid_slugs.each do |slug|
-        context "with the slug #{slug}" do
-          it "doesn't rewrite a (.) relative link" do
-            filtered_link = filter(
-              "<a href='.#{link}'>Link</a>",
-              project_wiki: wiki,
-              page_slug: slug).children[0]
-
-            expect(filtered_link.attribute('href').value).not_to include(slug)
-          end
-
-          it "doesn't rewrite a (..) relative link" do
-            filtered_link = filter(
-              "<a href='..#{link}'>Link</a>",
-              project_wiki: wiki,
-              page_slug: slug).children[0]
-
-            expect(filtered_link.attribute('href').value).not_to include(slug)
-          end
-        end
-      end
-    end
   end
 end

spec/lib/banzai/pipeline/wiki_pipeline_spec.rb

@@ -177,6 +177,85 @@ describe Banzai::Pipeline::WikiPipeline do
         end
       end
     end
+
+    describe "checking slug validity when assembling links" do
+      context "with a valid slug" do
+        let(:valid_slug) { "http://example.com" }
+
+        it "includes the slug in a (.) relative link" do
+          output = described_class.to_html(
+            "[Link](./alert(1);)",
+            project: project,
+            project_wiki: project_wiki,
+            page_slug: valid_slug
+          )
+
+          expect(output).to include(valid_slug)
+        end
+
+        it "includeds the slug in a (..) relative link" do
+          output = described_class.to_html(
+            "[Link](../alert(1);)",
+            project: project,
+            project_wiki: project_wiki,
+            page_slug: valid_slug
+          )
+
+          expect(output).to include(valid_slug)
+        end
+      end
+
+      context "when the slug is deemed unsafe or invalid" do
+        invalid_slugs = [
+          "javascript:",
+          "JaVaScRiPt:",
+          "\u0001java\u0003script:",
+          "javascript    :",
+          "javascript:    ",
+          "javascript    :   ",
+          ":javascript:",
+          "javascript:",
+          "javascript:",
+          "javascript:",
+          "javascript:",
+          "java\0script:",
+          "   javascript:"
+        ]
+
+        invalid_js_links = [
+          "alert(1);",
+          "alert(document.location);"
+        ]
+
+        invalid_slugs.each do |slug|
+          context "with the invalid slug #{slug}" do
+            invalid_js_links.each do |link|
+              it "doesn't include a prohibited slug in a (.) relative link '#{link}'" do
+                output = described_class.to_html(
+                  "[Link](./#{link})",
+                  project: project,
+                  project_wiki: project_wiki,
+                  page_slug: slug
+                )
+
+                expect(output).not_to include(slug)
+              end
+
+              it "doesn't include a prohibited slug in a (..) relative link '#{link}'" do
+                output = described_class.to_html(
+                  "[Link](../#{link})",
+                  project: project,
+                  project_wiki: project_wiki,
+                  page_slug: slug
+                )
+
+                expect(output).not_to include(slug)
+              end
+            end
+          end
+        end
+      end
+    end
   end
 
   describe 'videos' do

spec/lib/gitlab/utils/sanitize_node_link_spec.rb

+require 'spec_helper'
+
+describe Gitlab::Utils::SanitizeNodeLink do
+  let(:klass) do
+    struct = Struct.new(:value)
+    struct.include(described_class)
+
+    struct
+  end
+
+  subject(:object) { klass.new(:value) }
+
+  invalid_schemes = [
+    "javascript:",
+    "JaVaScRiPt:",
+    "\u0001java\u0003script:",
+    "javascript    :",
+    "javascript:    ",
+    "javascript    :   ",
+    ":javascript:",
+    "javascript:",
+    "javascript:",
+    "   javascript:"
+  ]
+
+  invalid_schemes.each do |scheme|
+    context "with the scheme: #{scheme}" do
+      describe "#remove_unsafe_links" do
+        tags = {
+          a: {
+            doc: HTML::Pipeline.parse("<a href='#{scheme}alert(1);'>foo</a>"),
+            attr: "href",
+            node_to_check: -> (doc) { doc.children.first }
+          },
+          img: {
+            doc: HTML::Pipeline.parse("<img src='#{scheme}alert(1);'>"),
+            attr: "src",
+            node_to_check: -> (doc) { doc.children.first }
+          },
+          video: {
+            doc: HTML::Pipeline.parse("<video><source src='#{scheme}alert(1);'></video>"),
+            attr: "src",
+            node_to_check: -> (doc) { doc.children.first.children.filter("source").first }
+          }
+        }
+
+        tags.each do |tag, opts|
+          context "<#{tag}> tags" do
+            it "removes the unsafe link" do
+              node = opts[:node_to_check].call(opts[:doc])
+
+              expect { object.remove_unsafe_links({ node: node }, remove_invalid_links: true) }
+                .to change { node[opts[:attr]] }
+
+              expect(node[opts[:attr]]).to be_blank
+            end
+          end
+        end
+      end
+
+      describe "#safe_protocol?" do
+        let(:doc) { HTML::Pipeline.parse("<a href='#{scheme}alert(1);'>foo</a>") }
+        let(:node) { doc.children.first }
+        let(:uri) { Addressable::URI.parse(node['href'])}
+
+        it "returns false" do
+          expect(object.safe_protocol?(scheme)).to be_falsy
+        end
+      end
+    end
+  end
+end