Set up methods for html migration

README updates as well closes LF-380 flag=none Test plan - Canvas migrations should still work properly with links as before - Quizzes people are okay with the methods Change-Id: I0c1816391e625c86ba531cff9394ea78915cf8d8 Reviewed-on: https://gerrit.instructure.com/c/canvas-lms/+/322237 Tested-by: Service Cloud Jenkins <svc.cloudjenkins@instructure.com> Reviewed-by: Paul Gray <paul.gray@instructure.com> QA-Review: Paul Gray <paul.gray@instructure.com> Product-Review: Paul Gray <paul.gray@instructure.com>
2023-07-13 16:45:22 -06:00 · 2023-07-13 16:45:22 -06:00 · 215f35e5e0
parent ebebf1e9f4
commit 215f35e5e0
15 changed files with 541 additions and 183 deletions
--- a/app/models/content_migration.rb
+++ b/app/models/content_migration.rb
@ -20,7 +20,7 @@

 class ContentMigration < ActiveRecord::Base
  include Workflow
-  include TextHelper
+  include HtmlTextHelper
  include Rails.application.routes.url_helpers
  include CanvasOutcomesHelper

@ -941,15 +941,15 @@ class ContentMigration < ActiveRecord::Base
  end

  def html_converter
-    @html_converter ||= ImportedHtmlConverter.new(self, Importers::DbMigrationQueryService.new(context, self))
+    @html_converter ||= CanvasImportedHtmlConverter.new(self)
  end

-  def convert_html(*args)
-    html_converter.convert(*args)
+  def convert_html(*args, **keyword_args)
+    html_converter.convert(*args, **keyword_args)
  end

-  def convert_text(*args)
-    html_converter.convert_text(*args)
+  def convert_text(text)
+    format_message(text || "")[0]
  end

  delegate :resolve_content_links!, to: :html_converter
--- a/app/models/importers/assessment_question_importer.rb
+++ b/app/models/importers/assessment_question_importer.rb
@ -194,7 +194,7 @@ module Importers
        next unless hash[field].present?

        hash[field] = migration.convert_html(
-          hash[field], item_type, hash[:migration_id], field, { remove_outer_nodes_if_one_child: true }
+          hash[field], item_type, hash[:migration_id], field, remove_outer_nodes_if_one_child: true
        )
      end

@ -218,7 +218,7 @@ module Importers
          next unless answer[field].present?

          answer[field] = migration.convert_html(
-            answer[field], item_type, hash[:migration_id], key, { remove_outer_nodes_if_one_child: true }
+            answer[field], item_type, hash[:migration_id], key, remove_outer_nodes_if_one_child: true
          )
        end
        if answer[:comments].present? && answer[:comments] == answer[:comments_html]
--- a/app/models/importers/db_migration_query_service.rb
+++ b/app/models/importers/db_migration_query_service.rb
@ -25,8 +25,8 @@ module Importers
  # Each function returns exactly one id (if available), and nil if an id
  # cannot be resolved
  class DbMigrationQueryService
-    def initialize(context, migration)
-      @context = context
+    def initialize(migration)
+      @context = migration.context
      @migration = migration
    end

--- a/gems/canvas_link_migrator/README.md
+++ b/gems/canvas_link_migrator/README.md
@ -1,14 +1,15 @@
 # CanvasLinkMigration

-CanvasLinkMigration handles importing links to other course content (ie Assignments, Quizzes, Content Pages etc, that are created by the RichContentService using Canvas APIs) exported from Canvas.
+CanvasLinkMigration processes html content from a Canvas export package and allows it to be retranslated into html with links to Canvas course content (ie Assignments, Quizzes, Content Pages etc).

 ## Usage

-Create a CanvasLinkMigrator::ImportedHtmlConvert using an asset_id_mapping (Get the asset_id_mapping from Canvas API after a migration) and then pass in the html_string data to be imported.
+Create a CanvasLinkMigrator::ImportedHtmlConverter using an asset_id_map (Get the asset_map from the Canvas API after a migration) and then pass in the html_string data to be imported.

+First create the ImportedHtmlConverter, and then pass the html string into the converter's convert_exported_html method to get the final html product
+
+If any links cannot be resolved appropriately, a list will be returned to you with the resolved html
 ```ruby
-converter = CanvasLinkMigrator::ImportedHtmlConvert.new(ResourceMapService.new(asset_id_mapping))
-converter.convert_and_replace(html_string)
+converter = CanvasLinkMigrator::ImportedHtmlConverter.new(resource_map: asset_id_map)
+new_html, missing_links = converter.convert_exported_html(html_string)
 ```
-
-Use convert and resolve_content_links! separately if desired
--- a/gems/canvas_link_migrator/lib/canvas_link_migrator.rb
+++ b/gems/canvas_link_migrator/lib/canvas_link_migrator.rb
@ -18,6 +18,7 @@
 # with this program. If not, see <http://www.gnu.org/licenses/>.

 require "canvas_link_migrator/resource_map_service"
+require "canvas_link_migrator/imported_html_converter"
 require "canvas_link_migrator/link_parser"
 require "canvas_link_migrator/link_replacer"
 require "canvas_link_migrator/link_resolver"
--- a/gems/canvas_link_migrator/lib/canvas_link_migrator/imported_html_converter.rb
+++ b/gems/canvas_link_migrator/lib/canvas_link_migrator/imported_html_converter.rb
@ -0,0 +1,68 @@
+# frozen_string_literal: true
+
+#
+# Copyright (C) 2023 - present Instructure, Inc.
+#
+# This file is part of Canvas.
+#
+# Canvas is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Affero General Public License as published by the Free
+# Software Foundation, version 3 of the License.
+#
+# Canvas is distributed in the hope that it will be useful, but WITHOUT ANY
+# WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
+# A PARTICULAR PURPOSE. See the GNU Affero General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Affero General Public License along
+# with this program. If not, see <http://www.gnu.org/licenses/>.
+
+require "active_support/core_ext/module"
+
+module CanvasLinkMigrator
+  class ImportedHtmlConverter
+    attr_reader :link_parser, :link_resolver, :migration_id_converter
+
+    def initialize(resource_map: nil, migration_id_converter: nil)
+      @migration_id_converter = migration_id_converter || ResourceMapService.new(resource_map)
+      @link_parser = LinkParser.new(@migration_id_converter)
+      @link_resolver = LinkResolver.new(@migration_id_converter)
+    end
+
+    delegate :convert, to: :link_parser
+    delegate :resolver_links!, to: :link_resolver
+
+    def convert_exported_html(input_html)
+      new_html = link_parser.convert(input_html, "type", "lookup_id", "field")
+      replace!(new_html)
+
+      # missing links comes back as a list for all types and fields, but if the user's only
+      # sending one piece of html at a time, we only need the first set of missing links,
+      # and only the actual missing links, not the look up information we set in this method
+      bad_links = missing_links&.first&.dig(:missing_links)
+      link_parser.reset!
+      [new_html, bad_links]
+    end
+
+    def replace!(placeholder_html)
+      link_map = link_parser.unresolved_link_map
+      return unless link_map.present?
+
+      link_resolver.resolve_links!(link_map)
+      LinkReplacer.sub_placeholders!(placeholder_html, link_map.values.map(&:values).flatten)
+      placeholder_html
+    end
+
+    def missing_links
+      link_parser.unresolved_link_map.each_with_object([]) do |(item_key, field_links), bad_links|
+        field_links.each do |field, links|
+          unresolved_links = links.select { |link| link[:replaced] && (link[:missing_url] || !link[:new_value]) }
+          unresolved_links = unresolved_links.map { |link| link.slice(:link_type, :missing_url) }
+          next unless unresolved_links.any?
+
+          bad_links << { object_lookup_id: item_key[:migration_id], object_type: item_key[:type], object_field: field, missing_links: unresolved_links }
+        end
+      end
+    end
+  end
+end
--- a/gems/canvas_link_migrator/lib/canvas_link_migrator/link_parser.rb
+++ b/gems/canvas_link_migrator/lib/canvas_link_migrator/link_parser.rb
@ -17,6 +17,9 @@
 # You should have received a copy of the GNU Affero General Public License along
 # with this program. If not, see <http://www.gnu.org/licenses/>.

+require "nokogiri"
+require "digest"
+
 module CanvasLinkMigrator
  class LinkParser
    REFERENCE_KEYWORDS = %w[CANVAS_COURSE_REFERENCE CANVAS_OBJECT_REFERENCE WIKI_REFERENCE IMS_CC_FILEBASE IMS-CC-FILEBASE].freeze
@ -76,7 +79,7 @@ module CanvasLinkMigrator
      "#{LINK_PLACEHOLDER}_#{Digest::MD5.hexdigest(old_value)}"
    end

-    def convert(html, item_type, mig_id, field, opts = {})
+    def convert(html, item_type, mig_id, field, remove_outer_nodes_if_one_child: nil)
      mig_id = mig_id.to_s
      doc = Nokogiri::HTML5(html || "")

@ -99,7 +102,7 @@ module CanvasLinkMigrator
      node = doc.at_css("body")
      return "" unless node

-      if opts[:remove_outer_nodes_if_one_child]
+      if remove_outer_nodes_if_one_child
        while node.children.size == 1 && node.child.child
          break unless CONTAINER_TYPES.member?(node.child.name) && node.child.attributes.blank?

--- a/gems/canvas_link_migrator/lib/canvas_link_migrator/link_replacer.rb
+++ b/gems/canvas_link_migrator/lib/canvas_link_migrator/link_replacer.rb
@ -20,7 +20,7 @@
 module CanvasLinkMigrator
  class LinkReplacer
    # returns false if no substitutions were made
-    def sub_placeholders!(html, links)
+    def self.sub_placeholders!(html, links)
      subbed = false
      links.each do |link|
        new_value = link[:new_value] || link[:old_value]
@ -32,7 +32,7 @@ module CanvasLinkMigrator
      subbed
    end

-    def recursively_sub_placeholders!(object, links)
+    def self.recursively_sub_placeholders!(object, links)
      subbed = false
      case object
      when Hash
--- a/gems/canvas_link_migrator/lib/canvas_link_migrator/resource_map_service.rb
+++ b/gems/canvas_link_migrator/lib/canvas_link_migrator/resource_map_service.rb
@ -35,10 +35,36 @@ module CanvasLinkMigrator
      migration_data["resource_mapping"]
    end

+    ### Overwritable methods
+    def supports_embedded_images
+      false
+    end
+
+    def fix_relative_urls?
+      true
+    end
+
+    def process_domain_substitutions(url)
+      url
+    end
+
+    def context_hosts
+      migration_data["destination_hosts"]
+    end
+
+    def attachment_path_id_lookup; end
+
+    def attachment_path_id_lookup_lower; end
+
+    def root_folder_name
+      migration_data["destination_root_folder"]
+    end
+    ### End of Ovewritable methods
+
    # Returns the path for the context, for a course, it should return something like
    # "courses/1"
    def context_path
-      "/courses/#{migration_data["source_course"]}"
+      "/courses/#{migration_data["destination_course"]}"
    end

    # Looks up a wiki page slug for a migration id
--- a/gems/canvas_link_migrator/spec/canvas_link_migrator/imported_html_converter_spec.rb
+++ b/gems/canvas_link_migrator/spec/canvas_link_migrator/imported_html_converter_spec.rb
@ -0,0 +1,279 @@
+# frozen_string_literal: true
+
+#
+# Copyright (C) 2012 - present Instructure, Inc.
+#
+# This file is part of Canvas.
+#
+# Canvas is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Affero General Public License as published by the Free
+# Software Foundation, version 3 of the License.
+#
+# Canvas is distributed in the hope that it will be useful, but WITHOUT ANY
+# WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
+# A PARTICULAR PURPOSE. See the GNU Affero General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Affero General Public License along
+# with this program. If not, see <http://www.gnu.org/licenses/>.
+#
+
+require "spec_helper"
+require "json"
+
+describe CanvasLinkMigrator::ImportedHtmlConverter do
+  # tests link_parser and link_resolver
+
+  describe ".convert" do
+    before do
+      @path = "/courses/2/"
+      @converter = CanvasLinkMigrator::ImportedHtmlConverter.new(resource_map: JSON.parse(File.read("spec/fixtures/canvas_resource_map.json")))
+    end
+
+    it "converts a wiki reference" do
+      test_string = %(<a href="%24WIKI_REFERENCE%24/wiki/test-wiki-page?query=blah">Test Wiki Page</a>)
+      html, bad_links = @converter.convert_exported_html(test_string)
+      expect(html).to eq %(<a href="#{@path}pages/test-wiki-page?query=blah">Test Wiki Page</a>)
+      expect(bad_links).to be_nil
+    end
+
+    context "when course attachments exist" do
+      subject { @converter.convert_exported_html(test_string) }
+
+      let(:migration_id) { "E" }
+
+      context "and a data-download-url attribute references an icon maker icon" do
+        let(:test_string) do
+          %(<img src="$CANVAS_COURSE_REFERENCE$/file_ref/#{migration_id}/download?download_frd=1" alt="" data-inst-icon-maker-icon="true" data-download-url="$CANVAS_COURSE_REFERENCE$/file_ref/#{migration_id}/download?download_frd=1&icon_maker_icon=1">)
+        end
+
+        it "converts data-download-url for files without appending a context" do
+          html, bad_links = subject
+          expect(html).to eq(
+            "<img src=\"#{@path}files/5/download?download_frd=1\" alt=\"\" data-inst-icon-maker-icon=\"true\" data-download-url=\"/files/5/download?download_frd=1&icon_maker_icon=1\">"
+          )
+          expect(bad_links).to be_nil
+        end
+      end
+
+      # it "finds an attachment by migration id" do
+      #   test_string = %{<p>This is an image: <br /><img src="%24CANVAS_OBJECT_REFERENCE%24/attachments/F" alt=":(" /></p>}
+      #   expect(@converter.convert_exported_html(test_string)).to eq([%{<p>This is an image: <br><img src="#{@path}files/6/preview" alt=":("></p>}, nil])
+      # end
+
+      # it "finds an attachment by path" do
+      #   test_string = %{<p>This is an image: <br /><img src="%24IMS_CC_FILEBASE%24/test.png" alt=":(" /></p>}
+
+      #   # if there isn't a path->migration id map it'll be a relative course file path
+      #   expect(@converter.convert_exported_html(test_string)).to eq %{<p>This is an image: <br><img src="#{@path}file_contents/course%20files/test.png" alt=":("></p>}
+
+      #   @migration.attachment_path_id_lookup = { "test.png" => att.migration_id }
+      #   expect(@converter.convert_exported_html(test_string)).to eq %{<p>This is an image: <br><img src="#{@path}files/#{att.id}/preview" alt=":("></p>}
+      # end
+
+      # it "finds an attachment by a path with a space" do
+      #   att = make_test_att
+      #   @migration.attachment_path_id_lookup = { "subfolder/with a space/test.png" => att.migration_id }
+
+      #   test_string = %(<img src="subfolder/with%20a%20space/test.png" alt="nope" />)
+      #   expect(@converter.convert_exported_html(test_string)).to eq %(<img src="#{@path}files/#{att.id}/preview" alt="nope">)
+
+      #   test_string = %(<img src="subfolder/with+a+space/test.png" alt="nope" />)
+      #   expect(@converter.convert_exported_html(test_string)).to eq %(<img src="#{@path}files/#{att.id}/preview" alt="nope">)
+      # end
+
+      # it "finds an attachment even if the link has an extraneous folder" do
+      #   att = make_test_att
+      #   @migration.attachment_path_id_lookup = { "subfolder/test.png" => att.migration_id }
+
+      #   test_string = %(<img src="anotherfolder/subfolder/test.png" alt="nope" />)
+      #   expect(@converter.convert_exported_html(test_string)).to eq %(<img src="#{@path}files/#{att.id}/preview" alt="nope">)
+      # end
+
+      # it "finds an attachment by path if capitalization is different" do
+      #   att = make_test_att
+      #   @migration.attachment_path_id_lookup = { "subfolder/withCapital/test.png" => "wrong!" }
+      #   @migration.attachment_path_id_lookup_lower = { "subfolder/withcapital/test.png" => att.migration_id }
+
+      #   test_string = %(<img src="subfolder/WithCapital/TEST.png" alt="nope" />)
+      #   expect(@converter.convert_exported_html(test_string)).to eq %(<img src="#{@path}files/#{att.id}/preview" alt="nope">)
+      # end
+
+      # it "finds an attachment with query params" do
+      #   att = make_test_att
+      #   @migration.attachment_path_id_lookup = { "test.png" => att.migration_id }
+
+      #   test_string = %(<img src="%24IMS_CC_FILEBASE%24/test.png?canvas_customaction=1&canvas_qs_customparam=1" alt="nope" />)
+      #   expect(@converter.convert_exported_html(test_string)).to eq %(<img src="#{@path}files/#{att.id}/customaction?customparam=1" alt="nope">)
+
+      #   test_string = %(<img src="%24IMS_CC_FILEBASE%24/test.png?canvas_qs_customparam2=3" alt="nope" />)
+      #   expect(@converter.convert_exported_html(test_string)).to eq %(<img src="#{@path}files/#{att.id}/preview?customparam2=3" alt="nope">)
+
+      #   test_string = %(<img src="%24IMS_CC_FILEBASE%24/test.png?notarelevantparam" alt="nope" />)
+      #   expect(@converter.convert_exported_html(test_string)).to eq %(<img src="#{@path}files/#{att.id}/preview" alt="nope">)
+      # end
+    end
+
+    it "converts picture source srcsets" do
+      test_string = %(<source srcset="$CANVAS_COURSE_REFERENCE$/img.src">)
+      expect(@converter.convert_exported_html(test_string)).to eq([%(<source srcset="/courses/2/img.src">), nil])
+    end
+
+    it "converts a wiki reference without $ escaped" do
+      test_string = %(<a href="$WIKI_REFERENCE$/wiki/test-wiki-page?query=blah">Test Wiki Page</a>)
+
+      expect(@converter.convert_exported_html(test_string)).to eq([%(<a href="#{@path}pages/test-wiki-page?query=blah">Test Wiki Page</a>), nil])
+    end
+
+    it "converts a wiki reference by migration id" do
+      test_string = %(<a href="wiki_page_migration_id=A">Test Wiki Page</a>)
+
+      expect(@converter.convert_exported_html(test_string)).to eq([%(<a href="#{@path}pages/slug-a">Test Wiki Page</a>), nil])
+    end
+
+    it "converts a discussion reference by migration id" do
+      test_string = %(<a href="discussion_topic_migration_id=G">Test topic</a>)
+
+      expect(@converter.convert_exported_html(test_string)).to eq([%(<a href="#{@path}discussion_topics/7">Test topic</a>), nil])
+    end
+
+    it "converts course section urls" do
+      test_string = %(<a href="%24CANVAS_COURSE_REFERENCE%24/discussion_topics">discussions</a>)
+      expect(@converter.convert_exported_html(test_string)).to eq([%(<a href="#{@path}discussion_topics">discussions</a>), nil])
+    end
+
+    it "leaves invalid and absolute urls alone" do
+      test_string = %(<a href="stupid &^%$ url">Linkage</a><br><a href="http://www.example.com/poop">Linkage</a>)
+      expect(@converter.convert_exported_html(test_string)).to eq([%(<a href="stupid &amp;^%$ url">Linkage</a><br><a href="http://www.example.com/poop">Linkage</a>), nil])
+    end
+
+    it "leaves invalid mailto addresses alone" do
+      test_string = %(<a href="mailto:.">Bad mailto</a><br><a href="mailto:test@example.com">Good mailto</a>)
+      expect(@converter.convert_exported_html(test_string)).to eq(
+        [
+          %(<a href="mailto:.">Bad mailto</a><br><a href="mailto:test@example.com">Good mailto</a>),
+          nil
+        ]
+      )
+    end
+
+    it "recognizes and relative-ize absolute links outside the course but in one of the course's domains" do
+      test_string = %(<a href="https://apple.edu/courses/123">Mine</a><br><a href="https://kiwi.edu/courses/456">Vain</a><br><a href="http://other-canvas.example.com/">Other Instance</a>)
+      expect(@converter.convert_exported_html(test_string)).to eq([%(<a href="/courses/123">Mine</a><br><a href="/courses/456">Vain</a><br><a href="http://other-canvas.example.com/">Other Instance</a>), nil])
+    end
+
+    it "prepends course files for unrecognized relative urls" do
+      test_string = %(<a href="/relative/path/to/file">Linkage</a>)
+      html, bad_links = @converter.convert_exported_html(test_string)
+      expect(html).to eq %(<a href="#{@path}file_contents/course%20files/relative/path/to/file">Linkage</a>)
+      expect(bad_links.length).to eq 1
+      expect(bad_links[0]).to include({ link_type: :file, missing_url: "/courses/2/file_contents/course%20files/relative/path/to/file" })
+
+      test_string = %(<a href="relative/path/to/file">Linkage</a>)
+      html, bad_links = @converter.convert_exported_html(test_string)
+      expect(html).to eq %(<a href="#{@path}file_contents/course%20files/relative/path/to/file">Linkage</a>)
+      expect(bad_links.length).to eq 1
+      expect(bad_links[0]).to include({ link_type: :file, missing_url: "/courses/2/file_contents/course%20files/relative/path/to/file" })
+
+      test_string = %(<a href="relative/path/to/file%20with%20space.html">Linkage</a>)
+      html, bad_links = @converter.convert_exported_html(test_string)
+      expect(html).to eq %(<a href="#{@path}file_contents/course%20files/relative/path/to/file%20with%20space.html">Linkage</a>)
+      expect(bad_links.length).to eq 1
+      expect(bad_links[0]).to include({ link_type: :file, missing_url: "/courses/2/file_contents/course%20files/relative/path/to/file%20with%20space.html" })
+    end
+
+    it "preserves media comment links" do
+      test_string = <<~HTML.strip
+        <p>
+          with media object url: <a id="media_comment_m-stuff" class="instructure_inline_media_comment video_comment" href="/media_objects/m-stuff">this is a media comment</a>
+          with file content url: <a id="media_comment_0_bq09qam2" class="instructure_inline_media_comment video_comment" href="/courses/2/file_contents/course%20files/media_objects/0_bq09qam2">this is a media comment</a>
+        </p>
+      HTML
+
+      expect(@converter.convert_exported_html(test_string)).to eq([test_string, nil])
+    end
+
+    it "handles and repair half broken media links" do
+      test_string = %(<p><a href="/courses/2/file_contents/%24IMS_CC_FILEBASE%24/#" class="instructure_inline_media_comment video_comment" id="media_comment_m-stuff">this is a media comment</a><br><br></p>)
+
+      expect(@converter.convert_exported_html(test_string)).to eq([%(<p><a href="/media_objects/m-stuff" class="instructure_inline_media_comment video_comment" id="media_comment_m-stuff">this is a media comment</a><br><br></p>), nil])
+    end
+
+    it "preserves new RCE media iframes" do
+      test_string = %(<iframe style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="video" src="/media_objects_iframe/m-stuff?type=video" allowfullscreen="allowfullscreen" allow="fullscreen" data-media-id="m-stuff"></iframe>)
+      expect(@converter.convert_exported_html(test_string)).to eq([test_string, nil])
+    end
+
+    it "handles and repair half broken new RCE media iframes" do
+      test_string = %(<iframe style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="video" src="%24IMS_CC_FILEBASE%24/#" allowfullscreen="allowfullscreen" allow="fullscreen" data-media-id="m-abcde"></iframe>)
+      repaired_string = %(<iframe style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="video" src="/media_objects_iframe/m-abcde?type=video" allowfullscreen="allowfullscreen" allow="fullscreen" data-media-id="m-abcde"></iframe>)
+      expect(@converter.convert_exported_html(test_string)).to eq([repaired_string, nil])
+    end
+
+    it "converts source tags to RCE media iframes" do
+      test_string = %(<video style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="video" allowfullscreen="allowfullscreen" allow="fullscreen" data-media-id="m-stuff"><source src="/media_objects_iframe/m-stuff?type=video" data-media-id="m-stuff" data-media-type="video"></video>)
+      converted_string = %(<iframe style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="video" allowfullscreen="allowfullscreen" allow="fullscreen" data-media-id="m-stuff" src="/media_objects_iframe/m-stuff?type=video"></iframe>)
+      expect(@converter.convert_exported_html(test_string)).to eq([converted_string, nil])
+
+      test_string = %(<audio style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="audio" data-media-id="m-stuff"><source src="/media_objects_iframe/m-stuff?type=audio" data-media-id="m-stuff" data-media-type="audio"></audio>)
+      converted_string = %(<iframe style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="audio" data-media-id="m-stuff" src="/media_objects_iframe/m-stuff?type=audio"></iframe>)
+      expect(@converter.convert_exported_html(test_string)).to eq([converted_string, nil])
+    end
+
+    it "converts source tags to RCE media attachment iframes" do
+      test_string = %(<video style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="video" allowfullscreen="allowfullscreen" allow="fullscreen" data-media-id="m-stuff"><source src="$CANVAS_OBJECT_REFERENCE$/media_attachments_iframe/E?type=video" data-media-id="m-stuff" data-media-type="video"></video>)
+      converted_string = %(<iframe style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="video" allowfullscreen="allowfullscreen" allow="fullscreen" data-media-id="m-stuff" src="/media_attachments_iframe/5?type=video"></iframe>)
+      expect(@converter.convert_exported_html(test_string)).to eq([converted_string, nil])
+
+      test_string = %(<audio  style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="audio" data-media-id="m-stuff"><source src="$CANVAS_OBJECT_REFERENCE$/media_attachments_iframe/E?type=audio" data-media-id="m-stuff" data-media-type="audio"></video>)
+      converted_string = %(<iframe style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="audio" data-media-id="m-stuff" src="/media_attachments_iframe/5?type=audio"></iframe>)
+      expect(@converter.convert_exported_html(test_string)).to eq([converted_string, nil])
+    end
+
+    it "converts source tags to RCE media attachment iframes when link is untranslated" do
+      test_string = %(<video style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="video" allowfullscreen="allowfullscreen" allow="fullscreen" data-media-id="m-stuff"><source src="/media_attachments_iframe/5?type=video" data-media-id="m-stuff" data-media-type="video"></video>)
+      converted_string = %(<iframe style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="video" allowfullscreen="allowfullscreen" allow="fullscreen" data-media-id="m-stuff" src="/media_attachments_iframe/5?type=video"></iframe>)
+      expect(@converter.convert_exported_html(test_string)).to eq([converted_string, nil])
+
+      test_string = %(<audio  style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="audio" data-media-id="m-stuff"><source src="/media_attachments_iframe/5?type=audio" data-media-id="m-stuff" data-media-type="audio"></video>)
+      converted_string = %(<iframe style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="audio" data-media-id="m-stuff" src="/media_attachments_iframe/5?type=audio"></iframe>)
+      expect(@converter.convert_exported_html(test_string)).to eq([converted_string, nil])
+    end
+
+    it "leaves source tags without data-media-id alone" do
+      test_string = %(<video style="width: 400px; height: 225px; display: inline-block;" title="this is a non-canvas video" allowfullscreen="allowfullscreen" allow="fullscreen"><source src="http://www.example.com/video.mov"></video>)
+      expect(@converter.convert_exported_html(test_string)).to eq([test_string, nil])
+    end
+
+    it "only converts url params" do
+      test_string = <<~HTML
+        <object>
+        <param name="controls" value="CONSOLE" />
+        <param name="controller" value="true" />
+        <param name="autostart" value="false" />
+        <param name="loop" value="false" />
+        <param name="src" value="%24IMS_CC_FILEBASE%24/test.mp3" />
+        <EMBED name="tag"  src="%24IMS_CC_FILEBASE%24/test.mp3" loop="false" autostart="false" controller="true" controls="CONSOLE" >
+        </EMBED>
+        </object>
+      HTML
+
+      expect(@converter.convert_exported_html(test_string)[0]).to match(<<~HTML.strip)
+        <object>
+        <param name="controls" value="CONSOLE">
+        <param name="controller" value="true">
+        <param name="autostart" value="false">
+        <param name="loop" value="false">
+        <param name="src" value="/courses/2/file_contents/course%20files/test.mp3">
+        <embed name="tag" src="/courses/2/file_contents/course%20files/test.mp3" loop="false" autostart="false" controller="true" controls="CONSOLE">
+
+        </object>
+      HTML
+    end
+
+    it "leaves an anchor tag alone" do
+      test_string = '<p><a href="#anchor_ref">ref</a></p>'
+      expect(@converter.convert_exported_html(test_string)).to eq([test_string, nil])
+    end
+  end
+end
--- a/gems/canvas_link_migrator/spec/canvas_link_migrator/link_resolver_spec.rb
+++ b/gems/canvas_link_migrator/spec/canvas_link_migrator/link_resolver_spec.rb
@ -22,45 +22,45 @@ require "spec_helper"
 require "json"

 describe CanvasLinkMigrator::LinkResolver do
-  def course_based_converter(assets = JSON.parse(File.read("spec/fixtures/canvas_resource_map.json")))
+  def resolver(assets = JSON.parse(File.read("spec/fixtures/canvas_resource_map.json")))
    CanvasLinkMigrator::LinkResolver.new(CanvasLinkMigrator::ResourceMapService.new(assets))
  end

  describe "resolve_link!" do
    it "converts wiki_pages links" do
      link = { link_type: :wiki_page, migration_id: "A", query: "?foo=bar" }
-      course_based_converter.resolve_link!(link)
-      expect(link[:new_value]).to eq("/courses/1/pages/slug-a?foo=bar")
+      resolver.resolve_link!(link)
+      expect(link[:new_value]).to eq("/courses/2/pages/slug-a?foo=bar")
    end

    it "converts module_item links" do
      link = { link_type: :module_item, migration_id: "C", query: "?foo=bar" }
-      course_based_converter.resolve_link!(link)
-      expect(link[:new_value]).to eq("/courses/1/modules/items/3?foo=bar")
+      resolver.resolve_link!(link)
+      expect(link[:new_value]).to eq("/courses/2/modules/items/3?foo=bar")
    end

    it "converts file_ref urls" do
      link = { link_type: :file_ref, migration_id: "F" }
-      course_based_converter.resolve_link!(link)
-      expect(link[:new_value]).to eq("/courses/1/files/6/preview")
+      resolver.resolve_link!(link)
+      expect(link[:new_value]).to eq("/courses/2/files/6/preview")
    end

    it "converts attachment urls" do
      link = { link_type: :object, type: "attachments", migration_id: "E", query: "?foo=bar" }
-      course_based_converter.resolve_link!(link)
-      expect(link[:new_value]).to eq("/courses/1/files/5/preview")
+      resolver.resolve_link!(link)
+      expect(link[:new_value]).to eq("/courses/2/files/5/preview")
    end

    it "converts media_attachments_iframe urls" do
      link = { link_type: :object, type: "media_attachments_iframe", migration_id: "F", query: "?foo=bar" }
-      course_based_converter.resolve_link!(link)
+      resolver.resolve_link!(link)
      expect(link[:new_value]).to eq("/media_attachments_iframe/6?foo=bar")
    end

    it "converts discussion_topic links" do
      link = { link_type: :discussion_topic, migration_id: "G", query: "?foo=bar" }
-      course_based_converter.resolve_link!(link)
-      expect(link[:new_value]).to eq("/courses/1/discussion_topics/7?foo=bar")
+      resolver.resolve_link!(link)
+      expect(link[:new_value]).to eq("/courses/2/discussion_topics/7?foo=bar")
    end
  end
 end
--- a/gems/canvas_link_migrator/spec/fixtures/canvas_resource_map.json
+++ b/gems/canvas_link_migrator/spec/fixtures/canvas_resource_map.json
@ -1,67 +1,73 @@
 {
-    "contains_migration_ids": true,
-    "destination_course": "2",
-    "destination_hosts": [
-        "apple.edu",
-        "kiwi.edu"
-    ],
-    "destination_root_folder": "course files/",
-    "resource_mapping": {
-        "discussion_topics": {
-            "G": {
-                "destination": {
-                    "id": "7"
-                },
-                "source": {
-                    "id": "6"
-                }
-            }
+  "contains_migration_ids": true,
+  "destination_course": "2",
+  "destination_hosts": [
+    "apple.edu",
+    "kiwi.edu"
+  ],
+  "destination_root_folder": "course files/",
+  "resource_mapping": {
+    "discussion_topics": {
+      "G": {
+        "destination": {
+          "id": "7"
        },
-        "files": {
-            "E": {
-                "destination": {
-                    "id": "5",
-                    "media_entry_id": "m-stuff"
-                },
-                "source": {
-                    "id": "3",
-                    "media_entry_id": "m-stuff"
-                }
-            },
-            "F": {
-                "destination": {
-                    "id": "6",
-                    "media_entry_id": "m-stuff"
-                },
-                "source": {
-                    "id": "4",
-                    "media_entry_id": "m-stuff"
-                }
-            }
-        },
-        "module_items": {
-            "C": {
-                "destination": {
-                    "id": "3"
-                },
-                "source": {
-                    "id": "2"
-                }
-            }
-        },
-        "wiki_pages": {
-            "A": {
-                "destination": {
-                    "id": "2",
-                    "url": "slug-a"
-                },
-                "source": {
-                    "id": "1",
-                    "url": "slug-a"
-                }
-            }
+        "source": {
+          "id": "6"
        }
+      }
    },
-    "source_course": "1",
-    "source_host": "pineapple.edu"
+    "files": {
+      "E": {
+        "destination": {
+          "id": "5",
+          "media_entry_id": "m-stuff"
+        },
+        "source": {
+          "id": "3",
+          "media_entry_id": "m-stuff"
+        }
+      },
+      "F": {
+        "destination": {
+          "id": "6",
+          "media_entry_id": "m-stuff"
+        },
+        "source": {
+          "id": "4",
+          "media_entry_id": "m-stuff"
+        }
+      }
+    },
+    "module_items": {
+      "C": {
+        "destination": {
+          "id": "3"
+        },
+        "source": {
+          "id": "2"
+        }
+      }
+    },
+    "wiki_pages": {
+      "A": {
+        "destination": {
+          "id": "2",
+          "url": "slug-a"
+        },
+        "source": {
+          "id": "1",
+          "url": "slug-a"
+        }
+      }
+    }
+  },
+  "file_tree": {
+    "folder 1": {
+      "id": "1",
+      "sub_folders": {}
+    }
+  },
+  "source_course": "1",
+  "source_host": "pineapple.edu"
 }
--- a/lib/canvas_imported_html_converter.rb
+++ b/lib/canvas_imported_html_converter.rb
@ -20,9 +20,8 @@

 require "nokogiri"

-class ImportedHtmlConverter
-  include TextHelper
-  include HtmlTextHelper
+class CanvasImportedHtmlConverter < CanvasLinkMigrator::ImportedHtmlConverter
+  include CanvasLinkMigrator

  LINK_TYPE_TO_CLASS = {
    announcement: Announcement,
@ -35,37 +34,9 @@ class ImportedHtmlConverter
    wiki_page: WikiPage
  }.freeze

-  attr_reader :link_parser, :link_resolver, :link_replacer
-
-  delegate :convert, to: :link_parser
-
-  def initialize(migration, migration_id_converter)
+  def initialize(migration)
    @migration = migration
-    @migration_id_converter = migration_id_converter
-    @link_parser = CanvasLinkMigrator::LinkParser.new(migration_id_converter)
-    @link_resolver = CanvasLinkMigrator::LinkResolver.new(migration_id_converter)
-    @link_replacer = CanvasLinkMigrator::LinkReplacer.new
-  end
-
-  def convert_text(text)
-    format_message(text || "")[0]
-  end
-
-  def resolve_content_links!
-    link_map = @link_parser.unresolved_link_map
-    return unless link_map.present?
-
-    @link_resolver.resolve_links!(link_map)
-    replace_placeholders!(link_map)
-    @link_parser.reset!
-  end
-
-  def self.relative_url?(url)
-    URI.parse(url).relative? && !url.to_s.start_with?("//")
-  rescue URI::Error
-    # leave the url as it was
-    Rails.logger.warn "attempting to translate invalid url: #{url}"
-    false
+    super(migration_id_converter: Importers::DbMigrationQueryService.new(migration))
  end

  def context
@ -76,6 +47,15 @@ class ImportedHtmlConverter
    @migration_id_converter.context_path
  end

+  def resolve_content_links!
+    link_map = link_parser.unresolved_link_map
+    return unless link_map.present?
+
+    link_resolver.resolve_links!(link_map)
+    replace_placeholders!(link_map)
+    link_parser.reset!
+  end
+
  def replace_placeholders!(link_map)
    load_questions!(link_map)

@ -150,7 +130,7 @@ class ImportedHtmlConverter
    case item_key[:type]
    when :syllabus
      syllabus = context.syllabus_body
-      if link_replacer.sub_placeholders!(syllabus, field_links.values.flatten)
+      if LinkReplacer.sub_placeholders!(syllabus, field_links.values.flatten)
        context.class.where(id: context.id).update_all(syllabus_body: syllabus)
      end
    when :assessment_question
@ -162,7 +142,7 @@ class ImportedHtmlConverter
      item_updates = {}
      field_links.each do |field, links|
        html = item.read_attribute(field)
-        if link_replacer.sub_placeholders!(html, links)
+        if LinkReplacer.sub_placeholders!(html, links)
          item_updates[field] = html
        end
      end
@ -210,7 +190,7 @@ class ImportedHtmlConverter
    # we have to do a little bit more here because the question_data can get copied all over
    quiz_ids = []
    Quizzes::QuizQuestion.where(assessment_question_id: aq.id).find_each do |qq|
-      if link_replacer.recursively_sub_placeholders!(qq["question_data"], links)
+      if LinkReplacer.recursively_sub_placeholders!(qq["question_data"], links)
        Quizzes::QuizQuestion.where(id: qq.id).update_all(question_data: qq["question_data"])
        quiz_ids << qq.quiz_id
      end
@ -218,7 +198,7 @@ class ImportedHtmlConverter

    if quiz_ids.any?
      Quizzes::Quiz.where(id: quiz_ids.uniq).where.not(quiz_data: nil).find_each do |quiz|
-        if link_replacer.recursively_sub_placeholders!(quiz["quiz_data"], links)
+        if LinkReplacer.recursively_sub_placeholders!(quiz["quiz_data"], links)
          Quizzes::Quiz.where(id: quiz.id).update_all(quiz_data: quiz["quiz_data"])
        end
      end
@ -233,18 +213,18 @@ class ImportedHtmlConverter
      link[:new_value] = aq.translate_file_link(link[:new_value])
    end

-    if link_replacer.recursively_sub_placeholders!(aq["question_data"], links)
+    if LinkReplacer.recursively_sub_placeholders!(aq["question_data"], links)
      AssessmentQuestion.where(id: aq.id).update_all(question_data: aq["question_data"])
    end
  end

  def process_quiz_question!(qq, links)
-    if link_replacer.recursively_sub_placeholders!(qq["question_data"], links)
+    if LinkReplacer.recursively_sub_placeholders!(qq["question_data"], links)
      Quizzes::QuizQuestion.where(id: qq.id).update_all(question_data: qq["question_data"])
    end

    quiz = Quizzes::Quiz.where(id: qq.quiz_id).where.not(quiz_data: nil).first
-    if quiz && link_replacer.recursively_sub_placeholders!(quiz["quiz_data"], links)
+    if quiz && LinkReplacer.recursively_sub_placeholders!(quiz["quiz_data"], links)
      Quizzes::Quiz.where(id: quiz.id).update_all(quiz_data: quiz["quiz_data"])
    end
  end
--- a/lib/course_link_validator.rb
+++ b/lib/course_link_validator.rb
@ -191,7 +191,7 @@ class CourseLinkValidator
    end
  end

-  # pretty much copied from ImportedHtmlConverter
+  # pretty much copied from CanvasImportedHtmlConverter
  def find_invalid_links(html)
    links = []
    doc = Nokogiri::HTML5(html || "")
--- a/spec/lib/canvas_imported_html_converter_spec.rb
+++ b/spec/lib/canvas_imported_html_converter_spec.rb
@ -18,7 +18,7 @@
 # with this program. If not, see <http://www.gnu.org/licenses/>.
 #

-describe ImportedHtmlConverter do
+describe CanvasImportedHtmlConverter do
  # tests link_parser and link_resolver
  before :once do
    course_factory
@ -27,26 +27,20 @@ describe ImportedHtmlConverter do
    @converter = @migration.html_converter
  end

-  context ".convert" do
-    def convert_and_replace(test_string)
-      html = @migration.convert_html(test_string, "sometype", "somemigid", "somefield")
-      link_map = @converter.link_parser.unresolved_link_map
-      @converter.link_resolver.resolve_links!(link_map)
-      if link_map.present?
-        @converter.link_replacer.sub_placeholders!(html, link_map.values.map(&:values).flatten)
-      end
-      html
+  describe ".convert" do
+    def convert_exported_html(*args)
+      @converter.convert_exported_html(*args)[0]
    end

    it "converts a wiki reference" do
      test_string = %(<a href="%24WIKI_REFERENCE%24/wiki/test-wiki-page?query=blah">Test Wiki Page</a>)
      @course.wiki_pages.create!(title: "Test Wiki Page", body: "stuff")

-      expect(convert_and_replace(test_string)).to eq %(<a href="#{@path}pages/test-wiki-page?query=blah">Test Wiki Page</a>)
+      expect(convert_exported_html(test_string)).to eq %(<a href="#{@path}pages/test-wiki-page?query=blah">Test Wiki Page</a>)
    end

    context "when course attachments exist" do
-      subject { convert_and_replace(test_string) }
+      subject { convert_exported_html(test_string) }

      let_once(:attachment) { attachment_model(context: course, migration_id:) }
      let(:course) { @course }
@ -67,14 +61,14 @@ describe ImportedHtmlConverter do

    it "converts picture source srcsets" do
      test_string = %(<source srcset="$CANVAS_COURSE_REFERENCE$/img.src">)
-      expect(convert_and_replace(test_string)).to eq %(<source srcset="/courses/#{@course.id}/img.src">)
+      expect(convert_exported_html(test_string)).to eq %(<source srcset="/courses/#{@course.id}/img.src">)
    end

    it "converts a wiki reference without $ escaped" do
      test_string = %(<a href="$WIKI_REFERENCE$/wiki/test-wiki-page?query=blah">Test Wiki Page</a>)
      @course.wiki_pages.create!(title: "Test Wiki Page", body: "stuff")

-      expect(convert_and_replace(test_string)).to eq %(<a href="#{@path}pages/test-wiki-page?query=blah">Test Wiki Page</a>)
+      expect(convert_exported_html(test_string)).to eq %(<a href="#{@path}pages/test-wiki-page?query=blah">Test Wiki Page</a>)
    end

    it "converts a wiki reference by migration id" do
@ -83,7 +77,7 @@ describe ImportedHtmlConverter do
      wiki.migration_id = "123456677788"
      wiki.save!

-      expect(convert_and_replace(test_string)).to eq %(<a href="#{@path}pages/test-wiki-page">Test Wiki Page</a>)
+      expect(convert_exported_html(test_string)).to eq %(<a href="#{@path}pages/test-wiki-page">Test Wiki Page</a>)
    end

    it "converts a discussion reference by migration id" do
@ -92,7 +86,7 @@ describe ImportedHtmlConverter do
      topic.migration_id = "123456677788"
      topic.save!

-      expect(convert_and_replace(test_string)).to eq %(<a href="#{@path}discussion_topics/#{topic.id}">Test topic</a>)
+      expect(convert_exported_html(test_string)).to eq %(<a href="#{@path}discussion_topics/#{topic.id}">Test topic</a>)
    end

    def make_test_att
@ -106,7 +100,7 @@ describe ImportedHtmlConverter do
      att = make_test_att

      test_string = %{<p>This is an image: <br /><img src="%24CANVAS_OBJECT_REFERENCE%24/attachments/1768525836051" alt=":(" /></p>}
-      expect(convert_and_replace(test_string)).to eq %{<p>This is an image: <br><img src="#{@path}files/#{att.id}/preview" alt=":("></p>}
+      expect(convert_exported_html(test_string)).to eq %{<p>This is an image: <br><img src="#{@path}files/#{att.id}/preview" alt=":("></p>}
    end

    it "finds an attachment by path" do
@ -115,10 +109,10 @@ describe ImportedHtmlConverter do
      test_string = %{<p>This is an image: <br /><img src="%24IMS_CC_FILEBASE%24/test.png" alt=":(" /></p>}

      # if there isn't a path->migration id map it'll be a relative course file path
-      expect(convert_and_replace(test_string)).to eq %{<p>This is an image: <br><img src="#{@path}file_contents/course%20files/test.png" alt=":("></p>}
+      expect(convert_exported_html(test_string)).to eq %{<p>This is an image: <br><img src="#{@path}file_contents/course%20files/test.png" alt=":("></p>}

      @migration.attachment_path_id_lookup = { "test.png" => att.migration_id }
-      expect(convert_and_replace(test_string)).to eq %{<p>This is an image: <br><img src="#{@path}files/#{att.id}/preview" alt=":("></p>}
+      expect(convert_exported_html(test_string)).to eq %{<p>This is an image: <br><img src="#{@path}files/#{att.id}/preview" alt=":("></p>}
    end

    it "finds an attachment by a path with a space" do
@ -126,10 +120,10 @@ describe ImportedHtmlConverter do
      @migration.attachment_path_id_lookup = { "subfolder/with a space/test.png" => att.migration_id }

      test_string = %(<img src="subfolder/with%20a%20space/test.png" alt="nope" />)
-      expect(convert_and_replace(test_string)).to eq %(<img src="#{@path}files/#{att.id}/preview" alt="nope">)
+      expect(convert_exported_html(test_string)).to eq %(<img src="#{@path}files/#{att.id}/preview" alt="nope">)

      test_string = %(<img src="subfolder/with+a+space/test.png" alt="nope" />)
-      expect(convert_and_replace(test_string)).to eq %(<img src="#{@path}files/#{att.id}/preview" alt="nope">)
+      expect(convert_exported_html(test_string)).to eq %(<img src="#{@path}files/#{att.id}/preview" alt="nope">)
    end

    it "finds an attachment even if the link has an extraneous folder" do
@ -137,7 +131,7 @@ describe ImportedHtmlConverter do
      @migration.attachment_path_id_lookup = { "subfolder/test.png" => att.migration_id }

      test_string = %(<img src="anotherfolder/subfolder/test.png" alt="nope" />)
-      expect(convert_and_replace(test_string)).to eq %(<img src="#{@path}files/#{att.id}/preview" alt="nope">)
+      expect(convert_exported_html(test_string)).to eq %(<img src="#{@path}files/#{att.id}/preview" alt="nope">)
    end

    it "finds an attachment by path if capitalization is different" do
@ -146,7 +140,7 @@ describe ImportedHtmlConverter do
      @migration.attachment_path_id_lookup_lower = { "subfolder/withcapital/test.png" => att.migration_id }

      test_string = %(<img src="subfolder/WithCapital/TEST.png" alt="nope" />)
-      expect(convert_and_replace(test_string)).to eq %(<img src="#{@path}files/#{att.id}/preview" alt="nope">)
+      expect(convert_exported_html(test_string)).to eq %(<img src="#{@path}files/#{att.id}/preview" alt="nope">)
    end

    it "finds an attachment with query params" do
@ -154,28 +148,28 @@ describe ImportedHtmlConverter do
      @migration.attachment_path_id_lookup = { "test.png" => att.migration_id }

      test_string = %(<img src="%24IMS_CC_FILEBASE%24/test.png?canvas_customaction=1&canvas_qs_customparam=1" alt="nope" />)
-      expect(convert_and_replace(test_string)).to eq %(<img src="#{@path}files/#{att.id}/customaction?customparam=1" alt="nope">)
+      expect(convert_exported_html(test_string)).to eq %(<img src="#{@path}files/#{att.id}/customaction?customparam=1" alt="nope">)

      test_string = %(<img src="%24IMS_CC_FILEBASE%24/test.png?canvas_qs_customparam2=3" alt="nope" />)
-      expect(convert_and_replace(test_string)).to eq %(<img src="#{@path}files/#{att.id}/preview?customparam2=3" alt="nope">)
+      expect(convert_exported_html(test_string)).to eq %(<img src="#{@path}files/#{att.id}/preview?customparam2=3" alt="nope">)

      test_string = %(<img src="%24IMS_CC_FILEBASE%24/test.png?notarelevantparam" alt="nope" />)
-      expect(convert_and_replace(test_string)).to eq %(<img src="#{@path}files/#{att.id}/preview" alt="nope">)
+      expect(convert_exported_html(test_string)).to eq %(<img src="#{@path}files/#{att.id}/preview" alt="nope">)
    end

    it "converts course section urls" do
      test_string = %(<a href="%24CANVAS_COURSE_REFERENCE%24/discussion_topics">discussions</a>)
-      expect(convert_and_replace(test_string)).to eq %(<a href="#{@path}discussion_topics">discussions</a>)
+      expect(convert_exported_html(test_string)).to eq %(<a href="#{@path}discussion_topics">discussions</a>)
    end

    it "leaves invalid and absolute urls alone" do
      test_string = %(<a href="stupid &^%$ url">Linkage</a><br><a href="http://www.example.com/poop">Linkage</a>)
-      expect(convert_and_replace(test_string)).to eq %(<a href="stupid &amp;^%$ url">Linkage</a><br><a href="http://www.example.com/poop">Linkage</a>)
+      expect(convert_exported_html(test_string)).to eq %(<a href="stupid &amp;^%$ url">Linkage</a><br><a href="http://www.example.com/poop">Linkage</a>)
    end

    it "leaves invalid mailto addresses alone" do
      test_string = %(<a href="mailto:.">Bad mailto</a><br><a href="mailto:test@example.com">Good mailto</a>)
-      expect(convert_and_replace(test_string)).to eq(
+      expect(convert_exported_html(test_string)).to eq(
        %(<a href="mailto:.">Bad mailto</a><br><a href="mailto:test@example.com">Good mailto</a>)
      )
    end
@ -183,16 +177,16 @@ describe ImportedHtmlConverter do
    it "recognizes and relative-ize absolute links outside the course but in one of the course's domains" do
      allow(HostUrl).to receive(:context_hosts).with(@course.root_account).and_return(["my-canvas.example.com", "vanity.my-canvas.edu"])
      test_string = %(<a href="https://my-canvas.example.com/courses/123">Mine</a><br><a href="https://vanity.my-canvas.edu/courses/456">Vain</a><br><a href="http://other-canvas.example.com/">Other Instance</a>)
-      expect(convert_and_replace(test_string)).to eq %(<a href="/courses/123">Mine</a><br><a href="/courses/456">Vain</a><br><a href="http://other-canvas.example.com/">Other Instance</a>)
+      expect(convert_exported_html(test_string)).to eq %(<a href="/courses/123">Mine</a><br><a href="/courses/456">Vain</a><br><a href="http://other-canvas.example.com/">Other Instance</a>)
    end

    it "prepends course files for unrecognized relative urls" do
      test_string = %(<a href="/relative/path/to/file">Linkage</a>)
-      expect(convert_and_replace(test_string)).to eq %(<a href="#{@path}file_contents/course%20files/relative/path/to/file">Linkage</a>)
+      expect(convert_exported_html(test_string)).to eq %(<a href="#{@path}file_contents/course%20files/relative/path/to/file">Linkage</a>)
      test_string = %(<a href="relative/path/to/file">Linkage</a>)
-      expect(convert_and_replace(test_string)).to eq %(<a href="#{@path}file_contents/course%20files/relative/path/to/file">Linkage</a>)
+      expect(convert_exported_html(test_string)).to eq %(<a href="#{@path}file_contents/course%20files/relative/path/to/file">Linkage</a>)
      test_string = %(<a href="relative/path/to/file%20with%20space.html">Linkage</a>)
-      expect(convert_and_replace(test_string)).to eq %(<a href="#{@path}file_contents/course%20files/relative/path/to/file%20with%20space.html">Linkage</a>)
+      expect(convert_exported_html(test_string)).to eq %(<a href="#{@path}file_contents/course%20files/relative/path/to/file%20with%20space.html">Linkage</a>)
    end

    it "preserves media comment links" do
@ -203,34 +197,34 @@ describe ImportedHtmlConverter do
        </p>
      HTML

-      expect(convert_and_replace(test_string)).to eq test_string
+      expect(convert_exported_html(test_string)).to eq test_string
    end

    it "handles and repair half broken media links" do
      test_string = %(<p><a href="/courses/#{@course.id}/file_contents/%24IMS_CC_FILEBASE%24/#" class="instructure_inline_media_comment video_comment" id="media_comment_0_l4l5n0wt">this is a media comment</a><br><br></p>)

-      expect(convert_and_replace(test_string)).to eq %(<p><a href="/media_objects/0_l4l5n0wt" class="instructure_inline_media_comment video_comment" id="media_comment_0_l4l5n0wt">this is a media comment</a><br><br></p>)
+      expect(convert_exported_html(test_string)).to eq %(<p><a href="/media_objects/0_l4l5n0wt" class="instructure_inline_media_comment video_comment" id="media_comment_0_l4l5n0wt">this is a media comment</a><br><br></p>)
    end

    it "preserves new RCE media iframes" do
      test_string = %(<iframe style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="video" src="/media_objects_iframe/0_l4l5n0wt?type=video" allowfullscreen="allowfullscreen" allow="fullscreen" data-media-id="0_l4l5n0wt"></iframe>)
-      expect(convert_and_replace(test_string)).to eq test_string
+      expect(convert_exported_html(test_string)).to eq test_string
    end

    it "handles and repair half broken new RCE media iframes" do
      test_string = %(<iframe style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="video" src="%24IMS_CC_FILEBASE%24/#" allowfullscreen="allowfullscreen" allow="fullscreen" data-media-id="m-abcde"></iframe>)
      repaired_string = %(<iframe style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="video" src="/media_objects_iframe/m-abcde?type=video" allowfullscreen="allowfullscreen" allow="fullscreen" data-media-id="m-abcde"></iframe>)
-      expect(convert_and_replace(test_string)).to eq repaired_string
+      expect(convert_exported_html(test_string)).to eq repaired_string
    end

    it "converts source tags to RCE media iframes" do
      test_string = %(<video style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="video" allowfullscreen="allowfullscreen" allow="fullscreen" data-media-id="0_l4l5n0wt"><source src="/media_objects_iframe/0_l4l5n0wt?type=video" data-media-id="0_l4l5n0wt" data-media-type="video"></video>)
      converted_string = %(<iframe style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="video" allowfullscreen="allowfullscreen" allow="fullscreen" data-media-id="0_l4l5n0wt" src="/media_objects_iframe/0_l4l5n0wt?type=video"></iframe>)
-      expect(convert_and_replace(test_string)).to eq converted_string
+      expect(convert_exported_html(test_string)).to eq converted_string

      test_string = %(<audio style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="audio" data-media-id="0_l4l5n0wt"><source src="/media_objects_iframe/0_l4l5n0wt?type=audio" data-media-id="0_l4l5n0wt" data-media-type="audio"></audio>)
      converted_string = %(<iframe style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="audio" data-media-id="0_l4l5n0wt" src="/media_objects_iframe/0_l4l5n0wt?type=audio"></iframe>)
-      expect(convert_and_replace(test_string)).to eq converted_string
+      expect(convert_exported_html(test_string)).to eq converted_string
    end

    it "converts source tags to RCE media attachment iframes" do
@ -238,11 +232,11 @@ describe ImportedHtmlConverter do

      test_string = %(<video style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="video" allowfullscreen="allowfullscreen" allow="fullscreen" data-media-id="0_l4l5n0wt"><source src="$CANVAS_OBJECT_REFERENCE$/media_attachments_iframe/#{att.migration_id}?type=video" data-media-id="0_l4l5n0wt" data-media-type="video"></video>)
      converted_string = %(<iframe style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="video" allowfullscreen="allowfullscreen" allow="fullscreen" data-media-id="0_l4l5n0wt" src="/media_attachments_iframe/#{att.id}?type=video"></iframe>)
-      expect(convert_and_replace(test_string)).to eq converted_string
+      expect(convert_exported_html(test_string)).to eq converted_string

      test_string = %(<audio  style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="audio" data-media-id="0_l4l5n0wt"><source src="$CANVAS_OBJECT_REFERENCE$/media_attachments_iframe/#{att.migration_id}?type=audio" data-media-id="0_l4l5n0wt" data-media-type="audio"></video>)
      converted_string = %(<iframe style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="audio" data-media-id="0_l4l5n0wt" src="/media_attachments_iframe/#{att.id}?type=audio"></iframe>)
-      expect(convert_and_replace(test_string)).to eq converted_string
+      expect(convert_exported_html(test_string)).to eq converted_string
    end

    it "converts source tags to RCE media attachment iframes when link is untranslated" do
@ -250,16 +244,16 @@ describe ImportedHtmlConverter do

      test_string = %(<video style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="video" allowfullscreen="allowfullscreen" allow="fullscreen" data-media-id="0_l4l5n0wt"><source src="/media_attachments_iframe/#{att.id}?type=video" data-media-id="0_l4l5n0wt" data-media-type="video"></video>)
      converted_string = %(<iframe style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="video" allowfullscreen="allowfullscreen" allow="fullscreen" data-media-id="0_l4l5n0wt" src="/media_attachments_iframe/#{att.id}?type=video"></iframe>)
-      expect(convert_and_replace(test_string)).to eq converted_string
+      expect(convert_exported_html(test_string)).to eq converted_string

      test_string = %(<audio  style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="audio" data-media-id="0_l4l5n0wt"><source src="/media_attachments_iframe/#{att.id}?type=audio" data-media-id="0_l4l5n0wt" data-media-type="audio"></video>)
      converted_string = %(<iframe style="width: 400px; height: 225px; display: inline-block;" title="this is a media comment" data-media-type="audio" data-media-id="0_l4l5n0wt" src="/media_attachments_iframe/#{att.id}?type=audio"></iframe>)
-      expect(convert_and_replace(test_string)).to eq converted_string
+      expect(convert_exported_html(test_string)).to eq converted_string
    end

    it "leaves source tags without data-media-id alone" do
      test_string = %(<video style="width: 400px; height: 225px; display: inline-block;" title="this is a non-canvas video" allowfullscreen="allowfullscreen" allow="fullscreen"><source src="http://www.example.com/video.mov"></video>)
-      expect(convert_and_replace(test_string)).to eq test_string
+      expect(convert_exported_html(test_string)).to eq test_string
    end

    it "only converts url params" do
@ -275,7 +269,7 @@ describe ImportedHtmlConverter do
        </object>
      HTML

-      expect(convert_and_replace(test_string)).to match_ignoring_whitespace(<<~HTML.strip)
+      expect(convert_exported_html(test_string)).to match_ignoring_whitespace(<<~HTML.strip)
        <object>
        <param name="controls" value="CONSOLE">
        <param name="controller" value="true">
@ -288,13 +282,13 @@ describe ImportedHtmlConverter do

    it "leaves an anchor tag alone" do
      test_string = '<p><a href="#anchor_ref">ref</a></p>'
-      expect(convert_and_replace(test_string)).to eq test_string
+      expect(convert_exported_html(test_string)).to eq test_string
    end

    it "converts base64 images to file links" do
      base64 = "R0lGODlhCQAJAIAAAICAgP///yH5BAEAAAEALAAAAAAJAAkAAAIQTGCZgGrc\nFIxvSuhwpsuFAgA7\n"
      test_string = "<p><img src=\"data:image/gif;base64,#{base64}\"></p>"
-      new_string = convert_and_replace(test_string)
+      new_string = convert_exported_html(test_string)
      attachment = Attachment.last
      expect(attachment.content_type).to eq "image/gif"
      expect(attachment.name).to eq "1d1fde3d669ed5c4fc68a49d643f140d.gif"