render images, audio and video in epubs

fixes CNVS-21798 Summary of changes: - Add files to epub as it's being built. - Update links to image files to be path specific to epub. - Update links to audio files to be epub-compatible `<audio>` tags. - Update links to video files to be epub-compatible `<video>` tags. test plan (requires console access): - Prereq: have a course with uploaded images, audio and video files, and reference that media in assignment bodies. - Make sure you have at least one completed content export in the db (can be triggered via console or via the content export UI). - open rails console - run the following commands: - - `ContentExport.include(CC::Exporter::Epub::Exportable)` - - `ContentExport.last.convert_to_epub` - This will output a path to the generated epub, something like: "/var/folders/7g/w8y0n7_j18v65h93xhmm9z4hxzj3r2/T/2052fbd0-860d-4114-9969-0adc7f5ecb79.Name of Course.epub" - Open up the folder of the file (assuming you're on Mac OSX) like so: `open -a finder /var/folders/7g/w8y0n7_j18v65h93xhmm9z4hxzj3r2/T`. - Find the file in that directory (in this case the file name is 2052fbd0-860d-4114-9969-0adc7f5ecb79.Name of Course.epub - Observe that images displayed inline work. - Observe that links to images work. - Observe that links to audio have been converted to audio players, and that the controls work. - Observe that links to video have been converted to video players, and that the controls work. - Repeat steps for discussion topics, wiki pages & quiz descriptions. Change-Id: I5a48c3300ccffc00230d94137a93731394ef0ebb Reviewed-on: https://gerrit.instructure.com/62842 Tested-by: Jenkins Reviewed-by: Matt Berns <mberns@instructure.com> QA-Review: Deepeeca Soundarrajan <dsoundarrajan@instructure.com> Product-Review: Cosme Salazar <cosme@instructure.com>
2015-09-09 10:49:24 -05:00 · 2015-09-09 10:49:24 -05:00 · c825ffc544
parent 6148c935a4
commit c825ffc544
16 changed files with 206 additions and 29 deletions
--- a/lib/cc/exporter/epub.rb
+++ b/lib/cc/exporter/epub.rb
@ -0,0 +1,5 @@
+module CC::Exporter
+  module Epub
+    FILE_PATH = 'media'
+  end
+end
--- a/lib/cc/exporter/epub/book.rb
+++ b/lib/cc/exporter/epub/book.rb
@ -1,12 +1,24 @@
 module CC::Exporter::Epub
  class Book
+    include CC::Exporter::Epub::Converters
+
    def initialize(content)
      @title = content.delete(:title)
+      @files = content.delete(:files)
      @content = content
    end
-    attr_reader :content, :title
+    attr_reader :content, :files, :title
+
+    def add_files
+      files.each do |file_data|
+        File.open(file_data[:full_path]) do |file|
+          epub.add_item(file_data[:local_path], file)
+        end
+      end
+    end

    def build
+      add_files
      content.each do |key, template|
        epub.add_ordered_item("#{key}.xhtml").
          add_content(StringIO.new(template.parse)).
--- a/lib/cc/exporter/epub/converters/assignment_epub_converter.rb
+++ b/lib/cc/exporter/epub/converters/assignment_epub_converter.rb
@ -12,7 +12,7 @@ module CC::Exporter::Epub::Converters
        html_path = File.join @unzipped_file_path, res.at_css('file[href$="html"]')['href']

        meta_node = open_file_xml(meta_path)
-        html_node = open_file(html_path)
+        html_node = convert_media_from_node!(open_file(html_path))

        next unless html_node

--- a/lib/cc/exporter/epub/converters/cartridge_converter.rb
+++ b/lib/cc/exporter/epub/converters/cartridge_converter.rb
@ -1,11 +1,14 @@
 module CC::Exporter::Epub::Converters
  class CartridgeConverter < Canvas::Migration::Migrator
+    include CC::CCHelper
    include Canvas::Migration::XMLHelper
    include WikiEpubConverter
    include AssignmentEpubConverter
    include TopicEpubConverter
    include QuizEpubConverter
    include ModuleEpubConverter
+    include FilesConverter
+    include MediaConverter

    MANIFEST_FILE = "imsmanifest.xml"

@ -17,10 +20,6 @@ module CC::Exporter::Epub::Converters
      @resource_nodes_for_flat_manifest = {}
    end

-    def export_directory
-      File.dirname(@archive.file)
-    end
-
    # exports the package into the intermediary json
    def export
      unzip_archive
@ -30,6 +29,8 @@ module CC::Exporter::Epub::Converters
      get_all_resources(@manifest)

      @course[:title] = get_node_val(@manifest, "string")
+      @course[:files] = convert_files
+
      set_progress(10)
      @course[:wikis] = convert_wikis
      set_progress(20)
@ -41,6 +42,7 @@ module CC::Exporter::Epub::Converters
      set_progress(50)
      @course[:modules] = convert_modules

+      # close up shop
      save_to_file
      set_progress(90)
      delete_unzipped_archive
--- a/lib/cc/exporter/epub/converters/files_converter.rb
+++ b/lib/cc/exporter/epub/converters/files_converter.rb
@ -0,0 +1,21 @@
+module CC::Exporter::Epub::Converters
+  module FilesConverter
+    include CC::CCHelper
+    include CC::Exporter
+
+    def convert_files
+      files = []
+      @manifest.css("resource[type=#{WEBCONTENT}][href^=#{WEB_RESOURCES_FOLDER}]").each do |res|
+        full_path = File.expand_path(get_full_path(res['href']))
+        local_path = res['href'].sub(WEB_RESOURCES_FOLDER, CC::Exporter::Epub::FILE_PATH)
+        files << {
+          migration_id: res['identifier'],
+          local_path: local_path,
+          file_name: File.basename(local_path),
+          full_path: full_path
+        }
+      end
+      files
+    end
+  end
+end
--- a/lib/cc/exporter/epub/converters/media_converter.rb
+++ b/lib/cc/exporter/epub/converters/media_converter.rb
@ -0,0 +1,42 @@
+module CC::Exporter::Epub::Converters
+  module MediaConverter
+    include CC::CCHelper
+
+    def convert_media_from_node!(html_node)
+      html_node.tap do |node|
+        convert_media_paths!(node)
+        convert_audio_tags!(node)
+        convert_video_tags!(node)
+      end
+    end
+
+    def convert_media_from_string!(html_string)
+      html_node = Nokogiri::HTML::DocumentFragment.parse(html_string)
+      convert_media_from_node!(html_node).to_s
+    end
+
+    def convert_media_paths!(html_node)
+      { a: 'href', img: 'src' }.each do |tag, attr|
+        html_node.search(tag).each do |match|
+          match[attr] = CGI.unescape(match[attr]).gsub(WEB_CONTENT_TOKEN, CC::Exporter::Epub::FILE_PATH)
+        end
+      end
+    end
+
+    def convert_audio_tags!(html_node)
+      html_node.search('a.instructure_audio_link').each do |audio_link|
+        audio_link.replace(<<-AUDIO_TAG)
+          <audio src="#{audio_link['href']}" controls="controls" />
+        AUDIO_TAG
+      end
+    end
+
+    def convert_video_tags!(html_node)
+      html_node.search('a.instructure_video_link').each do |video_link|
+        video_link.replace(<<-VIDEO_TAG)
+          <video src="#{video_link['href']}" controls="controls" />
+        VIDEO_TAG
+      end
+    end
+  end
+end
--- a/lib/cc/exporter/epub/converters/quiz_epub_converter.rb
+++ b/lib/cc/exporter/epub/converters/quiz_epub_converter.rb
@ -23,7 +23,7 @@ module CC::Exporter::Epub::Converters
      quiz_meta_data = open_file_xml(quiz_meta_link)

      quiz[:title] = get_node_val(quiz_meta_data, "title")
-      quiz[:description] = get_node_val(quiz_meta_data, "description")
+      quiz[:description] = convert_media_from_string!(get_node_val(quiz_meta_data, "description"))
      quiz[:due_at] = get_node_val(quiz_meta_data, "due_at")
      quiz[:lock_at] = get_node_val(quiz_meta_data, "lock_at")
      quiz[:unlock_at] = get_node_val(quiz_meta_data, "unlock_at")
--- a/lib/cc/exporter/epub/converters/topic_epub_converter.rb
+++ b/lib/cc/exporter/epub/converters/topic_epub_converter.rb
@ -9,7 +9,7 @@ module CC::Exporter::Epub::Converters
        cc_path = File.join @unzipped_file_path, res.at_css('file')['href']

        canvas_id = get_node_att(res, 'dependency', 'identifierref')
-        if canvas_id && meta_res = @manifest.at_css(%{resource[identifier="#{canvas_id}"]})
+        if canvas_id && (meta_res = @manifest.at_css(%{resource[identifier="#{canvas_id}"]}))
          canvas_path = File.join @unzipped_file_path, meta_res.at_css('file')['href']
          meta_node = open_file_xml(canvas_path)
        else
@ -25,7 +25,7 @@ module CC::Exporter::Epub::Converters

    def convert_topic(cc_doc, meta_doc)
      topic = {"resource_type" => :topics}
-      topic['description'] = get_node_val(cc_doc, 'text')
+      topic['description'] = convert_media_from_string!(get_node_val(cc_doc, 'text'))
      topic['title'] = get_node_val(cc_doc, 'title')
      if meta_doc
        topic['title'] = get_node_val(meta_doc, 'title')
@ -45,4 +45,4 @@ module CC::Exporter::Epub::Converters
      topic
    end
  end
-end
+end
--- a/lib/cc/exporter/epub/converters/wiki_epub_converter.rb
+++ b/lib/cc/exporter/epub/converters/wiki_epub_converter.rb
@ -21,10 +21,10 @@ module CC::Exporter::Epub::Converters
      title, body, meta = get_html_title_and_body_and_meta_fields(doc)
      wiki[:title] = title
      wiki[:front_page] = meta['front_page'] == 'true'
-      wiki[:text] = body
+      wiki[:text] = convert_media_from_string!(body)
      wiki[:identifier] = wiki_name
      wiki
    end

  end
-end
+end
--- a/lib/cc/exporter/epub/exporter.rb
+++ b/lib/cc/exporter/epub/exporter.rb
@ -30,7 +30,10 @@ module CC::Exporter::Epub
    end

    def templates
-      @_templates ||= {title: cartridge_json[:title]}.tap do |hash|
+      @_templates ||= {
+        title: cartridge_json[:title],
+        files: cartridge_json[:files]
+      }.tap do |hash|
        resources = sort_by_content ? [:assignments, :topics, :quizzes, :wikis] : [:modules]
        resources.each do |type|
          hash.merge!(type => create_template(type))
--- a/lib/cc/exporter/epub/template.rb
+++ b/lib/cc/exporter/epub/template.rb
@ -1,6 +1,7 @@
 module CC::Exporter::Epub
  class Template
    include TextHelper
+
    def initialize(content, base_template)
      @content = content[:resources] || []
      @base_template = base_template
@ -18,9 +19,7 @@ module CC::Exporter::Epub
    end

    def parse
-      Nokogiri::XML(build, &:noblanks).to_xml({
-        save_with: Nokogiri::XML::Node::SaveOptions::AS_XML | Nokogiri::XML::Node::SaveOptions::NO_DECLARATION
-      }).strip
+      Nokogiri::HTML(build, &:noblanks).to_xhtml.strip
    end

    def module_item(item)
@ -57,4 +56,4 @@ module CC::Exporter::Epub
      resources.find{|resource| resource[:identifier] == identifier}
    end
  end
-end
+end
--- a/lib/cc/exporter/epub/templates/assignment_epub_template.html.erb
+++ b/lib/cc/exporter/epub/templates/assignment_epub_template.html.erb
@ -14,4 +14,4 @@
 <hr/>
 <% if content_type_sorting %>
  <a href="assignments.xhtml">Back to Assignment Index</a>
-<% end %>
+<% end %>
--- a/lib/cc/exporter/epub/templates/content_sorting_template.html.erb
+++ b/lib/cc/exporter/epub/templates/content_sorting_template.html.erb
@ -1,8 +1,6 @@
-<?xml version='1.0' encoding='utf-8'?>
-<!DOCTYPE html PUBLIC '-//W3C//DTD XHTML 1.1//EN' 'http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd'>
-<html xmlns="http://www.w3.org/1999/xhtml">
+<!DOCTYPE html>
+<html>
 <head>
-  <meta http-equiv="Content-Type" content="application/xhtml+xml; charset=utf-8"/>
  <title><%= title %></title>
  <!-- embedded styling will be inserted here -->
 </head>
@ -19,4 +17,4 @@
      <div style="page-break-before:always;"></div>
    <% end %>
  </body>
-</html>
+</html>
--- a/lib/cc/exporter/epub/templates/module_sorting_template.html.erb
+++ b/lib/cc/exporter/epub/templates/module_sorting_template.html.erb
@ -1,8 +1,6 @@
-<?xml version='1.0' encoding='utf-8'?>
-<!DOCTYPE html PUBLIC '-//W3C//DTD XHTML 1.1//EN' 'http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd'>
-<html xmlns="http://www.w3.org/1999/xhtml">
+<!DOCTYPE html>
+<html>
 <head>
-  <meta http-equiv="Content-Type" content="application/xhtml+xml; charset=utf-8"/>
  <title>Course Modules</title>
 </head>
  <body>
@ -33,4 +31,4 @@
      <% end %>
    <% end %>
  </body>
-</html>
+</html>
--- a/lib/cc/exporter/epub/templates/topic_epub_template.html.erb
+++ b/lib/cc/exporter/epub/templates/topic_epub_template.html.erb
@ -17,4 +17,4 @@
 <hr/>
 <% if content_type_sorting %>
  <a href="topics.xhtml">Back to Discussion Topic Index</a>
-<% end %>
+<% end %>
--- a/spec/lib/cc/exporter/epub/converters/media_converter_spec.rb
+++ b/spec/lib/cc/exporter/epub/converters/media_converter_spec.rb
@ -0,0 +1,97 @@
+require File.expand_path(File.dirname(__FILE__) + '/../../../cc_spec_helper')
+
+describe "MediaConverter" do
+  class MediaConverterTest
+    include CC::Exporter::Epub::Converters::MediaConverter
+  end
+
+  describe "#convert_media_paths!" do
+    let(:doc) do
+      Nokogiri::HTML::DocumentFragment.parse(<<-HTML)
+        <div>
+          <a href="#{CGI.escape(MediaConverterTest::WEB_CONTENT_TOKEN)}/path/to/img.jpg">
+            Image Link
+          </a>
+          <img src="#{CGI.escape(MediaConverterTest::WEB_CONTENT_TOKEN)}/path/to/img.jpg" />
+        </div>
+      HTML
+    end
+    subject(:test_instance) { MediaConverterTest.new }
+
+    it "should update link hrefs containing WEB_CONTENT_TOKEN" do
+      expect(doc.search('a').all? do |element|
+        element['href'].match(CGI.escape(MediaConverterTest::WEB_CONTENT_TOKEN))
+      end).to be_truthy, 'precondition'
+
+      test_instance.convert_media_paths!(doc)
+
+      expect(doc.search('a').all? do |element|
+        element['href'].match(CC::Exporter::Epub::FILE_PATH)
+      end).to be_truthy
+
+      expect(doc.search('a').all? do |element|
+        element['href'].match(CGI.escape(MediaConverterTest::WEB_CONTENT_TOKEN))
+      end).to be_falsey
+    end
+
+    it "should update img srcs containing WEB_CONTENT_TOKEN" do
+      expect(doc.search('img').all? do |element|
+        element['src'].match(CGI.escape(MediaConverterTest::WEB_CONTENT_TOKEN))
+      end).to be_truthy, 'precondition'
+
+      test_instance.convert_media_paths!(doc)
+
+      expect(doc.search('img').all? do |element|
+        element['src'].match(CC::Exporter::Epub::FILE_PATH)
+      end).to be_truthy
+
+      expect(doc.search('img').all? do |element|
+        element['src'].match(CGI.escape(MediaConverterTest::WEB_CONTENT_TOKEN))
+      end).to be_falsey
+    end
+  end
+
+  describe "#convert_audio_tags!" do
+    let(:doc) do
+      Nokogiri::HTML::DocumentFragment.parse(<<-HTML)
+        <a href="#{CC::Exporter::Epub::FILE_PATH}/path/to/audio.mp3"
+          class="instructure_audio_link">
+          Audio Link
+        </a>
+      HTML
+    end
+    subject(:test_instance) { MediaConverterTest.new }
+
+    it "should change a tags to audio tags" do
+      expect(doc.search('a').any?).to be_truthy, 'precondition'
+      expect(doc.search('audio').empty?).to be_truthy, 'precondition'
+
+      test_instance.convert_audio_tags!(doc)
+
+      expect(doc.search('a').empty?).to be_truthy
+      expect(doc.search('audio').any?).to be_truthy
+    end
+  end
+
+  describe "#convert_video_tags!" do
+    let(:doc) do
+      Nokogiri::HTML::DocumentFragment.parse(<<-HTML)
+        <a href="#{CC::Exporter::Epub::FILE_PATH}/path/to/audio.mp3"
+          class="instructure_video_link">
+          Video Link
+        </a>
+      HTML
+    end
+    subject(:test_instance) { MediaConverterTest.new }
+
+    it "should change a tags to audio tags" do
+      expect(doc.search('a').any?).to be_truthy, 'precondition'
+      expect(doc.search('video').empty?).to be_truthy, 'precondition'
+
+      test_instance.convert_video_tags!(doc)
+
+      expect(doc.search('a').empty?).to be_truthy
+      expect(doc.search('video').any?).to be_truthy
+    end
+  end
+end