Make transcripts indexable for LMS search (#294)

* Make transcripts indexable for LMS search

* Fix javascript unit tests

video_xblock/mixins.py

@@ -99,6 +99,18 @@ def convert_caps_to_vtt(caps):
                 return WebVTTWriter().write(reader().read(caps))
         return u''
 
+    @staticmethod
+    def vtt_to_text(vtt_content):
+        """
+        Utility method to extract text from WebVTT format transcript.
+        """
+        text_lines = []
+        for line in vtt_content.splitlines():
+            if '-->' in line or line == '':
+                continue
+            text_lines.append(line)
+        return ' '.join(text_lines)
+
     def route_transcripts(self):
         """
         Re-route transcripts to appropriate handler.

video_xblock/static/video-xblock-karma.conf.js

@@ -15,15 +15,15 @@ module.exports = function (config) {
 
         // list of files / patterns to load in the browser
         files: [
-            'https://ajax.googleapis.com/ajax/libs/jquery/3.1.0/jquery.min.js',
+            'vendor/js/jquery.min.js',
             'vendor/js/video.min.js',
             'js/spec/test-context.js',
             'js/base.js',
             'js/runtime-handlers.js',
             'js/studio-edit/utils.js',
-            'js/studio-edit/transcripts-manual-upload.js',
             'js/spec/*spec.js',
-            'js/spec/studio-edit/*spec.js'
+            'js/spec/studio-edit/*spec.js',
+            'js/studio-edit/transcripts-manual-upload.js'
         ],
         plugins: [
             'karma-chrome-launcher',

video_xblock/video_xblock.py

@@ -834,3 +834,48 @@ def get_enabled_managed_transcripts(self):
         except ValueError:
             log.exception("JSON parser can't handle 'self.transcripts' field value: {}".format(self.transcripts))
             return []
+
+    def index_dictionary(self):
+        """
+        Part of edx-platform search index API.
+
+        Is invoked during course [re]index operation.
+        Takes enabled transcripts' content and puts it to search index.
+        """
+        xblock_body = super(VideoXBlock, self).index_dictionary()
+        video_body = {"display_name": self.display_name}
+
+        content = None
+        enabled_transcripts = self.route_transcripts()
+        for transcript in enabled_transcripts:
+            asset_file_name = transcript[u'url'].split('@')[-1]
+            try:
+                if transcript['source'] in [TranscriptSource.MANUAL, TranscriptSource.DEFAULT]:
+                    asset_location = self.static_content.compute_location(self.course_key, asset_file_name)
+                    asset = self.contentstore().find(asset_location)  # pylint: disable=not-callable
+                    content = asset.data
+                elif transcript['source'] == TranscriptSource.THREE_PLAY_MEDIA:
+                    external_transcript = self.fetch_single_3pm_translation({
+                        'id': transcript['id'], 'language_id': transcript['lang_id']
+                    })
+                    content = external_transcript and external_transcript.content
+            except IOError:
+                log.exception("Transcript indexing failure: can't fetch external transcript[{}]".format(transcript))
+            except (ValueError, KeyError, TypeError, AttributeError):
+                log.exception(
+                    "Transcript indexing failure: can't parse transcript for indexing: [{}]".format(transcript)
+                )
+            else:
+                if content:
+                    content_ = self.vtt_to_text(content)
+                    video_body.update({transcript[u'lang']: content_})
+            finally:
+                content = None
+
+        if "content" in xblock_body:
+            xblock_body["content"].update(video_body)
+        else:
+            xblock_body["content"] = video_body
+        xblock_body["content_type"] = "Video"
+
+        return xblock_body