Merge pull request #23910 from open-craft/samuel/fix-get-video-subs

SE-2589 fix AttributeError: 'LibraryLocatorV2' object has no attribute 'make_asset_key'

common/lib/xmodule/xmodule/video_module/transcripts_utils.py

@@ -784,49 +784,6 @@ class VideoTranscriptsMixin(object):
         # to clean redundant language codes.
         return list(set(translations))
 
-    def get_transcript(self, transcripts, transcript_format='srt', lang=None):
-        """
-        Returns transcript, filename and MIME type.
-
-        transcripts (dict): A dict with all transcripts and a sub.
-
-        Raises:
-            - NotFoundError if cannot find transcript file in storage.
-            - ValueError if transcript file is empty or incorrect JSON.
-            - KeyError if transcript file has incorrect format.
-
-        If language is 'en', self.sub should be correct subtitles name.
-        If language is 'en', but if self.sub is not defined, this means that we
-        should search for video name in order to get proper transcript (old style courses).
-        If language is not 'en', give back transcript in proper language and format.
-        """
-        if not lang:
-            lang = self.get_default_transcript_language(transcripts)
-
-        sub, other_lang = transcripts["sub"], transcripts["transcripts"]
-        if lang == 'en':
-            if sub:  # HTML5 case and (Youtube case for new style videos)
-                transcript_name = sub
-            elif self.youtube_id_1_0:  # old courses
-                transcript_name = self.youtube_id_1_0
-            else:
-                log.debug("No subtitles for 'en' language")
-                raise ValueError
-
-            data = Transcript.asset(self.location, transcript_name, lang).data.decode('utf-8')
-            filename = u'{}.{}'.format(transcript_name, transcript_format)
-            content = Transcript.convert(data, 'sjson', transcript_format)
-        else:
-            data = Transcript.asset(self.location, None, None, other_lang[lang]).data.decode('utf-8')
-            filename = u'{}.{}'.format(os.path.splitext(other_lang[lang])[0], transcript_format)
-            content = Transcript.convert(data, 'srt', transcript_format)
-
-        if not content:
-            log.debug('no subtitles produced in get_transcript')
-            raise ValueError
-
-        return content, filename, Transcript.mime_types[transcript_format]
-
     def get_default_transcript_language(self, transcripts):
         """
         Returns the default transcript language for this video module.
@@ -910,7 +867,10 @@ def get_transcript_from_val(edx_video_id, lang=None, output_format=Transcript.SR
 
 def get_transcript_for_video(video_location, subs_id, file_name, language):
     """
-    Get video transcript from content store.
+    Get video transcript from content store. This is a lower level function and is used by
+    `get_transcript_from_contentstore`. Prefer that function instead where possible. If you
+    need to support getting transcripts from VAL or Blockstore as well, use the `get_transcript`
+    function instead.
 
     NOTE: Transcripts can be searched from content store by two ways:
     1. by an id(a.k.a subs_id) which will be used to construct transcript filename

common/lib/xmodule/xmodule/video_module/video_module.py

@@ -56,7 +56,7 @@ from .transcripts_utils import (
     VideoTranscriptsMixin,
     clean_video_id,
     get_html5_ids,
-    get_transcript_for_video,
+    get_transcript,
     subs_filename
 )
 from .video_handlers import VideoStudentViewHandlers, VideoStudioViewHandlers
@@ -593,12 +593,7 @@ class VideoBlock(
         possible_sub_ids = [self.sub, self.youtube_id_1_0] + get_html5_ids(self.html5_sources)
         for sub_id in possible_sub_ids:
             try:
-                get_transcript_for_video(
-                    self.location,
-                    subs_id=sub_id,
-                    file_name=sub_id,
-                    language=u'en'
-                )
+                _, sub_id, _ = get_transcript(self, lang=u'en', output_format=Transcript.TXT)
                 transcripts_info['transcripts'] = dict(transcripts_info['transcripts'], en=sub_id)
                 break
             except NotFoundError:
@@ -1029,10 +1024,7 @@ class VideoBlock(
         def _update_transcript_for_index(language=None):
             """ Find video transcript - if not found, don't update index """
             try:
-                transcripts = self.get_transcripts_info()
-                transcript = self.get_transcript(
-                    transcripts, transcript_format='txt', lang=language
-                )[0].replace("\n", " ")
+                transcript = get_transcript(self, lang=language, output_format=Transcript.TXT)[0].replace("\n", " ")
                 transcript_index_name = "transcript_{}".format(language if language else self.transcript_language)
                 video_body.update({transcript_index_name: transcript})
             except NotFoundError: