Merge pull request #17696 from edx/ammar/update-transcripts-util

update get_transcript util

cms/djangoapps/contentstore/tests/test_transcripts_utils.py

@@ -760,7 +760,11 @@ class TestGetTranscript(SharedModuleStoreTestCase):
 
         self.user = UserFactory.create()
         self.vertical = ItemFactory.create(category='vertical', parent_location=self.course.location)
-        self.video = ItemFactory.create(category='video', parent_location=self.vertical.location)
+        self.video = ItemFactory.create(
+            category='video',
+            parent_location=self.vertical.location,
+            edx_video_id=u'1234-5678-90'
+        )
 
     def create_transcript(self, subs_id, language=u'en', filename='video.srt'):
         """
@@ -774,7 +778,8 @@ class TestGetTranscript(SharedModuleStoreTestCase):
             category='video',
             parent_location=self.vertical.location,
             sub=subs_id,
-            transcripts=transcripts
+            transcripts=transcripts,
+            edx_video_id=u'1234-5678-90'
         )
 
         if subs_id:
@@ -823,7 +828,10 @@ class TestGetTranscript(SharedModuleStoreTestCase):
         Verify that `NotFoundError` exception is raised when transcript is not found in both the content store and val.
         """
         with self.assertRaises(NotFoundError):
-            transcripts_utils.get_transcript(self.course.id, self.video.location.block_id, lang=lang)
+            transcripts_utils.get_transcript(
+                self.video,
+                lang=lang
+            )
 
     @ddt.data(
         {
@@ -845,8 +853,7 @@ class TestGetTranscript(SharedModuleStoreTestCase):
         self.upload_file(self.create_srt_file(self.subs_srt), self.video.location, filename)
         self.create_transcript(subs_id, language, filename)
         content, filename, mimetype = transcripts_utils.get_transcript(
-            self.course.id,
-            self.video.location.block_id,
+            self.video,
             language
         )
 
@@ -861,8 +868,7 @@ class TestGetTranscript(SharedModuleStoreTestCase):
         language = u'ur'
         self.create_transcript(self.subs_id, language)
         content, filename, mimetype = transcripts_utils.get_transcript(
-            self.course.id,
-            self.video.location.block_id,
+            self.video,
             language,
             output_format=transcripts_utils.Transcript.SJSON
         )
@@ -871,10 +877,6 @@ class TestGetTranscript(SharedModuleStoreTestCase):
         self.assertEqual(filename, 'ur_video_101.sjson')
         self.assertEqual(mimetype, self.sjson_mime_type)
 
-    @patch(
-        'openedx.core.djangoapps.video_config.models.VideoTranscriptEnabledFlag.feature_enabled',
-        Mock(return_value=True),
-    )
     @patch('xmodule.video_module.transcripts_utils.get_video_transcript_content')
     def test_get_transcript_from_val(self, mock_get_video_transcript_content):
         """
@@ -886,8 +888,7 @@ class TestGetTranscript(SharedModuleStoreTestCase):
         }
 
         content, filename, mimetype = transcripts_utils.get_transcript(
-            self.course.id,
-            self.video.location.block_id,
+            self.video,
         )
         self.assertEqual(content, self.subs_srt)
         self.assertEqual(filename, 'edx.srt')
@@ -899,8 +900,7 @@ class TestGetTranscript(SharedModuleStoreTestCase):
         """
         with self.assertRaises(NotFoundError) as invalid_format_exception:
             transcripts_utils.get_transcript(
-                self.course.id,
-                self.video.location.block_id,
+                self.video,
                 'ur',
                 output_format='mpeg'
             )
@@ -917,8 +917,7 @@ class TestGetTranscript(SharedModuleStoreTestCase):
 
         with self.assertRaises(NotFoundError) as no_content_exception:
             transcripts_utils.get_transcript(
-                self.course.id,
-                self.video.location.block_id,
+                self.video,
                 'ur'
             )
 
@@ -933,8 +932,7 @@ class TestGetTranscript(SharedModuleStoreTestCase):
         self.store.update_item(self.video, self.user.id)
         with self.assertRaises(NotFoundError) as no_en_transcript_exception:
             transcripts_utils.get_transcript(
-                self.course.id,
-                self.video.location.block_id,
+                self.video,
                 'en'
             )
 

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

@@ -904,6 +904,8 @@ def get_transcript_for_video(video_location, subs_id, file_name, language):
         tuple containing transcript input_format, basename, content
     """
     try:
+        if subs_id is None:
+            raise NotFoundError
         content = Transcript.asset(video_location, subs_id, language).data
         base_name = subs_id
         input_format = Transcript.SJSON
@@ -915,7 +917,7 @@ def get_transcript_for_video(video_location, subs_id, file_name, language):
     return input_format, base_name, content
 
 
-def get_transcript_from_contentstore(video, language, output_format, youtube_id=None, is_bumper=False):
+def get_transcript_from_contentstore(video, language, output_format, transcripts_info, youtube_id=None):
     """
     Get video transcript from content store.
 
@@ -923,8 +925,8 @@ def get_transcript_from_contentstore(video, language, output_format, youtube_id=
         video (Video Descriptor): Video descriptor
         language (unicode): transcript language
         output_format (unicode): transcript output format
+        transcripts_info (dict): transcript info for a video
         youtube_id (unicode): youtube video id
-        is_bumper (bool): indicates bumper video
 
     Returns:
         tuple containing content, filename, mimetype
@@ -932,7 +934,6 @@ def get_transcript_from_contentstore(video, language, output_format, youtube_id=
     if output_format not in (Transcript.SRT, Transcript.SJSON, Transcript.TXT):
         raise NotFoundError('Invalid transcript format `{output_format}`'.format(output_format=output_format))
 
-    transcripts_info = video.get_transcripts_info(is_bumper=is_bumper)
     sub, other_languages = transcripts_info['sub'], transcripts_info['transcripts']
     transcripts = dict(other_languages)
 
@@ -949,8 +950,8 @@ def get_transcript_from_contentstore(video, language, output_format, youtube_id=
     try:
         input_format, base_name, transcript_content = get_transcript_for_video(
             video.location,
-            subs_id=transcripts['en'],
-            file_name=language and transcripts[language],
+            subs_id=transcripts.get('en'),
+            file_name=transcripts[language],
             language=language
         )
     except KeyError:
@@ -973,29 +974,33 @@ def get_transcript_from_contentstore(video, language, output_format, youtube_id=
     return transcript_content, transcript_name, Transcript.mime_types[output_format]
 
 
-def get_transcript(course_id, block_id, lang=None, output_format=Transcript.SRT, is_bumper=False):
+def get_transcript(video, lang=None, output_format=Transcript.SRT, youtube_id=None):
     """
     Get video transcript from edx-val or content store.
 
     Arguments:
-        course_id (CourseLocator): course identifier
-        block_id (unicode): a unique identifier for an item in modulestore
+        video (Video Descriptor): Video Descriptor
         lang (unicode): transcript language
         output_format (unicode): transcript output format
-        is_bumper (bool): indicates bumper video
+        youtube_id (unicode): youtube video id
 
     Returns:
         tuple containing content, filename, mimetype
     """
-    usage_key = BlockUsageLocator(course_id, block_type='video', block_id=block_id)
-    video_descriptor = modulestore().get_item(usage_key)
+    transcripts_info = video.get_transcripts_info()
+    if not lang:
+        lang = video.get_default_transcript_language(transcripts_info)
 
     try:
-        return get_transcript_from_val(video_descriptor.edx_video_id, lang, output_format)
+        edx_video_id = clean_video_id(video.edx_video_id)
+        if not edx_video_id:
+            raise NotFoundError
+        return get_transcript_from_val(edx_video_id, lang, output_format)
     except NotFoundError:
         return get_transcript_from_contentstore(
-            video_descriptor,
+            video,
             lang,
+            youtube_id=youtube_id,
             output_format=output_format,
-            is_bumper=is_bumper
+            transcripts_info=transcripts_info
         )