def test_migrated_transcripts_without_commit(self):
"""
Test migrating transcripts as a dry-run
"""
# check that transcripts do not exist
languages = api.get_available_transcript_languages(
self.video_descriptor.edx_video_id)
self.assertEqual(len(languages), 0)
self.assertFalse(
api.is_transcript_available(self.video_descriptor.edx_video_id,
'hr'))
self.assertFalse(
api.is_transcript_available(self.video_descriptor.edx_video_id,
'ge'))
# now call migrate_transcripts command and check the transcript availability
call_command('migrate_transcripts', '--course-id',
unicode(self.course.id))
# check that transcripts still do not exist
languages = api.get_available_transcript_languages(
self.video_descriptor.edx_video_id)
self.assertEqual(len(languages), 0)
self.assertFalse(
api.is_transcript_available(self.video_descriptor.edx_video_id,
'hr'))
self.assertFalse(
api.is_transcript_available(self.video_descriptor.edx_video_id,
'ge'))
def test_migrated_transcripts_count_with_commit(self):
"""
Test migrating transcripts with commit
"""
# check that transcript does not exist
languages = api.get_available_transcript_languages(
self.video_descriptor.edx_video_id)
self.assertEqual(len(languages), 0)
self.assertFalse(
api.is_transcript_available(self.video_descriptor.edx_video_id,
'hr'))
self.assertFalse(
api.is_transcript_available(self.video_descriptor.edx_video_id,
'ge'))
# now call migrate_transcripts command and check the transcript availability
call_command('migrate_transcripts', '--course-id',
unicode(self.course.id), '--commit')
languages = api.get_available_transcript_languages(
self.video_descriptor.edx_video_id)
self.assertEqual(len(languages), 2)
self.assertTrue(
api.is_transcript_available(self.video_descriptor.edx_video_id,
'hr'))
self.assertTrue(
api.is_transcript_available(self.video_descriptor.edx_video_id,
'ge'))
def _get_videos(course, pagination_conf=None):
"""
Retrieves the list of videos from VAL corresponding to this course.
"""
videos, pagination_context = get_videos_for_course(
unicode(course.id),
VideoSortField.created,
SortDirection.desc,
pagination_conf
)
videos = list(videos)
# This is required to see if edx video pipeline is enabled while converting the video status.
course_video_upload_token = course.video_upload_pipeline.get('course_video_upload_token')
# convert VAL's status to studio's Video Upload feature status.
for video in videos:
# If we are using "new video workflow" and status is `transcription_in_progress` then video encodes are ready.
# This is because Transcription starts once all the encodes are complete except for YT, but according to
# "new video workflow" YT is disabled as well as deprecated. So, Its precise to say that the Transcription
# starts once all the encodings are complete *for the new video workflow*.
is_video_encodes_ready = not course_video_upload_token and video['status'] == 'transcription_in_progress'
# Update with transcript languages
video['transcripts'] = get_available_transcript_languages(video_id=video['edx_video_id'])
# Transcription status should only be visible if 3rd party transcripts are pending.
video['transcription_status'] = (
StatusDisplayStrings.get(video['status'])
if not video['transcripts'] and is_video_encodes_ready else
''
)
# Convert the video status.
video['status'] = convert_video_status(video, is_video_encodes_ready)
return videos, pagination_context
def validate_transcript_upload_data(data, files):
"""
Validates video transcript file.
Arguments:
data: A request's data part.
files: A request's files part.
Returns:
None or String
If there is error returns error message otherwise None.
"""
error = None
# Validate the must have attributes - this error is unlikely to be faced by common users.
must_have_attrs = ['edx_video_id', 'language_code', 'new_language_code']
missing = [attr for attr in must_have_attrs if attr not in data]
if missing:
error = _(u'The following parameters are required: {missing}.').format(missing=', '.join(missing))
elif (
data['language_code'] != data['new_language_code'] and
data['new_language_code'] in get_available_transcript_languages(video_id=data['edx_video_id'])
):
error = _(u'A transcript with the "{language_code}" language code already exists.'.format(
language_code=data['new_language_code']
))
elif 'file' not in files:
error = _(u'A transcript file is required.')
return error
def validate_transcript_upload_data(data, files):
"""
Validates video transcript file.
Arguments:
data: A request's data part.
files: A request's files part.
Returns:
None or String
If there is error returns error message otherwise None.
"""
error = None
# Validate the must have attributes - this error is unlikely to be faced by common users.
must_have_attrs = ['edx_video_id', 'language_code', 'new_language_code']
missing = [attr for attr in must_have_attrs if attr not in data]
if missing:
error = _(u'The following parameters are required: {missing}.').format(missing=', '.join(missing))
elif (
data['language_code'] != data['new_language_code'] and
data['new_language_code'] in get_available_transcript_languages(video_id=data['edx_video_id'])
):
error = _(u'A transcript with the "{language_code}" language code already exists.'.format(
language_code=data['new_language_code']
))
elif 'file' not in files:
error = _(u'A transcript file is required.')
return error
def _get_videos(course):
"""
Retrieves the list of videos from VAL corresponding to this course.
"""
videos = list(get_videos_for_course(unicode(course.id), VideoSortField.created, SortDirection.desc))
# This is required to see if edx video pipeline is enabled while converting the video status.
course_video_upload_token = course.video_upload_pipeline.get('course_video_upload_token')
# convert VAL's status to studio's Video Upload feature status.
for video in videos:
# If we are using "new video workflow" and status is `transcription_in_progress` then video encodes are ready.
# This is because Transcription starts once all the encodes are complete except for YT, but according to
# "new video workflow" YT is disabled as well as deprecated. So, Its precise to say that the Transcription
# starts once all the encodings are complete *for the new video workflow*.
is_video_encodes_ready = not course_video_upload_token and video['status'] == 'transcription_in_progress'
# Update with transcript languages
video['transcripts'] = get_available_transcript_languages(video_id=video['edx_video_id'])
# Transcription status should only be visible if 3rd party transcripts are pending.
video['transcription_status'] = (
StatusDisplayStrings.get(video['status'])
if not video['transcripts'] and is_video_encodes_ready else
''
)
# Convert the video status.
video['status'] = convert_video_status(video, is_video_encodes_ready)
return videos
def test_migrated_transcripts_count_with_commit(self):
"""
Test migrating transcripts with commit
"""
# check that transcript does not exist
languages = api.get_available_transcript_languages(self.video_descriptor.edx_video_id)
self.assertEqual(len(languages), 0)
self.assertFalse(api.is_transcript_available(self.video_descriptor.edx_video_id, 'hr'))
self.assertFalse(api.is_transcript_available(self.video_descriptor.edx_video_id, 'ge'))
# now call migrate_transcripts command and check the transcript availability
call_command('migrate_transcripts', '--course-id', unicode(self.course.id), '--commit')
languages = api.get_available_transcript_languages(self.video_descriptor.edx_video_id)
self.assertEqual(len(languages), 2)
self.assertTrue(api.is_transcript_available(self.video_descriptor.edx_video_id, 'hr'))
self.assertTrue(api.is_transcript_available(self.video_descriptor.edx_video_id, 'ge'))
def test_migrated_transcripts_without_commit(self):
"""
Test migrating transcripts as a dry-run
"""
# check that transcripts do not exist
languages = api.get_available_transcript_languages(self.video_descriptor.edx_video_id)
self.assertEqual(len(languages), 0)
self.assertFalse(api.is_transcript_available(self.video_descriptor.edx_video_id, 'hr'))
self.assertFalse(api.is_transcript_available(self.video_descriptor.edx_video_id, 'ge'))
# now call migrate_transcripts command and check the transcript availability
call_command('migrate_transcripts', '--course-id', unicode(self.course.id))
# check that transcripts still do not exist
languages = api.get_available_transcript_languages(self.video_descriptor.edx_video_id)
self.assertEqual(len(languages), 0)
self.assertFalse(api.is_transcript_available(self.video_descriptor.edx_video_id, 'hr'))
self.assertFalse(api.is_transcript_available(self.video_descriptor.edx_video_id, 'ge'))
def _get_videos(course):
"""
Retrieves the list of videos from VAL corresponding to this course.
"""
videos = list(get_videos_for_course(unicode(course.id), VideoSortField.created, SortDirection.desc))
# convert VAL's status to studio's Video Upload feature status.
for video in videos:
video["status"] = convert_video_status(video)
video['transcripts'] = get_available_transcript_languages(video_id=video['edx_video_id'])
return videos
def _get_videos(course):
"""
Retrieves the list of videos from VAL corresponding to this course.
"""
videos = list(get_videos_for_course(unicode(course.id), VideoSortField.created, SortDirection.desc))
# convert VAL's status to studio's Video Upload feature status.
for video in videos:
video["status"] = convert_video_status(video)
video['transcripts'] = get_available_transcript_languages(video_id=video['edx_video_id'])
return videos
def get_available_transcript_languages(edx_video_id):
"""
Gets available transcript languages for a video.
Arguments:
edx_video_id(unicode): edx-val's video identifier
Returns:
A list containing distinct transcript language codes against all the passed video ids.
"""
available_languages = []
edx_video_id = clean_video_id(edx_video_id)
if edxval_api and edx_video_id:
available_languages = edxval_api.get_available_transcript_languages(video_id=edx_video_id)
return available_languages
def get_available_transcript_languages(edx_video_id):
"""
Gets available transcript languages for a video.
Arguments:
edx_video_id(unicode): edx-val's video identifier
Returns:
A list containing distinct transcript language codes against all the passed video ids.
"""
available_languages = []
edx_video_id = clean_video_id(edx_video_id)
if edxval_api and edx_video_id:
available_languages = edxval_api.get_available_transcript_languages(video_id=edx_video_id)
return available_languages
def get_available_transcript_languages(edx_video_id, youtube_id_1_0, html5_sources):
"""
Gets available transcript languages from edx-val.
Arguments:
edx_video_id(unicode): edx-val's video identifier
youtube_id_1_0(unicode): A youtube source identifier
html5_sources(list): A list containing html5 sources
Returns:
A list containing distinct transcript language codes against all the passed video ids.
"""
available_languages = []
if edxval_api:
__, video_candidate_ids = get_video_ids_info(edx_video_id, youtube_id_1_0, html5_sources)
available_languages = edxval_api.get_available_transcript_languages(video_candidate_ids)
return available_languages
def get_available_transcript_languages(edx_video_id, youtube_id_1_0, html5_sources):
"""
Gets available transcript languages from edx-val.
Arguments:
edx_video_id(unicode): edx-val's video identifier
youtube_id_1_0(unicode): A youtube source identifier
html5_sources(list): A list containing html5 sources
Returns:
A list containing distinct transcript language codes against all the passed video ids.
"""
available_languages = []
if edxval_api:
__, video_candidate_ids = get_video_ids_info(edx_video_id, youtube_id_1_0, html5_sources)
available_languages = edxval_api.get_available_transcript_languages(video_candidate_ids)
return available_languages
def _get_videos(course, pagination_conf=None):
"""
Retrieves the list of videos from VAL corresponding to this course.
"""
videos, pagination_context = get_videos_for_course(
six.text_type(course.id),
VideoSortField.created,
SortDirection.desc,
pagination_conf
)
videos = list(videos)
# This is required to see if edx video pipeline is enabled while converting the video status.
course_video_upload_token = course.video_upload_pipeline.get('course_video_upload_token')
# TODO: add 'transcript_ready' when we have moved to VEM to keep transcript and encode status separate
transcription_statuses = ['partial_failure', 'transcription_in_progress']
# convert VAL's status to studio's Video Upload feature status.
for video in videos:
# If we are using "new video workflow" and status is `transcription_in_progress` then video encodes are ready.
# This is because Transcription starts once all the encodes are complete except for YT, but according to
# "new video workflow" YT is disabled as well as deprecated. So, Its precise to say that the Transcription
# starts once all the encodings are complete *for the new video workflow*.
# If the video status is 'partial_failure', it means during the transcription flow, some transcription jobs
# failed. As mentioned, transcript jobs start only when the encodes have finished
is_video_encodes_ready = not course_video_upload_token and (video['status'] in transcription_statuses)
# Update with transcript languages
video['transcripts'] = get_available_transcript_languages(video_id=video['edx_video_id'])
# Transcription status should only be visible if 3rd party transcripts are pending.
# TODO: change logic to separate transcript status from video status
video['transcription_status'] = (
StatusDisplayStrings.get(video['status'])
if not video['transcripts'] and is_video_encodes_ready else
''
)
# Convert the video status.
video['status'] = convert_video_status(video, is_video_encodes_ready)
return videos, pagination_context
def _get_videos(course):
"""
Retrieves the list of videos from VAL corresponding to this course.
"""
is_video_transcript_enabled = VideoTranscriptEnabledFlag.feature_enabled(
course.id)
videos = list(
get_videos_for_course(unicode(course.id), VideoSortField.created,
SortDirection.desc))
# convert VAL's status to studio's Video Upload feature status.
for video in videos:
video["status"] = convert_video_status(video)
if is_video_transcript_enabled:
all_languages = get_all_transcript_languages()
video['transcripts'] = {
lang_code: all_languages[lang_code]
for lang_code in get_available_transcript_languages(
[video['edx_video_id']])
}
return videos