def save_transcript_to_storage(command_run, edx_video_id, language_code,
transcript_content, file_format, force_update):
"""
Pushes a given transcript's data to django storage.
Arguments:
command_run: A positive integer indicating the current run
edx_video_id: video ID
language_code: language code
transcript_content: content of the transcript
file_format: format of the transcript file
force_update: tells whether it needs to perform force update in
case of an existing transcript for the given video.
"""
transcript_present = is_transcript_available(video_id=edx_video_id,
language_code=language_code)
if transcript_present and force_update:
create_or_update_video_transcript(edx_video_id, language_code,
dict({'file_format': file_format}),
ContentFile(transcript_content))
elif not transcript_present:
create_video_transcript(edx_video_id, language_code, file_format,
ContentFile(transcript_content))
else:
LOGGER.info(
"[%s] [run=%s] [do-not-override-existing-transcript] [edx_video_id=%s] [language_code=%s]",
MIGRATION_LOGS_PREFIX, command_run, edx_video_id, language_code)
def save_transcript_to_storage(command_run, edx_video_id, language_code, transcript_content, file_format, force_update):
"""
Pushes a given transcript's data to django storage.
Arguments:
command_run: A positive integer indicating the current run
edx_video_id: video ID
language_code: language code
transcript_content: content of the transcript
file_format: format of the transcript file
force_update: tells whether it needs to perform force update in
case of an existing transcript for the given video.
"""
transcript_present = is_transcript_available(video_id=edx_video_id, language_code=language_code)
if transcript_present and force_update:
create_or_update_video_transcript(
edx_video_id,
language_code,
dict({'file_format': file_format}),
ContentFile(transcript_content)
)
elif not transcript_present:
create_video_transcript(
edx_video_id,
language_code,
file_format,
ContentFile(transcript_content)
)
else:
LOGGER.info(
"[%s] [run=%s] [do-not-override-existing-transcript] [edx_video_id=%s] [language_code=%s]",
MIGRATION_LOGS_PREFIX, command_run, edx_video_id, language_code
)
def test_transcript_delete_handler(self, is_staff, is_course_staff):
"""
Tests that transcript delete handler works as expected with combinations of staff and course's staff.
"""
# Setup user's roles
self.user.is_staff = is_staff
self.user.save()
course_staff_role = CourseStaffRole(self.course.id)
if is_course_staff:
course_staff_role.add_users(self.user)
else:
course_staff_role.remove_users(self.user)
# Assert the user role
self.assertEqual(self.user.is_staff, is_staff)
self.assertEqual(CourseStaffRole(self.course.id).has_user(self.user), is_course_staff)
video_id, language_code = u'1234', u'en'
# Create a real transcript in VAL.
api.create_or_update_video_transcript(
video_id=video_id,
language_code=language_code,
metadata={'file_format': 'srt'}
)
# Make request to transcript deletion handler
response = self.client.delete(self.get_url_for_course_key(
self.course.id,
edx_video_id=video_id,
language_code=language_code
))
self.assertEqual(response.status_code, 200)
self.assertFalse(api.get_video_transcript_data(video_id=video_id, language_code=language_code))
def save_transcript_to_storage(edx_video_id, language_code, transcript_content, file_format, force_update):
"""
Pushes a given transcript's data to django storage.
Arguments:
edx_video_id: video ID
language_code: language code
transcript_content: content of the transcript
file_format: format of the transcript file
force_update: tells whether it needs to perform force update in
case of an existing transcript for the given video.
"""
transcript_present = is_transcript_available(video_id=edx_video_id, language_code=language_code)
if transcript_present and force_update:
create_or_update_video_transcript(
edx_video_id,
language_code,
dict({'file_format': file_format}),
ContentFile(transcript_content)
)
elif not transcript_present:
create_video_transcript(
edx_video_id,
language_code,
file_format,
ContentFile(transcript_content)
)
def post(self, request):
"""
Creates a video transcript instance with the given information.
Arguments:
request: A WSGI request.
"""
attrs = ('video_id', 'name', 'language_code', 'provider',
'file_format')
missing = [attr for attr in attrs if attr not in request.data]
if missing:
LOGGER.warn('[VAL] Required transcript params are missing. %s',
' and '.join(missing))
return Response(
status=status.HTTP_400_BAD_REQUEST,
data=dict(message=u'{missing} must be specified.'.format(
missing=' and '.join(missing))))
video_id = request.data['video_id']
language_code = request.data['language_code']
transcript_name = request.data['name']
provider = request.data['provider']
file_format = request.data['file_format']
supported_formats = sorted(dict(TranscriptFormat.CHOICES).keys())
if file_format not in supported_formats:
message = (
u'"{format}" transcript file type is not supported. Supported formats are "{supported_formats}"'
).format(format=file_format, supported_formats=supported_formats)
return Response(status=status.HTTP_400_BAD_REQUEST,
data={'message': message})
supported_providers = sorted(
dict(TranscriptProviderType.CHOICES).keys())
if provider not in supported_providers:
message = (
u'"{provider}" provider is not supported. Supported transcription providers are "{supported_providers}"'
).format(provider=provider,
supported_providers=supported_providers)
return Response(status=status.HTTP_400_BAD_REQUEST,
data={'message': message})
transcript = VideoTranscript.get_or_none(video_id, language_code)
if transcript is None:
create_or_update_video_transcript(video_id,
language_code,
metadata={
'provider': provider,
'file_name': transcript_name,
'file_format': file_format
})
response = Response(status=status.HTTP_200_OK)
else:
message = (
u'Can not override existing transcript for video "{video_id}" and language code "{language}".'
).format(video_id=video_id, language=language_code)
response = Response(status=status.HTTP_400_BAD_REQUEST,
data={'message': message})
return response
def save_video_transcript(edx_video_id, input_format, transcript_content,
language_code):
"""
Saves a video transcript to the VAL and its content to the configured django storage(DS).
Arguments:
edx_video_id: A Video ID to associate the transcript.
input_format: Input transcript format for content being passed.
transcript_content: Content of the transcript file
language_code: transcript language code
Returns:
A boolean indicating whether the transcript was saved or not.
"""
try:
# Convert the transcript into the 'sjson' and upload it to
# configured transcript storage. For example, S3.
sjson_subs = Transcript.convert(content=transcript_content,
input_format=input_format,
output_format=Transcript.SJSON)
create_or_update_video_transcript(
video_id=edx_video_id,
language_code=language_code,
metadata={
'provider': TranscriptProvider.CUSTOM,
'file_format': Transcript.SJSON,
'language_code': language_code
},
file_data=ContentFile(sjson_subs),
)
result = True
except (TranscriptsGenerationException, UnicodeDecodeError):
result = False
return result
def test_transcript_delete_handler(self, is_staff, is_course_staff):
"""
Tests that transcript delete handler works as expected with combinations of staff and course's staff.
"""
# Setup user's roles
self.user.is_staff = is_staff
self.user.save()
course_staff_role = CourseStaffRole(self.course.id)
if is_course_staff:
course_staff_role.add_users(self.user)
else:
course_staff_role.remove_users(self.user)
# Assert the user role
self.assertEqual(self.user.is_staff, is_staff)
self.assertEqual(
CourseStaffRole(self.course.id).has_user(self.user),
is_course_staff)
video_id, language_code = u'1234', u'en'
# Create a real transcript in VAL.
api.create_or_update_video_transcript(video_id=video_id,
language_code=language_code,
metadata={'file_format': 'srt'})
# Make request to transcript deletion handler
response = self.client.delete(
self.get_url_for_course_key(self.course.id,
edx_video_id=video_id,
language_code=language_code))
self.assertEqual(response.status_code, 200)
self.assertFalse(
api.get_video_transcript_data(video_id=video_id,
language_code=language_code))
def transcript_upload_handler(request, course_key_string):
"""
View to upload a transcript file.
Arguments:
request: A WSGI request object
course_key_string: Course key identifying a course
Transcript file, edx video id and transcript language are required.
Transcript file should be in SRT(SubRip) format.
Returns
- A 400 if any of the validation fails
- A 404 if the corresponding feature flag is disabled
- A 200 if transcript has been uploaded successfully
"""
# Check whether the feature is available for this course.
course_key = CourseKey.from_string(course_key_string)
if not VideoTranscriptEnabledFlag.feature_enabled(course_key):
return HttpResponseNotFound()
error = validate_transcript_upload_data(data=request.POST, files=request.FILES)
if error:
response = JsonResponse({'error': error}, status=400)
else:
edx_video_id = request.POST['edx_video_id']
language_code = request.POST['language_code']
new_language_code = request.POST['new_language_code']
transcript_file = request.FILES['file']
try:
# Convert SRT transcript into an SJSON format
# and upload it to S3.
sjson_subs = Transcript.convert(
content=transcript_file.read(),
input_format=Transcript.SRT,
output_format=Transcript.SJSON
)
create_or_update_video_transcript(
video_id=edx_video_id,
language_code=language_code,
metadata={
'provider': TranscriptProvider.CUSTOM,
'file_format': Transcript.SJSON,
'language_code': new_language_code
},
file_data=ContentFile(sjson_subs),
)
response = JsonResponse(status=201)
except (TranscriptsGenerationException, UnicodeDecodeError):
response = JsonResponse(
{'error': _(u'There is a problem with this transcript file. Try to upload a different file.')},
status=400
)
return response
def post(self, request):
"""
Creates a video transcript instance with the given information.
Arguments:
request: A WSGI request.
"""
attrs = ('video_id', 'name', 'language_code', 'provider', 'file_format')
missing = [attr for attr in attrs if attr not in request.data]
if missing:
LOGGER.warn(
'[VAL] Required transcript params are missing. %s', ' and '.join(missing)
)
return Response(
status=status.HTTP_400_BAD_REQUEST,
data=dict(message=u'{missing} must be specified.'.format(missing=' and '.join(missing)))
)
video_id = request.data['video_id']
language_code = request.data['language_code']
transcript_name = request.data['name']
provider = request.data['provider']
file_format = request.data['file_format']
supported_formats = sorted(dict(TranscriptFormat.CHOICES).keys())
if file_format not in supported_formats:
message = (
u'"{format}" transcript file type is not supported. Supported formats are "{supported_formats}"'
).format(format=file_format, supported_formats=supported_formats)
return Response(status=status.HTTP_400_BAD_REQUEST, data={'message': message})
supported_providers = sorted(dict(TranscriptProviderType.CHOICES).keys())
if provider not in supported_providers:
message = (
u'"{provider}" provider is not supported. Supported transcription providers are "{supported_providers}"'
).format(provider=provider, supported_providers=supported_providers)
return Response(status=status.HTTP_400_BAD_REQUEST, data={'message': message})
transcript = VideoTranscript.get_or_none(video_id, language_code)
if transcript is None:
create_or_update_video_transcript(video_id, language_code, metadata={
'provider': provider,
'file_name': transcript_name,
'file_format': file_format
})
response = Response(status=status.HTTP_200_OK)
else:
message = (
u'Can not override existing transcript for video "{video_id}" and language code "{language}".'
).format(video_id=video_id, language=language_code)
response = Response(status=status.HTTP_400_BAD_REQUEST, data={'message': message})
return response
def transcript_upload_handler(request):
"""
View to upload a transcript file.
Arguments:
request: A WSGI request object
Transcript file, edx video id and transcript language are required.
Transcript file should be in SRT(SubRip) format.
Returns
- A 400 if any of the validation fails
- A 200 if transcript has been uploaded successfully
"""
error = validate_transcript_upload_data(data=request.POST,
files=request.FILES)
if error:
response = JsonResponse({'error': error}, status=400)
else:
edx_video_id = request.POST['edx_video_id']
language_code = request.POST['language_code']
new_language_code = request.POST['new_language_code']
transcript_file = request.FILES['file']
try:
# Convert SRT transcript into an SJSON format
# and upload it to S3.
sjson_subs = Transcript.convert(
content=transcript_file.read().decode('utf-8'),
input_format=Transcript.SRT,
output_format=Transcript.SJSON)
create_or_update_video_transcript(
video_id=edx_video_id,
language_code=language_code,
metadata={
'provider': TranscriptProvider.CUSTOM,
'file_format': Transcript.SJSON,
'language_code': new_language_code
},
file_data=ContentFile(sjson_subs),
)
response = JsonResponse(status=201)
except (TranscriptsGenerationException, UnicodeDecodeError):
response = JsonResponse(
{
'error':
_(u'There is a problem with this transcript file. Try to upload a different file.'
)
},
status=400)
return response
def transcript_upload_handler(request):
"""
View to upload a transcript file.
Arguments:
request: A WSGI request object
Transcript file, edx video id and transcript language are required.
Transcript file should be in SRT(SubRip) format.
Returns
- A 400 if any of the validation fails
- A 200 if transcript has been uploaded successfully
"""
error = validate_transcript_upload_data(data=request.POST, files=request.FILES)
if error:
response = JsonResponse({'error': error}, status=400)
else:
edx_video_id = request.POST['edx_video_id']
language_code = request.POST['language_code']
new_language_code = request.POST['new_language_code']
transcript_file = request.FILES['file']
try:
# Convert SRT transcript into an SJSON format
# and upload it to S3.
sjson_subs = Transcript.convert(
content=transcript_file.read(),
input_format=Transcript.SRT,
output_format=Transcript.SJSON
)
create_or_update_video_transcript(
video_id=edx_video_id,
language_code=language_code,
metadata={
'provider': TranscriptProvider.CUSTOM,
'file_format': Transcript.SJSON,
'language_code': new_language_code
},
file_data=ContentFile(sjson_subs),
)
response = JsonResponse(status=201)
except (TranscriptsGenerationException, UnicodeDecodeError):
response = JsonResponse(
{'error': _(u'There is a problem with this transcript file. Try to upload a different file.')},
status=400
)
return response
def save_transcript_to_storage(edx_video_id,
language_code,
transcript_content,
file_format=Transcript.SJSON,
force_update=False):
"""
Pushes a given transcript's data to django storage.
"""
try:
result = None
edx_video_id = clean_video_id(edx_video_id)
if force_update:
result = create_or_update_video_transcript(
edx_video_id, language_code, dict({'file_format':
file_format}),
ContentFile(transcript_content))
LOGGER.info(
"[Transcript migration] save_transcript_to_storage %s for %s with create_or_update method",
True if result else False, edx_video_id)
else:
result = create_video_transcript(edx_video_id, language_code,
file_format,
ContentFile(transcript_content))
LOGGER.info(
"[Transcript migration] save_transcript_to_storage %s for %s with create method",
result, edx_video_id)
return result
except ValCannotCreateError as err:
LOGGER.exception(
"[Transcript migration] save_transcript_to_storage_failed: %s",
err)
raise
def upload_transcript(request):
"""
Upload a transcript file
Arguments:
request: A WSGI request object
Transcript file in SRT format
"""
edx_video_id = request.POST['edx_video_id']
language_code = request.POST['language_code']
new_language_code = request.POST['new_language_code']
transcript_file = request.FILES['file']
try:
# Convert SRT transcript into an SJSON format
# and upload it to S3.
sjson_subs = Transcript.convert(
content=transcript_file.read().decode('utf-8'),
input_format=Transcript.SRT,
output_format=Transcript.SJSON).encode()
create_or_update_video_transcript(
video_id=edx_video_id,
language_code=language_code,
metadata={
'provider': TranscriptProvider.CUSTOM,
'file_format': Transcript.SJSON,
'language_code': new_language_code
},
file_data=ContentFile(sjson_subs),
)
response = JsonResponse(status=201)
except (TranscriptsGenerationException, UnicodeDecodeError):
LOGGER.error(
"Unable to update transcript on edX video %s for language %s",
edx_video_id, new_language_code)
response = JsonResponse(
{
'error':
_('There is a problem with this transcript file. Try to upload a different file.'
)
},
status=400)
finally:
LOGGER.info("Updated transcript on edX video %s for language %s",
edx_video_id, new_language_code)
return response
def upload_transcripts(request):
"""
Upload transcripts for current module.
returns: response dict::
status: 'Success' and HTTP 200 or 'Error' and HTTP 400.
subs: Value of uploaded and saved html5 sub field in video item.
"""
error, validated_data = validate_transcript_upload_data(request)
if error:
response = JsonResponse({'status': error}, status=400)
else:
video = validated_data['video']
edx_video_id = validated_data['edx_video_id']
transcript_file = validated_data['transcript_file']
# check if we need to create an external VAL video to associate the transcript
# and save its ID on the video component.
if not edx_video_id:
edx_video_id = create_external_video(display_name=u'external video')
video.edx_video_id = edx_video_id
video.save_with_metadata(request.user)
response = JsonResponse({'edx_video_id': edx_video_id, 'status': 'Success'}, status=200)
try:
# Convert 'srt' transcript into the 'sjson' and upload it to
# configured transcript storage. For example, S3.
sjson_subs = Transcript.convert(
content=transcript_file.read(),
input_format=Transcript.SRT,
output_format=Transcript.SJSON
)
transcript_created = create_or_update_video_transcript(
video_id=edx_video_id,
language_code=u'en',
metadata={
'provider': TranscriptProvider.CUSTOM,
'file_format': Transcript.SJSON,
'language_code': u'en'
},
file_data=ContentFile(sjson_subs),
)
if transcript_created is None:
response = JsonResponse({'status': 'Invalid Video ID'}, status=400)
except (TranscriptsGenerationException, UnicodeDecodeError):
response = JsonResponse({
'status': _(u'There is a problem with this transcript file. Try to upload a different file.')
}, status=400)
return response
def upload_transcripts(request):
"""
Upload transcripts for current module.
returns: response dict::
status: 'Success' and HTTP 200 or 'Error' and HTTP 400.
subs: Value of uploaded and saved html5 sub field in video item.
"""
error, validated_data = validate_transcript_upload_data(request)
if error:
response = JsonResponse({'status': error}, status=400)
else:
video = validated_data['video']
edx_video_id = validated_data['edx_video_id']
transcript_file = validated_data['transcript_file']
# check if we need to create an external VAL video to associate the transcript
# and save its ID on the video component.
if not edx_video_id:
edx_video_id = create_external_video(display_name=u'external video')
video.edx_video_id = edx_video_id
video.save_with_metadata(request.user)
response = JsonResponse({'edx_video_id': edx_video_id, 'status': 'Success'}, status=200)
try:
# Convert 'srt' transcript into the 'sjson' and upload it to
# configured transcript storage. For example, S3.
sjson_subs = Transcript.convert(
content=transcript_file.read(),
input_format=Transcript.SRT,
output_format=Transcript.SJSON
)
transcript_created = create_or_update_video_transcript(
video_id=edx_video_id,
language_code=u'en',
metadata={
'provider': TranscriptProvider.CUSTOM,
'file_format': Transcript.SJSON,
'language_code': u'en'
},
file_data=ContentFile(sjson_subs),
)
if transcript_created is None:
response = JsonResponse({'status': 'Invalid Video ID'}, status=400)
except (TranscriptsGenerationException, UnicodeDecodeError):
response = JsonResponse({
'status': _(u'There is a problem with this transcript file. Try to upload a different file.')
}, status=400)
return response
def save_video_transcript(edx_video_id, input_format, transcript_content, language_code):
"""
Saves a video transcript to the VAL and its content to the configured django storage(DS).
Arguments:
edx_video_id: A Video ID to associate the transcript.
input_format: Input transcript format for content being passed.
transcript_content: Content of the transcript file
language_code: transcript language code
Returns:
A boolean indicating whether the transcript was saved or not.
"""
try:
# Convert the transcript into the 'sjson' and upload it to
# configured transcript storage. For example, S3.
sjson_subs = Transcript.convert(
content=transcript_content,
input_format=input_format,
output_format=Transcript.SJSON
)
create_or_update_video_transcript(
video_id=edx_video_id,
language_code=language_code,
metadata={
'provider': TranscriptProvider.CUSTOM,
'file_format': Transcript.SJSON,
'language_code': language_code
},
file_data=ContentFile(sjson_subs),
)
result = True
except (TranscriptsGenerationException, UnicodeDecodeError):
result = False
return result
def save_transcript_to_storage(
edx_video_id,
language_code,
transcript_content,
file_format=Transcript.SJSON,
force_update=False
):
"""
Pushes a given transcript's data to django storage.
"""
try:
result = None
edx_video_id = clean_video_id(edx_video_id)
if force_update:
result = create_or_update_video_transcript(
edx_video_id,
language_code,
dict({'file_format': file_format}),
ContentFile(transcript_content)
)
LOGGER.info("[Transcript migration] save_transcript_to_storage %s for %s with create_or_update method",
True if result else False, edx_video_id)
else:
result = create_video_transcript(
edx_video_id,
language_code,
file_format,
ContentFile(transcript_content)
)
LOGGER.info(
"[Transcript migration] save_transcript_to_storage %s for %s with create method",
result,
edx_video_id
)
return result
except ValCannotCreateError as err:
LOGGER.exception("[Transcript migration] save_transcript_to_storage_failed: %s", err)
raise
def studio_transcript(self, request, dispatch):
"""
Entry point for Studio transcript handlers.
Dispatches:
/translation/[language_id] - language_id sould be in url.
`translation` dispatch support following HTTP methods:
`POST`:
Upload srt file. Check possibility of generation of proper sjson files.
For now, it works only for self.transcripts, not for `en`.
Do not update self.transcripts, as fields are updated on save in Studio.
`GET:
Return filename from storage. SRT format is sent back on success. Filename should be in GET dict.
We raise all exceptions right in Studio:
NotFoundError:
Video or asset was deleted from module/contentstore, but request came later.
Seems impossible to be raised. module_render.py catches NotFoundErrors from here.
/translation POST:
TypeError:
Unjsonable filename or content.
TranscriptsGenerationException, TranscriptException:
no SRT extension or not parse-able by PySRT
UnicodeDecodeError: non-UTF8 uploaded file content encoding.
"""
_ = self.runtime.service(self, "i18n").ugettext
if dispatch.startswith('translation'):
if request.method == 'POST':
error = self.validate_transcript_upload_data(data=request.POST)
if error:
response = Response(json={'error': error}, status=400)
else:
edx_video_id = clean_video_id(request.POST['edx_video_id'])
language_code = request.POST['language_code']
new_language_code = request.POST['new_language_code']
transcript_file = request.POST['file'].file
if not edx_video_id:
# Back-populate the video ID for an external video.
# pylint: disable=attribute-defined-outside-init
self.edx_video_id = edx_video_id = create_external_video(display_name=u'external video')
try:
# Convert SRT transcript into an SJSON format
# and upload it to S3.
sjson_subs = Transcript.convert(
content=transcript_file.read(),
input_format=Transcript.SRT,
output_format=Transcript.SJSON
)
create_or_update_video_transcript(
video_id=edx_video_id,
language_code=language_code,
metadata={
'file_format': Transcript.SJSON,
'language_code': new_language_code
},
file_data=ContentFile(sjson_subs),
)
payload = {
'edx_video_id': edx_video_id,
'language_code': new_language_code
}
response = Response(json.dumps(payload), status=201)
except (TranscriptsGenerationException, UnicodeDecodeError):
response = Response(
json={
'error': _(
u'There is a problem with this transcript file. Try to upload a different file.'
)
},
status=400
)
elif request.method == 'DELETE':
request_data = request.json
if 'lang' not in request_data or 'edx_video_id' not in request_data:
return Response(status=400)
language = request_data['lang']
edx_video_id = clean_video_id(request_data['edx_video_id'])
if edx_video_id:
delete_video_transcript(video_id=edx_video_id, language_code=language)
if language == u'en':
# remove any transcript file from content store for the video ids
possible_sub_ids = [
self.sub, # pylint: disable=access-member-before-definition
self.youtube_id_1_0
] + get_html5_ids(self.html5_sources)
for sub_id in possible_sub_ids:
remove_subs_from_store(sub_id, self, language)
# update metadata as `en` can also be present in `transcripts` field
remove_subs_from_store(self.transcripts.pop(language, None), self, language)
# also empty `sub` field
self.sub = '' # pylint: disable=attribute-defined-outside-init
else:
remove_subs_from_store(self.transcripts.pop(language, None), self, language)
return Response(status=200)
elif request.method == 'GET':
language = request.GET.get('language_code')
if not language:
return Response(json={'error': _(u'Language is required.')}, status=400)
try:
transcript_content, transcript_name, mime_type = get_transcript(
video=self, lang=language, output_format=Transcript.SRT
)
response = Response(transcript_content, headerlist=[
(
'Content-Disposition',
'attachment; filename="{}"'.format(
transcript_name.encode('utf8') if six.PY2 else transcript_name
)
),
('Content-Language', language),
('Content-Type', mime_type)
])
except (UnicodeDecodeError, TranscriptsGenerationException, NotFoundError):
response = Response(status=404)
else:
# Any other HTTP method is not allowed.
response = Response(status=404)
else: # unknown dispatch
log.debug("Dispatch is not allowed")
response = Response(status=404)
return response
def studio_transcript(self, request, dispatch):
"""
Entry point for Studio transcript handlers.
Dispatches:
/translation/[language_id] - language_id sould be in url.
`translation` dispatch support following HTTP methods:
`POST`:
Upload srt file. Check possibility of generation of proper sjson files.
For now, it works only for self.transcripts, not for `en`.
Do not update self.transcripts, as fields are updated on save in Studio.
`GET:
Return filename from storage. SRT format is sent back on success. Filename should be in GET dict.
We raise all exceptions right in Studio:
NotFoundError:
Video or asset was deleted from module/contentstore, but request came later.
Seems impossible to be raised. module_render.py catches NotFoundErrors from here.
/translation POST:
TypeError:
Unjsonable filename or content.
TranscriptsGenerationException, TranscriptException:
no SRT extension or not parse-able by PySRT
UnicodeDecodeError: non-UTF8 uploaded file content encoding.
"""
_ = self.runtime.service(self, "i18n").ugettext
if dispatch.startswith('translation'):
if request.method == 'POST':
error = self.validate_transcript_upload_data(data=request.POST)
if error:
response = Response(json={'error': error}, status=400)
else:
edx_video_id = clean_video_id(request.POST['edx_video_id'])
language_code = request.POST['language_code']
new_language_code = request.POST['new_language_code']
transcript_file = request.POST['file'].file
if not edx_video_id:
# Back-populate the video ID for an external video.
# pylint: disable=attribute-defined-outside-init
self.edx_video_id = edx_video_id = create_external_video(display_name=u'external video')
try:
# Convert SRT transcript into an SJSON format
# and upload it to S3.
sjson_subs = Transcript.convert(
content=transcript_file.read(),
input_format=Transcript.SRT,
output_format=Transcript.SJSON
)
create_or_update_video_transcript(
video_id=edx_video_id,
language_code=language_code,
metadata={
'file_format': Transcript.SJSON,
'language_code': new_language_code
},
file_data=ContentFile(sjson_subs),
)
payload = {
'edx_video_id': edx_video_id,
'language_code': new_language_code
}
response = Response(json.dumps(payload), status=201)
except (TranscriptsGenerationException, UnicodeDecodeError):
response = Response(
json={
'error': _(
u'There is a problem with this transcript file. Try to upload a different file.'
)
},
status=400
)
elif request.method == 'DELETE':
request_data = request.json
if 'lang' not in request_data or 'edx_video_id' not in request_data:
return Response(status=400)
language = request_data['lang']
edx_video_id = clean_video_id(request_data['edx_video_id'])
if edx_video_id:
delete_video_transcript(video_id=edx_video_id, language_code=language)
if language == u'en':
# remove any transcript file from content store for the video ids
possible_sub_ids = [
self.sub, # pylint: disable=access-member-before-definition
self.youtube_id_1_0
] + get_html5_ids(self.html5_sources)
for sub_id in possible_sub_ids:
remove_subs_from_store(sub_id, self, language)
# update metadata as `en` can also be present in `transcripts` field
remove_subs_from_store(self.transcripts.pop(language, None), self, language)
# also empty `sub` field
self.sub = '' # pylint: disable=attribute-defined-outside-init
else:
remove_subs_from_store(self.transcripts.pop(language, None), self, language)
return Response(status=200)
elif request.method == 'GET':
language = request.GET.get('language_code')
if not language:
return Response(json={'error': _(u'Language is required.')}, status=400)
try:
transcript_content, transcript_name, mime_type = get_transcript(
video=self, lang=language, output_format=Transcript.SRT
)
response = Response(transcript_content, headerlist=[
('Content-Disposition', 'attachment; filename="{}"'.format(transcript_name.encode('utf8'))),
('Content-Language', language),
('Content-Type', mime_type)
])
except (UnicodeDecodeError, TranscriptsGenerationException, NotFoundError):
response = Response(status=404)
else:
# Any other HTTP method is not allowed.
response = Response(status=404)
else: # unknown dispatch
log.debug("Dispatch is not allowed")
response = Response(status=404)
return response
