def reconstruct_files():
count = 0
for file in File.objects(fragments__is_clean=False):
logger.debug('Reconstructing {}'.format(file))
file.reconstruct()
count += 1
return count
def post_file(request):
files = File.objects(Q(source=request.parameters['source']) &
Q(collection=request.parameters['collection']) &
Q(filename=request.parameters['name'])).all()
if len(files) == 0:
file = File()
file.collection = request.parameters['collection']
file.filename = request.parameters['name']
file.encoding = Encoding(**DEFAULT_ENCODING)
hubs = Hub.objects(cumulus_id=request.parameters['source'])
if len(hubs) == 0:
raise ObjectDoesNotExist('Source does not exist')
elif len(hubs) == 1:
hub = hubs[0]
else:
raise MultipleObjectsFound('Multiple files found for source')
file.source = hub
elif len(files) == 1:
file = files[0]
else:
raise MultipleObjectsFound('Multiple objects found for the search query')
with file as f:
f.write(request.data)
return FileSerializer(file).data
def get_file(request):
files = File.objects(Q(source=request.parameters['source']) &
Q(collection=request.parameters['collection']) &
Q(filename=request.parameters['name'])).all()
if len(files) == 0:
raise ObjectDoesNotExist('File does not exist')
elif len(files) == 1:
file = files[0]
else:
raise MultipleObjectsFound('Multiple files found for the search query')
return FileContentSerializer(file).data
def v_random(func):
pipeline = [{
'$sample': {
'size': int(len(File.objects) * VERIFY_FRACTION)
}
}]
files_to_reconstruct = list()
for file_dict in File.objects.aggregate(*pipeline):
file = one(File.objects(uuid=file_dict['uuid']))
if not getattr(file, func)():
files_to_reconstruct.append(file.uuid)
logger.debug('{} check failed: {}: {}/{}/{}'.format(
func, file.uid, file.source, file.collection, file.filename))
logger.info('Files to reconstruct: {}'.format(len(files_to_reconstruct)))
