def get(self, request, *args, **kwargs): # pylint: disable=unused-argument
version = self.get_object()
logger.debug(
'Export requested for %s version %s - Requesting AWS-S3 key',
self.entity.lower(), version.version)
if version.is_head:
return Response(status=status.HTTP_405_METHOD_NOT_ALLOWED)
key = version.export_path
url = S3.url_for(key)
if url:
logger.debug(' URL and Key retrieved for %s version %s',
self.entity.lower(), version.version)
else:
logger.debug(' Key does not exist for %s version %s',
self.entity.lower(), version.version)
return Response(status=status.HTTP_204_NO_CONTENT)
response = Response(status=status.HTTP_303_SEE_OTHER)
response['Location'] = url
# Set headers to ensure sure response is not cached by a client
response['Cache-Control'] = 'no-cache, no-store, must-revalidate'
response['Pragma'] = 'no-cache'
response['Expires'] = '0'
response['Last-Updated'] = version.last_child_update.isoformat()
response['Last-Updated-Timezone'] = settings.TIME_ZONE_PLACE
return response
def get_csv_from_s3(filename, is_owner): # pragma: no cover
filename = get_downloads_path(is_owner) + filename + '.csv.zip'
if S3.exists(filename):
return S3.url_for(filename)
return None
def write_csv_to_s3(data, is_owner, **kwargs): # pragma: no cover
cwd = cd_temp()
csv_file = csv_file_for(data, **kwargs)
csv_file.close()
zip_file_name = csv_file.name + '.zip'
with zipfile.ZipFile(zip_file_name, 'w', zipfile.ZIP_DEFLATED) as zip_file:
zip_file.write(csv_file.name)
file_path = get_downloads_path(is_owner) + zip_file_name
S3.upload_file(file_path)
os.chdir(cwd)
return S3.url_for(file_path)
def write_csv_to_s3(data, is_owner, **kwargs): # pragma: no cover
cwd = cd_temp()
csv_file = csv_file_for(data, **kwargs)
csv_file.close()
zip_file_name = csv_file.name + '.zip'
with zipfile.ZipFile(zip_file_name, 'w', zipfile.ZIP_DEFLATED) as zip_file:
zip_file.write(csv_file.name)
key = get_downloads_path(is_owner) + zip_file.filename
S3.upload_file(key=key,
file_path=os.path.abspath(zip_file.filename),
binary=True,
metadata=dict(ContentType='application/zip'),
headers={'content-type': 'application/zip'})
os.chdir(cwd)
return S3.url_for(key)
def write_export_file(version, resource_type, resource_serializer_type,
logger): # pylint: disable=too-many-statements,too-many-locals
cwd = cd_temp()
logger.info('Writing export file to tmp directory: %s' % cwd)
logger.info('Found %s version %s. Looking up resource...' %
(resource_type, version.version))
resource = version.head
logger.info('Found %s %s. Serializing attributes...' %
(resource_type, resource.mnemonic))
resource_serializer = get_class(resource_serializer_type)(version)
data = resource_serializer.data
resource_string = json.dumps(data, cls=encoders.JSONEncoder)
logger.info('Done serializing attributes.')
batch_size = 1000
concepts_qs = version.concepts
mappings_qs = version.mappings
if resource_type != 'collection':
concepts_qs = concepts_qs.filter(is_active=True)
mappings_qs = mappings_qs.filter(is_active=True)
total_concepts = concepts_qs.count()
total_mappings = mappings_qs.count()
with open('export.json', 'w') as out:
out.write('%s, "concepts": [' % resource_string[:-1])
if total_concepts:
logger.info('%s has %d concepts. Getting them in batches of %d...' %
(resource_type.title(), total_concepts, batch_size))
concept_serializer_class = get_class(
'core.concepts.serializers.ConceptVersionDetailSerializer')
for start in range(0, total_concepts, batch_size):
end = min(start + batch_size, total_concepts)
logger.info('Serializing concepts %d - %d...' % (start + 1, end))
concept_versions = concepts_qs.prefetch_related(
'names',
'descriptions').select_related('parent__organization',
'parent__user')[start:end]
concept_serializer = concept_serializer_class(concept_versions,
many=True)
concept_data = concept_serializer.data
concept_string = json.dumps(concept_data, cls=encoders.JSONEncoder)
concept_string = concept_string[1:-1]
with open('export.json', 'a') as out:
out.write(concept_string)
if end != total_concepts:
out.write(', ')
logger.info('Done serializing concepts.')
else:
logger.info('%s has no concepts to serialize.' %
(resource_type.title()))
with open('export.json', 'a') as out:
out.write('], "mappings": [')
if total_mappings:
logger.info('%s has %d mappings. Getting them in batches of %d...' %
(resource_type.title(), total_mappings, batch_size))
mapping_serializer_class = get_class(
'core.mappings.serializers.MappingDetailSerializer')
for start in range(0, total_mappings, batch_size):
end = min(start + batch_size, total_mappings)
logger.info('Serializing mappings %d - %d...' % (start + 1, end))
mappings = mappings_qs.select_related(
'parent__organization',
'parent__user',
'from_concept',
'to_concept',
'from_source__organization',
'from_source__user',
'to_source__organization',
'to_source__user',
)[start:end]
mapping_serializer = mapping_serializer_class(mappings, many=True)
mapping_data = mapping_serializer.data
mapping_string = json.dumps(mapping_data, cls=encoders.JSONEncoder)
mapping_string = mapping_string[1:-1]
with open('export.json', 'a') as out:
out.write(mapping_string)
if end != total_mappings:
out.write(', ')
logger.info('Done serializing mappings.')
else:
logger.info('%s has no mappings to serialize.' %
(resource_type.title()))
with open('export.json', 'a') as out:
out.write(']}')
with zipfile.ZipFile('export.zip', 'w', zipfile.ZIP_DEFLATED) as _zip:
_zip.write('export.json')
file_path = os.path.abspath('export.zip')
logger.info(file_path)
logger.info('Done compressing. Uploading...')
s3_key = version.export_path
S3.upload_file(key=s3_key, file_path=file_path, binary=True)
uploaded_path = S3.url_for(s3_key)
logger.info('Uploaded to %s.' % uploaded_path)
os.chdir(cwd)
def get_export_url(self):
return S3.url_for(self.export_path)
def get_csv_from_s3(filename, is_owner):
filename = get_downloads_path(is_owner) + filename + '.csv.zip'
return S3.url_for(filename)