def upload(dataset):
dataset = Dataset.find(dataset)
authz.require(authz.dataset_edit(dataset))
file_ = request.files.get('file')
if not file_ or not file_.filename:
err = {'file': "You need to upload a file"}
raise Invalid("No file.", None, None, error_dict=err)
upload = Upload.create(dataset, request.account, file_)
db.session.commit()
return jsonify(upload)
def parse_upload(dataset, id):
upload = Upload.by_id(id)
if upload is None:
return None, None
fh = open(upload.path, 'rb')
fn, ext = os.path.splitext(upload.filename)
ext = ext[1:] if ext else None
table_set = AnyTableSet.from_fileobj(fh,
mimetype=upload.mimetype,
extension=ext[1:])
return upload, table_set.tables[0]
def process(dataset, id):
dataset = Dataset.find(dataset)
authz.require(authz.dataset_edit(dataset))
upload = Upload.find(dataset, id)
mapping = request_data()
mapping['reviewed'] = mapping.get('reviewed') or False
mapping['columns'] = mapping.get('columns', {})
fields = mapping['columns'].values()
for header in mapping['columns'].keys():
if header not in upload.tab.headers:
raise Invalid("Invalid header: %s" % header, None, None)
if 'name' not in fields and 'id' not in fields:
raise Invalid("You have not selected a field that definies entity names.", None, None)
import_upload.delay(upload.id, request.account.id, mapping)
return jsonify({'status': 'Loading data...'})
def import_upload(upload_id, account_id, mapping):
upload = Upload.all().filter_by(id=upload_id).first()
account = Account.by_id(account_id)
mapped = mapping['columns'].values()
rows = [apply_mapping(r, mapping) for r in upload.tab.dict]
# put aliases second.
rows = sorted(rows, key=lambda r: 2 if r.get('canonical') else 1)
for i, row in enumerate(rows):
try:
entity = None
if row.get('id'):
entity = Entity.by_id(row.get('id'))
if entity is None:
entity = Entity.by_name(upload.dataset, row.get('name'))
if entity is None:
entity = Entity.create(upload.dataset, row, account)
# restore some defaults:
if entity.canonical_id and 'canonical' not in mapped:
row['canonical'] = entity.canonical_id
if entity.invalid and 'invalid' not in mapped:
row['invalid'] = entity.invalid
if entity.attributes:
attributes = entity.attributes.copy()
else:
attributes = {}
attributes.update(row['attributes'])
row['attributes'] = attributes
entity.update(row, account)
print(entity)
if i % 100 == 0:
db.session.commit()
logging.debug('Commit')
except Invalid as inv:
logging.warning('Exception during import: {}'.format(str(inv)))
db.session.commit()
logging.info('Import Completed')
def import_upload(upload_id, account_id, mapping):
upload = Upload.all().filter_by(id=upload_id).first()
account = Account.by_id(account_id)
mapped = mapping['columns'].values()
rows = [apply_mapping(r, mapping) for r in upload.tab.dict]
# put aliases second.
rows = sorted(rows, key=lambda r: 2 if r.get('canonical') else 1)
for i, row in enumerate(rows):
try:
entity = None
if row.get('id'):
entity = Entity.by_id(row.get('id'))
if entity is None:
entity = Entity.by_name(upload.dataset, row.get('name'))
if entity is None:
entity = Entity.create(upload.dataset, row, account)
# restore some defaults:
if entity.canonical_id and 'canonical' not in mapped:
row['canonical'] = entity.canonical_id
if entity.invalid and 'invalid' not in mapped:
row['invalid'] = entity.invalid
if entity.attributes:
attributes = entity.attributes.copy()
else:
attributes = {}
attributes.update(row['attributes'])
row['attributes'] = attributes
entity.update(row, account)
print entity
if i % 100 == 0:
print 'COMMIT'
db.session.commit()
except Invalid, inv:
# TODO: logging.
print inv
def view(dataset, id):
dataset = Dataset.find(dataset)
authz.require(authz.dataset_edit(dataset))
upload = Upload.find(dataset, id)
return jsonify(upload)
def upload_file(dataset, file_, account):
upload = Upload.create(dataset, account,
file_.filename, file_.mimetype)
file_.save(upload.path)
db.session.commit()
return upload
