def main():
args = parse_arguments([{'name': 'fields', 'type': str}])
# extract arguments
dataset = load_dataset(args.dataset_path)
original_dataset = load_dataset(args.origin_path)
fields = json.loads(args.fields)
usernames = []
iterate_and_apply(original_dataset, fields, usernames.append)
# detect names by traversing the dataset
matches = []
find_names(dataset, usernames, matches)
# notify the detected user names
if len(matches) > 0:
messages = ["Detected user names: "] + \
[' - %s' % m for m in matches[:10]]
if len(matches) > 10:
messages.append(' - ...')
notify_validation_result(ValidationOutcome.FAILURE,
'\n'.join(messages))
else:
notify_validation_result(ValidationOutcome.SUCCESS,
'No user name detected')
def main():
# Prepares the parameters for the algorithm (dataset_path and output_path
# come by default). You can then use them with args.parameter_name.
# Avoid editing the parameters here, use the dedicated utility instead and
# the code will change accordingly.
args = parse_arguments()
# load the json dataset, available as a python dictionary
dataset = load_dataset(args.dataset_path)
# write your validation code here
# notify the validation outcome
# outcome should be one of ValidationOutcome.SUCCESS,
# ValidationOutcome.WARNING or ValidationOutcome.FAILURE
# write the information to display in info
outcome = ValidationOutcome.SUCCESS
info = ""
notify_validation_result(outcome, info)
def main():
args = parse_arguments([
{'name': 'quasi_identifiers', 'type': str},
{'name': 'k', 'type': int}
])
dataset = load_dataset(args.dataset_path)
quasi_identifiers = json.loads(args.quasi_identifiers)
relevant_arrays = find_selected_arrays(dataset, quasi_identifiers)
if len(relevant_arrays) == 0:
notify_validation_result(
ValidationOutcome.WARNING, "No quasi identifier was found")
else:
# compute k-anonymity for each array containing QI's.
# final k-anonymity will be the lowest of them
ks = [compute_kanonymity(a['array'], a['field_selection'])
for a in relevant_arrays]
k = min(ks)
if k >= args.k:
notify_validation_result(
ValidationOutcome.SUCCESS,
'%d-anonymous (>= %d)' % (k, args.k))
else:
notify_validation_result(
ValidationOutcome.FAILURE,
'%d-anonymous (< %d)' % (k, args.k))
def main():
args = parse_arguments([
{'name': 'quasi_identifiers', 'type': str},
{'name': 'sensitive_attributes', 'type': str},
{'name': 'l', 'type': int}
])
dataset = load_dataset(args.dataset_path)
quasi_identifiers = json.loads(args.quasi_identifiers)
sensitive_attributes = json.loads(args.sensitive_attributes)
relevant_arrays = find_selected_arrays(
dataset, quasi_identifiers, sensitive_attributes)
if len(relevant_arrays) == 0:
notify_validation_result(
ValidationOutcome.WARNING, "No sensitive attribute was found")
else:
# compute l-diversity for each array with SA's
# final l-diversity will be the smallest of them
ls = [compute_ldiversity(array['array'],
array['quasi_identifiers'],
array['sensitive_attributes'])
for array in relevant_arrays]
l = min(ls)
if l >= args.l:
notify_validation_result(
ValidationOutcome.SUCCESS,
'%d-diversified (>= %d)' % (l, args.l))
else:
notify_validation_result(
ValidationOutcome.FAILURE,
'%d-diversified (< %d)' % (l, args.l))
def main():
notify_validation_result(ValidationOutcome.SUCCESS, "Successful")
def main():
notify_validation_result(ValidationOutcome.FAILURE, "Failure")
def main():
args = parse_arguments()
dataset = load_dataset(args.dataset_path)
has_user_name = False
has_AP_settings = False
has_API_data = False
# verify in 'actions' if 'data' and 'geolocation' are present
has_action_data = False
has_action_geolocation = False
actions = dataset['data']['actions']
for action in actions:
if 'data' in action:
data = action['data']
if data != None and data != '' and data != {}:
has_action_data = True
if 'geolocation' in action:
geolocation = action['geolocation']
if geolocation != None and geolocation != '' and geolocation != {}:
has_action_geolocation = True
# verify in 'users' if 'name' is present
users = dataset['data']['users']
for user in users:
if 'name' in user:
name = user['name']
if name != None and name != '':
has_user_name = True
# verify in 'appInstances' if 'settings' is present
appInstances = dataset['data']['appInstances']
for ar in appInstances:
if 'settings' in ar:
settings = ar['settings']
if settings != None and settings != '' and settings != {}:
has_AP_settings = True
# verify in 'appInstanceResources' if 'data' is present
appInstanceResources = dataset['data']['appInstanceResources']
for air in appInstanceResources:
if 'data' in air:
data = air['data']
if data != None and data != '' and data != {}:
has_API_data = True
# issue a warning if any of these potentially dangerousattributes
# are present
potentially_dangerous = (has_action_data or has_action_geolocation
or has_user_name or has_AP_settings
or has_API_data)
if potentially_dangerous:
messages = ['Potentially dangerous attributes: ']
if has_action_data:
messages.append('- actions > data')
if has_action_geolocation:
messages.append('- actions > geolocation')
if has_user_name:
messages.append('- users > name')
if has_AP_settings:
messages.append('- appInstances > settings')
if has_API_data:
messages.append('- appInstanceResources > data')
notify_validation_result(ValidationOutcome.WARNING,
'\n'.join(messages))
else:
notify_validation_result(ValidationOutcome.SUCCESS,
'No potentially dangerous attributes')
def main():
notify_validation_result(ValidationOutcome.WARNING, "Warning")