def update_collection(self, uuid, record):
if (HubmapConst.UUID_ATTRIBUTE in record
or HubmapConst.DOI_ATTRIBUTE in record
or HubmapConst.DISPLAY_DOI_ATTRIBUTE in record
or HubmapConst.ENTITY_TYPE_ATTRIBUTE in record):
raise HTTPException("ID attributes cannot be changed", 400)
not_allowed = []
for attrib in record.keys():
if not attrib in self.allowed_collection_update_attributes:
not_allowed.append(attrib)
if len(not_allowed) > 0:
raise HTTPException(
"Attribute(s) not allowed: " +
string_helper.listToDelimited(not_allowed, " "), 400)
if HubmapConst.COLLECTION_CREATORS_ATTRIBUTE in record:
creators = record[HubmapConst.COLLECTION_CREATORS_ATTRIBUTE]
for creator in creators:
for attrib in creator.keys():
if not attrib in self.allowed_creator_attributes and not attrib in not_allowed:
not_allowed.append(attrib)
if len(not_allowed) > 0:
raise HTTPException(
"Creator Aattribute(s) not allowed: " +
string_helper.listToDelimited(not_allowed, " "), 400)
save_record = {}
for attrib in record.keys():
if attrib == HubmapConst.COLLECTION_CREATORS_ATTRIBUTE:
save_record[attrib] = json.dumps(record[attrib])
else:
save_record[attrib] = record[attrib]
rval = self.get_py2neo_conn().run(
"match(c:Collection {uuid: {uuid}}) set c += {params} return c.uuid",
uuid=uuid,
params=save_record).data()
if len(rval) == 0:
raise HTTPException(
"Update failed for collection with uuid " + uuid +
". UUID possibly not found.", 400)
else:
if rval[0]['c.uuid'] != uuid:
raise HTTPException(
"Update failed, wrong uuid returned while trying to update "
+ uuid + " returned: " + rval[0]['c.uuid'])
def check_write_privs(self, groups_token, group_uuid):
user_info = self.getUserInfo(groups_token, getGroups=True)
if isinstance(user_info, Response):
raise HTTPException(user_info.text, user_info.status_code)
groups_by_id = self.getHMGroupsById()
if not group_uuid in groups_by_id:
raise HTTPException(f"{group_uuid} is not a valid group uuid", 400)
grp = groups_by_id[group_uuid]
if not 'data_provider' in grp or not grp['data_provider']:
raise HTTPException(f"grop with uuid {group_uuid} is not a valid data provider group", 400)
if 'hmgroupids' in user_info:
if data_admin_group_uuid in user_info['hmgroupids']:
return True
elif group_uuid not in user_info['hmgroupids']:
raise HTTPException("User not authorized for group.", 403)
else:
return True
else:
raise HTTPException("User is not authorized, no group membership", 403)
def move_dataset_files_for_publishing(self,
uuid,
group_uuid,
dataset_access_level,
trial_run=False):
from_path = self.__dataset_directory_absolute_path(
dataset_access_level, group_uuid, uuid, False)
if not os.path.isdir(from_path):
raise HTTPException(
f"{uuid}: path not found to dataset will not publish, path is {from_path}",
500)
data_access_level = 'protected'
if not dataset_access_level == 'protected':
data_access_level = 'public'
to_path = self.__dataset_directory_absolute_path(
data_access_level, group_uuid, uuid, True)
if not trial_run:
shutil.move(from_path, to_path)
else:
print(f"mv {from_path} {to_path}")
return None
def get_user_roles_deprecated(self, token):
authcache = None
if AuthHelper.isInitialized() == False:
authcache = AuthHelper.create(
self.entity_config['APP_CLIENT_ID'], self.entity_config['APP_CLIENT_SECRET'])
else:
authcache = AuthHelper.instance()
userinfo = authcache.getUserInfo(token, True)
if type(userinfo) == Response and userinfo.status_code == 401:
raise HTTPException('token is invalid.', 401)
if 'hmgroupids' not in userinfo:
raise ValueError("Cannot find Hubmap Group information for token")
return_list = []
role_list = AuthCache.getHMRoles()
for role_uuid in userinfo['hmroleids']:
for role_name in role_list.keys():
if role_list[role_name]['uuid'] == role_uuid:
return_list.append(role_list[role_name])
break
return return_list
def get_write_group_uuid(self, request_or_token, group_uuid = None):
if isinstance(request_or_token, str):
user_info = self.getUserInfo(request_or_token, getGroups=True)
else:
user_info = self.getUserInfoUsingRequest(request_or_token, getGroups=True)
if isinstance(user_info, Response):
raise HTTPException("Error while getting user information from token. " + user_info.get_data(as_text=True), user_info.status_code)
if len(AuthCache.groupsById) == 0:
AuthCache.getHMGroups()
groups_by_id = AuthCache.groupsById
if not group_uuid is None:
if group_uuid in groups_by_id:
if not 'data_provider' in groups_by_id[group_uuid] or not groups_by_id[group_uuid]['data_provider']:
raise HTTPException(f"Group {groups_by_id[group_uuid]['displayname']} is not a valid group for submitting data.", 403)
#user must be a member of the group or a member of the data admin group
elif not (group_uuid in user_info['hmgroupids'] or data_admin_group_uuid in user_info['hmgroupids']):
raise HTTPException(f"User is not a member of the group {groups_by_id[group_uuid]['displayname']}", 403)
else:
return group_uuid
else:
raise HTTPException("Invalid group_uuid", 400)
else:
count = 0
found_group_uuid = None
for grp_id in groups_by_id.keys():
grp_info = groups_by_id[grp_id]
if grp_id in user_info['hmgroupids'] and 'data_provider' in grp_info and grp_info['data_provider'] == True:
count = count + 1
found_group_uuid = grp_id
if count == 0:
if data_admin_group_uuid in user_info['hmgroupids']:
raise HTTPException("User is not a member of any groups that can provide data, but is a member of the data admin group. Please specify which group in the group_uuid field")
else:
raise HTTPException("User is not a member of any groups that can provide data.", 403)
elif count > 1:
raise HTTPException("The user is a member of multiple groups that can provide data. Please specify which group in the group_uuid field", 400)
else:
return found_group_uuid
def create_derived_datastage(self, nexus_token, json_data):
auth_header = {'Authorization': 'Bearer ' + nexus_token}
app_header = {'X-Hubmap-Application': 'ingest-api'}
source_dataset_uuids = json_data['source_dataset_uuids']
source_uuids = []
if isinstance(source_dataset_uuids, str):
# Create a list from this string
source_uuids = [source_dataset_uuids]
elif isinstance(source_dataset_uuids, list):
source_uuids = source_dataset_uuids
else:
raise TypeError(
"json_data['source_dataset_uuids'] must either be a string or a list"
)
# All of the source datasets come from the same data provider
# Get the group_uuid based on the first source dataset via entity-api
first_source_uuid = source_uuids[0]
get_url = file_helper.ensureTrailingSlashURL(
self.confdata['ENTITY_WEBSERVICE_URL']
) + 'entities/' + first_source_uuid
response = requests.get(get_url, headers=auth_header, verify=False)
if response.status_code != 200:
raise HTTPException(
"Error retrieving source dataset " + first_source_uuid,
response.status_code)
first_source_dataset = response.json()
# Create the derived dataset via entity-api
# The entity-api validates each of the provided source dataset uuid for existenace check
# The derived dataset will have the same group_uuid as the source datasets
derived_dataset_to_post = {
'title': json_data['derived_dataset_name'],
'data_types': json_data['derived_dataset_types'],
'direct_ancestor_uuids': source_uuids,
'contains_human_genetic_sequences': False,
'group_uuid': first_source_dataset['group_uuid']
}
post_url = file_helper.ensureTrailingSlashURL(
self.confdata['ENTITY_WEBSERVICE_URL']) + 'entities/dataset'
# Merge the auth_header and app_header for creating new Dataset
response = requests.post(post_url,
json=derived_dataset_to_post,
headers={
**auth_header,
**app_header
},
verify=False)
if response.status_code != 200:
raise HTTPException(
"Error creating derived dataset: " + response.text,
response.status_code)
derived_dataset = response.json()
file_help = IngestFileHelper(self.confdata)
sym_path = os.path.join(
str(self.confdata['HUBMAP_WEBSERVICE_FILEPATH']),
derived_dataset['uuid'])
new_directory_path = file_help.get_dataset_directory_absolute_path(
derived_dataset, derived_dataset['group_uuid'],
derived_dataset['uuid'])
new_path = IngestFileHelper.make_directory(new_directory_path,
sym_path)
try:
x = threading.Thread(target=file_help.set_dir_permissions,
args=['consortium', new_path])
x.start()
except Exception as e:
logger.error(e, exc_info=True)
response_data = {
'derived_dataset_uuid': derived_dataset['uuid'],
'group_uuid': derived_dataset['group_uuid'],
'group_display_name': derived_dataset['group_name'],
'full_path': new_path
}
return response_data