mirror of https://github.com/jumpserver/jumpserver
49 lines
1.6 KiB
Python
49 lines
1.6 KiB
Python
|
from collections import defaultdict
|
||
|
|
||
|
from django.db.models import Q
|
||
|
|
||
|
from common.utils import get_logger
|
||
|
from perms.models import AssetPermission
|
||
|
from perms.hands import Asset, User, UserGroup
|
||
|
from perms.models.base import BasePermissionQuerySet
|
||
|
|
||
|
logger = get_logger(__file__)
|
||
|
|
||
|
|
||
|
def get_asset_system_users_id_with_actions(asset_perm_queryset: BasePermissionQuerySet, asset: Asset):
|
||
|
nodes = asset.get_nodes()
|
||
|
node_keys = set()
|
||
|
for node in nodes:
|
||
|
ancestor_keys = node.get_ancestor_keys(with_self=True)
|
||
|
node_keys.update(ancestor_keys)
|
||
|
|
||
|
queryset = asset_perm_queryset.filter(
|
||
|
Q(assets=asset) |
|
||
|
Q(nodes__key__in=node_keys)
|
||
|
)
|
||
|
asset_protocols = asset.protocols_as_dict.keys()
|
||
|
values = queryset.filter(
|
||
|
system_users__protocol__in=asset_protocols
|
||
|
).distinct().values_list('system_users', 'actions')
|
||
|
system_users_actions = defaultdict(int)
|
||
|
|
||
|
for system_user_id, actions in values:
|
||
|
if None in (system_user_id, actions):
|
||
|
continue
|
||
|
system_users_actions[system_user_id] |= actions
|
||
|
return system_users_actions
|
||
|
|
||
|
|
||
|
def get_asset_system_users_id_with_actions_by_user(user: User, asset: Asset):
|
||
|
queryset = AssetPermission.objects.filter(
|
||
|
Q(users=user) | Q(user_groups__users=user)
|
||
|
).valid()
|
||
|
return get_asset_system_users_id_with_actions(queryset, asset)
|
||
|
|
||
|
|
||
|
def get_asset_system_users_id_with_actions_by_group(group: UserGroup, asset: Asset):
|
||
|
queryset = AssetPermission.objects.filter(
|
||
|
user_groups=group
|
||
|
).valid()
|
||
|
return get_asset_system_users_id_with_actions(queryset, asset)
|