mirror of
https://github.com/meta-llama/llama-stack.git
synced 2025-06-27 18:50:41 +00:00
7c1998db25
This allows a set of rules to be defined for determining access to
resources. The rules are (loosely) based on the cedar policy format.
A rule defines a list of action either to permit or to forbid. It may
specify a principal or a resource that must match for the rule to take
effect. It may also specify a condition, either a 'when' or an 'unless',
with additional constraints as to where the rule applies.
A list of rules is held for each type to be protected and tried in order
to find a match. If a match is found, the request is permitted or
forbidden depening on the type of rule. If no match is found, the
request is denied. If no rules are specified for a given type, a rule
that allows any action as long as the resource attributes match the user
attributes is added (i.e. the previous behaviour is the default.
Some examples in yaml:
```
model:
- permit:
principal: user-1
actions: [create, read, delete]
comment: user-1 has full access to all models
- permit:
principal: user-2
actions: [read]
resource: model-1
comment: user-2 has read access to model-1 only
- permit:
actions: [read]
when:
user_in: resource.namespaces
comment: any user has read access to models with matching attributes
vector_db:
- forbid:
actions: [create, read, delete]
unless:
user_in: role::admin
comment: only user with admin role can use vector_db resources
```
---------
Signed-off-by: Gordon Sim <gsim@redhat.com>
93 lines
2.9 KiB
Python
93 lines
2.9 KiB
Python
# Copyright (c) Meta Platforms, Inc. and affiliates.
|
|
# All rights reserved.
|
|
#
|
|
# This source code is licensed under the terms described in the LICENSE file in
|
|
# the root directory of this source tree.
|
|
|
|
import uuid
|
|
from typing import Any
|
|
|
|
from llama_stack.apis.datasets import (
|
|
Dataset,
|
|
DatasetPurpose,
|
|
Datasets,
|
|
DatasetType,
|
|
DataSource,
|
|
ListDatasetsResponse,
|
|
RowsDataSource,
|
|
URIDataSource,
|
|
)
|
|
from llama_stack.apis.resource import ResourceType
|
|
from llama_stack.distribution.datatypes import (
|
|
DatasetWithOwner,
|
|
)
|
|
from llama_stack.log import get_logger
|
|
|
|
from .common import CommonRoutingTableImpl
|
|
|
|
logger = get_logger(name=__name__, category="core")
|
|
|
|
|
|
class DatasetsRoutingTable(CommonRoutingTableImpl, Datasets):
|
|
async def list_datasets(self) -> ListDatasetsResponse:
|
|
return ListDatasetsResponse(data=await self.get_all_with_type(ResourceType.dataset.value))
|
|
|
|
async def get_dataset(self, dataset_id: str) -> Dataset:
|
|
dataset = await self.get_object_by_identifier("dataset", dataset_id)
|
|
if dataset is None:
|
|
raise ValueError(f"Dataset '{dataset_id}' not found")
|
|
return dataset
|
|
|
|
async def register_dataset(
|
|
self,
|
|
purpose: DatasetPurpose,
|
|
source: DataSource,
|
|
metadata: dict[str, Any] | None = None,
|
|
dataset_id: str | None = None,
|
|
) -> Dataset:
|
|
if isinstance(source, dict):
|
|
if source["type"] == "uri":
|
|
source = URIDataSource.parse_obj(source)
|
|
elif source["type"] == "rows":
|
|
source = RowsDataSource.parse_obj(source)
|
|
|
|
if not dataset_id:
|
|
dataset_id = f"dataset-{str(uuid.uuid4())}"
|
|
|
|
provider_dataset_id = dataset_id
|
|
|
|
# infer provider from source
|
|
if metadata:
|
|
if metadata.get("provider_id"):
|
|
provider_id = metadata.get("provider_id") # pass through from nvidia datasetio
|
|
elif source.type == DatasetType.rows.value:
|
|
provider_id = "localfs"
|
|
elif source.type == DatasetType.uri.value:
|
|
# infer provider from uri
|
|
if source.uri.startswith("huggingface"):
|
|
provider_id = "huggingface"
|
|
else:
|
|
provider_id = "localfs"
|
|
else:
|
|
raise ValueError(f"Unknown data source type: {source.type}")
|
|
|
|
if metadata is None:
|
|
metadata = {}
|
|
|
|
dataset = DatasetWithOwner(
|
|
identifier=dataset_id,
|
|
provider_resource_id=provider_dataset_id,
|
|
provider_id=provider_id,
|
|
purpose=purpose,
|
|
source=source,
|
|
metadata=metadata,
|
|
)
|
|
|
|
await self.register_object(dataset)
|
|
return dataset
|
|
|
|
async def unregister_dataset(self, dataset_id: str) -> None:
|
|
dataset = await self.get_dataset(dataset_id)
|
|
if dataset is None:
|
|
raise ValueError(f"Dataset {dataset_id} not found")
|
|
await self.unregister_object(dataset)
|