Skip to main content

Aserto API client

Project description

Aserto API client

High-level client interface to Aserto's APIs.

Authorizer

The client can be used for interacting with Aserto's Authorizer service.

Installation

Using Pip

pip install aserto

Using Poetry

poetry add aserto

Usage

from aserto.client import AuthorizerOptions, Identity
from aserto.client.authorizer import AuthorizerClient


client = AuthorizerClient(
    identity=Identity(type="NONE"),
    options=AuthorizerOptions(
        api_key=ASERTO_API_KEY,
        tenant_id=ASERTO_TENANT_ID,
    ),
)

result = await client.decision_tree(
    decisions=["visible", "enabled", "allowed"],
    policy_instance_name=ASERTO_POLICY_INSTANCE_NAME,
    policy_instance_label=ASERTO_POLICY_INSTANCE_LABEL,
    policy_path_root=ASERTO_POLICY_PATH_ROOT,
    policy_path_separator="DOT",
)

assert result == {
    "GET.your.policy.path": {
        "visible": True,
        "enabled": True,
        "allowed": False,
    },
}

Directory

The Directory APIs can be used to interact with the aserto directory services. It provides CRUD operations on objects and relations, including bulk import and export. The client can also be used to check whether a user has a permission or relation on an object instance.

Directory Client

You can initialize a directory client as follows:

from aserto.client.directory.v3 import Directory

ds = Directory(api_key="my_api_key", tenant_id="1234", address="localhost:9292")
  • address: hostname:port of directory service (required)
  • api_key: API key for directory service (required if using hosted directory)
  • tenant_id: Aserto tenant ID (required if using hosted directory)
  • cert: Path to the grpc service certificate when connecting to local topaz instance.

get_object

Get a directory object instance with the type and the id, optionally with the object's relations.

# without relations:
user = ds.get_object(object_type="user", object_id="euang@acmecorp.com")

# with relations:
page = PaginationRequest(size=10)
while True:
    resp = ds.get_object(object_type="user", object_id="euang@acmecorp.com", with_relations=True, page=page)
    user = resp.result               # The returned object.
    relations_page = resp.relations  # A page of relations.

    if not resp.page.next_token:
        # we've reached the last page.
        break

    # request the next page.
    page.token = resp.page.next_token

get_objects_many

Similar to get_object but can retrieve multiple object instances in a single request.

objects = ds.get_object_many(
    [
        ObjectIdentifier(type="user", id="euan@acmecorp.com"),
        ObjectIdentifier(type="group", id="marketing"),
    ]
)

get_objects

Get object instances with an object type type pagination info (page size and pagination token).

from aserto.client.directory.v3 import PaginationRequest

users = ds.get_objects(object_type="user", page=PaginationRequest(size=10))

set_object

Create an object instance with the specified properties. If an etag is specified and is different from the current object's etag, the call raises an ETagMismatchError.

# pass object fields as arguments:
user = ds.set_object(
    object_type="user",
    object_id="new-user@acmecorp.com",
    display_name="John Doe",
    "properties": {"active": True, "department": "Engineering"},
}

# set_object can also take an Object parameter:
user.display_name = "Jane Doe"
user.properties["title"] = "Senior Engineer"
updated_user = ds.set_object(object=user)

delete_object

Delete an object instance and optionally its relations, using its type and id:

# delete an object
ds.delete_object(object_type="user", object_id="test-object")

# delete an object and all its relations
ds.delete_object(object_type="user", object_id="test-object", with_relations=True)

get_relation

Retrieve a single relation from the directory or raise a NotFoundError if no matching relation exists.

# get the manager of euang@acmecorp.com:
relation = ds.get_relation(
    object_type="user",
    relation="manager",
    subject_type="user",
    subject_id="euang@acmecorp.com",
)

assert relation.object_id

# include the relation's object and subject in the response:
response = ds.get_relation(
    object_type="user",
    relation="manager",
    subject_type="user",
    subject_id="euang@acmecorp.com",
    with_relations=True,
)

assert response.relation.object_id
assert response.subject.display_name == "Euan Garden"
assert response.object.properties["department"] == "Sales"
#

get_relations

Searches the directory for relations matching the specified criteria, optionally including the object and subject of each returned relation.

# find all groups a user is a member of:
page = PaginationRequest(size=10)

while True:
    response = ds.get_relations(
        object_type="group",
        "relation"="member",
        "subject_type": "user",
        "subject_id": "euang@acmecorp.com",
        with_objects=True,
        page=page,
    )

    if not response.page.next_token:
        break

    page.token = response.page.next_token

set_relation

Create a new relation.

ds.set_relation(
    object_type="group",
    object_id="admin",
    relation="member",
    subject_type="user",
    subject_id="euang@acmecorp.com",
)

delete_relation

Delete a relation.

ds.delete_relation(
    object_type="group",
    object_id="admin",
    relation="member",
    subject_type="user",
    subject_id="euang@acmecorp.com",
)

check

Check if a subject has a given relation or permission on an object.

allowed = ds.check(
    object_type="folder",
    object_id="/path/to/folder",
    relation="can_delete",
    subject_type="user",
    subject_id="euang@acmecorp.com",
)

find_subjects

Find subjects that have a given relation to or permission on a specified object.

reponse = ds.find_subjects(
    object_type="folder",
    object_id="/path/to/folder",
    relation="can_delete",
    subject_type="user"
)

assert ObjectIdentifier("user", "euang@acmecorp.com") in response.results

find_objects

Find objects that a given subject has a specified relation to or permission on.

reponse = ds.find_objects(
    object_type="folder",
    relation="can_delete",
    subject_type="user"
    subjecct_id="euang@acmecorp.com"
)

assert ObjectIdentifier("folder", "/path/to/folder") in response.results

get_manifest

Download the directory manifest.

manifest = ds.get_manifest()

print(manifest.body)    # yaml manifest

# conditionally get the manifest if its etag has changed
new_manifest = ds.get_manifest(etag=manifest.etag)

assert new_manifest is None   # the manifest hasn't changed

set_manifest

Upload a new directory manifest.

with open("manifest.yaml", "rb") as f:
    manifest = f.read()

ds.set_manifest(manifest)

import_data

Bulk-insert objects and/or relations to the directory. Returns a summary of the number of objects/relations affected.

# import an object and a relation.
data = [
    Object(type="user", id="test@acmecorp.com"),
    Relation(
        object_type="user",
        object_id="euang@acmecorp.com",
        relation="manager",
        subject_type="user",
        subject_id="test@acmecorp.com",
    ),
]

response = ds.import_data(data)

assert response.objects.set == 1
assert response.object.error == 0
assert response.relations.set == 1
assert response.relations.error == 0

export_data

Bulk-retrieve objects and/or relations from the directory.

from aserto.client.directory.v3 import ExportOption, Object, Relation

# export all objects and relations
for item in ds.export(ExportOption.OPTION_DATA):
    if isinstance(item, Object):
        print("object:", item)
    elif isinstance(item, Relation):
        print("relation:", item)

Async Directory Client

You can initialize an asynchronous directory client as follows:

from aserto.client.directory.v3.aio import Directory

ds = Directory(api_key="my_api_key", tenant_id="1234", address="localhost:9292")

The methods on the async directory have the same signatures as their synchronous counterparts.

License

This project is licensed under the MIT license. See the LICENSE file for more info.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

aserto-0.32.0.tar.gz (19.8 kB view details)

Uploaded Source

Built Distribution

aserto-0.32.0-py3-none-any.whl (24.4 kB view details)

Uploaded Python 3

File details

Details for the file aserto-0.32.0.tar.gz.

File metadata

  • Download URL: aserto-0.32.0.tar.gz
  • Upload date:
  • Size: 19.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: poetry/1.8.3 CPython/3.12.7 Linux/6.5.0-1025-azure

File hashes

Hashes for aserto-0.32.0.tar.gz
Algorithm Hash digest
SHA256 058127fce79b67dca6802180e0289653e5eece4ae92859ac7303e4715c7d97bc
MD5 bd79be1fcc999b3039ba329407d156b9
BLAKE2b-256 de0f027e0dff9236fa57e15b4a4d76ee6dfe28d188dac804a8c2de8d18ca8521

See more details on using hashes here.

File details

Details for the file aserto-0.32.0-py3-none-any.whl.

File metadata

  • Download URL: aserto-0.32.0-py3-none-any.whl
  • Upload date:
  • Size: 24.4 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: poetry/1.8.3 CPython/3.12.7 Linux/6.5.0-1025-azure

File hashes

Hashes for aserto-0.32.0-py3-none-any.whl
Algorithm Hash digest
SHA256 f7e57b9ce8f5f76335e2f7f2741200eca26c9fafa12fbf532c98297d2f464667
MD5 f93794920b18aeedceb0c5f2bbeaf799
BLAKE2b-256 22898c2c5c92bf246453d76081f87243a6114f9d5ad1edc5fb2fd3a8cfbafd02

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page