Fakes

The fakes module exists to help application developers using the OpenStack SDK to unit test their applications. It provides a number of helper utilities to generate fake Resource and Proxy instances. These fakes do not require an established connection and allow you to validate that your application using valid attributes and methods for both Resource and Proxy instances.

openstack.test.fakes.generate_fake_resource(resource_type, **attrs)

Generate a fake resource

Example usage:

>>> from openstack.compute.v2 import server
>>> from openstack.test import fakes
>>> fakes.generate_fake_resource(server.Server)
openstack.compute.v2.server.Server(...)
Parameters:

  • resource_type – Object class

  • attrs – Optional attributes to be set on resource

Returns:

Instance of resource_type class populated with fake values of expected types

Raises:

NotImplementedError – If a resource attribute specifies a type or list_type that cannot be automatically generated

openstack.test.fakes.generate_fake_resources(resource_type, count=1, attrs=None)

Generate a given number of fake resource entities

Example usage:

>>> from openstack.compute.v2 import server
>>> from openstack.test import fakes
>>> fakes.generate_fake_resources(server.Server, count=3)
<generator object generate_fake_resources at 0x7f075dc65040>
Parameters:

  • resource_type – Object class

  • count – Number of objects to return

  • attrs – Attribute values to set into each instance

Returns:

Generator of resource_type class instances populated with fake values of expected types.

openstack.test.fakes.generate_fake_proxy(service, api_version=None)

Generate a fake proxy for the given service type

Proxy methods whose return type annotation is a Resource subclass (or a Generator / optional thereof) are automatically configured to return fake resource instances via generate_fake_resource() or generate_fake_resources(). Return values can still be overridden on the returned mock after the fact.

Example usage:

>>> from openstack.compute import compute_service
>>> from openstack.compute.v2 import server
>>> from openstack.test import fakes
>>> fake_compute_proxy = fakes.generate_fake_proxy(
...    compute_service.ComputeService,
... )
>>> # get_server is pre-configured; returns a fake Server
>>> fake_server = fake_compute_proxy.get_server('some-uuid')
>>> isinstance(fake_server, server.Server)
True
>>> # servers() is pre-configured; yields fake Server instances
>>> fake_servers = list(fake_compute_proxy.servers())
>>> isinstance(fake_servers[0], server.Server)
True
>>> fake_compute_proxy.serverssss()
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/lib64/python3.11/unittest/mock.py", line 653, in __getattr__
    raise AttributeError("Mock object has no attribute %r" % name)
AttributeError: Mock object has no attribute 'serverssss'. Did you mean: 'server_ips'?
Parameters:

  • service (ServiceDescription) – The service to generate the fake proxy for.

  • api_version (int or None) – The API version to generate the fake proxy for. This should be a major version must be supported by openstacksdk, as specified in the supported_versions attribute of the provided service. This is only required if openstacksdk supports multiple API versions for the given service.

Raises:

ValueError – if the service is not a valid ServiceDescription or if api_version is not supported

Returns:

An autospecced mock of the Proxy implementation for the specified service type and API version