wehub-resource-sync
59a0a3844c
PR Test AMD / cancel-on-close (push) Has been skipped
PR Test NVIDIA ARM / scan (push) Has been skipped
PR Test NVIDIA / cancel-on-close (push) Has been skipped
PR Test AMD / scan (push) Has been skipped
PR Test NVIDIA ARM / cancel-on-close (push) Has been skipped
PR Test NVIDIA / scan (push) Has been skipped
Release Docker Images / build (cu129-torch-2.11.0) (push) Has been skipped
Release Docker Images / build (cu130-torch-2.11.0) (push) Has been skipped
Release PyPI / publish (push) Has been skipped
Scheduler Python Test / test (push) Successful in 27m19s
Docs / build (push) Successful in 28m8s
Scheduler C++ Test / test (push) Successful in 28m19s
Scheduler C++ Test / test-flat (push) Successful in 28m18s
Docs / deploy (push) Has been cancelled
PR Test AMD / finish (push) Has been cancelled
PR Test NVIDIA / finish (push) Has been cancelled
PR Test NVIDIA ARM / finish (push) Has been cancelled
PR Test NVIDIA ARM / ${{ matrix.name }} (${{ matrix.runner }}) (push) Has been cancelled
PR Test AMD / ${{ matrix.name }} (${{ matrix.runner }}) (push) Has been cancelled
PR Test NVIDIA / ${{ matrix.name }} (${{ matrix.runner }}) (push) Has been cancelled
59 lines
2.2 KiB
Python
59 lines
2.2 KiB
Python
"""Protocol-conformance guard: ``AsyncLLM`` explicitly inherits
|
|
``EngineClient``.
|
|
|
|
Locks the inheritance as a class-level invariant so a future
|
|
refactor cannot silently break protocol conformance by dropping a
|
|
base class, renaming a method, or removing an attribute the OpenAI
|
|
serving layer depends on.
|
|
|
|
The check uses ``EngineClient in AsyncLLM.__mro__`` rather than
|
|
``issubclass(AsyncLLM, EngineClient)``: ``runtime_checkable``
|
|
``Protocol`` classes with attribute members raise ``TypeError:
|
|
Protocols with non-method members don't support issubclass()``, and
|
|
``EngineClient`` mixes attributes with methods. A direct MRO
|
|
membership check works regardless of protocol shape.
|
|
|
|
Registered on ``runtime-1gpu`` only because the tokenspeed
|
|
import graph currently requires a CUDA-capable environment
|
|
(``triton`` import); the test itself does not touch the GPU.
|
|
"""
|
|
|
|
import os
|
|
import sys
|
|
import unittest
|
|
|
|
# CI registration (AST-parsed, runtime no-op).
|
|
sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
|
|
from ci_system.ci_register import register_cuda_ci # noqa: E402
|
|
|
|
register_cuda_ci(est_time=30, suite="runtime-1gpu")
|
|
|
|
from tokenspeed.runtime.engine.async_llm import AsyncLLM # noqa: E402
|
|
from tokenspeed.runtime.engine.protocol import EngineClient # noqa: E402
|
|
|
|
|
|
class TestAsyncLLMSatisfiesEngineClient(unittest.TestCase):
|
|
"""D.3 invariant: ``AsyncLLM`` inherits ``EngineClient`` explicitly.
|
|
|
|
MRO membership check keeps this test purely structural — no
|
|
live engine needed — and sidesteps ``issubclass``'s refusal to
|
|
operate on ``@runtime_checkable Protocol`` classes that carry
|
|
attribute members.
|
|
"""
|
|
|
|
def test_async_llm_mro_contains_engine_client(self) -> None:
|
|
mro_names = [cls.__name__ for cls in AsyncLLM.__mro__]
|
|
self.assertIn(
|
|
EngineClient,
|
|
AsyncLLM.__mro__,
|
|
"AsyncLLM must explicitly inherit EngineClient so the "
|
|
"OpenAI serving layer's EngineClient-typed callsites stay "
|
|
"structurally sound. If this fails, check AsyncLLM's base "
|
|
"class declaration in runtime/engine/async_llm.py. "
|
|
f"Current MRO: {mro_names}",
|
|
)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
unittest.main(verbosity=2)
|