litellm/tests/load_tests/test_datadog_load_test.py
yuneng-jiang 6a0d03914c
test: drop the cwd-relative sys.path.insert calls from the test suite (#37802)
* test: drop the cwd-relative sys.path.insert calls from the test suite

TQ003 stands at 1,077 across 1,058 files, and 1,015 of them are the same shape:
sys.path.insert(0, os.path.abspath("../..")) and its deeper siblings. The
argument resolves against the working directory rather than the file, so from
the repo root, where every job runs pytest, it inserts the directory two levels
above the checkout. It has never pointed at litellm. The package is installed
into the environment anyway, which is what actually makes the import work, and
what the rule's message has said all along.

Removing them leaves 1,634 imports of sys and os with no remaining reference,
and those go too, except where another test module imports the name back out of
the file. The rest of TQ003 is 62 call sites that resolve against __file__ or a
variable, which are a different question and are left alone.

Collection is identical either way: 45,871 tests and the same 51 pre-existing
collection errors before and after, and ruff reports no new undefined name.

* test: drop the duplicate imports the sys.path sweep exposed to F811

* test(pre-call-utils): restore the os import the new bedrock tests need
2026-08-22 09:25:58 -07:00

102 lines
3.2 KiB
Python

import os
import asyncio
import litellm
import pytest
import logging
from litellm._logging import verbose_logger
def test_datadog_logging_async():
try:
# litellm.set_verbose = True
os.environ["DD_API_KEY"] = "anything"
os.environ["_DATADOG_BASE_URL"] = (
"https://exampleopenaiendpoint-production.up.railway.app"
)
os.environ["DD_SITE"] = "us5.datadoghq.com"
os.environ["DD_API_KEY"] = "xxxxxx"
litellm.success_callback = ["datadog"]
percentage_diffs = []
for run in range(1):
print(f"\nRun {run + 1}:")
# Test with empty success_callback
litellm.success_callback = []
litellm.callbacks = []
start_time_empty_callback = asyncio.run(make_async_calls())
print("Done with no callback test")
# Test with datadog callback
print("Starting datadog test")
litellm.success_callback = ["datadog"]
start_time_datadog = asyncio.run(make_async_calls())
print("Done with datadog test")
# Compare times and calculate percentage difference
print(f"Time with success_callback='datadog': {start_time_datadog}")
print(f"Time with empty success_callback: {start_time_empty_callback}")
percentage_diff = (
abs(start_time_datadog - start_time_empty_callback)
/ start_time_empty_callback
* 100
)
percentage_diffs.append(percentage_diff)
print(f"Performance difference: {percentage_diff:.2f}%")
print("percentage_diffs", percentage_diffs)
avg_percentage_diff = sum(percentage_diffs) / len(percentage_diffs)
print(f"\nAverage performance difference: {avg_percentage_diff:.2f}%")
assert (
avg_percentage_diff < 10
), f"Average performance difference of {avg_percentage_diff:.2f}% exceeds 10% threshold"
except litellm.Timeout:
pass
except Exception as e:
pytest.fail(f"An exception occurred - {e}")
async def make_async_calls(metadata=None, **completion_kwargs):
total_tasks = 300
batch_size = 100
total_time = 0
for batch in range(1):
tasks = [create_async_task() for _ in range(batch_size)]
start_time = asyncio.get_event_loop().time()
responses = await asyncio.gather(*tasks)
for idx, response in enumerate(responses):
print(f"Response from Task {batch * batch_size + idx + 1}: {response}")
await asyncio.sleep(7)
batch_time = asyncio.get_event_loop().time() - start_time
total_time += batch_time
return total_time
def create_async_task(**completion_kwargs):
litellm.set_verbose = True
completion_args = {
"model": "openai/chatgpt-v-3",
"api_version": "2024-02-01",
"messages": [{"role": "user", "content": "This is a test"}],
"max_tokens": 5,
"temperature": 0.7,
"timeout": 5,
"user": "datadog_latency_test_user",
"mock_response": "hello from my load test",
}
completion_args.update(completion_kwargs)
return asyncio.create_task(litellm.acompletion(**completion_args))