Spaces:
Configuration error
Configuration error
File size: 3,559 Bytes
447ebeb |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 |
# What this tests ?
## Tests /config/update + Test /chat/completions -> assert logs are sent to Langfuse
import pytest
import asyncio
import aiohttp
import os
import dotenv
from dotenv import load_dotenv
import pytest
load_dotenv()
async def config_update(session):
url = "http://0.0.0.0:4000/config/update"
headers = {"Authorization": "Bearer sk-1234", "Content-Type": "application/json"}
data = {
"litellm_settings": {
"success_callback": ["langfuse"],
},
"environment_variables": {
"LANGFUSE_HOST": os.environ["LANGFUSE_HOST"],
"LANGFUSE_PUBLIC_KEY": os.environ["LANGFUSE_PUBLIC_KEY"],
"LANGFUSE_SECRET_KEY": os.environ["LANGFUSE_SECRET_KEY"],
},
}
async with session.post(url, headers=headers, json=data) as response:
status = response.status
response_text = await response.text()
print(response_text)
print()
if status != 200:
raise Exception(f"Request did not return a 200 status code: {status}")
return await response.json()
async def chat_completion(session, key, model="azure-gpt-3.5", request_metadata=None):
url = "http://0.0.0.0:4000/chat/completions"
headers = {
"Authorization": f"Bearer {key}",
"Content-Type": "application/json",
}
data = {
"model": model,
"messages": [
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "Hello!"},
],
"metadata": request_metadata,
}
print("data sent in test=", data)
async with session.post(url, headers=headers, json=data) as response:
status = response.status
response_text = await response.text()
print(response_text)
print()
if status != 200:
raise Exception(f"Request did not return a 200 status code: {status}")
@pytest.mark.asyncio
@pytest.mark.skip(
reason="langfuse apis are flaky, we unit test team / key based logging in test_langfuse_unit_tests.py"
)
async def test_team_logging():
"""
1. Add Langfuse as a callback with /config/update
2. Call /chat/completions
3. Assert the logs are sent to Langfuse
"""
try:
async with aiohttp.ClientSession() as session:
# Add Langfuse as a callback with /config/update
await config_update(session)
# 2. Call /chat/completions with a specific trace id
import uuid
_trace_id = f"trace-{uuid.uuid4()}"
_request_metadata = {
"trace_id": _trace_id,
}
await chat_completion(
session,
key="sk-1234",
model="fake-openai-endpoint",
request_metadata=_request_metadata,
)
# Test - if the logs were sent to the correct team on langfuse
import langfuse
langfuse_client = langfuse.Langfuse(
host=os.getenv("LANGFUSE_HOST"),
public_key=os.getenv("LANGFUSE_PUBLIC_KEY"),
secret_key=os.getenv("LANGFUSE_SECRET_KEY"),
)
await asyncio.sleep(10)
print(f"searching for trace_id={_trace_id} on langfuse")
generations = langfuse_client.get_generations(trace_id=_trace_id).data
# 1 generation with this trace id
assert len(generations) == 1
except Exception as e:
pytest.fail("Team 2 logging failed: " + str(e))
|