From 13d20036cf3ac351be613e61ab9551678af22992 Mon Sep 17 00:00:00 2001 From: mateo Date: Thu, 17 Sep 2026 20:08:08 +0000 Subject: [PATCH] chore(tests): remove fully commented-out proxy test files and their CI entries Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --- .github/workflows/test-unit-proxy-db.yml | 4 - .../test_model_response_typing/server.py | 23 -- .../test_model_response_typing/test.py | 14 - .../test_model_response_typing/server.py | 23 -- .../test_model_response_typing/test.py | 14 - tests/proxy_unit_tests/test_proxy_gunicorn.py | 61 ---- .../test_proxy_server_keys.py | 269 ------------------ .../test_proxy_server_spend.py | 82 ------ 8 files changed, 490 deletions(-) delete mode 100644 tests/local_testing/test_model_response_typing/server.py delete mode 100644 tests/local_testing/test_model_response_typing/test.py delete mode 100644 tests/proxy_unit_tests/test_model_response_typing/server.py delete mode 100644 tests/proxy_unit_tests/test_model_response_typing/test.py delete mode 100644 tests/proxy_unit_tests/test_proxy_gunicorn.py delete mode 100644 tests/proxy_unit_tests/test_proxy_server_keys.py delete mode 100644 tests/proxy_unit_tests/test_proxy_server_spend.py diff --git a/.github/workflows/test-unit-proxy-db.yml b/.github/workflows/test-unit-proxy-db.yml index 3725e0f5805..32080dfec7f 100644 --- a/.github/workflows/test-unit-proxy-db.yml +++ b/.github/workflows/test-unit-proxy-db.yml @@ -110,8 +110,6 @@ jobs: - test-group: proxy-server-core test-path: >- tests/proxy_unit_tests/test_proxy_server.py - tests/proxy_unit_tests/test_proxy_server_keys.py - tests/proxy_unit_tests/test_proxy_server_spend.py tests/proxy_unit_tests/test_aproxy_startup.py workers: 4 dist: loadscope @@ -120,7 +118,6 @@ jobs: test-path: >- tests/proxy_unit_tests/test_proxy_config_unit_test.py tests/proxy_unit_tests/test_proxy_routes.py - tests/proxy_unit_tests/test_proxy_gunicorn.py tests/proxy_unit_tests/test_server_root_path.py tests/proxy_unit_tests/test_proxy_pass_user_config.py tests/proxy_unit_tests/test_proxy_token_counter.py @@ -198,7 +195,6 @@ jobs: tests/proxy_unit_tests/test_realtime_cache.py tests/proxy_unit_tests/test_proxy_exception_mapping.py tests/proxy_unit_tests/test_custom_tokenizer_bug.py - tests/proxy_unit_tests/test_model_response_typing workers: 4 dist: loadscope timeout: 15 diff --git a/tests/local_testing/test_model_response_typing/server.py b/tests/local_testing/test_model_response_typing/server.py deleted file mode 100644 index 80dbc33affd..00000000000 --- a/tests/local_testing/test_model_response_typing/server.py +++ /dev/null @@ -1,23 +0,0 @@ -# #### What this tests #### -# # This tests if the litellm model response type is returnable in a flask app - -# import sys, os -# import traceback -# from flask import Flask, request, jsonify, abort, Response -# sys.path.insert(0, os.path.abspath('../../..')) # Adds the parent directory to the system path - -# import litellm -# from litellm import completion - -# litellm.set_verbose = False - -# app = Flask(__name__) - -# @app.route('/') -# def hello(): -# data = request.json -# return completion(**data) - -# if __name__ == '__main__': -# from waitress import serve -# serve(app, host='localhost', port=8080, threads=10) diff --git a/tests/local_testing/test_model_response_typing/test.py b/tests/local_testing/test_model_response_typing/test.py deleted file mode 100644 index 46bf5fbb44b..00000000000 --- a/tests/local_testing/test_model_response_typing/test.py +++ /dev/null @@ -1,14 +0,0 @@ -# import requests, json - -# BASE_URL = 'http://localhost:8080' - -# def test_hello_route(): -# data = {"model": "claude-3-5-haiku-20241022", "messages": [{"role": "user", "content": "hey, how's it going?"}]} -# headers = {'Content-Type': 'application/json'} -# response = requests.get(BASE_URL, headers=headers, data=json.dumps(data)) -# print(response.text) -# assert response.status_code == 200 -# print("Hello route test passed!") - -# if __name__ == '__main__': -# test_hello_route() diff --git a/tests/proxy_unit_tests/test_model_response_typing/server.py b/tests/proxy_unit_tests/test_model_response_typing/server.py deleted file mode 100644 index 80dbc33affd..00000000000 --- a/tests/proxy_unit_tests/test_model_response_typing/server.py +++ /dev/null @@ -1,23 +0,0 @@ -# #### What this tests #### -# # This tests if the litellm model response type is returnable in a flask app - -# import sys, os -# import traceback -# from flask import Flask, request, jsonify, abort, Response -# sys.path.insert(0, os.path.abspath('../../..')) # Adds the parent directory to the system path - -# import litellm -# from litellm import completion - -# litellm.set_verbose = False - -# app = Flask(__name__) - -# @app.route('/') -# def hello(): -# data = request.json -# return completion(**data) - -# if __name__ == '__main__': -# from waitress import serve -# serve(app, host='localhost', port=8080, threads=10) diff --git a/tests/proxy_unit_tests/test_model_response_typing/test.py b/tests/proxy_unit_tests/test_model_response_typing/test.py deleted file mode 100644 index 46bf5fbb44b..00000000000 --- a/tests/proxy_unit_tests/test_model_response_typing/test.py +++ /dev/null @@ -1,14 +0,0 @@ -# import requests, json - -# BASE_URL = 'http://localhost:8080' - -# def test_hello_route(): -# data = {"model": "claude-3-5-haiku-20241022", "messages": [{"role": "user", "content": "hey, how's it going?"}]} -# headers = {'Content-Type': 'application/json'} -# response = requests.get(BASE_URL, headers=headers, data=json.dumps(data)) -# print(response.text) -# assert response.status_code == 200 -# print("Hello route test passed!") - -# if __name__ == '__main__': -# test_hello_route() diff --git a/tests/proxy_unit_tests/test_proxy_gunicorn.py b/tests/proxy_unit_tests/test_proxy_gunicorn.py deleted file mode 100644 index 73e368d35a5..00000000000 --- a/tests/proxy_unit_tests/test_proxy_gunicorn.py +++ /dev/null @@ -1,61 +0,0 @@ -# #### What this tests #### -# # Allow the user to easily run the local proxy server with Gunicorn -# # LOCAL TESTING ONLY -# import sys, os, subprocess -# import traceback -# from dotenv import load_dotenv - -# load_dotenv() -# import os, io - -# # this file is to test litellm/proxy - -# sys.path.insert( -# 0, os.path.abspath("../..") -# ) # Adds the parent directory to the system path -# import pytest -# import litellm - -# ### LOCAL Proxy Server INIT ### -# from litellm.proxy.proxy_server import save_worker_config # Replace with the actual module where your FastAPI router is defined -# filepath = os.path.dirname(os.path.abspath(__file__)) -# config_fp = f"{filepath}/test_configs/test_config_custom_auth.yaml" -# def get_openai_info(): -# return { -# "api_key": os.getenv("AZURE_API_KEY"), -# "api_base": os.getenv("AZURE_API_BASE"), -# } - -# def run_server(host="0.0.0.0",port=8008,num_workers=None): -# if num_workers is None: -# # Set it to min(8,cpu_count()) -# import multiprocessing -# num_workers = min(4,multiprocessing.cpu_count()) - -# ### LOAD KEYS ### - -# # Load the Azure keys. For now get them from openai-usage -# azure_info = get_openai_info() -# print(f"Azure info:{azure_info}") -# os.environ["AZURE_API_KEY"] = azure_info['api_key'] -# os.environ["AZURE_API_BASE"] = azure_info['api_base'] -# os.environ["AZURE_API_VERSION"] = "2023-09-01-preview" - -# ### SAVE CONFIG ### - -# os.environ["WORKER_CONFIG"] = config_fp - -# # In order for the app to behave well with signals, run it with gunicorn -# # The first argument must be the "name of the command run" -# cmd = f"gunicorn litellm.proxy.proxy_server:app --workers {num_workers} --worker-class uvicorn.workers.UvicornWorker --bind {host}:{port}" -# cmd = cmd.split() -# print(f"Running command: {cmd}") -# import sys -# sys.stdout.flush() -# sys.stderr.flush() - -# # Make sure to propage env variables -# subprocess.run(cmd) # This line actually starts Gunicorn - -# if __name__ == "__main__": -# run_server() diff --git a/tests/proxy_unit_tests/test_proxy_server_keys.py b/tests/proxy_unit_tests/test_proxy_server_keys.py deleted file mode 100644 index 717eec921b7..00000000000 --- a/tests/proxy_unit_tests/test_proxy_server_keys.py +++ /dev/null @@ -1,269 +0,0 @@ -# import sys, os, time -# import traceback -# from dotenv import load_dotenv - -# load_dotenv() -# import os, io - -# # this file is to test litellm/proxy - -# sys.path.insert( -# 0, os.path.abspath("../..") -# ) # Adds the parent directory to the system path -# import pytest, logging -# import litellm -# from litellm import embedding, completion, completion_cost, Timeout -# from litellm import RateLimitError - - -# import sys, os, time -# import traceback -# from dotenv import load_dotenv - -# load_dotenv() -# import os, io - -# # this file is to test litellm/proxy -# from concurrent.futures import ThreadPoolExecutor - -# sys.path.insert( -# 0, os.path.abspath("../..") -# ) # Adds the parent directory to the system path - -# import pytest, logging, requests -# import litellm -# from litellm import embedding, completion, completion_cost, Timeout -# from litellm import RateLimitError -# from github import Github -# import subprocess - - -# # Function to execute a command and return the output -# def run_command(command): -# process = subprocess.Popen(command, stdout=subprocess.PIPE, shell=True) -# output, _ = process.communicate() -# return output.decode().strip() - - -# # Retrieve the current branch name -# branch_name = run_command("git rev-parse --abbrev-ref HEAD") - -# # GitHub personal access token (with repo scope) or use username and password -# access_token = os.getenv("GITHUB_ACCESS_TOKEN") -# # Instantiate the PyGithub library's Github object -# g = Github(access_token) - -# # Provide the owner and name of the repository where the pull request is located -# repository_owner = "BerriAI" -# repository_name = "litellm" - -# # Get the repository object -# repo = g.get_repo(f"{repository_owner}/{repository_name}") - -# # Iterate through the pull requests to find the one related to your branch -# for pr in repo.get_pulls(): -# print(f"in here! {pr.head.ref}") -# if pr.head.ref == branch_name: -# pr_number = pr.number -# break - -# print(f"The pull request number for branch {branch_name} is: {pr_number}") - - -# def test_add_new_key(): -# max_retries = 3 -# retry_delay = 10 # seconds - -# for retry in range(max_retries + 1): -# try: -# # Your test data -# test_data = { -# "models": ["gpt-3.5-turbo", "gpt-4", "claude-2", "azure-model"], -# "aliases": {"mistral-7b": "gpt-3.5-turbo"}, -# "duration": "20m", -# } -# print("testing proxy server") - -# # Your bearer token -# token = os.getenv("PROXY_MASTER_KEY") -# headers = {"Authorization": f"Bearer {token}"} - -# endpoint = f"https://litellm-litellm-pr-{pr_number}.up.railway.app" - -# # Make a request to the staging endpoint -# response = requests.post( -# endpoint + "/key/generate", json=test_data, headers=headers -# ) - -# print(f"response: {response.text}") - -# if response.status_code == 200: -# result = response.json() -# break # Successful response, exit the loop -# elif response.status_code == 503 and retry < max_retries: -# print( -# f"Retrying in {retry_delay} seconds... (Retry {retry + 1}/{max_retries})" -# ) -# time.sleep(retry_delay) -# else: -# assert False, f"Unexpected response status code: {response.status_code}" - -# except Exception as e: -# print(traceback.format_exc()) -# pytest.fail(f"An error occurred {e}") - - -# def test_update_new_key(): -# try: -# # Your test data -# test_data = { -# "models": ["gpt-3.5-turbo", "gpt-4", "claude-2", "azure-model"], -# "aliases": {"mistral-7b": "gpt-3.5-turbo"}, -# "duration": "20m", -# } -# print("testing proxy server") -# # Your bearer token -# token = os.getenv("PROXY_MASTER_KEY") -# headers = {"Authorization": f"Bearer {token}"} - -# endpoint = f"https://litellm-litellm-pr-{pr_number}.up.railway.app" - -# # Make a request to the staging endpoint -# response = requests.post( -# endpoint + "/key/generate", json=test_data, headers=headers -# ) -# assert response.status_code == 200 -# result = response.json() -# assert result["key"].startswith("sk-") - -# def _post_data(): -# json_data = {"models": ["bedrock-models"], "key": result["key"]} -# response = requests.post( -# endpoint + "/key/generate", json=json_data, headers=headers -# ) -# print(f"response text: {response.text}") -# assert response.status_code == 200 -# return response - -# _post_data() -# print(f"Received response: {result}") -# except Exception as e: -# pytest.fail(f"LiteLLM Proxy test failed. Exception: {str(e)}") - -# def test_add_new_key_max_parallel_limit(): -# try: -# # Your test data -# test_data = {"duration": "20m", "max_parallel_requests": 1} -# # Your bearer token -# token = os.getenv("PROXY_MASTER_KEY") -# headers = {"Authorization": f"Bearer {token}"} - -# endpoint = f"https://litellm-litellm-pr-{pr_number}.up.railway.app" -# print(f"endpoint: {endpoint}") -# # Make a request to the staging endpoint -# response = requests.post( -# endpoint + "/key/generate", json=test_data, headers=headers -# ) -# assert response.status_code == 200 -# result = response.json() - -# # load endpoint with model -# model_data = { -# "model_name": "azure-model", -# "litellm_params": { -# "model": "azure/gpt-4.1-mini", -# "api_key": os.getenv("AZURE_API_KEY"), -# "api_base": os.getenv("AZURE_API_BASE"), -# "api_version": os.getenv("AZURE_API_VERSION") -# } -# } -# response = requests.post(endpoint + "/model/new", json=model_data, headers=headers) -# assert response.status_code == 200 -# print(f"response text: {response.text}") - - -# def _post_data(): -# json_data = { -# "model": "azure-model", -# "messages": [ -# { -# "role": "user", -# "content": f"this is a test request, write a short poem {time.time()}", -# } -# ], -# } -# # Your bearer token -# response = requests.post( -# endpoint + "/chat/completions", json=json_data, headers={"Authorization": f"Bearer {result['key']}"} -# ) -# return response - -# def _run_in_parallel(): -# with ThreadPoolExecutor(max_workers=2) as executor: -# future1 = executor.submit(_post_data) -# future2 = executor.submit(_post_data) - -# # Obtain the results from the futures -# response1 = future1.result() -# print(f"response1 text: {response1.text}") -# response2 = future2.result() -# print(f"response2 text: {response2.text}") -# if response1.status_code == 429 or response2.status_code == 429: -# pass -# else: -# raise Exception() - -# _run_in_parallel() -# except Exception as e: -# pytest.fail(f"LiteLLM Proxy test failed. Exception: {str(e)}") - -# def test_add_new_key_max_parallel_limit_streaming(): -# try: -# # Your test data -# test_data = {"duration": "20m", "max_parallel_requests": 1} -# # Your bearer token -# token = os.getenv("PROXY_MASTER_KEY") -# headers = {"Authorization": f"Bearer {token}"} - -# endpoint = f"https://litellm-litellm-pr-{pr_number}.up.railway.app" - -# # Make a request to the staging endpoint -# response = requests.post( -# endpoint + "/key/generate", json=test_data, headers=headers -# ) -# print(f"response: {response.text}") -# assert response.status_code == 200 -# result = response.json() - -# def _post_data(): -# json_data = { -# "model": "azure-model", -# "messages": [ -# { -# "role": "user", -# "content": f"this is a test request, write a short poem {time.time()}", -# } -# ], -# "stream": True, -# } -# response = requests.post( -# endpoint + "/chat/completions", json=json_data, headers={"Authorization": f"Bearer {result['key']}"} -# ) -# return response - -# def _run_in_parallel(): -# with ThreadPoolExecutor(max_workers=2) as executor: -# future1 = executor.submit(_post_data) -# future2 = executor.submit(_post_data) - -# # Obtain the results from the futures -# response1 = future1.result() -# response2 = future2.result() -# if response1.status_code == 429 or response2.status_code == 429: -# pass -# else: -# raise Exception() - -# _run_in_parallel() -# except Exception as e: -# pytest.fail(f"LiteLLM Proxy test failed. Exception: {str(e)}") diff --git a/tests/proxy_unit_tests/test_proxy_server_spend.py b/tests/proxy_unit_tests/test_proxy_server_spend.py deleted file mode 100644 index 9fed60412ce..00000000000 --- a/tests/proxy_unit_tests/test_proxy_server_spend.py +++ /dev/null @@ -1,82 +0,0 @@ -# import openai, json, time, asyncio -# client = openai.AsyncOpenAI( -# api_key="sk-1234", -# base_url="http://0.0.0.0:8000" -# ) - -# super_fake_messages = [ -# { -# "role": "user", -# "content": f"What's the weather like in San Francisco, Tokyo, and Paris? {time.time()}" -# }, -# { -# "content": None, -# "role": "assistant", -# "tool_calls": [ -# { -# "id": "1", -# "function": { -# "arguments": "{\"location\": \"San Francisco\", \"unit\": \"celsius\"}", -# "name": "get_current_weather" -# }, -# "type": "function" -# }, -# { -# "id": "2", -# "function": { -# "arguments": "{\"location\": \"Tokyo\", \"unit\": \"celsius\"}", -# "name": "get_current_weather" -# }, -# "type": "function" -# }, -# { -# "id": "3", -# "function": { -# "arguments": "{\"location\": \"Paris\", \"unit\": \"celsius\"}", -# "name": "get_current_weather" -# }, -# "type": "function" -# } -# ] -# }, -# { -# "tool_call_id": "1", -# "role": "tool", -# "name": "get_current_weather", -# "content": "{\"location\": \"San Francisco\", \"temperature\": \"90\", \"unit\": \"celsius\"}" -# }, -# { -# "tool_call_id": "2", -# "role": "tool", -# "name": "get_current_weather", -# "content": "{\"location\": \"Tokyo\", \"temperature\": \"30\", \"unit\": \"celsius\"}" -# }, -# { -# "tool_call_id": "3", -# "role": "tool", -# "name": "get_current_weather", -# "content": "{\"location\": \"Paris\", \"temperature\": \"50\", \"unit\": \"celsius\"}" -# } -# ] - -# async def chat_completions(): -# super_fake_response = await client.chat.completions.create( -# model="gpt-3.5-turbo", -# messages=super_fake_messages, -# seed=1337, -# stream=False -# ) # get a new response from the model where it can see the function response -# await asyncio.sleep(1) -# return super_fake_response - -# async def loadtest_fn(n = 1): -# global num_task_cancelled_errors, exception_counts, chat_completions -# start = time.time() -# tasks = [chat_completions() for _ in range(n)] -# chat_completions = await asyncio.gather(*tasks) -# successful_completions = [c for c in chat_completions if c is not None] -# print(n, time.time() - start, len(successful_completions)) - -# # print(json.dumps(super_fake_response.model_dump(), indent=4)) - -# asyncio.run(loadtest_fn())