File size: 15,538 Bytes
5f25733 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 | from __future__ import annotations
import json
import os
import queue
import shutil
import threading
import time
from dataclasses import dataclass
from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer
from pathlib import Path
from typing import Any
from openai_codex import CodexConfig
Json = dict[str, Any]
@dataclass(frozen=True)
class CapturedResponsesRequest:
"""Recorded request sent by app-server to the mock Responses API."""
method: str
path: str
headers: dict[str, str]
body: bytes
def body_json(self) -> Json:
"""Decode the request body as JSON."""
return json.loads(self.body.decode("utf-8"))
def input(self) -> list[Json]:
"""Return the Responses API input array from the request."""
value = self.body_json().get("input")
if not isinstance(value, list):
raise AssertionError(f"expected input list, got {value!r}")
return value
def message_input_texts(self, role: str) -> list[str]:
"""Return all input_text strings for message inputs matching one role."""
texts: list[str] = []
for item in self.input():
if item.get("type") != "message" or item.get("role") != role:
continue
content = item.get("content")
if isinstance(content, str):
texts.append(content)
continue
if not isinstance(content, list):
continue
for span in content:
if isinstance(span, dict) and span.get("type") == "input_text":
text = span.get("text")
if isinstance(text, str):
texts.append(text)
return texts
def message_content_items(self, role: str) -> list[Json]:
"""Return structured content items for message inputs matching one role."""
items: list[Json] = []
for item in self.input():
if item.get("type") != "message" or item.get("role") != role:
continue
content = item.get("content")
if not isinstance(content, list):
continue
items.extend(part for part in content if isinstance(part, dict))
return items
def message_image_urls(self, role: str) -> list[str]:
"""Return all input_image URLs for message inputs matching one role."""
urls: list[str] = []
for item in self.message_content_items(role):
if item.get("type") != "input_image":
continue
image_url = item.get("image_url")
if isinstance(image_url, str):
urls.append(image_url)
return urls
def header(self, name: str) -> str | None:
"""Return a captured request header by case-insensitive name."""
return self.headers.get(name.lower())
@dataclass(frozen=True)
class MockSseResponse:
"""One queued SSE response served by the mock Responses API."""
body: str
delay_between_events_s: float = 0.0
def chunks(self) -> list[bytes]:
"""Split an SSE body into event chunks while preserving framing."""
chunks: list[bytes] = []
for part in self.body.split("\n\n"):
if not part:
continue
chunks.append(f"{part}\n\n".encode("utf-8"))
return chunks
class MockResponsesServer:
"""Local HTTP server that records `/v1/responses` requests and returns SSE."""
def __init__(self) -> None:
self._responses: queue.Queue[MockSseResponse] = queue.Queue()
self._requests: list[CapturedResponsesRequest] = []
self._requests_lock = threading.Lock()
self._server = _ResponsesHttpServer(("127.0.0.1", 0), _ResponsesHandler, self)
self._thread = threading.Thread(
target=self._server.serve_forever,
name="mock-responses-api",
daemon=True,
)
def __enter__(self) -> MockResponsesServer:
self._thread.start()
return self
def __exit__(self, _exc_type: object, _exc: object, _tb: object) -> None:
self.close()
@property
def url(self) -> str:
"""Return the base URL for app-server config."""
host, port = self._server.server_address
return f"http://{host}:{port}"
def close(self) -> None:
"""Stop the background HTTP server thread."""
self._server.shutdown()
self._server.server_close()
self._thread.join(timeout=2)
def enqueue_sse(
self,
body: str,
*,
delay_between_events_s: float = 0.0,
) -> None:
"""Queue one SSE body for the next `/v1/responses` request."""
self._responses.put(
MockSseResponse(
body=body,
delay_between_events_s=delay_between_events_s,
)
)
def enqueue_assistant_message(self, text: str, *, response_id: str = "resp-1") -> None:
"""Queue a completed assistant-message model response."""
self.enqueue_sse(
sse(
[
ev_response_created(response_id),
ev_assistant_message(f"msg-{response_id}", text),
ev_completed(response_id),
]
)
)
def requests(self) -> list[CapturedResponsesRequest]:
"""Return all recorded Responses API requests."""
with self._requests_lock:
return list(self._requests)
def single_request(self) -> CapturedResponsesRequest:
"""Return the only recorded request, failing if the count differs."""
requests = self.requests()
if len(requests) != 1:
raise AssertionError(f"expected 1 request, got {len(requests)}")
return requests[0]
def wait_for_requests(
self,
count: int,
*,
timeout_s: float = 5.0,
) -> list[CapturedResponsesRequest]:
"""Wait until at least `count` requests have been recorded."""
deadline = time.monotonic() + timeout_s
while time.monotonic() < deadline:
requests = self.requests()
if len(requests) >= count:
return requests
time.sleep(0.01)
requests = self.requests()
raise AssertionError(f"expected {count} requests, got {len(requests)}")
def _record_request(self, handler: BaseHTTPRequestHandler, body: bytes) -> None:
"""Record one inbound HTTP request from app-server."""
headers = {key.lower(): value for key, value in handler.headers.items()}
request = CapturedResponsesRequest(
method=handler.command,
path=handler.path,
headers=headers,
body=body,
)
with self._requests_lock:
self._requests.append(request)
def _next_response(self) -> MockSseResponse:
"""Return the next queued SSE response or fail the HTTP request."""
return self._responses.get_nowait()
class AppServerHarness:
"""Test fixture that points the checkout's app-server at MockResponsesServer."""
def __init__(self, tmp_path: Path, *, requires_openai_auth: bool = False) -> None:
self.tmp_path = tmp_path
self.codex_home = tmp_path / "codex-home"
self.workspace = tmp_path / "workspace"
self.requires_openai_auth = requires_openai_auth
self.responses = MockResponsesServer()
def __enter__(self) -> AppServerHarness:
self.codex_home.mkdir()
self.workspace.mkdir()
self.responses.__enter__()
self._write_config()
return self
def __exit__(self, _exc_type: object, _exc: object, _tb: object) -> None:
self.responses.__exit__(_exc_type, _exc, _tb)
shutil.rmtree(self.codex_home, ignore_errors=True)
shutil.rmtree(self.workspace, ignore_errors=True)
def app_server_config(self) -> CodexConfig:
"""Prefer the CI binary, then a local debug build, then the installed runtime."""
binary_name = "codex.exe" if os.name == "nt" else "codex"
debug_binary = Path(__file__).resolve().parents[3] / "codex-rs/target/debug" / binary_name
codex_bin = os.environ.get("CODEX_EXEC_PATH")
if codex_bin is None and debug_binary.is_file():
codex_bin = str(debug_binary)
return CodexConfig(
codex_bin=codex_bin,
cwd=str(self.workspace),
env={
"CODEX_HOME": str(self.codex_home),
"CODEX_APP_SERVER_DISABLE_MANAGED_CONFIG": "1",
"RUST_LOG": "warn",
},
)
def _write_config(self) -> None:
"""Write config.toml that routes model calls to the mock server."""
config_toml = self.codex_home / "config.toml"
requires_openai_auth = "requires_openai_auth = true\n" if self.requires_openai_auth else ""
config_toml.write_text(
f"""
model = "mock-model"
approval_policy = "never"
sandbox_mode = "read-only"
model_provider = "mock_provider"
[model_providers.mock_provider]
name = "Mock provider for Python SDK tests"
base_url = "{self.responses.url}/v1"
wire_api = "responses"
request_max_retries = 0
stream_max_retries = 0
{requires_openai_auth}
""".lstrip()
)
class _ResponsesHttpServer(ThreadingHTTPServer):
"""ThreadingHTTPServer carrying a reference to the owning mock."""
def __init__(
self,
server_address: tuple[str, int],
handler_class: type[BaseHTTPRequestHandler],
mock: MockResponsesServer,
) -> None:
super().__init__(server_address, handler_class)
self.mock = mock
class _ResponsesHandler(BaseHTTPRequestHandler):
"""HTTP handler for the subset of the Responses API used by SDK tests."""
server: _ResponsesHttpServer
def log_message(self, _format: str, *_args: object) -> None:
"""Silence default stderr logging; pytest failures print captured requests."""
return None
def do_GET(self) -> None:
"""Serve a minimal `/v1/models` response if app-server asks for models."""
if self.path.endswith("/v1/models") or self.path.endswith("/models"):
self._send_json(
{
"object": "list",
"data": [
{
"id": "mock-model",
"object": "model",
"created": 0,
"owned_by": "openai",
}
],
}
)
return
self.send_error(404, f"unexpected GET {self.path}")
def do_POST(self) -> None:
"""Serve queued SSE responses for `/v1/responses` requests."""
length = int(self.headers.get("content-length", "0"))
body = self.rfile.read(length)
if self.path.endswith("/analytics/codex/turn-costs"):
# Optional cost probes are not model requests.
self.send_error(404, "turn costs are unavailable for the mock provider")
return
self.server.mock._record_request(self, body)
if not (self.path.endswith("/v1/responses") or self.path.endswith("/responses")):
self.send_error(404, f"unexpected POST {self.path}")
return
try:
response = self.server.mock._next_response()
except queue.Empty:
self.send_error(500, "no queued SSE response")
return
self.send_response(200)
self.send_header("content-type", "text/event-stream")
self.end_headers()
for chunk in response.chunks():
self.wfile.write(chunk)
self.wfile.flush()
if response.delay_between_events_s:
time.sleep(response.delay_between_events_s)
def _send_json(self, payload: Json) -> None:
"""Write one JSON response."""
body = json.dumps(payload).encode("utf-8")
self.send_response(200)
self.send_header("content-type", "application/json")
self.send_header("content-length", str(len(body)))
self.end_headers()
self.wfile.write(body)
def sse(events: list[Json]) -> str:
"""Build an SSE body from Responses API event JSON objects."""
chunks: list[str] = []
for event in events:
event_type = event["type"]
chunks.append(f"event: {event_type}\ndata: {json.dumps(event)}\n")
return "\n".join(chunks) + "\n"
def ev_response_created(response_id: str) -> Json:
"""Return a minimal `response.created` event."""
return {"type": "response.created", "response": {"id": response_id}}
def ev_completed(response_id: str) -> Json:
"""Return a minimal `response.completed` event with usage."""
return {
"type": "response.completed",
"response": {
"id": response_id,
"usage": {
"input_tokens": 1,
"input_tokens_details": None,
"output_tokens": 1,
"output_tokens_details": None,
"total_tokens": 2,
},
},
}
def ev_completed_with_usage(
response_id: str,
*,
input_tokens: int,
cached_input_tokens: int,
output_tokens: int,
reasoning_output_tokens: int,
total_tokens: int,
) -> Json:
"""Return `response.completed` with explicit token accounting."""
return {
"type": "response.completed",
"response": {
"id": response_id,
"usage": {
"input_tokens": input_tokens,
"input_tokens_details": {"cached_tokens": cached_input_tokens},
"output_tokens": output_tokens,
"output_tokens_details": {
"reasoning_tokens": reasoning_output_tokens,
},
"total_tokens": total_tokens,
},
},
}
def ev_assistant_message(item_id: str, text: str) -> Json:
"""Return a completed assistant message output item."""
return {
"type": "response.output_item.done",
"item": {
"type": "message",
"role": "assistant",
"id": item_id,
"content": [{"type": "output_text", "text": text}],
},
}
def ev_message_item_added(item_id: str, text: str = "") -> Json:
"""Return an assistant message added event before streaming deltas."""
return {
"type": "response.output_item.added",
"item": {
"type": "message",
"role": "assistant",
"id": item_id,
"content": [{"type": "output_text", "text": text}],
},
}
def ev_output_text_delta(delta: str) -> Json:
"""Return an output-text delta event."""
return {
"type": "response.output_text.delta",
"delta": delta,
}
def ev_function_call(call_id: str, name: str, arguments: str) -> Json:
"""Return a completed function-call output item."""
return {
"type": "response.output_item.done",
"item": {
"type": "function_call",
"call_id": call_id,
"name": name,
"arguments": arguments,
},
}
def ev_failed(response_id: str, message: str) -> Json:
"""Return a failed model response event."""
return {
"type": "response.failed",
"response": {
"id": response_id,
"error": {"code": "server_error", "message": message},
},
}
|