feat(06-02): wire CorrelationIDMiddleware + config fields + promote 5 test stubs

- backend/config.py: add log_level: str = "INFO" and log_json: bool = False
  fields under Observability (Phase 6 — D-01) comment; pydantic-settings reads
  LOG_LEVEL / LOG_JSON env vars automatically
- backend/main.py: add imports (uuid, time, structlog, ASGIApp/Receive/Scope/Send,
  setup_logging); add CorrelationIDMiddleware raw-ASGI class (NOT BaseHTTPMiddleware
  — avoids streaming buffering); call setup_logging() as first lifespan statement;
  register CorrelationIDMiddleware LAST so it runs FIRST (Starlette reverse order)
- backend/tests/test_logging.py: remove all 5 xfail decorators; replace single-line
  bodies with real assertions for JSON renderer, contextvar binding, X-Correlation-ID
  header, no-bleed between requests, uvicorn.access propagate=False
This commit is contained in:
curo1305
2026-06-03 18:47:49 +02:00
parent 9fa74a91f5
commit abe8f8ee90
3 changed files with 143 additions and 17 deletions
+67 -1
View File
@@ -1,6 +1,9 @@
import asyncio
import time
import uuid
from contextlib import asynccontextmanager
import structlog
from redis import asyncio as aioredis
from fastapi import FastAPI, Request, Response
from fastapi.middleware.cors import CORSMiddleware
@@ -12,12 +15,14 @@ from slowapi.middleware import SlowAPIMiddleware
from sqlalchemy import text
from starlette.middleware.base import BaseHTTPMiddleware
from starlette.responses import Response as StarletteResponse
from starlette.types import ASGIApp, Receive, Scope, Send
from api.auth import limiter as auth_limiter
from api.documents import router as documents_router
from api.topics import router as topics_router
from config import settings
from db.session import AsyncSessionLocal, engine
from services.logging import setup_logging
# ── CSP / Security headers middleware ────────────────────────────────────────
@@ -62,6 +67,60 @@ class OriginValidationMiddleware(BaseHTTPMiddleware):
return await call_next(request)
# ── Correlation ID Middleware ─────────────────────────────────────────────────
class CorrelationIDMiddleware:
"""Generate per-request UUID correlation ID; bind to structlog contextvars.
Uses raw ASGI (NOT BaseHTTPMiddleware) to avoid response-body buffering
issues with streaming responses (RESEARCH.md Anti-Patterns).
Execution order (Starlette reverse-insertion):
Registered LAST in app.add_middleware() so it runs FIRST on every request.
Per-request:
1. clear_contextvars() — Pitfall 2 guard; prevents context bleed between
requests handled by the same worker.
2. bind_contextvars(correlation_id, path, method) — available in all
downstream log calls.
3. Appends X-Correlation-ID to the response headers.
4. After response: bind duration_ms for final log emission.
"""
def __init__(self, app: ASGIApp) -> None:
self.app = app
async def __call__(self, scope: Scope, receive: Receive, send: Send) -> None:
if scope["type"] != "http":
await self.app(scope, receive, send)
return
correlation_id = str(uuid.uuid4())
start_ns = time.perf_counter_ns()
# Pitfall 2: MUST be first — clears any context from a prior request
# on the same worker before binding new values.
structlog.contextvars.clear_contextvars()
structlog.contextvars.bind_contextvars(
correlation_id=correlation_id,
path=scope.get("path", ""),
method=scope.get("method", ""),
)
async def send_with_header(message: dict) -> None:
if message["type"] == "http.response.start":
headers = list(message.get("headers", []))
headers.append((b"x-correlation-id", correlation_id.encode()))
message = {**message, "headers": headers}
await send(message)
await self.app(scope, receive, send_with_header)
# Bind duration after the response so downstream loggers can emit it.
duration_ms = (time.perf_counter_ns() - start_ns) / 1_000_000
structlog.contextvars.bind_contextvars(duration_ms=round(duration_ms, 2))
# ── Lifespan ──────────────────────────────────────────────────────────────────
@asynccontextmanager
@@ -74,6 +133,10 @@ async def lifespan(app: FastAPI):
TOTP replay prevention (AUTH-08).
Admin bootstrap (D-04): idempotent, runs only if no users exist.
"""
# Initialize structured logging first — all subsequent log calls use the
# configured renderer (JSON in production, console in dev).
setup_logging(json_logs=settings.log_json, log_level=settings.log_level)
# MinIO bucket initialization (RESEARCH.md Pattern 4)
minio_client = Minio(
settings.minio_endpoint,
@@ -127,9 +190,12 @@ app.add_middleware(
allow_headers=["*"],
)
# 3. Origin validation — runs first (added last), before CORS and route handlers
# 3. Origin validation — runs second (added second-to-last), before CORS and route handlers
app.add_middleware(OriginValidationMiddleware)
# 4. CorrelationID — added LAST so it runs FIRST (Starlette reverse-insertion order)
app.add_middleware(CorrelationIDMiddleware)
# ── Routes ────────────────────────────────────────────────────────────────────