Files
hyungi_document_server/tests/test_worker_recap_compaction.py
T
Hyungi Ahn eae1f48d62 feat(worker-pool): Registry-1C cap 1MB + deterministic compaction
사용자 결정 2026-05-19: 100KB cap 이 운영 7d 데이터 1.36MB 대비 부족 →
cap 상향만으로 raw 비대화 위험. cap 1MB + payload compaction 병행.

fetch_recap_context() 변경:
- memo payload item field 축소 = id/title/ai_tldr/ai_event_kind/created_at (5 필드)
  (ai_bullets/file_type/source_channel/category/extracted_text 등 제외)
- memo top-N = RECAP_MEMO_TOP_N env (default 200) — 초과분은 aggregate 로
- aggregate = memos_by_day + memos_by_kind + omitted_memos
- payload_compacted flag = aggregate fallback 발현 여부
- events 는 raw (운영 7d 데이터에서 통상 0~소량)

internal_worker.py:
- PAYLOAD_MAX_BYTES → _payload_max_bytes() env override
  (WORKER_RECAP_PAYLOAD_MAX_BYTES default 1_000_000)
- JobsRecapResponse 에 payload_compacted / omitted_memos 노출
- 413 detail 에 "after compaction" 명시 + RECAP_MEMO_TOP_N 조정 안내

테스트 3 항목 신규 + 기존 endpoint 413 test 업데이트:
- 700 memo → 200 kept + 500 omitted + compacted=true + < 1MB
- 10 memo → compacted=false + omitted=0
- 비정상 큰 title (compaction 후에도 cap 초과) → 413 유지

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-19 12:55:51 +09:00

203 lines
6.4 KiB
Python

"""PR-Worker-Pool-Registry-1C compaction — deterministic top-N + aggregate.
3 항목 (사용자 결정 2026-05-19 spec):
1. 700+ memo → /jobs/recap 200 + payload_size < 1MB + payload_compacted=true + omitted > 0
2. 작은 fixture (10 memos) → compacted=false + omitted=0
3. 거대 title (1MB 초과 시뮬레이션) → 413 유지
monkeypatch 로 fetch_recap_context 또는 _memo_top_n 조정.
"""
from __future__ import annotations
import json
import os
import sys
import pytest
import pytest_asyncio
sys.path.insert(0, os.path.join(os.path.dirname(__file__), "..", "app"))
from httpx import ASGITransport, AsyncClient
from _worker_pool_helpers import (
cleanup_worker_jobs,
ensure_user,
fetch_worker_job,
mint_access_token,
)
@pytest_asyncio.fixture
async def env_setup(monkeypatch):
monkeypatch.setenv("LAPTOP_WORKER_BOT_USERNAME", "laptop-worker-bot")
def _make_fake_context(total_memos: int, top_n: int = 200) -> dict:
"""compaction 모사 — _compact_memo 형식 (5 필드) 의 row."""
from datetime import datetime, timezone
from zoneinfo import ZoneInfo
kst = ZoneInfo("Asia/Seoul")
now = datetime.now(timezone.utc)
kept_count = min(total_memos, top_n)
omitted = max(0, total_memos - top_n)
memos = [
{
"id": i,
"title": f"memo {i}",
"ai_tldr": f"tldr {i}",
"ai_event_kind": "note",
"created_at": now.astimezone(kst).isoformat(),
}
for i in range(kept_count)
]
return {
"user_id": 1,
"days": 7,
"period_start": now.astimezone(kst).isoformat(),
"period_end": now.astimezone(kst).isoformat(),
"timezone": "Asia/Seoul",
"memos": memos,
"events": [],
"memo_count": total_memos,
"event_count": 0,
"summary_stats": {
"total_memos": total_memos,
"memos_kept": kept_count,
"omitted_memos": omitted,
"top_n": top_n,
"memos_by_day": {"2026-05-19": omitted} if omitted else {},
"memos_by_kind": {"note": omitted} if omitted else {},
},
"payload_compacted": omitted > 0,
}
@pytest.mark.asyncio
async def test_compaction_kicks_in_at_top_n(env_setup, monkeypatch):
"""700 memo → 200 kept + 500 omitted + payload_compacted=true + payload < 1MB."""
from api import internal_worker as iw_mod
from main import app
fake = _make_fake_context(total_memos=700, top_n=200)
async def fake_fetch(session, user_id, days=7):
# user_id 채워 줌
fake["user_id"] = user_id
return fake
monkeypatch.setattr(iw_mod, "fetch_recap_context", fake_fetch)
await ensure_user("test-recap-compact-1c")
token = mint_access_token("test-recap-compact-1c")
try:
async with AsyncClient(
transport=ASGITransport(app=app), base_url="http://test"
) as c:
r = await c.post(
"/internal/worker/jobs/recap",
json={"days": 7},
headers={"Authorization": f"Bearer {token}"},
)
assert r.status_code == 200, r.text
js = r.json()
assert js["memo_count"] == 700
assert js["payload_compacted"] is True
assert js["omitted_memos"] == 500
assert js["payload_bytes"] < 1_000_000
# DB verify — payload 안에 compacted memos + summary_stats
job = await fetch_worker_job(js["job_id"])
assert job is not None
finally:
await cleanup_worker_jobs("recap")
@pytest.mark.asyncio
async def test_no_compaction_when_under_top_n(env_setup, monkeypatch):
"""10 memo → kept 10 + omitted 0 + payload_compacted=false."""
from api import internal_worker as iw_mod
from main import app
fake = _make_fake_context(total_memos=10, top_n=200)
async def fake_fetch(session, user_id, days=7):
fake["user_id"] = user_id
return fake
monkeypatch.setattr(iw_mod, "fetch_recap_context", fake_fetch)
await ensure_user("test-recap-uncompact-1c")
token = mint_access_token("test-recap-uncompact-1c")
try:
async with AsyncClient(
transport=ASGITransport(app=app), base_url="http://test"
) as c:
r = await c.post(
"/internal/worker/jobs/recap",
json={"days": 7},
headers={"Authorization": f"Bearer {token}"},
)
assert r.status_code == 200, r.text
js = r.json()
assert js["memo_count"] == 10
assert js["payload_compacted"] is False
assert js["omitted_memos"] == 0
finally:
await cleanup_worker_jobs("recap")
@pytest.mark.asyncio
async def test_413_when_compacted_still_over_cap(env_setup, monkeypatch):
"""비정상 큰 title (compaction 후에도 1MB 초과) → 413 유지."""
from api import internal_worker as iw_mod
from main import app
async def fake_fetch(session, user_id, days=7):
huge_title = "x" * 6000 # per memo
return {
"user_id": user_id,
"days": days,
"period_start": "2026-05-12T00:00:00+09:00",
"period_end": "2026-05-19T00:00:00+09:00",
"timezone": "Asia/Seoul",
"memos": [
{
"id": i,
"title": huge_title,
"ai_tldr": None,
"ai_event_kind": None,
"created_at": "2026-05-19T00:00:00+09:00",
}
for i in range(200)
], # ~1.2MB
"events": [],
"memo_count": 200,
"event_count": 0,
"summary_stats": {
"total_memos": 200,
"memos_kept": 200,
"omitted_memos": 0,
"top_n": 200,
"memos_by_day": {},
"memos_by_kind": {},
},
"payload_compacted": False,
}
monkeypatch.setattr(iw_mod, "fetch_recap_context", fake_fetch)
await ensure_user("test-recap-413-large-1c")
token = mint_access_token("test-recap-413-large-1c")
async with AsyncClient(
transport=ASGITransport(app=app), base_url="http://test"
) as c:
r = await c.post(
"/internal/worker/jobs/recap",
json={"days": 7},
headers={"Authorization": f"Bearer {token}"},
)
assert r.status_code == 413, r.text
assert "after compaction" in r.json()["detail"]