homelab-codex-ws/packages/kb-mail/src/kb_mail/archive.py
oskar 1666511475 feat(kb-mail): fundament — pgvector spine, koperta, archiwum, pakiet domeny
- services/kb-postgres: pgvector/pgvector:pg16 na SOLARIA (:5433), named
  volume, init/001_envelope.sql (CREATE EXTENSION vector + zamrożona tabela
  envelope: id/source/ts/geo/raw_ref/entities), service.yaml, healthcheck,
  README z poprawnym mechanizmem deploy (deploy-node.sh składa dwa -f)
- hosts/solaria/runtime/kb-postgres/docker-compose.override.yml: mem_limit 4g
- inventory/topology.yaml + hosts/solaria/services.yaml: kb-postgres wpisany
- packages/kb-mail: nowa konwencja shared lib (pip install /repo/packages/<lib>/)
  envelope.py — @dataclass Envelope, walidacja tz-aware ts
  db.py       — insert_envelope / get_envelope (asyncpg, ON CONFLICT DO NOTHING)
  archive.py  — save_eml append-only (asyncio.to_thread, FileExistsError na dup)
  tests: 15 unit pass + 5 integration (@pytest.mark.integration, wymaga KB_TEST_DSN)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-19 20:02:25 +02:00

40 lines
1.1 KiB
Python

from __future__ import annotations
import asyncio
from datetime import datetime
from pathlib import Path
import structlog
_log = structlog.get_logger(__name__)
_UNSAFE = str.maketrans({"/": "_", "\\": "_", ":": "_", "<": "", ">": ""})
async def save_eml(
archive_root: Path,
envelope_id: str,
source: str,
ts: datetime,
raw: bytes,
) -> str:
"""Write raw .eml bytes to the append-only archive.
Layout: {archive_root}/{source}/{YYYY}/{MM}/{sanitized_id}.eml
Returns the relative raw_ref path (store this in Envelope.raw_ref).
Raises FileExistsError if envelope_id is already archived (append-only guarantee).
"""
safe_id = envelope_id.translate(_UNSAFE)
rel_path = f"{source}/{ts.year:04d}/{ts.month:02d}/{safe_id}.eml"
dest = archive_root / rel_path
dest.parent.mkdir(parents=True, exist_ok=True)
if dest.exists():
raise FileExistsError(f"archive: {rel_path} already exists — archive is append-only")
await asyncio.to_thread(dest.write_bytes, raw)
_log.info("archive.saved", raw_ref=rel_path, bytes=len(raw))
return rel_path