webmail-client/backend/scripts/backfill_delivery_id.py

"""One-off: recover delivery_id for messages whose stored Message-ID was
synthesized (§7.1 cross-account duplicate). The real Message-ID is still in the
raw blob, so re-parse it. Non-synthetic rows were already backfilled by migration
0011; here we only touch rows left with delivery_id IS NULL that have a real
Message-ID in their raw bytes."""

import asyncio

from sqlalchemy import select

from app.db import session_scope
from app.indexer import blobstore
from app.indexer.mime import parse_message
from app.models import Message


async def main() -> None:
    async with session_scope() as session:
        rows = list(await session.scalars(select(Message).where(Message.delivery_id.is_(None))))
        fixed = 0
        for m in rows:
            try:
                raw = blobstore.load(m.user_id, m.blob_sha256)
            except FileNotFoundError:
                continue
            parsed = parse_message(raw)
            if parsed.message_id:
                m.delivery_id = parsed.message_id
                fixed += 1
        print(f"scanned {len(rows)} null rows, set delivery_id on {fixed}")


asyncio.run(main())