app/build_sqlite.py

"""Build or refresh the SQLite FTS fee index (no OpenSearch / no torch)."""
from __future__ import annotations

import argparse
import logging
import sys

from .config import settings
from .schedule_docs import build_documents
from . import sqlite_store

logger = logging.getLogger(__name__)


def build_sqlite_index(*, force: bool = False) -> dict:
    path = sqlite_store.db_path()
    if path.exists() and sqlite_store.count_codes() > 0 and not force:
        n = sqlite_store.count_codes()
        logger.info("SQLite index already present (%d codes) at %s", n, path)
        return {"status": "exists", "codes": n, "path": str(path)}

    docs = build_documents()
    n = sqlite_store.replace_all(docs)
    return {"status": "built", "codes": n, "path": str(path)}


def main(argv: list[str] | None = None) -> int:
    parser = argparse.ArgumentParser(description=__doc__)
    parser.add_argument(
        "--force",
        action="store_true",
        help="Rebuild even if an index already exists",
    )
    args = parser.parse_args(argv)
    logging.basicConfig(level=logging.INFO)
    # Pilot / Spaces often use /tmp for writable caches.
    settings.ohip_data_dir = settings.ohip_data_dir or "/tmp/ohip"
    summary = build_sqlite_index(force=args.force)
    print(summary)
    return 0 if summary.get("codes", 0) > 0 else 1


if __name__ == "__main__":
    sys.exit(main())