"""CLI: ingest course material.

  python -m scripts.ingest --course imm1033 --site ../content/imm1033/dist   # generated website
  python -m scripts.ingest --course imm1033 --path ../content/imm1033/        # pdf/docx/pptx/md
"""

from __future__ import annotations

import argparse
import asyncio
from pathlib import Path

from app.db import init_db
from app.rag import retriever
from app.rag.ingest import ingest_file, ingest_site


async def main() -> None:
    ap = argparse.ArgumentParser()
    ap.add_argument("--course", required=True)
    ap.add_argument("--site", help="dist/<course> directory of the generated course website")
    ap.add_argument("--path", help="directory of pdf/docx/pptx/md files")
    ap.add_argument("--visibility", default="students", choices=["students", "professor_only"])
    args = ap.parse_args()
    await init_db()
    total = 0
    if args.site:
        total += await ingest_site(args.course, Path(args.site), args.visibility)
    if args.path:
        for p in sorted(Path(args.path).rglob("*")):
            if p.suffix.lower() in {".pdf", ".docx", ".pptx", ".md", ".txt"} and p.is_file():
                n = await ingest_file(args.course, p, args.visibility)
                print(f"  {p.name}: {n} chunks")
                total += n
    n = await retriever.rebuild_index()
    print(f"Ingested {total} chunks; index now holds {n} chunks.")


if __name__ == "__main__":
    asyncio.run(main())