"""Make referenced dbt source datasets requestable in the DAX inventory.

The command is dry-run by default. Added rows are active requestable ownership
metadata; existing active referenced source rows are promoted to requestable
without changing their owner.

Example from ``server/``::

    .venv/bin/python scripts/sync_data_access_lineage_roots.py \
      app/domains/data_access/lineage_catalogs/gcp-digital-bi-dbt.json \
      ../deploy/data-access/datasets/gcp-digital-bi-dbt.json --commit
"""
from __future__ import annotations

import argparse
import json
from pathlib import Path

from app.domains.data_access.lineage_inventory import (
    DEFAULT_LINEAGE_ROOT_OWNER_EMAIL,
    sync_lineage_root_datasets,
)


def main() -> None:
    parser = argparse.ArgumentParser(description=__doc__)
    parser.add_argument("catalog", type=Path)
    parser.add_argument("inventory", type=Path)
    parser.add_argument(
        "--default-owner-email",
        default=DEFAULT_LINEAGE_ROOT_OWNER_EMAIL,
    )
    parser.add_argument("--commit", action="store_true")
    args = parser.parse_args()

    catalog = _load_object(args.catalog, label="lineage catalog")
    inventory = _load_object(args.inventory, label="dataset inventory")
    requestable_before = {
        (str(row["project_id"]), str(row["dataset_id"])): (
            row.get("requestable") is True
        )
        for row in inventory.get("datasets", [])
        if isinstance(row, dict)
    }
    projected, additions = sync_lineage_root_datasets(
        inventory,
        catalog,
        default_owner_email=args.default_owner_email,
    )
    for addition in additions:
        print(
            "ADD "
            f"{addition.project_id}.{addition.dataset_id} "
            f"owner={addition.owner_email} requestable=true",
        )
    for row in projected["datasets"]:
        locator = (str(row["project_id"]), str(row["dataset_id"]))
        if (
            requestable_before.get(locator) is False
            and row.get("requestable") is True
        ):
            print(f"PROMOTE {locator[0]}.{locator[1]} requestable=true")
    prefix = "COMMIT" if args.commit else "DRY RUN"
    print(f"{prefix}: additions={len(additions)} total={len(projected['datasets'])}")
    if args.commit:
        args.inventory.write_text(
            json.dumps(projected, ensure_ascii=False, indent=2) + "\n",
            encoding="utf-8",
        )


def _load_object(path: Path, *, label: str) -> dict:
    value = json.loads(path.read_text(encoding="utf-8"))
    if not isinstance(value, dict):
        raise ValueError(f"{label} must be a JSON object")
    return value


if __name__ == "__main__":
    main()