content-find-agent-new/scripts/validate_config_excel_sync.py

"""Validate Excel↔JSON byte-equal sync (V2-M1D).

Thin wrapper over build_config_from_excel: regenerate JSON from Excel and assert
byte-equal with the committed runtime JSON. Equivalent to
`build_config_from_excel.py --check`, exposed as a uniform validator for the
config gate.
"""

from __future__ import annotations

import json
import sys
from pathlib import Path

ROOT = Path(__file__).resolve().parents[1]
if str(ROOT) not in sys.path:
    sys.path.insert(0, str(ROOT))

from scripts.build_config_from_excel import build, governance_report
from scripts.check_config_json_canonical import canonical_dumps


def main() -> int:
    findings = []
    for path_str, obj in build().items():
        path = Path(path_str)
        ok = canonical_dumps(obj) == path.read_text(encoding="utf-8")
        finding = {"level": "fail" if not ok else "info", "check_id": "excel_json_sync",
                   "config_path": str(path.relative_to(ROOT)), "byte_equal": ok}
        if not ok:
            finding["message"] = "Excel diverges from JSON; run build_config_from_excel.py --sync-excel (JSON wins) or --write (Excel wins)"
        findings.append(finding)
    governance = governance_report()
    for row in governance:
        findings.append(
            {
                "level": "info",
                "check_id": "excel_governance_sheet",
                "config_path": row["workbook"],
                "sheet": row["sheet"],
                "mode": row["mode"],
                "row_count": row["row_count"],
                "runtime_row_count": row["runtime_row_count"],
            }
        )
    status = "fail" if any(f["level"] == "fail" for f in findings) else "pass"
    print(json.dumps({"status": status, "findings": findings}, ensure_ascii=False, indent=2))
    return 1 if status == "fail" else 0


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