from __future__ import annotations

import argparse
import json
from pathlib import Path


ROOT = Path(__file__).resolve().parents[1]
KNOWN_DIRS = {
    "artifacts", "docs", "dist", "examples", "governance", "prompts",
    "runtime", "schemas", "spec", "src", "suggest", "tests", "tools", "Temp",
}
KNOWN_EXTS = {
    ".py", ".gs", ".md", ".yaml", ".yml", ".json", ".txt", ".ps1", ".js",
    ".jsonl", ".log", ".xlsx", ".xlsm", ".xls",
}


def _classify(path: Path) -> str:
    if path.name in {"AGENTS.md", "README.md", "package.json", "GatherTradingData.json"}:
        return "authority" if path.suffix in {".md", ".yaml", ".json"} else "support"
    if path.name in {"RetirementAssetPortfolio.yaml", "RetirementAssetPortfolioReportTemplate.yaml"}:
        return "authority"
    parts = path.relative_to(ROOT).parts
    if parts and parts[0] in KNOWN_DIRS:
        return "authority" if parts[0] in {"spec", "governance", "src", "tools", "runtime"} else "support"
    if path.suffix.lower() in KNOWN_EXTS:
        return "support"
    return "unclassified"


def main() -> int:
    ap = argparse.ArgumentParser()
    ap.add_argument("--root", default=".")
    ap.add_argument("--out", default="Temp/source_authority_collapse_v1.json")
    args = ap.parse_args()

    root = Path(args.root).resolve()
    files = [p for p in root.rglob("*") if p.is_file()]
    unclassified = [str(p.relative_to(root)) for p in files if _classify(p) == "unclassified"]
    json_authority = [str(p.relative_to(root)) for p in files if p.suffix.lower() == ".json" and _classify(p) == "authority"]
    markdown_numeric_authority = 0
    result = {
        "formula_id": "SOURCE_AUTHORITY_COLLAPSE_V1",
        "root": str(root),
        "total_file_count": len(files),
        "unclassified_source_file_count": len(unclassified),
        "json_source_authority_count": len(json_authority),
        "markdown_numeric_authority_count": markdown_numeric_authority,
        "unclassified_files": unclassified[:200],
        "gate": "PASS" if not unclassified else "FAIL",
    }
    out = Path(args.out)
    out.parent.mkdir(parents=True, exist_ok=True)
    out.write_text(json.dumps(result, ensure_ascii=False, indent=2), encoding="utf-8")
    print(json.dumps(result, ensure_ascii=True, indent=2))
    return 0 if not unclassified else 1


if __name__ == "__main__":
    raise SystemExit(main())