from __future__ import annotations

import json
import re
from pathlib import Path
from typing import Any


ROOT = Path(__file__).resolve().parent
SAMPLES = ROOT / "samples"
INDEX = ROOT / "index" / "catalog.json"


def flatten_keys(value: Any) -> set[str]:
    if isinstance(value, dict):
        return set(value) | set().union(*(flatten_keys(item) for item in value.values()))
    if isinstance(value, list):
        return set().union(*(flatten_keys(item) for item in value)) if value else set()
    return set()


def build_index() -> dict[str, Any]:
    records: list[dict[str, Any]] = []
    for source in sorted(SAMPLES.glob("*/model.cdsl.json")):
        cdsl = json.loads(source.read_text(encoding="utf-8"))
        profiles = [
            str(sketch.get("profile", {}).get("type"))
            for sketch in cdsl.get("geometry", {}).get("sketches", [])
            if sketch.get("profile", {}).get("type")
        ]
        features = [str(feature.get("atomic_id")) for feature in cdsl.get("features", [])]
        parameter_names = sorted(flatten_keys(cdsl.get("geometry", {})) | flatten_keys(cdsl.get("features", [])))
        part_id = str(cdsl.get("part_id") or source.parent.name)
        source_name = str(cdsl.get("meta", {}).get("source") or part_id)
        records.append({
            "part_id": part_id,
            "source_name": source_name,
            "profiles": profiles,
            "features": features,
            "parameters": parameter_names,
            "summary": f"{part_id}: {', '.join(profiles)}; {', '.join(features)}",
        })
    payload = {"schema_version": "1.0", "samples": records}
    INDEX.parent.mkdir(parents=True, exist_ok=True)
    INDEX.write_text(json.dumps(payload, ensure_ascii=False, indent=2) + "\n", encoding="utf-8")
    return payload


if __name__ == "__main__":
    result = build_index()
    print(json.dumps({"samples": len(result["samples"])}, ensure_ascii=False))