#!/usr/bin/env python3
"""Find every JSON field across public/resources that holds a unit/building id reference."""
from __future__ import annotations
import json, glob, collections, sys
from pathlib import Path

REPO = Path(__file__).resolve().parents[1]


def load_known_ids() -> set[str]:
    ids: set[str] = set()
    for sub in ("units", "buildings", "improvements", "items", "techs"):
        for fp in (REPO / "public" / "resources" / sub).glob("*.json"):
            if fp.name.endswith(".schema.json"):
                continue
            try:
                d = json.loads(fp.read_text())
            except Exception:
                continue
            items = d if isinstance(d, list) else [d]
            for it in items:
                if isinstance(it, dict) and isinstance(it.get("id"), str):
                    ids.add(it["id"])
    return ids


def main() -> int:
    known_ids = load_known_ids()
    print(f"loaded {len(known_ids)} ids from units/buildings/improvements/items/techs")

    field_count: collections.Counter[str] = collections.Counter()
    by_field_examples: dict[str, set[str]] = collections.defaultdict(set)

    for fp in sorted(glob.glob(str(REPO / "public" / "**" / "*.json"), recursive=True)):
        if fp.endswith(".schema.json"):
            continue
        try:
            data = json.loads(Path(fp).read_text())
        except Exception:
            continue
        rel = Path(fp).relative_to(REPO).as_posix()
        # Skip the resource files themselves to focus on cross-refs
        if rel.startswith("public/resources/units/") or rel.startswith("public/resources/buildings/"):
            continue

        def walk(o):
            if isinstance(o, dict):
                for k, v in o.items():
                    if isinstance(v, str) and v in known_ids:
                        field_count[k] += 1
                        if len(by_field_examples[k]) < 3:
                            by_field_examples[k].add(v)
                    elif isinstance(v, list):
                        for x in v:
                            if isinstance(x, str) and x in known_ids:
                                field_count[f"{k}[]"] += 1
                                if len(by_field_examples[f"{k}[]"]) < 3:
                                    by_field_examples[f"{k}[]"].add(x)
                            else:
                                walk(x)
                    elif isinstance(v, dict):
                        walk(v)

        walk(data)

    print("\n=== id-bearing fields (top 30, across non-units/buildings JSON) ===")
    for field, n in field_count.most_common(30):
        examples = ", ".join(sorted(by_field_examples[field])[:3])
        print(f"  {n:5d}  {field:32s}  e.g. {examples}")

    return 0


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