import os
import re
import json
import argparse
from pathlib import Path

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

def main():
    parser = argparse.ArgumentParser()
    parser.add_argument("--root", default=".")
    parser.add_argument("--out", default="Temp/version_sprawl_audit_v1.json")
    args = parser.parse_args()

    root_path = ROOT / args.root
    version_pattern = re.compile(r"(.+)_v(\d+)\.(.+)")
    
    groups = {}
    for dirpath, _, filenames in os.walk(root_path):
        if ".git" in dirpath or "node_modules" in dirpath or "__pycache__" in dirpath:
            continue
        
        for f in filenames:
            match = version_pattern.match(f)
            if match:
                base, version, ext = match.groups()
                rel_dir = os.path.relpath(dirpath, root_path)
                group_key = (rel_dir, base, ext)
                if group_key not in groups:
                    groups[group_key] = []
                groups[group_key].append({
                    "version": int(version),
                    "filename": f,
                    "path": os.path.join(rel_dir, f)
                })

    report = {
        "formula_id": "VERSION_SPRAWL_AUDIT_V1",
        "version_groups": []
    }

    for (rel_dir, base, ext), versions in groups.items():
        if len(versions) >= 3:
            versions.sort(key=lambda x: x["version"], reverse=True)
            report["version_groups"].append({
                "directory": rel_dir,
                "base": base,
                "extension": ext,
                "count": len(versions),
                "versions": versions
            })

    out_file = ROOT / args.out
    out_file.parent.mkdir(parents=True, exist_ok=True)
    with open(out_file, "w", encoding="utf-8") as f:
        json.dump(report, f, indent=2)

    print(f"Version sprawl audit completed. Found {len(report['version_groups'])} groups with 3+ versions.")
    print(json.dumps(report, indent=2))

if __name__ == "__main__":
    main()