#!/usr/bin/env python3
from __future__ import annotations

import re
from pathlib import Path


ROOT = Path(__file__).resolve().parent
ALLOWED_ROOT = ROOT / "source/lerobot/tests/artifacts"
POINTER_PATTERN = re.compile(
    rb"^version https://git-lfs\.github\.com/spec/v1\n"
    rb"oid sha256:([0-9a-f]{64})\n"
    rb"size ([0-9]+)\n?$"
)


def main() -> None:
    pointers = []
    for path in sorted(ROOT.rglob("*")):
        if not path.is_file() or path.is_symlink() or path.stat().st_size > 1024:
            continue
        data = path.read_bytes()
        match = POINTER_PATTERN.fullmatch(data)
        if match:
            pointers.append((path, match.group(1).decode(), int(match.group(2))))
    if not pointers:
        print("No unmaterialized Git LFS pointers found.")
        return
    for path, _, _ in pointers:
        if ALLOWED_ROOT not in path.parents:
            raise RuntimeError(f"refusing to remove LFS pointer outside test artifacts: {path}")
    report = ROOT / "source/LFS_POINTERS_OMITTED.txt"
    lines = [
        "Unmaterialized Git LFS test fixtures omitted for Hugging Face upload.",
        "They are unrelated to the SO101 training outputs and their pointers remain in the Git bundle history.",
        "",
        "path\tsha256_oid\tdeclared_size_bytes",
    ]
    for path, oid, size in pointers:
        lines.append(f"{path.relative_to(ROOT)}\t{oid}\t{size}")
    report.write_text("\n".join(lines) + "\n")
    for path, _, _ in pointers:
        path.unlink()
    print(f"Removed {len(pointers)} unmaterialized Git LFS test-fixture pointers.")
    print(f"Recorded omissions in {report}")


if __name__ == "__main__":
    main()