"""Record dependency, GPU and checkpoint identities used for experiments."""

from __future__ import annotations

import hashlib
import importlib.metadata
import json
from pathlib import Path
import subprocess
import sys

import torch

from pallet.devices import cuda_devices


PACKAGES = ("torch", "torchvision", "ultralytics", "rfdetr", "numpy", "opencv-python-headless",
            "pycocotools", "sam3", "scikit-learn", "onnx", "onnxruntime")
WEIGHTS = ("yolo26s.pt", "yolov8s.pt", "/root/.roboflow/models/rf-detr-small.pth",
           "/root/private/project/zero_shot_segmentation/models/sam3/sam3.pt")


def _hash_file(path):
    digest = hashlib.sha256()
    with Path(path).open("rb") as source:
        for chunk in iter(lambda: source.read(4 * 1024 * 1024), b""):
            digest.update(chunk)
    return digest.hexdigest()


def capture(output="outputs/environment.json"):
    versions = {}
    for package in PACKAGES:
        try:
            versions[package] = importlib.metadata.version(package)
        except importlib.metadata.PackageNotFoundError:
            versions[package] = None
    try:
        driver = subprocess.check_output(["nvidia-smi", "--query-gpu=index,uuid,name,driver_version,memory.total",
                                          "--format=csv,noheader"], text=True).splitlines()
    except (OSError, subprocess.CalledProcessError):
        driver = []
    result = {"python": sys.version, "versions": versions, "torch_cuda": torch.version.cuda,
              "gpus": driver, "torch_cuda_devices": cuda_devices(),
              "weights": {str(path): _hash_file(path) for path in WEIGHTS if Path(path).is_file()}}
    Path(output).parent.mkdir(parents=True, exist_ok=True)
    Path(output).write_text(json.dumps(result, indent=2), encoding="utf-8")
    return result


if __name__ == "__main__":
    print(json.dumps(capture(), indent=2))
