mirror of
https://github.com/nprasad2077/MNIST.git
synced 2026-11-04 09:15:29 +00:00
nprasad2077
2d2f7dd007
Lint notebooks (ruff), secret scan (gitleaks), dependency audit (pip-audit), workflow lint (actionlint) and notebook hygiene check, with a results table in the job summary. Fork PRs are skipped since the repo is public and the runner is self-hosted. Also removes an unused TensorDataset import from notebook 04.
49 lines
1.4 KiB
Python
49 lines
1.4 KiB
Python
"""Check that notebooks are valid JSON and report their size and embedded outputs.
|
|
|
|
Exits 1 only for unreadable notebooks; oversized files are warnings.
|
|
"""
|
|
|
|
import json
|
|
import sys
|
|
from pathlib import Path
|
|
|
|
SIZE_WARN_BYTES = 500 * 1024
|
|
|
|
|
|
def count_png_outputs(nb: dict) -> int:
|
|
return sum(
|
|
1
|
|
for cell in nb.get("cells", [])
|
|
for out in cell.get("outputs", [])
|
|
if "image/png" in out.get("data", {})
|
|
)
|
|
|
|
|
|
def main(root: str = "notebooks") -> int:
|
|
paths = sorted(Path(root).glob("*.ipynb"))
|
|
if not paths:
|
|
print(f"::error::no notebooks found under {root}/")
|
|
return 1
|
|
|
|
failed = False
|
|
print(f"{'notebook':<28}{'KB':>8}{'cells':>7}{'png':>5}")
|
|
for path in paths:
|
|
size = path.stat().st_size
|
|
try:
|
|
nb = json.loads(path.read_text(encoding="utf-8"))
|
|
except (OSError, json.JSONDecodeError) as exc:
|
|
print(f"::error file={path}::invalid notebook JSON: {exc}")
|
|
failed = True
|
|
continue
|
|
print(
|
|
f"{path.name:<28}{size / 1024:>8.0f}"
|
|
f"{len(nb.get('cells', [])):>7}{count_png_outputs(nb):>5}"
|
|
)
|
|
if size > SIZE_WARN_BYTES:
|
|
print(f"::warning file={path}::{size / 1024:.0f} KB exceeds {SIZE_WARN_BYTES // 1024} KB")
|
|
return 1 if failed else 0
|
|
|
|
|
|
if __name__ == "__main__":
|
|
sys.exit(main(*sys.argv[1:2]))
|