"""The dependency policy, enforced. Every module under align/ and scripts/ is AST-scanned. Any import whose top-level package is not (a) stdlib, (b) the four allowed runtime deps, or (c) the package itself, fails this test. Lazy imports inside functions count too -- a lazy `import transformers` is still a dependency, just a hidden one. Whitelist is per-FILE and per-PACKAGE, so an optional extra can exist in exactly one place and nowhere else. """ from __future__ import annotations import ast import sys from pathlib import Path ROOT = Path(__file__).resolve().parents[1] ALLOWED = {"torch", "triton", "numpy", "safetensors", "align"} # file (relative to ROOT) -> extra packages it alone may import lazily FILE_WHITELIST: dict[str, set[str]] = { "align/data/ingest_parquet.py": {"pyarrow"}, "scripts/teachers/download_teacher.py": {"huggingface_hub"}, # offline weight download, never imported at runtime } STDLIB = set(sys.stdlib_module_names) def _imports(path: Path): tree = ast.parse(path.read_text(), filename=str(path)) for node in ast.walk(tree): if isinstance(node, ast.Import): for a in node.names: yield node.lineno, a.name.split(".")[0] elif isinstance(node, ast.ImportFrom): if node.level: # relative import -> the package itself continue if node.module: yield node.lineno, node.module.split(".")[0] def test_no_forbidden_imports(): bad = [] for sub in ("align", "scripts"): for py in sorted((ROOT / sub).rglob("*.py")): rel = str(py.relative_to(ROOT)) extra = FILE_WHITELIST.get(rel, set()) for lineno, top in _imports(py): if top in STDLIB or top in ALLOWED or top in extra: continue bad.append(f"{rel}:{lineno}: {top}") assert not bad, "forbidden third-party imports:\n" + "\n".join(bad) def test_extras_are_not_reachable_from_core(): """extras/ may import anything, but nothing under align/ may import extras/.""" for py in (ROOT / "align").rglob("*.py"): for _, top in _imports(py): assert top != "extras", f"{py} imports extras/"