| """Shared data loading, file integrity and subprocess helpers.""" |
| import hashlib |
| import csv |
| import json |
| from pathlib import Path |
| import subprocess |
|
|
| ROOT = Path(__file__).resolve().parents[1] |
|
|
|
|
| def sha256(path): |
| digest = hashlib.sha256() |
| with Path(path).open("rb") as stream: |
| for chunk in iter(lambda: stream.read(1024 * 1024), b""): |
| digest.update(chunk) |
| return digest.hexdigest() |
|
|
|
|
| def contained(root, relative): |
| path = Path(relative) |
| if path.is_absolute() or ".." in path.parts or not path.parts: |
| raise ValueError(f"Unsafe relative path: {relative}") |
| resolved = (Path(root) / path).resolve() |
| if not resolved.is_relative_to(Path(root).resolve()): |
| raise ValueError(f"Path escapes root: {relative}") |
| return resolved |
|
|
|
|
| def resolve_instance_id(instance_id): |
| |
| contained(ROOT / "instances", instance_id) |
| mapping_path = ROOT / "data/instance-id-map.tsv" |
| if mapping_path.is_file(): |
| with mapping_path.open() as stream: |
| rows = list(csv.DictReader(stream, delimiter="\t")) |
| aliases = {} |
| for row in rows: |
| canonical = row["instance_id"] |
| for alias in (row["original_instance_id"], canonical): |
| if alias in aliases and aliases[alias] != canonical: |
| raise ValueError("Ambiguous instance ID mapping") |
| aliases[alias] = canonical |
| if instance_id not in aliases: |
| raise ValueError(f"Unknown instance ID: {instance_id}") |
| return aliases[instance_id] |
| return instance_id |
|
|
|
|
| def load_task(instance_id): |
| canonical = resolve_instance_id(instance_id) |
| path = contained(ROOT / "instances", canonical) / "task.json" |
| task = json.loads(path.read_text()) |
| if task["instance_id"] != canonical: |
| raise ValueError("Task ID does not match directory") |
| return task |
|
|
|
|
| def run(*args, cwd=None, capture=False): |
| return subprocess.run(args, cwd=cwd, check=True, text=True, |
| stdout=subprocess.PIPE if capture else None).stdout |
|
|
