Clean Code and Best Practices Questions

import json, subprocess, hashlib, os
from datetime import datetime

def git_sha():
    return subprocess.check_output(["git","rev-parse","HEAD"]).decode().strip()

def file_checksum(path):
    h = hashlib.sha256()
    with open(path,"rb") as f:
        for chunk in iter(lambda: f.read(8192), b""):
            h.update(chunk)
    return h.hexdigest()

def save_model_with_manifest(model_bytes, out_dir, hyperparams, metrics, dataset_path, extra=None):
    os.makedirs(out_dir, exist_ok=True)
    model_path = os.path.join(out_dir, "model.pt")
    with open(model_path,"wb") as f:
        f.write(model_bytes)  # e.g., torch.save bytes or file write

    manifest = {
        "git_sha": git_sha(),
        "timestamp": datetime.utcnow().isoformat()+"Z",
        "hyperparams": hyperparams,
        "metrics": metrics,
        "data_version": file_checksum(dataset_path),
        "dataset_path": os.path.basename(dataset_path),
    }
    if extra: manifest.update(extra)
    with open(os.path.join(out_dir,"manifest.json"), "w") as f:
        json.dump(manifest, f, indent=2)
    print("Saved model and manifest:", model_path)

# validate_manifest.py
import json, sys
m = json.load(open(sys.argv[1]))
required = ["git_sha","data_version","hyperparams","metrics"]
missing = [k for k in required if k not in m]
if missing:
    print("Missing metadata fields:", missing); sys.exit(1)
# Optionally compare m["git_sha"] to env var CI_COMMIT

import re

EMAIL_RE = re.compile(r"[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[A-Za-z]{2,}")
SSN_RE = re.compile(r"\b\d{3}-\d{2}-\d{4}\b|\b\d{9}\b")

def mask_emails(text, mask="[REDACTED_EMAIL]"):
    return EMAIL_RE.sub(mask, text)

def mask_ssns(text, mask="[REDACTED_SSN]"):
    return SSN_RE.sub(mask, text)

import csv, re, pytest
from pathlib import Path

EMAIL_RE = re.compile(r"[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[A-Za-z]{2,}")
SSN_RE = re.compile(r"\b\d{3}-\d{2}-\d{4}\b|\b\d{9}\b")

def scan_file_for_pii(path):
    with open(path, newline='', encoding='utf-8') as f:
        for i, line in enumerate(f, 1):
            if EMAIL_RE.search(line) or SSN_RE.search(line):
                return i, line.strip()
    return None, None

def test_fixtures_contain_no_pii():
    fixtures = Path("tests/fixtures").glob("*.csv")
    found=[]
    for f in fixtures:
        lineno, snippet = scan_file_for_pii(f)
        if lineno:
            found.append(f"{f}:{lineno}: {snippet}")
    assert not found, "PII found in fixtures:\n" + "\n".join(found)

import numpy as np
import pandas as pd
import json, hashlib

def make_fixture(n=200, seed=42, class_ratio=0.1, feat_stats=None, out_csv='fixture_v1.csv'):
    np.random.seed(seed)
    # labels: exact counts
    n_pos = int(round(n * class_ratio))
    labels = np.array([1]*n_pos + [0]*(n-n_pos))
    np.random.shuffle(labels)
    X = {}
    for fname, (mu, sigma, frac_miss) in feat_stats.items():
        vals = np.random.randn(n) * sigma + mu
        mask = np.random.rand(n) < frac_miss
        vals[mask] = np.nan
        X[fname] = vals
    df = pd.DataFrame(X)
    df['label'] = labels
    df.to_csv(out_csv, index=False)
    # metadata + checksum
    with open(out_csv,'rb') as f:
        checksum = hashlib.sha256(f.read()).hexdigest()
    meta = {'seed':seed, 'n':n, 'class_ratio':class_ratio,
            'feat_stats':feat_stats, 'checksum':checksum}
    with open(out_csv.replace('.csv','.json'),'w') as f:
        json.dump(meta, f, indent=2)
    return df, meta