Clean a list

import { readFile, writeFile } from "node:fs/promises";

const MAGPIE = "https://api.magpie-cloud.com/v1";
const sleep = (ms) => new Promise((resolve) => setTimeout(resolve, ms));

async function verify(email) {
  for (let attempt = 1; ; attempt++) {
    const res = await fetch(`${MAGPIE}/verify`, {
      method: "POST",
      headers: {
        Authorization: `Bearer ${process.env.MAGPIE_API_KEY}`,
        "Content-Type": "application/json",
      },
      body: JSON.stringify({ email }),
      signal: AbortSignal.timeout(90_000), // a slow mail server can take a minute
    });
    if (res.ok) return res.json();
    const retry = res.status === 429 || res.status >= 500;
    if (!retry || attempt === 5) {
      // 400, 401 or 402 (out of credits): stop, rather than fail every address.
      throw new Error(`Stopped at ${email}: HTTP ${res.status}`);
    }
    const seconds = Number(res.headers.get("Retry-After")) || 2 ** attempt;
    await sleep(seconds * 1000);
  }
}

// One address a line. Drop blanks and repeats first: each check is charged.
const text = await readFile("list.txt", "utf8");
const emails = [...new Set(text.split("\n").map((l) => l.trim().toLowerCase()).filter(Boolean))];

const where = { valid: "keep", invalid: "remove", catch_all: "careful", risky: "careful", unknown: "later" };
const files = { keep: [], remove: [], careful: [], later: [] };

for (const email of emails) {
  const { verdict } = await verify(email);
  files[where[verdict]].push(email);
}
for (const [name, list] of Object.entries(files)) {
  await writeFile(`${name}.txt`, list.join("\n"));
}

import os
import time
import requests

MAGPIE = "https://api.magpie-cloud.com/v1"

def verify(email):
    for attempt in range(1, 6):
        res = requests.post(
            f"{MAGPIE}/verify",
            json={"email": email},
            headers={"Authorization": f"Bearer {os.environ['MAGPIE_API_KEY']}"},
            timeout=90,  # a slow mail server can take a minute
        )
        if res.ok:
            return res.json()
        retry = res.status_code == 429 or res.status_code >= 500
        if not retry or attempt == 5:
            # 400, 401 or 402 (out of credits): stop, rather than fail every address.
            raise RuntimeError(f"Stopped at {email}: HTTP {res.status_code}")
        time.sleep(int(res.headers.get("Retry-After", 2 ** attempt)))

# One address a line. Drop blanks and repeats first: each check is charged.
with open("list.txt") as f:
    emails = list(dict.fromkeys(line.strip().lower() for line in f if line.strip()))

where = {"valid": "keep", "invalid": "remove", "catch_all": "careful", "risky": "careful", "unknown": "later"}
files = {"keep": [], "remove": [], "careful": [], "later": []}

for email in emails:
    files[where[verify(email)["verdict"]]].append(email)

for name, found in files.items():
    with open(f"{name}.txt", "w") as f:
        f.write("\n".join(found))