Proxies for SEO tools that keep running all night long

#!/usr/bin/env python3
"""Proxy list builder for Scrapebox, GSA and ZennoPoster.

Reads the CSV export from Advanced.name, keeps the country, speed and uptime you
want, removes duplicates and writes a plain ip:port file that any SEO tool loads.
Optionally verifies each address before writing. Standard library only.
Run: python3 build_list.py
"""

# ================================ CONFIG ====================================
CSV_URL = "https://get.advanced.name/YOUR_LIST_ID.csv"  # CSV export link from your dashboard
COUNTRY = ""            # "" for the whole pool, or an ISO code like DE, US
MIN_SPEED = 5           # keep proxies with speed >= this (higher is better)
MIN_UPTIME = 80         # keep proxies with uptime >= this percent
VERIFY = False          # True: open TEST_URL through each proxy and drop the dead ones
TEST_URL = "https://www.google.com/"
OUT_FILE = "proxies.txt"                 # ip:port per line, ready for the tool
THREADS = 200
TIMEOUT = 10
# ============================================================================

import csv
import io
import sys
import urllib.request
from concurrent.futures import ThreadPoolExecutor

COL_IP, COL_PORT, COL_COUNTRY, COL_SPEED, COL_UPTIME = 0, 1, 2, 3, 4


def load_proxies():
    raw = urllib.request.urlopen(CSV_URL, timeout=30).read().decode("utf-8", "replace")
    seen, pool = set(), []
    for row in csv.reader(io.StringIO(raw), delimiter=";"):
        if len(row) <= COL_UPTIME or not row[COL_PORT].strip().isdigit():
            continue
        addr = row[COL_IP].strip() + ":" + row[COL_PORT].strip()
        if addr in seen:
            continue
        if COUNTRY and row[COL_COUNTRY].strip().upper() != COUNTRY.upper():
            continue
        if int(row[COL_SPEED] or 0) < MIN_SPEED or int(row[COL_UPTIME] or 0) < MIN_UPTIME:
            continue
        seen.add(addr)
        pool.append(addr)
    return pool


def alive(proxy):
    handler = urllib.request.ProxyHandler({"http": "http://" + proxy, "https": "http://" + proxy})
    try:
        urllib.request.build_opener(handler).open(TEST_URL, timeout=TIMEOUT).read(1)
        return True
    except Exception:
        return False


def main():
    proxies = load_proxies()
    print("matched %d proxies (%s, speed>=%d, uptime>=%d%%)"
          % (len(proxies), COUNTRY or "any", MIN_SPEED, MIN_UPTIME))
    if VERIFY and proxies:
        with ThreadPoolExecutor(max_workers=THREADS) as pool:
            flags = list(pool.map(alive, proxies))
        proxies = [p for p, ok in zip(proxies, flags) if ok]
        print("verified alive: %d" % len(proxies))
    if not proxies:
        sys.exit("Nothing to write. Loosen MIN_SPEED / MIN_UPTIME or clear COUNTRY.")
    with open(OUT_FILE, "w", encoding="utf-8") as out:
        out.write("\n".join(proxies) + "\n")
    print("written %s (%d proxies)" % (OUT_FILE, len(proxies)))


if __name__ == "__main__":
    main()