Proxies para equipes de e-commerce que monitoram o mercado

#!/usr/bin/env python3
"""Marketplace stock and price watcher for Advanced.name proxies.

For each product URL it fetches the page through a proxy from the target country
and pulls both the price and an in-stock flag, writing a CSV you can chart over
time. Standard library only. Edit CONFIG and run: python3 stock.py
"""

# ================================ CONFIG ====================================
CSV_URL = "https://get.advanced.name/YOUR_LIST_ID.csv"  # CSV export link from your dashboard
COUNTRY = "IT"          # marketplace country
MIN_SPEED = 3
PRODUCTS_FILE = "products.txt"           # one product URL per line
PRICE_RE = r'"price"\s*:\s*"?([0-9][0-9.,]*)'          # group 1 = price
INSTOCK_RE = r'"availability"\s*:\s*"[^"]*InStock'     # matches when in stock
OUT_FILE = "stock.csv"                    # url;price;in_stock;proxy is written here
THREADS = 40
TIMEOUT = 20
RETRIES = 3
# ============================================================================

import csv
import io
import itertools
import re
import sys
import threading
import time
import urllib.request
from concurrent.futures import ThreadPoolExecutor, as_completed

COL_IP, COL_PORT, COL_COUNTRY, COL_SPEED = 0, 1, 2, 3
UA = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 Chrome/124.0 Safari/537.36"
PRICE = re.compile(PRICE_RE)
INSTOCK = re.compile(INSTOCK_RE)


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


def main():
    proxies = load_proxies()
    if not proxies:
        sys.exit("No proxies for COUNTRY=%r." % COUNTRY)
    with open(PRODUCTS_FILE, encoding="utf-8") as fh:
        products = [line.strip() for line in fh if line.strip() and not line.startswith("#")]
    print("proxies: %d (%s), products: %d" % (len(proxies), COUNTRY, len(products)))

    rotation = itertools.cycle(proxies)
    lock = threading.Lock()

    def watch(url):
        for _ in range(RETRIES):
            with lock:
                proxy = next(rotation)
            handler = urllib.request.ProxyHandler({"http": "http://" + proxy, "https": "http://" + proxy})
            req = urllib.request.Request(url, headers={"User-Agent": UA, "Accept-Language": COUNTRY.lower()})
            try:
                html = urllib.request.build_opener(handler).open(req, timeout=TIMEOUT).read().decode("utf-8", "replace")
            except Exception:
                continue
            price = PRICE.search(html)
            return (price.group(1) if price else ""), ("yes" if INSTOCK.search(html) else "no"), proxy
        return "", "unreachable", ""

    rows = []
    with ThreadPoolExecutor(max_workers=THREADS) as pool:
        futures = {pool.submit(watch, u): u for u in products}
        for fut in as_completed(futures):
            url = futures[fut]
            price, stock, proxy = fut.result()
            rows.append((time.strftime("%Y-%m-%d %H:%M"), url, price, stock, proxy))
            print("%-4s %-9s %s" % (stock, price, url[:56]))

    header = not _exists(OUT_FILE)
    with open(OUT_FILE, "a", encoding="utf-8", newline="") as out:
        writer = csv.writer(out, delimiter=";")
        if header:
            writer.writerow(["checked_at", "url", "price", "in_stock", "proxy"])
        writer.writerows(rows)
    print("appended %d rows to %s" % (len(rows), OUT_FILE))


def _exists(path):
    try:
        open(path).close()
        return True
    except OSError:
        return False


if __name__ == "__main__":
    main()