פרוקסי למעקב דירוגים במנועי חיפוש, עם המדינה בתוך הקישור

#!/usr/bin/env python3
"""Search-position checker for Advanced.name proxies.

Reads the CSV proxy export, keeps one country, then asks a search engine for each
keyword through a proxy from that country and reports where your domain sits in
the results. Standard library only. Edit CONFIG and run: python3 rank.py
"""

# ================================ CONFIG ====================================
CSV_URL = "https://get.advanced.name/YOUR_LIST_ID.csv"  # CSV export link from your dashboard
COUNTRY = "DE"          # ISO 3166-1 alpha-2: positions are read from this country
MIN_SPEED = 3           # keep proxies with speed >= this value
DOMAIN = "example.com"  # the domain whose position you track
KEYWORDS_FILE = "keywords.txt"           # one keyword per line
SEARCH_URL = "https://www.bing.com/search?q={q}&count=50&setlang={cc}"  # {q}=keyword, {cc}=country
OUT_FILE = "positions.csv"               # keyword;position;proxy is written here
TIMEOUT = 20
# ============================================================================

import csv
import io
import re
import sys
import urllib.parse
import urllib.request

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"
HREF = re.compile(r'href="https?://([^/"]+)', re.I)


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 position(html, domain):
    """First place the domain shows up among the result links, or 0 if absent."""
    rank, seen = 0, set()
    for host in HREF.findall(html):
        host = host.lower()
        if "bing.com" in host or "microsoft.com" in host or host in seen:
            continue
        seen.add(host)
        rank += 1
        if host == domain or host.endswith("." + domain):
            return rank
    return 0


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

    with open(OUT_FILE, "w", encoding="utf-8", newline="") as out:
        writer = csv.writer(out, delimiter=";")
        writer.writerow(["keyword", "position", "proxy"])
        for i, keyword in enumerate(keywords):
            proxy = proxies[i % len(proxies)]
            url = SEARCH_URL.format(q=urllib.parse.quote(keyword), cc=COUNTRY.lower())
            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")
                pos = position(html, DOMAIN.lower())
            except Exception as exc:
                pos = "err: %s" % exc.__class__.__name__
            writer.writerow([keyword, pos, proxy])
            print("%-40s %s" % (keyword[:40], pos))
    print("written %s" % OUT_FILE)


if __name__ == "__main__":
    main()

#!/usr/bin/env python3
"""AI Overview tracker for Advanced.name residential proxies.

Opens Google results for every keyword through a Resi Core or Resi Prime line
in a real Chromium, because the AI block is drawn by JavaScript. Records
whether the page shows the AI block and which domains it cites.
Setup: pip install playwright && python -m playwright install chromium
"""

# ================================ CONFIG ====================================
PROXY = "res.advanced.name:9000:USER-type-residential-country-us:PASS"  # host:port:login:password line from your dashboard
DOMAIN = "example.com"  # the domain you want to see among the cited sources
KEYWORDS_FILE = "keywords.txt"  # one keyword per line
SEARCH_URL = "https://www.google.com/search?q={q}&hl={hl}&gl={gl}"  # {q}=keyword
HL = "en"  # interface language of the results page
GL = "us"  # country of the results, keep it equal to the country in the login
LABELS = "AI Overview|AI Mode"  # heading text of the AI block, separated by |
CONSENT = "Reject all|Accept all"  # cookie banner buttons in EU countries, in the HL language
OUT_FILE = "ai_overview.csv"  # keyword;ai_block;cited;your_domain is written here
WAIT_MS = 4000  # time for the AI block to finish drawing
HEADLESS = True  # set False to watch the browser
# ============================================================================

import csv
import sys
import urllib.parse

from playwright.sync_api import sync_playwright

FIND_BLOCK = """(labels) => {
  const heads = [...document.querySelectorAll('h1,h2,h3,div,span')]
    .filter(e => e.children.length === 0 && labels.includes(e.textContent.trim()));
  if (!heads.length) return null;
  let box = heads[0];
  for (let i = 0; i < 8 && box.parentElement; i++) {
    box = box.parentElement;
    const links = [...box.querySelectorAll('a[href^="http"]')];
    if (links.length >= 2 && box.innerText.length > 200) return links.map(a => new URL(a.href).hostname);
  }
  return [];
}"""


def proxy_of(line):
    host, port, user, password = line.strip().split(":", 3)
    return {"server": "http://%s:%s" % (host, port), "username": user, "password": password}


def bare(host):
    host = host.lower()
    return host[4:] if host.startswith("www.") else host


def main():
    keywords = [k.strip() for k in open(KEYWORDS_FILE, encoding="utf-8") if k.strip()]
    labels = [x.strip() for x in LABELS.split("|") if x.strip()]
    consent = [x.strip() for x in CONSENT.split("|") if x.strip()]
    rows = []
    with sync_playwright() as pw:
        browser = pw.chromium.launch(headless=HEADLESS, proxy=proxy_of(PROXY))
        page = browser.new_page(locale=HL)
        for kw in keywords:
            url = SEARCH_URL.format(q=urllib.parse.quote_plus(kw), hl=HL, gl=GL)
            try:
                resp = page.goto(url, wait_until="domcontentloaded", timeout=60000)
                if resp is None or resp.status >= 400:
                    status = "http %s" % (resp.status if resp else "none")
                    print("%-40s %s" % (kw[:40], status), file=sys.stderr)
                    rows.append([kw, status, "", ""])
                    continue
                for name in consent:
                    button = page.get_by_role("button", name=name)
                    if button.count():
                        button.first.click()
                        page.wait_for_load_state("domcontentloaded")
                        break
                if "/sorry/" in page.url:
                    print("%-40s captcha: change the session id in the login" % kw[:40], file=sys.stderr)
                    rows.append([kw, "captcha", "", ""])
                    continue
                page.wait_for_timeout(WAIT_MS)
                hosts = page.evaluate(FIND_BLOCK, labels)
            except Exception as exc:
                print("%-40s error: %s" % (kw[:40], str(exc).splitlines()[0]), file=sys.stderr)
                rows.append([kw, "error", "", ""])
                continue
            if hosts is None:
                print("%-40s no AI block" % kw[:40])
                rows.append([kw, "no", "", ""])
                continue
            cited = sorted({bare(h) for h in hosts if "google." not in h})
            mine = any(h == DOMAIN or h.endswith("." + DOMAIN) for h in cited)
            print("%-40s AI block, %d sources, %s %s" % (kw[:40], len(cited), DOMAIN, "cited" if mine else "not cited"))
            rows.append([kw, "yes", " ".join(cited), "yes" if mine else "no"])
        browser.close()
    with open(OUT_FILE, "w", encoding="utf-8", newline="") as f:
        out = csv.writer(f, delimiter=";")
        out.writerow(["keyword", "ai_block", "cited", "your_domain"])
        out.writerows(rows)
    print("saved", OUT_FILE)


if __name__ == "__main__":
    main()