48 lines
1.9 KiB
Python
48 lines
1.9 KiB
Python
|
# -*- coding: utf-8 -*-
|
||
|
|
"""单页测试: Playwright + stealth + 系统Chrome 抓 TPU"""
|
||
|
|
import sys, time
|
||
|
|
from playwright.sync_api import sync_playwright
|
||
|
|
from playwright_stealth import Stealth
|
||
|
|
|
||
|
|
URL = sys.argv[1] if len(sys.argv) > 1 else "https://www.techpowerup.com/"
|
||
|
|
|
||
|
|
with sync_playwright() as p:
|
||
|
|
browser = p.chromium.launch(
|
||
|
|
headless=True,
|
||
|
|
executable_path="/usr/bin/google-chrome",
|
||
|
|
args=["--disable-blink-features=AutomationControlled", "--no-sandbox", "--disable-gpu"],
|
||
|
|
)
|
||
|
|
ctx = browser.new_context(
|
||
|
|
user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/125.0.0.0 Safari/537.36",
|
||
|
|
viewport={"width": 1920, "height": 1080},
|
||
|
|
locale="en-US",
|
||
|
|
)
|
||
|
|
Stealth().apply_stealth_sync(ctx) # 隐藏 webdriver/headless 特征
|
||
|
|
page = ctx.new_page()
|
||
|
|
try:
|
||
|
|
page.goto(URL, wait_until="domcontentloaded", timeout=60000)
|
||
|
|
# 等 JS 挑战自动通过: 最多等 40s, 轮询检查标题/内容特征
|
||
|
|
for i in range(40):
|
||
|
|
time.sleep(1)
|
||
|
|
try:
|
||
|
|
title = page.title()
|
||
|
|
html = page.content()
|
||
|
|
except Exception:
|
||
|
|
continue # 页面正在跳转, 稍后再试
|
||
|
|
if "Access Denied" in title or "403" in title:
|
||
|
|
print(f"[{i}s] 403 Access Denied"); break
|
||
|
|
if "bot check" in html.lower() and len(html) < 5000:
|
||
|
|
continue
|
||
|
|
if len(html) > 30000 or "news" in html.lower():
|
||
|
|
print(f"[{i}s] 通过! title={title!r} html_size={len(html)}")
|
||
|
|
break
|
||
|
|
else:
|
||
|
|
print("超时未通过, 当前 title:", page.title())
|
||
|
|
# 打印正文摘要作为证据
|
||
|
|
text = page.inner_text("body")[:800]
|
||
|
|
print("--- body 前800字符 ---")
|
||
|
|
print(text)
|
||
|
|
except Exception as e:
|
||
|
|
print("ERROR:", e)
|
||
|
|
finally:
|
||
|
|
browser.close()
|