debug_live.py

# -*- encoding: utf-8 -*-
import sys, io
sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding='utf-8', errors='replace')

import asyncio
from playwright.async_api import async_playwright
import time
import json

async def main():
    danmaku_ws_url = None
    captured_messages = []

    async with async_playwright() as p:
        browser = await p.chromium.launch(headless=True)
        context = await browser.new_context(
            storage_state=None  # 不使用存储状态，使用当前浏览器cookie
        )
        page = await context.new_page()

        # 拦截WebSocket
        async def on_websocket(ws):
            print(f"[WebSocket连接] {ws.url[:200]}")
            nonlocal danmaku_ws_url
            if 'webcast' in ws.url or 'chat' in ws.url or 'im' in ws.url:
                danmaku_ws_url = ws.url
            try:
                async for msg in ws.incoming_messages():
                    msg_text = str(msg)[:200]
                    print(f"[WS消息] {msg_text}")
                    captured_messages.append(msg_text)
            except Exception as e:
                print(f"[WS错误] {e}")

        page.on('websocket', on_websocket)

        print("访问直播间...")
        await page.goto('https://live.douyin.com/YOUR_ROOM_ID', timeout=60000)  # replace with your live room ID

        print("\n等待页面加载(30秒)...")
        await page.wait_for_timeout(30000)

        print("\n截图...")
        await page.screenshot(path='./test_danmaku.png')

        print(f"\n弹幕WebSocket URL: {danmaku_ws_url}")
        print(f"捕获消息数: {len(captured_messages)}")
        for i, msg in enumerate(captured_messages[:10]):
            print(f"  [{i}] {msg}")

        # 尝试执行JS查看页面状态
        print("\n尝试获取Vue/React状态...")
        try:
            state = await page.evaluate("""
                JSON.stringify(window.__INITIAL_STATE__ || {})
            """)
            if state and state != '{}':
                print("INITIAL_STATE:", state[:500])
        except:
            pass

        # 查找弹幕相关元素
        print("\n搜索页面中的弹幕内容...")
        try:
            # 搜索页面中所有包含用户名的元素
            all_text = await page.evaluate("""
                Array.from(document.querySelectorAll('span, div, p'))
                    .map(el => el.innerText)
                    .filter(t => t.length > 1 && t.length < 100)
                    .filter(t => !t.includes('\\n'))
                    .slice(0, 50)
            """)
            print("找到的文本片段:")
            for t in all_text[:20]:
                print(f"  {t[:80]}")
        except Exception as e:
            print(f"错误: {e}")

        await browser.close()
        print("\n完成")

asyncio.run(main())