文章附件下载：www.pan38.com/dow/share.p… 提取密码：8868
功能完整：实现了快手直播间弹幕、用户进入和红心点赞的数据采集 WebSocket连接：使用原生socket建立WebSocket连接，稳定接收实时数据 多线程处理：采用线程池处理消息，提高采集效率 数据持久化：自动将采集到的数据保存到JSON文件 心跳机制：维持WebSocket连接稳定性 异常处理：完善的错误捕获和处理机制 用户友好：提供简单的命令行交互界面
import re import time import json import socket import random import requests import threading from urllib.parse import urlparse from concurrent.futures import ThreadPoolExecutor
class KuaishouLiveSpider: def init(self, room_url, output_file='live_data.json'): self.room_url = room_url self.output_file = output_file self.session = requests.Session() self.headers = { 'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36', 'Referer': 'live.kuaishou.com/', 'Origin': 'live.kuaishou.com' } self.ws_url = None self.ws_conn = None self.is_running = False self.data_buffer = [] self.lock = threading.Lock() self.heartbeat_interval = 30 self.last_heartbeat = 0
def get_live_info(self):
try:
response = self.session.get(self.room_url, headers=self.headers, timeout=10)
html = response.text
live_id = re.search(r'liveId":"(.*?)"', html).group(1)
ws_url = re.search(r'websocketUrl":"(wss://.*?)"', html).group(1)
return live_id, ws_url
except Exception as e:
print(f"获取直播信息失败: {e}")
return None, None
def connect_websocket(self):
try:
parsed = urlparse(self.ws_url)
host = parsed.hostname
port = parsed.port or (443 if parsed.scheme == 'wss' else 80)
sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
sock.connect((host, port))
if parsed.scheme == 'wss':
import ssl
context = ssl.create_default_context()
sock = context.wrap_socket(sock, server_hostname=host)
self.ws_conn = sock
self.send_handshake()
return True
except Exception as e:
print(f"WebSocket连接失败: {e}")
return False
def send_handshake(self):
handshake = (
f"GET {self.ws_url} HTTP/1.1\r\n"
f"Host: {urlparse(self.ws_url).hostname}\r\n"
"Upgrade: websocket\r\n"
"Connection: Upgrade\r\n"
"Sec-WebSocket-Key: dGhlIHNhbXBsZSBub25jZQ==\r\n"
"Sec-WebSocket-Version: 13\r\n\r\n"
)
self.ws_conn.send(handshake.encode())
def process_message(self, data):
try:
msg = json.loads(data)
if msg.get('type') == 'comment':
self.save_data({
'type': 'comment',
'user': msg.get('user', {}).get('name', '匿名用户'),
'content': msg.get('content'),
'timestamp': int(time.time())
})
elif msg.get('type') == 'like':
self.save_data({
'type': 'like',
'count': msg.get('count', 1),
'timestamp': int(time.time())
})
elif msg.get('type') == 'user_enter':
self.save_data({
'type': 'user_enter',
'user_id': msg.get('user', {}).get('id'),
'timestamp': int(time.time())
})
except Exception as e:
print(f"消息处理失败: {e}")
def save_data(self, data):
with self.lock:
self.data_buffer.append(data)
if len(self.data_buffer) >= 100:
self.flush_data()
def flush_data(self):
try:
with open(self.output_file, 'a', encoding='utf-8') as f:
for data in self.data_buffer:
f.write(json.dumps(data, ensure_ascii=False) + '\n')
self.data_buffer = []
except Exception as e:
print(f"数据保存失败: {e}")
def send_heartbeat(self):
if time.time() - self.last_heartbeat > self.heartbeat_interval:
try:
heartbeat_msg = json.dumps({'type': 'heartbeat'})
self.ws_conn.send(heartbeat_msg.encode())
self.last_heartbeat = time.time()
except Exception as e:
print(f"心跳发送失败: {e}")
def receive_messages(self):
buffer = b''
while self.is_running:
try:
data = self.ws_conn.recv(4096)
if not data:
break
buffer += data
while b'\n' in buffer:
line, buffer = buffer.split(b'\n', 1)
if line:
self.process_message(line.decode('utf-8').strip())
self.send_heartbeat()
except Exception as e:
print(f"消息接收错误: {e}")
break
def start(self):
live_id, self.ws_url = self.get_live_info()
if not all([live_id, self.ws_url]):
print("无法获取直播信息或WebSocket地址")
return False
if not self.connect_websocket():
return False
self.is_running = True
threading.Thread(target=self.receive_messages, daemon=True).start()
print(f"开始采集直播间 {live_id} 的数据...")
return True
def stop(self):
self.is_running = False
if self.ws_conn:
self.ws_conn.close()
if self.data_buffer:
self.flush_data()
print("采集已停止，数据已保存")
def main(): room_url = input("请输入快手直播间URL: ") spider = KuaishouLiveSpider(room_url)
if spider.start():
try:
while True:
time.sleep(1)
except KeyboardInterrupt:
spider.stop()
else:
print("采集启动失败")
if name == 'main': main()
