douyinliveMan.py
發(fā)布日期:2025/11/20 19:19:16 瀏覽量:
#!/usr/bin/python
# coding:utf-8
# @FileName: liveMan.py
# @Time: 2024/1/2 21:51
# @Author: bubu
# @Project: douyinLiveWebFetcher
import codecs
import gzip
import hashlib
import random
import re
import ssl
import string
import subprocess
import urllib.parse
from contextlib import contextmanager
from unittest.mock import patch
import requests
import websocket
from py_mini_racer import MiniRacer
from protobuf.douyin import *
@contextmanager
def patched_popen_encoding(encoding=’utf-8’):
original_popen_init = subprocess.Popen.__init__
def new_popen_init(self, *args, **kwargs):
kwargs[’encoding’] = encoding
original_popen_init(self, *args, **kwargs)
with patch.object(subprocess.Popen, ’__init__’, new_popen_init):
yield
def generateSignature(wss, script_file=’sign.js’):
"""
出現(xiàn)gbk編碼問題則修改 python模塊subprocess.py的源碼中Popen類的__init__函數(shù)參數(shù)encoding值為 "utf-8"
"""
params = ("live_id,aid,version_code,webcast_sdk_version,"
"room_id,sub_room_id,sub_channel_id,did_rule,"
"user_unique_id,device_platform,device_type,ac,"
"identity").split(’,’)
wss_params = urllib.parse.urlparse(wss).query.split(’&’)
wss_maps = {i.split(’=’)[0]: i.split("=")[-1] for i in wss_params}
tpl_params = [f"{i}={wss_maps.get(i, ’’)}" for i in params]
param = ’,’.join(tpl_params)
md5 = hashlib.md5()
md5.update(param.encode())
md5_param = md5.hexdigest()
with codecs.open(script_file, ’r’, encoding=’utf8’) as f:
script = f.read()
ctx = MiniRacer()
ctx.eval(script)
try:
signature = ctx.call("get_sign", md5_param)
return signature
except Exception as e:
print(e)
# 以下代碼對(duì)應(yīng)js腳本為sign_v0.js
# context = execjs.compile(script)
# with patched_popen_encoding(encoding=’utf-8’):
# ret = context.call(’getSign’, {’X-MS-STUB’: md5_param})
# return ret.get(’X-Bogus’)
def generateMsToken(length=107):
"""
產(chǎn)生請(qǐng)求頭部cookie中的msToken字段,其實(shí)為隨機(jī)的107位字符
:param length:字符位數(shù)
:return:msToken
"""
random_str = ’’
base_str = string.ascii_letters + string.digits + ’=_’
_len = len(base_str) - 1
for _ in range(length):
random_str += base_str[random.randint(0, _len)]
return random_str
class DouyinLiveWebFetcher:
def __init__(self, live_id):
"""
直播間彈幕抓取對(duì)象
:param live_id: 直播間的直播id,打開直播間web首頁的鏈接如:https://live.douyin.com/261378947940,
其中的261378947940即是live_id
"""
self.__ttwid = None
self.__room_id = None
self.live_id = live_id
self.live_url = "https://live.douyin.com/"
self.user_agent = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) " \
"Chrome/120.0.0.0 Safari/537.36"
self.ball_game = None # 添加一個(gè)屬性來存儲(chǔ)BallGame實(shí)例
def set_ball_game(self, ball_game):
self.ball_game = ball_game # 設(shè)置BallGame實(shí)例
def start(self):
self._connectWebSocket()
def stop(self):
self.ws.close()
@property
def ttwid(self):
"""
產(chǎn)生請(qǐng)求頭部cookie中的ttwid字段,訪問抖音網(wǎng)頁版直播間首頁可以獲取到響應(yīng)cookie中的ttwid
:return: ttwid
"""
if self.__ttwid:
return self.__ttwid
headers = {
"User-Agent": self.user_agent,
}
try:
response = requests.get(self.live_url, headers=headers)
response.raise_for_status()
except Exception as err:
print("【X】Request the live url error: ", err)
else:
self.__ttwid = response.cookies.get(’ttwid’)
return self.__ttwid
@property
def room_id(self):
"""
根據(jù)直播間的地址獲取到真正的直播間roomId,有時(shí)會(huì)有錯(cuò)誤,可以重試請(qǐng)求解決
:return:room_id
"""
if self.__room_id:
return self.__room_id
url = self.live_url + self.live_id
headers = {
"User-Agent": self.user_agent,
"cookie": f"ttwid={self.ttwid}&msToken={generateMsToken()}; __ac_nonce=0123407cc00a9e438deb4",
}
try:
response = requests.get(url, headers=headers)
response.raise_for_status()
except Exception as err:
print("【X】Request the live room url error: ", err)
else:
match = re.search(r’roomId\\":\\"(\d+)\\"’, response.text)
if match is None or len(match.groups()) < 1:
print("【X】No match found for roomId")
self.__room_id = match.group(1)
return self.__room_id
def _connectWebSocket(self):
"""
連接抖音直播間websocket服務(wù)器,請(qǐng)求直播間數(shù)據(jù)
"""
wss = ("wss://webcast5-ws-web-hl.douyin.com/webcast/im/push/v2/?app_name=douyin_web"
"&version_code=180800&webcast_sdk_version=1.0.14-beta.0"
"&update_version_code=1.0.14-beta.0&compress=gzip&device_platform=web&cookie_enabled=true"
"&screen_width=1536&screen_height=864&browser_language=zh-CN&browser_platform=Win32"
"&browser_name=Mozilla"
"&browser_version=5.0%20(Windows%20NT%2010.0;%20Win64;%20x64)%20AppleWebKit/537.36%20(KHTML,"
"%20like%20Gecko)%20Chrome/126.0.0.0%20Safari/537.36"
"&browser_online=true&tz_name=Asia/Shanghai"
"&cursor=d-1_u-1_fh-7392091211001140287_t-1721106114633_r-1"
f"&internal_ext=internal_src:dim|wss_push_room_id:{self.room_id}|wss_push_did:7319483754668557238"
f"|first_req_ms:1721106114541|fetch_time:1721106114633|seq:1|wss_info:0-1721106114633-0-0|"
f"wrds_v:7392094459690748497"
f"&host=https://live.douyin.com&aid=6383&live_id=1&did_rule=3&endpoint=live_pc&support_wrds=1"
f"&user_unique_id=7319483754668557238&im_path=/webcast/im/fetch/&identity=audience"
f"&need_persist_msg_count=15&insert_task_id=&live_reason=&room_id={self.room_id}&heartbeatDuration=0")
signature = generateSignature(wss)
wss += f"&signature={signature}"
headers = {
"cookie": f"ttwid={self.ttwid}",
’user-agent’: self.user_agent,
}
self.ws = websocket.WebSocketApp(wss,
header=headers,
on_open=self._wsOnOpen,
on_message=self._wsOnMessage,
on_error=self._wsOnError,
on_close=self._wsOnClose)
try:
# 禁用 SSL 驗(yàn)證
self.ws.run_forever(sslopt={"cert_reqs": ssl.CERT_NONE})
except Exception:
self.stop()
raise
def _wsOnOpen(self, ws):
"""
連接建立成功
"""
print("WebSocket connected.")
def _wsOnMessage(self, ws, message):
"""
接收到數(shù)據(jù)
:param ws: websocket實(shí)例
:param message: 數(shù)據(jù)
"""
# 根據(jù)proto結(jié)構(gòu)體解析對(duì)象
package = PushFrame().parse(message)
response = Response().parse(gzip.decompress(package.payload))
# 返回直播間服務(wù)器鏈接存活確認(rèn)消息,便于持續(xù)獲取數(shù)據(jù)
if response.need_ack:
ack = PushFrame(log_id=package.log_id,
payload_type=’ack’,
payload=response.internal_ext.encode(’utf-8’)
).SerializeToString()
ws.send(ack, websocket.ABNF.OPCODE_BINARY)
# 根據(jù)消息類別解析消息體
for msg in response.messages_list:
method = msg.method
try:
{
’WebcastChatMessage’: self._parseChatMsg, # 聊天消息
’WebcastGiftMessage’: self._parseGiftMsg, # 禮物消息
’WebcastLikeMessage’: self._parseLikeMsg, # 點(diǎn)贊消息
’WebcastMemberMessage’: self._parseMemberMsg, # 進(jìn)入直播間消息
’WebcastSocialMessage’: self._parseSocialMsg, # 關(guān)注消息
’WebcastRoomUserSeqMessage’: self._parseRoomUserSeqMsg, # 直播間統(tǒng)計(jì)
’WebcastFansclubMessage’: self._parseFansclubMsg, # 粉絲團(tuán)消息
’WebcastControlMessage’: self._parseControlMsg, # 直播間狀態(tài)消息
’WebcastEmojiChatMessage’: self._parseEmojiChatMsg, # 聊天表情包消息
’WebcastRoomStatsMessage’: self._parseRoomStatsMsg, # 直播間統(tǒng)計(jì)信息
’WebcastRoomMessage’: self._parseRoomMsg, # 直播間信息
’WebcastRoomRankMessage’: self._parseRankMsg, # 直播間排行榜信息
’WebcastRoomStreamAdaptationMessage’: self._parseRoomStreamAdaptationMsg, # 直播間流配置
}.get(method)(msg.payload)
except Exception:
pass
def _wsOnError(self, ws, error):
print("WebSocket error: ", error)
def _wsOnClose(self, ws, *args):
print("WebSocket connection closed.")
def _parseChatMsg(self, payload):
"""聊天消息"""
message = ChatMessage().parse(payload)
user_name = message.user.nick_name
user_id = message.user.id
content = message.content
print(f"【聊天msg】[{user_id}]{user_name}: {content}")
if self.ball_game:
if content == ’1’:
self.ball_game.move_left(None)
elif content == ’2’:
self.ball_game.move_right(None)
def _parseGiftMsg(self, payload):
"""禮物消息"""
message = GiftMessage().parse(payload)
user_name = message.user.nick_name
gift_name = message.gift.name
gift_cnt = message.combo_count
print(f"【禮物msg】{user_name} 送出了 {gift_name}x{gift_cnt}")
def _parseLikeMsg(self, payload):
’’’點(diǎn)贊消息’’’
message = LikeMessage().parse(payload)
user_name = message.user.nick_name
count = message.count
print(f"【點(diǎn)贊msg】{user_name} 點(diǎn)了{(lán)count}個(gè)贊")
def _parseMemberMsg(self, payload):
’’’進(jìn)入直播間消息’’’
message = MemberMessage().parse(payload)
user_name = message.user.nick_name
user_id = message.user.id
gender = ["女", "男"][message.user.gender]
print(f"【進(jìn)場(chǎng)msg】[{user_id}][{gender}]{user_name} 進(jìn)入了直播間")
def _parseSocialMsg(self, payload):
’’’關(guān)注消息’’’
message = SocialMessage().parse(payload)
user_name = message.user.nick_name
user_id = message.user.id
print(f"【關(guān)注msg】[{user_id}]{user_name} 關(guān)注了主播")
def _parseRoomUserSeqMsg(self, payload):
’’’直播間統(tǒng)計(jì)’’’
message = RoomUserSeqMessage().parse(payload)
current = message.total
total = message.total_pv_for_anchor
print(f"【統(tǒng)計(jì)msg】當(dāng)前觀看人數(shù): {current}, 累計(jì)觀看人數(shù): {total}")
def _parseFansclubMsg(self, payload):
’’’粉絲團(tuán)消息’’’
message = FansclubMessage().parse(payload)
content = message.content
print(f"【粉絲團(tuán)msg】 {content}")
def _parseEmojiChatMsg(self, payload):
’’’聊天表情包消息’’’
message = EmojiChatMessage().parse(payload)
emoji_id = message.emoji_id
user = message.user
common = message.common
default_content = message.default_content
print(f"【聊天表情包id】 {emoji_id},user:{user},common:{common},default_content:{default_content}")
def _parseRoomMsg(self, payload):
message = RoomMessage().parse(payload)
common = message.common
room_id = common.room_id
print(f"【直播間msg】直播間id:{room_id}")
def _parseRoomStatsMsg(self, payload):
message = RoomStatsMessage().parse(payload)
display_long = message.display_long
print(f"【直播間統(tǒng)計(jì)msg】{display_long}")
def _parseRankMsg(self, payload):
message = RoomRankMessage().parse(payload)
ranks_list = message.ranks_list
print(f"【直播間排行榜msg】{ranks_list}")
def _parseControlMsg(self, payload):
’’’直播間狀態(tài)消息’’’
message = ControlMessage().parse(payload)
if message.status == 3:
print("直播間已結(jié)束")
self.stop()
def _parseRoomStreamAdaptationMsg(self, payload):
message = RoomStreamAdaptationMessage().parse(payload)
adaptationType = message.adaptation_type
print(f’直播間adaptation: {adaptationType}’)
馬上咨詢: 如果您有業(yè)務(wù)方面的問題或者需求,歡迎您咨詢!我們帶來的不僅僅是技術(shù),還有行業(yè)經(jīng)驗(yàn)積累。
QQ: 39764417/308460098 Phone: 13 9800 1 9844 / 135 6887 9550 聯(lián)系人:石先生/雷先生