0x01 背景介绍
平时刷抖音有不少有趣的视频，有的随手点个喜欢有的点喜欢收藏留着今后再看，但久而久之这些碎片化内容没被整理出来，看完也就忘了导致收藏夹越来越多，想把这些视频都下载下来做个整理，把有用小视频存档素材或者转化成摘录，但抖音本身没有批量下载视频的功能，因此需要脚本实现如下两个我的需求：
- 批量下载个人中心点赞收藏的无水印视频
- 边下载边取消喜欢收藏信息
0x02 需求实现
本着代码能白嫖就白嫖的理念，写之前在Github发现TikTokDownload可以批量下载视频，于是在这个脚本基础上进行加工，主要去掉了一些冗余代码和增加了取消喜欢的模块；批量下载接口只要没有开隐私保护不需要认证，取消喜欢接口需要携带认证的headers。在抓取请求时发现抖音做了SSL Pinning限制正常bp同网段抓包不行，放进虚拟机抓包或rvictl即可正常抓包：
调整测试脚本如下：
#!/usr/bin/env python
# -*- encoding: utf-8 -*-
# @Modify : https://github.com/Johnserf-Seed/TikTokDownload
# @Summary : Batch download tiktok video and cancel like
# @Usage : Configure login_header and personal_home_page and run
import os
import re
import json
import time
import requests
import traceback
class TikTok(object):
def __init__(self):
self.per_page_download_count = 34
self.number = 0
self.flag = True
self.sec_uid = None
self.mode = "like" # post|like
self.headers = {
'user-agent': 'Mozilla/5.0 (Linux; Android 8.0; Pixel 2 Build/OPD3.170816.012) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/87.0.4280.88 Mobile Safari/537.36 Edg/87.0.664.66'
}
self.login_header = {
'sdk-version': '2',
'passport-sdk-version': '18',
'x-Tt-Token': 'xxx-1.0.1',
'User-Agent': 'okhttp/3.10.0.1',
'X-SS-REQ-TICKET': 'xxx',
'Accept-Encoding': 'gzip, deflate',
'Cookie': 'xxx',
'X-Ladon': 'xxxx/xxxx',
'X-Khronos': 'xxxx',
'X-Gorgon': 'xxxx'
}
self.personal_home_page = "https://v.douyin.com/xxxx/"
self.init_url()
def init_url(self, max_cursor=0):
try:
r = requests.get(url=self.personal_home_page)
self.sec_uid = re.findall('/user/(.*?)\?', str(r.url))[0] if \
re.findall(r'/user/(.*?)\?', str(r.url))[0] else r.url[28:83]
api_url = 'https://www.iesdouyin.com/web/api/v2/aweme/%s/?sec_uid=' \
'%s&count=%s&max_cursor=%s&aid=1128&_signature=PDHVOQAAX' \
'MfFyj02QEpGaDwx1S&dytk=' % (
self.mode, self.sec_uid, str(self.per_page_download_count), max_cursor)
response = requests.get(
url=api_url, headers=self.headers, timeout=30)
html = json.loads(response.content.decode())
max_cursor = html['max_cursor']
result = html['aweme_list']
self.parseVideo(result, max_cursor)
except BaseException:
traceback.print_exc()
def parseVideo(self, video_result, max_cursor):
author_list, video_list, aweme_id, nickname = [], [], [], []
for _ in range(self.per_page_download_count):
try:
author_list.append(str(video_result[_]['desc']))
video_list.append(
str(video_result[_]['video']['play_addr']['url_list'][0]))
aweme_id.append(str(video_result[_]['aweme_id']))
nickname.append(str(video_result[_]['author']['nickname']))
except BaseException:
pass
self.downloadVideo(
author_list,
video_list,
nickname,
aweme_id,
max_cursor)
def downloadVideo(
self,
author_list,
video_list,
nickname,
aweme_id,
max_cursor):
for _ in range(self.per_page_download_count):
try:
video = requests.get(
video_list[_], headers=self.headers, timeout=30)
start = time.time()
init_size, chunk_size, content_size = 0, 1024, int(
video.headers["content-length"])
try:
os.makedirs(
os.path.join(
"./download/",
self.mode,
nickname[_]))
except BaseException:
pass
if video.status_code == 200:
self.number += 1
print('*' * 80)
print(
'\r' + '[ No ]: %s | %s | %s' %
(self.number, max_cursor, aweme_id[_]))
print('[ Video ]: ' +
''.join(author_list[_][:30]) +
'...' +
'({size:.2f} MB)'.format(size=content_size /
chunk_size /
1024))
video_url = os.path.join(
"./download/",
self.mode,
nickname[_],
re.sub(
r'[\\/:*?"<>|\r\n]+',
"_",
author_list[_]) +
".mp4")
if not os.path.exists(video_url):
with open(video_url, 'wb') as file:
for data in video.iter_content(
chunk_size=chunk_size):
file.write(data)
init_size += len(data)
print('\r' + '[Progress]: %s%.2f%%' % (
'>' * int(init_size * 50 / content_size), float(init_size / content_size * 100)),
end=' ')
end = time.time()
print('\n[Done]: Use: %.2fs' % (end - start))
else:
print('\n' + '[Pass]: File exists')
self.cancel_favorite(aweme_id[_])
except Exception as e:
print("Error({}),download next\r".format(e))
break
self.next_data(max_cursor)
def next_data(self, max_cursor):
api_naxt_post_url = 'https://www.iesdouyin.com/web/api/v2/aweme/%s/?sec_uid=%s' \
'&count=%s&max_cursor=%s&aid=1128&_signature=RuMN1wAAJu7w0' \
'.6HdIeO2EbjDc&dytk=' % (self.mode, self.sec_uid,
str(self.per_page_download_count), max_cursor)
index, result = 0, []
while self.flag:
if max_cursor == 0:
self.flag = False
return
index += 1
time.sleep(0.3)
try:
response = requests.get(
url=api_naxt_post_url, headers=self.headers)
html = json.loads(response.content.decode())
max_cursor = html['max_cursor']
result = html['aweme_list']
except BaseException:
traceback.print_exc()
self.parseVideo(result, max_cursor)
def cancel_favorite(self, aweme_id):
try:
self.login_header['x-common-params-v2'] = 'channel=App%20Store&version_code=13.3.0&app_name=aweme&vid=' \
'C8CAFFA5-41C9-47A5-92B9-88A9265FEB4A&app_version=13.3.0&mcc' \
'_mnc=46011&device_id=69577004385&aid=1128&screen_width=1125' \
'&openudid=33c4e33cbc64128d9e7f3c3c4c8e04ed94b6c827&os_api=1' \
'8&os_version=13.5.1&device_platform=iphone&build_number=133' \
'014&device_type=iPhone11,2&iid=2744012310333677&idfa=784369' \
'4E-DFBB-40CF-9628-D1846C97CB7D&js_sdk_version=1.85.0.16&cdi' \
'd=4BC5D665-979C-4834-933F-C79A0C6948DD'
data = {
'aweme_id': aweme_id,
'channel_id': 3,
'type': 0
}
cancel_url = "https://api26-normal-lq.amemv.com/aweme/v1/commit/item/digg/?tma_jssdk_version=1.85.0.16&ac" \
"=WIFI&is_vcd=1&"
req = requests.post(
cancel_url,
data=data,
headers=self.login_header)
res = json.loads(req.content)
if res['is_digg']:
print(
"[Cancel]: {} removed the favorite list success".format(aweme_id))
else:
print(
"[Cancel]: {} removed the favorite list fail".format(aweme_id))
except BaseException:
traceback.print_exc()
if __name__ == "__main__":
RTK = TikTok()
input('[ ALL Done ]: Batch download has been completed. Enter any key to exit')
sys.exit()
0x03 最后随笔
抖音批量下载、批量取消点赞这样的小功能发现网上有不少和我有着同样的需求，甚至在淘宝发现有人卖这样的服务，看来又是一枚可以躺着赚钱，三年起步最高七年的姿势了，猜测对于像批量取消点赞这样需要认证的服务，那些卖家拿到买家的认证信息后可能还会去刷赞刷评论等，花了钱可能还要帮别人"助纣为虐"..
---The END---