视频解析 | Lyh's Blog

抖音无水印视频解析原理及实现

抖音无水印视频下载其实很简单，自己手动也可以获取无水印视频。
浏览器调式模式F12打开链接，查看网络请求，如图

首先获取分享短链接
抖音复制链接，得到：「时间也不早啦，洗漱完就早点休息吧~ https://v.douyin.com/eRD1x1v/ 复制此lian接，打开Dou音搜索，矗接观看視频！」
将短链接在网页上打开
首先复制你要下载的视频链接在电脑的浏览器打开，在浏览器上输入这个链接以后，可以看到页面进行了一次跳转，刚才的短连接也变成了另一个长链接(这里其实是通过 302 进行了重定向)：
https://www.iesdouyin.com/share/video/6535309780305579272/?region=CN&mid=6535310190042942211&u_code=lgmf23ik&titleType=title&did=40591980672&iid=2252231182654638&timestamp=1615379860&app=aweme&utm_campaign=client_share&utm_medium=ios&tt_from=copy&utm_source=copy
这时按一下键盘的F12或者右键选择审查元素，选择 Network 这个标签，然后选择 Media 按 clear 之后按 F5 刷新。
得到播放地址：
https://aweme.snssdk.com/aweme/v1/playwm/?video_id=aabe577a58794e00a53895f73881dc12&ratio=720p&line=0

然后把地址中playwm中的wm去掉得到地址，wm就是 watermark 水印的意思。
https://aweme.snssdk.com/aweme/v1/play/?video_id=aabe577a58794e00a53895f73881dc12&ratio=720p&line=0

复制浏览器中打开，得到的视频就是无水印视频，下载右键视频另存为即可。
许多APP，小程序或在线工具解析原理就是把以上流程用程序实现

程序实现

抓包分析接口

获取视频信息：
https://www.iesdouyin.com/web/api/v2/aweme/iteminfo/?item_ids=**视频ID**&dytk=**dytk**
其中参数dytk获取方式如下（使用手机端UA）：
https://www.iesdouyin.com/share/video/**视频ID**/?mid=1

Python实现功能

基于Python + Flask 的去水印 API

# -*- encoding: utf-8 -*-

# 导入需求模块
import requests
from flask import Flask, request, jsonify
import re

headers = {
    'User-Agent':
    'Mozilla/5.0 (Linux; Android 5.0; SM-G900P Build/LRX21T) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.25 Mobile Safari/537.36'
}

app = Flask(__name__)

# jsonify返回中文直接显示，不加这一句的话会编码Unicode
app.config['JSON_AS_ASCII'] = False

# 设置允许跨域
@app.after_request
def cors(environ):
    environ.headers['Access-Control-Allow-Origin'] = '*'
    environ.headers['Access-Control-Allow-Method'] = '*'
    environ.headers[
        'Access-Control-Allow-Headers'] = 'x-requested-with,content-type'
    return environ

# 设置404返回模板
@app.errorhandler(404)
def not_found(error):
    resp = {'status_code': '404'}
    return jsonify(resp), 404

# 无参数
@app.route('/', methods=['GET', 'POST'])
def index():
    resp = {'status_code': '404'}
    return jsonify(resp)

# @NAME :   抖音解析去水印视频
# @URL  :   /douyin/parse?id=6535309780305579272
# @URL  :   /douyin/parse?url=https://v.douyin.com/eRD1x1v/ 
@app.route('/douyin/parse', methods=['GET'])
def douyin_parsing():
    id = request.args.get('id')
    url = request.args.get('url')
    if id:
        resp = parse_by_id(id)
    elif url:
        resp = parse_by_url(url)
    else:
        resp = {'error': '参数错误！'}
    return jsonify(resp)

def get_real_addr(uri):
    url = 'https://aweme.snssdk.com/aweme/v1/play/?video_id={}&ratio=720p&line=0'.format(
        uri)
    try:
        r = requests.head(url, headers=headers, allow_redirects=False)
        addr = r.headers['Location']
        return addr
    except Exception:
        return 'error'


def parse_by_url(url):
    if '/share/video/' in url:
        id = re.findall(r'share/video/(\d+)/?\??', url)[0]
    else:
        try:
            r = requests.head(url, headers=headers, allow_redirects=False)
            id = re.findall(r'share/video/(\d+)/?\??',
                            r.headers['Location'])[0]
        except Exception:
            return {'error': 'URL错误！'}
    return parse_by_id(id)


def parse_by_id(id):
    url = 'https://www.iesdouyin.com/web/api/v2/aweme/iteminfo/?item_ids=' + id
    try:
        r = requests.get(url).json()
        video = r['item_list'][0]
        res = {}
        res['desc'] = video['desc']
        res['comment'] = video['statistics']['comment_count']
        res['digg'] = video['statistics']['digg_count']
        res['author'] = video['author']['nickname']
        res['music'] = video['music']['play_url']['uri']
        res['cover'] = video['video']['cover']['url_list'][0]
        videouri = video['video']['play_addr']['uri']
        res['video'] = get_real_addr(videouri)
        return res
    except Exception:
        return {'error': '出错了！'}


# 本地调试
if __name__ == "__main__":
    app.run(host='127.0.0.1', port=2222)