How to Retrieve and Download Douyin Dynamic Wallpaper Videos Using mitmproxy and Python

The article shows how to obtain the MP4 download links of Douyin (TikTok) dynamic wallpaper videos without writing a custom request simulator. First, install the Windows version of Douyin, start mitmweb , and open the wallpaper page; mitmproxy will display a request starting with https://www.douyin.com/aweme/v1/web/wallpaper/ . Clicking the "Download" button saves a content.data file that contains the video URLs.

Next, a Python script reads content.data , parses the JSON, removes entries with anchor_info , extracts the play_addr URL list, and uses the video_id to deduplicate identical videos. The script builds a dictionary of unique URLs and finally creates a list of download links.

<code>import json
url_list = []
url_dict = {}
with open('content.data', 'r', encoding='utf-8') as f:
    x = json.load(f)
    for i in x['aweme_list']:
        if 'anchor_info' in i:
            continue
        else:
            for i in i['video']['bit_rate']:
                url = i['play_addr']['url_list'][2]
                a = url.split('video_id=')[1].split('&line=')[0]
                url_dict[a] = url
for k, v in url_dict.items():
    url_list.append(v)
print(len(url_list))
print(url_list)
</code>

After obtaining the list of MP4 URLs, another Python script downloads each video. It uses requests.get with streaming, shows a progress bar via tqdm , and saves the files under a videos directory.

<code>import os
import requests
from tqdm import tqdm

VIDEO_PATH = r'videos'

def download(url, fname):
    resp = requests.get(url, stream=True, verify=False)
    total = int(resp.headers.get('content-length', 0))
    with open(fname, 'wb') as file, tqdm(desc=fname, total=total, unit='iB', unit_scale=True, unit_divisor=1024) as bar:
        for data in resp.iter_content(chunk_size=1024):
            size = file.write(data)
            bar.update(size)

if __name__ == "__main__":
    url_list = [
        'https://www.douyin.com/aweme/v1/play/?video_id=v0d00fg10000cagm35rc77u3k4nb0430&line=0&file_id=fec3f8eeb45e48a18f30dfd96922f659&sign=4450c5609c69d0a5c1100e6801cf25dd&is_play_url=1&source=PackSourceEnum_AWEME_DETAIL',
        # ... more URLs ...
    ]
    for url in url_list:
        video_name = url[47:67]
        video_full_path = os.path.join(VIDEO_PATH, f"{video_name}.mp4")
        download(url, video_full_path)
</code>

Running the downloader fetches the dynamic wallpaper videos, which can then be set as desktop wallpapers, providing a visually appealing and constantly changing background.