A Selenium-based automated TikTok video uploader

• Installation
  • MacOS, Windows and Linux
    • Downloading from PyPI (Recommended)
    • Building from source
• Usage
  • 💻 Command Line Interface (CLI)
  • ⬆ Uploading Videos
  • 🫵 Mentions and Hashtags
  • 🪡 Stitches, Duets and Comments
  • 🔐 Authentication
  • 👀 Browser Selection
  • 🚲 Custom WebDriver Driver Options
  • 🤯 Headless Browsers
  • 🔨 Initial Setup
• ♻️ Examples
• 📝 Notes
• Accounts made with

A perquisite to using this program is the installation of a Selenium-compatible web browser. Google Chrome is recommended.

Install Python 3 or greater from python.org

Install tiktok-uploader using pip

pip install tiktok-uploader

Installing from source allows greater flexibility to modify the module's code to extend default behavior.

First, clone and move into the repository. Next, install hatch, the build tool used for this project ¹. Then, build the project. Finally, install the project with the -e or editable flag.

git clone https://github.com/wkaisertexas/tiktok-uploader.git
cd tiktok-uploader
pip install hatch
hatch build
pip install -e . 

tiktok-uploader works by duplicating your browser's cookies which tricks TikTok into believing you are logged in on a remote-controlled browser.

Using the CLI is as simple as calling tiktok-uploader with your videos: path (-v), description(-d) and cookies (-c)

tiktok-uploader -v video.mp4 -d "this is my escaped \"description\"" -c cookies.txt

from tiktok_uploader.upload import upload_video, upload_videos
from tiktok_uploader.auth import AuthBackend

# single video
upload_video('video.mp4', 
            description='this is my description', 
            cookies='cookies.txt')

# Multiple Videos
videos = [
    {
        'path': 'video.mp4', 
        'description': 'this is my description'
    },
    {
        'path': 'video2.mp4', 
        'description': 'this is also my description'
    }
]

auth = AuthBackend(cookies='cookies.txt')
upload_videos(videos=videos, auth=auth)

This library revolves around the upload_videos function which takes in a list of videos which have filenames and descriptions and are passed as follows:

from tiktok_uploader.upload import upload_videos
from tiktok_uploader.auth import AuthBackend

videos = [
    {
        'video': 'video0.mp4',
        'description': 'Video 1 is about ...'
    },
    {
        'video': 'video1.mp4',
        'description': 'Video 2 is about ...'
    }
]

auth = AuthBackend(cookies='cookies.txt')
failed_videos = upload_videos(videos=videos, auth=auth)

for video in failed_videos: # each input video object which failed
    print(f'{video['video']} with description "{video['description']}" failed')

Mentions and Hashtags now work so long as they are followed by a space. However, you as the user are responsible for verifying a mention or hashtag exists before posting

Example:

from tiktok_uploader.upload import upload_video

upload_video('video.mp4', '#fyp @icespicee', 'cookies.txt')

To set whether or not a video uploaded allows stitches, comments or duet, simply specify comment, stitch and/or duet as keyword arguments to upload_video or upload_videos.

upload_video(..., comment=True, stitch=True, duet=True)

Comments, Stitches and Duets are allowed by default

To set a proxy, currently only works with chrome as the browser, allow user:pass auth.

# proxy = {'user': 'myuser', 'pass': 'mypass', 'host': '111.111.111', 'port': '99'}  # user:pass
proxy = {'host': '111.111.111', 'port': '99'}
upload_video(..., proxy=proxy)

The datetime to schedule the video will be treated with the UTC timezone.
The scheduled datetime must be at least 20 minutes in the future and a maximum of 10 days.

import datetime
schedule = datetime.datetime(2020, 12, 20, 13, 00)
upload_video(..., schedule=schedule)

Authentication uses your browser's cookies. This workaround was done due to TikTok's stricter stance on authentication by a Selenium-controlled browser.

Your sessionid is all that is required for authentication and can be passed as an argument to nearly any function

🍪 Get cookies.txt makes getting cookies in a NetScape cookies format.

After installing, open the extensions menu on TikTok.com and click 🍪 Get cookies.txt to reveal your cookies. Select Export As ⇩ and specify a location and name to save.

Optionally, cookies_list is a list of dictionaries with keys name, value, domain, path and expiry which allow you to pass your own browser cookies.

Example:

cookies_list = [
    {
        'name': 'sessionid',
        'value': '**your session id**',
        'domain': 'https://tiktok.com',
        'path': '/',
        'expiry': '10/8/2023, 12:18:58 PM'
    }
]

upload_video(..., cookies_list=cookies_list)

Google Chrome is the preferred browser for TikTokUploader. The default anti-detection techniques used in this packaged are optimized for this. However, if you wish to use a different browser you may specify the browser in upload_video or upload_videos.

from tiktok_uploader.upload import upload_video

from random import choice

BROWSERS = [
    'chrome',
    'safari',
    'chromium',
    'edge',
    'firefox'
]

# randomly picks a web browser 
upload_video(..., browser=choice(BROWSERS))

✅ Supported Browsers:

• Chrome (Recommended)
• Safari
• Chromium
• Edge
• FireFox

Default modifications to Selenium are applied which help it avoid being detected by TikTok.

However, you may pass a custom driver configuration options. Simply pass options as a keyword argument to either upload_video or upload_videos.

from selenium.webdriver.chrome.options import Options

options = Options()

options.add_argument('start-maximized')

upload_videos(..., options=options)

Note: Make sure to use the right selenium options for your browser

Headless browsing only works on Chrome. When using Chrome, adding the --headless flag using the CLI or passing headless as a keyword argument to upload_video or upload_videos is all that is required.

upload_video(..., headless=True)
upload_videos(..., headless=True)

WebDriverManager is used to manage driver versions.

On initial startup, you may be prompted to install the correct driver for your selected browser. However, for Chrome and Edge the driver is automatically installed.

• Basic Upload Example: Uses upload_video to make one post.

• Multiple Videos At Once: Uploads the same video multiple times using upload_videos.

• Series Upload Example: Videos are read from a CSV file using Pandas. A video upload attempt is made and if and only if it is successful will the video be marked as uploaded.

This bot is not fool proof. Though I have not gotten an official ban, the video will fail to upload after too many uploads. In testing, waiting several hours was sufficient to fix this problem. For this reason, please thing of this more as a scheduled uploader for TikTok videos, rather than a spam bot.

• @C_Span - A split-screen channel with mobile games below featuring clips from C-Span's YouTube channel
• @habit_track - A Reddit bot to see which SubReddit is most viral on TikTok