This is an unofficial api wrapper for TikTok.com in python. With this api you are able to call most trending and fetch specific user information as well as much more.
This api is designed to retrieve data TikTok. It can not be used post or upload content to TikTok on the behalf of a user. It has no support any user-authenticated routes, if you can't access it while being logged out on their website you can't access it here.
These sponsors have paid to be placed here and beyond that I do not have any affiliation with them, the TikTokAPI package will always be free and open-source. If you wish to be a sponsor of this project check out my GitHub sponsors page.
![TikApi](https://raw.githubusercontent.com/davidteather/TikTok-Api/master/imgs/tikapi.png)
![Ensemble Data](https://raw.githubusercontent.com/andrearama/TikTok-Api/master/imgs/EnsembleData.png)
We provide 100+ Million posts / day to the largest Marketing and Social listening platforms.
You can find the full documentation here, the TikTokApi Class is where you'll probably spend most of your time.
To get started using this api follow the instructions below.
Note: If you want to learn how to web scrape websites check my free and open-source course for learning everything web scraping
- Star the repo 😎
- Consider sponsoring me on GitHub
- Send me an email or a LinkedIn message telling me what you're using the API for, I really like hearing what people are using it for.
- Submit PRs for issues :)
If you run into an issue please check the closed issues on the github, although feel free to re-open a new issue if you find an issue that's been closed for a few months. The codebase can and does run into similar issues as it has before, because TikTok changes things up.
pip install TikTokApi
python -m playwright install
If you would prefer a video walk through of setting up this package YouTube video just for that.
If you want a quick video to listen for TikTok Live events in python.
Clone this repository onto a local machine (or just the Dockerfile since it installs TikTokApi from pip) then run the following commands.
docker pull mcr.microsoft.com/playwright:focal
docker build . -t tiktokapi:latest
docker run -v TikTokApi --rm tiktokapi:latest python3 your_script.py
Note this assumes your script is named your_script.py and lives in the root of this directory.
Please don't open an issue if you're experiencing one of these just comment if the provided solution do not work for you.
- Browser Has no Attribute - make sure you ran
python3 -m playwright install
, if your error persists try the playwright-python quickstart guide and diagnose issues from there.
Here's a quick bit of code to get the most recent trending videos on TikTok. There's more examples in the examples directory.
Note: If you want to learn how to web scrape websites check my free and open-source course for web scraping
from TikTokApi import TikTokApi
# Watch https://www.youtube.com/watch?v=-uCt1x8kINQ for a brief setup tutorial
with TikTokApi() as api:
for trending_video in api.trending.videos(count=50):
# Prints the author's username of the trending video.
print(trending_video.author.username)
Note: Jupyter (ipynb) only works on linux, see microsoft/playwright-python #178
To run the example scripts from the repository root, make sure you use the -m
option on python.
python -m examples.get_trending
You can access the dictionary type of an object using .as_dict
. On a video this may look like
this, although TikTok changes their structure from time to time so it's worth investigating the structure of the dictionary when you use this package.
All changes will be noted on #803 if you want more information.
This package has been difficult to maintain due to it's structure, difficult to work with since the user of the package must write parsing methods to extract information from dictionaries, more memory intensive than it needs to be (although this can be further improved), and in general just difficult to work with for new users.
As a result, I've decided to at least attempt to remedy some of these issues, the biggest changes are that
- The package has shifted to using classes for different TikTok objects resulting in an easier, higher-level programming experience.
- All methods that used to return a list of objects have been switched to using iterators, to hopefully decrease memory utilization for most users.
You'll probably need to use this beyond just for legacy support, since not all attributes are parsed out and attached to the different objects.
You may want to use this as a workaround for legacy applications while you upgrade the rest of the app. I'd suggest that you do eventually upgrade to using the higher-level approach fully.
user = api.user(username='therock')
user.as_dict # -> dict of the user_object
for video in user.videos():
video.as_dict # -> dict of TikTok's video object as found when requesting the videos endpoint
Here's a few more examples that help illustrate the differences in the flow of the usage of the package with V5.
# V4
api = TikTokApi.get_instance()
trending_videos = api.by_trending()
#V5.1
with TikTokApi() as api: # .get_instance no longer exists
for trending_video in api.trending.videos():
# do something
Where in V4 you had to extract information yourself, the package now handles that for you. So it's much easier to do chained related function calls.
# V4
trending_videos = api.by_trending()
for video in trending_videos:
# The dictionary responses are also different depending on what endpoint you got them from
# So, it's usually more painful than this to deal with
trending_user = api.get_user(id=video['author']['id'], secUid=video['author']['secUid'])
# V5
# This is more complicated than above, but it illustrates the simplified approach
for trending_video in api.trending.videos():
user_stats = trending_video.author.info_full['stats']
if user_stats['followerCount'] >= 10000:
# maybe save the user in a database