Streamlit Cognito Auth
A Streamlit component for authenticating users with AWS Cognito
Installation
To install the package, you can use pip:
pip install streamlit-cognito-auth
Usage
To use the package, you can import the CognitoAuthenticator
class from the streamlit_cognito_auth
package, and create an instance of it, passing your pool_id
, app_client_id
and app_client_secret
as the arguments.
from streamlit_cognito_auth import CognitoAuthenticator
To perform login, you can call the login() method on the authenticator instance and check the returned value, if it's True then the user is logged in, otherwise the login process failed.
is_logged_in = authenticator.login()
if not is_logged_in:
print("Login failed")
To perform logout, you can call the logout() method on the authenticator instance
authenticator.logout()
You can also get the logged in user's username by calling get_username() method on the authenticator instance
username = authenticator.get_username()
You can find the full example code in examples/example.py
file.
Example
You can find an example of how to use the package in the examples directory.
To run the example file, you can use the following command, while replacing your_pool_id
, your_app_client_id
and your_app_client_secret
with the actual values:
export POOL_ID="your_pool_id"
export APP_CLIENT_ID="your_app_client_id"
export APP_CLIENT_SECRET="your_app_client_secret"
cd examples
streamlit run example.py
or in windows
set POOL_ID="your_pool_id"
set APP_CLIENT_ID="your_app_client_id"
set APP_CLIENT_SECRET="your_app_client_secret"
cd examples
streamlit run example.py
login
logout
Login with temporary password
Reset password and login
Limitations
- This package has been tested and known to work with Amazon Cognito pools that have an app client with a secret enabled and using the SRP protocol. Other configurations of Cognito pools may not be supported and have not been tested.
- This package has been tested and known to work with python 3.8 in Linux environment. It may not work with other versions of python or other operating systems.
Features
Credits
- This package is inspired by the work of mkhorasani/Streamlit-Authenticator, and we would like to thank the author for their work.
- JWT verification functionality is based on the work of awslabs/aws-support-tools, and we would like to thank the author for their work.
- This package uses the
pycognito
library for authentication, which is a Python library that provides a simple interface for working with AWS Cognito. We would like to express our gratitude to the authors of pycognito
for their work and for providing an excellent library for working with AWS Cognito. - We would also like to thank OpenAI's ChatGPT for providing helpful suggestions and examples throughout the development of this project.
- Special thanks to mrtj for the nice hosted UI feature. (#6)