Engineering Privacy in Software - Intimate Partner Violence (IPV) Project
Spring 2020, Professor Tim Libert
Group Members:
Chenxiao Guan (chenxiag@andrew.cmu.edu)
Ziyuan Zhu (ziyuanz@andrew.cmu.edu)
Justin Souvenir Niweteto (jniwetet@andrew.cmu.edu)
Sophia Maciel (smaciel@andrew.cmu.edu)
Xiaoyong (Sheryl) Liu (xioayonl@andrew.cmu.edu)
Project URL
You can visit here for the content. API documentations can be found at here. It is automatically generated by swagger.
Technology Stack
Front-end
The client app uses React as the primary framework. It uses the Material-UI component library for some basic functionalities and stylings. It also uses Redux for state management. A more detailed explanation of the structure can be found in the client directory's README here.
Back-end
The backend was built by using Django, which is a Python framework. Detailed structure explanation can be found below in the Backend structure section.
Build
This project uses Docker and docker-compose as the build tool. We have a far-end docker server and everytime we commit to the project, we build the docker image again.
Develop
To spin-up the development server, you can either do it in a local environment by starting two development servers separately by following the guide down below, or you can you docker-compose.yml as the docker-compose specification to start the development server as a docker container.
Front-end development server
Since the front-end was scaffolded by using create-react-app, you can start the development server by navigating into the client folder (which is the root directory of the front-end), run npm install if this is the first time running the application, and once finished just run npm start and it will start a development server at localhost:3000.
Back-end development server
Navigate into server directory, use pipenv to install all dependency needed and start a virtual python environment, and run python manage.py runserver to start the backend development server at localhost:8000. Make sure you have good internet connection in order to get connected to the database, since the database was deployed using AWS RDS and needs internet to be connected.
AWS connection and credentials
We specified all third-party connections in a .env file and it is not here on the github repo. You need to configure those proper environment variables in order to make the application work locally. Add a .env file under server/ directory with proper values.
Backend Structure
All dependencies are specified in server/Pipfile.
The starting point of the project is eps_ipv_server.
Since it is a Django Project, it generally follows the Django structure, which allows several applications to exist and interact. All files are separated into different applications under server/eps_ipv_server.
eps_ipv_server
This is the root directory, which works to connect the project with Django. It contants root files such as settings.py, wsgi.py, and asgi.py, which defines the basic settings and functions of the project. Also, it contains urls.py to perform routing based on defined url patterns.
api
This is the core application of the project, which defines models and most operations of the project. Also, the directory contains serialiers.py for model instances and urls.py for viewsets redirections. Besides basic .py files, the directory contains three sub directories api, migrations, and models for more information and operations.
api/api
This directory defines viewsets contains required data for the registration and authentication process. It also contains the image upload API to securely upload files and images to the AWS cloud storage service. The subfolder auth contains two APIs to perform registration and authentication operations.
JWT authentication related logic lives here as well. This backend application uses JWT as the primary tool for securing endpoints, and all JWT related files are in the utils directory.
api/models
This is where all models created and defined. The models will determine the table structures stored in the Mysql database.
api/migrations
This directory is generated automatically when models are created and linked to the database. All .py files under the directory is to trace any changes of the models and record them in the database.