🔴ATTENTION🔴

Project is deprecated. There are a lot of alternatives to ChatGPT, even in Russia. So there is no need for such a proxy.

At least try to create frontend with Svelte was funny

Gpt-django

It's some sort of pet-project. The main purpose is to give access to interaction with ChatGPT for my friends and relatives. So it's actualy simple django project with DRF and SvelteKit for frontend.

At this moment it's hosted on https://urf4cknmt.space

What's next?

Frontend is made now with SvelteKit. I don't know much Svelte, but I like it enough 😅

Feel free to make pull requests if you have any enhancement both for backend and frontend parts.

Django project

This project backend is bootstrapped using fandsdev/django template.

Project structure

The main django app is called app. It contains .env file for django-environ. For examples see src/app/.env.ci. Here are some usefull app-wide tools:

• app.admin — app-wide django-admin customizations (empty yet), check out usage examples
• app.test.api_client (available as api and anon fixtures within pytest) — a convinient DRF test client.

Django user model is located in the separate users app.

Also, feel free to add as much django apps as you want.

If you prefer develop from container or need it for frontend

Run the docker-compose:

$ cp .env.example .env  # default environment variables
$ cd backend && cp app/.env.ci app/.env  # default environment variables
$ docker-compose up -d
$ docker-compose exec app bash  # entering backend
$ ./manage.py createsuperuser  # creating admin user

There is volume between your and container src directories, so just do what you want.

Don't forget to run tests inside app container in such case.

Testing:

For backend

# run formatter
$ make fmt

# run lint
$ make lint

# run unit tests
$ make test

For frontend

# run formatter
$ npm format

# run lint
$ npm lint

# run unit tests
$ npm test:unit

Installing on a local machine

This project requires python 3.11. Deps are managed by pip-tools

If you prefer develop from venv

Create venv and install requirements:

$ python3 -m venv venv
$ source ./venv/bin/activate
$ pip install --upgrade pip pip-tools
$ make

Run the server:

$ cp .env.example .env  # default environment variables
$ cd backend && cp app/.env.ci app/.env  # default environment variables
$ ./manage.py migrate
$ ./manage.py createsuperuser
$ ./manage.py runserver

Backend Code requirements

Style

• Obey django's style guide.
• Configure your IDE to use flake8 for checking your python code. To run our linters manualy, do make lint
• Prefer English over your native language in comments and commit messages.
• Commit messages should contain the unique id of issue they are linked to (refs #100500)
• Every model, service and model method should have a docstring.

Code organisation

• KISS and DRY.
• Obey django best practices.
• No logic is allowed within the views or serializers. Only services and models. When a model grows beyond 500 lines of code — go create some services.
• Use PEP-484 type hints when possible.
• Prefer composition over inheritance.
• Never use signals or GenericRelations in your own code.
• No l10n is allowed in python code, use django translation.