dify/api

Dify Backend API

Usage

1. Start the docker-compose stack

The backend require some middleware, including PostgreSQL, Redis, and Weaviate, which can be started together using docker-compose.

   cd ../docker
   docker-compose -f docker-compose.middleware.yaml -p dify up -d
   cd ../api

1. Copy .env.example to .env
2. Generate a SECRET_KEY in the .env file.

   sed -i "/^SECRET_KEY=/c\SECRET_KEY=$(openssl rand -base64 42)" .env

   secret_key=$(openssl rand -base64 42)
   sed -i '' "/^SECRET_KEY=/c\\
   SECRET_KEY=${secret_key}" .env

1. Create environment.

Dify API service uses Poetry to manage dependencies. You can execute poetry shell to activate the environment.

Using pip can be found below.

1. Install dependencies

=======

   poetry env use 3.10
   poetry install

In case of contributors missing to update dependencies for pyproject.toml, you can perform the following shell instead.

   poetry shell                                               # activate current environment
   poetry add $(cat requirements.txt)           # install dependencies of production and update pyproject.toml
   poetry add $(cat requirements-dev.txt) --group dev    # install dependencies of development and update pyproject.toml

1. Run migrate

Before the first launch, migrate the database to the latest version.

   poetry run python -m flask db upgrade

1. Start backend

   poetry run python -m flask run --host 0.0.0.0 --port=5001 --debug

1. Start Dify web service.
2. Setup your application by visiting http://localhost:3000...
3. If you need to debug local async processing, please start the worker service.

   poetry run python -m celery -A app.celery worker -P gevent -c 1 --loglevel INFO -Q dataset,generation,mail

The started celery app handles the async tasks, e.g. dataset importing and documents indexing.

Testing

1. Install dependencies for both the backend and the test environment

   poetry install --with dev

1. Run the tests locally with mocked system environment variables in tool.pytest_env section in pyproject.toml

   cd ../
   poetry run -C api bash dev/pytest/pytest_all_tests.sh

Usage with pip

[!NOTE]
In the next version, we will deprecate pip as the primary package management tool for dify api service, currently Poetry and pip coexist.

1. Start the docker-compose stack

The backend require some middleware, including PostgreSQL, Redis, and Weaviate, which can be started together using docker-compose.

   cd ../docker
   docker-compose -f docker-compose.middleware.yaml -p dify up -d
   cd ../api

1. Copy .env.example to .env
2. Generate a SECRET_KEY in the .env file.

   sed -i "/^SECRET_KEY=/c\SECRET_KEY=$(openssl rand -base64 42)" .env

1. Create environment.

If you use Anaconda, create a new environment and activate it

   conda create --name dify python=3.10
   conda activate dify

1. Install dependencies

   pip install -r requirements.txt

1. Run migrate

Before the first launch, migrate the database to the latest version.

   flask db upgrade

1. Start backend:

   flask run --host 0.0.0.0 --port=5001 --debug

1. Setup your application by visiting http://localhost:5001/console/api/setup or other apis...
2. If you need to debug local async processing, please start the worker service.

   celery -A app.celery worker -P gevent -c 1 --loglevel INFO -Q dataset,generation,mail

The started celery app handles the async tasks, e.g. dataset importing and documents indexing.