ferelight 1.1.0

Extremely lightweight and purpose-built feature extraction and retrieval engine (FERE).

FERElight | ˈferēlīt |

Extremely lightweight and purpose-built feature extraction and retrieval engine (FERE).

Installation

From PyPI

pip install ferelight

From Source

pip install git+https://github.com/FEREorg/ferelight.git

Usage

To configure the pgvector PostgreSQL connection, create a file config.json in the root directory with the following content:

{
  "DBHOST": "<host>",
  "DBPORT": "<port>",
  "DBUSER": "<user>",
  "DBPASSWORD": "<password>"
}

To run the server, please execute the following from the root directory:

pip3 install -r requirements.txt
python3 -m ferelight

You can also specify a custom path to the configuration file:

python3 -m ferelight --config /path/to/your/config.json
# or using the short option
python3 -m ferelight -c /path/to/your/config.json

Running with Docker

Using Docker Hub

The Docker image is available on Docker Hub. You can pull and run it directly:

docker pull florianspiess/ferelight:latest
docker run -p 8080:8080 florianspiess/ferelight:latest

Building Locally

To build and run the server on a Docker container locally, execute the following from the root directory:

# building the image
docker build -t ferelight .

# starting up a container
docker run -p 8080:8080 ferelight

Development

Releasing New Versions

To release a new version:

1. Update the version number in ferelight/__init__.py
2. Create a new GitHub release or tag with a version number (e.g., v1.0.1)
3. The GitHub Actions workflows will automatically:
   • Build and publish the package to PyPI
   • Build and publish the Docker image to Docker Hub (as both latest and version-specific tags)