DANE

The Distributed Annotation 'n' Enrichment (DANE) system handles compute task assignment and file storage for the automatic annotation of content.

This repository contains contains the building blocks for with DANE, such as creating custom analysis workers or submitting new task.

Installation

This package can be installed through pip:

pip install dane

Configuration

DANE components are configured through the dane.config module, which is described here: https://dane.readthedocs.io/en/latest/intro.html#configuration It is however noteable that, because all DANE components are expected to rely on it, some of the DANE-server, ElasticSearch and RabbitMQ configuration are included in the default config. As such it is recommended that you create a $HOME/.dane/config.yml or $DANE_HOME/config.yml which contain machine-wide settings for how to connect to these services, which involves specifying the following settings:

DANE:
    API_URL: 'http://localhost:5500/DANE/'
    MANAGE_URL: 'http://localhost:5500/manage/'
RABBITMQ:
    HOST: 'localhost'
    PORT: 5672
    EXCHANGE: 'DANE-exchange'
    RESPONSE_QUEUE: 'DANE-response-queue'
    USER: 'guest'
    PASSWORD: 'guest'
ELASTICSEARCH:
    HOST: ['localhost']
    PORT: 9200
    USER: 'elastic'
    PASSWORD: 'changeme'
    SCHEME: 'http'
    INDEX: 'your_dane_index'

The values given here are the default values.

Usage

Examples of how to use DANE can be found in the examples/ directory.

Local Development

We moved from setup.py & requirements.txt to a single pyproject.toml. For local builds and publishing we use poetry.

For local installation:

poetry install
poetry shell

After installation the following unit test should succeed:

python -m test.test_dane

To build a wheel + source package (will end up in dist directory):

poetry build

The wheel can be conveniently tested in e.g. your own DANE worker by installing it e.g. using pip:

pip install path_to_dane_wheel_file

or with poetry

poetry add path_to_dane_wheel_file

Breaking changes after 0.3.1

Since version 0.3.1 DANE must be imported in lowercase letters:

import dane

Before version 0.3.1 you should import using uppercase letters:

import DANE