README.md

RAMP bundle

This repository contains the following RAMP modules:

• ramp-database - RAMP database module
• ramp-engine - RAMP runner service
• databoard - RAMP frontend server

The modules can be installed independently but have been added to the same repository so they can be kept in sync.

Contents

• Installation
• Configuration
• Testing

Installation

1. Retrieve the main repository

   git clone https://github.com/paris-saclay-cds/ramp-board
   cd ramp-board

2. Install Python dependencies using conda or pip

   • with conda

     # Make sure you run the latest version of conda
     conda update conda
     # Set up the virtual environment
     conda env create -f environment.yml
     # Activate it
     conda activate ramp-server

   • with pip

     pip install -r requirements.txt

3. Install all the elements at once

   make install

   or each project independently

   cd ramp-<project>
   pip install .

Configuration

databoard comes with a default configuration that can be overwritten via environment variables or a JSON configuration file.

Note: If you use both (e.g. general JSON config file AND environment variables for sensitive parameters like passwords), know that the user configuration file is loaded last so it will take precedence over the environment variables. You should therefore remove these entries from the JSON file.

JSON configuration file

The DATABOARD_USER_CONFIG environment variable must be set to the absolute path of the config file in order to be used by the ramp-server.

export DATABOARD_USER_CONFIG=/path/to/userconfig.json

Environment variables

export DATABOARD_STAGE=TESTING (or PRODUCTION)

# used when DATABOARD_STAGE=TESTING
export DATABOARD_DB_URL_TEST=postgresql://<db_user>:<db_password>@localhost/<db_name_test>
export DATABOARD_DEPLOYMENT_PATH_TEST=/path/to/<db_name_test>

# used when DATABOARD_STAGE=PRODUCTION
export DATABOARD_DB_URL=postgresql://<db_user>:<db_password>@localhost/<db_name_prod>
export DATABOARD_DEPLOYMENT_PATH=/path/to/<db_name_prod>

Testing

All modules can be tested with pytest

pytest <module>

Note that testing requires a database to be setup and running.