Developer Guide

Welcome to the developer’s guide to AiiDAlab applications for the ALC.

This document contains all the information required to develop AiiDA/AiiDAlab applications for use with ALC software projects. It includes details on how to create and run the required docker images to package additional components on top of the core AiiDAlab application, as well as how to deploy these to users either locally or via a cloud host such as ADA.

Introduction

AiiDAlab is a web based graphical user interface (GUI) for the popular scientific workflow management software package AiiDA. It presents a step-by-step approach to defining complex workflows utilising the relevant AiiDA plugins for the core computation. It is based on the appmode plugin for jupyter notebooks, enabling complex UI elements to be hosted within a web hosted jupyter notebook environment. This abstracts away many of the more challenging concepts of the AiiDA workflow itself whilst providing user friendly UI controls to interact with the provided plugins.

This guide aims to help developers get started with AiiDAlab plugin design and deployment across the wide variety of software applications developed within the ALC and STFC.

Any questions can be directed to the discussions page on GitHub or to benjamin.speake@stfc.ac.uk.

Contents

• Plugin Design
  • Core Requirements
  • Making Your App Discoverable
  • General Plugin Design Concepts
• Developing With Docker Containers
  • Developing Within a Container
  • Python Limitations
  • Docker Images For Distribution
• Creating Startup Scripts For AiiDAlab
  • Startup Script
  • Linux Desktop Entries
  • ADA Deployment (Ansible)