This is a baseline example that should be useful to start projects involving a relational database schema. For this particular case, a PostgreSQL database is assumed. Other database engines can be supported by making the relevant changes in the enclosed Makefile.
This skeleton is language agnostic as it only concerns with deploying a static database schema. In particular, this skeleton has been used without changes to support projects written in Perl and Go.
To start your project, simply clone this and start adding your database / programming code.
See this article for a detailed discussion on how to channel your database changes through deltas.
The first versions of this skeleton were based in GNU Make. While this can still be used, I am gravitating towards a script based deployment for various reasons explained in the post.
The accompanying db/README.md file provides a more detailed discussion
about this repo.
Support for PostgREST introduces a dependency on pgjwt, btree_gist and pgcrypto. Please ensure these extensions are availablein your environment.
Testing requires the pgtap extension as well as the pg_prove command line utility.
Roles, users and a few tables are provided to assist with deployments that plan to use PostgREST for API provisioning. Base functions to control configuration, store JWT secrets and authenticate users are provided.
The provided implementation is sufficient for you to add your views, tables and functions that will then be automatically be available via the PostgREST generated API.
This schema sckeleton has been augmented so as to create two namespaces
(SCHEMA in PostgreSQL parlance). This is meant to assist with the use of the
PostgREST tool.
This distribution includes a simplistic Docker-based deployment helper that will launch PostgREST and Redoc, pointing at your local database. This should help get you started for your own deployments.