This project is intended to become a Clojure library that scaffolds GraphQL schema and resolvers, with queries and mutations working out of the box, based on an existing Datomic db schema.

The purpose of this (future) library is to auto-generate the repetitive boilerplate needed for GraphQL endpoint based on famous Lacinia and Pedestal libraries, and the idea had come from the library called Stillsuit that already had implemented foundation code for this task under Apache license.

Stillsuit has to be used in conjunction with another library made by the same people from Workframe called Catchpocket and requires manual preparation of the config file. Schema produced with Stillsuit had only some basic queries and no mutations at all. It also has limitations to work only with Datomic peer API (no Client API support).

I wanted to make a fully automated and customizable library that works on both client and peer Datomic API. To achieve this I've decided to incorporate Stillsuit and Catchpocket code in my project as to be able to modify their code as needed.

At the same time, I was actively searching the internet for a similar kind of library (not necessarily for Datomic) to see what is possible and desirable. Recently I've stumbled upon the library called Mongo-graphql-starter that has implemented a lot of automation I was thinking of.

My goal would be to implement as much as possible of this library has already implemented. Also, this library has helped me to make my decision about the name of the project.

This project is intended to be used/tested with Datomic Pro Starter with Peer or Client API-s. I am using and datomic-pro-0.9.5703 versions.

Datomic Pro Starter from this link. After downloading, unzip it at any place that is convenient for you.

Datomic Pro Starter is the free version with all features of Datomic Pro but is limited to 1 year of maintenance and updates. Datomic Pro Starter supports both Datomic Peer API and Client API.

Before running Datomic Pro Starter transactor you will need to:

• copy ./config/samples/dev-transactor-template.properties to the root of folder where your Datomic installation resides.
• modify ./dev-transactor-template.properties file by inserting your license key at the beginning of this file (license-key= tag). You can get your License key from my.datomic.com by mail when you press "Send License Key" button.
• Install Datomic Pro locally using bin/maven-install from the installation folder
• After that you can run Datomic transactor with:

bin/transactor dev-transactor-template.properties

That's enough to be able to communicate with db using Datomic Peer API.

If you want to communicate with DB using Datomic Client API it's needed to run Datomic Peer Server with command like this:

bin/run -m datomic.peer-server -h localhost -p 8998 -a "access-key","secret" -d mbrainz,datomic:dev://localhost:4334/mbrainz

Of course, you should replace strings "access-key" and "secret" with something you use in your system.

For testing purposes, I am using Datomic MusicBrainz sample database that is a subset of the Mbrainz database covering the period 1968-1973. You can download the backup of this database from this location.

Then you will unzip it in folder and restore it to your local Datomic Pro instance (first run Datomic Pro transactor in terminal) with this command:

bin/datomic restore-db file:///full-path/to/backup/mbrainz datomic:dev://localhost:4334/mbrainz

Windows users should specify a basis-t to restore, like this:

bin/datomic restore-db file:///path/to/backup/mbrainz datomic:dev://localhost:4334/mbrainz 148348

In examples we use, we've renamed database "mbrainz-1968-1973" for simplicity to "mbrainz".

Finally, you should clone/download/fork this project.

Basic project configuration file is config.fern in root project folder:

{
;full paths
 api-conf               (fern/lit concat @root @api-conf-file)
 stillsuit-conf         (fern/lit concat @root @stillsuit-conf-file)
 catchpocket-conf       (fern/lit concat @root @catchpocket-conf-file)
 refs-conf              (fern/lit concat @root @refs-conf-file)

;root
 root                   (fern/lit concat @config-root @db-name "/")

; datomic db
 db-name                "mbrainz"
 secret "enes"
 db-dev                     (fern/lit concat "datomic:dev://localhost:4334/" @db-name)
 db-free                    (fern/lit concat "datomic:free://localhost:4334/" @db-name)

;files
 api-conf-file          "api-config.edn"
 refs-conf-file         "refs-config.edn"
 catchpocket-conf-file  "catchpocket-config.edn"
 stillsuit-conf-file    "stillsuit-config.edn"

;paths
  config-root            "resources/config/"
 }

config.fern file is written in fern format and defines names and locations for other configuration files, current database name and secret, Datomic Free and Dev database uris. This is the only config file that should be created manually .

Normally only fields here that needs to be changed are db-name and secret, but you can experiment with others too.

db-name defines the name of database we are connecting to with Graphql server

secret is the secret key that we use in script that runs Datomic Peer Server

Other configuration files will be created automatically in resources/config/<db-name> folder when we run project from REPL.

This file contains map with five keys:

:entities

:queries

:inserts

:deletions

:updates

For Mbrainz DB without manual editing it looks like this:

{:entities
 ["country"
  "artist"
  "language"
  "release"
  "medium"
  "label"
  "abstractRelease"
  "track"
  "script"],
 :queries [],
 :inserts []
 :deletions []
 :updates []}

Depends on refs-config.edn. If delete this file it will be recreated together and then stillsuit-congfig.edn will be recreated.

Depends on catchpocket-config.edn