Bring Your Own Schema (BYOS)

Build Pipeline

Bring Your Own Schema (BYOS) is an implementation based on the concepts presented in my thesis, focusing on transpiling GraphQL queries to SQL. This repository provides a practical example of how to bring your own GraphQL schema and use it to generate efficient SQL queries.

Features

• One SQL query for every root level query.
• GraphQL schema and SQL schema are not generated and do not contain BYOS specific code.
• jOOQ is used to ensure correctness of the SQL configuration.
• Supports filter arguments, pagination and sorting.

limit

The number of results can be restricted by the limit argument.

query { filmByIds(limit: 3) { edges { node { film_id title } } } }

where

The result can be restricted to a condition by the where argument.

query { filmByIds(where: {film_id: {_in: [1,2]}}) { edges { node { film_id title } } } }

And with variables { "firstNames": ["ED", "JOE"] }:

query Actors_with_FirstName($firstNames: [String!]!) {
 Actor(where: {firstName: { _in: $firstNames} }) {
   actorId
   firstName
 }
}

order

The order of the result can be defined by the orderBy argument.

query { allFilms(orderBy: [{title: asc}]) { edges { node { film_id title } } } }

Project structure

• src/main/kotlin/byos contains the BYOS implementation.
• src/main/kotlin/example contains the exemplary Spring Boot application.
• src/main/resources/graphql contains the GraphQL schema.
• src/test/kotlin/example contains the tests for the example application.

If you want to try it out...

... you will have to have a local psql version of the sakila database with this small addition:

ALTER TABLE category
    ADD COLUMN parent_category_id INTEGER;

You could also set up your own database by:

• changing the connection of jOOQ (and the language if you do not want to use Postgres).
• using a GraphQL schema matching the SQL schema.
• defining relationships and how to resolve them.

To run the example run the main method in ByosApplication.kt. After that you can query the GraphQL endpoint at http://localhost:8080 or open the GraphiQL interface at http://localhost:8080/graphiql.

---

For more details about the approach and implementation, refer to the associated thesis.

Happy querying and transpiling!