MS SQL Server: Relationships between tables/views

Introduction

To make nested object queries, the tables/views in your database need to be connected via relationships.

Let’s say we have the following tables in our database: authors, passport_infos, articles and tags.

Table relationships

The tables/views in any relational database are typically related to each other via one of the following types of table relationships:

Type Example Meaning
one-to-one authors and passport_infos
  • an author can have one passport_info
  • a passport_info can have one owner
one-to-many authors and articles
  • an author can have many articles
  • an article can have one author
many-to-many articles and tags
  • an article can have many tags
  • a tag can have many articles

GraphQL schema relationships

Each table relationship, as you can see from the above section, will have two component relationships (one in either direction) in the GraphQL schema. These relationships can be one of the following types:

Type Example Meaning
object relationship (one-to-one) an article can have one author an article object will have a single nested author object called author
array relationship (one-to-many) an author can have many articles an author object will have an array of nested article objects called articles

Note

The relationship name is used to refer to the nested objects in queries. For example, “articles” of an author and “author” of an article.

Managing GraphQL relationships

See the following to manage the object/array relationships between tables/views for the GraphQL schema:

Table relationships modelling guides

The following guides will help you model the different types of table relationships in the database: