GraphQL: all data on one query

This language was introduced by Facebook in 2012 as a more efficient alternative to the REST API. The company wanted to reduce the load on its servers. The solution was GraphQL. It was designed to overcome the limitations of the REST architecture.

GraphQL consists of 3 main components

• Queries – queries used by the client to request data from the server;
• resolvers – resolvers that GraphQL uses to understand how and from where to get the requested data;
• schema – a schema that describes what types of data GraphQL works with and what types of data GraphQL can return when responding to a query.

Both GraphQL and REST APIs use the HTTP protocol to communicate between client and server. The main difference between them is that GraphQL allows you to get all the necessary data with one request, even if they are on different endpoints (servers). In the REST API, this would require sending a separate request to each server.

Advantages over REST API:

• One link for all requests;
• One request is enough to get all the information you need;
• Only the requested data is presented in the response. No unnecessary information is transmitted. This increases the speed of processing network requests. This is an important difference to the REST API;
• The server response is defined by the request body.

Features of the GraphQL language:

• Strict typing – you can evaluate the correctness of a query before it is executed.
• Versatility – can be used with any framework, programming language. The source of the data does not matter.
• Easy scaling – allows you to add new types and fields without touching existing queries and without creating multiple versions of the same API.