Frequent Issues I Face When Working With GraphQL and How to Overcome Them

What is GraphQL?

GraphQL is a query language for APIs that’s becoming increasingly popular. It stands for “Graph Query Language” and was created and open-sourced by Facebook in 2012 to mitigate the flaws that REST presented. GraphQL is also “a server-side runtime for executing queries using a type system you define for your data.”

GraphQL works at both the front and back end and is a more flexible technology than REST for developers who want to exceed the common limitations of this architecture: documentation, filtering, argument validation, and joins.

Why use GraphQL?

The main difference between REST and GraphQL is that the first is software architecture while the second is a query language. REST functions with multiple endpoints, returning fixed data structures, while GraphQL allows you to access specific data. This means fewer journeys to the servers to fetch data, so both product iterations and developments occur faster.

However, when it comes to using GraphQL, there are two common issues that can arise: schema duplication and superfluous database calls. Luckily, the fixes for these issues are easy. Let’s take a closer look.

Schema Duplication

When you start coding GraphQL on the back end, you’ll inevitably need two very similar schemas: one for the GraphQL endpoint and another for the database. This similarity can commonly result in schema duplication.

However, you can solve this problem by using libraries like Prisma, TypeGraphQL + Typegoose and/or TypeGraphQL + TypeORM. These enable you to define both schemas in only one location instead of duplicating your schemas in different locations.

• Prisma: A Typescript and Nodejs ORM which provides a type-safe database client. (Extremely useful tool for prototyping)
• TypeGraphQL: This is a library that allows you to keep your GraphQL schema and your Typescript static types in sync.
• Typegoose: This is a wrapper to write mongoose schemas easily using Typescript.
• TypeORM: This is an ORM (object-relational-mapping) tool that helps you develop applications that use databases, no matter if they’re small apps or large scalable applications that need multiple databases.

Superfluous Database Calls

Superfluous database calls can cause a webpage to load slowly.

To avoid excessive trip requests to your database, you’ll need DataLoader. DataLoader batches your cache objects and queries. This way, they’re only fetched once from the database. DataLoader can detect previously used objects in its memory, avoiding making new database calls.

GraphQL Musts: Apollo and Node.js

When implementing GraphQL for your JavaScript applications, Apollo enters the picture. The server provides several open-source libraries that will simplify your work when you take your first steps with GraphQL.

Apollo is a comprehensive state management library designed for JavaScript, allowing you to administer data, either locally or remotely, with GraphQL. It can be used to manage application data and, at the same time, update your UI automatically.

Apollo provides you with open source components, client services, and commercial plugins. The Apollo server is your go-to resource if your code is already built around these features.

Typescript Merged with GraphQL

To avoid schema headaches like the ones described above, you can automatically create GraphQL schema definitions from TypeScript classes with decorators. Combining GraphQL with Typescript allows you to create a strongly typed end-end request-response lifecycle.

Tools such as TypeGraphQL allow you to define schema, create resolvers, and more. CodeGen is a tool that generates code out of your GraphQL schema.

These are just a couple of examples of the many tools available for merging GraphQL with Typescript; both are easy to find at GraphQL.org.

Conclusion

To sum up, GraphQL is a promising technology that you can choose for your APIs if you’re experiencing limitations with REST. Even though it can present some issues, none of the obstacles are serious or insurmountable, making it a good option to try out.