Build a GraphQL app in Node.js with TypeScript and graphql-request - LogRocket Blog

Hussain is a CS student in Pakistan whose biggest interest is learning and teaching programming to make the world a better place.

In this article, you will build a full-stack app using GraphQL and Node.js in the backend. Meanwhile, our frontend will use the graphql-request library to perform network operations on our backend.

We will cover the following steps:

Nội Dung Chính

Why use graphql-request and TypeScript?

Whenever developers build a GraphQL server using Apollo, the library generates a “frontend” which looks like so:

Frontend Developed By GraphQL And Apollo

This interface allows users to make query or mutation requests to the server via code. However, let’s address the elephant in the room: it doesn’t look very user friendly. Since the frontend doesn’t feature any buttons or any helpful interface elements, it might be hard for many users to navigate around your app. Consequently, this shrinks your user base. So how do we solve this problem?

This is where graphql-request comes in. It is an open source library which lets users perform queries on a GraphQL server. It boasts the following features:

Lightweight — This library is just over 21 kilobytes minified, which ensures your app stays performant
Promise-based API — This brings in support for asynchronous applications
TypeScript support — graphql-request is one of many libraries which allows for TypeScript. One major advantage of Typescript is that it allows for stable and predictable code

For example, look at the following program:

let myNumber = 9; //here, myNumber is an integer
myNumber = 'hello'; //now it is a string.
myNumber = myNumber + 10; //even though we are adding a string to an integer,
//JavaScript won't return an error. In the real world, it might bring unexpected outputs.
//However, in Typescript, we can tell the compiler..
//what data types we need to choose.
let myNumber:number = 39; //tell TS that we want to declare an integer.
myNumber = 9+'hello'; //returns an error. Therefore, it's easier to debug the program
//this promises stability and security.

In this article, we will build a full-stack app using GraphQL and TypeScript. Here, we will use the apollo-server-express package to build a backend server. Furthermore, for the frontend, we will use Next and graphql-request to consume our GraphQL API.

Building our server

Project initialization

To initialize a blank Node.js project, run these terminal commands:

mkdir graphql-ts-tutorial #create project folder 
cd graphql-ts-tutorial 
npm init -y #initialize the app

When that’s done, we now have to tell Node that we need to use TypeScript in our codebase:

#configure our Typescript:
npx tsc --init --rootDir app --outDir dist --esModuleInterop --resolveJsonModule --lib es6 --module commonjs --allowJs true --noImplicitAny true
mkdir app #our main code folder
mkdir dist #Typescript will use this folder to compile our program.

Next, install these dependencies:

#development dependencies. Will tell Node that we will use Typescript
npm install -d ts-node @types/node typescript @types/express nodemon
#Installing Apollo Server and its associated modules. Will help us build our GraphQL
#server
npm install apollo-server-express apollo-server-core express graphql

After this step, navigate to your app folder. Here, create the following files:

index.ts: Our main file. This will execute and run our Express GraphQL server
dataset.ts: This will serve as our database, which will be served to the client
Resolvers.ts: This module will handle user commands. We will learn about resolvers later in this article
Schema.ts: As the name suggests, this file will store the schematics needed to send data to the client

In the end, your folder structure should look like so:

Folder Structure

Creating our database

In this section, we will create a dummy database which will be used to send requested data. To do so, go to app/dataset.ts and write the following code:

let people: { id: number; name: string }[] = [
  { id: 1, name: "Cassie" },
  { id: 2, name: "Rue" },
  { id: 3, name: "Lexi" },
];
export default people;

First, we created an array of objects called people
This array will have two fields: id of type number, and name of type string

Defining our schema

Here, we will now create a schema for our GraphQL server.

To put it simply, a GraphQL schema is a description of the dataset that clients can request from an API. This concept is similar to that of the Mongoose library.
To build a schema, navigate to the app/Schema.ts file. There, write the following code:

import { gql } from "apollo-server-express"; //will create a schema
const Schema = gql`
  type Person {
    id: ID!
    name: String
  }
  #handle user commands
  type Query {
    getAllPeople: [Person] #will return multiple Person instances
    getPerson(id: Int): Person #has an argument of 'id` of type Integer.
  }
`;
export default Schema; 
//export this Schema so we can use it in our project

Let’s break down this code piece by piece:

The Schema variable contains our GraphQL schema
First, we created a Person schema. It will have two fields: id of type ID and name of type String
Later on, we instructed GraphQL that if the client runs the getAllPeople command, the server will return an array of Person objects
Furthermore, if the user uses the getPerson command, GraphQL will return a single Person instance

Creating resolvers

Now that we have coded our schema, our next step is to define our resolvers.
In simple terms, a resolver is a group of functions that generate response for a GraphQL query. In other words, a resolver serves as a GraphQL query handler.
In Resolvers.ts, write the following code:

import people from "./dataset"; //get all of the available data from our database.
const Resolvers = {
  Query: {
    getAllPeople: () => people, //if the user runs the getAllPeople command
    //if the user runs the getPerson command:
    getPerson: (_: any, args: any) => { 
      console.log(args);
      //get the object that contains the specified ID.
      return people.find((person) => person.id === args.id);
    },
  },
};
export default Resolvers;

Here, we created a Query object that handles all the incoming queries going to the server
If the user executes the getAllPeople command, the program will return all the objects present in our database
Moreover, the getPerson command requires an argument id. This will return a Person instance with the matching ID
In the end, we exported our resolver so that it could be linked with our app

Configuring our server

We’re almost done! Now that we have built both our schema and resolver, our next step is to link them together.

In index.js, write this block of code:

import { ApolloServer } from "apollo-server-express";
import Schema from "./Schema";
import Resolvers from "./Resolvers";
import express from "express";
import { ApolloServerPluginDrainHttpServer } from "apollo-server-core";
import http from "http";

async function startApolloServer(schema: any, resolvers: any) {
  const app = express();
  const httpServer = http.createServer(app);
  const server = new ApolloServer({
    typeDefs: schema,
    resolvers,
    //tell Express to attach GraphQL functionality to the server
    plugins: [ApolloServerPluginDrainHttpServer({ httpServer })],
  }) as any;
  await server.start(); //start the GraphQL server.
  server.applyMiddleware({ app });
  await new Promise<void>((resolve) =>
    httpServer.listen({ port: 4000 }, resolve) //run the server on port 4000
  );
  console.log(`Server ready at http://localhost:4000${server.graphqlPath}`);
}
//in the end, run the server and pass in our Schema and Resolver.
startApolloServer(Schema, Resolvers);

Let’s test it out! To run the code, use this Bash command:

npx nodemon app/index.ts

This will create a server at the localhost:4000/graphql URL.

Here, you can see your available schemas within the UI:

Available Schemas Within The UI

This means that our code works!

All of our GraphQL queries will go within the Operation panel. To see it in action, type this snippet within this box:

#make a query:
query {
  #get all of the people available in the server
  getAllPeople {
    #procure their IDs and names.
    id
    name
  }
}

To see the result, click on the Run button:

We can even search for a specific entity via the getPerson query:

query ($getPersonId: Int) { #the argument will be of type Integer
  getPerson(id: 1) {
    #get the person with the ID of 1
    name
    id
  }
}

Getperson Query

Creating mutations

In the GraphQL world, mutations are commands that perform side effects on the database. Common examples of this include:

Adding a user to the database — When a client signs up for a website, the user performs a mutation to save their data in their database
Editing or deleting an object — If a user modifies or removes data from a database, they are essentially creating a mutation on the server

To handle mutations, go to your Schema.ts module. Here, within the Schema variable, add the following lines of code:

const Schema = gql`
  #other code..
  type Mutation {
    #the addPerson commmand will accept an argument of type String.
    #it will return a 'Person' instance. 
    addPerson(name: String): Person
  }
`;

Our next step is to create a resolver to handle this mutation. To do so, within the Resolvers.ts file, add this block of code:

const Resolvers = {
  Query: {
    //..further code..
  },
  //code to add:
  //all our mutations go here.
  Mutation: {
    //create our mutation:
    addPerson: (_: any, args: any) => {
      const newPerson = {
        id: people.length + 1, //id field
        name: args.name, //name field
      };
      people.push(newPerson);
      return newPerson; //return the new object's result
    },
  },
};

The addPerson mutation accepts a name argument
When a name is passed, the program will create a new object with a matching name key
Next, it will use the push method to add this object to the people dataset
Finally, it will return the new object’s properties to the client

That’s it! To test it out, run this code within the Operations window:

#perform a mutation on the server
mutation($name: String) {
  addPerson(name:"Hussain") { #add a new person with the name "Hussain"
    #if the execution succeeds, return its 'id' and 'name` to the user.
    id
    name
  }
}

Addperson

Let’s verify if GraphQL has added the new entry to the database:

query {
  getAllPeople { #get all the results within the 'people' database. 
  #return only their names
  name 
  }
}

Verify That GraphQL Added A New Entry

Building our client

We have successfully built our server. In this section, we will build a client app using Next that will listen to the server and render data to the UI.

As a first step, initialize a blank Next.js app like so:

npx [email protected] graphql-client --ts
touch constants.tsx #our query variables go here.

To perform GraphQL operations, we will use the graphql-request library. This is a minimal, open source module that will help us make mutations and queries on our server:

npm install graphql-request graphql
npm install react-hook-form #to capture user input

Creating query variables

In this section, we will code our queries and mutations to help us make GraphQL operations. To do so, go to constants.tsx and add the following code:

import { gql } from "graphql-request";
//create our query
const getAllPeopleQuery = gql`
  query {
    getAllPeople { #run the getAllPeople command
      id
      name
    }
  }
`;
//Next, declare a mutation
const addPersonMutation = gql`
  mutation addPeople($name: String!) {
    addPerson(name: $name) { #add a new entry. Argument will be 'name'
      id
      name
    }
  }
`;
export { getAllPeopleQuery, addPersonMutation };

In the first part, we created the getAllPeopleQuery variable. When the user runs this query, the program will instruct the server to get all the entries present in the database
Later on, the addPerson mutation tells GraphQL to add a new entry with its respected name field
In the end, we used the export keyword to link our variables with the rest of the project

Performing queries

In pages/index.ts, write the following code:

import type { NextPage, GetStaticProps, InferGetStaticPropsType } from "next";
import { request } from "graphql-request"; //allows us to perform a request on our server
import { getAllPeopleQuery } from "../constants"; 
import Link from "next/link";
const Home: NextPage = ({
  result, //extract the 'result' prop 
}: InferGetStaticPropsType<typeof getStaticProps>) => {
  return (
    <div className={styles.container}>
      {result.map((item: any) => { //render the 'result' array to the UI 
        return <p key={item.id}>{item.name}</p>;
      })}
    <Link href="/addpage">Add a new entry </Link>
    </div>
  );
};
//fetch data from the server
export const getStaticProps: GetStaticProps = async () => {
  //the first argument is the URL of our GraphQL server
  const res = await request("http://localhost:4000/graphql", getAllPeopleQuery);
  const result = res.getAllPeople;
  return {
    props: {
      result,
    }, // will be passed to the page component as props
  };
};
export default Home;

Here is a breakdown of this code piece by piece:

In the getStaticProps method, we instructed Next to run the getAllPeople command on our GraphQL server
Later on, we returned its response to the Home functional component. This means that we can now render the result to the UI
Next, the program used the map method to render all of the results of the getAllPeople command to the UI. Each paragraph element will display the name fields of each entry
Furthermore, we also used a Link component to redirect the user to the addpage route. This will allow the user to add a new Person instance to the table

To test out the code, run the following terminal command:

npm run dev

This will be the result:

Addpage Route

Our GraphQL server even updates in real time.

GraphQL Updating In Real Time

Performing mutations

Now that we have successfully performed a query, we can even perform mutations via the graphql-request library.

Within your pages folder, create a new file called addpage.tsx. As the name suggests, this component will allow the user to add a new entry to the database. Here, start by writing the following block of code:

import type { NextPage, GetStaticProps, InferGetStaticPropsType } from "next";
import { request } from "graphql-request";
import { addPersonMutation } from "../constants";
const AddPage: NextPage = () => {
  return (
    <div>
      <p>We will add a new entry here. </p>
    </div>
  );
};
export default AddPage;

In this piece of code, we are creating a blank page with a piece of text. We are doing this to ensure whether our URL routing system works.

This means that we used routing successfully! Next, write this snippet in your addpage.tsx file:

import { useForm } from "react-hook-form";
const { register, handleSubmit } = useForm();
//if the user submits the form, then the program will output the value of their input.
const onSubmit = (data: any) => console.log(data);
return (
  <div>
    <form onSubmit={handleSubmit(onSubmit)}> {/*Bind our handler to this form.*/}
      {/* The user's input will be saved within the 'name' property */}
      <input defaultValue="test" {...register("name")} />
      <input type="submit" />
    </form>
  </div>
);

This will be the output:

Output Output

Now that we have successfully captured the user’s input, our last step is to add their entry to the server.

To do so, change the onSubmit handler located in pages/addpage.tsx file like so:

const onSubmit = async (data: any) => {
  const response = await request(
    "http://localhost:4000/graphql",
    addPersonMutation,
    data
  );
  console.log(response);
};

Here, we’re performing a mutation request to our GraphQL server via the request function
Furthermore, we also passed in the addPerson mutation command to our request header. This will tell GraphQL to perform the addMutation action on our server

This will be the result:

Result Of Addmutation Action

And we’re done!

Conclusion

Here is the full source code of this project.

In this article, you learned how to create a full-stack app using GraphQL and TypeScript. They both are extremely crucial skills within the programming world since they are in high demand nowadays.

If you encountered any difficulty in this code, I advise you to deconstruct the code and play with it so that you can fully grasp this concept.

Thank you so much for reading! Happy coding!

200’s only Monitor failed and slow network requests in production

Deploying a Node-based web app or website is the easy part. Making sure your Node instance continues to serve resources to your app is where things get tougher. If you’re interested in ensuring requests to the backend or third party services are successful, LogRocket Network Request Monitoring https://logrocket.com/signup/

LogRocket is like a DVR for web and mobile apps, recording literally everything that happens while a user interacts with your app. Instead of guessing why problems happen, you can aggregate and report on problematic network requests to quickly understand the root cause.

LogRocket instruments your app to record baseline performance timings such as page load time, time to first byte, slow network requests, and also logs Redux, NgRx, and Vuex actions/state.

LogRocket instruments your app to record baseline performance timings such as page load time, time to first byte, slow network requests, and also logs Redux, NgRx, and Vuex actions/state. Start monitoring for free

Build a GraphQL app in Node.js with TypeScript and graphql-request – LogRocket Blog