Getting Started with Backend Development in Kotlin Using Spring Boot 3 & MongoDB
Rate this tutorial
This is an introduction article on how to build a RESTful application in Kotlin using Spring Boot 3 and MongoDB Atlas.
Today, we are going to build a basic RESTful application that does a little more than a CRUD operation, and for that, we will use:
Spring Boot 3, which is one of the popular frameworks based on Spring, allowing developers to build production grades quickly.MongoDB, which is a document oriented database, allowing developers to focus on building apps rather than on database schema.
This is a getting-started article, so nothing much is needed as a prerequisite. But familiarity with Kotlin as a programming language, plus a basic understanding of Rest API and HTTP methods, would be helpful.
Building a HelloWorld app in any programming language/technology, I believe, is the quickest and easiest way to get familiar with it. This helps you cover the basic concepts, like how to build, run, debug, deploy, etc.
Since we are using the community version of IDEA, we cannot create the
HelloWorld project directly from IDE itself using the New Project. But we can use the Spring initializer app instead, which allows us to create a Spring
project out of the box.Once you are on the website, you can update the default selected parameters for the project, like the name of the project, language, version of
Spring Boot, etc., to something similar as shown below.
And since we want to create REST API with MongoDB as a database, let's add the dependency using the Add Dependency button on the right.
After all the updates, our project settings will look like this.
Now we can download the project folder using the generate button and open it using the IDE. If we scan the project folder, we will only find one class — i.e.,
HelloBackendWorldApplication.kt, which has the main function, as well.
The next step is to print HelloWorld on the screen. Since we are building a restful
application, we will create a
GET request API. So, let's add a function to act as a GET API call.1 2 fun hello( name: String?): String { 3 return String.format("Hello %s!", name) 4 }
We also need to add an annotation of
@RestController to our class to make it a Restful client.1 2 3 class HelloBackendWorldApplication { 4 5 fun hello(): String { 6 return "Hello World!" 7 } 8 } 9 10 fun main(args: Array<String>) { 11 runApplication<HelloBackendWorldApplication>(*args) 12 }
Now, let's run our project using the run icon from the toolbar.
Now load https://localhost:8080/hello on the browser once the build is complete, and that will print Hello World on your screen.
And on cross-validating this from Postman, we can clearly understand that our
Get API is working perfectly.It's time to understand the basics of
Spring Boot that made it so easy to create our first API call.As per official docs, Spring Boot makes it easy to create stand-alone, production-grade, Spring-based applications that you can "just run."
This implies that it's a tool built on top of the Spring framework, allowing us to build web applications quickly.
Spring Boot uses annotations, which do the heavy lifting in the background. A few of them, we have used already, like:@SpringBootApplication: This annotation is marked at class level, and declares to the code reader (developer) and Spring that it's a Spring Boot project. It allows an enabling feature, which can also be done using@EnableAutoConfiguration,@ComponentScan, and@Configuration.@RequestMappingand@RestController: This annotation provides the routing information. Routing is nothing but a mapping of aHTTPrequest path (text afterhost/) to classes that have the implementation of these across variousHTTPmethods.
These annotations are sufficient for building a basic application. Using Spring Boot, we will create a RESTful web service with all business logic, but we don't have a data container that can store or provide data to run these operations.
For our app, we will be using MongoDB as the database. MongoDB is an open-source, cross-platform, and distributed document database, which allows building apps with flexible schema. This is great as we can focus on building the app rather than defining the schema.
We can get started with MongoDB really quickly using MongoDB Atlas, which is a database as a service in the cloud and has a free forever tier.
I recommend that you explore the MongoDB Jumpstart series to get familiar with MongoDB and its various services in under 10 minutes.
With the basics of MongoDB covered, now let's connect our Spring Boot project to it. Connecting with MongoDB is really simple, thanks to the Spring Data MongoDB plugin.
To connect with MongoDB Atlas, we just need a database URL that can be added
as a
spring.data.mongodb.uri property in application.properties file. The connection string can be found as shown below.
The format for the connection string is:
1 spring.data.mongodb.uri = mongodb + srv ://<username>:<pwd>@<cluster>.mongodb.net/<dbname>
With all the basics covered, now let's build a more complex application than HelloWorld! In this app, we will be covering all CRUD operations and tweaking them along the way to make it a more realistic app. So, let's create a new project similar to the HelloWorld app we created earlier. And for this app, we will use one of the sample datasets provided by MongoDB — one of my favourite features that enables quick learning.
We will be using the
sample_restaurants collection for our CRUD application. Before we start with the actual CRUD operation, let's create the restaurant model class equivalent to it in the collection.
1 2 data class Restaurant( 3 4 val id: ObjectId = ObjectId(), 5 val address: Address = Address(), 6 val borough: String = "", 7 val cuisine: String = "", 8 val grades: List<Grade> = emptyList(), 9 val name: String = "", 10 11 val restaurantId: String = "" 12 ) 13 14 data class Address( 15 val building: String = "", 16 val street: String = "", 17 val zipcode: String = "", 18 19 val coordinate: List<Double> = emptyList() 20 ) 21 22 data class Grade( 23 val date: Date = Date(), 24 25 val rating: String = "", 26 val score: Int = 0 27 )
You will notice there is nothing fancy about this class except for the annotation. These annotations help us to connect or co-relate classes with databases like:
@Document: This declares that this data class represents a document in Atlas.@Field: This is used to define an alias name for a property in the document, likecoordfor coordinate inAddressmodel.
Now let's create a repository class where we can define all methods through which we can access data.
Spring Boot has interface MongoRepository, which helps us with this.1 interface Repo : MongoRepository<Restaurant, String> { 2 3 fun findByRestaurantId(id: String): Restaurant? 4 }
After that, we create a controller through which we can call these queries. Since this is a bigger project, unlike the HelloWorld app, we will create a separate controller where the
MongoRepository instance is passed using @Autowired, which provides annotations-driven dependency injection.1 2 3 class Controller( val repo: Repo) { 4 5 }
Now our project is ready to do some action, so let's count the number of restaurants in the collection using
GetMapping.1 2 3 class Controller( val repo: Repo) { 4 5 6 fun getCount(): Int { 7 return repo.findAll().count() 8 } 9 }
Taking a step further to read the restaurant-based
restaurantId. We will have to add a method in our repo as restaurantId is not marked @Id in the restaurant class.1 interface Repo : MongoRepository<Restaurant, String> { 2 fun findByRestaurantId(restaurantId: String): Restaurant? 3 }
1 2 fun getRestaurantById( id: String): Restaurant? { 3 return repo.findByRestaurantId(id) 4 }
And again, we will be using Postman to validate the output against a random
restaurantId from the sample dataset.
Let's also validate this against a non-existing
restaurantId.
As expected, we haven't gotten any results, but the API response code is still 200, which is incorrect! So, let's fix this.
In order to have the correct response code, we will have to check the result before sending it back with the correct response code.
1 2 fun getRestaurantById( id: String): ResponseEntity<Restaurant> { 3 val restaurant = repo.findByRestaurantId(id) 4 return if (restaurant != null) ResponseEntity.ok(restaurant) else ResponseEntity 5 .notFound().build() 6 }
To add a new object to the collection, we can add a
write function in the repo we created earlier, or we can use the inbuilt method insert provided by MongoRepository. Since we will be adding a new object to the collection, we'll be using @PostMapping for this.1 2 fun postRestaurant(): Restaurant { 3 val restaurant = Restaurant().copy(name = "sample", restaurantId = "33332") 4 return repo.insert(restaurant) 5 }
Spring doesn't have any specific in-built update similar to other CRUD operations, so we will be using the read and write operation in combination to perform the update function.
1 2 fun updateRestaurant( id: String): Restaurant? { 3 return repo.findByRestaurantId(restaurantId = id)?.let { 4 repo.save(it.copy(name = "Update")) 5 } 6 }
This is not an ideal way of updating items in the collection as it requires two operations and can be improved further if we use the MongoDB native driver, which allows us to perform complicated operations with the minimum number of steps.
Deleting a restaurant is also similar. We can use the
MongoRepository delete function of the item from the collection, which takes the item as input.1 2 fun deleteRestaurant( id: String) { 3 repo.findByRestaurantId(id)?.let { 4 repo.delete(it) 5 } 6 }
Thank you for reading and hopefully you find this article informative! The complete source code of the app can be found on GitHub.
If you have any queries or comments, you can share them on the MongoDB forum or tweet me @codeWithMohit.