Usage

About

Unlike RestTemplate, which relies on thread-per-request execution, WebClient uses reactive streams (Mono and Flux) to handle requests and responses—enabling higher throughput with fewer threads. This makes it particularly suitable for microservices, API gateways, and cloud-native applications where scalability and performance matter.

Dependency

To use WebClient in a Spring Boot application, we need to include the appropriate Spring WebFlux dependency. This dependency provides the WebClient class along with reactive features like Mono, Flux, and support for non-blocking HTTP communication.

Spring WebClient is part of the spring-webflux module, not spring-web.

For Maven Projects

Add the following dependency in our pom.xml:

<dependency>
    <groupId>org.springframework.boot</groupId>
    <artifactId>spring-boot-starter-webflux</artifactId>
</dependency>

For Gradle Projects

In build.gradle:

implementation 'org.springframework.boot:spring-boot-starter-webflux'

This brings in:

  • WebClient class

  • Reactor Core (Mono, Flux)

  • Netty as the default HTTP client

  • Jackson support for JSON (if already using it in Spring Boot)

Compatibility

  • WebClient is available in Spring 5+

  • It works in both reactive (WebFlux) and non-reactive (Spring MVC) applications

  • Even if we are using Spring MVC (with spring-boot-starter-web), we can still add spring-boot-starter-webflux to use WebClient

  • However, we don’t need to move our entire app to WebFlux to use WebClient it can be used as a standalone client in any Spring Boot app

Declarative and Fluent API

One of WebClient's most powerful traits is its fluent API design, which allows for declarative-style HTTP request building. This leads to highly readable, concise, and maintainable code—especially for composing dynamic or chained HTTP interactions.

What Does “Fluent” Mean ?

The fluent style means that each method call returns an intermediate builder object, allowing multiple configuration steps to be chained in a single, continuous statement.

Instead of imperatively setting headers, bodies, and HTTP methods through separate steps, WebClient allows we to declare everything in a single flow.

Example (Declarative Style)

WebClient client = WebClient.create();

Mono<User> result = client.get()
    .uri("http://user-service/api/users/{id}", 42)
    .header("Authorization", "Bearer some-token")
    .accept(MediaType.APPLICATION_JSON)
    .retrieve()
    .bodyToMono(User.class);

Here’s what is happening in a flow:

  1. GET request is made

  2. URI is constructed with path variable

  3. Authorization header is added

  4. Expected content type is declared

  5. Response is retrieved and body mapped to a Java class (User)

WebClient.create()
    .post()
    .uri(uriBuilder -> uriBuilder
        .scheme("https")
        .host("payment.service")
        .path("/api/payments")
        .queryParam("region", "IN")
        .build()
    )
    .header("Authorization", "Bearer token")
    .contentType(MediaType.APPLICATION_JSON)
    .bodyValue(new PaymentRequest(...))
    .retrieve()
    .onStatus(HttpStatus::is4xxClientError, response -> {
        return Mono.error(new ClientErrorException("Client error"));
    })
    .bodyToMono(PaymentResponse.class);

This demonstrates the fluent API even for:

  • Dynamic URI building

  • Error handling

  • Sending request bodies

  • Mapping to POJO responses

Creating a WebClient Instance

WebClient is the non-blocking, reactive HTTP client introduced in Spring WebFlux. To use it effectively, the first step is creating an instance—which can be done in multiple ways, depending on how we want to configure and reuse the client.

1. Default WebClient Instance (Minimal Setup)

If no special configuration is needed, we can directly use WebClient.create().

import org.springframework.web.reactive.function.client.WebClient;

public class SimpleClient {

    private final WebClient webClient = WebClient.create();

    public void callApi() {
        webClient.get()
            .uri("https://api.example.com/data")
            .retrieve()
            .bodyToMono(String.class)
            .subscribe(System.out::println);
    }
}

  • No base URL is set

  • Used for quick, lightweight HTTP requests

2. With a Base URL

When we are interacting with a fixed host (e.g., internal microservice), it's common to initialize WebClient with a base URL.

import org.springframework.web.reactive.function.client.WebClient;

public class BaseUrlClient {

    private final WebClient webClient = WebClient.create("http://user-service");

    public void getUser(int id) {
        webClient.get()
            .uri("/api/users/{id}", id)
            .retrieve()
            .bodyToMono(String.class)
            .subscribe(System.out::println);
    }
}

  • Helps avoid repeating the base URI

  • Ideal for service-to-service calls

3. Using WebClient.Builder (Custom Configuration)

For headers, timeouts, filters, or other settings, use WebClient.Builder.

import org.springframework.web.reactive.function.client.WebClient;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;

@Configuration
public class WebClientConfig {

    @Bean
    public WebClient customWebClient() {
        return WebClient.builder()
            .baseUrl("http://inventory-service")
            .defaultHeader("Authorization", "Bearer my-token")
            .build();
    }
}

We can then inject this bean wherever needed:

import org.springframework.stereotype.Service;
import org.springframework.web.reactive.function.client.WebClient;

@Service
public class InventoryClient {

    private final WebClient webClient;

    public InventoryClient(WebClient webClient) {
        this.webClient = webClient;
    }

    public void getInventory() {
        webClient.get()
            .uri("/api/items")
            .retrieve()
            .bodyToMono(String.class)
            .subscribe(System.out::println);
    }
}

  • Fully configurable

  • Reusable across classes via Spring dependency injection

  • Recommended for real-world applications

4. Per-Request WebClient Builder (When We Need Dynamic Behavior)

Sometimes, we may need to build the client differently for each request (e.g., dynamic headers or SSL certificates).

import org.springframework.web.reactive.function.client.WebClient;

public class DynamicClient {

    public void callWithDynamicToken(String token) {
        WebClient webClient = WebClient.builder()
            .baseUrl("http://auth-service")
            .defaultHeader("Authorization", "Bearer " + token)
            .build();

        webClient.get()
            .uri("/api/token/check")
            .retrieve()
            .bodyToMono(String.class)
            .subscribe(System.out::println);
    }
}

  • Not reused across requests

  • Useful for token rotation or multi-tenant scenarios

5. Using ExchangeStrategies (e.g., custom message converters)

We may also need custom serialization or message readers/writers.

import org.springframework.http.codec.json.Jackson2JsonDecoder;
import org.springframework.http.codec.json.Jackson2JsonEncoder;
import org.springframework.web.reactive.function.client.ExchangeStrategies;
import org.springframework.web.reactive.function.client.WebClient;
import com.fasterxml.jackson.databind.ObjectMapper;

public class CustomStrategyClient {

    public WebClient createClientWithCustomJson(ObjectMapper mapper) {
        ExchangeStrategies strategies = ExchangeStrategies.builder()
            .codecs(config -> {
                config.defaultCodecs().jackson2JsonDecoder(new Jackson2JsonDecoder(mapper));
                config.defaultCodecs().jackson2JsonEncoder(new Jackson2JsonEncoder(mapper));
            })
            .build();

        return WebClient.builder()
            .baseUrl("http://analytics-service")
            .exchangeStrategies(strategies)
            .build();
    }
}

  • Ideal for controlling JSON serialization/deserialization

  • Works well in systems using customized Jackson configurations

Making a HTTP GET Request

The most common use of WebClient is to perform HTTP GET requests to fetch data from a remote service. WebClient provides a clean, fluent API that allows developers to define the request, handle the response, and optionally map it into Java objects all in a non-blocking, reactive style.

Let’s say we want to call a user service to retrieve details of a user by ID.

Class: UserClient

package com.example.client;

import org.springframework.stereotype.Service;
import org.springframework.web.reactive.function.client.WebClient;
import reactor.core.publisher.Mono;

@Service
public class UserClient {

    private final WebClient webClient;

    public UserClient(WebClient.Builder builder) {
        this.webClient = builder
            .baseUrl("http://user-service")
            .build();
    }

    public Mono<User> getUserById(Long userId) {
        return webClient
            .get()
            .uri("/api/users/{id}", userId)
            .retrieve()
            .bodyToMono(User.class);
    }
}

Class: User

package com.example.client;

public class User {
    private Long id;
    private String name;
    private String email;

    // Getters and Setters
}

  • .get() – Specifies that the HTTP method is GET

  • .uri("/api/users/{id}", userId) – Templated URI with path variable replacement

  • .retrieve() – Triggers the request and prepares for response handling

  • .bodyToMono(User.class) – Converts the response body into a Mono<User> (i.e., an asynchronous result)

This approach is non-blocking and reactive. The method will return immediately and complete once the response is received and processed.

Consuming the Response

We can subscribe to the response like this:

userClient.getUserById(101L)
    .subscribe(user -> System.out.println("User name: " + user.getName()));

Or in a controller method, we can return the Mono directly:

@GetMapping("/user/{id}")
public Mono<User> getUser(@PathVariable Long id) {
    return userClient.getUserById(id);
}

Handling Path and Query Parameters

When making API calls, it is common to pass values via path variables or query parameters. WebClient offers flexible and readable ways to handle both types, allowing us to construct URIs dynamically and cleanly.

1. Handling Path Parameters

Path parameters are part of the URI itself (e.g., /api/users/{id}). WebClient supports templated URIs where placeholders are replaced with actual values at runtime.

Example: Path Parameter

Mono<User> response = webClient
    .get()
    .uri("/api/users/{id}", 42)
    .retrieve()
    .bodyToMono(User.class);

We can also use a map for named parameters:

Map<String, Object> uriVars = Map.of("id", 42);

Mono<User> response = webClient
    .get()
    .uri("/api/users/{id}", uriVars)
    .retrieve()
    .bodyToMono(User.class);

2. Handling Query Parameters

Query parameters are appended to the URI as key-value pairs (e.g., ?status=active&page=2). WebClient provides a way to add these via UriBuilder.

Mono<UserList> response = webClient
    .get()
    .uri(uriBuilder -> uriBuilder
        .path("/api/users")
        .queryParam("status", "active")
        .queryParam("page", 2)
        .build()
    )
    .retrieve()
    .bodyToMono(UserList.class);

This approach is clean, avoids string concatenation, and supports conditional or optional parameters easily.

3. Combining Path and Query Parameters

We can combine both approaches to form complex URIs.

Mono<OrderList> response = webClient
    .get()
    .uri(uriBuilder -> uriBuilder
        .path("/api/users/{id}/orders")
        .queryParam("status", "delivered")
        .build(42)
    )
    .retrieve()
    .bodyToMono(OrderList.class);

In this example:

  • {id} is replaced with 42

  • Query parameter status=delivered is appended

Previous2. WebClientNextRequest Customization

Last updated