Spring in Action (1/5) - Foundational Spring

(This are my notes taken from the book Spring in Action 5th Edition)

Spring parts

  • Spring core Provides the core container, dependency injection framework, Spring MVC (Rest APIs), Spring’s web framework and support for template-based JDBC and reactive programming with Spring WebFlux.
  • Spring boot Starter dependencies and autoconfiguration.
  • Spring data Provides the ability to define your application’s data repository as Java interfaces. Works with relational (JPA), document (Mongo) and graph (Neo4j) databases.
  • Spring security Authentication, authorization and API security.
  • Spring integration & Batch helps integrate with other applications
  • Spring cloud helps with microservices

Project structure

src/main/java
  ApplicationStarter.java

src/test/java
  ApplicationStarterTest.java

src/main/resources
  static
  templates
  application.properties
  schema.sql  
  data.sql

mvnw.cmd
pom.xml
  • mvnw.cmd maven wrapper scripts. Used to build the project even without maven installed on your computer
  • static folder to place images, stylesheets, js resources to serve to the browser
  • templates folder to place template files that will be used to render content to the browser
  • schema.sql & data.sql they will be executed on start. Useful to create the structure of databases and fill them with data.

Main’s config

@SpringBootApplication is used at the main jar’s Starter as in

@SpringBootApplication
public class ApplicationStarter {
    // main
}

It’s a composition Tag which contains another three

  • @SpringBootConfiguration Designates the class as a configuration one. is a specialized form of @Configuration
  • @EnableAutoConfiguration Enables Spring to auto-configure any components that it thinks you may need
  • @ComponentScan It lets you use another annotations such as @Component to declare Beans

Web requests

The basic structure is an application which will be started with the embedded tomcat. A controller marked with @Controller tag and a @GetMapping("/") tag, which returns the name of the Thymeleaf view to give to the browser.

Controller class. It handles HTTP Requests and either gives it to a view to return HTML or writes data directly to the body (RESTful)

basic controller

@Controller
@RequestMapping("/design")
public class MyController {

    @GetMapping("/")
    public String home() {
        return "home"; // view name
    }

}

Thymeleaf will automatically search for /templates/home.html and return it as the view if it exists.

view controller

For a @Controller which is used just to redirect a path to a view and it doesn’t adds any kind of input to the Model, we may implement it with just a single a line at a WebConfig file.
The example implemented on top would be better as the following

@Configuration
public class WebConfig implements WebMvcConfig {

  @Override
  public void addViewControllers(final
            ViewControllerRegistry registry) {
    registry.addViewController("/")
            .setViewName("home");
  }

}

request mapping tags

  • @RequestMapping(method = RequestMethod.GET) general tag prior to Spring 4.3
  • @GetMapping handles HTTP GET request
  • @PostMapping handles HTTP POST request
  • @PutMapping handles …
  • @DeleteMapping
  • @PatchMapping

thymeleaf

(See thymeleaf notes)

Full Validation Example

Java’s validation API works with Spring and Hibernate, this last adds a bunch of annotations to use and may work together between Model, View and Controller.

declare validation rules

This is done at the model.

@Data
public class Taco {

  @NotNull
  @Size(min=5,
        message="must be at least 5 chars long")
  private String name;

  @Size(min=1,
        message="must choose at least 1 ing")
  private List<String> ingredients;

}

We also have other tags such as

@NotBlank(message="name is required")
private String name;

@CreditCardNumber(message="card not valid")
private String ccNumber;

@Pattern(regexp="here comes the regex",
         message="must be MM/YY")
private String ccExpiration;

@Digits(integer=3, fraction=0,
        message="invalid CVV")
private String ccCVV;

activate validation

This is done at the controller w. @Valid.

@PostMapping
public String process(@Valid
@ModelAttribute("design") final Taco design,
                          final Errors errors)
{
  if(errors.hasErrors()) {
    return "design";
  }
  // do something
  return "viewX";
}

If an error is found, the details of this error will be loaded into the Errors object

access the errors

This is done at the view.

<form method="POST" th:object="${design}">

<div th:if="${#fields.hasErrors()}">
    <p class="validationError">
      Please, correct following problems
    </p>
</div>

<div>
  <h3>Name your Taco creation:</h3>

  <span class="validationError"
        th:if="${#fields.hasErrors('name')}"
        th:errors="*{name}">Placeholder</span>
  <br/>
  <input type="text" th:field="*{name}"/>

  <button>Submit your taco</button>
</div>
</form>

all broken together

  1. The tags @NotNull declare what’s wrong and when an Error should be thrown.
  2. At the controller with the line @Valid @ModelAttribute("design") final Taco design we declare what the model is and we assign it an id = design.
  3. At the view form:
    • we access this model with th:object="${design}"
    • we check if the form has any error(s) with th:if="${#fields.hasErrors()}" to show a general error
    • we check a specific field with th:if="${#fields.hasErrors('name')}" and access the error message with th:errors="*{name}" to overwrite the placeholder.
      This name is a field of the model we gave. At this example Taco has a private String name field which is the one we’re checking here.

## Store Model attributes in a HTTP session (cache) This is used to store values which have to be used in between requests.

The @SessionAttributes is used in a @Controller class and whose value matches the one in @ModelAttribute.
When a request comes in it checks if there’s an existing value in the HTTP session, if it does, it retrieves it directly without calling the method again.

  @Controller
  @RequestMapping("/design")
  @SessionAttributes("ingredients")
  public class Controller {

    @ModelAttribute("ingredients")
    public List<Ingredient> addIng(final Model model) {
      final List<Ingredient> ings = // ...
      // first time this will be executed
      // second time it will retrieve the value from the cache
      return ings;
    }

  }

Databases

JDBC (Java DataBase Connection)

(This is the old way without Spring JPA. Do not implement this).

Its support is rooted at JdbcTemplate.java. It provides a mean for developers to perform SQL operations against a relation database without that much boilerplate code.

Working with JdbcTemplate

Dependencies to add:

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

For every Bean to support, we should have an interface and an implementation of it, which will delegate all the operations to JDBC.

public interface TacoRepository {

  public Taco save(final Taco taco);

}

@Repository
public class JdbcTacoRepository
  implements TacoRepository {

    @Autowired
    private final JdbcTemplate jdbc;  

    @Override
    public Taco save(final Taco taco) {

      // ...

      final PreparedStatementCreator psc = new
        PreparedStatementCreatorFactory(
          "insert into Taco (name, createdAt)
          values (?, ?)",
          Types.VARCHAR, Types.TIMESTAMP
        ).newPreparedStatementCreator(
        Arrays.asList(taco.getName(),
        new Timestamp(taco.getCreatedAt().getTime())));

      final KeyHolder kh =
        new GeneratedKeyHolder();
      this.jdbc.update(psc, kh);
      return keyHolder.getKey().longValue();
    }
  }

The code is hard to read, and it still requires quite an amount of boilerplate for just an operation.

Spring Data JPA

Spring Data JPA is not a JPA Provider. It allows us to automatically create repositories removing the most boilerplate possible, as opposed to using only a JPA Provider (Hibernate). It adds an extra layer of abstraction on top of it.

It’s comprised of several subsprojects

  • Spring data JPA - JPA persistence to a relational database
  • Spring Data MongoDB - against a Mongo document DBB
  • Spring Data Redis - against a Redis key-value store
    (etc)

Annotations

A JPA entity needs the tags @Entity and @Id.
Then, @GeneratedValue marks we rely on the database to generate the id.
@ManyToMany indicates relationship between Entities.
@PrePersist calls a method just before it’s persisted.
@Table specifies the database table name where our object’s information will be saved.

@Entity
@Table(name="TacoCustom")
public class Taco {

  @Id
  @GeneratedValue(strategy = GenerationType.AUTO)
  private Long id;

  private Date createdAt;

  // ...

  @ManyToMany(targetEntity = Ingredient.class)
  @Size(min = 1,
  message = "Must select at least 1 ingredient")
  private List<Ingredient> ingredients;

  @PrePersist
  private void createdAt() {
      this.createdAt =
        Date.valueOf(LocalDate.now());
  }

}

Repository

The main advantage of using Spring JPA vs JDBC, is we don’t need to implement repositories’ methods. We just create an interface, which extends CrudRepository<Object,ID> and Spring will write the implementation at runtime.

This is the whole implementation for a Taco repository and we may use directly its parent’s methods.

  public interface TacoRepository
    extends CrudRepository<Taco, Long> {

    List<Taco> findAllByCreatedAt(
      final Date createdAt);

  }
Repository methods

Spring is able to parse the name of the previous method, to generate the SQL statement, which will retrieve the info we need. Another example of this would be

List<Order> findByDeliveryZipAndPlacedAtBetween(deliveryZip,
 date1, date2);

There’s a list of valid operators at page 82.

The alternative to this for larger queries would be to use @Query tag which may use SQL

@Query("Order o WHERE o.deliveryCity='Atlanta'")
List<Order> findWhatever();