Enterprise Spring Best Practices – Part 2 – Application Architecture

In part 2, Application Architecture.

This blog presents a look at the overall application components and architecture.


Sections




Application Domains



Our application components break down into two fundamental categories, the System and Problem Domains.

  • System Domain – infrastructure components, the plumbing, this is Spring’s sweet spot!
  • Problem Domain – business components, typically use-case driven, this is what most of developers are paid to solve



Application Layers



Application components (beans) should be separated into distinct layers, and categories.





Bean Layers
  1. Controllers (for MVC, System Domain)
  2. Services (Problem Domain)
  3. Repository (System Domain)
Other Bean Categories
  • Data Transfer Objects (Problem Domain)
  • System Functions (System Domain)



Controller Beans



More on Controllers in an upcoming blog on Enterprise Spring Best Practices MVC blog – TBD



Service Beans



Service Beans are Problem Domain components. These are the MOST significant in the application. Service beans are the Fundamental component of SOA.

  • These are POJOs
  • Always defined from interfaces
  • NEVER include infrastructure components
  • NO import of Spring or utility libraries
  • NO infrastructure annotations
  • Always declare transaction boundaries by public functions
  • Create implementation/concrete classes in a sub-package named internal

Spring annotations can be very useful for services, useful annotations are @Service and @Transactional. To abstract away the infrastructure from the business services, create a project specific meta annotation.

Custom Meta Annotation
package com.gordondickens.service.annotation;

import org.springframework.stereotype.Service;
import org.springframework.transaction.annotation.Transactional;

import java.lang.annotation.ElementType;
import java.lang.annotation.Retention;
import java.lang.annotation.RetentionPolicy;
import java.lang.annotation.Target;

@Retention(RetentionPolicy.RUNTIME)
@Target(ElementType.TYPE)
@Service
@Transactional
public @interface AppService {
    String value() default "";
}
Meta Annotation Use
...
@AppService
public class MyClass() {
  ...
}



Repository Beans



Repository beans are in the System Domain. More on this in Enterprise Spring Best Practices ORM blog – TBD

  • DO NOT contain business logic
  • Do use Spring and JPA annotations
  • Should be considered disposable



Data Transfer Beans



Data Transfer Objects (DTO) are the fundamental objects in and out of our system. DTOs are simple public POJOs that receive and send data as a logical set.

  • Always Public Beans
  • Annotate with JAXB2 Annotations



Conversion Beans



Spring provides a rich conversion registry at it’s core. The conversion service in Spring is based on the original Bean Specification PropertyEditor.

PropertyEditors are focussed on String data into and out of our application.

The Spring class org.springframework.beans.PropertyEditorRegistrySupport shows the built in String <–> object classes. Which we use, usually without knowledge, in our applications. When we configure our applications with XML and send in property values, Spring uses reflection to determine the argument type, if that type is not a String, Spring looks for a PropertyEditor that can convert from String to the target type. We also can create our own PropertyEditor’s and register them for types such as US Social Security Number or Telephone Number. See: Craig Wall’s Spring in Action, 3rd Ed for examples.

  • Primitive wrapper types: Long, Integer, etc
  • Collection types: List, Property, Set, Map, etc.
  • Arrays
  • Utility types: URL, TimeZone, Locale, etc.

Spring 3.0 introduced the Conversion service providing us the ability to register a conversion service for object <–> other conversion. To use the conversion registry, we can register a conversion class that will automatically convert to/from MyObject <–> MyOtherObject. See Using Spring Customer Type Converter Blog.



Further Reading




Social Me



Twitter – twitter.com/gdickens
LinkedIn – linkedin.com/in/gordondickens
GitHub: github.com/gordonad
gordon@gordondickens.com


About Gordon

Technology enthusiast primarily focused on Java and Open Source projects. Spring Certified Professional and Trainer. http://twitter.com/gdickens http://linkedin.com/in/gordondickens http://github.com/gordonad
This entry was posted in Spring, Spring Framework and tagged , , , , . Bookmark the permalink.

3 Responses to Enterprise Spring Best Practices – Part 2 – Application Architecture

  1. Pingback: Enterprise Spring Best Practices – Part 1 – Project Config | Technophile Blog

  2. Pingback: Enterprise Spring Framework Best Practices – Part 3 – XML Config | Technophile Blog

  3. Pingback: Enterprise Spring Best Practices – Part 4 – Annotation Config | Technophile Blog

Leave a Reply

Your email address will not be published. Required fields are marked *


*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>