Looking for Java best practices to improve your coding skills? Or maybe you want to discover some tips and tricks to optimise your work? Read on, as we have something for you!
What comes to your mind when you hear the word Java? Do you at least associate it with a picture of a cup of freshly made coffee?
Or maybe you just felt offended because you are an experienced Java developer?
Either way, keep reading if you want to find out about Java best practices programmers should know, no matter what their actual skills are.
Ready? Let’s go!
I will start by dividing the whole article into smaller parts:
- General Technological Concepts
- Code Quality
- Performance
- Frameworks and tools
- Architecture
- Other useful tips and tricks
Java Best Practices: General Concepts
Let’s assume that you want to develop efficient and flexible applications using Java. Otherwise, you wouldn’t be reading this article, right?
As you may already know, there are some proven design patterns and coding standards that will help you to be a better programmer.
Using the best coding practices and following coding guidelines improves code quality – it makes code much more understandable, flexible, reusable and maintainable.
It also helps to increase code performance and employee productivity.
Can I stop you for a second?
If you’re here for the first time, you need to know that we have more articles like this.
Looking fornon BS articles written by software developers? You can find them in the Tech category.
Code Quality
The most important thing regarding code quality is to keep your applications as secure and stable as possible from the beginning of the development process. To do this, always use the latest Java version in the source code.
What to do and what not to do?
- Give priority to generic types, methods, and wildcards
- Implement tight encapsulation, loose coupling and high cohesion in classes
- Avoid memory leaks by using final blocks whenever possible
- Always close IO streams
- Using Date and Time API (JSR 310) is better than using data.util.date and Joda-Time
- Expression lambdas are better than block lambdas
- Catching specific subclasses of the Exception class is better than catching the class itself
- Returning empty collections is better than null
- Remember about using interface type when you are declaring a collection
- Every catch block needs the Exception class
- If you need fast iteration and random access – use ArrayList
- If you need fast insertion and deletion – use LinkedList
Additionally, using Java code style formatters and source code analysers like FindBugs, CheckStyle or Sonar will improve your code’s readability.
Performance
Making an application functional is one thing but developing it quickly is another. To make both possible, every Java developer should be aware of common performance issues and avoid them whenever possible.
- database
issues
- N + 1 select – trap of incomprehensible lazy fetching strategy
- no cache – it is faster to get content from memory than from a database (possibly on a different server)
- insufficient size of pool connections
- memory problems
- garbage collection procedure occurs too often or takes too much time
- memory leaks and OutOfMemory error
- concurrency problems
- thread deadlocks
- thread gridlocks
Logging levels in our code will allow us to avoid printing unnecessary logs in non-development mode. It also protects the production environment from slowing down. Logs and return codes can give you detailed information about what exactly happened. Thanks to this, code consumers won’t waste their time.
If you want to avoid the recreation of objects and reduce the amount of garbage collection, cache frequently-used objects.
Rememberto use connection pools and cached-prepared statements to gain database access.
Frameworks and tools
If you are about to choose the application framework, I strongly recommend choosing Spring because of its popularity, flexibility, and capabilities. It’s an open-source framework that makes J2EE development easier to manage and helps to “wire” different components together.
Above all, Spring is helpful if you have a lot of components that you want to combine in different ways or if you want to facilitate swapping one component for another. This framework also provides many modules which support functionalities like:
- Dependency Injection
- Security
- Database access
- Webservices
- System integration
- Batch processing
What are the key benefits of Spring?
- Well-designed MVC framework
- Modular architecture
- Capability of implementing enterprise applications using POJOs
- Set of conventional technologies, i.e. ORM framework, logging framework, J2EE
- Consistent transaction management system
- Flexibility in building different kinds of applications, i.e. GUI applications
- Support for managing business objects and exposing their services to presentation-tier components
- Capability of eliminating the creation of singleton and factory classes
Use JPA standard implementation as the data persistence framework. The most popular one is Hibernate, which provides several useful modules like:
- ehcache – for defining second level cache
- envers – for tracking change history
- search – for indexing results in local storage
The key benefits of Hibernate are:
- Advantages of OOP concepts, i.e. inheritance, encapsulation
- Avoidance of repetitive code from JDBC API
- Caching mechanism
- Database engine independence
- Support of lazy loading
Architecture
Following SOLID principles is crucial to designing a good architecture. The acronym stands for:
- S – Single Responsibility Principle
- O – Open/Closed Principle
- L – Liskov’s Substitution Principle
- I – Interface Segregation Principle
- D – Dependency Inversion Principle
I’m pretty sure that you know them all, but do you follow them consistently?
Ask yourself and answer sincerely.
The next thing you should remember is to divide the application into different logical areas from the business perspective. What about implementation? It needs to be divided into logical tiers from a functional standpoint, i.e. presentation, business logic and data layer. If architecture is layered correctly, it means that it is built on separate layers that are not affected by specific changes in other layers.
If you want to offer some sophisticated and extensive functionalities for the users of your application, you should avoid developing one monolithic application and consider using microservices architecture. As a result, you will be able to build many small programs, and then you can create a small application that implements a new feature in every instance.
In addition, having a microkernel is another good practice. It helps us to implement extra features in plug-in architecture on top of the kernel module.
Deployment server
The construction of dynamic pages and implementation of services are supported by application servers, like clustering, fail-over and load balancing. Thanks to this, developers can just focus on fulfilling business requirements. What is the most popular application server for Java?
- GlassFish
- JBoss
- Tomcat
- Geronimo
- WebSphere
Other useful tips and tricks
Authorization and authentication
It is quite easy to configure access authentication as the applications are often created by using the Spring Framework.
Spring Security provides the capability of setting objects like Role, Permission, and User to Role assignment.
It also provides a comprehensive security solution that handles authentication and authorization, available at the web request level and the method invocation level.
Unit and integration testing
One way of verifying if the logging and behavior of the source code are correct is to do unit tests.
That is to say, tests make it possible to identify software regressions or bugs created as a result of the latest changes in the source code.
If you want to continue implementing features without having to perform a vast amount of manual tests, high test coverage of code will come in handy.
Technologies that support unit and integration tests:
- JUnit
- TestNG
- AspectJ
- AssertJ
- Mockito
- PowerMock
If you want to visualize code coverage, use the JaCoCo library.
Build and deployment
If you want to build applications faster, use build tools. Every Java developer should know about tools like Ant, Maven, and Gradle.
Maven is the most popular one, and it allows for downloading remote dependencies and automation of manual tasks often needed for deployment, like copying files, cleaning cache and injecting properties.
Learn more about Java software development
