Java DataSource, JDBC DataSource Example

Java DataSource and JDBC DataSource programming is the way to work with database in our java programs. We have already seen that JDBC DriverManager can be used to get relational database connections. But when it comes to actual programming, we want more than just connections.

Java DataSource

Most of the times we are looking for loose coupling for connectivity so that we can switch databases easily, connection pooling for transaction management and distributed systems support. JDBC DataSource is the preferred approach if you are looking for any of these features in your application. Java DataSource interface is present in javax.sql package and it only declare two overloaded methods getConnection() and getConnection(String str1,String str2).

JDBC DataSource

It is the responsibility of different Database vendors to provide different kinds of implementation of DataSource interface. For example MySQL JDBC Driver provides basic implementation of DataSource interface with com.mysql.jdbc.jdbc2.optional.MysqlDataSource class and Oracle database driver implements it with oracle.jdbc.pool.OracleDataSource class.

These implementation classes provide methods through which we can provide database server details with user credentials. Some of the other common features provided by these JDBC DataSource implementation classes are;

  • Caching of PreparedStatement for faster processing
  • Connection timeout settings
  • Logging features
  • ResultSet maximum size threshold

JDBC DataSource Example

Let’s create a simple JDBC DataSource example project and learn how to use MySQL and Oracle DataSource basic implementation classes to get the database connection. Our final project will look like below image.

JDBC DataSource Example

Java JDBC DataSource – Database Setup

Before we get into our example programs, we need some database setup with table and sample data. Installation of MySQL or Oracle database is out of scope of this tutorial, so I will just go ahead and setup table with sample data.

Now let’s move on to our java programs. For having database configuration loosely coupled, I will read them from property file. file:

Make sure that above configurations match with your local setup. Also make sure you have MySQL and Oracle DB JDBC jars included in the build path of the project.

Java JDBC DataSource – MySQL, Oracle Example

Let’s write a factory class that we can use to get MySQL or Oracle DataSource.

Notice that both Oracle and MySQL DataSource implementation classes are very similar, let’s write a simple test program to use these methods and run some test.

Notice that the client class is totally independent of any Database specific classes. This helps us in hiding the underlying implementation details from client program and achieve loose coupling and abstraction benefits.

When we run above test program, we will get below output.

Apache Commons DBCP Example

If you look at above Java DataSource factory class, there are two major issues with it.

  1. The factory class methods to create the MySQL and Oracle DataSource are tightly coupled with respective driver API. If we want to remove support for Oracle database in future or want to add some other database support, it will require code change.
  2. Most of the code to get the MySQL and Oracle DataSource is similar, the only different is the implementation class that we are using.

Apache Commons DBCP API helps us in getting rid of these issues by providing Java DataSource implementation that works as an abstraction layer between our program and different JDBC drivers.

Apache DBCP library depends on Commons Pool library, so make sure they both are in the build path as shown in the image.

Here is the DataSource factory class using BasicDataSource that is the simple implementation of DataSource.

As you can see that depending on user input, either MySQL or Oracle DataSource is created. If you are supporting only one database in the application then you don’t even need these logic. Just change the properties and you can switch from one database server to another. The key point through which Apache DBCP provide abstraction is setDriverClassName() method.

Here is the client program using above factory method to get different types of connection.

When you run above program, the output will be same as earlier program.

If you look at the Java JDBC DataSource and above usage, it can be done with normal DriverManager too. The major benefit of Java DataSource is when it’s used within a Context and with JNDI.

With simple configurations we can create a Database Connection Pool that is maintained by the Container itself. Most of the servlet containers such as Tomcat and JBoss provide it’s own Java DataSource implementation and all we need is to configure it through simple XML based configurations and then use JNDI context lookup to get the Java DataSource and work with it. This helps us by taking care of connection pooling and management from our application side to server side and thus giving us more time to write business logic for the application.

In next tutorial, we will learn how we can configure DataSource in Tomcat Container and use it in Web Application.

By admin

Leave a Reply