Skip Headers
Oracle® Database JDBC Developer's Guide and Reference,
11g Release 1 (11.1)

Part Number B31224-01
Go to Documentation Home
Go to Book List
Book List
Go to Table of Contents
Go to Index
Go to Master Index
Master Index
Go to Feedback page
Contact Us

Go to previous page
Go to next page
View PDF

2 Getting Started

This chapter discusses the compatibility of Oracle Java Database Connectivity (JDBC) driver versions, database versions, and Java Development Kit (JDK) versions. It also describes the basics of testing a client installation and configuration and running a simple application. This chapter contains the following sections:

Version Compatibility for Oracle JDBC Drivers

This section discusses the general JDBC version compatibility issues.

Backward Compatibility

The JDBC drivers are certified to work with the currently supported versions of Oracle Database. For example, the JDBC Thin drivers in Oracle Database 11g Release 1 (11.1) are certified to work with the 10.2.x, 10.1.x, 9.2.x, and 9.0.1.x Oracle Database releases. However, they are not certified to work with older, unsupported database releases, such as 8.0.x and 7.x.

Forward Compatibility

Existing and supported JDBC drivers are certified to work with Oracle Database 11g Release 1 (11.1).


Verification of a JDBC Client Installation

To verify a JDBC client installation, you must do all of the following:

Installation of an Oracle JDBC driver is platform-specific. Follow the installation instructions for the driver you want to install in your platform-specific documentation.

This section describes the steps for verifying an Oracle client installation of the JDBC drivers, assuming that you have already installed the driver of your choice.

If you have installed the JDBC Thin driver, then no further installation on the client computer is necessary.


The JDBC Thin driver requires a TCP/IP listener to be running on the computer where the database is installed.

If you have installed the JDBC Oracle Call Interface (OCI) driver, then you must also install the Oracle client software. This includes Oracle Net and the OCI libraries.

Check the Installed Directories and Files

Installing the Oracle Java products creates, among other things, the following directories:

  • ORACLE_HOME/jdbc

  • ORACLE_HOME /jlib

Check whether or not the following directories and files have been created and populated in the ORACLE_HOME/jdbc directory:

  • demo

    This directory contains a compressed file, or demo.tar. When you uncompress this compressed file, the samples directory and the Samples-Readme.txt file are created. The samples directory contains sample programs, including examples of how to use SQL92 and Oracle SQL syntax, PL/SQL blocks, streams, user-defined types, additional Oracle type extensions, and Oracle performance extensions.

  • doc

    This directory contains the file, which is the Oracle JDBC application programming interface (API) documentation.

  • lib

    The lib directory contains the following required Java classes:

    • orai18n.jar and orai18n-mapping.jar

      Contain classes for globalization and multibyte character sets support

    • ojdbc5.jar, ojdbc5_g.jar, ojdbc6.jar, and ojdbc6_g.jar

      Contain the JDBC driver classes for use with JDK 1.5 and JDK 1.6


      • In Oracle Database 11g Release 1 (11.1), support for a version of JDK earlier than version 1.5 has been removed. Also, the ojdbc14.jar and classes12.jar files are no longer shipped. Instead, you can use the ojdbc5.jar and ojdbc6.jar files, which are shipped with Oracle Database 11g.

      • If you are using JSE 6 and later, then there is no need to explicitly load the JDBC driver. This means that the Java run-time loads the driver when needed and you need not include Class.forName("oracle.jdbc.OracleDriver") or new oracle.jdbc.OracleDriver() in your code. But if you are using J2SE 5, then you need to load the JDBC driver explicitly.

  • Readme.txt

    This file contains late-breaking and release-specific information about the drivers, which may not have been included in other documentation on the product.

Check whether or not the following directories have been created and populated in the ORACLE_HOME /jlib directory:

  • jta.jar and jndi.jar

    These files contain classes for the Java Transaction API (JTA) and the Java Naming and Directory Interface (JNDI). These are required only if you are using JTA features for distributed transaction management or JNDI features for naming services.


    These files can also be obtained from the Sun Microsystems Web site. However, it is recommended that you use the versions supplied by Oracle, which have been tested with the Oracle drivers.

Check the Environment Variables

This section describes the environment variables that must be set for the JDBC OCI driver and the JDBC Thin driver, focusing on the Sun Solaris, Linux, and Microsoft Windows platforms.

You must set the CLASSPATH environment variable for your installed JDBC OCI or Thin driver. Include the following in the CLASSPATH environment variable:



If you use the JTA features and the JNDI features, then you must specify jta.jar and jndi.jar in your CLASSPATH environment variable.


If you are installing the JDBC OCI driver, then you must also set the following value for the library path environment variable:

  • On Sun Solaris or Linux, set the LD_LIBRARY_PATH environment variable as follows:


    This directory contains the shared object library.


    If you are running a 32-bit Java Virtual Machine (JVM) against a 64-bit client or database, then you must also add ORACLE_HOME/lib32 to the LD_LIBRARY_PATH environment variable.
  • On Microsoft Windows, set the PATH environment variable as follows:


    This directory contains the ocijdbc11.dll dynamic link library.

All of the JDBC OCI demonstration programs can be run in the Instant Client mode by including the JDBC OCI Instant Client data shared library on the library path environment variable.

JDBC Thin Driver

If you are installing the JDBC Thin driver, then you do not have to set any other environment variables. However, to use the JDBC server-side Thin driver, you need to set permission.

Setting Permission for the Server-Side Thin Driver

The JDBC server-side Thin driver opens a socket for its connection to the database. Because Oracle Database enforces the Java security model, a check is performed for a SocketPermission object.

To use the JDBC server-side Thin driver, the connecting user must be granted the appropriate permission. The following is an example of how the permission can be granted for the user SCOTT:

CREATE ROLE jdbcthin;
CALL dbms_java.grant_permission('JDBCTHIN', '', '*', 'connect');
GRANT jdbcthin TO SCOTT;

Note that JDBCTHIN in the grant_permission call must be in uppercase. The asterisk (*) is a pattern. You can restrict the user by granting permission to connect to only specific computers or ports.

Ensure that the Java Code Can Be Compiled and Run

To further ensure that Java is set up properly on your client system, go to the samples directory under ORACLE_HOME/jdbc/demo. Now, type the following commands on the command line, one after the other, to see if the Java compiler and the Java interpreter run without error. :



Each of the preceding commands should display a list of options and parameters and then exit. Ideally, verify that you can compile and run a simple test program, such as jdbc/demo/samples/generic/SelectExample.

Determine the Version of the JDBC Driver

You can determine the version of the JDBC driver that you installed, by calling the getDriverVersion method of the OracleDatabaseMetaData class.

The following sample code shows how to determine the driver version:

import java.sql.*;
import oracle.jdbc.*;
import oracle.jdbc.pool.OracleDataSource;

class JDBCVersion
  public static void main (String args[]) throws SQLException
    OracleDataSource ods = new OracleDataSource();
    Connection conn = ods.getConnection();

    // Create Oracle DatabaseMetaData object
    DatabaseMetaData meta = conn.getMetaData();

    // gets driver info:
    System.out.println("JDBC driver version is " + meta.getDriverVersion());

You can also determine the version of the JDBC driver by executing the following commands:

  • java -jar ojdbc5.jar

  • java -jar ojdbc6.jar

Test JDBC and the Database Connection

The samples directory contains sample programs for a particular Oracle JDBC driver. One of the programs,, is designed to test JDBC and the database connection. The program queries for the user name, password, and the name of the database to which you want to connect. The program connects to the database, queries for the string "Hello World", and prints it to the screen.

Go to the samples directory, and compile and run the program. If the results of the query print without error, then your Java and JDBC installations are correct.

Although is a simple program, it demonstrates several important functions by performing the following:

  • Imports the necessary Java classes, including JDBC classes

  • Creates a DataSource instance

  • Connects to the database

  • Runs a simple query

  • Prints the query results to your screen

The program, which uses the JDBC OCI driver, is as follows:

 * This sample can be used to check the JDBC installation.
 * Just run it and provide the connect information. It will select
 * "Hello World" from the database.

// You need to import the java.sql and JDBC packages to use JDBC
import java.sql.*;
import oracle.jdbc.*;
import oracle.jdbc.pool.OracleDataSource;

// We import to be able to read from the command line

class JdbcCheckup
  public static void main(String args[]) throws SQLException, IOException

    // Prompt the user for connect information
    System.out.println("Please enter information to test connection to 
                          the database");
    String user;
    String password;
    String database;

    user = readEntry("user: ");
    int slash_index = user.indexOf('/');
    if (slash_index != -1)
      password = user.substring(slash_index + 1);
      user = user.substring(0, slash_index);
      password = readEntry("password: ");
    database = readEntry("database(a TNSNAME entry): ");

    System.out.print("Connecting to the database...");
    // Open an OracleDataSource and get a connection
    OracleDataSource ods = new OracleDataSource();
    ods.setURL("jdbc:oracle:oci:@" + database);
    Connection conn = ods.getConnection();

    // Create a statement
    Statement stmt = conn.createStatement();

    // Do the SQL "Hello World" thing
    ResultSet rset = stmt.executeQuery("select 'Hello World' from dual");

    while (
    // close the result set, the statement and connect
    System.out.println("Your JDBC installation is correct.");

  // Utility function to read a line from standard input
  static String readEntry(String prompt)
      StringBuffer buffer = new StringBuffer();
      int c =;
      while (c != '\n' && c != -1)
        c =;
      return buffer.toString().trim();
    catch(IOException e)
      return "";

Basic Steps in JDBC

After verifying the JDBC client installation, you can start creating your JDBC applications. When using Oracle JDBC drivers, you must include certain driver-specific information in your programs. This section describes, in the form of a tutorial, where and how to add the information. The tutorial guides you through the steps to create code that connects to and queries a database from the client.

You must write code to perform the following tasks:

  1. Importing Packages

  2. Opening a Connection to a Database

  3. Creating a Statement Object

  4. Running a Query and Retrieving a ResultSet Object

  5. Processing the ResultSet Object

  6. Closing the ResultSet and Statement Objects

  7. Making Changes to the Database

  8. Committing Changes

  9. Closing the Connection


You must supply Oracle driver-specific information for the first three tasks, which allow your program to use the JDBC application programming interface (API) to access a database. For the other tasks, you can use standard JDBC Java code, as you would for any Java application.

Importing Packages

Regardless of which Oracle JDBC driver you use, include the import statements shown in Table 2-1 at the beginning of your program.

Table 2-1 Import Statements for JDBC Driver

Import statement Provides

import java.sql.*;

Standard JDBC packages.

import java.math.*;

The BigDecimal and BigInteger classes. You can omit this package if you are not going to use these classes in your application.

import oracle.jdbc.*;

import oracle.jdbc.pool.*;

import oracle.sql.*;

Oracle extensions to JDBC. This is optional.


Oracle type extensions. This is optional.

The Oracle packages listed as optional provide access to the extended functionality provided by Oracle JDBC drivers, but are not required for the example presented in this section.


It is better to import only the classes your application needs, rather than using the wildcard asterisk (*). This guide uses the asterisk (*) for simplicity, but this is not the recommended way of importing classes and interfaces.

Opening a Connection to a Database

First, you must create an OracleDataSource instance. Then, open a connection to the database using the OracleDataSource.getConnection method. The properties of the retrieved connection are derived from the OracleDataSource instance. If you set the URL connection property, then all other properties, including TNSEntryName, DatabaseName, ServiceName, ServerName, PortNumber, Network Protocol, and driver type are ignored.

Specifying a Database URL, User Name, and Password

The following code sets the URL, user name, and password for a data source:

OracleDataSource ods = new OracleDataSource();

The following example connects user scott with password tiger to a database with service orcl through port 1521 of the host myhost, using the JDBC Thin driver:

OracleDataSource ods = new OracleDataSource();
String url = "jdbc:oracle:thin:@//myhost:1521/orcl",
Connection conn = ods.getConnection();


The user name and password specified in the arguments override any user name and password specified in the URL.

Specifying a Database URL that Includes User Name and Password

The following example connects user scott with password tiger to a database host whose Transparent Network Substrate (TNS) entry is myTNSEntry, using the JDBC Oracle Call Interface (OCI) driver. In this case, the URL includes the user name and password and is the only input parameter.

String url = "jdbc:oracle:oci:scott/tiger@myTNSEntry");
Connection conn = ods.getConnection();

If you want to connect using the Thin driver, then you must specify the port number. For example, if you want to connect to the database on the host myhost that has a TCP/IP listener on port 1521 and the service identifier is orcl, then provide the following code:

String URL = "jdbc:oracle:thin:scott/tiger@//myhost:1521/orcl");
Connection conn = ods.getConnection();

Creating a Statement Object

Once you connect to the database and, in the process, create a Connection object, the next step is to create a Statement object. The createStatement method of the JDBC Connection object returns an object of the JDBC Statement type. To continue the example from the previous section, where the Connection object conn was created, here is an example of how to create the Statement object:

Statement stmt = conn.createStatement();

Running a Query and Retrieving a ResultSet Object

To query the database, use the executeQuery method of the Statement object. This method takes a SQL statement as input and returns a JDBC ResultSet object.


  • The method used to execute a Statement object depends on the type of SQL statement being executed. If the Statement object represents a SQL query returning a ResultSet object, the executeQuery method should be used. If the SQL is known to be a DDL statement or a DML statement returning an update count, the executeUpdate method should be used. If the type of the SQL statement is not known, the execute method should be used.

  • In case of a standard JDBC driver, if the SQL string being executed does not return a ResultSet object, then the executeQuery method throws a SQLException. In case of an Oracle JDBC driver, the executeQuery method does not throw a SQLException even if the SQL string being executed does not return a ResultSet object.

To continue the example, once you create the Statement object stmt, the next step is to run a query that returns a ResultSet object with the contents of the ename column of a table of employees named EMP:

ResultSet rset = stmt.executeQuery ("SELECT ename FROM emp");

Processing the ResultSet Object

Once you run your query, use the next() method of the ResultSet object to iterate through the results. This method steps through the result set row by row, detecting the end of the result set when it is reached.

To pull data out of the result set as you iterate through it, use the appropriate getXXX methods of the ResultSet object, where XXX corresponds to a Java data type.

For example, the following code will iterate through the ResultSet object, rset, from the previous section and will retrieve and print each employee name:

while (
   System.out.println (rset.getString(1));

The next() method returns false when it reaches the end of the result set. The employee names are materialized as Java String values.

Closing the ResultSet and Statement Objects

You must explicitly close the ResultSet and Statement objects after you finish using them. This applies to all ResultSet and Statement objects you create when using Oracle JDBC drivers. The drivers do not have finalizer methods. The cleanup routines are performed by the close method of the ResultSet and Statement classes. If you do not explicitly close the ResultSet and Statement objects, serious memory leaks could occur. You could also run out of cursors in the database. Closing both the result set and the statement releases the corresponding cursor in the database. If you close only the result set, then the cursor is not released.

For example, if your ResultSet object is rset and your Statement object is stmt, then close the result set and statement with the following lines of code:


When you close a Statement object that a given Connection object creates, the connection itself remains open.


Typically, you should put close statements in a finally clause.

Making Changes to the Database

To write changes to the database, such as for INSERT or UPDATE operations, you create a PreparedStatement object. This enables you to run a statement with varying sets of input parameters. The prepareStatement method of the JDBC Connection object lets you define a statement that takes variable bind parameters and returns a JDBC PreparedStatement object with your statement definition.

Use the setXXX methods on the PreparedStatement object to bind data to the prepared statement to be sent to the database.

The following example shows how to use a prepared statement to run INSERT operations that add two rows to the EMP table.

// Prepare to insert new names in the EMP table
PreparedStatement pstmt = null;
    pstmt = conn.prepareStatement ("insert into EMP (EMPNO, ENAME) values (?, ?)");

    // Add LESLIE as employee number 1500
    pstmt.setInt (1, 1500);          // The first ? is for EMPNO
    pstmt.setString (2, "LESLIE");   // The second ? is for ENAME
    // Do the insertion
    pstmt.execute ();

    // Add MARSHA as employee number 507
    pstmt.setInt (1, 507);           // The first ? is for EMPNO
    pstmt.setString (2, "MARSHA");   // The second ? is for ENAME
    // Do the insertion
    pstmt.execute ();


    // Close the statement

Committing Changes

By default, data manipulation language (DML) operations are committed automatically as soon as they are run. This is known as the auto-commit mode. However, you can disable auto-commit mode with the following method call on the Connection object:


If you disable the auto-commit mode, then you must manually commit or roll back changes with the appropriate method call on the Connection object:




A COMMIT or ROLLBACK operation affects all DML statements run since the last COMMIT or ROLLBACK.


  • If the auto-commit mode is disabled and you close the connection without explicitly committing or rolling back your last changes, then an implicit COMMIT operation is run.

  • Any data definition language (DDL) operation always causes an implicit COMMIT. If the auto-commit mode is disabled, then this implicit COMMIT will commit any pending DML operations that had not yet been explicitly committed or rolled back.

Closing the Connection

You must close the connection to the database after you have performed all the required operations and no longer require the connection. You can close the connection by using the close method of the Connection object, as follows:



Typically, you should put close statements in a finally clause.

Sample: Connecting, Querying, and Processing the Results

The steps in the preceding sections are illustrated in the following example, which uses the Oracle JDBC Thin driver to create a data source, connects to the database, creates a Statement object, runs a query, and processes the result set.

Note that the code for creating the Statement object, running the query, returning and processing the ResultSet object, and closing the statement and connection uses the standard JDBC API.

import java.sql.Connection;
import java.sql.Resultset;
import java.sql.Statement;
import oracle.jdbc.pool.OracleDataSource;

class JdbcTest
   public static void main (String args []) throws SQLException

OracleDataSource ods = null;
Connection conn = null;
Statement stmt = null;
ResultSet rset = null;

      // Create DataSource and connect to the local database
      ods = new OracleDataSource();
      conn = ods.getConnection();

      // Query the employee names 
      stmt = conn.createStatement (); 
      rset = stmt.executeQuery ("SELECT ename FROM emp");

      // Print the name out 
      while ( ())
         System.out.println (rset.getString (1));

      //Close the result set, statement, and the connection

      if(rset!=null) rset.close();
      if(stmt!=null) stmt.close();
      if(conn!=null) conn.close();

If you want to adapt the code for the OCI driver, then replace the call to the OracleDataSource.setURL method with the following:


where, MyHostString is an entry in the TNSNAMES.ORA file.

Stored Procedure Calls in JDBC Programs

This section describes how Oracle JDBC drivers support the following kinds of stored procedures:

PL/SQL Stored Procedures

Oracle JDBC drivers support the processing of PL/SQL stored procedures and anonymous blocks. They support PL/SQL block syntax and most of SQL92 escape syntax. The following PL/SQL calls would work with any Oracle JDBC driver:

// SQL92 syntax
CallableStatement cs1 = conn.prepareCall
                       ( "{call proc (?,?)}" ) ; // stored proc
CallableStatement cs2 = conn.prepareCall
                       ( "{? = call func (?,?)}" ) ; // stored func
// PL/SQL block syntax
CallableStatement cs3 = conn.prepareCall
                       ( "begin proc (?,?); end;" ) ; // stored proc
CallableStatement cs4 = conn.prepareCall
                       ( "begin ? := func(?,?); end;" ) ; // stored func

As an example of using the Oracle syntax, here is a PL/SQL code snippet that creates a stored function. The PL/SQL function gets a character sequence and concatenates a suffix to it:

create or replace function foo (val1 char)
return char as
   return val1 || 'suffix';

The function invocation in your JDBC program should look like the following:

OracleDataSource ods = new OracleDataSource();
Connection conn = ods.getConnection();

CallableStatement cs = conn.prepareCall ("begin ? := foo(?); end;");
cs.setString(2, "aa");
String result = cs.getString(1);

Java Stored Procedures

You can use JDBC to call Java stored procedures through the SQL and PL/SQL engines. The syntax for calling Java stored procedures is the same as the syntax for calling PL/SQL stored procedures, presuming they have been properly published. That is, you have written call specifications to publish them to the Oracle data dictionary. Applications can call Java stored procedures using the Native Java Interface for direct invocation of static Java methods.

Processing SQL Exceptions

To handle error conditions, Oracle JDBC drivers throw SQL exceptions, producing instances of the java.sql.SQLException class or its subclass. Errors can originate either in the JDBC driver or in the database itself. Resulting messages describe the error and identify the method that threw the error. Additional run-time information can also be appended.

JDBC 3.0 defines only a single exception, SQLException. However, there are large categories of errors and it is useful to distinguish them. Therefore, in JDBC 4.0, a set of subclasses of SQLException is introduced to identify the different categories of errors. To know more about this feature, see Support for JDBC 4.0 Standard.

Basic exception handling can include retrieving the error message, retrieving the error code, retrieving the SQL state, and printing the stack trace. The SQLException class includes functionality to retrieve all of this information, when available.

Retrieving Error Information

You can retrieve basic error information with the following methods of the SQLException class:

The following example prints output from a getMessage method call:

catch(SQLException e)
   System.out.println("exception: " + e.getMessage());

This would print the output, such as the following, for an error originating in the JDBC driver:

exception: Invalid column type


Error message text is available in alternative languages and character sets supported by Oracle.

Printing the Stack Trace

The SQLException class provides the printStackTrace() method for printing a stack trace. This method prints the stack trace of the throwable object to the standard error stream. You can also specify a object or object for output.

The following code fragment illustrates how you can catch SQL exceptions and print the stack trace.

try { <some code> } 
catch(SQLException e) { e.printStackTrace (); } 

To illustrate how the JDBC drivers handle errors, assume the following code uses an incorrect column index:

// Iterate through the result and print the employee names 
// of the code 
try { 
  while ( ()) 
      System.out.println (rset.getString (5));  // incorrect column index
catch(SQLException e) { e.printStackTrace (); } 

Assuming the column index is incorrect, running the program would produce the following error text:

java.sql.SQLException: Invalid column index
at oracle.jdbc.driver.DatabaseError.throwSqlException(
at oracle.jdbc.driver.DatabaseError.throwSqlException(
at oracle.jdbc.driver.DatabaseError.throwSqlException(
at oracle.jdbc.driver.OracleResultSetImpl.getDate(
at Employee.main(