Microsoft Extensibility SDK for Java for SQL Server
Applies to: 
SQL Server 2019 (15.x) and later versions
Learn how you can implement a Java program for SQL Server using the Microsoft Extensibility SDK for Java. The SDK is an interface for the Java language extension that is used to exchange data with SQL Server and to execute Java code from SQL Server.
The SDK is installed as part of SQL Server 2019 (15.x) and later versions, on both Windows and Linux:
- Default installation path on Windows:
<instance installation home directory>\MSSQL\Binn\mssql-java-lang-extension.jar - Default installation path on Linux:
/opt/mssql/lib/mssql-java-lang-extension.jar
The code is open source and can be found on the SQL Server Language Extensions GitHub repository.
Implementation requirements
The SDK interface defines a set of requirements that need to be fulfilled for SQL Server to communicate with the Java runtime. To use the SDK, you need to follow some implementation rules in your main class. SQL Server can then execute a specific method in the Java class and exchange data using the Java language extension.
For an example of how you can use the SDK, see Tutorial: Search for a string using regular expressions (regex) in Java.
SDK Classes
The SDK consists of three classes.
Two abstract classes that define the interface the Java extension uses to exchange data with SQL Server:
AbstractSqlServerExtensionExecutorAbstractSqlServerExtensionDataset
The third class is a helper class, which contains an implementation of a data set object. It's an optional class you can use, which makes it easier to get started. You can also use your own implementation of such a class instead.
PrimitiveDataset
In the following section, you find descriptions of each class in the SDK. The source code of the SDK classes is available in the SQL Server Language Extensions GitHub repository.
Class: AbstractSqlServerExtensionExecutor
The abstract class AbstractSqlServerExtensionExecutor contains the interface used to execute Java code by the Java language extension for SQL Server.
Your main Java class needs to inherit from this class. Inheriting from this class means that there are certain methods in the class you need to implement in your own class.
To inherit from this abstract class, you extend with the abstract class name in the class declaration:
public class <MyClass> extends AbstractSqlServerExtensionExecutor {}
At a minimum, your main class needs to implement the execute(...) method.
Method execute
The execute method is the method that is called from SQL Server via the Java language extension, to invoke Java code from SQL Server. It's a key method where you include the main operations you wish to execute from SQL Server.
To pass method arguments to Java from SQL Server, use the @param parameter in sp_execute_external_script. The method execute takes its arguments that way.
public AbstractSqlServerExtensionDataset execute(AbstractSqlServerExtensionDataset input, LinkedHashMap<String, Object> params) {}
Method init
The init method is executed after the constructor, and before the execute method. Any operations that need to be performed before execute(...) can be done in this method.
public void init(String sessionId, int taskId, int numtask) {}
Class: AbstractSqlServerExtensionDataset
The abstract class AbstractSqlServerExtensionDataset contains the interface for handling input and output data used by the Java extension.
Class: PrimitiveDataset
The class PrimitiveDataset is an implementation of AbstractSqlServerExtensionDataset that stores simple types as primitives arrays.
PrimitiveDataset is provided in the SDK as an optional helper class. If you don't use this class, you need to implement your own class that inherits from AbstractSqlServerExtensionDataset.