In this part 2, I will share the topologies that a Java Integration Stage supports it’s structure and it’s features.
Structure of User Code:
Processor is the main class that is extended by the user code. The processor class consists of methods that are invoked by the Java Integration Stage. At minimum, user code must implement the following two abstract methods.
- public abstract boolean validateConfiguration()
- public abstract void process()
Other supported functionalities:
- Custom Properties: This the user can set from the tool and the Java code can access this using getUserPropertyDefinitions(). These values (like DB connection properties etc) can be edited in stage or configuration editor.
- Java Beans: Stage automatically converts between bean properties and DataStage column. These bean properties are imported to column definitions.
- Column Based Implementation: Column data can be passed from DataStage to Java via a column based interface. Input columns can be transferred to output columns. Supports adding columns to output links dynamically at runtime.
- User-Defined Functions: The Java Integration Stage supports the execution of existing user defined functions that use Java Beans or Primitive types in their calling interface. User does not have to write any code at all to invoke such existing methods.
- Logging: User’s code calls logger method to log informational, warning, and error messages. These calls write message to job log (runtime) or to CAS log (design time).
- Meta-Data Import: One can map java columns to existing DataStage columns in the column mapping editor.