Raising errors from Java stored procedures without "ORA-29532 Java call terminated by uncaught Java exception"

25,404

As I know, Oracle can't catch the Java errors directly. And the Oracle docs note of the ORA-29532 validate this idea:

ORA-29532: Java call terminated by uncaught Java exception: string

Cause: A Java exception or error was signaled and could not be resolved by the Java code.

Action: Modify Java code, if this behavior is not intended.

According to this text, I think you should handle the exception in the Java code.

You can solve that with these things:

  1. The return value of the Java function may be String - not void, and you may send the error description in the return variable.
  2. You may save the error to an Oracle table in the Java code and read it in the PL/SQL code.
  3. You just send the error to the Oracle user dump file with the System.out.println();.

Edit: Adam's final solution

This is roughly what I implemented based on the answer:

package mypackage;

public class MyClass {

    public static final int SUCCESS = 1;
    public static final int FAILURE = 0;

    /**
     * This method actually performs the business logic.
     */
    public static void doSomething(String arg1, String arg2) throws SQLException {
        // Actually do something...
    }

    /**
     * This method is called from PL/SQL.
     */
    public static int doSomething(String arg1, String arg2, int[] errorCode, String[] errorMessage) {
        try {
            doSomething(arg1, arg2);
            return success();
        } catch (SQLException e) {
            return failure(e, errorCode, errorMessage);
        }
    }

    private static int success() {
        return SUCCESS;
    }

    private static int failure(SQLException e, int[] errorCode, String[] errorMessage) {
        errorCode[0] = e.getErrorCode();
        errorMessage[0] = e.getMessage();
        return FAILURE;
    }

}

Then in PL/SQL:

SUCCESS CONSTANT BINARY_INTEGER := 1;
FAILURE CONSTANT BINARY_INTEGER := 0;

SUBTYPE error_code_type  IS BINARY_INTEGER;
SUBTYPE error_message_type IS VARCHAR2(1000);

PROCEDURE
  raise_error_if_failure
  (
    status        BINARY_INTEGER,
    error_code    ERROR_CODE_TYPE,
    error_message ERROR_MESSAGE_TYPE
  )
  IS
BEGIN
  IF status = FAILURE THEN
    raise_application_error(error_code, error_message);
  END IF;
END;


FUNCTION
  do_something_in_java
  (
    arg1          VARCHAR2,
    arg2          VARCHAR2,
    error_code    OUT ERROR_CODE_TYPE,
    error_message OUT ERROR_MESSAGE_TYPE
  )
  RETURN BINARY_INTEGER
    AS LANGUAGE JAVA
    NAME 'mypackage.MyClass.doSomething(java.lang.String, java.lang.String, int[], java.lang.String[]) return int';


PROCEDURE
  do_something
  (
    arg1 VARCHAR2,
    arg2 VARCHAR2
  )
IS
  error_code    ERROR_CODE_TYPE;
  error_message ERROR_MESSAGE_TYPE;
BEGIN
  raise_error_if_failure(
      do_something_in_java(arg1, arg2, error_code, error_message),
      error_code,
      error_message
    );
END;
Share:
25,404
Adam Paynter
Author by

Adam Paynter

I was born in Prince Edward Island, Canada and have been experimenting with programming since grade five. In 2007, I graduated with first class honors from the Bachelor of Computer Science program at the University of New Brunswick, Canada. I currently live in Plano, Texas with my wife and two daughters. I work as a software developer for Capital One. I still pursue a handful of personal projects during lunch hours and weekends.

Updated on June 14, 2020

Comments

  • Adam Paynter
    Adam Paynter almost 4 years

    Suppose you have a Java class which defines a copyFile(String, String) method:

    public class FileSystem {
    
        public static void copyFile(String sourcePath, String destinationPath)
            throws IOException
        {
            // ignore the fact that I'm not properly managing resources...
            FileInputStream source = new FileInputStream(sourcePath);
            FileOutputStream destination = new FileOutputStream(destinationPath);
    
            copyStream(source, destination);
    
            source.close();
            destination.close();
        }
    
        private static void copyStream(InputStream source, OutputStream destination)
            throws IOException
        {
            byte[] buffer = new byte[1024];
    
            int length;
            while ( (length = source.read(buffer)) != -1 ) {
                destination.write(buffer, 0, length);
            }
        }
    
    }
    

    And suppose you wrap this in a Java stored procedure:

    CREATE PROCEDURE copy_file (source_path VARCHAR2, destination_path VARCHAR2)
    AS LANGUAGE JAVA
    NAME 'FileSystem.copyFile(String, String)';
    

    Now suppose you call the copy_file stored procedure and a Java IOException is thrown:

    BEGIN
      copy_file( '/some/file/that/does/not/exist.txt', '/path/to/destination.txt' );
    END;
    

    In the PL/SQL block, the error raised is:

    ORA-29532 Java call terminated by uncaught Java exception

    The error message also contains a description of the uncaught Exception, but it is still an ORA-29532. Is there a way to throw an Exception from Java that dictates both the error code and the error message raised in PL/SQL?