null indicator values in cobol db2

June 29, 2025 sulawesi

In COBOL programming, when working with relational databases like DB2, one of the key concepts that developers must understand is how to handle null values. These values indicate that data is missing or undefined in the database. To manage this effectively, COBOL uses null indicators to track whether a field contains valid data or a null value. In this topic, we will explore the concept of null indicator values in COBOL DB2, how they work, and how they are used in real-world applications.

Table of Contents

What is a Null Indicator?

A null indicator in COBOL is a special variable that is used to track the presence of null values when interacting with a DB2 database. A null value is different from a blank space, zero, or any other default value. It is used to represent missing, undefined, or unknown data in a database field. Without the use of null indicators, COBOL programs would not be able to distinguish between a valid value and a missing one.

The null indicator helps a COBOL program determine whether the data returned from the DB2 database is null or non-null. This is especially important when dealing with SQL SELECT statements that fetch data from a table.

How Do Null Indicators Work in COBOL?

When working with DB2 in COBOL, you will define a null indicator variable to track whether a database field contains a null value. In COBOL, null indicators are typically defined as signed 2-byte fields. These indicators help differentiate between actual data and null values when retrieving or updating records.

Zero (0): If the null indicator is set to 0, it means the data fetched from the DB2 table is not null.
Negative One (-1): If the null indicator is set to -1, it means the data fetched from the DB2 table is null.

In other words, when a null value is present in the database field, the corresponding null indicator will be set to -1. If the field contains valid data, the null indicator will be set to 0.

Declaring a Null Indicator in COBOL

In COBOL, defining a null indicator is an essential step when dealing with DB2 databases. It is typically declared in the WORKING-STORAGE SECTION of the COBOL program. Here is an example of how a null indicator is defined in COBOL:

01  NULL-INDICATOR        PIC S9(4) COMP-5.

In this example:

NULL-INDICATOR is the name of the variable used to track the null status of a field.
PIC S9(4) specifies that it is a signed 4-digit field.
COMP-5 indicates that the null indicator is a binary field, which is recommended for optimal performance when using DB2.

When you fetch data from the DB2 database, the value of the null indicator will be set automatically by DB2. The program can then check the value of the null indicator to determine if the field contains data or is null.

Example of Using Null Indicators in COBOL with DB2

Let’s walk through an example where a COBOL program fetches customer information from a DB2 database. One of the fields in the database, CUSTOMER_EMAIL, may contain null values, so we will use a null indicator to track the status of this field.

Example Database Table

Assume you have a CUSTOMER table in your DB2 database with the following columns:

CUSTOMER_ID (INTEGER)
CUSTOMER_NAME (VARCHAR)
CUSTOMER_EMAIL (VARCHAR) – can be null

COBOL Program Code

Here is an example of how a null indicator might be used in a COBOL program to handle the CUSTOMER_EMAIL field, which may be null.

       IDENTIFICATION DIVISION.PROGRAM-ID. FETCH-CUSTOMER.ENVIRONMENT DIVISION.INPUT-OUTPUT SECTION.FILE-CONTROL.SELECT CUSTOMER-FILE ASSIGN TO DATABASE.DATA DIVISION.FILE SECTION.FD  CUSTOMER-FILE.01  CUSTOMER-RECORD.05  CUSTOMER-ID        PIC 9(5).05  CUSTOMER-NAME      PIC X(50).05  CUSTOMER-EMAIL     PIC X(100).05  NULL-INDICATOR     PIC S9(4) COMP-5.WORKING-STORAGE SECTION.01  WS-CUSTOMER-ID        PIC 9(5).01  WS-CUSTOMER-NAME      PIC X(50).01  WS-CUSTOMER-EMAIL     PIC X(100).01  WS-NUMBER-OF-RECORDS PIC 9(5) VALUE 0.EXEC SQLDECLARE C1 CURSOR FOR CUSTOMER_CURSOREND-EXEC.EXEC SQLOPEN C1END-EXEC.PERFORM FETCH-CUSTOMER-RECORDS.STOP RUN.FETCH-CUSTOMER-RECORDS.EXEC SQLFETCH C1 INTO :WS-CUSTOMER-ID, :WS-CUSTOMER-NAME, :WS-CUSTOMER-EMAIL, :NULL-INDICATOREND-EXECIF SQLCODE = 100DISPLAY 'End of Records'ELSEPERFORM CHECK-NULLEMAILEND-IF.CHECK-NULLEMAIL.IF NULL-INDICATOR = -1DISPLAY 'Customer Email is Null.'ELSEDISPLAY 'Customer Email: ' WS-CUSTOMER-EMAILEND-IF.

Explanation of the Example

File Section:
- The CUSTOMER-RECORD structure defines the fields that will be fetched from the DB2 database, including the NULL-INDICATOR.
Working-Storage Section:
- Temporary variables like WS-CUSTOMER-ID, WS-CUSTOMER-NAME, and WS-CUSTOMER-EMAIL are defined to hold the data fetched from the database.
SQL Section:
- The EXEC SQL statements declare and open a cursor for fetching data from the CUSTOMER table.
Fetching Data:
- The FETCH statement retrieves data from the table, including the NULL-INDICATOR for the CUSTOMER_EMAIL field.
Checking for Null:
- The CHECK-NULLEMAIL paragraph checks the value of the NULL-INDICATOR. If the indicator is -1, it means the CUSTOMER_EMAIL field is null, and the program displays a message indicating this. If the indicator is 0, it means the field contains valid data, and the program displays the email.

Common Issues When Using Null Indicators

1. Incorrect Null Indicator Definition

If the null indicator is not defined correctly, it may result in incorrect handling of null values. Always use the correct data type, such as PIC S9(4) COMP-5, to define the null indicator.

2. Null Indicator and SQLCODE Handling

If the FETCH statement encounters an error, the SQLCODE will not be zero. Always check the SQLCODE for error handling to ensure data is fetched successfully.

3. Invalid or Unexpected Null Indicator Values

Ensure that the null indicator is set to 0 or -1. Any other value may lead to unexpected results, making it essential to verify its validity before proceeding with any logic based on its value.

Handling null values in DB2 COBOL programs is essential for ensuring the correctness of data processing and retrieval. By using null indicators, COBOL programs can efficiently manage database fields that contain null data. In this topic, we discussed the concept of null indicators, how they work, and provided an example of how to use them in a COBOL program. By understanding and correctly implementing null indicators, you can improve the reliability and accuracy of your COBOL programs when interacting with DB2 databases.