COBOL Data Types Concepts

Data types in COBOL define how data is stored, interpreted, and manipulated within programs. Understanding data types is fundamental to COBOL programming as it determines memory allocation, arithmetic operations, and data representation. COBOL provides various data types including numeric, character, binary, and specialized types for different business requirements.

Understanding Data Types

Data types in COBOL are defined using PICTURE clauses that specify the format, size, and characteristics of data items. Each data type has specific storage requirements, arithmetic capabilities, and display formats. Proper understanding of data types ensures efficient memory usage and correct data processing.

Numeric Data Types

1. DISPLAY Format (PIC 9)

DISPLAY format is the default numeric format in COBOL, storing numbers as character strings. This format is human-readable and suitable for input/output operations but less efficient for arithmetic calculations.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
       DATA DIVISION.
       WORKING-STORAGE SECTION.
       01 DISPLAY-NUMBERS.
           05 INTEGER-VALUE     PIC 9(5) VALUE 12345.
           05 DECIMAL-VALUE    PIC 9(5)V99 VALUE 123.45.
           05 SIGNED-VALUE     PIC S9(4) VALUE -1234.
           05 ZERO-FILLED      PIC 9(6) VALUE ZEROS.
       
       PROCEDURE DIVISION.
           DISPLAY 'DISPLAY Format Examples:'
           DISPLAY 'Integer: ' INTEGER-VALUE
           DISPLAY 'Decimal: ' DECIMAL-VALUE
           DISPLAY 'Signed: ' SIGNED-VALUE
           DISPLAY 'Zero-filled: ' ZERO-FILLED

DISPLAY format stores numbers as character strings, making them human-readable but requiring conversion for arithmetic operations. The V indicates decimal point position, and S indicates signed numbers.

2. BINARY Format (COMP)

BINARY format stores numbers in binary representation, providing efficient arithmetic operations and compact storage. This format is ideal for calculations and counters.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
       DATA DIVISION.
       WORKING-STORAGE SECTION.
       01 BINARY-NUMBERS.
           05 COUNTER          PIC 9(5) COMP VALUE 1000.
           05 CALCULATION      PIC 9(8)V99 COMP VALUE 12345.67.
           05 INDEX-VALUE      PIC 9(3) COMP VALUE 50.
       
       PROCEDURE DIVISION.
           DISPLAY 'BINARY Format Examples:'
           DISPLAY 'Counter: ' COUNTER
           DISPLAY 'Calculation: ' CALCULATION
           DISPLAY 'Index: ' INDEX-VALUE
           
           *> Efficient arithmetic operations
           ADD 100 TO COUNTER
           MULTIPLY 2 BY CALCULATION

BINARY format provides efficient storage and arithmetic operations. Numbers are stored in binary representation, making calculations faster but requiring conversion for display.

3. PACKED DECIMAL Format (COMP-3)

PACKED DECIMAL format stores decimal numbers efficiently while maintaining decimal precision. This format is commonly used for financial calculations where precision is critical.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
       DATA DIVISION.
       WORKING-STORAGE SECTION.
       01 PACKED-DECIMALS.
           05 AMOUNT          PIC 9(7)V99 COMP-3 VALUE 12345.67.
           05 BALANCE         PIC S9(8)V99 COMP-3 VALUE -5000.00.
           05 RATE            PIC 9(3)V99 COMP-3 VALUE 5.25.
       
       PROCEDURE DIVISION.
           DISPLAY 'PACKED DECIMAL Format Examples:'
           DISPLAY 'Amount: ' AMOUNT
           DISPLAY 'Balance: ' BALANCE
           DISPLAY 'Rate: ' RATE
           
           *> Precise decimal arithmetic
           COMPUTE AMOUNT = AMOUNT * RATE / 100

PACKED DECIMAL format maintains decimal precision while providing efficient storage. Each digit is stored in half a byte, with the sign stored in the rightmost half-byte.

Character Data Types

1. Alphanumeric (PIC X)

Alphanumeric data type stores character data including letters, numbers, spaces, and special characters. This is the most flexible character type in COBOL.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
       DATA DIVISION.
       WORKING-STORAGE SECTION.
       01 CHARACTER-DATA.
           05 NAME            PIC X(20) VALUE 'John Smith'.
           05 ADDRESS         PIC X(50) VALUE '123 Main Street'.
           05 PHONE           PIC X(12) VALUE '555-123-4567'.
           05 EMPTY-FIELD     PIC X(10) VALUE SPACES.
       
       PROCEDURE DIVISION.
           DISPLAY 'Character Data Examples:'
           DISPLAY 'Name: ' NAME
           DISPLAY 'Address: ' ADDRESS
           DISPLAY 'Phone: ' PHONE
           DISPLAY 'Empty: "' EMPTY-FIELD '"'

Alphanumeric fields can store any character and are commonly used for names, addresses, descriptions, and other text data. Spaces are used to pad shorter values.

2. Alphabetic (PIC A)

Alphabetic data type stores only letters and spaces, providing validation for fields that should contain only alphabetic characters.

1
2
3
4
5
6
7
8
9
10
11
12
       DATA DIVISION.
       WORKING-STORAGE SECTION.
       01 ALPHABETIC-DATA.
           05 FIRST-NAME     PIC A(10) VALUE 'John'.
           05 LAST-NAME      PIC A(15) VALUE 'Smith'.
           05 CITY           PIC A(20) VALUE 'New York'.
       
       PROCEDURE DIVISION.
           DISPLAY 'Alphabetic Data Examples:'
           DISPLAY 'First Name: ' FIRST-NAME
           DISPLAY 'Last Name: ' LAST-NAME
           DISPLAY 'City: ' CITY

Alphabetic fields automatically validate that only letters and spaces are stored, preventing numeric or special characters from being entered.

Specialized Data Types

1. USAGE Clauses

USAGE clauses specify how data is stored in memory, affecting performance and storage efficiency. Different usage types are optimized for different operations.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
       DATA DIVISION.
       WORKING-STORAGE SECTION.
       01 USAGE-EXAMPLES.
           05 DISPLAY-NUM    PIC 9(5) USAGE DISPLAY VALUE 12345.
           05 BINARY-NUM     PIC 9(5) USAGE COMP VALUE 12345.
           05 PACKED-NUM     PIC 9(5)V99 USAGE COMP-3 VALUE 123.45.
           05 INDEX-NUM      PIC 9(5) USAGE INDEX VALUE 100.
           05 POINTER-VAL    USAGE POINTER VALUE NULL.
       
       PROCEDURE DIVISION.
           DISPLAY 'USAGE Clause Examples:'
           DISPLAY 'Display: ' DISPLAY-NUM
           DISPLAY 'Binary: ' BINARY-NUM
           DISPLAY 'Packed: ' PACKED-NUM

USAGE clauses control memory storage format. DISPLAY is default, COMP provides binary storage, COMP-3 provides packed decimal, INDEX is optimized for array indexing, and POINTER stores memory addresses.

2. Data Type Conversion

Data type conversion is essential when working with different data types. COBOL provides automatic conversion in many cases, but explicit conversion may be necessary for optimal performance.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
       DATA DIVISION.
       WORKING-STORAGE SECTION.
       01 CONVERSION-DATA.
           05 DISPLAY-VAL    PIC 9(5) VALUE 12345.
           05 BINARY-VAL     PIC 9(5) COMP.
           05 CHAR-VAL       PIC X(5).
       
       PROCEDURE DIVISION.
           DISPLAY 'Data Type Conversion Examples:'
           
           *> Convert DISPLAY to BINARY
           MOVE DISPLAY-VAL TO BINARY-VAL
           DISPLAY 'Display to Binary: ' BINARY-VAL
           
           *> Convert numeric to character
           MOVE DISPLAY-VAL TO CHAR-VAL
           DISPLAY 'Numeric to Character: ' CHAR-VAL
           
           *> Convert character to numeric
           MOVE '98765' TO CHAR-VAL
           MOVE CHAR-VAL TO DISPLAY-VAL
           DISPLAY 'Character to Numeric: ' DISPLAY-VAL

Data type conversion allows different data types to work together. COBOL automatically converts between compatible types, but explicit conversion ensures proper handling and performance optimization.

Best Practices for Data Types

• Choose Appropriate Types: Select data types that match your data and operations
• Use COMP for Calculations: Use COMP format for arithmetic operations and counters
• Use COMP-3 for Decimals: Use COMP-3 for financial calculations requiring precision
• Validate Input: Use appropriate PIC clauses to validate input data
• Consider Storage: Balance storage efficiency with processing requirements
• Document Types: Document data type choices and their rationale

Common Data Type Patterns

• Financial Data: Use COMP-3 for monetary amounts and rates
• Counters and Indexes: Use COMP for loop counters and array indexes
• Text Data: Use PIC X for names, addresses, and descriptions
• Numeric Input: Use PIC 9 for user input and display
• Flags and Codes: Use PIC X(1) for single-character codes
• Dates and Times: Use PIC 9(8) for dates and PIC 9(6) for times