How do you read a record in COBOL?

You read a record using the READ statement. For sequential files: READ file-name AT END imperative-statement END-READ. For indexed files with random access: set the RECORD KEY, then READ file-name INVALID KEY imperative-statement NOT INVALID KEY imperative-statement END-READ. For sequential access to indexed files: READ file-name NEXT AT END imperative-statement END-READ. Always check file status codes and handle end-of-file conditions appropriately.

How do you write a record in COBOL?

You write a record using the WRITE statement. For output files: WRITE record-name [FROM identifier] [AFTER ADVANCING lines]. For indexed files: set the RECORD KEY, then WRITE record-name INVALID KEY imperative-statement END-WRITE. For relative files: set the RELATIVE KEY, then WRITE record-name INVALID KEY imperative-statement END-WRITE. Always check file status codes to ensure the write was successful.

How do you update a record in COBOL?

You update a record using REWRITE after reading it. First, read the record you want to update. Then modify the record data in memory. Finally, use REWRITE record-name [INVALID KEY imperative-statement] END-REWRITE to write the updated record back to the file. REWRITE replaces the record that was last read. For indexed files, the record key cannot be changed. Always check file status codes after REWRITE.

How do you delete a record in COBOL?

You delete a record using the DELETE statement. For indexed files: set the RECORD KEY to the record to delete, then DELETE file-name RECORD INVALID KEY imperative-statement END-DELETE. For relative files: set the RELATIVE KEY, then DELETE file-name RECORD INVALID KEY imperative-statement END-DELETE. DELETE removes the record from the file. Sequential files (and ESDS) do not support DELETE. Always check file status codes after DELETE.

What is a record structure in COBOL?

A record structure in COBOL defines the layout and fields of a record in a file. It's defined in the FILE SECTION under the FD (File Description) for the file. The record structure specifies field names, data types (PIC clauses), and field positions. Example: 01 CUSTOMER-RECORD. 05 CUSTOMER-ID PIC 9(5). 05 CUSTOMER-NAME PIC X(30). Record structures determine how data is stored and accessed in files.

What is file status in COBOL?

File status is a two-character code that indicates the result of a file operation. It's defined in FILE-CONTROL with FILE STATUS IS status-field. Common status codes: '00' = successful, '10' = end of file, '23' = record not found (for random access), '30' = permanent I/O error. Checking file status after each I/O operation allows programs to handle errors and end-of-file conditions appropriately.

What is the difference between READ and READ NEXT?

READ (without NEXT) is used for random access to indexed or relative files - you set the key first, then read. READ NEXT is used for sequential access - it reads the next record in sequence. For sequential files, use READ (NEXT is implied). For indexed files with sequential access, explicitly use READ NEXT to read records in key order. READ NEXT advances through records sequentially, while READ accesses a specific record by key.

MainframeMaster

COBOL Tutorial

Progress0 of 0 lessons

COBOL Record Management

Record management in COBOL involves operations for reading, writing, updating, and deleting records in files. This includes defining record structures, opening and closing files, performing I/O operations (READ, WRITE, REWRITE, DELETE), handling file status codes, managing record keys for indexed files, and processing records sequentially or randomly. Understanding record management is essential for file processing, data manipulation, and building file-based applications in mainframe COBOL programs.

What is Record Management?

Record management encompasses all operations for working with file records:

Reading records: Retrieving records from files
Writing records: Adding new records to files
Updating records: Modifying existing records
Deleting records: Removing records from files
Record structure: Defining the layout of records
File operations: Opening, closing, and managing files
Error handling: Managing file status codes and errors

Effective record management ensures data integrity, proper file handling, and reliable file operations.

Defining Record Structures

Record structures define the layout of records in files:

Basic Record Definition

cobol

1
2
3
4
5
6
7
8
9
       DATA DIVISION.
       FILE SECTION.
       FD  CUSTOMER-FILE.
       01  CUSTOMER-RECORD.
           05  CUSTOMER-ID      PIC 9(5).
           05  CUSTOMER-NAME    PIC X(30).
           05  CUSTOMER-ADDRESS PIC X(50).
           05  CUSTOMER-BALANCE PIC 9(8)V99.
           05  FILLER           PIC X(35).  *> Padding to fixed length

The record structure defines:

Field names: Identifiers for each data element
Data types: PIC clauses specifying format and size
Field positions: Order and position within the record
Record length: Total size of the record

Hierarchical Record Structures

cobol

1
2
3
4
5
6
7
8
9
10
11
12
       FD  EMPLOYEE-FILE.
       01  EMPLOYEE-RECORD.
           05  EMPLOYEE-ID      PIC 9(6).
           05  EMPLOYEE-NAME.
               10  FIRST-NAME   PIC X(15).
               10  LAST-NAME    PIC X(15).
           05  EMPLOYEE-ADDRESS.
               10  STREET       PIC X(30).
               10  CITY         PIC X(20).
               10  STATE        PIC X(2).
               10  ZIP-CODE    PIC 9(5).
           05  EMPLOYEE-SALARY  PIC 9(7)V99.

Hierarchical structures organize related fields into groups, improving readability and organization.

Opening and Closing Files

Files must be opened before use and closed when done:

File Open Modes

cobol

1
2
3
4
5
6
7
8
9
10
11
       *> INPUT - Read only
       OPEN INPUT CUSTOMER-FILE
       
       *> OUTPUT - Write only (creates new file, deletes existing)
       OPEN OUTPUT CUSTOMER-FILE
       
       *> I-O - Input-Output (read and write, for updates)
       OPEN I-O CUSTOMER-FILE
       
       *> EXTEND - Append to existing file
       OPEN EXTEND CUSTOMER-FILE

Choose the appropriate mode based on your operations:

INPUT: When you only need to read records
OUTPUT: When creating a new file (deletes existing file)
I-O: When you need to read, update, or delete records
EXTEND: When adding records to an existing file

Closing Files

cobol

1
2
3
4
5
6
7
8
9
10
11
12
13
       *> Close single file
       CLOSE CUSTOMER-FILE
       
       *> Close multiple files
       CLOSE CUSTOMER-FILE
             EMPLOYEE-FILE
             TRANSACTION-FILE
       
       *> Close with error handling
       CLOSE CUSTOMER-FILE
       IF CUSTOMER-STATUS NOT = '00'
           DISPLAY "Error closing file: " CUSTOMER-STATUS
       END-IF

Always close files when done to ensure data is written and resources are released.

Reading Records

Reading records depends on file organization and access mode:

Sequential File Reading

cobol

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
       PROCEDURE DIVISION.
       MAIN-PARA.
           OPEN INPUT CUSTOMER-FILE
           
           PERFORM UNTIL END-OF-FILE
               READ CUSTOMER-FILE
                   AT END
                       SET END-OF-FILE TO TRUE
                       DISPLAY "End of file reached"
                   NOT AT END
                       DISPLAY "Customer: " CUSTOMER-NAME
                       DISPLAY "Balance: " CUSTOMER-BALANCE
                       *> Process record
               END-READ
           END-PERFORM
           
           CLOSE CUSTOMER-FILE
           STOP RUN.

Sequential reading processes records one after another in the order they appear in the file.

Random Access Reading (Indexed Files)

cobol

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
       WORKING-STORAGE SECTION.
       01  SEARCH-CUSTOMER-ID  PIC 9(5).
       
       PROCEDURE DIVISION.
       MAIN-PARA.
           OPEN INPUT CUSTOMER-FILE
           
           *> Set key for random access
           MOVE 12345 TO CUSTOMER-ID
           READ CUSTOMER-FILE
               INVALID KEY
                   DISPLAY "Customer not found: " CUSTOMER-ID
               NOT INVALID KEY
                   DISPLAY "Customer found: " CUSTOMER-NAME
                   DISPLAY "Balance: " CUSTOMER-BALANCE
           END-READ
           
           CLOSE CUSTOMER-FILE
           STOP RUN.

For random access, set the RECORD KEY before reading. The key must match an existing record.

Sequential Reading (Indexed Files)

cobol

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
       PROCEDURE DIVISION.
       MAIN-PARA.
           OPEN INPUT CUSTOMER-FILE
           
           *> Read records in key order
           PERFORM UNTIL END-OF-FILE
               READ CUSTOMER-FILE NEXT
                   AT END
                       SET END-OF-FILE TO TRUE
                   NOT AT END
                       DISPLAY "Customer ID: " CUSTOMER-ID
                       DISPLAY "Name: " CUSTOMER-NAME
           END-READ
           END-PERFORM
           
           CLOSE CUSTOMER-FILE
           STOP RUN.

READ NEXT reads records sequentially in key order for indexed files.

Writing Records

Writing records adds new records to files:

Writing to Sequential Files

cobol

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
       PROCEDURE DIVISION.
       MAIN-PARA.
           OPEN OUTPUT CUSTOMER-FILE
           
           *> Prepare record data
           MOVE 12345 TO CUSTOMER-ID
           MOVE 'JOHN SMITH' TO CUSTOMER-NAME
           MOVE '123 MAIN ST' TO CUSTOMER-ADDRESS
           MOVE 1000.50 TO CUSTOMER-BALANCE
           
           *> Write record
           WRITE CUSTOMER-RECORD
           
           *> Or write from another field
           MOVE WS-CUSTOMER-DATA TO CUSTOMER-RECORD
           WRITE CUSTOMER-RECORD
           
           CLOSE CUSTOMER-FILE
           STOP RUN.

Writing to Indexed Files

cobol

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
       PROCEDURE DIVISION.
       MAIN-PARA.
           OPEN OUTPUT CUSTOMER-FILE
           
           *> Set key and record data
           MOVE 12345 TO CUSTOMER-ID
           MOVE 'JOHN SMITH' TO CUSTOMER-NAME
           MOVE 1000.50 TO CUSTOMER-BALANCE
           
           *> Write record
           WRITE CUSTOMER-RECORD
               INVALID KEY
                   DISPLAY "Error: Duplicate key or invalid key"
                   DISPLAY "Customer ID: " CUSTOMER-ID
               NOT INVALID KEY
                   DISPLAY "Record written successfully"
           END-WRITE
           
           CLOSE CUSTOMER-FILE
           STOP RUN.

For indexed files, set the RECORD KEY before writing. INVALID KEY handles duplicate keys or invalid key values.

Updating Records

Updating records requires reading first, then rewriting:

Update Process

cobol

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
       PROCEDURE DIVISION.
       MAIN-PARA.
           OPEN I-O CUSTOMER-FILE
           
           *> Step 1: Read the record to update
           MOVE 12345 TO CUSTOMER-ID
           READ CUSTOMER-FILE
               INVALID KEY
                   DISPLAY "Customer not found: " CUSTOMER-ID
                   STOP RUN
               NOT INVALID KEY
                   *> Step 2: Modify the record data
                   ADD 100.00 TO CUSTOMER-BALANCE
                   MOVE 'UPDATED ADDRESS' TO CUSTOMER-ADDRESS
                   
                   *> Step 3: Write the updated record back
                   REWRITE CUSTOMER-RECORD
                       INVALID KEY
                           DISPLAY "Error updating record"
                       NOT INVALID KEY
                           DISPLAY "Record updated successfully"
                   END-REWRITE
           END-READ
           
           CLOSE CUSTOMER-FILE
           STOP RUN.

The update process:

1. Read the record: Retrieve the record you want to update
2. Modify data: Change field values in memory
3. REWRITE: Write the updated record back to the file

Important: For indexed files, you cannot change the RECORD KEY when updating. The key must remain the same.

Deleting Records

Deleting records removes them from the file:

Deleting from Indexed Files

cobol

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
       PROCEDURE DIVISION.
       MAIN-PARA.
           OPEN I-O CUSTOMER-FILE
           
           *> Set key of record to delete
           MOVE 12345 TO CUSTOMER-ID
           
           *> Delete the record
           DELETE CUSTOMER-FILE RECORD
               INVALID KEY
                   DISPLAY "Record not found: " CUSTOMER-ID
               NOT INVALID KEY
                   DISPLAY "Record deleted successfully"
           END-DELETE
           
           CLOSE CUSTOMER-FILE
           STOP RUN.

DELETE removes the record with the specified key from the file. The record is permanently removed and cannot be recovered.

Deleting from Relative Files

cobol

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
       WORKING-STORAGE SECTION.
       01  RECORD-NUMBER      PIC 9(4).
       
       PROCEDURE DIVISION.
       MAIN-PARA.
           OPEN I-O LOOKUP-TABLE
           
           *> Set relative record number
           MOVE 10 TO RECORD-NUMBER
           
           *> Delete record at that position
           DELETE LOOKUP-TABLE RECORD
               INVALID KEY
                   DISPLAY "Record not found at position: " RECORD-NUMBER
           END-DELETE
           
           CLOSE LOOKUP-TABLE
           STOP RUN.

For relative files, set the RELATIVE KEY (record number) before deleting.

File Status Codes

File status codes indicate the result of file operations:

Common File Status Codes
Status Code	Meaning	Description
'00'	Success	Operation completed successfully
'10'	End of file	No more records available (sequential read)
'23'	Record not found	Record with specified key not found (random access)
'30'	Permanent I/O error	Hardware or file system error
'22'	Duplicate key	Attempt to write duplicate key (indexed files)
'24'	Boundary violation	Attempt to write beyond file limits

Using File Status

cobol

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
       ENVIRONMENT DIVISION.
       INPUT-OUTPUT SECTION.
       FILE-CONTROL.
           SELECT CUSTOMER-FILE ASSIGN TO CUSTOMER.DAT
               FILE STATUS IS CUSTOMER-STATUS.
       
       WORKING-STORAGE SECTION.
       01  CUSTOMER-STATUS     PIC XX.
           88  FILE-OK          VALUE '00'.
           88  END-OF-FILE      VALUE '10'.
           88  RECORD-NOT-FOUND VALUE '23'.
           88  FILE-ERROR       VALUE '30'.
       
       PROCEDURE DIVISION.
       MAIN-PARA.
           OPEN INPUT CUSTOMER-FILE
           
           IF NOT FILE-OK
               EVALUATE CUSTOMER-STATUS
                   WHEN '30'
                       DISPLAY "Permanent I/O error"
                   WHEN OTHER
                       DISPLAY "File open error: " CUSTOMER-STATUS
               END-EVALUATE
               STOP RUN
           END-IF
           
           *> Process file...
           CLOSE CUSTOMER-FILE
           STOP RUN.

Record Processing Patterns

Pattern 1: Sequential Processing

cobol

1
2
3
4
5
6
7
8
9
10
       *> Process all records sequentially
       OPEN INPUT INPUT-FILE
       PERFORM UNTIL END-OF-FILE
           READ INPUT-FILE
               AT END SET END-OF-FILE TO TRUE
               NOT AT END
                   PERFORM PROCESS-RECORD
           END-READ
       END-PERFORM
       CLOSE INPUT-FILE

Pattern 2: Lookup and Update

cobol

1
2
3
4
5
6
7
8
9
10
11
12
13
       *> Look up record and update
       MOVE SEARCH-KEY TO RECORD-KEY
       READ FILE-NAME
           INVALID KEY
               DISPLAY "Record not found"
           NOT INVALID KEY
               *> Modify record
               ADD AMOUNT TO BALANCE-FIELD
               REWRITE RECORD-NAME
                   INVALID KEY
                       DISPLAY "Update failed"
               END-REWRITE
       END-READ

Pattern 3: Add New Record

cobol

1
2
3
4
5
6
7
8
9
10
11
12
13
       *> Add new record with validation
       MOVE NEW-KEY TO RECORD-KEY
       MOVE NEW-DATA TO RECORD-FIELDS
       WRITE RECORD-NAME
           INVALID KEY
               IF FILE-STATUS = '22'
                   DISPLAY "Duplicate key - record exists"
               ELSE
                   DISPLAY "Write error: " FILE-STATUS
               END-IF
           NOT INVALID KEY
               DISPLAY "Record added successfully"
       END-WRITE

Best Practices for Record Management

Follow these best practices:

Always check file status: Check FILE STATUS after every file operation
Handle end of file: Use AT END for sequential reads to detect end of file
Use INVALID KEY: Handle INVALID KEY for random access operations
Close files properly: Always close files when done, even on errors
Validate before write: Ensure data is valid before writing records
Read before update: Always read a record before updating it
Set keys correctly: Ensure RECORD KEY or RELATIVE KEY is set before random operations
Handle errors gracefully: Provide meaningful error messages and recovery options
Use appropriate open mode: Choose INPUT, OUTPUT, I-O, or EXTEND based on operations needed
Document record structure: Comment record definitions to explain field purposes

Common Mistakes in Record Management

Avoid these common mistakes:

Mistake 1: Not Checking File Status

cobol

1
2
3
4
5
6
7
8
9
10
11
       *> WRONG: No error checking
       READ CUSTOMER-FILE
       DISPLAY CUSTOMER-NAME
       
       *> CORRECT: Check file status
       READ CUSTOMER-FILE
           AT END
               DISPLAY "End of file"
           NOT AT END
               DISPLAY CUSTOMER-NAME
       END-READ

Mistake 2: Forgetting to Set Key

cobol

1
2
3
4
5
6
7
8
9
       *> WRONG: Key not set
       READ CUSTOMER-FILE
           INVALID KEY...
       
       *> CORRECT: Set key first
       MOVE 12345 TO CUSTOMER-ID
       READ CUSTOMER-FILE
           INVALID KEY...
       END-READ

Mistake 3: Updating Without Reading

cobol

1
2
3
4
5
6
7
8
9
10
       *> WRONG: REWRITE without READ
       MOVE NEW-DATA TO CUSTOMER-RECORD
       REWRITE CUSTOMER-RECORD
       
       *> CORRECT: Read first, then REWRITE
       READ CUSTOMER-FILE
           NOT INVALID KEY
               MOVE NEW-DATA TO CUSTOMER-RECORD
               REWRITE CUSTOMER-RECORD
           END-READ

Explain Like I'm 5: Record Management

Think of record management like managing a filing cabinet:

Opening a file is like opening the filing cabinet drawer - you need to open it before you can use it
Reading a record is like taking a file out of the cabinet to look at it
Writing a record is like putting a new file into the cabinet
Updating a record is like taking out a file, changing something in it, and putting it back
Deleting a record is like throwing away a file from the cabinet
Closing a file is like closing the drawer when you're done - you always close it when finished

So record management is like organizing and managing files in a filing cabinet - you open it, work with files, and close it when done!

Practice Exercises

Complete these exercises to reinforce your understanding:

Exercise 1: Sequential Reading

Create a program that opens a sequential file, reads all records, and displays each record's key information. Handle end of file appropriately.

Exercise 2: Random Access Lookup

Create a program that opens an indexed file, accepts a key value, and looks up the record. Display the record if found, or an error message if not found.

Exercise 3: Record Update

Create a program that reads a record from an indexed file, modifies specific fields, and uses REWRITE to update the record. Include proper error handling.

Exercise 4: Add New Record

Create a program that accepts input for a new record, sets the key and data fields, and writes the record to an indexed file. Handle duplicate key errors.

Exercise 5: Delete Record

Create a program that accepts a key value, deletes the corresponding record from an indexed file, and handles cases where the record is not found.

Test Your Knowledge

1. What statement is used to update an existing record?

UPDATE
REWRITE
MODIFY
CHANGE

2. What must you do before reading a record with random access?

Open the file
Set the RECORD KEY
Set file status
Initialize the record

3. What file status code indicates end of file?

'00'
'10'
'23'
'30'

4. What file status code indicates record not found?

'00'
'10'
'23'
'30'

5. What is the correct sequence for updating a record?

REWRITE, then READ
READ, modify data, then REWRITE
WRITE, then READ
DELETE, then WRITE

6. Which file organization does NOT support DELETE?

KSDS
ESDS
RRDS
All support DELETE

COBOL Tutorial

COBOL Record Management

What is Record Management?

Defining Record Structures

Basic Record Definition

Hierarchical Record Structures

Opening and Closing Files

File Open Modes

Closing Files

Reading Records

Sequential File Reading

Random Access Reading (Indexed Files)

Sequential Reading (Indexed Files)

Writing Records

Writing to Sequential Files

Writing to Indexed Files

Updating Records

Update Process

Deleting Records

Deleting from Indexed Files

Deleting from Relative Files

File Status Codes

Using File Status

Record Processing Patterns

Pattern 1: Sequential Processing

Pattern 2: Lookup and Update

Pattern 3: Add New Record

Best Practices for Record Management

Common Mistakes in Record Management

Mistake 1: Not Checking File Status

Mistake 2: Forgetting to Set Key

Mistake 3: Updating Without Reading

Explain Like I'm 5: Record Management

Practice Exercises

Exercise 1: Sequential Reading

Exercise 2: Random Access Lookup

Exercise 3: Record Update

Exercise 4: Add New Record

Exercise 5: Delete Record

Test Your Knowledge

Related Concepts

Related Concepts

File Operations

File Organization

VSAM Concepts

Error Handling

Related Pages