COBOL Tutorial

COBOL Indexed File Operations

Progress0 of 0 lessons

OPEN for Indexed Files

Indexed files in COBOL use key-based access methods to store and retrieve data, allowing for both sequential and random access. The OPEN statement for indexed files is similar to sequential files but requires appropriate file declarations in the ENVIRONMENT DIVISION to specify the key fields and organization.

Declaration and OPEN Syntax

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
31
32
33
34
35
36
ENVIRONMENT DIVISION.
INPUT-OUTPUT SECTION.
FILE-CONTROL.
    SELECT CUSTOMER-FILE
        ASSIGN TO CUSTFILE
        ORGANIZATION IS INDEXED
        ACCESS MODE IS DYNAMIC
        RECORD KEY IS CUSTOMER-ID
        ALTERNATE RECORD KEY IS CUSTOMER-NAME WITH DUPLICATES
        FILE STATUS IS CUSTOMER-STATUS.
 
DATA DIVISION.
FILE SECTION.
FD  CUSTOMER-FILE
    LABEL RECORDS ARE STANDARD
    RECORD CONTAINS 100 CHARACTERS.
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 S9(7)V99.
    05  FILLER           PIC X(6).
 
WORKING-STORAGE SECTION.
01  CUSTOMER-STATUS      PIC XX.
 
PROCEDURE DIVISION.
MAIN-PROCESS.
    OPEN I-O CUSTOMER-FILE
    IF CUSTOMER-STATUS = "00"
       PERFORM PROCESS-CUSTOMERS
    ELSE
       PERFORM FILE-ERROR-HANDLER
    END-IF
    CLOSE CUSTOMER-FILE
    STOP RUN.

ORGANIZATION IS INDEXED specifies an indexed file organization
RECORD KEY IS identifies the primary key field
ALTERNATE RECORD KEY IS defines secondary access paths
WITH DUPLICATES allows non-unique values for alternate keys
ACCESS MODE IS DYNAMIC allows both sequential and random access

OPEN Modes for Indexed Files

Mode	Description	Operations Allowed
INPUT	Opens for reading only	READ, START
OUTPUT	Creates a new file	WRITE
I-O	Opens for reading and writing	READ, WRITE, REWRITE, DELETE, START
EXTEND	Opens for adding records	WRITE (adding new records only)

Access Modes for Indexed Files

Access Mode	Description	Common Operations
SEQUENTIAL	Records are accessed in key sequence	READ NEXT, WRITE (in key order)
RANDOM	Records are accessed directly by key	READ with specific key values
DYNAMIC	Combines sequential and random access	READ NEXT and direct READ operations

DYNAMIC is the most flexible access mode, allowing both sequential processing and direct record access as needed.

Common File Status Codes for OPEN

Status Code	Meaning	Typical Solution
"00"	Successful operation	Proceed with file operations
"35"	File not found	Create file or check path/name
"37"	Open mode not permitted	Check file permissions
"41"	File already open	Close file before reopening
"93"	Resource not available	Check resource allocation
"96"	No file space in directory	Free space or use different location

Indexed File Creation Example

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
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
IDENTIFICATION DIVISION.
PROGRAM-ID. INDEXCRT.
 
ENVIRONMENT DIVISION.
INPUT-OUTPUT SECTION.
FILE-CONTROL.
    SELECT CUSTOMER-FILE
        ASSIGN TO "CUSTOMER.IDX"
        ORGANIZATION IS INDEXED
        ACCESS MODE IS SEQUENTIAL
        RECORD KEY IS CUST-ID
        ALTERNATE RECORD KEY IS CUST-NAME WITH DUPLICATES
        FILE STATUS IS FILE-STATUS.
    
    SELECT INPUT-FILE
        ASSIGN TO "CUSTDATA.DAT"
        ORGANIZATION IS SEQUENTIAL
        FILE STATUS IS INPUT-STATUS.
 
DATA DIVISION.
FILE SECTION.
FD  CUSTOMER-FILE
    LABEL RECORDS ARE STANDARD
    RECORD CONTAINS 100 CHARACTERS.
01  CUSTOMER-RECORD.
    05  CUST-ID          PIC 9(5).
    05  CUST-NAME        PIC X(30).
    05  CUST-ADDRESS     PIC X(50).
    05  CUST-BALANCE     PIC S9(7)V99.
    05  FILLER           PIC X(6).
 
FD  INPUT-FILE
    LABEL RECORDS ARE STANDARD
    RECORD CONTAINS 100 CHARACTERS.
01  INPUT-RECORD         PIC X(100).
 
WORKING-STORAGE SECTION.
01  FILE-STATUS          PIC XX.
01  INPUT-STATUS         PIC XX.
01  EOF-FLAG             PIC X VALUE "N".
    88  END-OF-FILE        VALUE "Y".
 
PROCEDURE DIVISION.
MAIN-PROCESS.
    PERFORM INITIALIZATION
    PERFORM PROCESS-RECORDS UNTIL END-OF-FILE
    PERFORM TERMINATION
    STOP RUN.
 
INITIALIZATION.
    OPEN INPUT INPUT-FILE
    IF INPUT-STATUS NOT = "00"
       DISPLAY "Error opening input file: " INPUT-STATUS
       MOVE "Y" TO EOF-FLAG
    ELSE
       OPEN OUTPUT CUSTOMER-FILE
       IF FILE-STATUS NOT = "00"
          DISPLAY "Error creating indexed file: " FILE-STATUS
          MOVE "Y" TO EOF-FLAG
          CLOSE INPUT-FILE
       END-IF
    END-IF.
 
PROCESS-RECORDS.
    READ INPUT-FILE INTO CUSTOMER-RECORD
       AT END
          MOVE "Y" TO EOF-FLAG
       NOT AT END
          WRITE CUSTOMER-RECORD
             INVALID KEY
                DISPLAY "Duplicate or invalid key: " CUST-ID
             NOT INVALID KEY
                DISPLAY "Record written for: " CUST-ID
          END-WRITE
    END-READ.
 
TERMINATION.
    CLOSE INPUT-FILE
    CLOSE CUSTOMER-FILE
    DISPLAY "File creation complete."
    DISPLAY "File status: " FILE-STATUS.

This example demonstrates creating an indexed file by reading data from a sequential file and writing it to a new indexed file. It includes proper error handling for both file opening and key validation.

READ with KEY Options

The READ statement for indexed files has several formats, allowing for sequential access in key order or direct access using specific key values. This flexibility is one of the primary advantages of indexed files over sequential files.

READ Statement Formats

cobol

1
2
3
4
5
6
7
8
9
10
11
12
* Sequential access (next record in key sequence)
READ file-name NEXT RECORD [INTO identifier-1]
     [AT END imperative-statement-1]
     [NOT AT END imperative-statement-2]
     [END-READ]
 
* Random access (specific record by key)
READ file-name RECORD [INTO identifier-1]
     KEY IS data-name-1
     [INVALID KEY imperative-statement-1]
     [NOT INVALID KEY imperative-statement-2]
     [END-READ]

For sequential access, use the NEXT keyword
For random access, specify the key with KEY IS
AT END applies to sequential reads when no more records exist
INVALID KEY applies to random reads when no matching record is found
INTO clause moves the record to a working storage area

Sequential Reading Example

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
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
IDENTIFICATION DIVISION.
PROGRAM-ID. INDEXSEQ.
 
ENVIRONMENT DIVISION.
INPUT-OUTPUT SECTION.
FILE-CONTROL.
    SELECT CUSTOMER-FILE
        ASSIGN TO "CUSTOMER.IDX"
        ORGANIZATION IS INDEXED
        ACCESS MODE IS DYNAMIC
        RECORD KEY IS CUST-ID
        FILE STATUS IS FILE-STATUS.
 
DATA DIVISION.
FILE SECTION.
FD  CUSTOMER-FILE
    LABEL RECORDS ARE STANDARD.
01  CUSTOMER-RECORD.
    05  CUST-ID          PIC 9(5).
    05  CUST-NAME        PIC X(30).
    05  CUST-ADDRESS     PIC X(50).
    05  CUST-BALANCE     PIC S9(7)V99.
 
WORKING-STORAGE SECTION.
01  FILE-STATUS          PIC XX.
01  EOF-FLAG             PIC X VALUE "N".
    88  END-OF-FILE        VALUE "Y".
01  RECORD-COUNT         PIC 9(5) VALUE ZERO.
 
PROCEDURE DIVISION.
MAIN-PROCESS.
    PERFORM INITIALIZATION
    PERFORM PROCESS-RECORDS UNTIL END-OF-FILE
    PERFORM TERMINATION
    STOP RUN.
 
INITIALIZATION.
    OPEN INPUT CUSTOMER-FILE
    IF FILE-STATUS NOT = "00"
       DISPLAY "Error opening file: " FILE-STATUS
       MOVE "Y" TO EOF-FLAG
    END-IF.
 
PROCESS-RECORDS.
    READ CUSTOMER-FILE NEXT RECORD
       AT END
          MOVE "Y" TO EOF-FLAG
       NOT AT END
          ADD 1 TO RECORD-COUNT
          DISPLAY "Record " RECORD-COUNT ": " CUST-ID " - " CUST-NAME
    END-READ.
 
TERMINATION.
    CLOSE CUSTOMER-FILE
    DISPLAY "Total records read: " RECORD-COUNT.

This example reads all records from an indexed file in key sequence (sorted by CUST-ID) using the READ NEXT format. This is similar to reading a sequential file, but the records are retrieved in primary key order.

Random Reading Example

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
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
IDENTIFICATION DIVISION.
PROGRAM-ID. INDEXRND.
 
ENVIRONMENT DIVISION.
INPUT-OUTPUT SECTION.
FILE-CONTROL.
    SELECT CUSTOMER-FILE
        ASSIGN TO "CUSTOMER.IDX"
        ORGANIZATION IS INDEXED
        ACCESS MODE IS RANDOM
        RECORD KEY IS CUST-ID
        ALTERNATE RECORD KEY IS CUST-NAME WITH DUPLICATES
        FILE STATUS IS FILE-STATUS.
 
DATA DIVISION.
FILE SECTION.
FD  CUSTOMER-FILE
    LABEL RECORDS ARE STANDARD.
01  CUSTOMER-RECORD.
    05  CUST-ID          PIC 9(5).
    05  CUST-NAME        PIC X(30).
    05  CUST-ADDRESS     PIC X(50).
    05  CUST-BALANCE     PIC S9(7)V99.
 
WORKING-STORAGE SECTION.
01  FILE-STATUS          PIC XX.
    88  RECORD-FOUND       VALUE "00".
    88  RECORD-NOT-FOUND   VALUE "23".
01  SEARCH-ID            PIC 9(5).
01  SEARCH-NAME          PIC X(30).
 
PROCEDURE DIVISION.
MAIN-PROCESS.
    PERFORM INITIALIZATION
    PERFORM UNTIL 1 = 2
       DISPLAY "Enter customer ID (or 0 to quit): "
       ACCEPT SEARCH-ID
       IF SEARCH-ID = 0
          EXIT PERFORM
       END-IF
       PERFORM FIND-BY-ID
       
       DISPLAY "Enter customer name (or X to quit): "
       ACCEPT SEARCH-NAME
       IF SEARCH-NAME = "X"
          EXIT PERFORM
       END-IF
       PERFORM FIND-BY-NAME
    END-PERFORM
    PERFORM TERMINATION
    STOP RUN.
 
INITIALIZATION.
    OPEN INPUT CUSTOMER-FILE
    IF FILE-STATUS NOT = "00"
       DISPLAY "Error opening file: " FILE-STATUS
       STOP RUN
    END-IF.
 
FIND-BY-ID.
    MOVE SEARCH-ID TO CUST-ID
    READ CUSTOMER-FILE
       INVALID KEY
          DISPLAY "No customer found with ID: " SEARCH-ID
       NOT INVALID KEY
          DISPLAY "Found: " CUST-ID " - " CUST-NAME
          DISPLAY "Balance: " CUST-BALANCE
    END-READ.
 
FIND-BY-NAME.
    MOVE SEARCH-NAME TO CUST-NAME
    READ CUSTOMER-FILE KEY IS CUST-NAME
       INVALID KEY
          DISPLAY "No customer found with name: " SEARCH-NAME
       NOT INVALID KEY
          DISPLAY "Found: " CUST-ID " - " CUST-NAME
          DISPLAY "Balance: " CUST-BALANCE
    END-READ.
 
TERMINATION.
    CLOSE CUSTOMER-FILE.

This example demonstrates direct record access using both primary and alternate keys. It allows the user to search for customers by either ID or name, demonstrating the flexibility of indexed file access.

Reading with Alternate Keys

cobol

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
* Reading with primary key (default)
MOVE 12345 TO CUST-ID
READ CUSTOMER-FILE
    INVALID KEY PERFORM RECORD-NOT-FOUND-ROUTINE
END-READ
 
* Reading with alternate key
MOVE "SMITH JOHN" TO CUST-NAME
READ CUSTOMER-FILE KEY IS CUST-NAME
    INVALID KEY PERFORM NAME-NOT-FOUND-ROUTINE
END-READ
 
* Handling duplicate alternate keys
MOVE "SMITH" TO CUST-NAME
* First read gets the first occurrence
READ CUSTOMER-FILE KEY IS CUST-NAME
    INVALID KEY PERFORM NAME-NOT-FOUND-ROUTINE
END-READ
 
* Subsequent reads need a START and READ NEXT combination
* See the START section for handling duplicates

When reading with alternate keys that allow duplicates, only the first occurrence is retrieved with a direct READ. To access all records with the same alternate key value, use the START statement to position at the first occurrence, then use READ NEXT to retrieve all matching records.

Common Status Codes for READ Operations

Status Code	Meaning	Condition
"00"	Successful operation	Record found and read
"10"	End of file	No more records (READ NEXT)
"23"	Record not found	Key value not in file
"46"	Read beyond file end	Sequential read after end
"47"	File not open in correct mode	File not open for input

Best Practices for READ Operations

Always check file status after READ operations
Use INVALID KEY for random reads and AT END for sequential reads
Handle potential duplicate key conditions when using alternate keys
Consider reading into working storage to preserve the record area
Use DYNAMIC access mode when mixing sequential and random reads
Implement proper error recovery procedures for missing records
For better performance, minimize switching between access patterns

WRITE with KEY Specification

The WRITE statement for indexed files adds records to the file based on the key values in the record. Unlike sequential files, the physical placement of the record is determined by the key values rather than the sequence of WRITE operations. This ensures that records can be retrieved efficiently using their keys.

WRITE Statement Syntax

cobol

1
2
3
4
WRITE record-name [FROM identifier-1]
      [INVALID KEY imperative-statement-1]
      [NOT INVALID KEY imperative-statement-2]
      [END-WRITE]

record-name must be defined in the FILE SECTION with the correct key fields
FROM clause allows writing from a separate working storage area
INVALID KEY clause handles duplicate key and other key-related errors
NOT INVALID KEY executes when the write is successful
END-WRITE terminates the scope of the WRITE statement

Key Considerations for WRITE

Aspect	Rule	Example
Primary Key	Must be unique; duplicate values cause error	CUSTOMER-ID, ISBN, ACCOUNT-NUMBER
Alternate Keys	Can allow duplicates if specified	CUSTOMER-NAME, ZIP-CODE, PRODUCT-CATEGORY
Key Initialization	All key fields must be properly populated	MOVE "12345" TO CUSTOMER-ID
OPEN Mode	Requires OUTPUT, I-O, or EXTEND mode	OPEN OUTPUT CUSTOMER-FILE
Record Order	Physical order determined by key values	Records are stored in primary key sequence

Basic WRITE Operations

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
31
32
33
34
35
36
37
38
39
* Simple WRITE with error handling
MOVE "12345" TO CUSTOMER-ID
MOVE "JOHN SMITH" TO CUSTOMER-NAME
MOVE "123 MAIN ST" TO CUSTOMER-ADDRESS
MOVE 1000.00 TO CUSTOMER-BALANCE
 
WRITE CUSTOMER-RECORD
   INVALID KEY
      DISPLAY "Error writing record: " CUSTOMER-ID
      PERFORM ERROR-ROUTINE
   NOT INVALID KEY
      DISPLAY "Record written successfully"
END-WRITE
 
* WRITE with FROM clause
MOVE "12345" TO WS-CUSTOMER-ID
MOVE "JOHN SMITH" TO WS-CUSTOMER-NAME
MOVE "123 MAIN ST" TO WS-CUSTOMER-ADDRESS
MOVE 1000.00 TO WS-CUSTOMER-BALANCE
 
WRITE CUSTOMER-RECORD FROM WORK-CUSTOMER-RECORD
   INVALID KEY
      DISPLAY "Error writing record: " WS-CUSTOMER-ID
END-WRITE
 
* Status checking approach
MOVE "12345" TO CUSTOMER-ID
MOVE "JOHN SMITH" TO CUSTOMER-NAME
MOVE "123 MAIN ST" TO CUSTOMER-ADDRESS
MOVE 1000.00 TO CUSTOMER-BALANCE
 
WRITE CUSTOMER-RECORD
IF FILE-STATUS = "00"
   DISPLAY "Record written successfully"
ELSE IF FILE-STATUS = "22"
   DISPLAY "Duplicate key: " CUSTOMER-ID
ELSE
   DISPLAY "Error writing record: " FILE-STATUS
END-IF

These examples demonstrate different approaches to writing records to an indexed file, including error handling for invalid key conditions. The FROM clause is useful when working with multiple record formats or when you need to preserve the original record.

Complete Example: Adding Records

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
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
IDENTIFICATION DIVISION.
PROGRAM-ID. INDEXADD.
 
ENVIRONMENT DIVISION.
INPUT-OUTPUT SECTION.
FILE-CONTROL.
    SELECT CUSTOMER-FILE
        ASSIGN TO "CUSTOMER.IDX"
        ORGANIZATION IS INDEXED
        ACCESS MODE IS RANDOM
        RECORD KEY IS CUST-ID
        ALTERNATE RECORD KEY IS CUST-NAME WITH DUPLICATES
        FILE STATUS IS FILE-STATUS.
 
DATA DIVISION.
FILE SECTION.
FD  CUSTOMER-FILE
    LABEL RECORDS ARE STANDARD.
01  CUSTOMER-RECORD.
    05  CUST-ID          PIC 9(5).
    05  CUST-NAME        PIC X(30).
    05  CUST-ADDRESS     PIC X(50).
    05  CUST-BALANCE     PIC S9(7)V99.
 
WORKING-STORAGE SECTION.
01  FILE-STATUS          PIC XX.
    88  SUCCESS            VALUE "00".
    88  DUPLICATE-KEY      VALUE "22".
01  WORK-FIELDS.
    05  MORE-RECORDS     PIC X VALUE "Y".
        88  NO-MORE-RECORDS VALUE "N".
    05  KEY-EXISTS       PIC X.
        88  KEY-ALREADY-EXISTS VALUE "Y".
        88  KEY-IS-NEW         VALUE "N".
 
PROCEDURE DIVISION.
MAIN-PROCESS.
    PERFORM INITIALIZATION
    PERFORM ADD-RECORDS UNTIL NO-MORE-RECORDS
    PERFORM TERMINATION
    STOP RUN.
 
INITIALIZATION.
    OPEN I-O CUSTOMER-FILE
    IF NOT SUCCESS
       DISPLAY "File does not exist, creating new file"
       CLOSE CUSTOMER-FILE
       OPEN OUTPUT CUSTOMER-FILE
       IF NOT SUCCESS
          DISPLAY "Could not create file: " FILE-STATUS
          MOVE "N" TO MORE-RECORDS
       END-IF
    END-IF.
 
ADD-RECORDS.
    INITIALIZE CUSTOMER-RECORD
    DISPLAY "Enter customer ID (or 0 to quit): "
    ACCEPT CUST-ID
    
    IF CUST-ID = 0
       MOVE "N" TO MORE-RECORDS
    ELSE
       PERFORM CHECK-IF-KEY-EXISTS
       IF NOT KEY-ALREADY-EXISTS
          DISPLAY "Enter customer name: "
          ACCEPT CUST-NAME
          DISPLAY "Enter customer address: "
          ACCEPT CUST-ADDRESS
          DISPLAY "Enter customer balance: "
          ACCEPT CUST-BALANCE
 
          WRITE CUSTOMER-RECORD
             INVALID KEY
                DISPLAY "Error writing record - status: " FILE-STATUS
             NOT INVALID KEY
                DISPLAY "Record added successfully"
          END-WRITE
       END-IF
    END-IF.
 
CHECK-IF-KEY-EXISTS.
    MOVE "N" TO KEY-EXISTS
    READ CUSTOMER-FILE
       INVALID KEY
          CONTINUE
       NOT INVALID KEY
          DISPLAY "Customer ID " CUST-ID " already exists"
          DISPLAY "Name: " CUST-NAME
          MOVE "Y" TO KEY-EXISTS
    END-READ.
 
TERMINATION.
    CLOSE CUSTOMER-FILE.

This example demonstrates a complete program for adding records to an indexed file, with proper error handling and validation. It includes checking for duplicate keys before attempting to write, which provides better user feedback.

Common Status Codes for WRITE Operations

Status Code	Meaning	Handling Approach
"00"	Successful operation	Continue processing
"22"	Duplicate key	Generate new key or update existing record
"24"	Boundary violation	Check record size against file definition
"34"	Disk full or space allocation exceeded	Free disk space or extend allocation
"48"	File not open in write mode	Check OPEN mode (needs OUTPUT, I-O, or EXTEND)

Status code "22" (duplicate key) is particularly important for indexed files and should always be handled explicitly when writing records.

Best Practices for WRITE Operations

Always initialize all key fields before writing a record
Check for duplicate keys before attempting to write when appropriate
Use the INVALID KEY clause to handle write errors
Implement a consistent key generation strategy for new records
Consider using the FROM clause to preserve the working record
For bulk operations, open the file once and perform multiple writes
Validate data before writing to ensure data integrity
Maintain proper key relationships when working with related files

START Statement for Positioning

The START statement positions the file pointer at a record with a specific key value or key condition. This is essential for accessing indexed files in a specific sequence or for finding groups of records that match certain criteria. After positioning with START, subsequent READ NEXT operations retrieve records starting from that position.

START Statement Syntax

cobol

1
2
3
4
5
6
7
8
9
10
11
12
START file-name
      KEY {IS EQUAL TO     |
           IS =            |
           IS GREATER THAN |
           IS >            |
           IS NOT LESS THAN|
           IS NOT <        |
           IS GREATER THAN OR EQUAL TO |
           IS >=           } data-name-1
      [INVALID KEY imperative-statement-1]
      [NOT INVALID KEY imperative-statement-2]
      [END-START]

data-name-1 must be a key field (primary or alternate)
Comparison operators determine the positioning criteria
INVALID KEY executes when no matching record is found
NOT INVALID KEY executes when positioning is successful
END-START terminates the scope of the START statement

Key Comparison Operations

Operation	Description	Usage
EQUAL TO (=)	Exact match only	Finding specific records
GREATER THAN (>)	First record with key > value	Processing ranges above a threshold
NOT LESS THAN (NOT <)	First record with key >= value	Processing inclusive ranges
GREATER THAN OR EQUAL TO (>=)	First record with key >= value	Same as NOT LESS THAN

If no KEY clause is specified, the default is EQUAL TO for the primary key. These comparison operators determine which record the file pointer is positioned at for subsequent READ NEXT operations.

START and READ NEXT Examples

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
31
32
33
34
35
36
37
38
39
* Position at an exact customer ID
MOVE "12345" TO CUSTOMER-ID
START CUSTOMER-FILE KEY IS = CUSTOMER-ID
   INVALID KEY
      DISPLAY "Customer " CUSTOMER-ID " not found"
   NOT INVALID KEY
      PERFORM READ-CUSTOMER-RECORD
END-START
 
* Position at customers with IDs >= 50000
MOVE "50000" TO CUSTOMER-ID
START CUSTOMER-FILE KEY IS NOT < CUSTOMER-ID
   INVALID KEY
      DISPLAY "No customers with ID >= 50000"
   NOT INVALID KEY
      PERFORM PROCESS-CUSTOMERS UNTIL END-OF-FILE
END-START
 
* Position at customers in a specific ZIP code
MOVE "90210" TO CUSTOMER-ZIP
START CUSTOMER-FILE KEY IS = CUSTOMER-ZIP
   INVALID KEY
      DISPLAY "No customers in ZIP code " CUSTOMER-ZIP
   NOT INVALID KEY
      PERFORM PROCESS-ZIP-CUSTOMERS UNTIL NO-MORE-IN-ZIP
END-START
 
* Example of reading duplicate alternate keys
PROCESS-ZIP-CUSTOMERS.
    READ CUSTOMER-FILE NEXT RECORD
       AT END
          SET END-OF-FILE TO TRUE
       NOT AT END
          IF CUSTOMER-ZIP NOT = "90210"
             SET NO-MORE-IN-ZIP TO TRUE
          ELSE
             DISPLAY CUSTOMER-NAME " lives in Beverly Hills"
          END-IF
    END-READ.

These examples demonstrate how START positions the file for subsequent READ NEXT operations. This combination is particularly useful for processing ranges of records or records with duplicate alternate keys.

Complete Example: Processing Date Ranges

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
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
IDENTIFICATION DIVISION.
PROGRAM-ID. DATERANGE.
 
ENVIRONMENT DIVISION.
INPUT-OUTPUT SECTION.
FILE-CONTROL.
    SELECT TRANSACTION-FILE
        ASSIGN TO "TRANSACT.IDX"
        ORGANIZATION IS INDEXED
        ACCESS MODE IS DYNAMIC
        RECORD KEY IS TRANS-ID
        ALTERNATE RECORD KEY IS TRANS-DATE
        FILE STATUS IS FILE-STATUS.
 
DATA DIVISION.
FILE SECTION.
FD  TRANSACTION-FILE
    LABEL RECORDS ARE STANDARD.
01  TRANSACTION-RECORD.
    05  TRANS-ID         PIC 9(8).
    05  TRANS-DATE       PIC 9(8).  * YYYYMMDD
    05  TRANS-AMOUNT     PIC S9(7)V99.
    05  TRANS-TYPE       PIC X.
        88  DEPOSIT        VALUE "D".
        88  WITHDRAWAL     VALUE "W".
    05  TRANS-ACCOUNT    PIC 9(10).
    05  TRANS-DESCRIPTION PIC X(50).
 
WORKING-STORAGE SECTION.
01  FILE-STATUS          PIC XX.
01  START-DATE           PIC 9(8).
01  END-DATE             PIC 9(8).
01  TOTAL-AMOUNT         PIC S9(9)V99 VALUE ZERO.
01  TRANS-COUNT          PIC 9(5) VALUE ZERO.
01  FLAGS.
    05  EOF-FLAG         PIC X VALUE "N".
        88  END-OF-FILE    VALUE "Y".
    05  RANGE-FLAG       PIC X VALUE "N".
        88  END-OF-RANGE   VALUE "Y".
 
PROCEDURE DIVISION.
MAIN-PROCESS.
    PERFORM INITIALIZATION
    PERFORM PROCESS-DATE-RANGE
    PERFORM TERMINATION
    STOP RUN.
 
INITIALIZATION.
    OPEN INPUT TRANSACTION-FILE
    IF FILE-STATUS NOT = "00"
       DISPLAY "Error opening file: " FILE-STATUS
       MOVE "Y" TO EOF-FLAG
    ELSE
       DISPLAY "Enter start date (YYYYMMDD): "
       ACCEPT START-DATE
       DISPLAY "Enter end date (YYYYMMDD): "
       ACCEPT END-DATE
    END-IF.
 
PROCESS-DATE-RANGE.
    MOVE START-DATE TO TRANS-DATE
    START TRANSACTION-FILE KEY IS NOT < TRANS-DATE
       INVALID KEY
          DISPLAY "No transactions on or after " START-DATE
          MOVE "Y" TO EOF-FLAG
    END-START
    
    PERFORM UNTIL END-OF-FILE OR END-OF-RANGE
       READ TRANSACTION-FILE NEXT RECORD
          AT END
             MOVE "Y" TO EOF-FLAG
          NOT AT END
             IF TRANS-DATE > END-DATE
                MOVE "Y" TO RANGE-FLAG
             ELSE
                ADD 1 TO TRANS-COUNT
                IF DEPOSIT
                   ADD TRANS-AMOUNT TO TOTAL-AMOUNT
                ELSE
                   SUBTRACT TRANS-AMOUNT FROM TOTAL-AMOUNT
                END-IF
                
                DISPLAY "Transaction on " TRANS-DATE ": " 
                         TRANS-AMOUNT " (" TRANS-TYPE ")"
             END-IF
       END-READ
    END-PERFORM.
 
TERMINATION.
    CLOSE TRANSACTION-FILE
    DISPLAY "Transactions processed: " TRANS-COUNT
    DISPLAY "Net amount: " TOTAL-AMOUNT.

This example uses START with the NOT LESS THAN operation to position at the beginning of a date range, then reads sequentially until the end date is exceeded. This demonstrates a common pattern for processing time-based data in indexed files.

Using START with Different Keys

cobol

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
* Using the primary key
MOVE "12345" TO CUSTOMER-ID
START CUSTOMER-FILE KEY IS = CUSTOMER-ID
   ...
 
* Using an alternate key
MOVE "SMITH" TO CUSTOMER-NAME
START CUSTOMER-FILE KEY IS = CUSTOMER-NAME
   ...
 
* Finding partial key matches (prefix search)
MOVE "SMITH" TO CUSTOMER-NAME
MOVE SPACES TO CUSTOMER-NAME(6:25) * Clear the rest of the field
START CUSTOMER-FILE KEY IS NOT < CUSTOMER-NAME
   ...
* Then read until no longer matching the prefix
READ CUSTOMER-FILE NEXT RECORD
...
IF CUSTOMER-NAME(1:5) NOT = "SMITH"
   * No longer in the SMITH records
   ...

These examples show how to use START with different key fields. The last example demonstrates a technique for prefix searching, which is useful for finding all records that begin with certain characters.

Common Status Codes for START Operations

Status Code	Meaning	Handling Approach
"00"	Successful operation	Proceed with READ NEXT operations
"23"	Record not found	No matching key; handle as end of processing
"46"	Read beyond file end	Key value exceeds highest in file
"47"	File not open in input mode	Check OPEN mode (needs INPUT or I-O)

Best Practices for START Operations

Use START before sequential reading to establish a starting point
Always check for INVALID KEY conditions
For prefix searching, pad partial keys appropriately
When reading after START, check that records still match criteria
For duplicate alternate keys, use START and READ NEXT together
Use greater than or equal comparisons for range processing
Consider performance implications for large files or frequent repositioning

DELETE and REWRITE Operations

The DELETE and REWRITE statements allow you to modify existing records in indexed files. DELETE removes a record from the file, while REWRITE updates an existing record. Both operations rely on proper record identification using key values.

DELETE Statement Syntax

cobol

1
2
3
4
DELETE file-name RECORD
      [INVALID KEY imperative-statement-1]
      [NOT INVALID KEY imperative-statement-2]
      [END-DELETE]

The record to be deleted must first be read into the record area
The file must be open in I-O mode
INVALID KEY handles the case where the record was not found
Deletion is based on the primary key of the record

REWRITE Statement Syntax

cobol

1
2
3
4
REWRITE record-name [FROM identifier-1]
      [INVALID KEY imperative-statement-1]
      [NOT INVALID KEY imperative-statement-2]
      [END-REWRITE]

The record must first be read into the record area before rewriting
The primary key value cannot be changed during REWRITE
Alternate key values can be modified
The file must be open in I-O mode
FROM clause allows updating from a different storage area

Basic DELETE and REWRITE Examples

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
31
* DELETE example
* First read the record
MOVE "12345" TO CUSTOMER-ID
READ CUSTOMER-FILE
   INVALID KEY
      DISPLAY "Customer " CUSTOMER-ID " not found"
   NOT INVALID KEY
      DELETE CUSTOMER-FILE RECORD
         INVALID KEY
            DISPLAY "Error deleting customer " CUSTOMER-ID
         NOT INVALID KEY
            DISPLAY "Customer " CUSTOMER-ID " deleted"
      END-DELETE
END-READ
 
* REWRITE example
* First read the record
MOVE "12345" TO CUSTOMER-ID
READ CUSTOMER-FILE
   INVALID KEY
      DISPLAY "Customer " CUSTOMER-ID " not found"
   NOT INVALID KEY
      * Modify fields (except primary key)
      MOVE "NEW ADDRESS" TO CUSTOMER-ADDRESS
      REWRITE CUSTOMER-RECORD
         INVALID KEY
            DISPLAY "Error updating customer " CUSTOMER-ID
         NOT INVALID KEY
            DISPLAY "Customer " CUSTOMER-ID " updated"
      END-REWRITE
END-READ

These examples demonstrate the basic pattern for DELETE and REWRITE operations: first read the record to establish currency, then perform the operation with appropriate error handling.

Complete Example: Record Maintenance

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
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
IDENTIFICATION DIVISION.
PROGRAM-ID. RECMAINT.
 
ENVIRONMENT DIVISION.
INPUT-OUTPUT SECTION.
FILE-CONTROL.
    SELECT CUSTOMER-FILE
        ASSIGN TO "CUSTOMER.IDX"
        ORGANIZATION IS INDEXED
        ACCESS MODE IS DYNAMIC
        RECORD KEY IS CUST-ID
        ALTERNATE RECORD KEY IS CUST-NAME WITH DUPLICATES
        FILE STATUS IS FILE-STATUS.
 
DATA DIVISION.
FILE SECTION.
FD  CUSTOMER-FILE
    LABEL RECORDS ARE STANDARD.
01  CUSTOMER-RECORD.
    05  CUST-ID          PIC 9(5).
    05  CUST-NAME        PIC X(30).
    05  CUST-ADDRESS     PIC X(50).
    05  CUST-BALANCE     PIC S9(7)V99.
 
WORKING-STORAGE SECTION.
01  FILE-STATUS          PIC XX.
    88  SUCCESS            VALUE "00".
    88  RECORD-NOT-FOUND   VALUE "23".
01  MENU-CHOICE          PIC 9.
01  SEARCH-ID            PIC 9(5).
01  CONTINUE-FLAG        PIC X VALUE "Y".
    88  EXIT-PROGRAM       VALUE "N".
 
PROCEDURE DIVISION.
MAIN-PROCESS.
    PERFORM INITIALIZATION
    PERFORM UNTIL EXIT-PROGRAM
       PERFORM DISPLAY-MENU
       EVALUATE MENU-CHOICE
          WHEN 1
             PERFORM UPDATE-CUSTOMER
          WHEN 2
             PERFORM DELETE-CUSTOMER
          WHEN 3
             MOVE "N" TO CONTINUE-FLAG
          WHEN OTHER
             DISPLAY "Invalid choice"
       END-EVALUATE
    END-PERFORM
    PERFORM TERMINATION
    STOP RUN.
 
INITIALIZATION.
    OPEN I-O CUSTOMER-FILE
    IF NOT SUCCESS
       DISPLAY "Error opening file: " FILE-STATUS
       MOVE "N" TO CONTINUE-FLAG
    END-IF.
 
DISPLAY-MENU.
    DISPLAY "CUSTOMER MAINTENANCE"
    DISPLAY "1. Update Customer"
    DISPLAY "2. Delete Customer"
    DISPLAY "3. Exit"
    DISPLAY "Enter choice: "
    ACCEPT MENU-CHOICE.
 
UPDATE-CUSTOMER.
    DISPLAY "Enter customer ID to update: "
    ACCEPT SEARCH-ID
    MOVE SEARCH-ID TO CUST-ID
    
    READ CUSTOMER-FILE
       INVALID KEY
          DISPLAY "Customer " SEARCH-ID " not found"
       NOT INVALID KEY
          DISPLAY "Current information:"
          DISPLAY "  Name: " CUST-NAME
          DISPLAY "  Address: " CUST-ADDRESS
          DISPLAY "  Balance: " CUST-BALANCE
          
          DISPLAY "Enter new name (or ENTER to keep current): "
          ACCEPT CUST-NAME
          
          DISPLAY "Enter new address (or ENTER to keep current): "
          ACCEPT CUST-ADDRESS
          
          DISPLAY "Enter new balance (or ENTER to keep current): "
          ACCEPT CUST-BALANCE
          
          REWRITE CUSTOMER-RECORD
             INVALID KEY
                DISPLAY "Error updating customer: " FILE-STATUS
             NOT INVALID KEY
                DISPLAY "Customer information updated"
          END-REWRITE
    END-READ.
 
DELETE-CUSTOMER.
    DISPLAY "Enter customer ID to delete: "
    ACCEPT SEARCH-ID
    MOVE SEARCH-ID TO CUST-ID
    
    READ CUSTOMER-FILE
       INVALID KEY
          DISPLAY "Customer " SEARCH-ID " not found"
       NOT INVALID KEY
          DISPLAY "Customer information:"
          DISPLAY "  Name: " CUST-NAME
          DISPLAY "  Address: " CUST-ADDRESS
          DISPLAY "  Balance: " CUST-BALANCE
          
          DISPLAY "Are you sure you want to delete this customer? (Y/N): "
          ACCEPT CONTINUE-FLAG
          
          IF CONTINUE-FLAG = "Y"
             DELETE CUSTOMER-FILE RECORD
                INVALID KEY
                   DISPLAY "Error deleting customer: " FILE-STATUS
                NOT INVALID KEY
                   DISPLAY "Customer " SEARCH-ID " deleted"
             END-DELETE
          ELSE
             DISPLAY "Deletion cancelled"
          END-IF
    END-READ.
 
TERMINATION.
    CLOSE CUSTOMER-FILE.

This comprehensive example demonstrates a maintenance program that allows updating and deleting customer records. It includes user confirmation for deletions and demonstrates the proper sequence of operations for both REWRITE and DELETE.

Key Constraints During REWRITE

When using REWRITE with indexed files, key constraints must be observed:

Key Type	Can Be Changed?	Notes
Primary Key	No	To change primary key, DELETE old record and WRITE new one
Alternate Key (Unique)	Yes	Cannot conflict with existing values
Alternate Key (Duplicates)	Yes	Can match existing values

Attempting to change the primary key during REWRITE will result in an error. Changing an alternate key to a value that already exists will fail unless the WITH DUPLICATES phrase was specified for that key.

Changing the Primary Key

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
31
32
33
34
35
* Process to change a primary key (ID change)
* 1. Read the old record
MOVE "12345" TO CUSTOMER-ID
READ CUSTOMER-FILE
   INVALID KEY
      DISPLAY "Customer " CUSTOMER-ID " not found"
   NOT INVALID KEY
      * 2. Save the record data
      MOVE CUSTOMER-RECORD TO WORK-RECORD
      
      * 3. Delete the old record
      DELETE CUSTOMER-FILE RECORD
         INVALID KEY
            DISPLAY "Error deleting old record"
         NOT INVALID KEY
            * 4. Change the key in the work record
            MOVE "54321" TO WORK-ID
            
            * 5. Write as a new record
            MOVE WORK-RECORD TO CUSTOMER-RECORD
            WRITE CUSTOMER-RECORD
               INVALID KEY
                  DISPLAY "Error writing new record"
                  * Recovery - try to restore the old record
                  MOVE "12345" TO CUSTOMER-ID
                  MOVE WORK-RECORD TO CUSTOMER-RECORD
                  WRITE CUSTOMER-RECORD
                     INVALID KEY
                        DISPLAY "CRITICAL ERROR: Lost original record"
                  END-WRITE
               NOT INVALID KEY
                  DISPLAY "ID changed from 12345 to 54321"
            END-WRITE
      END-DELETE
END-READ

This example demonstrates the process for changing a primary key, which requires deleting the old record and writing a new one with the updated key. Note the error recovery logic that attempts to restore the original record if the new record cannot be written.

Common Status Codes for DELETE/REWRITE

Status Code	Meaning	Applies To
"00"	Successful operation	DELETE, REWRITE
"22"	Duplicate key	REWRITE (alternate key conflict)
"23"	Record not found	DELETE, REWRITE
"43"	Record not read before DELETE/REWRITE	DELETE, REWRITE
"47"	File not open in I-O mode	DELETE, REWRITE
"92"	Record locked	DELETE, REWRITE (multi-user)

Best Practices for DELETE and REWRITE

Always READ the record before DELETE or REWRITE operations
Implement appropriate error handling for all operations
Consider using transactions for related operations to maintain data integrity
For high-value data, create backup records before deletion
Never assume DELETE or REWRITE will succeed; always check status codes
When changing primary keys, use the DELETE/WRITE pattern with error recovery
Maintain audit trails of record changes for critical applications
Consider locking strategies in multi-user environments

Test Your Knowledge

1. What is the primary advantage of using indexed files over sequential files in COBOL?

Indexed files use less disk space
Indexed files allow direct access to specific records using keys
Indexed files are faster for all operations
Indexed files do not require OPEN statements

2. In COBOL, which clause in the SELECT statement defines the primary key for an indexed file?

PRIMARY KEY IS
RECORD KEY IS
ACCESS KEY IS
INDEX KEY IS

3. What happens if you attempt to WRITE a record with a key value that already exists in an indexed file?

The new record overwrites the existing one
The system merges the two records
A duplicate key error (status "22") occurs
The new record is appended to the end of the file

4. Which statement is used to position the file pointer at a specific location in an indexed file before reading?

POSITION
SEEK
START
LOCATE

5. What is the purpose of the ALTERNATE RECORD KEY clause in indexed file declarations?

It defines a fallback key if the primary key fails
It provides additional access paths to the data with different keys
It creates a backup copy of the primary key
It establishes temporary keys for sorting operations

COBOL Tutorial

COBOL Indexed File Operations

OPEN for Indexed Files

Declaration and OPEN Syntax

OPEN Modes for Indexed Files

Access Modes for Indexed Files

Common File Status Codes for OPEN

Indexed File Creation Example

READ with KEY Options

READ Statement Formats

Sequential Reading Example

Random Reading Example

Reading with Alternate Keys

Common Status Codes for READ Operations

Best Practices for READ Operations

WRITE with KEY Specification

WRITE Statement Syntax

Key Considerations for WRITE

Basic WRITE Operations

Complete Example: Adding Records

Common Status Codes for WRITE Operations

Best Practices for WRITE Operations

START Statement for Positioning

START Statement Syntax

Key Comparison Operations

START and READ NEXT Examples

Complete Example: Processing Date Ranges

Using START with Different Keys

Common Status Codes for START Operations

Best Practices for START Operations

DELETE and REWRITE Operations

DELETE Statement Syntax

REWRITE Statement Syntax

Basic DELETE and REWRITE Examples

Complete Example: Record Maintenance

Key Constraints During REWRITE

Changing the Primary Key

Common Status Codes for DELETE/REWRITE

Best Practices for DELETE and REWRITE

Test Your Knowledge

Frequently Asked Questions

What is the difference between a primary key and an alternate key in indexed files?

How do I handle duplicate key errors when writing to indexed files?

What are the performance considerations when working with indexed files?

How do I convert between sequential and indexed files?

What is the best practice for handling file status codes with indexed files?

Related Concepts

File Organization Types

File Description (FD)

Sequential File Operations

ENVIRONMENT DIVISION

File Status and Error Handling

Related Pages