SQL/MP Programming Manual for COBOL

Data Retrieval and Modification

HP NonStop SQL/MP Programming Manual for COBOL—529758-003

4-12

Using SQL Cursors

An SQL cursor is a named pointer that a host language program (C, COBOL, Pascal,

or TAL) can use to access a set of rows in a table or view, one row at a time. Using a

cursor, a program can process rows in the same way it might process records in a

sequential file. The program can test the data in each row at the current cursor position

and then, if the data meets certain criteria, the program can display, update, delete, or

ignore the row.

Figure 4-1 shows the steps you follow to declare and use a static SQL cursor in a

COBOL program. A cursor operation must execute each statement in this specified

order. All steps are required, even if you execute the FETCH statement only once to

retrieve a single row.

Figure 4-1. Using a Static SQL Cursor in a COBOL Program

DATA DIVISION.

• • •

EXEC SQL BEGIN DECLARE SECTION END-EXEC.

* Declare host variable(s).

• • •

EXEC SQL END DECLARE SECTION END-EXEC.

• • •

EXEC SQL DECLARE CURSOR1 CURSOR FOR

SELECT COLUMN1, COLUMN2, COLUMN n

FROM =TABLE

WHERE COLUMN1 = :HOSTVAR-FIND-ROW

END-EXEC.

• • •

PROCEDURE DIVISION.

• • •

* Initialize the host variable(s).

MOVE INITIAL-VALUE TO HOSTVAR-FIND-ROW.

• • •

EXEC SQL OPEN CURSOR1 END-EXEC.

* Fetch data from a row into the host variable(s).

EXEC SQL FETCH CURSOR1

INTO :HOSTVAR_1, :HOSTVAR_2, :HOSTVAR n

END-EXEC.

* Process the row values in the host variable(s).

• • •

* Branch back to fetch another row.

• • •

EXEC SQL CLOSE CURSOR1 END-EXEC.

VST009.vsd