COMPUTE

Structured Mode Syntax

COMPUTE
ASSIGN

[ROUNDED] {operand1 [:]= } ...

arithmetic-expression
operand2
SUBSTRING (operand2,operand3,operand4)

 

{operand1 := } ...

arithmetic-expression
operand2
SUBSTRING (operand2,operand3,operand4)

                 

Reporting Mode Syntax

COMPUTE
ASSIGN

[ROUNDED]

{operand1 [:]= }

arithmetic-expression
operand2
SUBSTRING (operand2,operand3,operand4)

This document covers the following topics:

For an explanation of the symbols used in the syntax diagram, see Syntax Symbols.

Related Statements: ADD | COMPRESS | DIVIDE | EXAMINE | MOVE | MOVE ALL | MULTIPLY | RESET | SEPARATE | SUBTRACT

Belongs to Function Group: Arithmetic and Data Movement Operations

COMPUTE Usage

The COMPUTE statement is used to perform an arithmetic or assignment operation.

A COMPUTE statement with multiple target operands (operand1) is identical to the corresponding individual COMPUTE statements if the source operand (operand2) is not an arithmetic expression. 

#TARGET1 := #TARGET2 := #SOURCE

is identical to

#TARGET1 := #SOURCE 
#TARGET2 := #SOURCE

Example:

DEFINE DATA LOCAL
1 #ARRAY(I4/1:3) INIT <3,0,9>
1 #INDEX(I4)
1 #RESULT(I4)
END-DEFINE
*
#INDEX := 1
*
#INDEX := 		    /* #INDEX is 3 
#RESULT := 		   /* #RESULT is 9  
#ARRAY(#INDEX)                                            
*
#INDEX := 2
*
#INDEX  := 		   /* #INDEX is 0
#ARRAY(3) := 	 	/* returns runtime error NAT1316 
#ARRAY(#INDEX)  
END

If the source operand is an arithmetic expression, the expression is evaluated and its result is stored in a temporary variable. Then the temporary variable is assigned to the target operands. 

#TARGET1 := #TARGET2 := #SOURCE1 + 1 
is identical to
#TEMP := #SOURCE1 + 1 
#TARGET1 := #TEMP 
#TARGET2 := #TEMP

Example:

DEFINE DATA LOCAL
1 #ARRAY(I4/1:3) INIT <2, 0, 9>
1 #INDEX(I4)
1 #RESULT(I4)
END-DEFINE
*
#INDEX := 1
* 
#INDEX := 		/* #INDEX is 3 
#RESULT := 		/* #RESULT is 3  
#ARRAY(#INDEX) + 1 
*
#INDEX := 2
* 
#INDEX  :=		/* #INDEX is 0
#ARRAY(3) := 		/* returns run time error NAT1316 
#ARRAY(#INDEX) 
END

For further information, see Rules for Arithmetic Assignment in the Programming Guide and particularly the following sections:

COMPUTE Syntax Description

Operand Definition Table:

Operand Possible Structure Possible Formats Referencing Permitted Dynamic Definition
operand1   S A   M   A U N P I F B D T L C G O yes yes
operand2 C S A   N E A U N P I F B D T L C G O yes no
operand3 C S             N P I   B*             yes no
operand4 C S             N P I   B*             yes no

* If operand3 or operand4 is a binary variable, it may be used only with a length of less than or equal to 4.

Syntax Element Description:

Syntax Element Description
COMPUTE | ASSIGN [:]=
Usage of Keywords:

This statement may be issued in short form by omitting the statement keyword COMPUTE (or ASSIGN).

In structured mode, when the statement keyword COMPUTE (or ASSIGN) is omitted, the equal sign (=) must be preceded by a colon (:).

However, when the ROUNDED option is used, the statement keyword COMPUTE (or ASSIGN) must be specified.

ROUNDED
ROUNDED Option:

If you specify the keyword ROUNDED, the value will be rounded before it is assigned to operand1.

For information on rounding, see Rules for Arithmetic Assignments, Field Truncation and Field Rounding in the Programming Guide.

operand1
Result Field:

operand1 will contain the result of the arithmetic/assignment operation.

For the precision of the result, see Precision of Results of Arithmetic Operations in the Programming Guide.

If operand1 is a database field, the field in the database is not updated.

If operand1 is a dynamic variable, it is filled with exactly the data and length of operand2 or the length of the result of the arithmetic-operation, including trailing blanks. The current length of a dynamic variable can be obtained by using the system variable *LENGTH.

For general information on dynamic variables, see Using Dynamic and Large Variables.

arithmetic-expression
Arithmetic Expression:

An arithmetic expression consists of one or more constants, database fields, and user-defined variables.

Natural mathematical functions (described in the System Functions documentation) may also be used as arithmetic operands.

Operands used in an arithmetic expression must be defined with format N, P, I, F, D, or T.

As for the formats of the operands, see also Performance Considerations for Mixed Formats in the Programming Guide.

The following connecting operators may be used:

Operator: Symbol:
Parentheses ( )
Exponentiation **
Multiplication *
Division /
Addition +
Subtraction -

Each operator should be preceded and followed by at least one blank so as to avoid any conflict with a variable name that contains any of the above characters.

The processing order of arithmetic operations is:

  1. Parentheses

  2. Exponentiation

  3. Multiplication/division (left to right as detected)

  4. Addition/subtraction (left to right as detected)

operand2
Source Field:

operand2 is the source field. If operand1 is of format C, operand2 may also be specified as an attribute constant.

See User-Defined Constants in the Programming Guide.

SUBSTRING (operand2,operand3,operand4)
SUBSTRING Option:

Without the substring option, the whole content of operand2 is moved.

If operand1 and operand2 are of alphanumeric, Unicode or binary format, you may use the SUBSTRING option to assign a certain part of operand2 to operand1.

After the field name (operand2) in the SUBSTRING clause, you specify the starting position (operand3) and then the length (operand4) of the field portion to be moved.

For example, to assign the 3rd to 6th position of field #B to field #A, you would specify: 

#A := SUBSTRING(#B,3,4)

If you omit operand3, the starting position is assumed to be 1. If you omit operand4, the length is assumed to range from the starting position to the end of the field.

Note:
ASSIGN with the SUBSTRING option is a byte-by-byte assignment (that is, the rules described under Rules for Arithmetic Assignment in the Programming Guide do not apply).

See also MOVE SUBSTRING.

Result Precision of a Division

The precision (number of decimal positions) of the result of a division in a COMPUTE statement is determined by the precision of either the first operand (dividend) or the first result field, whichever is greater.

For a division of integer operands, however, the following applies: For a division of two integer constants, the precision of the result is determined by the precision of the first result field; however, if at least one of the two integer operands is a variable, the result is also of integer format (that is, without decimal positions, regardless of the precision of the result field).

COMPUTE Examples

Example 1 - ASSIGN Statement

** Example 'ASGEX1S': ASSIGN (structured mode)                          
************************************************************************
DEFINE DATA LOCAL                                                       
1 #A (N3)                                                               
1 #B (A6)                                                               
1 #C (N0.3)                                                             
1 #D (N0.5)                                                             
1 #E (N1.3)                                                             
1 #F (N5)                                                               
1 #G (A25)                                                              
1 #H (A3/1:3)                                                           
END-DEFINE                                                              
*                                                                       
ASSIGN #A = 5                           WRITE NOTITLE '=' #A            
ASSIGN #B = 'ABC'                       WRITE '=' #B                    
ASSIGN #C = .45                         WRITE '=' #C                    
ASSIGN #D = #E = -0.12345               WRITE '=' #D / '=' #E           
ASSIGN ROUNDED #F = 199.999             WRITE '=' #F                    
#G     := 'HELLO'                       WRITE '=' #G                    
#H (1) := 'UVW'                                           
#H (3) := 'XYZ'                         WRITE '=' #H (1:3)
*                                                         
END

Output of Program ASGEX1S:

#A:    5       
#B: ABC        
#C:  .450      
#D: -.12345    
#E: -0.123     
#F:    200     
#G: HELLO      
#H: UVW     XYZ

Equivalent reporting-mode example: ASGEX1R.

Example 2 - COMPUTE Statement

** Example 'CPTEX1': COMPUTE                                            
************************************************************************
DEFINE DATA LOCAL                                                       
1 EMPLOY-VIEW VIEW OF EMPLOYEES                                         
  2 PERSONNEL-ID                                                        
  2 SALARY    (1:2)                                                     
*                                                                       
1 #A          (P4)                                                      
1 #B          (N3.4)                                                    
1 #C          (N3.4)                                                    
1 #CUM-SALARY (P10)                                                     
1 #I          (P2)                                                      
END-DEFINE                                                              
*                                                                       
COMPUTE #A = 3 * 2 + 4 / 2 - 1
WRITE NOTITLE 'COMPUTE #A = 3 * 2 + 4 / 2 - 1' 10X '=' #A               
*                                                                       
COMPUTE ROUNDED #B = 3 -4 / 2 * .89
WRITE 'COMPUTE ROUNDED #B = 3 -4 / 2 * .89' 5X '=' #B                   
*                                                        
COMPUTE #C = SQRT (#B)
WRITE 'COMPUTE #C = SQRT (#B)' 18X '=' #C                
*                                                        
LIMIT 1                                                  
READ EMPLOY-VIEW BY PERSONNEL-ID STARTING FROM '20017000'
  WRITE / 'CURRENT SALARY: '  4X SALARY (1)              
        / 'PREVIOUS SALARY:'  4X SALARY (2)              
  FOR #I = 1 TO 2                                        
    COMPUTE #CUM-SALARY = #CUM-SALARY + SALARY (#I)      
  END-FOR                                                
  WRITE 'CUMULATIVE SALARY:' #CUM-SALARY                 
END-READ                                                 
*                                                        
END

Output of Program CPTEX1:

COMPUTE #A = 3 * 2 + 4 / 2 - 1          #A:     7    
COMPUTE ROUNDED #B = 3 -4 / 2 * .89     #B:    1.2200
COMPUTE #C = SQRT (#B)                  #C:    1.1045
                                                     
CURRENT SALARY:          34000                       
PREVIOUS SALARY:         32300                       
CUMULATIVE SALARY:       66300