Procedure Modules › Converting Floating Point Values to Packed Decimal(CULLUS36) › Helpful Hints

Helpful Hints

• The COBOL equivalents for floating point values and the corresponding field lengths are:

• The number of decimal places specified for output (decimal-place-vn) should agree with the DP= specification of the work field that receives the converted value (result-v).
• If the result overflows the space allocated:
  • The rightmost digits are truncated to the right of the decimal point.
  • The leftmost digits are truncated to the left of the decimal point.

Example

This example shows the CA Culprit parameters required to convert single- and double-precision values to packed decimal format.

The following code:

• Defines the single- (FIELD1) and double- (FIELD2) precision fields on REC parameters
• Defines work fields for the result fields (PACKED-SINGLE and PACKED-DOUBLE) and an error code field for each conversion (CODE-1 and CODE-2)
• Issues separate calls to CULLUS36 to convert the single-precision field and then the double-precision field
• Tests for successful conversion after each call. If the conversion is not successful, an error handling routine executes.

   INPUT 80 F 2960
   REC FIELD1 1 4
   REC FIELD2 5 8
   990 PACKED-SINGLE DP=2
   990 PACKED-DOUBLE DP=3
   990 CODE-1 '1234'
   990 CODE-2 '    '
             .
             .
             .
   997010 CALL US36 (FIELD1 'S' PACKED-SINGLE 2 CODE-1) 
   997015 CODE-1 NE ' ' 200
   997020 CALL US36 (FIELD2 'D' PACKED-DOUBLE 3 CODE-2) 
   997025 CODE-2 NE ' ' 200
   997100 TAKE
   997200 $ERROR HANDLING ROUTINE IS CODED HERE