SAS/C Library Reference, Vol. 1

pdval -- Packed Decimal Conversion: Packed Decimal to Double

SYNOPSIS

 #include <packed.h>

 double pdval(const char (*val)[], unsigned int scale);

DESCRIPTION

pdval converts val, a number in packed-decimal format, to its floating-point representation. The maximum length of val is 8 bytes (15 decimal digits). After conversion, val is multiplied by pow(10.0, -scale). scale, which specifies a scaling factor, must be a nonnegative integer less than or equal to 15.

RETURN VALUE

The return value is the double-precision, floating-point representation of val, appropriately scaled.

ERRORS

If val does not contain valid packed-decimal data, an 0C7 ABEND results.

DIAGNOSTICS

If scale is not positive and less than or equal to 15, then pdval returns HUGE_VAL and sets errno to EARG.

IMPLEMENTATION

pdval is defined in <packed.h> as

 #define pdval(val, scale) _pdval(val, sizeof(*val), scale)

EXAMPLE

  #include <packed.h>
  #include <stdio.h>

  void printamt(char (*amount)[6])

     /* expected COBOL data declaration: */
     /* AMOUNT PIC 9(9)V99 COMP-3.       */
  {
     double dollars;

        /* Convert to dollars and cents. */
     dollars = pdval(amount, 2);
     printf("Amount is $ % 12.2fn", dollars);
     return;
  }

RELATED FUNCTIONS

pdset