void mi_fp_setretprec(fparam_ptr, ret_pos, ret_prec)
   MI_FPARAM *fparam_ptr;
   mi_integer ret_pos;
   mi_integer ret_prec;

fparam_ptr

is a pointer to the associated MI_FPARAM structure.

ret_pos

is the index position into the return-precision array for the return value whose precision you want to set. For C user-defined functions, the only valid value is 0.

ret_prec

is the precision, in number of digits, to set for the ret_pos+1 return value.

Valid in Client LIBMI Application?	Valid in User-Defined Routine?
Yes	Yes

The mi_fp_setretprec( ) function sets the precision of the function return value at position ret_pos in the MI_FPARAM structure that fparam_ptr references.

The precision is an attribute of the data type that represents the total number of digits the routine return value can hold, as follows.

Data Type

Meaning

DECIMAL,
MONEY

Number of significant digits in the fixed-point or floating-point (DECIMAL) column

DATETIME,
INTERVAL

Number of digits that are stored in the date and/or time column with the specified qualifier

Character,
Varying-character

Maximum number of characters in the column

The MI_FPARAM structure stores information about the precision of function return values in the zero-based return-precision array. To set information for the nth return value, use a ret_pos value of n-1.

For example, the following call to mi_fp_setretprec( ) sets the precision of 10 for the first return value of the my_func( ) UDR, with which fparam1 is associated:

mi_integer my_func(..., fparam1)
   ...
   MI_FPARAM *fparam1;
{
   ...
   mi_fp_setretprec(fparam1, 0, 10);

None.

See also the descriptions of mi_fp_argprec( ), mi_fp_setargprec( ), mi_fp_setretlen( ), mi_fp_setretscale( ), mi_fp_setrettype( ), and mi_fp_setreturnisnull( ).

For more information about return-value information in an MI_FPARAM structure or about the precision of a fixed-point data type, see the IBM Informix: DataBlade API Programmer's Guide.