joe@petsd.UUCP (Joe Orost) (08/13/87)
We had a customer enter the following to our (validated) compiler:
function F(ARG : INTEGER) return INTEGER is
begin
return INTEGER(ARG * 33.3);
end F;
He was wondering why he was getting NUMERIC_ERROR whenever ARG > 3936. I
was wondering why his "FORTRAN-LIKE" code compiled at all!!! What he meant
to say was INTEGER(ARG * FLOAT(33.3)).
So then I looked up the implicit conversion rules (4.6 (15)), and it turns
out that this is legal Ada code, and the type of the constant is "DURATION",
as long as no other fixed point types are declared. If other fixed point
types are declared, then this expression is in error:
"An implicit conversion of a convertable universal operand is
applied if and only if the innermost complete context (see 8.7)
determines a unique (numeric) target type for the implicit
conversion, and there is no legal interpretation of this
context without this conversion."
This, along with:
function "*" (LEFT : INTEGER; RIGHT: any_fixed_point_type) return
any_fixed_point_type;
along with:
type DURATION is ...;
causes the compiler to interpret "33.3" as type DURATION !!!!!
The NUMERIC_ERROR came about because DURATION'LAST / 33.3 = 3936.096
And I thought I knew the Ada language!
regards,
joe
--
Full-Name: Joseph M. Orost
UUCP: ihnp4!vax135!petsd!joe
ARPA: petsd!joe@RUTGERS.EDU
Phone: (201) 758-7284
US Mail: MS 313; Concurrent Computer Corporation; 106 Apple St
Tinton Falls, NJ 07724