On Wednesday, 7 February 2018 at 20:10:10 UTC, dekevin wrote:
On Wednesday, 7 February 2018 at 16:26:16 UTC, Dominikus Dittes
Scherkl wrote:
On Wednesday, 7 February 2018 at 12:10:38 UTC, dekevin wrote:
struct ℚ{
ℤ num, den; //cannot call constructors on these, since
they require gmp_init, which requires runtime code
//Default initialiser disabled, since else num=0,den=0
You can use a different default initializer:
ℤ num = 0, den = 1;
Thus avoiding the bad denominator.
I appreciate the help, but I cannot instantiate these variables
with a constructor.
Whenever I do, the compiler tries to figure out its values at
compile-time.
This it cannot do however, since the code requires gmp_init for
constructing an explicit value, which is only available after
dynamically linking GMP.
Here is the error I get:
source/gmp/z.d(117): Error: __gmpz_init_set_si cannot be
interpreted at compile time, because it has no available source
code
source/util.d(575): called from here:
_MpZ(__mpz_struct(0, 0, null)).this(0)
source/gmp/z.d(117): Error: __gmpz_init_set_si cannot be
interpreted at compile time, because it has no available source
code
source/util.d(575): called from here:
_MpZ(__mpz_struct(0, 0, null)).this(1)
I don't really know what to do now, since I have another struct
which already uses static variables with the same naming
convention and I'll also need it a second time when I write the
bindings to MPFR.
I should perhaps add how ℤ looks like.
Here is a stripped-down version:
struct ℤ {
@disable this();
this(T)(T value) @trusted
if (isArithmetic!T)
{
version(ccc) { ++_ccc; }
static if (isUnsigned!T)
__gmpz_init_set_ui(_ptr, value);
else static if (isFloating!T)
__gmpz_init_set_d(_ptr, value);
else // isSigned integral
__gmpz_init_set_si(_ptr, value);
}
~this() @trusted
{
assert(_ptr, "Pointer is null");
//if(hasBeenInit)
__gmpz_clear(_ptr); version(ccc) { ++_ccc; }
}
inout(__mpz_struct)* _ptr() inout return @system
{
return &_z;
}
}
