Re: [libertyeiffel] Constraint genericity and REAL_GENERAL
- Cyril Adrian
- 2010-08-01 @ 10:48
It is a SmartEiffel-specific question. You should ask on their mailing list.
If you use Liberty's libraries you can try
expanded class MY_CLASS[T -> FLOAT]
On Sun, Aug 1, 2010 at 09:37, Christophe HARO <email@example.com>wrote:
> I do not understand why it is not legal to declare :
> expanded class MY_CLASS[T -> REAL_GENERAL]
> but it is legal to write :
> expanded class MY_CLASS[T -> NUMERIC]
> and I do not understand the message : << Actually this is only a compiler
> implementation facility. >> that the compiler outputs in the first case :
> compile -debug test_math_macros_real.e -o test_math_macros_real.exe
> ****** Fatal Error: The REAL_GENERAL type cannot be used here.
> Actually this is only a compiler implementation facility.
> Line 17 column 22 in MATH_MACROS
> MATH_MACROS[T -> REAL_GENERAL]
> Error occurs while compiling. Compilation process aborted.
> Thanks in advance.
> Christophe HARO