On Tue, Oct 6, 2009 at 4:03 PM, Robert Bradshaw
<rober...@math.washington.edu> wrote:
>
> c99_complex is an internal directive (and I (intentionally) didn't
> add it to the list of directives because I wasn't sure if that's the
> final interface we wanted), so I think we're safe to change it. I'm
> starting to wonder if we even need it at all.
I think we need it... you know how much I hate second guessing... I've
never liked that of c99 complex being on depending of "complex.h"
being included... Moreover, going C++, I do not know how to detect if
<complex> was #include's ...
> Rather, we could use
> struct-based complexes by default, but if complex.h is included
> (detected by looking for the specific macros it defines) either c99
> or C++ complexes would be used.
I have an alternative proposal: Let's call the directive 'ccomplex'
(in the spirit of cdivision, which applies both to C and C++)...
Cython generates code to support struct-based complexes, C99 _Complex
in C, and std::complex<> in C++... there is a C macro PYX_USE_CCOMPLEX
that let choose the struct based implementation or the C99/C++
(depending on defined(__cplusplus))...
Then the ONLY effect of the directive 'ccomplex=True' would be to make
the macro PYX_USE_CCOMPLEX be 1 be default... BUT if at C compile time
PYX_USE_CCOMPLEX is set to 0, then the struct-based impl is used....
Similarly, if 'ccomplex=False' (the default), then PYX_USE_CCOMPLEX
defaults to 0, BUT again if at C compile time PYX_USE_CCOMPLEX is
defined to 1, then the C99 or C++ versions are used depending on
__cplusplus...
What do you think? I really like this..
> All arithmetic (including getting/
> setting the real/imag parts) would be done via macros.
>
Of course. Though I anticipate problems the setter/getter for
real/imag ... I do think we should have {SET|GET}_{REAL|IMAG} part, if
not we will not be able to support C++0x ... Also, currently the C99
support is using __real__/__imag__
> This is essentially the current default behavior, except we'd add
> support for C++ complexes in place of c99 complexes if, at C compile
> time, the language is determined to be C++.
>
--
Lisandro Dalcín
---------------
Centro Internacional de Métodos Computacionales en Ingeniería (CIMEC)
Instituto de Desarrollo Tecnológico para la Industria Química (INTEC)
Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET)
PTLC - Güemes 3450, (3000) Santa Fe, Argentina
Tel/Fax: +54-(0)342-451.1594
_______________________________________________
Cython-dev mailing list
Cython-dev@codespeak.net
http://codespeak.net/mailman/listinfo/cython-dev
