On Wed, Oct 7, 2009 at 6:59 PM, Robert Bradshaw <rober...@math.washington.edu> wrote: > > Don't reopen the ticket, make a new one for C++ complex support. >
Done: http://trac.cython.org/cython_trac/ticket/398 >> For the case of C, C99 complexes will be automatically used ONLY if >> _Complex_I is defined... of if the user explicitly pass >> "-DPYX_CCOMPLEX=1" at C-compile time (or a Cython-included C header >> #define PYX_CCOMPLEX=1), or if the directive "ccomplex=True". >> >> I'm still in doubt about how to handle 'ccomplex=True'... Should it >> just enable by default using "#define PYX_CCOMPLEX=1", or actually >> impact in code generation (as currently does)... > > I would be OK with it just changing the flag, with the code > generation remaining unchanged. > OK, this is work TBD in my patch... >> Ah! I forgot... I do not know how to implement z.conjugate(); the >> C/C99/C++ support is there though... BTW, GCC seems to accept "~z" for >> complex conjugation (no idea if it is C99-standard or a >> GCC-extension)... Perhaps Cython should accept this (non-Python) >> syntax? I think it is nice. > > I know is sage we map ~z to 1/z (it is the __invert__ operator after > all). Python doesn't define an __invert__ for their native complex > type. I could go either way. > IMHO, my brain can parse "~z" as z.conjugate(), but would have a hard time realizing ~z is "1/z"... Anyway, you are the BDLF, it's up to you... Or the champion who actually implement support for z.conjugate()... Hey, Dag. Are you there? ;-) -- 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
