crossplattform: enable math constants on all plattforms

[TheJJ]

This introduces namespaced math constants available on all plattforms.

closes #208.

[franciscod]

boooo 👎

In file included from C:/msys/home/Windows/openage/cpp/unit/unit.cpp:8:0:
C:/msys/home/Windows/openage/cpp/unit/../crossplatform/math_constants.h:13:33: e                                                      rror: 'M_E' was not declared in this scope
 constexpr double E            = M_E;        //!< e
                                 ^
C:/msys/home/Windows/openage/cpp/unit/../crossplatform/math_constants.h:14:33: e                                                      rror: 'M_LOG2E' was not declared in this scope
 constexpr double LOG2E        = M_LOG2E;    //!< log_2 e
                                 ^
C:/msys/home/Windows/openage/cpp/unit/../crossplatform/math_constants.h:15:33: e                                                      rror: 'M_LOG10E' was not declared in this scope
 constexpr double LOG10E       = M_LOG10E;   //!< log_10 e
                                 ^
C:/msys/home/Windows/openage/cpp/unit/../crossplatform/math_constants.h:16:33: e                                                      rror: 'M_LN2' was not declared in this scope
 constexpr double LN2          = M_LN2;      //!< log_e 2
                                 ^
C:/msys/home/Windows/openage/cpp/unit/../crossplatform/math_constants.h:17:33: e                                                      rror: 'M_LN10' was not declared in this scope
 constexpr double LN10         = M_LN10;     //!< log_e 10
                                 ^
C:/msys/home/Windows/openage/cpp/unit/../crossplatform/math_constants.h:18:33: e                                                      rror: 'M_PI' was not declared in this scope
 constexpr double PI           = M_PI;       //!< pi
                                 ^
C:/msys/home/Windows/openage/cpp/unit/../crossplatform/math_constants.h:19:33: e                                                      rror: 'M_PI_2' was not declared in this scope
 constexpr double PI_2         = M_PI_2;     //!< pi/2
                                 ^
C:/msys/home/Windows/openage/cpp/unit/../crossplatform/math_constants.h:20:33: e                                                      rror: 'M_PI_4' was not declared in this scope
 constexpr double PI_4         = M_PI_4;     //!< pi/4
                                 ^
C:/msys/home/Windows/openage/cpp/unit/../crossplatform/math_constants.h:21:33: e                                                      rror: 'M_1_PI' was not declared in this scope
 constexpr double INV_PI       = M_1_PI;     //!< 1/pi
                                 ^
C:/msys/home/Windows/openage/cpp/unit/../crossplatform/math_constants.h:22:33: e                                                      rror: 'M_2_PI' was not declared in this scope
 constexpr double INV2_PI      = M_2_PI;     //!< 2/pi
                                 ^
C:/msys/home/Windows/openage/cpp/unit/../crossplatform/math_constants.h:23:33: e                                                      rror: 'M_2_SQRTPI' was not declared in this scope
 constexpr double INV2_SQRT_PI = M_2_SQRTPI; //!< 2/sqrt(pi)
                                 ^
C:/msys/home/Windows/openage/cpp/unit/../crossplatform/math_constants.h:24:33: e                                                      rror: 'M_SQRT2' was not declared in this scope
 constexpr double SQRT_2       = M_SQRT2;    //!< sqrt(2)
                                 ^
C:/msys/home/Windows/openage/cpp/unit/../crossplatform/math_constants.h:25:33: e                                                      rror: 'M_SQRT1_2' was not declared in this scope
 constexpr double INV_SQRT_2   = M_SQRT1_2;  //!< 1/sqrt(2)
                                 ^
C:/msys/home/Windows/openage/cpp/unit/unit.cpp: In function 'unsigned int openag                                                      e::dir_group(openage::coord::phys3_delta, unsigned int, unsigned int)':
C:/msys/home/Windows/openage/cpp/unit/unit.cpp:160:1: warning: control reaches e                                                      nd of non-void function [-Wreturn-type]
 }
 ^
cpp/CMakeFiles/openage.dir/build.make:1388: recipe for target 'cpp/CMakeFiles/op                                                      enage.dir/unit/unit.cpp.obj' failed
make[2]: *** [cpp/CMakeFiles/openage.dir/unit/unit.cpp.obj] Error 1
CMakeFiles/Makefile2:1108: recipe for target 'cpp/CMakeFiles/openage.dir/all' fa                                                      iled
make[1]: *** [cpp/CMakeFiles/openage.dir/all] Error 2
Makefile:127: recipe for target 'all' failed
make: *** [all] Error 2

[franciscod]

see the previous comment. I think it breaks (on msys) because cmath gets imported earlier in the build process without the _USE_MATH_DEFINES macro defined.

however, on linux it builds fine, lol

i'm poor man's travis

[TheJJ]

Hmm, but then i see no other solution than defining the constants manually to the actual numbers.

We can't manipulate the inclusion order and devs will want to include <cmath> without thinking about breaking the constant definitions..

[franciscod]

These guys were defining that macro on CMakeLists, but since then migrated too boost:

ddemidov/vexcl@696d74a

---

The math library normally defines M_PI to a double approximation of pi. If strict ISO and/or POSIX >compliance are requested this constant is not defined, but you can easily define it yourself:

#define M_PI 3.14159265358979323846264338327
You can also compute the value of pi with the expression acos (-1.0).

http://www.sbin.org/doc/glibc/libc_19.html

---

My vote goes to hardcode the constants without relying on the cmath behaviour.

[franciscod]

TheJJ#1

[franciscod]

typo on crossplattform!

[TheJJ]

oh lol 👊

but the idea of pullrequesting at forks is pretty good actually

[franciscod]

builds fine on msys2 now, 👍

[franciscod]

half typos remaining!

crossplatform: namespaced math constants on all plattforms
        ^                                          ^
        |                                          |-NOT OK
        |-OK                                          

not a big issue... you can decide how nazi is our grammar policy :D

[andrekupka]

👍