pinhole.hpp compiler error with std::max

[rgolovanov]

Describe the bug
Compiler error pointing to std::max function in pinhole.hpp

const float focal_length_y = 0.5f / std::tan(std::max(fov_y * 0.5f, EPSILON));

Compiler errors:

[build] XXX\rerun\archetypes/pinhole.hpp(156,63): error C2589: '(': illegal token on right side of '::' (compiling source file YYY.cpp) [ZZZ.vcxproj]
[build] XXX\rerun\archetypes/pinhole.hpp(156): error C2144: syntax error: 'unknown-type' should be preceded by ')' (compiling source file YYY.cpp) [ZZZ.vcxproj]
[build] XXX\rerun\archetypes/pinhole.hpp(156,1): error C2661: 'tan': no overloaded function takes 0 arguments (compiling source file YYY.cpp) [ZZZ.vcxproj]
[build] XXX\rerun\archetypes/pinhole.hpp(156,25): error C2737: 'focal_length_y': const object must be initialized (compiling source file YYY.cpp) [ZZZ.vcxproj]
[build] XXX\rerun\archetypes/pinhole.hpp(156): error C2144: syntax error: 'unknown-type' should be preceded by ';' (compiling source file YYY.cpp) [ZZZ.vcxproj]
[build] XXX\rerun\archetypes/pinhole.hpp(156,63): error C2143: syntax error: missing ')' before '*' (compiling source file YYY.cpp) [ZZZ.vcxproj]
[build] XXX\rerun\archetypes/pinhole.hpp(156,63): error C2143: syntax error: missing ';' before '*' (compiling source file YYY.cpp) [ZZZ.vcxproj]
[build] XXX\rerun\archetypes/pinhole.hpp(156,1): error C2082: redefinition of formal parameter 'fov_y' (compiling source file YYY.cpp) [ZZZ.vcxproj]
[build] XXX\rerun\archetypes/pinhole.hpp(156,1): error C2059: syntax error: ')' (compiling source file YYY.cpp) [ZZZ.vcxproj]
[build] XXX\rerun\archetypes/pinhole.hpp(156,63): error C2100: illegal indirection (compiling source file YYY.cpp) [ZZZ.vcxproj]
[build] Build finished with exit code 1

To Reproduce
Steps to reproduce the behavior:

1. Compile C++ project linking Rerun

Expected behavior
No compiler errors

Desktop (please complete the following information):

• OS: Windows, VS 2019

Rerun version
0.14.1

Additional context
Workaround is manually patch this line by adding braces:

const float focal_length_y = 0.5f / std::tan((std::max)(fov_y * 0.5f, EPSILON));

[Wumpf]

The problem is that you include windows.h before including rerun. Windows has macros for min and max, imho the better solution is to add #define NOMINMAX before including windows.h.
Putting (std::max) in parenthesis is an interesting solution I haven't thought of before, we can add that in!

[rgolovanov]

@Wumpf you are right. It is coming from windows.h and NOMINMAX resolves the issue.

[nasteyi]

explicitly passing a template parameter can also fix compilation
const float focal_length_y = 0.5f / std::tan(std::max<float>(fov_y * 0.5f, EPSILON));

[Wumpf]

went with @nasteyi suggestion and added a test to validate rerun.hpp on our ci for this. Fixed in next release then!