Move constructor and move assignment is now noexcept and it just swaps
the data of the two instances instead of calling GL API, thus it can be
inline. Also removed unneded limitations in BufferTexture.
The point is that we accept everything except const BufferImage&, as the
buffer needs to be bound, which is not an constant operation, as it
modifies internal GL state tracker.
Move constructor and move assignment now behaves similarly to Image
(not only the buffer but also size is swapped). Added constructor taking
size + data, reordered setData() parameters to match order in Image. The
old setData() function is now alias to the new one, is marked as
deprecated and will be removed in future release.
Before this commit the default context creation in tryCreateContext
would create the OpenGL 2.1 context.After this commit by default,
the OpenGL 3.2 context is created by default, if this fails
tryCreateContext will fall back to creating a OpenGL 2.1 context.
Recommended fix:
https://github.com/mosra/magnum/blob/master/modules/FindSDL2.cmake#L46 to list only SDL_scancode.h (i.e. remove SDL.h)
Test:
Add SDL2 Lib to the /Libraries/Frameworks/
Builds file.
Add SDL2 Lib and SDL1 to the /Libraries/Frameworks/
Build fails.
Remove SDL.h from at line 46.
Builds fine.
Remove SDL 1 from the /Libraries/Frameworks/
Builds fine.
Previously this was done in DebugMarker only via GREMEDY_string_marker
on desktop and EXT_debug_marker on ES, now supporting both three on
desktop and KHR/EXT version on ES. The old DebugMarker is now only a
thin wrapper around DebugMessage, is marked as deprecated and will be
removed in future release.
Can't test EXT_debug_label, as that is apparently OSX 10.9-only. Added
GL tests for all implemented objects. KHR_debug is selected first, if
that is not available, fall back to EXT_debug_label. If neither is
available, the functions are no-op.
I hope EXT_debug_label gets replaced by KHR_debug later, thus it is now
only "emulated" through KHR_debug enums.
In NaCl the wheel event is something completely different than mouse
event and in my opinion overly complicated (the scrolled distance is
measured in pixel precision!).
To preserve at least some compatibility with other toolkits, the events
are shoved into normal MouseEvent. Sadly the event doesn't contain any
position information.