Sun, 17 Aug 2014 15:07:00 -0700Fixed 2680 - OSX: Replace NSAutoreleasePool with @autoreleasepool
Sam Lantinga <slouken@libsdl.org> [Sun, 17 Aug 2014 15:07:00 -0700] rev 9087
Fixed 2680 - OSX: Replace NSAutoreleasePool with @autoreleasepool

Tim McDaniel

This patch replaces all use of NSAutoreleasePool with the Apple recommended @autoreleasepool. @autoreleasepool is supposedly more efficient, and since it is scope based it can't be accidentally not released.

Sun, 17 Aug 2014 14:57:52 -0700Fixed bug 2655 - OSX: Window position and global mouse coord spaces are different
Sam Lantinga <slouken@libsdl.org> [Sun, 17 Aug 2014 14:57:52 -0700] rev 9086
Fixed bug 2655 - OSX: Window position and global mouse coord spaces are different

Tim McDaniel

On OSX, with revision 8729, the coordinate space for window position and the coordinate space for global mouse position don't match. For a non-fullscreen window, the window position is global relative to the bottom of the menubar. The global mouse position is relative to the top of the screen. This affects Cocoa_WarpMouse and potentially other things as well. Further, the coordinate system for window position is now affected by what screen it is on. For example, if I have two equal size screens oriented side by side such that the tops of the screens are equal in global space, with the menubar on one screen, and a window straddles the two screens, the window's y position makes no sense. The window's y position depends on what screen "most" of the window is on. So if I move the window horizontally just a bit, the y position of my window is now different by the size of the menubar, even though the window was not moved vertically.

I'd like to reiterate that this was a fairly fundamental change (and a breaking change for us). If SDL OSX is to really support multi-display configurations, this is especially problematic.

If the real concern is preventing windows from going under the menubar, then perhaps a solution involving something like overriding [NSWindow constrainFrameRect] would be less problematic than redefining the global window coord space for the main display.

Sun, 17 Aug 2014 14:44:53 -0700Fixed bug 2685 - SDL_RenderReadPixels() doesn't work with offscreen targets
Sam Lantinga <slouken@libsdl.org> [Sun, 17 Aug 2014 14:44:53 -0700] rev 9085
Fixed bug 2685 - SDL_RenderReadPixels() doesn't work with offscreen targets

Andreas Falkenhahn

SDL_RenderReadPixels() doesn't seem to work when trying to read pixels from a texture that has been created using SDL_TEXTUREACCESS_TARGET and has been selected as the render target using SDL_SetRenderTarget().

I am attaching a small program that demonstrates the issue. I get the following result here:

READ PIXEL RETURN: 0 --- COLOR CHECK: ff000000

But it should be:

READ PIXEL RETURN: 0 --- COLOR CHECK: ffff0000

Tested with SDL 2.0.3 on Windows 7.

Sun, 17 Aug 2014 14:34:41 -0700Fixed bug where the render target is updated instead of the default output when the window is resized.
Sam Lantinga <slouken@libsdl.org> [Sun, 17 Aug 2014 14:34:41 -0700] rev 9084
Fixed bug where the render target is updated instead of the default output when the window is resized.

Sun, 17 Aug 2014 13:49:53 -0700Reset CMAKE_REQUIRED_FLAGS after test
Sam Lantinga <slouken@libsdl.org> [Sun, 17 Aug 2014 13:49:53 -0700] rev 9083
Reset CMAKE_REQUIRED_FLAGS after test

Sun, 17 Aug 2014 13:15:45 -0700cmake: add -Wl,--no-undefined to GCC linker flags
Sam Lantinga <slouken@libsdl.org> [Sun, 17 Aug 2014 13:15:45 -0700] rev 9082
cmake: add -Wl,--no-undefined to GCC linker flags

This way unresolved symbols will be detected when linking the shared
library version.

Sun, 17 Aug 2014 13:15:09 -0700cmake: add -Wall/-Wshadow to GCC compilation flags
Sam Lantinga <slouken@libsdl.org> [Sun, 17 Aug 2014 13:15:09 -0700] rev 9081
cmake: add -Wall/-Wshadow to GCC compilation flags

Sun, 17 Aug 2014 13:11:55 -0700Removed SDL_round() because the license wasn't compatible with zlib
Sam Lantinga <slouken@libsdl.org> [Sun, 17 Aug 2014 13:11:55 -0700] rev 9080
Removed SDL_round() because the license wasn't compatible with zlib

Sun, 17 Aug 2014 10:10:41 -0700Fixed bug 2688 - failure to build test/loopwavequeue.c on Linux
Sam Lantinga <slouken@libsdl.org> [Sun, 17 Aug 2014 10:10:41 -0700] rev 9079
Fixed bug 2688 - failure to build test/loopwavequeue.c on Linux

Sat, 16 Aug 2014 23:30:44 -0700Fixed bug 2681 - dereference a NULL pointer dst_fmt in SDL_CreateTextureFromSurface function
Sam Lantinga <slouken@libsdl.org> [Sat, 16 Aug 2014 23:30:44 -0700] rev 9078
Fixed bug 2681 - dereference a NULL pointer dst_fmt in SDL_CreateTextureFromSurface function

Nitz

In SDL_CreateTextureFromSurface:

SDL_PixelFormat *dst_fmt;
/* Set up a destination surface for the texture update */
dst_fmt = SDL_AllocFormat(format);
temp = SDL_ConvertSurface(surface, dst_fmt, 0);

Here is need of NULL check for dst_fmt because there are chances of NULL return from SDL_AllocFormat(format);