Fri, 05 Jul 2013 00:41:01 -0400Added a FIXME.
Ryan C. Gordon <icculus@icculus.org> [Fri, 05 Jul 2013 00:41:01 -0400] rev 7343
Added a FIXME.

Fri, 05 Jul 2013 00:30:23 -0400Don't close already-closed audio devices during SDL_Quit().
Ryan C. Gordon <icculus@icculus.org> [Fri, 05 Jul 2013 00:30:23 -0400] rev 7342
Don't close already-closed audio devices during SDL_Quit().

Otherwise, we spam a bunch of unnecessary SDL_SetError() calls.

Fixes Bugzilla #1791.

Wed, 03 Jul 2013 12:31:40 -0700Added SDL_HINT_JOYSTICK_ALLOW_BACKGROUND_EVENTS to the documentation for the joystick and game controller APIs
Sam Lantinga <slouken@libsdl.org> [Wed, 03 Jul 2013 12:31:40 -0700] rev 7341
Added SDL_HINT_JOYSTICK_ALLOW_BACKGROUND_EVENTS to the documentation for the joystick and game controller APIs

Tue, 04 Jun 2013 12:02:18 -0400add in alt+enter keyboard shortcut to test common to toggle fullscreen desktop
Edward Rudd <urkle@outoforder.cc> [Tue, 04 Jun 2013 12:02:18 -0400] rev 7340
add in alt+enter keyboard shortcut to test common to toggle fullscreen desktop

Sat, 29 Jun 2013 14:40:55 -0700Fixed the logical size for rendering to texture, thanks to Mason Wheeler.
Sam Lantinga <slouken@libsdl.org> [Sat, 29 Jun 2013 14:40:55 -0700] rev 7339
Fixed the logical size for rendering to texture, thanks to Mason Wheeler.

Mason Wheeler

The SDL_RenderGetLogicalSize function should always return the amount of pixels that are currently available for rendering to. But after updating to the latest SDL2, I started getting crashes because it was returning (0,0) as the logical size! After a bit of debugging, I tracked it down to the following code in SDL_SetRenderTarget:

if (texture) {
renderer->viewport.x = 0;
renderer->viewport.y = 0;
renderer->viewport.w = texture->w;
renderer->viewport.h = texture->h;
renderer->scale.x = 1.0f;
renderer->scale.y = 1.0f;
renderer->logical_w = 0;
renderer->logical_h = 0;
}

This is obviously wrong; 0 is never the correct value for a valid renderer. Those last two lines should read:

renderer->logical_w = texture->w;
renderer->logical_h = texture->h;

Sat, 29 Jun 2013 22:08:38 +0200Changed paths and name of header file in text of VisualC.html.
Philipp Wiesemann <philipp.wiesemann@arcor.de> [Sat, 29 Jun 2013 22:08:38 +0200] rev 7338
Changed paths and name of header file in text of VisualC.html.

Fri, 28 Jun 2013 23:29:13 -0700Only check for keyboard focus if the video subsystem was initialized.
Sam Lantinga <slouken@libsdl.org> [Fri, 28 Jun 2013 23:29:13 -0700] rev 7337
Only check for keyboard focus if the video subsystem was initialized.
Check the hint at initialization time, as an optimization. This isn't something we expect the application to change at runtime, and if it is we should add an API for it.

Fri, 28 Jun 2013 22:49:03 -0700Removing old dangerous signal catching code.
Sam Lantinga <slouken@libsdl.org> [Fri, 28 Jun 2013 22:49:03 -0700] rev 7336
Removing old dangerous signal catching code.
* None of the supported platforms are hosed if the program crashes without cleaning up the graphics display connection.
* It's very likely to be unsafe to call SDL_Quit() at an arbitrary point from an arbitrary thread.
* Catching signals can prevent applications from getting correct stack traces from crashes.

Fri, 28 Jun 2013 22:44:49 -0700Not using SDLmain on Windows is just fine, don't fail the initialization if the program implements WinMain() and then doesn't call SDL_SetMainReady().
Sam Lantinga <slouken@libsdl.org> [Fri, 28 Jun 2013 22:44:49 -0700] rev 7335
Not using SDLmain on Windows is just fine, don't fail the initialization if the program implements WinMain() and then doesn't call SDL_SetMainReady().
This broke Steam's Big Picture game controller support.

I also added some more documentation so people know why main() is overridden on various platforms.

Fri, 28 Jun 2013 22:42:10 -0700Added testgamecontroller to the Visual Studio 2012 solution.
Sam Lantinga <slouken@libsdl.org> [Fri, 28 Jun 2013 22:42:10 -0700] rev 7334
Added testgamecontroller to the Visual Studio 2012 solution.