Commit Graph

6588 Commits

Author SHA1 Message Date
Pierre-Loup A. Griffais e0771482f6 Fix regression causing crashes when running without a window manager. 2013-10-23 13:43:17 -07:00
Ryan C. Gordon 965a0da901 Moved a SDL_SetError() call elsewhere to avoid triggering it needlessly.
Otherwise, the XInput path would always trigger it in a harmless manner.
2013-10-23 15:54:12 -04:00
Ryan C. Gordon 28aa076c8b SDL_GetPrefPath() now uses the organization on all platforms.
Even if that's not the general convention for a given platform.
2013-10-23 00:58:20 -04:00
Sam Lantinga 298d3d627a Fixed duplicate const warning 2013-10-22 21:54:52 -07:00
Sam Lantinga 30ce2bc723 SDL 2.0.0 defined __inline__ and some code relies on it now. 2013-10-22 21:53:58 -07:00
Sam Lantinga fa73a3c8f6 Testing fullscreen window in the accelerometer test. 2013-10-22 21:53:42 -07:00
Andreas Schiffler 09b3b7e78b Fix error assert for negative cases of pixels_getPixelFormatName test 2013-10-21 23:17:37 -07:00
Sam Lantinga 6a71d99233 Added patch notes for 2.0.1 2013-10-21 22:27:07 -07:00
Sam Lantinga 2348e426c7 Fixed whitespace 2013-10-21 22:08:56 -07:00
Sam Lantinga 3d1c3b19c3 Revert fix for bug 1392 because it breaks linking with mingw32 2013-10-21 10:28:57 -07:00
Sam Lantinga 06811ecf18 Fixed bug 2168 - Mac: SDL_filesystem.h not exported to public header folder when building framework
Alex Szpakowski

I actually made a mistake when creating the previous patch file... I forgot to include a crucial line which changed.

I've attached a new patch which just changes the line I forgot, since the other part has already been applied to the repository.
I tested this one by doing a clean rebuild of SDL, and it works with the new patch.
2013-10-21 02:46:16 -07:00
Sam Lantinga 50da48eebf We don't have API changes documented yet. 2013-10-21 02:38:23 -07:00
Sam Lantinga bcf5472d16 Fixed bug 2073 - Mac: window moves unexpectedly when exiting SDL_WINDOW_FULLSCREEN_DESKTOP mode
Alex Szpakowski

In Mac OS X, when SDL_SetWindowFullscreen(window, 0) is called on a window which was in SDL_WINDOW_FULLSCREEN_DESKTOP mode, its original size is restored but its position is moved to the bottom of the screen.

I tracked down the issue to these two lines: http://hg.libsdl.org/SDL/file/fdd5cc9e9f90/src/video/cocoa/SDL_cocoawindow.m#l1034

I believe [nswindow setFrameOrigin:rect.origin] implicitly calls [nswindow constrainFrameRect:rect toScreen:screen], which will attempt to constrain the window to the screen, but at that point the window size is still full-screen rather than the restored window size, so the constrainFrameRect function operates on the wrong window size.

https://developer.apple.com/library/mac/documentation/cocoa/reference/applicationkit/classes/NSWindow_Class/Reference/Reference.html#//apple_ref/occ/instm/NSWindow/constrainFrameRect:toScreen:

I resolved the issue by swapping the order of the function calls, like so:
[nswindow setContentSize:rect.size];
[nswindow setFrameOrigin:rect.origin];
2013-10-21 02:37:03 -07:00
Sam Lantinga d08634e2c6 Better fix for bug 2129 - fix for bug 2121 breaks linking for mingw and throws multiple warnings
J?nis R?cis

Reopening as compilation with ANSI C throws lots of unnecessary warnings, both using MinGW and using Linux GCC. (BTW, what happened? MinGW is broken to all hell. sdl2-config does not even link SDLMain anymore?)

I think this may have been lost somewhere, so again: GCC supports inlining via __inline__ in all known versions of GCC, regardless of the C standard in use. Please don't assume that __STRICT_ANSI__ implies no inlining support.
2013-10-21 02:32:34 -07:00
Sam Lantinga c10389cf5b Fixed bug 2168 - Mac: SDL_filesystem.h not exported to public header folder when building framework 2013-10-21 02:20:39 -07:00
Sam Lantinga b8b9bfbe5b Fixed some warnings building for 64-bit Windows 2013-10-21 01:16:16 -07:00
Sam Lantinga 11d38d696e Fixed bug 2015 - surface getting leak in SDLTest_ScreenShot function
Nitz 2013-08-06 03:53:41 UTC

In function SDLTest_ScreenShot() surface pointer getting leak while return
2013-10-21 00:30:05 -07:00
Sam Lantinga 80930559a3 Fixed bug 2012 - Algorithm logic getting wrong in ComputeOutCode
Nitz

I was going through the SDL_IntersectRectAndLine function and wondered to see the ComputeOutCode function implementation.

The problem in this algo is, x and y axis are getting check with respect to 0, Which is wrong, it should be get checked with respect to rectangle x and y axis.
2013-10-21 00:25:43 -07:00
Sam Lantinga 6f76bc290a Fixed bug 2061 - SDL_filesystem API not built using cmake
Marcus von Appen

The SDL_filesystem API does not get built properly on Unix and BeOS using cmake due to not including the correct C files.
2013-10-21 00:22:07 -07:00
Sam Lantinga deb8086b08 Fixed bug 2072 - Make OS X's filesystem calls use -[NSString fileSystemRepresentation]
C.W. Betts

The recommended way of getting a file name that POSIX file APIs can open in OS X when using an NSString is -[NSString fileSystemRepresentation]. However, the current filesystem API in hg uses -[NSString UTF8String].
2013-10-21 00:20:27 -07:00
Sam Lantinga 360c3d853a Better cleanup if OpenGL initialization fails 2013-10-21 00:15:24 -07:00
Sam Lantinga 5697089d98 Fixed bug 2156 - Android: Accelerometer values fed to joystick driver are inverted for the X axis and wrong for Z
Denis Bernard

Background information: http://developer.android.com/reference/android/hardware/SensorEvent.html#values

Steps to reproduce: compile testjoystick.c as an android app (change screen size according to your device). While running the app, also run:

adb logcat -c; adb logcat -s 'SDL:*' 'SDL/APP:*'

When tilting the device left/right, the joystick moves in the opposite direction of what one would expect. Or at least, the behaviour is not consistent with the Y axis.

Also when the device sits on a table (obviously not moving), the Z axis value oscillates between -32000 and +32000 (by overflow):

I/SDL/APP ( 1994): Joystick 0 axis 2 value: 32511
I/SDL/APP ( 1994): Joystick 0 axis 2 value: 32575
I/SDL/APP ( 1994): Joystick 0 axis 2 value: 32383
I/SDL/APP ( 1994): Joystick 0 axis 2 value: -32386
I/SDL/APP ( 1994): Joystick 0 axis 2 value: -32450
I/SDL/APP ( 1994): Joystick 0 axis 2 value: -32578

This is caused by the accelerometer yielding a constant value around 9.81 for Z and feeding something like 0.9 to 1.1 to the joystick driver, resulting in the overflow.

Proposed fix in SDLActivity.java (swap X and subtract G from Z reading)
2013-10-20 23:52:02 -07:00
Sam Lantinga a503dabe01 Fixed a crash initializing Android touch IDs 2013-10-20 23:23:25 -07:00
Sam Lantinga f8c11bb955 Fixed the status bar visibility in iOS 7 2013-10-20 23:05:53 -07:00
Sam Lantinga 85c2e2367c Fixed Y axis inversion on iOS; positive is up, negative is down. 2013-10-20 22:23:09 -07:00
Sam Lantinga 08fa8da77c Fixed bug 2129 - fix for bug 2121 breaks linking for mingw and throws multiple warnings
Andreas Ertelt

The problem in question is caused by changeset 7771 (http://hg.libsdl.org/SDL/rev/5486e579872e / https://bugzilla.libsdl.org/show_bug.cgi?id=2121)

The redefinition of __inline__ (introduced by the addition of begin_code.h:128's "|| __STRICT_ANSI__") results in mingw's gcc throwing multiple

   warning: always_inline function might not be inlinable [-Wattributes]

as well as a whole bunch of redefinitions of mingw internals which break linking of projects including the SDL2 headers.
2013-10-20 21:56:15 -07:00
Sam Lantinga 4ca34ad4a4 Prevent conflicts when linking both SDL2 and SDL2_gfx 2013-10-20 21:34:38 -07:00
Sam Lantinga aa86e05d21 Fixed windows compile 2013-10-20 20:49:36 -07:00
Sam Lantinga f5fa492e26 Added a macro SDL_TICKS_PASSED() to correctly compare two 32-bit tick values.
Went through the code and used the macro and fixed a couple places that were using incorrect timestamp comparisons.
2013-10-20 20:42:55 -07:00
Sam Lantinga 04e170ceb7 Added __WINDOWS__ to reflect both 32 and 64-bit windows platforms 2013-10-20 20:41:30 -07:00
Ryan C. Gordon d710399671 Fix some edge cases in XInput haptic timing. 2013-10-20 23:38:19 -04:00
Ryan C. Gordon dddb878761 Disable Win32 thread naming again. See Bugzilla #2089. 2013-10-20 23:08:45 -04:00
Ryan C. Gordon 0e699eb5bb Added SDL_GL_FRAMEBUFFER_SRGB_CAPABLE (thanks, David!).
Fixes Bugzilla #1985.
2013-10-20 21:18:05 -04:00
Ryan C. Gordon d1bde399af Backed out changeset 737771c47c6f, done testing Buildbot changes. 2013-10-20 20:24:00 -04:00
Ryan C. Gordon df3680846b Intentionally breaking Windows and Raspberry Pi builds to test Buildbot change. 2013-10-20 20:02:02 -04:00
Ryan C. Gordon f28c7fe8a5 Backed out hg changeset add2dbe99fe2. REFGUID is apparently a const type. 2013-10-20 16:03:11 -04:00
Ryan C. Gordon c521ae84bc Patched to compile on non-C99 Microsoft compiler. 2013-10-20 16:02:24 -04:00
Ryan C. Gordon a614f18666 Patched to compile on Windows. 2013-10-20 16:01:10 -04:00
Ryan C. Gordon b285b60092 Allow XInput haptics to run for SDL_HAPTIC_INFINITY time (thanks, Mitchell!).
Partially fixes Bugzilla #2126.
2013-10-20 15:55:47 -04:00
Ryan C. Gordon 6e1169ddc2 Fixed goofy logic in haptic device comparison code.
Mitchell Keith Bloch did the research on this bug, and came up with a slightly
 different patch than this.

Partially fixes Bugzilla #2126.
2013-10-20 15:49:52 -04:00
Ryan C. Gordon 54dce3f53a Don't bother calling SDL_SYS_HapticEffectType() at all for XInput devices. 2013-10-20 15:45:48 -04:00
Ryan C. Gordon 5e74299a83 Fix adding of XInput devices (thanks, Mitchell!)
Partially fixes Bugzilla #2126.
2013-10-20 15:40:20 -04:00
Sam Lantinga ade13fd741 It's okay to quit things more than once. 2013-10-20 10:39:26 -07:00
Sam Lantinga 20f5167d27 Use vertex arrays for drawing points in addition to lines 2013-10-20 10:35:51 -07:00
Sam Lantinga 8fc4fdf10b Fixed bug 1392 - Debian patch: do not propagate -lpthread
manuel.montezelo

Since the bug report[1] in 2006 Debian is shipping the patch attached.

[1] http://bugs.debian.org/375822

Maybe nowadays you don't propagate that library for linking, so maybe the patch should be dropped, but at the moment I do not have an easy/quick way to check it.

So I am submitting this report in the case that you consider it useful (even if maybe the patch itself has to be reworked), or otherwise learn if the patch is unneeded or even harmful.
2013-10-20 10:19:51 -07:00
Sam Lantinga 82b8e6df87 Fixed bug 2158 - Pixel missing in SDL_RenderDrawLines
Sean McKean

I am running Ubuntu 12.04 (GL version 1.4 Mesa 8.0.4) , and on drawing a set of lines through the renderer through SDL_RenderDrawLines() (looped or not) or SDL_RenderDrawRect() I notice a pixel missing. For RenderDrawLines() it seems to be the second point in the sequence; for RenderDrawRect() it is the lower-right. This can be fixed by specifying SDL_RenderDrawPoint(s), but wouldn't it be easier to specify each pixel in a GL_POINTS glBegin/End loop in the OpenGL code, just to make sure?

I also ran the same program on Android; the rendering seemed to be correct, which uses glDrawArrays.
2013-10-20 10:10:14 -07:00
Sam Lantinga a25b51923a Fixed compiling with the new X11 symbol wrapping 2013-10-20 09:58:37 -07:00
Stefanos Apostolopoulos 6a3478c235 Fix bug 1300 by querying current border size in ConfigureNotify, and adjusting window coordinates accordingly. 2013-10-20 17:23:43 +02:00
Sam Lantinga e343273abb Fixed bug 2162 - SDL_RenderClear not clearing entire render target
Kevin Wells

Overview:
SDL_RenderClear is only clearing part of a texture when it is the render target and a different size than the screen.

Steps to Reproduce:
1) This only occurs with the render driver set to direct3d, so: SDL_SetHint(SDL_HINT_RENDER_DRIVER,"direct3d")
Also, my window was 1280x720.

2) Create a texture for a render target with a resolution of 1024x1024:
texture=SDL_CreateTexture(main_window.renderer,SDL_PIXELFORMAT_RGBA8888,SDL_TEXTUREACCESS_TARGET,1024,1024);
SDL_SetTextureBlendMode(texture,SDL_BLENDMODE_BLEND);

3) Target the texture for rendering: SDL_SetRenderTarget(main_window.renderer,texture);

4) Set the draw color to whatever you want (problem occurs with both 0,0,0,0 and 0,0,0,255 among others) and then clear the render target:
SDL_SetRenderDrawColor(main_window.renderer,0,0,0,0);
SDL_RenderClear(main_window.renderer);

Actual Results:
Only about the top 3/4s of the texture gets cleared on calling SDL_RenderClear. The bottom 1/4 or so does not clear.

Expected Results:
Entire render target should be cleared.
2013-10-19 01:29:23 -07:00
Ryan C. Gordon b4a00144fb Fixed the XInput2 X11 symbols. 2013-10-18 10:56:45 -04:00