Call SDL_OutOfMemory() if SDL_malloc() fails.

Merged SDL 1.3 revision 5424, fixing a crash in the joystick code on recent kernels.

Fixed bug #901 Tim Angus 2009-12-11 11:45:46 PST Disable mouse event generation when state is not SDL_APPMOUSEFOCUS If a Windows SDL application is minimised by using alt-tab, SDL_APPMOUSEFOCUS is lost as part of the minimisation. Unfortunately, the directx driver doesn't pay any attention to this state when generating mouse button events, so clicking on the Desktop can cause mouse clicks in the SDL application, while it's still minimised. The attached patch fixes this. It looks much more complicated than it actually is due to indentation; here it is ignoring whitespace: tma@abraxas:~/sources/SDL-1.2-svn$ svn diff -x -b Index: src/video/windx5/SDL_dx5events.c =================================================================== --- src/video/windx5/SDL_dx5events.c (revision 5376) +++ src/video/windx5/SDL_dx5events.c (working copy) @@ -374,10 +374,9 @@ if ( !(SDL_GetAppState() & SDL_APPMOUSEFOCUS) ) { mouse_lost = 1; ClipCursor(NULL); - } - + } else { /* If the mouse was lost, regain some sense of mouse state */ - if ( mouse_lost && (SDL_GetAppState() & SDL_APPMOUSEFOCUS) ) { + if ( mouse_lost ) { POINT mouse_pos; Uint8 old_state; Uint8 new_state; @@ -548,6 +547,7 @@ if ( xrel || yrel ) { post_mouse_motion(1, xrel, yrel); } + } } /* The main Win32 event handler */

Fixed bug #898 Jeremiah Morris 2009-12-09 16:07:17 PST No-op GlobalToLocal translations in fullscreen mode On my MacBook Pro running 10.6, I noticed a small upward bias on mouse movement in a fullscreen SDL application. The app uses WarpCursor and GetMouseState in a loop to measure relative movement. I tracked it down to NSWindow's convertBaseToScreen: routine, which added a 2-pixel offset on the Y coordinate instead of the expected (+0,+0) translation. In fullscreen mode, QZ_PrivateWarpCursor() does not translate the desired position through QZ_PrivateGlobalToLocal() before passing it to the Core Graphics system. However, QZ_GetMouseLocation() does call the reverse QZ_PrivateLocalToGlobal() even in fullscreen mode. This asymmetry caused problems each time the mouse was moved.

Fixed bug #849 some more: Tim Angus 2009-11-26 14:41:04 PST Fix to the cursor not being responsive when the app doesn't have SDL_APPINPUTFOCUS The problems with the directx driver are similar to the ones I introduced in the windib driver with r4478. Basically if the application did not have focus, the mouse position is not updated. It's not really that the mouse cursor was invisible, it's that it is stuck underneath another window where you can't see it. This behaviour predates my r4478 changes and is the reason I unwittingly broke the windib driver as I had been replicating the way the directx driver deals with focus. Prior to r4478 the directx driver could not be used in windowed mode, so the broken focusing would not have actually been observable. Anyway, the attached patch makes the directx driver behaves like the windib driver in terms of focus. Time for 1.2.15? ;) I've added an additional change of moving the calls to WIN_GrabInput that are made on WM_ACTIVATE messages so that they only occur when the state is SDL_APPINPUTFOCUS. When a fullscreen application is minimised using alt-tab, it still receives WM_ACTIVATE messages when other applications are selected. If WIN_GrabInput is called when the SDL application doesn't have input focus, bad things happen; it shouldn't be being called at all. I've also added a line to make sure that SDL_APPMOUSEFOCUS state is dropped when the application is minimised following an alt-tab.

Fixed bug #895 Markus Rothe 2009-12-07 09:32:46 PST Check if libspe2 is available Building on PowerPC64 results in: [...] In file included from ./src/video/ps3/SDL_ps3events_c.h:33, from ./src/video/ps3/SDL_ps3video.c:32: ./src/video/ps3/SDL_ps3video.h:34:21: error: libspe2.h: No such file or directory In file included from ./src/video/ps3/SDL_ps3events_c.h:33, from ./src/video/ps3/SDL_ps3video.c:32: ./src/video/ps3/SDL_ps3video.h:75: error: expected specifier-qualifier-list before 'spe_context_ptr_t' ./src/video/ps3/SDL_ps3video.c:78: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'fb_writer_spu' ./src/video/ps3/SDL_ps3video.c: In function 'PS3_VideoInit': ./src/video/ps3/SDL_ps3video.c:167: error: 'spu_data_t' has no member named 'program' ./src/video/ps3/SDL_ps3video.c:167: error: 'fb_writer_spu' undeclared (first use in this function) [...] The problem is that the ps3 video driver is being build regardless if libspe2 is installed or not. The attached patch fixes this problem. [ Note: The other headers that are being checked for are coming from the linux headers package on powerpc64. As a result these are also available if you're on a generic powerpc64 machine and not a PS3 only. ]

Fixed bug #899 Jeremiah Morris 2009-12-09 16:23:50 PST Re-enable mouseLocation workaround on 10.4, 10.5 OS X systems before 10.6 have a bug with [NSEvent mouseLocation] if the screen resolution changes. SDL_QuartzVideo.m contains a workaround for this bug, but it was placed inside an #ifdef in revision 4762. The comment reads, "I'm gambling they fixed this by 10.4." After seeing this bug on several Tiger and Leopard systems (both PPC and Intel), I can confirm that it's not fixed until 10.6. The workaround doesn't compile for x86_64/10.6, so I can understand why it was segregated, but it needs to remain in place for the i386 and ppc versions. The workaround causes no problems under 10.6, even though it's not necessary there. Patch is attached with one method of keeping the workaround active on the systems that need it.

Fixed bug #896 John Popplewell 2009-12-08 23:05:50 PST Originally reported by AKFoerster on the mailing list. Error decoding UTF8 Russian text to UTF-16LE on Windows, but specifically on platforms without iconv support (the default on Windows). Valid UTF8 characters are flagged as being overlong and then substituted by the UNKNOWN_UNICODE character. After studying the testiconv.c example program, reading the RFCs and putting some printf statements in SDL_iconv.c the problem is in a test for 'Maximum overlong sequences', specifically 4.2.1, which is carried out by the following code: } else if ( p[0] >= 0xC0 ) { if ( (p[0] & 0xE0) != 0xC0 ) { /* Skip illegal sequences return SDL_ICONV_EILSEQ; */ ch = UNKNOWN_UNICODE; } else { if ( (p[0] & 0xCE) == 0xC0 ) { <<<<<<<< here overlong = SDL_TRUE; } ch = (Uint32)(p[0] & 0x1F); left = 1; } } else { Here is the 2-byte encoding of a character in range 00000080 - 000007FF 110xxxxx 10xxxxxx The line in question is supposed to be checking for an overlong sequence which would be less than 11000001 10111111 which should be represented as a single byte. BUT, the mask value (0xCE) is wrong, it isn't checking the top-most bit: 11000001 value 11001110 mask (incorrect) ^ and should be (0xDE): 11000001 value 11011110 mask (correct) making the above code: } else if ( p[0] >= 0xC0 ) { if ( (p[0] & 0xE0) != 0xC0 ) { /* Skip illegal sequences return SDL_ICONV_EILSEQ; */ ch = UNKNOWN_UNICODE; } else { if ( (p[0] & 0xDE) == 0xC0 ) { <<<<<<<< here overlong = SDL_TRUE; } ch = (Uint32)(p[0] & 0x1F); left = 1; } } else { I can supply a test program and/or a patch if required, best regards, John Popplewell

Missing pop over a jump

Fixed crash - need to save and restore rbx around cpuid, since the compiler may be assuming the stack pointer isn't being modified when filling in %0. I did it around each call to cpuid which isn't strictly necessary, but is definitely future proof. :)