diff options
| -rw-r--r-- | engines/sci/engine/script_patches.cpp | 16 |
1 files changed, 11 insertions, 5 deletions
diff --git a/engines/sci/engine/script_patches.cpp b/engines/sci/engine/script_patches.cpp index a5679fac54..33ec5ef5ea 100644 --- a/engines/sci/engine/script_patches.cpp +++ b/engines/sci/engine/script_patches.cpp @@ -826,16 +826,22 @@ const uint16 qfg1vgaPatchFightEvents[] = { PATCH_END }; +// FIXME: // When QFG1VGA and QFG3 dispose of a child window. For example, when choosing // a spell (parent window), if the spell can't be casted, a subsequent window // opens, notifying that it can't be casted. When showing the child window, the // scripts restore the area below the parent window, draw the child window, and // then attempt to redraw the parent window, which leads to the background -// picture (which has just been restored) overwriting the child window. It -// appers that kGraph(redrawBox) is different in QFG1VGA and QFG3. However, we -// can just remove the window redraw and update calls when the window is -// supposed to be disposed, and the window is disposed of correctly. Fixes bug -// #3053093. +// picture (which has just been restored) overwriting the child window. +// +// This faulty redraw is caused by a used and freed SaveBits handle that is +// still stored in spellWin::pUnderBits being re-assigned with a SaveBits +// call for the child window. We should ensure that invalidated SaveBits handles +// can't (soon?) become valid again. +// +// However, we can just remove the window redraw and update calls when the +// window is supposed to be disposed, and the window is disposed of correctly. +// This is a workaround for bug #3053093. const byte qfg1vgaWindowDispose[] = { 17, 0x39, 0x05, // pushi 05 |