From dd13fdfe1740ac847f7dfb4a332d3f5948b1a782 Mon Sep 17 00:00:00 2001
From: Colin Snover
Date: Thu, 12 Jan 2017 12:41:27 -0600
Subject: SCI32: "Fix" renderer for PQ4CD

PQ4CD and several other games contain a hack in two renderer
methods to avoid rendering invalid screen items with zero or
negative-dimension target rects. This prevents PQ4CD from
crashing during the fifth phase of target practice.
---
 engines/sci/graphics/frameout.cpp | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

(limited to 'engines/sci/graphics/frameout.cpp')

diff --git a/engines/sci/graphics/frameout.cpp b/engines/sci/graphics/frameout.cpp
index 749403093f..464b28a2ba 100644
--- a/engines/sci/graphics/frameout.cpp
+++ b/engines/sci/graphics/frameout.cpp
@@ -1168,10 +1168,13 @@ void GfxFrameout::showBits() {
 
 		byte *sourceBuffer = (byte *)_currentBuffer.getPixels() + rounded.top * _currentBuffer.screenWidth + rounded.left;
 
-		// TODO: Sometimes transition screen items generate zero-dimension
-		// show rectangles. Is this a bug?
+		// Sometimes screen items (especially from SCI2.1early transitions, like
+		// in the asteroids minigame in PQ4) generate zero-dimension show
+		// rectangles. In SSCI, zero-dimension rectangles are OK (they just
+		// result in no copy), but OSystem::copyRectToScreen will assert on
+		// them, so we need to check for zero-dimensions rectangles and ignore
+		// them explicitly
 		if (rounded.width() == 0 || rounded.height() == 0) {
-			warning("Zero-dimension show rectangle ignored");
 			continue;
 		}
 
-- 
cgit v1.2.3