I was copying Frame, MainMenuFrame, TextureFrame...

Here you go.

CMakeLists.txt

@@ -5,7 +5,6 @@ set(CMAKE_MODULE_PATH ${CMAKE_MODULE_PATH} ${CMAKE_CURRENT_SOURCE_DIR}/cmake)
 
 set(CMAKE_INCLUDE_PATH ${CMAKE_INCLUDE_PATH} "${PROJECT_SOURCE_DIR}/deps/include")
 
-
 if(MINGW)
     set(CMAKE_LIBRARY_PATH ${CMAKE_LIBRARY_PATH} "${PROJECT_SOURCE_DIR}/deps/lib/mingw/x64")
 elseif(MSVC)
@@ -16,8 +15,10 @@ endif()
 
 if(CMAKE_COMPILER_IS_GNUCXX)
     set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -std=c++14 -fpermissive")
-    # Needed to link with debug version of bgfx
-    #set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} -ldl")
+    if(UNIX)
+        # Needed to link with debug version of bgfx on linux
+        set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} -ldl")
+    endif()
 endif()
 
 set(BUILD_SHARED_LIBS OFF)

assets/Models/Core/UnitQuad.obj

@@ -0,0 +1,12 @@
+o Plane
+v -0.5 -0.5 0.0
+v 0.5 -0.5 0.0
+v -0.5 0.5 0.0
+v 0.5 0.5 0.0
+vt 0 0
+vt 1 0
+vt 1 1
+vt 0 1
+vn 0 0 1
+s off
+f 1/1/1 2/2/1 4/3/1 3/4/1

assets/Models/Test/Ball/Ballopus.mtl

@@ -0,0 +1,22 @@
+# Blender MTL File: 'octopus.blend'
+# Material Count: 2
+
+newmtl Material.001
+Ns 96.078431
+Ka 0.000000 0.000000 0.000000
+Kd 0.640000 0.640000 0.640000
+Ks 0.500000 0.500000 0.500000
+Ni 1.000000
+d 1.000000
+illum 2
+map_Kd octotex.png
+
+newmtl Material.001_NONE
+Ns 96.078431
+Ka 0.000000 0.000000 0.000000
+Kd 0.640000 0.640000 0.640000
+Ks 0.500000 0.500000 0.500000
+Ni 1.000000
+d 1.000000
+illum 2
+map_Kd octotex.png

assets/Models/Test/Brick/Brick.mtl

@@ -0,0 +1,14 @@
+# Blender MTL File: 'None'
+# Material Count: 1
+
+newmtl Material
+Ns 96.078431
+Ka 0.000000 0.000000 0.000000
+Kd 0.640000 0.640000 0.640000
+Ks 0.500000 0.500000 0.500000
+Ni 1.000000
+d 1.000000
+illum 2
+map_Kd Brick_d.png
+map_Bump Brick_n.png
+map_Ks Brick_s.png

assets/Models/Test/Brick/Brick.obj

@@ -0,0 +1,46 @@
+# Blender v2.75 (sub 0) OBJ File: ''
+# www.blender.org
+mtllib Brick.mtl
+o Cube
+v 0.500000 -0.500000 -0.500000
+v 0.500000 -0.500000 0.500000
+v -0.500000 -0.500000 0.500000
+v -0.500000 -0.500000 -0.500000
+v 0.500000 0.500000 -0.499999
+v 0.499999 0.500000 0.500000
+v -0.500000 0.500000 0.500000
+v -0.500000 0.500000 -0.500000
+vt 0.566439 0.503009
+vt 0.567048 0.728848
+vt 0.341210 0.729457
+vt 0.339992 0.277781
+vt 0.339383 0.051943
+vt 0.565221 0.051334
+vt 0.793495 0.954077
+vt 0.567657 0.954686
+vt 0.341819 0.955295
+vt 0.115981 0.955904
+vt 0.115372 0.730066
+vt 0.340601 0.503618
+vt 0.565830 0.277172
+vt 0.792886 0.728239
+vn 0.000000 -1.000000 0.000000
+vn 0.000000 1.000000 0.000000
+vn 1.000000 0.000000 0.000000
+vn -0.000000 0.000000 1.000000
+vn -1.000000 -0.000000 -0.000000
+vn 0.000000 0.000000 -1.000000
+usemtl Material
+s off
+f 1/1/1 2/2/1 3/3/1
+f 8/4/2 7/5/2 6/6/2
+f 5/7/3 6/8/3 2/2/3
+f 6/8/4 7/9/4 3/3/4
+f 7/9/5 8/10/5 4/11/5
+f 1/1/6 4/12/6 8/4/6
+f 4/12/1 1/1/1 3/3/1
+f 5/13/2 8/4/2 6/6/2
+f 1/14/3 5/7/3 2/2/3
+f 2/2/4 6/8/4 3/3/4
+f 3/3/5 7/9/5 4/11/5
+f 5/13/6 1/1/6 8/4/6

assets/Models/Test/Brick/High_Poly_Brick.mtl

@@ -0,0 +1,14 @@
+# Blender MTL File: 'Brick.blend'
+# Material Count: 1
+
+newmtl Brick
+Ns 17.647059
+Ka 0.000000 0.000000 0.000000
+Kd 0.640000 0.154050 0.039202
+Ks 0.000000 0.000000 0.000000
+Ni 1.000000
+d 1.000000
+illum 1
+map_Kd Brick_d.png
+map_Bump Brick_n.png
+map_Ks Brick_s.png

assets/Models/Test/halfpipe/Halfpipe.mtl

@@ -0,0 +1,14 @@
+# Blender MTL File: 'None'
+# Material Count: 1
+
+newmtl Material.001
+Ns 96.078431
+Ka 0.000000 0.000000 0.000000
+Kd 0.640000 0.640000 0.640000
+Ks 0.500000 0.500000 0.500000
+Ni 1.000000
+d 1.000000
+illum 2
+map_Kd Brick_Diffuse.png
+map_Bump Brick_Normal.png
+map_Ks Brick_Specular.png

assets/Models/Test/halfpipe/Halfpipe.obj

@@ -0,0 +1,124 @@
+# Blender v2.75 (sub 0) OBJ File: ''
+# www.blender.org
+mtllib Halfpipe.mtl
+o Cylinder
+v 0.000000 -2.908792 -1.000000
+v 0.000000 2.908792 -1.000000
+v 0.195090 -2.908792 -0.980785
+v 0.195090 2.908792 -0.980785
+v 0.382683 -2.908792 -0.923880
+v 0.382683 2.908792 -0.923880
+v 0.555570 -2.908792 -0.831470
+v 0.555570 2.908792 -0.831470
+v 0.707107 -2.908792 -0.707107
+v 0.707107 2.908792 -0.707107
+v 0.831470 -2.908792 -0.555570
+v 0.831470 2.908792 -0.555570
+v 0.923880 -2.908792 -0.382683
+v 0.923880 2.908792 -0.382683
+v 0.980785 -2.908792 -0.195090
+v 0.980785 2.908792 -0.195090
+v 1.000000 -2.908792 -0.000000
+v 1.000000 2.908792 -0.000000
+v -1.000000 -2.908792 -0.000001
+v -1.000000 2.908792 -0.000001
+v -0.980785 -2.908792 -0.195091
+v -0.980785 2.908792 -0.195091
+v -0.923879 -2.908792 -0.382684
+v -0.923879 2.908792 -0.382684
+v -0.831469 -2.908792 -0.555571
+v -0.831469 2.908792 -0.555571
+v -0.707106 -2.908792 -0.707108
+v -0.707106 2.908792 -0.707108
+v -0.555569 -2.908792 -0.831470
+v -0.555569 2.908792 -0.831470
+v -0.382682 -2.908792 -0.923880
+v -0.382682 2.908792 -0.923880
+v -0.195089 -2.908792 -0.980786
+v -0.195089 2.908792 -0.980786
+vt -1.156714 -1.419990
+vt -1.158346 2.336063
+vt -1.031779 2.336118
+vt -1.283288 -1.420045
+vt -1.284913 2.336008
+vt -1.409862 -1.420100
+vt -1.411480 2.335954
+vt -1.536436 -1.420155
+vt -1.538048 2.335899
+vt 2.387354 -1.418355
+vt 2.385534 2.337692
+vt 2.512101 2.337753
+vt 2.260782 -1.418416
+vt 2.258967 2.337631
+vt 2.134208 -1.418478
+vt 2.132400 2.337570
+vt 2.007634 -1.418538
+vt 2.005832 2.337510
+vt -0.144122 -1.419541
+vt -0.145809 2.336510
+vt -0.019241 2.336567
+vt -0.270696 -1.419598
+vt -0.272376 2.336454
+vt -0.397270 -1.419654
+vt -0.398943 2.336397
+vt -0.523844 -1.419711
+vt -0.525511 2.336341
+vt -0.650418 -1.419767
+vt -0.652078 2.336285
+vt -0.776992 -1.419823
+vt -0.778645 2.336229
+vt -1.030140 -1.419935
+vt -0.905212 2.336174
+vt -0.903566 -1.419879
+vt 2.513929 -1.418293
+vt -0.017548 -1.419484
+vn -0.098000 0.000000 0.995200
+vn -0.290300 0.000000 0.956900
+vn -0.471400 0.000000 0.881900
+vn -0.634400 0.000000 0.773000
+vn -0.773000 0.000000 0.634400
+vn -0.881900 0.000000 0.471400
+vn -0.956900 0.000000 0.290300
+vn -0.995200 0.000000 0.098000
+vn 0.995200 0.000000 0.098000
+vn 0.956900 0.000000 0.290300
+vn 0.881900 0.000000 0.471400
+vn 0.773000 0.000000 0.634400
+vn 0.634400 0.000000 0.773000
+vn 0.471400 0.000000 0.881900
+vn 0.098000 0.000000 0.995200
+vn 0.290300 0.000000 0.956900
+usemtl Material.001
+s off
+f 3/1/1 4/2/1 2/3/1
+f 5/4/2 6/5/2 4/2/2
+f 7/6/3 8/7/3 6/5/3
+f 9/8/4 10/9/4 8/7/4
+f 11/10/5 12/11/5 10/12/5
+f 13/13/6 14/14/6 12/11/6
+f 15/15/7 16/16/7 14/14/7
+f 17/17/8 18/18/8 16/16/8
+f 21/19/9 22/20/9 20/21/9
+f 23/22/10 24/23/10 22/20/10
+f 25/24/11 26/25/11 24/23/11
+f 27/26/12 28/27/12 26/25/12
+f 29/28/13 30/29/13 28/27/13
+f 31/30/14 32/31/14 30/29/14
+f 1/32/15 2/3/15 34/33/15
+f 33/34/16 34/33/16 32/31/16
+f 1/32/1 3/1/1 2/3/1
+f 3/1/2 5/4/2 4/2/2
+f 5/4/3 7/6/3 6/5/3
+f 7/6/4 9/8/4 8/7/4
+f 9/35/5 11/10/5 10/12/5
+f 11/10/6 13/13/6 12/11/6
+f 13/13/7 15/15/7 14/14/7
+f 15/15/8 17/17/8 16/16/8
+f 19/36/9 21/19/9 20/21/9
+f 21/19/10 23/22/10 22/20/10
+f 23/22/11 25/24/11 24/23/11
+f 25/24/12 27/26/12 26/25/12
+f 27/26/13 29/28/13 28/27/13
+f 29/28/14 31/30/14 30/29/14
+f 33/34/15 1/32/15 34/33/15
+f 31/30/16 33/34/16 32/31/16

cmake/Findliquidfun.cmake

@@ -0,0 +1,8 @@
+find_path(liquidfun_INCLUDE_DIR Box2D/Box2D.h)
+find_library(liquidfun_LIBRARY NAMES liquidfun)
+
+set(liquidfun_INCLUDE_DIRS ${liquidfun_INCLUDE_DIR})
+set(liquidfun_LIBRARIES ${liquidfun_LIBRARY})
+
+find_package_handle_standard_args(liquidfun DEFAULT_MSG liquidfun_LIBRARY liquidfun_INCLUDE_DIR)
+mark_as_advanced(liquidfun_FOUND liquidfun_INCLUDE_DIR liquidfun_LIBRARY)

deps/include/Box2D/Box2D.h

@@ -0,0 +1,68 @@
+/*
+* Copyright (c) 2006-2009 Erin Catto http://www.box2d.org
+* Copyright (c) 2013 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#ifndef BOX2D_H
+#define BOX2D_H
+
+/**
+\mainpage LiquidFun API Documentation
+
+*/
+
+// These include files constitute the main Box2D API
+
+#include <Box2D/Common/b2Settings.h>
+#include <Box2D/Common/b2Draw.h>
+#include <Box2D/Common/b2Stat.h>
+#include <Box2D/Common/b2Timer.h>
+
+#include <Box2D/Collision/Shapes/b2CircleShape.h>
+#include <Box2D/Collision/Shapes/b2EdgeShape.h>
+#include <Box2D/Collision/Shapes/b2ChainShape.h>
+#include <Box2D/Collision/Shapes/b2PolygonShape.h>
+
+#include <Box2D/Collision/b2BroadPhase.h>
+#include <Box2D/Collision/b2Distance.h>
+#include <Box2D/Collision/b2DynamicTree.h>
+#include <Box2D/Collision/b2TimeOfImpact.h>
+
+#include <Box2D/Dynamics/b2Body.h>
+#include <Box2D/Dynamics/b2Fixture.h>
+#include <Box2D/Dynamics/b2WorldCallbacks.h>
+#include <Box2D/Dynamics/b2TimeStep.h>
+#include <Box2D/Dynamics/b2World.h>
+
+#include <Box2D/Dynamics/Contacts/b2Contact.h>
+
+#include <Box2D/Dynamics/Joints/b2DistanceJoint.h>
+#include <Box2D/Dynamics/Joints/b2FrictionJoint.h>
+#include <Box2D/Dynamics/Joints/b2GearJoint.h>
+#include <Box2D/Dynamics/Joints/b2MotorJoint.h>
+#include <Box2D/Dynamics/Joints/b2MouseJoint.h>
+#include <Box2D/Dynamics/Joints/b2PrismaticJoint.h>
+#include <Box2D/Dynamics/Joints/b2PulleyJoint.h>
+#include <Box2D/Dynamics/Joints/b2RevoluteJoint.h>
+#include <Box2D/Dynamics/Joints/b2RopeJoint.h>
+#include <Box2D/Dynamics/Joints/b2WeldJoint.h>
+#include <Box2D/Dynamics/Joints/b2WheelJoint.h>
+
+#include <Box2D/Particle/b2Particle.h>
+#include <Box2D/Particle/b2ParticleGroup.h>
+
+#endif

deps/include/Box2D/Collision/Shapes/b2ChainShape.cpp

@@ -0,0 +1,203 @@
+/*
+* Copyright (c) 2006-2010 Erin Catto http://www.box2d.org
+* Copyright (c) 2013 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#include <Box2D/Collision/Shapes/b2ChainShape.h>
+#include <Box2D/Collision/Shapes/b2EdgeShape.h>
+#include <new>
+#include <memory.h>
+#include <string.h>
+
+b2ChainShape::~b2ChainShape()
+{
+	b2Free(m_vertices);
+	m_vertices = NULL;
+	m_count = 0;
+}
+
+void b2ChainShape::CreateLoop(const b2Vec2* vertices, int32 count)
+{
+	b2Assert(m_vertices == NULL && m_count == 0);
+	b2Assert(count >= 3);
+	for (int32 i = 1; i < count; ++i)
+	{
+#if B2_ASSERT_ENABLED
+		b2Vec2 v1 = vertices[i-1];
+		b2Vec2 v2 = vertices[i];
+		// If the code crashes here, it means your vertices are too close together.
+		b2Assert(b2DistanceSquared(v1, v2) > b2_linearSlop * b2_linearSlop);
+#endif // B2_ASSERT_ENABLED
+	}
+
+	m_count = count + 1;
+	m_vertices = (b2Vec2*)b2Alloc(m_count * sizeof(b2Vec2));
+	memcpy(m_vertices, vertices, count * sizeof(b2Vec2));
+	m_vertices[count] = m_vertices[0];
+	m_prevVertex = m_vertices[m_count - 2];
+	m_nextVertex = m_vertices[1];
+	m_hasPrevVertex = true;
+	m_hasNextVertex = true;
+}
+
+void b2ChainShape::CreateChain(const b2Vec2* vertices, int32 count)
+{
+	b2Assert(m_vertices == NULL && m_count == 0);
+	b2Assert(count >= 2);
+	for (int32 i = 1; i < count; ++i)
+	{
+#if B2_ASSERT_ENABLED
+		b2Vec2 v1 = vertices[i-1];
+		b2Vec2 v2 = vertices[i];
+		// If the code crashes here, it means your vertices are too close together.
+		b2Assert(b2DistanceSquared(v1, v2) > b2_linearSlop * b2_linearSlop);
+#endif // B2_ASSERT_ENABLED
+	}
+
+	m_count = count;
+	m_vertices = (b2Vec2*)b2Alloc(count * sizeof(b2Vec2));
+	memcpy(m_vertices, vertices, m_count * sizeof(b2Vec2));
+
+	m_hasPrevVertex = false;
+	m_hasNextVertex = false;
+
+	m_prevVertex.SetZero();
+	m_nextVertex.SetZero();
+}
+
+void b2ChainShape::SetPrevVertex(const b2Vec2& prevVertex)
+{
+	m_prevVertex = prevVertex;
+	m_hasPrevVertex = true;
+}
+
+void b2ChainShape::SetNextVertex(const b2Vec2& nextVertex)
+{
+	m_nextVertex = nextVertex;
+	m_hasNextVertex = true;
+}
+
+b2Shape* b2ChainShape::Clone(b2BlockAllocator* allocator) const
+{
+	void* mem = allocator->Allocate(sizeof(b2ChainShape));
+	b2ChainShape* clone = new (mem) b2ChainShape;
+	clone->CreateChain(m_vertices, m_count);
+	clone->m_prevVertex = m_prevVertex;
+	clone->m_nextVertex = m_nextVertex;
+	clone->m_hasPrevVertex = m_hasPrevVertex;
+	clone->m_hasNextVertex = m_hasNextVertex;
+	return clone;
+}
+
+int32 b2ChainShape::GetChildCount() const
+{
+	// edge count = vertex count - 1
+	return m_count - 1;
+}
+
+void b2ChainShape::GetChildEdge(b2EdgeShape* edge, int32 index) const
+{
+	b2Assert(0 <= index && index < m_count - 1);
+	edge->m_type = b2Shape::e_edge;
+	edge->m_radius = m_radius;
+
+	edge->m_vertex1 = m_vertices[index + 0];
+	edge->m_vertex2 = m_vertices[index + 1];
+
+	if (index > 0)
+	{
+		edge->m_vertex0 = m_vertices[index - 1];
+		edge->m_hasVertex0 = true;
+	}
+	else
+	{
+		edge->m_vertex0 = m_prevVertex;
+		edge->m_hasVertex0 = m_hasPrevVertex;
+	}
+
+	if (index < m_count - 2)
+	{
+		edge->m_vertex3 = m_vertices[index + 2];
+		edge->m_hasVertex3 = true;
+	}
+	else
+	{
+		edge->m_vertex3 = m_nextVertex;
+		edge->m_hasVertex3 = m_hasNextVertex;
+	}
+}
+
+void b2ChainShape::ComputeDistance(const b2Transform& xf, const b2Vec2& p, float32* distance, b2Vec2* normal, int32 childIndex) const
+{
+	b2EdgeShape edge;
+	GetChildEdge(&edge, childIndex);
+	edge.ComputeDistance(xf, p, distance, normal, 0);
+}
+
+bool b2ChainShape::TestPoint(const b2Transform& xf, const b2Vec2& p) const
+{
+	B2_NOT_USED(xf);
+	B2_NOT_USED(p);
+	return false;
+}
+
+bool b2ChainShape::RayCast(b2RayCastOutput* output, const b2RayCastInput& input,
+							const b2Transform& xf, int32 childIndex) const
+{
+	b2Assert(childIndex < m_count);
+
+	b2EdgeShape edgeShape;
+
+	int32 i1 = childIndex;
+	int32 i2 = childIndex + 1;
+	if (i2 == m_count)
+	{
+		i2 = 0;
+	}
+
+	edgeShape.m_vertex1 = m_vertices[i1];
+	edgeShape.m_vertex2 = m_vertices[i2];
+
+	return edgeShape.RayCast(output, input, xf, 0);
+}
+
+void b2ChainShape::ComputeAABB(b2AABB* aabb, const b2Transform& xf, int32 childIndex) const
+{
+	b2Assert(childIndex < m_count);
+
+	int32 i1 = childIndex;
+	int32 i2 = childIndex + 1;
+	if (i2 == m_count)
+	{
+		i2 = 0;
+	}
+
+	b2Vec2 v1 = b2Mul(xf, m_vertices[i1]);
+	b2Vec2 v2 = b2Mul(xf, m_vertices[i2]);
+
+	aabb->lowerBound = b2Min(v1, v2);
+	aabb->upperBound = b2Max(v1, v2);
+}
+
+void b2ChainShape::ComputeMass(b2MassData* massData, float32 density) const
+{
+	B2_NOT_USED(density);
+
+	massData->mass = 0.0f;
+	massData->center.SetZero();
+	massData->I = 0.0f;
+}

deps/include/Box2D/Collision/Shapes/b2ChainShape.h

@@ -0,0 +1,106 @@
+/*
+* Copyright (c) 2006-2010 Erin Catto http://www.box2d.org
+* Copyright (c) 2013 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#ifndef B2_CHAIN_SHAPE_H
+#define B2_CHAIN_SHAPE_H
+
+#include <Box2D/Collision/Shapes/b2Shape.h>
+
+class b2EdgeShape;
+
+/// A chain shape is a free form sequence of line segments.
+/// The chain has two-sided collision, so you can use inside and outside collision.
+/// Therefore, you may use any winding order.
+/// Since there may be many vertices, they are allocated using b2Alloc.
+/// Connectivity information is used to create smooth collisions.
+/// WARNING: The chain will not collide properly if there are self-intersections.
+class b2ChainShape : public b2Shape
+{
+public:
+	b2ChainShape();
+
+	/// The destructor frees the vertices using b2Free.
+	~b2ChainShape();
+
+	/// Create a loop. This automatically adjusts connectivity.
+	/// @param vertices an array of vertices, these are copied
+	/// @param count the vertex count
+	void CreateLoop(const b2Vec2* vertices, int32 count);
+
+	/// Create a chain with isolated end vertices.
+	/// @param vertices an array of vertices, these are copied
+	/// @param count the vertex count
+	void CreateChain(const b2Vec2* vertices, int32 count);
+
+	/// Establish connectivity to a vertex that precedes the first vertex.
+	/// Don't call this for loops.
+	void SetPrevVertex(const b2Vec2& prevVertex);
+
+	/// Establish connectivity to a vertex that follows the last vertex.
+	/// Don't call this for loops.
+	void SetNextVertex(const b2Vec2& nextVertex);
+
+	/// Implement b2Shape. Vertices are cloned using b2Alloc.
+	b2Shape* Clone(b2BlockAllocator* allocator) const;
+
+	/// @see b2Shape::GetChildCount
+	int32 GetChildCount() const;
+
+	/// Get a child edge.
+	void GetChildEdge(b2EdgeShape* edge, int32 index) const;
+
+	/// This always return false.
+	/// @see b2Shape::TestPoint
+	bool TestPoint(const b2Transform& transform, const b2Vec2& p) const;
+
+	// @see b2Shape::ComputeDistance
+	void ComputeDistance(const b2Transform& xf, const b2Vec2& p, float32* distance, b2Vec2* normal, int32 childIndex) const;
+
+	/// Implement b2Shape.
+	bool RayCast(b2RayCastOutput* output, const b2RayCastInput& input,
+					const b2Transform& transform, int32 childIndex) const;
+
+	/// @see b2Shape::ComputeAABB
+	void ComputeAABB(b2AABB* aabb, const b2Transform& transform, int32 childIndex) const;
+
+	/// Chains have zero mass.
+	/// @see b2Shape::ComputeMass
+	void ComputeMass(b2MassData* massData, float32 density) const;
+
+	/// The vertices. Owned by this class.
+	b2Vec2* m_vertices;
+
+	/// The vertex count.
+	int32 m_count;
+
+	b2Vec2 m_prevVertex, m_nextVertex;
+	bool m_hasPrevVertex, m_hasNextVertex;
+};
+
+inline b2ChainShape::b2ChainShape()
+{
+	m_type = e_chain;
+	m_radius = b2_polygonRadius;
+	m_vertices = NULL;
+	m_count = 0;
+	m_hasPrevVertex = false;
+	m_hasNextVertex = false;
+}
+
+#endif

deps/include/Box2D/Collision/Shapes/b2CircleShape.cpp

@@ -0,0 +1,111 @@
+/*
+* Copyright (c) 2006-2009 Erin Catto http://www.box2d.org
+* Copyright (c) 2013 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#include <Box2D/Collision/Shapes/b2CircleShape.h>
+#include <new>
+
+b2Shape* b2CircleShape::Clone(b2BlockAllocator* allocator) const
+{
+	void* mem = allocator->Allocate(sizeof(b2CircleShape));
+	b2CircleShape* clone = new (mem) b2CircleShape;
+	*clone = *this;
+	return clone;
+}
+
+int32 b2CircleShape::GetChildCount() const
+{
+	return 1;
+}
+
+bool b2CircleShape::TestPoint(const b2Transform& transform, const b2Vec2& p) const
+{
+	b2Vec2 center = transform.p + b2Mul(transform.q, m_p);
+	b2Vec2 d = p - center;
+	return b2Dot(d, d) <= m_radius * m_radius;
+}
+
+void b2CircleShape::ComputeDistance(const b2Transform& transform, const b2Vec2& p, float32* distance, b2Vec2* normal, int32 childIndex) const
+{
+	B2_NOT_USED(childIndex);
+
+	b2Vec2 center = transform.p + b2Mul(transform.q, m_p);
+	b2Vec2 d = p - center;
+	float32 d1 = d.Length();
+	*distance = d1 - m_radius;
+	*normal = 1 / d1 * d;
+}
+
+// Collision Detection in Interactive 3D Environments by Gino van den Bergen
+// From Section 3.1.2
+// x = s + a * r
+// norm(x) = radius
+bool b2CircleShape::RayCast(b2RayCastOutput* output, const b2RayCastInput& input,
+							const b2Transform& transform, int32 childIndex) const
+{
+	B2_NOT_USED(childIndex);
+
+	b2Vec2 position = transform.p + b2Mul(transform.q, m_p);
+	b2Vec2 s = input.p1 - position;
+	float32 b = b2Dot(s, s) - m_radius * m_radius;
+
+	// Solve quadratic equation.
+	b2Vec2 r = input.p2 - input.p1;
+	float32 c =  b2Dot(s, r);
+	float32 rr = b2Dot(r, r);
+	float32 sigma = c * c - rr * b;
+
+	// Check for negative discriminant and short segment.
+	if (sigma < 0.0f || rr < b2_epsilon)
+	{
+		return false;
+	}
+
+	// Find the point of intersection of the line with the circle.
+	float32 a = -(c + b2Sqrt(sigma));
+
+	// Is the intersection point on the segment?
+	if (0.0f <= a && a <= input.maxFraction * rr)
+	{
+		a /= rr;
+		output->fraction = a;
+		output->normal = s + a * r;
+		output->normal.Normalize();
+		return true;
+	}
+
+	return false;
+}
+
+void b2CircleShape::ComputeAABB(b2AABB* aabb, const b2Transform& transform, int32 childIndex) const
+{
+	B2_NOT_USED(childIndex);
+
+	b2Vec2 p = transform.p + b2Mul(transform.q, m_p);
+	aabb->lowerBound.Set(p.x - m_radius, p.y - m_radius);
+	aabb->upperBound.Set(p.x + m_radius, p.y + m_radius);
+}
+
+void b2CircleShape::ComputeMass(b2MassData* massData, float32 density) const
+{
+	massData->mass = density * b2_pi * m_radius * m_radius;
+	massData->center = m_p;
+
+	// inertia about the local origin
+	massData->I = massData->mass * (0.5f * m_radius * m_radius + b2Dot(m_p, m_p));
+}

deps/include/Box2D/Collision/Shapes/b2CircleShape.h

@@ -0,0 +1,107 @@
+/*
+* Copyright (c) 2006-2009 Erin Catto http://www.box2d.org
+* Copyright (c) 2013 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#ifndef B2_CIRCLE_SHAPE_H
+#define B2_CIRCLE_SHAPE_H
+
+#include <Box2D/Collision/Shapes/b2Shape.h>
+
+/// A circle shape.
+class b2CircleShape : public b2Shape
+{
+public:
+	b2CircleShape();
+
+	/// Implement b2Shape.
+	b2Shape* Clone(b2BlockAllocator* allocator) const;
+
+	/// @see b2Shape::GetChildCount
+	int32 GetChildCount() const;
+
+	/// Implement b2Shape.
+	bool TestPoint(const b2Transform& transform, const b2Vec2& p) const;
+
+	// @see b2Shape::ComputeDistance
+	void ComputeDistance(const b2Transform& xf, const b2Vec2& p, float32* distance, b2Vec2* normal, int32 childIndex) const;
+
+	/// Implement b2Shape.
+	bool RayCast(b2RayCastOutput* output, const b2RayCastInput& input,
+				const b2Transform& transform, int32 childIndex) const;
+
+	/// @see b2Shape::ComputeAABB
+	void ComputeAABB(b2AABB* aabb, const b2Transform& transform, int32 childIndex) const;
+
+	/// @see b2Shape::ComputeMass
+	void ComputeMass(b2MassData* massData, float32 density) const;
+
+	/// Get the supporting vertex index in the given direction.
+	int32 GetSupport(const b2Vec2& d) const;
+
+	/// Get the supporting vertex in the given direction.
+	const b2Vec2& GetSupportVertex(const b2Vec2& d) const;
+
+	/// Get the vertex count.
+	int32 GetVertexCount() const { return 1; }
+
+	/// Get a vertex by index. Used by b2Distance.
+	const b2Vec2& GetVertex(int32 index) const;
+
+#if LIQUIDFUN_EXTERNAL_LANGUAGE_API
+public:
+	/// Set position with direct floats.
+	void SetPosition(float32 x, float32 y) { m_p.Set(x, y); }
+
+	/// Get x-coordinate of position.
+	float32 GetPositionX() const { return m_p.x; }
+
+	/// Get y-coordinate of position.
+	float32 GetPositionY() const { return m_p.y; }
+#endif // LIQUIDFUN_EXTERNAL_LANGUAGE_API
+
+	/// Position
+	b2Vec2 m_p;
+};
+
+inline b2CircleShape::b2CircleShape()
+{
+	m_type = e_circle;
+	m_radius = 0.0f;
+	m_p.SetZero();
+}
+
+inline int32 b2CircleShape::GetSupport(const b2Vec2 &d) const
+{
+	B2_NOT_USED(d);
+	return 0;
+}
+
+inline const b2Vec2& b2CircleShape::GetSupportVertex(const b2Vec2 &d) const
+{
+	B2_NOT_USED(d);
+	return m_p;
+}
+
+inline const b2Vec2& b2CircleShape::GetVertex(int32 index) const
+{
+	B2_NOT_USED(index);
+	b2Assert(index == 0);
+	return m_p;
+}
+
+#endif

deps/include/Box2D/Collision/Shapes/b2EdgeShape.cpp

@@ -0,0 +1,168 @@
+/*
+* Copyright (c) 2006-2010 Erin Catto http://www.box2d.org
+* Copyright (c) 2013 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#include <Box2D/Collision/Shapes/b2EdgeShape.h>
+#include <new>
+
+void b2EdgeShape::Set(const b2Vec2& v1, const b2Vec2& v2)
+{
+	m_vertex1 = v1;
+	m_vertex2 = v2;
+	m_hasVertex0 = false;
+	m_hasVertex3 = false;
+}
+
+b2Shape* b2EdgeShape::Clone(b2BlockAllocator* allocator) const
+{
+	void* mem = allocator->Allocate(sizeof(b2EdgeShape));
+	b2EdgeShape* clone = new (mem) b2EdgeShape;
+	*clone = *this;
+	return clone;
+}
+
+int32 b2EdgeShape::GetChildCount() const
+{
+	return 1;
+}
+
+bool b2EdgeShape::TestPoint(const b2Transform& xf, const b2Vec2& p) const
+{
+	B2_NOT_USED(xf);
+	B2_NOT_USED(p);
+	return false;
+}
+
+void b2EdgeShape::ComputeDistance(const b2Transform& xf, const b2Vec2& p, float32* distance, b2Vec2* normal, int32 childIndex) const
+{
+	B2_NOT_USED(childIndex);
+
+	b2Vec2 v1 = b2Mul(xf, m_vertex1);
+	b2Vec2 v2 = b2Mul(xf, m_vertex2);
+
+	b2Vec2 d = p - v1;
+	b2Vec2 s = v2 - v1;
+	float32 ds = b2Dot(d, s);
+	if (ds > 0)
+	{
+		float32 s2 = b2Dot(s, s);
+		if (ds > s2)
+		{
+			d = p - v2;
+		}
+		else
+		{
+			d -= ds / s2 * s;
+		}
+	}
+
+	float32 d1 = d.Length();
+	*distance = d1;
+	*normal = d1 > 0 ? 1 / d1 * d : b2Vec2_zero;
+
+}
+
+// p = p1 + t * d
+// v = v1 + s * e
+// p1 + t * d = v1 + s * e
+// s * e - t * d = p1 - v1
+bool b2EdgeShape::RayCast(b2RayCastOutput* output, const b2RayCastInput& input,
+							const b2Transform& xf, int32 childIndex) const
+{
+	B2_NOT_USED(childIndex);
+
+	// Put the ray into the edge's frame of reference.
+	b2Vec2 p1 = b2MulT(xf.q, input.p1 - xf.p);
+	b2Vec2 p2 = b2MulT(xf.q, input.p2 - xf.p);
+	b2Vec2 d = p2 - p1;
+
+	b2Vec2 v1 = m_vertex1;
+	b2Vec2 v2 = m_vertex2;
+	b2Vec2 e = v2 - v1;
+	b2Vec2 normal(e.y, -e.x);
+	normal.Normalize();
+
+	// q = p1 + t * d
+	// dot(normal, q - v1) = 0
+	// dot(normal, p1 - v1) + t * dot(normal, d) = 0
+	float32 numerator = b2Dot(normal, v1 - p1);
+	float32 denominator = b2Dot(normal, d);
+
+	if (denominator == 0.0f)
+	{
+		return false;
+	}
+
+	float32 t = numerator / denominator;
+	if (t < 0.0f || input.maxFraction < t)
+	{
+		return false;
+	}
+
+	b2Vec2 q = p1 + t * d;
+
+	// q = v1 + s * r
+	// s = dot(q - v1, r) / dot(r, r)
+	b2Vec2 r = v2 - v1;
+	float32 rr = b2Dot(r, r);
+	if (rr == 0.0f)
+	{
+		return false;
+	}
+
+	float32 s = b2Dot(q - v1, r) / rr;
+	if (s < 0.0f || 1.0f < s)
+	{
+		return false;
+	}
+
+	output->fraction = t;
+	if (numerator > 0.0f)
+	{
+		output->normal = -b2Mul(xf.q, normal);
+	}
+	else
+	{
+		output->normal = b2Mul(xf.q, normal);
+	}
+	return true;
+}
+
+void b2EdgeShape::ComputeAABB(b2AABB* aabb, const b2Transform& xf, int32 childIndex) const
+{
+	B2_NOT_USED(childIndex);
+
+	b2Vec2 v1 = b2Mul(xf, m_vertex1);
+	b2Vec2 v2 = b2Mul(xf, m_vertex2);
+
+	b2Vec2 lower = b2Min(v1, v2);
+	b2Vec2 upper = b2Max(v1, v2);
+
+	b2Vec2 r(m_radius, m_radius);
+	aabb->lowerBound = lower - r;
+	aabb->upperBound = upper + r;
+}
+
+void b2EdgeShape::ComputeMass(b2MassData* massData, float32 density) const
+{
+	B2_NOT_USED(density);
+
+	massData->mass = 0.0f;
+	massData->center = 0.5f * (m_vertex1 + m_vertex2);
+	massData->I = 0.0f;
+}

deps/include/Box2D/Collision/Shapes/b2EdgeShape.h

@@ -0,0 +1,94 @@
+/*
+* Copyright (c) 2006-2010 Erin Catto http://www.box2d.org
+* Copyright (c) 2013 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#ifndef B2_EDGE_SHAPE_H
+#define B2_EDGE_SHAPE_H
+
+#include <Box2D/Collision/Shapes/b2Shape.h>
+
+/// A line segment (edge) shape. These can be connected in chains or loops
+/// to other edge shapes. The connectivity information is used to ensure
+/// correct contact normals.
+class b2EdgeShape : public b2Shape
+{
+public:
+	b2EdgeShape();
+
+	/// Set this as an isolated edge.
+	void Set(const b2Vec2& v1, const b2Vec2& v2);
+
+	/// Implement b2Shape.
+	b2Shape* Clone(b2BlockAllocator* allocator) const;
+
+	/// @see b2Shape::GetChildCount
+	int32 GetChildCount() const;
+
+	/// @see b2Shape::TestPoint
+	bool TestPoint(const b2Transform& transform, const b2Vec2& p) const;
+
+	// @see b2Shape::ComputeDistance
+	void ComputeDistance(const b2Transform& xf, const b2Vec2& p, float32* distance, b2Vec2* normal, int32 childIndex) const;
+
+	/// Implement b2Shape.
+	bool RayCast(b2RayCastOutput* output, const b2RayCastInput& input,
+				const b2Transform& transform, int32 childIndex) const;
+
+	/// @see b2Shape::ComputeAABB
+	void ComputeAABB(b2AABB* aabb, const b2Transform& transform, int32 childIndex) const;
+
+	/// @see b2Shape::ComputeMass
+	void ComputeMass(b2MassData* massData, float32 density) const;
+
+#if LIQUIDFUN_EXTERNAL_LANGUAGE_API
+public:
+	/// Set this as an isolated edge, with direct floats.
+	void Set(float32 vx1, float32 vy1, float32 vx2, float32 vy2);
+#endif // LIQUIDFUN_EXTERNAL_LANGUAGE_API
+
+	/// These are the edge vertices
+	b2Vec2 m_vertex1, m_vertex2;
+
+	/// Optional adjacent vertices. These are used for smooth collision.
+	b2Vec2 m_vertex0, m_vertex3;
+	bool m_hasVertex0, m_hasVertex3;
+};
+
+inline b2EdgeShape::b2EdgeShape()
+{
+	m_type = e_edge;
+	m_radius = b2_polygonRadius;
+	m_vertex0.x = 0.0f;
+	m_vertex0.y = 0.0f;
+	m_vertex3.x = 0.0f;
+	m_vertex3.y = 0.0f;
+	m_hasVertex0 = false;
+	m_hasVertex3 = false;
+}
+
+#if LIQUIDFUN_EXTERNAL_LANGUAGE_API
+inline void b2EdgeShape::Set(float32 vx1,
+														 float32 vy1,
+														 float32 vx2,
+														 float32 vy2) {
+	Set(b2Vec2(vx1, vy1), b2Vec2(vx2, vy2));
+}
+#endif // LIQUIDFUN_EXTERNAL_LANGUAGE_API
+
+
+#endif

deps/include/Box2D/Collision/Shapes/b2PolygonShape.cpp

@@ -0,0 +1,504 @@
+/*
+* Copyright (c) 2006-2009 Erin Catto http://www.box2d.org
+* Copyright (c) 2013 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#include <Box2D/Collision/Shapes/b2PolygonShape.h>
+#include <new>
+
+b2Shape* b2PolygonShape::Clone(b2BlockAllocator* allocator) const
+{
+	void* mem = allocator->Allocate(sizeof(b2PolygonShape));
+	b2PolygonShape* clone = new (mem) b2PolygonShape;
+	*clone = *this;
+	return clone;
+}
+
+void b2PolygonShape::SetAsBox(float32 hx, float32 hy)
+{
+	m_count = 4;
+	m_vertices[0].Set(-hx, -hy);
+	m_vertices[1].Set( hx, -hy);
+	m_vertices[2].Set( hx,  hy);
+	m_vertices[3].Set(-hx,  hy);
+	m_normals[0].Set(0.0f, -1.0f);
+	m_normals[1].Set(1.0f, 0.0f);
+	m_normals[2].Set(0.0f, 1.0f);
+	m_normals[3].Set(-1.0f, 0.0f);
+	m_centroid.SetZero();
+}
+
+void b2PolygonShape::SetAsBox(float32 hx, float32 hy, const b2Vec2& center, float32 angle)
+{
+	m_count = 4;
+	m_vertices[0].Set(-hx, -hy);
+	m_vertices[1].Set( hx, -hy);
+	m_vertices[2].Set( hx,  hy);
+	m_vertices[3].Set(-hx,  hy);
+	m_normals[0].Set(0.0f, -1.0f);
+	m_normals[1].Set(1.0f, 0.0f);
+	m_normals[2].Set(0.0f, 1.0f);
+	m_normals[3].Set(-1.0f, 0.0f);
+	m_centroid = center;
+
+	b2Transform xf;
+	xf.p = center;
+	xf.q.Set(angle);
+
+	// Transform vertices and normals.
+	for (int32 i = 0; i < m_count; ++i)
+	{
+		m_vertices[i] = b2Mul(xf, m_vertices[i]);
+		m_normals[i] = b2Mul(xf.q, m_normals[i]);
+	}
+}
+
+int32 b2PolygonShape::GetChildCount() const
+{
+	return 1;
+}
+
+static b2Vec2 ComputeCentroid(const b2Vec2* vs, int32 count)
+{
+	b2Assert(count >= 3);
+
+	b2Vec2 c; c.Set(0.0f, 0.0f);
+	float32 area = 0.0f;
+
+	// pRef is the reference point for forming triangles.
+	// It's location doesn't change the result (except for rounding error).
+	b2Vec2 pRef(0.0f, 0.0f);
+#if 0
+	// This code would put the reference point inside the polygon.
+	for (int32 i = 0; i < count; ++i)
+	{
+		pRef += vs[i];
+	}
+	pRef *= 1.0f / count;
+#endif
+
+	const float32 inv3 = 1.0f / 3.0f;
+
+	for (int32 i = 0; i < count; ++i)
+	{
+		// Triangle vertices.
+		b2Vec2 p1 = pRef;
+		b2Vec2 p2 = vs[i];
+		b2Vec2 p3 = i + 1 < count ? vs[i+1] : vs[0];
+
+		b2Vec2 e1 = p2 - p1;
+		b2Vec2 e2 = p3 - p1;
+
+		float32 D = b2Cross(e1, e2);
+
+		float32 triangleArea = 0.5f * D;
+		area += triangleArea;
+
+		// Area weighted centroid
+		c += triangleArea * inv3 * (p1 + p2 + p3);
+	}
+
+	// Centroid
+	b2Assert(area > b2_epsilon);
+	c *= 1.0f / area;
+	return c;
+}
+
+void b2PolygonShape::Set(const b2Vec2* vertices, int32 count)
+{
+	b2Assert(3 <= count && count <= b2_maxPolygonVertices);
+	if (count < 3)
+	{
+		SetAsBox(1.0f, 1.0f);
+		return;
+	}
+	
+	int32 n = b2Min(count, b2_maxPolygonVertices);
+
+	// Perform welding and copy vertices into local buffer.
+	b2Vec2 ps[b2_maxPolygonVertices];
+	int32 tempCount = 0;
+	for (int32 i = 0; i < n; ++i)
+	{
+		b2Vec2 v = vertices[i];
+
+		bool unique = true;
+		for (int32 j = 0; j < tempCount; ++j)
+		{
+			if (b2DistanceSquared(v, ps[j]) < 0.5f * b2_linearSlop)
+			{
+				unique = false;
+				break;
+			}
+		}
+
+		if (unique)
+		{
+			ps[tempCount++] = v;
+		}
+	}
+
+	n = tempCount;
+	if (n < 3)
+	{
+		// Polygon is degenerate.
+		b2Assert(false);
+		SetAsBox(1.0f, 1.0f);
+		return;
+	}
+
+	// Create the convex hull using the Gift wrapping algorithm
+	// http://en.wikipedia.org/wiki/Gift_wrapping_algorithm
+
+	// Find the right most point on the hull
+	int32 i0 = 0;
+	float32 x0 = ps[0].x;
+	for (int32 i = 1; i < n; ++i)
+	{
+		float32 x = ps[i].x;
+		if (x > x0 || (x == x0 && ps[i].y < ps[i0].y))
+		{
+			i0 = i;
+			x0 = x;
+		}
+	}
+
+	int32 hull[b2_maxPolygonVertices];
+	int32 m = 0;
+	int32 ih = i0;
+
+	for (;;)
+	{
+		hull[m] = ih;
+
+		int32 ie = 0;
+		for (int32 j = 1; j < n; ++j)
+		{
+			if (ie == ih)
+			{
+				ie = j;
+				continue;
+			}
+
+			b2Vec2 r = ps[ie] - ps[hull[m]];
+			b2Vec2 v = ps[j] - ps[hull[m]];
+			float32 c = b2Cross(r, v);
+			if (c < 0.0f)
+			{
+				ie = j;
+			}
+
+			// Collinearity check
+			if (c == 0.0f && v.LengthSquared() > r.LengthSquared())
+			{
+				ie = j;
+			}
+		}
+
+		++m;
+		ih = ie;
+
+		if (ie == i0)
+		{
+			break;
+		}
+	}
+	
+	m_count = m;
+
+	// Copy vertices.
+	for (int32 i = 0; i < m; ++i)
+	{
+		m_vertices[i] = ps[hull[i]];
+	}
+
+	// Compute normals. Ensure the edges have non-zero length.
+	for (int32 i = 0; i < m; ++i)
+	{
+		int32 i1 = i;
+		int32 i2 = i + 1 < m ? i + 1 : 0;
+		b2Vec2 edge = m_vertices[i2] - m_vertices[i1];
+		b2Assert(edge.LengthSquared() > b2_epsilon * b2_epsilon);
+		m_normals[i] = b2Cross(edge, 1.0f);
+		m_normals[i].Normalize();
+	}
+
+	// Compute the polygon centroid.
+	m_centroid = ComputeCentroid(m_vertices, m);
+}
+
+bool b2PolygonShape::TestPoint(const b2Transform& xf, const b2Vec2& p) const
+{
+	b2Vec2 pLocal = b2MulT(xf.q, p - xf.p);
+
+	for (int32 i = 0; i < m_count; ++i)
+	{
+		float32 dot = b2Dot(m_normals[i], pLocal - m_vertices[i]);
+		if (dot > 0.0f)
+		{
+			return false;
+		}
+	}
+
+	return true;
+}
+
+void b2PolygonShape::ComputeDistance(const b2Transform& xf, const b2Vec2& p, float32* distance, b2Vec2* normal, int32 childIndex) const
+{
+	B2_NOT_USED(childIndex);
+
+	b2Vec2 pLocal = b2MulT(xf.q, p - xf.p);
+	float32 maxDistance = -FLT_MAX;
+	b2Vec2 normalForMaxDistance = pLocal;
+
+	for (int32 i = 0; i < m_count; ++i)
+	{
+		float32 dot = b2Dot(m_normals[i], pLocal - m_vertices[i]);
+		if (dot > maxDistance)
+		{
+			maxDistance = dot;
+			normalForMaxDistance = m_normals[i];
+		}
+	}
+
+	if (maxDistance > 0)
+	{
+		b2Vec2 minDistance = normalForMaxDistance;
+		float32 minDistance2 = maxDistance * maxDistance;
+		for (int32 i = 0; i < m_count; ++i)
+		{
+			b2Vec2 distance = pLocal - m_vertices[i];
+			float32 distance2 = distance.LengthSquared();
+			if (minDistance2 > distance2)
+			{
+				minDistance = distance;
+				minDistance2 = distance2;
+			}
+		}
+
+		*distance = b2Sqrt(minDistance2);
+		*normal = b2Mul(xf.q, minDistance);
+		normal->Normalize();
+	}
+	else
+	{
+		*distance = maxDistance;
+		*normal = b2Mul(xf.q, normalForMaxDistance);
+	}
+}
+
+bool b2PolygonShape::RayCast(b2RayCastOutput* output, const b2RayCastInput& input,
+								const b2Transform& xf, int32 childIndex) const
+{
+	B2_NOT_USED(childIndex);
+
+	// Put the ray into the polygon's frame of reference.
+	b2Vec2 p1 = b2MulT(xf.q, input.p1 - xf.p);
+	b2Vec2 p2 = b2MulT(xf.q, input.p2 - xf.p);
+	b2Vec2 d = p2 - p1;
+
+	float32 lower = 0.0f, upper = input.maxFraction;
+
+	int32 index = -1;
+
+	for (int32 i = 0; i < m_count; ++i)
+	{
+		// p = p1 + a * d
+		// dot(normal, p - v) = 0
+		// dot(normal, p1 - v) + a * dot(normal, d) = 0
+		float32 numerator = b2Dot(m_normals[i], m_vertices[i] - p1);
+		float32 denominator = b2Dot(m_normals[i], d);
+
+		if (denominator == 0.0f)
+		{	
+			if (numerator < 0.0f)
+			{
+				return false;
+			}
+		}
+		else
+		{
+			// Note: we want this predicate without division:
+			// lower < numerator / denominator, where denominator < 0
+			// Since denominator < 0, we have to flip the inequality:
+			// lower < numerator / denominator <==> denominator * lower > numerator.
+			if (denominator < 0.0f && numerator < lower * denominator)
+			{
+				// Increase lower.
+				// The segment enters this half-space.
+				lower = numerator / denominator;
+				index = i;
+			}
+			else if (denominator > 0.0f && numerator < upper * denominator)
+			{
+				// Decrease upper.
+				// The segment exits this half-space.
+				upper = numerator / denominator;
+			}
+		}
+
+		// The use of epsilon here causes the assert on lower to trip
+		// in some cases. Apparently the use of epsilon was to make edge
+		// shapes work, but now those are handled separately.
+		//if (upper < lower - b2_epsilon)
+		if (upper < lower)
+		{
+			return false;
+		}
+	}
+
+	b2Assert(0.0f <= lower && lower <= input.maxFraction);
+
+	if (index >= 0)
+	{
+		output->fraction = lower;
+		output->normal = b2Mul(xf.q, m_normals[index]);
+		return true;
+	}
+
+	return false;
+}
+
+void b2PolygonShape::ComputeAABB(b2AABB* aabb, const b2Transform& xf, int32 childIndex) const
+{
+	B2_NOT_USED(childIndex);
+
+	b2Vec2 lower = b2Mul(xf, m_vertices[0]);
+	b2Vec2 upper = lower;
+
+	for (int32 i = 1; i < m_count; ++i)
+	{
+		b2Vec2 v = b2Mul(xf, m_vertices[i]);
+		lower = b2Min(lower, v);
+		upper = b2Max(upper, v);
+	}
+
+	b2Vec2 r(m_radius, m_radius);
+	aabb->lowerBound = lower - r;
+	aabb->upperBound = upper + r;
+}
+
+void b2PolygonShape::ComputeMass(b2MassData* massData, float32 density) const
+{
+	// Polygon mass, centroid, and inertia.
+	// Let rho be the polygon density in mass per unit area.
+	// Then:
+	// mass = rho * int(dA)
+	// centroid.x = (1/mass) * rho * int(x * dA)
+	// centroid.y = (1/mass) * rho * int(y * dA)
+	// I = rho * int((x*x + y*y) * dA)
+	//
+	// We can compute these integrals by summing all the integrals
+	// for each triangle of the polygon. To evaluate the integral
+	// for a single triangle, we make a change of variables to
+	// the (u,v) coordinates of the triangle:
+	// x = x0 + e1x * u + e2x * v
+	// y = y0 + e1y * u + e2y * v
+	// where 0 <= u && 0 <= v && u + v <= 1.
+	//
+	// We integrate u from [0,1-v] and then v from [0,1].
+	// We also need to use the Jacobian of the transformation:
+	// D = cross(e1, e2)
+	//
+	// Simplification: triangle centroid = (1/3) * (p1 + p2 + p3)
+	//
+	// The rest of the derivation is handled by computer algebra.
+
+	b2Assert(m_count >= 3);
+
+	b2Vec2 center; center.Set(0.0f, 0.0f);
+	float32 area = 0.0f;
+	float32 I = 0.0f;
+
+	// s is the reference point for forming triangles.
+	// It's location doesn't change the result (except for rounding error).
+	b2Vec2 s(0.0f, 0.0f);
+
+	// This code would put the reference point inside the polygon.
+	for (int32 i = 0; i < m_count; ++i)
+	{
+		s += m_vertices[i];
+	}
+	s *= 1.0f / m_count;
+
+	const float32 k_inv3 = 1.0f / 3.0f;
+
+	for (int32 i = 0; i < m_count; ++i)
+	{
+		// Triangle vertices.
+		b2Vec2 e1 = m_vertices[i] - s;
+		b2Vec2 e2 = i + 1 < m_count ? m_vertices[i+1] - s : m_vertices[0] - s;
+
+		float32 D = b2Cross(e1, e2);
+
+		float32 triangleArea = 0.5f * D;
+		area += triangleArea;
+
+		// Area weighted centroid
+		center += triangleArea * k_inv3 * (e1 + e2);
+
+		float32 ex1 = e1.x, ey1 = e1.y;
+		float32 ex2 = e2.x, ey2 = e2.y;
+
+		float32 intx2 = ex1*ex1 + ex2*ex1 + ex2*ex2;
+		float32 inty2 = ey1*ey1 + ey2*ey1 + ey2*ey2;
+
+		I += (0.25f * k_inv3 * D) * (intx2 + inty2);
+	}
+
+	// Total mass
+	massData->mass = density * area;
+
+	// Center of mass
+	b2Assert(area > b2_epsilon);
+	center *= 1.0f / area;
+	massData->center = center + s;
+
+	// Inertia tensor relative to the local origin (point s).
+	massData->I = density * I;
+	
+	// Shift to center of mass then to original body origin.
+	massData->I += massData->mass * (b2Dot(massData->center, massData->center) - b2Dot(center, center));
+}
+
+bool b2PolygonShape::Validate() const
+{
+	for (int32 i = 0; i < m_count; ++i)
+	{
+		int32 i1 = i;
+		int32 i2 = i < m_count - 1 ? i1 + 1 : 0;
+		b2Vec2 p = m_vertices[i1];
+		b2Vec2 e = m_vertices[i2] - p;
+
+		for (int32 j = 0; j < m_count; ++j)
+		{
+			if (j == i1 || j == i2)
+			{
+				continue;
+			}
+
+			b2Vec2 v = m_vertices[j] - p;
+			float32 c = b2Cross(e, v);
+			if (c < 0.0f)
+			{
+				return false;
+			}
+		}
+	}
+
+	return true;
+}

deps/include/Box2D/Collision/Shapes/b2PolygonShape.h

@@ -0,0 +1,134 @@
+/*
+* Copyright (c) 2006-2009 Erin Catto http://www.box2d.org
+* Copyright (c) 2013 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#ifndef B2_POLYGON_SHAPE_H
+#define B2_POLYGON_SHAPE_H
+
+#include <Box2D/Collision/Shapes/b2Shape.h>
+
+/// A convex polygon. It is assumed that the interior of the polygon is to
+/// the left of each edge.
+/// Polygons have a maximum number of vertices equal to b2_maxPolygonVertices.
+/// In most cases you should not need many vertices for a convex polygon.
+class b2PolygonShape : public b2Shape
+{
+public:
+	b2PolygonShape();
+
+	/// Implement b2Shape.
+	b2Shape* Clone(b2BlockAllocator* allocator) const;
+
+	/// @see b2Shape::GetChildCount
+	int32 GetChildCount() const;
+
+	/// Create a convex hull from the given array of local points.
+	/// The count must be in the range [3, b2_maxPolygonVertices].
+	/// @warning the points may be re-ordered, even if they form a convex polygon
+	/// @warning collinear points are handled but not removed. Collinear points
+	/// may lead to poor stacking behavior.
+	void Set(const b2Vec2* points, int32 count);
+
+	/// Build vertices to represent an axis-aligned box centered on the local origin.
+	/// @param hx the half-width.
+	/// @param hy the half-height.
+	void SetAsBox(float32 hx, float32 hy);
+
+	/// Build vertices to represent an oriented box.
+	/// @param hx the half-width.
+	/// @param hy the half-height.
+	/// @param center the center of the box in local coordinates.
+	/// @param angle the rotation of the box in local coordinates.
+	void SetAsBox(float32 hx, float32 hy, const b2Vec2& center, float32 angle);
+
+	/// @see b2Shape::TestPoint
+	bool TestPoint(const b2Transform& transform, const b2Vec2& p) const;
+
+	// @see b2Shape::ComputeDistance
+	void ComputeDistance(const b2Transform& xf, const b2Vec2& p, float32* distance, b2Vec2* normal, int32 childIndex) const;
+
+	/// Implement b2Shape.
+	bool RayCast(b2RayCastOutput* output, const b2RayCastInput& input,
+					const b2Transform& transform, int32 childIndex) const;
+
+	/// @see b2Shape::ComputeAABB
+	void ComputeAABB(b2AABB* aabb, const b2Transform& transform, int32 childIndex) const;
+
+	/// @see b2Shape::ComputeMass
+	void ComputeMass(b2MassData* massData, float32 density) const;
+
+	/// Get the vertex count.
+	int32 GetVertexCount() const { return m_count; }
+
+	/// Get a vertex by index.
+	const b2Vec2& GetVertex(int32 index) const;
+
+	/// Validate convexity. This is a very time consuming operation.
+	/// @returns true if valid
+	bool Validate() const;
+
+#if LIQUIDFUN_EXTERNAL_LANGUAGE_API
+public:
+	/// Set centroid with direct floats.
+	void SetCentroid(float32 x, float32 y);
+
+	/// SetAsBox with direct floats for center.
+	/// @see b2Shape::SetAsBox
+	void SetAsBox(float32 hx,
+								float32 hy,
+								float32 centerX,
+								float32 centerY,
+								float32 angle);
+#endif // LIQUIDFUN_EXTERNAL_LANGUAGE_API
+
+	b2Vec2 m_centroid;
+	b2Vec2 m_vertices[b2_maxPolygonVertices];
+	b2Vec2 m_normals[b2_maxPolygonVertices];
+	int32 m_count;
+};
+
+inline b2PolygonShape::b2PolygonShape()
+{
+	m_type = e_polygon;
+	m_radius = b2_polygonRadius;
+	m_count = 0;
+	m_centroid.SetZero();
+}
+
+inline const b2Vec2& b2PolygonShape::GetVertex(int32 index) const
+{
+	b2Assert(0 <= index && index < m_count);
+	return m_vertices[index];
+}
+
+#if LIQUIDFUN_EXTERNAL_LANGUAGE_API
+inline void b2PolygonShape::SetCentroid(float32 x, float32 y)
+{
+	m_centroid.Set(x, y);
+}
+
+inline void b2PolygonShape::SetAsBox(float32 hx,
+										 float32 hy,
+										 float32 centerX,
+										 float32 centerY,
+										 float32 angle) {
+	SetAsBox(hx, hy, b2Vec2(centerX, centerY), angle);
+}
+#endif // LIQUIDFUN_EXTERNAL_LANGUAGE_API
+
+#endif

deps/include/Box2D/Collision/Shapes/b2Shape.h

@@ -0,0 +1,109 @@
+/*
+* Copyright (c) 2006-2009 Erin Catto http://www.box2d.org
+* Copyright (c) 2013 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#ifndef B2_SHAPE_H
+#define B2_SHAPE_H
+
+#include <Box2D/Common/b2BlockAllocator.h>
+#include <Box2D/Common/b2Math.h>
+#include <Box2D/Collision/b2Collision.h>
+
+/// This holds the mass data computed for a shape.
+struct b2MassData
+{
+	/// The mass of the shape, usually in kilograms.
+	float32 mass;
+
+	/// The position of the shape's centroid relative to the shape's origin.
+	b2Vec2 center;
+
+	/// The rotational inertia of the shape about the local origin.
+	float32 I;
+};
+
+/// A shape is used for collision detection. You can create a shape however you like.
+/// Shapes used for simulation in b2World are created automatically when a b2Fixture
+/// is created. Shapes may encapsulate a one or more child shapes.
+class b2Shape
+{
+public:
+	
+	enum Type
+	{
+		e_circle = 0,
+		e_edge = 1,
+		e_polygon = 2,
+		e_chain = 3,
+		e_typeCount = 4
+	};
+
+	virtual ~b2Shape() {}
+
+	/// Clone the concrete shape using the provided allocator.
+	virtual b2Shape* Clone(b2BlockAllocator* allocator) const = 0;
+
+	/// Get the type of this shape. You can use this to down cast to the concrete shape.
+	/// @return the shape type.
+	Type GetType() const;
+
+	/// Get the number of child primitives.
+	virtual int32 GetChildCount() const = 0;
+
+	/// Test a point for containment in this shape. This only works for convex shapes.
+	/// @param xf the shape world transform.
+	/// @param p a point in world coordinates.
+	virtual bool TestPoint(const b2Transform& xf, const b2Vec2& p) const = 0;
+
+	/// Compute the distance from the current shape to the specified point. This only works for convex shapes.
+	/// @param xf the shape world transform.
+	/// @param p a point in world coordinates.
+	/// @param distance returns the distance from the current shape.
+	/// @param normal returns the direction in which the distance increases.
+	virtual void ComputeDistance(const b2Transform& xf, const b2Vec2& p, float32* distance, b2Vec2* normal, int32 childIndex) const= 0;
+
+	/// Cast a ray against a child shape.
+	/// @param output the ray-cast results.
+	/// @param input the ray-cast input parameters.
+	/// @param transform the transform to be applied to the shape.
+	/// @param childIndex the child shape index
+	virtual bool RayCast(b2RayCastOutput* output, const b2RayCastInput& input,
+						const b2Transform& transform, int32 childIndex) const = 0;
+
+	/// Given a transform, compute the associated axis aligned bounding box for a child shape.
+	/// @param aabb returns the axis aligned box.
+	/// @param xf the world transform of the shape.
+	/// @param childIndex the child shape
+	virtual void ComputeAABB(b2AABB* aabb, const b2Transform& xf, int32 childIndex) const = 0;
+
+	/// Compute the mass properties of this shape using its dimensions and density.
+	/// The inertia tensor is computed about the local origin.
+	/// @param massData returns the mass data for this shape.
+	/// @param density the density in kilograms per meter squared.
+	virtual void ComputeMass(b2MassData* massData, float32 density) const = 0;
+
+	Type m_type;
+	float32 m_radius;
+};
+
+inline b2Shape::Type b2Shape::GetType() const
+{
+	return m_type;
+}
+
+#endif

deps/include/Box2D/Collision/b2BroadPhase.cpp

@@ -0,0 +1,119 @@
+/*
+* Copyright (c) 2006-2009 Erin Catto http://www.box2d.org
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#include <Box2D/Collision/b2BroadPhase.h>
+
+b2BroadPhase::b2BroadPhase()
+{
+	m_proxyCount = 0;
+
+	m_pairCapacity = 16;
+	m_pairCount = 0;
+	m_pairBuffer = (b2Pair*)b2Alloc(m_pairCapacity * sizeof(b2Pair));
+
+	m_moveCapacity = 16;
+	m_moveCount = 0;
+	m_moveBuffer = (int32*)b2Alloc(m_moveCapacity * sizeof(int32));
+}
+
+b2BroadPhase::~b2BroadPhase()
+{
+	b2Free(m_moveBuffer);
+	b2Free(m_pairBuffer);
+}
+
+int32 b2BroadPhase::CreateProxy(const b2AABB& aabb, void* userData)
+{
+	int32 proxyId = m_tree.CreateProxy(aabb, userData);
+	++m_proxyCount;
+	BufferMove(proxyId);
+	return proxyId;
+}
+
+void b2BroadPhase::DestroyProxy(int32 proxyId)
+{
+	UnBufferMove(proxyId);
+	--m_proxyCount;
+	m_tree.DestroyProxy(proxyId);
+}
+
+void b2BroadPhase::MoveProxy(int32 proxyId, const b2AABB& aabb, const b2Vec2& displacement)
+{
+	bool buffer = m_tree.MoveProxy(proxyId, aabb, displacement);
+	if (buffer)
+	{
+		BufferMove(proxyId);
+	}
+}
+
+void b2BroadPhase::TouchProxy(int32 proxyId)
+{
+	BufferMove(proxyId);
+}
+
+void b2BroadPhase::BufferMove(int32 proxyId)
+{
+	if (m_moveCount == m_moveCapacity)
+	{
+		int32* oldBuffer = m_moveBuffer;
+		m_moveCapacity *= 2;
+		m_moveBuffer = (int32*)b2Alloc(m_moveCapacity * sizeof(int32));
+		memcpy(m_moveBuffer, oldBuffer, m_moveCount * sizeof(int32));
+		b2Free(oldBuffer);
+	}
+
+	m_moveBuffer[m_moveCount] = proxyId;
+	++m_moveCount;
+}
+
+void b2BroadPhase::UnBufferMove(int32 proxyId)
+{
+	for (int32 i = 0; i < m_moveCount; ++i)
+	{
+		if (m_moveBuffer[i] == proxyId)
+		{
+			m_moveBuffer[i] = e_nullProxy;
+		}
+	}
+}
+
+// This is called from b2DynamicTree::Query when we are gathering pairs.
+bool b2BroadPhase::QueryCallback(int32 proxyId)
+{
+	// A proxy cannot form a pair with itself.
+	if (proxyId == m_queryProxyId)
+	{
+		return true;
+	}
+
+	// Grow the pair buffer as needed.
+	if (m_pairCount == m_pairCapacity)
+	{
+		b2Pair* oldBuffer = m_pairBuffer;
+		m_pairCapacity *= 2;
+		m_pairBuffer = (b2Pair*)b2Alloc(m_pairCapacity * sizeof(b2Pair));
+		memcpy(m_pairBuffer, oldBuffer, m_pairCount * sizeof(b2Pair));
+		b2Free(oldBuffer);
+	}
+
+	m_pairBuffer[m_pairCount].proxyIdA = b2Min(proxyId, m_queryProxyId);
+	m_pairBuffer[m_pairCount].proxyIdB = b2Max(proxyId, m_queryProxyId);
+	++m_pairCount;
+
+	return true;
+}

deps/include/Box2D/Collision/b2BroadPhase.h

@@ -0,0 +1,257 @@
+/*
+* Copyright (c) 2006-2009 Erin Catto http://www.box2d.org
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#ifndef B2_BROAD_PHASE_H
+#define B2_BROAD_PHASE_H
+
+#include <Box2D/Common/b2Settings.h>
+#include <Box2D/Collision/b2Collision.h>
+#include <Box2D/Collision/b2DynamicTree.h>
+#include <algorithm>
+
+struct b2Pair
+{
+	int32 proxyIdA;
+	int32 proxyIdB;
+};
+
+/// The broad-phase is used for computing pairs and performing volume queries and ray casts.
+/// This broad-phase does not persist pairs. Instead, this reports potentially new pairs.
+/// It is up to the client to consume the new pairs and to track subsequent overlap.
+class b2BroadPhase
+{
+public:
+
+	enum
+	{
+		e_nullProxy = -1
+	};
+
+	b2BroadPhase();
+	~b2BroadPhase();
+
+	/// Create a proxy with an initial AABB. Pairs are not reported until
+	/// UpdatePairs is called.
+	int32 CreateProxy(const b2AABB& aabb, void* userData);
+
+	/// Destroy a proxy. It is up to the client to remove any pairs.
+	void DestroyProxy(int32 proxyId);
+
+	/// Call MoveProxy as many times as you like, then when you are done
+	/// call UpdatePairs to finalized the proxy pairs (for your time step).
+	void MoveProxy(int32 proxyId, const b2AABB& aabb, const b2Vec2& displacement);
+
+	/// Call to trigger a re-processing of it's pairs on the next call to UpdatePairs.
+	void TouchProxy(int32 proxyId);
+
+	/// Get the fat AABB for a proxy.
+	const b2AABB& GetFatAABB(int32 proxyId) const;
+
+	/// Get user data from a proxy. Returns NULL if the id is invalid.
+	void* GetUserData(int32 proxyId) const;
+
+	/// Test overlap of fat AABBs.
+	bool TestOverlap(int32 proxyIdA, int32 proxyIdB) const;
+
+	/// Get the number of proxies.
+	int32 GetProxyCount() const;
+
+	/// Update the pairs. This results in pair callbacks. This can only add pairs.
+	template <typename T>
+	void UpdatePairs(T* callback);
+
+	/// Query an AABB for overlapping proxies. The callback class
+	/// is called for each proxy that overlaps the supplied AABB.
+	template <typename T>
+	void Query(T* callback, const b2AABB& aabb) const;
+
+	/// Ray-cast against the proxies in the tree. This relies on the callback
+	/// to perform a exact ray-cast in the case were the proxy contains a shape.
+	/// The callback also performs the any collision filtering. This has performance
+	/// roughly equal to k * log(n), where k is the number of collisions and n is the
+	/// number of proxies in the tree.
+	/// @param input the ray-cast input data. The ray extends from p1 to p1 + maxFraction * (p2 - p1).
+	/// @param callback a callback class that is called for each proxy that is hit by the ray.
+	template <typename T>
+	void RayCast(T* callback, const b2RayCastInput& input) const;
+
+	/// Get the height of the embedded tree.
+	int32 GetTreeHeight() const;
+
+	/// Get the balance of the embedded tree.
+	int32 GetTreeBalance() const;
+
+	/// Get the quality metric of the embedded tree.
+	float32 GetTreeQuality() const;
+
+	/// Shift the world origin. Useful for large worlds.
+	/// The shift formula is: position -= newOrigin
+	/// @param newOrigin the new origin with respect to the old origin
+	void ShiftOrigin(const b2Vec2& newOrigin);
+
+private:
+
+	friend class b2DynamicTree;
+
+	void BufferMove(int32 proxyId);
+	void UnBufferMove(int32 proxyId);
+
+	bool QueryCallback(int32 proxyId);
+
+	b2DynamicTree m_tree;
+
+	int32 m_proxyCount;
+
+	int32* m_moveBuffer;
+	int32 m_moveCapacity;
+	int32 m_moveCount;
+
+	b2Pair* m_pairBuffer;
+	int32 m_pairCapacity;
+	int32 m_pairCount;
+
+	int32 m_queryProxyId;
+};
+
+/// This is used to sort pairs.
+inline bool b2PairLessThan(const b2Pair& pair1, const b2Pair& pair2)
+{
+	if (pair1.proxyIdA < pair2.proxyIdA)
+	{
+		return true;
+	}
+
+	if (pair1.proxyIdA == pair2.proxyIdA)
+	{
+		return pair1.proxyIdB < pair2.proxyIdB;
+	}
+
+	return false;
+}
+
+inline void* b2BroadPhase::GetUserData(int32 proxyId) const
+{
+	return m_tree.GetUserData(proxyId);
+}
+
+inline bool b2BroadPhase::TestOverlap(int32 proxyIdA, int32 proxyIdB) const
+{
+	const b2AABB& aabbA = m_tree.GetFatAABB(proxyIdA);
+	const b2AABB& aabbB = m_tree.GetFatAABB(proxyIdB);
+	return b2TestOverlap(aabbA, aabbB);
+}
+
+inline const b2AABB& b2BroadPhase::GetFatAABB(int32 proxyId) const
+{
+	return m_tree.GetFatAABB(proxyId);
+}
+
+inline int32 b2BroadPhase::GetProxyCount() const
+{
+	return m_proxyCount;
+}
+
+inline int32 b2BroadPhase::GetTreeHeight() const
+{
+	return m_tree.GetHeight();
+}
+
+inline int32 b2BroadPhase::GetTreeBalance() const
+{
+	return m_tree.GetMaxBalance();
+}
+
+inline float32 b2BroadPhase::GetTreeQuality() const
+{
+	return m_tree.GetAreaRatio();
+}
+
+template <typename T>
+void b2BroadPhase::UpdatePairs(T* callback)
+{
+	// Reset pair buffer
+	m_pairCount = 0;
+
+	// Perform tree queries for all moving proxies.
+	for (int32 i = 0; i < m_moveCount; ++i)
+	{
+		m_queryProxyId = m_moveBuffer[i];
+		if (m_queryProxyId == e_nullProxy)
+		{
+			continue;
+		}
+
+		// We have to query the tree with the fat AABB so that
+		// we don't fail to create a pair that may touch later.
+		const b2AABB& fatAABB = m_tree.GetFatAABB(m_queryProxyId);
+
+		// Query tree, create pairs and add them pair buffer.
+		m_tree.Query(this, fatAABB);
+	}
+
+	// Reset move buffer
+	m_moveCount = 0;
+
+	// Sort the pair buffer to expose duplicates.
+	std::sort(m_pairBuffer, m_pairBuffer + m_pairCount, b2PairLessThan);
+
+	// Send the pairs back to the client.
+	int32 i = 0;
+	while (i < m_pairCount)
+	{
+		b2Pair* primaryPair = m_pairBuffer + i;
+		void* userDataA = m_tree.GetUserData(primaryPair->proxyIdA);
+		void* userDataB = m_tree.GetUserData(primaryPair->proxyIdB);
+
+		callback->AddPair(userDataA, userDataB);
+		++i;
+
+		// Skip any duplicate pairs.
+		while (i < m_pairCount)
+		{
+			b2Pair* pair = m_pairBuffer + i;
+			if (pair->proxyIdA != primaryPair->proxyIdA || pair->proxyIdB != primaryPair->proxyIdB)
+			{
+				break;
+			}
+			++i;
+		}
+	}
+
+	// Try to keep the tree balanced.
+	//m_tree.Rebalance(4);
+}
+
+template <typename T>
+inline void b2BroadPhase::Query(T* callback, const b2AABB& aabb) const
+{
+	m_tree.Query(callback, aabb);
+}
+
+template <typename T>
+inline void b2BroadPhase::RayCast(T* callback, const b2RayCastInput& input) const
+{
+	m_tree.RayCast(callback, input);
+}
+
+inline void b2BroadPhase::ShiftOrigin(const b2Vec2& newOrigin)
+{
+	m_tree.ShiftOrigin(newOrigin);
+}
+
+#endif

deps/include/Box2D/Collision/b2CollideCircle.cpp

@@ -0,0 +1,154 @@
+/*
+* Copyright (c) 2007-2009 Erin Catto http://www.box2d.org
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#include <Box2D/Collision/b2Collision.h>
+#include <Box2D/Collision/Shapes/b2CircleShape.h>
+#include <Box2D/Collision/Shapes/b2PolygonShape.h>
+
+void b2CollideCircles(
+	b2Manifold* manifold,
+	const b2CircleShape* circleA, const b2Transform& xfA,
+	const b2CircleShape* circleB, const b2Transform& xfB)
+{
+	manifold->pointCount = 0;
+
+	b2Vec2 pA = b2Mul(xfA, circleA->m_p);
+	b2Vec2 pB = b2Mul(xfB, circleB->m_p);
+
+	b2Vec2 d = pB - pA;
+	float32 distSqr = b2Dot(d, d);
+	float32 rA = circleA->m_radius, rB = circleB->m_radius;
+	float32 radius = rA + rB;
+	if (distSqr > radius * radius)
+	{
+		return;
+	}
+
+	manifold->type = b2Manifold::e_circles;
+	manifold->localPoint = circleA->m_p;
+	manifold->localNormal.SetZero();
+	manifold->pointCount = 1;
+
+	manifold->points[0].localPoint = circleB->m_p;
+	manifold->points[0].id.key = 0;
+}
+
+void b2CollidePolygonAndCircle(
+	b2Manifold* manifold,
+	const b2PolygonShape* polygonA, const b2Transform& xfA,
+	const b2CircleShape* circleB, const b2Transform& xfB)
+{
+	manifold->pointCount = 0;
+
+	// Compute circle position in the frame of the polygon.
+	b2Vec2 c = b2Mul(xfB, circleB->m_p);
+	b2Vec2 cLocal = b2MulT(xfA, c);
+
+	// Find the min separating edge.
+	int32 normalIndex = 0;
+	float32 separation = -b2_maxFloat;
+	float32 radius = polygonA->m_radius + circleB->m_radius;
+	int32 vertexCount = polygonA->m_count;
+	const b2Vec2* vertices = polygonA->m_vertices;
+	const b2Vec2* normals = polygonA->m_normals;
+
+	for (int32 i = 0; i < vertexCount; ++i)
+	{
+		float32 s = b2Dot(normals[i], cLocal - vertices[i]);
+
+		if (s > radius)
+		{
+			// Early out.
+			return;
+		}
+
+		if (s > separation)
+		{
+			separation = s;
+			normalIndex = i;
+		}
+	}
+
+	// Vertices that subtend the incident face.
+	int32 vertIndex1 = normalIndex;
+	int32 vertIndex2 = vertIndex1 + 1 < vertexCount ? vertIndex1 + 1 : 0;
+	b2Vec2 v1 = vertices[vertIndex1];
+	b2Vec2 v2 = vertices[vertIndex2];
+
+	// If the center is inside the polygon ...
+	if (separation < b2_epsilon)
+	{
+		manifold->pointCount = 1;
+		manifold->type = b2Manifold::e_faceA;
+		manifold->localNormal = normals[normalIndex];
+		manifold->localPoint = 0.5f * (v1 + v2);
+		manifold->points[0].localPoint = circleB->m_p;
+		manifold->points[0].id.key = 0;
+		return;
+	}
+
+	// Compute barycentric coordinates
+	float32 u1 = b2Dot(cLocal - v1, v2 - v1);
+	float32 u2 = b2Dot(cLocal - v2, v1 - v2);
+	if (u1 <= 0.0f)
+	{
+		if (b2DistanceSquared(cLocal, v1) > radius * radius)
+		{
+			return;
+		}
+
+		manifold->pointCount = 1;
+		manifold->type = b2Manifold::e_faceA;
+		manifold->localNormal = cLocal - v1;
+		manifold->localNormal.Normalize();
+		manifold->localPoint = v1;
+		manifold->points[0].localPoint = circleB->m_p;
+		manifold->points[0].id.key = 0;
+	}
+	else if (u2 <= 0.0f)
+	{
+		if (b2DistanceSquared(cLocal, v2) > radius * radius)
+		{
+			return;
+		}
+
+		manifold->pointCount = 1;
+		manifold->type = b2Manifold::e_faceA;
+		manifold->localNormal = cLocal - v2;
+		manifold->localNormal.Normalize();
+		manifold->localPoint = v2;
+		manifold->points[0].localPoint = circleB->m_p;
+		manifold->points[0].id.key = 0;
+	}
+	else
+	{
+		b2Vec2 faceCenter = 0.5f * (v1 + v2);
+		float32 separation = b2Dot(cLocal - faceCenter, normals[vertIndex1]);
+		if (separation > radius)
+		{
+			return;
+		}
+
+		manifold->pointCount = 1;
+		manifold->type = b2Manifold::e_faceA;
+		manifold->localNormal = normals[vertIndex1];
+		manifold->localPoint = faceCenter;
+		manifold->points[0].localPoint = circleB->m_p;
+		manifold->points[0].id.key = 0;
+	}
+}

deps/include/Box2D/Collision/b2CollideEdge.cpp

@@ -0,0 +1,698 @@
+/*
+ * Copyright (c) 2007-2009 Erin Catto http://www.box2d.org
+ *
+ * This software is provided 'as-is', without any express or implied
+ * warranty.  In no event will the authors be held liable for any damages
+ * arising from the use of this software.
+ * Permission is granted to anyone to use this software for any purpose,
+ * including commercial applications, and to alter it and redistribute it
+ * freely, subject to the following restrictions:
+ * 1. The origin of this software must not be misrepresented; you must not
+ * claim that you wrote the original software. If you use this software
+ * in a product, an acknowledgment in the product documentation would be
+ * appreciated but is not required.
+ * 2. Altered source versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ * 3. This notice may not be removed or altered from any source distribution.
+ */
+
+#include <Box2D/Collision/b2Collision.h>
+#include <Box2D/Collision/Shapes/b2CircleShape.h>
+#include <Box2D/Collision/Shapes/b2EdgeShape.h>
+#include <Box2D/Collision/Shapes/b2PolygonShape.h>
+
+
+// Compute contact points for edge versus circle.
+// This accounts for edge connectivity.
+void b2CollideEdgeAndCircle(b2Manifold* manifold,
+							const b2EdgeShape* edgeA, const b2Transform& xfA,
+							const b2CircleShape* circleB, const b2Transform& xfB)
+{
+	manifold->pointCount = 0;
+	
+	// Compute circle in frame of edge
+	b2Vec2 Q = b2MulT(xfA, b2Mul(xfB, circleB->m_p));
+	
+	b2Vec2 A = edgeA->m_vertex1, B = edgeA->m_vertex2;
+	b2Vec2 e = B - A;
+	
+	// Barycentric coordinates
+	float32 u = b2Dot(e, B - Q);
+	float32 v = b2Dot(e, Q - A);
+	
+	float32 radius = edgeA->m_radius + circleB->m_radius;
+	
+	b2ContactFeature cf;
+	cf.indexB = 0;
+	cf.typeB = b2ContactFeature::e_vertex;
+	
+	// Region A
+	if (v <= 0.0f)
+	{
+		b2Vec2 P = A;
+		b2Vec2 d = Q - P;
+		float32 dd = b2Dot(d, d);
+		if (dd > radius * radius)
+		{
+			return;
+		}
+		
+		// Is there an edge connected to A?
+		if (edgeA->m_hasVertex0)
+		{
+			b2Vec2 A1 = edgeA->m_vertex0;
+			b2Vec2 B1 = A;
+			b2Vec2 e1 = B1 - A1;
+			float32 u1 = b2Dot(e1, B1 - Q);
+			
+			// Is the circle in Region AB of the previous edge?
+			if (u1 > 0.0f)
+			{
+				return;
+			}
+		}
+		
+		cf.indexA = 0;
+		cf.typeA = b2ContactFeature::e_vertex;
+		manifold->pointCount = 1;
+		manifold->type = b2Manifold::e_circles;
+		manifold->localNormal.SetZero();
+		manifold->localPoint = P;
+		manifold->points[0].id.key = 0;
+		manifold->points[0].id.cf = cf;
+		manifold->points[0].localPoint = circleB->m_p;
+		return;
+	}
+	
+	// Region B
+	if (u <= 0.0f)
+	{
+		b2Vec2 P = B;
+		b2Vec2 d = Q - P;
+		float32 dd = b2Dot(d, d);
+		if (dd > radius * radius)
+		{
+			return;
+		}
+		
+		// Is there an edge connected to B?
+		if (edgeA->m_hasVertex3)
+		{
+			b2Vec2 B2 = edgeA->m_vertex3;
+			b2Vec2 A2 = B;
+			b2Vec2 e2 = B2 - A2;
+			float32 v2 = b2Dot(e2, Q - A2);
+			
+			// Is the circle in Region AB of the next edge?
+			if (v2 > 0.0f)
+			{
+				return;
+			}
+		}
+		
+		cf.indexA = 1;
+		cf.typeA = b2ContactFeature::e_vertex;
+		manifold->pointCount = 1;
+		manifold->type = b2Manifold::e_circles;
+		manifold->localNormal.SetZero();
+		manifold->localPoint = P;
+		manifold->points[0].id.key = 0;
+		manifold->points[0].id.cf = cf;
+		manifold->points[0].localPoint = circleB->m_p;
+		return;
+	}
+	
+	// Region AB
+	float32 den = b2Dot(e, e);
+	b2Assert(den > 0.0f);
+	b2Vec2 P = (1.0f / den) * (u * A + v * B);
+	b2Vec2 d = Q - P;
+	float32 dd = b2Dot(d, d);
+	if (dd > radius * radius)
+	{
+		return;
+	}
+	
+	b2Vec2 n(-e.y, e.x);
+	if (b2Dot(n, Q - A) < 0.0f)
+	{
+		n.Set(-n.x, -n.y);
+	}
+	n.Normalize();
+	
+	cf.indexA = 0;
+	cf.typeA = b2ContactFeature::e_face;
+	manifold->pointCount = 1;
+	manifold->type = b2Manifold::e_faceA;
+	manifold->localNormal = n;
+	manifold->localPoint = A;
+	manifold->points[0].id.key = 0;
+	manifold->points[0].id.cf = cf;
+	manifold->points[0].localPoint = circleB->m_p;
+}
+
+// This structure is used to keep track of the best separating axis.
+struct b2EPAxis
+{
+	enum Type
+	{
+		e_unknown,
+		e_edgeA,
+		e_edgeB
+	};
+	
+	Type type;
+	int32 index;
+	float32 separation;
+};
+
+// This holds polygon B expressed in frame A.
+struct b2TempPolygon
+{
+	b2Vec2 vertices[b2_maxPolygonVertices];
+	b2Vec2 normals[b2_maxPolygonVertices];
+	int32 count;
+};
+
+// Reference face used for clipping
+struct b2ReferenceFace
+{
+	int32 i1, i2;
+	
+	b2Vec2 v1, v2;
+	
+	b2Vec2 normal;
+	
+	b2Vec2 sideNormal1;
+	float32 sideOffset1;
+	
+	b2Vec2 sideNormal2;
+	float32 sideOffset2;
+};
+
+// This class collides and edge and a polygon, taking into account edge adjacency.
+struct b2EPCollider
+{
+	void Collide(b2Manifold* manifold, const b2EdgeShape* edgeA, const b2Transform& xfA,
+				 const b2PolygonShape* polygonB, const b2Transform& xfB);
+	b2EPAxis ComputeEdgeSeparation();
+	b2EPAxis ComputePolygonSeparation();
+	
+	enum VertexType
+	{
+		e_isolated,
+		e_concave,
+		e_convex
+	};
+	
+	b2TempPolygon m_polygonB;
+	
+	b2Transform m_xf;
+	b2Vec2 m_centroidB;
+	b2Vec2 m_v0, m_v1, m_v2, m_v3;
+	b2Vec2 m_normal0, m_normal1, m_normal2;
+	b2Vec2 m_normal;
+	VertexType m_type1, m_type2;
+	b2Vec2 m_lowerLimit, m_upperLimit;
+	float32 m_radius;
+	bool m_front;
+};
+
+// Algorithm:
+// 1. Classify v1 and v2
+// 2. Classify polygon centroid as front or back
+// 3. Flip normal if necessary
+// 4. Initialize normal range to [-pi, pi] about face normal
+// 5. Adjust normal range according to adjacent edges
+// 6. Visit each separating axes, only accept axes within the range
+// 7. Return if _any_ axis indicates separation
+// 8. Clip
+void b2EPCollider::Collide(b2Manifold* manifold, const b2EdgeShape* edgeA, const b2Transform& xfA,
+						   const b2PolygonShape* polygonB, const b2Transform& xfB)
+{
+	m_xf = b2MulT(xfA, xfB);
+	
+	m_centroidB = b2Mul(m_xf, polygonB->m_centroid);
+	
+	m_v0 = edgeA->m_vertex0;
+	m_v1 = edgeA->m_vertex1;
+	m_v2 = edgeA->m_vertex2;
+	m_v3 = edgeA->m_vertex3;
+	
+	bool hasVertex0 = edgeA->m_hasVertex0;
+	bool hasVertex3 = edgeA->m_hasVertex3;
+	
+	b2Vec2 edge1 = m_v2 - m_v1;
+	edge1.Normalize();
+	m_normal1.Set(edge1.y, -edge1.x);
+	float32 offset1 = b2Dot(m_normal1, m_centroidB - m_v1);
+	float32 offset0 = 0.0f, offset2 = 0.0f;
+	bool convex1 = false, convex2 = false;
+	
+	// Is there a preceding edge?
+	if (hasVertex0)
+	{
+		b2Vec2 edge0 = m_v1 - m_v0;
+		edge0.Normalize();
+		m_normal0.Set(edge0.y, -edge0.x);
+		convex1 = b2Cross(edge0, edge1) >= 0.0f;
+		offset0 = b2Dot(m_normal0, m_centroidB - m_v0);
+	}
+	
+	// Is there a following edge?
+	if (hasVertex3)
+	{
+		b2Vec2 edge2 = m_v3 - m_v2;
+		edge2.Normalize();
+		m_normal2.Set(edge2.y, -edge2.x);
+		convex2 = b2Cross(edge1, edge2) > 0.0f;
+		offset2 = b2Dot(m_normal2, m_centroidB - m_v2);
+	}
+	
+	// Determine front or back collision. Determine collision normal limits.
+	if (hasVertex0 && hasVertex3)
+	{
+		if (convex1 && convex2)
+		{
+			m_front = offset0 >= 0.0f || offset1 >= 0.0f || offset2 >= 0.0f;
+			if (m_front)
+			{
+				m_normal = m_normal1;
+				m_lowerLimit = m_normal0;
+				m_upperLimit = m_normal2;
+			}
+			else
+			{
+				m_normal = -m_normal1;
+				m_lowerLimit = -m_normal1;
+				m_upperLimit = -m_normal1;
+			}
+		}
+		else if (convex1)
+		{
+			m_front = offset0 >= 0.0f || (offset1 >= 0.0f && offset2 >= 0.0f);
+			if (m_front)
+			{
+				m_normal = m_normal1;
+				m_lowerLimit = m_normal0;
+				m_upperLimit = m_normal1;
+			}
+			else
+			{
+				m_normal = -m_normal1;
+				m_lowerLimit = -m_normal2;
+				m_upperLimit = -m_normal1;
+			}
+		}
+		else if (convex2)
+		{
+			m_front = offset2 >= 0.0f || (offset0 >= 0.0f && offset1 >= 0.0f);
+			if (m_front)
+			{
+				m_normal = m_normal1;
+				m_lowerLimit = m_normal1;
+				m_upperLimit = m_normal2;
+			}
+			else
+			{
+				m_normal = -m_normal1;
+				m_lowerLimit = -m_normal1;
+				m_upperLimit = -m_normal0;
+			}
+		}
+		else
+		{
+			m_front = offset0 >= 0.0f && offset1 >= 0.0f && offset2 >= 0.0f;
+			if (m_front)
+			{
+				m_normal = m_normal1;
+				m_lowerLimit = m_normal1;
+				m_upperLimit = m_normal1;
+			}
+			else
+			{
+				m_normal = -m_normal1;
+				m_lowerLimit = -m_normal2;
+				m_upperLimit = -m_normal0;
+			}
+		}
+	}
+	else if (hasVertex0)
+	{
+		if (convex1)
+		{
+			m_front = offset0 >= 0.0f || offset1 >= 0.0f;
+			if (m_front)
+			{
+				m_normal = m_normal1;
+				m_lowerLimit = m_normal0;
+				m_upperLimit = -m_normal1;
+			}
+			else
+			{
+				m_normal = -m_normal1;
+				m_lowerLimit = m_normal1;
+				m_upperLimit = -m_normal1;
+			}
+		}
+		else
+		{
+			m_front = offset0 >= 0.0f && offset1 >= 0.0f;
+			if (m_front)
+			{
+				m_normal = m_normal1;
+				m_lowerLimit = m_normal1;
+				m_upperLimit = -m_normal1;
+			}
+			else
+			{
+				m_normal = -m_normal1;
+				m_lowerLimit = m_normal1;
+				m_upperLimit = -m_normal0;
+			}
+		}
+	}
+	else if (hasVertex3)
+	{
+		if (convex2)
+		{
+			m_front = offset1 >= 0.0f || offset2 >= 0.0f;
+			if (m_front)
+			{
+				m_normal = m_normal1;
+				m_lowerLimit = -m_normal1;
+				m_upperLimit = m_normal2;
+			}
+			else
+			{
+				m_normal = -m_normal1;
+				m_lowerLimit = -m_normal1;
+				m_upperLimit = m_normal1;
+			}
+		}
+		else
+		{
+			m_front = offset1 >= 0.0f && offset2 >= 0.0f;
+			if (m_front)
+			{
+				m_normal = m_normal1;
+				m_lowerLimit = -m_normal1;
+				m_upperLimit = m_normal1;
+			}
+			else
+			{
+				m_normal = -m_normal1;
+				m_lowerLimit = -m_normal2;
+				m_upperLimit = m_normal1;
+			}
+		}		
+	}
+	else
+	{
+		m_front = offset1 >= 0.0f;
+		if (m_front)
+		{
+			m_normal = m_normal1;
+			m_lowerLimit = -m_normal1;
+			m_upperLimit = -m_normal1;
+		}
+		else
+		{
+			m_normal = -m_normal1;
+			m_lowerLimit = m_normal1;
+			m_upperLimit = m_normal1;
+		}
+	}
+	
+	// Get polygonB in frameA
+	m_polygonB.count = polygonB->m_count;
+	for (int32 i = 0; i < polygonB->m_count; ++i)
+	{
+		m_polygonB.vertices[i] = b2Mul(m_xf, polygonB->m_vertices[i]);
+		m_polygonB.normals[i] = b2Mul(m_xf.q, polygonB->m_normals[i]);
+	}
+	
+	m_radius = 2.0f * b2_polygonRadius;
+	
+	manifold->pointCount = 0;
+	
+	b2EPAxis edgeAxis = ComputeEdgeSeparation();
+	
+	// If no valid normal can be found than this edge should not collide.
+	if (edgeAxis.type == b2EPAxis::e_unknown)
+	{
+		return;
+	}
+	
+	if (edgeAxis.separation > m_radius)
+	{
+		return;
+	}
+	
+	b2EPAxis polygonAxis = ComputePolygonSeparation();
+	if (polygonAxis.type != b2EPAxis::e_unknown && polygonAxis.separation > m_radius)
+	{
+		return;
+	}
+	
+	// Use hysteresis for jitter reduction.
+	const float32 k_relativeTol = 0.98f;
+	const float32 k_absoluteTol = 0.001f;
+	
+	b2EPAxis primaryAxis;
+	if (polygonAxis.type == b2EPAxis::e_unknown)
+	{
+		primaryAxis = edgeAxis;
+	}
+	else if (polygonAxis.separation > k_relativeTol * edgeAxis.separation + k_absoluteTol)
+	{
+		primaryAxis = polygonAxis;
+	}
+	else
+	{
+		primaryAxis = edgeAxis;
+	}
+	
+	b2ClipVertex ie[2];
+	b2ReferenceFace rf;
+	if (primaryAxis.type == b2EPAxis::e_edgeA)
+	{
+		manifold->type = b2Manifold::e_faceA;
+		
+		// Search for the polygon normal that is most anti-parallel to the edge normal.
+		int32 bestIndex = 0;
+		float32 bestValue = b2Dot(m_normal, m_polygonB.normals[0]);
+		for (int32 i = 1; i < m_polygonB.count; ++i)
+		{
+			float32 value = b2Dot(m_normal, m_polygonB.normals[i]);
+			if (value < bestValue)
+			{
+				bestValue = value;
+				bestIndex = i;
+			}
+		}
+		
+		int32 i1 = bestIndex;
+		int32 i2 = i1 + 1 < m_polygonB.count ? i1 + 1 : 0;
+		
+		ie[0].v = m_polygonB.vertices[i1];
+		ie[0].id.cf.indexA = 0;
+		ie[0].id.cf.indexB = static_cast<uint8>(i1);
+		ie[0].id.cf.typeA = b2ContactFeature::e_face;
+		ie[0].id.cf.typeB = b2ContactFeature::e_vertex;
+		
+		ie[1].v = m_polygonB.vertices[i2];
+		ie[1].id.cf.indexA = 0;
+		ie[1].id.cf.indexB = static_cast<uint8>(i2);
+		ie[1].id.cf.typeA = b2ContactFeature::e_face;
+		ie[1].id.cf.typeB = b2ContactFeature::e_vertex;
+		
+		if (m_front)
+		{
+			rf.i1 = 0;
+			rf.i2 = 1;
+			rf.v1 = m_v1;
+			rf.v2 = m_v2;
+			rf.normal = m_normal1;
+		}
+		else
+		{
+			rf.i1 = 1;
+			rf.i2 = 0;
+			rf.v1 = m_v2;
+			rf.v2 = m_v1;
+			rf.normal = -m_normal1;
+		}		
+	}
+	else
+	{
+		manifold->type = b2Manifold::e_faceB;
+		
+		ie[0].v = m_v1;
+		ie[0].id.cf.indexA = 0;
+		ie[0].id.cf.indexB = static_cast<uint8>(primaryAxis.index);
+		ie[0].id.cf.typeA = b2ContactFeature::e_vertex;
+		ie[0].id.cf.typeB = b2ContactFeature::e_face;
+		
+		ie[1].v = m_v2;
+		ie[1].id.cf.indexA = 0;
+		ie[1].id.cf.indexB = static_cast<uint8>(primaryAxis.index);		
+		ie[1].id.cf.typeA = b2ContactFeature::e_vertex;
+		ie[1].id.cf.typeB = b2ContactFeature::e_face;
+		
+		rf.i1 = primaryAxis.index;
+		rf.i2 = rf.i1 + 1 < m_polygonB.count ? rf.i1 + 1 : 0;
+		rf.v1 = m_polygonB.vertices[rf.i1];
+		rf.v2 = m_polygonB.vertices[rf.i2];
+		rf.normal = m_polygonB.normals[rf.i1];
+	}
+	
+	rf.sideNormal1.Set(rf.normal.y, -rf.normal.x);
+	rf.sideNormal2 = -rf.sideNormal1;
+	rf.sideOffset1 = b2Dot(rf.sideNormal1, rf.v1);
+	rf.sideOffset2 = b2Dot(rf.sideNormal2, rf.v2);
+	
+	// Clip incident edge against extruded edge1 side edges.
+	b2ClipVertex clipPoints1[2];
+	b2ClipVertex clipPoints2[2];
+	int32 np;
+	
+	// Clip to box side 1
+	np = b2ClipSegmentToLine(clipPoints1, ie, rf.sideNormal1, rf.sideOffset1, rf.i1);
+	
+	if (np < b2_maxManifoldPoints)
+	{
+		return;
+	}
+	
+	// Clip to negative box side 1
+	np = b2ClipSegmentToLine(clipPoints2, clipPoints1, rf.sideNormal2, rf.sideOffset2, rf.i2);
+	
+	if (np < b2_maxManifoldPoints)
+	{
+		return;
+	}
+	
+	// Now clipPoints2 contains the clipped points.
+	if (primaryAxis.type == b2EPAxis::e_edgeA)
+	{
+		manifold->localNormal = rf.normal;
+		manifold->localPoint = rf.v1;
+	}
+	else
+	{
+		manifold->localNormal = polygonB->m_normals[rf.i1];
+		manifold->localPoint = polygonB->m_vertices[rf.i1];
+	}
+	
+	int32 pointCount = 0;
+	for (int32 i = 0; i < b2_maxManifoldPoints; ++i)
+	{
+		float32 separation;
+		
+		separation = b2Dot(rf.normal, clipPoints2[i].v - rf.v1);
+		
+		if (separation <= m_radius)
+		{
+			b2ManifoldPoint* cp = manifold->points + pointCount;
+			
+			if (primaryAxis.type == b2EPAxis::e_edgeA)
+			{
+				cp->localPoint = b2MulT(m_xf, clipPoints2[i].v);
+				cp->id = clipPoints2[i].id;
+			}
+			else
+			{
+				cp->localPoint = clipPoints2[i].v;
+				cp->id.cf.typeA = clipPoints2[i].id.cf.typeB;
+				cp->id.cf.typeB = clipPoints2[i].id.cf.typeA;
+				cp->id.cf.indexA = clipPoints2[i].id.cf.indexB;
+				cp->id.cf.indexB = clipPoints2[i].id.cf.indexA;
+			}
+			
+			++pointCount;
+		}
+	}
+	
+	manifold->pointCount = pointCount;
+}
+
+b2EPAxis b2EPCollider::ComputeEdgeSeparation()
+{
+	b2EPAxis axis;
+	axis.type = b2EPAxis::e_edgeA;
+	axis.index = m_front ? 0 : 1;
+	axis.separation = FLT_MAX;
+	
+	for (int32 i = 0; i < m_polygonB.count; ++i)
+	{
+		float32 s = b2Dot(m_normal, m_polygonB.vertices[i] - m_v1);
+		if (s < axis.separation)
+		{
+			axis.separation = s;
+		}
+	}
+	
+	return axis;
+}
+
+b2EPAxis b2EPCollider::ComputePolygonSeparation()
+{
+	b2EPAxis axis;
+	axis.type = b2EPAxis::e_unknown;
+	axis.index = -1;
+	axis.separation = -FLT_MAX;
+
+	b2Vec2 perp(-m_normal.y, m_normal.x);
+
+	for (int32 i = 0; i < m_polygonB.count; ++i)
+	{
+		b2Vec2 n = -m_polygonB.normals[i];
+		
+		float32 s1 = b2Dot(n, m_polygonB.vertices[i] - m_v1);
+		float32 s2 = b2Dot(n, m_polygonB.vertices[i] - m_v2);
+		float32 s = b2Min(s1, s2);
+		
+		if (s > m_radius)
+		{
+			// No collision
+			axis.type = b2EPAxis::e_edgeB;
+			axis.index = i;
+			axis.separation = s;
+			return axis;
+		}
+		
+		// Adjacency
+		if (b2Dot(n, perp) >= 0.0f)
+		{
+			if (b2Dot(n - m_upperLimit, m_normal) < -b2_angularSlop)
+			{
+				continue;
+			}
+		}
+		else
+		{
+			if (b2Dot(n - m_lowerLimit, m_normal) < -b2_angularSlop)
+			{
+				continue;
+			}
+		}
+		
+		if (s > axis.separation)
+		{
+			axis.type = b2EPAxis::e_edgeB;
+			axis.index = i;
+			axis.separation = s;
+		}
+	}
+	
+	return axis;
+}
+
+void b2CollideEdgeAndPolygon(	b2Manifold* manifold,
+							 const b2EdgeShape* edgeA, const b2Transform& xfA,
+							 const b2PolygonShape* polygonB, const b2Transform& xfB)
+{
+	b2EPCollider collider;
+	collider.Collide(manifold, edgeA, xfA, polygonB, xfB);
+}

deps/include/Box2D/Collision/b2CollidePolygon.cpp

@@ -0,0 +1,239 @@
+/*
+* Copyright (c) 2006-2009 Erin Catto http://www.box2d.org
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#include <Box2D/Collision/b2Collision.h>
+#include <Box2D/Collision/Shapes/b2PolygonShape.h>
+
+// Find the max separation between poly1 and poly2 using edge normals from poly1.
+static float32 b2FindMaxSeparation(int32* edgeIndex,
+								 const b2PolygonShape* poly1, const b2Transform& xf1,
+								 const b2PolygonShape* poly2, const b2Transform& xf2)
+{
+	int32 count1 = poly1->m_count;
+	int32 count2 = poly2->m_count;
+	const b2Vec2* n1s = poly1->m_normals;
+	const b2Vec2* v1s = poly1->m_vertices;
+	const b2Vec2* v2s = poly2->m_vertices;
+	b2Transform xf = b2MulT(xf2, xf1);
+
+	int32 bestIndex = 0;
+	float32 maxSeparation = -b2_maxFloat;
+	for (int32 i = 0; i < count1; ++i)
+	{
+		// Get poly1 normal in frame2.
+		b2Vec2 n = b2Mul(xf.q, n1s[i]);
+		b2Vec2 v1 = b2Mul(xf, v1s[i]);
+
+		// Find deepest point for normal i.
+		float32 si = b2_maxFloat;
+		for (int32 j = 0; j < count2; ++j)
+		{
+			float32 sij = b2Dot(n, v2s[j] - v1);
+			if (sij < si)
+			{
+				si = sij;
+			}
+		}
+
+		if (si > maxSeparation)
+		{
+			maxSeparation = si;
+			bestIndex = i;
+		}
+	}
+
+	*edgeIndex = bestIndex;
+	return maxSeparation;
+}
+
+static void b2FindIncidentEdge(b2ClipVertex c[2],
+							 const b2PolygonShape* poly1, const b2Transform& xf1, int32 edge1,
+							 const b2PolygonShape* poly2, const b2Transform& xf2)
+{
+	const b2Vec2* normals1 = poly1->m_normals;
+
+	int32 count2 = poly2->m_count;
+	const b2Vec2* vertices2 = poly2->m_vertices;
+	const b2Vec2* normals2 = poly2->m_normals;
+
+	b2Assert(0 <= edge1 && edge1 < poly1->m_count);
+
+	// Get the normal of the reference edge in poly2's frame.
+	b2Vec2 normal1 = b2MulT(xf2.q, b2Mul(xf1.q, normals1[edge1]));
+
+	// Find the incident edge on poly2.
+	int32 index = 0;
+	float32 minDot = b2_maxFloat;
+	for (int32 i = 0; i < count2; ++i)
+	{
+		float32 dot = b2Dot(normal1, normals2[i]);
+		if (dot < minDot)
+		{
+			minDot = dot;
+			index = i;
+		}
+	}
+
+	// Build the clip vertices for the incident edge.
+	int32 i1 = index;
+	int32 i2 = i1 + 1 < count2 ? i1 + 1 : 0;
+
+	c[0].v = b2Mul(xf2, vertices2[i1]);
+	c[0].id.cf.indexA = (uint8)edge1;
+	c[0].id.cf.indexB = (uint8)i1;
+	c[0].id.cf.typeA = b2ContactFeature::e_face;
+	c[0].id.cf.typeB = b2ContactFeature::e_vertex;
+
+	c[1].v = b2Mul(xf2, vertices2[i2]);
+	c[1].id.cf.indexA = (uint8)edge1;
+	c[1].id.cf.indexB = (uint8)i2;
+	c[1].id.cf.typeA = b2ContactFeature::e_face;
+	c[1].id.cf.typeB = b2ContactFeature::e_vertex;
+}
+
+// Find edge normal of max separation on A - return if separating axis is found
+// Find edge normal of max separation on B - return if separation axis is found
+// Choose reference edge as min(minA, minB)
+// Find incident edge
+// Clip
+
+// The normal points from 1 to 2
+void b2CollidePolygons(b2Manifold* manifold,
+					  const b2PolygonShape* polyA, const b2Transform& xfA,
+					  const b2PolygonShape* polyB, const b2Transform& xfB)
+{
+	manifold->pointCount = 0;
+	float32 totalRadius = polyA->m_radius + polyB->m_radius;
+
+	int32 edgeA = 0;
+	float32 separationA = b2FindMaxSeparation(&edgeA, polyA, xfA, polyB, xfB);
+	if (separationA > totalRadius)
+		return;
+
+	int32 edgeB = 0;
+	float32 separationB = b2FindMaxSeparation(&edgeB, polyB, xfB, polyA, xfA);
+	if (separationB > totalRadius)
+		return;
+
+	const b2PolygonShape* poly1;	// reference polygon
+	const b2PolygonShape* poly2;	// incident polygon
+	b2Transform xf1, xf2;
+	int32 edge1;					// reference edge
+	uint8 flip;
+	const float32 k_tol = 0.1f * b2_linearSlop;
+
+	if (separationB > separationA + k_tol)
+	{
+		poly1 = polyB;
+		poly2 = polyA;
+		xf1 = xfB;
+		xf2 = xfA;
+		edge1 = edgeB;
+		manifold->type = b2Manifold::e_faceB;
+		flip = 1;
+	}
+	else
+	{
+		poly1 = polyA;
+		poly2 = polyB;
+		xf1 = xfA;
+		xf2 = xfB;
+		edge1 = edgeA;
+		manifold->type = b2Manifold::e_faceA;
+		flip = 0;
+	}
+
+	b2ClipVertex incidentEdge[2];
+	b2FindIncidentEdge(incidentEdge, poly1, xf1, edge1, poly2, xf2);
+
+	int32 count1 = poly1->m_count;
+	const b2Vec2* vertices1 = poly1->m_vertices;
+
+	int32 iv1 = edge1;
+	int32 iv2 = edge1 + 1 < count1 ? edge1 + 1 : 0;
+
+	b2Vec2 v11 = vertices1[iv1];
+	b2Vec2 v12 = vertices1[iv2];
+
+	b2Vec2 localTangent = v12 - v11;
+	localTangent.Normalize();
+	
+	b2Vec2 localNormal = b2Cross(localTangent, 1.0f);
+	b2Vec2 planePoint = 0.5f * (v11 + v12);
+
+	b2Vec2 tangent = b2Mul(xf1.q, localTangent);
+	b2Vec2 normal = b2Cross(tangent, 1.0f);
+	
+	v11 = b2Mul(xf1, v11);
+	v12 = b2Mul(xf1, v12);
+
+	// Face offset.
+	float32 frontOffset = b2Dot(normal, v11);
+
+	// Side offsets, extended by polytope skin thickness.
+	float32 sideOffset1 = -b2Dot(tangent, v11) + totalRadius;
+	float32 sideOffset2 = b2Dot(tangent, v12) + totalRadius;
+
+	// Clip incident edge against extruded edge1 side edges.
+	b2ClipVertex clipPoints1[2];
+	b2ClipVertex clipPoints2[2];
+	int np;
+
+	// Clip to box side 1
+	np = b2ClipSegmentToLine(clipPoints1, incidentEdge, -tangent, sideOffset1, iv1);
+
+	if (np < 2)
+		return;
+
+	// Clip to negative box side 1
+	np = b2ClipSegmentToLine(clipPoints2, clipPoints1,  tangent, sideOffset2, iv2);
+
+	if (np < 2)
+	{
+		return;
+	}
+
+	// Now clipPoints2 contains the clipped points.
+	manifold->localNormal = localNormal;
+	manifold->localPoint = planePoint;
+
+	int32 pointCount = 0;
+	for (int32 i = 0; i < b2_maxManifoldPoints; ++i)
+	{
+		float32 separation = b2Dot(normal, clipPoints2[i].v) - frontOffset;
+
+		if (separation <= totalRadius)
+		{
+			b2ManifoldPoint* cp = manifold->points + pointCount;
+			cp->localPoint = b2MulT(xf2, clipPoints2[i].v);
+			cp->id = clipPoints2[i].id;
+			if (flip)
+			{
+				// Swap features
+				b2ContactFeature cf = cp->id.cf;
+				cp->id.cf.indexA = cf.indexB;
+				cp->id.cf.indexB = cf.indexA;
+				cp->id.cf.typeA = cf.typeB;
+				cp->id.cf.typeB = cf.typeA;
+			}
+			++pointCount;
+		}
+	}
+
+	manifold->pointCount = pointCount;
+}

deps/include/Box2D/Collision/b2Collision.cpp

@@ -0,0 +1,252 @@
+/*
+* Copyright (c) 2007-2009 Erin Catto http://www.box2d.org
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#include <Box2D/Collision/b2Collision.h>
+#include <Box2D/Collision/b2Distance.h>
+
+void b2WorldManifold::Initialize(const b2Manifold* manifold,
+						  const b2Transform& xfA, float32 radiusA,
+						  const b2Transform& xfB, float32 radiusB)
+{
+	if (manifold->pointCount == 0)
+	{
+		return;
+	}
+
+	switch (manifold->type)
+	{
+	case b2Manifold::e_circles:
+		{
+			normal.Set(1.0f, 0.0f);
+			b2Vec2 pointA = b2Mul(xfA, manifold->localPoint);
+			b2Vec2 pointB = b2Mul(xfB, manifold->points[0].localPoint);
+			if (b2DistanceSquared(pointA, pointB) > b2_epsilon * b2_epsilon)
+			{
+				normal = pointB - pointA;
+				normal.Normalize();
+			}
+
+			b2Vec2 cA = pointA + radiusA * normal;
+			b2Vec2 cB = pointB - radiusB * normal;
+			points[0] = 0.5f * (cA + cB);
+			separations[0] = b2Dot(cB - cA, normal);
+		}
+		break;
+
+	case b2Manifold::e_faceA:
+		{
+			normal = b2Mul(xfA.q, manifold->localNormal);
+			b2Vec2 planePoint = b2Mul(xfA, manifold->localPoint);
+			
+			for (int32 i = 0; i < manifold->pointCount; ++i)
+			{
+				b2Vec2 clipPoint = b2Mul(xfB, manifold->points[i].localPoint);
+				b2Vec2 cA = clipPoint + (radiusA - b2Dot(clipPoint - planePoint, normal)) * normal;
+				b2Vec2 cB = clipPoint - radiusB * normal;
+				points[i] = 0.5f * (cA + cB);
+				separations[i] = b2Dot(cB - cA, normal);
+			}
+		}
+		break;
+
+	case b2Manifold::e_faceB:
+		{
+			normal = b2Mul(xfB.q, manifold->localNormal);
+			b2Vec2 planePoint = b2Mul(xfB, manifold->localPoint);
+
+			for (int32 i = 0; i < manifold->pointCount; ++i)
+			{
+				b2Vec2 clipPoint = b2Mul(xfA, manifold->points[i].localPoint);
+				b2Vec2 cB = clipPoint + (radiusB - b2Dot(clipPoint - planePoint, normal)) * normal;
+				b2Vec2 cA = clipPoint - radiusA * normal;
+				points[i] = 0.5f * (cA + cB);
+				separations[i] = b2Dot(cA - cB, normal);
+			}
+
+			// Ensure normal points from A to B.
+			normal = -normal;
+		}
+		break;
+	}
+}
+
+void b2GetPointStates(b2PointState state1[b2_maxManifoldPoints], b2PointState state2[b2_maxManifoldPoints],
+					  const b2Manifold* manifold1, const b2Manifold* manifold2)
+{
+	for (int32 i = 0; i < b2_maxManifoldPoints; ++i)
+	{
+		state1[i] = b2_nullState;
+		state2[i] = b2_nullState;
+	}
+
+	// Detect persists and removes.
+	for (int32 i = 0; i < manifold1->pointCount; ++i)
+	{
+		b2ContactID id = manifold1->points[i].id;
+
+		state1[i] = b2_removeState;
+
+		for (int32 j = 0; j < manifold2->pointCount; ++j)
+		{
+			if (manifold2->points[j].id.key == id.key)
+			{
+				state1[i] = b2_persistState;
+				break;
+			}
+		}
+	}
+
+	// Detect persists and adds.
+	for (int32 i = 0; i < manifold2->pointCount; ++i)
+	{
+		b2ContactID id = manifold2->points[i].id;
+
+		state2[i] = b2_addState;
+
+		for (int32 j = 0; j < manifold1->pointCount; ++j)
+		{
+			if (manifold1->points[j].id.key == id.key)
+			{
+				state2[i] = b2_persistState;
+				break;
+			}
+		}
+	}
+}
+
+// From Real-time Collision Detection, p179.
+bool b2AABB::RayCast(b2RayCastOutput* output, const b2RayCastInput& input) const
+{
+	float32 tmin = -b2_maxFloat;
+	float32 tmax = b2_maxFloat;
+
+	b2Vec2 p = input.p1;
+	b2Vec2 d = input.p2 - input.p1;
+	b2Vec2 absD = b2Abs(d);
+
+	b2Vec2 normal;
+
+	for (int32 i = 0; i < 2; ++i)
+	{
+		if (absD(i) < b2_epsilon)
+		{
+			// Parallel.
+			if (p(i) < lowerBound(i) || upperBound(i) < p(i))
+			{
+				return false;
+			}
+		}
+		else
+		{
+			float32 inv_d = 1.0f / d(i);
+			float32 t1 = (lowerBound(i) - p(i)) * inv_d;
+			float32 t2 = (upperBound(i) - p(i)) * inv_d;
+
+			// Sign of the normal vector.
+			float32 s = -1.0f;
+
+			if (t1 > t2)
+			{
+				b2Swap(t1, t2);
+				s = 1.0f;
+			}
+
+			// Push the min up
+			if (t1 > tmin)
+			{
+				normal.SetZero();
+				normal(i) = s;
+				tmin = t1;
+			}
+
+			// Pull the max down
+			tmax = b2Min(tmax, t2);
+
+			if (tmin > tmax)
+			{
+				return false;
+			}
+		}
+	}
+
+	// Does the ray start inside the box?
+	// Does the ray intersect beyond the max fraction?
+	if (tmin < 0.0f || input.maxFraction < tmin)
+	{
+		return false;
+	}
+
+	// Intersection.
+	output->fraction = tmin;
+	output->normal = normal;
+	return true;
+}
+
+// Sutherland-Hodgman clipping.
+int32 b2ClipSegmentToLine(b2ClipVertex vOut[2], const b2ClipVertex vIn[2],
+						const b2Vec2& normal, float32 offset, int32 vertexIndexA)
+{
+	// Start with no output points
+	int32 numOut = 0;
+
+	// Calculate the distance of end points to the line
+	float32 distance0 = b2Dot(normal, vIn[0].v) - offset;
+	float32 distance1 = b2Dot(normal, vIn[1].v) - offset;
+
+	// If the points are behind the plane
+	if (distance0 <= 0.0f) vOut[numOut++] = vIn[0];
+	if (distance1 <= 0.0f) vOut[numOut++] = vIn[1];
+
+	// If the points are on different sides of the plane
+	if (distance0 * distance1 < 0.0f)
+	{
+		// Find intersection point of edge and plane
+		float32 interp = distance0 / (distance0 - distance1);
+		vOut[numOut].v = vIn[0].v + interp * (vIn[1].v - vIn[0].v);
+
+		// VertexA is hitting edgeB.
+		vOut[numOut].id.cf.indexA = static_cast<uint8>(vertexIndexA);
+		vOut[numOut].id.cf.indexB = vIn[0].id.cf.indexB;
+		vOut[numOut].id.cf.typeA = b2ContactFeature::e_vertex;
+		vOut[numOut].id.cf.typeB = b2ContactFeature::e_face;
+		++numOut;
+	}
+
+	return numOut;
+}
+
+bool b2TestOverlap(	const b2Shape* shapeA, int32 indexA,
+					const b2Shape* shapeB, int32 indexB,
+					const b2Transform& xfA, const b2Transform& xfB)
+{
+	b2DistanceInput input;
+	input.proxyA.Set(shapeA, indexA);
+	input.proxyB.Set(shapeB, indexB);
+	input.transformA = xfA;
+	input.transformB = xfB;
+	input.useRadii = true;
+
+	b2SimplexCache cache;
+	cache.count = 0;
+
+	b2DistanceOutput output;
+
+	b2Distance(&output, &cache, &input);
+
+	return output.distance < 10.0f * b2_epsilon;
+}

deps/include/Box2D/Collision/b2Collision.h

@@ -0,0 +1,277 @@
+/*
+* Copyright (c) 2006-2009 Erin Catto http://www.box2d.org
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#ifndef B2_COLLISION_H
+#define B2_COLLISION_H
+
+#include <Box2D/Common/b2Math.h>
+#include <limits.h>
+
+/// @file
+/// Structures and functions used for computing contact points, distance
+/// queries, and TOI queries.
+
+class b2Shape;
+class b2CircleShape;
+class b2EdgeShape;
+class b2PolygonShape;
+
+const uint8 b2_nullFeature = UCHAR_MAX;
+
+/// The features that intersect to form the contact point
+/// This must be 4 bytes or less.
+struct b2ContactFeature
+{
+	enum Type
+	{
+		e_vertex = 0,
+		e_face = 1
+	};
+
+	uint8 indexA;		///< Feature index on shapeA
+	uint8 indexB;		///< Feature index on shapeB
+	uint8 typeA;		///< The feature type on shapeA
+	uint8 typeB;		///< The feature type on shapeB
+};
+
+/// Contact ids to facilitate warm starting.
+union b2ContactID
+{
+	b2ContactFeature cf;
+	uint32 key;					///< Used to quickly compare contact ids.
+};
+
+/// A manifold point is a contact point belonging to a contact
+/// manifold. It holds details related to the geometry and dynamics
+/// of the contact points.
+/// The local point usage depends on the manifold type:
+/// -e_circles: the local center of circleB
+/// -e_faceA: the local center of cirlceB or the clip point of polygonB
+/// -e_faceB: the clip point of polygonA
+/// This structure is stored across time steps, so we keep it small.
+/// Note: the impulses are used for internal caching and may not
+/// provide reliable contact forces, especially for high speed collisions.
+struct b2ManifoldPoint
+{
+	b2Vec2 localPoint;		///< usage depends on manifold type
+	float32 normalImpulse;	///< the non-penetration impulse
+	float32 tangentImpulse;	///< the friction impulse
+	b2ContactID id;			///< uniquely identifies a contact point between two shapes
+};
+
+/// A manifold for two touching convex shapes.
+/// Box2D supports multiple types of contact:
+/// - clip point versus plane with radius
+/// - point versus point with radius (circles)
+/// The local point usage depends on the manifold type:
+/// -e_circles: the local center of circleA
+/// -e_faceA: the center of faceA
+/// -e_faceB: the center of faceB
+/// Similarly the local normal usage:
+/// -e_circles: not used
+/// -e_faceA: the normal on polygonA
+/// -e_faceB: the normal on polygonB
+/// We store contacts in this way so that position correction can
+/// account for movement, which is critical for continuous physics.
+/// All contact scenarios must be expressed in one of these types.
+/// This structure is stored across time steps, so we keep it small.
+struct b2Manifold
+{
+	enum Type
+	{
+		e_circles,
+		e_faceA,
+		e_faceB
+	};
+
+	b2ManifoldPoint points[b2_maxManifoldPoints];	///< the points of contact
+	b2Vec2 localNormal;								///< not use for Type::e_points
+	b2Vec2 localPoint;								///< usage depends on manifold type
+	Type type;
+	int32 pointCount;								///< the number of manifold points
+};
+
+/// This is used to compute the current state of a contact manifold.
+struct b2WorldManifold
+{
+	/// Evaluate the manifold with supplied transforms. This assumes
+	/// modest motion from the original state. This does not change the
+	/// point count, impulses, etc. The radii must come from the shapes
+	/// that generated the manifold.
+	void Initialize(const b2Manifold* manifold,
+					const b2Transform& xfA, float32 radiusA,
+					const b2Transform& xfB, float32 radiusB);
+
+	b2Vec2 normal;								///< world vector pointing from A to B
+	b2Vec2 points[b2_maxManifoldPoints];		///< world contact point (point of intersection)
+	float32 separations[b2_maxManifoldPoints];	///< a negative value indicates overlap, in meters
+};
+
+/// This is used for determining the state of contact points.
+enum b2PointState
+{
+	b2_nullState,		///< point does not exist
+	b2_addState,		///< point was added in the update
+	b2_persistState,	///< point persisted across the update
+	b2_removeState		///< point was removed in the update
+};
+
+/// Compute the point states given two manifolds. The states pertain to the transition from manifold1
+/// to manifold2. So state1 is either persist or remove while state2 is either add or persist.
+void b2GetPointStates(b2PointState state1[b2_maxManifoldPoints], b2PointState state2[b2_maxManifoldPoints],
+					  const b2Manifold* manifold1, const b2Manifold* manifold2);
+
+/// Used for computing contact manifolds.
+struct b2ClipVertex
+{
+	b2Vec2 v;
+	b2ContactID id;
+};
+
+/// Ray-cast input data. The ray extends from p1 to p1 + maxFraction * (p2 - p1).
+struct b2RayCastInput
+{
+	b2Vec2 p1, p2;
+	float32 maxFraction;
+};
+
+/// Ray-cast output data. The ray hits at p1 + fraction * (p2 - p1), where p1 and p2
+/// come from b2RayCastInput.
+struct b2RayCastOutput
+{
+	b2Vec2 normal;
+	float32 fraction;
+};
+
+/// An axis aligned bounding box.
+struct b2AABB
+{
+	/// Verify that the bounds are sorted.
+	bool IsValid() const;
+
+	/// Get the center of the AABB.
+	b2Vec2 GetCenter() const
+	{
+		return 0.5f * (lowerBound + upperBound);
+	}
+
+	/// Get the extents of the AABB (half-widths).
+	b2Vec2 GetExtents() const
+	{
+		return 0.5f * (upperBound - lowerBound);
+	}
+
+	/// Get the perimeter length
+	float32 GetPerimeter() const
+	{
+		float32 wx = upperBound.x - lowerBound.x;
+		float32 wy = upperBound.y - lowerBound.y;
+		return 2.0f * (wx + wy);
+	}
+
+	/// Combine an AABB into this one.
+	void Combine(const b2AABB& aabb)
+	{
+		lowerBound = b2Min(lowerBound, aabb.lowerBound);
+		upperBound = b2Max(upperBound, aabb.upperBound);
+	}
+
+	/// Combine two AABBs into this one.
+	void Combine(const b2AABB& aabb1, const b2AABB& aabb2)
+	{
+		lowerBound = b2Min(aabb1.lowerBound, aabb2.lowerBound);
+		upperBound = b2Max(aabb1.upperBound, aabb2.upperBound);
+	}
+
+	/// Does this aabb contain the provided AABB.
+	bool Contains(const b2AABB& aabb) const
+	{
+		bool result = true;
+		result = result && lowerBound.x <= aabb.lowerBound.x;
+		result = result && lowerBound.y <= aabb.lowerBound.y;
+		result = result && aabb.upperBound.x <= upperBound.x;
+		result = result && aabb.upperBound.y <= upperBound.y;
+		return result;
+	}
+
+	bool RayCast(b2RayCastOutput* output, const b2RayCastInput& input) const;
+
+	b2Vec2 lowerBound;	///< the lower vertex
+	b2Vec2 upperBound;	///< the upper vertex
+};
+
+/// Compute the collision manifold between two circles.
+void b2CollideCircles(b2Manifold* manifold,
+					  const b2CircleShape* circleA, const b2Transform& xfA,
+					  const b2CircleShape* circleB, const b2Transform& xfB);
+
+/// Compute the collision manifold between a polygon and a circle.
+void b2CollidePolygonAndCircle(b2Manifold* manifold,
+							   const b2PolygonShape* polygonA, const b2Transform& xfA,
+							   const b2CircleShape* circleB, const b2Transform& xfB);
+
+/// Compute the collision manifold between two polygons.
+void b2CollidePolygons(b2Manifold* manifold,
+					   const b2PolygonShape* polygonA, const b2Transform& xfA,
+					   const b2PolygonShape* polygonB, const b2Transform& xfB);
+
+/// Compute the collision manifold between an edge and a circle.
+void b2CollideEdgeAndCircle(b2Manifold* manifold,
+							   const b2EdgeShape* polygonA, const b2Transform& xfA,
+							   const b2CircleShape* circleB, const b2Transform& xfB);
+
+/// Compute the collision manifold between an edge and a circle.
+void b2CollideEdgeAndPolygon(b2Manifold* manifold,
+							   const b2EdgeShape* edgeA, const b2Transform& xfA,
+							   const b2PolygonShape* circleB, const b2Transform& xfB);
+
+/// Clipping for contact manifolds.
+int32 b2ClipSegmentToLine(b2ClipVertex vOut[2], const b2ClipVertex vIn[2],
+							const b2Vec2& normal, float32 offset, int32 vertexIndexA);
+
+/// Determine if two generic shapes overlap.
+bool b2TestOverlap(	const b2Shape* shapeA, int32 indexA,
+					const b2Shape* shapeB, int32 indexB,
+					const b2Transform& xfA, const b2Transform& xfB);
+
+// ---------------- Inline Functions ------------------------------------------
+
+inline bool b2AABB::IsValid() const
+{
+	b2Vec2 d = upperBound - lowerBound;
+	bool valid = d.x >= 0.0f && d.y >= 0.0f;
+	valid = valid && lowerBound.IsValid() && upperBound.IsValid();
+	return valid;
+}
+
+inline bool b2TestOverlap(const b2AABB& a, const b2AABB& b)
+{
+	b2Vec2 d1, d2;
+	d1 = b.lowerBound - a.upperBound;
+	d2 = a.lowerBound - b.upperBound;
+
+	if (d1.x > 0.0f || d1.y > 0.0f)
+		return false;
+
+	if (d2.x > 0.0f || d2.y > 0.0f)
+		return false;
+
+	return true;
+}
+
+#endif

deps/include/Box2D/Collision/b2Distance.cpp

@@ -0,0 +1,614 @@
+/*
+* Copyright (c) 2007-2009 Erin Catto http://www.box2d.org
+* Copyright (c) 2014 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+#include <string.h>
+#include <memory.h>
+
+#include <Box2D/Collision/b2Distance.h>
+#include <Box2D/Collision/Shapes/b2CircleShape.h>
+#include <Box2D/Collision/Shapes/b2EdgeShape.h>
+#include <Box2D/Collision/Shapes/b2ChainShape.h>
+#include <Box2D/Collision/Shapes/b2PolygonShape.h>
+
+// GJK using Voronoi regions (Christer Ericson) and Barycentric coordinates.
+int32 b2_gjkCalls, b2_gjkIters, b2_gjkMaxIters;
+
+void b2DistanceProxy::Set(const b2Shape* shape, int32 index)
+{
+	switch (shape->GetType())
+	{
+	case b2Shape::e_circle:
+		{
+			const b2CircleShape* circle = static_cast<const b2CircleShape*>(shape);
+			m_vertices = &circle->m_p;
+			m_count = 1;
+			m_radius = circle->m_radius;
+		}
+		break;
+
+	case b2Shape::e_polygon:
+		{
+			const b2PolygonShape* polygon = static_cast<const b2PolygonShape*>(shape);
+			m_vertices = polygon->m_vertices;
+			m_count = polygon->m_count;
+			m_radius = polygon->m_radius;
+		}
+		break;
+
+	case b2Shape::e_chain:
+		{
+			const b2ChainShape* chain = static_cast<const b2ChainShape*>(shape);
+			b2Assert(0 <= index && index < chain->m_count);
+
+			m_buffer[0] = chain->m_vertices[index];
+			if (index + 1 < chain->m_count)
+			{
+				m_buffer[1] = chain->m_vertices[index + 1];
+			}
+			else
+			{
+				m_buffer[1] = chain->m_vertices[0];
+			}
+
+			m_vertices = m_buffer;
+			m_count = 2;
+			m_radius = chain->m_radius;
+		}
+		break;
+
+	case b2Shape::e_edge:
+		{
+			const b2EdgeShape* edge = static_cast<const b2EdgeShape*>(shape);
+			m_vertices = &edge->m_vertex1;
+			m_count = 2;
+			m_radius = edge->m_radius;
+		}
+		break;
+
+	default:
+		b2Assert(false);
+	}
+}
+
+
+struct b2SimplexVertex
+{
+	b2Vec2 wA;		// support point in proxyA
+	b2Vec2 wB;		// support point in proxyB
+	b2Vec2 w;		// wB - wA
+	float32 a;		// barycentric coordinate for closest point
+	int32 indexA;	// wA index
+	int32 indexB;	// wB index
+};
+
+struct b2Simplex
+{
+	void ReadCache(	const b2SimplexCache* cache,
+					const b2DistanceProxy* proxyA, const b2Transform& transformA,
+					const b2DistanceProxy* proxyB, const b2Transform& transformB)
+	{
+		b2Assert(cache->count <= 3);
+		
+		// Copy data from cache.
+		m_count = cache->count;
+		b2SimplexVertex* vertices = &m_v1;
+		for (int32 i = 0; i < m_count; ++i)
+		{
+			b2SimplexVertex* v = vertices + i;
+			v->indexA = cache->indexA[i];
+			v->indexB = cache->indexB[i];
+			b2Vec2 wALocal = proxyA->GetVertex(v->indexA);
+			b2Vec2 wBLocal = proxyB->GetVertex(v->indexB);
+			v->wA = b2Mul(transformA, wALocal);
+			v->wB = b2Mul(transformB, wBLocal);
+			v->w = v->wB - v->wA;
+			v->a = 0.0f;
+		}
+
+		// Compute the new simplex metric, if it is substantially different than
+		// old metric then flush the simplex.
+		if (m_count > 1)
+		{
+			float32 metric1 = cache->metric;
+			float32 metric2 = GetMetric();
+			if (metric2 < 0.5f * metric1 || 2.0f * metric1 < metric2 || metric2 < b2_epsilon)
+			{
+				// Reset the simplex.
+				m_count = 0;
+			}
+		}
+
+		// If the cache is empty or invalid ...
+		if (m_count == 0)
+		{
+			b2SimplexVertex* v = vertices + 0;
+			v->indexA = 0;
+			v->indexB = 0;
+			b2Vec2 wALocal = proxyA->GetVertex(0);
+			b2Vec2 wBLocal = proxyB->GetVertex(0);
+			v->wA = b2Mul(transformA, wALocal);
+			v->wB = b2Mul(transformB, wBLocal);
+			v->w = v->wB - v->wA;
+			v->a = 1.0f;
+			m_count = 1;
+		}
+	}
+
+	void WriteCache(b2SimplexCache* cache) const
+	{
+		cache->metric = GetMetric();
+		cache->count = uint16(m_count);
+		const b2SimplexVertex* vertices = &m_v1;
+		for (int32 i = 0; i < m_count; ++i)
+		{
+			cache->indexA[i] = uint8(vertices[i].indexA);
+			cache->indexB[i] = uint8(vertices[i].indexB);
+		}
+	}
+
+	b2Vec2 GetSearchDirection() const
+	{
+		switch (m_count)
+		{
+		case 1:
+			return -m_v1.w;
+
+		case 2:
+			{
+				b2Vec2 e12 = m_v2.w - m_v1.w;
+				float32 sgn = b2Cross(e12, -m_v1.w);
+				if (sgn > 0.0f)
+				{
+					// Origin is left of e12.
+					return b2Cross(1.0f, e12);
+				}
+				else
+				{
+					// Origin is right of e12.
+					return b2Cross(e12, 1.0f);
+				}
+			}
+
+		default:
+			b2Assert(false);
+			return b2Vec2_zero;
+		}
+	}
+
+	b2Vec2 GetClosestPoint() const
+	{
+		switch (m_count)
+		{
+		case 0:
+			b2Assert(false);
+			return b2Vec2_zero;
+
+		case 1:
+			return m_v1.w;
+
+		case 2:
+			return m_v1.a * m_v1.w + m_v2.a * m_v2.w;
+
+		case 3:
+			return b2Vec2_zero;
+
+		default:
+			b2Assert(false);
+			return b2Vec2_zero;
+		}
+	}
+
+	void GetWitnessPoints(b2Vec2* pA, b2Vec2* pB) const
+	{
+		switch (m_count)
+		{
+		case 0:
+			b2Assert(false);
+			break;
+
+		case 1:
+			*pA = m_v1.wA;
+			*pB = m_v1.wB;
+			break;
+
+		case 2:
+			*pA = m_v1.a * m_v1.wA + m_v2.a * m_v2.wA;
+			*pB = m_v1.a * m_v1.wB + m_v2.a * m_v2.wB;
+			break;
+
+		case 3:
+			*pA = m_v1.a * m_v1.wA + m_v2.a * m_v2.wA + m_v3.a * m_v3.wA;
+			*pB = *pA;
+			break;
+
+		default:
+			b2Assert(false);
+			break;
+		}
+	}
+
+	float32 GetMetric() const
+	{
+		switch (m_count)
+		{
+		case 0:
+			b2Assert(false);
+			return 0.0f;
+
+		case 1:
+			return 0.0f;
+
+		case 2:
+			return b2Distance(m_v1.w, m_v2.w);
+
+		case 3:
+			return b2Cross(m_v2.w - m_v1.w, m_v3.w - m_v1.w);
+
+		default:
+			b2Assert(false);
+			return 0.0f;
+		}
+	}
+
+	void Solve2();
+	void Solve3();
+
+	b2SimplexVertex m_v1, m_v2, m_v3;
+	int32 m_count;
+};
+
+
+// Solve a line segment using barycentric coordinates.
+//
+// p = a1 * w1 + a2 * w2
+// a1 + a2 = 1
+//
+// The vector from the origin to the closest point on the line is
+// perpendicular to the line.
+// e12 = w2 - w1
+// dot(p, e) = 0
+// a1 * dot(w1, e) + a2 * dot(w2, e) = 0
+//
+// 2-by-2 linear system
+// [1      1     ][a1] = [1]
+// [w1.e12 w2.e12][a2] = [0]
+//
+// Define
+// d12_1 =  dot(w2, e12)
+// d12_2 = -dot(w1, e12)
+// d12 = d12_1 + d12_2
+//
+// Solution
+// a1 = d12_1 / d12
+// a2 = d12_2 / d12
+void b2Simplex::Solve2()
+{
+	b2Vec2 w1 = m_v1.w;
+	b2Vec2 w2 = m_v2.w;
+	b2Vec2 e12 = w2 - w1;
+
+	// w1 region
+	float32 d12_2 = -b2Dot(w1, e12);
+	if (d12_2 <= 0.0f)
+	{
+		// a2 <= 0, so we clamp it to 0
+		m_v1.a = 1.0f;
+		m_count = 1;
+		return;
+	}
+
+	// w2 region
+	float32 d12_1 = b2Dot(w2, e12);
+	if (d12_1 <= 0.0f)
+	{
+		// a1 <= 0, so we clamp it to 0
+		m_v2.a = 1.0f;
+		m_count = 1;
+		m_v1 = m_v2;
+		return;
+	}
+
+	// Must be in e12 region.
+	float32 inv_d12 = 1.0f / (d12_1 + d12_2);
+	m_v1.a = d12_1 * inv_d12;
+	m_v2.a = d12_2 * inv_d12;
+	m_count = 2;
+}
+
+// Possible regions:
+// - points[2]
+// - edge points[0]-points[2]
+// - edge points[1]-points[2]
+// - inside the triangle
+void b2Simplex::Solve3()
+{
+	b2Vec2 w1 = m_v1.w;
+	b2Vec2 w2 = m_v2.w;
+	b2Vec2 w3 = m_v3.w;
+
+	// Edge12
+	// [1      1     ][a1] = [1]
+	// [w1.e12 w2.e12][a2] = [0]
+	// a3 = 0
+	b2Vec2 e12 = w2 - w1;
+	float32 w1e12 = b2Dot(w1, e12);
+	float32 w2e12 = b2Dot(w2, e12);
+	float32 d12_1 = w2e12;
+	float32 d12_2 = -w1e12;
+
+	// Edge13
+	// [1      1     ][a1] = [1]
+	// [w1.e13 w3.e13][a3] = [0]
+	// a2 = 0
+	b2Vec2 e13 = w3 - w1;
+	float32 w1e13 = b2Dot(w1, e13);
+	float32 w3e13 = b2Dot(w3, e13);
+	float32 d13_1 = w3e13;
+	float32 d13_2 = -w1e13;
+
+	// Edge23
+	// [1      1     ][a2] = [1]
+	// [w2.e23 w3.e23][a3] = [0]
+	// a1 = 0
+	b2Vec2 e23 = w3 - w2;
+	float32 w2e23 = b2Dot(w2, e23);
+	float32 w3e23 = b2Dot(w3, e23);
+	float32 d23_1 = w3e23;
+	float32 d23_2 = -w2e23;
+	
+	// Triangle123
+	float32 n123 = b2Cross(e12, e13);
+
+	float32 d123_1 = n123 * b2Cross(w2, w3);
+	float32 d123_2 = n123 * b2Cross(w3, w1);
+	float32 d123_3 = n123 * b2Cross(w1, w2);
+
+	// w1 region
+	if (d12_2 <= 0.0f && d13_2 <= 0.0f)
+	{
+		m_v1.a = 1.0f;
+		m_count = 1;
+		return;
+	}
+
+	// e12
+	if (d12_1 > 0.0f && d12_2 > 0.0f && d123_3 <= 0.0f)
+	{
+		float32 inv_d12 = 1.0f / (d12_1 + d12_2);
+		m_v1.a = d12_1 * inv_d12;
+		m_v2.a = d12_2 * inv_d12;
+		m_count = 2;
+		return;
+	}
+
+	// e13
+	if (d13_1 > 0.0f && d13_2 > 0.0f && d123_2 <= 0.0f)
+	{
+		float32 inv_d13 = 1.0f / (d13_1 + d13_2);
+		m_v1.a = d13_1 * inv_d13;
+		m_v3.a = d13_2 * inv_d13;
+		m_count = 2;
+		m_v2 = m_v3;
+		return;
+	}
+
+	// w2 region
+	if (d12_1 <= 0.0f && d23_2 <= 0.0f)
+	{
+		m_v2.a = 1.0f;
+		m_count = 1;
+		m_v1 = m_v2;
+		return;
+	}
+
+	// w3 region
+	if (d13_1 <= 0.0f && d23_1 <= 0.0f)
+	{
+		m_v3.a = 1.0f;
+		m_count = 1;
+		m_v1 = m_v3;
+		return;
+	}
+
+	// e23
+	if (d23_1 > 0.0f && d23_2 > 0.0f && d123_1 <= 0.0f)
+	{
+		float32 inv_d23 = 1.0f / (d23_1 + d23_2);
+		m_v2.a = d23_1 * inv_d23;
+		m_v3.a = d23_2 * inv_d23;
+		m_count = 2;
+		m_v1 = m_v3;
+		return;
+	}
+
+	// Must be in triangle123
+	float32 inv_d123 = 1.0f / (d123_1 + d123_2 + d123_3);
+	m_v1.a = d123_1 * inv_d123;
+	m_v2.a = d123_2 * inv_d123;
+	m_v3.a = d123_3 * inv_d123;
+	m_count = 3;
+}
+
+void b2Distance(b2DistanceOutput* output,
+				b2SimplexCache* cache,
+				const b2DistanceInput* input)
+{
+	++b2_gjkCalls;
+
+	const b2DistanceProxy* proxyA = &input->proxyA;
+	const b2DistanceProxy* proxyB = &input->proxyB;
+
+	b2Transform transformA = input->transformA;
+	b2Transform transformB = input->transformB;
+
+	// Initialize the simplex.
+	b2Simplex simplex;
+	simplex.ReadCache(cache, proxyA, transformA, proxyB, transformB);
+
+	// Get simplex vertices as an array.
+	b2SimplexVertex* vertices = &simplex.m_v1;
+	const int32 k_maxIters = 20;
+
+	// These store the vertices of the last simplex so that we
+	// can check for duplicates and prevent cycling.
+	int32 saveA[3], saveB[3];
+	int32 saveCount = 0;
+
+	// Work around spurious gcc-4.8.2 warnings when -Wmaybe-uninitialized is
+	// enabled by initializing saveA / saveB arrays when they're referenced
+	// at the end of the main iteration loop below even though saveCount
+	// entries of each array are initialized at the start of the main
+	// iteration loop.
+	memset(saveA, 0, sizeof(saveA));
+	memset(saveB, 0, sizeof(saveB));
+
+	float32 distanceSqr1 = b2_maxFloat;
+	float32 distanceSqr2;
+
+	// Main iteration loop.
+	int32 iter = 0;
+	while (iter < k_maxIters)
+	{
+		// Copy simplex so we can identify duplicates.
+		saveCount = simplex.m_count;
+		for (int32 i = 0; i < saveCount; ++i)
+		{
+			saveA[i] = vertices[i].indexA;
+			saveB[i] = vertices[i].indexB;
+		}
+
+		switch (simplex.m_count)
+		{
+		case 1:
+			break;
+
+		case 2:
+			simplex.Solve2();
+			break;
+
+		case 3:
+			simplex.Solve3();
+			break;
+
+		default:
+			b2Assert(false);
+		}
+
+		// If we have 3 points, then the origin is in the corresponding triangle.
+		if (simplex.m_count == 3)
+		{
+			break;
+		}
+
+		// Compute closest point.
+		b2Vec2 p = simplex.GetClosestPoint();
+		distanceSqr2 = p.LengthSquared();
+
+		// Ensure progress
+		if (distanceSqr2 >= distanceSqr1)
+		{
+			//break;
+		}
+		distanceSqr1 = distanceSqr2;
+
+		// Get search direction.
+		b2Vec2 d = simplex.GetSearchDirection();
+
+		// Ensure the search direction is numerically fit.
+		if (d.LengthSquared() < b2_epsilon * b2_epsilon)
+		{
+			// The origin is probably contained by a line segment
+			// or triangle. Thus the shapes are overlapped.
+
+			// We can't return zero here even though there may be overlap.
+			// In case the simplex is a point, segment, or triangle it is difficult
+			// to determine if the origin is contained in the CSO or very close to it.
+			break;
+		}
+
+		// Compute a tentative new simplex vertex using support points.
+		b2SimplexVertex* vertex = vertices + simplex.m_count;
+		vertex->indexA = proxyA->GetSupport(b2MulT(transformA.q, -d));
+		vertex->wA = b2Mul(transformA, proxyA->GetVertex(vertex->indexA));
+		b2Vec2 wBLocal;
+		vertex->indexB = proxyB->GetSupport(b2MulT(transformB.q, d));
+		vertex->wB = b2Mul(transformB, proxyB->GetVertex(vertex->indexB));
+		vertex->w = vertex->wB - vertex->wA;
+
+		// Iteration count is equated to the number of support point calls.
+		++iter;
+		++b2_gjkIters;
+
+		// Check for duplicate support points. This is the main termination criteria.
+		bool duplicate = false;
+		for (int32 i = 0; i < saveCount; ++i)
+		{
+			if (vertex->indexA == saveA[i] && vertex->indexB == saveB[i])
+			{
+				duplicate = true;
+				break;
+			}
+		}
+
+		// If we found a duplicate support point we must exit to avoid cycling.
+		if (duplicate)
+		{
+			break;
+		}
+
+		// New vertex is ok and needed.
+		++simplex.m_count;
+	}
+
+	b2_gjkMaxIters = b2Max(b2_gjkMaxIters, iter);
+
+	// Prepare output.
+	simplex.GetWitnessPoints(&output->pointA, &output->pointB);
+	output->distance = b2Distance(output->pointA, output->pointB);
+	output->iterations = iter;
+
+	// Cache the simplex.
+	simplex.WriteCache(cache);
+
+	// Apply radii if requested.
+	if (input->useRadii)
+	{
+		float32 rA = proxyA->m_radius;
+		float32 rB = proxyB->m_radius;
+
+		if (output->distance > rA + rB && output->distance > b2_epsilon)
+		{
+			// Shapes are still no overlapped.
+			// Move the witness points to the outer surface.
+			output->distance -= rA + rB;
+			b2Vec2 normal = output->pointB - output->pointA;
+			normal.Normalize();
+			output->pointA += rA * normal;
+			output->pointB -= rB * normal;
+		}
+		else
+		{
+			// Shapes are overlapped when radii are considered.
+			// Move the witness points to the middle.
+			b2Vec2 p = 0.5f * (output->pointA + output->pointB);
+			output->pointA = p;
+			output->pointB = p;
+			output->distance = 0.0f;
+		}
+	}
+}

deps/include/Box2D/Collision/b2Distance.h

@@ -0,0 +1,141 @@
+
+/*
+* Copyright (c) 2006-2009 Erin Catto http://www.box2d.org
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#ifndef B2_DISTANCE_H
+#define B2_DISTANCE_H
+
+#include <Box2D/Common/b2Math.h>
+
+class b2Shape;
+
+/// A distance proxy is used by the GJK algorithm.
+/// It encapsulates any shape.
+struct b2DistanceProxy
+{
+	b2DistanceProxy() : m_vertices(NULL), m_count(0), m_radius(0.0f) {}
+
+	/// Initialize the proxy using the given shape. The shape
+	/// must remain in scope while the proxy is in use.
+	void Set(const b2Shape* shape, int32 index);
+
+	/// Get the supporting vertex index in the given direction.
+	int32 GetSupport(const b2Vec2& d) const;
+
+	/// Get the supporting vertex in the given direction.
+	const b2Vec2& GetSupportVertex(const b2Vec2& d) const;
+
+	/// Get the vertex count.
+	int32 GetVertexCount() const;
+
+	/// Get a vertex by index. Used by b2Distance.
+	const b2Vec2& GetVertex(int32 index) const;
+
+	b2Vec2 m_buffer[2];
+	const b2Vec2* m_vertices;
+	int32 m_count;
+	float32 m_radius;
+};
+
+/// Used to warm start b2Distance.
+/// Set count to zero on first call.
+struct b2SimplexCache
+{
+	float32 metric;		///< length or area
+	uint16 count;
+	uint8 indexA[3];	///< vertices on shape A
+	uint8 indexB[3];	///< vertices on shape B
+};
+
+/// Input for b2Distance.
+/// You have to option to use the shape radii
+/// in the computation. Even 
+struct b2DistanceInput
+{
+	b2DistanceProxy proxyA;
+	b2DistanceProxy proxyB;
+	b2Transform transformA;
+	b2Transform transformB;
+	bool useRadii;
+};
+
+/// Output for b2Distance.
+struct b2DistanceOutput
+{
+	b2Vec2 pointA;		///< closest point on shapeA
+	b2Vec2 pointB;		///< closest point on shapeB
+	float32 distance;
+	int32 iterations;	///< number of GJK iterations used
+};
+
+/// Compute the closest points between two shapes. Supports any combination of:
+/// b2CircleShape, b2PolygonShape, b2EdgeShape. The simplex cache is input/output.
+/// On the first call set b2SimplexCache.count to zero.
+void b2Distance(b2DistanceOutput* output,
+				b2SimplexCache* cache, 
+				const b2DistanceInput* input);
+
+
+//////////////////////////////////////////////////////////////////////////
+
+inline int32 b2DistanceProxy::GetVertexCount() const
+{
+	return m_count;
+}
+
+inline const b2Vec2& b2DistanceProxy::GetVertex(int32 index) const
+{
+	b2Assert(0 <= index && index < m_count);
+	return m_vertices[index];
+}
+
+inline int32 b2DistanceProxy::GetSupport(const b2Vec2& d) const
+{
+	int32 bestIndex = 0;
+	float32 bestValue = b2Dot(m_vertices[0], d);
+	for (int32 i = 1; i < m_count; ++i)
+	{
+		float32 value = b2Dot(m_vertices[i], d);
+		if (value > bestValue)
+		{
+			bestIndex = i;
+			bestValue = value;
+		}
+	}
+
+	return bestIndex;
+}
+
+inline const b2Vec2& b2DistanceProxy::GetSupportVertex(const b2Vec2& d) const
+{
+	int32 bestIndex = 0;
+	float32 bestValue = b2Dot(m_vertices[0], d);
+	for (int32 i = 1; i < m_count; ++i)
+	{
+		float32 value = b2Dot(m_vertices[i], d);
+		if (value > bestValue)
+		{
+			bestIndex = i;
+			bestValue = value;
+		}
+	}
+
+	return m_vertices[bestIndex];
+}
+
+#endif

deps/include/Box2D/Collision/b2DynamicTree.cpp

@@ -0,0 +1,784 @@
+/*
+* Copyright (c) 2009 Erin Catto http://www.box2d.org
+* Copyright (c) 2014 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#include <Box2D/Collision/b2DynamicTree.h>
+#include <memory.h>
+#include <string.h>
+
+b2DynamicTree::b2DynamicTree()
+{
+	m_root = b2_nullNode;
+
+	m_nodeCapacity = 16;
+	m_nodeCount = 0;
+	m_nodes = (b2TreeNode*)b2Alloc(m_nodeCapacity * sizeof(b2TreeNode));
+	memset(m_nodes, 0, m_nodeCapacity * sizeof(b2TreeNode));
+
+	// Build a linked list for the free list.
+	for (int32 i = 0; i < m_nodeCapacity - 1; ++i)
+	{
+		m_nodes[i].next = i + 1;
+		m_nodes[i].height = -1;
+	}
+	m_nodes[m_nodeCapacity-1].next = b2_nullNode;
+	m_nodes[m_nodeCapacity-1].height = -1;
+	m_freeList = 0;
+
+	m_path = 0;
+
+	m_insertionCount = 0;
+}
+
+b2DynamicTree::~b2DynamicTree()
+{
+	// This frees the entire tree in one shot.
+	b2Free(m_nodes);
+}
+
+// Allocate a node from the pool. Grow the pool if necessary.
+int32 b2DynamicTree::AllocateNode()
+{
+	// Expand the node pool as needed.
+	if (m_freeList == b2_nullNode)
+	{
+		b2Assert(m_nodeCount == m_nodeCapacity);
+
+		// The free list is empty. Rebuild a bigger pool.
+		b2TreeNode* oldNodes = m_nodes;
+		m_nodeCapacity *= 2;
+		m_nodes = (b2TreeNode*)b2Alloc(m_nodeCapacity * sizeof(b2TreeNode));
+		memcpy(m_nodes, oldNodes, m_nodeCount * sizeof(b2TreeNode));
+		b2Free(oldNodes);
+
+		// Build a linked list for the free list. The parent
+		// pointer becomes the "next" pointer.
+		for (int32 i = m_nodeCount; i < m_nodeCapacity - 1; ++i)
+		{
+			m_nodes[i].next = i + 1;
+			m_nodes[i].height = -1;
+		}
+		m_nodes[m_nodeCapacity-1].next = b2_nullNode;
+		m_nodes[m_nodeCapacity-1].height = -1;
+		m_freeList = m_nodeCount;
+	}
+
+	// Peel a node off the free list.
+	int32 nodeId = m_freeList;
+	m_freeList = m_nodes[nodeId].next;
+	m_nodes[nodeId].parent = b2_nullNode;
+	m_nodes[nodeId].child1 = b2_nullNode;
+	m_nodes[nodeId].child2 = b2_nullNode;
+	m_nodes[nodeId].height = 0;
+	m_nodes[nodeId].userData = NULL;
+	++m_nodeCount;
+	return nodeId;
+}
+
+// Return a node to the pool.
+void b2DynamicTree::FreeNode(int32 nodeId)
+{
+	b2Assert(0 <= nodeId && nodeId < m_nodeCapacity);
+	b2Assert(0 < m_nodeCount);
+	m_nodes[nodeId].next = m_freeList;
+	m_nodes[nodeId].height = -1;
+	m_freeList = nodeId;
+	--m_nodeCount;
+}
+
+// Create a proxy in the tree as a leaf node. We return the index
+// of the node instead of a pointer so that we can grow
+// the node pool.
+int32 b2DynamicTree::CreateProxy(const b2AABB& aabb, void* userData)
+{
+	int32 proxyId = AllocateNode();
+
+	// Fatten the aabb.
+	b2Vec2 r(b2_aabbExtension, b2_aabbExtension);
+	m_nodes[proxyId].aabb.lowerBound = aabb.lowerBound - r;
+	m_nodes[proxyId].aabb.upperBound = aabb.upperBound + r;
+	m_nodes[proxyId].userData = userData;
+	m_nodes[proxyId].height = 0;
+
+	InsertLeaf(proxyId);
+
+	return proxyId;
+}
+
+void b2DynamicTree::DestroyProxy(int32 proxyId)
+{
+	b2Assert(0 <= proxyId && proxyId < m_nodeCapacity);
+	b2Assert(m_nodes[proxyId].IsLeaf());
+
+	RemoveLeaf(proxyId);
+	FreeNode(proxyId);
+}
+
+bool b2DynamicTree::MoveProxy(int32 proxyId, const b2AABB& aabb, const b2Vec2& displacement)
+{
+	b2Assert(0 <= proxyId && proxyId < m_nodeCapacity);
+
+	b2Assert(m_nodes[proxyId].IsLeaf());
+
+	if (m_nodes[proxyId].aabb.Contains(aabb))
+	{
+		return false;
+	}
+
+	RemoveLeaf(proxyId);
+
+	// Extend AABB.
+	b2AABB b = aabb;
+	b2Vec2 r(b2_aabbExtension, b2_aabbExtension);
+	b.lowerBound = b.lowerBound - r;
+	b.upperBound = b.upperBound + r;
+
+	// Predict AABB displacement.
+	b2Vec2 d = b2_aabbMultiplier * displacement;
+
+	if (d.x < 0.0f)
+	{
+		b.lowerBound.x += d.x;
+	}
+	else
+	{
+		b.upperBound.x += d.x;
+	}
+
+	if (d.y < 0.0f)
+	{
+		b.lowerBound.y += d.y;
+	}
+	else
+	{
+		b.upperBound.y += d.y;
+	}
+
+	m_nodes[proxyId].aabb = b;
+
+	InsertLeaf(proxyId);
+	return true;
+}
+
+void b2DynamicTree::InsertLeaf(int32 leaf)
+{
+	++m_insertionCount;
+
+	if (m_root == b2_nullNode)
+	{
+		m_root = leaf;
+		m_nodes[m_root].parent = b2_nullNode;
+		return;
+	}
+
+	// Find the best sibling for this node
+	b2AABB leafAABB = m_nodes[leaf].aabb;
+	int32 index = m_root;
+	while (m_nodes[index].IsLeaf() == false)
+	{
+		int32 child1 = m_nodes[index].child1;
+		int32 child2 = m_nodes[index].child2;
+
+		float32 area = m_nodes[index].aabb.GetPerimeter();
+
+		b2AABB combinedAABB;
+		combinedAABB.Combine(m_nodes[index].aabb, leafAABB);
+		float32 combinedArea = combinedAABB.GetPerimeter();
+
+		// Cost of creating a new parent for this node and the new leaf
+		float32 cost = 2.0f * combinedArea;
+
+		// Minimum cost of pushing the leaf further down the tree
+		float32 inheritanceCost = 2.0f * (combinedArea - area);
+
+		// Cost of descending into child1
+		float32 cost1;
+		if (m_nodes[child1].IsLeaf())
+		{
+			b2AABB aabb;
+			aabb.Combine(leafAABB, m_nodes[child1].aabb);
+			cost1 = aabb.GetPerimeter() + inheritanceCost;
+		}
+		else
+		{
+			b2AABB aabb;
+			aabb.Combine(leafAABB, m_nodes[child1].aabb);
+			float32 oldArea = m_nodes[child1].aabb.GetPerimeter();
+			float32 newArea = aabb.GetPerimeter();
+			cost1 = (newArea - oldArea) + inheritanceCost;
+		}
+
+		// Cost of descending into child2
+		float32 cost2;
+		if (m_nodes[child2].IsLeaf())
+		{
+			b2AABB aabb;
+			aabb.Combine(leafAABB, m_nodes[child2].aabb);
+			cost2 = aabb.GetPerimeter() + inheritanceCost;
+		}
+		else
+		{
+			b2AABB aabb;
+			aabb.Combine(leafAABB, m_nodes[child2].aabb);
+			float32 oldArea = m_nodes[child2].aabb.GetPerimeter();
+			float32 newArea = aabb.GetPerimeter();
+			cost2 = newArea - oldArea + inheritanceCost;
+		}
+
+		// Descend according to the minimum cost.
+		if (cost < cost1 && cost < cost2)
+		{
+			break;
+		}
+
+		// Descend
+		if (cost1 < cost2)
+		{
+			index = child1;
+		}
+		else
+		{
+			index = child2;
+		}
+	}
+
+	int32 sibling = index;
+
+	// Create a new parent.
+	int32 oldParent = m_nodes[sibling].parent;
+	int32 newParent = AllocateNode();
+	m_nodes[newParent].parent = oldParent;
+	m_nodes[newParent].userData = NULL;
+	m_nodes[newParent].aabb.Combine(leafAABB, m_nodes[sibling].aabb);
+	m_nodes[newParent].height = m_nodes[sibling].height + 1;
+
+	if (oldParent != b2_nullNode)
+	{
+		// The sibling was not the root.
+		if (m_nodes[oldParent].child1 == sibling)
+		{
+			m_nodes[oldParent].child1 = newParent;
+		}
+		else
+		{
+			m_nodes[oldParent].child2 = newParent;
+		}
+
+		m_nodes[newParent].child1 = sibling;
+		m_nodes[newParent].child2 = leaf;
+		m_nodes[sibling].parent = newParent;
+		m_nodes[leaf].parent = newParent;
+	}
+	else
+	{
+		// The sibling was the root.
+		m_nodes[newParent].child1 = sibling;
+		m_nodes[newParent].child2 = leaf;
+		m_nodes[sibling].parent = newParent;
+		m_nodes[leaf].parent = newParent;
+		m_root = newParent;
+	}
+
+	// Walk back up the tree fixing heights and AABBs
+	index = m_nodes[leaf].parent;
+	while (index != b2_nullNode)
+	{
+		index = Balance(index);
+
+		int32 child1 = m_nodes[index].child1;
+		int32 child2 = m_nodes[index].child2;
+
+		b2Assert(child1 != b2_nullNode);
+		b2Assert(child2 != b2_nullNode);
+
+		m_nodes[index].height = 1 + b2Max(m_nodes[child1].height, m_nodes[child2].height);
+		m_nodes[index].aabb.Combine(m_nodes[child1].aabb, m_nodes[child2].aabb);
+
+		index = m_nodes[index].parent;
+	}
+
+	//Validate();
+}
+
+void b2DynamicTree::RemoveLeaf(int32 leaf)
+{
+	if (leaf == m_root)
+	{
+		m_root = b2_nullNode;
+		return;
+	}
+
+	int32 parent = m_nodes[leaf].parent;
+	int32 grandParent = m_nodes[parent].parent;
+	int32 sibling;
+	if (m_nodes[parent].child1 == leaf)
+	{
+		sibling = m_nodes[parent].child2;
+	}
+	else
+	{
+		sibling = m_nodes[parent].child1;
+	}
+
+	if (grandParent != b2_nullNode)
+	{
+		// Destroy parent and connect sibling to grandParent.
+		if (m_nodes[grandParent].child1 == parent)
+		{
+			m_nodes[grandParent].child1 = sibling;
+		}
+		else
+		{
+			m_nodes[grandParent].child2 = sibling;
+		}
+		m_nodes[sibling].parent = grandParent;
+		FreeNode(parent);
+
+		// Adjust ancestor bounds.
+		int32 index = grandParent;
+		while (index != b2_nullNode)
+		{
+			index = Balance(index);
+
+			int32 child1 = m_nodes[index].child1;
+			int32 child2 = m_nodes[index].child2;
+
+			m_nodes[index].aabb.Combine(m_nodes[child1].aabb, m_nodes[child2].aabb);
+			m_nodes[index].height = 1 + b2Max(m_nodes[child1].height, m_nodes[child2].height);
+
+			index = m_nodes[index].parent;
+		}
+	}
+	else
+	{
+		m_root = sibling;
+		m_nodes[sibling].parent = b2_nullNode;
+		FreeNode(parent);
+	}
+
+	//Validate();
+}
+
+// Perform a left or right rotation if node A is imbalanced.
+// Returns the new root index.
+int32 b2DynamicTree::Balance(int32 iA)
+{
+	b2Assert(iA != b2_nullNode);
+
+	b2TreeNode* A = m_nodes + iA;
+	if (A->IsLeaf() || A->height < 2)
+	{
+		return iA;
+	}
+
+	int32 iB = A->child1;
+	int32 iC = A->child2;
+	b2Assert(0 <= iB && iB < m_nodeCapacity);
+	b2Assert(0 <= iC && iC < m_nodeCapacity);
+
+	b2TreeNode* B = m_nodes + iB;
+	b2TreeNode* C = m_nodes + iC;
+
+	int32 balance = C->height - B->height;
+
+	// Rotate C up
+	if (balance > 1)
+	{
+		int32 iF = C->child1;
+		int32 iG = C->child2;
+		b2TreeNode* F = m_nodes + iF;
+		b2TreeNode* G = m_nodes + iG;
+		b2Assert(0 <= iF && iF < m_nodeCapacity);
+		b2Assert(0 <= iG && iG < m_nodeCapacity);
+
+		// Swap A and C
+		C->child1 = iA;
+		C->parent = A->parent;
+		A->parent = iC;
+
+		// A's old parent should point to C
+		if (C->parent != b2_nullNode)
+		{
+			if (m_nodes[C->parent].child1 == iA)
+			{
+				m_nodes[C->parent].child1 = iC;
+			}
+			else
+			{
+				b2Assert(m_nodes[C->parent].child2 == iA);
+				m_nodes[C->parent].child2 = iC;
+			}
+		}
+		else
+		{
+			m_root = iC;
+		}
+
+		// Rotate
+		if (F->height > G->height)
+		{
+			C->child2 = iF;
+			A->child2 = iG;
+			G->parent = iA;
+			A->aabb.Combine(B->aabb, G->aabb);
+			C->aabb.Combine(A->aabb, F->aabb);
+
+			A->height = 1 + b2Max(B->height, G->height);
+			C->height = 1 + b2Max(A->height, F->height);
+		}
+		else
+		{
+			C->child2 = iG;
+			A->child2 = iF;
+			F->parent = iA;
+			A->aabb.Combine(B->aabb, F->aabb);
+			C->aabb.Combine(A->aabb, G->aabb);
+
+			A->height = 1 + b2Max(B->height, F->height);
+			C->height = 1 + b2Max(A->height, G->height);
+		}
+
+		return iC;
+	}
+	
+	// Rotate B up
+	if (balance < -1)
+	{
+		int32 iD = B->child1;
+		int32 iE = B->child2;
+		b2TreeNode* D = m_nodes + iD;
+		b2TreeNode* E = m_nodes + iE;
+		b2Assert(0 <= iD && iD < m_nodeCapacity);
+		b2Assert(0 <= iE && iE < m_nodeCapacity);
+
+		// Swap A and B
+		B->child1 = iA;
+		B->parent = A->parent;
+		A->parent = iB;
+
+		// A's old parent should point to B
+		if (B->parent != b2_nullNode)
+		{
+			if (m_nodes[B->parent].child1 == iA)
+			{
+				m_nodes[B->parent].child1 = iB;
+			}
+			else
+			{
+				b2Assert(m_nodes[B->parent].child2 == iA);
+				m_nodes[B->parent].child2 = iB;
+			}
+		}
+		else
+		{
+			m_root = iB;
+		}
+
+		// Rotate
+		if (D->height > E->height)
+		{
+			B->child2 = iD;
+			A->child1 = iE;
+			E->parent = iA;
+			A->aabb.Combine(C->aabb, E->aabb);
+			B->aabb.Combine(A->aabb, D->aabb);
+
+			A->height = 1 + b2Max(C->height, E->height);
+			B->height = 1 + b2Max(A->height, D->height);
+		}
+		else
+		{
+			B->child2 = iE;
+			A->child1 = iD;
+			D->parent = iA;
+			A->aabb.Combine(C->aabb, D->aabb);
+			B->aabb.Combine(A->aabb, E->aabb);
+
+			A->height = 1 + b2Max(C->height, D->height);
+			B->height = 1 + b2Max(A->height, E->height);
+		}
+
+		return iB;
+	}
+
+	return iA;
+}
+
+int32 b2DynamicTree::GetHeight() const
+{
+	if (m_root == b2_nullNode)
+	{
+		return 0;
+	}
+
+	return m_nodes[m_root].height;
+}
+
+//
+float32 b2DynamicTree::GetAreaRatio() const
+{
+	if (m_root == b2_nullNode)
+	{
+		return 0.0f;
+	}
+
+	const b2TreeNode* root = m_nodes + m_root;
+	float32 rootArea = root->aabb.GetPerimeter();
+
+	float32 totalArea = 0.0f;
+	for (int32 i = 0; i < m_nodeCapacity; ++i)
+	{
+		const b2TreeNode* node = m_nodes + i;
+		if (node->height < 0)
+		{
+			// Free node in pool
+			continue;
+		}
+
+		totalArea += node->aabb.GetPerimeter();
+	}
+
+	return totalArea / rootArea;
+}
+
+// Compute the height of a sub-tree.
+int32 b2DynamicTree::ComputeHeight(int32 nodeId) const
+{
+	b2Assert(0 <= nodeId && nodeId < m_nodeCapacity);
+	b2TreeNode* node = m_nodes + nodeId;
+
+	if (node->IsLeaf())
+	{
+		return 0;
+	}
+
+	int32 height1 = ComputeHeight(node->child1);
+	int32 height2 = ComputeHeight(node->child2);
+	return 1 + b2Max(height1, height2);
+}
+
+int32 b2DynamicTree::ComputeHeight() const
+{
+	int32 height = ComputeHeight(m_root);
+	return height;
+}
+
+void b2DynamicTree::ValidateStructure(int32 index) const
+{
+	if (index == b2_nullNode)
+	{
+		return;
+	}
+
+	if (index == m_root)
+	{
+		b2Assert(m_nodes[index].parent == b2_nullNode);
+	}
+
+	const b2TreeNode* node = m_nodes + index;
+
+#if B2_ASSERT_ENABLED || DEBUG
+	int32 child1 = node->child1;
+	int32 child2 = node->child2;
+#endif  // B2_ASSERT_ENABLED || DEBUG
+
+	if (node->IsLeaf())
+	{
+		b2Assert(child1 == b2_nullNode);
+		b2Assert(child2 == b2_nullNode);
+		b2Assert(node->height == 0);
+		return;
+	}
+
+	b2Assert(0 <= child1 && child1 < m_nodeCapacity);
+	b2Assert(0 <= child2 && child2 < m_nodeCapacity);
+
+	b2Assert(m_nodes[child1].parent == index);
+	b2Assert(m_nodes[child2].parent == index);
+
+	B2_DEBUG_STATEMENT(ValidateStructure(child1));
+	B2_DEBUG_STATEMENT(ValidateStructure(child2));
+}
+
+void b2DynamicTree::ValidateMetrics(int32 index) const
+{
+	if (index == b2_nullNode)
+	{
+		return;
+	}
+
+	const b2TreeNode* node = m_nodes + index;
+
+	int32 child1 = node->child1;
+	int32 child2 = node->child2;
+
+	if (node->IsLeaf())
+	{
+		b2Assert(child1 == b2_nullNode);
+		b2Assert(child2 == b2_nullNode);
+		b2Assert(node->height == 0);
+		return;
+	}
+
+	b2Assert(0 <= child1 && child1 < m_nodeCapacity);
+	b2Assert(0 <= child2 && child2 < m_nodeCapacity);
+
+#if B2_ASSERT_ENABLED
+	int32 height1 = m_nodes[child1].height;
+	int32 height2 = m_nodes[child2].height;
+	int32 height;
+	height = 1 + b2Max(height1, height2);
+#endif // B2_ASSERT_ENABLED
+	b2Assert(node->height == height);
+
+	b2AABB aabb;
+	aabb.Combine(m_nodes[child1].aabb, m_nodes[child2].aabb);
+
+	b2Assert(aabb.lowerBound == node->aabb.lowerBound);
+	b2Assert(aabb.upperBound == node->aabb.upperBound);
+
+	ValidateMetrics(child1);
+	ValidateMetrics(child2);
+}
+
+void b2DynamicTree::Validate() const
+{
+	B2_DEBUG_STATEMENT(ValidateStructure(m_root));
+	B2_DEBUG_STATEMENT(ValidateMetrics(m_root));
+
+	int32 freeCount = 0;
+	int32 freeIndex = m_freeList;
+	while (freeIndex != b2_nullNode)
+	{
+		b2Assert(0 <= freeIndex && freeIndex < m_nodeCapacity);
+		freeIndex = m_nodes[freeIndex].next;
+		++freeCount;
+	}
+
+	b2Assert(GetHeight() == ComputeHeight());
+
+	b2Assert(m_nodeCount + freeCount == m_nodeCapacity);
+}
+
+int32 b2DynamicTree::GetMaxBalance() const
+{
+	int32 maxBalance = 0;
+	for (int32 i = 0; i < m_nodeCapacity; ++i)
+	{
+		const b2TreeNode* node = m_nodes + i;
+		if (node->height <= 1)
+		{
+			continue;
+		}
+
+		b2Assert(node->IsLeaf() == false);
+
+		int32 child1 = node->child1;
+		int32 child2 = node->child2;
+		int32 balance = b2Abs(m_nodes[child2].height - m_nodes[child1].height);
+		maxBalance = b2Max(maxBalance, balance);
+	}
+
+	return maxBalance;
+}
+
+void b2DynamicTree::RebuildBottomUp()
+{
+	int32* nodes = (int32*)b2Alloc(m_nodeCount * sizeof(int32));
+	int32 count = 0;
+
+	// Build array of leaves. Free the rest.
+	for (int32 i = 0; i < m_nodeCapacity; ++i)
+	{
+		if (m_nodes[i].height < 0)
+		{
+			// free node in pool
+			continue;
+		}
+
+		if (m_nodes[i].IsLeaf())
+		{
+			m_nodes[i].parent = b2_nullNode;
+			nodes[count] = i;
+			++count;
+		}
+		else
+		{
+			FreeNode(i);
+		}
+	}
+
+	while (count > 1)
+	{
+		float32 minCost = b2_maxFloat;
+		int32 iMin = -1, jMin = -1;
+		for (int32 i = 0; i < count; ++i)
+		{
+			b2AABB aabbi = m_nodes[nodes[i]].aabb;
+
+			for (int32 j = i + 1; j < count; ++j)
+			{
+				b2AABB aabbj = m_nodes[nodes[j]].aabb;
+				b2AABB b;
+				b.Combine(aabbi, aabbj);
+				float32 cost = b.GetPerimeter();
+				if (cost < minCost)
+				{
+					iMin = i;
+					jMin = j;
+					minCost = cost;
+				}
+			}
+		}
+
+		int32 index1 = nodes[iMin];
+		int32 index2 = nodes[jMin];
+		b2TreeNode* child1 = m_nodes + index1;
+		b2TreeNode* child2 = m_nodes + index2;
+
+		int32 parentIndex = AllocateNode();
+		b2TreeNode* parent = m_nodes + parentIndex;
+		parent->child1 = index1;
+		parent->child2 = index2;
+		parent->height = 1 + b2Max(child1->height, child2->height);
+		parent->aabb.Combine(child1->aabb, child2->aabb);
+		parent->parent = b2_nullNode;
+
+		child1->parent = parentIndex;
+		child2->parent = parentIndex;
+
+		nodes[jMin] = nodes[count-1];
+		nodes[iMin] = parentIndex;
+		--count;
+	}
+
+	m_root = nodes[0];
+	b2Free(nodes);
+
+	B2_DEBUG_STATEMENT(Validate());
+}
+
+void b2DynamicTree::ShiftOrigin(const b2Vec2& newOrigin)
+{
+	// Build array of leaves. Free the rest.
+	for (int32 i = 0; i < m_nodeCapacity; ++i)
+	{
+		m_nodes[i].aabb.lowerBound -= newOrigin;
+		m_nodes[i].aabb.upperBound -= newOrigin;
+	}
+}

deps/include/Box2D/Collision/b2DynamicTree.h

@@ -0,0 +1,289 @@
+/*
+* Copyright (c) 2009 Erin Catto http://www.box2d.org
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#ifndef B2_DYNAMIC_TREE_H
+#define B2_DYNAMIC_TREE_H
+
+#include <Box2D/Collision/b2Collision.h>
+#include <Box2D/Common/b2GrowableStack.h>
+
+#define b2_nullNode (-1)
+
+/// A node in the dynamic tree. The client does not interact with this directly.
+struct b2TreeNode
+{
+	bool IsLeaf() const
+	{
+		return child1 == b2_nullNode;
+	}
+
+	/// Enlarged AABB
+	b2AABB aabb;
+
+	void* userData;
+
+	union
+	{
+		int32 parent;
+		int32 next;
+	};
+
+	int32 child1;
+	int32 child2;
+
+	// leaf = 0, free node = -1
+	int32 height;
+};
+
+/// A dynamic AABB tree broad-phase, inspired by Nathanael Presson's btDbvt.
+/// A dynamic tree arranges data in a binary tree to accelerate
+/// queries such as volume queries and ray casts. Leafs are proxies
+/// with an AABB. In the tree we expand the proxy AABB by b2_fatAABBFactor
+/// so that the proxy AABB is bigger than the client object. This allows the client
+/// object to move by small amounts without triggering a tree update.
+///
+/// Nodes are pooled and relocatable, so we use node indices rather than pointers.
+class b2DynamicTree
+{
+public:
+	/// Constructing the tree initializes the node pool.
+	b2DynamicTree();
+
+	/// Destroy the tree, freeing the node pool.
+	~b2DynamicTree();
+
+	/// Create a proxy. Provide a tight fitting AABB and a userData pointer.
+	int32 CreateProxy(const b2AABB& aabb, void* userData);
+
+	/// Destroy a proxy. This asserts if the id is invalid.
+	void DestroyProxy(int32 proxyId);
+
+	/// Move a proxy with a swepted AABB. If the proxy has moved outside of its fattened AABB,
+	/// then the proxy is removed from the tree and re-inserted. Otherwise
+	/// the function returns immediately.
+	/// @return true if the proxy was re-inserted.
+	bool MoveProxy(int32 proxyId, const b2AABB& aabb1, const b2Vec2& displacement);
+
+	/// Get proxy user data.
+	/// @return the proxy user data or 0 if the id is invalid.
+	void* GetUserData(int32 proxyId) const;
+
+	/// Get the fat AABB for a proxy.
+	const b2AABB& GetFatAABB(int32 proxyId) const;
+
+	/// Query an AABB for overlapping proxies. The callback class
+	/// is called for each proxy that overlaps the supplied AABB.
+	template <typename T>
+	void Query(T* callback, const b2AABB& aabb) const;
+
+	/// Ray-cast against the proxies in the tree. This relies on the callback
+	/// to perform a exact ray-cast in the case were the proxy contains a shape.
+	/// The callback also performs the any collision filtering. This has performance
+	/// roughly equal to k * log(n), where k is the number of collisions and n is the
+	/// number of proxies in the tree.
+	/// @param input the ray-cast input data. The ray extends from p1 to p1 + maxFraction * (p2 - p1).
+	/// @param callback a callback class that is called for each proxy that is hit by the ray.
+	template <typename T>
+	void RayCast(T* callback, const b2RayCastInput& input) const;
+
+	/// Validate this tree. For testing.
+	void Validate() const;
+
+	/// Compute the height of the binary tree in O(N) time. Should not be
+	/// called often.
+	int32 GetHeight() const;
+
+	/// Get the maximum balance of an node in the tree. The balance is the difference
+	/// in height of the two children of a node.
+	int32 GetMaxBalance() const;
+
+	/// Get the ratio of the sum of the node areas to the root area.
+	float32 GetAreaRatio() const;
+
+	/// Build an optimal tree. Very expensive. For testing.
+	void RebuildBottomUp();
+
+	/// Shift the world origin. Useful for large worlds.
+	/// The shift formula is: position -= newOrigin
+	/// @param newOrigin the new origin with respect to the old origin
+	void ShiftOrigin(const b2Vec2& newOrigin);
+
+private:
+
+	int32 AllocateNode();
+	void FreeNode(int32 node);
+
+	void InsertLeaf(int32 node);
+	void RemoveLeaf(int32 node);
+
+	int32 Balance(int32 index);
+
+	int32 ComputeHeight() const;
+	int32 ComputeHeight(int32 nodeId) const;
+
+	void ValidateStructure(int32 index) const;
+	void ValidateMetrics(int32 index) const;
+
+	int32 m_root;
+
+	b2TreeNode* m_nodes;
+	int32 m_nodeCount;
+	int32 m_nodeCapacity;
+
+	int32 m_freeList;
+
+	/// This is used to incrementally traverse the tree for re-balancing.
+	uint32 m_path;
+
+	int32 m_insertionCount;
+};
+
+inline void* b2DynamicTree::GetUserData(int32 proxyId) const
+{
+	b2Assert(0 <= proxyId && proxyId < m_nodeCapacity);
+	return m_nodes[proxyId].userData;
+}
+
+inline const b2AABB& b2DynamicTree::GetFatAABB(int32 proxyId) const
+{
+	b2Assert(0 <= proxyId && proxyId < m_nodeCapacity);
+	return m_nodes[proxyId].aabb;
+}
+
+template <typename T>
+inline void b2DynamicTree::Query(T* callback, const b2AABB& aabb) const
+{
+	b2GrowableStack<int32, 256> stack;
+	stack.Push(m_root);
+
+	while (stack.GetCount() > 0)
+	{
+		int32 nodeId = stack.Pop();
+		if (nodeId == b2_nullNode)
+		{
+			continue;
+		}
+
+		const b2TreeNode* node = m_nodes + nodeId;
+
+		if (b2TestOverlap(node->aabb, aabb))
+		{
+			if (node->IsLeaf())
+			{
+				bool proceed = callback->QueryCallback(nodeId);
+				if (proceed == false)
+				{
+					return;
+				}
+			}
+			else
+			{
+				stack.Push(node->child1);
+				stack.Push(node->child2);
+			}
+		}
+	}
+}
+
+template <typename T>
+inline void b2DynamicTree::RayCast(T* callback, const b2RayCastInput& input) const
+{
+	b2Vec2 p1 = input.p1;
+	b2Vec2 p2 = input.p2;
+	b2Vec2 r = p2 - p1;
+	b2Assert(r.LengthSquared() > 0.0f);
+	r.Normalize();
+
+	// v is perpendicular to the segment.
+	b2Vec2 v = b2Cross(1.0f, r);
+	b2Vec2 abs_v = b2Abs(v);
+
+	// Separating axis for segment (Gino, p80).
+	// |dot(v, p1 - c)| > dot(|v|, h)
+
+	float32 maxFraction = input.maxFraction;
+
+	// Build a bounding box for the segment.
+	b2AABB segmentAABB;
+	{
+		b2Vec2 t = p1 + maxFraction * (p2 - p1);
+		segmentAABB.lowerBound = b2Min(p1, t);
+		segmentAABB.upperBound = b2Max(p1, t);
+	}
+
+	b2GrowableStack<int32, 256> stack;
+	stack.Push(m_root);
+
+	while (stack.GetCount() > 0)
+	{
+		int32 nodeId = stack.Pop();
+		if (nodeId == b2_nullNode)
+		{
+			continue;
+		}
+
+		const b2TreeNode* node = m_nodes + nodeId;
+
+		if (b2TestOverlap(node->aabb, segmentAABB) == false)
+		{
+			continue;
+		}
+
+		// Separating axis for segment (Gino, p80).
+		// |dot(v, p1 - c)| > dot(|v|, h)
+		b2Vec2 c = node->aabb.GetCenter();
+		b2Vec2 h = node->aabb.GetExtents();
+		float32 separation = b2Abs(b2Dot(v, p1 - c)) - b2Dot(abs_v, h);
+		if (separation > 0.0f)
+		{
+			continue;
+		}
+
+		if (node->IsLeaf())
+		{
+			b2RayCastInput subInput;
+			subInput.p1 = input.p1;
+			subInput.p2 = input.p2;
+			subInput.maxFraction = maxFraction;
+
+			float32 value = callback->RayCastCallback(subInput, nodeId);
+
+			if (value == 0.0f)
+			{
+				// The client has terminated the ray cast.
+				return;
+			}
+
+			if (value > 0.0f)
+			{
+				// Update segment bounding box.
+				maxFraction = value;
+				b2Vec2 t = p1 + maxFraction * (p2 - p1);
+				segmentAABB.lowerBound = b2Min(p1, t);
+				segmentAABB.upperBound = b2Max(p1, t);
+			}
+		}
+		else
+		{
+			stack.Push(node->child1);
+			stack.Push(node->child2);
+		}
+	}
+}
+
+#endif

deps/include/Box2D/Collision/b2TimeOfImpact.cpp

@@ -0,0 +1,487 @@
+/*
+* Copyright (c) 2007-2009 Erin Catto http://www.box2d.org
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#include <Box2D/Collision/b2Collision.h>
+#include <Box2D/Collision/b2Distance.h>
+#include <Box2D/Collision/b2TimeOfImpact.h>
+#include <Box2D/Collision/Shapes/b2CircleShape.h>
+#include <Box2D/Collision/Shapes/b2PolygonShape.h>
+#include <Box2D/Common/b2Timer.h>
+
+#include <stdio.h>
+
+float32 b2_toiTime, b2_toiMaxTime;
+int32 b2_toiCalls, b2_toiIters, b2_toiMaxIters;
+int32 b2_toiRootIters, b2_toiMaxRootIters;
+
+//
+struct b2SeparationFunction
+{
+	enum Type
+	{
+		e_points,
+		e_faceA,
+		e_faceB
+	};
+
+	// TODO_ERIN might not need to return the separation
+
+	float32 Initialize(const b2SimplexCache* cache,
+		const b2DistanceProxy* proxyA, const b2Sweep& sweepA,
+		const b2DistanceProxy* proxyB, const b2Sweep& sweepB,
+		float32 t1)
+	{
+		m_proxyA = proxyA;
+		m_proxyB = proxyB;
+		int32 count = cache->count;
+		b2Assert(0 < count && count < 3);
+
+		m_sweepA = sweepA;
+		m_sweepB = sweepB;
+
+		b2Transform xfA, xfB;
+		m_sweepA.GetTransform(&xfA, t1);
+		m_sweepB.GetTransform(&xfB, t1);
+
+		if (count == 1)
+		{
+			m_type = e_points;
+			b2Vec2 localPointA = m_proxyA->GetVertex(cache->indexA[0]);
+			b2Vec2 localPointB = m_proxyB->GetVertex(cache->indexB[0]);
+			b2Vec2 pointA = b2Mul(xfA, localPointA);
+			b2Vec2 pointB = b2Mul(xfB, localPointB);
+			m_axis = pointB - pointA;
+			float32 s = m_axis.Normalize();
+			m_localPoint = b2Vec2_zero;
+			return s;
+		}
+		else if (cache->indexA[0] == cache->indexA[1])
+		{
+			// Two points on B and one on A.
+			m_type = e_faceB;
+			b2Vec2 localPointB1 = proxyB->GetVertex(cache->indexB[0]);
+			b2Vec2 localPointB2 = proxyB->GetVertex(cache->indexB[1]);
+
+			m_axis = b2Cross(localPointB2 - localPointB1, 1.0f);
+			m_axis.Normalize();
+			b2Vec2 normal = b2Mul(xfB.q, m_axis);
+
+			m_localPoint = 0.5f * (localPointB1 + localPointB2);
+			b2Vec2 pointB = b2Mul(xfB, m_localPoint);
+
+			b2Vec2 localPointA = proxyA->GetVertex(cache->indexA[0]);
+			b2Vec2 pointA = b2Mul(xfA, localPointA);
+
+			float32 s = b2Dot(pointA - pointB, normal);
+			if (s < 0.0f)
+			{
+				m_axis = -m_axis;
+				s = -s;
+			}
+			return s;
+		}
+		else
+		{
+			// Two points on A and one or two points on B.
+			m_type = e_faceA;
+			b2Vec2 localPointA1 = m_proxyA->GetVertex(cache->indexA[0]);
+			b2Vec2 localPointA2 = m_proxyA->GetVertex(cache->indexA[1]);
+			
+			m_axis = b2Cross(localPointA2 - localPointA1, 1.0f);
+			m_axis.Normalize();
+			b2Vec2 normal = b2Mul(xfA.q, m_axis);
+
+			m_localPoint = 0.5f * (localPointA1 + localPointA2);
+			b2Vec2 pointA = b2Mul(xfA, m_localPoint);
+
+			b2Vec2 localPointB = m_proxyB->GetVertex(cache->indexB[0]);
+			b2Vec2 pointB = b2Mul(xfB, localPointB);
+
+			float32 s = b2Dot(pointB - pointA, normal);
+			if (s < 0.0f)
+			{
+				m_axis = -m_axis;
+				s = -s;
+			}
+			return s;
+		}
+	}
+
+	//
+	float32 FindMinSeparation(int32* indexA, int32* indexB, float32 t) const
+	{
+		b2Transform xfA, xfB;
+		m_sweepA.GetTransform(&xfA, t);
+		m_sweepB.GetTransform(&xfB, t);
+
+		switch (m_type)
+		{
+		case e_points:
+			{
+				b2Vec2 axisA = b2MulT(xfA.q,  m_axis);
+				b2Vec2 axisB = b2MulT(xfB.q, -m_axis);
+
+				*indexA = m_proxyA->GetSupport(axisA);
+				*indexB = m_proxyB->GetSupport(axisB);
+
+				b2Vec2 localPointA = m_proxyA->GetVertex(*indexA);
+				b2Vec2 localPointB = m_proxyB->GetVertex(*indexB);
+				
+				b2Vec2 pointA = b2Mul(xfA, localPointA);
+				b2Vec2 pointB = b2Mul(xfB, localPointB);
+
+				float32 separation = b2Dot(pointB - pointA, m_axis);
+				return separation;
+			}
+
+		case e_faceA:
+			{
+				b2Vec2 normal = b2Mul(xfA.q, m_axis);
+				b2Vec2 pointA = b2Mul(xfA, m_localPoint);
+
+				b2Vec2 axisB = b2MulT(xfB.q, -normal);
+				
+				*indexA = -1;
+				*indexB = m_proxyB->GetSupport(axisB);
+
+				b2Vec2 localPointB = m_proxyB->GetVertex(*indexB);
+				b2Vec2 pointB = b2Mul(xfB, localPointB);
+
+				float32 separation = b2Dot(pointB - pointA, normal);
+				return separation;
+			}
+
+		case e_faceB:
+			{
+				b2Vec2 normal = b2Mul(xfB.q, m_axis);
+				b2Vec2 pointB = b2Mul(xfB, m_localPoint);
+
+				b2Vec2 axisA = b2MulT(xfA.q, -normal);
+
+				*indexB = -1;
+				*indexA = m_proxyA->GetSupport(axisA);
+
+				b2Vec2 localPointA = m_proxyA->GetVertex(*indexA);
+				b2Vec2 pointA = b2Mul(xfA, localPointA);
+
+				float32 separation = b2Dot(pointA - pointB, normal);
+				return separation;
+			}
+
+		default:
+			b2Assert(false);
+			*indexA = -1;
+			*indexB = -1;
+			return 0.0f;
+		}
+	}
+
+	//
+	float32 Evaluate(int32 indexA, int32 indexB, float32 t) const
+	{
+		b2Transform xfA, xfB;
+		m_sweepA.GetTransform(&xfA, t);
+		m_sweepB.GetTransform(&xfB, t);
+
+		switch (m_type)
+		{
+		case e_points:
+			{
+				b2Vec2 localPointA = m_proxyA->GetVertex(indexA);
+				b2Vec2 localPointB = m_proxyB->GetVertex(indexB);
+
+				b2Vec2 pointA = b2Mul(xfA, localPointA);
+				b2Vec2 pointB = b2Mul(xfB, localPointB);
+				float32 separation = b2Dot(pointB - pointA, m_axis);
+
+				return separation;
+			}
+
+		case e_faceA:
+			{
+				b2Vec2 normal = b2Mul(xfA.q, m_axis);
+				b2Vec2 pointA = b2Mul(xfA, m_localPoint);
+
+				b2Vec2 localPointB = m_proxyB->GetVertex(indexB);
+				b2Vec2 pointB = b2Mul(xfB, localPointB);
+
+				float32 separation = b2Dot(pointB - pointA, normal);
+				return separation;
+			}
+
+		case e_faceB:
+			{
+				b2Vec2 normal = b2Mul(xfB.q, m_axis);
+				b2Vec2 pointB = b2Mul(xfB, m_localPoint);
+
+				b2Vec2 localPointA = m_proxyA->GetVertex(indexA);
+				b2Vec2 pointA = b2Mul(xfA, localPointA);
+
+				float32 separation = b2Dot(pointA - pointB, normal);
+				return separation;
+			}
+
+		default:
+			b2Assert(false);
+			return 0.0f;
+		}
+	}
+
+	const b2DistanceProxy* m_proxyA;
+	const b2DistanceProxy* m_proxyB;
+	b2Sweep m_sweepA, m_sweepB;
+	Type m_type;
+	b2Vec2 m_localPoint;
+	b2Vec2 m_axis;
+};
+
+// CCD via the local separating axis method. This seeks progression
+// by computing the largest time at which separation is maintained.
+void b2TimeOfImpact(b2TOIOutput* output, const b2TOIInput* input)
+{
+	b2Timer timer;
+
+	++b2_toiCalls;
+
+	output->state = b2TOIOutput::e_unknown;
+	output->t = input->tMax;
+
+	const b2DistanceProxy* proxyA = &input->proxyA;
+	const b2DistanceProxy* proxyB = &input->proxyB;
+
+	b2Sweep sweepA = input->sweepA;
+	b2Sweep sweepB = input->sweepB;
+
+	// Large rotations can make the root finder fail, so we normalize the
+	// sweep angles.
+	sweepA.Normalize();
+	sweepB.Normalize();
+
+	float32 tMax = input->tMax;
+
+	float32 totalRadius = proxyA->m_radius + proxyB->m_radius;
+	float32 target = b2Max(b2_linearSlop, totalRadius - 3.0f * b2_linearSlop);
+	float32 tolerance = 0.25f * b2_linearSlop;
+	b2Assert(target > tolerance);
+
+	float32 t1 = 0.0f;
+	const int32 k_maxIterations = 20;	// TODO_ERIN b2Settings
+	int32 iter = 0;
+
+	// Prepare input for distance query.
+	b2SimplexCache cache;
+	cache.count = 0;
+	b2DistanceInput distanceInput;
+	distanceInput.proxyA = input->proxyA;
+	distanceInput.proxyB = input->proxyB;
+	distanceInput.useRadii = false;
+
+	// The outer loop progressively attempts to compute new separating axes.
+	// This loop terminates when an axis is repeated (no progress is made).
+	for(;;)
+	{
+		b2Transform xfA, xfB;
+		sweepA.GetTransform(&xfA, t1);
+		sweepB.GetTransform(&xfB, t1);
+
+		// Get the distance between shapes. We can also use the results
+		// to get a separating axis.
+		distanceInput.transformA = xfA;
+		distanceInput.transformB = xfB;
+		b2DistanceOutput distanceOutput;
+		b2Distance(&distanceOutput, &cache, &distanceInput);
+
+		// If the shapes are overlapped, we give up on continuous collision.
+		if (distanceOutput.distance <= 0.0f)
+		{
+			// Failure!
+			output->state = b2TOIOutput::e_overlapped;
+			output->t = 0.0f;
+			break;
+		}
+
+		if (distanceOutput.distance < target + tolerance)
+		{
+			// Victory!
+			output->state = b2TOIOutput::e_touching;
+			output->t = t1;
+			break;
+		}
+
+		// Initialize the separating axis.
+		b2SeparationFunction fcn;
+		fcn.Initialize(&cache, proxyA, sweepA, proxyB, sweepB, t1);
+#if 0
+		// Dump the curve seen by the root finder
+		{
+			const int32 N = 100;
+			float32 dx = 1.0f / N;
+			float32 xs[N+1];
+			float32 fs[N+1];
+
+			float32 x = 0.0f;
+
+			for (int32 i = 0; i <= N; ++i)
+			{
+				sweepA.GetTransform(&xfA, x);
+				sweepB.GetTransform(&xfB, x);
+				float32 f = fcn.Evaluate(xfA, xfB) - target;
+
+				printf("%g %g\n", x, f);
+
+				xs[i] = x;
+				fs[i] = f;
+
+				x += dx;
+			}
+		}
+#endif
+
+		// Compute the TOI on the separating axis. We do this by successively
+		// resolving the deepest point. This loop is bounded by the number of vertices.
+		bool done = false;
+		float32 t2 = tMax;
+		int32 pushBackIter = 0;
+		for (;;)
+		{
+			// Find the deepest point at t2. Store the witness point indices.
+			int32 indexA, indexB;
+			float32 s2 = fcn.FindMinSeparation(&indexA, &indexB, t2);
+
+			// Is the final configuration separated?
+			if (s2 > target + tolerance)
+			{
+				// Victory!
+				output->state = b2TOIOutput::e_separated;
+				output->t = tMax;
+				done = true;
+				break;
+			}
+
+			// Has the separation reached tolerance?
+			if (s2 > target - tolerance)
+			{
+				// Advance the sweeps
+				t1 = t2;
+				break;
+			}
+
+			// Compute the initial separation of the witness points.
+			float32 s1 = fcn.Evaluate(indexA, indexB, t1);
+
+			// Check for initial overlap. This might happen if the root finder
+			// runs out of iterations.
+			if (s1 < target - tolerance)
+			{
+				output->state = b2TOIOutput::e_failed;
+				output->t = t1;
+				done = true;
+				break;
+			}
+
+			// Check for touching
+			if (s1 <= target + tolerance)
+			{
+				// Victory! t1 should hold the TOI (could be 0.0).
+				output->state = b2TOIOutput::e_touching;
+				output->t = t1;
+				done = true;
+				break;
+			}
+
+			// Compute 1D root of: f(x) - target = 0
+			int32 rootIterCount = 0;
+			float32 a1 = t1, a2 = t2;
+			for (;;)
+			{
+				// Use a mix of the secant rule and bisection.
+				float32 t;
+				if (rootIterCount & 1)
+				{
+					// Secant rule to improve convergence.
+					t = a1 + (target - s1) * (a2 - a1) / (s2 - s1);
+				}
+				else
+				{
+					// Bisection to guarantee progress.
+					t = 0.5f * (a1 + a2);
+				}
+
+				++rootIterCount;
+				++b2_toiRootIters;
+
+				float32 s = fcn.Evaluate(indexA, indexB, t);
+
+				if (b2Abs(s - target) < tolerance)
+				{
+					// t2 holds a tentative value for t1
+					t2 = t;
+					break;
+				}
+
+				// Ensure we continue to bracket the root.
+				if (s > target)
+				{
+					a1 = t;
+					s1 = s;
+				}
+				else
+				{
+					a2 = t;
+					s2 = s;
+				}
+				
+				if (rootIterCount == 50)
+				{
+					break;
+				}
+			}
+
+			b2_toiMaxRootIters = b2Max(b2_toiMaxRootIters, rootIterCount);
+
+			++pushBackIter;
+
+			if (pushBackIter == b2_maxPolygonVertices)
+			{
+				break;
+			}
+		}
+
+		++iter;
+		++b2_toiIters;
+
+		if (done)
+		{
+			break;
+		}
+
+		if (iter == k_maxIterations)
+		{
+			// Root finder got stuck. Semi-victory.
+			output->state = b2TOIOutput::e_failed;
+			output->t = t1;
+			break;
+		}
+	}
+
+	b2_toiMaxIters = b2Max(b2_toiMaxIters, iter);
+
+	float32 time = timer.GetMilliseconds();
+	b2_toiMaxTime = b2Max(b2_toiMaxTime, time);
+	b2_toiTime += time;
+}

deps/include/Box2D/Collision/b2TimeOfImpact.h

@@ -0,0 +1,58 @@
+/*
+* Copyright (c) 2006-2009 Erin Catto http://www.box2d.org
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#ifndef B2_TIME_OF_IMPACT_H
+#define B2_TIME_OF_IMPACT_H
+
+#include <Box2D/Common/b2Math.h>
+#include <Box2D/Collision/b2Distance.h>
+
+/// Input parameters for b2TimeOfImpact
+struct b2TOIInput
+{
+	b2DistanceProxy proxyA;
+	b2DistanceProxy proxyB;
+	b2Sweep sweepA;
+	b2Sweep sweepB;
+	float32 tMax;		// defines sweep interval [0, tMax]
+};
+
+// Output parameters for b2TimeOfImpact.
+struct b2TOIOutput
+{
+	enum State
+	{
+		e_unknown,
+		e_failed,
+		e_overlapped,
+		e_touching,
+		e_separated
+	};
+
+	State state;
+	float32 t;
+};
+
+/// Compute the upper bound on time before two shapes penetrate. Time is represented as
+/// a fraction between [0,tMax]. This uses a swept separating axis and may miss some intermediate,
+/// non-tunneling collision. If you change the time interval, you should call this function
+/// again.
+/// Note: use b2Distance to compute the contact point and normal at the time of impact.
+void b2TimeOfImpact(b2TOIOutput* output, const b2TOIInput* input);
+
+#endif

deps/include/Box2D/Common/b2BlockAllocator.cpp

@@ -0,0 +1,225 @@
+/*
+* Copyright (c) 2006-2009 Erin Catto http://www.box2d.org
+* Copyright (c) 2013 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#include <Box2D/Common/b2BlockAllocator.h>
+#include <limits.h>
+#include <memory.h>
+#include <stddef.h>
+#include <string.h>
+#include <new> // For placement new
+
+int32 b2BlockAllocator::s_blockSizes[b2_blockSizes] =
+{
+	16,		// 0
+	32,		// 1
+	64,		// 2
+	96,		// 3
+	128,	// 4
+	160,	// 5
+	192,	// 6
+	224,	// 7
+	256,	// 8
+	320,	// 9
+	384,	// 10
+	448,	// 11
+	512,	// 12
+	640,	// 13
+};
+uint8 b2BlockAllocator::s_blockSizeLookup[b2_maxBlockSize + 1];
+bool b2BlockAllocator::s_blockSizeLookupInitialized;
+
+struct b2Chunk
+{
+	int32 blockSize;
+	b2Block* blocks;
+};
+
+struct b2Block
+{
+	b2Block* next;
+};
+
+b2BlockAllocator::b2BlockAllocator()
+{
+	b2Assert((uint32)b2_blockSizes < UCHAR_MAX);
+
+	m_chunkSpace = b2_chunkArrayIncrement;
+	m_chunkCount = 0;
+	m_chunks = (b2Chunk*)b2Alloc(m_chunkSpace * sizeof(b2Chunk));
+
+	memset(m_chunks, 0, m_chunkSpace * sizeof(b2Chunk));
+	memset(m_freeLists, 0, sizeof(m_freeLists));
+
+	if (s_blockSizeLookupInitialized == false)
+	{
+		int32 j = 0;
+		for (int32 i = 1; i <= b2_maxBlockSize; ++i)
+		{
+			b2Assert(j < b2_blockSizes);
+			if (i <= s_blockSizes[j])
+			{
+				s_blockSizeLookup[i] = (uint8)j;
+			}
+			else
+			{
+				++j;
+				s_blockSizeLookup[i] = (uint8)j;
+			}
+		}
+
+		s_blockSizeLookupInitialized = true;
+	}
+}
+
+b2BlockAllocator::~b2BlockAllocator()
+{
+	for (int32 i = 0; i < m_chunkCount; ++i)
+	{
+		b2Free(m_chunks[i].blocks);
+	}
+
+	b2Free(m_chunks);
+}
+
+uint32 b2BlockAllocator::GetNumGiantAllocations() const
+{
+	return m_giants.GetList().GetLength();
+}
+
+void* b2BlockAllocator::Allocate(int32 size)
+{
+	if (size == 0)
+		return NULL;
+
+	b2Assert(0 < size);
+
+	if (size > b2_maxBlockSize)
+	{
+		return m_giants.Allocate(size);
+	}
+
+	int32 index = s_blockSizeLookup[size];
+	b2Assert(0 <= index && index < b2_blockSizes);
+
+	if (m_freeLists[index])
+	{
+		b2Block* block = m_freeLists[index];
+		m_freeLists[index] = block->next;
+		return block;
+	}
+	else
+	{
+		if (m_chunkCount == m_chunkSpace)
+		{
+			b2Chunk* oldChunks = m_chunks;
+			m_chunkSpace += b2_chunkArrayIncrement;
+			m_chunks = (b2Chunk*)b2Alloc(m_chunkSpace * sizeof(b2Chunk));
+			memcpy(m_chunks, oldChunks, m_chunkCount * sizeof(b2Chunk));
+			memset(m_chunks + m_chunkCount, 0, b2_chunkArrayIncrement * sizeof(b2Chunk));
+			b2Free(oldChunks);
+		}
+
+		b2Chunk* chunk = m_chunks + m_chunkCount;
+		chunk->blocks = (b2Block*)b2Alloc(b2_chunkSize);
+#if DEBUG
+		memset(chunk->blocks, 0xcd, b2_chunkSize);
+#endif
+		int32 blockSize = s_blockSizes[index];
+		chunk->blockSize = blockSize;
+		int32 blockCount = b2_chunkSize / blockSize;
+		b2Assert(blockCount * blockSize <= b2_chunkSize);
+		for (int32 i = 0; i < blockCount - 1; ++i)
+		{
+			b2Block* block = (b2Block*)((int8*)chunk->blocks + blockSize * i);
+			b2Block* next = (b2Block*)((int8*)chunk->blocks + blockSize * (i + 1));
+			block->next = next;
+		}
+		b2Block* last = (b2Block*)((int8*)chunk->blocks + blockSize * (blockCount - 1));
+		last->next = NULL;
+
+		m_freeLists[index] = chunk->blocks->next;
+		++m_chunkCount;
+
+		return chunk->blocks;
+	}
+}
+
+void b2BlockAllocator::Free(void* p, int32 size)
+{
+	if (size == 0)
+	{
+		return;
+	}
+
+	b2Assert(0 < size);
+
+	if (size > b2_maxBlockSize)
+	{
+		m_giants.Free(p);
+		return;
+	}
+
+	int32 index = s_blockSizeLookup[size];
+	b2Assert(0 <= index && index < b2_blockSizes);
+
+#if B2_ASSERT_ENABLED
+	// Verify the memory address and size is valid.
+	int32 blockSize = s_blockSizes[index];
+	bool found = false;
+	for (int32 i = 0; i < m_chunkCount; ++i)
+	{
+		b2Chunk* chunk = m_chunks + i;
+		if (chunk->blockSize != blockSize)
+		{
+			b2Assert(	(int8*)p + blockSize <= (int8*)chunk->blocks ||
+						(int8*)chunk->blocks + b2_chunkSize <= (int8*)p);
+		}
+		else
+		{
+			if ((int8*)chunk->blocks <= (int8*)p && (int8*)p + blockSize <= (int8*)chunk->blocks + b2_chunkSize)
+			{
+				found = true;
+			}
+		}
+	}
+
+	b2Assert(found);
+#endif // B2_ASSERT_ENABLED
+
+#if DEBUG
+	memset(p, 0xfd, s_blockSizes[index]);
+#endif
+
+	b2Block* block = (b2Block*)p;
+	block->next = m_freeLists[index];
+	m_freeLists[index] = block;
+}
+
+void b2BlockAllocator::Clear()
+{
+	for (int32 i = 0; i < m_chunkCount; ++i)
+	{
+		b2Free(m_chunks[i].blocks);
+	}
+
+	m_chunkCount = 0;
+	memset(m_chunks, 0, m_chunkSpace * sizeof(b2Chunk));
+
+	memset(m_freeLists, 0, sizeof(m_freeLists));
+}

deps/include/Box2D/Common/b2BlockAllocator.h

@@ -0,0 +1,68 @@
+/*
+* Copyright (c) 2006-2009 Erin Catto http://www.box2d.org
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#ifndef B2_BLOCK_ALLOCATOR_H
+#define B2_BLOCK_ALLOCATOR_H
+
+#include <Box2D/Common/b2Settings.h>
+#include <Box2D/Common/b2TrackedBlock.h>
+
+const int32 b2_chunkSize = 16 * 1024;
+const int32 b2_maxBlockSize = 640;
+const int32 b2_blockSizes = 14;
+const int32 b2_chunkArrayIncrement = 128;
+
+struct b2Block;
+struct b2Chunk;
+
+/// This is a small object allocator used for allocating small
+/// objects that persist for more than one time step.
+/// See: http://www.codeproject.com/useritems/Small_Block_Allocator.asp
+class b2BlockAllocator
+{
+public:
+	b2BlockAllocator();
+	~b2BlockAllocator();
+
+	/// Allocate memory. This uses b2Alloc if the size is larger than b2_maxBlockSize.
+	void* Allocate(int32 size);
+
+	/// Free memory. This uses b2Free if the size is larger than b2_maxBlockSize.
+	void Free(void* p, int32 size);
+
+	void Clear();
+
+	/// Returns the number of allocations larger than the max block size.
+	uint32 GetNumGiantAllocations() const;
+
+private:
+	b2Chunk* m_chunks;
+	int32 m_chunkCount;
+	int32 m_chunkSpace;
+
+	b2Block* m_freeLists[b2_blockSizes];
+
+	// Record giant allocations--ones bigger than the max block size
+	b2TrackedBlockAllocator m_giants;
+
+	static int32 s_blockSizes[b2_blockSizes];
+	static uint8 s_blockSizeLookup[b2_maxBlockSize + 1];
+	static bool s_blockSizeLookupInitialized;
+};
+
+#endif

deps/include/Box2D/Common/b2Draw.cpp

@@ -0,0 +1,44 @@
+/*
+* Copyright (c) 2011 Erin Catto http://box2d.org
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#include <Box2D/Common/b2Draw.h>
+
+b2Draw::b2Draw()
+{
+	m_drawFlags = 0;
+}
+
+void b2Draw::SetFlags(uint32 flags)
+{
+	m_drawFlags = flags;
+}
+
+uint32 b2Draw::GetFlags() const
+{
+	return m_drawFlags;
+}
+
+void b2Draw::AppendFlags(uint32 flags)
+{
+	m_drawFlags |= flags;
+}
+
+void b2Draw::ClearFlags(uint32 flags)
+{
+	m_drawFlags &= ~flags;
+}

deps/include/Box2D/Common/b2Draw.h

@@ -0,0 +1,92 @@
+/*
+* Copyright (c) 2011 Erin Catto http://box2d.org
+* Copyright (c) 2014 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#ifndef B2_DRAW_H
+#define B2_DRAW_H
+
+#include <Box2D/Common/b2Math.h>
+#include <Box2D/Particle/b2Particle.h>
+
+/// Color for debug drawing. Each value has the range [0,1].
+struct b2Color
+{
+	b2Color() {}
+	b2Color(float32 r, float32 g, float32 b) : r(r), g(g), b(b) {}
+	void Set(float32 ri, float32 gi, float32 bi) { r = ri; g = gi; b = bi; }
+	float32 r, g, b;
+};
+
+/// Implement and register this class with a b2World to provide debug drawing of physics
+/// entities in your game.
+class b2Draw
+{
+public:
+	b2Draw();
+
+	virtual ~b2Draw() {}
+
+	enum
+	{
+		e_shapeBit				= 0x0001,	///< draw shapes
+		e_jointBit				= 0x0002,	///< draw joint connections
+		e_aabbBit				= 0x0004,	///< draw axis aligned bounding boxes
+		e_pairBit				= 0x0008,	///< draw broad-phase pairs
+		e_centerOfMassBit			= 0x0010,	///< draw center of mass frame
+		e_particleBit				= 0x0020  ///< draw particles
+	};
+
+	/// Set the drawing flags.
+	void SetFlags(uint32 flags);
+
+	/// Get the drawing flags.
+	uint32 GetFlags() const;
+
+	/// Append flags to the current flags.
+	void AppendFlags(uint32 flags);
+
+	/// Clear flags from the current flags.
+	void ClearFlags(uint32 flags);
+
+	/// Draw a closed polygon provided in CCW order.
+	virtual void DrawPolygon(const b2Vec2* vertices, int32 vertexCount, const b2Color& color) = 0;
+
+	/// Draw a solid closed polygon provided in CCW order.
+	virtual void DrawSolidPolygon(const b2Vec2* vertices, int32 vertexCount, const b2Color& color) = 0;
+
+	/// Draw a circle.
+	virtual void DrawCircle(const b2Vec2& center, float32 radius, const b2Color& color) = 0;
+
+	/// Draw a solid circle.
+	virtual void DrawSolidCircle(const b2Vec2& center, float32 radius, const b2Vec2& axis, const b2Color& color) = 0;
+
+	/// Draw a particle array
+	virtual void DrawParticles(const b2Vec2 *centers, float32 radius, const b2ParticleColor *colors, int32 count) = 0;
+
+	/// Draw a line segment.
+	virtual void DrawSegment(const b2Vec2& p1, const b2Vec2& p2, const b2Color& color) = 0;
+
+	/// Draw a transform. Choose your own length scale.
+	/// @param xf a transform.
+	virtual void DrawTransform(const b2Transform& xf) = 0;
+
+protected:
+	uint32 m_drawFlags;
+};
+
+#endif

deps/include/Box2D/Common/b2FreeList.cpp

@@ -0,0 +1,57 @@
+/*
+* Copyright (c) 2014 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+#include <Box2D/Common/b2FreeList.h>
+#include <Box2D/Common/b2IntrusiveList.h>
+#include <Box2D/Common/b2Settings.h>
+
+/// Allocate an item from the freelist.
+b2IntrusiveListNode* b2FreeList::Allocate()
+{
+	if (m_free.IsEmpty()) return NULL;
+	b2IntrusiveListNode * const node = m_free.GetNext();
+	node->Remove();
+	m_allocated.InsertBefore(node);
+	return node;
+}
+
+void b2FreeList::Free(b2IntrusiveListNode* node)
+{
+	b2Assert(node);
+#if B2_FREE_LIST_CHECK_ALLOCATED_ON_FREE
+	b2Assert(m_allocated.FindNodeInList(node));
+#endif // B2_FREE_LIST_CHECK_ALLOCATED_ON_FREE
+	node->Remove();
+	m_free.InsertAfter(node);
+}
+
+void b2FreeList::AddToFreeList(b2IntrusiveListNode* node)
+{
+	b2Assert(node);
+	b2Assert(!node->InList());
+	m_free.InsertBefore(node);
+}
+
+void b2FreeList::RemoveAll()
+{
+	while (!m_allocated.IsEmpty()) {
+		m_allocated.GetNext()->Remove();
+	}
+	while (!m_free.IsEmpty()) {
+		m_free.GetNext()->Remove();
+	}
+}

deps/include/Box2D/Common/b2FreeList.h

@@ -0,0 +1,113 @@
+/*
+* Copyright (c) 2014 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+#ifndef B2_FREE_LIST_H
+#define B2_FREE_LIST_H
+
+#include <Box2D/Common/b2IntrusiveList.h>
+#include <Box2D/Common/b2Settings.h>
+
+/// When B2_FREE_LIST_CHECK_ALLOCATED_ON_FREE is 1, b2FreeList::Free() will
+/// check that the deallocated node was allocated from the freelist.
+#ifndef B2_FREE_LIST_CHECK_ALLOCATED_ON_FREE
+#define B2_FREE_LIST_CHECK_ALLOCATED_ON_FREE 0
+#endif // B2_FREE_LIST_CHECK_ALLOCATED_ON_FREE
+
+
+/// Fast - O(1) - list based allocator for items that can be inserted into
+/// b2IntrusiveListNode lists.
+class b2FreeList
+{
+public:
+	/// Construct the free list.
+	b2FreeList() { }
+
+	/// Destroy the free list.
+	~b2FreeList() { }
+
+	/// Allocate an item from the freelist.
+	b2IntrusiveListNode* Allocate();
+
+	/// Free an item from the freelist.
+	void Free(b2IntrusiveListNode* node);
+
+	/// Add an item to the freelist so that it can be allocated using
+	/// b2FreeList::Allocate().
+	void AddToFreeList(b2IntrusiveListNode* node);
+
+	/// Remove all items (allocated and free) from the freelist.
+	void RemoveAll();
+
+	/// Get the list which tracks allocated items.
+	const b2IntrusiveListNode& GetAllocatedList() const {
+		return m_allocated;
+	}
+
+	/// Get the list which tracks free items.
+	const b2IntrusiveListNode& GetFreeList() const {
+		return m_free;
+	}
+
+protected:
+	/// List of allocated items.
+	b2IntrusiveListNode m_allocated;
+	/// List of free items.
+	b2IntrusiveListNode m_free;
+};
+
+
+/// Typed b2FreeList which manages items of type T assuming T implements
+/// the GetInstanceFromListNode() and GetListNode() methods.
+template<typename T>
+class b2TypedFreeList {
+public:
+	/// Construct the free list.
+	b2TypedFreeList() { }
+
+	/// Destroy the free list.
+	~b2TypedFreeList() { }
+
+	/// Allocate an item from the free list.
+	T* Allocate() {
+		b2IntrusiveListNode* const node = m_freeList.Allocate();
+		if (!node) return NULL;
+		return T::GetInstanceFromListNode(node);
+	}
+
+	/// Free an item.
+	void Free(T* instance) {
+		b2Assert(instance);
+		m_freeList.Free(instance->GetListNode());
+	}
+
+	/// Add an item to the freelist so that it can be allocated with
+	/// b2TypedFreeList::Allocate().
+	void AddToFreeList(T* instance)
+	{
+		b2Assert(instance);
+		m_freeList.AddToFreeList(instance->GetListNode());
+	}
+
+	// Get the underlying b2FreeList.
+	b2FreeList* GetFreeList() { return &m_freeList; }
+	const b2FreeList* GetFreeList() const { return &m_freeList; }
+
+protected:
+	b2FreeList m_freeList;
+};
+
+#endif  // B2_FREE_LIST_H

deps/include/Box2D/Common/b2GrowableBuffer.h

@@ -0,0 +1,200 @@
+/*
+* Copyright (c) 2014 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+#ifndef B2_GROWABLE_BUFFER_H
+#define B2_GROWABLE_BUFFER_H
+
+#include <Box2D/Common/b2BlockAllocator.h>
+#include <string.h>
+#include <memory.h>
+#include <algorithm>
+
+
+/// A simple array-like container, similar to std::vector.
+/// If we ever start using stl, we should replace this with std::vector.
+template <typename T>
+class b2GrowableBuffer
+{
+public:
+	b2GrowableBuffer(b2BlockAllocator& allocator) :
+		data(NULL),
+		count(0),
+		capacity(0),
+		allocator(&allocator)
+	{
+	#if defined(LIQUIDFUN_SIMD_NEON)
+		// b2ParticleAssembly.neon.s assumes these values are at fixed offsets.
+        // If this assert fails, be sure to update the assembly offsets!
+		// ldr r3, [r9, #0] @ r3 = out = contacts.data
+        // ldr r6, [r9, #8] @ r6 = contacts.capacity
+		b2Assert((intptr_t)&data - (intptr_t)(this) == 0
+			  && (intptr_t)&capacity - (intptr_t)(this) == 8);
+	#endif // defined(LIQUIDFUN_SIMD_NEON)
+	}
+
+	b2GrowableBuffer(const b2GrowableBuffer<T>& rhs) :
+		data(NULL),
+		count(rhs.count),
+		capacity(rhs.capacity),
+		allocator(rhs.allocator)
+	{
+		if (rhs.data != NULL)
+		{
+			data = (T*) allocator->Allocate(sizeof(T) * capacity);
+			memcpy(data, rhs.data, sizeof(T) * count);
+		}
+	}
+
+	~b2GrowableBuffer()
+	{
+		Free();
+	}
+
+	T& Append()
+	{
+		if (count >= capacity)
+		{
+			Grow();
+		}
+		return data[count++];
+	}
+
+	void Reserve(int32 newCapacity)
+	{
+		if (capacity >= newCapacity)
+			return;
+
+		// Reallocate and copy.
+		T* newData = (T*) allocator->Allocate(sizeof(T) * newCapacity);
+		if (data)
+		{
+			memcpy(newData, data, sizeof(T) * count);
+			allocator->Free(data, sizeof(T) * capacity);
+		}
+
+		// Update pointer and capacity.
+		capacity = newCapacity;
+		data = newData;
+	}
+
+	void Grow()
+	{
+		// Double the capacity.
+		int32 newCapacity = capacity ? 2 * capacity
+						  : b2_minParticleSystemBufferCapacity;
+		b2Assert(newCapacity > capacity);
+		Reserve(newCapacity);
+	}
+
+	void Free()
+	{
+		if (data == NULL)
+			return;
+
+		allocator->Free(data, sizeof(data[0]) * capacity);
+		data = NULL;
+		capacity = 0;
+		count = 0;
+	}
+
+	void Shorten(const T* newEnd)
+	{
+		b2Assert(newEnd >= data);
+		count = (int32) (newEnd - data);
+	}
+
+	T& operator[](int i)
+	{
+		return data[i];
+	}
+
+	const T& operator[](int i) const
+	{
+		return data[i];
+	}
+
+	T* Data()
+	{
+		return data;
+	}
+
+	const T* Data() const
+	{
+		return data;
+	}
+
+	T* Begin()
+	{
+		return data;
+	}
+
+	const T* Begin() const
+	{
+		return data;
+	}
+
+	T* End()
+	{
+		return &data[count];
+	}
+
+	const T* End() const
+	{
+		return &data[count];
+	}
+
+	int32 GetCount() const
+	{
+		return count;
+	}
+
+	void SetCount(int32 newCount)
+	{
+		b2Assert(0 <= newCount && newCount <= capacity);
+		count = newCount;
+	}
+
+	int32 GetCapacity() const
+	{
+		return capacity;
+	}
+
+	template<class UnaryPredicate>
+	T* RemoveIf(UnaryPredicate pred)
+	{
+		T* newEnd = std::remove_if(data, data + count, pred);
+		Shorten(newEnd);
+		return newEnd;
+	}
+
+	template<class BinaryPredicate>
+	T* Unique(BinaryPredicate pred)
+	{
+		T* newEnd = std::unique(data, data + count, pred);
+		Shorten(newEnd);
+		return newEnd;
+	}
+
+private:
+	T* data;
+	int32 count;
+	int32 capacity;
+	b2BlockAllocator* allocator;
+};
+
+#endif // B2_GROWABLE_BUFFER_H
+

deps/include/Box2D/Common/b2GrowableStack.h

@@ -0,0 +1,87 @@
+/*
+* Copyright (c) 2010 Erin Catto http://www.box2d.org
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#ifndef B2_GROWABLE_STACK_H
+#define B2_GROWABLE_STACK_H
+
+#include <Box2D/Common/b2Settings.h>
+#include <string.h>
+#include <memory.h>
+
+/// This is a growable LIFO stack with an initial capacity of N.
+/// If the stack size exceeds the initial capacity, the heap is used
+/// to increase the size of the stack.
+template <typename T, int32 N>
+class b2GrowableStack
+{
+public:
+	b2GrowableStack()
+	{
+		m_stack = m_array;
+		m_count = 0;
+		m_capacity = N;
+	}
+
+	~b2GrowableStack()
+	{
+		if (m_stack != m_array)
+		{
+			b2Free(m_stack);
+			m_stack = NULL;
+		}
+	}
+
+	void Push(const T& element)
+	{
+		if (m_count == m_capacity)
+		{
+			T* old = m_stack;
+			m_capacity *= 2;
+			m_stack = (T*)b2Alloc(m_capacity * sizeof(T));
+			memcpy(m_stack, old, m_count * sizeof(T));
+			if (old != m_array)
+			{
+				b2Free(old);
+			}
+		}
+
+		m_stack[m_count] = element;
+		++m_count;
+	}
+
+	T Pop()
+	{
+		b2Assert(m_count > 0);
+		--m_count;
+		return m_stack[m_count];
+	}
+
+	int32 GetCount()
+	{
+		return m_count;
+	}
+
+private:
+	T* m_stack;
+	T m_array[N];
+	int32 m_count;
+	int32 m_capacity;
+};
+
+
+#endif

deps/include/Box2D/Common/b2IntrusiveList.h

@@ -0,0 +1,369 @@
+/*
+* Copyright (c) 2014 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+#ifndef B2_INTRUSIVE_LIST
+#define B2_INTRUSIVE_LIST
+
+#include <Box2D/Common/b2Settings.h>
+
+// Whether to enable b2IntrusiveList::ValidateList().
+// Be careful when enabling this since this changes the size of
+// b2IntrusiveListNode so make sure *all* projects that include Box2D.h
+// also define this value in the same way to avoid data corruption.
+#ifndef B2_INTRUSIVE_LIST_VALIDATE
+#define B2_INTRUSIVE_LIST_VALIDATE 0
+#endif  // B2_INTRUSIVE_LIST_VALIDATE
+
+/// b2IntrusiveListNode is used to implement an intrusive doubly-linked
+/// list.
+///
+/// For example:
+///
+/// class MyClass {
+/// public:
+/// 	MyClass(const char *msg) : m_msg(msg) {}
+/// 	const char* GetMessage() const { return m_msg; }
+/// 	B2_INTRUSIVE_LIST_GET_NODE(m_node);
+/// 	B2_INTRUSIVE_LIST_NODE_GET_CLASS(MyClass, m_node);
+/// private:
+/// 	b2IntrusiveListNode m_node;
+/// 	const char *m_msg;
+/// };
+///
+/// int main(int argc, char *argv[]) {
+/// 	b2IntrusiveListNode list; // NOTE: type is NOT MyClass
+/// 	MyClass a("this");
+/// 	MyClass b("is");
+/// 	MyClass c("a");
+/// 	MyClass d("test");
+/// 	list.InsertBefore(a.GetListNode());
+/// 	list.InsertBefore(b.GetListNode());
+/// 	list.InsertBefore(c.GetListNode());
+/// 	list.InsertBefore(d.GetListNode());
+/// 	for (b2IntrusiveListNode* node = list.GetNext();
+/// 		 node != list.GetTerminator(); node = node->GetNext()) {
+/// 		MyClass *cls = MyClass::GetInstanceFromListNode(node);
+/// 		printf("%s\n", cls->GetMessage());
+/// 	}
+/// 	return 0;
+/// }
+class b2IntrusiveListNode
+{
+public:
+	/// Initialize the node.
+	b2IntrusiveListNode()
+	{
+		Initialize();
+#if B2_INTRUSIVE_LIST_VALIDATE
+		m_magic = k_magic;
+#endif // B2_INTRUSIVE_LIST_VALIDATE
+	}
+
+	/// If the node is in a list, remove it from the list.
+	~b2IntrusiveListNode()
+	{
+		Remove();
+#if B2_INTRUSIVE_LIST_VALIDATE
+		m_magic = 0;
+#endif // B2_INTRUSIVE_LIST_VALIDATE
+	}
+
+	/// Insert this node after the specified node.
+	void InsertAfter(b2IntrusiveListNode* const node)
+	{
+		b2Assert(!node->InList());
+		node->m_next = m_next;
+		node->m_prev = this;
+		m_next->m_prev = node;
+		m_next = node;
+	}
+
+	/// Insert this node before the specified node.
+	void InsertBefore(b2IntrusiveListNode* const node)
+	{
+		b2Assert(!node->InList());
+		node->m_next = this;
+		node->m_prev = m_prev;
+		m_prev->m_next = node;
+		m_prev = node;
+	}
+
+	/// Get the terminator of the list.
+	const b2IntrusiveListNode* GetTerminator() const
+	{
+		return this;
+	}
+
+	/// Remove this node from the list it's currently in.
+	b2IntrusiveListNode* Remove()
+	{
+		m_prev->m_next = m_next;
+		m_next->m_prev = m_prev;
+		Initialize();
+		return this;
+	}
+
+	/// Determine whether this list is empty or the node isn't in a list.
+	bool IsEmpty() const
+	{
+	  return GetNext() == this;
+	}
+
+	/// Determine whether this node is in a list or the list contains nodes.
+	bool InList() const
+	{
+	  return !IsEmpty();
+	}
+
+	/// Calculate the length of the list.
+	uint32 GetLength() const
+	{
+		uint32 length = 0;
+		const b2IntrusiveListNode * const terminator = GetTerminator();
+		for (const b2IntrusiveListNode* node = GetNext();
+			 node != terminator; node = node->GetNext())
+		{
+			length++;
+		}
+		return length;
+	}
+
+	/// Get the next node in the list.
+	b2IntrusiveListNode* GetNext() const
+	{
+		return m_next;
+	}
+
+	/// Get the previous node in the list.
+	b2IntrusiveListNode* GetPrevious() const
+	{
+		return m_prev;
+	}
+
+	/// If B2_INTRUSIVE_LIST_VALIDATE is 1 perform a very rough validation
+	/// of all nodes in the list.
+	bool ValidateList() const
+	{
+#if B2_INTRUSIVE_LIST_VALIDATE
+	  if (m_magic != k_magic) return false;
+	  const b2IntrusiveListNode * const terminator = GetTerminator();
+	  for (b2IntrusiveListNode *node = GetNext(); node != terminator;
+		   node = node->GetNext()) {
+		if (node->m_magic != k_magic) return false;
+	  }
+#endif  // B2_INTRUSIVE_LIST_VALIDATE
+	  return true;
+	}
+
+	/// Determine whether the specified node is present in this list.
+	bool FindNodeInList(b2IntrusiveListNode* const nodeToFind) const
+	{
+		const b2IntrusiveListNode * const terminator = GetTerminator();
+		for (b2IntrusiveListNode *node = GetNext(); node != terminator;
+			 node = node->GetNext())
+		{
+			if (nodeToFind == node) return true;
+		}
+		return false;
+	}
+
+private:
+	/// Initialize the list node.
+	void Initialize()
+	{
+		m_next = this;
+		m_prev = this;
+	}
+
+private:
+#if B2_INTRUSIVE_LIST_VALIDATE
+	uint32 m_magic;
+#endif  // B2_INTRUSIVE_LIST_VALIDATE
+	/// The next node in the list.
+	b2IntrusiveListNode *m_prev;
+	/// The previous node in the list.
+	b2IntrusiveListNode *m_next;
+
+private:
+#if B2_INTRUSIVE_LIST_VALIDATE
+	static const uint32 k_magic = 0x7157ac01;
+#endif  // B2_INTRUSIVE_LIST_VALIDATE
+};
+
+/// Declares the member function GetListNode() of Class to retrieve a pointer
+/// to NodeMemberName.
+/// See #B2_INTRUSIVE_LIST_NODE_GET_CLASS_ACCESSOR()
+#define B2_INTRUSIVE_LIST_GET_NODE(NodeMemberName) \
+	b2IntrusiveListNode* GetListNode() { return &NodeMemberName; } \
+	const b2IntrusiveListNode* GetListNode() const { return &NodeMemberName; }
+
+/// Declares the member function FunctionName of Class to retrieve a pointer
+/// to a Class instance from a list node pointer.   NodeMemberName references
+/// the name of the b2IntrusiveListNode member of Class.
+#define B2_INTRUSIVE_LIST_NODE_GET_CLASS_ACCESSOR( \
+	Class, NodeMemberName, FunctionName) \
+	static Class* FunctionName(b2IntrusiveListNode *node) \
+	{ \
+		Class *cls = NULL; \
+		/* This effectively performs offsetof(Class, NodeMemberName) */ \
+		/* which ends up in the undefined behavior realm of C++ but in */ \
+		/* practice this works with most compilers. */ \
+		return reinterpret_cast<Class*>((uint8*)(node) - \
+										(uint8*)(&cls->NodeMemberName)); \
+	} \
+	\
+	static const Class* FunctionName(const b2IntrusiveListNode *node) \
+	{ \
+		return FunctionName(const_cast<b2IntrusiveListNode*>(node)); \
+	}
+
+/// Declares the member function GetInstanceFromListNode() of Class to retrieve
+/// a pointer to a Class instance from a list node pointer.  NodeMemberName
+/// reference the name of the b2IntrusiveListNode member of Class.
+#define B2_INTRUSIVE_LIST_NODE_GET_CLASS(Class, NodeMemberName) \
+	B2_INTRUSIVE_LIST_NODE_GET_CLASS_ACCESSOR(Class, NodeMemberName, \
+											  GetInstanceFromListNode)
+
+/// b2TypedIntrusiveListNode which supports inserting an object into a single
+/// doubly linked list.  For objects that need to be inserted in multiple
+/// doubly linked lists, use b2IntrusiveListNode.
+///
+/// For example:
+///
+/// class IntegerItem : public b2TypedIntrusiveListNode<IntegerItem>
+/// {
+/// public:
+/// 	IntegerItem(int32 value) : m_value(value) { }
+/// 	~IntegerItem() { }
+/// 	int32 GetValue() const { return m_value; }
+/// private:
+/// 	int32 m_value;
+/// };
+///
+/// int main(int argc, const char *arvg[]) {
+/// 	b2TypedIntrusiveListNode<IntegerItem> list;
+/// 	IntegerItem a(1);
+/// 	IntegerItem b(2);
+/// 	IntegerItem c(3);
+/// 	list.InsertBefore(&a);
+/// 	list.InsertBefore(&b);
+/// 	list.InsertBefore(&c);
+/// 	for (IntegerItem* item = list.GetNext();
+/// 		 item != list.GetTerminator(); item = item->GetNext())
+/// 	{
+/// 		printf("%d\n", item->GetValue());
+/// 	}
+/// }
+template<typename T>
+class b2TypedIntrusiveListNode
+{
+public:
+	b2TypedIntrusiveListNode() { }
+	~b2TypedIntrusiveListNode() { }
+
+	/// Insert this object after the specified object.
+	void InsertAfter(T* const obj)
+	{
+		b2Assert(obj);
+		GetListNode()->InsertAfter(obj->GetListNode());
+	}
+
+	/// Insert this object before the specified object.
+	void InsertBefore(T* const obj)
+	{
+		b2Assert(obj);
+		GetListNode()->InsertBefore(obj->GetListNode());
+	}
+
+	/// Get the next object in the list.
+	/// Check against GetTerminator() before deferencing the object.
+	T* GetNext() const
+	{
+		return GetInstanceFromListNode(GetListNode()->GetNext());
+	}
+
+	/// Get the previous object in the list.
+	/// Check against GetTerminator() before deferencing the object.
+	T* GetPrevious() const
+	{
+		return GetInstanceFromListNode(GetListNode()->GetPrevious());
+	}
+
+	/// Get the terminator of the list.
+	/// This should not be dereferenced as it is a pointer to
+	/// b2TypedIntrusiveListNode<T> *not* T.
+	T* GetTerminator() const
+	{
+		return (T*)GetListNode();
+	}
+
+	/// Remove this object from the list it's currently in.
+	T* Remove()
+	{
+		GetListNode()->Remove();
+		return GetInstanceFromListNode(GetListNode());
+	}
+
+	/// Determine whether this object is in a list.
+	bool InList() const
+	{
+		return GetListNode()->InList();
+	}
+
+	// Determine whether this list is empty.
+	bool IsEmpty() const
+	{
+		return GetListNode()->IsEmpty();
+	}
+
+	/// Calculate the length of the list.
+	uint32 GetLength() const
+	{
+		return GetListNode()->GetLength();
+	}
+
+	B2_INTRUSIVE_LIST_GET_NODE(m_node);
+
+private:
+	// Node within an intrusive list.
+	b2IntrusiveListNode m_node;
+
+public:
+	/// Get a pointer to the instance of T that contains "node".
+	static T* GetInstanceFromListNode(b2IntrusiveListNode* const node)
+	{
+		b2Assert(node);
+		// Calculate the pointer to T from the offset.
+		return (T*)((uint8*)node - GetNodeOffset(node));
+	}
+
+private:
+	// Get the offset of m_node within this class.
+	static int32 GetNodeOffset(b2IntrusiveListNode* const node)
+	{
+		b2Assert(node);
+		// Perform some type punning to calculate the offset of m_node in T.
+		// WARNING: This could result in undefined behavior with some C++
+		// compilers.
+		T* obj = (T*)node;
+		int32 nodeOffset = (int32)((uint8*)&obj->m_node - (uint8*)obj);
+		return nodeOffset;
+	}
+};
+
+#endif // B2_INTRUSIVE_LIST
+

deps/include/Box2D/Common/b2Math.cpp

@@ -0,0 +1,94 @@
+/*
+* Copyright (c) 2007-2009 Erin Catto http://www.box2d.org
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#include <Box2D/Common/b2Math.h>
+
+const b2Vec2 b2Vec2_zero(0.0f, 0.0f);
+
+/// Solve A * x = b, where b is a column vector. This is more efficient
+/// than computing the inverse in one-shot cases.
+b2Vec3 b2Mat33::Solve33(const b2Vec3& b) const
+{
+	float32 det = b2Dot(ex, b2Cross(ey, ez));
+	if (det != 0.0f)
+	{
+		det = 1.0f / det;
+	}
+	b2Vec3 x;
+	x.x = det * b2Dot(b, b2Cross(ey, ez));
+	x.y = det * b2Dot(ex, b2Cross(b, ez));
+	x.z = det * b2Dot(ex, b2Cross(ey, b));
+	return x;
+}
+
+/// Solve A * x = b, where b is a column vector. This is more efficient
+/// than computing the inverse in one-shot cases.
+b2Vec2 b2Mat33::Solve22(const b2Vec2& b) const
+{
+	float32 a11 = ex.x, a12 = ey.x, a21 = ex.y, a22 = ey.y;
+	float32 det = a11 * a22 - a12 * a21;
+	if (det != 0.0f)
+	{
+		det = 1.0f / det;
+	}
+	b2Vec2 x;
+	x.x = det * (a22 * b.x - a12 * b.y);
+	x.y = det * (a11 * b.y - a21 * b.x);
+	return x;
+}
+
+///
+void b2Mat33::GetInverse22(b2Mat33* M) const
+{
+	float32 a = ex.x, b = ey.x, c = ex.y, d = ey.y;
+	float32 det = a * d - b * c;
+	if (det != 0.0f)
+	{
+		det = 1.0f / det;
+	}
+
+	M->ex.x =  det * d;	M->ey.x = -det * b; M->ex.z = 0.0f;
+	M->ex.y = -det * c;	M->ey.y =  det * a; M->ey.z = 0.0f;
+	M->ez.x = 0.0f; M->ez.y = 0.0f; M->ez.z = 0.0f;
+}
+
+/// Returns the zero matrix if singular.
+void b2Mat33::GetSymInverse33(b2Mat33* M) const
+{
+	float32 det = b2Dot(ex, b2Cross(ey, ez));
+	if (det != 0.0f)
+	{
+		det = 1.0f / det;
+	}
+
+	float32 a11 = ex.x, a12 = ey.x, a13 = ez.x;
+	float32 a22 = ey.y, a23 = ez.y;
+	float32 a33 = ez.z;
+
+	M->ex.x = det * (a22 * a33 - a23 * a23);
+	M->ex.y = det * (a13 * a23 - a12 * a33);
+	M->ex.z = det * (a12 * a23 - a13 * a22);
+
+	M->ey.x = M->ex.y;
+	M->ey.y = det * (a11 * a33 - a13 * a13);
+	M->ey.z = det * (a13 * a12 - a11 * a23);
+
+	M->ez.x = M->ex.z;
+	M->ez.y = M->ey.z;
+	M->ez.z = det * (a11 * a22 - a12 * a12);
+}

deps/include/Box2D/Common/b2Math.h

@@ -0,0 +1,800 @@
+/*
+* Copyright (c) 2006-2009 Erin Catto http://www.box2d.org
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#ifndef B2_MATH_H
+#define B2_MATH_H
+
+#include <Box2D/Common/b2Settings.h>
+#include <math.h>
+
+/// This function is used to ensure that a floating point number is not a NaN or infinity.
+inline bool b2IsValid(float32 x)
+{
+	union {
+		float32 f;
+		int32 i;
+	} v = { x };
+	return (v.i & 0x7f800000) != 0x7f800000;
+}
+
+/// This is a approximate yet fast inverse square-root.
+inline float32 b2InvSqrt(float32 x)
+{
+	union
+	{
+		float32 x;
+		int32 i;
+	} convert;
+
+	convert.x = x;
+	float32 xhalf = 0.5f * x;
+	convert.i = 0x5f3759df - (convert.i >> 1);
+	x = convert.x;
+	x = x * (1.5f - xhalf * x * x);
+	return x;
+}
+
+#define	b2Sqrt(x)	sqrtf(x)
+#define	b2Atan2(y, x)	atan2f(y, x)
+
+/// A 2D column vector.
+struct b2Vec2
+{
+	/// Default constructor does nothing (for performance).
+	b2Vec2() {}
+
+	/// Construct using coordinates.
+	b2Vec2(float32 x, float32 y) : x(x), y(y) {}
+
+	/// Set this vector to all zeros.
+	void SetZero() { x = 0.0f; y = 0.0f; }
+
+	/// Set this vector to some specified coordinates.
+	void Set(float32 x_, float32 y_) { x = x_; y = y_; }
+
+	/// Negate this vector.
+	b2Vec2 operator -() const { b2Vec2 v; v.Set(-x, -y); return v; }
+
+	/// Read from and indexed element.
+	float32 operator () (int32 i) const
+	{
+		return (&x)[i];
+	}
+
+	/// Write to an indexed element.
+	float32& operator () (int32 i)
+	{
+		return (&x)[i];
+	}
+
+	/// Add a vector to this vector.
+	void operator += (const b2Vec2& v)
+	{
+		x += v.x; y += v.y;
+	}
+
+	/// Subtract a vector from this vector.
+	void operator -= (const b2Vec2& v)
+	{
+		x -= v.x; y -= v.y;
+	}
+
+	/// Multiply this vector by a scalar.
+	void operator *= (float32 a)
+	{
+		x *= a; y *= a;
+	}
+
+	/// Get the length of this vector (the norm).
+	float32 Length() const
+	{
+		return b2Sqrt(x * x + y * y);
+	}
+
+	/// Get the length squared. For performance, use this instead of
+	/// b2Vec2::Length (if possible).
+	float32 LengthSquared() const
+	{
+		return x * x + y * y;
+	}
+
+	/// Convert this vector into a unit vector. Returns the length.
+	float32 Normalize()
+	{
+		float32 length = Length();
+		if (length < b2_epsilon)
+		{
+			return 0.0f;
+		}
+		float32 invLength = 1.0f / length;
+		x *= invLength;
+		y *= invLength;
+
+		return length;
+	}
+
+	/// Does this vector contain finite coordinates?
+	bool IsValid() const
+	{
+		return b2IsValid(x) && b2IsValid(y);
+	}
+
+	/// Get the skew vector such that dot(skew_vec, other) == cross(vec, other)
+	b2Vec2 Skew() const
+	{
+		return b2Vec2(-y, x);
+	}
+
+	float32 x, y;
+};
+
+/// Add a float to a vector.
+inline b2Vec2 operator + (const b2Vec2& v, float f)
+{
+	return b2Vec2(v.x + f, v.y + f);
+}
+
+/// Substract a float from a vector.
+inline b2Vec2 operator - (const b2Vec2& v, float f)
+{
+	return b2Vec2(v.x - f, v.y - f);
+}
+
+/// Multiply a float with a vector.
+inline b2Vec2 operator * (const b2Vec2& v, float f)
+{
+	return b2Vec2(v.x * f, v.y * f);
+}
+
+/// Divide a vector by a float.
+inline b2Vec2 operator / (const b2Vec2& v, float f)
+{
+	return b2Vec2(v.x / f, v.y / f);
+}
+
+/// A 3D column vector with 3 elements.
+struct b2Vec3
+{
+	/// Default constructor does nothing (for performance).
+	b2Vec3() {}
+
+	/// Construct using coordinates.
+	b2Vec3(float32 x, float32 y, float32 z) : x(x), y(y), z(z) {}
+
+	/// Set this vector to all zeros.
+	void SetZero() { x = 0.0f; y = 0.0f; z = 0.0f; }
+
+	/// Set this vector to some specified coordinates.
+	void Set(float32 x_, float32 y_, float32 z_) { x = x_; y = y_; z = z_; }
+
+	/// Negate this vector.
+	b2Vec3 operator -() const { b2Vec3 v; v.Set(-x, -y, -z); return v; }
+
+	/// Add a vector to this vector.
+	void operator += (const b2Vec3& v)
+	{
+		x += v.x; y += v.y; z += v.z;
+	}
+
+	/// Subtract a vector from this vector.
+	void operator -= (const b2Vec3& v)
+	{
+		x -= v.x; y -= v.y; z -= v.z;
+	}
+
+	/// Multiply this vector by a scalar.
+	void operator *= (float32 s)
+	{
+		x *= s; y *= s; z *= s;
+	}
+
+		/// Get the length of this vector (the norm).
+	float32 Length() const
+	{
+		return b2Sqrt(x * x + y * y + z * z);
+	}
+
+	/// Convert this vector into a unit vector. Returns the length.
+	float32 Normalize()
+	{
+		float32 length = Length();
+		if (length < b2_epsilon)
+		{
+			return 0.0f;
+		}
+		float32 invLength = 1.0f / length;
+		x *= invLength;
+		y *= invLength;
+		z *= invLength;
+
+		return length;
+	}
+
+	float32 x, y, z;
+};
+
+/// A 4D column vector with 4 elements.
+struct b2Vec4
+{
+	/// Default constructor does nothing (for performance).
+	b2Vec4() {}
+
+	/// Construct using coordinates.
+	b2Vec4(float32 x, float32 y, float32 z, float32 w) : x(x), y(y), z(z), w(w) {}
+
+	float32 x, y, z, w;
+};
+
+/// A 2-by-2 matrix. Stored in column-major order.
+struct b2Mat22
+{
+	/// The default constructor does nothing (for performance).
+	b2Mat22() {}
+
+	/// Construct this matrix using columns.
+	b2Mat22(const b2Vec2& c1, const b2Vec2& c2)
+	{
+		ex = c1;
+		ey = c2;
+	}
+
+	/// Construct this matrix using scalars.
+	b2Mat22(float32 a11, float32 a12, float32 a21, float32 a22)
+	{
+		ex.x = a11; ex.y = a21;
+		ey.x = a12; ey.y = a22;
+	}
+
+	/// Initialize this matrix using columns.
+	void Set(const b2Vec2& c1, const b2Vec2& c2)
+	{
+		ex = c1;
+		ey = c2;
+	}
+
+	/// Set this to the identity matrix.
+	void SetIdentity()
+	{
+		ex.x = 1.0f; ey.x = 0.0f;
+		ex.y = 0.0f; ey.y = 1.0f;
+	}
+
+	/// Set this matrix to all zeros.
+	void SetZero()
+	{
+		ex.x = 0.0f; ey.x = 0.0f;
+		ex.y = 0.0f; ey.y = 0.0f;
+	}
+
+	b2Mat22 GetInverse() const
+	{
+		float32 a = ex.x, b = ey.x, c = ex.y, d = ey.y;
+		b2Mat22 B;
+		float32 det = a * d - b * c;
+		if (det != 0.0f)
+		{
+			det = 1.0f / det;
+		}
+		B.ex.x =  det * d;	B.ey.x = -det * b;
+		B.ex.y = -det * c;	B.ey.y =  det * a;
+		return B;
+	}
+
+	/// Solve A * x = b, where b is a column vector. This is more efficient
+	/// than computing the inverse in one-shot cases.
+	b2Vec2 Solve(const b2Vec2& b) const
+	{
+		float32 a11 = ex.x, a12 = ey.x, a21 = ex.y, a22 = ey.y;
+		float32 det = a11 * a22 - a12 * a21;
+		if (det != 0.0f)
+		{
+			det = 1.0f / det;
+		}
+		b2Vec2 x;
+		x.x = det * (a22 * b.x - a12 * b.y);
+		x.y = det * (a11 * b.y - a21 * b.x);
+		return x;
+	}
+
+	b2Vec2 ex, ey;
+};
+
+/// A 3-by-3 matrix. Stored in column-major order.
+struct b2Mat33
+{
+	/// The default constructor does nothing (for performance).
+	b2Mat33() {}
+
+	/// Construct this matrix using columns.
+	b2Mat33(const b2Vec3& c1, const b2Vec3& c2, const b2Vec3& c3)
+	{
+		ex = c1;
+		ey = c2;
+		ez = c3;
+	}
+
+	/// Set this matrix to all zeros.
+	void SetZero()
+	{
+		ex.SetZero();
+		ey.SetZero();
+		ez.SetZero();
+	}
+
+	/// Solve A * x = b, where b is a column vector. This is more efficient
+	/// than computing the inverse in one-shot cases.
+	b2Vec3 Solve33(const b2Vec3& b) const;
+
+	/// Solve A * x = b, where b is a column vector. This is more efficient
+	/// than computing the inverse in one-shot cases. Solve only the upper
+	/// 2-by-2 matrix equation.
+	b2Vec2 Solve22(const b2Vec2& b) const;
+
+	/// Get the inverse of this matrix as a 2-by-2.
+	/// Returns the zero matrix if singular.
+	void GetInverse22(b2Mat33* M) const;
+
+	/// Get the symmetric inverse of this matrix as a 3-by-3.
+	/// Returns the zero matrix if singular.
+	void GetSymInverse33(b2Mat33* M) const;
+
+	b2Vec3 ex, ey, ez;
+};
+
+/// Rotation
+struct b2Rot
+{
+	b2Rot() {}
+
+	/// Initialize from an angle in radians
+	explicit b2Rot(float32 angle)
+	{
+		/// TODO_ERIN optimize
+		s = sinf(angle);
+		c = cosf(angle);
+	}
+
+	/// Set using an angle in radians.
+	void Set(float32 angle)
+	{
+		/// TODO_ERIN optimize
+		s = sinf(angle);
+		c = cosf(angle);
+	}
+
+	/// Set to the identity rotation
+	void SetIdentity()
+	{
+		s = 0.0f;
+		c = 1.0f;
+	}
+
+	/// Get the angle in radians
+	float32 GetAngle() const
+	{
+		return b2Atan2(s, c);
+	}
+
+	/// Get the x-axis
+	b2Vec2 GetXAxis() const
+	{
+		return b2Vec2(c, s);
+	}
+
+	/// Get the u-axis
+	b2Vec2 GetYAxis() const
+	{
+		return b2Vec2(-s, c);
+	}
+
+	/// Sine and cosine
+	float32 s, c;
+};
+
+/// A transform contains translation and rotation. It is used to represent
+/// the position and orientation of rigid frames.
+struct b2Transform
+{
+	/// The default constructor does nothing.
+	b2Transform() {}
+
+	/// Initialize using a position vector and a rotation.
+	b2Transform(const b2Vec2& position, const b2Rot& rotation) : p(position), q(rotation) {}
+
+	/// Set this to the identity transform.
+	void SetIdentity()
+	{
+		p.SetZero();
+		q.SetIdentity();
+	}
+
+	/// Set this based on the position and angle.
+	void Set(const b2Vec2& position, float32 angle)
+	{
+		p = position;
+		q.Set(angle);
+	}
+
+#if LIQUIDFUN_EXTERNAL_LANGUAGE_API
+	/// Get x-coordinate of p.
+	float32 GetPositionX() const { return p.x; }
+
+	/// Get y-coordinate of p.
+	float32 GetPositionY() const { return p.y; }
+
+	/// Get sine-component of q.
+	float32 GetRotationSin() const { return q.s; }
+
+	/// Get cosine-component of q.
+	float32 GetRotationCos() const { return q.c; }
+#endif // LIQUIDFUN_EXTERNAL_LANGUAGE_API
+
+	b2Vec2 p;
+	b2Rot q;
+};
+
+/// This describes the motion of a body/shape for TOI computation.
+/// Shapes are defined with respect to the body origin, which may
+/// no coincide with the center of mass. However, to support dynamics
+/// we must interpolate the center of mass position.
+struct b2Sweep
+{
+	/// Get the interpolated transform at a specific time.
+	/// @param beta is a factor in [0,1], where 0 indicates alpha0.
+	void GetTransform(b2Transform* xfb, float32 beta) const;
+
+	/// Advance the sweep forward, yielding a new initial state.
+	/// @param alpha the new initial time.
+	void Advance(float32 alpha);
+
+	/// Normalize the angles.
+	void Normalize();
+
+	b2Vec2 localCenter;	///< local center of mass position
+	b2Vec2 c0, c;		///< center world positions
+	float32 a0, a;		///< world angles
+
+	/// Fraction of the current time step in the range [0,1]
+	/// c0 and a0 are the positions at alpha0.
+	float32 alpha0;
+};
+
+/// Useful constant
+extern const b2Vec2 b2Vec2_zero;
+
+/// Perform the dot product on two vectors.
+inline float32 b2Dot(const b2Vec2& a, const b2Vec2& b)
+{
+	return a.x * b.x + a.y * b.y;
+}
+
+/// Perform the cross product on two vectors. In 2D this produces a scalar.
+inline float32 b2Cross(const b2Vec2& a, const b2Vec2& b)
+{
+	return a.x * b.y - a.y * b.x;
+}
+
+/// Perform the cross product on a vector and a scalar. In 2D this produces
+/// a vector.
+inline b2Vec2 b2Cross(const b2Vec2& a, float32 s)
+{
+	return b2Vec2(s * a.y, -s * a.x);
+}
+
+/// Perform the cross product on a scalar and a vector. In 2D this produces
+/// a vector.
+inline b2Vec2 b2Cross(float32 s, const b2Vec2& a)
+{
+	return b2Vec2(-s * a.y, s * a.x);
+}
+
+/// Multiply a matrix times a vector. If a rotation matrix is provided,
+/// then this transforms the vector from one frame to another.
+inline b2Vec2 b2Mul(const b2Mat22& A, const b2Vec2& v)
+{
+	return b2Vec2(A.ex.x * v.x + A.ey.x * v.y, A.ex.y * v.x + A.ey.y * v.y);
+}
+
+/// Multiply a matrix transpose times a vector. If a rotation matrix is provided,
+/// then this transforms the vector from one frame to another (inverse transform).
+inline b2Vec2 b2MulT(const b2Mat22& A, const b2Vec2& v)
+{
+	return b2Vec2(b2Dot(v, A.ex), b2Dot(v, A.ey));
+}
+
+/// Add two vectors component-wise.
+inline b2Vec2 operator + (const b2Vec2& a, const b2Vec2& b)
+{
+	return b2Vec2(a.x + b.x, a.y + b.y);
+}
+
+/// Subtract two vectors component-wise.
+inline b2Vec2 operator - (const b2Vec2& a, const b2Vec2& b)
+{
+	return b2Vec2(a.x - b.x, a.y - b.y);
+}
+
+inline b2Vec2 operator * (float32 s, const b2Vec2& a)
+{
+	return b2Vec2(s * a.x, s * a.y);
+}
+
+inline bool operator == (const b2Vec2& a, const b2Vec2& b)
+{
+	return a.x == b.x && a.y == b.y;
+}
+
+inline bool operator != (const b2Vec2& a, const b2Vec2& b)
+{
+	return !operator==(a, b);
+}
+
+inline float32 b2Distance(const b2Vec2& a, const b2Vec2& b)
+{
+	b2Vec2 c = a - b;
+	return c.Length();
+}
+
+inline float32 b2DistanceSquared(const b2Vec2& a, const b2Vec2& b)
+{
+	b2Vec2 c = a - b;
+	return b2Dot(c, c);
+}
+
+inline b2Vec3 operator * (float32 s, const b2Vec3& a)
+{
+	return b2Vec3(s * a.x, s * a.y, s * a.z);
+}
+
+/// Add two vectors component-wise.
+inline b2Vec3 operator + (const b2Vec3& a, const b2Vec3& b)
+{
+	return b2Vec3(a.x + b.x, a.y + b.y, a.z + b.z);
+}
+
+/// Subtract two vectors component-wise.
+inline b2Vec3 operator - (const b2Vec3& a, const b2Vec3& b)
+{
+	return b2Vec3(a.x - b.x, a.y - b.y, a.z - b.z);
+}
+
+/// Perform the dot product on two vectors.
+inline float32 b2Dot(const b2Vec3& a, const b2Vec3& b)
+{
+	return a.x * b.x + a.y * b.y + a.z * b.z;
+}
+
+/// Perform the cross product on two vectors.
+inline b2Vec3 b2Cross(const b2Vec3& a, const b2Vec3& b)
+{
+	return b2Vec3(a.y * b.z - a.z * b.y, a.z * b.x - a.x * b.z, a.x * b.y - a.y * b.x);
+}
+
+inline b2Mat22 operator + (const b2Mat22& A, const b2Mat22& B)
+{
+	return b2Mat22(A.ex + B.ex, A.ey + B.ey);
+}
+
+// A * B
+inline b2Mat22 b2Mul(const b2Mat22& A, const b2Mat22& B)
+{
+	return b2Mat22(b2Mul(A, B.ex), b2Mul(A, B.ey));
+}
+
+// A^T * B
+inline b2Mat22 b2MulT(const b2Mat22& A, const b2Mat22& B)
+{
+	b2Vec2 c1(b2Dot(A.ex, B.ex), b2Dot(A.ey, B.ex));
+	b2Vec2 c2(b2Dot(A.ex, B.ey), b2Dot(A.ey, B.ey));
+	return b2Mat22(c1, c2);
+}
+
+/// Multiply a matrix times a vector.
+inline b2Vec3 b2Mul(const b2Mat33& A, const b2Vec3& v)
+{
+	return v.x * A.ex + v.y * A.ey + v.z * A.ez;
+}
+
+/// Multiply a matrix times a vector.
+inline b2Vec2 b2Mul22(const b2Mat33& A, const b2Vec2& v)
+{
+	return b2Vec2(A.ex.x * v.x + A.ey.x * v.y, A.ex.y * v.x + A.ey.y * v.y);
+}
+
+/// Multiply two rotations: q * r
+inline b2Rot b2Mul(const b2Rot& q, const b2Rot& r)
+{
+	// [qc -qs] * [rc -rs] = [qc*rc-qs*rs -qc*rs-qs*rc]
+	// [qs  qc]   [rs  rc]   [qs*rc+qc*rs -qs*rs+qc*rc]
+	// s = qs * rc + qc * rs
+	// c = qc * rc - qs * rs
+	b2Rot qr;
+	qr.s = q.s * r.c + q.c * r.s;
+	qr.c = q.c * r.c - q.s * r.s;
+	return qr;
+}
+
+/// Transpose multiply two rotations: qT * r
+inline b2Rot b2MulT(const b2Rot& q, const b2Rot& r)
+{
+	// [ qc qs] * [rc -rs] = [qc*rc+qs*rs -qc*rs+qs*rc]
+	// [-qs qc]   [rs  rc]   [-qs*rc+qc*rs qs*rs+qc*rc]
+	// s = qc * rs - qs * rc
+	// c = qc * rc + qs * rs
+	b2Rot qr;
+	qr.s = q.c * r.s - q.s * r.c;
+	qr.c = q.c * r.c + q.s * r.s;
+	return qr;
+}
+
+/// Rotate a vector
+inline b2Vec2 b2Mul(const b2Rot& q, const b2Vec2& v)
+{
+	return b2Vec2(q.c * v.x - q.s * v.y, q.s * v.x + q.c * v.y);
+}
+
+/// Inverse rotate a vector
+inline b2Vec2 b2MulT(const b2Rot& q, const b2Vec2& v)
+{
+	return b2Vec2(q.c * v.x + q.s * v.y, -q.s * v.x + q.c * v.y);
+}
+
+inline b2Vec2 b2Mul(const b2Transform& T, const b2Vec2& v)
+{
+	float32 x = (T.q.c * v.x - T.q.s * v.y) + T.p.x;
+	float32 y = (T.q.s * v.x + T.q.c * v.y) + T.p.y;
+
+	return b2Vec2(x, y);
+}
+
+inline b2Vec2 b2MulT(const b2Transform& T, const b2Vec2& v)
+{
+	float32 px = v.x - T.p.x;
+	float32 py = v.y - T.p.y;
+	float32 x = (T.q.c * px + T.q.s * py);
+	float32 y = (-T.q.s * px + T.q.c * py);
+
+	return b2Vec2(x, y);
+}
+
+// v2 = A.q.Rot(B.q.Rot(v1) + B.p) + A.p
+//    = (A.q * B.q).Rot(v1) + A.q.Rot(B.p) + A.p
+inline b2Transform b2Mul(const b2Transform& A, const b2Transform& B)
+{
+	b2Transform C;
+	C.q = b2Mul(A.q, B.q);
+	C.p = b2Mul(A.q, B.p) + A.p;
+	return C;
+}
+
+// v2 = A.q' * (B.q * v1 + B.p - A.p)
+//    = A.q' * B.q * v1 + A.q' * (B.p - A.p)
+inline b2Transform b2MulT(const b2Transform& A, const b2Transform& B)
+{
+	b2Transform C;
+	C.q = b2MulT(A.q, B.q);
+	C.p = b2MulT(A.q, B.p - A.p);
+	return C;
+}
+
+template <typename T>
+inline T b2Abs(T a)
+{
+	return a > T(0) ? a : -a;
+}
+
+inline b2Vec2 b2Abs(const b2Vec2& a)
+{
+	return b2Vec2(b2Abs(a.x), b2Abs(a.y));
+}
+
+inline b2Mat22 b2Abs(const b2Mat22& A)
+{
+	return b2Mat22(b2Abs(A.ex), b2Abs(A.ey));
+}
+
+template <typename T>
+inline T b2Min(T a, T b)
+{
+	return a < b ? a : b;
+}
+
+inline b2Vec2 b2Min(const b2Vec2& a, const b2Vec2& b)
+{
+	return b2Vec2(b2Min(a.x, b.x), b2Min(a.y, b.y));
+}
+
+template <typename T>
+inline T b2Max(T a, T b)
+{
+	return a > b ? a : b;
+}
+
+inline b2Vec2 b2Max(const b2Vec2& a, const b2Vec2& b)
+{
+	return b2Vec2(b2Max(a.x, b.x), b2Max(a.y, b.y));
+}
+
+template <typename T>
+inline T b2Clamp(T a, T low, T high)
+{
+	return b2Max(low, b2Min(a, high));
+}
+
+inline b2Vec2 b2Clamp(const b2Vec2& a, const b2Vec2& low, const b2Vec2& high)
+{
+	return b2Max(low, b2Min(a, high));
+}
+
+template<typename T> inline void b2Swap(T& a, T& b)
+{
+	T tmp = a;
+	a = b;
+	b = tmp;
+}
+
+/// "Next Largest Power of 2
+/// Given a binary integer value x, the next largest power of 2 can be computed by a SWAR algorithm
+/// that recursively "folds" the upper bits into the lower bits. This process yields a bit vector with
+/// the same most significant 1 as x, but all 1's below it. Adding 1 to that value yields the next
+/// largest power of 2. For a 32-bit value:"
+inline uint32 b2NextPowerOfTwo(uint32 x)
+{
+	x |= (x >> 1);
+	x |= (x >> 2);
+	x |= (x >> 4);
+	x |= (x >> 8);
+	x |= (x >> 16);
+	return x + 1;
+}
+
+inline bool b2IsPowerOfTwo(uint32 x)
+{
+	bool result = x > 0 && (x & (x - 1)) == 0;
+	return result;
+}
+
+inline void b2Sweep::GetTransform(b2Transform* xf, float32 beta) const
+{
+	xf->p = (1.0f - beta) * c0 + beta * c;
+	float32 angle = (1.0f - beta) * a0 + beta * a;
+	xf->q.Set(angle);
+
+	// Shift to origin
+	xf->p -= b2Mul(xf->q, localCenter);
+}
+
+inline void b2Sweep::Advance(float32 alpha)
+{
+	b2Assert(alpha0 < 1.0f);
+	float32 beta = (alpha - alpha0) / (1.0f - alpha0);
+	c0 += beta * (c - c0);
+	a0 += beta * (a - a0);
+	alpha0 = alpha;
+}
+
+/// Normalize an angle in radians to be between -pi and pi
+inline void b2Sweep::Normalize()
+{
+	float32 twoPi = 2.0f * b2_pi;
+	float32 d =  twoPi * floorf(a0 / twoPi);
+	a0 -= d;
+	a -= d;
+}
+
+#endif

deps/include/Box2D/Common/b2Settings.cpp

@@ -0,0 +1,136 @@
+/*
+* Copyright (c) 2006-2009 Erin Catto http://www.box2d.org
+* Copyright (c) 2013 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#include <Box2D/Common/b2Settings.h>
+#include <stdio.h>
+#include <stdarg.h>
+#include <stdlib.h>
+
+b2Version b2_version = {2, 3, 0};
+
+#define LIQUIDFUN_VERSION_MAJOR 1
+#define LIQUIDFUN_VERSION_MINOR 1
+#define LIQUIDFUN_VERSION_REVISION 0
+#define LIQUIDFUN_STRING_EXPAND(X) #X
+#define LIQUIDFUN_STRING(X) LIQUIDFUN_STRING_EXPAND(X)
+
+static void* b2AllocDefault(int32 size, void* callbackData);
+static void b2FreeDefault(void* mem, void* callbackData);
+
+const b2Version b2_liquidFunVersion = {
+	LIQUIDFUN_VERSION_MAJOR, LIQUIDFUN_VERSION_MINOR,
+	LIQUIDFUN_VERSION_REVISION,
+};
+
+const char *b2_liquidFunVersionString =
+	"LiquidFun "
+	LIQUIDFUN_STRING(LIQUIDFUN_VERSION_MAJOR) "."
+	LIQUIDFUN_STRING(LIQUIDFUN_VERSION_MINOR) "."
+	LIQUIDFUN_STRING(LIQUIDFUN_VERSION_REVISION);
+
+static int32 b2_numAllocs = 0;
+
+// Initialize default allocator.
+static b2AllocFunction b2_allocCallback = b2AllocDefault;
+static b2FreeFunction b2_freeCallback = b2FreeDefault;
+static void *b2_callbackData = NULL;
+
+// Default implementation of b2AllocFunction.
+static void* b2AllocDefault(int32 size, void* callbackData)
+{
+	B2_NOT_USED(callbackData);
+	return malloc(size);
+}
+
+// Default implementation of b2FreeFunction.
+static void b2FreeDefault(void* mem, void* callbackData)
+{
+	B2_NOT_USED(callbackData);
+	free(mem);
+}
+
+/// Set alloc and free callbacks to override the default behavior of using
+/// malloc() and free() for dynamic memory allocation.
+/// Set allocCallback and freeCallback to NULL to restore the default
+/// allocator (malloc / free).
+void b2SetAllocFreeCallbacks(b2AllocFunction allocCallback,
+							 b2FreeFunction freeCallback, void* callbackData)
+{
+	b2Assert((allocCallback && freeCallback) ||
+			 (!allocCallback && !freeCallback));
+	b2Assert(0 == b2GetNumAllocs());
+	if (allocCallback && freeCallback)
+	{
+		b2_allocCallback = allocCallback;
+		b2_freeCallback = freeCallback;
+		b2_callbackData = callbackData;
+	}
+	else
+	{
+		b2_allocCallback = b2AllocDefault;
+		b2_freeCallback = b2FreeDefault;
+		b2_callbackData = NULL;
+	}
+}
+
+// Memory allocators. Modify these to use your own allocator.
+void* b2Alloc(int32 size)
+{
+	b2_numAllocs++;
+	return b2_allocCallback(size, b2_callbackData);
+}
+
+void b2Free(void* mem)
+{
+	b2_numAllocs--;
+	b2_freeCallback(mem, b2_callbackData);
+}
+
+void b2SetNumAllocs(const int32 numAllocs)
+{
+	b2_numAllocs = numAllocs;
+}
+
+int32 b2GetNumAllocs()
+{
+	return b2_numAllocs;
+}
+
+// You can modify this to use your logging facility.
+void b2Log(const char* string, ...)
+{
+#if DEBUG
+	va_list args;
+	va_start(args, string);
+	vprintf(string, args);
+	va_end(args);
+#else
+	B2_NOT_USED(string);
+#endif
+}
+
+class Validator
+{
+public:
+	Validator()
+	{
+		b2Assert(sizeof(uint64)==8);
+		b2Assert(sizeof(int64)==8);
+	}
+} validate;

deps/include/Box2D/Common/b2Settings.h

@@ -0,0 +1,267 @@
+/*
+* Copyright (c) 2006-2009 Erin Catto http://www.box2d.org
+* Copyright (c) 2013 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#ifndef B2_SETTINGS_H
+#define B2_SETTINGS_H
+
+#include <stddef.h>
+#include <assert.h>
+#include <float.h>
+
+#define B2_NOT_USED(x) ((void)(x))
+#if DEBUG && !defined(NDEBUG)
+#define b2Assert(A) assert(A)
+#define B2_ASSERT_ENABLED 1
+#else
+#define b2Assert(A)
+#define B2_ASSERT_ENABLED 0
+#endif
+
+// Statement which is compiled out when DEBUG isn't defined.
+#if DEBUG
+#define B2_DEBUG_STATEMENT(A) A
+#else
+#define B2_DEBUG_STATEMENT(A)
+#endif  // DEBUG
+
+// Calculate the size of a static array.
+#define B2_ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
+
+typedef signed char	int8;
+typedef signed short int16;
+typedef signed int int32;
+typedef unsigned char uint8;
+typedef unsigned short uint16;
+typedef unsigned int uint32;
+typedef float float32;
+typedef double float64;
+
+#ifdef WIN32
+typedef __int64   int64;
+typedef unsigned __int64   uint64;
+#else // !WIN32
+typedef long long int64;
+typedef unsigned long long uint64;
+#endif
+
+#define	b2_maxFloat		FLT_MAX
+#define	b2_epsilon		FLT_EPSILON
+#define b2_pi			3.14159265359f
+
+#if !defined(b2Inline)
+#if defined(__GNUC__)
+#define b2Inline __attribute__((always_inline))
+#else
+#define b2Inline inline
+#endif // defined(__GNUC__)
+#endif // !defined(b2Inline)
+
+// We expand the API so that other languages (e.g. Java) can call into
+// our C++ more easily. Only set if when the flag is not externally defined.
+#if !defined(LIQUIDFUN_EXTERNAL_LANGUAGE_API)
+#if SWIG || LIQUIDFUN_UNIT_TESTS
+#define LIQUIDFUN_EXTERNAL_LANGUAGE_API 1
+#else
+#define LIQUIDFUN_EXTERNAL_LANGUAGE_API 0
+#endif
+#endif
+
+/// @file
+/// Global tuning constants based on meters-kilograms-seconds (MKS) units.
+///
+
+// Collision
+
+/// The maximum number of contact points between two convex shapes. Do
+/// not change this value.
+#define b2_maxManifoldPoints	2
+
+/// The maximum number of vertices on a convex polygon. You cannot increase
+/// this too much because b2BlockAllocator has a maximum object size.
+#define b2_maxPolygonVertices	8
+
+/// This is used to fatten AABBs in the dynamic tree. This allows proxies
+/// to move by a small amount without triggering a tree adjustment.
+/// This is in meters.
+#define b2_aabbExtension		0.1f
+
+/// This is used to fatten AABBs in the dynamic tree. This is used to predict
+/// the future position based on the current displacement.
+/// This is a dimensionless multiplier.
+#define b2_aabbMultiplier		2.0f
+
+/// A small length used as a collision and constraint tolerance. Usually it is
+/// chosen to be numerically significant, but visually insignificant.
+#define b2_linearSlop			0.005f
+
+/// A small angle used as a collision and constraint tolerance. Usually it is
+/// chosen to be numerically significant, but visually insignificant.
+#define b2_angularSlop			(2.0f / 180.0f * b2_pi)
+
+/// The radius of the polygon/edge shape skin. This should not be modified. Making
+/// this smaller means polygons will have an insufficient buffer for continuous collision.
+/// Making it larger may create artifacts for vertex collision.
+#define b2_polygonRadius		(2.0f * b2_linearSlop)
+
+/// Maximum number of sub-steps per contact in continuous physics simulation.
+#define b2_maxSubSteps			8
+
+
+// Dynamics
+
+/// Maximum number of contacts to be handled to solve a TOI impact.
+#define b2_maxTOIContacts			32
+
+/// A velocity threshold for elastic collisions. Any collision with a relative linear
+/// velocity below this threshold will be treated as inelastic.
+#define b2_velocityThreshold		1.0f
+
+/// The maximum linear position correction used when solving constraints. This helps to
+/// prevent overshoot.
+#define b2_maxLinearCorrection		0.2f
+
+/// The maximum angular position correction used when solving constraints. This helps to
+/// prevent overshoot.
+#define b2_maxAngularCorrection		(8.0f / 180.0f * b2_pi)
+
+/// The maximum linear velocity of a body. This limit is very large and is used
+/// to prevent numerical problems. You shouldn't need to adjust this.
+#define b2_maxTranslation			2.0f
+#define b2_maxTranslationSquared	(b2_maxTranslation * b2_maxTranslation)
+
+/// The maximum angular velocity of a body. This limit is very large and is used
+/// to prevent numerical problems. You shouldn't need to adjust this.
+#define b2_maxRotation				(0.5f * b2_pi)
+#define b2_maxRotationSquared		(b2_maxRotation * b2_maxRotation)
+
+/// This scale factor controls how fast overlap is resolved. Ideally this would be 1 so
+/// that overlap is removed in one time step. However using values close to 1 often lead
+/// to overshoot.
+#define b2_baumgarte				0.2f
+#define b2_toiBaugarte				0.75f
+
+
+// Particle
+
+/// NEON SIMD requires 16-bit particle indices
+#if !defined(B2_USE_16_BIT_PARTICLE_INDICES) && defined(LIQUIDFUN_SIMD_NEON)
+#define B2_USE_16_BIT_PARTICLE_INDICES
+#endif
+
+/// A symbolic constant that stands for particle allocation error.
+#define b2_invalidParticleIndex		(-1)
+
+#ifdef B2_USE_16_BIT_PARTICLE_INDICES
+#define b2_maxParticleIndex			0x7FFF
+#else
+#define b2_maxParticleIndex			0x7FFFFFFF
+#endif
+
+/// The default distance between particles, multiplied by the particle diameter.
+#define b2_particleStride			0.75f
+
+/// The minimum particle weight that produces pressure.
+#define b2_minParticleWeight			1.0f
+
+/// The upper limit for particle pressure.
+#define b2_maxParticlePressure		0.25f
+
+/// The upper limit for force between particles.
+#define b2_maxParticleForce		0.5f
+
+/// The maximum distance between particles in a triad, multiplied by the
+/// particle diameter.
+#define b2_maxTriadDistance			2
+#define b2_maxTriadDistanceSquared		(b2_maxTriadDistance * b2_maxTriadDistance)
+
+/// The initial size of particle data buffers.
+#define b2_minParticleSystemBufferCapacity	256
+
+/// The time into the future that collisions against barrier particles will be detected.
+#define b2_barrierCollisionTime 2.5f
+
+// Sleep
+
+/// The time that a body must be still before it will go to sleep.
+#define b2_timeToSleep				0.5f
+
+/// A body cannot sleep if its linear velocity is above this tolerance.
+#define b2_linearSleepTolerance		0.01f
+
+/// A body cannot sleep if its angular velocity is above this tolerance.
+#define b2_angularSleepTolerance	(2.0f / 180.0f * b2_pi)
+
+// Memory Allocation
+
+/// Implement this function to use your own memory allocator.
+void* b2Alloc(int32 size);
+
+/// If you implement b2Alloc, you should also implement this function.
+void b2Free(void* mem);
+
+/// Use this function to override b2Alloc() without recompiling this library.
+typedef void* (*b2AllocFunction)(int32 size, void* callbackData);
+/// Use this function to override b2Free() without recompiling this library.
+typedef void (*b2FreeFunction)(void* mem, void* callbackData);
+
+/// Set alloc and free callbacks to override the default behavior of using
+/// malloc() and free() for dynamic memory allocation.
+/// Set allocCallback and freeCallback to NULL to restore the default
+/// allocator (malloc / free).
+void b2SetAllocFreeCallbacks(b2AllocFunction allocCallback,
+							 b2FreeFunction freeCallback,
+							 void* callbackData);
+
+/// Set the number of calls to b2Alloc minus the number of calls to b2Free.
+/// This can be used to disable the empty heap check in
+/// b2SetAllocFreeCallbacks() which can be useful for testing.
+void b2SetNumAllocs(const int32 numAllocs);
+
+/// Get number of calls to b2Alloc minus number of calls to b2Free.
+int32 b2GetNumAllocs();
+
+/// Logging function.
+void b2Log(const char* string, ...);
+
+/// Version numbering scheme.
+/// See http://en.wikipedia.org/wiki/Software_versioning
+struct b2Version
+{
+	int32 major;		///< significant changes
+	int32 minor;		///< incremental changes
+	int32 revision;		///< bug fixes
+};
+
+/// Current version.
+/// Version of Box2D, LiquidFun is based upon.
+extern b2Version b2_version;
+
+/// Global variable is used to identify the version of LiquidFun.
+extern const b2Version b2_liquidFunVersion;
+/// String which identifies the current version of LiquidFun.
+/// b2_liquidFunVersionString is used by Google developers to identify which
+/// applications uploaded to Google Play are using this library.  This allows
+/// the development team at Google to determine the popularity of the library.
+/// How it works: Applications that are uploaded to the Google Play Store are
+/// scanned for this version string.  We track which applications are using it
+/// to measure popularity.  You are free to remove it (of course) but we would
+/// appreciate if you left it in.
+extern const char *b2_liquidFunVersionString;
+
+#endif

deps/include/Box2D/Common/b2SlabAllocator.h

@@ -0,0 +1,244 @@
+/*
+* Copyright (c) 2014 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+#ifndef B2_SLAB_ALLOCATOR_H
+#define B2_SLAB_ALLOCATOR_H
+
+#include <stddef.h>
+#include <stdint.h>
+#include <new>
+#include <Box2D/Common/b2IntrusiveList.h>
+#include <Box2D/Common/b2FreeList.h>
+#include <Box2D/Common/b2Settings.h>
+#include <Box2D/Common/b2TrackedBlock.h>
+
+/// Freelist based allocator for fixed sized items from slabs (memory
+/// preallocated from the heap).
+/// T should be a class which has a default constructor and implements the
+/// member function "b2IntrusiveList* GetListNode()".
+/// All objects in a slab are constructed when a slab is created and destructed
+/// when a slab is freed.
+template<typename T>
+class b2SlabAllocator
+{
+private:
+	// Information about a slab.
+	class Slab
+	{
+	public:
+		/// Initialize a slab with the number of items it contains.
+		Slab(uint32 numberOfItems) :
+			m_numberOfItems(numberOfItems)
+		{
+			B2_NOT_USED(m_padding);
+			// This assumes that this class is packed on at least a 4-byte
+			// boundary with no padding.  Verify the assumption.
+			b2Assert(sizeof(*this) == b2_mallocAlignment);
+		}
+
+		/// Empty destructor.
+		~Slab() { }
+
+		/// Get the number of items in this slab.
+		uint32 GetNumberOfItems() const { return m_numberOfItems; }
+
+		/// Get a pointer to the first item in the slab.
+		T* GetFirstItem() const
+		{
+			return (T*)((uint8*)(this + 1));
+		}
+
+		/// Get a pointer to the end of the slab.
+		/// NOTE: This is a pointer after the last byte of the slab not the
+		/// last item in the slab.
+		T* GetItemEnd() const { return GetFirstItem() + GetNumberOfItems(); }
+
+	private:
+		/// Number of items in the slab.
+		uint32 m_numberOfItems;
+		/// Padding to align the first item in the slab to b2_mallocAlignment.
+		uint8 m_padding[b2_mallocAlignment - sizeof(uint32)];
+	};
+
+public:
+	/// Initialize the allocator to allocate itemsPerSlab of type T for each
+	/// slab that is allocated.
+	b2SlabAllocator(const uint32 itemsPerSlab) :
+		m_itemsPerSlab(itemsPerSlab)
+	{
+	}
+
+	/// Free all allocated slabs.
+	~b2SlabAllocator()
+	{
+		FreeAllSlabs();
+	}
+
+	/// Set size of the next allocated slab using the number of items per
+	/// slab.  Setting this value to zero disables further slab allocation.
+	void SetItemsPerSlab(uint32 itemsPerSlab)
+	{
+		m_itemsPerSlab = itemsPerSlab;
+	}
+
+	// Get the size of the next allocated slab.
+	uint32 GetItemsPerSlab() const
+	{
+		return m_itemsPerSlab;
+	}
+
+	/// Allocate a item from the slab.
+	T* Allocate()
+	{
+		// Allocate a slab if needed here.
+		if (m_freeList.GetFreeList()->GetFreeList().IsEmpty() &&
+			!AllocateSlab())
+			return NULL;
+		return m_freeList.Allocate();
+	}
+
+	/// Free an item from the slab.
+	void Free(T *object)
+	{
+		m_freeList.Free(object);
+	}
+
+	/// Allocate a slab, construct instances of T and add them to the free
+	/// pool.
+	bool AllocateSlab()
+	{
+		if (!m_itemsPerSlab) return false;
+		const uint32 slabSize = sizeof(Slab) + (sizeof(T) * m_itemsPerSlab);
+		void* const memory = m_slabs.Allocate(slabSize);
+		if (!memory) return false;
+
+		Slab* const slab = new (BlockGetSlab(memory)) Slab(m_itemsPerSlab);
+		T* item = slab->GetFirstItem();
+		for (uint32 i = 0; i < m_itemsPerSlab; ++i, ++item)
+		{
+			m_freeList.AddToFreeList(new (item) T);
+		}
+		return true;
+	}
+
+	/// Free all slabs.
+	void FreeAllSlabs()
+	{
+		const b2TypedIntrusiveListNode<b2TrackedBlock>& slabList =
+			m_slabs.GetList();
+		while (!slabList.IsEmpty())
+		{
+			FreeSlab(BlockGetSlab(slabList.GetNext()->GetMemory()));
+		}
+	}
+
+	/// Free all empty slabs.
+	/// This method is slow - O(M^N) - since this class doesn't track
+	/// the association between each item and slab.
+	void FreeEmptySlabs()
+	{
+		const b2IntrusiveListNode& freeItemList =
+			m_freeList.GetFreeList()->GetFreeList();
+		const b2IntrusiveListNode* freeItemListTerminator =
+			freeItemList.GetTerminator();
+		const b2TypedIntrusiveListNode<b2TrackedBlock>& slabList =
+			m_slabs.GetList();
+		const b2TypedIntrusiveListNode<b2TrackedBlock>* slabListTerminator =
+			slabList.GetTerminator();
+		b2TrackedBlock* block = slabList.GetNext();
+		while (block != slabListTerminator)
+		{
+			// Get the Slab from the memory associated with the block.
+			Slab* const slab = BlockGetSlab(block->GetMemory());
+			block = block->GetNext();
+
+			// Determine the range of memory the Slab owns.
+			const uint8* const slabItemStart = (uint8*)slab->GetFirstItem();
+			const uint8* const slabItemEnd = (uint8*)slab->GetItemEnd();
+
+			// Count all free items that are owned by the current slab.
+			uint8 freeItems = 0;
+			bool empty = false;
+			for (b2IntrusiveListNode* itemNode = freeItemList.GetNext();
+				 itemNode != freeItemListTerminator;
+				 itemNode = itemNode->GetNext())
+			{
+				const uint8* itemNodeAddress = (uint8*)itemNode;
+				if (itemNodeAddress >= slabItemStart &&
+					itemNodeAddress <= slabItemEnd)
+				{
+					++freeItems;
+					if (slab->GetNumberOfItems() == freeItems)
+					{
+						empty = true;
+						break;
+					}
+				}
+			}
+			// If a slab is empty, free it.
+			if (empty)
+			{
+				FreeSlab(slab);
+			}
+		}
+	}
+
+	/// Get the item allocator freelist.
+	const b2TypedFreeList<T>& GetFreeList() const
+	{
+		return m_freeList;
+	}
+
+private:
+	/// Destroy all objects in a slab and free the slab.
+	void FreeSlab(Slab * const slab)
+	{
+		b2Assert(slab);
+		const uint32 numberOfItems = slab->GetNumberOfItems();
+		T* item = slab->GetFirstItem();
+		for (uint32 i = 0; i < numberOfItems; ++i, ++item)
+		{
+			item->~T();
+		}
+		slab->~Slab();
+		m_slabs.Free(slab);
+	}
+
+	/// Get a pointer to a Slab from a block of memory in m_slabs.
+	Slab* BlockGetSlab(void *memory)
+	{
+		return (Slab*)memory;
+	}
+
+	/// Get a pointer to the first item in the array of items referenced by a
+	/// Slab.
+	T* SlabGetFirstItem(Slab* slab)
+	{
+		return (T*)(slab + 1);
+	}
+
+private:
+	/// Contains a list of b2TrackedBlock instances where each b2TrackedBlock's
+	/// associated user memory contains a Slab followed by instances of T.
+	b2TrackedBlockAllocator m_slabs;
+	/// Number of items to allocate in the next allocated slab.
+	uint32 m_itemsPerSlab;
+	/// Freelist which contains instances of T.
+	b2TypedFreeList<T> m_freeList;
+};
+
+#endif  // B2_SLAB_ALLOCATOR_H

deps/include/Box2D/Common/b2StackAllocator.cpp

@@ -0,0 +1,120 @@
+/*
+* Copyright (c) 2006-2009 Erin Catto http://www.box2d.org
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#include <Box2D/Common/b2StackAllocator.h>
+#include <Box2D/Common/b2Math.h>
+#include <string.h>
+
+b2StackAllocator::b2StackAllocator()
+{
+	m_index = 0;
+	m_allocation = 0;
+	m_maxAllocation = 0;
+	m_entryCount = 0;
+}
+
+b2StackAllocator::~b2StackAllocator()
+{
+	b2Assert(m_index == 0);
+	b2Assert(m_entryCount == 0);
+}
+
+void* b2StackAllocator::Allocate(int32 size)
+{
+	b2Assert(m_entryCount < b2_maxStackEntries);
+	const int32 roundedSize = (size + ALIGN_MASK) & ~ALIGN_MASK;
+	b2StackEntry* entry = m_entries + m_entryCount;
+	entry->size = roundedSize;
+	if (m_index + roundedSize > b2_stackSize)
+	{
+		entry->data = (char*)b2Alloc(roundedSize);
+		entry->usedMalloc = true;
+	}
+	else
+	{
+		entry->data = m_data + m_index;
+		entry->usedMalloc = false;
+		m_index += roundedSize;
+	}
+
+	m_allocation += roundedSize;
+	m_maxAllocation = b2Max(m_maxAllocation, m_allocation);
+	++m_entryCount;
+
+	return entry->data;
+}
+
+void* b2StackAllocator::Reallocate(void* p, int32 size)
+{
+	b2Assert(m_entryCount > 0);
+	b2StackEntry* entry = m_entries + m_entryCount - 1;
+	b2Assert(p == entry->data);
+	B2_NOT_USED(p);
+	int32 incrementSize = size - entry->size;
+	if (incrementSize > 0)
+	{
+		if (entry->usedMalloc)
+		{
+			void* data = b2Alloc(size);
+			memcpy(data, entry->data, entry->size);
+			b2Free(entry->data);
+			entry->data = (char*)data;
+		}
+		else if (m_index + incrementSize > b2_stackSize)
+		{
+			void* data = b2Alloc(size);
+			memcpy(data, entry->data, entry->size);
+			m_index -= entry->size;
+			entry->data = (char*)data;
+			entry->usedMalloc = true;
+		}
+		else
+		{
+			m_index += incrementSize;
+			m_allocation += incrementSize;
+			m_maxAllocation = b2Max(m_maxAllocation, m_allocation);
+		}
+		entry->size = size;
+	}
+
+	return entry->data;
+}
+
+void b2StackAllocator::Free(void* p)
+{
+	b2Assert(m_entryCount > 0);
+	b2StackEntry* entry = m_entries + m_entryCount - 1;
+	b2Assert(p == entry->data);
+	if (entry->usedMalloc)
+	{
+		b2Free(p);
+	}
+	else
+	{
+		m_index -= entry->size;
+	}
+	m_allocation -= entry->size;
+	--m_entryCount;
+
+	p = NULL;
+}
+
+int32 b2StackAllocator::GetMaxAllocation() const
+{
+	return m_maxAllocation;
+}

deps/include/Box2D/Common/b2StackAllocator.h

@@ -0,0 +1,64 @@
+/*
+* Copyright (c) 2006-2009 Erin Catto http://www.box2d.org
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#ifndef B2_STACK_ALLOCATOR_H
+#define B2_STACK_ALLOCATOR_H
+
+#include <Box2D/Common/b2Settings.h>
+
+const int32 b2_stackSize = 100 * 1024;	// 100k
+const int32 b2_maxStackEntries = 32;
+
+struct b2StackEntry
+{
+	char* data;
+	int32 size;
+	bool usedMalloc;
+};
+
+// This is a stack allocator used for fast per step allocations.
+// You must nest allocate/free pairs. The code will assert
+// if you try to interleave multiple allocate/free pairs.
+class b2StackAllocator
+{
+public:
+	enum { MIN_ALIGNMENT = sizeof(void*) }; // Must be a power of 2
+	enum { ALIGN_MASK = MIN_ALIGNMENT - 1 };
+
+	b2StackAllocator();
+	~b2StackAllocator();
+
+	void* Allocate(int32 size);
+	void* Reallocate(void* p, int32 size);
+	void Free(void* p);
+
+	int32 GetMaxAllocation() const;
+
+private:
+
+	char m_data[b2_stackSize];
+	int32 m_index;
+
+	int32 m_allocation;
+	int32 m_maxAllocation;
+
+	b2StackEntry m_entries[b2_maxStackEntries];
+	int32 m_entryCount;
+};
+
+#endif

deps/include/Box2D/Common/b2Stat.cpp

@@ -0,0 +1,67 @@
+/*
+* Copyright (c) 2013 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+#include "b2Stat.h"
+
+#include <algorithm>
+#include <cfloat>
+
+b2Stat::b2Stat()
+{
+	Clear();
+}
+
+void b2Stat::Record( float32 t )
+{
+	m_total += t;
+	m_min = std::min(m_min,t);
+	m_max = std::max(m_max,t);
+	m_count++;
+}
+
+int b2Stat::GetCount() const
+{
+	return m_count;
+}
+
+float32 b2Stat::GetMean() const
+{
+	if (m_count == 0)
+	{
+		return 0.0f;
+	}
+	return (float32)(m_total / m_count);
+}
+
+float32 b2Stat::GetMin() const
+{
+	return m_min;
+}
+
+float32 b2Stat::GetMax() const
+{
+	return m_max;
+}
+
+void b2Stat::Clear()
+{
+	m_count = 0;
+	m_total = 0;
+	m_min = FLT_MAX;
+	m_max = -FLT_MAX;
+}
+

deps/include/Box2D/Common/b2Stat.h

@@ -0,0 +1,57 @@
+/*
+* Copyright (c) 2013 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+#ifndef B2_STAT
+#define B2_STAT
+
+#include <Box2D/Common/b2Settings.h>
+
+/// Calculates min/max/mean of a set of samples
+class b2Stat
+{
+public:
+	b2Stat();
+
+	/// Record a sample
+	void Record( float32 t );
+
+	/// Returns the number of recorded samples
+	int GetCount() const;
+
+	/// Returns the mean of all recorded samples,
+	/// Returns 0 if there are no recorded samples
+	float32 GetMean() const;
+
+	/// Returns the min of all recorded samples,
+	/// FLT_MAX if there are no recorded samples
+	float32 GetMin() const;
+
+	/// Returns the max of all recorded samples,
+	/// -FLT_MAX if there are no recorded samples
+	float32 GetMax() const;
+
+	/// Erase all recorded samples
+	void Clear();
+private:
+
+	int m_count;
+	float64 m_total;
+	float32 m_min;
+	float32 m_max;
+};
+
+#endif

deps/include/Box2D/Common/b2Timer.cpp

@@ -0,0 +1,134 @@
+/*
+* Copyright (c) 2011 Erin Catto http://box2d.org
+* Copyright (c) 2014 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#include <Box2D/Common/b2Timer.h>
+
+#if defined(_WIN32)
+
+float64 b2Timer::s_invFrequency = 0.0f;
+
+#define WIN32_LEAN_AND_MEAN
+#include <windows.h>
+
+typedef BOOL (WINAPI *SystemGetTimeFunc)(_Out_ LARGE_INTEGER *lpFrequency);
+SystemGetTimeFunc systemGetTimeFunc = ::QueryPerformanceCounter;
+SystemGetTimeFunc systemGetFreqFunc = ::QueryPerformanceFrequency;
+
+int64 b2Timer::GetTicks()
+{
+	LARGE_INTEGER largeInteger;
+	systemGetTimeFunc(&largeInteger);
+	return largeInteger.QuadPart;
+}
+
+b2Timer::b2Timer()
+{
+	LARGE_INTEGER largeInteger;
+
+	if (s_invFrequency == 0.0f)
+	{
+		systemGetFreqFunc(&largeInteger);
+		s_invFrequency = float64(largeInteger.QuadPart);
+		if (s_invFrequency > 0.0f)
+		{
+			s_invFrequency = 1000.0f / s_invFrequency;
+		}
+	}
+
+	m_start = GetTicks();
+}
+
+void b2Timer::Reset()
+{
+	m_start = GetTicks();
+}
+
+float32 b2Timer::GetMilliseconds() const
+{
+	int64 elapsed = GetTicks() - m_start;
+	return (float32)(s_invFrequency * elapsed);
+}
+
+#elif defined(__linux__) || defined (__APPLE__)
+
+#include <sys/time.h>
+#include <time.h>
+
+// systemGetTimeFunc is defined with external linkage to allow unit
+// test to mock out the system time function
+
+#if defined(__linux__)
+
+typedef int (*SystemGetTimeFunc)(clockid_t clk_id, struct timespec *tp);
+SystemGetTimeFunc systemGetTimeFunc = ::clock_gettime;
+
+#elif defined(__APPLE__)
+
+typedef int (*SystemGetTimeFunc)(struct timeval * tp, void * tzp);
+SystemGetTimeFunc systemGetTimeFunc = ::gettimeofday;
+
+#endif
+
+int64 b2Timer::GetTicks()
+{
+	static const int NSEC_PER_SEC = 1000000000;
+
+#ifdef __linux__
+	timespec ts;
+	systemGetTimeFunc(CLOCK_MONOTONIC,&ts);
+	return ((int64)ts.tv_sec) * NSEC_PER_SEC + ts.tv_nsec;
+#else
+	timeval t;
+	systemGetTimeFunc(&t, 0);
+	return ((int64)t.tv_sec) * NSEC_PER_SEC + t.tv_usec * 1000;
+#endif
+}
+
+b2Timer::b2Timer()
+{
+	Reset();
+}
+
+void b2Timer::Reset()
+{
+	m_start = GetTicks();
+}
+
+float32 b2Timer::GetMilliseconds() const
+{
+	static const float32 kTicksToMs = 0.000001f;
+	return kTicksToMs * (float32)(GetTicks() - m_start);
+}
+
+#else
+
+b2Timer::b2Timer()
+{
+}
+
+void b2Timer::Reset()
+{
+}
+
+float32 b2Timer::GetMilliseconds() const
+{
+	return 0.0f;
+}
+
+#endif

deps/include/Box2D/Common/b2Timer.h

@@ -0,0 +1,50 @@
+/*
+* Copyright (c) 2011 Erin Catto http://box2d.org
+* Copyright (c) 2014 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#ifndef B2_TIMER_H
+#define B2_TIMER_H
+
+#include <Box2D/Common/b2Settings.h>
+
+/// Timer for profiling. This has platform specific code and may
+/// not work on every platform.
+class b2Timer
+{
+public:
+
+	/// Constructor
+	b2Timer();
+
+	/// Reset the timer.
+	void Reset();
+
+	/// Get the time since construction or the last reset.
+	float32 GetMilliseconds() const;
+
+private:
+	/// Get platform specific tick count
+	static int64 GetTicks();
+
+#if defined(_WIN32)
+	static float64 s_invFrequency;
+#endif
+	int64 m_start;
+};
+
+#endif

deps/include/Box2D/Common/b2TrackedBlock.cpp

@@ -0,0 +1,111 @@
+/*
+* Copyright (c) 2014 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+
+#include <Box2D/Common/b2TrackedBlock.h>
+#include <stddef.h>
+#include <stdint.h>
+#include <new>
+
+// Initialize this block with a reference to "this".
+b2TrackedBlock::b2TrackedBlock()
+{
+	b2TrackedBlock** pointerToThis =
+		(b2TrackedBlock**)((uint8*)GetMemory() - sizeof(b2TrackedBlock**));
+	*pointerToThis = this;
+}
+
+/// Get the allocated memory associated with this block.
+void* b2TrackedBlock::GetMemory() const
+{
+	// The size of data in this without padding.
+	static const uint32 kSizeOfThisWithNoPadding =
+		sizeof(*this) - sizeof(m_padding) + sizeof(b2TrackedBlock**);
+
+	// Make sure b2_mallocAlignment is base2.
+	b2Assert(((b2_mallocAlignment - 1) & b2_mallocAlignment) == 0);
+
+	// Round the pointer following data in this to b2_mallocAlignment.
+	uint8* const aligned = (uint8*)(
+		((uintptr_t)this + kSizeOfThisWithNoPadding + b2_mallocAlignment - 1) &
+		~((uintptr_t)b2_mallocAlignment - 1));
+	// Verify offset doesn't overlap data in this.
+	b2Assert((uintptr_t)aligned - (uintptr_t)this >= kSizeOfThisWithNoPadding);
+	return aligned;
+}
+
+/// Allocate a b2TrackedBlock returning a pointer to memory of size
+/// bytes that can be used by the caller.
+void* b2TrackedBlock::Allocate(uint32 size)
+{
+	void* memory = (b2TrackedBlock*)b2Alloc(sizeof(b2TrackedBlock) +
+											size);
+	if (!memory)
+	{
+		return NULL;
+	}
+	return (new(memory) b2TrackedBlock)->GetMemory();
+}
+
+/// Get a b2TrackedBlock from a pointer to memory returned by
+/// b2TrackedBlock::Allocate().
+b2TrackedBlock* b2TrackedBlock::GetFromMemory(void *memory)
+{
+	uint8* const aligned = (uint8*)memory;
+	b2Assert(memory);
+	b2TrackedBlock **blockPtr = (b2TrackedBlock**)(aligned -
+												   sizeof(b2TrackedBlock**));
+	b2Assert(*blockPtr);
+	return *blockPtr;
+}
+
+/// Free a block of memory returned by b2TrackedBlock::Allocate()
+void b2TrackedBlock::Free(void *memory)
+{
+	Free(GetFromMemory(memory));
+}
+
+/// Free a b2TrackedBlock.
+void b2TrackedBlock::Free(b2TrackedBlock *block)
+{
+	b2Assert(block);
+	block->~b2TrackedBlock();
+	b2Free(block);
+}
+
+/// Allocate a block of size bytes using b2TrackedBlock::Allocate().
+void* b2TrackedBlockAllocator::Allocate(uint32 size)
+{
+	void *memory = b2TrackedBlock::Allocate(size);
+	m_blocks.InsertBefore(b2TrackedBlock::GetFromMemory(memory));
+	return memory;
+}
+
+/// Free a block returned by Allocate().
+void b2TrackedBlockAllocator::Free(void *memory)
+{
+	b2TrackedBlock::Free(memory);
+}
+
+/// Free all allocated blocks.
+void b2TrackedBlockAllocator::FreeAll()
+{
+	while (!m_blocks.IsEmpty())
+	{
+		b2TrackedBlock::Free(m_blocks.GetNext());
+	}
+}

deps/include/Box2D/Common/b2TrackedBlock.h

@@ -0,0 +1,92 @@
+/*
+* Copyright (c) 2014 Google, Inc.
+*
+* This software is provided 'as-is', without any express or implied
+* warranty.  In no event will the authors be held liable for any damages
+* arising from the use of this software.
+* Permission is granted to anyone to use this software for any purpose,
+* including commercial applications, and to alter it and redistribute it
+* freely, subject to the following restrictions:
+* 1. The origin of this software must not be misrepresented; you must not
+* claim that you wrote the original software. If you use this software
+* in a product, an acknowledgment in the product documentation would be
+* appreciated but is not required.
+* 2. Altered source versions must be plainly marked as such, and must not be
+* misrepresented as being the original software.
+* 3. This notice may not be removed or altered from any source distribution.
+*/
+#ifndef B2_TRACKED_BLOCK_H
+#define B2_TRACKED_BLOCK_H
+
+#include <Box2D/Common/b2IntrusiveList.h>
+#include <Box2D/Common/b2Settings.h>
+
+/// Alignment (in bytes) of user memory associated with b2TrackedBlock.
+const int32 b2_mallocAlignment = 32;
+
+/// Allocated block of memory that can be tracked in a b2IntrusiveList.
+class b2TrackedBlock : public b2TypedIntrusiveListNode<b2TrackedBlock>
+{
+private:
+	// Initialize this block with a reference to "this".
+	b2TrackedBlock();
+	// Remove the block from the list.
+	~b2TrackedBlock() { }
+
+public:
+	/// Get the allocated memory associated with this block.
+	void* GetMemory() const;
+
+private:
+	// Padding required to align the pointer to user memory in the block
+	// to b2_mallocAlignment.
+	uint8 m_padding[b2_mallocAlignment + sizeof(b2TrackedBlock**)];
+
+public:
+	/// Allocate a b2TrackedBlock returning a pointer to memory of size
+	/// bytes that can be used by the caller.
+	static void* Allocate(uint32 size);
+
+	/// Get a b2TrackedBlock from a pointer to memory returned by
+	/// b2TrackedBlock::Allocate().
+	static b2TrackedBlock* GetFromMemory(void *memory);
+
+	/// Free a block of memory returned by b2TrackedBlock::Allocate()
+	static void Free(void *memory);
+
+	/// Free a b2TrackedBlock.
+	static void Free(b2TrackedBlock *block);
+};
+
+/// Allocator of blocks which are tracked in a list.
+class b2TrackedBlockAllocator
+{
+public:
+	/// Initialize.
+	b2TrackedBlockAllocator() {}
+	/// Free all allocated blocks.
+	~b2TrackedBlockAllocator()
+	{
+		FreeAll();
+	}
+
+	/// Allocate a block of size bytes using b2TrackedBlock::Allocate().
+	void* Allocate(uint32 size);
+
+	/// Free a block returned by Allocate().
+	void Free(void *memory);
+
+	/// Free all allocated blocks.
+	void FreeAll();
+
+	// Get the list of allocated blocks.
+	const b2TypedIntrusiveListNode<b2TrackedBlock>& GetList() const
+	{
+		return m_blocks;
+	}
+
+private:
+	b2TypedIntrusiveListNode<b2TrackedBlock> m_blocks;
+};
+
+#endif  // B2_TRACKED_BLOCK_H

deps/include/Box2D/Documentation/Building/Building.md

@@ -0,0 +1,19 @@
+Building LiquidFun and Running Examples {#mainpage}
+=======================================
+
+LiquidFun is an extension of [Box2D](http://box2d.org), a 2D physics engine
+for games.
+
+API documentation is located in the Documentation/ folder and referenced
+by our [landing page](../../index.html).
+
+If your project already uses Box2D, please see these instructions for
+[Porting from Box2D](md__porting_from_box2_d.html).
+
+Instructions on how to build for:
+- [Android](md__building_android.html)
+- [iOS](md__buildingi_o_s.html)
+- [Linux](md__building_linux.html)
+- [Windows](md__building_windows.html)
+- [OS X](md__building_o_s_x.html)
+- [JavaScript version](md__building_java_script.html)

deps/include/Box2D/Documentation/Building/BuildingAndroid.md

@@ -0,0 +1,117 @@
+# Building for Android
+
+### Version Requirements
+
+Following are the minimum required versions for the tools and libraries you
+need for building LiquidFun for Android:
+
+-   Android SDK:  Android 2.3.3 (API Level 10)
+-   ADT: 20130917
+-   NDK: android-ndk-r9
+-   NDK plugn for Eclipse: bundled with ADT
+-   cmake (when building from source): 2.8.12.1
+
+### Before Building
+
+-   Install the [Android SDK].
+-   Install the [Android NDK].
+
+### Building
+
+Each LiquidFun project for Android has an associated `AndroidManifest.xml`
+file and `jni` subdirectory.  For samples, `AndroidManifest.xml` contains
+details about how to build an Android package (apk).  For libraries, such as
+LiquidFun and freeglut, the `AndroidManifest.xml` file informs the `ndk-build`
+tool that the directory contains NDK makefiles under the `jni` subdirectory.
+
+The following [Android NDK][] projects live in the liquidfun directory tree.
+
+-   `liquidfun/Box2D` - liquidfun dynamic and static libraries.
+-   `liquidfun/Box2D/HelloWorld` - Non-interactive Hello World sample
+     application.
+-   `liquidfun/Box2D/Testbed` - Interactive Testbed sample application.
+
+To build an [Android NDK][] project:
+
+-   Open a command line window.
+-   Go to the working directory containing the project to build.
+-   Execute ndk-build.
+
+For example, to build the LiquidFun dynamic and static libraries:
+
+    cd liquidfun/Box2D
+    ndk-build
+
+### Executing a Sample
+
+Running a sample requires the Android Development Tools ([ADT][]) plugin and
+the [NDK Eclipse plugin][].
+
+#### Running an application using Eclipse:
+
+-   Build a project using `ndk-build` (see above).
+-   Open [ADT][] Eclipse.
+-   Select "File->Import..." from the menu.
+-   Select "Android > Existing Android Code Into Workspace", and click "Next".
+-   Click the "Browse..." button next to `Root Directory:` and select the
+    project folder (e.g. `liquidfun/Box2D/Testbed`).
+-   Click "Finish". Eclipse imports the project, and displays it in the
+    Package Explorer pane.
+-   Right-click the project, and select "Run->Run As->Android Application"
+    from the menu.
+-   If you do not have a physical device, you must define a virtual one.
+    For details about how to define a virtual device, see [managing avds][].
+    We don’t recommend a virtual device for development.
+
+-   If the target is a physical device, unlock the device and observe the application executing.
+
+#### Build and Run from the Command Line:
+
+It's also possible to use `build_apk.sh` to build, deploy, and execute an application.
+`build_apk.sh` requires the installation of the [Android NDK][] and [Android SDK][].
+On Windows the script requires that you install [Cygwin][], as it
+makes use of Bash and common UNIX utilities.  In addition, `build_apk.sh`
+requires the addition of the [Android SDK][] `sdk/tools`, [Android SDK][]
+`sdk/platform-tools` and [Android NDK][] root directories to the `PATH`
+environment variable.
+
+To build and run an application using `build_apk.sh`:
+
+-   Change into the directory containing the project to build.
+-   Run build\_apk.sh.
+
+For example, the following will build the Testbed application, deploy it to
+the connected device, and start it:
+
+    cd liquidfun/Box2D/Testbed
+    ../AndroidUtil/build_apk.sh
+
+### Code Generation
+
+By default, code is generated for devices that support the `armeabi-v7a` ABI.
+Alternatively, you can generate a fat `.apk` that includes code for all ABIs.
+To do so, override APP\_ABI on ndk-build's command line:
+
+    ndk-build APP_ABI=all
+
+### Running Unit Tests
+
+Developers modifying the LiquidFun library can verify that it is still working
+correctly by executing the provided unit tests.
+
+Use the run\_tests\_android.sh script to execute unit tests:
+
+    ./liquidfun/Box2D/Unittests/run_tests_android.sh
+
+`run_tests_android.sh` uses `build_apk.sh` to build, deploy, and
+execute each unit test on the connected device.
+The "Build and Run from the Command Line" section above describes
+the setup requirements for `build_apk.sh`.
+
+  [Android SDK]: http://developer.android.com/sdk/index.html
+  [Android NDK]: http://developer.android.com/tools/sdk/ndk/index.html
+  [NDK Eclipse plugin]: http://developer.android.com/sdk/index.html
+  [managing avds]: http://developer.android.com/tools/devices/managing-avds.html
+  [Cygwin installation]: http://www.cygwin.com/
+  [ADT]: http://developer.android.com/tools/sdk/eclipse-adt.html
+

deps/include/Box2D/Documentation/Building/BuildingJavaScript.md

@@ -0,0 +1,92 @@
+# Building JavaScript version
+
+### Background
+
+LiquidFun can be translated from C++ into JavaScript by
+[Emscripten](http://github.com/kripken/emscripten/wiki).
+The entire JavaScript translation is in one file: liquidfun.js.
+
+The LiquidFun source includes a pre-made version of liquidfun.js.
+You can reference that liquidfun.js to start writing LiquidFun programs
+in JavaScript right away.
+
+However, if you make changes to the LiquidFun C++ code, or if you need
+to expose more of the LiquidFun API, you will need to remake liquidfun.js
+by following the steps on this page.
+
+### Before Building
+
+We use Emscripten on Linux, but you should be able to use the Emscripten SDK
+on Mac or Windows too, if you prefer. Note that Mac and Windows build
+environments have not been tested.
+
+-   Get Emscripten. See [Installing from Source](http://github.com/kripken/emscripten/wiki/Emscripten-SDK#installing-from-source). 
+    -   You also need [Fastcomp](http://github.com/kripken/emscripten/wiki/LLVM-Backend#getting-fastcomp)
+    -   And [Node.js](http://nodejs.org/download/)
+-   Setup Emscripten.
+    -   As suggested in the Emscripten docs, run "emcc -v"
+    -   Update your ~/.emscripten file. You'll probably want something like this:
+        -   LLVM_ROOT = os.path.expanduser('~/Code/emscripten-fastcomp/build/Release/bin')
+        -   NODE_JS = os.path.expanduser('~/Code/node-v0.10.29-linux-x64/bin/node')
+-   Get the [Closure Compiler](http://developers.google.com/closure/compiler/).
+-   Set environment variables. Depending on your setup, you may want to add
+    some lines like this to your ~/.bashrc file.
+    - export EMSCRIPTEN=~/Code/emscripten
+    - export CLOSURE_JAR=~/Code/closure/compiler.jar
+
+### Building
+
+The following commands run Emscripten to translate C++ to lf_core.js,
+then combine lf_core.js with the bindings in jsBindings into liquidfun.js.
+
+    cd Box2D/lfjs
+    make
+    ./uglify.sh
+
+### Executing the Testbed
+
+For an example program, try openning `Box2D/lfjs/index.html' in your browser.
+This will load a JavaScript version of the LiquidFun Testbed. It uses
+the liquidfun.js file that you built above.
+
+### Debugging
+
+The default output of `make` and `uglify.sh` is whitespace-optimized
+JavaScript that is difficult to debug. To create an un-optimized version
+of the testbed, please follow the steps below.
+
++ In `lfjs/Makefile` change -O2 to -O0, to disable Emscripten optimizations.
++ In `lfjs/index.html` replace `<script src="liquidfun.js"></script>` with
+  a similar line for every .js file in `lfjs/uglify.sh`. It should look
+  something like,
+
+        <script src="lf_core.js"></script>
+        <script src="jsBindings/offsets.js"></script>
+        <script src="jsBindings/Common/b2Math.js"></script>
+        <script src="jsBindings/Collision/b2Collision.js"></script>
+        <script src="jsBindings/Collision/Shapes/b2EdgeShape.js"></script>
+        <script src="jsBindings/Collision/Shapes/b2PolygonShape.js"></script>
+        <script src="jsBindings/Collision/Shapes/b2Shape.js"></script>
+        <script src="jsBindings/Collision/Shapes/b2ChainShape.js"></script>
+        <script src="jsBindings/Collision/Shapes/b2CircleShape.js"></script>
+        <script src="jsBindings/Dynamics/b2Body.js"></script>
+        <script src="jsBindings/Dynamics/b2World.js"></script>
+        <script src="jsBindings/Dynamics/Joints/b2WheelJoint.js"></script>
+        <script src="jsBindings/Dynamics/Joints/b2WeldJoint.js"></script>
+        <script src="jsBindings/Dynamics/Joints/b2GearJoint.js"></script>
+        <script src="jsBindings/Dynamics/Joints/b2Joint.js"></script>
+        <script src="jsBindings/Dynamics/Joints/b2FrictionJoint.js"></script>
+        <script src="jsBindings/Dynamics/Joints/b2RevoluteJoint.js"></script>
+        <script src="jsBindings/Dynamics/Joints/b2MotorJoint.js"></script>
+        <script src="jsBindings/Dynamics/Joints/b2PulleyJoint.js"></script>
+        <script src="jsBindings/Dynamics/Joints/b2DistanceJoint.js"></script>
+        <script src="jsBindings/Dynamics/Joints/b2PrismaticJoint.js"></script>
+        <script src="jsBindings/Dynamics/Joints/b2RopeJoint.js"></script>
+        <script src="jsBindings/Dynamics/Joints/b2MouseJoint.js"></script>
+        <script src="jsBindings/Dynamics/Contacts/b2Contact.js"></script>
+        <script src="jsBindings/Dynamics/b2Fixture.js"></script>
+        <script src="jsBindings/Dynamics/b2WorldCallbacks.js"></script>
+        <script src="jsBindings/Particle/b2ParticleSystem.js"></script>
+        <script src="jsBindings/Particle/b2ParticleGroup.js"></script>
+        <script src="jsBindings/Particle/b2Particle.js"></script>
+

deps/include/Box2D/Documentation/Building/BuildingLinux.md

@@ -0,0 +1,79 @@
+# Building for Linux
+
+### Version Requirements
+
+Following are the minimum required versions for the tools and libraries you
+need for building LiquidFun for Linux:
+
+-   OpenGL: libglapi-mesa 8.0.4 (tested with libglapi-mesa 8.0.4-0ubuntu0)
+-   GLU: libglu1-mesa-dev 8.0.4 (tested with libglu1-mesa-dev 8.0.4.0ubuntu0)
+-   cmake (when building from source): 2.8.12.1
+
+### Before Building
+
+Prior to building, install the following components using the [Linux][]
+distribution's package manager:
+-    [cmake][]. You can also manually install from [cmake.org]
+     (http://cmake.org).
+-    OpenGL (`libglapi-mesa`).
+-    GLU (`libglu1-mesa-dev`).
+
+For example, on Ubuntu:
+
+    sudo apt-get install cmake
+    sudo apt-get install libglapi-mesa
+    sudo apt-get install libglu1-mesa-dev
+
+The sample applications require OpenGL and GLU.
+
+### Building
+
+-   Generate makefiles from the [cmake][] project in `liquidfun/Box2D`.
+-   Execute `make` to build the library and sample applications.
+
+For example:
+
+    cd liquidfun/Box2D
+    cmake -G'Unix Makefiles'
+    make
+
+To perform a debug build:
+
+    cd liquidfun/Box2D
+    cmake -G'Unix Makefiles' -DCMAKE_BUILD_TYPE=Debug
+    make
+
+Build targets can be configured using options exposed in
+`liquidfun/Box2D/CMakeLists.txt` by using cmake's `-D` option.
+Build configuration set using the `-D` option is sticky across subsequent
+builds.
+
+For example, if a build is performed using:
+
+    cmake -G"Unix Makefiles" -DCMAKE_BUILD_TYPE=Debug
+    make
+
+to switch to a release build CMAKE_BUILD_TYPE must be explicitly specified:
+
+    cmake -G"Unix Makefiles" -DCMAKE_BUILD_TYPE=Release
+    make
+
+### Executing a Sample
+
+After building the project, you can execute the samples from the command line.
+For example:
+
+    ./liquidfun/Box2D/Testbed/Release/Testbed
+
+### Running Unit Tests
+
+Developers modifying the LiquidFun library can verify that it still working
+correctly by executing the provided unit tests.
+
+Use the run\_tests.sh script to execute unit tests:
+
+    ./liquidfun/Box2D/Unittests/run_tests.sh
+
+  [cmake]: http://www.cmake.org
+  [Linux]: http://en.wikipedia.org/wiki/Linux
+  [Ubuntu]: http://www.ubuntu.com

deps/include/Box2D/Documentation/Building/BuildingOSX.md

@@ -0,0 +1,62 @@
+# Building for OS X
+
+You can use [cmake][] to generate an [Xcode][] project for LiquidFun's
+Testbed and Unittests on [OS X][].
+
+Alternatively, you can download a pre-generated Xcode
+project from the [release page][]. The [Xcode][] project is free of
+host-specific dependencies.
+
+### Version Requirements
+
+These are the minimum required versions for building LiquidFun on OS X:
+
+-   OS X: Mavericks 10.9.1.
+-   Xcode: 5.0.1
+-   Xquartz: 2.7.5 (xorg-server 1.14.4)
+-   cmake (when building from source) 2.8.12.1
+
+### Before Building
+
+-   Install [Xquartz][] to run LiquidFun sample applications.
+    The Testbed uses [freeglut][], which requires [Xquartz][].
+-   Reboot your machine.  Rebooting sets the `DISPLAY` environment variable for
+    [Xquartz][], which enables sample applications to run correctly.
+
+### Creating the Xcode project using [cmake][]
+
+When working directly with the source, you can generate the [Xcode][]
+project using [cmake][].  [cmake][] version 2.8.12.1 or above is required to
+build this project on [OS X][] 10.9 (Mavericks).  For example, the following
+generates the Xcode project in the liquidfun/Box2D directory.
+
+    cd liquidfun/Box2D
+    cmake -G "Xcode"
+
+### Building with [Xcode][]
+
+-   Double-click on `liquidfun/Box2D/Box2D.xcodeproj` to open the project in
+    [Xcode][].
+-   Select "Product-->Build" from the menu.
+
+### Executing a Sample
+
+-   Select a sample `Scheme`, for example "Testbed-->My Mac 64-bit", from the
+    combo box to the right of the "Run" button.
+-   Click the "Run" button.
+
+### Running Unit Tests
+
+Developers modifying the LiquidFun library can verify that it is still working
+correctly by executing the provided unit tests.
+
+Use the run\_tests.sh script to execute unit tests:
+
+    ./liquidfun/Box2D/Unittests/run_tests.sh
+
+  [Xquartz]: http://xquartz.macosforge.org/
+  [cmake]: http://www.cmake.org
+  [Xcode]: http://developer.apple.com/xcode/
+  [OS X]: http://www.apple.com/osx/
+  [freeglut]: http://freeglut.sourceforge.net/
+  [release page]: http://github.com/google/liquidfun/releases

deps/include/Box2D/Documentation/Building/BuildingWindows.md

@@ -0,0 +1,64 @@
+# Building for Windows
+
+You can use [cmake][] to generate a [Visual Studio][] project for
+LiquidFun's Testbed and Unittests on [Windows][].
+
+Alternatively, you can download a pre-generated Visual Studio solution
+from the [release page][]. The Visual Studio solution is free of
+host-specific dependencies.
+
+### Version Requirements
+
+These are the minimum required versions for building LiquidFun for Windows:
+
+-   Windows: 7
+-   Visual Studio: 2010 or 2012
+-   cmake: 2.8.12.1
+
+### Creating the Visual Studio solution using [cmake][]
+
+When working directly with the source, use [cmake][] to generate the
+[Visual Studio][] solution and project files.  For example, the following
+generates the [Visual Studio][] 2012 solution in the `liquidfun/Box2D`
+directory:
+
+    cd liquidfun\Box2D
+    cmake -G "Visual Studio 11"
+
+To generate a [Visual Studio][] 2010 solution, use this commend:
+
+    cd liquidfun\Box2D
+    cmake -G "Visual Studio 10"
+
+Running [cmake][] under [cygwin][] requires empty TMP, TEMP, tmp and temp
+variables.  To generate a [Visual Studio][] solution from a [cygwin][]
+bash shell use:
+
+    $ cd liquidfun/Box2D
+    $ ( unset {temp,tmp,TEMP,TMP} ; cmake -G "Visual Studio 11" )
+
+### Building with [Visual Studio][]
+
+-   Double-click on `liquidfun/Box2D/Box2D.sln` to open the solution.
+-   Select "Build-->Build Solution" from the menu.
+
+### Executing a Sample
+
+-   Right-click on an example project (e.g Testbed) in the Solution Explorer
+    pane, and select "Set as StartUp Project".
+-   Select "Debug-->Start Debugging" from the menu.
+
+### Running Unit Tests
+
+Developers modifying the LiquidFun library can verify that it is still working
+correctly by executing the provided unit tests.
+
+Use the run\_tests.bat batch file to execute unit tests:
+
+    liquidfun\Box2D\Unittests\run_tests.bat
+
+  [cmake]: http://www.cmake.org
+  [Visual Studio]: http://www.visualstudio.com/
+  [Windows]: http://windows.microsoft.com/
+  [cygwin]: http://www.cygwin.com/
+  [release page]: http://github.com/google/liquidfun/releases