Move other lone thirdparty files to thirdparty/misc

Also move Box2D ConvexDecomposition contrib code to
thirdparty/b2d_convexdecomp.

8 files changed, 21 insertions, 5308 deletions

diff --git a/scene/SCsub b/scene/SCsub
index bd2da1eab9..df52e9c6cb 100644
--- a/scene/SCsub
+++ b/scene/SCsub
@@ -3,10 +3,23 @@
 Import('env')
 
 env.scene_sources = []
+
+
+# Thirdparty code
+thirdparty_dir = "#thirdparty/misc/"
+thirdparty_sources = [
+	# C sources
+	"mikktspace.c",
+]
+thirdparty_sources = [thirdparty_dir + file for file in thirdparty_sources]
+env.add_source_files(env.scene_sources, thirdparty_sources)
+
+
+# Godot's own sources
 env.add_source_files(env.scene_sources, "*.cpp")
 
-Export('env')
 
+# Chain load SCsubs
 SConscript('main/SCsub')
 SConscript('gui/SCsub')
 SConscript('3d/SCsub')
@@ -17,6 +30,8 @@ SConscript('resources/SCsub')
 SConscript('io/SCsub')
 
 
+# Build it all as a library
 lib = env.Library("scene", env.scene_sources)
-
 env.Prepend(LIBS=[lib])
+
+Export('env')
diff --git a/scene/resources/SCsub b/scene/resources/SCsub
index 60b16cd0d4..2ad90247a7 100644
--- a/scene/resources/SCsub
+++ b/scene/resources/SCsub
@@ -3,7 +3,6 @@
 Import('env')
 
 env.add_source_files(env.scene_sources, "*.cpp")
-env.add_source_files(env.scene_sources, "*.c")
 
 Export('env')
 
diff --git a/scene/resources/dynamic_font_stb.cpp b/scene/resources/dynamic_font_stb.cpp
index 0a784c6496..397349bbf9 100644
--- a/scene/resources/dynamic_font_stb.cpp
+++ b/scene/resources/dynamic_font_stb.cpp
@@ -33,7 +33,6 @@
 
 #define STB_TRUETYPE_IMPLEMENTATION
 #include "os/file_access.h"
-#include "stb_truetype.h"
 
 void DynamicFontData::lock() {
 
diff --git a/scene/resources/dynamic_font_stb.h b/scene/resources/dynamic_font_stb.h
index 3862607ab5..a2c081a925 100644
--- a/scene/resources/dynamic_font_stb.h
+++ b/scene/resources/dynamic_font_stb.h
@@ -34,7 +34,8 @@
 
 #include "font.h"
 #include "io/resource_loader.h"
-#include "stb_truetype.h"
+
+#include "thirdparty/misc/stb_truetype.h"
 
 class DynamicFontAtSize;
 class DynamicFont;
diff --git a/scene/resources/mikktspace.c b/scene/resources/mikktspace.c
deleted file mode 100644
index 62aa2da251..0000000000
--- a/scene/resources/mikktspace.c
+++ /dev/null
@@ -1,1890 +0,0 @@
-/** \file mikktspace/mikktspace.c
- *  \ingroup mikktspace
- */
-/**
- *  Copyright (C) 2011 by Morten S. Mikkelsen
- *
- *  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 <assert.h>
-#include <stdio.h>
-#include <math.h>
-#include <string.h>
-#include <float.h>
-#include <stdlib.h>
-
-#include "mikktspace.h"
-
-#define TFALSE		0
-#define TTRUE		1
-
-#ifndef M_PI
-#define M_PI	3.1415926535897932384626433832795
-#endif
-
-#define INTERNAL_RND_SORT_SEED		39871946
-
-// internal structure
-typedef struct {
-	float x, y, z;
-} SVec3;
-
-static tbool			veq( const SVec3 v1, const SVec3 v2 )
-{
-	return (v1.x == v2.x) && (v1.y == v2.y) && (v1.z == v2.z);
-}
-
-static SVec3		vadd( const SVec3 v1, const SVec3 v2 )
-{
-	SVec3 vRes;
-
-	vRes.x = v1.x + v2.x;
-	vRes.y = v1.y + v2.y;
-	vRes.z = v1.z + v2.z;
-
-	return vRes;
-}
-
-
-static SVec3		vsub( const SVec3 v1, const SVec3 v2 )
-{
-	SVec3 vRes;
-
-	vRes.x = v1.x - v2.x;
-	vRes.y = v1.y - v2.y;
-	vRes.z = v1.z - v2.z;
-
-	return vRes;
-}
-
-static SVec3		vscale(const float fS, const SVec3 v)
-{
-	SVec3 vRes;
-
-	vRes.x = fS * v.x;
-	vRes.y = fS * v.y;
-	vRes.z = fS * v.z;
-
-	return vRes;
-}
-
-static float			LengthSquared( const SVec3 v )
-{
-	return v.x*v.x + v.y*v.y + v.z*v.z;
-}
-
-static float			Length( const SVec3 v )
-{
-	return sqrtf(LengthSquared(v));
-}
-
-static SVec3		Normalize( const SVec3 v )
-{
-	return vscale(1 / Length(v), v);
-}
-
-static float		vdot( const SVec3 v1, const SVec3 v2)
-{
-	return v1.x*v2.x + v1.y*v2.y + v1.z*v2.z;
-}
-
-
-static tbool NotZero(const float fX)
-{
-	// could possibly use FLT_EPSILON instead
-	return fabsf(fX) > FLT_MIN;
-}
-
-static tbool VNotZero(const SVec3 v)
-{
-	// might change this to an epsilon based test
-	return NotZero(v.x) || NotZero(v.y) || NotZero(v.z);
-}
-
-
-
-typedef struct {
-	int iNrFaces;
-	int * pTriMembers;
-} SSubGroup;
-
-typedef struct {
-	int iNrFaces;
-	int * pFaceIndices;
-	int iVertexRepresentitive;
-	tbool bOrientPreservering;
-} SGroup;
-
-// 
-#define MARK_DEGENERATE				1
-#define QUAD_ONE_DEGEN_TRI			2
-#define GROUP_WITH_ANY				4
-#define ORIENT_PRESERVING			8
-
-
-
-typedef struct {
-	int FaceNeighbors[3];
-	SGroup * AssignedGroup[3];
-	
-	// normalized first order face derivatives
-	SVec3 vOs, vOt;
-	float fMagS, fMagT;	// original magnitudes
-
-	// determines if the current and the next triangle are a quad.
-	int iOrgFaceNumber;
-	int iFlag, iTSpacesOffs;
-	unsigned char vert_num[4];
-} STriInfo;
-
-typedef struct {
-	SVec3 vOs;
-	float fMagS;
-	SVec3 vOt;
-	float fMagT;
-	int iCounter;	// this is to average back into quads.
-	tbool bOrient;
-} STSpace;
-
-static int GenerateInitialVerticesIndexList(STriInfo pTriInfos[], int piTriList_out[], const SMikkTSpaceContext * pContext, const int iNrTrianglesIn);
-static void GenerateSharedVerticesIndexList(int piTriList_in_and_out[], const SMikkTSpaceContext * pContext, const int iNrTrianglesIn);
-static void InitTriInfo(STriInfo pTriInfos[], const int piTriListIn[], const SMikkTSpaceContext * pContext, const int iNrTrianglesIn);
-static int Build4RuleGroups(STriInfo pTriInfos[], SGroup pGroups[], int piGroupTrianglesBuffer[], const int piTriListIn[], const int iNrTrianglesIn);
-static tbool GenerateTSpaces(STSpace psTspace[], const STriInfo pTriInfos[], const SGroup pGroups[],
-                             const int iNrActiveGroups, const int piTriListIn[], const float fThresCos,
-                             const SMikkTSpaceContext * pContext);
-
-static int MakeIndex(const int iFace, const int iVert)
-{
-	assert(iVert>=0 && iVert<4 && iFace>=0);
-	return (iFace<<2) | (iVert&0x3);
-}
-
-static void IndexToData(int * piFace, int * piVert, const int iIndexIn)
-{
-	piVert[0] = iIndexIn&0x3;
-	piFace[0] = iIndexIn>>2;
-}
-
-static STSpace AvgTSpace(const STSpace * pTS0, const STSpace * pTS1)
-{
-	STSpace ts_res;
-
-	// this if is important. Due to floating point precision
-	// averaging when ts0==ts1 will cause a slight difference
-	// which results in tangent space splits later on
-	if (pTS0->fMagS==pTS1->fMagS && pTS0->fMagT==pTS1->fMagT &&
-	   veq(pTS0->vOs,pTS1->vOs)	&& veq(pTS0->vOt, pTS1->vOt))
-	{
-		ts_res.fMagS = pTS0->fMagS;
-		ts_res.fMagT = pTS0->fMagT;
-		ts_res.vOs = pTS0->vOs;
-		ts_res.vOt = pTS0->vOt;
-	}
-	else
-	{
-		ts_res.fMagS = 0.5f*(pTS0->fMagS+pTS1->fMagS);
-		ts_res.fMagT = 0.5f*(pTS0->fMagT+pTS1->fMagT);
-		ts_res.vOs = vadd(pTS0->vOs,pTS1->vOs);
-		ts_res.vOt = vadd(pTS0->vOt,pTS1->vOt);
-		if ( VNotZero(ts_res.vOs) ) ts_res.vOs = Normalize(ts_res.vOs);
-		if ( VNotZero(ts_res.vOt) ) ts_res.vOt = Normalize(ts_res.vOt);
-	}
-
-	return ts_res;
-}
-
-
-
-static SVec3 GetPosition(const SMikkTSpaceContext * pContext, const int index);
-static SVec3 GetNormal(const SMikkTSpaceContext * pContext, const int index);
-static SVec3 GetTexCoord(const SMikkTSpaceContext * pContext, const int index);
-
-
-// degen triangles
-static void DegenPrologue(STriInfo pTriInfos[], int piTriList_out[], const int iNrTrianglesIn, const int iTotTris);
-static void DegenEpilogue(STSpace psTspace[], STriInfo pTriInfos[], int piTriListIn[], const SMikkTSpaceContext * pContext, const int iNrTrianglesIn, const int iTotTris);
-
-
-tbool genTangSpaceDefault(const SMikkTSpaceContext * pContext)
-{
-	return genTangSpace(pContext, 180.0f);
-}
-
-tbool genTangSpace(const SMikkTSpaceContext * pContext, const float fAngularThreshold)
-{
-	// count nr_triangles
-	int * piTriListIn = NULL, * piGroupTrianglesBuffer = NULL;
-	STriInfo * pTriInfos = NULL;
-	SGroup * pGroups = NULL;
-	STSpace * psTspace = NULL;
-	int iNrTrianglesIn = 0, f=0, t=0, i=0;
-	int iNrTSPaces = 0, iTotTris = 0, iDegenTriangles = 0, iNrMaxGroups = 0;
-	int iNrActiveGroups = 0, index = 0;
-	const int iNrFaces = pContext->m_pInterface->m_getNumFaces(pContext);
-	tbool bRes = TFALSE;
-	const float fThresCos = (float) cos((fAngularThreshold*(float)M_PI)/180.0f);
-
-	// verify all call-backs have been set
-	if ( pContext->m_pInterface->m_getNumFaces==NULL ||
-		pContext->m_pInterface->m_getNumVerticesOfFace==NULL ||
-		pContext->m_pInterface->m_getPosition==NULL ||
-		pContext->m_pInterface->m_getNormal==NULL ||
-		pContext->m_pInterface->m_getTexCoord==NULL )
-		return TFALSE;
-
-	// count triangles on supported faces
-	for (f=0; f<iNrFaces; f++)
-	{
-		const int verts = pContext->m_pInterface->m_getNumVerticesOfFace(pContext, f);
-		if (verts==3) ++iNrTrianglesIn;
-		else if (verts==4) iNrTrianglesIn += 2;
-	}
-	if (iNrTrianglesIn<=0) return TFALSE;
-
-	// allocate memory for an index list
-	piTriListIn = (int *) malloc(sizeof(int)*3*iNrTrianglesIn);
-	pTriInfos = (STriInfo *) malloc(sizeof(STriInfo)*iNrTrianglesIn);
-	if (piTriListIn==NULL || pTriInfos==NULL)
-	{
-		if (piTriListIn!=NULL) free(piTriListIn);
-		if (pTriInfos!=NULL) free(pTriInfos);
-		return TFALSE;
-	}
-
-	// make an initial triangle --> face index list
-	iNrTSPaces = GenerateInitialVerticesIndexList(pTriInfos, piTriListIn, pContext, iNrTrianglesIn);
-
-	// make a welded index list of identical positions and attributes (pos, norm, texc)
-	//printf("gen welded index list begin\n");
-	GenerateSharedVerticesIndexList(piTriListIn, pContext, iNrTrianglesIn);
-	//printf("gen welded index list end\n");
-
-	// Mark all degenerate triangles
-	iTotTris = iNrTrianglesIn;
-	iDegenTriangles = 0;
-	for (t=0; t<iTotTris; t++)
-	{
-		const int i0 = piTriListIn[t*3+0];
-		const int i1 = piTriListIn[t*3+1];
-		const int i2 = piTriListIn[t*3+2];
-		const SVec3 p0 = GetPosition(pContext, i0);
-		const SVec3 p1 = GetPosition(pContext, i1);
-		const SVec3 p2 = GetPosition(pContext, i2);
-		if (veq(p0,p1) || veq(p0,p2) || veq(p1,p2))	// degenerate
-		{
-			pTriInfos[t].iFlag |= MARK_DEGENERATE;
-			++iDegenTriangles;
-		}
-	}
-	iNrTrianglesIn = iTotTris - iDegenTriangles;
-
-	// mark all triangle pairs that belong to a quad with only one
-	// good triangle. These need special treatment in DegenEpilogue().
-	// Additionally, move all good triangles to the start of
-	// pTriInfos[] and piTriListIn[] without changing order and
-	// put the degenerate triangles last.
-	DegenPrologue(pTriInfos, piTriListIn, iNrTrianglesIn, iTotTris);
-
-	
-	// evaluate triangle level attributes and neighbor list
-	//printf("gen neighbors list begin\n");
-	InitTriInfo(pTriInfos, piTriListIn, pContext, iNrTrianglesIn);
-	//printf("gen neighbors list end\n");
-
-	
-	// based on the 4 rules, identify groups based on connectivity
-	iNrMaxGroups = iNrTrianglesIn*3;
-	pGroups = (SGroup *) malloc(sizeof(SGroup)*iNrMaxGroups);
-	piGroupTrianglesBuffer = (int *) malloc(sizeof(int)*iNrTrianglesIn*3);
-	if (pGroups==NULL || piGroupTrianglesBuffer==NULL)
-	{
-		if (pGroups!=NULL) free(pGroups);
-		if (piGroupTrianglesBuffer!=NULL) free(piGroupTrianglesBuffer);
-		free(piTriListIn);
-		free(pTriInfos);
-		return TFALSE;
-	}
-	//printf("gen 4rule groups begin\n");
-	iNrActiveGroups =
-		Build4RuleGroups(pTriInfos, pGroups, piGroupTrianglesBuffer, piTriListIn, iNrTrianglesIn);
-	//printf("gen 4rule groups end\n");
-
-	//
-
-	psTspace = (STSpace *) malloc(sizeof(STSpace)*iNrTSPaces);
-	if (psTspace==NULL)
-	{
-		free(piTriListIn);
-		free(pTriInfos);
-		free(pGroups);
-		free(piGroupTrianglesBuffer);
-		return TFALSE;
-	}
-	memset(psTspace, 0, sizeof(STSpace)*iNrTSPaces);
-	for (t=0; t<iNrTSPaces; t++)
-	{
-		psTspace[t].vOs.x=1.0f; psTspace[t].vOs.y=0.0f; psTspace[t].vOs.z=0.0f; psTspace[t].fMagS = 1.0f;
-		psTspace[t].vOt.x=0.0f; psTspace[t].vOt.y=1.0f; psTspace[t].vOt.z=0.0f; psTspace[t].fMagT = 1.0f;
-	}
-
-	// make tspaces, each group is split up into subgroups if necessary
-	// based on fAngularThreshold. Finally a tangent space is made for
-	// every resulting subgroup
-	//printf("gen tspaces begin\n");
-	bRes = GenerateTSpaces(psTspace, pTriInfos, pGroups, iNrActiveGroups, piTriListIn, fThresCos, pContext);
-	//printf("gen tspaces end\n");
-	
-	// clean up
-	free(pGroups);
-	free(piGroupTrianglesBuffer);
-
-	if (!bRes)	// if an allocation in GenerateTSpaces() failed
-	{
-		// clean up and return false
-		free(pTriInfos); free(piTriListIn); free(psTspace);
-		return TFALSE;
-	}
-
-
-	// degenerate quads with one good triangle will be fixed by copying a space from
-	// the good triangle to the coinciding vertex.
-	// all other degenerate triangles will just copy a space from any good triangle
-	// with the same welded index in piTriListIn[].
-	DegenEpilogue(psTspace, pTriInfos, piTriListIn, pContext, iNrTrianglesIn, iTotTris);
-
-	free(pTriInfos); free(piTriListIn);
-
-	index = 0;
-	for (f=0; f<iNrFaces; f++)
-	{
-		const int verts = pContext->m_pInterface->m_getNumVerticesOfFace(pContext, f);
-		if (verts!=3 && verts!=4) continue;
-		
-
-		// I've decided to let degenerate triangles and group-with-anythings
-		// vary between left/right hand coordinate systems at the vertices.
-		// All healthy triangles on the other hand are built to always be either or.
-
-		/*// force the coordinate system orientation to be uniform for every face.
-		// (this is already the case for good triangles but not for
-		// degenerate ones and those with bGroupWithAnything==true)
-		bool bOrient = psTspace[index].bOrient;
-		if (psTspace[index].iCounter == 0)	// tspace was not derived from a group
-		{
-			// look for a space created in GenerateTSpaces() by iCounter>0
-			bool bNotFound = true;
-			int i=1;
-			while (i<verts && bNotFound)
-			{
-				if (psTspace[index+i].iCounter > 0) bNotFound=false;
-				else ++i;
-			}
-			if (!bNotFound) bOrient = psTspace[index+i].bOrient;
-		}*/
-
-		// set data
-		for (i=0; i<verts; i++)
-		{
-			const STSpace * pTSpace = &psTspace[index];
-			float tang[] = {pTSpace->vOs.x, pTSpace->vOs.y, pTSpace->vOs.z};
-			float bitang[] = {pTSpace->vOt.x, pTSpace->vOt.y, pTSpace->vOt.z};
-			if (pContext->m_pInterface->m_setTSpace!=NULL)
-				pContext->m_pInterface->m_setTSpace(pContext, tang, bitang, pTSpace->fMagS, pTSpace->fMagT, pTSpace->bOrient, f, i);
-			if (pContext->m_pInterface->m_setTSpaceBasic!=NULL)
-				pContext->m_pInterface->m_setTSpaceBasic(pContext, tang, pTSpace->bOrient==TTRUE ? 1.0f : (-1.0f), f, i);
-
-			++index;
-		}
-	}
-
-	free(psTspace);
-
-	
-	return TTRUE;
-}
-
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-
-typedef struct {
-	float vert[3];
-	int index;
-} STmpVert;
-
-static const int g_iCells = 2048;
-
-#ifdef _MSC_VER
-	#define NOINLINE __declspec(noinline)
-#else
-	#define NOINLINE __attribute__ ((noinline))
-#endif
-
-// it is IMPORTANT that this function is called to evaluate the hash since
-// inlining could potentially reorder instructions and generate different
-// results for the same effective input value fVal.
-static NOINLINE int FindGridCell(const float fMin, const float fMax, const float fVal)
-{
-	const float fIndex = g_iCells * ((fVal-fMin)/(fMax-fMin));
-	const int iIndex = (int)fIndex;
-	return iIndex < g_iCells ? (iIndex >= 0 ? iIndex : 0) : (g_iCells - 1);
-}
-
-static void MergeVertsFast(int piTriList_in_and_out[], STmpVert pTmpVert[], const SMikkTSpaceContext * pContext, const int iL_in, const int iR_in);
-static void MergeVertsSlow(int piTriList_in_and_out[], const SMikkTSpaceContext * pContext, const int pTable[], const int iEntries);
-static void GenerateSharedVerticesIndexListSlow(int piTriList_in_and_out[], const SMikkTSpaceContext * pContext, const int iNrTrianglesIn);
-
-static void GenerateSharedVerticesIndexList(int piTriList_in_and_out[], const SMikkTSpaceContext * pContext, const int iNrTrianglesIn)
-{
-
-	// Generate bounding box
-	int * piHashTable=NULL, * piHashCount=NULL, * piHashOffsets=NULL, * piHashCount2=NULL;
-	STmpVert * pTmpVert = NULL;
-	int i=0, iChannel=0, k=0, e=0;
-	int iMaxCount=0;
-	SVec3 vMin = GetPosition(pContext, 0), vMax = vMin, vDim;
-	float fMin, fMax;
-	for (i=1; i<(iNrTrianglesIn*3); i++)
-	{
-		const int index = piTriList_in_and_out[i];
-
-		const SVec3 vP = GetPosition(pContext, index);
-		if (vMin.x > vP.x) vMin.x = vP.x;
-		else if (vMax.x < vP.x) vMax.x = vP.x;
-		if (vMin.y > vP.y) vMin.y = vP.y;
-		else if (vMax.y < vP.y) vMax.y = vP.y;
-		if (vMin.z > vP.z) vMin.z = vP.z;
-		else if (vMax.z < vP.z) vMax.z = vP.z;
-	}
-
-	vDim = vsub(vMax,vMin);
-	iChannel = 0;
-	fMin = vMin.x; fMax=vMax.x;
-	if (vDim.y>vDim.x && vDim.y>vDim.z)
-	{
-		iChannel=1;
-		fMin = vMin.y, fMax=vMax.y;
-	}
-	else if (vDim.z>vDim.x)
-	{
-		iChannel=2;
-		fMin = vMin.z, fMax=vMax.z;
-	}
-
-	// make allocations
-	piHashTable = (int *) malloc(sizeof(int)*iNrTrianglesIn*3);
-	piHashCount = (int *) malloc(sizeof(int)*g_iCells);
-	piHashOffsets = (int *) malloc(sizeof(int)*g_iCells);
-	piHashCount2 = (int *) malloc(sizeof(int)*g_iCells);
-
-	if (piHashTable==NULL || piHashCount==NULL || piHashOffsets==NULL || piHashCount2==NULL)
-	{
-		if (piHashTable!=NULL) free(piHashTable);
-		if (piHashCount!=NULL) free(piHashCount);
-		if (piHashOffsets!=NULL) free(piHashOffsets);
-		if (piHashCount2!=NULL) free(piHashCount2);
-		GenerateSharedVerticesIndexListSlow(piTriList_in_and_out, pContext, iNrTrianglesIn);
-		return;
-	}
-	memset(piHashCount, 0, sizeof(int)*g_iCells);
-	memset(piHashCount2, 0, sizeof(int)*g_iCells);
-
-	// count amount of elements in each cell unit
-	for (i=0; i<(iNrTrianglesIn*3); i++)
-	{
-		const int index = piTriList_in_and_out[i];
-		const SVec3 vP = GetPosition(pContext, index);
-		const float fVal = iChannel==0 ? vP.x : (iChannel==1 ? vP.y : vP.z);
-		const int iCell = FindGridCell(fMin, fMax, fVal);
-		++piHashCount[iCell];
-	}
-
-	// evaluate start index of each cell.
-	piHashOffsets[0]=0;
-	for (k=1; k<g_iCells; k++)
-		piHashOffsets[k]=piHashOffsets[k-1]+piHashCount[k-1];
-
-	// insert vertices
-	for (i=0; i<(iNrTrianglesIn*3); i++)
-	{
-		const int index = piTriList_in_and_out[i];
-		const SVec3 vP = GetPosition(pContext, index);
-		const float fVal = iChannel==0 ? vP.x : (iChannel==1 ? vP.y : vP.z);
-		const int iCell = FindGridCell(fMin, fMax, fVal);
-		int * pTable = NULL;
-
-		assert(piHashCount2[iCell]<piHashCount[iCell]);
-		pTable = &piHashTable[piHashOffsets[iCell]];
-		pTable[piHashCount2[iCell]] = i;	// vertex i has been inserted.
-		++piHashCount2[iCell];
-	}
-	for (k=0; k<g_iCells; k++)
-		assert(piHashCount2[k] == piHashCount[k]);	// verify the count
-	free(piHashCount2);
-
-	// find maximum amount of entries in any hash entry
-	iMaxCount = piHashCount[0];
-	for (k=1; k<g_iCells; k++)
-		if (iMaxCount<piHashCount[k])
-			iMaxCount=piHashCount[k];
-	pTmpVert = (STmpVert *) malloc(sizeof(STmpVert)*iMaxCount);
-	
-
-	// complete the merge
-	for (k=0; k<g_iCells; k++)
-	{
-		// extract table of cell k and amount of entries in it
-		int * pTable = &piHashTable[piHashOffsets[k]];
-		const int iEntries = piHashCount[k];
-		if (iEntries < 2) continue;
-
-		if (pTmpVert!=NULL)
-		{
-			for (e=0; e<iEntries; e++)
-			{
-				int i = pTable[e];
-				const SVec3 vP = GetPosition(pContext, piTriList_in_and_out[i]);
-				pTmpVert[e].vert[0] = vP.x; pTmpVert[e].vert[1] = vP.y;
-				pTmpVert[e].vert[2] = vP.z; pTmpVert[e].index = i;
-			}
-			MergeVertsFast(piTriList_in_and_out, pTmpVert, pContext, 0, iEntries-1);
-		}
-		else
-			MergeVertsSlow(piTriList_in_and_out, pContext, pTable, iEntries);
-	}
-
-	if (pTmpVert!=NULL) { free(pTmpVert); }
-	free(piHashTable);
-	free(piHashCount);
-	free(piHashOffsets);
-}
-
-static void MergeVertsFast(int piTriList_in_and_out[], STmpVert pTmpVert[], const SMikkTSpaceContext * pContext, const int iL_in, const int iR_in)
-{
-	// make bbox
-	int c=0, l=0, channel=0;
-	float fvMin[3], fvMax[3];
-	float dx=0, dy=0, dz=0, fSep=0;
-	for (c=0; c<3; c++)
-	{	fvMin[c]=pTmpVert[iL_in].vert[c]; fvMax[c]=fvMin[c];	}
-	for (l=(iL_in+1); l<=iR_in; l++)
-		for (c=0; c<3; c++)
-			if (fvMin[c]>pTmpVert[l].vert[c]) fvMin[c]=pTmpVert[l].vert[c];
-			else if (fvMax[c]<pTmpVert[l].vert[c]) fvMax[c]=pTmpVert[l].vert[c];
-
-	dx = fvMax[0]-fvMin[0];
-	dy = fvMax[1]-fvMin[1];
-	dz = fvMax[2]-fvMin[2];
-
-	channel = 0;
-	if (dy>dx && dy>dz) channel=1;
-	else if (dz>dx) channel=2;
-
-	fSep = 0.5f*(fvMax[channel]+fvMin[channel]);
-
-	// terminate recursion when the separation/average value
-	// is no longer strictly between fMin and fMax values.
-	if (fSep>=fvMax[channel] || fSep<=fvMin[channel])
-	{
-		// complete the weld
-		for (l=iL_in; l<=iR_in; l++)
-		{
-			int i = pTmpVert[l].index;
-			const int index = piTriList_in_and_out[i];
-			const SVec3 vP = GetPosition(pContext, index);
-			const SVec3 vN = GetNormal(pContext, index);
-			const SVec3 vT = GetTexCoord(pContext, index);
-
-			tbool bNotFound = TTRUE;
-			int l2=iL_in, i2rec=-1;
-			while (l2<l && bNotFound)
-			{
-				const int i2 = pTmpVert[l2].index;
-				const int index2 = piTriList_in_and_out[i2];
-				const SVec3 vP2 = GetPosition(pContext, index2);
-				const SVec3 vN2 = GetNormal(pContext, index2);
-				const SVec3 vT2 = GetTexCoord(pContext, index2);
-				i2rec=i2;
-
-				//if (vP==vP2 && vN==vN2 && vT==vT2)
-				if (vP.x==vP2.x && vP.y==vP2.y && vP.z==vP2.z &&
-					vN.x==vN2.x && vN.y==vN2.y && vN.z==vN2.z &&
-					vT.x==vT2.x && vT.y==vT2.y && vT.z==vT2.z)
-					bNotFound = TFALSE;
-				else
-					++l2;
-			}
-			
-			// merge if previously found
-			if (!bNotFound)
-				piTriList_in_and_out[i] = piTriList_in_and_out[i2rec];
-		}
-	}
-	else
-	{
-		int iL=iL_in, iR=iR_in;
-		assert((iR_in-iL_in)>0);	// at least 2 entries
-
-		// separate (by fSep) all points between iL_in and iR_in in pTmpVert[]
-		while (iL < iR)
-		{
-			tbool bReadyLeftSwap = TFALSE, bReadyRightSwap = TFALSE;
-			while ((!bReadyLeftSwap) && iL<iR)
-			{
-				assert(iL>=iL_in && iL<=iR_in);
-				bReadyLeftSwap = !(pTmpVert[iL].vert[channel]<fSep);
-				if (!bReadyLeftSwap) ++iL;
-			}
-			while ((!bReadyRightSwap) && iL<iR)
-			{
-				assert(iR>=iL_in && iR<=iR_in);
-				bReadyRightSwap = pTmpVert[iR].vert[channel]<fSep;
-				if (!bReadyRightSwap) --iR;
-			}
-			assert( (iL<iR) || !(bReadyLeftSwap && bReadyRightSwap) );
-
-			if (bReadyLeftSwap && bReadyRightSwap)
-			{
-				const STmpVert sTmp = pTmpVert[iL];
-				assert(iL<iR);
-				pTmpVert[iL] = pTmpVert[iR];
-				pTmpVert[iR] = sTmp;
-				++iL; --iR;
-			}
-		}
-
-		assert(iL==(iR+1) || (iL==iR));
-		if (iL==iR)
-		{
-			const tbool bReadyRightSwap = pTmpVert[iR].vert[channel]<fSep;
-			if (bReadyRightSwap) ++iL;
-			else --iR;
-		}
-
-		// only need to weld when there is more than 1 instance of the (x,y,z)
-		if (iL_in < iR)
-			MergeVertsFast(piTriList_in_and_out, pTmpVert, pContext, iL_in, iR);	// weld all left of fSep
-		if (iL < iR_in)
-			MergeVertsFast(piTriList_in_and_out, pTmpVert, pContext, iL, iR_in);	// weld all right of (or equal to) fSep
-	}
-}
-
-static void MergeVertsSlow(int piTriList_in_and_out[], const SMikkTSpaceContext * pContext, const int pTable[], const int iEntries)
-{
-	// this can be optimized further using a tree structure or more hashing.
-	int e=0;
-	for (e=0; e<iEntries; e++)
-	{
-		int i = pTable[e];
-		const int index = piTriList_in_and_out[i];
-		const SVec3 vP = GetPosition(pContext, index);
-		const SVec3 vN = GetNormal(pContext, index);
-		const SVec3 vT = GetTexCoord(pContext, index);
-
-		tbool bNotFound = TTRUE;
-		int e2=0, i2rec=-1;
-		while (e2<e && bNotFound)
-		{
-			const int i2 = pTable[e2];
-			const int index2 = piTriList_in_and_out[i2];
-			const SVec3 vP2 = GetPosition(pContext, index2);
-			const SVec3 vN2 = GetNormal(pContext, index2);
-			const SVec3 vT2 = GetTexCoord(pContext, index2);
-			i2rec = i2;
-
-			if (veq(vP,vP2) && veq(vN,vN2) && veq(vT,vT2))
-				bNotFound = TFALSE;
-			else
-				++e2;
-		}
-		
-		// merge if previously found
-		if (!bNotFound)
-			piTriList_in_and_out[i] = piTriList_in_and_out[i2rec];
-	}
-}
-
-static void GenerateSharedVerticesIndexListSlow(int piTriList_in_and_out[], const SMikkTSpaceContext * pContext, const int iNrTrianglesIn)
-{
-	int iNumUniqueVerts = 0, t=0, i=0;
-	for (t=0; t<iNrTrianglesIn; t++)
-	{
-		for (i=0; i<3; i++)
-		{
-			const int offs = t*3 + i;
-			const int index = piTriList_in_and_out[offs];
-
-			const SVec3 vP = GetPosition(pContext, index);
-			const SVec3 vN = GetNormal(pContext, index);
-			const SVec3 vT = GetTexCoord(pContext, index);
-
-			tbool bFound = TFALSE;
-			int t2=0, index2rec=-1;
-			while (!bFound && t2<=t)
-			{
-				int j=0;
-				while (!bFound && j<3)
-				{
-					const int index2 = piTriList_in_and_out[t2*3 + j];
-					const SVec3 vP2 = GetPosition(pContext, index2);
-					const SVec3 vN2 = GetNormal(pContext, index2);
-					const SVec3 vT2 = GetTexCoord(pContext, index2);
-					
-					if (veq(vP,vP2) && veq(vN,vN2) && veq(vT,vT2))
-						bFound = TTRUE;
-					else
-						++j;
-				}
-				if (!bFound) ++t2;
-			}
-
-			assert(bFound);
-			// if we found our own
-			if (index2rec == index) { ++iNumUniqueVerts; }
-
-			piTriList_in_and_out[offs] = index2rec;
-		}
-	}
-}
-
-static int GenerateInitialVerticesIndexList(STriInfo pTriInfos[], int piTriList_out[], const SMikkTSpaceContext * pContext, const int iNrTrianglesIn)
-{
-	int iTSpacesOffs = 0, f=0, t=0;
-	int iDstTriIndex = 0;
-	for (f=0; f<pContext->m_pInterface->m_getNumFaces(pContext); f++)
-	{
-		const int verts = pContext->m_pInterface->m_getNumVerticesOfFace(pContext, f);
-		if (verts!=3 && verts!=4) continue;
-
-		pTriInfos[iDstTriIndex].iOrgFaceNumber = f;
-		pTriInfos[iDstTriIndex].iTSpacesOffs = iTSpacesOffs;
-
-		if (verts==3)
-		{
-			unsigned char * pVerts = pTriInfos[iDstTriIndex].vert_num;
-			pVerts[0]=0; pVerts[1]=1; pVerts[2]=2;
-			piTriList_out[iDstTriIndex*3+0] = MakeIndex(f, 0);
-			piTriList_out[iDstTriIndex*3+1] = MakeIndex(f, 1);
-			piTriList_out[iDstTriIndex*3+2] = MakeIndex(f, 2);
-			++iDstTriIndex;	// next
-		}
-		else
-		{
-			{
-				pTriInfos[iDstTriIndex+1].iOrgFaceNumber = f;
-				pTriInfos[iDstTriIndex+1].iTSpacesOffs = iTSpacesOffs;
-			}
-
-			{
-				// need an order independent way to evaluate
-				// tspace on quads. This is done by splitting
-				// along the shortest diagonal.
-				const int i0 = MakeIndex(f, 0);
-				const int i1 = MakeIndex(f, 1);
-				const int i2 = MakeIndex(f, 2);
-				const int i3 = MakeIndex(f, 3);
-				const SVec3 T0 = GetTexCoord(pContext, i0);
-				const SVec3 T1 = GetTexCoord(pContext, i1);
-				const SVec3 T2 = GetTexCoord(pContext, i2);
-				const SVec3 T3 = GetTexCoord(pContext, i3);
-				const float distSQ_02 = LengthSquared(vsub(T2,T0));
-				const float distSQ_13 = LengthSquared(vsub(T3,T1));
-				tbool bQuadDiagIs_02;
-				if (distSQ_02<distSQ_13)
-					bQuadDiagIs_02 = TTRUE;
-				else if (distSQ_13<distSQ_02)
-					bQuadDiagIs_02 = TFALSE;
-				else
-				{
-					const SVec3 P0 = GetPosition(pContext, i0);
-					const SVec3 P1 = GetPosition(pContext, i1);
-					const SVec3 P2 = GetPosition(pContext, i2);
-					const SVec3 P3 = GetPosition(pContext, i3);
-					const float distSQ_02 = LengthSquared(vsub(P2,P0));
-					const float distSQ_13 = LengthSquared(vsub(P3,P1));
-
-					bQuadDiagIs_02 = distSQ_13<distSQ_02 ? TFALSE : TTRUE;
-				}
-
-				if (bQuadDiagIs_02)
-				{
-					{
-						unsigned char * pVerts_A = pTriInfos[iDstTriIndex].vert_num;
-						pVerts_A[0]=0; pVerts_A[1]=1; pVerts_A[2]=2;
-					}
-					piTriList_out[iDstTriIndex*3+0] = i0;
-					piTriList_out[iDstTriIndex*3+1] = i1;
-					piTriList_out[iDstTriIndex*3+2] = i2;
-					++iDstTriIndex;	// next
-					{
-						unsigned char * pVerts_B = pTriInfos[iDstTriIndex].vert_num;
-						pVerts_B[0]=0; pVerts_B[1]=2; pVerts_B[2]=3;
-					}
-					piTriList_out[iDstTriIndex*3+0] = i0;
-					piTriList_out[iDstTriIndex*3+1] = i2;
-					piTriList_out[iDstTriIndex*3+2] = i3;
-					++iDstTriIndex;	// next
-				}
-				else
-				{
-					{
-						unsigned char * pVerts_A = pTriInfos[iDstTriIndex].vert_num;
-						pVerts_A[0]=0; pVerts_A[1]=1; pVerts_A[2]=3;
-					}
-					piTriList_out[iDstTriIndex*3+0] = i0;
-					piTriList_out[iDstTriIndex*3+1] = i1;
-					piTriList_out[iDstTriIndex*3+2] = i3;
-					++iDstTriIndex;	// next
-					{
-						unsigned char * pVerts_B = pTriInfos[iDstTriIndex].vert_num;
-						pVerts_B[0]=1; pVerts_B[1]=2; pVerts_B[2]=3;
-					}
-					piTriList_out[iDstTriIndex*3+0] = i1;
-					piTriList_out[iDstTriIndex*3+1] = i2;
-					piTriList_out[iDstTriIndex*3+2] = i3;
-					++iDstTriIndex;	// next
-				}
-			}
-		}
-
-		iTSpacesOffs += verts;
-		assert(iDstTriIndex<=iNrTrianglesIn);
-	}
-
-	for (t=0; t<iNrTrianglesIn; t++)
-		pTriInfos[t].iFlag = 0;
-
-	// return total amount of tspaces
-	return iTSpacesOffs;
-}
-
-static SVec3 GetPosition(const SMikkTSpaceContext * pContext, const int index)
-{
-	int iF, iI;
-	SVec3 res; float pos[3];
-	IndexToData(&iF, &iI, index);
-	pContext->m_pInterface->m_getPosition(pContext, pos, iF, iI);
-	res.x=pos[0]; res.y=pos[1]; res.z=pos[2];
-	return res;
-}
-
-static SVec3 GetNormal(const SMikkTSpaceContext * pContext, const int index)
-{
-	int iF, iI;
-	SVec3 res; float norm[3];
-	IndexToData(&iF, &iI, index);
-	pContext->m_pInterface->m_getNormal(pContext, norm, iF, iI);
-	res.x=norm[0]; res.y=norm[1]; res.z=norm[2];
-	return res;
-}
-
-static SVec3 GetTexCoord(const SMikkTSpaceContext * pContext, const int index)
-{
-	int iF, iI;
-	SVec3 res; float texc[2];
-	IndexToData(&iF, &iI, index);
-	pContext->m_pInterface->m_getTexCoord(pContext, texc, iF, iI);
-	res.x=texc[0]; res.y=texc[1]; res.z=1.0f;
-	return res;
-}
-
-/////////////////////////////////////////////////////////////////////////////////////////////////////
-/////////////////////////////////////////////////////////////////////////////////////////////////////
-
-typedef union {
-	struct
-	{
-		int i0, i1, f;
-	};
-	int array[3];
-} SEdge;
-
-static void BuildNeighborsFast(STriInfo pTriInfos[], SEdge * pEdges, const int piTriListIn[], const int iNrTrianglesIn);
-static void BuildNeighborsSlow(STriInfo pTriInfos[], const int piTriListIn[], const int iNrTrianglesIn);
-
-// returns the texture area times 2
-static float CalcTexArea(const SMikkTSpaceContext * pContext, const int indices[])
-{
-	const SVec3 t1 = GetTexCoord(pContext, indices[0]);
-	const SVec3 t2 = GetTexCoord(pContext, indices[1]);
-	const SVec3 t3 = GetTexCoord(pContext, indices[2]);
-
-	const float t21x = t2.x-t1.x;
-	const float t21y = t2.y-t1.y;
-	const float t31x = t3.x-t1.x;
-	const float t31y = t3.y-t1.y;
-
-	const float fSignedAreaSTx2 = t21x*t31y - t21y*t31x;
-
-	return fSignedAreaSTx2<0 ? (-fSignedAreaSTx2) : fSignedAreaSTx2;
-}
-
-static void InitTriInfo(STriInfo pTriInfos[], const int piTriListIn[], const SMikkTSpaceContext * pContext, const int iNrTrianglesIn)
-{
-	int f=0, i=0, t=0;
-	// pTriInfos[f].iFlag is cleared in GenerateInitialVerticesIndexList() which is called before this function.
-
-	// generate neighbor info list
-	for (f=0; f<iNrTrianglesIn; f++)
-		for (i=0; i<3; i++)
-		{
-			pTriInfos[f].FaceNeighbors[i] = -1;
-			pTriInfos[f].AssignedGroup[i] = NULL;
-
-			pTriInfos[f].vOs.x=0.0f; pTriInfos[f].vOs.y=0.0f; pTriInfos[f].vOs.z=0.0f;
-			pTriInfos[f].vOt.x=0.0f; pTriInfos[f].vOt.y=0.0f; pTriInfos[f].vOt.z=0.0f;
-			pTriInfos[f].fMagS = 0;
-			pTriInfos[f].fMagT = 0;
-
-			// assumed bad
-			pTriInfos[f].iFlag |= GROUP_WITH_ANY;
-		}
-
-	// evaluate first order derivatives
-	for (f=0; f<iNrTrianglesIn; f++)
-	{
-		// initial values
-		const SVec3 v1 = GetPosition(pContext, piTriListIn[f*3+0]);
-		const SVec3 v2 = GetPosition(pContext, piTriListIn[f*3+1]);
-		const SVec3 v3 = GetPosition(pContext, piTriListIn[f*3+2]);
-		const SVec3 t1 = GetTexCoord(pContext, piTriListIn[f*3+0]);
-		const SVec3 t2 = GetTexCoord(pContext, piTriListIn[f*3+1]);
-		const SVec3 t3 = GetTexCoord(pContext, piTriListIn[f*3+2]);
-
-		const float t21x = t2.x-t1.x;
-		const float t21y = t2.y-t1.y;
-		const float t31x = t3.x-t1.x;
-		const float t31y = t3.y-t1.y;
-		const SVec3 d1 = vsub(v2,v1);
-		const SVec3 d2 = vsub(v3,v1);
-
-		const float fSignedAreaSTx2 = t21x*t31y - t21y*t31x;
-		//assert(fSignedAreaSTx2!=0);
-		SVec3 vOs = vsub(vscale(t31y,d1), vscale(t21y,d2));	// eq 18
-		SVec3 vOt = vadd(vscale(-t31x,d1), vscale(t21x,d2)); // eq 19
-
-		pTriInfos[f].iFlag |= (fSignedAreaSTx2>0 ? ORIENT_PRESERVING : 0);
-
-		if ( NotZero(fSignedAreaSTx2) )
-		{
-			const float fAbsArea = fabsf(fSignedAreaSTx2);
-			const float fLenOs = Length(vOs);
-			const float fLenOt = Length(vOt);
-			const float fS = (pTriInfos[f].iFlag&ORIENT_PRESERVING)==0 ? (-1.0f) : 1.0f;
-			if ( NotZero(fLenOs) ) pTriInfos[f].vOs = vscale(fS/fLenOs, vOs);
-			if ( NotZero(fLenOt) ) pTriInfos[f].vOt = vscale(fS/fLenOt, vOt);
-
-			// evaluate magnitudes prior to normalization of vOs and vOt
-			pTriInfos[f].fMagS = fLenOs / fAbsArea;
-			pTriInfos[f].fMagT = fLenOt / fAbsArea;
-
-			// if this is a good triangle
-			if ( NotZero(pTriInfos[f].fMagS) && NotZero(pTriInfos[f].fMagT))
-				pTriInfos[f].iFlag &= (~GROUP_WITH_ANY);
-		}
-	}
-
-	// force otherwise healthy quads to a fixed orientation
-	while (t<(iNrTrianglesIn-1))
-	{
-		const int iFO_a = pTriInfos[t].iOrgFaceNumber;
-		const int iFO_b = pTriInfos[t+1].iOrgFaceNumber;
-		if (iFO_a==iFO_b)	// this is a quad
-		{
-			const tbool bIsDeg_a = (pTriInfos[t].iFlag&MARK_DEGENERATE)!=0 ? TTRUE : TFALSE;
-			const tbool bIsDeg_b = (pTriInfos[t+1].iFlag&MARK_DEGENERATE)!=0 ? TTRUE : TFALSE;
-			
-			// bad triangles should already have been removed by
-			// DegenPrologue(), but just in case check bIsDeg_a and bIsDeg_a are false
-			if ((bIsDeg_a||bIsDeg_b)==TFALSE)
-			{
-				const tbool bOrientA = (pTriInfos[t].iFlag&ORIENT_PRESERVING)!=0 ? TTRUE : TFALSE;
-				const tbool bOrientB = (pTriInfos[t+1].iFlag&ORIENT_PRESERVING)!=0 ? TTRUE : TFALSE;
-				// if this happens the quad has extremely bad mapping!!
-				if (bOrientA!=bOrientB)
-				{
-					//printf("found quad with bad mapping\n");
-					tbool bChooseOrientFirstTri = TFALSE;
-					if ((pTriInfos[t+1].iFlag&GROUP_WITH_ANY)!=0) bChooseOrientFirstTri = TTRUE;
-					else if ( CalcTexArea(pContext, &piTriListIn[t*3+0]) >= CalcTexArea(pContext, &piTriListIn[(t+1)*3+0]) )
-						bChooseOrientFirstTri = TTRUE;
-
-					// force match
-					{
-						const int t0 = bChooseOrientFirstTri ? t : (t+1);
-						const int t1 = bChooseOrientFirstTri ? (t+1) : t;
-						pTriInfos[t1].iFlag &= (~ORIENT_PRESERVING);	// clear first
-						pTriInfos[t1].iFlag |= (pTriInfos[t0].iFlag&ORIENT_PRESERVING);	// copy bit
-					}
-				}
-			}
-			t += 2;
-		}
-		else
-			++t;
-	}
-	
-	// match up edge pairs
-	{
-		SEdge * pEdges = (SEdge *) malloc(sizeof(SEdge)*iNrTrianglesIn*3);
-		if (pEdges==NULL)
-			BuildNeighborsSlow(pTriInfos, piTriListIn, iNrTrianglesIn);
-		else
-		{
-			BuildNeighborsFast(pTriInfos, pEdges, piTriListIn, iNrTrianglesIn);
-	
-			free(pEdges);
-		}
-	}
-}
-
-/////////////////////////////////////////////////////////////////////////////////////////////////////
-/////////////////////////////////////////////////////////////////////////////////////////////////////
-
-static tbool AssignRecur(const int piTriListIn[], STriInfo psTriInfos[], const int iMyTriIndex, SGroup * pGroup);
-static void AddTriToGroup(SGroup * pGroup, const int iTriIndex);
-
-static int Build4RuleGroups(STriInfo pTriInfos[], SGroup pGroups[], int piGroupTrianglesBuffer[], const int piTriListIn[], const int iNrTrianglesIn)
-{
-	const int iNrMaxGroups = iNrTrianglesIn*3;
-	int iNrActiveGroups = 0;
-	int iOffset = 0, f=0, i=0;
-	(void)iNrMaxGroups;  /* quiet warnings in non debug mode */
-	for (f=0; f<iNrTrianglesIn; f++)
-	{
-		for (i=0; i<3; i++)
-		{
-			// if not assigned to a group
-			if ((pTriInfos[f].iFlag&GROUP_WITH_ANY)==0 && pTriInfos[f].AssignedGroup[i]==NULL)
-			{
-				tbool bOrPre;
-				int neigh_indexL, neigh_indexR;
-				const int vert_index = piTriListIn[f*3+i];
-				assert(iNrActiveGroups<iNrMaxGroups);
-				pTriInfos[f].AssignedGroup[i] = &pGroups[iNrActiveGroups];
-				pTriInfos[f].AssignedGroup[i]->iVertexRepresentitive = vert_index;
-				pTriInfos[f].AssignedGroup[i]->bOrientPreservering = (pTriInfos[f].iFlag&ORIENT_PRESERVING)!=0;
-				pTriInfos[f].AssignedGroup[i]->iNrFaces = 0;
-				pTriInfos[f].AssignedGroup[i]->pFaceIndices = &piGroupTrianglesBuffer[iOffset];
-				++iNrActiveGroups;
-
-				AddTriToGroup(pTriInfos[f].AssignedGroup[i], f);
-				bOrPre = (pTriInfos[f].iFlag&ORIENT_PRESERVING)!=0 ? TTRUE : TFALSE;
-				neigh_indexL = pTriInfos[f].FaceNeighbors[i];
-				neigh_indexR = pTriInfos[f].FaceNeighbors[i>0?(i-1):2];
-				if (neigh_indexL>=0) // neighbor
-				{
-					const tbool bAnswer =
-						AssignRecur(piTriListIn, pTriInfos, neigh_indexL,
-									pTriInfos[f].AssignedGroup[i] );
-					
-					const tbool bOrPre2 = (pTriInfos[neigh_indexL].iFlag&ORIENT_PRESERVING)!=0 ? TTRUE : TFALSE;
-					const tbool bDiff = bOrPre!=bOrPre2 ? TTRUE : TFALSE;
-					assert(bAnswer || bDiff);
-					(void)bAnswer, (void)bDiff;  /* quiet warnings in non debug mode */
-				}
-				if (neigh_indexR>=0) // neighbor
-				{
-					const tbool bAnswer =
-						AssignRecur(piTriListIn, pTriInfos, neigh_indexR,
-									pTriInfos[f].AssignedGroup[i] );
-
-					const tbool bOrPre2 = (pTriInfos[neigh_indexR].iFlag&ORIENT_PRESERVING)!=0 ? TTRUE : TFALSE;
-					const tbool bDiff = bOrPre!=bOrPre2 ? TTRUE : TFALSE;
-					assert(bAnswer || bDiff);
-					(void)bAnswer, (void)bDiff;  /* quiet warnings in non debug mode */
-				}
-
-				// update offset
-				iOffset += pTriInfos[f].AssignedGroup[i]->iNrFaces;
-				// since the groups are disjoint a triangle can never
-				// belong to more than 3 groups. Subsequently something
-				// is completely screwed if this assertion ever hits.
-				assert(iOffset <= iNrMaxGroups);
-			}
-		}
-	}
-
-	return iNrActiveGroups;
-}
-
-static void AddTriToGroup(SGroup * pGroup, const int iTriIndex)
-{
-	pGroup->pFaceIndices[pGroup->iNrFaces] = iTriIndex;
-	++pGroup->iNrFaces;
-}
-
-static tbool AssignRecur(const int piTriListIn[], STriInfo psTriInfos[],
-				 const int iMyTriIndex, SGroup * pGroup)
-{
-	STriInfo * pMyTriInfo = &psTriInfos[iMyTriIndex];
-
-	// track down vertex
-	const int iVertRep = pGroup->iVertexRepresentitive;
-	const int * pVerts = &piTriListIn[3*iMyTriIndex+0];
-	int i=-1;
-	if (pVerts[0]==iVertRep) i=0;
-	else if (pVerts[1]==iVertRep) i=1;
-	else if (pVerts[2]==iVertRep) i=2;
-	assert(i>=0 && i<3);
-
-	// early out
-	if (pMyTriInfo->AssignedGroup[i] == pGroup) return TTRUE;
-	else if (pMyTriInfo->AssignedGroup[i]!=NULL) return TFALSE;
-	if ((pMyTriInfo->iFlag&GROUP_WITH_ANY)!=0)
-	{
-		// first to group with a group-with-anything triangle
-		// determines it's orientation.
-		// This is the only existing order dependency in the code!!
-		if ( pMyTriInfo->AssignedGroup[0] == NULL &&
-			pMyTriInfo->AssignedGroup[1] == NULL &&
-			pMyTriInfo->AssignedGroup[2] == NULL )
-		{
-			pMyTriInfo->iFlag &= (~ORIENT_PRESERVING);
-			pMyTriInfo->iFlag |= (pGroup->bOrientPreservering ? ORIENT_PRESERVING : 0);
-		}
-	}
-	{
-		const tbool bOrient = (pMyTriInfo->iFlag&ORIENT_PRESERVING)!=0 ? TTRUE : TFALSE;
-		if (bOrient != pGroup->bOrientPreservering) return TFALSE;
-	}
-
-	AddTriToGroup(pGroup, iMyTriIndex);
-	pMyTriInfo->AssignedGroup[i] = pGroup;
-
-	{
-		const int neigh_indexL = pMyTriInfo->FaceNeighbors[i];
-		const int neigh_indexR = pMyTriInfo->FaceNeighbors[i>0?(i-1):2];
-		if (neigh_indexL>=0)
-			AssignRecur(piTriListIn, psTriInfos, neigh_indexL, pGroup);
-		if (neigh_indexR>=0)
-			AssignRecur(piTriListIn, psTriInfos, neigh_indexR, pGroup);
-	}
-
-
-
-	return TTRUE;
-}
-
-/////////////////////////////////////////////////////////////////////////////////////////////////////
-/////////////////////////////////////////////////////////////////////////////////////////////////////
-
-static tbool CompareSubGroups(const SSubGroup * pg1, const SSubGroup * pg2);
-static void QuickSort(int* pSortBuffer, int iLeft, int iRight, unsigned int uSeed);
-static STSpace EvalTspace(int face_indices[], const int iFaces, const int piTriListIn[], const STriInfo pTriInfos[], const SMikkTSpaceContext * pContext, const int iVertexRepresentitive);
-
-static tbool GenerateTSpaces(STSpace psTspace[], const STriInfo pTriInfos[], const SGroup pGroups[],
-                             const int iNrActiveGroups, const int piTriListIn[], const float fThresCos,
-                             const SMikkTSpaceContext * pContext)
-{
-	STSpace * pSubGroupTspace = NULL;
-	SSubGroup * pUniSubGroups = NULL;
-	int * pTmpMembers = NULL;
-	int iMaxNrFaces=0, iUniqueTspaces=0, g=0, i=0;
-	for (g=0; g<iNrActiveGroups; g++)
-		if (iMaxNrFaces < pGroups[g].iNrFaces)
-			iMaxNrFaces = pGroups[g].iNrFaces;
-
-	if (iMaxNrFaces == 0) return TTRUE;
-
-	// make initial allocations
-	pSubGroupTspace = (STSpace *) malloc(sizeof(STSpace)*iMaxNrFaces);
-	pUniSubGroups = (SSubGroup *) malloc(sizeof(SSubGroup)*iMaxNrFaces);
-	pTmpMembers = (int *) malloc(sizeof(int)*iMaxNrFaces);
-	if (pSubGroupTspace==NULL || pUniSubGroups==NULL || pTmpMembers==NULL)
-	{
-		if (pSubGroupTspace!=NULL) free(pSubGroupTspace);
-		if (pUniSubGroups!=NULL) free(pUniSubGroups);
-		if (pTmpMembers!=NULL) free(pTmpMembers);
-		return TFALSE;
-	}
-
-
-	iUniqueTspaces = 0;
-	for (g=0; g<iNrActiveGroups; g++)
-	{
-		const SGroup * pGroup = &pGroups[g];
-		int iUniqueSubGroups = 0, s=0;
-
-		for (i=0; i<pGroup->iNrFaces; i++)	// triangles
-		{
-			const int f = pGroup->pFaceIndices[i];	// triangle number
-			int index=-1, iVertIndex=-1, iOF_1=-1, iMembers=0, j=0, l=0;
-			SSubGroup tmp_group;
-			tbool bFound;
-			SVec3 n, vOs, vOt;
-			if (pTriInfos[f].AssignedGroup[0]==pGroup) index=0;
-			else if (pTriInfos[f].AssignedGroup[1]==pGroup) index=1;
-			else if (pTriInfos[f].AssignedGroup[2]==pGroup) index=2;
-			assert(index>=0 && index<3);
-
-			iVertIndex = piTriListIn[f*3+index];
-			assert(iVertIndex==pGroup->iVertexRepresentitive);
-
-			// is normalized already
-			n = GetNormal(pContext, iVertIndex);
-			
-			// project
-			vOs = vsub(pTriInfos[f].vOs, vscale(vdot(n,pTriInfos[f].vOs), n));
-			vOt = vsub(pTriInfos[f].vOt, vscale(vdot(n,pTriInfos[f].vOt), n));
-			if ( VNotZero(vOs) ) vOs = Normalize(vOs);
-			if ( VNotZero(vOt) ) vOt = Normalize(vOt);
-
-			// original face number
-			iOF_1 = pTriInfos[f].iOrgFaceNumber;
-			
-			iMembers = 0;
-			for (j=0; j<pGroup->iNrFaces; j++)
-			{
-				const int t = pGroup->pFaceIndices[j];	// triangle number
-				const int iOF_2 = pTriInfos[t].iOrgFaceNumber;
-
-				// project
-				SVec3 vOs2 = vsub(pTriInfos[t].vOs, vscale(vdot(n,pTriInfos[t].vOs), n));
-				SVec3 vOt2 = vsub(pTriInfos[t].vOt, vscale(vdot(n,pTriInfos[t].vOt), n));
-				if ( VNotZero(vOs2) ) vOs2 = Normalize(vOs2);
-				if ( VNotZero(vOt2) ) vOt2 = Normalize(vOt2);
-
-				{
-					const tbool bAny = ( (pTriInfos[f].iFlag | pTriInfos[t].iFlag) & GROUP_WITH_ANY )!=0 ? TTRUE : TFALSE;
-					// make sure triangles which belong to the same quad are joined.
-					const tbool bSameOrgFace = iOF_1==iOF_2 ? TTRUE : TFALSE;
-
-					const float fCosS = vdot(vOs,vOs2);
-					const float fCosT = vdot(vOt,vOt2);
-
-					assert(f!=t || bSameOrgFace);	// sanity check
-					if (bAny || bSameOrgFace || (fCosS>fThresCos && fCosT>fThresCos))
-						pTmpMembers[iMembers++] = t;
-				}
-			}
-
-			// sort pTmpMembers
-			tmp_group.iNrFaces = iMembers;
-			tmp_group.pTriMembers = pTmpMembers;
-			if (iMembers>1)
-			{
-				unsigned int uSeed = INTERNAL_RND_SORT_SEED;	// could replace with a random seed?
-				QuickSort(pTmpMembers, 0, iMembers-1, uSeed);
-			}
-
-			// look for an existing match
-			bFound = TFALSE;
-			l=0;
-			while (l<iUniqueSubGroups && !bFound)
-			{
-				bFound = CompareSubGroups(&tmp_group, &pUniSubGroups[l]);
-				if (!bFound) ++l;
-			}
-			
-			// assign tangent space index
-			assert(bFound || l==iUniqueSubGroups);
-			//piTempTangIndices[f*3+index] = iUniqueTspaces+l;
-
-			// if no match was found we allocate a new subgroup
-			if (!bFound)
-			{
-				// insert new subgroup
-				int * pIndices = (int *) malloc(sizeof(int)*iMembers);
-				if (pIndices==NULL)
-				{
-					// clean up and return false
-					int s=0;
-					for (s=0; s<iUniqueSubGroups; s++)
-						free(pUniSubGroups[s].pTriMembers);
-					free(pUniSubGroups);
-					free(pTmpMembers);
-					free(pSubGroupTspace);
-					return TFALSE;
-				}
-				pUniSubGroups[iUniqueSubGroups].iNrFaces = iMembers;
-				pUniSubGroups[iUniqueSubGroups].pTriMembers = pIndices;
-				memcpy(pIndices, tmp_group.pTriMembers, iMembers*sizeof(int));
-				pSubGroupTspace[iUniqueSubGroups] =
-					EvalTspace(tmp_group.pTriMembers, iMembers, piTriListIn, pTriInfos, pContext, pGroup->iVertexRepresentitive);
-				++iUniqueSubGroups;
-			}
-
-			// output tspace
-			{
-				const int iOffs = pTriInfos[f].iTSpacesOffs;
-				const int iVert = pTriInfos[f].vert_num[index];
-				STSpace * pTS_out = &psTspace[iOffs+iVert];
-				assert(pTS_out->iCounter<2);
-				assert(((pTriInfos[f].iFlag&ORIENT_PRESERVING)!=0) == pGroup->bOrientPreservering);
-				if (pTS_out->iCounter==1)
-				{
-					*pTS_out = AvgTSpace(pTS_out, &pSubGroupTspace[l]);
-					pTS_out->iCounter = 2;	// update counter
-					pTS_out->bOrient = pGroup->bOrientPreservering;
-				}
-				else
-				{
-					assert(pTS_out->iCounter==0);
-					*pTS_out = pSubGroupTspace[l];
-					pTS_out->iCounter = 1;	// update counter
-					pTS_out->bOrient = pGroup->bOrientPreservering;
-				}
-			}
-		}
-
-		// clean up and offset iUniqueTspaces
-		for (s=0; s<iUniqueSubGroups; s++)
-			free(pUniSubGroups[s].pTriMembers);
-		iUniqueTspaces += iUniqueSubGroups;
-	}
-
-	// clean up
-	free(pUniSubGroups);
-	free(pTmpMembers);
-	free(pSubGroupTspace);
-
-	return TTRUE;
-}
-
-static STSpace EvalTspace(int face_indices[], const int iFaces, const int piTriListIn[], const STriInfo pTriInfos[],
-                          const SMikkTSpaceContext * pContext, const int iVertexRepresentitive)
-{
-	STSpace res;
-	float fAngleSum = 0;
-	int face=0;
-	res.vOs.x=0.0f; res.vOs.y=0.0f; res.vOs.z=0.0f;
-	res.vOt.x=0.0f; res.vOt.y=0.0f; res.vOt.z=0.0f;
-	res.fMagS = 0; res.fMagT = 0;
-
-	for (face=0; face<iFaces; face++)
-	{
-		const int f = face_indices[face];
-
-		// only valid triangles get to add their contribution
-		if ( (pTriInfos[f].iFlag&GROUP_WITH_ANY)==0 )
-		{
-			SVec3 n, vOs, vOt, p0, p1, p2, v1, v2;
-			float fCos, fAngle, fMagS, fMagT;
-			int i=-1, index=-1, i0=-1, i1=-1, i2=-1;
-			if (piTriListIn[3*f+0]==iVertexRepresentitive) i=0;
-			else if (piTriListIn[3*f+1]==iVertexRepresentitive) i=1;
-			else if (piTriListIn[3*f+2]==iVertexRepresentitive) i=2;
-			assert(i>=0 && i<3);
-
-			// project
-			index = piTriListIn[3*f+i];
-			n = GetNormal(pContext, index);
-			vOs = vsub(pTriInfos[f].vOs, vscale(vdot(n,pTriInfos[f].vOs), n));
-			vOt = vsub(pTriInfos[f].vOt, vscale(vdot(n,pTriInfos[f].vOt), n));
-			if ( VNotZero(vOs) ) vOs = Normalize(vOs);
-			if ( VNotZero(vOt) ) vOt = Normalize(vOt);
-
-			i2 = piTriListIn[3*f + (i<2?(i+1):0)];
-			i1 = piTriListIn[3*f + i];
-			i0 = piTriListIn[3*f + (i>0?(i-1):2)];
-
-			p0 = GetPosition(pContext, i0);
-			p1 = GetPosition(pContext, i1);
-			p2 = GetPosition(pContext, i2);
-			v1 = vsub(p0,p1);
-			v2 = vsub(p2,p1);
-
-			// project
-			v1 = vsub(v1, vscale(vdot(n,v1),n)); if ( VNotZero(v1) ) v1 = Normalize(v1);
-			v2 = vsub(v2, vscale(vdot(n,v2),n)); if ( VNotZero(v2) ) v2 = Normalize(v2);
-
-			// weight contribution by the angle
-			// between the two edge vectors
-			fCos = vdot(v1,v2); fCos=fCos>1?1:(fCos<(-1) ? (-1) : fCos);
-			fAngle = (float) acos(fCos);
-			fMagS = pTriInfos[f].fMagS;
-			fMagT = pTriInfos[f].fMagT;
-
-			res.vOs=vadd(res.vOs, vscale(fAngle,vOs));
-			res.vOt=vadd(res.vOt,vscale(fAngle,vOt));
-			res.fMagS+=(fAngle*fMagS);
-			res.fMagT+=(fAngle*fMagT);
-			fAngleSum += fAngle;
-		}
-	}
-
-	// normalize
-	if ( VNotZero(res.vOs) ) res.vOs = Normalize(res.vOs);
-	if ( VNotZero(res.vOt) ) res.vOt = Normalize(res.vOt);
-	if (fAngleSum>0)
-	{
-		res.fMagS /= fAngleSum;
-		res.fMagT /= fAngleSum;
-	}
-
-	return res;
-}
-
-static tbool CompareSubGroups(const SSubGroup * pg1, const SSubGroup * pg2)
-{
-	tbool bStillSame=TTRUE;
-	int i=0;
-	if (pg1->iNrFaces!=pg2->iNrFaces) return TFALSE;
-	while (i<pg1->iNrFaces && bStillSame)
-	{
-		bStillSame = pg1->pTriMembers[i]==pg2->pTriMembers[i] ? TTRUE : TFALSE;
-		if (bStillSame) ++i;
-	}
-	return bStillSame;
-}
-
-static void QuickSort(int* pSortBuffer, int iLeft, int iRight, unsigned int uSeed)
-{
-	int iL, iR, n, index, iMid, iTmp;
-
-	// Random
-	unsigned int t=uSeed&31;
-	t=(uSeed<<t)|(uSeed>>(32-t));
-	uSeed=uSeed+t+3;
-	// Random end
-
-	iL=iLeft; iR=iRight;
-	n = (iR-iL)+1;
-	assert(n>=0);
-	index = (int) (uSeed%n);
-
-	iMid=pSortBuffer[index + iL];
-
-
-	do
-	{
-		while (pSortBuffer[iL] < iMid)
-			++iL;
-		while (pSortBuffer[iR] > iMid)
-			--iR;
-
-		if (iL <= iR)
-		{
-			iTmp = pSortBuffer[iL];
-			pSortBuffer[iL] = pSortBuffer[iR];
-			pSortBuffer[iR] = iTmp;
-			++iL; --iR;
-		}
-	}
-	while (iL <= iR);
-
-	if (iLeft < iR)
-		QuickSort(pSortBuffer, iLeft, iR, uSeed);
-	if (iL < iRight)
-		QuickSort(pSortBuffer, iL, iRight, uSeed);
-}
-
-/////////////////////////////////////////////////////////////////////////////////////////////
-/////////////////////////////////////////////////////////////////////////////////////////////
-
-static void QuickSortEdges(SEdge * pSortBuffer, int iLeft, int iRight, const int channel, unsigned int uSeed);
-static void GetEdge(int * i0_out, int * i1_out, int * edgenum_out, const int indices[], const int i0_in, const int i1_in);
-
-static void BuildNeighborsFast(STriInfo pTriInfos[], SEdge * pEdges, const int piTriListIn[], const int iNrTrianglesIn)
-{
-	// build array of edges
-	unsigned int uSeed = INTERNAL_RND_SORT_SEED;				// could replace with a random seed?
-	int iEntries=0, iCurStartIndex=-1, f=0, i=0;
-	for (f=0; f<iNrTrianglesIn; f++)
-		for (i=0; i<3; i++)
-		{
-			const int i0 = piTriListIn[f*3+i];
-			const int i1 = piTriListIn[f*3+(i<2?(i+1):0)];
-			pEdges[f*3+i].i0 = i0 < i1 ? i0 : i1;			// put minimum index in i0
-			pEdges[f*3+i].i1 = !(i0 < i1) ? i0 : i1;		// put maximum index in i1
-			pEdges[f*3+i].f = f;							// record face number
-		}
-
-	// sort over all edges by i0, this is the pricy one.
-	QuickSortEdges(pEdges, 0, iNrTrianglesIn*3-1, 0, uSeed);	// sort channel 0 which is i0
-
-	// sub sort over i1, should be fast.
-	// could replace this with a 64 bit int sort over (i0,i1)
-	// with i0 as msb in the quicksort call above.
-	iEntries = iNrTrianglesIn*3;
-	iCurStartIndex = 0;
-	for (i=1; i<iEntries; i++)
-	{
-		if (pEdges[iCurStartIndex].i0 != pEdges[i].i0)
-		{
-			const int iL = iCurStartIndex;
-			const int iR = i-1;
-			//const int iElems = i-iL;
-			iCurStartIndex = i;
-			QuickSortEdges(pEdges, iL, iR, 1, uSeed);	// sort channel 1 which is i1
-		}
-	}
-
-	// sub sort over f, which should be fast.
-	// this step is to remain compliant with BuildNeighborsSlow() when
-	// more than 2 triangles use the same edge (such as a butterfly topology).
-	iCurStartIndex = 0;
-	for (i=1; i<iEntries; i++)
-	{
-		if (pEdges[iCurStartIndex].i0 != pEdges[i].i0 || pEdges[iCurStartIndex].i1 != pEdges[i].i1)
-		{
-			const int iL = iCurStartIndex;
-			const int iR = i-1;
-			//const int iElems = i-iL;
-			iCurStartIndex = i;
-			QuickSortEdges(pEdges, iL, iR, 2, uSeed);	// sort channel 2 which is f
-		}
-	}
-
-	// pair up, adjacent triangles
-	for (i=0; i<iEntries; i++)
-	{
-		const int i0=pEdges[i].i0;
-		const int i1=pEdges[i].i1;
-		const int f = pEdges[i].f;
-		tbool bUnassigned_A;
-
-		int i0_A, i1_A;
-		int edgenum_A, edgenum_B=0;	// 0,1 or 2
-		GetEdge(&i0_A, &i1_A, &edgenum_A, &piTriListIn[f*3], i0, i1);	// resolve index ordering and edge_num
-		bUnassigned_A = pTriInfos[f].FaceNeighbors[edgenum_A] == -1 ? TTRUE : TFALSE;
-
-		if (bUnassigned_A)
-		{
-			// get true index ordering
-			int j=i+1, t;
-			tbool bNotFound = TTRUE;
-			while (j<iEntries && i0==pEdges[j].i0 && i1==pEdges[j].i1 && bNotFound)
-			{
-				tbool bUnassigned_B;
-				int i0_B, i1_B;
-				t = pEdges[j].f;
-				// flip i0_B and i1_B
-				GetEdge(&i1_B, &i0_B, &edgenum_B, &piTriListIn[t*3], pEdges[j].i0, pEdges[j].i1);	// resolve index ordering and edge_num
-				//assert(!(i0_A==i1_B && i1_A==i0_B));
-				bUnassigned_B =  pTriInfos[t].FaceNeighbors[edgenum_B]==-1 ? TTRUE : TFALSE;
-				if (i0_A==i0_B && i1_A==i1_B && bUnassigned_B)
-					bNotFound = TFALSE;
-				else
-					++j;
-			}
-
-			if (!bNotFound)
-			{
-				int t = pEdges[j].f;
-				pTriInfos[f].FaceNeighbors[edgenum_A] = t;
-				//assert(pTriInfos[t].FaceNeighbors[edgenum_B]==-1);
-				pTriInfos[t].FaceNeighbors[edgenum_B] = f;
-			}
-		}
-	}
-}
-
-static void BuildNeighborsSlow(STriInfo pTriInfos[], const int piTriListIn[], const int iNrTrianglesIn)
-{
-	int f=0, i=0;
-	for (f=0; f<iNrTrianglesIn; f++)
-	{
-		for (i=0; i<3; i++)
-		{
-			// if unassigned
-			if (pTriInfos[f].FaceNeighbors[i] == -1)
-			{
-				const int i0_A = piTriListIn[f*3+i];
-				const int i1_A = piTriListIn[f*3+(i<2?(i+1):0)];
-
-				// search for a neighbor
-				tbool bFound = TFALSE;
-				int t=0, j=0;
-				while (!bFound && t<iNrTrianglesIn)
-				{
-					if (t!=f)
-					{
-						j=0;
-						while (!bFound && j<3)
-						{
-							// in rev order
-							const int i1_B = piTriListIn[t*3+j];
-							const int i0_B = piTriListIn[t*3+(j<2?(j+1):0)];
-							//assert(!(i0_A==i1_B && i1_A==i0_B));
-							if (i0_A==i0_B && i1_A==i1_B)
-								bFound = TTRUE;
-							else
-								++j;
-						}
-					}
-					
-					if (!bFound) ++t;
-				}
-
-				// assign neighbors
-				if (bFound)
-				{
-					pTriInfos[f].FaceNeighbors[i] = t;
-					//assert(pTriInfos[t].FaceNeighbors[j]==-1);
-					pTriInfos[t].FaceNeighbors[j] = f;
-				}
-			}
-		}
-	}
-}
-
-static void QuickSortEdges(SEdge * pSortBuffer, int iLeft, int iRight, const int channel, unsigned int uSeed)
-{
-	unsigned int t;
-	int iL, iR, n, index, iMid;
-
-	// early out
-	SEdge sTmp;
-	const int iElems = iRight-iLeft+1;
-	if (iElems<2) return;
-	else if (iElems==2)
-	{
-		if (pSortBuffer[iLeft].array[channel] > pSortBuffer[iRight].array[channel])
-		{
-			sTmp = pSortBuffer[iLeft];
-			pSortBuffer[iLeft] = pSortBuffer[iRight];
-			pSortBuffer[iRight] = sTmp;
-		}
-		return;
-	}
-
-	// Random
-	t=uSeed&31;
-	t=(uSeed<<t)|(uSeed>>(32-t));
-	uSeed=uSeed+t+3;
-	// Random end
-
-	iL=iLeft, iR=iRight;
-	n = (iR-iL)+1;
-	assert(n>=0);
-	index = (int) (uSeed%n);
-
-	iMid=pSortBuffer[index + iL].array[channel];
-
-	do
-	{
-		while (pSortBuffer[iL].array[channel] < iMid)
-			++iL;
-		while (pSortBuffer[iR].array[channel] > iMid)
-			--iR;
-
-		if (iL <= iR)
-		{
-			sTmp = pSortBuffer[iL];
-			pSortBuffer[iL] = pSortBuffer[iR];
-			pSortBuffer[iR] = sTmp;
-			++iL; --iR;
-		}
-	}
-	while (iL <= iR);
-
-	if (iLeft < iR)
-		QuickSortEdges(pSortBuffer, iLeft, iR, channel, uSeed);
-	if (iL < iRight)
-		QuickSortEdges(pSortBuffer, iL, iRight, channel, uSeed);
-}
-
-// resolve ordering and edge number
-static void GetEdge(int * i0_out, int * i1_out, int * edgenum_out, const int indices[], const int i0_in, const int i1_in)
-{
-	*edgenum_out = -1;
-	
-	// test if first index is on the edge
-	if (indices[0]==i0_in || indices[0]==i1_in)
-	{
-		// test if second index is on the edge
-		if (indices[1]==i0_in || indices[1]==i1_in)
-		{
-			edgenum_out[0]=0;	// first edge
-			i0_out[0]=indices[0];
-			i1_out[0]=indices[1];
-		}
-		else
-		{
-			edgenum_out[0]=2;	// third edge
-			i0_out[0]=indices[2];
-			i1_out[0]=indices[0];
-		}
-	}
-	else
-	{
-		// only second and third index is on the edge
-		edgenum_out[0]=1;	// second edge
-		i0_out[0]=indices[1];
-		i1_out[0]=indices[2];
-	}
-}
-
-
-/////////////////////////////////////////////////////////////////////////////////////////////
-/////////////////////////////////// Degenerate triangles ////////////////////////////////////
-
-static void DegenPrologue(STriInfo pTriInfos[], int piTriList_out[], const int iNrTrianglesIn, const int iTotTris)
-{
-	int iNextGoodTriangleSearchIndex=-1;
-	tbool bStillFindingGoodOnes;
-
-	// locate quads with only one good triangle
-	int t=0;
-	while (t<(iTotTris-1))
-	{
-		const int iFO_a = pTriInfos[t].iOrgFaceNumber;
-		const int iFO_b = pTriInfos[t+1].iOrgFaceNumber;
-		if (iFO_a==iFO_b)	// this is a quad
-		{
-			const tbool bIsDeg_a = (pTriInfos[t].iFlag&MARK_DEGENERATE)!=0 ? TTRUE : TFALSE;
-			const tbool bIsDeg_b = (pTriInfos[t+1].iFlag&MARK_DEGENERATE)!=0 ? TTRUE : TFALSE;
-			if ((bIsDeg_a^bIsDeg_b)!=0)
-			{
-				pTriInfos[t].iFlag |= QUAD_ONE_DEGEN_TRI;
-				pTriInfos[t+1].iFlag |= QUAD_ONE_DEGEN_TRI;
-			}
-			t += 2;
-		}
-		else
-			++t;
-	}
-
-	// reorder list so all degen triangles are moved to the back
-	// without reordering the good triangles
-	iNextGoodTriangleSearchIndex = 1;
-	t=0;
-	bStillFindingGoodOnes = TTRUE;
-	while (t<iNrTrianglesIn && bStillFindingGoodOnes)
-	{
-		const tbool bIsGood = (pTriInfos[t].iFlag&MARK_DEGENERATE)==0 ? TTRUE : TFALSE;
-		if (bIsGood)
-		{
-			if (iNextGoodTriangleSearchIndex < (t+2))
-				iNextGoodTriangleSearchIndex = t+2;
-		}
-		else
-		{
-			int t0, t1;
-			// search for the first good triangle.
-			tbool bJustADegenerate = TTRUE;
-			while (bJustADegenerate && iNextGoodTriangleSearchIndex<iTotTris)
-			{
-				const tbool bIsGood = (pTriInfos[iNextGoodTriangleSearchIndex].iFlag&MARK_DEGENERATE)==0 ? TTRUE : TFALSE;
-				if (bIsGood) bJustADegenerate=TFALSE;
-				else ++iNextGoodTriangleSearchIndex;
-			}
-
-			t0 = t;
-			t1 = iNextGoodTriangleSearchIndex;
-			++iNextGoodTriangleSearchIndex;
-			assert(iNextGoodTriangleSearchIndex > (t+1));
-
-			// swap triangle t0 and t1
-			if (!bJustADegenerate)
-			{
-				int i=0;
-				for (i=0; i<3; i++)
-				{
-					const int index = piTriList_out[t0*3+i];
-					piTriList_out[t0*3+i] = piTriList_out[t1*3+i];
-					piTriList_out[t1*3+i] = index;
-				}
-				{
-					const STriInfo tri_info = pTriInfos[t0];
-					pTriInfos[t0] = pTriInfos[t1];
-					pTriInfos[t1] = tri_info;
-				}
-			}
-			else
-				bStillFindingGoodOnes = TFALSE;	// this is not supposed to happen
-		}
-
-		if (bStillFindingGoodOnes) ++t;
-	}
-
-	assert(bStillFindingGoodOnes);	// code will still work.
-	assert(iNrTrianglesIn == t);
-}
-
-static void DegenEpilogue(STSpace psTspace[], STriInfo pTriInfos[], int piTriListIn[], const SMikkTSpaceContext * pContext, const int iNrTrianglesIn, const int iTotTris)
-{
-	int t=0, i=0;
-	// deal with degenerate triangles
-	// punishment for degenerate triangles is O(N^2)
-	for (t=iNrTrianglesIn; t<iTotTris; t++)
-	{
-		// degenerate triangles on a quad with one good triangle are skipped
-		// here but processed in the next loop
-		const tbool bSkip = (pTriInfos[t].iFlag&QUAD_ONE_DEGEN_TRI)!=0 ? TTRUE : TFALSE;
-
-		if (!bSkip)
-		{
-			for (i=0; i<3; i++)
-			{
-				const int index1 = piTriListIn[t*3+i];
-				// search through the good triangles
-				tbool bNotFound = TTRUE;
-				int j=0;
-				while (bNotFound && j<(3*iNrTrianglesIn))
-				{
-					const int index2 = piTriListIn[j];
-					if (index1==index2) bNotFound=TFALSE;
-					else ++j;
-				}
-
-				if (!bNotFound)
-				{
-					const int iTri = j/3;
-					const int iVert = j%3;
-					const int iSrcVert=pTriInfos[iTri].vert_num[iVert];
-					const int iSrcOffs=pTriInfos[iTri].iTSpacesOffs;
-					const int iDstVert=pTriInfos[t].vert_num[i];
-					const int iDstOffs=pTriInfos[t].iTSpacesOffs;
-					
-					// copy tspace
-					psTspace[iDstOffs+iDstVert] = psTspace[iSrcOffs+iSrcVert];
-				}
-			}
-		}
-	}
-
-	// deal with degenerate quads with one good triangle
-	for (t=0; t<iNrTrianglesIn; t++)
-	{
-		// this triangle belongs to a quad where the
-		// other triangle is degenerate
-		if ( (pTriInfos[t].iFlag&QUAD_ONE_DEGEN_TRI)!=0 )
-		{
-			SVec3 vDstP;
-			int iOrgF=-1, i=0;
-			tbool bNotFound;
-			unsigned char * pV = pTriInfos[t].vert_num;
-			int iFlag = (1<<pV[0]) | (1<<pV[1]) | (1<<pV[2]);
-			int iMissingIndex = 0;
-			if ((iFlag&2)==0) iMissingIndex=1;
-			else if ((iFlag&4)==0) iMissingIndex=2;
-			else if ((iFlag&8)==0) iMissingIndex=3;
-
-			iOrgF = pTriInfos[t].iOrgFaceNumber;
-			vDstP = GetPosition(pContext, MakeIndex(iOrgF, iMissingIndex));
-			bNotFound = TTRUE;
-			i=0;
-			while (bNotFound && i<3)
-			{
-				const int iVert = pV[i];
-				const SVec3 vSrcP = GetPosition(pContext, MakeIndex(iOrgF, iVert));
-				if (veq(vSrcP, vDstP)==TTRUE)
-				{
-					const int iOffs = pTriInfos[t].iTSpacesOffs;
-					psTspace[iOffs+iMissingIndex] = psTspace[iOffs+iVert];
-					bNotFound=TFALSE;
-				}
-				else
-					++i;
-			}
-			assert(!bNotFound);
-		}
-	}
-}
diff --git a/scene/resources/mikktspace.h b/scene/resources/mikktspace.h
deleted file mode 100644
index 52c44a713c..0000000000
--- a/scene/resources/mikktspace.h
+++ /dev/null
@@ -1,145 +0,0 @@
-/** \file mikktspace/mikktspace.h
- *  \ingroup mikktspace
- */
-/**
- *  Copyright (C) 2011 by Morten S. Mikkelsen
- *
- *  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 __MIKKTSPACE_H__
-#define __MIKKTSPACE_H__
-
-
-#ifdef __cplusplus
-extern "C" {
-#endif
-
-/* Author: Morten S. Mikkelsen
- * Version: 1.0
- *
- * The files mikktspace.h and mikktspace.c are designed to be
- * stand-alone files and it is important that they are kept this way.
- * Not having dependencies on structures/classes/libraries specific
- * to the program, in which they are used, allows them to be copied
- * and used as is into any tool, program or plugin.
- * The code is designed to consistently generate the same
- * tangent spaces, for a given mesh, in any tool in which it is used.
- * This is done by performing an internal welding step and subsequently an order-independent evaluation
- * of tangent space for meshes consisting of triangles and quads.
- * This means faces can be received in any order and the same is true for
- * the order of vertices of each face. The generated result will not be affected
- * by such reordering. Additionally, whether degenerate (vertices or texture coordinates)
- * primitives are present or not will not affect the generated results either.
- * Once tangent space calculation is done the vertices of degenerate primitives will simply
- * inherit tangent space from neighboring non degenerate primitives.
- * The analysis behind this implementation can be found in my master's thesis
- * which is available for download --> http://image.diku.dk/projects/media/morten.mikkelsen.08.pdf
- * Note that though the tangent spaces at the vertices are generated in an order-independent way,
- * by this implementation, the interpolated tangent space is still affected by which diagonal is
- * chosen to split each quad. A sensible solution is to have your tools pipeline always
- * split quads by the shortest diagonal. This choice is order-independent and works with mirroring.
- * If these have the same length then compare the diagonals defined by the texture coordinates.
- * XNormal which is a tool for baking normal maps allows you to write your own tangent space plugin
- * and also quad triangulator plugin.
- */
-
-
-typedef int tbool;
-typedef struct SMikkTSpaceContext SMikkTSpaceContext;
-
-typedef struct {
-	// Returns the number of faces (triangles/quads) on the mesh to be processed.
-	int (*m_getNumFaces)(const SMikkTSpaceContext * pContext);
-
-	// Returns the number of vertices on face number iFace
-	// iFace is a number in the range {0, 1, ..., getNumFaces()-1}
-	int (*m_getNumVerticesOfFace)(const SMikkTSpaceContext * pContext, const int iFace);
-
-	// returns the position/normal/texcoord of the referenced face of vertex number iVert.
-	// iVert is in the range {0,1,2} for triangles and {0,1,2,3} for quads.
-	void (*m_getPosition)(const SMikkTSpaceContext * pContext, float fvPosOut[], const int iFace, const int iVert);
-	void (*m_getNormal)(const SMikkTSpaceContext * pContext, float fvNormOut[], const int iFace, const int iVert);
-	void (*m_getTexCoord)(const SMikkTSpaceContext * pContext, float fvTexcOut[], const int iFace, const int iVert);
-
-	// either (or both) of the two setTSpace callbacks can be set.
-	// The call-back m_setTSpaceBasic() is sufficient for basic normal mapping.
-
-	// This function is used to return the tangent and fSign to the application.
-	// fvTangent is a unit length vector.
-	// For normal maps it is sufficient to use the following simplified version of the bitangent which is generated at pixel/vertex level.
-	// bitangent = fSign * cross(vN, tangent);
-	// Note that the results are returned unindexed. It is possible to generate a new index list
-	// But averaging/overwriting tangent spaces by using an already existing index list WILL produce INCRORRECT results.
-	// DO NOT! use an already existing index list.
-	void (*m_setTSpaceBasic)(const SMikkTSpaceContext * pContext, const float fvTangent[], const float fSign, const int iFace, const int iVert);
-
-	// This function is used to return tangent space results to the application.
-	// fvTangent and fvBiTangent are unit length vectors and fMagS and fMagT are their
-	// true magnitudes which can be used for relief mapping effects.
-	// fvBiTangent is the "real" bitangent and thus may not be perpendicular to fvTangent.
-	// However, both are perpendicular to the vertex normal.
-	// For normal maps it is sufficient to use the following simplified version of the bitangent which is generated at pixel/vertex level.
-	// fSign = bIsOrientationPreserving ? 1.0f : (-1.0f);
-	// bitangent = fSign * cross(vN, tangent);
-	// Note that the results are returned unindexed. It is possible to generate a new index list
-	// But averaging/overwriting tangent spaces by using an already existing index list WILL produce INCRORRECT results.
-	// DO NOT! use an already existing index list.
-	void (*m_setTSpace)(const SMikkTSpaceContext * pContext, const float fvTangent[], const float fvBiTangent[], const float fMagS, const float fMagT,
-						const tbool bIsOrientationPreserving, const int iFace, const int iVert);
-} SMikkTSpaceInterface;
-
-struct SMikkTSpaceContext
-{
-	SMikkTSpaceInterface * m_pInterface;	// initialized with callback functions
-	void * m_pUserData;						// pointer to client side mesh data etc. (passed as the first parameter with every interface call)
-};
-
-// these are both thread safe!
-tbool genTangSpaceDefault(const SMikkTSpaceContext * pContext);	// Default (recommended) fAngularThreshold is 180 degrees (which means threshold disabled)
-tbool genTangSpace(const SMikkTSpaceContext * pContext, const float fAngularThreshold);
-
-
-// To avoid visual errors (distortions/unwanted hard edges in lighting), when using sampled normal maps, the
-// normal map sampler must use the exact inverse of the pixel shader transformation.
-// The most efficient transformation we can possibly do in the pixel shader is
-// achieved by using, directly, the "unnormalized" interpolated tangent, bitangent and vertex normal: vT, vB and vN.
-// pixel shader (fast transform out)
-// vNout = normalize( vNt.x * vT + vNt.y * vB + vNt.z * vN );
-// where vNt is the tangent space normal. The normal map sampler must likewise use the
-// interpolated and "unnormalized" tangent, bitangent and vertex normal to be compliant with the pixel shader.
-// sampler does (exact inverse of pixel shader):
-// float3 row0 = cross(vB, vN);
-// float3 row1 = cross(vN, vT);
-// float3 row2 = cross(vT, vB);
-// float fSign = dot(vT, row0)<0 ? -1 : 1;
-// vNt = normalize( fSign * float3(dot(vNout,row0), dot(vNout,row1), dot(vNout,row2)) );
-// where vNout is the sampled normal in some chosen 3D space.
-//
-// Should you choose to reconstruct the bitangent in the pixel shader instead
-// of the vertex shader, as explained earlier, then be sure to do this in the normal map sampler also.
-// Finally, beware of quad triangulations. If the normal map sampler doesn't use the same triangulation of
-// quads as your renderer then problems will occur since the interpolated tangent spaces will differ
-// eventhough the vertex level tangent spaces match. This can be solved either by triangulating before
-// sampling/exporting or by using the order-independent choice of diagonal for splitting quads suggested earlier.
-// However, this must be used both by the sampler and your tools/rendering pipeline.
-
-#ifdef __cplusplus
-}
-#endif
-
-#endif
diff --git a/scene/resources/stb_truetype.h b/scene/resources/stb_truetype.h
deleted file mode 100644
index 016972785a..0000000000
--- a/scene/resources/stb_truetype.h
+++ /dev/null
@@ -1,3267 +0,0 @@
-// stb_truetype.h - v1.11 - public domain
-// authored from 2009-2015 by Sean Barrett / RAD Game Tools
-//
-//   This library processes TrueType files:
-//        parse files
-//        extract glyph metrics
-//        extract glyph shapes
-//        render glyphs to one-channel bitmaps with antialiasing (box filter)
-//
-//   Todo:
-//        non-MS cmaps
-//        crashproof on bad data
-//        hinting? (no longer patented)
-//        cleartype-style AA?
-//        optimize: use simple memory allocator for intermediates
-//        optimize: build edge-list directly from curves
-//        optimize: rasterize directly from curves?
-//
-// ADDITIONAL CONTRIBUTORS
-//
-//   Mikko Mononen: compound shape support, more cmap formats
-//   Tor Andersson: kerning, subpixel rendering
-//
-//   Misc other:
-//       Ryan Gordon
-//       Simon Glass
-//
-//   Bug/warning reports/fixes:
-//       "Zer" on mollyrocket (with fix)
-//       Cass Everitt
-//       stoiko (Haemimont Games)
-//       Brian Hook
-//       Walter van Niftrik
-//       David Gow
-//       David Given
-//       Ivan-Assen Ivanov
-//       Anthony Pesch
-//       Johan Duparc
-//       Hou Qiming
-//       Fabian "ryg" Giesen
-//       Martins Mozeiko
-//       Cap Petschulat
-//       Omar Cornut
-//       github:aloucks
-//       Peter LaValle
-//       Sergey Popov
-//       Giumo X. Clanjor
-//       Higor Euripedes
-//       Thomas Fields
-//       Derek Vinyard
-//
-// VERSION HISTORY
-//
-//   1.11 (2016-04-02) fix unused-variable warning
-//   1.10 (2016-04-02) user-defined fabs(); rare memory leak; remove duplicate typedef
-//   1.09 (2016-01-16) warning fix; avoid crash on outofmem; use allocation userdata properly
-//   1.08 (2015-09-13) document stbtt_Rasterize(); fixes for vertical & horizontal edges
-//   1.07 (2015-08-01) allow PackFontRanges to accept arrays of sparse codepoints;
-//                     variant PackFontRanges to pack and render in separate phases;
-//                     fix stbtt_GetFontOFfsetForIndex (never worked for non-0 input?);
-//                     fixed an assert() bug in the new rasterizer
-//                     replace assert() with STBTT_assert() in new rasterizer
-//   1.06 (2015-07-14) performance improvements (~35% faster on x86 and x64 on test machine)
-//                     also more precise AA rasterizer, except if shapes overlap
-//                     remove need for STBTT_sort
-//   1.05 (2015-04-15) fix misplaced definitions for STBTT_STATIC
-//   1.04 (2015-04-15) typo in example
-//   1.03 (2015-04-12) STBTT_STATIC, fix memory leak in new packing, various fixes
-//
-//   Full history can be found at the end of this file.
-//
-// LICENSE
-//
-//   This software is dual-licensed to the public domain and under the following
-//   license: you are granted a perpetual, irrevocable license to copy, modify,
-//   publish, and distribute this file as you see fit.
-//
-// USAGE
-//
-//   Include this file in whatever places neeed to refer to it. In ONE C/C++
-//   file, write:
-//      #define STB_TRUETYPE_IMPLEMENTATION
-//   before the #include of this file. This expands out the actual
-//   implementation into that C/C++ file.
-//
-//   To make the implementation private to the file that generates the implementation,
-//      #define STBTT_STATIC
-//
-//   Simple 3D API (don't ship this, but it's fine for tools and quick start)
-//           stbtt_BakeFontBitmap()               -- bake a font to a bitmap for use as texture
-//           stbtt_GetBakedQuad()                 -- compute quad to draw for a given char
-//
-//   Improved 3D API (more shippable):
-//           #include "stb_rect_pack.h"           -- optional, but you really want it
-//           stbtt_PackBegin()
-//           stbtt_PackSetOversample()            -- for improved quality on small fonts
-//           stbtt_PackFontRanges()               -- pack and renders
-//           stbtt_PackEnd()
-//           stbtt_GetPackedQuad()
-//
-//   "Load" a font file from a memory buffer (you have to keep the buffer loaded)
-//           stbtt_InitFont()
-//           stbtt_GetFontOffsetForIndex()        -- use for TTC font collections
-//
-//   Render a unicode codepoint to a bitmap
-//           stbtt_GetCodepointBitmap()           -- allocates and returns a bitmap
-//           stbtt_MakeCodepointBitmap()          -- renders into bitmap you provide
-//           stbtt_GetCodepointBitmapBox()        -- how big the bitmap must be
-//
-//   Character advance/positioning
-//           stbtt_GetCodepointHMetrics()
-//           stbtt_GetFontVMetrics()
-//           stbtt_GetCodepointKernAdvance()
-//
-//   Starting with version 1.06, the rasterizer was replaced with a new,
-//   faster and generally-more-precise rasterizer. The new rasterizer more
-//   accurately measures pixel coverage for anti-aliasing, except in the case
-//   where multiple shapes overlap, in which case it overestimates the AA pixel
-//   coverage. Thus, anti-aliasing of intersecting shapes may look wrong. If
-//   this turns out to be a problem, you can re-enable the old rasterizer with
-//        #define STBTT_RASTERIZER_VERSION 1
-//   which will incur about a 15% speed hit.
-//
-// ADDITIONAL DOCUMENTATION
-//
-//   Immediately after this block comment are a series of sample programs.
-//
-//   After the sample programs is the "header file" section. This section
-//   includes documentation for each API function.
-//
-//   Some important concepts to understand to use this library:
-//
-//      Codepoint
-//         Characters are defined by unicode codepoints, e.g. 65 is
-//         uppercase A, 231 is lowercase c with a cedilla, 0x7e30 is
-//         the hiragana for "ma".
-//
-//      Glyph
-//         A visual character shape (every codepoint is rendered as
-//         some glyph)
-//
-//      Glyph index
-//         A font-specific integer ID representing a glyph
-//
-//      Baseline
-//         Glyph shapes are defined relative to a baseline, which is the
-//         bottom of uppercase characters. Characters extend both above
-//         and below the baseline.
-//
-//      Current Point
-//         As you draw text to the screen, you keep track of a "current point"
-//         which is the origin of each character. The current point's vertical
-//         position is the baseline. Even "baked fonts" use this model.
-//
-//      Vertical Font Metrics
-//         The vertical qualities of the font, used to vertically position
-//         and space the characters. See docs for stbtt_GetFontVMetrics.
-//
-//      Font Size in Pixels or Points
-//         The preferred interface for specifying font sizes in stb_truetype
-//         is to specify how tall the font's vertical extent should be in pixels.
-//         If that sounds good enough, skip the next paragraph.
-//
-//         Most font APIs instead use "points", which are a common typographic
-//         measurement for describing font size, defined as 72 points per inch.
-//         stb_truetype provides a point API for compatibility. However, true
-//         "per inch" conventions don't make much sense on computer displays
-//         since they different monitors have different number of pixels per
-//         inch. For example, Windows traditionally uses a convention that
-//         there are 96 pixels per inch, thus making 'inch' measurements have
-//         nothing to do with inches, and thus effectively defining a point to
-//         be 1.333 pixels. Additionally, the TrueType font data provides
-//         an explicit scale factor to scale a given font's glyphs to points,
-//         but the author has observed that this scale factor is often wrong
-//         for non-commercial fonts, thus making fonts scaled in points
-//         according to the TrueType spec incoherently sized in practice.
-//
-// ADVANCED USAGE
-//
-//   Quality:
-//
-//    - Use the functions with Subpixel at the end to allow your characters
-//      to have subpixel positioning. Since the font is anti-aliased, not
-//      hinted, this is very import for quality. (This is not possible with
-//      baked fonts.)
-//
-//    - Kerning is now supported, and if you're supporting subpixel rendering
-//      then kerning is worth using to give your text a polished look.
-//
-//   Performance:
-//
-//    - Convert Unicode codepoints to glyph indexes and operate on the glyphs;
-//      if you don't do this, stb_truetype is forced to do the conversion on
-//      every call.
-//
-//    - There are a lot of memory allocations. We should modify it to take
-//      a temp buffer and allocate from the temp buffer (without freeing),
-//      should help performance a lot.
-//
-// NOTES
-//
-//   The system uses the raw data found in the .ttf file without changing it
-//   and without building auxiliary data structures. This is a bit inefficient
-//   on little-endian systems (the data is big-endian), but assuming you're
-//   caching the bitmaps or glyph shapes this shouldn't be a big deal.
-//
-//   It appears to be very hard to programmatically determine what font a
-//   given file is in a general way. I provide an API for this, but I don't
-//   recommend it.
-//
-//
-// SOURCE STATISTICS (based on v0.6c, 2050 LOC)
-//
-//   Documentation & header file        520 LOC  \___ 660 LOC documentation
-//   Sample code                        140 LOC  /
-//   Truetype parsing                   620 LOC  ---- 620 LOC TrueType
-//   Software rasterization             240 LOC  \                           .
-//   Curve tesselation                  120 LOC   \__ 550 LOC Bitmap creation
-//   Bitmap management                  100 LOC   /
-//   Baked bitmap interface              70 LOC  /
-//   Font name matching & access        150 LOC  ---- 150
-//   C runtime library abstraction       60 LOC  ----  60
-//
-//
-// PERFORMANCE MEASUREMENTS FOR 1.06:
-//
-//                      32-bit     64-bit
-//   Previous release:  8.83 s     7.68 s
-//   Pool allocations:  7.72 s     6.34 s
-//   Inline sort     :  6.54 s     5.65 s
-//   New rasterizer  :  5.63 s     5.00 s
-
-//////////////////////////////////////////////////////////////////////////////
-//////////////////////////////////////////////////////////////////////////////
-////
-////  SAMPLE PROGRAMS
-////
-//
-//  Incomplete text-in-3d-api example, which draws quads properly aligned to be lossless
-//
-#if 0
-#define STB_TRUETYPE_IMPLEMENTATION  // force following include to generate implementation
-#include "stb_truetype.h"
-
-unsigned char ttf_buffer[1<<20];
-unsigned char temp_bitmap[512*512];
-
-stbtt_bakedchar cdata[96]; // ASCII 32..126 is 95 glyphs
-GLuint ftex;
-
-void my_stbtt_initfont(void)
-{
-   fread(ttf_buffer, 1, 1<<20, fopen("c:/windows/fonts/times.ttf", "rb"));
-   stbtt_BakeFontBitmap(ttf_buffer,0, 32.0, temp_bitmap,512,512, 32,96, cdata); // no guarantee this fits!
-   // can free ttf_buffer at this point
-   glGenTextures(1, &ftex);
-   glBindTexture(GL_TEXTURE_2D, ftex);
-   glTexImage2D(GL_TEXTURE_2D, 0, GL_ALPHA, 512,512, 0, GL_ALPHA, GL_UNSIGNED_BYTE, temp_bitmap);
-   // can free temp_bitmap at this point
-   glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
-}
-
-void my_stbtt_print(float x, float y, char *text)
-{
-   // assume orthographic projection with units = screen pixels, origin at top left
-   glEnable(GL_TEXTURE_2D);
-   glBindTexture(GL_TEXTURE_2D, ftex);
-   glBegin(GL_QUADS);
-   while (*text) {
-      if (*text >= 32 && *text < 128) {
-         stbtt_aligned_quad q;
-         stbtt_GetBakedQuad(cdata, 512,512, *text-32, &x,&y,&q,1);//1=opengl & d3d10+,0=d3d9
-         glTexCoord2f(q.s0,q.t1); glVertex2f(q.x0,q.y0);
-         glTexCoord2f(q.s1,q.t1); glVertex2f(q.x1,q.y0);
-         glTexCoord2f(q.s1,q.t0); glVertex2f(q.x1,q.y1);
-         glTexCoord2f(q.s0,q.t0); glVertex2f(q.x0,q.y1);
-      }
-      ++text;
-   }
-   glEnd();
-}
-#endif
-//
-//
-//////////////////////////////////////////////////////////////////////////////
-//
-// Complete program (this compiles): get a single bitmap, print as ASCII art
-//
-#if 0
-#include <stdio.h>
-#define STB_TRUETYPE_IMPLEMENTATION  // force following include to generate implementation
-#include "stb_truetype.h"
-
-char ttf_buffer[1<<25];
-
-int main(int argc, char **argv)
-{
-   stbtt_fontinfo font;
-   unsigned char *bitmap;
-   int w,h,i,j,c = (argc > 1 ? atoi(argv[1]) : 'a'), s = (argc > 2 ? atoi(argv[2]) : 20);
-
-   fread(ttf_buffer, 1, 1<<25, fopen(argc > 3 ? argv[3] : "c:/windows/fonts/arialbd.ttf", "rb"));
-
-   stbtt_InitFont(&font, ttf_buffer, stbtt_GetFontOffsetForIndex(ttf_buffer,0));
-   bitmap = stbtt_GetCodepointBitmap(&font, 0,stbtt_ScaleForPixelHeight(&font, s), c, &w, &h, 0,0);
-
-   for (j=0; j < h; ++j) {
-      for (i=0; i < w; ++i)
-         putchar(" .:ioVM@"[bitmap[j*w+i]>>5]);
-      putchar('\n');
-   }
-   return 0;
-}
-#endif
-//
-// Output:
-//
-//     .ii.
-//    @@@@@@.
-//   V@Mio@@o
-//   :i.  V@V
-//     :oM@@M
-//   :@@@MM@M
-//   @@o  o@M
-//  :@@.  M@M
-//   @@@o@@@@
-//   :M@@V:@@.
-//
-//////////////////////////////////////////////////////////////////////////////
-//
-// Complete program: print "Hello World!" banner, with bugs
-//
-#if 0
-char buffer[24<<20];
-unsigned char screen[20][79];
-
-int main(int arg, char **argv)
-{
-   stbtt_fontinfo font;
-   int i,j,ascent,baseline,ch=0;
-   float scale, xpos=2; // leave a little padding in case the character extends left
-   char *text = "Heljo World!"; // intentionally misspelled to show 'lj' brokenness
-
-   fread(buffer, 1, 1000000, fopen("c:/windows/fonts/arialbd.ttf", "rb"));
-   stbtt_InitFont(&font, buffer, 0);
-
-   scale = stbtt_ScaleForPixelHeight(&font, 15);
-   stbtt_GetFontVMetrics(&font, &ascent,0,0);
-   baseline = (int) (ascent*scale);
-
-   while (text[ch]) {
-      int advance,lsb,x0,y0,x1,y1;
-      float x_shift = xpos - (float) floor(xpos);
-      stbtt_GetCodepointHMetrics(&font, text[ch], &advance, &lsb);
-      stbtt_GetCodepointBitmapBoxSubpixel(&font, text[ch], scale,scale,x_shift,0, &x0,&y0,&x1,&y1);
-      stbtt_MakeCodepointBitmapSubpixel(&font, &screen[baseline + y0][(int) xpos + x0], x1-x0,y1-y0, 79, scale,scale,x_shift,0, text[ch]);
-      // note that this stomps the old data, so where character boxes overlap (e.g. 'lj') it's wrong
-      // because this API is really for baking character bitmaps into textures. if you want to render
-      // a sequence of characters, you really need to render each bitmap to a temp buffer, then
-      // "alpha blend" that into the working buffer
-      xpos += (advance * scale);
-      if (text[ch+1])
-         xpos += scale*stbtt_GetCodepointKernAdvance(&font, text[ch],text[ch+1]);
-      ++ch;
-   }
-
-   for (j=0; j < 20; ++j) {
-      for (i=0; i < 78; ++i)
-         putchar(" .:ioVM@"[screen[j][i]>>5]);
-      putchar('\n');
-   }
-
-   return 0;
-}
-#endif
-
-
-//////////////////////////////////////////////////////////////////////////////
-//////////////////////////////////////////////////////////////////////////////
-////
-////   INTEGRATION WITH YOUR CODEBASE
-////
-////   The following sections allow you to supply alternate definitions
-////   of C library functions used by stb_truetype.
-
-#ifdef STB_TRUETYPE_IMPLEMENTATION
-   // #define your own (u)stbtt_int8/16/32 before including to override this
-   #ifndef stbtt_uint8
-   typedef unsigned char   stbtt_uint8;
-   typedef signed   char   stbtt_int8;
-   typedef unsigned short  stbtt_uint16;
-   typedef signed   short  stbtt_int16;
-   typedef unsigned int    stbtt_uint32;
-   typedef signed   int    stbtt_int32;
-   #endif
-
-   typedef char stbtt__check_size32[sizeof(stbtt_int32)==4 ? 1 : -1];
-   typedef char stbtt__check_size16[sizeof(stbtt_int16)==2 ? 1 : -1];
-
-   // #define your own STBTT_ifloor/STBTT_iceil() to avoid math.h
-   #ifndef STBTT_ifloor
-   #include <math.h>
-   #define STBTT_ifloor(x)   ((int) floor(x))
-   #define STBTT_iceil(x)    ((int) ceil(x))
-   #endif
-
-   #ifndef STBTT_sqrt
-   #include <math.h>
-   #define STBTT_sqrt(x)      sqrt(x)
-   #endif
-
-   #ifndef STBTT_fabs
-   #include <math.h>
-   #define STBTT_fabs(x)      fabs(x)
-   #endif
-
-   // #define your own functions "STBTT_malloc" / "STBTT_free" to avoid malloc.h
-   #ifndef STBTT_malloc
-   #include <stdlib.h>
-   #define STBTT_malloc(x,u)  ((void)(u),malloc(x))
-   #define STBTT_free(x,u)    ((void)(u),free(x))
-   #endif
-
-   #ifndef STBTT_assert
-   #include <assert.h>
-   #define STBTT_assert(x)    assert(x)
-   #endif
-
-   #ifndef STBTT_strlen
-   #include <string.h>
-   #define STBTT_strlen(x)    strlen(x)
-   #endif
-
-   #ifndef STBTT_memcpy
-   #include <memory.h>
-   #define STBTT_memcpy       memcpy
-   #define STBTT_memset       memset
-   #endif
-#endif
-
-///////////////////////////////////////////////////////////////////////////////
-///////////////////////////////////////////////////////////////////////////////
-////
-////   INTERFACE
-////
-////
-
-#ifndef __STB_INCLUDE_STB_TRUETYPE_H__
-#define __STB_INCLUDE_STB_TRUETYPE_H__
-
-#ifdef STBTT_STATIC
-#define STBTT_DEF static
-#else
-#define STBTT_DEF extern
-#endif
-
-#ifdef __cplusplus
-extern "C" {
-#endif
-
-//////////////////////////////////////////////////////////////////////////////
-//
-// TEXTURE BAKING API
-//
-// If you use this API, you only have to call two functions ever.
-//
-
-typedef struct
-{
-   unsigned short x0,y0,x1,y1; // coordinates of bbox in bitmap
-   float xoff,yoff,xadvance;
-} stbtt_bakedchar;
-
-STBTT_DEF int stbtt_BakeFontBitmap(const unsigned char *data, int offset,  // font location (use offset=0 for plain .ttf)
-                                float pixel_height,                     // height of font in pixels
-                                unsigned char *pixels, int pw, int ph,  // bitmap to be filled in
-                                int first_char, int num_chars,          // characters to bake
-                                stbtt_bakedchar *chardata);             // you allocate this, it's num_chars long
-// if return is positive, the first unused row of the bitmap
-// if return is negative, returns the negative of the number of characters that fit
-// if return is 0, no characters fit and no rows were used
-// This uses a very crappy packing.
-
-typedef struct
-{
-   float x0,y0,s0,t0; // top-left
-   float x1,y1,s1,t1; // bottom-right
-} stbtt_aligned_quad;
-
-STBTT_DEF void stbtt_GetBakedQuad(stbtt_bakedchar *chardata, int pw, int ph,  // same data as above
-                               int char_index,             // character to display
-                               float *xpos, float *ypos,   // pointers to current position in screen pixel space
-                               stbtt_aligned_quad *q,      // output: quad to draw
-                               int opengl_fillrule);       // true if opengl fill rule; false if DX9 or earlier
-// Call GetBakedQuad with char_index = 'character - first_char', and it
-// creates the quad you need to draw and advances the current position.
-//
-// The coordinate system used assumes y increases downwards.
-//
-// Characters will extend both above and below the current position;
-// see discussion of "BASELINE" above.
-//
-// It's inefficient; you might want to c&p it and optimize it.
-
-
-
-//////////////////////////////////////////////////////////////////////////////
-//
-// NEW TEXTURE BAKING API
-//
-// This provides options for packing multiple fonts into one atlas, not
-// perfectly but better than nothing.
-
-typedef struct
-{
-   unsigned short x0,y0,x1,y1; // coordinates of bbox in bitmap
-   float xoff,yoff,xadvance;
-   float xoff2,yoff2;
-} stbtt_packedchar;
-
-typedef struct stbtt_pack_context stbtt_pack_context;
-typedef struct stbtt_fontinfo stbtt_fontinfo;
-#ifndef STB_RECT_PACK_VERSION
-typedef struct stbrp_rect stbrp_rect;
-#endif
-
-STBTT_DEF int  stbtt_PackBegin(stbtt_pack_context *spc, unsigned char *pixels, int width, int height, int stride_in_bytes, int padding, void *alloc_context);
-// Initializes a packing context stored in the passed-in stbtt_pack_context.
-// Future calls using this context will pack characters into the bitmap passed
-// in here: a 1-channel bitmap that is weight x height. stride_in_bytes is
-// the distance from one row to the next (or 0 to mean they are packed tightly
-// together). "padding" is the amount of padding to leave between each
-// character (normally you want '1' for bitmaps you'll use as textures with
-// bilinear filtering).
-//
-// Returns 0 on failure, 1 on success.
-
-STBTT_DEF void stbtt_PackEnd  (stbtt_pack_context *spc);
-// Cleans up the packing context and frees all memory.
-
-#define STBTT_POINT_SIZE(x)   (-(x))
-
-STBTT_DEF int  stbtt_PackFontRange(stbtt_pack_context *spc, unsigned char *fontdata, int font_index, float font_size,
-                                int first_unicode_char_in_range, int num_chars_in_range, stbtt_packedchar *chardata_for_range);
-// Creates character bitmaps from the font_index'th font found in fontdata (use
-// font_index=0 if you don't know what that is). It creates num_chars_in_range
-// bitmaps for characters with unicode values starting at first_unicode_char_in_range
-// and increasing. Data for how to render them is stored in chardata_for_range;
-// pass these to stbtt_GetPackedQuad to get back renderable quads.
-//
-// font_size is the full height of the character from ascender to descender,
-// as computed by stbtt_ScaleForPixelHeight. To use a point size as computed
-// by stbtt_ScaleForMappingEmToPixels, wrap the point size in STBTT_POINT_SIZE()
-// and pass that result as 'font_size':
-//       ...,                  20 , ... // font max minus min y is 20 pixels tall
-//       ..., STBTT_POINT_SIZE(20), ... // 'M' is 20 pixels tall
-
-typedef struct
-{
-   float font_size;
-   int first_unicode_codepoint_in_range;  // if non-zero, then the chars are continuous, and this is the first codepoint
-   int *array_of_unicode_codepoints;       // if non-zero, then this is an array of unicode codepoints
-   int num_chars;
-   stbtt_packedchar *chardata_for_range; // output
-   unsigned char h_oversample, v_oversample; // don't set these, they're used internally
-} stbtt_pack_range;
-
-STBTT_DEF int  stbtt_PackFontRanges(stbtt_pack_context *spc, unsigned char *fontdata, int font_index, stbtt_pack_range *ranges, int num_ranges);
-// Creates character bitmaps from multiple ranges of characters stored in
-// ranges. This will usually create a better-packed bitmap than multiple
-// calls to stbtt_PackFontRange. Note that you can call this multiple
-// times within a single PackBegin/PackEnd.
-
-STBTT_DEF void stbtt_PackSetOversampling(stbtt_pack_context *spc, unsigned int h_oversample, unsigned int v_oversample);
-// Oversampling a font increases the quality by allowing higher-quality subpixel
-// positioning, and is especially valuable at smaller text sizes.
-//
-// This function sets the amount of oversampling for all following calls to
-// stbtt_PackFontRange(s) or stbtt_PackFontRangesGatherRects for a given
-// pack context. The default (no oversampling) is achieved by h_oversample=1
-// and v_oversample=1. The total number of pixels required is
-// h_oversample*v_oversample larger than the default; for example, 2x2
-// oversampling requires 4x the storage of 1x1. For best results, render
-// oversampled textures with bilinear filtering. Look at the readme in
-// stb/tests/oversample for information about oversampled fonts
-//
-// To use with PackFontRangesGather etc., you must set it before calls
-// call to PackFontRangesGatherRects.
-
-STBTT_DEF void stbtt_GetPackedQuad(stbtt_packedchar *chardata, int pw, int ph,  // same data as above
-                               int char_index,             // character to display
-                               float *xpos, float *ypos,   // pointers to current position in screen pixel space
-                               stbtt_aligned_quad *q,      // output: quad to draw
-                               int align_to_integer);
-
-STBTT_DEF int  stbtt_PackFontRangesGatherRects(stbtt_pack_context *spc, stbtt_fontinfo *info, stbtt_pack_range *ranges, int num_ranges, stbrp_rect *rects);
-STBTT_DEF void stbtt_PackFontRangesPackRects(stbtt_pack_context *spc, stbrp_rect *rects, int num_rects);
-STBTT_DEF int  stbtt_PackFontRangesRenderIntoRects(stbtt_pack_context *spc, stbtt_fontinfo *info, stbtt_pack_range *ranges, int num_ranges, stbrp_rect *rects);
-// Calling these functions in sequence is roughly equivalent to calling
-// stbtt_PackFontRanges(). If you more control over the packing of multiple
-// fonts, or if you want to pack custom data into a font texture, take a look
-// at the source to of stbtt_PackFontRanges() and create a custom version
-// using these functions, e.g. call GatherRects multiple times,
-// building up a single array of rects, then call PackRects once,
-// then call RenderIntoRects repeatedly. This may result in a
-// better packing than calling PackFontRanges multiple times
-// (or it may not).
-
-// this is an opaque structure that you shouldn't mess with which holds
-// all the context needed from PackBegin to PackEnd.
-struct stbtt_pack_context {
-   void *user_allocator_context;
-   void *pack_info;
-   int   width;
-   int   height;
-   int   stride_in_bytes;
-   int   padding;
-   unsigned int   h_oversample, v_oversample;
-   unsigned char *pixels;
-   void  *nodes;
-};
-
-//////////////////////////////////////////////////////////////////////////////
-//
-// FONT LOADING
-//
-//
-
-STBTT_DEF int stbtt_GetFontOffsetForIndex(const unsigned char *data, int index);
-// Each .ttf/.ttc file may have more than one font. Each font has a sequential
-// index number starting from 0. Call this function to get the font offset for
-// a given index; it returns -1 if the index is out of range. A regular .ttf
-// file will only define one font and it always be at offset 0, so it will
-// return '0' for index 0, and -1 for all other indices. You can just skip
-// this step if you know it's that kind of font.
-
-
-// The following structure is defined publically so you can declare one on
-// the stack or as a global or etc, but you should treat it as opaque.
-struct stbtt_fontinfo
-{
-   void           * userdata;
-   unsigned char  * data;              // pointer to .ttf file
-   int              fontstart;         // offset of start of font
-
-   int numGlyphs;                     // number of glyphs, needed for range checking
-
-   int loca,head,glyf,hhea,hmtx,kern; // table locations as offset from start of .ttf
-   int index_map;                     // a cmap mapping for our chosen character encoding
-   int indexToLocFormat;              // format needed to map from glyph index to glyph
-};
-
-STBTT_DEF int stbtt_InitFont(stbtt_fontinfo *info, const unsigned char *data, int offset);
-// Given an offset into the file that defines a font, this function builds
-// the necessary cached info for the rest of the system. You must allocate
-// the stbtt_fontinfo yourself, and stbtt_InitFont will fill it out. You don't
-// need to do anything special to free it, because the contents are pure
-// value data with no additional data structures. Returns 0 on failure.
-
-
-//////////////////////////////////////////////////////////////////////////////
-//
-// CHARACTER TO GLYPH-INDEX CONVERSIOn
-
-STBTT_DEF int stbtt_FindGlyphIndex(const stbtt_fontinfo *info, int unicode_codepoint);
-// If you're going to perform multiple operations on the same character
-// and you want a speed-up, call this function with the character you're
-// going to process, then use glyph-based functions instead of the
-// codepoint-based functions.
-
-
-//////////////////////////////////////////////////////////////////////////////
-//
-// CHARACTER PROPERTIES
-//
-
-STBTT_DEF float stbtt_ScaleForPixelHeight(const stbtt_fontinfo *info, float pixels);
-// computes a scale factor to produce a font whose "height" is 'pixels' tall.
-// Height is measured as the distance from the highest ascender to the lowest
-// descender; in other words, it's equivalent to calling stbtt_GetFontVMetrics
-// and computing:
-//       scale = pixels / (ascent - descent)
-// so if you prefer to measure height by the ascent only, use a similar calculation.
-
-STBTT_DEF float stbtt_ScaleForMappingEmToPixels(const stbtt_fontinfo *info, float pixels);
-// computes a scale factor to produce a font whose EM size is mapped to
-// 'pixels' tall. This is probably what traditional APIs compute, but
-// I'm not positive.
-
-STBTT_DEF void stbtt_GetFontVMetrics(const stbtt_fontinfo *info, int *ascent, int *descent, int *lineGap);
-// ascent is the coordinate above the baseline the font extends; descent
-// is the coordinate below the baseline the font extends (i.e. it is typically negative)
-// lineGap is the spacing between one row's descent and the next row's ascent...
-// so you should advance the vertical position by "*ascent - *descent + *lineGap"
-//   these are expressed in unscaled coordinates, so you must multiply by
-//   the scale factor for a given size
-
-STBTT_DEF void stbtt_GetFontBoundingBox(const stbtt_fontinfo *info, int *x0, int *y0, int *x1, int *y1);
-// the bounding box around all possible characters
-
-STBTT_DEF void stbtt_GetCodepointHMetrics(const stbtt_fontinfo *info, int codepoint, int *advanceWidth, int *leftSideBearing);
-// leftSideBearing is the offset from the current horizontal position to the left edge of the character
-// advanceWidth is the offset from the current horizontal position to the next horizontal position
-//   these are expressed in unscaled coordinates
-
-STBTT_DEF int  stbtt_GetCodepointKernAdvance(const stbtt_fontinfo *info, int ch1, int ch2);
-// an additional amount to add to the 'advance' value between ch1 and ch2
-
-STBTT_DEF int stbtt_GetCodepointBox(const stbtt_fontinfo *info, int codepoint, int *x0, int *y0, int *x1, int *y1);
-// Gets the bounding box of the visible part of the glyph, in unscaled coordinates
-
-STBTT_DEF void stbtt_GetGlyphHMetrics(const stbtt_fontinfo *info, int glyph_index, int *advanceWidth, int *leftSideBearing);
-STBTT_DEF int  stbtt_GetGlyphKernAdvance(const stbtt_fontinfo *info, int glyph1, int glyph2);
-STBTT_DEF int  stbtt_GetGlyphBox(const stbtt_fontinfo *info, int glyph_index, int *x0, int *y0, int *x1, int *y1);
-// as above, but takes one or more glyph indices for greater efficiency
-
-
-//////////////////////////////////////////////////////////////////////////////
-//
-// GLYPH SHAPES (you probably don't need these, but they have to go before
-// the bitmaps for C declaration-order reasons)
-//
-
-#ifndef STBTT_vmove // you can predefine these to use different values (but why?)
-   enum {
-      STBTT_vmove=1,
-      STBTT_vline,
-      STBTT_vcurve
-   };
-#endif
-
-#ifndef stbtt_vertex // you can predefine this to use different values
-                   // (we share this with other code at RAD)
-   #define stbtt_vertex_type short // can't use stbtt_int16 because that's not visible in the header file
-   typedef struct
-   {
-      stbtt_vertex_type x,y,cx,cy;
-      unsigned char type,padding;
-   } stbtt_vertex;
-#endif
-
-STBTT_DEF int stbtt_IsGlyphEmpty(const stbtt_fontinfo *info, int glyph_index);
-// returns non-zero if nothing is drawn for this glyph
-
-STBTT_DEF int stbtt_GetCodepointShape(const stbtt_fontinfo *info, int unicode_codepoint, stbtt_vertex **vertices);
-STBTT_DEF int stbtt_GetGlyphShape(const stbtt_fontinfo *info, int glyph_index, stbtt_vertex **vertices);
-// returns # of vertices and fills *vertices with the pointer to them
-//   these are expressed in "unscaled" coordinates
-//
-// The shape is a series of countours. Each one starts with
-// a STBTT_moveto, then consists of a series of mixed
-// STBTT_lineto and STBTT_curveto segments. A lineto
-// draws a line from previous endpoint to its x,y; a curveto
-// draws a quadratic bezier from previous endpoint to
-// its x,y, using cx,cy as the bezier control point.
-
-STBTT_DEF void stbtt_FreeShape(const stbtt_fontinfo *info, stbtt_vertex *vertices);
-// frees the data allocated above
-
-//////////////////////////////////////////////////////////////////////////////
-//
-// BITMAP RENDERING
-//
-
-STBTT_DEF void stbtt_FreeBitmap(unsigned char *bitmap, void *userdata);
-// frees the bitmap allocated below
-
-STBTT_DEF unsigned char *stbtt_GetCodepointBitmap(const stbtt_fontinfo *info, float scale_x, float scale_y, int codepoint, int *width, int *height, int *xoff, int *yoff);
-// allocates a large-enough single-channel 8bpp bitmap and renders the
-// specified character/glyph at the specified scale into it, with
-// antialiasing. 0 is no coverage (transparent), 255 is fully covered (opaque).
-// *width & *height are filled out with the width & height of the bitmap,
-// which is stored left-to-right, top-to-bottom.
-//
-// xoff/yoff are the offset it pixel space from the glyph origin to the top-left of the bitmap
-
-STBTT_DEF unsigned char *stbtt_GetCodepointBitmapSubpixel(const stbtt_fontinfo *info, float scale_x, float scale_y, float shift_x, float shift_y, int codepoint, int *width, int *height, int *xoff, int *yoff);
-// the same as stbtt_GetCodepoitnBitmap, but you can specify a subpixel
-// shift for the character
-
-STBTT_DEF void stbtt_MakeCodepointBitmap(const stbtt_fontinfo *info, unsigned char *output, int out_w, int out_h, int out_stride, float scale_x, float scale_y, int codepoint);
-// the same as stbtt_GetCodepointBitmap, but you pass in storage for the bitmap
-// in the form of 'output', with row spacing of 'out_stride' bytes. the bitmap
-// is clipped to out_w/out_h bytes. Call stbtt_GetCodepointBitmapBox to get the
-// width and height and positioning info for it first.
-
-STBTT_DEF void stbtt_MakeCodepointBitmapSubpixel(const stbtt_fontinfo *info, unsigned char *output, int out_w, int out_h, int out_stride, float scale_x, float scale_y, float shift_x, float shift_y, int codepoint);
-// same as stbtt_MakeCodepointBitmap, but you can specify a subpixel
-// shift for the character
-
-STBTT_DEF void stbtt_GetCodepointBitmapBox(const stbtt_fontinfo *font, int codepoint, float scale_x, float scale_y, int *ix0, int *iy0, int *ix1, int *iy1);
-// get the bbox of the bitmap centered around the glyph origin; so the
-// bitmap width is ix1-ix0, height is iy1-iy0, and location to place
-// the bitmap top left is (leftSideBearing*scale,iy0).
-// (Note that the bitmap uses y-increases-down, but the shape uses
-// y-increases-up, so CodepointBitmapBox and CodepointBox are inverted.)
-
-STBTT_DEF void stbtt_GetCodepointBitmapBoxSubpixel(const stbtt_fontinfo *font, int codepoint, float scale_x, float scale_y, float shift_x, float shift_y, int *ix0, int *iy0, int *ix1, int *iy1);
-// same as stbtt_GetCodepointBitmapBox, but you can specify a subpixel
-// shift for the character
-
-// the following functions are equivalent to the above functions, but operate
-// on glyph indices instead of Unicode codepoints (for efficiency)
-STBTT_DEF unsigned char *stbtt_GetGlyphBitmap(const stbtt_fontinfo *info, float scale_x, float scale_y, int glyph, int *width, int *height, int *xoff, int *yoff);
-STBTT_DEF unsigned char *stbtt_GetGlyphBitmapSubpixel(const stbtt_fontinfo *info, float scale_x, float scale_y, float shift_x, float shift_y, int glyph, int *width, int *height, int *xoff, int *yoff);
-STBTT_DEF void stbtt_MakeGlyphBitmap(const stbtt_fontinfo *info, unsigned char *output, int out_w, int out_h, int out_stride, float scale_x, float scale_y, int glyph);
-STBTT_DEF void stbtt_MakeGlyphBitmapSubpixel(const stbtt_fontinfo *info, unsigned char *output, int out_w, int out_h, int out_stride, float scale_x, float scale_y, float shift_x, float shift_y, int glyph);
-STBTT_DEF void stbtt_GetGlyphBitmapBox(const stbtt_fontinfo *font, int glyph, float scale_x, float scale_y, int *ix0, int *iy0, int *ix1, int *iy1);
-STBTT_DEF void stbtt_GetGlyphBitmapBoxSubpixel(const stbtt_fontinfo *font, int glyph, float scale_x, float scale_y,float shift_x, float shift_y, int *ix0, int *iy0, int *ix1, int *iy1);
-
-
-// @TODO: don't expose this structure
-typedef struct
-{
-   int w,h,stride;
-   unsigned char *pixels;
-} stbtt__bitmap;
-
-// rasterize a shape with quadratic beziers into a bitmap
-STBTT_DEF void stbtt_Rasterize(stbtt__bitmap *result,        // 1-channel bitmap to draw into
-                               float flatness_in_pixels,     // allowable error of curve in pixels
-                               stbtt_vertex *vertices,       // array of vertices defining shape
-                               int num_verts,                // number of vertices in above array
-                               float scale_x, float scale_y, // scale applied to input vertices
-                               float shift_x, float shift_y, // translation applied to input vertices
-                               int x_off, int y_off,         // another translation applied to input
-                               int invert,                   // if non-zero, vertically flip shape
-                               void *userdata);              // context for to STBTT_MALLOC
-
-//////////////////////////////////////////////////////////////////////////////
-//
-// Finding the right font...
-//
-// You should really just solve this offline, keep your own tables
-// of what font is what, and don't try to get it out of the .ttf file.
-// That's because getting it out of the .ttf file is really hard, because
-// the names in the file can appear in many possible encodings, in many
-// possible languages, and e.g. if you need a case-insensitive comparison,
-// the details of that depend on the encoding & language in a complex way
-// (actually underspecified in truetype, but also gigantic).
-//
-// But you can use the provided functions in two possible ways:
-//     stbtt_FindMatchingFont() will use *case-sensitive* comparisons on
-//             unicode-encoded names to try to find the font you want;
-//             you can run this before calling stbtt_InitFont()
-//
-//     stbtt_GetFontNameString() lets you get any of the various strings
-//             from the file yourself and do your own comparisons on them.
-//             You have to have called stbtt_InitFont() first.
-
-
-STBTT_DEF int stbtt_FindMatchingFont(const unsigned char *fontdata, const char *name, int flags);
-// returns the offset (not index) of the font that matches, or -1 if none
-//   if you use STBTT_MACSTYLE_DONTCARE, use a font name like "Arial Bold".
-//   if you use any other flag, use a font name like "Arial"; this checks
-//     the 'macStyle' header field; i don't know if fonts set this consistently
-#define STBTT_MACSTYLE_DONTCARE     0
-#define STBTT_MACSTYLE_BOLD         1
-#define STBTT_MACSTYLE_ITALIC       2
-#define STBTT_MACSTYLE_UNDERSCORE   4
-#define STBTT_MACSTYLE_NONE         8   // <= not same as 0, this makes us check the bitfield is 0
-
-STBTT_DEF int stbtt_CompareUTF8toUTF16_bigendian(const char *s1, int len1, const char *s2, int len2);
-// returns 1/0 whether the first string interpreted as utf8 is identical to
-// the second string interpreted as big-endian utf16... useful for strings from next func
-
-STBTT_DEF const char *stbtt_GetFontNameString(const stbtt_fontinfo *font, int *length, int platformID, int encodingID, int languageID, int nameID);
-// returns the string (which may be big-endian double byte, e.g. for unicode)
-// and puts the length in bytes in *length.
-//
-// some of the values for the IDs are below; for more see the truetype spec:
-//     http://developer.apple.com/textfonts/TTRefMan/RM06/Chap6name.html
-//     http://www.microsoft.com/typography/otspec/name.htm
-
-enum { // platformID
-   STBTT_PLATFORM_ID_UNICODE   =0,
-   STBTT_PLATFORM_ID_MAC       =1,
-   STBTT_PLATFORM_ID_ISO       =2,
-   STBTT_PLATFORM_ID_MICROSOFT =3
-};
-
-enum { // encodingID for STBTT_PLATFORM_ID_UNICODE
-   STBTT_UNICODE_EID_UNICODE_1_0    =0,
-   STBTT_UNICODE_EID_UNICODE_1_1    =1,
-   STBTT_UNICODE_EID_ISO_10646      =2,
-   STBTT_UNICODE_EID_UNICODE_2_0_BMP=3,
-   STBTT_UNICODE_EID_UNICODE_2_0_FULL=4
-};
-
-enum { // encodingID for STBTT_PLATFORM_ID_MICROSOFT
-   STBTT_MS_EID_SYMBOL        =0,
-   STBTT_MS_EID_UNICODE_BMP   =1,
-   STBTT_MS_EID_SHIFTJIS      =2,
-   STBTT_MS_EID_UNICODE_FULL  =10
-};
-
-enum { // encodingID for STBTT_PLATFORM_ID_MAC; same as Script Manager codes
-   STBTT_MAC_EID_ROMAN        =0,   STBTT_MAC_EID_ARABIC       =4,
-   STBTT_MAC_EID_JAPANESE     =1,   STBTT_MAC_EID_HEBREW       =5,
-   STBTT_MAC_EID_CHINESE_TRAD =2,   STBTT_MAC_EID_GREEK        =6,
-   STBTT_MAC_EID_KOREAN       =3,   STBTT_MAC_EID_RUSSIAN      =7
-};
-
-enum { // languageID for STBTT_PLATFORM_ID_MICROSOFT; same as LCID...
-       // problematic because there are e.g. 16 english LCIDs and 16 arabic LCIDs
-   STBTT_MS_LANG_ENGLISH     =0x0409,   STBTT_MS_LANG_ITALIAN     =0x0410,
-   STBTT_MS_LANG_CHINESE     =0x0804,   STBTT_MS_LANG_JAPANESE    =0x0411,
-   STBTT_MS_LANG_DUTCH       =0x0413,   STBTT_MS_LANG_KOREAN      =0x0412,
-   STBTT_MS_LANG_FRENCH      =0x040c,   STBTT_MS_LANG_RUSSIAN     =0x0419,
-   STBTT_MS_LANG_GERMAN      =0x0407,   STBTT_MS_LANG_SPANISH     =0x0409,
-   STBTT_MS_LANG_HEBREW      =0x040d,   STBTT_MS_LANG_SWEDISH     =0x041D
-};
-
-enum { // languageID for STBTT_PLATFORM_ID_MAC
-   STBTT_MAC_LANG_ENGLISH      =0 ,   STBTT_MAC_LANG_JAPANESE     =11,
-   STBTT_MAC_LANG_ARABIC       =12,   STBTT_MAC_LANG_KOREAN       =23,
-   STBTT_MAC_LANG_DUTCH        =4 ,   STBTT_MAC_LANG_RUSSIAN      =32,
-   STBTT_MAC_LANG_FRENCH       =1 ,   STBTT_MAC_LANG_SPANISH      =6 ,
-   STBTT_MAC_LANG_GERMAN       =2 ,   STBTT_MAC_LANG_SWEDISH      =5 ,
-   STBTT_MAC_LANG_HEBREW       =10,   STBTT_MAC_LANG_CHINESE_SIMPLIFIED =33,
-   STBTT_MAC_LANG_ITALIAN      =3 ,   STBTT_MAC_LANG_CHINESE_TRAD =19
-};
-
-#ifdef __cplusplus
-}
-#endif
-
-#endif // __STB_INCLUDE_STB_TRUETYPE_H__
-
-///////////////////////////////////////////////////////////////////////////////
-///////////////////////////////////////////////////////////////////////////////
-////
-////   IMPLEMENTATION
-////
-////
-
-#ifdef STB_TRUETYPE_IMPLEMENTATION
-
-#ifndef STBTT_MAX_OVERSAMPLE
-#define STBTT_MAX_OVERSAMPLE   8
-#endif
-
-#if STBTT_MAX_OVERSAMPLE > 255
-#error "STBTT_MAX_OVERSAMPLE cannot be > 255"
-#endif
-
-typedef int stbtt__test_oversample_pow2[(STBTT_MAX_OVERSAMPLE & (STBTT_MAX_OVERSAMPLE-1)) == 0 ? 1 : -1];
-
-#ifndef STBTT_RASTERIZER_VERSION
-#define STBTT_RASTERIZER_VERSION 2
-#endif
-
-#ifdef _MSC_VER
-#define STBTT__NOTUSED(v)  (void)(v)
-#else
-#define STBTT__NOTUSED(v)  (void)sizeof(v)
-#endif
-
-//////////////////////////////////////////////////////////////////////////
-//
-// accessors to parse data from file
-//
-
-// on platforms that don't allow misaligned reads, if we want to allow
-// truetype fonts that aren't padded to alignment, define ALLOW_UNALIGNED_TRUETYPE
-
-#define ttBYTE(p)     (* (stbtt_uint8 *) (p))
-#define ttCHAR(p)     (* (stbtt_int8 *) (p))
-#define ttFixed(p)    ttLONG(p)
-
-#if defined(STB_TRUETYPE_BIGENDIAN) && !defined(ALLOW_UNALIGNED_TRUETYPE)
-
-   #define ttUSHORT(p)   (* (stbtt_uint16 *) (p))
-   #define ttSHORT(p)    (* (stbtt_int16 *) (p))
-   #define ttULONG(p)    (* (stbtt_uint32 *) (p))
-   #define ttLONG(p)     (* (stbtt_int32 *) (p))
-
-#else
-
-   static stbtt_uint16 ttUSHORT(const stbtt_uint8 *p) { return p[0]*256 + p[1]; }
-   static stbtt_int16 ttSHORT(const stbtt_uint8 *p)   { return p[0]*256 + p[1]; }
-   static stbtt_uint32 ttULONG(const stbtt_uint8 *p)  { return (p[0]<<24) + (p[1]<<16) + (p[2]<<8) + p[3]; }
-   static stbtt_int32 ttLONG(const stbtt_uint8 *p)    { return (p[0]<<24) + (p[1]<<16) + (p[2]<<8) + p[3]; }
-
-#endif
-
-#define stbtt_tag4(p,c0,c1,c2,c3) ((p)[0] == (c0) && (p)[1] == (c1) && (p)[2] == (c2) && (p)[3] == (c3))
-#define stbtt_tag(p,str)           stbtt_tag4(p,str[0],str[1],str[2],str[3])
-
-static int stbtt__isfont(const stbtt_uint8 *font)
-{
-   // check the version number
-   if (stbtt_tag4(font, '1',0,0,0))  return 1; // TrueType 1
-   if (stbtt_tag(font, "typ1"))   return 1; // TrueType with type 1 font -- we don't support this!
-   if (stbtt_tag(font, "OTTO"))   return 1; // OpenType with CFF
-   if (stbtt_tag4(font, 0,1,0,0)) return 1; // OpenType 1.0
-   return 0;
-}
-
-// @OPTIMIZE: binary search
-static stbtt_uint32 stbtt__find_table(stbtt_uint8 *data, stbtt_uint32 fontstart, const char *tag)
-{
-   stbtt_int32 num_tables = ttUSHORT(data+fontstart+4);
-   stbtt_uint32 tabledir = fontstart + 12;
-   stbtt_int32 i;
-   for (i=0; i < num_tables; ++i) {
-      stbtt_uint32 loc = tabledir + 16*i;
-      if (stbtt_tag(data+loc+0, tag))
-         return ttULONG(data+loc+8);
-   }
-   return 0;
-}
-
-STBTT_DEF int stbtt_GetFontOffsetForIndex(const unsigned char *font_collection, int index)
-{
-   // if it's just a font, there's only one valid index
-   if (stbtt__isfont(font_collection))
-      return index == 0 ? 0 : -1;
-
-   // check if it's a TTC
-   if (stbtt_tag(font_collection, "ttcf")) {
-      // version 1?
-      if (ttULONG(font_collection+4) == 0x00010000 || ttULONG(font_collection+4) == 0x00020000) {
-         stbtt_int32 n = ttLONG(font_collection+8);
-         if (index >= n)
-            return -1;
-         return ttULONG(font_collection+12+index*4);
-      }
-   }
-   return -1;
-}
-
-STBTT_DEF int stbtt_InitFont(stbtt_fontinfo *info, const unsigned char *data2, int fontstart)
-{
-   stbtt_uint8 *data = (stbtt_uint8 *) data2;
-   stbtt_uint32 cmap, t;
-   stbtt_int32 i,numTables;
-
-   info->data = data;
-   info->fontstart = fontstart;
-
-   cmap = stbtt__find_table(data, fontstart, "cmap");       // required
-   info->loca = stbtt__find_table(data, fontstart, "loca"); // required
-   info->head = stbtt__find_table(data, fontstart, "head"); // required
-   info->glyf = stbtt__find_table(data, fontstart, "glyf"); // required
-   info->hhea = stbtt__find_table(data, fontstart, "hhea"); // required
-   info->hmtx = stbtt__find_table(data, fontstart, "hmtx"); // required
-   info->kern = stbtt__find_table(data, fontstart, "kern"); // not required
-   if (!cmap || !info->loca || !info->head || !info->glyf || !info->hhea || !info->hmtx)
-      return 0;
-
-   t = stbtt__find_table(data, fontstart, "maxp");
-   if (t)
-      info->numGlyphs = ttUSHORT(data+t+4);
-   else
-      info->numGlyphs = 0xffff;
-
-   // find a cmap encoding table we understand *now* to avoid searching
-   // later. (todo: could make this installable)
-   // the same regardless of glyph.
-   numTables = ttUSHORT(data + cmap + 2);
-   info->index_map = 0;
-   for (i=0; i < numTables; ++i) {
-      stbtt_uint32 encoding_record = cmap + 4 + 8 * i;
-      // find an encoding we understand:
-      switch(ttUSHORT(data+encoding_record)) {
-         case STBTT_PLATFORM_ID_MICROSOFT:
-            switch (ttUSHORT(data+encoding_record+2)) {
-               case STBTT_MS_EID_UNICODE_BMP:
-               case STBTT_MS_EID_UNICODE_FULL:
-                  // MS/Unicode
-                  info->index_map = cmap + ttULONG(data+encoding_record+4);
-                  break;
-            }
-            break;
-        case STBTT_PLATFORM_ID_UNICODE:
-            // Mac/iOS has these
-            // all the encodingIDs are unicode, so we don't bother to check it
-            info->index_map = cmap + ttULONG(data+encoding_record+4);
-            break;
-      }
-   }
-   if (info->index_map == 0)
-      return 0;
-
-   info->indexToLocFormat = ttUSHORT(data+info->head + 50);
-   return 1;
-}
-
-STBTT_DEF int stbtt_FindGlyphIndex(const stbtt_fontinfo *info, int unicode_codepoint)
-{
-   stbtt_uint8 *data = info->data;
-   stbtt_uint32 index_map = info->index_map;
-
-   stbtt_uint16 format = ttUSHORT(data + index_map + 0);
-   if (format == 0) { // apple byte encoding
-      stbtt_int32 bytes = ttUSHORT(data + index_map + 2);
-      if (unicode_codepoint < bytes-6)
-         return ttBYTE(data + index_map + 6 + unicode_codepoint);
-      return 0;
-   } else if (format == 6) {
-      stbtt_uint32 first = ttUSHORT(data + index_map + 6);
-      stbtt_uint32 count = ttUSHORT(data + index_map + 8);
-      if ((stbtt_uint32) unicode_codepoint >= first && (stbtt_uint32) unicode_codepoint < first+count)
-         return ttUSHORT(data + index_map + 10 + (unicode_codepoint - first)*2);
-      return 0;
-   } else if (format == 2) {
-      STBTT_assert(0); // @TODO: high-byte mapping for japanese/chinese/korean
-      return 0;
-   } else if (format == 4) { // standard mapping for windows fonts: binary search collection of ranges
-      stbtt_uint16 segcount = ttUSHORT(data+index_map+6) >> 1;
-      stbtt_uint16 searchRange = ttUSHORT(data+index_map+8) >> 1;
-      stbtt_uint16 entrySelector = ttUSHORT(data+index_map+10);
-      stbtt_uint16 rangeShift = ttUSHORT(data+index_map+12) >> 1;
-
-      // do a binary search of the segments
-      stbtt_uint32 endCount = index_map + 14;
-      stbtt_uint32 search = endCount;
-
-      if (unicode_codepoint > 0xffff)
-         return 0;
-
-      // they lie from endCount .. endCount + segCount
-      // but searchRange is the nearest power of two, so...
-      if (unicode_codepoint >= ttUSHORT(data + search + rangeShift*2))
-         search += rangeShift*2;
-
-      // now decrement to bias correctly to find smallest
-      search -= 2;
-      while (entrySelector) {
-         stbtt_uint16 end;
-         searchRange >>= 1;
-         end = ttUSHORT(data + search + searchRange*2);
-         if (unicode_codepoint > end)
-            search += searchRange*2;
-         --entrySelector;
-      }
-      search += 2;
-
-      {
-         stbtt_uint16 offset, start;
-         stbtt_uint16 item = (stbtt_uint16) ((search - endCount) >> 1);
-
-         STBTT_assert(unicode_codepoint <= ttUSHORT(data + endCount + 2*item));
-         start = ttUSHORT(data + index_map + 14 + segcount*2 + 2 + 2*item);
-         if (unicode_codepoint < start)
-            return 0;
-
-         offset = ttUSHORT(data + index_map + 14 + segcount*6 + 2 + 2*item);
-         if (offset == 0)
-            return (stbtt_uint16) (unicode_codepoint + ttSHORT(data + index_map + 14 + segcount*4 + 2 + 2*item));
-
-         return ttUSHORT(data + offset + (unicode_codepoint-start)*2 + index_map + 14 + segcount*6 + 2 + 2*item);
-      }
-   } else if (format == 12 || format == 13) {
-      stbtt_uint32 ngroups = ttULONG(data+index_map+12);
-      stbtt_int32 low,high;
-      low = 0; high = (stbtt_int32)ngroups;
-      // Binary search the right group.
-      while (low < high) {
-         stbtt_int32 mid = low + ((high-low) >> 1); // rounds down, so low <= mid < high
-         stbtt_uint32 start_char = ttULONG(data+index_map+16+mid*12);
-         stbtt_uint32 end_char = ttULONG(data+index_map+16+mid*12+4);
-         if ((stbtt_uint32) unicode_codepoint < start_char)
-            high = mid;
-         else if ((stbtt_uint32) unicode_codepoint > end_char)
-            low = mid+1;
-         else {
-            stbtt_uint32 start_glyph = ttULONG(data+index_map+16+mid*12+8);
-            if (format == 12)
-               return start_glyph + unicode_codepoint-start_char;
-            else // format == 13
-               return start_glyph;
-         }
-      }
-      return 0; // not found
-   }
-   // @TODO
-   STBTT_assert(0);
-   return 0;
-}
-
-STBTT_DEF int stbtt_GetCodepointShape(const stbtt_fontinfo *info, int unicode_codepoint, stbtt_vertex **vertices)
-{
-   return stbtt_GetGlyphShape(info, stbtt_FindGlyphIndex(info, unicode_codepoint), vertices);
-}
-
-static void stbtt_setvertex(stbtt_vertex *v, stbtt_uint8 type, stbtt_int32 x, stbtt_int32 y, stbtt_int32 cx, stbtt_int32 cy)
-{
-   v->type = type;
-   v->x = (stbtt_int16) x;
-   v->y = (stbtt_int16) y;
-   v->cx = (stbtt_int16) cx;
-   v->cy = (stbtt_int16) cy;
-}
-
-static int stbtt__GetGlyfOffset(const stbtt_fontinfo *info, int glyph_index)
-{
-   int g1,g2;
-
-   if (glyph_index >= info->numGlyphs) return -1; // glyph index out of range
-   if (info->indexToLocFormat >= 2)    return -1; // unknown index->glyph map format
-
-   if (info->indexToLocFormat == 0) {
-      g1 = info->glyf + ttUSHORT(info->data + info->loca + glyph_index * 2) * 2;
-      g2 = info->glyf + ttUSHORT(info->data + info->loca + glyph_index * 2 + 2) * 2;
-   } else {
-      g1 = info->glyf + ttULONG (info->data + info->loca + glyph_index * 4);
-      g2 = info->glyf + ttULONG (info->data + info->loca + glyph_index * 4 + 4);
-   }
-
-   return g1==g2 ? -1 : g1; // if length is 0, return -1
-}
-
-STBTT_DEF int stbtt_GetGlyphBox(const stbtt_fontinfo *info, int glyph_index, int *x0, int *y0, int *x1, int *y1)
-{
-   int g = stbtt__GetGlyfOffset(info, glyph_index);
-   if (g < 0) return 0;
-
-   if (x0) *x0 = ttSHORT(info->data + g + 2);
-   if (y0) *y0 = ttSHORT(info->data + g + 4);
-   if (x1) *x1 = ttSHORT(info->data + g + 6);
-   if (y1) *y1 = ttSHORT(info->data + g + 8);
-   return 1;
-}
-
-STBTT_DEF int stbtt_GetCodepointBox(const stbtt_fontinfo *info, int codepoint, int *x0, int *y0, int *x1, int *y1)
-{
-   return stbtt_GetGlyphBox(info, stbtt_FindGlyphIndex(info,codepoint), x0,y0,x1,y1);
-}
-
-STBTT_DEF int stbtt_IsGlyphEmpty(const stbtt_fontinfo *info, int glyph_index)
-{
-   stbtt_int16 numberOfContours;
-   int g = stbtt__GetGlyfOffset(info, glyph_index);
-   if (g < 0) return 1;
-   numberOfContours = ttSHORT(info->data + g);
-   return numberOfContours == 0;
-}
-
-static int stbtt__close_shape(stbtt_vertex *vertices, int num_vertices, int was_off, int start_off,
-    stbtt_int32 sx, stbtt_int32 sy, stbtt_int32 scx, stbtt_int32 scy, stbtt_int32 cx, stbtt_int32 cy)
-{
-   if (start_off) {
-      if (was_off)
-         stbtt_setvertex(&vertices[num_vertices++], STBTT_vcurve, (cx+scx)>>1, (cy+scy)>>1, cx,cy);
-      stbtt_setvertex(&vertices[num_vertices++], STBTT_vcurve, sx,sy,scx,scy);
-   } else {
-      if (was_off)
-         stbtt_setvertex(&vertices[num_vertices++], STBTT_vcurve,sx,sy,cx,cy);
-      else
-         stbtt_setvertex(&vertices[num_vertices++], STBTT_vline,sx,sy,0,0);
-   }
-   return num_vertices;
-}
-
-STBTT_DEF int stbtt_GetGlyphShape(const stbtt_fontinfo *info, int glyph_index, stbtt_vertex **pvertices)
-{
-   stbtt_int16 numberOfContours;
-   stbtt_uint8 *endPtsOfContours;
-   stbtt_uint8 *data = info->data;
-   stbtt_vertex *vertices=0;
-   int num_vertices=0;
-   int g = stbtt__GetGlyfOffset(info, glyph_index);
-
-   *pvertices = NULL;
-
-   if (g < 0) return 0;
-
-   numberOfContours = ttSHORT(data + g);
-
-   if (numberOfContours > 0) {
-      stbtt_uint8 flags=0,flagcount;
-      stbtt_int32 ins, i,j=0,m,n, next_move, was_off=0, off, start_off=0;
-      stbtt_int32 x,y,cx,cy,sx,sy, scx,scy;
-      stbtt_uint8 *points;
-      endPtsOfContours = (data + g + 10);
-      ins = ttUSHORT(data + g + 10 + numberOfContours * 2);
-      points = data + g + 10 + numberOfContours * 2 + 2 + ins;
-
-      n = 1+ttUSHORT(endPtsOfContours + numberOfContours*2-2);
-
-      m = n + 2*numberOfContours;  // a loose bound on how many vertices we might need
-      vertices = (stbtt_vertex *) STBTT_malloc(m * sizeof(vertices[0]), info->userdata);
-      if (vertices == 0)
-         return 0;
-
-      next_move = 0;
-      flagcount=0;
-
-      // in first pass, we load uninterpreted data into the allocated array
-      // above, shifted to the end of the array so we won't overwrite it when
-      // we create our final data starting from the front
-
-      off = m - n; // starting offset for uninterpreted data, regardless of how m ends up being calculated
-
-      // first load flags
-
-      for (i=0; i < n; ++i) {
-         if (flagcount == 0) {
-            flags = *points++;
-            if (flags & 8)
-               flagcount = *points++;
-         } else
-            --flagcount;
-         vertices[off+i].type = flags;
-      }
-
-      // now load x coordinates
-      x=0;
-      for (i=0; i < n; ++i) {
-         flags = vertices[off+i].type;
-         if (flags & 2) {
-            stbtt_int16 dx = *points++;
-            x += (flags & 16) ? dx : -dx; // ???
-         } else {
-            if (!(flags & 16)) {
-               x = x + (stbtt_int16) (points[0]*256 + points[1]);
-               points += 2;
-            }
-         }
-         vertices[off+i].x = (stbtt_int16) x;
-      }
-
-      // now load y coordinates
-      y=0;
-      for (i=0; i < n; ++i) {
-         flags = vertices[off+i].type;
-         if (flags & 4) {
-            stbtt_int16 dy = *points++;
-            y += (flags & 32) ? dy : -dy; // ???
-         } else {
-            if (!(flags & 32)) {
-               y = y + (stbtt_int16) (points[0]*256 + points[1]);
-               points += 2;
-            }
-         }
-         vertices[off+i].y = (stbtt_int16) y;
-      }
-
-      // now convert them to our format
-      num_vertices=0;
-      sx = sy = cx = cy = scx = scy = 0;
-      for (i=0; i < n; ++i) {
-         flags = vertices[off+i].type;
-         x     = (stbtt_int16) vertices[off+i].x;
-         y     = (stbtt_int16) vertices[off+i].y;
-
-         if (next_move == i) {
-            if (i != 0)
-               num_vertices = stbtt__close_shape(vertices, num_vertices, was_off, start_off, sx,sy,scx,scy,cx,cy);
-
-            // now start the new one
-            start_off = !(flags & 1);
-            if (start_off) {
-               // if we start off with an off-curve point, then when we need to find a point on the curve
-               // where we can start, and we need to save some state for when we wraparound.
-               scx = x;
-               scy = y;
-               if (!(vertices[off+i+1].type & 1)) {
-                  // next point is also a curve point, so interpolate an on-point curve
-                  sx = (x + (stbtt_int32) vertices[off+i+1].x) >> 1;
-                  sy = (y + (stbtt_int32) vertices[off+i+1].y) >> 1;
-               } else {
-                  // otherwise just use the next point as our start point
-                  sx = (stbtt_int32) vertices[off+i+1].x;
-                  sy = (stbtt_int32) vertices[off+i+1].y;
-                  ++i; // we're using point i+1 as the starting point, so skip it
-               }
-            } else {
-               sx = x;
-               sy = y;
-            }
-            stbtt_setvertex(&vertices[num_vertices++], STBTT_vmove,sx,sy,0,0);
-            was_off = 0;
-            next_move = 1 + ttUSHORT(endPtsOfContours+j*2);
-            ++j;
-         } else {
-            if (!(flags & 1)) { // if it's a curve
-               if (was_off) // two off-curve control points in a row means interpolate an on-curve midpoint
-                  stbtt_setvertex(&vertices[num_vertices++], STBTT_vcurve, (cx+x)>>1, (cy+y)>>1, cx, cy);
-               cx = x;
-               cy = y;
-               was_off = 1;
-            } else {
-               if (was_off)
-                  stbtt_setvertex(&vertices[num_vertices++], STBTT_vcurve, x,y, cx, cy);
-               else
-                  stbtt_setvertex(&vertices[num_vertices++], STBTT_vline, x,y,0,0);
-               was_off = 0;
-            }
-         }
-      }
-      num_vertices = stbtt__close_shape(vertices, num_vertices, was_off, start_off, sx,sy,scx,scy,cx,cy);
-   } else if (numberOfContours == -1) {
-      // Compound shapes.
-      int more = 1;
-      stbtt_uint8 *comp = data + g + 10;
-      num_vertices = 0;
-      vertices = 0;
-      while (more) {
-         stbtt_uint16 flags, gidx;
-         int comp_num_verts = 0, i;
-         stbtt_vertex *comp_verts = 0, *tmp = 0;
-         float mtx[6] = {1,0,0,1,0,0}, m, n;
-
-         flags = ttSHORT(comp); comp+=2;
-         gidx = ttSHORT(comp); comp+=2;
-
-         if (flags & 2) { // XY values
-            if (flags & 1) { // shorts
-               mtx[4] = ttSHORT(comp); comp+=2;
-               mtx[5] = ttSHORT(comp); comp+=2;
-            } else {
-               mtx[4] = ttCHAR(comp); comp+=1;
-               mtx[5] = ttCHAR(comp); comp+=1;
-            }
-         }
-         else {
-            // @TODO handle matching point
-            STBTT_assert(0);
-         }
-         if (flags & (1<<3)) { // WE_HAVE_A_SCALE
-            mtx[0] = mtx[3] = ttSHORT(comp)/16384.0f; comp+=2;
-            mtx[1] = mtx[2] = 0;
-         } else if (flags & (1<<6)) { // WE_HAVE_AN_X_AND_YSCALE
-            mtx[0] = ttSHORT(comp)/16384.0f; comp+=2;
-            mtx[1] = mtx[2] = 0;
-            mtx[3] = ttSHORT(comp)/16384.0f; comp+=2;
-         } else if (flags & (1<<7)) { // WE_HAVE_A_TWO_BY_TWO
-            mtx[0] = ttSHORT(comp)/16384.0f; comp+=2;
-            mtx[1] = ttSHORT(comp)/16384.0f; comp+=2;
-            mtx[2] = ttSHORT(comp)/16384.0f; comp+=2;
-            mtx[3] = ttSHORT(comp)/16384.0f; comp+=2;
-         }
-
-         // Find transformation scales.
-         m = (float) STBTT_sqrt(mtx[0]*mtx[0] + mtx[1]*mtx[1]);
-         n = (float) STBTT_sqrt(mtx[2]*mtx[2] + mtx[3]*mtx[3]);
-
-         // Get indexed glyph.
-         comp_num_verts = stbtt_GetGlyphShape(info, gidx, &comp_verts);
-         if (comp_num_verts > 0) {
-            // Transform vertices.
-            for (i = 0; i < comp_num_verts; ++i) {
-               stbtt_vertex* v = &comp_verts[i];
-               stbtt_vertex_type x,y;
-               x=v->x; y=v->y;
-               v->x = (stbtt_vertex_type)(m * (mtx[0]*x + mtx[2]*y + mtx[4]));
-               v->y = (stbtt_vertex_type)(n * (mtx[1]*x + mtx[3]*y + mtx[5]));
-               x=v->cx; y=v->cy;
-               v->cx = (stbtt_vertex_type)(m * (mtx[0]*x + mtx[2]*y + mtx[4]));
-               v->cy = (stbtt_vertex_type)(n * (mtx[1]*x + mtx[3]*y + mtx[5]));
-            }
-            // Append vertices.
-            tmp = (stbtt_vertex*)STBTT_malloc((num_vertices+comp_num_verts)*sizeof(stbtt_vertex), info->userdata);
-            if (!tmp) {
-               if (vertices) STBTT_free(vertices, info->userdata);
-               if (comp_verts) STBTT_free(comp_verts, info->userdata);
-               return 0;
-            }
-            if (num_vertices > 0) STBTT_memcpy(tmp, vertices, num_vertices*sizeof(stbtt_vertex));
-            STBTT_memcpy(tmp+num_vertices, comp_verts, comp_num_verts*sizeof(stbtt_vertex));
-            if (vertices) STBTT_free(vertices, info->userdata);
-            vertices = tmp;
-            STBTT_free(comp_verts, info->userdata);
-            num_vertices += comp_num_verts;
-         }
-         // More components ?
-         more = flags & (1<<5);
-      }
-   } else if (numberOfContours < 0) {
-      // @TODO other compound variations?
-      STBTT_assert(0);
-   } else {
-      // numberOfCounters == 0, do nothing
-   }
-
-   *pvertices = vertices;
-   return num_vertices;
-}
-
-STBTT_DEF void stbtt_GetGlyphHMetrics(const stbtt_fontinfo *info, int glyph_index, int *advanceWidth, int *leftSideBearing)
-{
-   stbtt_uint16 numOfLongHorMetrics = ttUSHORT(info->data+info->hhea + 34);
-   if (glyph_index < numOfLongHorMetrics) {
-      if (advanceWidth)     *advanceWidth    = ttSHORT(info->data + info->hmtx + 4*glyph_index);
-      if (leftSideBearing)  *leftSideBearing = ttSHORT(info->data + info->hmtx + 4*glyph_index + 2);
-   } else {
-      if (advanceWidth)     *advanceWidth    = ttSHORT(info->data + info->hmtx + 4*(numOfLongHorMetrics-1));
-      if (leftSideBearing)  *leftSideBearing = ttSHORT(info->data + info->hmtx + 4*numOfLongHorMetrics + 2*(glyph_index - numOfLongHorMetrics));
-   }
-}
-
-STBTT_DEF int  stbtt_GetGlyphKernAdvance(const stbtt_fontinfo *info, int glyph1, int glyph2)
-{
-   stbtt_uint8 *data = info->data + info->kern;
-   stbtt_uint32 needle, straw;
-   int l, r, m;
-
-   // we only look at the first table. it must be 'horizontal' and format 0.
-   if (!info->kern)
-      return 0;
-   if (ttUSHORT(data+2) < 1) // number of tables, need at least 1
-      return 0;
-   if (ttUSHORT(data+8) != 1) // horizontal flag must be set in format
-      return 0;
-
-   l = 0;
-   r = ttUSHORT(data+10) - 1;
-   needle = glyph1 << 16 | glyph2;
-   while (l <= r) {
-      m = (l + r) >> 1;
-      straw = ttULONG(data+18+(m*6)); // note: unaligned read
-      if (needle < straw)
-         r = m - 1;
-      else if (needle > straw)
-         l = m + 1;
-      else
-         return ttSHORT(data+22+(m*6));
-   }
-   return 0;
-}
-
-STBTT_DEF int  stbtt_GetCodepointKernAdvance(const stbtt_fontinfo *info, int ch1, int ch2)
-{
-   if (!info->kern) // if no kerning table, don't waste time looking up both codepoint->glyphs
-      return 0;
-   return stbtt_GetGlyphKernAdvance(info, stbtt_FindGlyphIndex(info,ch1), stbtt_FindGlyphIndex(info,ch2));
-}
-
-STBTT_DEF void stbtt_GetCodepointHMetrics(const stbtt_fontinfo *info, int codepoint, int *advanceWidth, int *leftSideBearing)
-{
-   stbtt_GetGlyphHMetrics(info, stbtt_FindGlyphIndex(info,codepoint), advanceWidth, leftSideBearing);
-}
-
-STBTT_DEF void stbtt_GetFontVMetrics(const stbtt_fontinfo *info, int *ascent, int *descent, int *lineGap)
-{
-   if (ascent ) *ascent  = ttSHORT(info->data+info->hhea + 4);
-   if (descent) *descent = ttSHORT(info->data+info->hhea + 6);
-   if (lineGap) *lineGap = ttSHORT(info->data+info->hhea + 8);
-}
-
-STBTT_DEF void stbtt_GetFontBoundingBox(const stbtt_fontinfo *info, int *x0, int *y0, int *x1, int *y1)
-{
-   *x0 = ttSHORT(info->data + info->head + 36);
-   *y0 = ttSHORT(info->data + info->head + 38);
-   *x1 = ttSHORT(info->data + info->head + 40);
-   *y1 = ttSHORT(info->data + info->head + 42);
-}
-
-STBTT_DEF float stbtt_ScaleForPixelHeight(const stbtt_fontinfo *info, float height)
-{
-   int fheight = ttSHORT(info->data + info->hhea + 4) - ttSHORT(info->data + info->hhea + 6);
-   return (float) height / fheight;
-}
-
-STBTT_DEF float stbtt_ScaleForMappingEmToPixels(const stbtt_fontinfo *info, float pixels)
-{
-   int unitsPerEm = ttUSHORT(info->data + info->head + 18);
-   return pixels / unitsPerEm;
-}
-
-STBTT_DEF void stbtt_FreeShape(const stbtt_fontinfo *info, stbtt_vertex *v)
-{
-   STBTT_free(v, info->userdata);
-}
-
-//////////////////////////////////////////////////////////////////////////////
-//
-// antialiasing software rasterizer
-//
-
-STBTT_DEF void stbtt_GetGlyphBitmapBoxSubpixel(const stbtt_fontinfo *font, int glyph, float scale_x, float scale_y,float shift_x, float shift_y, int *ix0, int *iy0, int *ix1, int *iy1)
-{
-   int x0=0,y0=0,x1,y1; // =0 suppresses compiler warning
-   if (!stbtt_GetGlyphBox(font, glyph, &x0,&y0,&x1,&y1)) {
-      // e.g. space character
-      if (ix0) *ix0 = 0;
-      if (iy0) *iy0 = 0;
-      if (ix1) *ix1 = 0;
-      if (iy1) *iy1 = 0;
-   } else {
-      // move to integral bboxes (treating pixels as little squares, what pixels get touched)?
-      if (ix0) *ix0 = STBTT_ifloor( x0 * scale_x + shift_x);
-      if (iy0) *iy0 = STBTT_ifloor(-y1 * scale_y + shift_y);
-      if (ix1) *ix1 = STBTT_iceil ( x1 * scale_x + shift_x);
-      if (iy1) *iy1 = STBTT_iceil (-y0 * scale_y + shift_y);
-   }
-}
-
-STBTT_DEF void stbtt_GetGlyphBitmapBox(const stbtt_fontinfo *font, int glyph, float scale_x, float scale_y, int *ix0, int *iy0, int *ix1, int *iy1)
-{
-   stbtt_GetGlyphBitmapBoxSubpixel(font, glyph, scale_x, scale_y,0.0f,0.0f, ix0, iy0, ix1, iy1);
-}
-
-STBTT_DEF void stbtt_GetCodepointBitmapBoxSubpixel(const stbtt_fontinfo *font, int codepoint, float scale_x, float scale_y, float shift_x, float shift_y, int *ix0, int *iy0, int *ix1, int *iy1)
-{
-   stbtt_GetGlyphBitmapBoxSubpixel(font, stbtt_FindGlyphIndex(font,codepoint), scale_x, scale_y,shift_x,shift_y, ix0,iy0,ix1,iy1);
-}
-
-STBTT_DEF void stbtt_GetCodepointBitmapBox(const stbtt_fontinfo *font, int codepoint, float scale_x, float scale_y, int *ix0, int *iy0, int *ix1, int *iy1)
-{
-   stbtt_GetCodepointBitmapBoxSubpixel(font, codepoint, scale_x, scale_y,0.0f,0.0f, ix0,iy0,ix1,iy1);
-}
-
-//////////////////////////////////////////////////////////////////////////////
-//
-//  Rasterizer
-
-typedef struct stbtt__hheap_chunk
-{
-   struct stbtt__hheap_chunk *next;
-} stbtt__hheap_chunk;
-
-typedef struct stbtt__hheap
-{
-   struct stbtt__hheap_chunk *head;
-   void   *first_free;
-   int    num_remaining_in_head_chunk;
-} stbtt__hheap;
-
-static void *stbtt__hheap_alloc(stbtt__hheap *hh, size_t size, void *userdata)
-{
-   if (hh->first_free) {
-      void *p = hh->first_free;
-      hh->first_free = * (void **) p;
-      return p;
-   } else {
-      if (hh->num_remaining_in_head_chunk == 0) {
-         int count = (size < 32 ? 2000 : size < 128 ? 800 : 100);
-         stbtt__hheap_chunk *c = (stbtt__hheap_chunk *) STBTT_malloc(sizeof(stbtt__hheap_chunk) + size * count, userdata);
-         if (c == NULL)
-            return NULL;
-         c->next = hh->head;
-         hh->head = c;
-         hh->num_remaining_in_head_chunk = count;
-      }
-      --hh->num_remaining_in_head_chunk;
-      return (char *) (hh->head) + size * hh->num_remaining_in_head_chunk;
-   }
-}
-
-static void stbtt__hheap_free(stbtt__hheap *hh, void *p)
-{
-   *(void **) p = hh->first_free;
-   hh->first_free = p;
-}
-
-static void stbtt__hheap_cleanup(stbtt__hheap *hh, void *userdata)
-{
-   stbtt__hheap_chunk *c = hh->head;
-   while (c) {
-      stbtt__hheap_chunk *n = c->next;
-      STBTT_free(c, userdata);
-      c = n;
-   }
-}
-
-typedef struct stbtt__edge {
-   float x0,y0, x1,y1;
-   int invert;
-} stbtt__edge;
-
-
-typedef struct stbtt__active_edge
-{
-   struct stbtt__active_edge *next;
-   #if STBTT_RASTERIZER_VERSION==1
-   int x,dx;
-   float ey;
-   int direction;
-   #elif STBTT_RASTERIZER_VERSION==2
-   float fx,fdx,fdy;
-   float direction;
-   float sy;
-   float ey;
-   #else
-   #error "Unrecognized value of STBTT_RASTERIZER_VERSION"
-   #endif
-} stbtt__active_edge;
-
-#if STBTT_RASTERIZER_VERSION == 1
-#define STBTT_FIXSHIFT   10
-#define STBTT_FIX        (1 << STBTT_FIXSHIFT)
-#define STBTT_FIXMASK    (STBTT_FIX-1)
-
-static stbtt__active_edge *stbtt__new_active(stbtt__hheap *hh, stbtt__edge *e, int off_x, float start_point, void *userdata)
-{
-   stbtt__active_edge *z = (stbtt__active_edge *) stbtt__hheap_alloc(hh, sizeof(*z), userdata);
-   float dxdy = (e->x1 - e->x0) / (e->y1 - e->y0);
-   STBTT_assert(z != NULL);
-   if (!z) return z;
-
-   // round dx down to avoid overshooting
-   if (dxdy < 0)
-      z->dx = -STBTT_ifloor(STBTT_FIX * -dxdy);
-   else
-      z->dx = STBTT_ifloor(STBTT_FIX * dxdy);
-
-   z->x = STBTT_ifloor(STBTT_FIX * e->x0 + z->dx * (start_point - e->y0)); // use z->dx so when we offset later it's by the same amount
-   z->x -= off_x * STBTT_FIX;
-
-   z->ey = e->y1;
-   z->next = 0;
-   z->direction = e->invert ? 1 : -1;
-   return z;
-}
-#elif STBTT_RASTERIZER_VERSION == 2
-static stbtt__active_edge *stbtt__new_active(stbtt__hheap *hh, stbtt__edge *e, int off_x, float start_point, void *userdata)
-{
-   stbtt__active_edge *z = (stbtt__active_edge *) stbtt__hheap_alloc(hh, sizeof(*z), userdata);
-   float dxdy = (e->x1 - e->x0) / (e->y1 - e->y0);
-   STBTT_assert(z != NULL);
-   //STBTT_assert(e->y0 <= start_point);
-   if (!z) return z;
-   z->fdx = dxdy;
-   z->fdy = dxdy != 0.0f ? (1.0f/dxdy) : 0.0f;
-   z->fx = e->x0 + dxdy * (start_point - e->y0);
-   z->fx -= off_x;
-   z->direction = e->invert ? 1.0f : -1.0f;
-   z->sy = e->y0;
-   z->ey = e->y1;
-   z->next = 0;
-   return z;
-}
-#else
-#error "Unrecognized value of STBTT_RASTERIZER_VERSION"
-#endif
-
-#if STBTT_RASTERIZER_VERSION == 1
-// note: this routine clips fills that extend off the edges... ideally this
-// wouldn't happen, but it could happen if the truetype glyph bounding boxes
-// are wrong, or if the user supplies a too-small bitmap
-static void stbtt__fill_active_edges(unsigned char *scanline, int len, stbtt__active_edge *e, int max_weight)
-{
-   // non-zero winding fill
-   int x0=0, w=0;
-
-   while (e) {
-      if (w == 0) {
-         // if we're currently at zero, we need to record the edge start point
-         x0 = e->x; w += e->direction;
-      } else {
-         int x1 = e->x; w += e->direction;
-         // if we went to zero, we need to draw
-         if (w == 0) {
-            int i = x0 >> STBTT_FIXSHIFT;
-            int j = x1 >> STBTT_FIXSHIFT;
-
-            if (i < len && j >= 0) {
-               if (i == j) {
-                  // x0,x1 are the same pixel, so compute combined coverage
-                  scanline[i] = scanline[i] + (stbtt_uint8) ((x1 - x0) * max_weight >> STBTT_FIXSHIFT);
-               } else {
-                  if (i >= 0) // add antialiasing for x0
-                     scanline[i] = scanline[i] + (stbtt_uint8) (((STBTT_FIX - (x0 & STBTT_FIXMASK)) * max_weight) >> STBTT_FIXSHIFT);
-                  else
-                     i = -1; // clip
-
-                  if (j < len) // add antialiasing for x1
-                     scanline[j] = scanline[j] + (stbtt_uint8) (((x1 & STBTT_FIXMASK) * max_weight) >> STBTT_FIXSHIFT);
-                  else
-                     j = len; // clip
-
-                  for (++i; i < j; ++i) // fill pixels between x0 and x1
-                     scanline[i] = scanline[i] + (stbtt_uint8) max_weight;
-               }
-            }
-         }
-      }
-
-      e = e->next;
-   }
-}
-
-static void stbtt__rasterize_sorted_edges(stbtt__bitmap *result, stbtt__edge *e, int n, int vsubsample, int off_x, int off_y, void *userdata)
-{
-   stbtt__hheap hh = { 0, 0, 0 };
-   stbtt__active_edge *active = NULL;
-   int y,j=0;
-   int max_weight = (255 / vsubsample);  // weight per vertical scanline
-   int s; // vertical subsample index
-   unsigned char scanline_data[512], *scanline;
-
-   if (result->w > 512)
-      scanline = (unsigned char *) STBTT_malloc(result->w, userdata);
-   else
-      scanline = scanline_data;
-
-   y = off_y * vsubsample;
-   e[n].y0 = (off_y + result->h) * (float) vsubsample + 1;
-
-   while (j < result->h) {
-      STBTT_memset(scanline, 0, result->w);
-      for (s=0; s < vsubsample; ++s) {
-         // find center of pixel for this scanline
-         float scan_y = y + 0.5f;
-         stbtt__active_edge **step = &active;
-
-         // update all active edges;
-         // remove all active edges that terminate before the center of this scanline
-         while (*step) {
-            stbtt__active_edge * z = *step;
-            if (z->ey <= scan_y) {
-               *step = z->next; // delete from list
-               STBTT_assert(z->direction);
-               z->direction = 0;
-               stbtt__hheap_free(&hh, z);
-            } else {
-               z->x += z->dx; // advance to position for current scanline
-               step = &((*step)->next); // advance through list
-            }
-         }
-
-         // resort the list if needed
-         for(;;) {
-            int changed=0;
-            step = &active;
-            while (*step && (*step)->next) {
-               if ((*step)->x > (*step)->next->x) {
-                  stbtt__active_edge *t = *step;
-                  stbtt__active_edge *q = t->next;
-
-                  t->next = q->next;
-                  q->next = t;
-                  *step = q;
-                  changed = 1;
-               }
-               step = &(*step)->next;
-            }
-            if (!changed) break;
-         }
-
-         // insert all edges that start before the center of this scanline -- omit ones that also end on this scanline
-         while (e->y0 <= scan_y) {
-            if (e->y1 > scan_y) {
-               stbtt__active_edge *z = stbtt__new_active(&hh, e, off_x, scan_y, userdata);
-               if (z != NULL) {
-                  // find insertion point
-                  if (active == NULL)
-                     active = z;
-                  else if (z->x < active->x) {
-                     // insert at front
-                     z->next = active;
-                     active = z;
-                  } else {
-                     // find thing to insert AFTER
-                     stbtt__active_edge *p = active;
-                     while (p->next && p->next->x < z->x)
-                        p = p->next;
-                     // at this point, p->next->x is NOT < z->x
-                     z->next = p->next;
-                     p->next = z;
-                  }
-               }
-            }
-            ++e;
-         }
-
-         // now process all active edges in XOR fashion
-         if (active)
-            stbtt__fill_active_edges(scanline, result->w, active, max_weight);
-
-         ++y;
-      }
-      STBTT_memcpy(result->pixels + j * result->stride, scanline, result->w);
-      ++j;
-   }
-
-   stbtt__hheap_cleanup(&hh, userdata);
-
-   if (scanline != scanline_data)
-      STBTT_free(scanline, userdata);
-}
-
-#elif STBTT_RASTERIZER_VERSION == 2
-
-// the edge passed in here does not cross the vertical line at x or the vertical line at x+1
-// (i.e. it has already been clipped to those)
-static void stbtt__handle_clipped_edge(float *scanline, int x, stbtt__active_edge *e, float x0, float y0, float x1, float y1)
-{
-   if (y0 == y1) return;
-   STBTT_assert(y0 < y1);
-   STBTT_assert(e->sy <= e->ey);
-   if (y0 > e->ey) return;
-   if (y1 < e->sy) return;
-   if (y0 < e->sy) {
-      x0 += (x1-x0) * (e->sy - y0) / (y1-y0);
-      y0 = e->sy;
-   }
-   if (y1 > e->ey) {
-      x1 += (x1-x0) * (e->ey - y1) / (y1-y0);
-      y1 = e->ey;
-   }
-
-   if (x0 == x)
-      STBTT_assert(x1 <= x+1);
-   else if (x0 == x+1)
-      STBTT_assert(x1 >= x);
-   else if (x0 <= x)
-      STBTT_assert(x1 <= x);
-   else if (x0 >= x+1)
-      STBTT_assert(x1 >= x+1);
-   else
-      STBTT_assert(x1 >= x && x1 <= x+1);
-
-   if (x0 <= x && x1 <= x)
-      scanline[x] += e->direction * (y1-y0);
-   else if (x0 >= x+1 && x1 >= x+1)
-      ;
-   else {
-      STBTT_assert(x0 >= x && x0 <= x+1 && x1 >= x && x1 <= x+1);
-      scanline[x] += e->direction * (y1-y0) * (1-((x0-x)+(x1-x))/2); // coverage = 1 - average x position
-   }
-}
-
-static void stbtt__fill_active_edges_new(float *scanline, float *scanline_fill, int len, stbtt__active_edge *e, float y_top)
-{
-   float y_bottom = y_top+1;
-
-   while (e) {
-      // brute force every pixel
-
-      // compute intersection points with top & bottom
-      STBTT_assert(e->ey >= y_top);
-
-      if (e->fdx == 0) {
-         float x0 = e->fx;
-         if (x0 < len) {
-            if (x0 >= 0) {
-               stbtt__handle_clipped_edge(scanline,(int) x0,e, x0,y_top, x0,y_bottom);
-               stbtt__handle_clipped_edge(scanline_fill-1,(int) x0+1,e, x0,y_top, x0,y_bottom);
-            } else {
-               stbtt__handle_clipped_edge(scanline_fill-1,0,e, x0,y_top, x0,y_bottom);
-            }
-         }
-      } else {
-         float x0 = e->fx;
-         float dx = e->fdx;
-         float xb = x0 + dx;
-         float x_top, x_bottom;
-         float sy0,sy1;
-         float dy = e->fdy;
-         STBTT_assert(e->sy <= y_bottom && e->ey >= y_top);
-
-         // compute endpoints of line segment clipped to this scanline (if the
-         // line segment starts on this scanline. x0 is the intersection of the
-         // line with y_top, but that may be off the line segment.
-         if (e->sy > y_top) {
-            x_top = x0 + dx * (e->sy - y_top);
-            sy0 = e->sy;
-         } else {
-            x_top = x0;
-            sy0 = y_top;
-         }
-         if (e->ey < y_bottom) {
-            x_bottom = x0 + dx * (e->ey - y_top);
-            sy1 = e->ey;
-         } else {
-            x_bottom = xb;
-            sy1 = y_bottom;
-         }
-
-         if (x_top >= 0 && x_bottom >= 0 && x_top < len && x_bottom < len) {
-            // from here on, we don't have to range check x values
-
-            if ((int) x_top == (int) x_bottom) {
-               float height;
-               // simple case, only spans one pixel
-               int x = (int) x_top;
-               height = sy1 - sy0;
-               STBTT_assert(x >= 0 && x < len);
-               scanline[x] += e->direction * (1-((x_top - x) + (x_bottom-x))/2)  * height;
-               scanline_fill[x] += e->direction * height; // everything right of this pixel is filled
-            } else {
-               int x,x1,x2;
-               float y_crossing, step, sign, area;
-               // covers 2+ pixels
-               if (x_top > x_bottom) {
-                  // flip scanline vertically; signed area is the same
-                  float t;
-                  sy0 = y_bottom - (sy0 - y_top);
-                  sy1 = y_bottom - (sy1 - y_top);
-                  t = sy0, sy0 = sy1, sy1 = t;
-                  t = x_bottom, x_bottom = x_top, x_top = t;
-                  dx = -dx;
-                  dy = -dy;
-                  t = x0, x0 = xb, xb = t;
-               }
-
-               x1 = (int) x_top;
-               x2 = (int) x_bottom;
-               // compute intersection with y axis at x1+1
-               y_crossing = (x1+1 - x0) * dy + y_top;
-
-               sign = e->direction;
-               // area of the rectangle covered from y0..y_crossing
-               area = sign * (y_crossing-sy0);
-               // area of the triangle (x_top,y0), (x+1,y0), (x+1,y_crossing)
-               scanline[x1] += area * (1-((x_top - x1)+(x1+1-x1))/2);
-
-               step = sign * dy;
-               for (x = x1+1; x < x2; ++x) {
-                  scanline[x] += area + step/2;
-                  area += step;
-               }
-               y_crossing += dy * (x2 - (x1+1));
-
-               STBTT_assert(STBTT_fabs(area) <= 1.01f);
-
-               scanline[x2] += area + sign * (1-((x2-x2)+(x_bottom-x2))/2) * (sy1-y_crossing);
-
-               scanline_fill[x2] += sign * (sy1-sy0);
-            }
-         } else {
-            // if edge goes outside of box we're drawing, we require
-            // clipping logic. since this does not match the intended use
-            // of this library, we use a different, very slow brute
-            // force implementation
-            int x;
-            for (x=0; x < len; ++x) {
-               // cases:
-               //
-               // there can be up to two intersections with the pixel. any intersection
-               // with left or right edges can be handled by splitting into two (or three)
-               // regions. intersections with top & bottom do not necessitate case-wise logic.
-               //
-               // the old way of doing this found the intersections with the left & right edges,
-               // then used some simple logic to produce up to three segments in sorted order
-               // from top-to-bottom. however, this had a problem: if an x edge was epsilon
-               // across the x border, then the corresponding y position might not be distinct
-               // from the other y segment, and it might ignored as an empty segment. to avoid
-               // that, we need to explicitly produce segments based on x positions.
-
-               // rename variables to clear pairs
-               float y0 = y_top;
-               float x1 = (float) (x);
-               float x2 = (float) (x+1);
-               float x3 = xb;
-               float y3 = y_bottom;
-               float y1,y2;
-
-               // x = e->x + e->dx * (y-y_top)
-               // (y-y_top) = (x - e->x) / e->dx
-               // y = (x - e->x) / e->dx + y_top
-               y1 = (x - x0) / dx + y_top;
-               y2 = (x+1 - x0) / dx + y_top;
-
-               if (x0 < x1 && x3 > x2) {         // three segments descending down-right
-                  stbtt__handle_clipped_edge(scanline,x,e, x0,y0, x1,y1);
-                  stbtt__handle_clipped_edge(scanline,x,e, x1,y1, x2,y2);
-                  stbtt__handle_clipped_edge(scanline,x,e, x2,y2, x3,y3);
-               } else if (x3 < x1 && x0 > x2) {  // three segments descending down-left
-                  stbtt__handle_clipped_edge(scanline,x,e, x0,y0, x2,y2);
-                  stbtt__handle_clipped_edge(scanline,x,e, x2,y2, x1,y1);
-                  stbtt__handle_clipped_edge(scanline,x,e, x1,y1, x3,y3);
-               } else if (x0 < x1 && x3 > x1) {  // two segments across x, down-right
-                  stbtt__handle_clipped_edge(scanline,x,e, x0,y0, x1,y1);
-                  stbtt__handle_clipped_edge(scanline,x,e, x1,y1, x3,y3);
-               } else if (x3 < x1 && x0 > x1) {  // two segments across x, down-left
-                  stbtt__handle_clipped_edge(scanline,x,e, x0,y0, x1,y1);
-                  stbtt__handle_clipped_edge(scanline,x,e, x1,y1, x3,y3);
-               } else if (x0 < x2 && x3 > x2) {  // two segments across x+1, down-right
-                  stbtt__handle_clipped_edge(scanline,x,e, x0,y0, x2,y2);
-                  stbtt__handle_clipped_edge(scanline,x,e, x2,y2, x3,y3);
-               } else if (x3 < x2 && x0 > x2) {  // two segments across x+1, down-left
-                  stbtt__handle_clipped_edge(scanline,x,e, x0,y0, x2,y2);
-                  stbtt__handle_clipped_edge(scanline,x,e, x2,y2, x3,y3);
-               } else {  // one segment
-                  stbtt__handle_clipped_edge(scanline,x,e, x0,y0, x3,y3);
-               }
-            }
-         }
-      }
-      e = e->next;
-   }
-}
-
-// directly AA rasterize edges w/o supersampling
-static void stbtt__rasterize_sorted_edges(stbtt__bitmap *result, stbtt__edge *e, int n, int vsubsample, int off_x, int off_y, void *userdata)
-{
-   stbtt__hheap hh = { 0, 0, 0 };
-   stbtt__active_edge *active = NULL;
-   int y,j=0, i;
-   float scanline_data[129], *scanline, *scanline2;
-
-   STBTT__NOTUSED(vsubsample);
-
-   if (result->w > 64)
-      scanline = (float *) STBTT_malloc((result->w*2+1) * sizeof(float), userdata);
-   else
-      scanline = scanline_data;
-
-   scanline2 = scanline + result->w;
-
-   y = off_y;
-   e[n].y0 = (float) (off_y + result->h) + 1;
-
-   while (j < result->h) {
-      // find center of pixel for this scanline
-      float scan_y_top    = y + 0.0f;
-      float scan_y_bottom = y + 1.0f;
-      stbtt__active_edge **step = &active;
-
-      STBTT_memset(scanline , 0, result->w*sizeof(scanline[0]));
-      STBTT_memset(scanline2, 0, (result->w+1)*sizeof(scanline[0]));
-
-      // update all active edges;
-      // remove all active edges that terminate before the top of this scanline
-      while (*step) {
-         stbtt__active_edge * z = *step;
-         if (z->ey <= scan_y_top) {
-            *step = z->next; // delete from list
-            STBTT_assert(z->direction);
-            z->direction = 0;
-            stbtt__hheap_free(&hh, z);
-         } else {
-            step = &((*step)->next); // advance through list
-         }
-      }
-
-      // insert all edges that start before the bottom of this scanline
-      while (e->y0 <= scan_y_bottom) {
-         if (e->y0 != e->y1) {
-            stbtt__active_edge *z = stbtt__new_active(&hh, e, off_x, scan_y_top, userdata);
-            if (z != NULL) {
-               STBTT_assert(z->ey >= scan_y_top);
-               // insert at front
-               z->next = active;
-               active = z;
-            }
-         }
-         ++e;
-      }
-
-      // now process all active edges
-      if (active)
-         stbtt__fill_active_edges_new(scanline, scanline2+1, result->w, active, scan_y_top);
-
-      {
-         float sum = 0;
-         for (i=0; i < result->w; ++i) {
-            float k;
-            int m;
-            sum += scanline2[i];
-            k = scanline[i] + sum;
-            k = (float) STBTT_fabs(k)*255 + 0.5f;
-            m = (int) k;
-            if (m > 255) m = 255;
-            result->pixels[j*result->stride + i] = (unsigned char) m;
-         }
-      }
-      // advance all the edges
-      step = &active;
-      while (*step) {
-         stbtt__active_edge *z = *step;
-         z->fx += z->fdx; // advance to position for current scanline
-         step = &((*step)->next); // advance through list
-      }
-
-      ++y;
-      ++j;
-   }
-
-   stbtt__hheap_cleanup(&hh, userdata);
-
-   if (scanline != scanline_data)
-      STBTT_free(scanline, userdata);
-}
-#else
-#error "Unrecognized value of STBTT_RASTERIZER_VERSION"
-#endif
-
-#define STBTT__COMPARE(a,b)  ((a)->y0 < (b)->y0)
-
-static void stbtt__sort_edges_ins_sort(stbtt__edge *p, int n)
-{
-   int i,j;
-   for (i=1; i < n; ++i) {
-      stbtt__edge t = p[i], *a = &t;
-      j = i;
-      while (j > 0) {
-         stbtt__edge *b = &p[j-1];
-         int c = STBTT__COMPARE(a,b);
-         if (!c) break;
-         p[j] = p[j-1];
-         --j;
-      }
-      if (i != j)
-         p[j] = t;
-   }
-}
-
-static void stbtt__sort_edges_quicksort(stbtt__edge *p, int n)
-{
-   /* threshhold for transitioning to insertion sort */
-   while (n > 12) {
-      stbtt__edge t;
-      int c01,c12,c,m,i,j;
-
-      /* compute median of three */
-      m = n >> 1;
-      c01 = STBTT__COMPARE(&p[0],&p[m]);
-      c12 = STBTT__COMPARE(&p[m],&p[n-1]);
-      /* if 0 >= mid >= end, or 0 < mid < end, then use mid */
-      if (c01 != c12) {
-         /* otherwise, we'll need to swap something else to middle */
-         int z;
-         c = STBTT__COMPARE(&p[0],&p[n-1]);
-         /* 0>mid && mid<n:  0>n => n; 0<n => 0 */
-         /* 0<mid && mid>n:  0>n => 0; 0<n => n */
-         z = (c == c12) ? 0 : n-1;
-         t = p[z];
-         p[z] = p[m];
-         p[m] = t;
-      }
-      /* now p[m] is the median-of-three */
-      /* swap it to the beginning so it won't move around */
-      t = p[0];
-      p[0] = p[m];
-      p[m] = t;
-
-      /* partition loop */
-      i=1;
-      j=n-1;
-      for(;;) {
-         /* handling of equality is crucial here */
-         /* for sentinels & efficiency with duplicates */
-         for (;;++i) {
-            if (!STBTT__COMPARE(&p[i], &p[0])) break;
-         }
-         for (;;--j) {
-            if (!STBTT__COMPARE(&p[0], &p[j])) break;
-         }
-         /* make sure we haven't crossed */
-         if (i >= j) break;
-         t = p[i];
-         p[i] = p[j];
-         p[j] = t;
-
-         ++i;
-         --j;
-      }
-      /* recurse on smaller side, iterate on larger */
-      if (j < (n-i)) {
-         stbtt__sort_edges_quicksort(p,j);
-         p = p+i;
-         n = n-i;
-      } else {
-         stbtt__sort_edges_quicksort(p+i, n-i);
-         n = j;
-      }
-   }
-}
-
-static void stbtt__sort_edges(stbtt__edge *p, int n)
-{
-   stbtt__sort_edges_quicksort(p, n);
-   stbtt__sort_edges_ins_sort(p, n);
-}
-
-typedef struct
-{
-   float x,y;
-} stbtt__point;
-
-static void stbtt__rasterize(stbtt__bitmap *result, stbtt__point *pts, int *wcount, int windings, float scale_x, float scale_y, float shift_x, float shift_y, int off_x, int off_y, int invert, void *userdata)
-{
-   float y_scale_inv = invert ? -scale_y : scale_y;
-   stbtt__edge *e;
-   int n,i,j,k,m;
-#if STBTT_RASTERIZER_VERSION == 1
-   int vsubsample = result->h < 8 ? 15 : 5;
-#elif STBTT_RASTERIZER_VERSION == 2
-   int vsubsample = 1;
-#else
-   #error "Unrecognized value of STBTT_RASTERIZER_VERSION"
-#endif
-   // vsubsample should divide 255 evenly; otherwise we won't reach full opacity
-
-   // now we have to blow out the windings into explicit edge lists
-   n = 0;
-   for (i=0; i < windings; ++i)
-      n += wcount[i];
-
-   e = (stbtt__edge *) STBTT_malloc(sizeof(*e) * (n+1), userdata); // add an extra one as a sentinel
-   if (e == 0) return;
-   n = 0;
-
-   m=0;
-   for (i=0; i < windings; ++i) {
-      stbtt__point *p = pts + m;
-      m += wcount[i];
-      j = wcount[i]-1;
-      for (k=0; k < wcount[i]; j=k++) {
-         int a=k,b=j;
-         // skip the edge if horizontal
-         if (p[j].y == p[k].y)
-            continue;
-         // add edge from j to k to the list
-         e[n].invert = 0;
-         if (invert ? p[j].y > p[k].y : p[j].y < p[k].y) {
-            e[n].invert = 1;
-            a=j,b=k;
-         }
-         e[n].x0 = p[a].x * scale_x + shift_x;
-         e[n].y0 = (p[a].y * y_scale_inv + shift_y) * vsubsample;
-         e[n].x1 = p[b].x * scale_x + shift_x;
-         e[n].y1 = (p[b].y * y_scale_inv + shift_y) * vsubsample;
-         ++n;
-      }
-   }
-
-   // now sort the edges by their highest point (should snap to integer, and then by x)
-   //STBTT_sort(e, n, sizeof(e[0]), stbtt__edge_compare);
-   stbtt__sort_edges(e, n);
-
-   // now, traverse the scanlines and find the intersections on each scanline, use xor winding rule
-   stbtt__rasterize_sorted_edges(result, e, n, vsubsample, off_x, off_y, userdata);
-
-   STBTT_free(e, userdata);
-}
-
-static void stbtt__add_point(stbtt__point *points, int n, float x, float y)
-{
-   if (!points) return; // during first pass, it's unallocated
-   points[n].x = x;
-   points[n].y = y;
-}
-
-// tesselate until threshhold p is happy... @TODO warped to compensate for non-linear stretching
-static int stbtt__tesselate_curve(stbtt__point *points, int *num_points, float x0, float y0, float x1, float y1, float x2, float y2, float objspace_flatness_squared, int n)
-{
-   // midpoint
-   float mx = (x0 + 2*x1 + x2)/4;
-   float my = (y0 + 2*y1 + y2)/4;
-   // versus directly drawn line
-   float dx = (x0+x2)/2 - mx;
-   float dy = (y0+y2)/2 - my;
-   if (n > 16) // 65536 segments on one curve better be enough!
-      return 1;
-   if (dx*dx+dy*dy > objspace_flatness_squared) { // half-pixel error allowed... need to be smaller if AA
-      stbtt__tesselate_curve(points, num_points, x0,y0, (x0+x1)/2.0f,(y0+y1)/2.0f, mx,my, objspace_flatness_squared,n+1);
-      stbtt__tesselate_curve(points, num_points, mx,my, (x1+x2)/2.0f,(y1+y2)/2.0f, x2,y2, objspace_flatness_squared,n+1);
-   } else {
-      stbtt__add_point(points, *num_points,x2,y2);
-      *num_points = *num_points+1;
-   }
-   return 1;
-}
-
-// returns number of contours
-static stbtt__point *stbtt_FlattenCurves(stbtt_vertex *vertices, int num_verts, float objspace_flatness, int **contour_lengths, int *num_contours, void *userdata)
-{
-   stbtt__point *points=0;
-   int num_points=0;
-
-   float objspace_flatness_squared = objspace_flatness * objspace_flatness;
-   int i,n=0,start=0, pass;
-
-   // count how many "moves" there are to get the contour count
-   for (i=0; i < num_verts; ++i)
-      if (vertices[i].type == STBTT_vmove)
-         ++n;
-
-   *num_contours = n;
-   if (n == 0) return 0;
-
-   *contour_lengths = (int *) STBTT_malloc(sizeof(**contour_lengths) * n, userdata);
-
-   if (*contour_lengths == 0) {
-      *num_contours = 0;
-      return 0;
-   }
-
-   // make two passes through the points so we don't need to realloc
-   for (pass=0; pass < 2; ++pass) {
-      float x=0,y=0;
-      if (pass == 1) {
-         points = (stbtt__point *) STBTT_malloc(num_points * sizeof(points[0]), userdata);
-         if (points == NULL) goto error;
-      }
-      num_points = 0;
-      n= -1;
-      for (i=0; i < num_verts; ++i) {
-         switch (vertices[i].type) {
-            case STBTT_vmove:
-               // start the next contour
-               if (n >= 0)
-                  (*contour_lengths)[n] = num_points - start;
-               ++n;
-               start = num_points;
-
-               x = vertices[i].x, y = vertices[i].y;
-               stbtt__add_point(points, num_points++, x,y);
-               break;
-            case STBTT_vline:
-               x = vertices[i].x, y = vertices[i].y;
-               stbtt__add_point(points, num_points++, x, y);
-               break;
-            case STBTT_vcurve:
-               stbtt__tesselate_curve(points, &num_points, x,y,
-                                        vertices[i].cx, vertices[i].cy,
-                                        vertices[i].x,  vertices[i].y,
-                                        objspace_flatness_squared, 0);
-               x = vertices[i].x, y = vertices[i].y;
-               break;
-         }
-      }
-      (*contour_lengths)[n] = num_points - start;
-   }
-
-   return points;
-error:
-   STBTT_free(points, userdata);
-   STBTT_free(*contour_lengths, userdata);
-   *contour_lengths = 0;
-   *num_contours = 0;
-   return NULL;
-}
-
-STBTT_DEF void stbtt_Rasterize(stbtt__bitmap *result, float flatness_in_pixels, stbtt_vertex *vertices, int num_verts, float scale_x, float scale_y, float shift_x, float shift_y, int x_off, int y_off, int invert, void *userdata)
-{
-   float scale = scale_x > scale_y ? scale_y : scale_x;
-   int winding_count, *winding_lengths;
-   stbtt__point *windings = stbtt_FlattenCurves(vertices, num_verts, flatness_in_pixels / scale, &winding_lengths, &winding_count, userdata);
-   if (windings) {
-      stbtt__rasterize(result, windings, winding_lengths, winding_count, scale_x, scale_y, shift_x, shift_y, x_off, y_off, invert, userdata);
-      STBTT_free(winding_lengths, userdata);
-      STBTT_free(windings, userdata);
-   }
-}
-
-STBTT_DEF void stbtt_FreeBitmap(unsigned char *bitmap, void *userdata)
-{
-   STBTT_free(bitmap, userdata);
-}
-
-STBTT_DEF unsigned char *stbtt_GetGlyphBitmapSubpixel(const stbtt_fontinfo *info, float scale_x, float scale_y, float shift_x, float shift_y, int glyph, int *width, int *height, int *xoff, int *yoff)
-{
-   int ix0,iy0,ix1,iy1;
-   stbtt__bitmap gbm;
-   stbtt_vertex *vertices;
-   int num_verts = stbtt_GetGlyphShape(info, glyph, &vertices);
-
-   if (scale_x == 0) scale_x = scale_y;
-   if (scale_y == 0) {
-      if (scale_x == 0) {
-         STBTT_free(vertices, info->userdata);
-         return NULL;
-      }
-      scale_y = scale_x;
-   }
-
-   stbtt_GetGlyphBitmapBoxSubpixel(info, glyph, scale_x, scale_y, shift_x, shift_y, &ix0,&iy0,&ix1,&iy1);
-
-   // now we get the size
-   gbm.w = (ix1 - ix0);
-   gbm.h = (iy1 - iy0);
-   gbm.pixels = NULL; // in case we error
-
-   if (width ) *width  = gbm.w;
-   if (height) *height = gbm.h;
-   if (xoff  ) *xoff   = ix0;
-   if (yoff  ) *yoff   = iy0;
-
-   if (gbm.w && gbm.h) {
-      gbm.pixels = (unsigned char *) STBTT_malloc(gbm.w * gbm.h, info->userdata);
-      if (gbm.pixels) {
-         gbm.stride = gbm.w;
-
-         stbtt_Rasterize(&gbm, 0.35f, vertices, num_verts, scale_x, scale_y, shift_x, shift_y, ix0, iy0, 1, info->userdata);
-      }
-   }
-   STBTT_free(vertices, info->userdata);
-   return gbm.pixels;
-}
-
-STBTT_DEF unsigned char *stbtt_GetGlyphBitmap(const stbtt_fontinfo *info, float scale_x, float scale_y, int glyph, int *width, int *height, int *xoff, int *yoff)
-{
-   return stbtt_GetGlyphBitmapSubpixel(info, scale_x, scale_y, 0.0f, 0.0f, glyph, width, height, xoff, yoff);
-}
-
-STBTT_DEF void stbtt_MakeGlyphBitmapSubpixel(const stbtt_fontinfo *info, unsigned char *output, int out_w, int out_h, int out_stride, float scale_x, float scale_y, float shift_x, float shift_y, int glyph)
-{
-   int ix0,iy0;
-   stbtt_vertex *vertices;
-   int num_verts = stbtt_GetGlyphShape(info, glyph, &vertices);
-   stbtt__bitmap gbm;
-
-   stbtt_GetGlyphBitmapBoxSubpixel(info, glyph, scale_x, scale_y, shift_x, shift_y, &ix0,&iy0,0,0);
-   gbm.pixels = output;
-   gbm.w = out_w;
-   gbm.h = out_h;
-   gbm.stride = out_stride;
-
-   if (gbm.w && gbm.h)
-      stbtt_Rasterize(&gbm, 0.35f, vertices, num_verts, scale_x, scale_y, shift_x, shift_y, ix0,iy0, 1, info->userdata);
-
-   STBTT_free(vertices, info->userdata);
-}
-
-STBTT_DEF void stbtt_MakeGlyphBitmap(const stbtt_fontinfo *info, unsigned char *output, int out_w, int out_h, int out_stride, float scale_x, float scale_y, int glyph)
-{
-   stbtt_MakeGlyphBitmapSubpixel(info, output, out_w, out_h, out_stride, scale_x, scale_y, 0.0f,0.0f, glyph);
-}
-
-STBTT_DEF unsigned char *stbtt_GetCodepointBitmapSubpixel(const stbtt_fontinfo *info, float scale_x, float scale_y, float shift_x, float shift_y, int codepoint, int *width, int *height, int *xoff, int *yoff)
-{
-   return stbtt_GetGlyphBitmapSubpixel(info, scale_x, scale_y,shift_x,shift_y, stbtt_FindGlyphIndex(info,codepoint), width,height,xoff,yoff);
-}
-
-STBTT_DEF void stbtt_MakeCodepointBitmapSubpixel(const stbtt_fontinfo *info, unsigned char *output, int out_w, int out_h, int out_stride, float scale_x, float scale_y, float shift_x, float shift_y, int codepoint)
-{
-   stbtt_MakeGlyphBitmapSubpixel(info, output, out_w, out_h, out_stride, scale_x, scale_y, shift_x, shift_y, stbtt_FindGlyphIndex(info,codepoint));
-}
-
-STBTT_DEF unsigned char *stbtt_GetCodepointBitmap(const stbtt_fontinfo *info, float scale_x, float scale_y, int codepoint, int *width, int *height, int *xoff, int *yoff)
-{
-   return stbtt_GetCodepointBitmapSubpixel(info, scale_x, scale_y, 0.0f,0.0f, codepoint, width,height,xoff,yoff);
-}
-
-STBTT_DEF void stbtt_MakeCodepointBitmap(const stbtt_fontinfo *info, unsigned char *output, int out_w, int out_h, int out_stride, float scale_x, float scale_y, int codepoint)
-{
-   stbtt_MakeCodepointBitmapSubpixel(info, output, out_w, out_h, out_stride, scale_x, scale_y, 0.0f,0.0f, codepoint);
-}
-
-//////////////////////////////////////////////////////////////////////////////
-//
-// bitmap baking
-//
-// This is SUPER-CRAPPY packing to keep source code small
-
-STBTT_DEF int stbtt_BakeFontBitmap(const unsigned char *data, int offset,  // font location (use offset=0 for plain .ttf)
-                                float pixel_height,                     // height of font in pixels
-                                unsigned char *pixels, int pw, int ph,  // bitmap to be filled in
-                                int first_char, int num_chars,          // characters to bake
-                                stbtt_bakedchar *chardata)
-{
-   float scale;
-   int x,y,bottom_y, i;
-   stbtt_fontinfo f;
-   f.userdata = NULL;
-   if (!stbtt_InitFont(&f, data, offset))
-      return -1;
-   STBTT_memset(pixels, 0, pw*ph); // background of 0 around pixels
-   x=y=1;
-   bottom_y = 1;
-
-   scale = stbtt_ScaleForPixelHeight(&f, pixel_height);
-
-   for (i=0; i < num_chars; ++i) {
-      int advance, lsb, x0,y0,x1,y1,gw,gh;
-      int g = stbtt_FindGlyphIndex(&f, first_char + i);
-      stbtt_GetGlyphHMetrics(&f, g, &advance, &lsb);
-      stbtt_GetGlyphBitmapBox(&f, g, scale,scale, &x0,&y0,&x1,&y1);
-      gw = x1-x0;
-      gh = y1-y0;
-      if (x + gw + 1 >= pw)
-         y = bottom_y, x = 1; // advance to next row
-      if (y + gh + 1 >= ph) // check if it fits vertically AFTER potentially moving to next row
-         return -i;
-      STBTT_assert(x+gw < pw);
-      STBTT_assert(y+gh < ph);
-      stbtt_MakeGlyphBitmap(&f, pixels+x+y*pw, gw,gh,pw, scale,scale, g);
-      chardata[i].x0 = (stbtt_int16) x;
-      chardata[i].y0 = (stbtt_int16) y;
-      chardata[i].x1 = (stbtt_int16) (x + gw);
-      chardata[i].y1 = (stbtt_int16) (y + gh);
-      chardata[i].xadvance = scale * advance;
-      chardata[i].xoff     = (float) x0;
-      chardata[i].yoff     = (float) y0;
-      x = x + gw + 1;
-      if (y+gh+1 > bottom_y)
-         bottom_y = y+gh+1;
-   }
-   return bottom_y;
-}
-
-STBTT_DEF void stbtt_GetBakedQuad(stbtt_bakedchar *chardata, int pw, int ph, int char_index, float *xpos, float *ypos, stbtt_aligned_quad *q, int opengl_fillrule)
-{
-   float d3d_bias = opengl_fillrule ? 0 : -0.5f;
-   float ipw = 1.0f / pw, iph = 1.0f / ph;
-   stbtt_bakedchar *b = chardata + char_index;
-   int round_x = STBTT_ifloor((*xpos + b->xoff) + 0.5f);
-   int round_y = STBTT_ifloor((*ypos + b->yoff) + 0.5f);
-
-   q->x0 = round_x + d3d_bias;
-   q->y0 = round_y + d3d_bias;
-   q->x1 = round_x + b->x1 - b->x0 + d3d_bias;
-   q->y1 = round_y + b->y1 - b->y0 + d3d_bias;
-
-   q->s0 = b->x0 * ipw;
-   q->t0 = b->y0 * iph;
-   q->s1 = b->x1 * ipw;
-   q->t1 = b->y1 * iph;
-
-   *xpos += b->xadvance;
-}
-
-//////////////////////////////////////////////////////////////////////////////
-//
-// rectangle packing replacement routines if you don't have stb_rect_pack.h
-//
-
-#ifndef STB_RECT_PACK_VERSION
-
-typedef int stbrp_coord;
-
-////////////////////////////////////////////////////////////////////////////////////
-//                                                                                //
-//                                                                                //
-// COMPILER WARNING ?!?!?                                                         //
-//                                                                                //
-//                                                                                //
-// if you get a compile warning due to these symbols being defined more than      //
-// once, move #include "stb_rect_pack.h" before #include "stb_truetype.h"         //
-//                                                                                //
-////////////////////////////////////////////////////////////////////////////////////
-
-typedef struct
-{
-   int width,height;
-   int x,y,bottom_y;
-} stbrp_context;
-
-typedef struct
-{
-   unsigned char x;
-} stbrp_node;
-
-struct stbrp_rect
-{
-   stbrp_coord x,y;
-   int id,w,h,was_packed;
-};
-
-static void stbrp_init_target(stbrp_context *con, int pw, int ph, stbrp_node *nodes, int num_nodes)
-{
-   con->width  = pw;
-   con->height = ph;
-   con->x = 0;
-   con->y = 0;
-   con->bottom_y = 0;
-   STBTT__NOTUSED(nodes);
-   STBTT__NOTUSED(num_nodes);
-}
-
-static void stbrp_pack_rects(stbrp_context *con, stbrp_rect *rects, int num_rects)
-{
-   int i;
-   for (i=0; i < num_rects; ++i) {
-      if (con->x + rects[i].w > con->width) {
-         con->x = 0;
-         con->y = con->bottom_y;
-      }
-      if (con->y + rects[i].h > con->height)
-         break;
-      rects[i].x = con->x;
-      rects[i].y = con->y;
-      rects[i].was_packed = 1;
-      con->x += rects[i].w;
-      if (con->y + rects[i].h > con->bottom_y)
-         con->bottom_y = con->y + rects[i].h;
-   }
-   for (   ; i < num_rects; ++i)
-      rects[i].was_packed = 0;
-}
-#endif
-
-//////////////////////////////////////////////////////////////////////////////
-//
-// bitmap baking
-//
-// This is SUPER-AWESOME (tm Ryan Gordon) packing using stb_rect_pack.h. If
-// stb_rect_pack.h isn't available, it uses the BakeFontBitmap strategy.
-
-STBTT_DEF int stbtt_PackBegin(stbtt_pack_context *spc, unsigned char *pixels, int pw, int ph, int stride_in_bytes, int padding, void *alloc_context)
-{
-   stbrp_context *context = (stbrp_context *) STBTT_malloc(sizeof(*context)            ,alloc_context);
-   int            num_nodes = pw - padding;
-   stbrp_node    *nodes   = (stbrp_node    *) STBTT_malloc(sizeof(*nodes  ) * num_nodes,alloc_context);
-
-   if (context == NULL || nodes == NULL) {
-      if (context != NULL) STBTT_free(context, alloc_context);
-      if (nodes   != NULL) STBTT_free(nodes  , alloc_context);
-      return 0;
-   }
-
-   spc->user_allocator_context = alloc_context;
-   spc->width = pw;
-   spc->height = ph;
-   spc->pixels = pixels;
-   spc->pack_info = context;
-   spc->nodes = nodes;
-   spc->padding = padding;
-   spc->stride_in_bytes = stride_in_bytes != 0 ? stride_in_bytes : pw;
-   spc->h_oversample = 1;
-   spc->v_oversample = 1;
-
-   stbrp_init_target(context, pw-padding, ph-padding, nodes, num_nodes);
-
-   if (pixels)
-      STBTT_memset(pixels, 0, pw*ph); // background of 0 around pixels
-
-   return 1;
-}
-
-STBTT_DEF void stbtt_PackEnd  (stbtt_pack_context *spc)
-{
-   STBTT_free(spc->nodes    , spc->user_allocator_context);
-   STBTT_free(spc->pack_info, spc->user_allocator_context);
-}
-
-STBTT_DEF void stbtt_PackSetOversampling(stbtt_pack_context *spc, unsigned int h_oversample, unsigned int v_oversample)
-{
-   STBTT_assert(h_oversample <= STBTT_MAX_OVERSAMPLE);
-   STBTT_assert(v_oversample <= STBTT_MAX_OVERSAMPLE);
-   if (h_oversample <= STBTT_MAX_OVERSAMPLE)
-      spc->h_oversample = h_oversample;
-   if (v_oversample <= STBTT_MAX_OVERSAMPLE)
-      spc->v_oversample = v_oversample;
-}
-
-#define STBTT__OVER_MASK  (STBTT_MAX_OVERSAMPLE-1)
-
-static void stbtt__h_prefilter(unsigned char *pixels, int w, int h, int stride_in_bytes, unsigned int kernel_width)
-{
-   unsigned char buffer[STBTT_MAX_OVERSAMPLE];
-   int safe_w = w - kernel_width;
-   int j;
-   STBTT_memset(buffer, 0, STBTT_MAX_OVERSAMPLE); // suppress bogus warning from VS2013 -analyze
-   for (j=0; j < h; ++j) {
-      int i;
-      unsigned int total;
-      STBTT_memset(buffer, 0, kernel_width);
-
-      total = 0;
-
-      // make kernel_width a constant in common cases so compiler can optimize out the divide
-      switch (kernel_width) {
-         case 2:
-            for (i=0; i <= safe_w; ++i) {
-               total += pixels[i] - buffer[i & STBTT__OVER_MASK];
-               buffer[(i+kernel_width) & STBTT__OVER_MASK] = pixels[i];
-               pixels[i] = (unsigned char) (total / 2);
-            }
-            break;
-         case 3:
-            for (i=0; i <= safe_w; ++i) {
-               total += pixels[i] - buffer[i & STBTT__OVER_MASK];
-               buffer[(i+kernel_width) & STBTT__OVER_MASK] = pixels[i];
-               pixels[i] = (unsigned char) (total / 3);
-            }
-            break;
-         case 4:
-            for (i=0; i <= safe_w; ++i) {
-               total += pixels[i] - buffer[i & STBTT__OVER_MASK];
-               buffer[(i+kernel_width) & STBTT__OVER_MASK] = pixels[i];
-               pixels[i] = (unsigned char) (total / 4);
-            }
-            break;
-         case 5:
-            for (i=0; i <= safe_w; ++i) {
-               total += pixels[i] - buffer[i & STBTT__OVER_MASK];
-               buffer[(i+kernel_width) & STBTT__OVER_MASK] = pixels[i];
-               pixels[i] = (unsigned char) (total / 5);
-            }
-            break;
-         default:
-            for (i=0; i <= safe_w; ++i) {
-               total += pixels[i] - buffer[i & STBTT__OVER_MASK];
-               buffer[(i+kernel_width) & STBTT__OVER_MASK] = pixels[i];
-               pixels[i] = (unsigned char) (total / kernel_width);
-            }
-            break;
-      }
-
-      for (; i < w; ++i) {
-         STBTT_assert(pixels[i] == 0);
-         total -= buffer[i & STBTT__OVER_MASK];
-         pixels[i] = (unsigned char) (total / kernel_width);
-      }
-
-      pixels += stride_in_bytes;
-   }
-}
-
-static void stbtt__v_prefilter(unsigned char *pixels, int w, int h, int stride_in_bytes, unsigned int kernel_width)
-{
-   unsigned char buffer[STBTT_MAX_OVERSAMPLE];
-   int safe_h = h - kernel_width;
-   int j;
-   STBTT_memset(buffer, 0, STBTT_MAX_OVERSAMPLE); // suppress bogus warning from VS2013 -analyze
-   for (j=0; j < w; ++j) {
-      int i;
-      unsigned int total;
-      STBTT_memset(buffer, 0, kernel_width);
-
-      total = 0;
-
-      // make kernel_width a constant in common cases so compiler can optimize out the divide
-      switch (kernel_width) {
-         case 2:
-            for (i=0; i <= safe_h; ++i) {
-               total += pixels[i*stride_in_bytes] - buffer[i & STBTT__OVER_MASK];
-               buffer[(i+kernel_width) & STBTT__OVER_MASK] = pixels[i*stride_in_bytes];
-               pixels[i*stride_in_bytes] = (unsigned char) (total / 2);
-            }
-            break;
-         case 3:
-            for (i=0; i <= safe_h; ++i) {
-               total += pixels[i*stride_in_bytes] - buffer[i & STBTT__OVER_MASK];
-               buffer[(i+kernel_width) & STBTT__OVER_MASK] = pixels[i*stride_in_bytes];
-               pixels[i*stride_in_bytes] = (unsigned char) (total / 3);
-            }
-            break;
-         case 4:
-            for (i=0; i <= safe_h; ++i) {
-               total += pixels[i*stride_in_bytes] - buffer[i & STBTT__OVER_MASK];
-               buffer[(i+kernel_width) & STBTT__OVER_MASK] = pixels[i*stride_in_bytes];
-               pixels[i*stride_in_bytes] = (unsigned char) (total / 4);
-            }
-            break;
-         case 5:
-            for (i=0; i <= safe_h; ++i) {
-               total += pixels[i*stride_in_bytes] - buffer[i & STBTT__OVER_MASK];
-               buffer[(i+kernel_width) & STBTT__OVER_MASK] = pixels[i*stride_in_bytes];
-               pixels[i*stride_in_bytes] = (unsigned char) (total / 5);
-            }
-            break;
-         default:
-            for (i=0; i <= safe_h; ++i) {
-               total += pixels[i*stride_in_bytes] - buffer[i & STBTT__OVER_MASK];
-               buffer[(i+kernel_width) & STBTT__OVER_MASK] = pixels[i*stride_in_bytes];
-               pixels[i*stride_in_bytes] = (unsigned char) (total / kernel_width);
-            }
-            break;
-      }
-
-      for (; i < h; ++i) {
-         STBTT_assert(pixels[i*stride_in_bytes] == 0);
-         total -= buffer[i & STBTT__OVER_MASK];
-         pixels[i*stride_in_bytes] = (unsigned char) (total / kernel_width);
-      }
-
-      pixels += 1;
-   }
-}
-
-static float stbtt__oversample_shift(int oversample)
-{
-   if (!oversample)
-      return 0.0f;
-
-   // The prefilter is a box filter of width "oversample",
-   // which shifts phase by (oversample - 1)/2 pixels in
-   // oversampled space. We want to shift in the opposite
-   // direction to counter this.
-   return (float)-(oversample - 1) / (2.0f * (float)oversample);
-}
-
-// rects array must be big enough to accommodate all characters in the given ranges
-STBTT_DEF int stbtt_PackFontRangesGatherRects(stbtt_pack_context *spc, stbtt_fontinfo *info, stbtt_pack_range *ranges, int num_ranges, stbrp_rect *rects)
-{
-   int i,j,k;
-
-   k=0;
-   for (i=0; i < num_ranges; ++i) {
-      float fh = ranges[i].font_size;
-      float scale = fh > 0 ? stbtt_ScaleForPixelHeight(info, fh) : stbtt_ScaleForMappingEmToPixels(info, -fh);
-      ranges[i].h_oversample = (unsigned char) spc->h_oversample;
-      ranges[i].v_oversample = (unsigned char) spc->v_oversample;
-      for (j=0; j < ranges[i].num_chars; ++j) {
-         int x0,y0,x1,y1;
-         int codepoint = ranges[i].array_of_unicode_codepoints == NULL ? ranges[i].first_unicode_codepoint_in_range + j : ranges[i].array_of_unicode_codepoints[j];
-         int glyph = stbtt_FindGlyphIndex(info, codepoint);
-         stbtt_GetGlyphBitmapBoxSubpixel(info,glyph,
-                                         scale * spc->h_oversample,
-                                         scale * spc->v_oversample,
-                                         0,0,
-                                         &x0,&y0,&x1,&y1);
-         rects[k].w = (stbrp_coord) (x1-x0 + spc->padding + spc->h_oversample-1);
-         rects[k].h = (stbrp_coord) (y1-y0 + spc->padding + spc->v_oversample-1);
-         ++k;
-      }
-   }
-
-   return k;
-}
-
-// rects array must be big enough to accommodate all characters in the given ranges
-STBTT_DEF int stbtt_PackFontRangesRenderIntoRects(stbtt_pack_context *spc, stbtt_fontinfo *info, stbtt_pack_range *ranges, int num_ranges, stbrp_rect *rects)
-{
-   int i,j,k, return_value = 1;
-
-   // save current values
-   int old_h_over = spc->h_oversample;
-   int old_v_over = spc->v_oversample;
-
-   k = 0;
-   for (i=0; i < num_ranges; ++i) {
-      float fh = ranges[i].font_size;
-      float scale = fh > 0 ? stbtt_ScaleForPixelHeight(info, fh) : stbtt_ScaleForMappingEmToPixels(info, -fh);
-      float recip_h,recip_v,sub_x,sub_y;
-      spc->h_oversample = ranges[i].h_oversample;
-      spc->v_oversample = ranges[i].v_oversample;
-      recip_h = 1.0f / spc->h_oversample;
-      recip_v = 1.0f / spc->v_oversample;
-      sub_x = stbtt__oversample_shift(spc->h_oversample);
-      sub_y = stbtt__oversample_shift(spc->v_oversample);
-      for (j=0; j < ranges[i].num_chars; ++j) {
-         stbrp_rect *r = &rects[k];
-         if (r->was_packed) {
-            stbtt_packedchar *bc = &ranges[i].chardata_for_range[j];
-            int advance, lsb, x0,y0,x1,y1;
-            int codepoint = ranges[i].array_of_unicode_codepoints == NULL ? ranges[i].first_unicode_codepoint_in_range + j : ranges[i].array_of_unicode_codepoints[j];
-            int glyph = stbtt_FindGlyphIndex(info, codepoint);
-            stbrp_coord pad = (stbrp_coord) spc->padding;
-
-            // pad on left and top
-            r->x += pad;
-            r->y += pad;
-            r->w -= pad;
-            r->h -= pad;
-            stbtt_GetGlyphHMetrics(info, glyph, &advance, &lsb);
-            stbtt_GetGlyphBitmapBox(info, glyph,
-                                    scale * spc->h_oversample,
-                                    scale * spc->v_oversample,
-                                    &x0,&y0,&x1,&y1);
-            stbtt_MakeGlyphBitmapSubpixel(info,
-                                          spc->pixels + r->x + r->y*spc->stride_in_bytes,
-                                          r->w - spc->h_oversample+1,
-                                          r->h - spc->v_oversample+1,
-                                          spc->stride_in_bytes,
-                                          scale * spc->h_oversample,
-                                          scale * spc->v_oversample,
-                                          0,0,
-                                          glyph);
-
-            if (spc->h_oversample > 1)
-               stbtt__h_prefilter(spc->pixels + r->x + r->y*spc->stride_in_bytes,
-                                  r->w, r->h, spc->stride_in_bytes,
-                                  spc->h_oversample);
-
-            if (spc->v_oversample > 1)
-               stbtt__v_prefilter(spc->pixels + r->x + r->y*spc->stride_in_bytes,
-                                  r->w, r->h, spc->stride_in_bytes,
-                                  spc->v_oversample);
-
-            bc->x0       = (stbtt_int16)  r->x;
-            bc->y0       = (stbtt_int16)  r->y;
-            bc->x1       = (stbtt_int16) (r->x + r->w);
-            bc->y1       = (stbtt_int16) (r->y + r->h);
-            bc->xadvance =                scale * advance;
-            bc->xoff     =       (float)  x0 * recip_h + sub_x;
-            bc->yoff     =       (float)  y0 * recip_v + sub_y;
-            bc->xoff2    =                (x0 + r->w) * recip_h + sub_x;
-            bc->yoff2    =                (y0 + r->h) * recip_v + sub_y;
-         } else {
-            return_value = 0; // if any fail, report failure
-         }
-
-         ++k;
-      }
-   }
-
-   // restore original values
-   spc->h_oversample = old_h_over;
-   spc->v_oversample = old_v_over;
-
-   return return_value;
-}
-
-STBTT_DEF void stbtt_PackFontRangesPackRects(stbtt_pack_context *spc, stbrp_rect *rects, int num_rects)
-{
-   stbrp_pack_rects((stbrp_context *) spc->pack_info, rects, num_rects);
-}
-
-STBTT_DEF int stbtt_PackFontRanges(stbtt_pack_context *spc, unsigned char *fontdata, int font_index, stbtt_pack_range *ranges, int num_ranges)
-{
-   stbtt_fontinfo info;
-   int i,j,n, return_value = 1;
-   //stbrp_context *context = (stbrp_context *) spc->pack_info;
-   stbrp_rect    *rects;
-
-   // flag all characters as NOT packed
-   for (i=0; i < num_ranges; ++i)
-      for (j=0; j < ranges[i].num_chars; ++j)
-         ranges[i].chardata_for_range[j].x0 =
-         ranges[i].chardata_for_range[j].y0 =
-         ranges[i].chardata_for_range[j].x1 =
-         ranges[i].chardata_for_range[j].y1 = 0;
-
-   n = 0;
-   for (i=0; i < num_ranges; ++i)
-      n += ranges[i].num_chars;
-
-   rects = (stbrp_rect *) STBTT_malloc(sizeof(*rects) * n, spc->user_allocator_context);
-   if (rects == NULL)
-      return 0;
-
-   info.userdata = spc->user_allocator_context;
-   stbtt_InitFont(&info, fontdata, stbtt_GetFontOffsetForIndex(fontdata,font_index));
-
-   n = stbtt_PackFontRangesGatherRects(spc, &info, ranges, num_ranges, rects);
-
-   stbtt_PackFontRangesPackRects(spc, rects, n);
-
-   return_value = stbtt_PackFontRangesRenderIntoRects(spc, &info, ranges, num_ranges, rects);
-
-   STBTT_free(rects, spc->user_allocator_context);
-   return return_value;
-}
-
-STBTT_DEF int stbtt_PackFontRange(stbtt_pack_context *spc, unsigned char *fontdata, int font_index, float font_size,
-            int first_unicode_codepoint_in_range, int num_chars_in_range, stbtt_packedchar *chardata_for_range)
-{
-   stbtt_pack_range range;
-   range.first_unicode_codepoint_in_range = first_unicode_codepoint_in_range;
-   range.array_of_unicode_codepoints = NULL;
-   range.num_chars                   = num_chars_in_range;
-   range.chardata_for_range          = chardata_for_range;
-   range.font_size                   = font_size;
-   return stbtt_PackFontRanges(spc, fontdata, font_index, &range, 1);
-}
-
-STBTT_DEF void stbtt_GetPackedQuad(stbtt_packedchar *chardata, int pw, int ph, int char_index, float *xpos, float *ypos, stbtt_aligned_quad *q, int align_to_integer)
-{
-   float ipw = 1.0f / pw, iph = 1.0f / ph;
-   stbtt_packedchar *b = chardata + char_index;
-
-   if (align_to_integer) {
-      float x = (float) STBTT_ifloor((*xpos + b->xoff) + 0.5f);
-      float y = (float) STBTT_ifloor((*ypos + b->yoff) + 0.5f);
-      q->x0 = x;
-      q->y0 = y;
-      q->x1 = x + b->xoff2 - b->xoff;
-      q->y1 = y + b->yoff2 - b->yoff;
-   } else {
-      q->x0 = *xpos + b->xoff;
-      q->y0 = *ypos + b->yoff;
-      q->x1 = *xpos + b->xoff2;
-      q->y1 = *ypos + b->yoff2;
-   }
-
-   q->s0 = b->x0 * ipw;
-   q->t0 = b->y0 * iph;
-   q->s1 = b->x1 * ipw;
-   q->t1 = b->y1 * iph;
-
-   *xpos += b->xadvance;
-}
-
-
-//////////////////////////////////////////////////////////////////////////////
-//
-// font name matching -- recommended not to use this
-//
-
-// check if a utf8 string contains a prefix which is the utf16 string; if so return length of matching utf8 string
-static stbtt_int32 stbtt__CompareUTF8toUTF16_bigendian_prefix(const stbtt_uint8 *s1, stbtt_int32 len1, const stbtt_uint8 *s2, stbtt_int32 len2)
-{
-   stbtt_int32 i=0;
-
-   // convert utf16 to utf8 and compare the results while converting
-   while (len2) {
-      stbtt_uint16 ch = s2[0]*256 + s2[1];
-      if (ch < 0x80) {
-         if (i >= len1) return -1;
-         if (s1[i++] != ch) return -1;
-      } else if (ch < 0x800) {
-         if (i+1 >= len1) return -1;
-         if (s1[i++] != 0xc0 + (ch >> 6)) return -1;
-         if (s1[i++] != 0x80 + (ch & 0x3f)) return -1;
-      } else if (ch >= 0xd800 && ch < 0xdc00) {
-         stbtt_uint32 c;
-         stbtt_uint16 ch2 = s2[2]*256 + s2[3];
-         if (i+3 >= len1) return -1;
-         c = ((ch - 0xd800) << 10) + (ch2 - 0xdc00) + 0x10000;
-         if (s1[i++] != 0xf0 + (c >> 18)) return -1;
-         if (s1[i++] != 0x80 + ((c >> 12) & 0x3f)) return -1;
-         if (s1[i++] != 0x80 + ((c >>  6) & 0x3f)) return -1;
-         if (s1[i++] != 0x80 + ((c      ) & 0x3f)) return -1;
-         s2 += 2; // plus another 2 below
-         len2 -= 2;
-      } else if (ch >= 0xdc00 && ch < 0xe000) {
-         return -1;
-      } else {
-         if (i+2 >= len1) return -1;
-         if (s1[i++] != 0xe0 + (ch >> 12)) return -1;
-         if (s1[i++] != 0x80 + ((ch >> 6) & 0x3f)) return -1;
-         if (s1[i++] != 0x80 + ((ch     ) & 0x3f)) return -1;
-      }
-      s2 += 2;
-      len2 -= 2;
-   }
-   return i;
-}
-
-STBTT_DEF int stbtt_CompareUTF8toUTF16_bigendian(const char *s1, int len1, const char *s2, int len2)
-{
-   return len1 == stbtt__CompareUTF8toUTF16_bigendian_prefix((const stbtt_uint8*) s1, len1, (const stbtt_uint8*) s2, len2);
-}
-
-// returns results in whatever encoding you request... but note that 2-byte encodings
-// will be BIG-ENDIAN... use stbtt_CompareUTF8toUTF16_bigendian() to compare
-STBTT_DEF const char *stbtt_GetFontNameString(const stbtt_fontinfo *font, int *length, int platformID, int encodingID, int languageID, int nameID)
-{
-   stbtt_int32 i,count,stringOffset;
-   stbtt_uint8 *fc = font->data;
-   stbtt_uint32 offset = font->fontstart;
-   stbtt_uint32 nm = stbtt__find_table(fc, offset, "name");
-   if (!nm) return NULL;
-
-   count = ttUSHORT(fc+nm+2);
-   stringOffset = nm + ttUSHORT(fc+nm+4);
-   for (i=0; i < count; ++i) {
-      stbtt_uint32 loc = nm + 6 + 12 * i;
-      if (platformID == ttUSHORT(fc+loc+0) && encodingID == ttUSHORT(fc+loc+2)
-          && languageID == ttUSHORT(fc+loc+4) && nameID == ttUSHORT(fc+loc+6)) {
-         *length = ttUSHORT(fc+loc+8);
-         return (const char *) (fc+stringOffset+ttUSHORT(fc+loc+10));
-      }
-   }
-   return NULL;
-}
-
-static int stbtt__matchpair(stbtt_uint8 *fc, stbtt_uint32 nm, stbtt_uint8 *name, stbtt_int32 nlen, stbtt_int32 target_id, stbtt_int32 next_id)
-{
-   stbtt_int32 i;
-   stbtt_int32 count = ttUSHORT(fc+nm+2);
-   stbtt_int32 stringOffset = nm + ttUSHORT(fc+nm+4);
-
-   for (i=0; i < count; ++i) {
-      stbtt_uint32 loc = nm + 6 + 12 * i;
-      stbtt_int32 id = ttUSHORT(fc+loc+6);
-      if (id == target_id) {
-         // find the encoding
-         stbtt_int32 platform = ttUSHORT(fc+loc+0), encoding = ttUSHORT(fc+loc+2), language = ttUSHORT(fc+loc+4);
-
-         // is this a Unicode encoding?
-         if (platform == 0 || (platform == 3 && encoding == 1) || (platform == 3 && encoding == 10)) {
-            stbtt_int32 slen = ttUSHORT(fc+loc+8);
-            stbtt_int32 off = ttUSHORT(fc+loc+10);
-
-            // check if there's a prefix match
-            stbtt_int32 matchlen = stbtt__CompareUTF8toUTF16_bigendian_prefix(name, nlen, fc+stringOffset+off,slen);
-            if (matchlen >= 0) {
-               // check for target_id+1 immediately following, with same encoding & language
-               if (i+1 < count && ttUSHORT(fc+loc+12+6) == next_id && ttUSHORT(fc+loc+12) == platform && ttUSHORT(fc+loc+12+2) == encoding && ttUSHORT(fc+loc+12+4) == language) {
-                  slen = ttUSHORT(fc+loc+12+8);
-                  off = ttUSHORT(fc+loc+12+10);
-                  if (slen == 0) {
-                     if (matchlen == nlen)
-                        return 1;
-                  } else if (matchlen < nlen && name[matchlen] == ' ') {
-                     ++matchlen;
-                     if (stbtt_CompareUTF8toUTF16_bigendian((char*) (name+matchlen), nlen-matchlen, (char*)(fc+stringOffset+off),slen))
-                        return 1;
-                  }
-               } else {
-                  // if nothing immediately following
-                  if (matchlen == nlen)
-                     return 1;
-               }
-            }
-         }
-
-         // @TODO handle other encodings
-      }
-   }
-   return 0;
-}
-
-static int stbtt__matches(stbtt_uint8 *fc, stbtt_uint32 offset, stbtt_uint8 *name, stbtt_int32 flags)
-{
-   stbtt_int32 nlen = (stbtt_int32) STBTT_strlen((char *) name);
-   stbtt_uint32 nm,hd;
-   if (!stbtt__isfont(fc+offset)) return 0;
-
-   // check italics/bold/underline flags in macStyle...
-   if (flags) {
-      hd = stbtt__find_table(fc, offset, "head");
-      if ((ttUSHORT(fc+hd+44) & 7) != (flags & 7)) return 0;
-   }
-
-   nm = stbtt__find_table(fc, offset, "name");
-   if (!nm) return 0;
-
-   if (flags) {
-      // if we checked the macStyle flags, then just check the family and ignore the subfamily
-      if (stbtt__matchpair(fc, nm, name, nlen, 16, -1))  return 1;
-      if (stbtt__matchpair(fc, nm, name, nlen,  1, -1))  return 1;
-      if (stbtt__matchpair(fc, nm, name, nlen,  3, -1))  return 1;
-   } else {
-      if (stbtt__matchpair(fc, nm, name, nlen, 16, 17))  return 1;
-      if (stbtt__matchpair(fc, nm, name, nlen,  1,  2))  return 1;
-      if (stbtt__matchpair(fc, nm, name, nlen,  3, -1))  return 1;
-   }
-
-   return 0;
-}
-
-STBTT_DEF int stbtt_FindMatchingFont(const unsigned char *font_collection, const char *name_utf8, stbtt_int32 flags)
-{
-   stbtt_int32 i;
-   for (i=0;;++i) {
-      stbtt_int32 off = stbtt_GetFontOffsetForIndex(font_collection, i);
-      if (off < 0) return off;
-      if (stbtt__matches((stbtt_uint8 *) font_collection, off, (stbtt_uint8*) name_utf8, flags))
-         return off;
-   }
-}
-
-#endif // STB_TRUETYPE_IMPLEMENTATION
-
-
-// FULL VERSION HISTORY
-//
-//   1.11 (2016-04-02) fix unused-variable warning
-//   1.10 (2016-04-02) allow user-defined fabs() replacement
-//                     fix memory leak if fontsize=0.0
-//                     fix warning from duplicate typedef
-//   1.09 (2016-01-16) warning fix; avoid crash on outofmem; use alloc userdata for PackFontRanges
-//   1.08 (2015-09-13) document stbtt_Rasterize(); fixes for vertical & horizontal edges
-//   1.07 (2015-08-01) allow PackFontRanges to accept arrays of sparse codepoints;
-//                     allow PackFontRanges to pack and render in separate phases;
-//                     fix stbtt_GetFontOFfsetForIndex (never worked for non-0 input?);
-//                     fixed an assert() bug in the new rasterizer
-//                     replace assert() with STBTT_assert() in new rasterizer
-//   1.06 (2015-07-14) performance improvements (~35% faster on x86 and x64 on test machine)
-//                     also more precise AA rasterizer, except if shapes overlap
-//                     remove need for STBTT_sort
-//   1.05 (2015-04-15) fix misplaced definitions for STBTT_STATIC
-//   1.04 (2015-04-15) typo in example
-//   1.03 (2015-04-12) STBTT_STATIC, fix memory leak in new packing, various fixes
-//   1.02 (2014-12-10) fix various warnings & compile issues w/ stb_rect_pack, C++
-//   1.01 (2014-12-08) fix subpixel position when oversampling to exactly match
-//                        non-oversampled; STBTT_POINT_SIZE for packed case only
-//   1.00 (2014-12-06) add new PackBegin etc. API, w/ support for oversampling
-//   0.99 (2014-09-18) fix multiple bugs with subpixel rendering (ryg)
-//   0.9  (2014-08-07) support certain mac/iOS fonts without an MS platformID
-//   0.8b (2014-07-07) fix a warning
-//   0.8  (2014-05-25) fix a few more warnings
-//   0.7  (2013-09-25) bugfix: subpixel glyph bug fixed in 0.5 had come back
-//   0.6c (2012-07-24) improve documentation
-//   0.6b (2012-07-20) fix a few more warnings
-//   0.6  (2012-07-17) fix warnings; added stbtt_ScaleForMappingEmToPixels,
-//                        stbtt_GetFontBoundingBox, stbtt_IsGlyphEmpty
-//   0.5  (2011-12-09) bugfixes:
-//                        subpixel glyph renderer computed wrong bounding box
-//                        first vertex of shape can be off-curve (FreeSans)
-//   0.4b (2011-12-03) fixed an error in the font baking example
-//   0.4  (2011-12-01) kerning, subpixel rendering (tor)
-//                    bugfixes for:
-//                        codepoint-to-glyph conversion using table fmt=12
-//                        codepoint-to-glyph conversion using table fmt=4
-//                        stbtt_GetBakedQuad with non-square texture (Zer)
-//                    updated Hello World! sample to use kerning and subpixel
-//                    fixed some warnings
-//   0.3  (2009-06-24) cmap fmt=12, compound shapes (MM)
-//                    userdata, malloc-from-userdata, non-zero fill (stb)
-//   0.2  (2009-03-11) Fix unsigned/signed char warnings
-//   0.1  (2009-03-09) First public release
-//
diff --git a/scene/resources/surface_tool.h b/scene/resources/surface_tool.h
index 6e52db218d..b143086e11 100644
--- a/scene/resources/surface_tool.h
+++ b/scene/resources/surface_tool.h
@@ -30,9 +30,10 @@
 #ifndef SURFACE_TOOL_H
 #define SURFACE_TOOL_H
 
-#include "mikktspace.h"
 #include "scene/resources/mesh.h"
 
+#include "thirdparty/misc/mikktspace.h"
+
 class SurfaceTool : public Reference {
 
 	GDCLASS(SurfaceTool, Reference);