Adding upstream version 4:24.2.0.upstream/4%24.2.0

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

11 files changed, 5927 insertions, 0 deletions

diff --git a/tools/source/generic/b3dtrans.cxx b/tools/source/generic/b3dtrans.cxx
new file mode 100644
index 0000000000..0cae750da0
--- /dev/null
+++ b/tools/source/generic/b3dtrans.cxx
@@ -0,0 +1,456 @@
+/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
+/*
+ * This file is part of the LibreOffice project.
+ *
+ * This Source Code Form is subject to the terms of the Mozilla Public
+ * License, v. 2.0. If a copy of the MPL was not distributed with this
+ * file, You can obtain one at http://mozilla.org/MPL/2.0/.
+ *
+ * This file incorporates work covered by the following license notice:
+ *
+ *   Licensed to the Apache Software Foundation (ASF) under one or more
+ *   contributor license agreements. See the NOTICE file distributed
+ *   with this work for additional information regarding copyright
+ *   ownership. The ASF licenses this file to you under the Apache
+ *   License, Version 2.0 (the "License"); you may not use this file
+ *   except in compliance with the License. You may obtain a copy of
+ *   the License at http://www.apache.org/licenses/LICENSE-2.0 .
+ */
+
+#include <tools/b3dtrans.hxx>
+
+#include <osl/diagnose.h>
+
+    // Near and far clipping planes
+constexpr double gfNearBound = 0.001;
+constexpr double gfFarBound = 1.001;
+
+
+// B3dTransformationSet --------------------------------------------------------
+// Transformations for all 3D output
+
+B3dTransformationSet::B3dTransformationSet()
+{
+    Reset();
+}
+
+B3dTransformationSet::~B3dTransformationSet()
+{
+}
+
+void B3dTransformationSet::Orientation(basegfx::B3DHomMatrix& rTarget, const basegfx::B3DPoint& aVRP, basegfx::B3DVector aVPN, basegfx::B3DVector aVUP)
+{
+    rTarget.translate( -aVRP.getX(), -aVRP.getY(), -aVRP.getZ());
+    aVUP.normalize();
+    aVPN.normalize();
+    basegfx::B3DVector aRx(aVUP);
+    basegfx::B3DVector aRy(aVPN);
+    aRx = aRx.getPerpendicular(aRy);
+    aRx.normalize();
+    aRy = aRy.getPerpendicular(aRx);
+    aRy.normalize();
+    basegfx::B3DHomMatrix aTemp;
+    aTemp.set(0, 0, aRx.getX());
+    aTemp.set(0, 1, aRx.getY());
+    aTemp.set(0, 2, aRx.getZ());
+    aTemp.set(1, 0, aRy.getX());
+    aTemp.set(1, 1, aRy.getY());
+    aTemp.set(1, 2, aRy.getZ());
+    aTemp.set(2, 0, aVPN.getX());
+    aTemp.set(2, 1, aVPN.getY());
+    aTemp.set(2, 2, aVPN.getZ());
+    rTarget *= aTemp;
+}
+
+void B3dTransformationSet::Frustum(basegfx::B3DHomMatrix& rTarget, double fLeft, double fRight, double fBottom, double fTop, double fNear, double fFar)
+{
+    if(!(fNear > 0.0))
+    {
+        fNear = 0.001;
+    }
+    if(!(fFar > 0.0))
+    {
+        fFar = 1.0;
+    }
+    if(fNear == fFar)
+    {
+        fFar = fNear + 1.0;
+    }
+    if(fLeft == fRight)
+    {
+        fLeft -= 1.0;
+        fRight += 1.0;
+    }
+    if(fTop == fBottom)
+    {
+        fBottom -= 1.0;
+        fTop += 1.0;
+    }
+    basegfx::B3DHomMatrix aTemp;
+
+    aTemp.set(0, 0, 2.0 * fNear / (fRight - fLeft));
+    aTemp.set(1, 1, 2.0 * fNear / (fTop - fBottom));
+    aTemp.set(0, 2, (fRight + fLeft) / (fRight - fLeft));
+    aTemp.set(1, 2, (fTop + fBottom) / (fTop - fBottom));
+    aTemp.set(2, 2, -1.0 * ((fFar + fNear) / (fFar - fNear)));
+    aTemp.set(3, 2, -1.0);
+    aTemp.set(2, 3, -1.0 * ((2.0 * fFar * fNear) / (fFar - fNear)));
+    aTemp.set(3, 3, 0.0);
+
+    rTarget *= aTemp;
+}
+
+void B3dTransformationSet::Ortho(basegfx::B3DHomMatrix& rTarget,
+                                 double fLeft, double fRight, double fBottom, double fTop,
+                                 double fNear, double fFar)
+{
+    if(fNear == fFar)
+    {
+        OSL_FAIL("Near and far clipping plane in Ortho definition are identical");
+        fFar = fNear + 1.0;
+    }
+    if(fLeft == fRight)
+    {
+        OSL_FAIL("Left and right in Ortho definition are identical");
+        fLeft -= 1.0;
+        fRight += 1.0;
+    }
+    if(fTop == fBottom)
+    {
+        OSL_FAIL("Top and bottom in Ortho definition are identical");
+        fBottom -= 1.0;
+        fTop += 1.0;
+    }
+    basegfx::B3DHomMatrix aTemp;
+
+    aTemp.set(0, 0, 2.0 / (fRight - fLeft));
+    aTemp.set(1, 1, 2.0 / (fTop - fBottom));
+    aTemp.set(2, 2, -1.0 * (2.0 / (fFar - fNear)));
+    aTemp.set(0, 3, -1.0 * ((fRight + fLeft) / (fRight - fLeft)));
+    aTemp.set(1, 3, -1.0 * ((fTop + fBottom) / (fTop - fBottom)));
+    aTemp.set(2, 3, -1.0 * ((fFar + fNear) / (fFar - fNear)));
+
+    rTarget *= aTemp;
+}
+
+/// reset values
+void B3dTransformationSet::Reset()
+{
+    // Reset matrices to identity matrices
+    maObjectTrans.identity();
+    PostSetObjectTrans();
+
+    Orientation(maOrientation);
+    PostSetOrientation();
+
+    maTexture.identity();
+
+    mfLeftBound = mfBottomBound = -1.0;
+    mfRightBound = mfTopBound = 1.0;
+
+    mfRatio = 0.0;
+
+    maViewportRectangle = tools::Rectangle(-1, -1, 2, 2);
+    maVisibleRectangle = maViewportRectangle;
+
+    mbPerspective = true;
+
+    mbProjectionValid = false;
+
+    CalcViewport();
+}
+
+/// Object transformation
+void B3dTransformationSet::PostSetObjectTrans()
+{
+    // Assign and compute inverse
+    maInvObjectTrans = maObjectTrans;
+    maInvObjectTrans.invert();
+}
+
+void B3dTransformationSet::SetOrientation(const basegfx::B3DPoint& rVRP, const basegfx::B3DVector& rVPN, const basegfx::B3DVector& rVUP)
+{
+    maOrientation.identity();
+    Orientation(maOrientation, rVRP, rVPN, rVUP);
+
+    PostSetOrientation();
+}
+
+void B3dTransformationSet::PostSetOrientation()
+{
+    // Assign and compute inverse
+    maInvOrientation = maOrientation;
+    maInvOrientation.invert();
+}
+
+/// Projections for transformations
+void B3dTransformationSet::SetProjection(const basegfx::B3DHomMatrix& mProject)
+{
+    maProjection = mProject;
+    PostSetProjection();
+}
+
+const basegfx::B3DHomMatrix& B3dTransformationSet::GetProjection()
+{
+    if(!mbProjectionValid)
+        CalcViewport();
+    return maProjection;
+}
+
+void B3dTransformationSet::PostSetProjection()
+{
+    // Assign and compute inverse
+    maInvProjection = GetProjection();
+    maInvProjection.invert();
+}
+
+/// Transformations for viewport
+void B3dTransformationSet::CalcViewport()
+{
+    // Parameters for projection
+    double fLeft(mfLeftBound);
+    double fRight(mfRightBound);
+    double fBottom(mfBottomBound);
+    double fTop(mfTopBound);
+
+    // Adjust projection to aspect ratio, if set
+    if(GetRatio() != 0.0)
+    {
+        // Compute current aspect ratio of boundaries
+        double fBoundWidth = static_cast<double>(maViewportRectangle.GetWidth() + 1);
+        double fBoundHeight = static_cast<double>(maViewportRectangle.GetHeight() + 1);
+        double fActRatio = 1;
+        double fFactor;
+
+        if(fBoundWidth != 0.0)
+            fActRatio = fBoundHeight / fBoundWidth;
+        // FIXME   else in this case has a lot of problems,  should this return.
+
+        // scale down larger part
+        if(fActRatio > mfRatio)
+        {
+            // scale down Y
+            fFactor = fActRatio;
+            fTop *= fFactor;
+            fBottom *= fFactor;
+        }
+        else
+        {
+            // scale down X
+            fFactor = 1.0 / fActRatio;
+            fRight  *= fFactor;
+            fLeft *= fFactor;
+        }
+    }
+
+    // Do projection and object areas overlap?
+    maSetBound = maViewportRectangle;
+
+    // Reset projection with new values
+    basegfx::B3DHomMatrix aNewProjection;
+
+    // #i36281#
+    // OpenGL needs a little more rough additional size to not let
+    // the front face vanish. Changed from SMALL_DVALUE to 0.000001,
+    // which is 1/10000th, compared with 1/tenth of a million from SMALL_DVALUE.
+    const double fDistPart((gfFarBound - gfNearBound) * 0.0001);
+
+    // To avoid critical clipping, set Near & Far generously
+    if(mbPerspective)
+    {
+        Frustum(aNewProjection, fLeft, fRight, fBottom, fTop, gfNearBound - fDistPart, gfFarBound + fDistPart);
+    }
+    else
+    {
+        Ortho(aNewProjection, fLeft, fRight, fBottom, fTop, gfNearBound - fDistPart, gfFarBound + fDistPart);
+    }
+
+    // Set to true to guarantee loop termination
+    mbProjectionValid = true;
+
+    // set new projection
+    SetProjection(aNewProjection);
+
+    // fill parameters for ViewportTransformation
+    // Translation
+    maTranslate.setX(static_cast<double>(maSetBound.Left()) + ((maSetBound.GetWidth() - 1) / 2.0));
+    maTranslate.setY(static_cast<double>(maSetBound.Top()) + ((maSetBound.GetHeight() - 1) / 2.0));
+    maTranslate.setZ(ZBUFFER_DEPTH_RANGE / 2.0);
+
+    // Scaling
+    maScale.setX((maSetBound.GetWidth() - 1) / 2.0);
+    maScale.setY((maSetBound.GetHeight() - 1) / -2.0);
+    maScale.setZ(ZBUFFER_DEPTH_RANGE / 2.0);
+}
+
+void B3dTransformationSet::SetRatio(double fNew)
+{
+    if(mfRatio != fNew)
+    {
+        mfRatio = fNew;
+        mbProjectionValid = false;
+    }
+}
+
+void B3dTransformationSet::SetDeviceRectangle(double fL, double fR, double fB, double fT)
+{
+    if(fL != mfLeftBound || fR != mfRightBound || fB != mfBottomBound || fT != mfTopBound)
+    {
+        mfLeftBound = fL;
+        mfRightBound = fR;
+        mfBottomBound = fB;
+        mfTopBound = fT;
+
+        mbProjectionValid = false;
+
+        // Broadcast changes
+        DeviceRectangleChange();
+    }
+}
+
+void B3dTransformationSet::DeviceRectangleChange()
+{
+}
+
+void B3dTransformationSet::SetPerspective(bool bNew)
+{
+    if(mbPerspective != bNew)
+    {
+        mbPerspective = bNew;
+        mbProjectionValid = false;
+    }
+}
+
+void B3dTransformationSet::SetViewportRectangle(tools::Rectangle const & rRect, tools::Rectangle const & rVisible)
+{
+    if(rRect != maViewportRectangle || rVisible != maVisibleRectangle)
+    {
+        maViewportRectangle = rRect;
+        maVisibleRectangle = rVisible;
+
+        mbProjectionValid = false;
+    }
+}
+
+// direct access to various transformations
+
+basegfx::B3DPoint B3dTransformationSet::WorldToEyeCoor(const basegfx::B3DPoint& rVec)
+{
+    basegfx::B3DPoint aVec(rVec);
+    aVec *= maOrientation;
+    return aVec;
+}
+
+basegfx::B3DPoint B3dTransformationSet::EyeToWorldCoor(const basegfx::B3DPoint& rVec)
+{
+    basegfx::B3DPoint aVec(rVec);
+    aVec *= maInvOrientation;
+    return aVec;
+}
+
+// B3dViewport -----------------------------------------------------------------
+
+B3dViewport::B3dViewport()
+:   aVRP(0, 0, 0),
+    aVPN(0, 0, 1),
+    aVUV(0, 1, 0)
+{
+    CalcOrientation();
+}
+
+B3dViewport::~B3dViewport()
+{
+}
+
+void B3dViewport::SetVUV(const basegfx::B3DVector& rNewVUV)
+{
+    aVUV = rNewVUV;
+    CalcOrientation();
+}
+
+void B3dViewport::SetViewportValues(
+    const basegfx::B3DPoint& rNewVRP,
+    const basegfx::B3DVector& rNewVPN,
+    const basegfx::B3DVector& rNewVUV)
+{
+    aVRP = rNewVRP;
+    aVPN = rNewVPN;
+    aVUV = rNewVUV;
+    CalcOrientation();
+}
+
+void B3dViewport::CalcOrientation()
+{
+    SetOrientation(aVRP, aVPN, aVUV);
+}
+
+// B3dCamera -------------------------------------------------------------------
+
+B3dCamera::B3dCamera(
+    const basegfx::B3DPoint& rPos, const basegfx::B3DVector& rLkAt,
+    double fFocLen, double fBnkAng)
+:   aPosition(rPos),
+    aLookAt(rLkAt),
+    fFocalLength(fFocLen),
+    fBankAngle(fBnkAng)
+{
+    CalcNewViewportValues();
+}
+
+B3dCamera::~B3dCamera()
+{
+}
+
+void B3dCamera::DeviceRectangleChange()
+{
+    // call parent
+    B3dViewport::DeviceRectangleChange();
+
+    // react to changes
+    CalcNewViewportValues();
+}
+
+void B3dCamera::CalcNewViewportValues()
+{
+    basegfx::B3DVector aNewVPN(aPosition - aLookAt);
+
+    basegfx::B3DVector aNewVUV(0.0, 1.0, 0.0);
+    if(aNewVPN.getLength() < aNewVPN.getY())
+        aNewVUV.setX(0.5);
+
+    aNewVUV.normalize();
+    aNewVPN.normalize();
+
+    basegfx::B3DVector aNewToTheRight = aNewVPN.getPerpendicular(aNewVUV);
+    aNewToTheRight.normalize();
+    aNewVUV = aNewToTheRight.getPerpendicular(aNewVPN);
+    aNewVUV.normalize();
+
+    SetViewportValues(aPosition, aNewVPN, aNewVUV);
+    CalcFocalLength();
+
+    if(fBankAngle != 0.0)
+    {
+        basegfx::B3DHomMatrix aRotMat;
+        aRotMat.rotate(0.0, 0.0, fBankAngle);
+        basegfx::B3DVector aUp(0.0, 1.0, 0.0);
+        aUp *= aRotMat;
+        aUp = EyeToWorldCoor(aUp);
+        aUp.normalize();
+        SetVUV(aUp);
+    }
+}
+
+void B3dCamera::CalcFocalLength()
+{
+    double fWidth = GetDeviceRectangleWidth();
+
+    // Adjust focal length based on given position
+    basegfx::B3DPoint aOldPosition = WorldToEyeCoor({});
+    if(fWidth != 0.0)
+        fFocalLength = aOldPosition.getZ() / fWidth * 35.0;
+    if(fFocalLength < 5.0)
+        fFocalLength = 5.0;
+}
+
+/* vim:set shiftwidth=4 softtabstop=4 expandtab: */
diff --git a/tools/source/generic/bigint.cxx b/tools/source/generic/bigint.cxx
new file mode 100644
index 0000000000..51810cab17
--- /dev/null
+++ b/tools/source/generic/bigint.cxx
@@ -0,0 +1,852 @@
+/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
+/*
+ * This file is part of the LibreOffice project.
+ *
+ * This Source Code Form is subject to the terms of the Mozilla Public
+ * License, v. 2.0. If a copy of the MPL was not distributed with this
+ * file, You can obtain one at http://mozilla.org/MPL/2.0/.
+ *
+ * This file incorporates work covered by the following license notice:
+ *
+ *   Licensed to the Apache Software Foundation (ASF) under one or more
+ *   contributor license agreements. See the NOTICE file distributed
+ *   with this work for additional information regarding copyright
+ *   ownership. The ASF licenses this file to you under the Apache
+ *   License, Version 2.0 (the "License"); you may not use this file
+ *   except in compliance with the License. You may obtain a copy of
+ *   the License at http://www.apache.org/licenses/LICENSE-2.0 .
+ */
+
+#include <math.h>
+
+#include <osl/diagnose.h>
+#include <tools/bigint.hxx>
+
+#include <algorithm>
+#include <string.h>
+
+/**
+ * The range in which we can perform add/sub without fear of overflow
+ */
+const sal_Int32 MY_MAXLONG  = 0x3fffffff;
+const sal_Int32 MY_MINLONG  = -MY_MAXLONG;
+
+/*
+ * The algorithms for Addition, Subtraction, Multiplication and Division
+ * of large numbers originate from SEMINUMERICAL ALGORITHMS by
+ * DONALD E. KNUTH in the series The Art of Computer Programming:
+ * chapter 4.3.1. The Classical Algorithms.
+ */
+
+// TODO: Needs conversion to sal_uInt16/INT16/sal_uInt32/sal_Int32
+void BigInt::MakeBigInt( const BigInt& rVal )
+{
+    if ( rVal.nLen != 0 )
+    {
+        memcpy( static_cast<void*>(this), static_cast<const void*>(&rVal), sizeof( BigInt ) );
+        while ( nLen > 1 && nNum[nLen-1] == 0 )
+            nLen--;
+    }
+    else
+    {
+        nVal = rVal.nVal;
+        sal_uInt32 nTmp;
+        if (nVal < 0)
+        {
+            bIsNeg = true;
+            nTmp = -static_cast<sal_Int64>(nVal);
+        }
+        else
+        {
+            bIsNeg = false;
+            nTmp = nVal;
+        }
+
+        nNum[0] = static_cast<sal_uInt16>(nTmp & 0xffffL);
+        nNum[1] = static_cast<sal_uInt16>(nTmp >> 16);
+        if ( nTmp & 0xffff0000L )
+            nLen = 2;
+        else
+            nLen = 1;
+    }
+}
+
+void BigInt::Normalize()
+{
+    if ( nLen != 0 )
+    {
+        while ( nLen > 1 && nNum[nLen-1] == 0 )
+            nLen--;
+
+        if ( nLen < 3 )
+        {
+            sal_Int32 newVal;
+            if ( nLen < 2 )
+                newVal = nNum[0];
+            else if ( nNum[1] & 0x8000 )
+                return;
+            else
+                newVal = (static_cast<sal_Int32>(nNum[1]) << 16) + nNum[0];
+
+            nLen = 0;
+            nVal = newVal;
+
+            if ( bIsNeg )
+                nVal = -nVal;
+        }
+        // else nVal is undefined !!! W.P.
+    }
+    // why? nVal is undefined ??? W.P.
+    else if ( nVal & 0xFFFF0000L )
+        nLen = 2;
+    else
+        nLen = 1;
+}
+
+void BigInt::Mult( const BigInt &rVal, sal_uInt16 nMul )
+{
+    sal_uInt16 nK = 0;
+    for ( int i = 0; i < rVal.nLen; i++ )
+    {
+        sal_uInt32 nTmp = static_cast<sal_uInt32>(rVal.nNum[i]) * static_cast<sal_uInt32>(nMul) + nK;
+        nK            = static_cast<sal_uInt16>(nTmp >> 16);
+        nNum[i] = static_cast<sal_uInt16>(nTmp);
+    }
+
+    if ( nK )
+    {
+        nNum[rVal.nLen] = nK;
+        nLen = rVal.nLen + 1;
+    }
+    else
+        nLen = rVal.nLen;
+
+    bIsNeg = rVal.bIsNeg;
+}
+
+void BigInt::Div( sal_uInt16 nDiv, sal_uInt16& rRem )
+{
+    sal_uInt32 nK = 0;
+    for ( int i = nLen - 1; i >= 0; i-- )
+    {
+        sal_uInt32 nTmp = static_cast<sal_uInt32>(nNum[i]) + (nK << 16);
+        nNum[i] = static_cast<sal_uInt16>(nTmp / nDiv);
+        nK            = nTmp % nDiv;
+    }
+    rRem = static_cast<sal_uInt16>(nK);
+
+    if ( nNum[nLen-1] == 0 )
+        nLen -= 1;
+}
+
+bool BigInt::IsLess( const BigInt& rVal ) const
+{
+    if ( rVal.nLen < nLen)
+        return true;
+    if ( rVal.nLen > nLen )
+        return false;
+
+    int i;
+    for ( i = nLen - 1; i > 0 && nNum[i] == rVal.nNum[i]; i-- )
+    {
+    }
+    return rVal.nNum[i] < nNum[i];
+}
+
+void BigInt::AddLong( BigInt& rB, BigInt& rErg )
+{
+    if ( bIsNeg == rB.bIsNeg )
+    {
+        int  i;
+        char len;
+
+        // if length of the two values differ, fill remaining positions
+        // of the smaller value with zeros.
+        if (nLen >= rB.nLen)
+        {
+            len = nLen;
+            for (i = rB.nLen; i < len; i++)
+                rB.nNum[i] = 0;
+        }
+        else
+        {
+            len = rB.nLen;
+            for (i = nLen; i < len; i++)
+                nNum[i] = 0;
+        }
+
+        // Add numerals, starting from the back
+        sal_Int32 k;
+        sal_Int32 nZ = 0;
+        for (i = 0, k = 0; i < len; i++) {
+            nZ = static_cast<sal_Int32>(nNum[i]) + static_cast<sal_Int32>(rB.nNum[i]) + k;
+            if (nZ & 0xff0000L)
+                k = 1;
+            else
+                k = 0;
+            rErg.nNum[i] = static_cast<sal_uInt16>(nZ & 0xffffL);
+        }
+        // If an overflow occurred, add to solution
+        if (nZ & 0xff0000L) // or if(k)
+        {
+            rErg.nNum[i] = 1;
+            len++;
+        }
+        // Set length and sign
+        rErg.nLen   = len;
+        rErg.bIsNeg = bIsNeg && rB.bIsNeg;
+    }
+    // If one of the values is negative, perform subtraction instead
+    else if (bIsNeg)
+    {
+        bIsNeg = false;
+        rB.SubLong(*this, rErg);
+        bIsNeg = true;
+    }
+    else
+    {
+        rB.bIsNeg = false;
+        SubLong(rB, rErg);
+        rB.bIsNeg = true;
+    }
+}
+
+void BigInt::SubLong( BigInt& rB, BigInt& rErg )
+{
+    if ( bIsNeg == rB.bIsNeg )
+    {
+        int  i;
+        char len;
+        sal_Int32 nZ, k;
+
+        // if length of the two values differ, fill remaining positions
+        // of the smaller value with zeros.
+        if (nLen >= rB.nLen)
+        {
+            len = nLen;
+            for (i = rB.nLen; i < len; i++)
+                rB.nNum[i] = 0;
+        }
+        else
+        {
+            len = rB.nLen;
+            for (i = nLen; i < len; i++)
+                nNum[i] = 0;
+        }
+
+        if ( IsLess(rB) )
+        {
+            for (i = 0, k = 0; i < len; i++)
+            {
+                nZ = static_cast<sal_Int32>(nNum[i]) - static_cast<sal_Int32>(rB.nNum[i]) + k;
+                if (nZ < 0)
+                    k = -1;
+                else
+                    k = 0;
+                rErg.nNum[i] = static_cast<sal_uInt16>(nZ & 0xffffL);
+            }
+            rErg.bIsNeg = bIsNeg;
+        }
+        else
+        {
+            for (i = 0, k = 0; i < len; i++)
+            {
+                nZ = static_cast<sal_Int32>(rB.nNum[i]) - static_cast<sal_Int32>(nNum[i]) + k;
+                if (nZ < 0)
+                    k = -1;
+                else
+                    k = 0;
+                rErg.nNum[i] = static_cast<sal_uInt16>(nZ & 0xffffL);
+            }
+            // if a < b, revert sign
+            rErg.bIsNeg = !bIsNeg;
+        }
+        rErg.nLen   = len;
+    }
+    // If one of the values is negative, perform addition instead
+    else if (bIsNeg)
+    {
+        bIsNeg = false;
+        AddLong(rB, rErg);
+        bIsNeg = true;
+        rErg.bIsNeg = true;
+    }
+    else
+    {
+        rB.bIsNeg = false;
+        AddLong(rB, rErg);
+        rB.bIsNeg = true;
+        rErg.bIsNeg = false;
+    }
+}
+
+void BigInt::MultLong( const BigInt& rB, BigInt& rErg ) const
+{
+    int        i, j;
+    sal_uInt32 nZ, k;
+
+    rErg.bIsNeg = bIsNeg != rB.bIsNeg;
+    rErg.nLen   = nLen + rB.nLen;
+
+    for (i = 0; i < rErg.nLen; i++)
+        rErg.nNum[i] = 0;
+
+    for (j = 0; j < rB.nLen; j++)
+    {
+        for (i = 0, k = 0; i < nLen; i++)
+        {
+            nZ = static_cast<sal_uInt32>(nNum[i]) * static_cast<sal_uInt32>(rB.nNum[j]) +
+                 static_cast<sal_uInt32>(rErg.nNum[i + j]) + k;
+            rErg.nNum[i + j] = static_cast<sal_uInt16>(nZ & 0xffffU);
+            k = nZ >> 16;
+        }
+        rErg.nNum[i + j] = static_cast<sal_uInt16>(k);
+    }
+}
+
+void BigInt::DivLong( const BigInt& rB, BigInt& rErg ) const
+{
+    int    i, j;
+    sal_uInt16 nK, nQ, nMult;
+    sal_uInt16  nLenB  = rB.nLen;
+    sal_uInt16  nLenB1 = rB.nLen - 1;
+    BigInt aTmpA, aTmpB;
+
+    nMult = static_cast<sal_uInt16>(0x10000L / (static_cast<sal_Int32>(rB.nNum[nLenB1]) + 1));
+
+    aTmpA.Mult( *this, nMult );
+    if ( aTmpA.nLen == nLen )
+    {
+        aTmpA.nNum[aTmpA.nLen] = 0;
+        aTmpA.nLen++;
+    }
+
+    aTmpB.Mult( rB, nMult );
+
+    for (j = aTmpA.nLen - 1; j >= nLenB; j--)
+    { // guess divisor
+        sal_uInt32 nTmp = ( static_cast<sal_uInt32>(aTmpA.nNum[j]) << 16 ) + aTmpA.nNum[j - 1];
+        if (aTmpA.nNum[j] == aTmpB.nNum[nLenB1])
+            nQ = 0xFFFF;
+        else
+            nQ = static_cast<sal_uInt16>(nTmp / aTmpB.nNum[nLenB1]);
+
+        if ( (static_cast<sal_uInt32>(aTmpB.nNum[nLenB1 - 1]) * nQ) >
+            ((nTmp - static_cast<sal_uInt32>(aTmpB.nNum[nLenB1]) * nQ) << 16) + aTmpA.nNum[j - 2])
+            nQ--;
+        // Start division
+        nK = 0;
+        for (i = 0; i < nLenB; i++)
+        {
+            nTmp = static_cast<sal_uInt32>(aTmpA.nNum[j - nLenB + i])
+                   - (static_cast<sal_uInt32>(aTmpB.nNum[i]) * nQ)
+                   - nK;
+            aTmpA.nNum[j - nLenB + i] = static_cast<sal_uInt16>(nTmp);
+            nK = static_cast<sal_uInt16>(nTmp >> 16);
+            if ( nK )
+                nK = static_cast<sal_uInt16>(0x10000U - nK);
+        }
+        sal_uInt16& rNum( aTmpA.nNum[j - nLenB + i] );
+        rNum -= nK;
+        if (aTmpA.nNum[j - nLenB + i] == 0)
+            rErg.nNum[j - nLenB] = nQ;
+        else
+        {
+            rErg.nNum[j - nLenB] = nQ - 1;
+            nK = 0;
+            for (i = 0; i < nLenB; i++)
+            {
+                nTmp = aTmpA.nNum[j - nLenB + i] + aTmpB.nNum[i] + nK;
+                aTmpA.nNum[j - nLenB + i] = static_cast<sal_uInt16>(nTmp & 0xFFFFL);
+                if (nTmp & 0xFFFF0000L)
+                    nK = 1;
+                else
+                    nK = 0;
+            }
+        }
+    }
+
+    rErg.bIsNeg = bIsNeg != rB.bIsNeg;
+    rErg.nLen   = nLen - rB.nLen + 1;
+}
+
+void BigInt::ModLong( const BigInt& rB, BigInt& rErg ) const
+{
+    sal_uInt16 i, j;
+    sal_uInt16 nK, nQ, nMult;
+    sal_Int16  nLenB  = rB.nLen;
+    sal_Int16  nLenB1 = rB.nLen - 1;
+    BigInt aTmpA, aTmpB;
+
+    nMult = static_cast<sal_uInt16>(0x10000L / (static_cast<sal_Int32>(rB.nNum[nLenB1]) + 1));
+
+    aTmpA.Mult( *this, nMult);
+    if ( aTmpA.nLen == nLen )
+    {
+        aTmpA.nNum[aTmpA.nLen] = 0;
+        aTmpA.nLen++;
+    }
+
+    aTmpB.Mult( rB, nMult);
+
+    for (j = aTmpA.nLen - 1; j >= nLenB; j--)
+    { // Guess divisor
+        sal_uInt32 nTmp = ( static_cast<sal_uInt32>(aTmpA.nNum[j]) << 16 ) + aTmpA.nNum[j - 1];
+        if (aTmpA.nNum[j] == aTmpB.nNum[nLenB1])
+            nQ = 0xFFFF;
+        else
+            nQ = static_cast<sal_uInt16>(nTmp / aTmpB.nNum[nLenB1]);
+
+        if ( (static_cast<sal_uInt32>(aTmpB.nNum[nLenB1 - 1]) * nQ) >
+            ((nTmp - aTmpB.nNum[nLenB1] * nQ) << 16) + aTmpA.nNum[j - 2])
+            nQ--;
+        // Start division
+        nK = 0;
+        for (i = 0; i < nLenB; i++)
+        {
+            nTmp = static_cast<sal_uInt32>(aTmpA.nNum[j - nLenB + i])
+                   - (static_cast<sal_uInt32>(aTmpB.nNum[i]) * nQ)
+                   - nK;
+            aTmpA.nNum[j - nLenB + i] = static_cast<sal_uInt16>(nTmp);
+            nK = static_cast<sal_uInt16>(nTmp >> 16);
+            if ( nK )
+                nK = static_cast<sal_uInt16>(0x10000U - nK);
+        }
+        sal_uInt16& rNum( aTmpA.nNum[j - nLenB + i] );
+        rNum = rNum - nK;
+        if (aTmpA.nNum[j - nLenB + i] == 0)
+            rErg.nNum[j - nLenB] = nQ;
+        else
+        {
+            rErg.nNum[j - nLenB] = nQ - 1;
+            nK = 0;
+            for (i = 0; i < nLenB; i++) {
+                nTmp = aTmpA.nNum[j - nLenB + i] + aTmpB.nNum[i] + nK;
+                aTmpA.nNum[j - nLenB + i] = static_cast<sal_uInt16>(nTmp & 0xFFFFL);
+                if (nTmp & 0xFFFF0000L)
+                    nK = 1;
+                else
+                    nK = 0;
+            }
+        }
+    }
+
+    rErg = aTmpA;
+    rErg.Div( nMult, nQ );
+}
+
+bool BigInt::ABS_IsLess( const BigInt& rB ) const
+{
+    if (nLen != 0 || rB.nLen != 0)
+    {
+        BigInt nA, nB;
+        nA.MakeBigInt( *this );
+        nB.MakeBigInt( rB );
+        if (nA.nLen == nB.nLen)
+        {
+            int i;
+            for (i = nA.nLen - 1; i > 0 && nA.nNum[i] == nB.nNum[i]; i--)
+            {
+            }
+            return nA.nNum[i] < nB.nNum[i];
+        }
+        else
+            return nA.nLen < nB.nLen;
+    }
+    if ( nVal < 0 )
+        if ( rB.nVal < 0 )
+            return nVal > rB.nVal;
+        else
+            return nVal > -rB.nVal;
+    else
+        if ( rB.nVal < 0 )
+            return nVal < -rB.nVal;
+        else
+            return nVal < rB.nVal;
+}
+
+BigInt::BigInt( const BigInt& rBigInt )
+    : nLen(0)
+    , bIsNeg(false)
+{
+    if ( rBigInt.nLen != 0 )
+        memcpy( static_cast<void*>(this), static_cast<const void*>(&rBigInt), sizeof( BigInt ) );
+    else
+        nVal = rBigInt.nVal;
+}
+
+BigInt::BigInt( std::u16string_view rString )
+    : nLen(0)
+{
+    bIsNeg = false;
+    nVal   = 0;
+
+    bool bNeg = false;
+    auto p = rString.begin();
+    auto pEnd = rString.end();
+    if (p == pEnd)
+        return;
+    if ( *p == '-' )
+    {
+        bNeg = true;
+        p++;
+    }
+    if (p == pEnd)
+        return;
+    while( p != pEnd && *p >= '0' && *p <= '9' )
+    {
+        *this *= 10;
+        *this += *p - '0';
+        p++;
+    }
+    if ( nLen != 0 )
+        bIsNeg = bNeg;
+    else if( bNeg )
+        nVal = -nVal;
+}
+
+BigInt::BigInt( double nValue )
+    : nVal(0)
+{
+    if ( nValue < 0 )
+    {
+        nValue *= -1;
+        bIsNeg  = true;
+    }
+    else
+    {
+        bIsNeg  = false;
+    }
+
+    if ( nValue < 1 )
+    {
+        nVal   = 0;
+        nLen   = 0;
+    }
+    else
+    {
+        int i=0;
+
+        while ( ( nValue > 65536.0 ) && ( i < MAX_DIGITS ) )
+        {
+            nNum[i] = static_cast<sal_uInt16>(fmod( nValue, 65536.0 ));
+            nValue -= nNum[i];
+            nValue /= 65536.0;
+            i++;
+        }
+        if ( i < MAX_DIGITS )
+            nNum[i++] = static_cast<sal_uInt16>(nValue);
+
+        nLen = i;
+
+        if ( i < 3 )
+            Normalize();
+    }
+}
+
+BigInt::BigInt( sal_uInt32 nValue )
+    : nVal(0)
+{
+    if ( nValue & 0x80000000U )
+    {
+        bIsNeg  = false;
+        nNum[0] = static_cast<sal_uInt16>(nValue & 0xffffU);
+        nNum[1] = static_cast<sal_uInt16>(nValue >> 16);
+        nLen    = 2;
+    }
+    else
+    {
+        bIsNeg = false;
+        nVal   = nValue;
+        nLen   = 0;
+    }
+}
+
+BigInt::BigInt( sal_Int64 nValue )
+    : nVal(0)
+{
+    bIsNeg = nValue < 0;
+    nLen = 0;
+
+    if ((nValue >= SAL_MIN_INT32) && (nValue <= SAL_MAX_INT32))
+    {
+        nVal = static_cast<sal_Int32>(nValue);
+    }
+    else
+    {
+        sal_uInt64 nUValue = static_cast<sal_uInt64>(bIsNeg ? -nValue : nValue);
+        for (int i = 0; (i != sizeof(sal_uInt64) / 2) && (nUValue != 0); ++i)
+        {
+            nNum[i] = static_cast<sal_uInt16>(nUValue & 0xffffUL);
+            nUValue = nUValue >> 16;
+            ++nLen;
+        }
+    }
+}
+
+BigInt::operator double() const
+{
+    if ( nLen == 0 )
+        return static_cast<double>(nVal);
+    else
+    {
+        int     i = nLen-1;
+        double  nRet = static_cast<double>(static_cast<sal_uInt32>(nNum[i]));
+
+        while ( i )
+        {
+            nRet *= 65536.0;
+            i--;
+            nRet += static_cast<double>(static_cast<sal_uInt32>(nNum[i]));
+        }
+
+        if ( bIsNeg )
+            nRet *= -1;
+
+        return nRet;
+    }
+}
+
+BigInt& BigInt::operator=( const BigInt& rBigInt )
+{
+    if (this == &rBigInt)
+        return *this;
+
+    if ( rBigInt.nLen != 0 )
+        memcpy( static_cast<void*>(this), static_cast<const void*>(&rBigInt), sizeof( BigInt ) );
+    else
+    {
+        nLen = 0;
+        nVal = rBigInt.nVal;
+    }
+    return *this;
+}
+
+BigInt& BigInt::operator+=( const BigInt& rVal )
+{
+    if ( nLen == 0 && rVal.nLen == 0 )
+    {
+        if( nVal <= MY_MAXLONG && rVal.nVal <= MY_MAXLONG
+            && nVal >= MY_MINLONG && rVal.nVal >= MY_MINLONG )
+        { // No overflows may occur here
+            nVal += rVal.nVal;
+            return *this;
+        }
+
+        if( (nVal < 0) != (rVal.nVal < 0) )
+        { // No overflows may occur here
+            nVal += rVal.nVal;
+            return *this;
+        }
+    }
+
+    BigInt aTmp1, aTmp2;
+    aTmp1.MakeBigInt( *this );
+    aTmp2.MakeBigInt( rVal );
+    aTmp1.AddLong( aTmp2, *this );
+    Normalize();
+    return *this;
+}
+
+BigInt& BigInt::operator-=( const BigInt& rVal )
+{
+    if ( nLen == 0 && rVal.nLen == 0 )
+    {
+        if ( nVal <= MY_MAXLONG && rVal.nVal <= MY_MAXLONG &&
+             nVal >= MY_MINLONG && rVal.nVal >= MY_MINLONG )
+        { // No overflows may occur here
+            nVal -= rVal.nVal;
+            return *this;
+        }
+
+        if ( (nVal < 0) == (rVal.nVal < 0) )
+        { // No overflows may occur here
+            nVal -= rVal.nVal;
+            return *this;
+        }
+    }
+
+    BigInt aTmp1, aTmp2;
+    aTmp1.MakeBigInt( *this );
+    aTmp2.MakeBigInt( rVal );
+    aTmp1.SubLong( aTmp2, *this );
+    Normalize();
+    return *this;
+}
+
+BigInt& BigInt::operator*=( const BigInt& rVal )
+{
+    static const sal_Int32 MY_MAXSHORT = 0x00007fff;
+    static const sal_Int32 MY_MINSHORT = -MY_MAXSHORT;
+
+    if ( nLen == 0 && rVal.nLen == 0
+         && nVal <= MY_MAXSHORT && rVal.nVal <= MY_MAXSHORT
+         && nVal >= MY_MINSHORT && rVal.nVal >= MY_MINSHORT )
+         // TODO: not optimal !!! W.P.
+    { // No overflows may occur here
+        nVal *= rVal.nVal;
+    }
+    else
+    {
+        BigInt aTmp1, aTmp2;
+        aTmp1.MakeBigInt( rVal );
+        aTmp2.MakeBigInt( *this );
+        aTmp1.MultLong(aTmp2, *this);
+        Normalize();
+    }
+    return *this;
+}
+
+BigInt& BigInt::operator/=( const BigInt& rVal )
+{
+    if ( rVal.nLen == 0 )
+    {
+        if ( rVal.nVal == 0 )
+        {
+            OSL_FAIL( "BigInt::operator/ --> divide by zero" );
+            return *this;
+        }
+
+        if ( nLen == 0 )
+        {
+            // No overflows may occur here
+            nVal /= rVal.nVal;
+            return *this;
+        }
+
+        if ( rVal.nVal == 1 )
+            return *this;
+
+        if ( rVal.nVal == -1 )
+        {
+            bIsNeg = !bIsNeg;
+            return *this;
+        }
+
+        if ( rVal.nVal <= 0xFFFF && rVal.nVal >= -0xFFFF )
+        {
+            // Divide BigInt with an sal_uInt16
+            sal_uInt16 nTmp;
+            if ( rVal.nVal < 0 )
+            {
+                nTmp = static_cast<sal_uInt16>(-rVal.nVal);
+                bIsNeg = !bIsNeg;
+            }
+            else
+                nTmp = static_cast<sal_uInt16>(rVal.nVal);
+
+            Div( nTmp, nTmp );
+            Normalize();
+            return *this;
+        }
+    }
+
+    if ( ABS_IsLess( rVal ) )
+    {
+        *this = BigInt( 0 );
+        return *this;
+    }
+
+    // Divide BigInt with BigInt
+    BigInt aTmp1, aTmp2;
+    aTmp1.MakeBigInt( *this );
+    aTmp2.MakeBigInt( rVal );
+    aTmp1.DivLong(aTmp2, *this);
+    Normalize();
+    return *this;
+}
+
+BigInt& BigInt::operator%=( const BigInt& rVal )
+{
+    if ( rVal.nLen == 0 )
+    {
+        if ( rVal.nVal == 0 )
+        {
+            OSL_FAIL( "BigInt::operator/ --> divide by zero" );
+            return *this;
+        }
+
+        if ( nLen == 0 )
+        {
+            // No overflows may occur here
+            nVal %= rVal.nVal;
+            return *this;
+        }
+
+        if ( rVal.nVal <= 0xFFFF && rVal.nVal >= -0xFFFF )
+        {
+            // Divide Bigint by int16
+            sal_uInt16 nTmp;
+            if ( rVal.nVal < 0 )
+            {
+                nTmp = static_cast<sal_uInt16>(-rVal.nVal);
+                bIsNeg = !bIsNeg;
+            }
+            else
+                nTmp = static_cast<sal_uInt16>(rVal.nVal);
+
+            Div( nTmp, nTmp );
+            *this = BigInt( nTmp );
+            return *this;
+        }
+    }
+
+    if ( ABS_IsLess( rVal ) )
+        return *this;
+
+    // Divide BigInt with BigInt
+    BigInt aTmp1, aTmp2;
+    aTmp1.MakeBigInt( *this );
+    aTmp2.MakeBigInt( rVal );
+    aTmp1.ModLong(aTmp2, *this);
+    Normalize();
+    return *this;
+}
+
+bool operator==( const BigInt& rVal1, const BigInt& rVal2 )
+{
+    if (rVal1.nLen == 0 && rVal2.nLen == 0)
+        return rVal1.nVal == rVal2.nVal;
+
+    BigInt nA, nB;
+    nA.MakeBigInt(rVal1);
+    nB.MakeBigInt(rVal2);
+    return nA.bIsNeg == nB.bIsNeg && nA.nLen == nB.nLen
+           && std::equal(nA.nNum, nA.nNum + nA.nLen, nB.nNum);
+}
+
+bool operator<( const BigInt& rVal1, const BigInt& rVal2 )
+{
+    if (rVal1.nLen == 0 && rVal2.nLen == 0)
+        return rVal1.nVal < rVal2.nVal;
+
+    BigInt nA, nB;
+    nA.MakeBigInt(rVal1);
+    nB.MakeBigInt(rVal2);
+    if (nA.bIsNeg != nB.bIsNeg)
+        return !nB.bIsNeg;
+    if (nA.nLen != nB.nLen)
+        return nA.bIsNeg ? (nA.nLen > nB.nLen) : (nA.nLen < nB.nLen);
+    int i = nA.nLen - 1;
+    while (i > 0 && nA.nNum[i] == nB.nNum[i])
+        --i;
+    return nA.bIsNeg ? (nA.nNum[i] > nB.nNum[i]) : (nA.nNum[i] < nB.nNum[i]);
+}
+
+tools::Long BigInt::Scale( tools::Long nVal, tools::Long nMul, tools::Long nDiv )
+{
+    BigInt aVal( nVal );
+
+    aVal *= nMul;
+
+    if ( aVal.IsNeg() != ( nDiv < 0 ) )
+        aVal -= nDiv / 2; // for correct rounding
+    else
+        aVal += nDiv / 2; // for correct rounding
+
+    aVal /= nDiv;
+
+    return tools::Long( aVal );
+}
+
+/* vim:set shiftwidth=4 softtabstop=4 expandtab: */
diff --git a/tools/source/generic/color.cxx b/tools/source/generic/color.cxx
new file mode 100644
index 0000000000..421045b588
--- /dev/null
+++ b/tools/source/generic/color.cxx
@@ -0,0 +1,266 @@
+/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
+/*
+ * This file is part of the LibreOffice project.
+ *
+ * This Source Code Form is subject to the terms of the Mozilla Public
+ * License, v. 2.0. If a copy of the MPL was not distributed with this
+ * file, You can obtain one at http://mozilla.org/MPL/2.0/.
+ *
+ * This file incorporates work covered by the following license notice:
+ *
+ *   Licensed to the Apache Software Foundation (ASF) under one or more
+ *   contributor license agreements. See the NOTICE file distributed
+ *   with this work for additional information regarding copyright
+ *   ownership. The ASF licenses this file to you under the Apache
+ *   License, Version 2.0 (the "License"); you may not use this file
+ *   except in compliance with the License. You may obtain a copy of
+ *   the License at http://www.apache.org/licenses/LICENSE-2.0 .
+ */
+
+#include <sal/config.h>
+
+#include <algorithm>
+#include <iomanip>
+#include <sstream>
+#include <stdlib.h>
+
+#include <tools/color.hxx>
+#include <tools/helpers.hxx>
+#include <tools/long.hxx>
+#include <o3tl/string_view.hxx>
+#include <basegfx/color/bcolortools.hxx>
+
+void Color::IncreaseLuminance(sal_uInt8 cLumInc)
+{
+    R = sal_uInt8(std::clamp(R + cLumInc, 0, 255));
+    G = sal_uInt8(std::clamp(G + cLumInc, 0, 255));
+    B = sal_uInt8(std::clamp(B + cLumInc, 0, 255));
+}
+
+void Color::DecreaseLuminance(sal_uInt8 cLumDec)
+{
+    R = sal_uInt8(std::clamp(R - cLumDec, 0, 255));
+    G = sal_uInt8(std::clamp(G - cLumDec, 0, 255));
+    B = sal_uInt8(std::clamp(B - cLumDec, 0, 255));
+}
+
+void Color::DecreaseContrast(sal_uInt8 nContDec)
+{
+    if (nContDec)
+    {
+        const double fM = (128.0 - 0.4985 * nContDec) / 128.0;
+        const double fOff = 128.0 - fM * 128.0;
+
+        R = sal_uInt8(std::clamp(FRound(R * fM + fOff), tools::Long(0), tools::Long(255)));
+        G = sal_uInt8(std::clamp(FRound(G * fM + fOff), tools::Long(0), tools::Long(255)));
+        B = sal_uInt8(std::clamp(FRound(B * fM + fOff), tools::Long(0), tools::Long(255)));
+    }
+}
+
+// color space conversion
+
+void Color::RGBtoHSB( sal_uInt16& nHue, sal_uInt16& nSat, sal_uInt16& nBri ) const
+{
+    sal_uInt8 c[3];
+    sal_uInt8 cMax, cMin;
+
+    c[0] = R;
+    c[1] = G;
+    c[2] = B;
+
+    cMax = c[0];
+    if( c[1] > cMax )
+        cMax = c[1];
+    if( c[2] > cMax )
+        cMax = c[2];
+
+    // Brightness = max(R, G, B);
+    nBri = cMax * 100 / 255;
+
+    cMin = c[0];
+    if( c[1] < cMin )
+        cMin = c[1];
+    if( c[2] < cMin )
+        cMin = c[2];
+
+    sal_uInt8 cDelta = cMax - cMin;
+
+    // Saturation = max - min / max
+    if( nBri > 0 )
+        nSat = cDelta * 100 / cMax;
+    else
+        nSat = 0;
+
+    if( nSat == 0 )
+        nHue = 0; // Default = undefined
+    else
+    {
+        double dHue = 0.0;
+
+        if( c[0] == cMax )
+        {
+            dHue = static_cast<double>( c[1] - c[2] ) / static_cast<double>(cDelta);
+        }
+        else if( c[1] == cMax )
+        {
+            dHue = 2.0 + static_cast<double>( c[2] - c[0] ) / static_cast<double>(cDelta);
+        }
+        else if ( c[2] == cMax )
+        {
+            dHue = 4.0 + static_cast<double>( c[0] - c[1] ) / static_cast<double>(cDelta);
+        }
+        dHue *= 60.0;
+
+        if( dHue < 0.0 )
+            dHue += 360.0;
+
+        nHue = static_cast<sal_uInt16>(dHue);
+    }
+}
+
+Color Color::HSBtoRGB( sal_uInt16 nHue, sal_uInt16 nSat, sal_uInt16 nBri )
+{
+    sal_uInt8 cR=0,cG=0,cB=0;
+    sal_uInt8 nB = static_cast<sal_uInt8>( nBri * 255 / 100 );
+
+    if( nSat == 0 )
+    {
+        cR = nB;
+        cG = nB;
+        cB = nB;
+    }
+    else
+    {
+        double dH = nHue;
+        double f;
+        sal_uInt16 n;
+        if( dH == 360.0 )
+            dH = 0.0;
+
+        dH /= 60.0;
+        n = static_cast<sal_uInt16>(dH);
+        f = dH - n;
+
+        sal_uInt8 a = static_cast<sal_uInt8>( nB * ( 100 - nSat ) / 100 );
+        sal_uInt8 b = static_cast<sal_uInt8>( nB * ( 100 - ( static_cast<double>(nSat) * f ) ) / 100 );
+        sal_uInt8 c = static_cast<sal_uInt8>( nB * ( 100 - ( static_cast<double>(nSat) * ( 1.0 - f ) ) ) / 100 );
+
+        switch( n )
+        {
+            case 0: cR = nB;    cG = c;     cB = a;     break;
+            case 1: cR = b;     cG = nB;    cB = a;     break;
+            case 2: cR = a;     cG = nB;    cB = c;     break;
+            case 3: cR = a;     cG = b;     cB = nB;    break;
+            case 4: cR = c;     cG = a;     cB = nB;    break;
+            case 5: cR = nB;    cG = a;     cB = b;     break;
+        }
+    }
+
+    return Color( cR, cG, cB );
+}
+
+Color Color::STRtoRGB(std::u16string_view colorname)
+{
+    Color col;
+    if(colorname.empty()) return col;
+
+    switch(colorname.size()){
+        case 7:
+            col.mValue = o3tl::toUInt32(colorname.substr(1,6), 16);
+            break;
+        case 6:
+            col.mValue = o3tl::toUInt32(colorname, 16);
+            break;
+        case 4:
+        {
+            sal_Unicode data[6] = { colorname[1], colorname[1], colorname[2],
+                                     colorname[2], colorname[3], colorname[3] };
+            col.mValue = o3tl::toUInt32(std::u16string_view(data,6), 16);
+            break;
+        }
+        case 3:
+        {
+            sal_Unicode data[6] = { colorname[0], colorname[0], colorname[1],
+                                     colorname[1], colorname[2], colorname[2] };
+            col.mValue = o3tl::toUInt32(std::u16string_view(data,6), 16);
+            break;
+        }
+        default:
+            break;
+    }
+    return col;
+}
+
+OUString Color::AsRGBHexString() const
+{
+    std::stringstream ss;
+    ss << std::hex << std::setfill ('0') << std::setw(6) << sal_uInt32(GetRGBColor());
+    return OUString::createFromAscii(ss.str());
+}
+
+OUString Color::AsRGBHEXString() const
+{
+    std::stringstream ss;
+    ss << std::hex << std::uppercase << std::setfill ('0') << std::setw(6) << sal_uInt32(GetRGBColor());
+    return OUString::createFromAscii(ss.str());
+}
+
+void Color::ApplyTintOrShade(sal_Int16 n100thPercent)
+{
+    if (n100thPercent == 0)
+        return;
+
+    basegfx::BColor aBColor = basegfx::utils::rgb2hsl(getBColor());
+    double fFactor = 1.0 - (std::abs(double(n100thPercent)) / 10000.0);
+    double fResult;
+
+    if (n100thPercent > 0) // tint
+    {
+        fResult = aBColor.getBlue() * fFactor + (1.0 - fFactor);
+    }
+    else // shade
+    {
+        fResult = aBColor.getBlue() * fFactor;
+    }
+
+    aBColor.setBlue(fResult);
+    aBColor = basegfx::utils::hsl2rgb(aBColor);
+
+    R = sal_uInt8(std::lround(aBColor.getRed()   * 255.0));
+    G = sal_uInt8(std::lround(aBColor.getGreen() * 255.0));
+    B = sal_uInt8(std::lround(aBColor.getBlue()  * 255.0));
+}
+
+void Color::ApplyLumModOff(sal_Int16 nMod, sal_Int16 nOff)
+{
+    if (nMod == 10000 && nOff == 0)
+    {
+        return;
+    }
+    // Switch to HSL, where applying these transforms is easier.
+    basegfx::BColor aBColor = basegfx::utils::rgb2hsl(getBColor());
+
+    // 50% is half luminance, 200% is double luminance. Unit is 100th percent.
+    aBColor.setBlue(std::clamp(aBColor.getBlue() * nMod / 10000, 0.0, 1.0));
+    // If color changes to black or white, it will stay gray if luminance changes again.
+    if ((aBColor.getBlue() == 0.0) || (aBColor.getBlue() == 1.0))
+    {
+        aBColor.setGreen(0.0);
+    }
+
+    // Luminance offset means hue and saturation is left unchanged. Unit is 100th percent.
+    aBColor.setBlue(std::clamp(aBColor.getBlue() + static_cast<double>(nOff) / 10000, 0.0, 1.0));
+    // If color changes to black or white, it will stay gray if luminance changes again.
+    if ((aBColor.getBlue() == 0.0) || (aBColor.getBlue() == 1.0))
+    {
+        aBColor.setGreen(0.0);
+    }
+
+    // Switch back to RGB.
+    aBColor = basegfx::utils::hsl2rgb(aBColor);
+    R = sal_uInt8(std::lround(aBColor.getRed()   * 255.0));
+    G = sal_uInt8(std::lround(aBColor.getGreen() * 255.0));
+    B = sal_uInt8(std::lround(aBColor.getBlue()  * 255.0));
+}
+
+/* vim:set shiftwidth=4 softtabstop=4 expandtab: */
diff --git a/tools/source/generic/config.cxx b/tools/source/generic/config.cxx
new file mode 100644
index 0000000000..cc5cce9c3b
--- /dev/null
+++ b/tools/source/generic/config.cxx
@@ -0,0 +1,944 @@
+/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
+/*
+ * This file is part of the LibreOffice project.
+ *
+ * This Source Code Form is subject to the terms of the Mozilla Public
+ * License, v. 2.0. If a copy of the MPL was not distributed with this
+ * file, You can obtain one at http://mozilla.org/MPL/2.0/.
+ *
+ * This file incorporates work covered by the following license notice:
+ *
+ *   Licensed to the Apache Software Foundation (ASF) under one or more
+ *   contributor license agreements. See the NOTICE file distributed
+ *   with this work for additional information regarding copyright
+ *   ownership. The ASF licenses this file to you under the Apache
+ *   License, Version 2.0 (the "License"); you may not use this file
+ *   except in compliance with the License. You may obtain a copy of
+ *   the License at http://www.apache.org/licenses/LICENSE-2.0 .
+ */
+
+#include <cstddef>
+#include <cstdlib>
+#include <string.h>
+
+#ifdef _WIN32
+#include <stdlib.h>
+#endif
+
+#include <osl/file.hxx>
+#include <tools/config.hxx>
+#include <sal/log.hxx>
+
+namespace {
+
+struct ImplKeyData
+{
+    ImplKeyData*    mpNext;
+    OString         maKey;
+    OString         maValue;
+    bool            mbIsComment;
+};
+
+}
+
+struct ImplGroupData
+{
+    ImplGroupData*  mpNext;
+    ImplKeyData*    mpFirstKey;
+    OString         maGroupName;
+    sal_uInt16      mnEmptyLines;
+};
+
+struct ImplConfigData
+{
+    ImplGroupData*  mpFirstGroup;
+    OUString        maFileName;
+    sal_uInt32      mnDataUpdateId;
+    sal_uInt32      mnTimeStamp;
+    bool            mbModified;
+    bool            mbRead;
+    bool            mbIsUTF8BOM;
+};
+
+static OUString toUncPath( const OUString& rPath )
+{
+    OUString aFileURL;
+
+    // check if rFileName is already a URL; if not make it so
+    if( rPath.startsWith( "file://"))
+    {
+        aFileURL = rPath;
+    }
+    else if( ::osl::FileBase::getFileURLFromSystemPath( rPath, aFileURL ) != ::osl::FileBase::E_None )
+    {
+        aFileURL = rPath;
+    }
+    return aFileURL;
+}
+
+static sal_uInt32 ImplSysGetConfigTimeStamp( const OUString& rFileName )
+{
+    sal_uInt32 nTimeStamp = 0;
+    ::osl::DirectoryItem aItem;
+    ::osl::FileStatus aStatus( osl_FileStatus_Mask_ModifyTime );
+
+    if( ::osl::DirectoryItem::get( rFileName, aItem ) == ::osl::FileBase::E_None &&
+        aItem.getFileStatus( aStatus ) == ::osl::FileBase::E_None )
+    {
+        nTimeStamp = aStatus.getModifyTime().Seconds;
+    }
+
+    return nTimeStamp;
+}
+
+static std::unique_ptr<sal_uInt8[]> ImplSysReadConfig( const OUString& rFileName,
+                                sal_uInt64& rRead, bool& rbRead, bool& rbIsUTF8BOM, sal_uInt32& rTimeStamp )
+{
+    std::unique_ptr<sal_uInt8[]> pBuf;
+    ::osl::File aFile( rFileName );
+
+    if( aFile.open( osl_File_OpenFlag_Read ) == ::osl::FileBase::E_None )
+    {
+        sal_uInt64 nPos = 0;
+        if( aFile.getSize( nPos ) == ::osl::FileBase::E_None )
+        {
+            if (nPos > SAL_MAX_SIZE) {
+                aFile.close();
+                return nullptr;
+            }
+            pBuf.reset(new sal_uInt8[static_cast< std::size_t >(nPos)]);
+            sal_uInt64 nRead = 0;
+            if( aFile.read( pBuf.get(), nPos, nRead ) == ::osl::FileBase::E_None && nRead == nPos )
+            {
+                //skip the byte-order-mark 0xEF 0xBB 0xBF, if it was UTF8 files
+                unsigned char const BOM[3] = {0xEF, 0xBB, 0xBF};
+                if (nRead > 2 && memcmp(pBuf.get(), BOM, 3) == 0)
+                {
+                    nRead -= 3;
+                    memmove(pBuf.get(), pBuf.get() + 3, sal::static_int_cast<std::size_t>(nRead * sizeof(sal_uInt8)) );
+                    rbIsUTF8BOM = true;
+                }
+
+                rTimeStamp = ImplSysGetConfigTimeStamp( rFileName );
+                rbRead = true;
+                rRead = nRead;
+            }
+            else
+            {
+                pBuf.reset();
+            }
+        }
+        aFile.close();
+    }
+
+    return pBuf;
+}
+
+static bool ImplSysWriteConfig( const OUString& rFileName,
+                                const sal_uInt8* pBuf, sal_uInt32 nBufLen, bool rbIsUTF8BOM, sal_uInt32& rTimeStamp )
+{
+    bool bSuccess = false;
+    bool bUTF8BOMSuccess = false;
+
+    ::osl::File aFile( rFileName );
+    ::osl::FileBase::RC eError = aFile.open( osl_File_OpenFlag_Write | osl_File_OpenFlag_Create );
+    if( eError != ::osl::FileBase::E_None )
+        eError = aFile.open( osl_File_OpenFlag_Write );
+    if( eError == ::osl::FileBase::E_None )
+    {
+        // truncate
+        aFile.setSize( 0 );
+        sal_uInt64 nWritten;
+
+        //write the byte-order-mark 0xEF 0xBB 0xBF first , if it was UTF8 files
+        if ( rbIsUTF8BOM )
+        {
+            unsigned char const BOM[3] = {0xEF, 0xBB, 0xBF};
+            sal_uInt64 nUTF8BOMWritten;
+            if( aFile.write( BOM, 3, nUTF8BOMWritten ) == ::osl::FileBase::E_None && 3 == nUTF8BOMWritten )
+            {
+                bUTF8BOMSuccess = true;
+            }
+        }
+
+        if( aFile.write( pBuf, nBufLen, nWritten ) == ::osl::FileBase::E_None && nWritten == nBufLen )
+        {
+            bSuccess = true;
+        }
+        if ( rbIsUTF8BOM ? bSuccess && bUTF8BOMSuccess : bSuccess )
+        {
+            rTimeStamp = ImplSysGetConfigTimeStamp( rFileName );
+        }
+    }
+
+    return rbIsUTF8BOM ? bSuccess && bUTF8BOMSuccess : bSuccess;
+}
+
+namespace {
+OString makeOString(const sal_uInt8* p, sal_uInt64 n)
+{
+    if (n > SAL_MAX_INT32)
+    {
+        #ifdef _WIN32
+        abort();
+        #else
+        ::std::abort(); //TODO: handle this gracefully
+        #endif
+    }
+    return OString(
+        reinterpret_cast< char const * >(p),
+        sal::static_int_cast< sal_Int32 >(n));
+}
+}
+
+static void ImplMakeConfigList( ImplConfigData* pData,
+                                const sal_uInt8* pBuf, sal_uInt64 nLen )
+{
+    if ( !nLen )
+        return;
+
+    // Parse buffer and build config list
+    sal_uInt64 nStart;
+    sal_uInt64 nLineLen;
+    sal_uInt64 nNameLen;
+    sal_uInt64 nKeyLen;
+    sal_uInt64 i;
+    const sal_uInt8*    pLine;
+    ImplKeyData*    pPrevKey = nullptr;
+    ImplGroupData*  pPrevGroup = nullptr;
+    ImplGroupData*  pGroup = nullptr;
+    i = 0;
+    while ( i < nLen )
+    {
+        // Ctrl+Z
+        if ( pBuf[i] == 0x1A )
+            break;
+
+        // Remove spaces and tabs
+        while ( (pBuf[i] == ' ') || (pBuf[i] == '\t') )
+            i++;
+
+        // remember line-starts
+        nStart = i;
+        pLine = pBuf+i;
+
+        // search line-endings
+        while (  (i < nLen) && pBuf[i] && (pBuf[i] != '\r') && (pBuf[i] != '\n') &&
+                (pBuf[i] != 0x1A) )
+            i++;
+
+        nLineLen = i-nStart;
+
+        // if Line-ending is found, continue once
+        if ( (i+1 < nLen) &&
+             (pBuf[i] != pBuf[i+1]) &&
+             ((pBuf[i+1] == '\r') || (pBuf[i+1] == '\n')) )
+            i++;
+        i++;
+
+        // evaluate line
+        if ( *pLine == '[' )
+        {
+            pGroup               = new ImplGroupData;
+            pGroup->mpNext       = nullptr;
+            pGroup->mpFirstKey   = nullptr;
+            pGroup->mnEmptyLines = 0;
+            if ( pPrevGroup )
+                pPrevGroup->mpNext = pGroup;
+            else
+                pData->mpFirstGroup = pGroup;
+            pPrevGroup  = pGroup;
+            pPrevKey    = nullptr;
+
+            // filter group names
+            pLine++;
+            nLineLen--;
+            // remove spaces and tabs
+            while ( (*pLine == ' ') || (*pLine == '\t') )
+            {
+                nLineLen--;
+                pLine++;
+            }
+            nNameLen = 0;
+            while ( (nNameLen < nLineLen) && (pLine[nNameLen] != ']') )
+                nNameLen++;
+            if ( nNameLen )
+            {
+                while ( (pLine[nNameLen-1] == ' ') || (pLine[nNameLen-1] == '\t') )
+                    nNameLen--;
+            }
+            pGroup->maGroupName = makeOString(pLine, nNameLen);
+        }
+        else
+        {
+            if ( nLineLen )
+            {
+                // If no group exists yet, add to default
+                if ( !pGroup )
+                {
+                    pGroup              = new ImplGroupData;
+                    pGroup->mpNext      = nullptr;
+                    pGroup->mpFirstKey  = nullptr;
+                    pGroup->mnEmptyLines = 0;
+                    pData->mpFirstGroup = pGroup;
+                    pPrevGroup  = pGroup;
+                    pPrevKey    = nullptr;
+                }
+
+                // if empty line, append it
+                if ( pPrevKey )
+                {
+                    while ( pGroup->mnEmptyLines )
+                    {
+                        ImplKeyData* pKey = new ImplKeyData;
+                        pKey->mbIsComment   = true;
+                        pPrevKey->mpNext    = pKey;
+                        pPrevKey            = pKey;
+                        pGroup->mnEmptyLines--;
+                    }
+                }
+
+                // Generate new key
+                ImplKeyData* pKey = new ImplKeyData;
+                pKey->mpNext = nullptr;
+                if ( pPrevKey )
+                    pPrevKey->mpNext = pKey;
+                else
+                    pGroup->mpFirstKey = pKey;
+                pPrevKey = pKey;
+                if ( pLine[0] == ';' )
+                {
+                    pKey->maValue = makeOString(pLine, nLineLen);
+                    pKey->mbIsComment = true;
+                }
+                else
+                {
+                    pKey->mbIsComment = false;
+                    nNameLen = 0;
+                    while ( (nNameLen < nLineLen) && (pLine[nNameLen] != '=') )
+                        nNameLen++;
+                    nKeyLen = nNameLen;
+                    // Remove spaces and tabs
+                    if ( nNameLen )
+                    {
+                        while ( (pLine[nNameLen-1] == ' ') || (pLine[nNameLen-1] == '\t') )
+                            nNameLen--;
+                    }
+                    pKey->maKey = makeOString(pLine, nNameLen);
+                    nKeyLen++;
+                    if ( nKeyLen < nLineLen )
+                    {
+                        pLine += nKeyLen;
+                        nLineLen -= nKeyLen;
+                        // Remove spaces and tabs
+                        while ( (*pLine == ' ') || (*pLine == '\t') )
+                        {
+                            nLineLen--;
+                            pLine++;
+                        }
+                        if ( nLineLen )
+                        {
+                            while ( (pLine[nLineLen-1] == ' ') || (pLine[nLineLen-1] == '\t') )
+                                nLineLen--;
+                            pKey->maValue = makeOString(pLine, nLineLen);
+                        }
+                    }
+                }
+            }
+            else
+            {
+                // Spaces are counted and appended only after key generation,
+                // as we want to store spaces even after adding new keys
+                if ( pGroup )
+                    pGroup->mnEmptyLines++;
+            }
+        }
+    }
+}
+
+static std::unique_ptr<sal_uInt8[]> ImplGetConfigBuffer( const ImplConfigData* pData, sal_uInt32& rLen )
+{
+    std::unique_ptr<sal_uInt8[]> pWriteBuf;
+    sal_uInt8*      pBuf;
+    sal_uInt8       aLineEndBuf[2] = {0, 0};
+    ImplKeyData*    pKey;
+    ImplGroupData*  pGroup;
+    sal_uInt32      nBufLen;
+    sal_uInt32      nValueLen;
+    sal_uInt32      nKeyLen;
+    sal_uInt32      nLineEndLen;
+
+    aLineEndBuf[0] = '\r';
+    aLineEndBuf[1] = '\n';
+    nLineEndLen = 2;
+
+    nBufLen = 0;
+    pGroup = pData->mpFirstGroup;
+    while ( pGroup )
+    {
+        // Don't write empty groups
+        if ( pGroup->mpFirstKey )
+        {
+            nBufLen += pGroup->maGroupName.getLength() + nLineEndLen + 2;
+            pKey = pGroup->mpFirstKey;
+            while ( pKey )
+            {
+                nValueLen = pKey->maValue.getLength();
+                if ( pKey->mbIsComment )
+                    nBufLen += nValueLen + nLineEndLen;
+                else
+                    nBufLen += pKey->maKey.getLength() + nValueLen + nLineEndLen + 1;
+
+                pKey = pKey->mpNext;
+            }
+
+            // Write empty lines after each group
+            if ( !pGroup->mnEmptyLines )
+                pGroup->mnEmptyLines = 1;
+            nBufLen += nLineEndLen * pGroup->mnEmptyLines;
+        }
+
+        pGroup = pGroup->mpNext;
+    }
+
+    // Output buffer length
+    rLen = nBufLen;
+    if ( !nBufLen )
+    {
+        pWriteBuf.reset(new sal_uInt8[nLineEndLen]);
+        pWriteBuf[0] = aLineEndBuf[0];
+        if ( nLineEndLen == 2 )
+            pWriteBuf[1] = aLineEndBuf[1];
+        return pWriteBuf;
+    }
+
+    // Allocate new write buffer (caller frees it)
+    pWriteBuf.reset(new sal_uInt8[nBufLen]);
+
+    // fill buffer
+    pBuf = pWriteBuf.get();
+    pGroup = pData->mpFirstGroup;
+    while ( pGroup )
+    {
+        // Don't write empty groups
+        if ( pGroup->mpFirstKey )
+        {
+            *pBuf = '[';    pBuf++;
+            memcpy( pBuf, pGroup->maGroupName.getStr(), pGroup->maGroupName.getLength() );
+            pBuf += pGroup->maGroupName.getLength();
+            *pBuf = ']';    pBuf++;
+            *pBuf = aLineEndBuf[0]; pBuf++;
+            if ( nLineEndLen == 2 )
+            {
+                *pBuf = aLineEndBuf[1]; pBuf++;
+            }
+            pKey = pGroup->mpFirstKey;
+            while ( pKey )
+            {
+                nValueLen = pKey->maValue.getLength();
+                if ( pKey->mbIsComment )
+                {
+                    if ( nValueLen )
+                    {
+                        memcpy( pBuf, pKey->maValue.getStr(), nValueLen );
+                        pBuf += nValueLen;
+                    }
+                    *pBuf = aLineEndBuf[0]; pBuf++;
+                    if ( nLineEndLen == 2 )
+                    {
+                        *pBuf = aLineEndBuf[1]; pBuf++;
+                    }
+                }
+                else
+                {
+                    nKeyLen = pKey->maKey.getLength();
+                    memcpy( pBuf, pKey->maKey.getStr(), nKeyLen );
+                    pBuf += nKeyLen;
+                    *pBuf = '=';    pBuf++;
+                    memcpy( pBuf, pKey->maValue.getStr(), nValueLen );
+                    pBuf += nValueLen;
+                    *pBuf = aLineEndBuf[0]; pBuf++;
+                    if ( nLineEndLen == 2 )
+                    {
+                        *pBuf = aLineEndBuf[1]; pBuf++;
+                    }
+                }
+
+                pKey = pKey->mpNext;
+            }
+
+            // Store empty line after each group
+            sal_uInt16 nEmptyLines = pGroup->mnEmptyLines;
+            while ( nEmptyLines )
+            {
+                *pBuf = aLineEndBuf[0]; pBuf++;
+                if ( nLineEndLen == 2 )
+                {
+                    *pBuf = aLineEndBuf[1]; pBuf++;
+                }
+                nEmptyLines--;
+            }
+        }
+
+        pGroup = pGroup->mpNext;
+    }
+
+    return pWriteBuf;
+}
+
+static void ImplReadConfig( ImplConfigData* pData )
+{
+    sal_uInt32   nTimeStamp = 0;
+    sal_uInt64  nRead = 0;
+    bool    bRead = false;
+    bool    bIsUTF8BOM = false;
+    std::unique_ptr<sal_uInt8[]> pBuf = ImplSysReadConfig( pData->maFileName, nRead, bRead, bIsUTF8BOM, nTimeStamp );
+
+    // Read config list from buffer
+    if ( pBuf )
+    {
+        ImplMakeConfigList( pData, pBuf.get(), nRead );
+        pBuf.reset();
+    }
+    pData->mnTimeStamp = nTimeStamp;
+    pData->mbModified  = false;
+    if ( bRead )
+        pData->mbRead = true;
+    if ( bIsUTF8BOM )
+        pData->mbIsUTF8BOM = true;
+}
+
+static void ImplWriteConfig( ImplConfigData* pData )
+{
+    SAL_WARN_IF( pData->mnTimeStamp != ImplSysGetConfigTimeStamp( pData->maFileName ),
+        "tools.generic", "Config overwrites modified configfile: " << pData->maFileName );
+
+    // Read config list from buffer
+    sal_uInt32 nBufLen;
+    std::unique_ptr<sal_uInt8[]> pBuf = ImplGetConfigBuffer( pData, nBufLen );
+    if ( pBuf )
+    {
+        if ( ImplSysWriteConfig( pData->maFileName, pBuf.get(), nBufLen, pData->mbIsUTF8BOM, pData->mnTimeStamp ) )
+            pData->mbModified = false;
+    }
+    else
+        pData->mbModified = false;
+}
+
+static void ImplDeleteConfigData( ImplConfigData* pData )
+{
+    ImplKeyData*    pTempKey;
+    ImplKeyData*    pKey;
+    ImplGroupData*  pTempGroup;
+    ImplGroupData*  pGroup = pData->mpFirstGroup;
+    while ( pGroup )
+    {
+        pTempGroup = pGroup->mpNext;
+
+        // remove all keys
+        pKey = pGroup->mpFirstKey;
+        while ( pKey )
+        {
+            pTempKey = pKey->mpNext;
+            delete pKey;
+            pKey = pTempKey;
+        }
+
+        // remove group and continue
+        delete pGroup;
+        pGroup = pTempGroup;
+    }
+
+    pData->mpFirstGroup = nullptr;
+}
+
+static std::unique_ptr<ImplConfigData> ImplGetConfigData( const OUString& rFileName )
+{
+    std::unique_ptr<ImplConfigData> pData(new ImplConfigData);
+    pData->maFileName       = rFileName;
+    pData->mpFirstGroup     = nullptr;
+    pData->mnDataUpdateId   = 0;
+    pData->mbRead           = false;
+    pData->mbIsUTF8BOM      = false;
+    ImplReadConfig( pData.get() );
+
+    return pData;
+}
+
+bool Config::ImplUpdateConfig() const
+{
+    // Re-read file if timestamp differs
+    if ( mpData->mnTimeStamp != ImplSysGetConfigTimeStamp( maFileName ) )
+    {
+        ImplDeleteConfigData( mpData.get() );
+        ImplReadConfig( mpData.get() );
+        mpData->mnDataUpdateId++;
+        return true;
+    }
+    else
+        return false;
+}
+
+ImplGroupData* Config::ImplGetGroup() const
+{
+    if ( !mpActGroup || (mnDataUpdateId != mpData->mnDataUpdateId) )
+    {
+        ImplGroupData* pPrevGroup = nullptr;
+        ImplGroupData* pGroup = mpData->mpFirstGroup;
+        while ( pGroup )
+        {
+            if ( pGroup->maGroupName.equalsIgnoreAsciiCase(maGroupName) )
+                break;
+
+            pPrevGroup = pGroup;
+            pGroup = pGroup->mpNext;
+        }
+
+        // Add group if not exists
+        if ( !pGroup )
+        {
+            pGroup               = new ImplGroupData;
+            pGroup->mpNext       = nullptr;
+            pGroup->mpFirstKey   = nullptr;
+            pGroup->mnEmptyLines = 1;
+            if ( pPrevGroup )
+                pPrevGroup->mpNext = pGroup;
+            else
+                mpData->mpFirstGroup = pGroup;
+        }
+
+        // Always inherit group names and update cache members
+        pGroup->maGroupName             = maGroupName;
+        const_cast<Config*>(this)->mnDataUpdateId = mpData->mnDataUpdateId;
+        const_cast<Config*>(this)->mpActGroup     = pGroup;
+    }
+
+    return mpActGroup;
+}
+
+Config::Config( const OUString& rFileName )
+{
+    // Initialize config data
+    maFileName      = toUncPath( rFileName );
+    mpData          = ImplGetConfigData( maFileName );
+    mpActGroup      = nullptr;
+    mnDataUpdateId  = 0;
+
+    SAL_INFO("tools.generic", "Config::Config( " << maFileName << " )");
+}
+
+Config::~Config()
+{
+    SAL_INFO("tools.generic", "Config::~Config()" );
+
+    Flush();
+    ImplDeleteConfigData( mpData.get() );
+}
+
+void Config::SetGroup(const OString& rGroup)
+{
+    // If group is to be reset, it needs to be updated on next call
+    if ( maGroupName != rGroup )
+    {
+        maGroupName     = rGroup;
+        mnDataUpdateId  = mpData->mnDataUpdateId-1;
+    }
+}
+
+void Config::DeleteGroup(std::string_view rGroup)
+{
+    // Update config data if necessary
+    if ( !mpData->mbRead )
+    {
+        ImplUpdateConfig();
+        mpData->mbRead = true;
+    }
+
+    ImplGroupData* pPrevGroup = nullptr;
+    ImplGroupData* pGroup = mpData->mpFirstGroup;
+    while ( pGroup )
+    {
+        if ( pGroup->maGroupName.equalsIgnoreAsciiCase(rGroup) )
+            break;
+
+        pPrevGroup = pGroup;
+        pGroup = pGroup->mpNext;
+    }
+
+    if ( !pGroup )
+        return;
+
+    // Remove all keys
+    ImplKeyData* pTempKey;
+    ImplKeyData* pKey = pGroup->mpFirstKey;
+    while ( pKey )
+    {
+        pTempKey = pKey->mpNext;
+        delete pKey;
+        pKey = pTempKey;
+    }
+
+    // Rewire pointers and remove group
+    if ( pPrevGroup )
+        pPrevGroup->mpNext = pGroup->mpNext;
+    else
+        mpData->mpFirstGroup = pGroup->mpNext;
+    delete pGroup;
+
+    // Rewrite config data
+    mpData->mbModified = true;
+
+    mnDataUpdateId = mpData->mnDataUpdateId;
+    mpData->mnDataUpdateId++;
+}
+
+OString Config::GetGroupName(sal_uInt16 nGroup) const
+{
+    ImplGroupData*  pGroup = mpData->mpFirstGroup;
+    sal_uInt16          nGroupCount = 0;
+    OString aGroupName;
+    while ( pGroup )
+    {
+        if ( nGroup == nGroupCount )
+        {
+            aGroupName = pGroup->maGroupName;
+            break;
+        }
+
+        nGroupCount++;
+        pGroup = pGroup->mpNext;
+    }
+
+    return aGroupName;
+}
+
+sal_uInt16 Config::GetGroupCount() const
+{
+    ImplGroupData*  pGroup = mpData->mpFirstGroup;
+    sal_uInt16          nGroupCount = 0;
+    while ( pGroup )
+    {
+        nGroupCount++;
+        pGroup = pGroup->mpNext;
+    }
+
+    return nGroupCount;
+}
+
+bool Config::HasGroup(std::string_view rGroup) const
+{
+    ImplGroupData*  pGroup = mpData->mpFirstGroup;
+    bool            bRet = false;
+
+    while( pGroup )
+    {
+        if( pGroup->maGroupName.equalsIgnoreAsciiCase(rGroup) )
+        {
+            bRet = true;
+            break;
+        }
+
+        pGroup = pGroup->mpNext;
+    }
+
+    return bRet;
+}
+
+OString Config::ReadKey(const OString& rKey) const
+{
+    return ReadKey(rKey, OString());
+}
+
+OString Config::ReadKey(const OString& rKey, const OString& rDefault) const
+{
+    SAL_INFO("tools.generic", "Config::ReadKey( " << rKey << " ) from " << GetGroup()
+                      << " in " << maFileName);
+
+    // Search key, return value if found
+    ImplGroupData* pGroup = ImplGetGroup();
+    if ( pGroup )
+    {
+        ImplKeyData* pKey = pGroup->mpFirstKey;
+        while ( pKey )
+        {
+            if ( !pKey->mbIsComment && pKey->maKey.equalsIgnoreAsciiCase(rKey) )
+                return pKey->maValue;
+
+            pKey = pKey->mpNext;
+        }
+    }
+
+    return rDefault;
+}
+
+void Config::WriteKey(const OString& rKey, const OString& rStr)
+{
+    SAL_INFO("tools.generic", "Config::WriteKey( " << rKey << ", " << rStr << " ) to "
+                       << GetGroup() << " in " << maFileName);
+
+    // Update config data if necessary
+    if ( !mpData->mbRead )
+    {
+        ImplUpdateConfig();
+        mpData->mbRead = true;
+    }
+
+    // Search key and update value if found
+    ImplGroupData* pGroup = ImplGetGroup();
+    if ( !pGroup )
+        return;
+
+    ImplKeyData* pPrevKey = nullptr;
+    ImplKeyData* pKey = pGroup->mpFirstKey;
+    while ( pKey )
+    {
+        if ( !pKey->mbIsComment && pKey->maKey.equalsIgnoreAsciiCase(rKey) )
+            break;
+
+        pPrevKey = pKey;
+        pKey = pKey->mpNext;
+    }
+
+    bool bNewValue;
+    if ( !pKey )
+    {
+        pKey              = new ImplKeyData;
+        pKey->mpNext      = nullptr;
+        pKey->maKey       = rKey;
+        pKey->mbIsComment = false;
+        if ( pPrevKey )
+            pPrevKey->mpNext = pKey;
+        else
+            pGroup->mpFirstKey = pKey;
+        bNewValue = true;
+    }
+    else
+        bNewValue = pKey->maValue != rStr;
+
+    if ( bNewValue )
+    {
+        pKey->maValue = rStr;
+
+        mpData->mbModified = true;
+    }
+}
+
+void Config::DeleteKey(std::string_view rKey)
+{
+    // Update config data if necessary
+    if ( !mpData->mbRead )
+    {
+        ImplUpdateConfig();
+        mpData->mbRead = true;
+    }
+
+    // Search key and update value
+    ImplGroupData* pGroup = ImplGetGroup();
+    if ( !pGroup )
+        return;
+
+    ImplKeyData* pPrevKey = nullptr;
+    ImplKeyData* pKey = pGroup->mpFirstKey;
+    while ( pKey )
+    {
+        if ( !pKey->mbIsComment && pKey->maKey.equalsIgnoreAsciiCase(rKey) )
+            break;
+
+        pPrevKey = pKey;
+        pKey = pKey->mpNext;
+    }
+
+    if ( pKey )
+    {
+        // Rewire group pointers and delete
+        if ( pPrevKey )
+            pPrevKey->mpNext = pKey->mpNext;
+        else
+            pGroup->mpFirstKey = pKey->mpNext;
+        delete pKey;
+
+        mpData->mbModified = true;
+    }
+}
+
+sal_uInt16 Config::GetKeyCount() const
+{
+    SAL_INFO("tools.generic", "Config::GetKeyCount() from " << GetGroup() << " in " << maFileName);
+
+    // Search key and update value
+    sal_uInt16 nCount = 0;
+    ImplGroupData* pGroup = ImplGetGroup();
+    if ( pGroup )
+    {
+        ImplKeyData* pKey = pGroup->mpFirstKey;
+        while ( pKey )
+        {
+            if ( !pKey->mbIsComment )
+                nCount++;
+
+            pKey = pKey->mpNext;
+        }
+    }
+
+    return nCount;
+}
+
+OString Config::GetKeyName(sal_uInt16 nKey) const
+{
+    SAL_INFO("tools.generic", "Config::GetKeyName( " << OString::number(static_cast<sal_Int32>(nKey))
+                      << " ) from " << GetGroup() << " in " << maFileName);
+
+    // search key and return name if found
+    ImplGroupData* pGroup = ImplGetGroup();
+    if ( pGroup )
+    {
+        ImplKeyData* pKey = pGroup->mpFirstKey;
+        while ( pKey )
+        {
+            if ( !pKey->mbIsComment )
+            {
+                if ( !nKey )
+                    return pKey->maKey;
+                nKey--;
+            }
+
+            pKey = pKey->mpNext;
+        }
+    }
+
+    return OString();
+}
+
+OString Config::ReadKey(sal_uInt16 nKey) const
+{
+    SAL_INFO("tools.generic", "Config::ReadKey( " << OString::number(static_cast<sal_Int32>(nKey))
+                       << " ) from " << GetGroup() << " in " << maFileName);
+
+    // Search key and return value if found
+    ImplGroupData* pGroup = ImplGetGroup();
+    if ( pGroup )
+    {
+        ImplKeyData* pKey = pGroup->mpFirstKey;
+        while ( pKey )
+        {
+            if ( !pKey->mbIsComment )
+            {
+                if ( !nKey )
+                    return pKey->maValue;
+                nKey--;
+            }
+
+            pKey = pKey->mpNext;
+        }
+    }
+
+    return OString();
+}
+
+void Config::Flush()
+{
+    if ( mpData->mbModified )
+        ImplWriteConfig( mpData.get() );
+}
+
+/* vim:set shiftwidth=4 softtabstop=4 expandtab: */
diff --git a/tools/source/generic/fract.cxx b/tools/source/generic/fract.cxx
new file mode 100644
index 0000000000..abe8ee41b2
--- /dev/null
+++ b/tools/source/generic/fract.cxx
@@ -0,0 +1,537 @@
+/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
+/*
+ * This file is part of the LibreOffice project.
+ *
+ * This Source Code Form is subject to the terms of the Mozilla Public
+ * License, v. 2.0. If a copy of the MPL was not distributed with this
+ * file, You can obtain one at http://mozilla.org/MPL/2.0/.
+ *
+ * This file incorporates work covered by the following license notice:
+ *
+ *   Licensed to the Apache Software Foundation (ASF) under one or more
+ *   contributor license agreements. See the NOTICE file distributed
+ *   with this work for additional information regarding copyright
+ *   ownership. The ASF licenses this file to you under the Apache
+ *   License, Version 2.0 (the "License"); you may not use this file
+ *   except in compliance with the License. You may obtain a copy of
+ *   the License at http://www.apache.org/licenses/LICENSE-2.0 .
+ */
+
+#include <tools/fract.hxx>
+#include <tools/debug.hxx>
+#include <o3tl/hash_combine.hxx>
+#include <o3tl/safeint.hxx>
+#include <sal/log.hxx>
+#include <osl/diagnose.h>
+
+#include <algorithm>
+#include <cmath>
+#include <numeric>
+
+#include <boost/rational.hpp>
+
+#ifdef _MSC_VER
+#include <intrin.h>
+#endif
+
+static boost::rational<sal_Int32> rational_FromDouble(double dVal);
+static void rational_ReduceInaccurate(boost::rational<sal_Int32>& rRational, unsigned nSignificantBits);
+static int impl_NumberOfBits( sal_uInt32 nNum );
+
+static boost::rational<sal_Int32> toRational(sal_Int32 n, sal_Int32 d)
+{
+    // https://github.com/boostorg/boost/issues/335 when these are std::numeric_limits<sal_Int32>::min
+    if (n == d)
+        return 1;
+    // tdf#144319 avoid boost::bad_rational e.g. if numerator=-476741369, denominator=-2147483648
+    if (d < -std::numeric_limits<sal_Int32>::max())
+        return 0;
+    return boost::rational<sal_Int32>(n, d);
+}
+
+static constexpr bool isOutOfRange(sal_Int64 nNum)
+{
+    return nNum < std::numeric_limits<sal_Int32>::min()
+            || nNum > std::numeric_limits<sal_Int32>::max();
+}
+
+Fraction::Fraction( sal_Int64 nNum, sal_Int64 nDen ) : mnNumerator(nNum), mnDenominator(nDen)
+{
+    if ( isOutOfRange(nNum) || isOutOfRange(nDen) )
+    {
+        // tdf#143200
+        if (const auto gcd = std::gcd(nNum, nDen); gcd > 1)
+        {
+            nNum /= gcd;
+            nDen /= gcd;
+        }
+        SAL_WARN_IF(isOutOfRange(nNum) || isOutOfRange(nDen),
+                    "tools.fraction", "values outside of range we can represent, doing reduction, which will reduce precision");
+        while (isOutOfRange(nNum) || isOutOfRange(nDen))
+        {
+            nNum /= 2;
+            nDen /= 2;
+        }
+        mnNumerator = nNum;
+        mnDenominator = nDen;
+    }
+    if ( mnDenominator == 0 )
+    {
+        mbValid = false;
+        SAL_WARN( "tools.fraction", "'Fraction(" << nNum << ",0)' invalid fraction created" );
+        return;
+    }
+    else if ((nDen == -1 && nNum == std::numeric_limits<sal_Int32>::min()) ||
+            (nNum == -1 && nDen == std::numeric_limits<sal_Int32>::min()))
+    {
+        mbValid = false;
+        SAL_WARN("tools.fraction", "'Fraction(" << nNum << "," << nDen << ")' invalid fraction created");
+        return;
+    }
+}
+
+/**
+ * only here to prevent passing of NaN
+ */
+Fraction::Fraction( double nNum, double nDen )
+    : Fraction(sal_Int64(nNum), sal_Int64(nDen))
+{}
+
+Fraction::Fraction( double dVal )
+{
+    try
+    {
+        boost::rational<sal_Int32> v = rational_FromDouble( dVal );
+        mnNumerator = v.numerator();
+        mnDenominator = v.denominator();
+    }
+    catch (const boost::bad_rational&)
+    {
+        mbValid = false;
+        SAL_WARN( "tools.fraction", "'Fraction(" << dVal << ")' invalid fraction created" );
+    }
+}
+
+Fraction::operator double() const
+{
+    if (!mbValid)
+    {
+        SAL_WARN( "tools.fraction", "'double()' on invalid fraction" );
+        return 0.0;
+    }
+
+    return boost::rational_cast<double>(toRational(mnNumerator, mnDenominator));
+}
+
+// This methods first validates both values.
+// If one of the arguments is invalid, the whole operation is invalid.
+// After computation detect if result overflows a sal_Int32 value
+// which cause the operation to be marked as invalid
+Fraction& Fraction::operator += ( const Fraction& rVal )
+{
+    if ( !rVal.mbValid )
+        mbValid = false;
+
+    if ( !mbValid )
+    {
+        SAL_WARN( "tools.fraction", "'operator +=' with invalid fraction" );
+        return *this;
+    }
+
+    boost::rational<sal_Int32> a = toRational(mnNumerator, mnDenominator);
+    a += toRational(rVal.mnNumerator, rVal.mnDenominator);
+    mnNumerator = a.numerator();
+    mnDenominator = a.denominator();
+
+    return *this;
+}
+
+Fraction& Fraction::operator -= ( const Fraction& rVal )
+{
+    if ( !rVal.mbValid )
+        mbValid = false;
+
+    if ( !mbValid )
+    {
+        SAL_WARN( "tools.fraction", "'operator -=' with invalid fraction" );
+        return *this;
+    }
+
+    boost::rational<sal_Int32> a = toRational(mnNumerator, mnDenominator);
+    a -= toRational(rVal.mnNumerator, rVal.mnDenominator);
+    mnNumerator = a.numerator();
+    mnDenominator = a.denominator();
+
+    return *this;
+}
+
+namespace
+{
+    bool checked_multiply_by(boost::rational<sal_Int32>& i, const boost::rational<sal_Int32>& r)
+    {
+        // Protect against self-modification
+        sal_Int32 num = r.numerator();
+        sal_Int32 den = r.denominator();
+
+        // Fast-path if the number of bits in input is < the number of bits in the output, overflow cannot happen
+        // This is considerably faster than repeated std::gcd() operations
+        if ((impl_NumberOfBits(std::abs(i.numerator())) + impl_NumberOfBits(std::abs(r.numerator()))) < 32 &&
+            (impl_NumberOfBits(std::abs(i.denominator())) + impl_NumberOfBits(std::abs(r.denominator()))) < 32)
+        {
+            i *= r;
+            return false;
+        }
+
+        // Avoid overflow and preserve normalization
+        sal_Int32 gcd1 = std::gcd(i.numerator(), den);
+        sal_Int32 gcd2 = std::gcd(num, i.denominator());
+
+        if (!gcd1 || !gcd2)
+            return true;
+
+        bool fail = false;
+        fail |= o3tl::checked_multiply(i.numerator() / gcd1, num / gcd2, num);
+        fail |= o3tl::checked_multiply(i.denominator() / gcd2, den / gcd1, den);
+
+        if (!fail)
+            i.assign(num, den);
+
+        return fail;
+    }
+}
+
+Fraction& Fraction::operator *= ( const Fraction& rVal )
+{
+    if ( !rVal.mbValid )
+        mbValid = false;
+
+    if ( !mbValid )
+    {
+        SAL_WARN( "tools.fraction", "'operator *=' with invalid fraction" );
+        return *this;
+    }
+
+    boost::rational<sal_Int32> a = toRational(mnNumerator, mnDenominator);
+    boost::rational<sal_Int32> b = toRational(rVal.mnNumerator, rVal.mnDenominator);
+    bool bFail = checked_multiply_by(a, b);
+    mnNumerator = a.numerator();
+    mnDenominator = a.denominator();
+
+    if (bFail)
+    {
+        mbValid = false;
+    }
+
+    return *this;
+}
+
+Fraction& Fraction::operator /= ( const Fraction& rVal )
+{
+    if ( !rVal.mbValid )
+        mbValid = false;
+
+    if ( !mbValid )
+    {
+        SAL_WARN( "tools.fraction", "'operator /=' with invalid fraction" );
+        return *this;
+    }
+
+    boost::rational<sal_Int32> a = toRational(mnNumerator, mnDenominator);
+    a /= toRational(rVal.mnNumerator, rVal.mnDenominator);
+    mnNumerator = a.numerator();
+    mnDenominator = a.denominator();
+
+    return *this;
+}
+
+/** Inaccurate cancellation for a fraction.
+
+    Clip both nominator and denominator to said number of bits. If
+    either of those already have equal or less number of bits used,
+    this method does nothing.
+
+    @param nSignificantBits denotes, how many significant binary
+    digits to maintain, in both nominator and denominator.
+
+    @example ReduceInaccurate(8) has an error <1% [1/2^(8-1)] - the
+    largest error occurs with the following pair of values:
+
+    binary    1000000011111111111111111111111b/1000000000000000000000000000000b
+    =         1082130431/1073741824
+    = approx. 1.007812499
+
+    A ReduceInaccurate(8) yields 1/1.
+*/
+void Fraction::ReduceInaccurate( unsigned nSignificantBits )
+{
+    if ( !mbValid )
+    {
+        SAL_WARN( "tools.fraction", "'ReduceInaccurate' on invalid fraction" );
+        return;
+    }
+
+    if ( !mnNumerator )
+        return;
+
+    auto a = toRational(mnNumerator, mnDenominator);
+    rational_ReduceInaccurate(a, nSignificantBits);
+    mnNumerator = a.numerator();
+    mnDenominator = a.denominator();
+}
+
+sal_Int32 Fraction::GetNumerator() const
+{
+    if ( !mbValid )
+    {
+        SAL_WARN( "tools.fraction", "'GetNumerator()' on invalid fraction" );
+        return 0;
+    }
+    return mnNumerator;
+}
+
+sal_Int32 Fraction::GetDenominator() const
+{
+    if ( !mbValid )
+    {
+        SAL_WARN( "tools.fraction", "'GetDenominator()' on invalid fraction" );
+        return -1;
+    }
+    return mnDenominator;
+}
+
+Fraction::operator sal_Int32() const
+{
+    if ( !mbValid )
+    {
+        SAL_WARN( "tools.fraction", "'operator sal_Int32()' on invalid fraction" );
+        return 0;
+    }
+    return boost::rational_cast<sal_Int32>(toRational(mnNumerator, mnDenominator));
+}
+
+Fraction operator+( const Fraction& rVal1, const Fraction& rVal2 )
+{
+    Fraction aErg( rVal1 );
+    aErg += rVal2;
+    return aErg;
+}
+
+Fraction operator-( const Fraction& rVal1, const Fraction& rVal2 )
+{
+    Fraction aErg( rVal1 );
+    aErg -= rVal2;
+    return aErg;
+}
+
+Fraction operator*( const Fraction& rVal1, const Fraction& rVal2 )
+{
+    Fraction aErg( rVal1 );
+    aErg *= rVal2;
+    return aErg;
+}
+
+Fraction operator/( const Fraction& rVal1, const Fraction& rVal2 )
+{
+    Fraction aErg( rVal1 );
+    aErg /= rVal2;
+    return aErg;
+}
+
+bool operator !=( const Fraction& rVal1, const Fraction& rVal2 )
+{
+    return !(rVal1 == rVal2);
+}
+
+bool operator <=( const Fraction& rVal1, const Fraction& rVal2 )
+{
+    return !(rVal1 > rVal2);
+}
+
+bool operator >=( const Fraction& rVal1, const Fraction& rVal2 )
+{
+    return !(rVal1 < rVal2);
+}
+
+bool operator == ( const Fraction& rVal1, const Fraction& rVal2 )
+{
+    if ( !rVal1.mbValid || !rVal2.mbValid )
+    {
+        SAL_WARN( "tools.fraction", "'operator ==' with an invalid fraction" );
+        return false;
+    }
+
+    return toRational(rVal1.mnNumerator, rVal1.mnDenominator) == toRational(rVal2.mnNumerator, rVal2.mnDenominator);
+}
+
+bool operator < ( const Fraction& rVal1, const Fraction& rVal2 )
+{
+    if ( !rVal1.mbValid || !rVal2.mbValid )
+    {
+        SAL_WARN( "tools.fraction", "'operator <' with an invalid fraction" );
+        return false;
+    }
+
+    return toRational(rVal1.mnNumerator, rVal1.mnDenominator) < toRational(rVal2.mnNumerator, rVal2.mnDenominator);
+}
+
+bool operator > ( const Fraction& rVal1, const Fraction& rVal2 )
+{
+    if ( !rVal1.mbValid || !rVal2.mbValid )
+    {
+        SAL_WARN( "tools.fraction", "'operator >' with an invalid fraction" );
+        return false;
+    }
+
+    return toRational(rVal1.mnNumerator, rVal1.mnDenominator) > toRational(rVal2.mnNumerator, rVal2.mnDenominator);
+}
+
+// If dVal > LONG_MAX or dVal < LONG_MIN, the rational throws a boost::bad_rational.
+// Otherwise, dVal and denominator are multiplied by 10, until one of them
+// is larger than (LONG_MAX / 10).
+//
+// NOTE: here we use 'sal_Int32' due that only values in sal_Int32 range are valid.
+static boost::rational<sal_Int32> rational_FromDouble(double dVal)
+{
+    if ( dVal > std::numeric_limits<sal_Int32>::max() ||
+         dVal < std::numeric_limits<sal_Int32>::min() ||
+         std::isnan(dVal) )
+        throw boost::bad_rational();
+
+    const sal_Int32 nMAX = std::numeric_limits<sal_Int32>::max() / 10;
+    sal_Int32 nDen = 1;
+    while ( std::abs( dVal ) < nMAX && nDen < nMAX ) {
+        dVal *= 10;
+        nDen *= 10;
+    }
+    return boost::rational<sal_Int32>( sal_Int32(dVal), nDen );
+}
+
+/**
+ * Find the number of bits required to represent this number, using the CLZ intrinsic
+ */
+static int impl_NumberOfBits( sal_uInt32 nNum )
+{
+    if (nNum == 0)
+        return 0;
+#ifdef _MSC_VER
+    unsigned long r = 0;
+    _BitScanReverse(&r, nNum);
+    return r + 1;
+#else
+    return 32 - __builtin_clz(nNum);
+#endif
+}
+
+/** Inaccurate cancellation for a fraction.
+
+    Clip both nominator and denominator to said number of bits. If
+    either of those already have equal or less number of bits used,
+    this method does nothing.
+
+    @param nSignificantBits denotes, how many significant binary
+    digits to maintain, in both nominator and denominator.
+
+    @example ReduceInaccurate(8) has an error <1% [1/2^(8-1)] - the
+    largest error occurs with the following pair of values:
+
+    binary    1000000011111111111111111111111b/1000000000000000000000000000000b
+    =         1082130431/1073741824
+    = approx. 1.007812499
+
+    A ReduceInaccurate(8) yields 1/1.
+*/
+static void rational_ReduceInaccurate(boost::rational<sal_Int32>& rRational, unsigned nSignificantBits)
+{
+    if ( !rRational )
+        return;
+
+    // http://www.boost.org/doc/libs/release/libs/rational/rational.html#Internal%20representation
+    sal_Int32 nMul = rRational.numerator();
+    if (nMul == std::numeric_limits<sal_Int32>::min())
+    {
+        // ofz#32973 Integer-overflow
+        return;
+    }
+    const bool bNeg = nMul < 0;
+    if (bNeg)
+        nMul = -nMul;
+    sal_Int32 nDiv = rRational.denominator();
+
+    DBG_ASSERT(nSignificantBits<65, "More than 64 bit of significance is overkill!");
+
+    // How much bits can we lose?
+    const int nMulBitsToLose = std::max( ( impl_NumberOfBits( nMul ) - int( nSignificantBits ) ), 0 );
+    const int nDivBitsToLose = std::max( ( impl_NumberOfBits( nDiv ) - int( nSignificantBits ) ), 0 );
+
+    const int nToLose = std::min( nMulBitsToLose, nDivBitsToLose );
+
+    // Remove the bits
+    nMul >>= nToLose;
+    nDiv >>= nToLose;
+
+    if ( !nMul || !nDiv ) {
+        // Return without reduction
+        OSL_FAIL( "Oops, we reduced too much..." );
+        return;
+    }
+
+    rRational.assign( bNeg ? -nMul : nMul, nDiv );
+}
+
+size_t Fraction::GetHashValue() const
+{
+    size_t hash = 0;
+    o3tl::hash_combine( hash, mnNumerator );
+    o3tl::hash_combine( hash, mnDenominator );
+    o3tl::hash_combine( hash, mbValid );
+    return hash;
+}
+
+Fraction Fraction::MakeFraction( tools::Long nN1, tools::Long nN2, tools::Long nD1, tools::Long nD2 )
+{
+    if( nD1 == 0 || nD2 == 0 ) //under these bad circumstances the following while loop will be endless
+    {
+        SAL_WARN("tools.fraction", "Invalid parameter for ImplMakeFraction");
+        return Fraction( 1, 1 );
+    }
+
+    tools::Long i = 1;
+
+    if ( nN1 < 0 ) { i = -i; nN1 = -nN1; }
+    if ( nN2 < 0 ) { i = -i; nN2 = -nN2; }
+    if ( nD1 < 0 ) { i = -i; nD1 = -nD1; }
+    if ( nD2 < 0 ) { i = -i; nD2 = -nD2; }
+    // all positive; i sign
+
+    assert( nN1 >= std::numeric_limits<sal_Int32>::min() );
+    assert( nN1 <= std::numeric_limits<sal_Int32>::max( ));
+    assert( nD1 >= std::numeric_limits<sal_Int32>::min() );
+    assert( nD1 <= std::numeric_limits<sal_Int32>::max( ));
+    assert( nN2 >= std::numeric_limits<sal_Int32>::min() );
+    assert( nN2 <= std::numeric_limits<sal_Int32>::max( ));
+    assert( nD2 >= std::numeric_limits<sal_Int32>::min() );
+    assert( nD2 <= std::numeric_limits<sal_Int32>::max( ));
+
+    boost::rational<sal_Int32> a = toRational(i*nN1, nD1);
+    boost::rational<sal_Int32> b = toRational(nN2, nD2);
+    bool bFail = checked_multiply_by(a, b);
+
+    while ( bFail ) {
+        if ( nN1 > nN2 )
+            nN1 = (nN1 + 1) / 2;
+        else
+            nN2 = (nN2 + 1) / 2;
+        if ( nD1 > nD2 )
+            nD1 = (nD1 + 1) / 2;
+        else
+            nD2 = (nD2 + 1) / 2;
+
+        a = toRational(i*nN1, nD1);
+        b = toRational(nN2, nD2);
+        bFail = checked_multiply_by(a, b);
+    }
+
+    return Fraction(a.numerator(), a.denominator());
+}
+
+/* vim:set shiftwidth=4 softtabstop=4 expandtab: */
diff --git a/tools/source/generic/gen.cxx b/tools/source/generic/gen.cxx
new file mode 100644
index 0000000000..012318c021
--- /dev/null
+++ b/tools/source/generic/gen.cxx
@@ -0,0 +1,225 @@
+/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
+/*
+ * This file is part of the LibreOffice project.
+ *
+ * This Source Code Form is subject to the terms of the Mozilla Public
+ * License, v. 2.0. If a copy of the MPL was not distributed with this
+ * file, You can obtain one at http://mozilla.org/MPL/2.0/.
+ *
+ * This file incorporates work covered by the following license notice:
+ *
+ *   Licensed to the Apache Software Foundation (ASF) under one or more
+ *   contributor license agreements. See the NOTICE file distributed
+ *   with this work for additional information regarding copyright
+ *   ownership. The ASF licenses this file to you under the Apache
+ *   License, Version 2.0 (the "License"); you may not use this file
+ *   except in compliance with the License. You may obtain a copy of
+ *   the License at http://www.apache.org/licenses/LICENSE-2.0 .
+ */
+
+#include <sal/config.h>
+#include <rtl/string.hxx>
+
+#include <algorithm>
+#include <tuple>
+#include <o3tl/hash_combine.hxx>
+#include <o3tl/safeint.hxx>
+#include <tools/gen.hxx>
+
+OString Pair::toString() const
+{
+    // Note that this is not just used for debugging output but the
+    // format is parsed by external code (passed in callbacks to
+    // LibreOfficeKit clients). So don't change.
+    return OString::number(A()) + ", " + OString::number(B());
+}
+
+size_t Pair::GetHashValue() const
+{
+    size_t hash = 0;
+    o3tl::hash_combine( hash, mnA );
+    o3tl::hash_combine( hash, mnB );
+    return hash;
+}
+
+void RectangleTemplateBase::SaturatingSetSize(const SizeTemplateBase& rSize)
+{
+    if (rSize.Width() < 0)
+        mnRight = o3tl::saturating_add(mnLeft, (rSize.Width() + 1));
+    else if ( rSize.Width() > 0 )
+        mnRight = o3tl::saturating_add(mnLeft, (rSize.Width() - 1));
+    else
+        SetWidthEmpty();
+
+    if ( rSize.Height() < 0 )
+        mnBottom = o3tl::saturating_add(mnTop, (rSize.Height() + 1));
+    else if ( rSize.Height() > 0 )
+        mnBottom = o3tl::saturating_add(mnTop, (rSize.Height() - 1));
+    else
+        SetHeightEmpty();
+}
+
+void RectangleTemplateBase::SaturatingSetPosX(tools::Long x)
+{
+    if (!IsWidthEmpty())
+        mnRight = o3tl::saturating_add(mnRight, x - mnLeft);
+    mnLeft = x;
+}
+
+void RectangleTemplateBase::SaturatingSetPosY(tools::Long y)
+{
+    if (!IsHeightEmpty())
+        mnBottom = o3tl::saturating_add(mnBottom, y - mnTop);
+    mnTop = y;
+}
+
+void RectangleTemplateBase::Union( const RectangleTemplateBase& rRect )
+{
+    if ( rRect.IsEmpty() )
+        return;
+
+    if ( IsEmpty() )
+        *this = rRect;
+    else
+    {
+        std::tie(mnLeft, mnRight) = std::minmax({ mnLeft, rRect.mnLeft, mnRight, rRect.mnRight });
+        std::tie(mnTop, mnBottom) = std::minmax({ mnTop, rRect.mnTop, mnBottom, rRect.mnBottom });
+    }
+}
+
+void RectangleTemplateBase::Intersection( const RectangleTemplateBase& rRect )
+{
+    if ( IsEmpty() )
+        return;
+    if ( rRect.IsEmpty() )
+    {
+        *this = tools::Rectangle();
+        return;
+    }
+
+    // Normalize rectangle
+    RectangleTemplateBase aTmpRect( rRect );
+    Normalize();
+    aTmpRect.Normalize();
+
+    // Perform intersection
+    mnLeft  = std::max( mnLeft, aTmpRect.mnLeft );
+    mnRight = std::min( mnRight, aTmpRect.mnRight );
+    mnTop   = std::max( mnTop, aTmpRect.mnTop );
+    mnBottom= std::min( mnBottom, aTmpRect.mnBottom );
+
+    // Determine if intersection is empty
+    if ( mnRight < mnLeft || mnBottom < mnTop )
+        *this = tools::Rectangle();
+}
+
+void RectangleTemplateBase::Normalize()
+{
+    if ((mnRight < mnLeft) && (!IsWidthEmpty()))
+    {
+        std::swap(mnLeft, mnRight);
+    }
+
+    if ((mnBottom < mnTop) && (!IsHeightEmpty()))
+    {
+        std::swap(mnBottom, mnTop);
+    }
+}
+
+bool RectangleTemplateBase::Contains( const PointTemplateBase& rPoint ) const
+{
+    if ( IsEmpty() )
+        return false;
+
+    if ( mnLeft <= mnRight )
+    {
+        if ( (rPoint.X() < mnLeft) || (rPoint.X() > mnRight) )
+            return false;
+    }
+    else
+    {
+        if ( (rPoint.X() > mnLeft) || (rPoint.X() < mnRight) )
+            return false;
+    }
+    if ( mnTop <= mnBottom )
+    {
+        if ( (rPoint.Y() < mnTop) || (rPoint.Y() > mnBottom) )
+            return false;
+    }
+    else
+    {
+        if ( (rPoint.Y() > mnTop) || (rPoint.Y() < mnBottom) )
+            return false;
+    }
+    return true;
+}
+
+bool RectangleTemplateBase::Contains( const RectangleTemplateBase& rRect ) const
+{
+    return Contains( PointTemplateBase{ rRect.Left(), rRect.Top() } )
+        && Contains( PointTemplateBase{ rRect.Right(), rRect.Bottom() } );
+}
+
+bool RectangleTemplateBase::Overlaps( const RectangleTemplateBase& rRect ) const
+{
+    // If there's no intersection, they don't overlap
+    RectangleTemplateBase aTmp(*this);
+    aTmp.Intersection(rRect);
+    return !aTmp.IsEmpty();
+}
+
+OString RectangleTemplateBase::toString() const
+{
+    // Note that this is not just used for debugging output but the
+    // format is parsed by external code (passed in callbacks to
+    // LibreOfficeKit clients). So don't change.
+    return OString::number(Left()) + ", "
+            + OString::number(Top()) + ", "
+            + OString::number(getOpenWidth()) + ", "
+            + OString::number(getOpenHeight());
+}
+
+void RectangleTemplateBase::expand(tools::Long nExpandBy)
+{
+    AdjustLeft(-nExpandBy);
+    AdjustTop(-nExpandBy);
+    AdjustRight(nExpandBy);
+    AdjustBottom(nExpandBy);
+}
+
+void RectangleTemplateBase::shrink(tools::Long nShrinkBy)
+{
+    mnLeft   += nShrinkBy;
+    mnTop    += nShrinkBy;
+    if (!IsWidthEmpty())
+        mnRight -= nShrinkBy;
+    if (!IsHeightEmpty())
+        mnBottom -= nShrinkBy;
+}
+
+tools::Long RectangleTemplateBase::AdjustRight(tools::Long nHorzMoveDelta)
+{
+    if (IsWidthEmpty())
+        mnRight = mnLeft + nHorzMoveDelta - 1;
+    else
+        mnRight += nHorzMoveDelta;
+    return mnRight;
+}
+
+tools::Long RectangleTemplateBase::AdjustBottom( tools::Long nVertMoveDelta )
+{
+    if (IsHeightEmpty())
+        mnBottom = mnTop + nVertMoveDelta - 1;
+    else
+        mnBottom += nVertMoveDelta;
+    return mnBottom;
+}
+
+static_assert( std::is_trivially_copyable< Pair >::value );
+static_assert( std::is_trivially_copyable< Point >::value );
+static_assert( std::is_trivially_copyable< Size >::value );
+static_assert( std::is_trivially_copyable< Range >::value );
+static_assert( std::is_trivially_copyable< Selection >::value );
+static_assert( std::is_trivially_copyable< tools::Rectangle >::value );
+
+/* vim:set shiftwidth=4 softtabstop=4 expandtab: */
diff --git a/tools/source/generic/line.cxx b/tools/source/generic/line.cxx
new file mode 100644
index 0000000000..ee9ad97979
--- /dev/null
+++ b/tools/source/generic/line.cxx
@@ -0,0 +1,140 @@
+/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
+/*
+ * This file is part of the LibreOffice project.
+ *
+ * This Source Code Form is subject to the terms of the Mozilla Public
+ * License, v. 2.0. If a copy of the MPL was not distributed with this
+ * file, You can obtain one at http://mozilla.org/MPL/2.0/.
+ *
+ * This file incorporates work covered by the following license notice:
+ *
+ *   Licensed to the Apache Software Foundation (ASF) under one or more
+ *   contributor license agreements. See the NOTICE file distributed
+ *   with this work for additional information regarding copyright
+ *   ownership. The ASF licenses this file to you under the Apache
+ *   License, Version 2.0 (the "License"); you may not use this file
+ *   except in compliance with the License. You may obtain a copy of
+ *   the License at http://www.apache.org/licenses/LICENSE-2.0 .
+ */
+
+#include <tools/line.hxx>
+#include <tools/helpers.hxx>
+
+#include <cmath>
+
+namespace tools
+{
+
+double Line::GetLength() const
+{
+    return hypot( maStart.X() - maEnd.X(), maStart.Y() - maEnd.Y() );
+}
+
+bool Line::Intersection( const Line& rLine, Point& rIntersection ) const
+{
+    double  fX, fY;
+    bool    bRet;
+
+    if( Intersection( rLine, fX, fY ) )
+    {
+        rIntersection.setX( FRound( fX ) );
+        rIntersection.setY( FRound( fY ) );
+        bRet = true;
+    }
+    else
+        bRet = false;
+
+    return bRet;
+}
+
+bool Line::Intersection( const tools::Line& rLine, double& rIntersectionX, double& rIntersectionY ) const
+{
+    const double    fAx = static_cast<double>(maEnd.X()) - maStart.X();
+    const double    fAy = static_cast<double>(maEnd.Y()) - maStart.Y();
+    const double    fBx = static_cast<double>(rLine.maStart.X()) - rLine.maEnd.X();
+    const double    fBy = static_cast<double>(rLine.maStart.Y()) - rLine.maEnd.Y();
+    const double    fDen = fAy * fBx - fAx * fBy;
+    bool            bOk = false;
+
+    if( fDen != 0. )
+    {
+        const double    fCx = static_cast<double>(maStart.X()) - rLine.maStart.X();
+        const double    fCy = static_cast<double>(maStart.Y()) - rLine.maStart.Y();
+        const double    fA = fBy * fCx - fBx * fCy;
+        const bool      bGreater = ( fDen > 0. );
+
+        bOk = true;
+
+        if ( bGreater )
+        {
+            if ( ( fA < 0. ) || ( fA > fDen ) )
+                bOk = false;
+        }
+        else if ( ( fA > 0. ) || ( fA < fDen ) )
+            bOk = false;
+
+        if ( bOk )
+        {
+            const double fB = fAx * fCy - fAy * fCx;
+
+            if ( bGreater )
+            {
+                if ( ( fB < 0. ) || ( fB > fDen ) )
+                    bOk = false;
+            }
+            else if ( ( fB > 0. ) || ( fB < fDen ) )
+                bOk = false;
+
+            if( bOk )
+            {
+                const double fAlpha = fA / fDen;
+
+                rIntersectionX = ( maStart.X() + fAlpha * fAx );
+                rIntersectionY = ( maStart.Y() + fAlpha * fAy );
+            }
+        }
+    }
+
+    return bOk;
+}
+
+double Line::GetDistance( const double& rPtX, const double& rPtY ) const
+{
+    double fDist;
+
+    if( maStart != maEnd )
+    {
+        const double fDistX = static_cast<double>(maEnd.X()) - maStart.X();
+        const double fDistY = static_cast<double>(maEnd.Y()) - maStart.Y();
+        const double fACX = static_cast<double>(maStart.X()) - rPtX;
+        const double fACY = static_cast<double>(maStart.Y()) - rPtY;
+        const double fL2 = fDistX * fDistX + fDistY * fDistY;
+        const double fR = ( fACY * -fDistY - fACX * fDistX ) / fL2;
+        const double fS = ( fACY * fDistX - fACX * fDistY ) / fL2;
+
+        if( fR < 0.0 )
+        {
+            fDist = hypot( maStart.X() - rPtX, maStart.Y() - rPtY );
+
+            if( fS < 0.0 )
+                fDist *= -1.0;
+        }
+        else if( fR <= 1.0 )
+            fDist = fS * sqrt( fL2 );
+        else
+        {
+            fDist = hypot( maEnd.X() - rPtX, maEnd.Y() - rPtY );
+
+            if( fS < 0.0 )
+                fDist *= -1.0;
+        }
+    }
+    else
+        fDist = hypot( maStart.X() - rPtX, maStart.Y() - rPtY );
+
+    return fDist;
+}
+
+} //namespace tools
+
+/* vim:set shiftwidth=4 softtabstop=4 expandtab: */
diff --git a/tools/source/generic/point.cxx b/tools/source/generic/point.cxx
new file mode 100644
index 0000000000..e71f60411a
--- /dev/null
+++ b/tools/source/generic/point.cxx
@@ -0,0 +1,86 @@
+/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
+/*
+ * This file is part of the LibreOffice project.
+ *
+ * This Source Code Form is subject to the terms of the Mozilla Public
+ * License, v. 2.0. If a copy of the MPL was not distributed with this
+ * file, You can obtain one at http://mozilla.org/MPL/2.0/.
+ *
+ * This file incorporates work covered by the following license notice:
+ *
+ *   Licensed to the Apache Software Foundation (ASF) under one or more
+ *   contributor license agreements. See the NOTICE file distributed
+ *   with this work for additional information regarding copyright
+ *   ownership. The ASF licenses this file to you under the Apache
+ *   License, Version 2.0 (the "License"); you may not use this file
+ *   except in compliance with the License. You may obtain a copy of
+ *   the License at http://www.apache.org/licenses/LICENSE-2.0 .
+ */
+
+#include <tools/gen.hxx>
+
+void PointTemplateBase::RotateAround( PointTemplateBase& rPoint,
+                          Degree10 nOrientation ) const
+{
+    tools::Long nX = rPoint.mnA;
+    tools::Long nY = rPoint.mnB;
+    RotateAround(nX, nY, nOrientation);
+    rPoint.mnA = nX;
+    rPoint.mnB = nY;
+}
+
+void PointTemplateBase::RotateAround( tools::Long& rX, tools::Long& rY,
+                          Degree10 nOrientation ) const
+{
+    const tools::Long nOriginX = mnA;
+    const tools::Long nOriginY = mnB;
+
+    if ( (nOrientation >= 0_deg10) && !(nOrientation % 900_deg10) )
+    {
+        if ( nOrientation >= 3600_deg10 )
+            nOrientation %= 3600_deg10;
+
+        if ( nOrientation )
+        {
+            rX -= nOriginX;
+            rY -= nOriginY;
+
+            if ( nOrientation == 900_deg10 )
+            {
+                tools::Long nTemp = rX;
+                rX = rY;
+                rY = -nTemp;
+            }
+            else if ( nOrientation == 1800_deg10 )
+            {
+                rX = -rX;
+                rY = -rY;
+            }
+            else /* ( nOrientation == 2700 ) */
+            {
+                tools::Long nTemp = rX;
+                rX = -rY;
+                rY = nTemp;
+            }
+
+            rX += nOriginX;
+            rY += nOriginY;
+        }
+    }
+    else
+    {
+        double nRealOrientation = toRadians(nOrientation);
+        double nCos = cos( nRealOrientation );
+        double nSin = sin( nRealOrientation );
+
+        // Translation...
+        tools::Long nX = rX-nOriginX;
+        tools::Long nY = rY-nOriginY;
+
+        // Rotation...
+        rX = + static_cast<tools::Long>(nCos*nX + nSin*nY) + nOriginX;
+        rY = - static_cast<tools::Long>(nSin*nX - nCos*nY) + nOriginY;
+    }
+}
+
+/* vim:set shiftwidth=4 softtabstop=4 expandtab: */
diff --git a/tools/source/generic/poly.cxx b/tools/source/generic/poly.cxx
new file mode 100644
index 0000000000..89405eea41
--- /dev/null
+++ b/tools/source/generic/poly.cxx
@@ -0,0 +1,1866 @@
+/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
+/*
+ * This file is part of the LibreOffice project.
+ *
+ * This Source Code Form is subject to the terms of the Mozilla Public
+ * License, v. 2.0. If a copy of the MPL was not distributed with this
+ * file, You can obtain one at http://mozilla.org/MPL/2.0/.
+ *
+ * This file incorporates work covered by the following license notice:
+ *
+ *   Licensed to the Apache Software Foundation (ASF) under one or more
+ *   contributor license agreements. See the NOTICE file distributed
+ *   with this work for additional information regarding copyright
+ *   ownership. The ASF licenses this file to you under the Apache
+ *   License, Version 2.0 (the "License"); you may not use this file
+ *   except in compliance with the License. You may obtain a copy of
+ *   the License at http://www.apache.org/licenses/LICENSE-2.0 .
+ */
+
+#include <osl/endian.h>
+#include <osl/diagnose.h>
+#include <sal/log.hxx>
+#include <tools/bigint.hxx>
+#include <tools/debug.hxx>
+#include <tools/helpers.hxx>
+#include <tools/stream.hxx>
+#include <tools/vcompat.hxx>
+#include <tools/gen.hxx>
+#include <poly.h>
+#include <o3tl/safeint.hxx>
+#include <tools/line.hxx>
+#include <tools/poly.hxx>
+#include <basegfx/polygon/b2dpolygon.hxx>
+#include <basegfx/point/b2dpoint.hxx>
+#include <basegfx/vector/b2dvector.hxx>
+#include <basegfx/polygon/b2dpolygontools.hxx>
+#include <basegfx/curve/b2dcubicbezier.hxx>
+
+#include <memory>
+#include <vector>
+#include <algorithm>
+#include <cassert>
+#include <cstring>
+#include <limits.h>
+#include <cmath>
+
+constexpr int EDGE_LEFT   = 1;
+constexpr int EDGE_TOP    = 2;
+constexpr int EDGE_RIGHT  = 4;
+constexpr int EDGE_BOTTOM = 8;
+constexpr int EDGE_HORZ   = EDGE_RIGHT | EDGE_LEFT;
+constexpr int EDGE_VERT   = EDGE_TOP | EDGE_BOTTOM;
+constexpr double SMALL_DVALUE = 0.0000001;
+#define FSQRT2          1.4142135623730950488016887242097
+
+static double ImplGetParameter( const Point& rCenter, const Point& rPt, double fWR, double fHR )
+{
+    const double nDX = static_cast<double>(rPt.X()) - rCenter.X();
+    const double nDY = static_cast<double>(rCenter.Y()) - rPt.Y();
+    double fAngle = atan2(nDY, nDX);
+
+    return atan2(fWR*sin(fAngle), fHR*cos(fAngle));
+}
+
+ImplPolygon::ImplPolygon( sal_uInt16 nInitSize  )
+{
+    ImplInitSize(nInitSize, false);
+}
+
+ImplPolygon::ImplPolygon( const ImplPolygon& rImpPoly )
+{
+    if ( rImpPoly.mnPoints )
+    {
+        mxPointAry.reset(new Point[rImpPoly.mnPoints]);
+        memcpy(mxPointAry.get(), rImpPoly.mxPointAry.get(), rImpPoly.mnPoints * sizeof(Point));
+
+        if( rImpPoly.mxFlagAry )
+        {
+            mxFlagAry.reset(new PolyFlags[rImpPoly.mnPoints]);
+            memcpy(mxFlagAry.get(), rImpPoly.mxFlagAry.get(), rImpPoly.mnPoints);
+        }
+    }
+
+    mnPoints   = rImpPoly.mnPoints;
+}
+
+ImplPolygon::ImplPolygon( sal_uInt16 nInitSize, const Point* pInitAry, const PolyFlags* pInitFlags )
+{
+    if ( nInitSize )
+    {
+        mxPointAry.reset(new Point[nInitSize]);
+        memcpy(mxPointAry.get(), pInitAry, nInitSize * sizeof(Point));
+
+        if( pInitFlags )
+        {
+            mxFlagAry.reset(new PolyFlags[nInitSize]);
+            memcpy(mxFlagAry.get(), pInitFlags, nInitSize);
+        }
+    }
+
+    mnPoints   = nInitSize;
+}
+
+ImplPolygon::ImplPolygon( const tools::Rectangle& rRect )
+{
+    if ( !rRect.IsEmpty() )
+    {
+         ImplInitSize(5);
+         mxPointAry[0] = rRect.TopLeft();
+         mxPointAry[1] = rRect.TopRight();
+         mxPointAry[2] = rRect.BottomRight();
+         mxPointAry[3] = rRect.BottomLeft();
+         mxPointAry[4] = rRect.TopLeft();
+    }
+    else
+        mnPoints = 0;
+}
+
+ImplPolygon::ImplPolygon( const tools::Rectangle& rRect, sal_uInt32 nHorzRound, sal_uInt32 nVertRound )
+{
+    if ( !rRect.IsEmpty() )
+    {
+        tools::Rectangle aRect( rRect );
+        aRect.Normalize();            // SJ: i9140
+
+        nHorzRound = std::min( nHorzRound, static_cast<sal_uInt32>(std::abs( aRect.GetWidth() >> 1 )) );
+        nVertRound = std::min( nVertRound, static_cast<sal_uInt32>(std::abs( aRect.GetHeight() >> 1 )) );
+
+        if( !nHorzRound && !nVertRound )
+        {
+            ImplInitSize(5);
+            mxPointAry[0] = aRect.TopLeft();
+            mxPointAry[1] = aRect.TopRight();
+            mxPointAry[2] = aRect.BottomRight();
+            mxPointAry[3] = aRect.BottomLeft();
+            mxPointAry[4] = aRect.TopLeft();
+        }
+        else
+        {
+            const Point     aTL( aRect.Left() + nHorzRound, aRect.Top() + nVertRound );
+            const Point     aTR( aRect.Right() - nHorzRound, aRect.Top() + nVertRound );
+            const Point     aBR( aRect.Right() - nHorzRound, aRect.Bottom() - nVertRound );
+            const Point     aBL( aRect.Left() + nHorzRound, aRect.Bottom() - nVertRound );
+            tools::Polygon aEllipsePoly( Point(), nHorzRound, nVertRound );
+            sal_uInt16 i, nEnd, nSize4 = aEllipsePoly.GetSize() >> 2;
+
+            ImplInitSize(aEllipsePoly.GetSize() + 1);
+
+            const Point* pSrcAry = aEllipsePoly.GetConstPointAry();
+            Point* pDstAry = mxPointAry.get();
+
+            for( i = 0, nEnd = nSize4; i < nEnd; i++ )
+                pDstAry[ i ] = pSrcAry[ i ] + aTR;
+
+            for( nEnd = nEnd + nSize4; i < nEnd; i++ )
+                pDstAry[ i ] = pSrcAry[ i ] + aTL;
+
+            for( nEnd = nEnd + nSize4; i < nEnd; i++ )
+                pDstAry[ i ] = pSrcAry[ i ] + aBL;
+
+            for( nEnd = nEnd + nSize4; i < nEnd; i++ )
+                pDstAry[ i ] = pSrcAry[ i ] + aBR;
+
+            pDstAry[ nEnd ] = pDstAry[ 0 ];
+        }
+    }
+    else
+        mnPoints = 0;
+}
+
+ImplPolygon::ImplPolygon( const Point& rCenter, tools::Long nRadX, tools::Long nRadY )
+{
+    if( nRadX && nRadY )
+    {
+        sal_uInt16 nPoints;
+        // Compute default (depends on size)
+        tools::Long nRadXY;
+        const bool bOverflow = o3tl::checked_multiply(nRadX, nRadY, nRadXY);
+        if (!bOverflow)
+        {
+            nPoints = std::clamp(
+                ( M_PI * ( 1.5 * ( nRadX + nRadY ) -
+                           sqrt( static_cast<double>(std::abs(nRadXY)) ) ) ),
+                32.0, 256.0 );
+        }
+        else
+        {
+           nPoints = 256;
+        }
+
+        if( ( nRadX > 32 ) && ( nRadY > 32 ) && ( nRadX + nRadY ) < 8192 )
+            nPoints >>= 1;
+
+        // Ceil number of points until divisible by four
+        nPoints = (nPoints + 3) & ~3;
+        ImplInitSize(nPoints);
+
+        sal_uInt16 i;
+        sal_uInt16 nPoints2 = nPoints >> 1;
+        sal_uInt16 nPoints4 = nPoints >> 2;
+        double nAngle;
+        double nAngleStep = M_PI_2 / ( nPoints4 - 1 );
+
+        for( i=0, nAngle = 0.0; i < nPoints4; i++, nAngle += nAngleStep )
+        {
+            tools::Long nX = FRound( nRadX * cos( nAngle ) );
+            tools::Long nY = FRound( -nRadY * sin( nAngle ) );
+
+            Point* pPt = &(mxPointAry[i]);
+            pPt->setX(  nX + rCenter.X() );
+            pPt->setY(  nY + rCenter.Y() );
+            pPt = &(mxPointAry[nPoints2-i-1]);
+            pPt->setX( -nX + rCenter.X() );
+            pPt->setY(  nY + rCenter.Y() );
+            pPt = &(mxPointAry[i+nPoints2]);
+            pPt->setX( -nX + rCenter.X() );
+            pPt->setY( -nY + rCenter.Y() );
+            pPt = &(mxPointAry[nPoints-i-1]);
+            pPt->setX(  nX + rCenter.X() );
+            pPt->setY( -nY + rCenter.Y() );
+        }
+    }
+    else
+        mnPoints = 0;
+}
+
+ImplPolygon::ImplPolygon(const tools::Rectangle& rBound, const Point& rStart, const Point& rEnd,
+                         PolyStyle eStyle, const bool bClockWiseArcDirection)
+{
+    const auto nWidth = rBound.GetWidth();
+    const auto nHeight = rBound.GetHeight();
+
+    if ((nWidth != 0) && (nHeight != 0))
+    {
+        const Point aCenter(rBound.Center());
+        // tdf#142268 Get Top Left corner of rectangle (the rectangle is not always correctly created)
+        const auto aBoundLeft = rBound.Left() < aCenter.X() ? rBound.Left() : rBound.Right();
+        const auto aBoundTop = rBound.Top() < aCenter.Y() ? rBound.Top() : rBound.Bottom();
+        const auto nRadX = o3tl::saturating_sub(aCenter.X(), aBoundLeft);
+        const auto nRadY = o3tl::saturating_sub(aCenter.Y(), aBoundTop);
+        sal_uInt16 nPoints;
+
+        tools::Long nRadXY;
+        const bool bOverflow = o3tl::checked_multiply(nRadX, nRadY, nRadXY);
+        if (!bOverflow)
+        {
+            nPoints = std::clamp(
+                ( M_PI * ( 1.5 * ( nRadX + nRadY ) -
+                           sqrt( static_cast<double>(std::abs(nRadXY)) ) ) ),
+                32.0, 256.0 );
+        }
+        else
+        {
+            nPoints = 256;
+        }
+
+
+        if (nRadX > 32 && nRadY > 32 && o3tl::saturating_add(nRadX, nRadY) < 8192)
+            nPoints >>= 1;
+
+        // compute threshold
+        const double    fRadX = nRadX;
+        const double    fRadY = nRadY;
+        const double    fCenterX = aCenter.X();
+        const double    fCenterY = aCenter.Y();
+        double          fStart = ImplGetParameter( aCenter, rStart, fRadX, fRadY );
+        double          fEnd = ImplGetParameter( aCenter, rEnd, fRadX, fRadY );
+        double          fDiff = fEnd - fStart;
+        double          fStep;
+        sal_uInt16      nStart;
+        sal_uInt16      nEnd;
+
+        if (bClockWiseArcDirection == false)
+        {
+            // #i73608# If startPoint is equal to endPoint, then draw full circle instead of nothing (as Metafiles spec)
+            if (fDiff <= 0.)
+                fDiff += 2. * M_PI;
+        }
+        else
+        {
+            fDiff = (2. * M_PI) - fDiff;
+            if (fDiff > 2. * M_PI)
+                fDiff -= 2. * M_PI;
+        }
+
+        // Proportionally shrink number of points( fDiff / (2PI) );
+        nPoints = std::max(static_cast<sal_uInt16>((fDiff / (2. * M_PI)) * nPoints), sal_uInt16(16));
+        fStep = fDiff / (nPoints - 1);
+        if (bClockWiseArcDirection == true)
+            fStep = -fStep;
+
+        if (PolyStyle::Pie == eStyle)
+        {
+            const Point aCenter2(FRound(fCenterX), FRound(fCenterY));
+
+            nStart = 1;
+            nEnd = nPoints + 1;
+            ImplInitSize(nPoints + 2);
+            mxPointAry[0] = aCenter2;
+            mxPointAry[nEnd] = aCenter2;
+        }
+        else
+        {
+            ImplInitSize( ( PolyStyle::Chord == eStyle ) ? ( nPoints + 1 ) : nPoints );
+            nStart = 0;
+            nEnd = nPoints;
+        }
+
+        for(; nStart < nEnd; nStart++, fStart += fStep )
+        {
+            Point& rPt = mxPointAry[nStart];
+
+            rPt.setX( FRound( fCenterX + fRadX * cos( fStart ) ) );
+            rPt.setY( FRound( fCenterY - fRadY * sin( fStart ) ) );
+        }
+
+        if( PolyStyle::Chord == eStyle )
+            mxPointAry[nPoints] = mxPointAry[0];
+    }
+    else
+        mnPoints = 0;
+}
+
+ImplPolygon::ImplPolygon( const Point& rBezPt1, const Point& rCtrlPt1,
+    const Point& rBezPt2, const Point& rCtrlPt2, sal_uInt16 nPoints )
+{
+    nPoints = ( 0 == nPoints ) ? 25 : ( ( nPoints < 2 ) ? 2 : nPoints );
+
+    const double    fInc = 1.0 / ( nPoints - 1 );
+    double          fK_1 = 0.0, fK1_1 = 1.0;
+    double          fK_2, fK_3, fK1_2, fK1_3;
+    const double    fX0 = rBezPt1.X();
+    const double    fY0 = rBezPt1.Y();
+    const double    fX1 = 3.0 * rCtrlPt1.X();
+    const double    fY1 = 3.0 * rCtrlPt1.Y();
+    const double    fX2 = 3.0 * rCtrlPt2.X();
+    const double    fY2 = 3.0 * rCtrlPt2.Y();
+    const double    fX3 = rBezPt2.X();
+    const double    fY3 = rBezPt2.Y();
+
+    ImplInitSize(nPoints);
+
+    for( sal_uInt16 i = 0; i < nPoints; i++, fK_1 += fInc, fK1_1 -= fInc )
+    {
+        Point& rPt = mxPointAry[i];
+
+        fK_2 = fK_1;
+        fK_2 *= fK_1;
+        fK_3 = fK_2;
+        fK_3 *= fK_1;
+        fK1_2 = fK1_1;
+        fK1_2 *= fK1_1;
+        fK1_3 = fK1_2;
+        fK1_3 *= fK1_1;
+        double fK12 = fK_1 * fK1_2;
+        double fK21 = fK_2 * fK1_1;
+
+        rPt.setX( FRound( fK1_3 * fX0 + fK12 * fX1 + fK21 * fX2 + fK_3 * fX3 ) );
+        rPt.setY( FRound( fK1_3 * fY0 + fK12 * fY1 + fK21 * fY2 + fK_3 * fY3 ) );
+    }
+}
+
+// constructor to convert from basegfx::B2DPolygon
+// #i76891# Needed to change from adding all control points (even for unused
+// edges) and creating a fixed-size Polygon in the first run to creating the
+// minimal Polygon. This requires a temporary Point- and Flag-Array for curves
+// and a memcopy at ImplPolygon creation, but contains no zero-controlpoints
+// for straight edges.
+ImplPolygon::ImplPolygon(const basegfx::B2DPolygon& rPolygon)
+    : mnPoints(0)
+{
+    const bool bCurve(rPolygon.areControlPointsUsed());
+    const bool bClosed(rPolygon.isClosed());
+    sal_uInt32 nB2DLocalCount(rPolygon.count());
+
+    if(bCurve)
+    {
+        // #127979# Reduce source point count hard to the limit of the tools Polygon
+        if(nB2DLocalCount > ((0x0000ffff / 3) - 1))
+        {
+            OSL_FAIL("Polygon::Polygon: Too many points in given B2DPolygon, need to reduce hard to maximum of tools Polygon (!)");
+            nB2DLocalCount = ((0x0000ffff / 3) - 1);
+        }
+
+        // calculate target point count
+        const sal_uInt32 nLoopCount(bClosed ? nB2DLocalCount : (nB2DLocalCount ? nB2DLocalCount - 1 : 0 ));
+
+        if(nLoopCount)
+        {
+            // calculate maximum array size and allocate; prepare insert index
+            const sal_uInt32 nMaxTargetCount((nLoopCount * 3) + 1);
+            ImplInitSize(static_cast< sal_uInt16 >(nMaxTargetCount), true);
+
+            // prepare insert index and current point
+            sal_uInt32 nArrayInsert(0);
+            basegfx::B2DCubicBezier aBezier;
+            aBezier.setStartPoint(rPolygon.getB2DPoint(0));
+
+            for(sal_uInt32 a(0); a < nLoopCount; a++)
+            {
+                // add current point (always) and remember StartPointIndex for evtl. later corrections
+                const Point aStartPoint(FRound(aBezier.getStartPoint().getX()), FRound(aBezier.getStartPoint().getY()));
+                const sal_uInt32 nStartPointIndex(nArrayInsert);
+                mxPointAry[nStartPointIndex] = aStartPoint;
+                mxFlagAry[nStartPointIndex] = PolyFlags::Normal;
+                nArrayInsert++;
+
+                // prepare next segment
+                const sal_uInt32 nNextIndex((a + 1) % nB2DLocalCount);
+                aBezier.setEndPoint(rPolygon.getB2DPoint(nNextIndex));
+                aBezier.setControlPointA(rPolygon.getNextControlPoint(a));
+                aBezier.setControlPointB(rPolygon.getPrevControlPoint(nNextIndex));
+
+                if(aBezier.isBezier())
+                {
+                    // if one is used, add always two control points due to the old schema
+                    mxPointAry[nArrayInsert] = Point(FRound(aBezier.getControlPointA().getX()), FRound(aBezier.getControlPointA().getY()));
+                    mxFlagAry[nArrayInsert] = PolyFlags::Control;
+                    nArrayInsert++;
+
+                    mxPointAry[nArrayInsert] = Point(FRound(aBezier.getControlPointB().getX()), FRound(aBezier.getControlPointB().getY()));
+                    mxFlagAry[nArrayInsert] = PolyFlags::Control;
+                    nArrayInsert++;
+                }
+
+                // test continuity with previous control point to set flag value
+                if(aBezier.getControlPointA() != aBezier.getStartPoint() && (bClosed || a))
+                {
+                    const basegfx::B2VectorContinuity eCont(rPolygon.getContinuityInPoint(a));
+
+                    if(basegfx::B2VectorContinuity::C1 == eCont)
+                    {
+                        mxFlagAry[nStartPointIndex] = PolyFlags::Smooth;
+                    }
+                    else if(basegfx::B2VectorContinuity::C2 == eCont)
+                    {
+                        mxFlagAry[nStartPointIndex] = PolyFlags::Symmetric;
+                    }
+                }
+
+                // prepare next polygon step
+                aBezier.setStartPoint(aBezier.getEndPoint());
+            }
+
+            if(bClosed)
+            {
+                // add first point again as closing point due to old definition
+                mxPointAry[nArrayInsert] = mxPointAry[0];
+                mxFlagAry[nArrayInsert] = PolyFlags::Normal;
+                nArrayInsert++;
+            }
+            else
+            {
+                // add last point as closing point
+                const basegfx::B2DPoint aClosingPoint(rPolygon.getB2DPoint(nB2DLocalCount - 1));
+                const Point aEnd(FRound(aClosingPoint.getX()), FRound(aClosingPoint.getY()));
+                mxPointAry[nArrayInsert] = aEnd;
+                mxFlagAry[nArrayInsert] = PolyFlags::Normal;
+                nArrayInsert++;
+            }
+
+            DBG_ASSERT(nArrayInsert <= nMaxTargetCount, "Polygon::Polygon from basegfx::B2DPolygon: wrong max point count estimation (!)");
+
+            if(nArrayInsert != nMaxTargetCount)
+            {
+                ImplSetSize(static_cast< sal_uInt16 >(nArrayInsert));
+            }
+        }
+    }
+    else
+    {
+        // #127979# Reduce source point count hard to the limit of the tools Polygon
+        if(nB2DLocalCount > (0x0000ffff - 1))
+        {
+            OSL_FAIL("Polygon::Polygon: Too many points in given B2DPolygon, need to reduce hard to maximum of tools Polygon (!)");
+            nB2DLocalCount = (0x0000ffff - 1);
+        }
+
+        if(nB2DLocalCount)
+        {
+            // point list creation
+            const sal_uInt32 nTargetCount(nB2DLocalCount + (bClosed ? 1 : 0));
+            ImplInitSize(static_cast< sal_uInt16 >(nTargetCount));
+            sal_uInt16 nIndex(0);
+
+            for(sal_uInt32 a(0); a < nB2DLocalCount; a++)
+            {
+                basegfx::B2DPoint aB2DPoint(rPolygon.getB2DPoint(a));
+                Point aPoint(FRound(aB2DPoint.getX()), FRound(aB2DPoint.getY()));
+                mxPointAry[nIndex++] = aPoint;
+            }
+
+            if(bClosed)
+            {
+                // add first point as closing point
+                mxPointAry[nIndex] = mxPointAry[0];
+            }
+        }
+    }
+}
+
+bool ImplPolygon::operator==( const ImplPolygon& rCandidate) const
+{
+    return mnPoints == rCandidate.mnPoints &&
+           mxFlagAry.get() == rCandidate.mxFlagAry.get() &&
+           mxPointAry.get() == rCandidate.mxPointAry.get();
+}
+
+void ImplPolygon::ImplInitSize(sal_uInt16 nInitSize, bool bFlags)
+{
+    if (nInitSize)
+    {
+        mxPointAry.reset(new Point[nInitSize]);
+    }
+
+    if (bFlags)
+    {
+        mxFlagAry.reset(new PolyFlags[nInitSize]);
+        memset(mxFlagAry.get(), 0, nInitSize);
+    }
+
+    mnPoints = nInitSize;
+}
+
+void ImplPolygon::ImplSetSize( sal_uInt16 nNewSize, bool bResize )
+{
+    if( mnPoints == nNewSize )
+        return;
+
+    std::unique_ptr<Point[]> xNewAry;
+
+    if (nNewSize)
+    {
+        const std::size_t nNewSz(static_cast<std::size_t>(nNewSize)*sizeof(Point));
+        xNewAry.reset(new Point[nNewSize]);
+
+        if ( bResize )
+        {
+            // Copy the old points
+            if ( mnPoints < nNewSize )
+            {
+                // New points are already implicitly initialized to zero
+                const std::size_t nOldSz(mnPoints * sizeof(Point));
+                if (mxPointAry)
+                    memcpy(xNewAry.get(), mxPointAry.get(), nOldSz);
+            }
+            else
+            {
+                if (mxPointAry)
+                    memcpy(xNewAry.get(), mxPointAry.get(), nNewSz);
+            }
+        }
+    }
+
+    mxPointAry = std::move(xNewAry);
+
+    // take FlagArray into account, if applicable
+    if( mxFlagAry )
+    {
+        std::unique_ptr<PolyFlags[]> xNewFlagAry;
+
+        if( nNewSize )
+        {
+            xNewFlagAry.reset(new PolyFlags[nNewSize]);
+
+            if( bResize )
+            {
+                // copy the old flags
+                if ( mnPoints < nNewSize )
+                {
+                    // initialize new flags to zero
+                    memset(xNewFlagAry.get() + mnPoints, 0, nNewSize-mnPoints);
+                    memcpy(xNewFlagAry.get(), mxFlagAry.get(), mnPoints);
+                }
+                else
+                    memcpy(xNewFlagAry.get(), mxFlagAry.get(), nNewSize);
+            }
+        }
+
+        mxFlagAry = std::move(xNewFlagAry);
+    }
+
+    mnPoints   = nNewSize;
+}
+
+bool ImplPolygon::ImplSplit( sal_uInt16 nPos, sal_uInt16 nSpace, ImplPolygon const * pInitPoly )
+{
+    //Can't fit this in :-(, throw ?
+    if (mnPoints + nSpace > USHRT_MAX)
+    {
+        SAL_WARN("tools", "Polygon needs " << mnPoints + nSpace << " points, but only " << USHRT_MAX << " possible");
+        return false;
+    }
+
+    const sal_uInt16    nNewSize = mnPoints + nSpace;
+    const std::size_t   nSpaceSize = static_cast<std::size_t>(nSpace) * sizeof(Point);
+
+    if( nPos >= mnPoints )
+    {
+        // Append at the back
+        nPos = mnPoints;
+        ImplSetSize( nNewSize );
+
+        if( pInitPoly )
+        {
+            memcpy(mxPointAry.get() + nPos, pInitPoly->mxPointAry.get(), nSpaceSize);
+
+            if (pInitPoly->mxFlagAry)
+                memcpy(mxFlagAry.get() + nPos, pInitPoly->mxFlagAry.get(), nSpace);
+        }
+    }
+    else
+    {
+        const sal_uInt16    nSecPos = nPos + nSpace;
+        const sal_uInt16    nRest = mnPoints - nPos;
+
+        std::unique_ptr<Point[]> xNewAry(new Point[nNewSize]);
+        memcpy(xNewAry.get(), mxPointAry.get(), nPos * sizeof(Point));
+
+        if( pInitPoly )
+            memcpy(xNewAry.get() + nPos, pInitPoly->mxPointAry.get(), nSpaceSize);
+
+        memcpy(xNewAry.get() + nSecPos, mxPointAry.get() + nPos, nRest * sizeof(Point));
+        mxPointAry = std::move(xNewAry);
+
+        // consider FlagArray
+        if (mxFlagAry)
+        {
+            std::unique_ptr<PolyFlags[]> xNewFlagAry(new PolyFlags[nNewSize]);
+
+            memcpy(xNewFlagAry.get(), mxFlagAry.get(), nPos);
+
+            if (pInitPoly && pInitPoly->mxFlagAry)
+                memcpy(xNewFlagAry.get() + nPos, pInitPoly->mxFlagAry.get(), nSpace);
+            else
+                memset(xNewFlagAry.get() + nPos, 0, nSpace);
+
+            memcpy(xNewFlagAry.get() + nSecPos, mxFlagAry.get() + nPos, nRest);
+            mxFlagAry = std::move(xNewFlagAry);
+        }
+
+        mnPoints   = nNewSize;
+    }
+
+    return true;
+}
+
+void ImplPolygon::ImplCreateFlagArray()
+{
+    if (!mxFlagAry)
+    {
+        mxFlagAry.reset(new PolyFlags[mnPoints]);
+        memset(mxFlagAry.get(), 0, mnPoints);
+    }
+}
+
+namespace {
+
+class ImplPointFilter
+{
+public:
+    virtual void LastPoint() = 0;
+    virtual void Input( const Point& rPoint ) = 0;
+
+protected:
+    ~ImplPointFilter() {}
+};
+
+class ImplPolygonPointFilter : public ImplPointFilter
+{
+    ImplPolygon maPoly;
+    sal_uInt16  mnSize;
+public:
+    explicit ImplPolygonPointFilter(sal_uInt16 nDestSize)
+        : maPoly(nDestSize)
+        , mnSize(0)
+    {
+    }
+
+    virtual ~ImplPolygonPointFilter()
+    {
+    }
+
+    virtual void    LastPoint() override;
+    virtual void    Input( const Point& rPoint ) override;
+
+    ImplPolygon&    get() { return maPoly; }
+};
+
+}
+
+void ImplPolygonPointFilter::Input( const Point& rPoint )
+{
+    if ( !mnSize || (rPoint != maPoly.mxPointAry[mnSize-1]) )
+    {
+        mnSize++;
+        if ( mnSize > maPoly.mnPoints )
+            maPoly.ImplSetSize( mnSize );
+        maPoly.mxPointAry[mnSize-1] = rPoint;
+    }
+}
+
+void ImplPolygonPointFilter::LastPoint()
+{
+    if ( mnSize < maPoly.mnPoints )
+        maPoly.ImplSetSize( mnSize );
+};
+
+namespace {
+
+class ImplEdgePointFilter : public ImplPointFilter
+{
+    Point               maFirstPoint;
+    Point               maLastPoint;
+    ImplPointFilter&    mrNextFilter;
+    const tools::Long          mnLow;
+    const tools::Long          mnHigh;
+    const int           mnEdge;
+    int                 mnLastOutside;
+    bool                mbFirst;
+
+public:
+                        ImplEdgePointFilter( int nEdge, tools::Long nLow, tools::Long nHigh,
+                                             ImplPointFilter& rNextFilter ) :
+                            mrNextFilter( rNextFilter ),
+                            mnLow( nLow ),
+                            mnHigh( nHigh ),
+                            mnEdge( nEdge ),
+                            mnLastOutside( 0 ),
+                            mbFirst( true )
+                        {
+                        }
+
+    virtual             ~ImplEdgePointFilter() {}
+
+    Point               EdgeSection( const Point& rPoint, int nEdge ) const;
+    int                 VisibleSide( const Point& rPoint ) const;
+    bool                IsPolygon() const
+                            { return maFirstPoint == maLastPoint; }
+
+    virtual void        Input( const Point& rPoint ) override;
+    virtual void        LastPoint() override;
+};
+
+}
+
+inline int ImplEdgePointFilter::VisibleSide( const Point& rPoint ) const
+{
+    if ( mnEdge & EDGE_HORZ )
+    {
+        return rPoint.X() < mnLow ? EDGE_LEFT :
+                                     rPoint.X() > mnHigh ? EDGE_RIGHT : 0;
+    }
+    else
+    {
+        return rPoint.Y() < mnLow ? EDGE_TOP :
+                                     rPoint.Y() > mnHigh ? EDGE_BOTTOM : 0;
+    }
+}
+
+Point ImplEdgePointFilter::EdgeSection( const Point& rPoint, int nEdge ) const
+{
+    tools::Long lx = maLastPoint.X();
+    tools::Long ly = maLastPoint.Y();
+    tools::Long md = rPoint.X() - lx;
+    tools::Long mn = rPoint.Y() - ly;
+    tools::Long nNewX;
+    tools::Long nNewY;
+
+    if ( nEdge & EDGE_VERT )
+    {
+        nNewY = (nEdge == EDGE_TOP) ? mnLow : mnHigh;
+        tools::Long dy = nNewY - ly;
+        if ( !md )
+            nNewX = lx;
+        else if ( (LONG_MAX / std::abs(md)) >= std::abs(dy) )
+            nNewX = (dy * md) / mn + lx;
+        else
+        {
+            BigInt ady = dy;
+            ady *= md;
+            if( ady.IsNeg() )
+                if( mn < 0 )
+                    ady += mn/2;
+                else
+                    ady -= (mn-1)/2;
+            else
+                if( mn < 0 )
+                    ady -= (mn+1)/2;
+                else
+                    ady += mn/2;
+            ady /= mn;
+            nNewX = static_cast<tools::Long>(ady) + lx;
+        }
+    }
+    else
+    {
+        nNewX = (nEdge == EDGE_LEFT) ? mnLow : mnHigh;
+        tools::Long dx = nNewX - lx;
+        if ( !mn )
+            nNewY = ly;
+        else if ( (LONG_MAX / std::abs(mn)) >= std::abs(dx) )
+            nNewY = (dx * mn) / md + ly;
+        else
+        {
+            BigInt adx = dx;
+            adx *= mn;
+            if( adx.IsNeg() )
+                if( md < 0 )
+                    adx += md/2;
+                else
+                    adx -= (md-1)/2;
+            else
+                if( md < 0 )
+                    adx -= (md+1)/2;
+                else
+                    adx += md/2;
+            adx /= md;
+            nNewY = static_cast<tools::Long>(adx) + ly;
+        }
+    }
+
+    return Point( nNewX, nNewY );
+}
+
+void ImplEdgePointFilter::Input( const Point& rPoint )
+{
+    int nOutside = VisibleSide( rPoint );
+
+    if ( mbFirst )
+    {
+        maFirstPoint = rPoint;
+        mbFirst      = false;
+        if ( !nOutside )
+            mrNextFilter.Input( rPoint );
+    }
+    else if ( rPoint == maLastPoint )
+        return;
+    else if ( !nOutside )
+    {
+        if ( mnLastOutside )
+            mrNextFilter.Input( EdgeSection( rPoint, mnLastOutside ) );
+        mrNextFilter.Input( rPoint );
+    }
+    else if ( !mnLastOutside )
+        mrNextFilter.Input( EdgeSection( rPoint, nOutside ) );
+    else if ( nOutside != mnLastOutside )
+    {
+        mrNextFilter.Input( EdgeSection( rPoint, mnLastOutside ) );
+        mrNextFilter.Input( EdgeSection( rPoint, nOutside ) );
+    }
+
+    maLastPoint    = rPoint;
+    mnLastOutside  = nOutside;
+}
+
+void ImplEdgePointFilter::LastPoint()
+{
+    if ( !mbFirst )
+    {
+        int nOutside = VisibleSide( maFirstPoint );
+
+        if ( nOutside != mnLastOutside )
+            Input( maFirstPoint );
+        mrNextFilter.LastPoint();
+    }
+}
+
+namespace tools
+{
+
+tools::Polygon Polygon::SubdivideBezier( const tools::Polygon& rPoly )
+{
+    tools::Polygon aPoly;
+
+    // #100127# Use adaptive subdivide instead of fixed 25 segments
+    rPoly.AdaptiveSubdivide( aPoly );
+
+    return aPoly;
+}
+
+Polygon::Polygon() : mpImplPolygon(ImplPolygon())
+{
+}
+
+Polygon::Polygon( sal_uInt16 nSize ) : mpImplPolygon(ImplPolygon(nSize))
+{
+}
+
+Polygon::Polygon( sal_uInt16 nPoints, const Point* pPtAry, const PolyFlags* pFlagAry ) : mpImplPolygon(ImplPolygon(nPoints, pPtAry, pFlagAry))
+{
+}
+
+Polygon::Polygon( const tools::Polygon& rPoly ) : mpImplPolygon(rPoly.mpImplPolygon)
+{
+}
+
+Polygon::Polygon( tools::Polygon&& rPoly) noexcept
+    : mpImplPolygon(std::move(rPoly.mpImplPolygon))
+{
+}
+
+Polygon::Polygon( const tools::Rectangle& rRect ) : mpImplPolygon(ImplPolygon(rRect))
+{
+}
+
+Polygon::Polygon( const tools::Rectangle& rRect, sal_uInt32 nHorzRound, sal_uInt32 nVertRound )
+    : mpImplPolygon(ImplPolygon(rRect, nHorzRound, nVertRound))
+{
+}
+
+Polygon::Polygon( const Point& rCenter, tools::Long nRadX, tools::Long nRadY )
+    : mpImplPolygon(ImplPolygon(rCenter, nRadX, nRadY))
+{
+}
+
+Polygon::Polygon(const tools::Rectangle& rBound, const Point& rStart, const Point& rEnd,
+                 PolyStyle eStyle, const bool bClockWiseArcDirection)
+    : mpImplPolygon(ImplPolygon(rBound, rStart, rEnd, eStyle, bClockWiseArcDirection))
+{
+}
+
+Polygon::Polygon( const Point& rBezPt1, const Point& rCtrlPt1,
+                  const Point& rBezPt2, const Point& rCtrlPt2,
+                  sal_uInt16 nPoints ) : mpImplPolygon(ImplPolygon(rBezPt1, rCtrlPt1, rBezPt2, rCtrlPt2, nPoints))
+{
+}
+
+Polygon::~Polygon()
+{
+}
+
+Point * Polygon::GetPointAry()
+{
+    return mpImplPolygon->mxPointAry.get();
+}
+
+const Point* Polygon::GetConstPointAry() const
+{
+    return mpImplPolygon->mxPointAry.get();
+}
+
+const PolyFlags* Polygon::GetConstFlagAry() const
+{
+    return mpImplPolygon->mxFlagAry.get();
+}
+
+void Polygon::SetPoint( const Point& rPt, sal_uInt16 nPos )
+{
+    DBG_ASSERT( nPos < mpImplPolygon->mnPoints,
+                "Polygon::SetPoint(): nPos >= nPoints" );
+
+    mpImplPolygon->mxPointAry[nPos] = rPt;
+}
+
+void Polygon::SetFlags( sal_uInt16 nPos, PolyFlags eFlags )
+{
+    DBG_ASSERT( nPos < mpImplPolygon->mnPoints,
+                "Polygon::SetFlags(): nPos >= nPoints" );
+
+    // we do only want to create the flag array if there
+    // is at least one flag different to PolyFlags::Normal
+    if ( eFlags != PolyFlags::Normal )
+    {
+        mpImplPolygon->ImplCreateFlagArray();
+        mpImplPolygon->mxFlagAry[ nPos ] = eFlags;
+    }
+}
+
+const Point& Polygon::GetPoint( sal_uInt16 nPos ) const
+{
+    DBG_ASSERT( nPos < mpImplPolygon->mnPoints,
+                "Polygon::GetPoint(): nPos >= nPoints" );
+
+    return mpImplPolygon->mxPointAry[nPos];
+}
+
+PolyFlags Polygon::GetFlags( sal_uInt16 nPos ) const
+{
+    DBG_ASSERT( nPos < mpImplPolygon->mnPoints,
+                "Polygon::GetFlags(): nPos >= nPoints" );
+    return mpImplPolygon->mxFlagAry
+           ? mpImplPolygon->mxFlagAry[ nPos ]
+           : PolyFlags::Normal;
+}
+
+bool Polygon::HasFlags() const
+{
+    return bool(mpImplPolygon->mxFlagAry);
+}
+
+bool Polygon::IsRect() const
+{
+    bool bIsRect = false;
+    if (!mpImplPolygon->mxFlagAry)
+    {
+        if ( ( ( mpImplPolygon->mnPoints == 5 ) && ( mpImplPolygon->mxPointAry[ 0 ] == mpImplPolygon->mxPointAry[ 4 ] ) ) ||
+             ( mpImplPolygon->mnPoints == 4 ) )
+        {
+            if ( ( mpImplPolygon->mxPointAry[ 0 ].X() == mpImplPolygon->mxPointAry[ 3 ].X() ) &&
+                 ( mpImplPolygon->mxPointAry[ 0 ].Y() == mpImplPolygon->mxPointAry[ 1 ].Y() ) &&
+                 ( mpImplPolygon->mxPointAry[ 1 ].X() == mpImplPolygon->mxPointAry[ 2 ].X() ) &&
+                 ( mpImplPolygon->mxPointAry[ 2 ].Y() == mpImplPolygon->mxPointAry[ 3 ].Y() ) )
+                bIsRect = true;
+        }
+    }
+    return bIsRect;
+}
+
+void Polygon::SetSize( sal_uInt16 nNewSize )
+{
+    if( nNewSize != mpImplPolygon->mnPoints )
+    {
+        mpImplPolygon->ImplSetSize( nNewSize );
+    }
+}
+
+sal_uInt16 Polygon::GetSize() const
+{
+    return mpImplPolygon->mnPoints;
+}
+
+void Polygon::Clear()
+{
+    mpImplPolygon = ImplType(ImplPolygon());
+}
+
+double Polygon::CalcDistance( sal_uInt16 nP1, sal_uInt16 nP2 ) const
+{
+    DBG_ASSERT( nP1 < mpImplPolygon->mnPoints,
+                "Polygon::CalcDistance(): nPos1 >= nPoints" );
+    DBG_ASSERT( nP2 < mpImplPolygon->mnPoints,
+                "Polygon::CalcDistance(): nPos2 >= nPoints" );
+
+    const Point& rP1 = mpImplPolygon->mxPointAry[ nP1 ];
+    const Point& rP2 = mpImplPolygon->mxPointAry[ nP2 ];
+    const double fDx = rP2.X() - rP1.X();
+    const double fDy = rP2.Y() - rP1.Y();
+
+    return std::hypot( fDx, fDy );
+}
+
+void Polygon::Optimize( PolyOptimizeFlags nOptimizeFlags )
+{
+    sal_uInt16 nSize = mpImplPolygon->mnPoints;
+
+    if( !(bool(nOptimizeFlags) && nSize) )
+        return;
+
+    if( nOptimizeFlags & PolyOptimizeFlags::EDGES )
+    {
+        const tools::Rectangle aBound( GetBoundRect() );
+        const double    fArea = ( aBound.GetWidth() + aBound.GetHeight() ) * 0.5;
+        const sal_uInt16 nPercent = 50;
+
+        Optimize( PolyOptimizeFlags::NO_SAME );
+        ImplReduceEdges( *this, fArea, nPercent );
+    }
+    else if( nOptimizeFlags & PolyOptimizeFlags::NO_SAME )
+    {
+        tools::Polygon aNewPoly;
+        const Point& rFirst = mpImplPolygon->mxPointAry[ 0 ];
+
+        while( nSize && ( mpImplPolygon->mxPointAry[ nSize - 1 ] == rFirst ) )
+            nSize--;
+
+        if( nSize > 1 )
+        {
+            sal_uInt16 nLast = 0, nNewCount = 1;
+
+            aNewPoly.SetSize( nSize );
+            aNewPoly[ 0 ] = rFirst;
+
+            for( sal_uInt16 i = 1; i < nSize; i++ )
+            {
+                if( mpImplPolygon->mxPointAry[ i ] != mpImplPolygon->mxPointAry[ nLast ])
+                {
+                    nLast = i;
+                    aNewPoly[ nNewCount++ ] = mpImplPolygon->mxPointAry[ i ];
+                }
+            }
+
+            if( nNewCount == 1 )
+                aNewPoly.Clear();
+            else
+                aNewPoly.SetSize( nNewCount );
+        }
+
+        *this = aNewPoly;
+    }
+
+    nSize = mpImplPolygon->mnPoints;
+
+    if( nSize > 1 )
+    {
+        if( ( nOptimizeFlags & PolyOptimizeFlags::CLOSE ) &&
+            ( mpImplPolygon->mxPointAry[ 0 ] != mpImplPolygon->mxPointAry[ nSize - 1 ] ) )
+        {
+            SetSize( mpImplPolygon->mnPoints + 1 );
+            mpImplPolygon->mxPointAry[ mpImplPolygon->mnPoints - 1 ] = mpImplPolygon->mxPointAry[ 0 ];
+        }
+    }
+}
+
+
+/** Recursively subdivide cubic bezier curve via deCasteljau.
+
+   @param rPoints
+   Output vector, where the subdivided polylines are written to.
+
+   @param d
+   Squared difference of curve to a straight line
+
+   @param P*
+   Exactly four points, interpreted as support and control points of
+   a cubic bezier curve. Must be in device coordinates, since stop
+   criterion is based on the following assumption: the device has a
+   finite resolution, it is thus sufficient to stop subdivision if the
+   curve does not deviate more than one pixel from a straight line.
+
+*/
+static void ImplAdaptiveSubdivide( std::vector<Point>& rPoints,
+                                   const double old_d2,
+                                   int recursionDepth,
+                                   const double d2,
+                                   const double P1x, const double P1y,
+                                   const double P2x, const double P2y,
+                                   const double P3x, const double P3y,
+                                   const double P4x, const double P4y )
+{
+    // Hard limit on recursion depth, empiric number.
+    enum {maxRecursionDepth=128};
+
+    // Perform bezier flatness test (lecture notes from R. Schaback,
+    // Mathematics of Computer-Aided Design, Uni Goettingen, 2000)
+
+    // ||P(t) - L(t)|| <= max     ||b_j - b_0 - j/n(b_n - b_0)||
+    //                    0<=j<=n
+
+    // What is calculated here is an upper bound to the distance from
+    // a line through b_0 and b_3 (P1 and P4 in our notation) and the
+    // curve. We can drop 0 and n from the running indices, since the
+    // argument of max becomes zero for those cases.
+    const double fJ1x( P2x - P1x - 1.0/3.0*(P4x - P1x) );
+    const double fJ1y( P2y - P1y - 1.0/3.0*(P4y - P1y) );
+    const double fJ2x( P3x - P1x - 2.0/3.0*(P4x - P1x) );
+    const double fJ2y( P3y - P1y - 2.0/3.0*(P4y - P1y) );
+    const double distance2( ::std::max( fJ1x*fJ1x + fJ1y*fJ1y,
+                                        fJ2x*fJ2x + fJ2y*fJ2y) );
+
+    // stop if error measure does not improve anymore. This is a
+    // safety guard against floating point inaccuracies.
+    // stop at recursion level 128. This is a safety guard against
+    // floating point inaccuracies.
+    // stop if distance from line is guaranteed to be bounded by d
+    if( old_d2 > d2 &&
+        recursionDepth < maxRecursionDepth &&
+        distance2 >= d2 &&
+        rPoints.size() < SAL_MAX_UINT16 )
+    {
+        // deCasteljau bezier arc, split at t=0.5
+        // Foley/vanDam, p. 508
+        const double L1x( P1x ),             L1y( P1y );
+        const double L2x( (P1x + P2x)*0.5 ), L2y( (P1y + P2y)*0.5 );
+        const double Hx ( (P2x + P3x)*0.5 ), Hy ( (P2y + P3y)*0.5 );
+        const double L3x( (L2x + Hx)*0.5 ),  L3y( (L2y + Hy)*0.5 );
+        const double R4x( P4x ),             R4y( P4y );
+        const double R3x( (P3x + P4x)*0.5 ), R3y( (P3y + P4y)*0.5 );
+        const double R2x( (Hx + R3x)*0.5 ),  R2y( (Hy + R3y)*0.5 );
+        const double R1x( (L3x + R2x)*0.5 ), R1y( (L3y + R2y)*0.5 );
+        const double L4x( R1x ),             L4y( R1y );
+
+        // subdivide further
+        ++recursionDepth;
+        ImplAdaptiveSubdivide(rPoints, distance2, recursionDepth, d2, L1x, L1y, L2x, L2y, L3x, L3y, L4x, L4y);
+        ImplAdaptiveSubdivide(rPoints, distance2, recursionDepth, d2, R1x, R1y, R2x, R2y, R3x, R3y, R4x, R4y);
+    }
+    else
+    {
+        // requested resolution reached.
+        // Add end points to output iterator.
+        // order is preserved, since this is so to say depth first traversal.
+        rPoints.push_back(Point(FRound(P1x), FRound(P1y)));
+    }
+}
+
+void Polygon::AdaptiveSubdivide( Polygon& rResult, const double d ) const
+{
+    if (!mpImplPolygon->mxFlagAry)
+    {
+        rResult = *this;
+    }
+    else
+    {
+        sal_uInt16 i;
+        sal_uInt16 nPts( GetSize() );
+        ::std::vector< Point > aPoints;
+        aPoints.reserve( nPts );
+
+        for(i=0; i<nPts;)
+        {
+            if( ( i + 3 ) < nPts )
+            {
+                PolyFlags P1( mpImplPolygon->mxFlagAry[ i ] );
+                PolyFlags P4( mpImplPolygon->mxFlagAry[ i + 3 ] );
+
+                if( ( PolyFlags::Normal == P1 || PolyFlags::Smooth == P1 || PolyFlags::Symmetric == P1 ) &&
+                    ( PolyFlags::Control == mpImplPolygon->mxFlagAry[ i + 1 ] ) &&
+                    ( PolyFlags::Control == mpImplPolygon->mxFlagAry[ i + 2 ] ) &&
+                    ( PolyFlags::Normal == P4 || PolyFlags::Smooth == P4 || PolyFlags::Symmetric == P4 ) )
+                {
+                    ImplAdaptiveSubdivide( aPoints, d*d+1.0, 0, d*d,
+                                           mpImplPolygon->mxPointAry[ i ].X(),   mpImplPolygon->mxPointAry[ i ].Y(),
+                                           mpImplPolygon->mxPointAry[ i+1 ].X(), mpImplPolygon->mxPointAry[ i+1 ].Y(),
+                                           mpImplPolygon->mxPointAry[ i+2 ].X(), mpImplPolygon->mxPointAry[ i+2 ].Y(),
+                                           mpImplPolygon->mxPointAry[ i+3 ].X(), mpImplPolygon->mxPointAry[ i+3 ].Y() );
+                    i += 3;
+                    continue;
+                }
+            }
+
+            aPoints.push_back(mpImplPolygon->mxPointAry[i++]);
+
+            if (aPoints.size() >= SAL_MAX_UINT16)
+            {
+                OSL_ENSURE(aPoints.size() < SAL_MAX_UINT16,
+                    "Polygon::AdaptiveSubdivision created polygon too many points;"
+                    " using original polygon instead");
+
+                // The resulting polygon can not hold all the points
+                // that we have created so far.  Stop the subdivision
+                // and return a copy of the unmodified polygon.
+                rResult = *this;
+                return;
+            }
+        }
+
+        // fill result polygon
+        rResult = tools::Polygon( static_cast<sal_uInt16>(aPoints.size()) ); // ensure sufficient size for copy
+        ::std::copy(aPoints.begin(), aPoints.end(), rResult.mpImplPolygon->mxPointAry.get());
+    }
+}
+
+namespace {
+
+class Vector2D
+{
+private:
+    double              mfX;
+    double              mfY;
+public:
+    explicit     Vector2D( const Point& rPoint ) : mfX( rPoint.X() ), mfY( rPoint.Y() ) {};
+    double       GetLength() const { return hypot( mfX, mfY ); }
+    Vector2D&    operator-=( const Vector2D& rVec ) { mfX -= rVec.mfX; mfY -= rVec.mfY; return *this; }
+    double       Scalar( const Vector2D& rVec ) const { return mfX * rVec.mfX + mfY * rVec.mfY ; }
+    Vector2D&    Normalize();
+    bool         IsPositive( Vector2D const & rVec ) const { return ( mfX * rVec.mfY - mfY * rVec.mfX ) >= 0.0; }
+    bool         IsNegative( Vector2D const & rVec ) const { return !IsPositive( rVec ); }
+};
+
+}
+
+Vector2D& Vector2D::Normalize()
+{
+    double fLen = Scalar( *this );
+
+    if( ( fLen != 0.0 ) && ( fLen != 1.0 ) )
+    {
+        fLen = sqrt( fLen );
+        if( fLen != 0.0 )
+        {
+            mfX /= fLen;
+            mfY /= fLen;
+        }
+    }
+
+    return *this;
+}
+
+void Polygon::ImplReduceEdges( tools::Polygon& rPoly, const double& rArea, sal_uInt16 nPercent )
+{
+    const double    fBound = 2000.0 * ( 100 - nPercent ) * 0.01;
+    sal_uInt16      nNumNoChange = 0,
+                    nNumRuns = 0;
+
+    while( nNumNoChange < 2 )
+    {
+        sal_uInt16  nPntCnt = rPoly.GetSize(), nNewPos = 0;
+        tools::Polygon aNewPoly( nPntCnt );
+        bool bChangeInThisRun = false;
+
+        for( sal_uInt16 n = 0; n < nPntCnt; n++ )
+        {
+            bool bDeletePoint = false;
+
+            if( ( n + nNumRuns ) % 2 )
+            {
+                sal_uInt16      nIndPrev = !n ? nPntCnt - 1 : n - 1;
+                sal_uInt16      nIndPrevPrev = !nIndPrev ? nPntCnt - 1 : nIndPrev - 1;
+                sal_uInt16      nIndNext = ( n == nPntCnt-1 ) ? 0 : n + 1;
+                sal_uInt16      nIndNextNext = ( nIndNext == nPntCnt - 1 ) ? 0 : nIndNext + 1;
+                Vector2D    aVec1( rPoly[ nIndPrev ] ); aVec1 -= Vector2D(rPoly[ nIndPrevPrev ]);
+                Vector2D    aVec2( rPoly[ n ] ); aVec2 -= Vector2D(rPoly[ nIndPrev ]);
+                Vector2D    aVec3( rPoly[ nIndNext ] ); aVec3 -= Vector2D(rPoly[ n ]);
+                Vector2D    aVec4( rPoly[ nIndNextNext ] ); aVec4 -= Vector2D(rPoly[ nIndNext ]);
+                double      fDist1 = aVec1.GetLength(), fDist2 = aVec2.GetLength();
+                double      fDist3 = aVec3.GetLength(), fDist4 = aVec4.GetLength();
+                double      fTurnB = aVec2.Normalize().Scalar( aVec3.Normalize() );
+
+                if( fabs( fTurnB ) < ( 1.0 + SMALL_DVALUE ) && fabs( fTurnB ) > ( 1.0 - SMALL_DVALUE ) )
+                    bDeletePoint = true;
+                else
+                {
+                    Vector2D    aVecB( rPoly[ nIndNext ] );
+                    aVecB -= Vector2D(rPoly[ nIndPrev ] );
+                    double      fDistB = aVecB.GetLength();
+                    double      fLenWithB = fDist2 + fDist3;
+                    double      fLenFact = ( fDistB != 0.0 ) ? fLenWithB / fDistB : 1.0;
+                    double      fTurnPrev = aVec1.Normalize().Scalar( aVec2 );
+                    double      fTurnNext = aVec3.Scalar( aVec4.Normalize() );
+                    double      fGradPrev, fGradB, fGradNext;
+
+                    if( fabs( fTurnPrev ) < ( 1.0 + SMALL_DVALUE ) && fabs( fTurnPrev ) > ( 1.0 - SMALL_DVALUE ) )
+                        fGradPrev = 0.0;
+                    else
+                        fGradPrev = basegfx::rad2deg(acos(fTurnPrev)) * (aVec1.IsNegative(aVec2) ? -1 : 1);
+
+                    fGradB = basegfx::rad2deg(acos(fTurnB)) * (aVec2.IsNegative(aVec3) ? -1 : 1);
+
+                    if( fabs( fTurnNext ) < ( 1.0 + SMALL_DVALUE ) && fabs( fTurnNext ) > ( 1.0 - SMALL_DVALUE ) )
+                        fGradNext = 0.0;
+                    else
+                        fGradNext = basegfx::rad2deg(acos(fTurnNext)) * (aVec3.IsNegative(aVec4) ? -1 : 1);
+
+                    if( ( fGradPrev > 0.0 && fGradB < 0.0 && fGradNext > 0.0 ) ||
+                        ( fGradPrev < 0.0 && fGradB > 0.0 && fGradNext < 0.0 ) )
+                    {
+                        if( ( fLenFact < ( FSQRT2 + SMALL_DVALUE ) ) &&
+                            ( ( ( fDist1 + fDist4 ) / ( fDist2 + fDist3 ) ) * 2000.0 ) > fBound )
+                        {
+                            bDeletePoint = true;
+                        }
+                    }
+                    else
+                    {
+                        double fRelLen = 1.0 - sqrt( fDistB / rArea );
+
+                        if( fRelLen < 0.0 )
+                            fRelLen = 0.0;
+                        else if( fRelLen > 1.0 )
+                            fRelLen = 1.0;
+
+                        if( ( std::round( ( fLenFact - 1.0 ) * 1000000.0 ) < fBound ) &&
+                            ( fabs( fGradB ) <= ( fRelLen * fBound * 0.01 ) ) )
+                        {
+                            bDeletePoint = true;
+                        }
+                    }
+                }
+            }
+
+            if( !bDeletePoint )
+                aNewPoly[ nNewPos++ ] = rPoly[ n ];
+            else
+                bChangeInThisRun = true;
+        }
+
+        if( bChangeInThisRun && nNewPos )
+        {
+            aNewPoly.SetSize( nNewPos );
+            rPoly = aNewPoly;
+            nNumNoChange = 0;
+        }
+        else
+            nNumNoChange++;
+
+        nNumRuns++;
+    }
+}
+
+void Polygon::Move( tools::Long nHorzMove, tools::Long nVertMove )
+{
+    // This check is required for DrawEngine
+    if ( !nHorzMove && !nVertMove )
+        return;
+
+    // Move points
+    sal_uInt16 nCount = mpImplPolygon->mnPoints;
+    for ( sal_uInt16 i = 0; i < nCount; i++ )
+    {
+        Point& rPt = mpImplPolygon->mxPointAry[i];
+        rPt.AdjustX(nHorzMove );
+        rPt.AdjustY(nVertMove );
+    }
+}
+
+void Polygon::Translate(const Point& rTrans)
+{
+    for ( sal_uInt16 i = 0, nCount = mpImplPolygon->mnPoints; i < nCount; i++ )
+        mpImplPolygon->mxPointAry[ i ] += rTrans;
+}
+
+void Polygon::Scale( double fScaleX, double fScaleY )
+{
+    for ( sal_uInt16 i = 0, nCount = mpImplPolygon->mnPoints; i < nCount; i++ )
+    {
+        Point& rPnt = mpImplPolygon->mxPointAry[i];
+        rPnt.setX( static_cast<tools::Long>( fScaleX * rPnt.X() ) );
+        rPnt.setY( static_cast<tools::Long>( fScaleY * rPnt.Y() ) );
+    }
+}
+
+void Polygon::Rotate( const Point& rCenter, Degree10 nAngle10 )
+{
+    nAngle10 %= 3600_deg10;
+
+    if( nAngle10 )
+    {
+        const double fAngle = toRadians(nAngle10);
+        Rotate( rCenter, sin( fAngle ), cos( fAngle ) );
+    }
+}
+
+void Polygon::Rotate( const Point& rCenter, double fSin, double fCos )
+{
+    tools::Long nCenterX = rCenter.X();
+    tools::Long nCenterY = rCenter.Y();
+
+    for( sal_uInt16 i = 0, nCount = mpImplPolygon->mnPoints; i < nCount; i++ )
+    {
+        Point& rPt = mpImplPolygon->mxPointAry[ i ];
+
+        const tools::Long nX = rPt.X() - nCenterX;
+        const tools::Long nY = rPt.Y() - nCenterY;
+        rPt.setX( FRound(fCos * nX + fSin * nY + nCenterX) );
+        rPt.setY( FRound(-(fSin * nX - fCos * nY - nCenterY)) );
+    }
+}
+
+void Polygon::Clip( const tools::Rectangle& rRect )
+{
+    // This algorithm is broken for bezier curves, they would get converted to lines.
+    // Use PolyPolygon::Clip().
+    assert( !HasFlags());
+
+    // #105251# Normalize rect before edge filtering
+    tools::Rectangle               aJustifiedRect( rRect );
+    aJustifiedRect.Normalize();
+
+    sal_uInt16              nSourceSize = mpImplPolygon->mnPoints;
+    ImplPolygonPointFilter  aPolygon( nSourceSize );
+    ImplEdgePointFilter     aHorzFilter( EDGE_HORZ, aJustifiedRect.Left(), aJustifiedRect.Right(),
+                                         aPolygon );
+    ImplEdgePointFilter     aVertFilter( EDGE_VERT, aJustifiedRect.Top(), aJustifiedRect.Bottom(),
+                                         aHorzFilter );
+
+    for ( sal_uInt16 i = 0; i < nSourceSize; i++ )
+        aVertFilter.Input( mpImplPolygon->mxPointAry[i] );
+    if ( aVertFilter.IsPolygon() )
+        aVertFilter.LastPoint();
+    else
+        aPolygon.LastPoint();
+
+    mpImplPolygon = ImplType(aPolygon.get());
+}
+
+tools::Rectangle Polygon::GetBoundRect() const
+{
+    // Removing the assert. Bezier curves have the attribute that each single
+    // curve segment defined by four points can not exit the four-point polygon
+    // defined by that points. This allows to say that the curve segment can also
+    // never leave the Range of its defining points.
+    // The result is that Polygon::GetBoundRect() may not create the minimal
+    // BoundRect of the Polygon (to get that, use basegfx::B2DPolygon classes),
+    // but will always create a valid BoundRect, at least as long as this method
+    // 'blindly' travels over all points, including control points.
+
+    // DBG_ASSERT( !mpImplPolygon->mxFlagAry.get(), "GetBoundRect could fail with beziers!" );
+
+    sal_uInt16  nCount = mpImplPolygon->mnPoints;
+    if( ! nCount )
+        return tools::Rectangle();
+
+    tools::Long    nXMin, nXMax, nYMin, nYMax;
+
+    const Point& pFirstPt = mpImplPolygon->mxPointAry[0];
+    nXMin = nXMax = pFirstPt.X();
+    nYMin = nYMax = pFirstPt.Y();
+
+    for ( sal_uInt16 i = 0; i < nCount; i++ )
+    {
+        const Point& rPt = mpImplPolygon->mxPointAry[i];
+
+        if (rPt.X() < nXMin)
+            nXMin = rPt.X();
+        if (rPt.X() > nXMax)
+            nXMax = rPt.X();
+        if (rPt.Y() < nYMin)
+            nYMin = rPt.Y();
+        if (rPt.Y() > nYMax)
+            nYMax = rPt.Y();
+    }
+
+    return tools::Rectangle( nXMin, nYMin, nXMax, nYMax );
+}
+
+bool Polygon::Contains( const Point& rPoint ) const
+{
+    DBG_ASSERT( !mpImplPolygon->mxFlagAry, "IsInside could fail with beziers!" );
+
+    const tools::Rectangle aBound( GetBoundRect() );
+    const Line      aLine( rPoint, Point( aBound.Right() + 100, rPoint.Y() ) );
+    sal_uInt16          nCount = mpImplPolygon->mnPoints;
+    sal_uInt16          nPCounter = 0;
+
+    if ( ( nCount > 2 ) && aBound.Contains( rPoint ) )
+    {
+        Point   aPt1( mpImplPolygon->mxPointAry[ 0 ] );
+        Point   aIntersection;
+        Point   aLastIntersection;
+
+        while ( ( aPt1 == mpImplPolygon->mxPointAry[ nCount - 1 ] ) && ( nCount > 3 ) )
+            nCount--;
+
+        for ( sal_uInt16 i = 1; i <= nCount; i++ )
+        {
+            const Point& rPt2 = mpImplPolygon->mxPointAry[ ( i < nCount ) ? i : 0 ];
+
+            if ( aLine.Intersection( Line( aPt1, rPt2 ), aIntersection ) )
+            {
+                // This avoids insertion of double intersections
+                if ( nPCounter )
+                {
+                    if ( aIntersection != aLastIntersection )
+                    {
+                        aLastIntersection = aIntersection;
+                        nPCounter++;
+                    }
+                }
+                else
+                {
+                    aLastIntersection = aIntersection;
+                    nPCounter++;
+                }
+            }
+
+            aPt1 = rPt2;
+        }
+    }
+
+    // is inside, if number of intersection points is odd
+    return ( ( nPCounter & 1 ) == 1 );
+}
+
+void Polygon::Insert( sal_uInt16 nPos, const Point& rPt )
+{
+    if( nPos >= mpImplPolygon->mnPoints )
+        nPos = mpImplPolygon->mnPoints;
+
+    if (mpImplPolygon->ImplSplit(nPos, 1))
+        mpImplPolygon->mxPointAry[ nPos ] = rPt;
+}
+
+void Polygon::Insert( sal_uInt16 nPos, const tools::Polygon& rPoly )
+{
+    const sal_uInt16 nInsertCount = rPoly.mpImplPolygon->mnPoints;
+
+    if( nInsertCount )
+    {
+        if( nPos >= mpImplPolygon->mnPoints )
+            nPos = mpImplPolygon->mnPoints;
+
+        if (rPoly.mpImplPolygon->mxFlagAry)
+            mpImplPolygon->ImplCreateFlagArray();
+
+        mpImplPolygon->ImplSplit( nPos, nInsertCount, rPoly.mpImplPolygon.get() );
+    }
+}
+
+Point& Polygon::operator[]( sal_uInt16 nPos )
+{
+    assert( nPos < mpImplPolygon->mnPoints );
+
+    return mpImplPolygon->mxPointAry[nPos];
+}
+
+tools::Polygon& Polygon::operator=( const tools::Polygon& rPoly )
+{
+    mpImplPolygon = rPoly.mpImplPolygon;
+    return *this;
+}
+
+tools::Polygon& Polygon::operator=( tools::Polygon&& rPoly ) noexcept
+{
+    mpImplPolygon = std::move(rPoly.mpImplPolygon);
+    return *this;
+}
+
+bool Polygon::operator==( const tools::Polygon& rPoly ) const
+{
+    return (mpImplPolygon == rPoly.mpImplPolygon);
+}
+
+bool Polygon::IsEqual( const tools::Polygon& rPoly ) const
+{
+    bool bIsEqual = true;
+    sal_uInt16 i;
+    if ( GetSize() != rPoly.GetSize() )
+        bIsEqual = false;
+    else
+    {
+        for ( i = 0; i < GetSize(); i++ )
+        {
+            if ( ( GetPoint( i ) != rPoly.GetPoint( i ) ) ||
+                ( GetFlags( i ) != rPoly.GetFlags( i ) ) )
+            {
+                bIsEqual = false;
+                break;
+            }
+        }
+    }
+    return bIsEqual;
+}
+
+SvStream& ReadPolygon( SvStream& rIStream, tools::Polygon& rPoly )
+{
+    sal_uInt16          nPoints(0);
+
+    // read all points and create array
+    rIStream.ReadUInt16( nPoints );
+
+    const size_t nMaxRecordsPossible = rIStream.remainingSize() / (2 * sizeof(sal_Int32));
+    if (nPoints > nMaxRecordsPossible)
+    {
+        SAL_WARN("tools", "Polygon claims " << nPoints << " records, but only " << nMaxRecordsPossible << " possible");
+        nPoints = nMaxRecordsPossible;
+    }
+
+    rPoly.mpImplPolygon->ImplSetSize( nPoints, false );
+
+    for (sal_uInt16 i = 0; i < nPoints; i++)
+    {
+        sal_Int32 nTmpX(0), nTmpY(0);
+        rIStream.ReadInt32(nTmpX).ReadInt32(nTmpY);
+        rPoly.mpImplPolygon->mxPointAry[i].setX(nTmpX);
+        rPoly.mpImplPolygon->mxPointAry[i].setY(nTmpY);
+    }
+
+    return rIStream;
+}
+
+SvStream& WritePolygon( SvStream& rOStream, const tools::Polygon& rPoly )
+{
+    sal_uInt16          nPoints = rPoly.GetSize();
+
+    // Write number of points
+    rOStream.WriteUInt16( nPoints );
+
+    for (sal_uInt16 i = 0; i < nPoints; i++)
+    {
+        rOStream.WriteInt32(rPoly.mpImplPolygon->mxPointAry[i].X())
+            .WriteInt32(rPoly.mpImplPolygon->mxPointAry[i].Y());
+    }
+
+    return rOStream;
+}
+
+void Polygon::ImplRead( SvStream& rIStream )
+{
+    sal_uInt8 bHasPolyFlags(0);
+
+    ReadPolygon( rIStream, *this );
+    rIStream.ReadUChar( bHasPolyFlags );
+
+    if ( bHasPolyFlags )
+    {
+        mpImplPolygon->mxFlagAry.reset(new PolyFlags[mpImplPolygon->mnPoints]);
+        auto nRead = rIStream.ReadBytes(mpImplPolygon->mxFlagAry.get(), mpImplPolygon->mnPoints);
+        if (nRead != mpImplPolygon->mnPoints)
+        {
+            SAL_WARN("tools", "Short read");
+            memset(mpImplPolygon->mxFlagAry.get() + nRead, 0, mpImplPolygon->mnPoints - nRead);
+        }
+    }
+}
+
+void Polygon::Read( SvStream& rIStream )
+{
+    VersionCompatRead aCompat(rIStream);
+
+    ImplRead( rIStream );
+}
+
+void Polygon::ImplWrite( SvStream& rOStream ) const
+{
+    bool bHasPolyFlags(mpImplPolygon->mxFlagAry);
+    WritePolygon( rOStream, *this );
+    rOStream.WriteBool(bHasPolyFlags);
+
+    if ( bHasPolyFlags )
+        rOStream.WriteBytes(mpImplPolygon->mxFlagAry.get(), mpImplPolygon->mnPoints);
+}
+
+void Polygon::Write( SvStream& rOStream ) const
+{
+    VersionCompatWrite aCompat(rOStream, 1);
+
+    ImplWrite( rOStream );
+}
+
+// #i74631#/#i115917# numerical correction method for B2DPolygon
+static void impCorrectContinuity(basegfx::B2DPolygon& roPolygon, sal_uInt32 nIndex, PolyFlags nCFlag)
+{
+    const sal_uInt32 nPointCount(roPolygon.count());
+    OSL_ENSURE(nIndex < nPointCount, "impCorrectContinuity: index access out of range (!)");
+
+    if(nIndex >= nPointCount || (PolyFlags::Smooth != nCFlag && PolyFlags::Symmetric != nCFlag))
+        return;
+
+    if(!roPolygon.isPrevControlPointUsed(nIndex) || !roPolygon.isNextControlPointUsed(nIndex))
+        return;
+
+    // #i115917# Patch from osnola (modified, thanks for showing the problem)
+
+    // The correction is needed because an integer polygon with control points
+    // is converted to double precision. When C1 or C2 is used the involved vectors
+    // may not have the same directions/lengths since these come from integer coordinates
+    //  and may have been snapped to different nearest integer coordinates. The snap error
+    // is in the range of +-1 in y and y, thus 0.0 <= error <= sqrt(2.0). Nonetheless,
+    // it needs to be corrected to be able to detect the continuity in this points
+    // correctly.
+
+    // We only have the integer data here (already in double precision form, but no mantissa
+    // used), so the best correction is to use:
+
+    // for C1: The longest vector since it potentially has best preserved the original vector.
+    //         Even better the sum of the vectors, weighted by their length. This gives the
+    //         normal vector addition to get the vector itself, lengths need to be preserved.
+    // for C2: The mediated vector(s) since both should be the same, but mirrored
+
+    // extract the point and vectors
+    const basegfx::B2DPoint aPoint(roPolygon.getB2DPoint(nIndex));
+    const basegfx::B2DVector aNext(roPolygon.getNextControlPoint(nIndex) - aPoint);
+    const basegfx::B2DVector aPrev(aPoint - roPolygon.getPrevControlPoint(nIndex));
+
+    // calculate common direction vector, normalize
+    const basegfx::B2DVector aDirection(aNext + aPrev);
+    const double fDirectionLen = aDirection.getLength();
+    if (fDirectionLen == 0.0)
+        return;
+
+    if (PolyFlags::Smooth == nCFlag)
+    {
+        // C1: apply common direction vector, preserve individual lengths
+        const double fInvDirectionLen(1.0 / fDirectionLen);
+        roPolygon.setNextControlPoint(nIndex, basegfx::B2DPoint(aPoint + (aDirection * (aNext.getLength() * fInvDirectionLen))));
+        roPolygon.setPrevControlPoint(nIndex, basegfx::B2DPoint(aPoint - (aDirection * (aPrev.getLength() * fInvDirectionLen))));
+    }
+    else // PolyFlags::Symmetric
+    {
+        // C2: get mediated length. Taking half of the unnormalized direction would be
+        // an approximation, but not correct.
+        const double fMedLength((aNext.getLength() + aPrev.getLength()) * (0.5 / fDirectionLen));
+        const basegfx::B2DVector aScaledDirection(aDirection * fMedLength);
+
+        // Bring Direction to correct length and apply
+        roPolygon.setNextControlPoint(nIndex, basegfx::B2DPoint(aPoint + aScaledDirection));
+        roPolygon.setPrevControlPoint(nIndex, basegfx::B2DPoint(aPoint - aScaledDirection));
+    }
+}
+
+// convert to basegfx::B2DPolygon and return
+basegfx::B2DPolygon Polygon::getB2DPolygon() const
+{
+    basegfx::B2DPolygon aRetval;
+    const sal_uInt16 nCount(mpImplPolygon->mnPoints);
+
+    if (nCount)
+    {
+        if (mpImplPolygon->mxFlagAry)
+        {
+            // handling for curves. Add start point
+            const Point aStartPoint(mpImplPolygon->mxPointAry[0]);
+            PolyFlags nPointFlag(mpImplPolygon->mxFlagAry[0]);
+            aRetval.append(basegfx::B2DPoint(aStartPoint.X(), aStartPoint.Y()));
+            Point aControlA, aControlB;
+
+            for(sal_uInt16 a(1); a < nCount;)
+            {
+                bool bControlA(false);
+                bool bControlB(false);
+
+                if(PolyFlags::Control == mpImplPolygon->mxFlagAry[a])
+                {
+                    aControlA = mpImplPolygon->mxPointAry[a++];
+                    bControlA = true;
+                }
+
+                if(a < nCount && PolyFlags::Control == mpImplPolygon->mxFlagAry[a])
+                {
+                    aControlB = mpImplPolygon->mxPointAry[a++];
+                    bControlB = true;
+                }
+
+                // assert invalid polygons
+                OSL_ENSURE(bControlA == bControlB, "Polygon::getB2DPolygon: Invalid source polygon (!)");
+
+                if(a < nCount)
+                {
+                    const Point aEndPoint(mpImplPolygon->mxPointAry[a]);
+
+                    if(bControlA)
+                    {
+                        // bezier edge, add
+                        aRetval.appendBezierSegment(
+                            basegfx::B2DPoint(aControlA.X(), aControlA.Y()),
+                            basegfx::B2DPoint(aControlB.X(), aControlB.Y()),
+                            basegfx::B2DPoint(aEndPoint.X(), aEndPoint.Y()));
+
+                        impCorrectContinuity(aRetval, aRetval.count() - 2, nPointFlag);
+                    }
+                    else
+                    {
+                        // no bezier edge, add end point
+                        aRetval.append(basegfx::B2DPoint(aEndPoint.X(), aEndPoint.Y()));
+                    }
+
+                    nPointFlag = mpImplPolygon->mxFlagAry[a++];
+                }
+            }
+
+            // if exist, remove double first/last points, set closed and correct control points
+            basegfx::utils::checkClosed(aRetval);
+
+            if(aRetval.isClosed())
+            {
+                // closeWithGeometryChange did really close, so last point(s) were removed.
+                // Correct the continuity in the changed point
+                impCorrectContinuity(aRetval, 0, mpImplPolygon->mxFlagAry[0]);
+            }
+        }
+        else
+        {
+            // extra handling for non-curves (most-used case) for speedup
+            for(sal_uInt16 a(0); a < nCount; a++)
+            {
+                // get point and add
+                const Point aPoint(mpImplPolygon->mxPointAry[a]);
+                aRetval.append(basegfx::B2DPoint(aPoint.X(), aPoint.Y()));
+            }
+
+            // set closed flag
+            basegfx::utils::checkClosed(aRetval);
+        }
+    }
+
+    return aRetval;
+}
+
+Polygon::Polygon(const basegfx::B2DPolygon& rPolygon) :  mpImplPolygon(ImplPolygon(rPolygon))
+{
+}
+
+} // namespace tools
+
+/* vim:set shiftwidth=4 softtabstop=4 expandtab: */
diff --git a/tools/source/generic/poly2.cxx b/tools/source/generic/poly2.cxx
new file mode 100644
index 0000000000..a00f64f41b
--- /dev/null
+++ b/tools/source/generic/poly2.cxx
@@ -0,0 +1,520 @@
+/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
+/*
+ * This file is part of the LibreOffice project.
+ *
+ * This Source Code Form is subject to the terms of the Mozilla Public
+ * License, v. 2.0. If a copy of the MPL was not distributed with this
+ * file, You can obtain one at http://mozilla.org/MPL/2.0/.
+ *
+ * This file incorporates work covered by the following license notice:
+ *
+ *   Licensed to the Apache Software Foundation (ASF) under one or more
+ *   contributor license agreements. See the NOTICE file distributed
+ *   with this work for additional information regarding copyright
+ *   ownership. The ASF licenses this file to you under the Apache
+ *   License, Version 2.0 (the "License"); you may not use this file
+ *   except in compliance with the License. You may obtain a copy of
+ *   the License at http://www.apache.org/licenses/LICENSE-2.0 .
+ */
+
+#include <sal/log.hxx>
+#include <osl/diagnose.h>
+#include <poly.h>
+#include <tools/poly.hxx>
+#include <tools/debug.hxx>
+#include <tools/stream.hxx>
+#include <tools/vcompat.hxx>
+#include <tools/gen.hxx>
+#include <basegfx/polygon/b2dpolypolygon.hxx>
+#include <basegfx/polygon/b2dpolypolygoncutter.hxx>
+#include <basegfx/polygon/b2dpolygonclipper.hxx>
+
+namespace tools {
+
+PolyPolygon::PolyPolygon( sal_uInt16 nInitSize )
+    : mpImplPolyPolygon( ImplPolyPolygon( nInitSize ) )
+{
+}
+
+PolyPolygon::PolyPolygon( const tools::Polygon& rPoly )
+    : mpImplPolyPolygon( rPoly )
+{
+}
+PolyPolygon::PolyPolygon( const tools::Rectangle& rRect )
+    : mpImplPolyPolygon( tools::Polygon(rRect) )
+{
+}
+
+PolyPolygon::PolyPolygon( const tools::PolyPolygon& rPolyPoly )
+    : mpImplPolyPolygon( rPolyPoly.mpImplPolyPolygon )
+{
+}
+
+PolyPolygon::PolyPolygon( tools::PolyPolygon&& rPolyPoly ) noexcept
+    : mpImplPolyPolygon( std::move(rPolyPoly.mpImplPolyPolygon) )
+{
+}
+
+PolyPolygon::~PolyPolygon()
+{
+}
+
+void PolyPolygon::Insert( const tools::Polygon& rPoly, sal_uInt16 nPos )
+{
+    assert ( mpImplPolyPolygon->mvPolyAry.size() < MAX_POLYGONS );
+
+    if ( nPos > mpImplPolyPolygon->mvPolyAry.size() )
+        nPos = mpImplPolyPolygon->mvPolyAry.size();
+
+    mpImplPolyPolygon->mvPolyAry.insert(mpImplPolyPolygon->mvPolyAry.begin() + nPos, rPoly);
+}
+
+void PolyPolygon::Remove( sal_uInt16 nPos )
+{
+    assert(nPos < Count() && "PolyPolygon::Remove(): nPos >= nSize");
+
+    mpImplPolyPolygon->mvPolyAry.erase(mpImplPolyPolygon->mvPolyAry.begin() + nPos);
+}
+
+void PolyPolygon::Replace( const tools::Polygon& rPoly, sal_uInt16 nPos )
+{
+    assert(nPos < Count() && "PolyPolygon::Replace(): nPos >= nSize");
+
+    mpImplPolyPolygon->mvPolyAry[nPos] = rPoly;
+}
+
+const tools::Polygon& PolyPolygon::GetObject( sal_uInt16 nPos ) const
+{
+    assert(nPos < Count() && "PolyPolygon::GetObject(): nPos >= nSize");
+
+    return mpImplPolyPolygon->mvPolyAry[nPos];
+}
+
+bool PolyPolygon::IsRect() const
+{
+    bool bIsRect = false;
+    if ( Count() == 1 )
+        bIsRect = mpImplPolyPolygon->mvPolyAry[ 0 ].IsRect();
+    return bIsRect;
+}
+
+void PolyPolygon::Clear()
+{
+    mpImplPolyPolygon->mvPolyAry.clear();
+}
+
+void PolyPolygon::Optimize( PolyOptimizeFlags nOptimizeFlags )
+{
+    if(!(bool(nOptimizeFlags) && Count()))
+        return;
+
+    // #115630# ImplDrawHatch does not work with beziers included in the polypolygon, take care of that
+    bool bIsCurve(false);
+
+    for(sal_uInt16 a(0); !bIsCurve && a < Count(); a++)
+    {
+        if((*this)[a].HasFlags())
+        {
+            bIsCurve = true;
+        }
+    }
+
+    if(bIsCurve)
+    {
+        OSL_ENSURE(false, "Optimize does *not* support curves, falling back to AdaptiveSubdivide()...");
+        tools::PolyPolygon aPolyPoly;
+
+        AdaptiveSubdivide(aPolyPoly);
+        aPolyPoly.Optimize(nOptimizeFlags);
+        *this = aPolyPoly;
+    }
+    else
+    {
+        double      fArea;
+        const bool  bEdges = ( nOptimizeFlags & PolyOptimizeFlags::EDGES ) == PolyOptimizeFlags::EDGES;
+        sal_uInt16      nPercent = 0;
+
+        if( bEdges )
+        {
+            const tools::Rectangle aBound( GetBoundRect() );
+
+            fArea = ( aBound.GetWidth() + aBound.GetHeight() ) * 0.5;
+            nPercent = 50;
+            nOptimizeFlags &= ~PolyOptimizeFlags::EDGES;
+        }
+
+        // Optimize polygons
+        for( sal_uInt16 i = 0, nPolyCount = mpImplPolyPolygon->mvPolyAry.size(); i < nPolyCount; i++ )
+        {
+            if( bEdges )
+            {
+                mpImplPolyPolygon->mvPolyAry[ i ].Optimize( PolyOptimizeFlags::NO_SAME );
+                tools::Polygon::ImplReduceEdges( mpImplPolyPolygon->mvPolyAry[ i ], fArea, nPercent );
+            }
+
+            if( bool(nOptimizeFlags) )
+                mpImplPolyPolygon->mvPolyAry[ i ].Optimize( nOptimizeFlags );
+        }
+    }
+}
+
+void PolyPolygon::AdaptiveSubdivide( tools::PolyPolygon& rResult ) const
+{
+    rResult.Clear();
+
+    tools::Polygon aPolygon;
+
+    for( size_t i = 0; i < mpImplPolyPolygon->mvPolyAry.size(); i++ )
+    {
+        mpImplPolyPolygon->mvPolyAry[ i ].AdaptiveSubdivide( aPolygon, 1.0 );
+        rResult.Insert( aPolygon );
+    }
+}
+
+tools::PolyPolygon PolyPolygon::SubdivideBezier( const tools::PolyPolygon& rPolyPoly )
+{
+    sal_uInt16 i, nPolys = rPolyPoly.Count();
+    tools::PolyPolygon aPolyPoly( nPolys );
+    for( i=0; i<nPolys; ++i )
+        aPolyPoly.Insert( tools::Polygon::SubdivideBezier( rPolyPoly.GetObject(i) ) );
+
+    return aPolyPoly;
+}
+
+
+void PolyPolygon::GetIntersection( const tools::PolyPolygon& rPolyPoly, tools::PolyPolygon& rResult ) const
+{
+    ImplDoOperation( rPolyPoly, rResult, PolyClipOp::INTERSECT );
+}
+
+void PolyPolygon::GetUnion( const tools::PolyPolygon& rPolyPoly, tools::PolyPolygon& rResult ) const
+{
+    ImplDoOperation( rPolyPoly, rResult, PolyClipOp::UNION );
+}
+
+void PolyPolygon::ImplDoOperation( const tools::PolyPolygon& rPolyPoly, tools::PolyPolygon& rResult, PolyClipOp nOperation ) const
+{
+    // Convert to B2DPolyPolygon, temporarily. It might be
+    // advantageous in the future, to have a tools::PolyPolygon adaptor that
+    // just simulates a B2DPolyPolygon here...
+    basegfx::B2DPolyPolygon aMergePolyPolygonA( getB2DPolyPolygon() );
+    basegfx::B2DPolyPolygon aMergePolyPolygonB( rPolyPoly.getB2DPolyPolygon() );
+
+    // normalize the two polypolygons before. Force properly oriented
+    // polygons.
+    aMergePolyPolygonA = basegfx::utils::prepareForPolygonOperation( aMergePolyPolygonA );
+    aMergePolyPolygonB = basegfx::utils::prepareForPolygonOperation( aMergePolyPolygonB );
+
+    switch( nOperation )
+    {
+        // All code extracted from svx/source/svdraw/svedtv2.cxx
+
+        case PolyClipOp::UNION:
+        {
+            // merge A and B (OR)
+            aMergePolyPolygonA = basegfx::utils::solvePolygonOperationOr(aMergePolyPolygonA, aMergePolyPolygonB);
+            break;
+        }
+
+        default:
+        case PolyClipOp::INTERSECT:
+        {
+            // cut poly 1 against polys 2..n (AND)
+            aMergePolyPolygonA = basegfx::utils::solvePolygonOperationAnd(aMergePolyPolygonA, aMergePolyPolygonB);
+            break;
+        }
+    }
+
+    rResult = tools::PolyPolygon( aMergePolyPolygonA );
+}
+
+sal_uInt16 PolyPolygon::Count() const
+{
+    return mpImplPolyPolygon->mvPolyAry.size();
+}
+
+void PolyPolygon::Move( tools::Long nHorzMove, tools::Long nVertMove )
+{
+    // Required for DrawEngine
+    if( nHorzMove || nVertMove )
+    {
+        // move points
+        sal_uInt16 nPolyCount = mpImplPolyPolygon->mvPolyAry.size();
+        for ( sal_uInt16 i = 0; i < nPolyCount; i++ )
+            mpImplPolyPolygon->mvPolyAry[i].Move( nHorzMove, nVertMove );
+    }
+}
+
+void PolyPolygon::Translate( const Point& rTrans )
+{
+    // move points
+    for ( sal_uInt16 i = 0, nCount = mpImplPolyPolygon->mvPolyAry.size(); i < nCount; i++ )
+        mpImplPolyPolygon->mvPolyAry[ i ].Translate( rTrans );
+}
+
+void PolyPolygon::Scale( double fScaleX, double fScaleY )
+{
+    // Move points
+    for ( sal_uInt16 i = 0, nCount = mpImplPolyPolygon->mvPolyAry.size(); i < nCount; i++ )
+        mpImplPolyPolygon->mvPolyAry[ i ].Scale( fScaleX, fScaleY );
+}
+
+void PolyPolygon::Rotate( const Point& rCenter, Degree10 nAngle10 )
+{
+    nAngle10 %= 3600_deg10;
+
+    if( nAngle10 )
+    {
+        const double fAngle = toRadians(nAngle10);
+        Rotate( rCenter, sin( fAngle ), cos( fAngle ) );
+    }
+}
+
+void PolyPolygon::Rotate( const Point& rCenter, double fSin, double fCos )
+{
+    // move points
+    for ( sal_uInt16 i = 0, nCount = mpImplPolyPolygon->mvPolyAry.size(); i < nCount; i++ )
+        mpImplPolyPolygon->mvPolyAry[ i ].Rotate( rCenter, fSin, fCos );
+}
+
+void PolyPolygon::Clip( const tools::Rectangle& rRect )
+{
+    sal_uInt16 nPolyCount = mpImplPolyPolygon->mvPolyAry.size();
+    sal_uInt16 i;
+
+    if ( !nPolyCount )
+        return;
+
+    // If there are bezier curves involved, Polygon::Clip() is broken.
+    // Use a temporary B2DPolyPolygon for the clipping.
+    for ( i = 0; i < nPolyCount; i++ )
+    {
+        if(mpImplPolyPolygon->mvPolyAry[i].HasFlags())
+        {
+            const basegfx::B2DPolyPolygon aPoly(
+                basegfx::utils::clipPolyPolygonOnRange(
+                    getB2DPolyPolygon(),
+                    basegfx::B2DRange(
+                        rRect.Left(),
+                        rRect.Top(),
+                        rRect.Right() + 1,
+                        rRect.Bottom() + 1),
+                    true,
+                    false));
+            *this = PolyPolygon( aPoly );
+            return;
+        }
+    }
+
+    // Clip every polygon, deleting the empty ones
+    for ( i = 0; i < nPolyCount; i++ )
+        mpImplPolyPolygon->mvPolyAry[i].Clip( rRect );
+    while ( nPolyCount )
+    {
+        if ( GetObject( nPolyCount-1 ).GetSize() <= 2 )
+            Remove( nPolyCount-1 );
+        nPolyCount--;
+    }
+}
+
+tools::Rectangle PolyPolygon::GetBoundRect() const
+{
+    tools::Long    nXMin=0, nXMax=0, nYMin=0, nYMax=0;
+    bool    bFirst = true;
+    sal_uInt16  nPolyCount = mpImplPolyPolygon->mvPolyAry.size();
+
+    for ( sal_uInt16 n = 0; n < nPolyCount; n++ )
+    {
+        const tools::Polygon*  pPoly = &mpImplPolyPolygon->mvPolyAry[n];
+        const Point*    pAry = pPoly->GetConstPointAry();
+        sal_uInt16          nPointCount = pPoly->GetSize();
+
+        for ( sal_uInt16 i = 0; i < nPointCount; i++ )
+        {
+            const Point* pPt = &pAry[ i ];
+
+            if ( bFirst )
+            {
+                nXMin = nXMax = pPt->X();
+                nYMin = nYMax = pPt->Y();
+                bFirst = false;
+            }
+            else
+            {
+                if ( pPt->X() < nXMin )
+                    nXMin = pPt->X();
+                if ( pPt->X() > nXMax )
+                    nXMax = pPt->X();
+                if ( pPt->Y() < nYMin )
+                    nYMin = pPt->Y();
+                if ( pPt->Y() > nYMax )
+                    nYMax = pPt->Y();
+            }
+        }
+    }
+
+    if ( !bFirst )
+        return tools::Rectangle( nXMin, nYMin, nXMax, nYMax );
+    else
+        return tools::Rectangle();
+}
+
+Polygon& PolyPolygon::operator[]( sal_uInt16 nPos )
+{
+    assert(nPos < Count() && "PolyPolygon::[](): nPos >= nSize");
+
+    return mpImplPolyPolygon->mvPolyAry[nPos];
+}
+
+PolyPolygon& PolyPolygon::operator=( const tools::PolyPolygon& rPolyPoly )
+{
+    mpImplPolyPolygon = rPolyPoly.mpImplPolyPolygon;
+    return *this;
+}
+
+PolyPolygon& PolyPolygon::operator=( tools::PolyPolygon&& rPolyPoly ) noexcept
+{
+    mpImplPolyPolygon = std::move(rPolyPoly.mpImplPolyPolygon);
+    return *this;
+}
+
+bool PolyPolygon::operator==( const tools::PolyPolygon& rPolyPoly ) const
+{
+    return rPolyPoly.mpImplPolyPolygon == mpImplPolyPolygon;
+}
+
+SvStream& ReadPolyPolygon( SvStream& rIStream, tools::PolyPolygon& rPolyPoly )
+{
+    sal_uInt16 nPolyCount(0);
+
+    // Read number of polygons
+    rIStream.ReadUInt16( nPolyCount );
+
+    const size_t nMinRecordSize = sizeof(sal_uInt16);
+    const size_t nMaxRecords = rIStream.remainingSize() / nMinRecordSize;
+    if (nPolyCount > nMaxRecords)
+    {
+        SAL_WARN("tools", "Parsing error: " << nMaxRecords <<
+                 " max possible entries, but " << nPolyCount << " claimed, truncating");
+        nPolyCount = nMaxRecords;
+    }
+
+    if( nPolyCount )
+    {
+        rPolyPoly.mpImplPolyPolygon->mvPolyAry.resize(nPolyCount);
+
+        tools::Polygon aTempPoly;
+        for ( sal_uInt16 i = 0; i < nPolyCount; i++ )
+        {
+            ReadPolygon( rIStream, aTempPoly );
+            rPolyPoly.mpImplPolyPolygon->mvPolyAry[i] = aTempPoly;
+        }
+    }
+    else
+        rPolyPoly = tools::PolyPolygon();
+
+    return rIStream;
+}
+
+SvStream& WritePolyPolygon( SvStream& rOStream, const tools::PolyPolygon& rPolyPoly )
+{
+    // Write number of polygons
+    sal_uInt16 nPolyCount = rPolyPoly.mpImplPolyPolygon->mvPolyAry.size();
+    rOStream.WriteUInt16( nPolyCount );
+
+    // output polygons
+    for ( sal_uInt16 i = 0; i < nPolyCount; i++ )
+        WritePolygon( rOStream, rPolyPoly.mpImplPolyPolygon->mvPolyAry[i] );
+
+    return rOStream;
+}
+
+void PolyPolygon::Read( SvStream& rIStream )
+{
+    VersionCompatRead aCompat(rIStream);
+
+    sal_uInt16 nPolyCount(0);
+
+    // Read number of polygons
+    rIStream.ReadUInt16( nPolyCount );
+
+    const size_t nMinRecordSize = sizeof(sal_uInt16);
+    const size_t nMaxRecords = rIStream.remainingSize() / nMinRecordSize;
+    if (nPolyCount > nMaxRecords)
+    {
+        SAL_WARN("tools", "Parsing error: " << nMaxRecords <<
+                 " max possible entries, but " << nPolyCount << " claimed, truncating");
+        nPolyCount = nMaxRecords;
+    }
+
+    if( nPolyCount )
+    {
+        mpImplPolyPolygon->mvPolyAry.clear();
+
+        for ( sal_uInt16 i = 0; i < nPolyCount; i++ )
+        {
+            tools::Polygon aTempPoly;
+            aTempPoly.ImplRead( rIStream );
+            mpImplPolyPolygon->mvPolyAry.emplace_back( aTempPoly );
+        }
+    }
+    else
+        *this = tools::PolyPolygon();
+}
+
+void PolyPolygon::Write( SvStream& rOStream ) const
+{
+    VersionCompatWrite aCompat(rOStream, 1);
+
+    // Write number of polygons
+    sal_uInt16 nPolyCount = mpImplPolyPolygon->mvPolyAry.size();
+    rOStream.WriteUInt16( nPolyCount );
+
+    // Output polygons
+    for ( sal_uInt16 i = 0; i < nPolyCount; i++ )
+        mpImplPolyPolygon->mvPolyAry[i].ImplWrite( rOStream );
+}
+
+// convert to basegfx::B2DPolyPolygon and return
+basegfx::B2DPolyPolygon PolyPolygon::getB2DPolyPolygon() const
+{
+    basegfx::B2DPolyPolygon aRetval;
+
+    for(size_t a(0); a < mpImplPolyPolygon->mvPolyAry.size(); a++)
+    {
+        tools::Polygon const & rCandidate = mpImplPolyPolygon->mvPolyAry[a];
+        aRetval.append(rCandidate.getB2DPolygon());
+    }
+
+    return aRetval;
+}
+
+// constructor to convert from basegfx::B2DPolyPolygon
+PolyPolygon::PolyPolygon(const basegfx::B2DPolyPolygon& rPolyPolygon)
+    : mpImplPolyPolygon(rPolyPolygon)
+{
+}
+
+} /* namespace tools */
+
+ImplPolyPolygon::ImplPolyPolygon(const basegfx::B2DPolyPolygon& rPolyPolygon)
+{
+    const sal_uInt16 nCount(sal_uInt16(rPolyPolygon.count()));
+    DBG_ASSERT(sal_uInt32(nCount) == rPolyPolygon.count(),
+        "PolyPolygon::PolyPolygon: Too many sub-polygons in given basegfx::B2DPolyPolygon (!)");
+
+    if ( nCount )
+    {
+        mvPolyAry.resize( nCount );
+
+        for(sal_uInt16 a(0); a < nCount; a++)
+        {
+            const basegfx::B2DPolygon& aCandidate(rPolyPolygon.getB2DPolygon(sal_uInt32(a)));
+            mvPolyAry[a] = tools::Polygon( aCandidate );
+        }
+    }
+    else
+       mvPolyAry.reserve(16);
+}
+
+/* vim:set shiftwidth=4 softtabstop=4 expandtab: */
diff --git a/tools/source/generic/svborder.cxx b/tools/source/generic/svborder.cxx
new file mode 100644
index 0000000000..a61c79b0e4
--- /dev/null
+++ b/tools/source/generic/svborder.cxx
@@ -0,0 +1,35 @@
+/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
+/*
+ * This file is part of the LibreOffice project.
+ *
+ * This Source Code Form is subject to the terms of the Mozilla Public
+ * License, v. 2.0. If a copy of the MPL was not distributed with this
+ * file, You can obtain one at http://mozilla.org/MPL/2.0/.
+ *
+ * This file incorporates work covered by the following license notice:
+ *
+ *   Licensed to the Apache Software Foundation (ASF) under one or more
+ *   contributor license agreements. See the NOTICE file distributed
+ *   with this work for additional information regarding copyright
+ *   ownership. The ASF licenses this file to you under the Apache
+ *   License, Version 2.0 (the "License"); you may not use this file
+ *   except in compliance with the License. You may obtain a copy of
+ *   the License at http://www.apache.org/licenses/LICENSE-2.0 .
+ */
+#include <tools/svborder.hxx>
+#include <tools/gen.hxx>
+
+tools::Rectangle& operator+=(tools::Rectangle& rRect, const SvBorder& rBorder)
+{
+    // call GetSize first due to Empty-Rect
+    Size aS(rRect.GetSize());
+    aS.AdjustWidth(rBorder.Left() + rBorder.Right());
+    aS.AdjustHeight(rBorder.Top() + rBorder.Bottom());
+
+    rRect.AdjustLeft(-(rBorder.Left()));
+    rRect.AdjustTop(-(rBorder.Top()));
+    rRect.SetSize(aS);
+    return rRect;
+}
+
+/* vim:set shiftwidth=4 softtabstop=4 expandtab: */