Files
loongoffice/include/basegfx/numeric/ftools.hxx
Armin Le Grand (allotropia) 604f6fc8fe tdf#143377 Correct maximum Skew for TextOrientation in Calc
For zero or 180 degree text orinentation errors can happen
in the Border visualization, theoretically also in the
Text rendering. Ths has to do with sin(0) and sin(180)
being zero and lead internal to numerical problems, e.g.
a very huge Skew that when applied show the reported
'errors'.

I limit this mechanism now to +/- 1/2 degree from the
critical mentioned places, for Border and Text - to
not risk to have different points of corrections. The
UI only allows angles of 1 degree steps, but UNO API
and pdf import may allow more.

Change-Id: Idbc68f6a7beab84df0672165c2a813d96eeff84e
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/141999
Tested-by: Jenkins
Reviewed-by: Armin Le Grand <Armin.Le.Grand@me.com>
2022-10-31 10:01:46 +01:00

214 lines
7.1 KiB
C++

/* -*- 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 .
*/
#pragma once
#include <rtl/math.h>
#include <cmath>
#include <math.h>
#include <basegfx/basegfxdllapi.h>
#include <limits>
#include <algorithm>
// fTools defines
namespace basegfx
{
/** Round double to nearest integer
@return the nearest integer
*/
inline sal_Int32 fround( double fVal )
{
if (fVal >= 0.0)
{
if (fVal >= std::numeric_limits<sal_Int32>::max() - .5)
return std::numeric_limits<sal_Int32>::max();
return static_cast<sal_Int32>(fVal + .5);
}
if (fVal <= std::numeric_limits<sal_Int32>::min() + .5)
return std::numeric_limits<sal_Int32>::min();
return static_cast<sal_Int32>(fVal - .5);
}
/** Round double to nearest integer
@return the nearest 64 bit integer
*/
inline sal_Int64 fround64( double fVal )
{
return fVal > 0.0 ? static_cast<sal_Int64>( fVal + .5 ) : -static_cast<sal_Int64>( -fVal + .5 );
}
/** Prune a small epsilon range around zero.
Use this method e.g. for calculating scale values. There, it
is usually advisable not to set a scaling to 0.0, because that
yields singular transformation matrices.
@param fVal
An arbitrary, but finite and valid number
@return either fVal, or a small value slightly above (when
fVal>0) or below (when fVal<0) zero.
*/
inline double pruneScaleValue( double fVal )
{
if(fVal < 0.0)
return std::min(fVal, -0.00001);
else
return std::max(fVal, 0.00001);
}
/** Convert value from degrees to radians
*/
template <int DegMultiple = 1> constexpr double deg2rad( double v )
{
// divide first, to get exact values for v being a multiple of
// 90 degrees
return v / (90.0 * DegMultiple) * M_PI_2;
}
/** Convert value radians to degrees
*/
template <int DegMultiple = 1> constexpr double rad2deg( double v )
{
// divide first, to get exact values for v being a multiple of
// pi/2
return v / M_PI_2 * (90.0 * DegMultiple);
}
/** Snap v to nearest multiple of fStep, from negative and
positive side.
Examples:
snapToNearestMultiple(-0.1, 0.5) = 0.0
snapToNearestMultiple(0.1, 0.5) = 0.0
snapToNearestMultiple(0.25, 0.5) = 0.0
snapToNearestMultiple(0.26, 0.5) = 0.5
*/
BASEGFX_DLLPUBLIC double snapToNearestMultiple(double v, const double fStep);
/** Snap v to the range [0.0 .. fWidth] using modulo
*/
BASEGFX_DLLPUBLIC double snapToZeroRange(double v, double fWidth);
/** Snap v to the range [fLow .. fHigh] using modulo
*/
double snapToRange(double v, double fLow, double fHigh);
/** return fValue with the sign of fSignCarrier, thus evtl. changed
*/
inline double copySign(double fValue, double fSignCarrier)
{
#ifdef _WIN32
return _copysign(fValue, fSignCarrier);
#else
return copysign(fValue, fSignCarrier);
#endif
}
/** RotateFlyFrame3: Normalize to range defined by [0.0 ... fRange[, independent
if v is positive or negative.
Examples:
normalizeToRange(0.5, -1.0) = 0.0
normalizeToRange(0.5, 0.0) = 0.0
normalizeToRange(0.5, 1.0) = 0.5
normalizeToRange(-0.5, 1.0) = 0.5
normalizeToRange(-0.3, 1.0) = 0.7
normalizeToRange(-0.7, 1.0) = 0.3
normalizeToRange(3.5, 1.0) = 0.5
normalizeToRange(3.3, 1.0) = 0.3
normalizeToRange(3.7, 1.0) = 0.7
normalizeToRange(-3.5, 1.0) = 0.5
normalizeToRange(-3.3, 1.0) = 0.7
normalizeToRange(-3.7, 1.0) = 0.3
*/
BASEGFX_DLLPUBLIC double normalizeToRange(double v, const double fRange);
namespace fTools
{
/// Get threshold value for equalZero and friends
inline double getSmallValue() { return 0.000000001f; }
/// Compare against small value
template <typename T, std::enable_if_t<std::is_floating_point_v<T>, int> = 0>
inline bool equalZero(const T& rfVal)
{
return (fabs(rfVal) <= getSmallValue());
}
/// Compare against given small value
template <typename T, std::enable_if_t<std::is_floating_point_v<T>, int> = 0>
inline bool equalZero(const T& rfVal, const T& rfSmallValue)
{
return (fabs(rfVal) <= rfSmallValue);
}
template <typename T, std::enable_if_t<std::is_floating_point_v<T>, int> = 0>
inline bool equal(T const& rfValA, T const& rfValB)
{
// changed to approxEqual usage for better numerical correctness
return rtl_math_approxEqual(rfValA, rfValB);
}
template <typename T, std::enable_if_t<std::is_floating_point_v<T>, int> = 0>
inline bool equal(const T& rfValA, const T& rfValB, const T& rfSmallValue)
{
return (fabs(rfValA - rfValB) <= rfSmallValue);
}
template <typename T, std::enable_if_t<std::is_floating_point_v<T>, int> = 0>
inline bool less(const T& rfValA, const T& rfValB)
{
return (rfValA < rfValB && !equal(rfValA, rfValB));
}
template <typename T, std::enable_if_t<std::is_floating_point_v<T>, int> = 0>
inline bool lessOrEqual(const T& rfValA, const T& rfValB)
{
return (rfValA < rfValB || equal(rfValA, rfValB));
}
template <typename T, std::enable_if_t<std::is_floating_point_v<T>, int> = 0>
inline bool more(const T& rfValA, const T& rfValB)
{
return (rfValA > rfValB && !equal(rfValA, rfValB));
}
template <typename T, std::enable_if_t<std::is_floating_point_v<T>, int> = 0>
inline bool moreOrEqual(const T& rfValA, const T& rfValB)
{
return (rfValA > rfValB || equal(rfValA, rfValB));
}
template <typename T, std::enable_if_t<std::is_floating_point_v<T>, int> = 0>
inline bool betweenOrEqualEither(const T& rfValA, const T& rfValB, const T& rfValC)
{
return (rfValA > rfValB && rfValA < rfValC) || equal(rfValA, rfValB) || equal(rfValA, rfValC);
}
};
} // end of namespace basegfx
/* vim:set shiftwidth=4 softtabstop=4 expandtab: */