#include "traster.h"
#include "trastercm.h"
#include "tpalette.h"
#include "tcolorstyles.h"
//#include "trop.h"
#include "tropcm.h"
#include "tpixelutils.h"
#define TMIN(a, b) (a < b ? a : b)
#define TMAX(a, b) (a > b ? a : b)
//------------------------------------------------------------------------------
namespace
{
//------------------------------------------------------------------------------
template <class PIXOUT, class PIXIN>
void doConvolve_row_9_i(PIXOUT *pixout, int n, PIXIN *pixarr[], long w[])
{
long w1, w2, w3, w4, w5, w6, w7, w8, w9;
PIXIN *p1, *p2, *p3, *p4, *p5, *p6, *p7, *p8, *p9;
w1 = w[0];
w2 = w[1];
w3 = w[2];
w4 = w[3];
w5 = w[4];
w6 = w[5];
w7 = w[6];
w8 = w[7];
w9 = w[8];
p1 = pixarr[0];
p2 = pixarr[1];
p3 = pixarr[2];
p4 = pixarr[3];
p5 = pixarr[4];
p6 = pixarr[5];
p7 = pixarr[6];
p8 = pixarr[7];
p9 = pixarr[8];
int rightShift = 16 + ((int)sizeof(typename PIXIN::Channel) - (int)sizeof(typename PIXOUT::Channel)) * 8;
while (n-- > 0) {
pixout->r = (typename PIXOUT::Channel)((p1->r * w1 + p2->r * w2 + p3->r * w3 +
p4->r * w4 + p5->r * w5 + p6->r * w6 +
p7->r * w7 + p8->r * w8 + p9->r * w9 + (1 << 15)) >>
rightShift);
pixout->g = (typename PIXOUT::Channel)((p1->g * w1 + p2->g * w2 + p3->g * w3 +
p4->g * w4 + p5->g * w5 + p6->g * w6 +
p7->g * w7 + p8->g * w8 + p9->g * w9 + (1 << 15)) >>
rightShift);
pixout->b = (typename PIXOUT::Channel)((p1->b * w1 + p2->b * w2 + p3->b * w3 +
p4->b * w4 + p5->b * w5 + p6->b * w6 +
p7->b * w7 + p8->b * w8 + p9->b * w9 + (1 << 15)) >>
rightShift);
pixout->m = (typename PIXOUT::Channel)((p1->m * w1 + p2->m * w2 + p3->m * w3 +
p4->m * w4 + p5->m * w5 + p6->m * w6 +
p7->m * w7 + p8->m * w8 + p9->m * w9 + (1 << 15)) >>
rightShift);
p1++;
p2++;
p3++;
p4++;
p5++;
p6++;
p7++;
p8++;
p9++;
pixout++;
}
}
//------------------------------------------------------------------------------
template <class PIXOUT>
void doConvolve_cm32_row_9_i(PIXOUT *pixout, int n,
TPixelCM32 *pixarr[], long w[],
const std::vector<TPixel32> &paints,
const std::vector<TPixel32> &inks)
{
long w1, w2, w3, w4, w5, w6, w7, w8, w9;
TPixelCM32 *p1, *p2, *p3, *p4, *p5, *p6, *p7, *p8, *p9;
TPixel32 val[9];
w1 = w[0];
w2 = w[1];
w3 = w[2];
w4 = w[3];
w5 = w[4];
w6 = w[5];
w7 = w[6];
w8 = w[7];
w9 = w[8];
p1 = pixarr[0];
p2 = pixarr[1];
p3 = pixarr[2];
p4 = pixarr[3];
p5 = pixarr[4];
p6 = pixarr[5];
p7 = pixarr[6];
p8 = pixarr[7];
p9 = pixarr[8];
while (n-- > 0) {
for (int i = 0; i < 9; ++i) {
int tone = p1->getTone();
int paint = p1->getPaint();
int ink = p1->getInk();
if (tone == TPixelCM32::getMaxTone())
val[i] = paints[paint];
else if (tone == 0)
val[i] = inks[ink];
else
val[i] = blend(inks[ink], paints[paint], tone, TPixelCM32::getMaxTone());
}
pixout->r = (typename PIXOUT::Channel)((val[1].r * w1 + val[2].r * w2 + val[3].r * w3 +
val[4].r * w4 + val[5].r * w5 + val[6].r * w6 +
val[7].r * w7 + val[8].r * w8 + val[9].r * w9 + (1 << 15)) >>
16);
pixout->g = (typename PIXOUT::Channel)((val[1].g * w1 + val[2].g * w2 + val[3].g * w3 +
val[4].g * w4 + val[5].g * w5 + val[6].g * w6 +
val[7].g * w7 + val[8].g * w8 + val[9].g * w9 + (1 << 15)) >>
16);
pixout->b = (typename PIXOUT::Channel)((val[1].b * w1 + val[2].b * w2 + val[3].b * w3 +
val[4].b * w4 + val[5].b * w5 + val[6].b * w6 +
val[7].b * w7 + val[8].b * w8 + val[9].b * w9 + (1 << 15)) >>
16);
pixout->m = (typename PIXOUT::Channel)((val[1].m * w1 + val[2].m * w2 + val[3].m * w3 +
val[4].m * w4 + val[5].m * w5 + val[6].m * w6 +
val[7].m * w7 + val[8].m * w8 + val[9].m * w9 + (1 << 15)) >>
16);
p1++;
p2++;
p3++;
p4++;
p5++;
p6++;
p7++;
p8++;
p9++;
pixout++;
}
}
//------------------------------------------------------------------------------
template <class PIXOUT, class PIXIN>
void doConvolve_row_i(PIXOUT *pixout, int n,
PIXIN *pixarr[], long w[], int pixn)
{
long ar, ag, ab, am;
int i;
int rightShift = 16 + ((int)sizeof(typename PIXIN::Channel) - (int)sizeof(typename PIXOUT::Channel)) * 8;
while (n-- > 0) {
ar = ag = ab = am = 0;
for (i = 0; i < pixn; i++) {
ar += pixarr[i]->r * w[i];
ag += pixarr[i]->g * w[i];
ab += pixarr[i]->b * w[i];
am += pixarr[i]->m * w[i];
pixarr[i]++;
}
pixout->r = (typename PIXOUT::Channel)((ar + (1 << 15)) >> rightShift);
pixout->g = (typename PIXOUT::Channel)((ag + (1 << 15)) >> rightShift);
pixout->b = (typename PIXOUT::Channel)((ab + (1 << 15)) >> rightShift);
pixout->m = (typename PIXOUT::Channel)((am + (1 << 15)) >> rightShift);
pixout++;
}
}
//------------------------------------------------------------------------------
template <class PIXOUT>
void doConvolve_cm32_row_i(PIXOUT *pixout, int n,
TPixelCM32 *pixarr[], long w[], int pixn,
const std::vector<TPixel32> &paints,
const std::vector<TPixel32> &inks)
{
long ar, ag, ab, am;
int i;
while (n-- > 0) {
ar = ag = ab = am = 0;
for (i = 0; i < pixn; i++) {
TPixel32 val;
int tone = pixarr[i]->getTone();
int paint = pixarr[i]->getPaint();
int ink = pixarr[i]->getInk();
if (tone == TPixelCM32::getMaxTone())
val = paints[paint];
else if (tone == 0)
val = inks[ink];
else
val = blend(inks[ink], paints[paint], tone, TPixelCM32::getMaxTone());
ar += val.r * w[i];
ag += val.g * w[i];
ab += val.b * w[i];
am += val.m * w[i];
pixarr[i]++;
}
pixout->r = (typename PIXOUT::Channel)((ar + (1 << 15)) >> 16);
pixout->g = (typename PIXOUT::Channel)((ag + (1 << 15)) >> 16);
pixout->b = (typename PIXOUT::Channel)((ab + (1 << 15)) >> 16);
pixout->m = (typename PIXOUT::Channel)((am + (1 << 15)) >> 16);
pixout++;
}
}
//------------------------------------------------------------------------------
template <class PIXOUT, class PIXIN>
void doConvolve_3_i(TRasterPT<PIXOUT> rout,
TRasterPT<PIXIN> rin,
int dx, int dy,
double conv[])
{
PIXIN *bufferin;
PIXOUT *bufferout;
PIXIN *pixin;
PIXOUT *pixout;
PIXIN *pixarr[9];
long w[9];
int pixn;
int wrapin, wrapout;
int x, y, n;
int x1, y1, x2, y2;
int fx1, fy1, fx2, fy2, fx, fy;
rout->clear();
wrapin = rin->getWrap();
wrapout = rout->getWrap();
/* calcolo l'area di output interessata */
x1 = TMAX(0, -dx - 1);
y1 = TMAX(0, -dy - 1);
x2 = TMIN(rout->getLx() - 1, -dx + rin->getLx());
y2 = TMIN(rout->getLy() - 1, -dy + rin->getLy());
rin->lock();
rout->lock();
bufferin = rin->pixels();
bufferout = rout->pixels();
for (y = y1; y <= y2; y++) {
fy1 = TMAX(-1, -dy - y);
fy2 = TMIN(1, -dy + rin->getLy() - 1 - y);
if (fy1 > fy2)
continue;
x = x1;
pixout = bufferout + wrapout * y + x;
pixin = bufferin + wrapin * (y + dy) + (x + dx);
while (x <= x2) {
fx1 = TMAX(-1, -dx - x);
fx2 = TMIN(1, -dx + rin->getLx() - 1 - x);
if (x > -dx && x < -dx + rin->getLx() - 1)
n = tmin(-dx + rin->getLx() - 1 - x, x2 - x + 1);
else
n = 1;
if (n < 1)
break;
pixn = 0;
for (fy = fy1; fy <= fy2; fy++)
for (fx = fx1; fx <= fx2; fx++) {
pixarr[pixn] = pixin + fy * wrapin + fx;
w[pixn] = (long)(conv[(fy + 1) * 3 + fx + 1] * (1 << 16));
pixn++;
}
if (pixn == 9)
doConvolve_row_9_i<PIXOUT, PIXIN>(pixout, n, pixarr, w);
else
doConvolve_row_i<PIXOUT, PIXIN>(pixout, n, pixarr, w, pixn);
x += n;
pixin += n;
pixout += n;
}
}
rin->unlock();
rout->unlock();
}
//------------------------------------------------------------------------------
template <class PIXOUT, class PIXIN>
void doConvolve_i(TRasterPT<PIXOUT> rout,
TRasterPT<PIXIN> rin,
int dx, int dy,
double conv[], int radius)
{
PIXIN *bufferin;
PIXOUT *bufferout;
PIXIN *pixin;
PIXOUT *pixout;
int radiusSquare = sq(radius);
std::unique_ptr<PIXIN*[]> pixarr(new PIXIN *[radiusSquare]);
std::unique_ptr<long[]> w(new long[radiusSquare]);
int pixn;
int wrapin, wrapout;
int x, y, n;
int x1, y1, x2, y2;
int fx1, fy1, fx2, fy2, fx, fy;
int radius1 = radius / 2;
int radius0 = radius1 - radius + 1;
rout->clear();
wrapin = rin->getWrap();
wrapout = rout->getWrap();
/* calcolo l'area di output interessata */
x1 = TMAX(0, -dx - 1);
y1 = TMAX(0, -dy - 1);
x2 = TMIN(rout->getLx() - 1, -dx + rin->getLx());
y2 = TMIN(rout->getLy() - 1, -dy + rin->getLy());
rin->lock();
rout->lock();
bufferin = rin->pixels();
bufferout = rout->pixels();
for (y = y1; y <= y2; y++) {
fy1 = TMAX(radius0, -dy - y);
fy2 = TMIN(radius1, -dy - y + rin->getLy() - 1);
if (fy1 > fy2)
continue;
x = x1;
pixout = bufferout + wrapout * y + x;
pixin = bufferin + wrapin * (y + dy) + (x + dx);
while (x <= x2) {
fx1 = TMAX(radius0, -dx - x);
fx2 = TMIN(radius1, -dx - x + rin->getLx() - 1);
if (x > -dx && x < -dx + rin->getLx() - 1)
n = tmin(-dx + rin->getLx() - 1 - x, x2 - x + 1);
else
n = 1;
if (n < 1)
break;
pixn = 0;
for (fy = fy1; fy <= fy2; fy++)
for (fx = fx1; fx <= fx2; fx++) {
pixarr[pixn] = pixin + fy * wrapin + fx;
w[pixn] = (long)(conv[(fy - radius0) * radius + fx - radius0] * (1 << 16));
pixn++;
}
doConvolve_row_i<PIXOUT, PIXIN>(pixout, n, pixarr.get(), w.get(), pixn);
x += n;
pixin += n;
pixout += n;
}
}
rin->unlock();
rout->unlock();
}
//------------------------------------------------------------------------------
template <class PIXOUT>
void doConvolve_cm32_3_i(TRasterPT<PIXOUT> rout,
TRasterCM32P rin,
const TPaletteP &palette,
int dx, int dy,
double conv[])
{
TPixelCM32 *pixin;
PIXOUT *pixout;
TPixelCM32 *pixarr[9];
long w[9];
int pixn;
int wrapin, wrapout;
int x, y, n;
int x1, y1, x2, y2;
int fx1, fy1, fx2, fy2, fx, fy;
rout->clear();
wrapin = rin->getWrap();
wrapout = rout->getWrap();
/* calcolo l'area di output interessata */
x1 = TMAX(0, -dx - 1);
y1 = TMAX(0, -dy - 1);
x2 = TMIN(rout->getLx() - 1, -dx + rin->getLx());
y2 = TMIN(rout->getLy() - 1, -dy + rin->getLy());
int colorCount = palette->getStyleCount();
colorCount = tmax(colorCount, TPixelCM32::getMaxInk(), TPixelCM32::getMaxPaint());
std::vector<TPixel32> paints(colorCount);
std::vector<TPixel32> inks(colorCount);
rin->lock();
rout->lock();
TPixelCM32 *bufferin = rin->pixels();
PIXOUT *bufferout = rout->pixels();
for (int i = 0; i < palette->getStyleCount(); i++)
paints[i] = inks[i] = palette->getStyle(i)->getAverageColor();
for (y = y1; y <= y2; y++) {
fy1 = TMAX(-1, -dy - y);
fy2 = TMIN(1, -dy + rin->getLy() - 1 - y);
if (fy1 > fy2)
continue;
x = x1;
pixout = bufferout + wrapout * y + x;
pixin = bufferin + wrapin * (y + dy) + (x + dx);
while (x <= x2) {
fx1 = TMAX(-1, -dx - x);
fx2 = TMIN(1, -dx + rin->getLx() - 1 - x);
if (x > -dx && x < -dx + rin->getLx() - 1)
n = TMIN(-dx + rin->getLx() - 1 - x, x2 - x + 1);
else
n = 1;
if (n < 1)
break;
pixn = 0;
for (fy = fy1; fy <= fy2; fy++)
for (fx = fx1; fx <= fx2; fx++) {
pixarr[pixn] = pixin + fy * wrapin + fx;
w[pixn] = (long)(conv[(fy + 1) * 3 + fx + 1] * (1 << 16));
pixn++;
}
if (pixn == 9)
doConvolve_cm32_row_9_i<PIXOUT>(pixout, n, pixarr, w, paints, inks);
else
doConvolve_cm32_row_i<PIXOUT>(pixout, n, pixarr, w, pixn, paints, inks);
x += n;
pixin += n;
pixout += n;
}
}
rin->unlock();
rout->unlock();
}
//------------------------------------------------------------------------------
template <class PIXOUT>
void doConvolve_cm32_i(TRasterPT<PIXOUT> rout,
TRasterCM32P rin,
const TPaletteP &palette,
int dx, int dy,
double conv[], int radius)
{
TPixelCM32 *pixin;
PIXOUT *pixout;
int radiusSquare = sq(radius);
std::unique_ptr<TPixelCM32*[]> pixarr(new TPixelCM32*[radiusSquare]);
std::unique_ptr<long[]> w(new long[radiusSquare]);
int pixn;
int wrapin, wrapout;
int x, y, n;
int x1, y1, x2, y2;
int fx1, fy1, fx2, fy2, fx, fy;
int radius1 = radius / 2;
int radius0 = radius1 - radius + 1;
rout->clear();
wrapin = rin->getWrap();
wrapout = rout->getWrap();
/* calcolo l'area di output interessata */
x1 = TMAX(0, -dx - 1);
y1 = TMAX(0, -dy - 1);
x2 = TMIN(rout->getLx() - 1, -dx + rin->getLx());
y2 = TMIN(rout->getLy() - 1, -dy + rin->getLy());
int colorCount = palette->getStyleCount();
colorCount = tmax(colorCount, TPixelCM32::getMaxInk(), TPixelCM32::getMaxPaint());
std::vector<TPixel32> paints(colorCount);
std::vector<TPixel32> inks(colorCount);
rin->lock();
rout->lock();
TPixelCM32 *bufferin = rin->pixels();
PIXOUT *bufferout = rout->pixels();
for (int i = 0; i < palette->getStyleCount(); i++)
paints[i] = inks[i] = palette->getStyle(i)->getAverageColor();
for (y = y1; y <= y2; y++) {
fy1 = TMAX(radius0, -dy - y);
fy2 = TMIN(radius1, -dy + rin->getLy() - 1 - y);
if (fy1 > fy2)
continue;
x = x1;
pixout = bufferout + wrapout * y + x;
pixin = bufferin + wrapin * (y + dy) + (x + dx);
while (x <= x2) {
fx1 = TMAX(radius0, -dx - x);
fx2 = TMIN(radius1, -dx + rin->getLx() - 1 - x);
if (x > -dx && x < -dx + rin->getLx() - 1)
n = TMIN(-dx + rin->getLx() - 1 - x, x2 - x + 1);
else
n = 1;
if (n < 1)
break;
pixn = 0;
for (fy = fy1; fy <= fy2; fy++)
for (fx = fx1; fx <= fx2; fx++) {
pixarr[pixn] = pixin + fy * wrapin + fx;
w[pixn] = (long)(conv[(fy - radius0) * radius + fx - radius0] * (1 << 16));
pixn++;
}
doConvolve_cm32_row_i<PIXOUT>(pixout, n, pixarr.get(), w.get(), pixn, paints, inks);
x += n;
pixin += n;
pixout += n;
}
}
rin->unlock();
rout->unlock();
}
} // anonymous namespace
//------------------------------------------------------------------------------
void TRop::convolve_3_i(TRasterP rout, TRasterP rin, int dx, int dy, double conv[])
{
TRaster32P rin32 = rin;
if (rin32) {
TRaster32P rout32 = rout;
if (rout32) {
doConvolve_3_i<TPixel32, TPixel32>(rout32, rin32, dx, dy, conv);
return;
}
TRaster64P rout64 = rout;
if (rout64) {
doConvolve_3_i<TPixel64, TPixel32>(rout64, rin32, dx, dy, conv);
return;
}
} else {
TRaster64P rin64 = rin;
if (rin64) {
TRaster32P rout32 = rout;
if (rout32) {
doConvolve_3_i<TPixel32, TPixel64>(rout32, rin64, dx, dy, conv);
return;
}
TRaster64P rout64 = rout;
if (rout64) {
doConvolve_3_i<TPixel64, TPixel64>(rout64, rin64, dx, dy, conv);
return;
}
}
}
throw TRopException("TRop::convolve_3_i: unsupported pixel type");
}
//------------------------------------------------------------------------------
void TRop::convolve_3_i(TRasterP rout, TRasterCM32P rin, const TPaletteP &palette,
int dx, int dy, double conv[])
{
TRaster32P rout32 = rout;
if (rout32) {
doConvolve_cm32_3_i<TPixel32>(rout32, rin, palette, dx, dy, conv);
return;
}
TRaster64P rout64 = rout;
if (rout64) {
doConvolve_cm32_3_i<TPixel64>(rout64, rin, palette, dx, dy, conv);
return;
}
throw TRopException("TRop::convolve_3_i: unsupported pixel type");
}
//------------------------------------------------------------------------------
void TRop::convolve_i(TRasterP rout, TRasterP rin, int dx, int dy, double conv[], int radius)
{
TRaster32P rin32 = rin;
if (rin32) {
TRaster32P rout32 = rout;
if (rout32) {
doConvolve_i<TPixel32, TPixel32>(rout32, rin32, dx, dy, conv, radius);
return;
}
TRaster64P rout64 = rout;
if (rout64) {
doConvolve_i<TPixel64, TPixel32>(rout64, rin32, dx, dy, conv, radius);
return;
}
} else {
TRaster64P rin64 = rin;
if (rin64) {
TRaster32P rout32 = rout;
if (rout32) {
doConvolve_i<TPixel32, TPixel64>(rout32, rin64, dx, dy, conv, radius);
return;
}
TRaster64P rout64 = rout;
if (rout64) {
doConvolve_i<TPixel64, TPixel64>(rout64, rin64, dx, dy, conv, radius);
return;
}
}
}
throw TRopException("TRop::convolve_i: unsupported pixel type");
}
//------------------------------------------------------------------------------
void TRop::convolve_i(TRasterP rout, TRasterCM32P rin, const TPaletteP &palette,
int dx, int dy, double conv[], int radius)
{
TRaster32P rout32 = rout;
if (rout32) {
doConvolve_cm32_i<TPixel32>(rout32, rin, palette, dx, dy, conv, radius);
return;
}
TRaster64P rout64 = rout;
if (rout64) {
doConvolve_cm32_i<TPixel64>(rout64, rin, palette, dx, dy, conv, radius);
return;
}
throw TRopException("TRop::convolve_i: unsupported pixel type");
}