bsnes/higan/nall/png.hpp

338 lines
9.0 KiB
C++
Raw Normal View History

#ifndef NALL_PNG_HPP
#define NALL_PNG_HPP
//PNG image decoder
//author: byuu
#include <nall/inflate.hpp>
#include <nall/string.hpp>
namespace nall {
struct png {
//colorType:
//0 = L
//2 = R,G,B
//3 = P
//4 = L,A
//6 = R,G,B,A
struct Info {
unsigned width;
unsigned height;
unsigned bitDepth;
unsigned colorType;
unsigned compressionMethod;
unsigned filterType;
unsigned interlaceMethod;
unsigned bytesPerPixel;
unsigned pitch;
uint8_t palette[256][3];
} info;
uint8_t *data;
unsigned size;
inline bool decode(const string &filename);
inline bool decode(const uint8_t *sourceData, unsigned sourceSize);
inline unsigned readbits(const uint8_t *&data);
unsigned bitpos;
inline png();
inline ~png();
protected:
enum class FourCC : unsigned {
IHDR = 0x49484452,
PLTE = 0x504c5445,
IDAT = 0x49444154,
IEND = 0x49454e44,
};
inline unsigned interlace(unsigned pass, unsigned index);
inline unsigned inflateSize();
inline bool deinterlace(const uint8_t *&inputData, unsigned pass);
inline bool filter(uint8_t *outputData, const uint8_t *inputData, unsigned width, unsigned height);
inline unsigned read(const uint8_t *data, unsigned length);
};
bool png::decode(const string &filename) {
Update to v088r03 release. byuu says: static vector<uint8_t> file::read(const string &filename); replaces: static bool file::read(const string &filename, uint8_t *&data, unsigned &size); This allows automatic deletion of the underlying data. Added vectorstream, which is obviously a vector<uint8_t> wrapper for a data stream. Plan is for all data accesses inside my emulation cores to take stream objects, especially MSU1. This lets you feed the core anything: memorystream, filestream, zipstream, gzipstream, httpstream, etc. There will still be exceptions for link and serial, those need actual library files on disk. But those aren't official hardware devices anyway. So to help with speed a bit, I'm rethinking the video rendering path. Previous system: - core outputs system-native samples (SNES = 19-bit LRGB, NES = 9-bit emphasis+palette, DMG = 2-bit grayscale, etc.) - interfaceSystem transforms samples to 30-bit via lookup table inside the emulation core - interfaceSystem masks off overscan areas, if enabled - interfaceUI runs filter to produce new target buffer, if enabled - interfaceUI transforms 30-bit video to native display depth (24-bit or 30-bit), and applies color-adjustments (gamma, etc) at the same time New system: - all cores now generate an internal palette, and call Interface::videoColor(uint32_t source, uint16_t red, uint16_t green, uint16_t blue) to get native display color post-adjusted (gamma, etc applied already.) - all cores output to uint32_t* buffer now (output video.palette[color] instead of just color) - interfaceUI runs filter to produce new target buffer, if enabled - interfaceUI memcpy()'s buffer to the video card videoColor() is pretty neat. source is the raw pixel (as per the old-format, 19-bit SNES, 9-bit NES, etc), and you can create a color from that if you really want to. Or return that value to get a buffer just like v088 and below. red, green, blue are 16-bits per channel, because why the hell not, right? Just lop off all the bits you don't want. If you have more bits on your display than that, fuck you :P The last step is extremely difficult to avoid. Video cards can and do have pitches that differ from the width of the texture. Trying to make the core account for this would be really awful. And even if we did that, the emulation routine would need to write directly to a video card RAM buffer. Some APIs require you to lock the video buffer while writing, so this would leave the video buffer locked for a long time. Probably not catastrophic, but still awful. And lastly, if the emulation core tried writing directly to the display texture, software filters would no longer be possible (unless you -really- jump through hooks and divert to a memory buffer when a filter is enabled, but ... fuck.) Anyway, the point of all that work was to eliminate an extra video copy, and the need for a really painful 30-bit to 24-bit conversion (three shifts, three masks, three array indexes.) So this basically reverts us, performance-wise, to where we were pre-30 bit support. [...] The downside to this is that we're going to need a filter for each output depth. Since the array type is uint32_t*, and I don't intend to support higher or lower depths, we really only need 24+30-bit versions of each filter. Kinda shitty, but oh well.
2012-04-27 12:12:53 +00:00
if(auto memory = file::read(filename)) {
return decode(memory.data(), memory.size());
}
return false;
}
bool png::decode(const uint8_t *sourceData, unsigned sourceSize) {
if(sourceSize < 8) return false;
if(read(sourceData + 0, 4) != 0x89504e47) return false;
if(read(sourceData + 4, 4) != 0x0d0a1a0a) return false;
uint8_t *compressedData = 0;
unsigned compressedSize = 0;
unsigned offset = 8;
while(offset < sourceSize) {
unsigned length = read(sourceData + offset + 0, 4);
unsigned fourCC = read(sourceData + offset + 4, 4);
unsigned checksum = read(sourceData + offset + 8 + length, 4);
if(fourCC == (unsigned)FourCC::IHDR) {
info.width = read(sourceData + offset + 8, 4);
info.height = read(sourceData + offset + 12, 4);
info.bitDepth = read(sourceData + offset + 16, 1);
info.colorType = read(sourceData + offset + 17, 1);
info.compressionMethod = read(sourceData + offset + 18, 1);
info.filterType = read(sourceData + offset + 19, 1);
info.interlaceMethod = read(sourceData + offset + 20, 1);
if(info.bitDepth == 0 || info.bitDepth > 16) return false;
if(info.bitDepth & (info.bitDepth - 1)) return false; //not a power of two
if(info.compressionMethod != 0) return false;
if(info.filterType != 0) return false;
if(info.interlaceMethod != 0 && info.interlaceMethod != 1) return false;
switch(info.colorType) {
case 0: info.bytesPerPixel = info.bitDepth * 1; break; //L
case 2: info.bytesPerPixel = info.bitDepth * 3; break; //R,G,B
case 3: info.bytesPerPixel = info.bitDepth * 1; break; //P
case 4: info.bytesPerPixel = info.bitDepth * 2; break; //L,A
case 6: info.bytesPerPixel = info.bitDepth * 4; break; //R,G,B,A
default: return false;
}
if(info.colorType == 2 || info.colorType == 4 || info.colorType == 6)
if(info.bitDepth != 8 && info.bitDepth != 16) return false;
if(info.colorType == 3 && info.bitDepth == 16) return false;
info.bytesPerPixel = (info.bytesPerPixel + 7) / 8;
info.pitch = (int)info.width * info.bytesPerPixel;
}
if(fourCC == (unsigned)FourCC::PLTE) {
if(length % 3) return false;
for(unsigned n = 0, p = offset + 8; n < length / 3; n++) {
info.palette[n][0] = sourceData[p++];
info.palette[n][1] = sourceData[p++];
info.palette[n][2] = sourceData[p++];
}
}
if(fourCC == (unsigned)FourCC::IDAT) {
compressedData = (uint8_t*)realloc(compressedData, compressedSize + length);
memcpy(compressedData + compressedSize, sourceData + offset + 8, length);
compressedSize += length;
}
if(fourCC == (unsigned)FourCC::IEND) {
break;
}
offset += 4 + 4 + length + 4;
}
unsigned interlacedSize = inflateSize();
uint8_t *interlacedData = new uint8_t[interlacedSize];
bool result = inflate(interlacedData, interlacedSize, compressedData + 2, compressedSize - 6);
delete[] compressedData;
if(result == false) {
delete[] interlacedData;
return false;
}
size = info.width * info.height * info.bytesPerPixel;
data = new uint8_t[size];
if(info.interlaceMethod == 0) {
if(filter(data, interlacedData, info.width, info.height) == false) {
delete[] interlacedData;
delete[] data;
data = 0;
return false;
}
} else {
const uint8_t *passData = interlacedData;
for(unsigned pass = 0; pass < 7; pass++) {
if(deinterlace(passData, pass) == false) {
delete[] interlacedData;
delete[] data;
data = 0;
return false;
}
}
}
delete[] interlacedData;
return true;
}
unsigned png::interlace(unsigned pass, unsigned index) {
static const unsigned data[7][4] = {
//x-distance, y-distance, x-origin, y-origin
{ 8, 8, 0, 0 },
{ 8, 8, 4, 0 },
{ 4, 8, 0, 4 },
{ 4, 4, 2, 0 },
{ 2, 4, 0, 2 },
{ 2, 2, 1, 0 },
{ 1, 2, 0, 1 },
};
return data[pass][index];
}
unsigned png::inflateSize() {
if(info.interlaceMethod == 0) {
return info.width * info.height * info.bytesPerPixel + info.height;
}
unsigned size = 0;
for(unsigned pass = 0; pass < 7; pass++) {
unsigned xd = interlace(pass, 0), yd = interlace(pass, 1);
unsigned xo = interlace(pass, 2), yo = interlace(pass, 3);
unsigned width = (info.width + (xd - xo - 1)) / xd;
unsigned height = (info.height + (yd - yo - 1)) / yd;
if(width == 0 || height == 0) continue;
size += width * height * info.bytesPerPixel + height;
}
return size;
}
bool png::deinterlace(const uint8_t *&inputData, unsigned pass) {
unsigned xd = interlace(pass, 0), yd = interlace(pass, 1);
unsigned xo = interlace(pass, 2), yo = interlace(pass, 3);
unsigned width = (info.width + (xd - xo - 1)) / xd;
unsigned height = (info.height + (yd - yo - 1)) / yd;
if(width == 0 || height == 0) return true;
unsigned outputSize = width * height * info.bytesPerPixel;
uint8_t *outputData = new uint8_t[outputSize];
bool result = filter(outputData, inputData, width, height);
const uint8_t *rd = outputData;
for(unsigned y = yo; y < info.height; y += yd) {
uint8_t *wr = data + y * info.pitch;
for(unsigned x = xo; x < info.width; x += xd) {
for(unsigned b = 0; b < info.bytesPerPixel; b++) {
wr[x * info.bytesPerPixel + b] = *rd++;
}
}
}
inputData += outputSize + height;
delete[] outputData;
return result;
}
bool png::filter(uint8_t *outputData, const uint8_t *inputData, unsigned width, unsigned height) {
uint8_t *wr = outputData;
const uint8_t *rd = inputData;
int bpp = info.bytesPerPixel, pitch = width * bpp;
for(int y = 0; y < height; y++) {
uint8_t filter = *rd++;
switch(filter) {
case 0x00: //None
for(int x = 0; x < pitch; x++) {
wr[x] = rd[x];
}
break;
case 0x01: //Subtract
for(int x = 0; x < pitch; x++) {
wr[x] = rd[x] + (x - bpp < 0 ? 0 : wr[x - bpp]);
}
break;
case 0x02: //Above
for(int x = 0; x < pitch; x++) {
wr[x] = rd[x] + (y - 1 < 0 ? 0 : wr[x - pitch]);
}
break;
case 0x03: //Average
for(int x = 0; x < pitch; x++) {
short a = x - bpp < 0 ? 0 : wr[x - bpp];
short b = y - 1 < 0 ? 0 : wr[x - pitch];
wr[x] = rd[x] + (uint8_t)((a + b) / 2);
}
break;
case 0x04: //Paeth
for(int x = 0; x < pitch; x++) {
short a = x - bpp < 0 ? 0 : wr[x - bpp];
short b = y - 1 < 0 ? 0 : wr[x - pitch];
short c = x - bpp < 0 || y - 1 < 0 ? 0 : wr[x - pitch - bpp];
short p = a + b - c;
short pa = p > a ? p - a : a - p;
short pb = p > b ? p - b : b - p;
short pc = p > c ? p - c : c - p;
uint8_t paeth = (uint8_t)((pa <= pb && pa <= pc) ? a : (pb <= pc) ? b : c);
wr[x] = rd[x] + paeth;
}
break;
default: //Invalid
return false;
}
rd += pitch;
wr += pitch;
}
return true;
}
unsigned png::read(const uint8_t *data, unsigned length) {
unsigned result = 0;
while(length--) result = (result << 8) | (*data++);
return result;
}
unsigned png::readbits(const uint8_t *&data) {
unsigned result = 0;
switch(info.bitDepth) {
case 1:
result = (*data >> bitpos) & 1;
bitpos++;
if(bitpos == 8) { data++; bitpos = 0; }
break;
case 2:
result = (*data >> bitpos) & 3;
bitpos += 2;
if(bitpos == 8) { data++; bitpos = 0; }
break;
case 4:
result = (*data >> bitpos) & 15;
bitpos += 4;
if(bitpos == 8) { data++; bitpos = 0; }
break;
case 8:
result = *data++;
break;
case 16:
result = (data[0] << 8) | (data[1] << 0);
data += 2;
break;
}
return result;
}
png::png() : data(nullptr) {
bitpos = 0;
}
png::~png() {
if(data) delete[] data;
}
}
#endif