pixel-count/decompress.c
2022-06-22 22:59:27 +02:00

131 lines
2.7 KiB
C

/*
Not implemented yet
Currently using zlib's decompress (sane)
uint8_t getCM(uint8_t b){
uint8_t r = b<<4;
return r>>4;
}
uint8_t getCINFO(uint8_t b){
return b>>4;
}
bool validateCMFFLG(uint8_t CMF, uint8_t FLG){
uint16_t CMFFLGcheck = (CMF<<8) + FLG;
return getCM(CMF) == 8 && CMFFLGcheck%31 == 0;
}
typedef bool bit;
bool getBit(int n, int bit){
return n & (1 << bit);
}
void printBinary(uint8_t n){
for(int i = 7; i >= 0; --i)
printf("%d", getBit(n, i));
}
uint8_t reverseBits(uint8_t n){
uint8_t r = 0;
for(int i = 0; i < 8; ++i)
r |= ((bool)(n&(1<<(7-i))))<<i;
return r;
}
uint8_t read8(bit *data, uint32_t *i){
// This function is used for decompression
// The first bit is the LSB
uint8_t r = 0;
for(int j = 0; j < 8; ++j){
r += data[(*i)+j]<<j;
}
(*i) += 8;
return r;
}
uint16_t read16(bit *data, uint32_t *i){
// This function is used for decompression
// The first bit is the LSB
uint16_t r = 0;
for(int j = 0; j < 16; ++j){
r += data[(*i)+j]<<j;
}
(*i) += 16;
return r;
}
typedef enum compressionType{
NONE, FIXED, DYNAMIC, ERROR,
} compressionType;
bit *decompressData(bit *data, uint32_t len){
// we have to decompress with zlib
// in zlib the first bit is the least significant
// (b0 b1 b2 b3)
// ftp://ftp.isi.edu/in-notes/rfc1950.txt
// https://www.ietf.org/rfc/rfc1950.txt
// https://stackoverflow.com/questions/42934720/understanding-the-zlib-header-cmf-cm-cinfo-flg-fdict-dictid-flevel-rfc
/*uint8_t CMF = data[0], FLG = data[1], FDICT, FLEVEL;
if(!validateCMFFLG(CMF, FLG)){
fprintf(stderr, "Invalid CMF+FLG check\n");
fclose(fp);
return 0;
}
uint32_t LZ77size = (1<<getCINFO(CMF))<<8;
FDICT = getBit(FLG, 5);
FLEVEL = getBit(FLG, 6) + (getBit(FLG, 7)<<1);
if(FDICT){
fprintf(stderr, "I dont know how to deal with DICT\n");
fclose(fp);
return 0;
}
// Deflate data
// https://www.ietf.org/rfc/rfc1951.txt
/*
Data elements are packed into bytes in order of
increasing bit number within the byte, i.e., starting
with the least-significant bit of the byte.
Data elements other than Huffman codes are packed
starting with the least-significant bit of the data
element.
Huffman codes are packed starting with the most-
significant bit of the code.
*
/
// Adler32
bit *out = malloc(len);
bit BFINAL;
uint8_t BTYPE;
uint32_t ii = 0, oi = 0;
while(!BFINAL){
BFINAL = data[ii++];
if(BFINAL){
// TODO: exit
}
BTYPE = (data[ii]<<1) + data[ii+1];
ii += 2;
if(BTYPE == NONE){
ii += 8-(ii%8);
}
uint16_t LEN, NLEN;
LEN = read16(data, &ii);
NLEN = read16(data, &ii);
for(uint16_t j = 0; j < LEN; ++j){
}
}
}
*/