git_clone/thorvg
1
0
Fork 0
mirror of https://github.com/thorvg/thorvg.git synced 2025-07-27 00:26:51 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

common: remove redundant lzw decoder

Browse source
This commit is contained in:
Hermet Park 2025-02-20 16:30:30 +09:00 • committed by Hermet Park
parent 42457a25c9
commit bed38a4521
2 changed files with 0 additions and 214 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

213
src/common/tvgCompressor.cpp
View file

@ -20,39 +20,6 @@
* SOFTWARE. * SOFTWARE.
*/ */
/*
* Lempel–Ziv–Welch (LZW) decoder by Guilherme R. Lampert(guilherme.ronaldo.lampert@gmail.com)
* This is the compression scheme used by the GIF image format and the Unix 'compress' tool.
* Main differences from this implementation is that End Of Input (EOI) and Clear Codes (CC)
* are not stored in the output and the max code length in bits is 12, vs 16 in compress.
*
* EOI is simply detected by the end of the data stream, while CC happens if the
* dictionary gets filled. Data is written/read from bit streams, which handle
* byte-alignment for us in a transparent way.
* The decoder relies on the hardcoded data layout produced by the encoder, since
* no additional reconstruction data is added to the output, so they must match.
* The nice thing about LZW is that we can reconstruct the dictionary directly from
* the stream of codes generated by the encoder, so this avoids storing additional
* headers in the bit stream.
* The output code length is variable. It starts with the minimum number of bits
* required to store the base byte-sized dictionary and automatically increases
* as the dictionary gets larger (it starts at 9-bits and grows to 10-bits when
* code 512 is added, then 11-bits when 1024 is added, and so on). If the dictionary
* is filled (4096 items for a 12-bits dictionary), the whole thing is cleared and
* the process starts over. This is the main reason why the encoder and the decoder
* must match perfectly, since the lengths of the codes will not be specified with
* the data itself.
* USEFUL LINKS:
* https://en.wikipedia.org/wiki/Lempel%E2%80%93Ziv%E2%80%93Welch
* http://rosettacode.org/wiki/LZW_compression
* http://www.cs.duke.edu/csed/curious/compression/lzw.html
* http://www.cs.cf.ac.uk/Dave/Multimedia/node214.html
* http://marknelson.us/1989/10/01/lzw-data-compression/
*/
#include "config.h" #include "config.h"
#include "tvgCommon.h" #include "tvgCommon.h"
#include "tvgCompressor.h" #include "tvgCompressor.h"
@ -60,186 +27,6 @@
namespace tvg { namespace tvg {
/************************************************************************/
/* LZW Implementation */
/************************************************************************/
//LZW Dictionary helper:
constexpr int Nil = -1;
constexpr int MaxDictBits = 12;
constexpr int StartBits = 9;
constexpr int FirstCode = (1 << (StartBits - 1)); // 256
constexpr int MaxDictEntries = (1 << MaxDictBits); // 4096
struct BitStreamReader
{
const uint8_t* stream; // Pointer to the external bit stream. Not owned by the reader.
const int sizeInBytes; // Size of the stream *in bytes*. Might include padding.
const int sizeInBits; // Size of the stream *in bits*, padding *not* include.
int currBytePos = 0; // Current byte being read in the stream.
int nextBitPos = 0; // Bit position within the current byte to access next. 0 to 7.
int numBitsRead = 0; // Total bits read from the stream so far. Never includes byte-rounding padding.
BitStreamReader(const uint8_t* bitStream, const int byteCount, const int bitCount) : stream(bitStream), sizeInBytes(byteCount), sizeInBits(bitCount)
{
}
bool readNextBit(int& bitOut)
{
if (numBitsRead >= sizeInBits) return false; //We are done.
const uint32_t mask = uint32_t(1) << nextBitPos;
bitOut = !!(stream[currBytePos] & mask);
++numBitsRead;
if (++nextBitPos == 8) {
nextBitPos = 0;
++currBytePos;
}
return true;
}
uint64_t readBitsU64(const int bitCount)
{
uint64_t num = 0;
for (int b = 0; b < bitCount; ++b) {
int bit;
if (!readNextBit(bit)) break;
/* Based on a "Stanford bit-hack":
http://graphics.stanford.edu/~seander/bithacks.html#ConditionalSetOrClearBitsWithoutBranching */
const uint64_t mask = uint64_t(1) << b;
num = (num & ~mask) | (-bit & mask);
}
return num;
}
bool isEndOfStream() const
{
return numBitsRead >= sizeInBits;
}
};
struct Dictionary
{
struct Entry
{
int code;
int value;
};
//Dictionary entries 0-255 are always reserved to the byte/ASCII range.
int size;
Entry entries[MaxDictEntries];
Dictionary()
{
/* First 256 dictionary entries are reserved to the byte/ASCII range.
Additional entries follow for the character sequences found in the input.
Up to 4096 - 256 (MaxDictEntries - FirstCode). */
size = FirstCode;
for (int i = 0; i < size; ++i) {
entries[i].code = Nil;
entries[i].value = i;
}
}
bool add(const int code, const int value)
{
if (size == MaxDictEntries) return false;
entries[size].code = code;
entries[size].value = value;
++size;
return true;
}
bool flush(int & codeBitsWidth)
{
if (size == (1 << codeBitsWidth)) {
++codeBitsWidth;
if (codeBitsWidth > MaxDictBits) {
//Clear the dictionary (except the first 256 byte entries).
codeBitsWidth = StartBits;
size = FirstCode;
return true;
}
}
return false;
}
};
static bool outputByte(int code, uint8_t*& output, int outputSizeBytes, int& bytesDecodedSoFar)
{
if (bytesDecodedSoFar >= outputSizeBytes) return false;
*output++ = static_cast<uint8_t>(code);
++bytesDecodedSoFar;
return true;
}
static bool outputSequence(const Dictionary& dict, int code, uint8_t*& output, int outputSizeBytes, int& bytesDecodedSoFar, int& firstByte)
{
/* A sequence is stored backwards, so we have to write
it to a temp then output the buffer in reverse. */
int i = 0;
uint8_t sequence[MaxDictEntries];
do {
sequence[i++] = dict.entries[code].value;
code = dict.entries[code].code;
} while (code >= 0);
firstByte = sequence[--i];
for (; i >= 0; --i) {
if (!outputByte(sequence[i], output, outputSizeBytes, bytesDecodedSoFar)) return false;
}
return true;
}
uint8_t* lzwDecode(const uint8_t* compressed, uint32_t compressedSizeBytes, uint32_t compressedSizeBits, uint32_t uncompressedSizeBytes)
{
int code = Nil;
int prevCode = Nil;
int firstByte = 0;
int bytesDecoded = 0;
int codeBitsWidth = StartBits;
auto uncompressed = tvg::malloc<uint8_t*>(sizeof(uint8_t) * uncompressedSizeBytes);
auto ptr = uncompressed;
/* We'll reconstruct the dictionary based on the bit stream codes.
Unlike Huffman encoding, we don't store the dictionary as a prefix to the data. */
Dictionary dictionary;
BitStreamReader bitStream(compressed, compressedSizeBytes, compressedSizeBits);
/* We check to avoid an overflow of the user buffer.
If the buffer is smaller than the decompressed size, we break the loop and return the current decompression count. */
while (!bitStream.isEndOfStream()) {
code = static_cast<int>(bitStream.readBitsU64(codeBitsWidth));
if (prevCode == Nil) {
if (!outputByte(code, ptr, uncompressedSizeBytes, bytesDecoded)) break;
firstByte = code;
prevCode = code;
continue;
}
if (code >= dictionary.size) {
if (!outputSequence(dictionary, prevCode, ptr, uncompressedSizeBytes, bytesDecoded, firstByte)) break;
if (!outputByte(firstByte, ptr, uncompressedSizeBytes, bytesDecoded)) break;
} else if (!outputSequence(dictionary, code, ptr, uncompressedSizeBytes, bytesDecoded, firstByte)) break;
dictionary.add(prevCode, firstByte);
if (dictionary.flush(codeBitsWidth)) prevCode = Nil;
else prevCode = code;
}
return uncompressed;
}
/************************************************************************/ /************************************************************************/
/* B64 Implementation */ /* B64 Implementation */
/************************************************************************/ /************************************************************************/

1
src/common/tvgCompressor.h
View file

@ -27,7 +27,6 @@
namespace tvg namespace tvg
{ {
uint8_t* lzwDecode(const uint8_t* compressed, uint32_t compressedSizeBytes, uint32_t compressedSizeBits, uint32_t uncompressedSizeBytes);
size_t b64Decode(const char* encoded, const size_t len, char** decoded); size_t b64Decode(const char* encoded, const size_t len, char** decoded);
unsigned long djb2Encode(const char* str); unsigned long djb2Encode(const char* str);
} }