How to use LZMA SDK in C++? - c++

Question

i have difficulties in using LZMA SDK in my application.
I would like to create a kind of single file compression tool. I dont need any directory support, just need only the LZMA2 stream. But i have no idea on how LZMA SDK is to be used for this.
Please can anyone give me a little example on how the LZMA SDK can be used under C++?

Answer 1

I think that it's a properly little example to use LZMA SDK.
/* LzmaUtil.c -- Test application for LZMA compression
2008-08-05
Igor Pavlov
public domain */
#define _CRT_SECURE_NO_WARNINGS
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include "../LzmaDec.h"
#include "../LzmaEnc.h"
#include "../Alloc.h"
const char *kCantReadMessage = "Can not read input file";
const char *kCantWriteMessage = "Can not write output file";
const char *kCantAllocateMessage = "Can not allocate memory";
const char *kDataErrorMessage = "Data error";
static void *SzAlloc(void *p, size_t size) { p = p; return MyAlloc(size); }
static void SzFree(void *p, void *address) { p = p; MyFree(address); }
static ISzAlloc g_Alloc = { SzAlloc, SzFree };
#define kInBufferSize (1 << 15)
#define kOutBufferSize (1 << 15)
unsigned char g_InBuffer[kInBufferSize];
unsigned char g_OutBuffer[kOutBufferSize];
size_t MyReadFile(FILE *file, void *data, size_t size)
{ return fread(data, 1, size, file); }
int MyReadFileAndCheck(FILE *file, void *data, size_t size)
{ return (MyReadFile(file, data, size) == size); }
size_t MyWriteFile(FILE *file, const void *data, size_t size)
{
if (size == 0)
return 0;
return fwrite(data, 1, size, file);
}
int MyWriteFileAndCheck(FILE *file, const void *data, size_t size)
{ return (MyWriteFile(file, data, size) == size); }
long MyGetFileLength(FILE *file)
{
long length;
fseek(file, 0, SEEK_END);
length = ftell(file);
fseek(file, 0, SEEK_SET);
return length;
}
void PrintHelp(char *buffer)
{
strcat(buffer, "\nLZMA Utility 4.58 Copyright (c) 1999-2008 Igor Pavlov 2008-04-11\n"
"\nUsage: lzma <e|d> inputFile outputFile\n"
" e: encode file\n"
" d: decode file\n");
}
int PrintError(char *buffer, const char *message)
{
strcat(buffer, "\nError: ");
strcat(buffer, message);
strcat(buffer, "\n");
return 1;
}
int PrintErrorNumber(char *buffer, SRes val)
{
sprintf(buffer + strlen(buffer), "\nError code: %x\n", (unsigned)val);
return 1;
}
int PrintUserError(char *buffer)
{
return PrintError(buffer, "Incorrect command");
}
#define IN_BUF_SIZE (1 << 16)
#define OUT_BUF_SIZE (1 << 16)
static int Decode(FILE *inFile, FILE *outFile, char *rs)
{
UInt64 unpackSize;
int thereIsSize; /* = 1, if there is uncompressed size in headers */
int i;
int res = 0;
CLzmaDec state;
/* header: 5 bytes of LZMA properties and 8 bytes of uncompressed size */
unsigned char header[LZMA_PROPS_SIZE + 8];
/* Read and parse header */
if (!MyReadFileAndCheck(inFile, header, sizeof(header)))
return PrintError(rs, kCantReadMessage);
unpackSize = 0;
thereIsSize = 0;
for (i = 0; i < 8; i++)
{
unsigned char b = header[LZMA_PROPS_SIZE + i];
if (b != 0xFF)
thereIsSize = 1;
unpackSize += (UInt64)b << (i * 8);
}
LzmaDec_Construct(&state);
res = LzmaDec_Allocate(&state, header, LZMA_PROPS_SIZE, &g_Alloc);
if (res != SZ_OK)
return res;
{
Byte inBuf[IN_BUF_SIZE];
Byte outBuf[OUT_BUF_SIZE];
size_t inPos = 0, inSize = 0, outPos = 0;
LzmaDec_Init(&state);
for (;;)
{
if (inPos == inSize)
{
inSize = MyReadFile(inFile, inBuf, IN_BUF_SIZE);
inPos = 0;
}
{
SizeT inProcessed = inSize - inPos;
SizeT outProcessed = OUT_BUF_SIZE - outPos;
ELzmaFinishMode finishMode = LZMA_FINISH_ANY;
ELzmaStatus status;
if (thereIsSize && outProcessed > unpackSize)
{
outProcessed = (SizeT)unpackSize;
finishMode = LZMA_FINISH_END;
}
res = LzmaDec_DecodeToBuf(&state, outBuf + outPos, &outProcessed,
inBuf + inPos, &inProcessed, finishMode, &status);
inPos += (UInt32)inProcessed;
outPos += outProcessed;
unpackSize -= outProcessed;
if (outFile != 0)
MyWriteFile(outFile, outBuf, outPos);
outPos = 0;
if (res != SZ_OK || thereIsSize && unpackSize == 0)
break;
if (inProcessed == 0 && outProcessed == 0)
{
if (thereIsSize || status != LZMA_STATUS_FINISHED_WITH_MARK)
res = SZ_ERROR_DATA;
break;
}
}
}
}
LzmaDec_Free(&state, &g_Alloc);
return res;
}
typedef struct _CFileSeqInStream
{
ISeqInStream funcTable;
FILE *file;
} CFileSeqInStream;
static SRes MyRead(void *p, void *buf, size_t *size)
{
if (*size == 0)
return SZ_OK;
*size = MyReadFile(((CFileSeqInStream*)p)->file, buf, *size);
/*
if (*size == 0)
return SZE_FAIL;
*/
return SZ_OK;
}
typedef struct _CFileSeqOutStream
{
ISeqOutStream funcTable;
FILE *file;
} CFileSeqOutStream;
static size_t MyWrite(void *pp, const void *buf, size_t size)
{
return MyWriteFile(((CFileSeqOutStream *)pp)->file, buf, size);
}
static SRes Encode(FILE *inFile, FILE *outFile, char *rs)
{
CLzmaEncHandle enc;
SRes res;
CFileSeqInStream inStream;
CFileSeqOutStream outStream;
CLzmaEncProps props;
enc = LzmaEnc_Create(&g_Alloc);
if (enc == 0)
return SZ_ERROR_MEM;
inStream.funcTable.Read = MyRead;
inStream.file = inFile;
outStream.funcTable.Write = MyWrite;
outStream.file = outFile;
LzmaEncProps_Init(&props);
res = LzmaEnc_SetProps(enc, &props);
if (res == SZ_OK)
{
Byte header[LZMA_PROPS_SIZE + 8];
size_t headerSize = LZMA_PROPS_SIZE;
UInt64 fileSize;
int i;
res = LzmaEnc_WriteProperties(enc, header, &headerSize);
fileSize = MyGetFileLength(inFile);
for (i = 0; i < 8; i++)
header[headerSize++] = (Byte)(fileSize >> (8 * i));
if (!MyWriteFileAndCheck(outFile, header, headerSize))
return PrintError(rs, "writing error");
if (res == SZ_OK)
res = LzmaEnc_Encode(enc, &outStream.funcTable, &inStream.funcTable,
NULL, &g_Alloc, &g_Alloc);
}
LzmaEnc_Destroy(enc, &g_Alloc, &g_Alloc);
return res;
}
int main2(int numArgs, const char *args[], char *rs)
{
FILE *inFile = 0;
FILE *outFile = 0;
char c;
int res;
int encodeMode;
if (numArgs == 1)
{
PrintHelp(rs);
return 0;
}
if (numArgs < 3 || numArgs > 4 || strlen(args[1]) != 1)
return PrintUserError(rs);
c = args[1][0];
encodeMode = (c == 'e' || c == 'E');
if (!encodeMode && c != 'd' && c != 'D')
return PrintUserError(rs);
{
size_t t4 = sizeof(UInt32);
size_t t8 = sizeof(UInt64);
if (t4 != 4 || t8 != 8)
return PrintError(rs, "LZMA UTil needs correct UInt32 and UInt64");
}
inFile = fopen(args[2], "rb");
if (inFile == 0)
return PrintError(rs, "Can not open input file");
if (numArgs > 3)
{
outFile = fopen(args[3], "wb+");
if (outFile == 0)
return PrintError(rs, "Can not open output file");
}
else if (encodeMode)
PrintUserError(rs);
if (encodeMode)
{
res = Encode(inFile, outFile, rs);
}
else
{
res = Decode(inFile, outFile, rs);
}
if (outFile != 0)
fclose(outFile);
fclose(inFile);
if (res != SZ_OK)
{
if (res == SZ_ERROR_MEM)
return PrintError(rs, kCantAllocateMessage);
else if (res == SZ_ERROR_DATA)
return PrintError(rs, kDataErrorMessage);
else
return PrintErrorNumber(rs, res);
}
return 0;
}
int MY_CDECL main(int numArgs, const char *args[])
{
char rs[800] = { 0 };
int res = main2(numArgs, args, rs);
printf(rs);
return res;
}
Also you can see it at:
http://read.pudn.com/downloads151/sourcecode/zip/656407/7z460/C/LzmaUtil/LzmaUtil.c__.htm
http://read.pudn.com/downloads157/sourcecode/zip/698262/LZMA/LzmaUtil.c__.htm

Answer 2

I recently found a nice example, written in C++. Credit goes to GH user Treeki who published the original gist:
// note: -D_7ZIP_ST is required when compiling on non-Windows platforms
// g++ -o lzma_sample -std=c++14 -D_7ZIP_ST lzma_sample.cpp LzmaDec.c LzmaEnc.c LzFind.c
#include <stdio.h>
#include <stdint.h>
#include <string.h>
#include <memory>
#include "LzmaEnc.h"
#include "LzmaDec.h"
static void *_lzmaAlloc(ISzAllocPtr, size_t size) {
return new uint8_t[size];
}
static void _lzmaFree(ISzAllocPtr, void *addr) {
if (!addr)
return;
delete[] reinterpret_cast<uint8_t *>(addr);
}
static ISzAlloc _allocFuncs = {
_lzmaAlloc, _lzmaFree
};
std::unique_ptr<uint8_t[]> lzmaCompress(const uint8_t *input, uint32_t inputSize, uint32_t *outputSize) {
std::unique_ptr<uint8_t[]> result;
// set up properties
CLzmaEncProps props;
LzmaEncProps_Init(&props);
if (inputSize >= (1 << 20))
props.dictSize = 1 << 20; // 1mb dictionary
else
props.dictSize = inputSize; // smaller dictionary = faster!
props.fb = 40;
// prepare space for the encoded properties
SizeT propsSize = 5;
uint8_t propsEncoded[5];
// allocate some space for the compression output
// this is way more than necessary in most cases...
// but better safe than sorry
// (a smarter implementation would use a growing buffer,
// but this requires a bunch of fuckery that is out of
/// scope for this simple example)
SizeT outputSize64 = inputSize * 1.5;
if (outputSize64 < 1024)
outputSize64 = 1024;
auto output = std::make_unique<uint8_t[]>(outputSize64);
int lzmaStatus = LzmaEncode(
output.get(), &outputSize64, input, inputSize,
&props, propsEncoded, &propsSize, 0,
NULL,
&_allocFuncs, &_allocFuncs);
*outputSize = outputSize64 + 13;
if (lzmaStatus == SZ_OK) {
// tricky: we have to generate the LZMA header
// 5 bytes properties + 8 byte uncompressed size
result = std::make_unique<uint8_t[]>(outputSize64 + 13);
uint8_t *resultData = result.get();
memcpy(resultData, propsEncoded, 5);
for (int i = 0; i < 8; i++)
resultData[5 + i] = (inputSize >> (i * 8)) & 0xFF;
memcpy(resultData + 13, output.get(), outputSize64);
}
return result;
}
std::unique_ptr<uint8_t[]> lzmaDecompress(const uint8_t *input, uint32_t inputSize, uint32_t *outputSize) {
if (inputSize < 13)
return NULL; // invalid header!
// extract the size from the header
UInt64 size = 0;
for (int i = 0; i < 8; i++)
size |= (input[5 + i] << (i * 8));
if (size <= (256 * 1024 * 1024)) {
auto blob = std::make_unique<uint8_t[]>(size);
ELzmaStatus lzmaStatus;
SizeT procOutSize = size, procInSize = inputSize - 13;
int status = LzmaDecode(blob.get(), &procOutSize, &input[13], &procInSize, input, 5, LZMA_FINISH_END, &lzmaStatus, &_allocFuncs);
if (status == SZ_OK && procOutSize == size) {
*outputSize = size;
return blob;
}
}
return NULL;
}
void hexdump(const uint8_t *buf, int size) {
int lines = (size + 15) / 16;
for (int i = 0; i < lines; i++) {
printf("%08x | ", i * 16);
int lineMin = i * 16;
int lineMax = lineMin + 16;
int lineCappedMax = (lineMax > size) ? size : lineMax;
for (int j = lineMin; j < lineCappedMax; j++)
printf("%02x ", buf[j]);
for (int j = lineCappedMax; j < lineMax; j++)
printf(" ");
printf("| ");
for (int j = lineMin; j < lineCappedMax; j++) {
if (buf[j] >= 32 && buf[j] <= 127)
printf("%c", buf[j]);
else
printf(".");
}
printf("\n");
}
}
void testIt(const uint8_t *input, int size) {
printf("Test Input:\n");
hexdump(input, size);
uint32_t compressedSize;
auto compressedBlob = lzmaCompress(input, size, &compressedSize);
if (compressedBlob) {
printf("Compressed:\n");
hexdump(compressedBlob.get(), compressedSize);
} else {
printf("Nope, we screwed it\n");
return;
}
// let's try decompressing it now
uint32_t decompressedSize;
auto decompressedBlob = lzmaDecompress(compressedBlob.get(), compressedSize, &decompressedSize);
if (decompressedBlob) {
printf("Decompressed:\n");
hexdump(decompressedBlob.get(), decompressedSize);
} else {
printf("Nope, we screwed it (part 2)\n");
return;
}
printf("----------\n");
}
void testIt(const char *string) {
testIt((const uint8_t *)string, strlen(string));
}
int main(int argc, char **argv) {
testIt("a");
testIt("here is a cool string");
testIt("here's something that should compress pretty well: abcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdef");
return 0;
}

Answer 3

You can refer to this file on how to use lzma2。
https://github.com/Tencent/libpag/blob/aab6391e455193c8ec5b8e2031b495b3fe77b034/test/framework/utils/LzmaUtil.cpp
/////////////////////////////////////////////////////////////////////////////////////////////////
//
// Tencent is pleased to support the open source community by making libpag available.
//
// Copyright (C) 2021 THL A29 Limited, a Tencent company. All rights reserved.
//
// Licensed 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
//
// unless required by applicable law or agreed to in writing, software distributed under the
// license is distributed on an "as is" basis, without warranties or conditions of any kind,
// either express or implied. see the license for the specific language governing permissions
// and limitations under the license.
//
/////////////////////////////////////////////////////////////////////////////////////////////////
#include "LzmaUtil.h"
#include "test/framework/lzma/Lzma2DecMt.h"
#include "test/framework/lzma/Lzma2Enc.h"
namespace pag {
static void* LzmaAlloc(ISzAllocPtr, size_t size) {
return new uint8_t[size];
}
static void LzmaFree(ISzAllocPtr, void* address) {
if (!address) {
return;
}
delete[] reinterpret_cast<uint8_t*>(address);
}
static ISzAlloc gAllocFuncs = {LzmaAlloc, LzmaFree};
class SequentialOutStream {
public:
virtual ~SequentialOutStream() = default;
virtual bool write(const void* data, size_t size) = 0;
};
class SequentialInStream {
public:
virtual ~SequentialInStream() = default;
virtual bool read(void* data, size_t size, size_t* processedSize) = 0;
};
struct CSeqInStreamWrap {
ISeqInStream vt;
std::unique_ptr<SequentialInStream> inStream;
};
struct CSeqOutStreamWrap {
ISeqOutStream vt;
std::unique_ptr<SequentialOutStream> outStream;
};
class BuffPtrInStream : public SequentialInStream {
public:
explicit BuffPtrInStream(const uint8_t* buffer, size_t bufferSize)
: buffer(buffer), bufferSize(bufferSize) {
}
bool read(void* data, size_t size, size_t* processedSize) override {
if (processedSize) {
*processedSize = 0;
}
if (size == 0 || position >= bufferSize) {
return true;
}
auto remain = bufferSize - position;
if (remain > size) {
remain = size;
}
memcpy(data, static_cast<const uint8_t*>(buffer) + position, remain);
position += remain;
if (processedSize) {
*processedSize = remain;
}
return true;
}
private:
const uint8_t* buffer = nullptr;
size_t bufferSize = 0;
size_t position = 0;
};
class VectorOutStream : public SequentialOutStream {
public:
explicit VectorOutStream(std::vector<uint8_t>* buffer) : buffer(buffer) {
}
bool write(const void* data, size_t size) override {
auto oldSize = buffer->size();
buffer->resize(oldSize + size);
memcpy(&(*buffer)[oldSize], data, size);
return true;
}
private:
std::vector<uint8_t>* buffer;
};
class BuffPtrSeqOutStream : public SequentialOutStream {
public:
BuffPtrSeqOutStream(uint8_t* buffer, size_t size) : buffer(buffer), bufferSize(size) {
}
bool write(const void* data, size_t size) override {
auto remain = bufferSize - position;
if (remain > size) {
remain = size;
}
if (remain != 0) {
memcpy(buffer + position, data, remain);
position += remain;
}
return remain != 0 || size == 0;
}
private:
uint8_t* buffer = nullptr;
size_t bufferSize = 0;
size_t position = 0;
};
static const size_t kStreamStepSize = 1 << 31;
static SRes MyRead(const ISeqInStream* p, void* data, size_t* size) {
CSeqInStreamWrap* wrap = CONTAINER_FROM_VTBL(p, CSeqInStreamWrap, vt);
auto curSize = (*size < kStreamStepSize) ? *size : kStreamStepSize;
if (!wrap->inStream->read(data, curSize, &curSize)) {
return SZ_ERROR_READ;
}
*size = curSize;
return SZ_OK;
}
static size_t MyWrite(const ISeqOutStream* p, const void* buf, size_t size) {
auto* wrap = CONTAINER_FROM_VTBL(p, CSeqOutStreamWrap, vt);
if (wrap->outStream->write(buf, size)) {
return size;
}
return 0;
}
class Lzma2Encoder {
public:
Lzma2Encoder() {
encoder = Lzma2Enc_Create(&gAllocFuncs, &gAllocFuncs);
}
~Lzma2Encoder() {
Lzma2Enc_Destroy(encoder);
}
std::shared_ptr<Data> code(const std::shared_ptr<Data>& inputData) {
if (encoder == nullptr || inputData == nullptr || inputData->size() == 0) {
return nullptr;
}
auto inputSize = inputData->size();
CLzma2EncProps lzma2Props;
Lzma2EncProps_Init(&lzma2Props);
lzma2Props.lzmaProps.dictSize = inputSize;
lzma2Props.lzmaProps.level = 9;
lzma2Props.numTotalThreads = 4;
Lzma2Enc_SetProps(encoder, &lzma2Props);
std::vector<uint8_t> outBuf;
outBuf.resize(1 + 8);
outBuf[0] = Lzma2Enc_WriteProperties(encoder);
for (int i = 0; i < 8; i++) {
outBuf[1 + i] = static_cast<uint8_t>(inputSize >> (8 * i));
}
CSeqInStreamWrap inWrap = {};
inWrap.vt.Read = MyRead;
inWrap.inStream = std::make_unique<BuffPtrInStream>(
static_cast<const uint8_t*>(inputData->data()), inputSize);
CSeqOutStreamWrap outStream = {};
outStream.vt.Write = MyWrite;
outStream.outStream = std::make_unique<VectorOutStream>(&outBuf);
auto status =
Lzma2Enc_Encode2(encoder, &outStream.vt, nullptr, nullptr, &inWrap.vt, nullptr, 0, nullptr);
if (status != SZ_OK) {
return nullptr;
}
return Data::MakeWithCopy(&outBuf[0], outBuf.size());
}
private:
CLzma2EncHandle encoder = nullptr;
};
std::shared_ptr<Data> LzmaUtil::Compress(const std::shared_ptr<Data>& pixelData) {
Lzma2Encoder encoder;
return encoder.code(pixelData);
}
class Lzma2Decoder {
public:
Lzma2Decoder() {
decoder = Lzma2DecMt_Create(&gAllocFuncs, &gAllocFuncs);
}
~Lzma2Decoder() {
if (decoder) {
Lzma2DecMt_Destroy(decoder);
}
}
std::shared_ptr<Data> code(const std::shared_ptr<Data>& inputData) {
if (decoder == nullptr || inputData == nullptr || inputData->size() == 0) {
return nullptr;
}
auto input = static_cast<const uint8_t*>(inputData->data());
auto inputSize = inputData->size() - 9;
Byte prop = static_cast<const Byte*>(input)[0];
CLzma2DecMtProps props;
Lzma2DecMtProps_Init(&props);
props.inBufSize_ST = inputSize;
props.numThreads = 1;
UInt64 outBufferSize = 0;
for (int i = 0; i < 8; i++) {
outBufferSize |= (input[1 + i] << (i * 8));
}
auto outBuffer = new uint8_t[outBufferSize];
CSeqInStreamWrap inWrap = {};
inWrap.vt.Read = MyRead;
inWrap.inStream = std::make_unique<BuffPtrInStream>(input + 9, inputSize);
CSeqOutStreamWrap outWrap = {};
outWrap.vt.Write = MyWrite;
outWrap.outStream = std::make_unique<BuffPtrSeqOutStream>(outBuffer, outBufferSize);
UInt64 inProcessed = 0;
int isMT = false;
auto res = Lzma2DecMt_Decode(decoder, prop, &props, &outWrap.vt, &outBufferSize, 1, &inWrap.vt,
&inProcessed, &isMT, nullptr);
if (res == SZ_OK && inputSize == inProcessed) {
return Data::MakeAdopted(outBuffer, outBufferSize, Data::DeleteProc);
}
delete[] outBuffer;
return nullptr;
}
private:
CLzma2DecMtHandle decoder = nullptr;
};
std::shared_ptr<Data> LzmaUtil::Decompress(const std::shared_ptr<Data>& data) {
Lzma2Decoder decoder;
return decoder.code(data);
}
} // namespace pag