ddidderr/zstd-rs
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Merge remote-tracking branch 'refs/remotes/Cyan4973/v05x' into v05x

Browse Source 
# Conflicts:
#	lib/zstd_compress.c
#	lib/zstd_static.h
This commit is contained in:
inikep
2016-02-03 17:29:40 +01:00
parent 78e5ea3a32 b010b3b627
commit 53721f134c
28 changed files with 5168 additions and 433 deletions
Download Patch File Download Diff File Expand all files Collapse all files
dictBuilder/dibcli.c
+10 -3
View File
@@ -184,7 +184,7 @@ int main(int argCount, const char** argv)
if (!strcmp(argument, "--verbose")) { g_displayLevel++; if (g_displayLevel<3) g_displayLevel=3; continue; }
if (!strcmp(argument, "--quiet")) { g_displayLevel--; continue; }
if (!strcmp(argument, "--maxdict")) { nextArgumentIsMaxDict=1; continue; }
if (!strcmp(argument, "--fast")) { selectionLevel=0; cLevel=1; continue; }
if (!strcmp(argument, "--fast")) { selectionLevel=1; cLevel=1; continue; }
/* Decode commands (note : aggregated commands are allowed) */
if (argument[0]=='-') {
@@ -247,8 +247,15 @@ int main(int argCount, const char** argv)
}
/* building ... */
DiB_setNotificationLevel(g_displayLevel);
operationResult = DiB_trainDictionary(dictFileName, maxDictSize, selectionLevel, cLevel, filenameTable, filenameIdx);
{
DiB_params_t param;
param.selectivityLevel = selectionLevel;
param.compressionLevel = cLevel;
DiB_setNotificationLevel(g_displayLevel);
operationResult = DiB_trainFromFiles(dictFileName, maxDictSize,
filenameTable, filenameIdx,
param);
}
if (main_pause) waitEnter();
free((void*)filenameTable);
dictBuilder/dictBuilder.c
+138 -96
View File
@@ -43,14 +43,14 @@
/*-*************************************
* Includes
***************************************/
#include <stdlib.h> /* malloc, free */
#include <string.h> /* memset */
#include <stdio.h> /* fprintf, fopen, ftello64 */
#include <sys/types.h> /* stat64 */
#include <sys/stat.h> /* stat64 */
#include <time.h> /* clock */
#include <stdlib.h> /* malloc, free */
#include <string.h> /* memset */
#include <stdio.h> /* fprintf, fopen, ftello64 */
#include <sys/types.h> /* stat64 */
#include <sys/stat.h> /* stat64 */
#include <time.h> /* clock */
#include "mem.h" /* read */
#include "mem.h" /* read */
#include "error_private.h"
#include "divsufsort.h"
#include "dictBuilder.h"
@@ -58,17 +58,13 @@
#include "huff0_static.h"
/* *************************************
/*-*************************************
* Compiler specifics
***************************************/
#if !defined(S_ISREG)
# define S_ISREG(x) (((x) & S_IFMT) == S_IFREG)
#endif
#ifdef _MSC_VER
#define snprintf sprintf_s
#endif
/*-*************************************
* Constants
@@ -87,6 +83,9 @@ static const size_t maxMemory = (sizeof(size_t) == 4) ? (2 GB - 64 MB) : ((size_
#define MINRATIO 4
static const U32 g_compressionLevel_default = 5;
static const U32 g_selectivity_default = 9;
static const size_t g_provision_entropySize = 200;
static const size_t g_min_fast_dictContent = 192;
/*-*************************************
@@ -146,6 +145,10 @@ static unsigned DiB_GetMilliSpan(clock_t nPrevious)
return nSpan;
}
unsigned DiB_isError(size_t errorCode) { return ERR_isError(errorCode); }
const char* DiB_getErrorName(size_t errorCode) { return ERR_getErrorName(errorCode); }
/* ********************************************************
* File related operations
@@ -563,9 +566,8 @@ static U32 DiB_dictSize(const dictItem* dictList)
static void DiB_trainBuffer(dictItem* dictList, U32 dictListSize,
const void* const buffer, const size_t bufferSize, /* buffer must end with noisy guard band */
const char* displayName,
const size_t* fileSizes, unsigned nbFiles, unsigned maxDictSize,
U32 shiftRatio)
const size_t* fileSizes, unsigned nbFiles,
U32 shiftRatio, unsigned maxDictSize)
{
saidx_t* const suffix0 = (saidx_t*)malloc((bufferSize+2)*sizeof(*suffix0));
saidx_t* const suffix = suffix0+1;
@@ -583,7 +585,7 @@ static void DiB_trainBuffer(dictItem* dictList, U32 dictListSize,
memset(doneMarks, 0, bufferSize+16);
/* sort */
DISPLAYLEVEL(2, "sorting %s ...\n", displayName);
DISPLAYLEVEL(2, "sorting %u files of total size %u MB ...\n", nbFiles, (U32)(bufferSize>>20));
errorCode = divsufsort((const sauchar_t*)buffer, suffix, (saidx_t)bufferSize);
if (errorCode != 0) EXM_THROW(2, "sort failed");
suffix[bufferSize] = (saidx_t)bufferSize; /* leads into noise */
@@ -699,7 +701,7 @@ static void DiB_countEStats(EStats_ress_t esr,
#define OFFCODE_MAX 18
static size_t DiB_analyzeEntropy(void* dstBuffer, size_t maxDstSize,
unsigned compressionLevel,
const void* srcBuffer, size_t* fileSizes, unsigned nbFiles,
const void* srcBuffer, const size_t* fileSizes, unsigned nbFiles,
const void* dictBuffer, size_t dictBufferSize)
{
U32 countLit[256];
@@ -793,8 +795,7 @@ static size_t DiB_analyzeEntropy(void* dstBuffer, size_t maxDstSize,
static void DiB_saveDict(const char* dictFileName,
const void* buff1, size_t buff1Size,
const void* buff2, size_t buff2Size)
const void* buff, size_t buffSize)
{
FILE* f;
size_t n;
@@ -802,11 +803,8 @@ static void DiB_saveDict(const char* dictFileName,
f = fopen(dictFileName, "wb");
if (f==NULL) EXM_THROW(3, "cannot open %s ", dictFileName);
n = fwrite(buff1, 1, buff1Size, f);
if (n!=buff1Size) EXM_THROW(4, "%s : write error", dictFileName)
n = fwrite(buff2, 1, buff2Size, f);
if (n!=buff2Size) EXM_THROW(4, "%s : write error", dictFileName)
n = fwrite(buff, 1, buffSize, f);
if (n!=buffSize) EXM_THROW(4, "%s : write error", dictFileName)
n = (size_t)fclose(f);
if (n!=0) EXM_THROW(5, "%s : flush error", dictFileName)
@@ -853,46 +851,35 @@ static size_t DiB_fastSampling(void* dictBuffer, size_t dictSize,
}
int DiB_trainDictionary(const char* dictFileName, unsigned maxDictSize,
unsigned shiftRatio, unsigned compressionLevel,
const char** fileNamesTable, unsigned nbFiles)
static size_t DiB_trainFromBuffer_internal(
void* dictBuffer, size_t maxDictSize,
const void* samplesBuffer, const size_t* sampleSizes, unsigned nbSamples,
DiB_params_t params)
{
void* srcBuffer;
size_t benchedSize;
size_t* fileSizes = (size_t*)malloc(nbFiles * sizeof(size_t));
unsigned long long totalSizeToLoad = DiB_getTotalFileSize(fileNamesTable, nbFiles);
const U32 dictListSize = MAX( MAX(DICTLISTSIZE, nbFiles), maxDictSize/16);
const U32 dictListSize = MAX( MAX(DICTLISTSIZE, nbSamples), (U32)(maxDictSize/16));
dictItem* dictList = (dictItem*)malloc(dictListSize * sizeof(*dictList));
char mfName[20] = {0};
const char* displayName = NULL;
unsigned selectivity = params.selectivityLevel;
unsigned compressionLevel = params.compressionLevel;
size_t targetDictSize = maxDictSize - g_provision_entropySize;
size_t sBuffSize;
size_t dictSize = 0;
/* checks */
if (maxDictSize <= g_provision_entropySize + g_min_fast_dictContent) return ERROR(dstSize_tooSmall);
/* init */
benchedSize = DiB_findMaxMem(totalSizeToLoad * MEMMULT) / MEMMULT;
if ((unsigned long long)benchedSize > totalSizeToLoad) benchedSize = (size_t)totalSizeToLoad;
if (benchedSize < totalSizeToLoad)
DISPLAYLEVEL(1, "Not enough memory; training on %u MB only...\n", (unsigned)(benchedSize >> 20));
/* Memory allocation & restrictions */
srcBuffer = malloc(benchedSize+NOISELENGTH); /* + noise */
if ((!fileSizes) || (!srcBuffer) || (!dictList)) EXM_THROW(12, "not enough memory for DiB_trainFiles"); /* should not happen */
{ unsigned u; for (u=0, sBuffSize=0; u<nbSamples; u++) sBuffSize += sampleSizes[u]; }
if (!dictList) { DISPLAYLEVEL(1, "not enough memory for DiB_trainFromBuffer"); return ERROR(memory_allocation); }
DiB_initDictItem(dictList);
if (selectivity==0) selectivity = g_selectivity_default;
if (compressionLevel==0) compressionLevel = g_compressionLevel_default;
/* Load input buffer */
DiB_loadFiles(srcBuffer, benchedSize, fileSizes, fileNamesTable, nbFiles);
DiB_fillNoise((char*)srcBuffer + benchedSize, NOISELENGTH); /* guard band, for end of buffer condition */
/* analyze sequences (non-fast mode) */
if (shiftRatio>0)
{
snprintf (mfName, sizeof(mfName), " %u files", nbFiles);
if (nbFiles > 1) displayName = mfName;
else displayName = fileNamesTable[0];
/* select stripes */
if (selectivity>1) {
DiB_trainBuffer(dictList, dictListSize,
srcBuffer, benchedSize,
displayName,
fileSizes, nbFiles, maxDictSize,
shiftRatio);
samplesBuffer, sBuffSize,
sampleSizes, nbSamples,
selectivity, (U32)targetDictSize);
/* display best matches */
if (g_displayLevel>= 3) {
@@ -907,72 +894,127 @@ int DiB_trainDictionary(const char* dictFileName, unsigned maxDictSize,
U32 d = MIN(40, l);
DISPLAYLEVEL(3, "%3u:%3u bytes at pos %8u, savings %7u bytes |",
u, l, p, dictList[u].savings);
DiB_printHex(3, (char*)srcBuffer+p, d);
DiB_printHex(3, (const char*)samplesBuffer+p, d);
DISPLAYLEVEL(3, "| \n");
} } }
/* create dictionary */
{
void* dictContent;
U32 dictContentSize = DiB_dictSize(dictList);
void* dictHeader;
size_t dictHeaderSize, hSize, addedContentLength;
size_t hSize;
BYTE* ptr;
U32 u;
/* build dict */
#define EBSIZE (2 KB)
dictHeaderSize = EBSIZE;
dictHeader = malloc(dictHeaderSize);
dictContent = malloc(maxDictSize);
if (!dictHeader || !dictContent) EXM_THROW(2, "not enough memory");
/* build dict content */
ptr = (BYTE*)dictContent + maxDictSize;
ptr = (BYTE*)dictBuffer + maxDictSize;
for (u=1; u<dictList->pos; u++) {
U32 l = dictList[u].length;
ptr -= l;
memcpy(ptr, (char*)srcBuffer+dictList[u].pos, l);
if (ptr<(BYTE*)dictBuffer) return ERROR(GENERIC); /* should not happen */
memcpy(ptr, (const char*)samplesBuffer+dictList[u].pos, l);
}
/* fast mode dict content */
if (shiftRatio==0) { /* note could also be used to complete a dictionary, but not necessarily better */
addedContentLength = ptr-(BYTE*)dictContent;
DISPLAYLEVEL(2, "\r%70s\r", ""); /* clean display line */
DISPLAYLEVEL(2, "Adding %u KB from fast sampling \n", (U32)(addedContentLength>>10));
addedContentLength = DiB_fastSampling(dictContent, addedContentLength, srcBuffer, benchedSize);
if (!ERR_isError(addedContentLength))
ptr -= addedContentLength, dictContentSize += addedContentLength;
if (selectivity==1) { /* note could also be used to complete a dictionary, but not necessarily better */
DISPLAYLEVEL(3, "\r%70s\r", ""); /* clean display line */
DISPLAYLEVEL(3, "Adding %u KB with fast sampling \n", (U32)(targetDictSize>>10));
dictContentSize = (U32)DiB_fastSampling((char*)dictBuffer + g_provision_entropySize,
targetDictSize, samplesBuffer, sBuffSize);
}
/* dictionary header */
MEM_writeLE32(dictHeader, ZSTD_DICT_MAGIC);
/* dictionary header */
MEM_writeLE32(dictBuffer, ZSTD_DICT_MAGIC);
hSize = 4;
dictHeaderSize -= 4;
/* entropic tables */
DISPLAYLEVEL(2, "\r%70s\r", ""); /* clean display line */
DISPLAYLEVEL(2, "statistics ... \n");
hSize += DiB_analyzeEntropy((char*)dictHeader+4, dictHeaderSize,
hSize += DiB_analyzeEntropy((char*)dictBuffer+4, maxDictSize-4,
compressionLevel,
srcBuffer, fileSizes, nbFiles,
ptr, dictContentSize);
samplesBuffer, sampleSizes, nbSamples,
(char*)dictBuffer + maxDictSize - dictContentSize, dictContentSize);
/* save dict */
{
size_t dictSize = hSize + dictContentSize;
DISPLAYLEVEL(2, "Save dictionary of size %u into file %s \n", (U32)dictSize, dictFileName);
DiB_saveDict(dictFileName, dictHeader, hSize, ptr, dictContentSize);
//DiB_saveDict(dictFileName, NULL, 0, dictContent, dictContentSize); // content only
}
/* clean */
free(dictHeader);
free(dictContent);
if (hSize + dictContentSize < maxDictSize)
memmove((char*)dictBuffer + hSize, (char*)dictBuffer + maxDictSize - dictContentSize, dictContentSize);
dictSize = MIN(maxDictSize, hSize+dictContentSize);
}
/* clean up */
free(srcBuffer);
free(fileSizes);
free(dictList);
return 0;
return dictSize;
}
/* issue : samplesBuffer need to be followed by a noisy guard band.
* work around : duplicate the buffer, and add the noise ? */
size_t DiB_trainFromBuffer(void* dictBuffer, size_t maxDictSize,
const void* samplesBuffer, const size_t* sampleSizes, unsigned nbSamples,
DiB_params_t params)
{
size_t sBuffSize;
void* newBuff;
size_t result;
{ unsigned u; for (u=0, sBuffSize=0; u<nbSamples; u++) sBuffSize += sampleSizes[u]; }
newBuff = malloc(sBuffSize + NOISELENGTH);
if (!newBuff) return ERROR(memory_allocation);
memcpy(newBuff, samplesBuffer, sBuffSize);
DiB_fillNoise((char*)newBuff + sBuffSize, NOISELENGTH); /* guard band, for end of buffer condition */
result = DiB_trainFromBuffer_internal(dictBuffer, maxDictSize,
newBuff, sampleSizes, nbSamples,
params);
free(newBuff);
return result;
}
int DiB_trainFromFiles(const char* dictFileName, unsigned maxDictSize,
const char** fileNamesTable, unsigned nbFiles,
DiB_params_t params)
{
void* srcBuffer;
size_t benchedSize;
size_t* fileSizes = (size_t*)malloc(nbFiles * sizeof(size_t));
unsigned long long totalSizeToLoad = DiB_getTotalFileSize(fileNamesTable, nbFiles);
void* dictBuffer = malloc(maxDictSize);
size_t dictSize;
int result = 0;
/* init */
benchedSize = DiB_findMaxMem(totalSizeToLoad * MEMMULT) / MEMMULT;
if ((unsigned long long)benchedSize > totalSizeToLoad) benchedSize = (size_t)totalSizeToLoad;
if (benchedSize < totalSizeToLoad)
DISPLAYLEVEL(1, "Not enough memory; training on %u MB only...\n", (unsigned)(benchedSize >> 20));
/* Memory allocation & restrictions */
srcBuffer = malloc(benchedSize+NOISELENGTH); /* + noise */
if ((!fileSizes) || (!srcBuffer) || (!dictBuffer)) EXM_THROW(12, "not enough memory for DiB_trainFiles"); /* should not happen */
/* Load input buffer */
DiB_loadFiles(srcBuffer, benchedSize, fileSizes, fileNamesTable, nbFiles);
DiB_fillNoise((char*)srcBuffer + benchedSize, NOISELENGTH); /* guard band, for end of buffer condition */
/* call buffer version */
dictSize = DiB_trainFromBuffer_internal(dictBuffer, maxDictSize,
srcBuffer, fileSizes, nbFiles,
params);
if (DiB_isError(dictSize))
{
DISPLAYLEVEL(1, "dictionary training failed : %s", DiB_getErrorName(dictSize)); /* should not happen */
result = 1;
goto _cleanup;
}
/* save dict */
DISPLAYLEVEL(2, "Save dictionary of size %u into file %s \n", (U32)dictSize, dictFileName);
DiB_saveDict(dictFileName, dictBuffer, dictSize);
/* clean up */
_cleanup:
free(srcBuffer);
free(dictBuffer);
free(fileSizes);
return result;
}
dictBuilder/dictBuilder.h
+45 -10
View File
@@ -26,6 +26,9 @@
/* This library is designed for a single-threaded console application.
* It exit() and printf() into stderr when it encounters an error condition. */
#ifndef DICTBUILDER_H_001
#define DICTBUILDER_H_001
/*-*************************************
* Version
***************************************/
@@ -36,24 +39,56 @@
unsigned DiB_versionNumber (void);
/*-*************************************
* Public type
***************************************/
typedef struct {
unsigned selectivityLevel; /* 0 means default; larger => bigger selection => larger dictionary */
unsigned compressionLevel; /* 0 means default; target a specific zstd compression level */
} DiB_params_t;
/*-*************************************
* Public functions
***************************************/
/*! DiB_trainDictionary
Train a dictionary from a set of files provided by @fileNamesTable
Resulting dictionary is written in file @dictFileName.
@selectivityLevel change criteria for insertion into the dictionary (more => bigger selection => larger dictionary)
@compressionLevel can be used to target a specific compression level of zstd. 0 means "default".
@result : 0 == ok
/*! DiB_trainFromBuffer
Train a dictionary from a memory buffer @samplesBuffer
where @nbSamples samples have been stored concatenated.
Each sample size is provided into an orderly table @sampleSizes.
Resulting dictionary will be saved into @dictBuffer.
@parameters is optional and can be provided with 0 values to mean "default".
@result : size of dictionary stored into @dictBuffer (<= @dictBufferSize)
or an error code, which can be tested by DiB_isError().
note : DiB_trainFromBuffer() will send notifications into stderr if instructed to, using DiB_setNotificationLevel()
*/
int DiB_trainDictionary(const char* dictFileName, unsigned maxDictSize,
unsigned selectivityLevel, unsigned compressionLevel,
const char** fileNamesTable, unsigned nbFiles);
size_t DiB_trainFromBuffer(void* dictBuffer, size_t dictBufferSize,
const void* samplesBuffer, const size_t* sampleSizes, unsigned nbSamples,
DiB_params_t parameters);
/*! DiB_trainFromFiles
Train a dictionary from a set of files provided by @fileNamesTable
Resulting dictionary is written into file @dictFileName.
@parameters is optional and can be provided with 0 values.
@result : 0 == ok. Any other : error.
*/
int DiB_trainFromFiles(const char* dictFileName, unsigned maxDictSize,
const char** fileNamesTable, unsigned nbFiles,
DiB_params_t parameters);
/*-*************************************
* Helper functions
***************************************/
unsigned DiB_isError(size_t errorCode);
const char* DiB_getErrorName(size_t errorCode);
/*! DiB_setNotificationLevel
Set amount of notification to be displayed on the console.
0 = no console notification (default).
default initial value : 0 = no console notification.
Note : not thread-safe (use a global constant)
*/
void DiB_setNotificationLevel(unsigned l);
#endif
lib/Makefile
+1 -1
View File
@@ -53,7 +53,7 @@ LIBDIR ?= $(PREFIX)/lib
INCLUDEDIR=$(PREFIX)/include
ZSTD_FILES := zstd_compress.c zstd_decompress.c fse.c huff0.c
ZSTD_LEGACY:= legacy/zstd_v01.c legacy/zstd_v02.c legacy/zstd_v03.c
ZSTD_LEGACY:= legacy/zstd_v01.c legacy/zstd_v02.c legacy/zstd_v03.c legacy/zstd_v04.c
ifeq ($(ZSTD_LEGACY_SUPPORT), 0)
CPPFLAGS += -DZSTD_LEGACY_SUPPORT=0
lib/bitstream.h
+27 -36
View File
@@ -1,8 +1,8 @@
/* ******************************************************************
bitstream
Part of NewGen Entropy library
Part of FSE library
header file (to include)
Copyright (C) 2013-2015, Yann Collet.
Copyright (C) 2013-2016, Yann Collet.
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
@@ -31,7 +31,6 @@
You can contact the author at :
- Source repository : https://github.com/Cyan4973/FiniteStateEntropy
- Public forum : https://groups.google.com/forum/#!forum/lz4c
****************************************************************** */
#ifndef BITSTREAM_H_MODULE
#define BITSTREAM_H_MODULE
@@ -47,17 +46,17 @@ extern "C" {
* these functions are defined into a .h to be included.
*/
/******************************************
* Includes
/*-****************************************
* Dependencies
******************************************/
#include "mem.h" /* unaligned access routines */
#include "error_private.h" /* error codes and messages */
/********************************************
* bitStream compression API (write forward)
/*-******************************************
* bitStream encoding API (write forward)
********************************************/
/*
/*!
* bitStream can mix input from multiple sources.
* A critical property of these streams is that they encode and decode in **reverse** direction.
* So the first bit sequence you add will be the last to be read, like a LIFO stack.
@@ -71,32 +70,32 @@ typedef struct
char* endPtr;
} BIT_CStream_t;
MEM_STATIC size_t BIT_initCStream(BIT_CStream_t* bitC, void* dstBuffer, size_t maxDstSize);
MEM_STATIC size_t BIT_initCStream(BIT_CStream_t* bitC, void* dstBuffer, size_t dstCapacity);
MEM_STATIC void BIT_addBits(BIT_CStream_t* bitC, size_t value, unsigned nbBits);
MEM_STATIC void BIT_flushBits(BIT_CStream_t* bitC);
MEM_STATIC size_t BIT_closeCStream(BIT_CStream_t* bitC);
/*
* Start by initCStream, providing the maximum size of write buffer to write into.
/*!
* Start by initCStream, providing the size of buffer to write into.
* bitStream will never write outside of this buffer.
* buffer must be at least as large as a size_t, otherwise function result will be an error code.
* @dstCapacity must be >= sizeof(size_t), otherwise @return will be an error code.
*
* bits are first added to a local register.
* Local register is size_t, hence 64-bits on 64-bits systems, or 32-bits on 32-bits systems.
* Writing data into memory is a manual operation, performed by the flushBits function.
* Writing data into memory is an explicit operation, performed by the flushBits function.
* Hence keep track how many bits are potentially stored into local register to avoid register overflow.
* After a flushBits, a maximum of 7 bits might still be stored into local register.
*
* Avoid storing elements of more than 25 bits if you want compatibility with 32-bits bitstream readers.
* Avoid storing elements of more than 24 bits if you want compatibility with 32-bits bitstream readers.
*
* Last operation is to close the bitStream.
* The function returns the final size of CStream in bytes.
* If data couldn't fit into dstBuffer, it will return a 0 ( == not storable)
* If data couldn't fit into @dstBuffer, it will return a 0 ( == not storable)
*/
/**********************************************
* bitStream decompression API (read backward)
/*-********************************************
* bitStream decoding API (read backward)
**********************************************/
typedef struct
{
@@ -118,19 +117,19 @@ MEM_STATIC BIT_DStream_status BIT_reloadDStream(BIT_DStream_t* bitD);
MEM_STATIC unsigned BIT_endOfDStream(const BIT_DStream_t* bitD);
/*
/*!
* Start by invoking BIT_initDStream().
* A chunk of the bitStream is then stored into a local register.
* Local register size is 64-bits on 64-bits systems, 32-bits on 32-bits systems (size_t).
* You can then retrieve bitFields stored into the local register, **in reverse order**.
* Local register is manually filled from memory by the BIT_reloadDStream() method.
* Local register is explicitly reloaded from memory by the BIT_reloadDStream() method.
* A reload guarantee a minimum of ((8*sizeof(size_t))-7) bits when its result is BIT_DStream_unfinished.
* Otherwise, it can be less than that, so proceed accordingly.
* Checking if DStream has reached its end can be performed with BIT_endOfDStream()
*/
/******************************************
/*-****************************************
* unsafe API
******************************************/
MEM_STATIC void BIT_addBitsFast(BIT_CStream_t* bitC, size_t value, unsigned nbBits);
@@ -144,7 +143,7 @@ MEM_STATIC size_t BIT_readBitsFast(BIT_DStream_t* bitD, unsigned nbBits);
/****************************************************************
/*-**************************************************************
* Helper functions
****************************************************************/
MEM_STATIC unsigned BIT_highbit32 (register U32 val)
@@ -170,10 +169,9 @@ MEM_STATIC unsigned BIT_highbit32 (register U32 val)
}
/****************************************************************
/*-**************************************************************
* bitStream encoding
****************************************************************/
MEM_STATIC size_t BIT_initCStream(BIT_CStream_t* bitC, void* startPtr, size_t maxSize)
{
bitC->bitContainer = 0;
@@ -240,10 +238,9 @@ MEM_STATIC size_t BIT_closeCStream(BIT_CStream_t* bitC)
}
/**********************************************************
/*-********************************************************
* bitStream decoding
**********************************************************/
/*!BIT_initDStream
* Initialize a BIT_DStream_t.
* @bitD : a pointer to an already allocated BIT_DStream_t structure
@@ -255,8 +252,7 @@ MEM_STATIC size_t BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, si
{
if (srcSize < 1) { memset(bitD, 0, sizeof(*bitD)); return ERROR(srcSize_wrong); }
if (srcSize >= sizeof(size_t)) /* normal case */
{
if (srcSize >= sizeof(size_t)) { /* normal case */
U32 contain32;
bitD->start = (const char*)srcBuffer;
bitD->ptr = (const char*)srcBuffer + srcSize - sizeof(size_t);
@@ -264,9 +260,7 @@ MEM_STATIC size_t BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, si
contain32 = ((const BYTE*)srcBuffer)[srcSize-1];
if (contain32 == 0) return ERROR(GENERIC); /* endMark not present */
bitD->bitsConsumed = 8 - BIT_highbit32(contain32);
}
else
{
} else {
U32 contain32;
bitD->start = (const char*)srcBuffer;
bitD->ptr = bitD->start;
@@ -342,23 +336,20 @@ MEM_STATIC BIT_DStream_status BIT_reloadDStream(BIT_DStream_t* bitD)
if (bitD->bitsConsumed > (sizeof(bitD->bitContainer)*8)) /* should never happen */
return BIT_DStream_overflow;
if (bitD->ptr >= bitD->start + sizeof(bitD->bitContainer))
{
if (bitD->ptr >= bitD->start + sizeof(bitD->bitContainer)) {
bitD->ptr -= bitD->bitsConsumed >> 3;
bitD->bitsConsumed &= 7;
bitD->bitContainer = MEM_readLEST(bitD->ptr);
return BIT_DStream_unfinished;
}
if (bitD->ptr == bitD->start)
{
if (bitD->ptr == bitD->start) {
if (bitD->bitsConsumed < sizeof(bitD->bitContainer)*8) return BIT_DStream_endOfBuffer;
return BIT_DStream_completed;
}
{
U32 nbBytes = bitD->bitsConsumed >> 3;
BIT_DStream_status result = BIT_DStream_unfinished;
if (bitD->ptr - nbBytes < bitD->start)
{
if (bitD->ptr - nbBytes < bitD->start) {
nbBytes = (U32)(bitD->ptr - bitD->start); /* ptr > start */
result = BIT_DStream_endOfBuffer;
}
lib/error_private.h
+34 -26
View File
@@ -40,14 +40,14 @@ extern "C" {
#endif
/* *****************************************
* Includes
/* ****************************************
* Dependencies
******************************************/
#include <stddef.h> /* size_t, ptrdiff_t */
#include <stddef.h> /* size_t */
#include "error_public.h" /* enum list */
/* *****************************************
/* ****************************************
* Compiler-specific
******************************************/
#if defined(__GNUC__)
@@ -61,44 +61,52 @@ extern "C" {
#endif
/* *****************************************
* Error Codes
/*-****************************************
* Customization
******************************************/
typedef ZSTD_errorCode ERR_enum;
#define PREFIX(name) ZSTD_error_##name
/*-****************************************
* Error codes handling
******************************************/
#ifdef ERROR
# undef ERROR /* reported already defined on VS 2015 by Rich Geldreich */
# undef ERROR /* reported already defined on VS 2015 (Rich Geldreich) */
#endif
#define ERROR(name) (size_t)-PREFIX(name)
ERR_STATIC unsigned ERR_isError(size_t code) { return (code > ERROR(maxCode)); }
ERR_STATIC ERR_enum ERR_getError(size_t code) { if (!ERR_isError(code)) return (ERR_enum)0; return (ERR_enum) (0-code); }
/* *****************************************
/*-****************************************
* Error Strings
******************************************/
ERR_STATIC const char* ERR_getErrorName(size_t code)
{
static const char* codeError = "Unspecified error code";
switch( (size_t)(0-code) )
static const char* notErrorCode = "Unspecified error code";
switch( ERR_getError(code) )
{
case ZSTD_error_No_Error: return "No error detected";
case ZSTD_error_GENERIC: return "Error (generic)";
case ZSTD_error_prefix_unknown: return "Unknown frame descriptor";
case ZSTD_error_frameParameter_unsupported: return "Unsupported frame parameter";
case ZSTD_error_frameParameter_unsupportedBy32bitsImplementation: return "Frame parameter unsupported in 32-bits mode";
case ZSTD_error_init_missing: return "Context should be init first";
case ZSTD_error_memory_allocation: return "Allocation error : not enough memory";
case ZSTD_error_dstSize_tooSmall: return "Destination buffer is too small";
case ZSTD_error_srcSize_wrong: return "Src size incorrect";
case ZSTD_error_corruption_detected: return "Corrupted block detected";
case ZSTD_error_tableLog_tooLarge: return "tableLog requires too much memory";
case ZSTD_error_maxSymbolValue_tooLarge: return "Unsupported max possible Symbol Value : too large";
case ZSTD_error_maxSymbolValue_tooSmall: return "Specified maxSymbolValue is too small";
case ZSTD_error_dictionary_corrupted: return "Dictionary is corrupted";
case ZSTD_error_maxCode:
default: return codeError;
case PREFIX(no_error): return "No error detected";
case PREFIX(GENERIC): return "Error (generic)";
case PREFIX(prefix_unknown): return "Unknown frame descriptor";
case PREFIX(frameParameter_unsupported): return "Unsupported frame parameter";
case PREFIX(frameParameter_unsupportedBy32bits): return "Frame parameter unsupported in 32-bits mode";
case PREFIX(init_missing): return "Context should be init first";
case PREFIX(memory_allocation): return "Allocation error : not enough memory";
case PREFIX(stage_wrong): return "Operation not authorized at current processing stage";
case PREFIX(dstSize_tooSmall): return "Destination buffer is too small";
case PREFIX(srcSize_wrong): return "Src size incorrect";
case PREFIX(corruption_detected): return "Corrupted block detected";
case PREFIX(tableLog_tooLarge): return "tableLog requires too much memory";
case PREFIX(maxSymbolValue_tooLarge): return "Unsupported max possible Symbol Value : too large";
case PREFIX(maxSymbolValue_tooSmall): return "Specified maxSymbolValue is too small";
case PREFIX(dictionary_corrupted): return "Dictionary is corrupted";
case PREFIX(maxCode):
default: return notErrorCode; /* should be impossible, due to ERR_getError() */
}
}
lib/error_public.h
+5 -5
View File
@@ -39,14 +39,14 @@ extern "C" {
/* ****************************************
* error list
* error codes list
******************************************/
enum {
ZSTD_error_No_Error,
typedef enum {
ZSTD_error_no_error,
ZSTD_error_GENERIC,
ZSTD_error_prefix_unknown,
ZSTD_error_frameParameter_unsupported,
ZSTD_error_frameParameter_unsupportedBy32bitsImplementation,
ZSTD_error_frameParameter_unsupportedBy32bits,
ZSTD_error_init_missing,
ZSTD_error_memory_allocation,
ZSTD_error_stage_wrong,
@@ -58,7 +58,7 @@ enum {
ZSTD_error_maxSymbolValue_tooSmall,
ZSTD_error_dictionary_corrupted,
ZSTD_error_maxCode
};
} ZSTD_errorCode;
/* note : functions provide error codes in reverse negative order,
so compare with (size_t)(0-enum) */
lib/fse.h
+10 -10
View File
@@ -170,18 +170,18 @@ void FSE_freeCTable (FSE_CTable* ct);
/*!
FSE_buildCTable():
Builds 'ct', which must be already allocated, using FSE_createCTable()
Builds @ct, which must be already allocated, using FSE_createCTable()
return : 0
or an errorCode, which can be tested using FSE_isError() */
size_t FSE_buildCTable(FSE_CTable* ct, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog);
/*!
FSE_compress_usingCTable():
Compress 'src' using 'ct' into 'dst' which must be already allocated
return : size of compressed data (<= maxDstSize)
or 0 if compressed data could not fit into 'dst'
Compress @src using @ct into @dst which must be already allocated
return : size of compressed data (<= @dstCapacity)
or 0 if compressed data could not fit into @dst
or an errorCode, which can be tested using FSE_isError() */
size_t FSE_compress_usingCTable (void* dst, size_t maxDstSize, const void* src, size_t srcSize, const FSE_CTable* ct);
size_t FSE_compress_usingCTable (void* dst, size_t dstCapacity, const void* src, size_t srcSize, const FSE_CTable* ct);
/*!
Tutorial :
@@ -221,7 +221,7 @@ If there is an error, both functions will return an ErrorCode (which can be test
'CTable' can then be used to compress 'src', with FSE_compress_usingCTable().
Similar to FSE_count(), the convention is that 'src' is assumed to be a table of char of size 'srcSize'
The function returns the size of compressed data (without header), necessarily <= maxDstSize.
The function returns the size of compressed data (without header), necessarily <= @dstCapacity.
If it returns '0', compressed data could not fit into 'dst'.
If there is an error, the function will return an ErrorCode (which can be tested using FSE_isError()).
*/
@@ -253,11 +253,11 @@ size_t FSE_buildDTable (FSE_DTable* dt, const short* normalizedCounter, unsigned
/*!
FSE_decompress_usingDTable():
Decompress compressed source 'cSrc' of size 'cSrcSize' using 'dt'
into 'dst' which must be already allocated.
return : size of regenerated data (necessarily <= maxDstSize)
Decompress compressed source @cSrc of size @cSrcSize using @dt
into @dst which must be already allocated.
return : size of regenerated data (necessarily <= @dstCapacity)
or an errorCode, which can be tested using FSE_isError() */
size_t FSE_decompress_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const FSE_DTable* dt);
size_t FSE_decompress_usingDTable(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, const FSE_DTable* dt);
/*!
Tutorial :
lib/legacy/zstd_legacy.h
+5 -1
View File
@@ -45,6 +45,7 @@ extern "C" {
#include "zstd_v01.h"
#include "zstd_v02.h"
#include "zstd_v03.h"
#include "zstd_v04.h"
MEM_STATIC unsigned ZSTD_isLegacy (U32 magicNumberLE)
{
@@ -52,7 +53,8 @@ MEM_STATIC unsigned ZSTD_isLegacy (U32 magicNumberLE)
{
case ZSTDv01_magicNumberLE :
case ZSTDv02_magicNumber :
case ZSTDv03_magicNumber : return 1;
case ZSTDv03_magicNumber :
case ZSTDv04_magicNumber : return 1;
default : return 0;
}
}
@@ -71,6 +73,8 @@ MEM_STATIC size_t ZSTD_decompressLegacy(
return ZSTDv02_decompress(dst, maxOriginalSize, src, compressedSize);
case ZSTDv03_magicNumber :
return ZSTDv03_decompress(dst, maxOriginalSize, src, compressedSize);
case ZSTDv04_magicNumber :
return ZSTDv04_decompress(dst, maxOriginalSize, src, compressedSize);
default :
return ERROR(prefix_unknown);
}
lib/legacy/zstd_v04.c
+4431
View File
@@ -0,0 +1,4431 @@
/* ******************************************************************
zstd_v04.c
Decompression module for ZSTD v0.4 legacy format
Copyright (C) 2016, Yann Collet.
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above
copyright notice, this list of conditions and the following disclaimer
in the documentation and/or other materials provided with the
distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
You can contact the author at :
- Homepage : http://www.zstd.net/
****************************************************************** */
/*- Dependencies -*/
#include "zstd_v04.h"
/* ******************************************************************
mem.h
low-level memory access routines
Copyright (C) 2013-2015, Yann Collet.
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above
copyright notice, this list of conditions and the following disclaimer
in the documentation and/or other materials provided with the
distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
You can contact the author at :
- FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
- Public forum : https://groups.google.com/forum/#!forum/lz4c
****************************************************************** */
#ifndef MEM_H_MODULE
#define MEM_H_MODULE
#if defined (__cplusplus)
extern "C" {
#endif
/******************************************
* Includes
******************************************/
#include <stddef.h> /* size_t, ptrdiff_t */
#include <string.h> /* memcpy */
/******************************************
* Compiler-specific
******************************************/
#if defined(__GNUC__)
# define MEM_STATIC static __attribute__((unused))
#elif defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */)
# define MEM_STATIC static inline
#elif defined(_MSC_VER)
# define MEM_STATIC static __inline
#else
# define MEM_STATIC static /* this version may generate warnings for unused static functions; disable the relevant warning */
#endif
/****************************************************************
* Basic Types
*****************************************************************/
#if defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */)
# include <stdint.h>
typedef uint8_t BYTE;
typedef uint16_t U16;
typedef int16_t S16;
typedef uint32_t U32;
typedef int32_t S32;
typedef uint64_t U64;
typedef int64_t S64;
#else
typedef unsigned char BYTE;
typedef unsigned short U16;
typedef signed short S16;
typedef unsigned int U32;
typedef signed int S32;
typedef unsigned long long U64;
typedef signed long long S64;
#endif
/****************************************************************
* Memory I/O
*****************************************************************/
/* MEM_FORCE_MEMORY_ACCESS
* By default, access to unaligned memory is controlled by `memcpy()`, which is safe and portable.
* Unfortunately, on some target/compiler combinations, the generated assembly is sub-optimal.
* The below switch allow to select different access method for improved performance.
* Method 0 (default) : use `memcpy()`. Safe and portable.
* Method 1 : `__packed` statement. It depends on compiler extension (ie, not portable).
* This method is safe if your compiler supports it, and *generally* as fast or faster than `memcpy`.
* Method 2 : direct access. This method is portable but violate C standard.
* It can generate buggy code on targets generating assembly depending on alignment.
* But in some circumstances, it's the only known way to get the most performance (ie GCC + ARMv6)
* See http://fastcompression.blogspot.fr/2015/08/accessing-unaligned-memory.html for details.
* Prefer these methods in priority order (0 > 1 > 2)
*/
#ifndef MEM_FORCE_MEMORY_ACCESS /* can be defined externally, on command line for example */
# if defined(__GNUC__) && ( defined(__ARM_ARCH_6__) || defined(__ARM_ARCH_6J__) || defined(__ARM_ARCH_6K__) || defined(__ARM_ARCH_6Z__) || defined(__ARM_ARCH_6ZK__) || defined(__ARM_ARCH_6T2__) )
# define MEM_FORCE_MEMORY_ACCESS 2
# elif defined(__INTEL_COMPILER) || \
(defined(__GNUC__) && ( defined(__ARM_ARCH_7__) || defined(__ARM_ARCH_7A__) || defined(__ARM_ARCH_7R__) || defined(__ARM_ARCH_7M__) || defined(__ARM_ARCH_7S__) ))
# define MEM_FORCE_MEMORY_ACCESS 1
# endif
#endif
MEM_STATIC unsigned MEM_32bits(void) { return sizeof(void*)==4; }
MEM_STATIC unsigned MEM_64bits(void) { return sizeof(void*)==8; }
MEM_STATIC unsigned MEM_isLittleEndian(void)
{
const union { U32 u; BYTE c[4]; } one = { 1 }; /* don't use static : performance detrimental */
return one.c[0];
}
#if defined(MEM_FORCE_MEMORY_ACCESS) && (MEM_FORCE_MEMORY_ACCESS==2)
/* violates C standard on structure alignment.
Only use if no other choice to achieve best performance on target platform */
MEM_STATIC U16 MEM_read16(const void* memPtr) { return *(const U16*) memPtr; }
MEM_STATIC U32 MEM_read32(const void* memPtr) { return *(const U32*) memPtr; }
MEM_STATIC U64 MEM_read64(const void* memPtr) { return *(const U64*) memPtr; }
MEM_STATIC void MEM_write16(void* memPtr, U16 value) { *(U16*)memPtr = value; }
MEM_STATIC void MEM_write32(void* memPtr, U32 value) { *(U32*)memPtr = value; }
MEM_STATIC void MEM_write64(void* memPtr, U64 value) { *(U64*)memPtr = value; }
#elif defined(MEM_FORCE_MEMORY_ACCESS) && (MEM_FORCE_MEMORY_ACCESS==1)
/* __pack instructions are safer, but compiler specific, hence potentially problematic for some compilers */
/* currently only defined for gcc and icc */
typedef union { U16 u16; U32 u32; U64 u64; } __attribute__((packed)) unalign;
MEM_STATIC U16 MEM_read16(const void* ptr) { return ((const unalign*)ptr)->u16; }
MEM_STATIC U32 MEM_read32(const void* ptr) { return ((const unalign*)ptr)->u32; }
MEM_STATIC U64 MEM_read64(const void* ptr) { return ((const unalign*)ptr)->u64; }
MEM_STATIC void MEM_write16(void* memPtr, U16 value) { ((unalign*)memPtr)->u16 = value; }
MEM_STATIC void MEM_write32(void* memPtr, U32 value) { ((unalign*)memPtr)->u32 = value; }
MEM_STATIC void MEM_write64(void* memPtr, U64 value) { ((unalign*)memPtr)->u64 = value; }
#else
/* default method, safe and standard.
can sometimes prove slower */
MEM_STATIC U16 MEM_read16(const void* memPtr)
{
U16 val; memcpy(&val, memPtr, sizeof(val)); return val;
}
MEM_STATIC U32 MEM_read32(const void* memPtr)
{
U32 val; memcpy(&val, memPtr, sizeof(val)); return val;
}
MEM_STATIC U64 MEM_read64(const void* memPtr)
{
U64 val; memcpy(&val, memPtr, sizeof(val)); return val;
}
MEM_STATIC void MEM_write16(void* memPtr, U16 value)
{
memcpy(memPtr, &value, sizeof(value));
}
MEM_STATIC void MEM_write32(void* memPtr, U32 value)
{
memcpy(memPtr, &value, sizeof(value));
}
MEM_STATIC void MEM_write64(void* memPtr, U64 value)
{
memcpy(memPtr, &value, sizeof(value));
}
#endif // MEM_FORCE_MEMORY_ACCESS
MEM_STATIC U16 MEM_readLE16(const void* memPtr)
{
if (MEM_isLittleEndian())
return MEM_read16(memPtr);
else
{
const BYTE* p = (const BYTE*)memPtr;
return (U16)(p[0] + (p[1]<<8));
}
}
MEM_STATIC void MEM_writeLE16(void* memPtr, U16 val)
{
if (MEM_isLittleEndian())
{
MEM_write16(memPtr, val);
}
else
{
BYTE* p = (BYTE*)memPtr;
p[0] = (BYTE)val;
p[1] = (BYTE)(val>>8);
}
}
MEM_STATIC U32 MEM_readLE32(const void* memPtr)
{
if (MEM_isLittleEndian())
return MEM_read32(memPtr);
else
{
const BYTE* p = (const BYTE*)memPtr;
return (U32)((U32)p[0] + ((U32)p[1]<<8) + ((U32)p[2]<<16) + ((U32)p[3]<<24));
}
}
MEM_STATIC void MEM_writeLE32(void* memPtr, U32 val32)
{
if (MEM_isLittleEndian())
{
MEM_write32(memPtr, val32);
}
else
{
BYTE* p = (BYTE*)memPtr;
p[0] = (BYTE)val32;
p[1] = (BYTE)(val32>>8);
p[2] = (BYTE)(val32>>16);
p[3] = (BYTE)(val32>>24);
}
}
MEM_STATIC U64 MEM_readLE64(const void* memPtr)
{
if (MEM_isLittleEndian())
return MEM_read64(memPtr);
else
{
const BYTE* p = (const BYTE*)memPtr;
return (U64)((U64)p[0] + ((U64)p[1]<<8) + ((U64)p[2]<<16) + ((U64)p[3]<<24)
+ ((U64)p[4]<<32) + ((U64)p[5]<<40) + ((U64)p[6]<<48) + ((U64)p[7]<<56));
}
}
MEM_STATIC void MEM_writeLE64(void* memPtr, U64 val64)
{
if (MEM_isLittleEndian())
{
MEM_write64(memPtr, val64);
}
else
{
BYTE* p = (BYTE*)memPtr;
p[0] = (BYTE)val64;
p[1] = (BYTE)(val64>>8);
p[2] = (BYTE)(val64>>16);
p[3] = (BYTE)(val64>>24);
p[4] = (BYTE)(val64>>32);
p[5] = (BYTE)(val64>>40);
p[6] = (BYTE)(val64>>48);
p[7] = (BYTE)(val64>>56);
}
}
MEM_STATIC size_t MEM_readLEST(const void* memPtr)
{
if (MEM_32bits())
return (size_t)MEM_readLE32(memPtr);
else
return (size_t)MEM_readLE64(memPtr);
}
MEM_STATIC void MEM_writeLEST(void* memPtr, size_t val)
{
if (MEM_32bits())
MEM_writeLE32(memPtr, (U32)val);
else
MEM_writeLE64(memPtr, (U64)val);
}
#if defined (__cplusplus)
}
#endif
#endif /* MEM_H_MODULE */
/* ******************************************************************
Error codes list
Copyright (C) 2016, Yann Collet
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above
copyright notice, this list of conditions and the following disclaimer
in the documentation and/or other materials provided with the
distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
You can contact the author at :
- Source repository : https://github.com/Cyan4973/zstd
****************************************************************** */
#ifndef ERROR_PUBLIC_H_MODULE
#define ERROR_PUBLIC_H_MODULE
#if defined (__cplusplus)
extern "C" {
#endif
/* ****************************************
* error list
******************************************/
enum {
ZSTD_error_No_Error,
ZSTD_error_GENERIC,
ZSTD_error_prefix_unknown,
ZSTD_error_frameParameter_unsupported,
ZSTD_error_frameParameter_unsupportedBy32bitsImplementation,
ZSTD_error_init_missing,
ZSTD_error_memory_allocation,
ZSTD_error_stage_wrong,
ZSTD_error_dstSize_tooSmall,
ZSTD_error_srcSize_wrong,
ZSTD_error_corruption_detected,
ZSTD_error_tableLog_tooLarge,
ZSTD_error_maxSymbolValue_tooLarge,
ZSTD_error_maxSymbolValue_tooSmall,
ZSTD_error_maxCode
};
/* note : functions provide error codes in reverse negative order,
so compare with (size_t)(0-enum) */
#if defined (__cplusplus)
}
#endif
#endif /* ERROR_PUBLIC_H_MODULE */
/*
zstd - standard compression library
Header File for static linking only
Copyright (C) 2014-2015, Yann Collet.
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above
copyright notice, this list of conditions and the following disclaimer
in the documentation and/or other materials provided with the
distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
You can contact the author at :
- zstd source repository : https://github.com/Cyan4973/zstd
- ztsd public forum : https://groups.google.com/forum/#!forum/lz4c
*/
#ifndef ZSTD_STATIC_H
#define ZSTD_STATIC_H
/* The objects defined into this file shall be considered experimental.
* They are not considered stable, as their prototype may change in the future.
* You can use them for tests, provide feedback, or if you can endure risks of future changes.
*/
#if defined (__cplusplus)
extern "C" {
#endif
/* *************************************
* Types
***************************************/
#define ZSTD_WINDOWLOG_MAX 26
#define ZSTD_WINDOWLOG_MIN 18
#define ZSTD_WINDOWLOG_ABSOLUTEMIN 11
#define ZSTD_CONTENTLOG_MAX (ZSTD_WINDOWLOG_MAX+1)
#define ZSTD_CONTENTLOG_MIN 4
#define ZSTD_HASHLOG_MAX 28
#define ZSTD_HASHLOG_MIN 4
#define ZSTD_SEARCHLOG_MAX (ZSTD_CONTENTLOG_MAX-1)
#define ZSTD_SEARCHLOG_MIN 1
#define ZSTD_SEARCHLENGTH_MAX 7
#define ZSTD_SEARCHLENGTH_MIN 4
/** from faster to stronger */
typedef enum { ZSTD_fast, ZSTD_greedy, ZSTD_lazy, ZSTD_lazy2, ZSTD_btlazy2 } ZSTD_strategy;
typedef struct
{
U64 srcSize; /* optional : tells how much bytes are present in the frame. Use 0 if not known. */
U32 windowLog; /* largest match distance : larger == more compression, more memory needed during decompression */
U32 contentLog; /* full search segment : larger == more compression, slower, more memory (useless for fast) */
U32 hashLog; /* dispatch table : larger == more memory, faster */
U32 searchLog; /* nb of searches : larger == more compression, slower */
U32 searchLength; /* size of matches : larger == faster decompression, sometimes less compression */
ZSTD_strategy strategy;
} ZSTD_parameters;
typedef ZSTDv04_Dctx ZSTD_DCtx;
/* *************************************
* Advanced functions
***************************************/
/** ZSTD_decompress_usingDict
* Same as ZSTD_decompressDCtx, using a Dictionary content as prefix
* Note : dict can be NULL, in which case, it's equivalent to ZSTD_decompressDCtx() */
static size_t ZSTD_decompress_usingDict(ZSTD_DCtx* ctx,
void* dst, size_t maxDstSize,
const void* src, size_t srcSize,
const void* dict,size_t dictSize);
/* **************************************
* Streaming functions (direct mode)
****************************************/
static size_t ZSTD_resetDCtx(ZSTD_DCtx* dctx);
static size_t ZSTD_getFrameParams(ZSTD_parameters* params, const void* src, size_t srcSize);
static void ZSTD_decompress_insertDictionary(ZSTD_DCtx* ctx, const void* src, size_t srcSize);
static size_t ZSTD_nextSrcSizeToDecompress(ZSTD_DCtx* dctx);
static size_t ZSTD_decompressContinue(ZSTD_DCtx* dctx, void* dst, size_t maxDstSize, const void* src, size_t srcSize);
/**
Streaming decompression, bufferless mode
A ZSTD_DCtx object is required to track streaming operations.
Use ZSTD_createDCtx() / ZSTD_freeDCtx() to manage it.
A ZSTD_DCtx object can be re-used multiple times. Use ZSTD_resetDCtx() to return to fresh status.
First operation is to retrieve frame parameters, using ZSTD_getFrameParams().
This function doesn't consume its input. It needs enough input data to properly decode the frame header.
Objective is to retrieve *params.windowlog, to know minimum amount of memory required during decoding.
Result : 0 when successful, it means the ZSTD_parameters structure has been filled.
>0 : means there is not enough data into src. Provides the expected size to successfully decode header.
errorCode, which can be tested using ZSTD_isError() (For example, if it's not a ZSTD header)
Then, you can optionally insert a dictionary.
This operation must mimic the compressor behavior, otherwise decompression will fail or be corrupted.
Then it's possible to start decompression.
Use ZSTD_nextSrcSizeToDecompress() and ZSTD_decompressContinue() alternatively.
ZSTD_nextSrcSizeToDecompress() tells how much bytes to provide as 'srcSize' to ZSTD_decompressContinue().
ZSTD_decompressContinue() requires this exact amount of bytes, or it will fail.
ZSTD_decompressContinue() needs previous data blocks during decompression, up to (1 << windowlog).
They should preferably be located contiguously, prior to current block. Alternatively, a round buffer is also possible.
@result of ZSTD_decompressContinue() is the number of bytes regenerated within 'dst'.
It can be zero, which is not an error; it just means ZSTD_decompressContinue() has decoded some header.
A frame is fully decoded when ZSTD_nextSrcSizeToDecompress() returns zero.
Context can then be reset to start a new decompression.
*/
#if defined (__cplusplus)
}
#endif
/* ******************************************************************
Error codes and messages
Copyright (C) 2013-2016, Yann Collet
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above
copyright notice, this list of conditions and the following disclaimer
in the documentation and/or other materials provided with the
distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
You can contact the author at :
- Source repository : https://github.com/Cyan4973/zstd
****************************************************************** */
/* Note : this module is expected to remain private, do not expose it */
#ifndef ERROR_H_MODULE
#define ERROR_H_MODULE
#if defined (__cplusplus)
extern "C" {
#endif
/* *****************************************
* Includes
******************************************/
#include <stddef.h> /* size_t, ptrdiff_t */
/* *****************************************
* Compiler-specific
******************************************/
#if defined(__GNUC__)
# define ERR_STATIC static __attribute__((unused))
#elif defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */)
# define ERR_STATIC static inline
#elif defined(_MSC_VER)
# define ERR_STATIC static __inline
#else
# define ERR_STATIC static /* this version may generate warnings for unused static functions; disable the relevant warning */
#endif
/* *****************************************
* Error Codes
******************************************/
#define PREFIX(name) ZSTD_error_##name
#ifdef ERROR
# undef ERROR /* reported already defined on VS 2015 by Rich Geldreich */
#endif
#define ERROR(name) (size_t)-PREFIX(name)
ERR_STATIC unsigned ERR_isError(size_t code) { return (code > ERROR(maxCode)); }
/* *****************************************
* Error Strings
******************************************/
ERR_STATIC const char* ERR_getErrorName(size_t code)
{
static const char* codeError = "Unspecified error code";
switch( (size_t)(0-code) )
{
case ZSTD_error_No_Error: return "No error detected";
case ZSTD_error_GENERIC: return "Error (generic)";
case ZSTD_error_prefix_unknown: return "Unknown frame descriptor";
case ZSTD_error_frameParameter_unsupported: return "Unsupported frame parameter";
case ZSTD_error_frameParameter_unsupportedBy32bitsImplementation: return "Frame parameter unsupported in 32-bits mode";
case ZSTD_error_init_missing: return "Context should be init first";
case ZSTD_error_memory_allocation: return "Allocation error : not enough memory";
case ZSTD_error_dstSize_tooSmall: return "Destination buffer is too small";
case ZSTD_error_srcSize_wrong: return "Src size incorrect";
case ZSTD_error_corruption_detected: return "Corrupted block detected";
case ZSTD_error_tableLog_tooLarge: return "tableLog requires too much memory";
case ZSTD_error_maxSymbolValue_tooLarge: return "Unsupported max possible Symbol Value : too large";
case ZSTD_error_maxSymbolValue_tooSmall: return "Specified maxSymbolValue is too small";
case ZSTD_error_maxCode:
default: return codeError;
}
}
#if defined (__cplusplus)
}
#endif
#endif /* ERROR_H_MODULE */
#endif /* ZSTD_STATIC_H */
/*
zstd_internal - common functions to include
Header File for include
Copyright (C) 2014-2015, Yann Collet.
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above
copyright notice, this list of conditions and the following disclaimer
in the documentation and/or other materials provided with the
distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
You can contact the author at :
- zstd source repository : https://github.com/Cyan4973/zstd
- ztsd public forum : https://groups.google.com/forum/#!forum/lz4c
*/
#ifndef ZSTD_CCOMMON_H_MODULE
#define ZSTD_CCOMMON_H_MODULE
#if defined (__cplusplus)
extern "C" {
#endif
/* *************************************
* Common macros
***************************************/
#define MIN(a,b) ((a)<(b) ? (a) : (b))
#define MAX(a,b) ((a)>(b) ? (a) : (b))
/* *************************************
* Common constants
***************************************/
#define ZSTD_MAGICNUMBER 0xFD2FB524 /* v0.4 */
#define KB *(1 <<10)
#define MB *(1 <<20)
#define GB *(1U<<30)
#define BLOCKSIZE (128 KB) /* define, for static allocation */
static const size_t ZSTD_blockHeaderSize = 3;
static const size_t ZSTD_frameHeaderSize_min = 5;
#define ZSTD_frameHeaderSize_max 5 /* define, for static allocation */
#define BIT7 128
#define BIT6 64
#define BIT5 32
#define BIT4 16
#define BIT1 2
#define BIT0 1
#define IS_RAW BIT0
#define IS_RLE BIT1
#define MINMATCH 4
#define REPCODE_STARTVALUE 4
#define MLbits 7
#define LLbits 6
#define Offbits 5
#define MaxML ((1<<MLbits) - 1)
#define MaxLL ((1<<LLbits) - 1)
#define MaxOff ((1<<Offbits)- 1)
#define MLFSELog 10
#define LLFSELog 10
#define OffFSELog 9
#define MaxSeq MAX(MaxLL, MaxML)
#define MIN_SEQUENCES_SIZE (2 /*seqNb*/ + 2 /*dumps*/ + 3 /*seqTables*/ + 1 /*bitStream*/)
#define MIN_CBLOCK_SIZE (3 /*litCSize*/ + MIN_SEQUENCES_SIZE)
typedef enum { bt_compressed, bt_raw, bt_rle, bt_end } blockType_t;
/* ******************************************
* Shared functions to include for inlining
********************************************/
static void ZSTD_copy8(void* dst, const void* src) { memcpy(dst, src, 8); }
#define COPY8(d,s) { ZSTD_copy8(d,s); d+=8; s+=8; }
/*! ZSTD_wildcopy : custom version of memcpy(), can copy up to 7-8 bytes too many */
static void ZSTD_wildcopy(void* dst, const void* src, size_t length)
{
const BYTE* ip = (const BYTE*)src;
BYTE* op = (BYTE*)dst;
BYTE* const oend = op + length;
do
COPY8(op, ip)
while (op < oend);
}
#if defined (__cplusplus)
}
#endif
/* ******************************************************************
FSE : Finite State Entropy coder
header file
Copyright (C) 2013-2015, Yann Collet.
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above
copyright notice, this list of conditions and the following disclaimer
in the documentation and/or other materials provided with the
distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
You can contact the author at :
- Source repository : https://github.com/Cyan4973/FiniteStateEntropy
- Public forum : https://groups.google.com/forum/#!forum/lz4c
****************************************************************** */
#ifndef FSE_H
#define FSE_H
#if defined (__cplusplus)
extern "C" {
#endif
/* *****************************************
* Includes
******************************************/
#include <stddef.h> /* size_t, ptrdiff_t */
/* *****************************************
* FSE simple functions
******************************************/
static size_t FSE_decompress(void* dst, size_t maxDstSize,
const void* cSrc, size_t cSrcSize);
/*!
FSE_decompress():
Decompress FSE data from buffer 'cSrc', of size 'cSrcSize',
into already allocated destination buffer 'dst', of size 'maxDstSize'.
return : size of regenerated data (<= maxDstSize)
or an error code, which can be tested using FSE_isError()
** Important ** : FSE_decompress() doesn't decompress non-compressible nor RLE data !!!
Why ? : making this distinction requires a header.
Header management is intentionally delegated to the user layer, which can better manage special cases.
*/
/* *****************************************
* Tool functions
******************************************/
/* Error Management */
static unsigned FSE_isError(size_t code); /* tells if a return value is an error code */
/* *****************************************
* FSE detailed API
******************************************/
/*!
FSE_compress() does the following:
1. count symbol occurrence from source[] into table count[]
2. normalize counters so that sum(count[]) == Power_of_2 (2^tableLog)
3. save normalized counters to memory buffer using writeNCount()
4. build encoding table 'CTable' from normalized counters
5. encode the data stream using encoding table 'CTable'
FSE_decompress() does the following:
1. read normalized counters with readNCount()
2. build decoding table 'DTable' from normalized counters
3. decode the data stream using decoding table 'DTable'
The following API allows targeting specific sub-functions for advanced tasks.
For example, it's possible to compress several blocks using the same 'CTable',
or to save and provide normalized distribution using external method.
*/
/* *** DECOMPRESSION *** */
/*!
FSE_readNCount():
Read compactly saved 'normalizedCounter' from 'rBuffer'.
return : size read from 'rBuffer'
or an errorCode, which can be tested using FSE_isError()
maxSymbolValuePtr[0] and tableLogPtr[0] will also be updated with their respective values */
static size_t FSE_readNCount (short* normalizedCounter, unsigned* maxSymbolValuePtr, unsigned* tableLogPtr, const void* rBuffer, size_t rBuffSize);
/*!
Constructor and Destructor of type FSE_DTable
Note that its size depends on 'tableLog' */
typedef unsigned FSE_DTable; /* don't allocate that. It's just a way to be more restrictive than void* */
/*!
FSE_buildDTable():
Builds 'dt', which must be already allocated, using FSE_createDTable()
return : 0,
or an errorCode, which can be tested using FSE_isError() */
static size_t FSE_buildDTable ( FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog);
/*!
FSE_decompress_usingDTable():
Decompress compressed source 'cSrc' of size 'cSrcSize' using 'dt'
into 'dst' which must be already allocated.
return : size of regenerated data (necessarily <= maxDstSize)
or an errorCode, which can be tested using FSE_isError() */
static size_t FSE_decompress_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const FSE_DTable* dt);
/*!
Tutorial :
----------
(Note : these functions only decompress FSE-compressed blocks.
If block is uncompressed, use memcpy() instead
If block is a single repeated byte, use memset() instead )
The first step is to obtain the normalized frequencies of symbols.
This can be performed by FSE_readNCount() if it was saved using FSE_writeNCount().
'normalizedCounter' must be already allocated, and have at least 'maxSymbolValuePtr[0]+1' cells of signed short.
In practice, that means it's necessary to know 'maxSymbolValue' beforehand,
or size the table to handle worst case situations (typically 256).
FSE_readNCount() will provide 'tableLog' and 'maxSymbolValue'.
The result of FSE_readNCount() is the number of bytes read from 'rBuffer'.
Note that 'rBufferSize' must be at least 4 bytes, even if useful information is less than that.
If there is an error, the function will return an error code, which can be tested using FSE_isError().
The next step is to build the decompression tables 'FSE_DTable' from 'normalizedCounter'.
This is performed by the function FSE_buildDTable().
The space required by 'FSE_DTable' must be already allocated using FSE_createDTable().
If there is an error, the function will return an error code, which can be tested using FSE_isError().
'FSE_DTable' can then be used to decompress 'cSrc', with FSE_decompress_usingDTable().
'cSrcSize' must be strictly correct, otherwise decompression will fail.
FSE_decompress_usingDTable() result will tell how many bytes were regenerated (<=maxDstSize).
If there is an error, the function will return an error code, which can be tested using FSE_isError(). (ex: dst buffer too small)
*/
#if defined (__cplusplus)
}
#endif
#endif /* FSE_H */
/* ******************************************************************
bitstream
Part of NewGen Entropy library
header file (to include)
Copyright (C) 2013-2015, Yann Collet.
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above
copyright notice, this list of conditions and the following disclaimer
in the documentation and/or other materials provided with the
distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
You can contact the author at :
- Source repository : https://github.com/Cyan4973/FiniteStateEntropy
- Public forum : https://groups.google.com/forum/#!forum/lz4c
****************************************************************** */
#ifndef BITSTREAM_H_MODULE
#define BITSTREAM_H_MODULE
#if defined (__cplusplus)
extern "C" {
#endif
/*
* This API consists of small unitary functions, which highly benefit from being inlined.
* Since link-time-optimization is not available for all compilers,
* these functions are defined into a .h to be included.
*/
/**********************************************
* bitStream decompression API (read backward)
**********************************************/
typedef struct
{
size_t bitContainer;
unsigned bitsConsumed;
const char* ptr;
const char* start;
} BIT_DStream_t;
typedef enum { BIT_DStream_unfinished = 0,
BIT_DStream_endOfBuffer = 1,
BIT_DStream_completed = 2,
BIT_DStream_overflow = 3 } BIT_DStream_status; /* result of BIT_reloadDStream() */
/* 1,2,4,8 would be better for bitmap combinations, but slows down performance a bit ... :( */
MEM_STATIC size_t BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, size_t srcSize);
MEM_STATIC size_t BIT_readBits(BIT_DStream_t* bitD, unsigned nbBits);
MEM_STATIC BIT_DStream_status BIT_reloadDStream(BIT_DStream_t* bitD);
MEM_STATIC unsigned BIT_endOfDStream(const BIT_DStream_t* bitD);
/*
* Start by invoking BIT_initDStream().
* A chunk of the bitStream is then stored into a local register.
* Local register size is 64-bits on 64-bits systems, 32-bits on 32-bits systems (size_t).
* You can then retrieve bitFields stored into the local register, **in reverse order**.
* Local register is manually filled from memory by the BIT_reloadDStream() method.
* A reload guarantee a minimum of ((8*sizeof(size_t))-7) bits when its result is BIT_DStream_unfinished.
* Otherwise, it can be less than that, so proceed accordingly.
* Checking if DStream has reached its end can be performed with BIT_endOfDStream()
*/
/******************************************
* unsafe API
******************************************/
MEM_STATIC size_t BIT_readBitsFast(BIT_DStream_t* bitD, unsigned nbBits);
/* faster, but works only if nbBits >= 1 */
/****************************************************************
* Helper functions
****************************************************************/
MEM_STATIC unsigned BIT_highbit32 (register U32 val)
{
# if defined(_MSC_VER) /* Visual */
unsigned long r=0;
_BitScanReverse ( &r, val );
return (unsigned) r;
# elif defined(__GNUC__) && (__GNUC__ >= 3) /* Use GCC Intrinsic */
return 31 - __builtin_clz (val);
# else /* Software version */
static const unsigned DeBruijnClz[32] = { 0, 9, 1, 10, 13, 21, 2, 29, 11, 14, 16, 18, 22, 25, 3, 30, 8, 12, 20, 28, 15, 17, 24, 7, 19, 27, 23, 6, 26, 5, 4, 31 };
U32 v = val;
unsigned r;
v |= v >> 1;
v |= v >> 2;
v |= v >> 4;
v |= v >> 8;
v |= v >> 16;
r = DeBruijnClz[ (U32) (v * 0x07C4ACDDU) >> 27];
return r;
# endif
}
/**********************************************************
* bitStream decoding
**********************************************************/
/*!BIT_initDStream
* Initialize a BIT_DStream_t.
* @bitD : a pointer to an already allocated BIT_DStream_t structure
* @srcBuffer must point at the beginning of a bitStream
* @srcSize must be the exact size of the bitStream
* @result : size of stream (== srcSize) or an errorCode if a problem is detected
*/
MEM_STATIC size_t BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, size_t srcSize)
{
if (srcSize < 1) { memset(bitD, 0, sizeof(*bitD)); return ERROR(srcSize_wrong); }
if (srcSize >= sizeof(size_t)) /* normal case */
{
U32 contain32;
bitD->start = (const char*)srcBuffer;
bitD->ptr = (const char*)srcBuffer + srcSize - sizeof(size_t);
bitD->bitContainer = MEM_readLEST(bitD->ptr);
contain32 = ((const BYTE*)srcBuffer)[srcSize-1];
if (contain32 == 0) return ERROR(GENERIC); /* endMark not present */
bitD->bitsConsumed = 8 - BIT_highbit32(contain32);
}
else
{
U32 contain32;
bitD->start = (const char*)srcBuffer;
bitD->ptr = bitD->start;
bitD->bitContainer = *(const BYTE*)(bitD->start);
switch(srcSize)
{
case 7: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[6]) << (sizeof(size_t)*8 - 16);
case 6: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[5]) << (sizeof(size_t)*8 - 24);
case 5: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[4]) << (sizeof(size_t)*8 - 32);
case 4: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[3]) << 24;
case 3: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[2]) << 16;
case 2: bitD->bitContainer += (size_t)(((const BYTE*)(bitD->start))[1]) << 8;
default:;
}
contain32 = ((const BYTE*)srcBuffer)[srcSize-1];
if (contain32 == 0) return ERROR(GENERIC); /* endMark not present */
bitD->bitsConsumed = 8 - BIT_highbit32(contain32);
bitD->bitsConsumed += (U32)(sizeof(size_t) - srcSize)*8;
}
return srcSize;
}
/*!BIT_lookBits
* Provides next n bits from local register
* local register is not modified (bits are still present for next read/look)
* On 32-bits, maxNbBits==25
* On 64-bits, maxNbBits==57
* @return : value extracted
*/
MEM_STATIC size_t BIT_lookBits(BIT_DStream_t* bitD, U32 nbBits)
{
const U32 bitMask = sizeof(bitD->bitContainer)*8 - 1;
return ((bitD->bitContainer << (bitD->bitsConsumed & bitMask)) >> 1) >> ((bitMask-nbBits) & bitMask);
}
/*! BIT_lookBitsFast :
* unsafe version; only works only if nbBits >= 1 */
MEM_STATIC size_t BIT_lookBitsFast(BIT_DStream_t* bitD, U32 nbBits)
{
const U32 bitMask = sizeof(bitD->bitContainer)*8 - 1;
return (bitD->bitContainer << (bitD->bitsConsumed & bitMask)) >> (((bitMask+1)-nbBits) & bitMask);
}
MEM_STATIC void BIT_skipBits(BIT_DStream_t* bitD, U32 nbBits)
{
bitD->bitsConsumed += nbBits;
}
/*!BIT_readBits
* Read next n bits from local register.
* pay attention to not read more than nbBits contained into local register.
* @return : extracted value.
*/
MEM_STATIC size_t BIT_readBits(BIT_DStream_t* bitD, U32 nbBits)
{
size_t value = BIT_lookBits(bitD, nbBits);
BIT_skipBits(bitD, nbBits);
return value;
}
/*!BIT_readBitsFast :
* unsafe version; only works only if nbBits >= 1 */
MEM_STATIC size_t BIT_readBitsFast(BIT_DStream_t* bitD, U32 nbBits)
{
size_t value = BIT_lookBitsFast(bitD, nbBits);
BIT_skipBits(bitD, nbBits);
return value;
}
MEM_STATIC BIT_DStream_status BIT_reloadDStream(BIT_DStream_t* bitD)
{
if (bitD->bitsConsumed > (sizeof(bitD->bitContainer)*8)) /* should never happen */
return BIT_DStream_overflow;
if (bitD->ptr >= bitD->start + sizeof(bitD->bitContainer))
{
bitD->ptr -= bitD->bitsConsumed >> 3;
bitD->bitsConsumed &= 7;
bitD->bitContainer = MEM_readLEST(bitD->ptr);
return BIT_DStream_unfinished;
}
if (bitD->ptr == bitD->start)
{
if (bitD->bitsConsumed < sizeof(bitD->bitContainer)*8) return BIT_DStream_endOfBuffer;
return BIT_DStream_completed;
}
{
U32 nbBytes = bitD->bitsConsumed >> 3;
BIT_DStream_status result = BIT_DStream_unfinished;
if (bitD->ptr - nbBytes < bitD->start)
{
nbBytes = (U32)(bitD->ptr - bitD->start); /* ptr > start */
result = BIT_DStream_endOfBuffer;
}
bitD->ptr -= nbBytes;
bitD->bitsConsumed -= nbBytes*8;
bitD->bitContainer = MEM_readLEST(bitD->ptr); /* reminder : srcSize > sizeof(bitD) */
return result;
}
}
/*! BIT_endOfDStream
* @return Tells if DStream has reached its exact end
*/
MEM_STATIC unsigned BIT_endOfDStream(const BIT_DStream_t* DStream)
{
return ((DStream->ptr == DStream->start) && (DStream->bitsConsumed == sizeof(DStream->bitContainer)*8));
}
#if defined (__cplusplus)
}
#endif
#endif /* BITSTREAM_H_MODULE */
/* ******************************************************************
FSE : Finite State Entropy coder
header file for static linking (only)
Copyright (C) 2013-2015, Yann Collet
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above
copyright notice, this list of conditions and the following disclaimer
in the documentation and/or other materials provided with the
distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
You can contact the author at :
- Source repository : https://github.com/Cyan4973/FiniteStateEntropy
- Public forum : https://groups.google.com/forum/#!forum/lz4c
****************************************************************** */
#ifndef FSE_STATIC_H
#define FSE_STATIC_H
#if defined (__cplusplus)
extern "C" {
#endif
/* *****************************************
* Static allocation
*******************************************/
/* FSE buffer bounds */
#define FSE_NCOUNTBOUND 512
#define FSE_BLOCKBOUND(size) (size + (size>>7))
#define FSE_COMPRESSBOUND(size) (FSE_NCOUNTBOUND + FSE_BLOCKBOUND(size)) /* Macro version, useful for static allocation */
/* It is possible to statically allocate FSE CTable/DTable as a table of unsigned using below macros */
#define FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue) (1 + (1<<(maxTableLog-1)) + ((maxSymbolValue+1)*2))
#define FSE_DTABLE_SIZE_U32(maxTableLog) (1 + (1<<maxTableLog))
/* *****************************************
* FSE advanced API
*******************************************/
static size_t FSE_buildDTable_raw (FSE_DTable* dt, unsigned nbBits);
/* build a fake FSE_DTable, designed to read an uncompressed bitstream where each symbol uses nbBits */
static size_t FSE_buildDTable_rle (FSE_DTable* dt, unsigned char symbolValue);
/* build a fake FSE_DTable, designed to always generate the same symbolValue */
/* *****************************************
* FSE symbol decompression API
*******************************************/
typedef struct
{
size_t state;
const void* table; /* precise table may vary, depending on U16 */
} FSE_DState_t;
static void FSE_initDState(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD, const FSE_DTable* dt);
static unsigned char FSE_decodeSymbol(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD);
static unsigned FSE_endOfDState(const FSE_DState_t* DStatePtr);
/*!
Let's now decompose FSE_decompress_usingDTable() into its unitary components.
You will decode FSE-encoded symbols from the bitStream,
and also any other bitFields you put in, **in reverse order**.
You will need a few variables to track your bitStream. They are :
BIT_DStream_t DStream; // Stream context
FSE_DState_t DState; // State context. Multiple ones are possible
FSE_DTable* DTablePtr; // Decoding table, provided by FSE_buildDTable()
The first thing to do is to init the bitStream.
errorCode = BIT_initDStream(&DStream, srcBuffer, srcSize);
You should then retrieve your initial state(s)
(in reverse flushing order if you have several ones) :
errorCode = FSE_initDState(&DState, &DStream, DTablePtr);
You can then decode your data, symbol after symbol.
For information the maximum number of bits read by FSE_decodeSymbol() is 'tableLog'.
Keep in mind that symbols are decoded in reverse order, like a LIFO stack (last in, first out).
unsigned char symbol = FSE_decodeSymbol(&DState, &DStream);
You can retrieve any bitfield you eventually stored into the bitStream (in reverse order)
Note : maximum allowed nbBits is 25, for 32-bits compatibility
size_t bitField = BIT_readBits(&DStream, nbBits);
All above operations only read from local register (which size depends on size_t).
Refueling the register from memory is manually performed by the reload method.
endSignal = FSE_reloadDStream(&DStream);
BIT_reloadDStream() result tells if there is still some more data to read from DStream.
BIT_DStream_unfinished : there is still some data left into the DStream.
BIT_DStream_endOfBuffer : Dstream reached end of buffer. Its container may no longer be completely filled.
BIT_DStream_completed : Dstream reached its exact end, corresponding in general to decompression completed.
BIT_DStream_tooFar : Dstream went too far. Decompression result is corrupted.
When reaching end of buffer (BIT_DStream_endOfBuffer), progress slowly, notably if you decode multiple symbols per loop,
to properly detect the exact end of stream.
After each decoded symbol, check if DStream is fully consumed using this simple test :
BIT_reloadDStream(&DStream) >= BIT_DStream_completed
When it's done, verify decompression is fully completed, by checking both DStream and the relevant states.
Checking if DStream has reached its end is performed by :
BIT_endOfDStream(&DStream);
Check also the states. There might be some symbols left there, if some high probability ones (>50%) are possible.
FSE_endOfDState(&DState);
*/
/* *****************************************
* FSE unsafe API
*******************************************/
static unsigned char FSE_decodeSymbolFast(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD);
/* faster, but works only if nbBits is always >= 1 (otherwise, result will be corrupted) */
/* *****************************************
* Implementation of inlined functions
*******************************************/
/* decompression */
typedef struct {
U16 tableLog;
U16 fastMode;
} FSE_DTableHeader; /* sizeof U32 */
typedef struct
{
unsigned short newState;
unsigned char symbol;
unsigned char nbBits;
} FSE_decode_t; /* size == U32 */
MEM_STATIC void FSE_initDState(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD, const FSE_DTable* dt)
{
const void* ptr = dt;
const FSE_DTableHeader* const DTableH = (const FSE_DTableHeader*)ptr;
DStatePtr->state = BIT_readBits(bitD, DTableH->tableLog);
BIT_reloadDStream(bitD);
DStatePtr->table = dt + 1;
}
MEM_STATIC BYTE FSE_decodeSymbol(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD)
{
const FSE_decode_t DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state];
const U32 nbBits = DInfo.nbBits;
BYTE symbol = DInfo.symbol;
size_t lowBits = BIT_readBits(bitD, nbBits);
DStatePtr->state = DInfo.newState + lowBits;
return symbol;
}
MEM_STATIC BYTE FSE_decodeSymbolFast(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD)
{
const FSE_decode_t DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state];
const U32 nbBits = DInfo.nbBits;
BYTE symbol = DInfo.symbol;
size_t lowBits = BIT_readBitsFast(bitD, nbBits);
DStatePtr->state = DInfo.newState + lowBits;
return symbol;
}
MEM_STATIC unsigned FSE_endOfDState(const FSE_DState_t* DStatePtr)
{
return DStatePtr->state == 0;
}
#if defined (__cplusplus)
}
#endif
#endif /* FSE_STATIC_H */
/* ******************************************************************
FSE : Finite State Entropy coder
Copyright (C) 2013-2015, Yann Collet.
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above
copyright notice, this list of conditions and the following disclaimer
in the documentation and/or other materials provided with the
distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
You can contact the author at :
- FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
- Public forum : https://groups.google.com/forum/#!forum/lz4c
****************************************************************** */
#ifndef FSE_COMMONDEFS_ONLY
/* **************************************************************
* Tuning parameters
****************************************************************/
/*!MEMORY_USAGE :
* Memory usage formula : N->2^N Bytes (examples : 10 -> 1KB; 12 -> 4KB ; 16 -> 64KB; 20 -> 1MB; etc.)
* Increasing memory usage improves compression ratio
* Reduced memory usage can improve speed, due to cache effect
* Recommended max value is 14, for 16KB, which nicely fits into Intel x86 L1 cache */
#define FSE_MAX_MEMORY_USAGE 14
#define FSE_DEFAULT_MEMORY_USAGE 13
/*!FSE_MAX_SYMBOL_VALUE :
* Maximum symbol value authorized.
* Required for proper stack allocation */
#define FSE_MAX_SYMBOL_VALUE 255
/* **************************************************************
* template functions type & suffix
****************************************************************/
#define FSE_FUNCTION_TYPE BYTE
#define FSE_FUNCTION_EXTENSION
#define FSE_DECODE_TYPE FSE_decode_t
#endif /* !FSE_COMMONDEFS_ONLY */
/* **************************************************************
* Compiler specifics
****************************************************************/
#ifdef _MSC_VER /* Visual Studio */
# define FORCE_INLINE static __forceinline
# include <intrin.h> /* For Visual 2005 */
# pragma warning(disable : 4127) /* disable: C4127: conditional expression is constant */
# pragma warning(disable : 4214) /* disable: C4214: non-int bitfields */
#else
# ifdef __GNUC__
# define GCC_VERSION (__GNUC__ * 100 + __GNUC_MINOR__)
# define FORCE_INLINE static inline __attribute__((always_inline))
# else
# define FORCE_INLINE static inline
# endif
#endif
/* **************************************************************
* Includes
****************************************************************/
#include <stdlib.h> /* malloc, free, qsort */
#include <string.h> /* memcpy, memset */
#include <stdio.h> /* printf (debug) */
/* ***************************************************************
* Constants
*****************************************************************/
#define FSE_MAX_TABLELOG (FSE_MAX_MEMORY_USAGE-2)
#define FSE_MAX_TABLESIZE (1U<<FSE_MAX_TABLELOG)
#define FSE_MAXTABLESIZE_MASK (FSE_MAX_TABLESIZE-1)
#define FSE_DEFAULT_TABLELOG (FSE_DEFAULT_MEMORY_USAGE-2)
#define FSE_MIN_TABLELOG 5
#define FSE_TABLELOG_ABSOLUTE_MAX 15
#if FSE_MAX_TABLELOG > FSE_TABLELOG_ABSOLUTE_MAX
#error "FSE_MAX_TABLELOG > FSE_TABLELOG_ABSOLUTE_MAX is not supported"
#endif
/* **************************************************************
* Error Management
****************************************************************/
#define FSE_STATIC_ASSERT(c) { enum { FSE_static_assert = 1/(int)(!!(c)) }; } /* use only *after* variable declarations */
/* **************************************************************
* Complex types
****************************************************************/
typedef U32 DTable_max_t[FSE_DTABLE_SIZE_U32(FSE_MAX_TABLELOG)];
/* **************************************************************
* Templates
****************************************************************/
/*
designed to be included
for type-specific functions (template emulation in C)
Objective is to write these functions only once, for improved maintenance
*/
/* safety checks */
#ifndef FSE_FUNCTION_EXTENSION
# error "FSE_FUNCTION_EXTENSION must be defined"
#endif
#ifndef FSE_FUNCTION_TYPE
# error "FSE_FUNCTION_TYPE must be defined"
#endif
/* Function names */
#define FSE_CAT(X,Y) X##Y
#define FSE_FUNCTION_NAME(X,Y) FSE_CAT(X,Y)
#define FSE_TYPE_NAME(X,Y) FSE_CAT(X,Y)
static U32 FSE_tableStep(U32 tableSize) { return (tableSize>>1) + (tableSize>>3) + 3; }
static size_t FSE_buildDTable(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog)
{
FSE_DTableHeader DTableH;
void* const tdPtr = dt+1; /* because dt is unsigned, 32-bits aligned on 32-bits */
FSE_DECODE_TYPE* const tableDecode = (FSE_DECODE_TYPE*) (tdPtr);
const U32 tableSize = 1 << tableLog;
const U32 tableMask = tableSize-1;
const U32 step = FSE_tableStep(tableSize);
U16 symbolNext[FSE_MAX_SYMBOL_VALUE+1];
U32 position = 0;
U32 highThreshold = tableSize-1;
const S16 largeLimit= (S16)(1 << (tableLog-1));
U32 noLarge = 1;
U32 s;
/* Sanity Checks */
if (maxSymbolValue > FSE_MAX_SYMBOL_VALUE) return ERROR(maxSymbolValue_tooLarge);
if (tableLog > FSE_MAX_TABLELOG) return ERROR(tableLog_tooLarge);
/* Init, lay down lowprob symbols */
DTableH.tableLog = (U16)tableLog;
for (s=0; s<=maxSymbolValue; s++)
{
if (normalizedCounter[s]==-1)
{
tableDecode[highThreshold--].symbol = (FSE_FUNCTION_TYPE)s;
symbolNext[s] = 1;
}
else
{
if (normalizedCounter[s] >= largeLimit) noLarge=0;
symbolNext[s] = normalizedCounter[s];
}
}
/* Spread symbols */
for (s=0; s<=maxSymbolValue; s++)
{
int i;
for (i=0; i<normalizedCounter[s]; i++)
{
tableDecode[position].symbol = (FSE_FUNCTION_TYPE)s;
position = (position + step) & tableMask;
while (position > highThreshold) position = (position + step) & tableMask; /* lowprob area */
}
}
if (position!=0) return ERROR(GENERIC); /* position must reach all cells once, otherwise normalizedCounter is incorrect */
/* Build Decoding table */
{
U32 i;
for (i=0; i<tableSize; i++)
{
FSE_FUNCTION_TYPE symbol = (FSE_FUNCTION_TYPE)(tableDecode[i].symbol);
U16 nextState = symbolNext[symbol]++;
tableDecode[i].nbBits = (BYTE) (tableLog - BIT_highbit32 ((U32)nextState) );
tableDecode[i].newState = (U16) ( (nextState << tableDecode[i].nbBits) - tableSize);
}
}
DTableH.fastMode = (U16)noLarge;
memcpy(dt, &DTableH, sizeof(DTableH));
return 0;
}
#ifndef FSE_COMMONDEFS_ONLY
/******************************************
* FSE helper functions
******************************************/
static unsigned FSE_isError(size_t code) { return ERR_isError(code); }
/****************************************************************
* FSE NCount encoding-decoding
****************************************************************/
static short FSE_abs(short a)
{
return a<0 ? -a : a;
}
static size_t FSE_readNCount (short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
const void* headerBuffer, size_t hbSize)
{
const BYTE* const istart = (const BYTE*) headerBuffer;
const BYTE* const iend = istart + hbSize;
const BYTE* ip = istart;
int nbBits;
int remaining;
int threshold;
U32 bitStream;
int bitCount;
unsigned charnum = 0;
int previous0 = 0;
if (hbSize < 4) return ERROR(srcSize_wrong);
bitStream = MEM_readLE32(ip);
nbBits = (bitStream & 0xF) + FSE_MIN_TABLELOG; /* extract tableLog */
if (nbBits > FSE_TABLELOG_ABSOLUTE_MAX) return ERROR(tableLog_tooLarge);
bitStream >>= 4;
bitCount = 4;
*tableLogPtr = nbBits;
remaining = (1<<nbBits)+1;
threshold = 1<<nbBits;
nbBits++;
while ((remaining>1) && (charnum<=*maxSVPtr))
{
if (previous0)
{
unsigned n0 = charnum;
while ((bitStream & 0xFFFF) == 0xFFFF)
{
n0+=24;
if (ip < iend-5)
{
ip+=2;
bitStream = MEM_readLE32(ip) >> bitCount;
}
else
{
bitStream >>= 16;
bitCount+=16;
}
}
while ((bitStream & 3) == 3)
{
n0+=3;
bitStream>>=2;
bitCount+=2;
}
n0 += bitStream & 3;
bitCount += 2;
if (n0 > *maxSVPtr) return ERROR(maxSymbolValue_tooSmall);
while (charnum < n0) normalizedCounter[charnum++] = 0;
if ((ip <= iend-7) || (ip + (bitCount>>3) <= iend-4))
{
ip += bitCount>>3;
bitCount &= 7;
bitStream = MEM_readLE32(ip) >> bitCount;
}
else
bitStream >>= 2;
}
{
const short max = (short)((2*threshold-1)-remaining);
short count;
if ((bitStream & (threshold-1)) < (U32)max)
{
count = (short)(bitStream & (threshold-1));
bitCount += nbBits-1;
}
else
{
count = (short)(bitStream & (2*threshold-1));
if (count >= threshold) count -= max;
bitCount += nbBits;
}
count--; /* extra accuracy */
remaining -= FSE_abs(count);
normalizedCounter[charnum++] = count;
previous0 = !count;
while (remaining < threshold)
{
nbBits--;
threshold >>= 1;
}
{
if ((ip <= iend-7) || (ip + (bitCount>>3) <= iend-4))
{
ip += bitCount>>3;
bitCount &= 7;
}
else
{
bitCount -= (int)(8 * (iend - 4 - ip));
ip = iend - 4;
}
bitStream = MEM_readLE32(ip) >> (bitCount & 31);
}
}
}
if (remaining != 1) return ERROR(GENERIC);
*maxSVPtr = charnum-1;
ip += (bitCount+7)>>3;
if ((size_t)(ip-istart) > hbSize) return ERROR(srcSize_wrong);
return ip-istart;
}
/*********************************************************
* Decompression (Byte symbols)
*********************************************************/
static size_t FSE_buildDTable_rle (FSE_DTable* dt, BYTE symbolValue)
{
void* ptr = dt;
FSE_DTableHeader* const DTableH = (FSE_DTableHeader*)ptr;
void* dPtr = dt + 1;
FSE_decode_t* const cell = (FSE_decode_t*)dPtr;
DTableH->tableLog = 0;
DTableH->fastMode = 0;
cell->newState = 0;
cell->symbol = symbolValue;
cell->nbBits = 0;
return 0;
}
static size_t FSE_buildDTable_raw (FSE_DTable* dt, unsigned nbBits)
{
void* ptr = dt;
FSE_DTableHeader* const DTableH = (FSE_DTableHeader*)ptr;
void* dPtr = dt + 1;
FSE_decode_t* const dinfo = (FSE_decode_t*)dPtr;
const unsigned tableSize = 1 << nbBits;
const unsigned tableMask = tableSize - 1;
const unsigned maxSymbolValue = tableMask;
unsigned s;
/* Sanity checks */
if (nbBits < 1) return ERROR(GENERIC); /* min size */
/* Build Decoding Table */
DTableH->tableLog = (U16)nbBits;
DTableH->fastMode = 1;
for (s=0; s<=maxSymbolValue; s++)
{
dinfo[s].newState = 0;
dinfo[s].symbol = (BYTE)s;
dinfo[s].nbBits = (BYTE)nbBits;
}
return 0;
}
FORCE_INLINE size_t FSE_decompress_usingDTable_generic(
void* dst, size_t maxDstSize,
const void* cSrc, size_t cSrcSize,
const FSE_DTable* dt, const unsigned fast)
{
BYTE* const ostart = (BYTE*) dst;
BYTE* op = ostart;
BYTE* const omax = op + maxDstSize;
BYTE* const olimit = omax-3;
BIT_DStream_t bitD;
FSE_DState_t state1;
FSE_DState_t state2;
size_t errorCode;
/* Init */
errorCode = BIT_initDStream(&bitD, cSrc, cSrcSize); /* replaced last arg by maxCompressed Size */
if (FSE_isError(errorCode)) return errorCode;
FSE_initDState(&state1, &bitD, dt);
FSE_initDState(&state2, &bitD, dt);
#define FSE_GETSYMBOL(statePtr) fast ? FSE_decodeSymbolFast(statePtr, &bitD) : FSE_decodeSymbol(statePtr, &bitD)
/* 4 symbols per loop */
for ( ; (BIT_reloadDStream(&bitD)==BIT_DStream_unfinished) && (op<olimit) ; op+=4)
{
op[0] = FSE_GETSYMBOL(&state1);
if (FSE_MAX_TABLELOG*2+7 > sizeof(bitD.bitContainer)*8) /* This test must be static */
BIT_reloadDStream(&bitD);
op[1] = FSE_GETSYMBOL(&state2);
if (FSE_MAX_TABLELOG*4+7 > sizeof(bitD.bitContainer)*8) /* This test must be static */
{ if (BIT_reloadDStream(&bitD) > BIT_DStream_unfinished) { op+=2; break; } }
op[2] = FSE_GETSYMBOL(&state1);
if (FSE_MAX_TABLELOG*2+7 > sizeof(bitD.bitContainer)*8) /* This test must be static */
BIT_reloadDStream(&bitD);
op[3] = FSE_GETSYMBOL(&state2);
}
/* tail */
/* note : BIT_reloadDStream(&bitD) >= FSE_DStream_partiallyFilled; Ends at exactly BIT_DStream_completed */
while (1)
{
if ( (BIT_reloadDStream(&bitD)>BIT_DStream_completed) || (op==omax) || (BIT_endOfDStream(&bitD) && (fast || FSE_endOfDState(&state1))) )
break;
*op++ = FSE_GETSYMBOL(&state1);
if ( (BIT_reloadDStream(&bitD)>BIT_DStream_completed) || (op==omax) || (BIT_endOfDStream(&bitD) && (fast || FSE_endOfDState(&state2))) )
break;
*op++ = FSE_GETSYMBOL(&state2);
}
/* end ? */
if (BIT_endOfDStream(&bitD) && FSE_endOfDState(&state1) && FSE_endOfDState(&state2))
return op-ostart;
if (op==omax) return ERROR(dstSize_tooSmall); /* dst buffer is full, but cSrc unfinished */
return ERROR(corruption_detected);
}
static size_t FSE_decompress_usingDTable(void* dst, size_t originalSize,
const void* cSrc, size_t cSrcSize,
const FSE_DTable* dt)
{
const void* ptr = dt;
const FSE_DTableHeader* DTableH = (const FSE_DTableHeader*)ptr;
const U32 fastMode = DTableH->fastMode;
/* select fast mode (static) */
if (fastMode) return FSE_decompress_usingDTable_generic(dst, originalSize, cSrc, cSrcSize, dt, 1);
return FSE_decompress_usingDTable_generic(dst, originalSize, cSrc, cSrcSize, dt, 0);
}
static size_t FSE_decompress(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize)
{
const BYTE* const istart = (const BYTE*)cSrc;
const BYTE* ip = istart;
short counting[FSE_MAX_SYMBOL_VALUE+1];
DTable_max_t dt; /* Static analyzer seems unable to understand this table will be properly initialized later */
unsigned tableLog;
unsigned maxSymbolValue = FSE_MAX_SYMBOL_VALUE;
size_t errorCode;
if (cSrcSize<2) return ERROR(srcSize_wrong); /* too small input size */
/* normal FSE decoding mode */
errorCode = FSE_readNCount (counting, &maxSymbolValue, &tableLog, istart, cSrcSize);
if (FSE_isError(errorCode)) return errorCode;
if (errorCode >= cSrcSize) return ERROR(srcSize_wrong); /* too small input size */
ip += errorCode;
cSrcSize -= errorCode;
errorCode = FSE_buildDTable (dt, counting, maxSymbolValue, tableLog);
if (FSE_isError(errorCode)) return errorCode;
/* always return, even if it is an error code */
return FSE_decompress_usingDTable (dst, maxDstSize, ip, cSrcSize, dt);
}
#endif /* FSE_COMMONDEFS_ONLY */
/* ******************************************************************
Huff0 : Huffman coder, part of New Generation Entropy library
header file
Copyright (C) 2013-2015, Yann Collet.
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above
copyright notice, this list of conditions and the following disclaimer
in the documentation and/or other materials provided with the
distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
You can contact the author at :
- Source repository : https://github.com/Cyan4973/FiniteStateEntropy
- Public forum : https://groups.google.com/forum/#!forum/lz4c
****************************************************************** */
#ifndef HUFF0_H
#define HUFF0_H
#if defined (__cplusplus)
extern "C" {
#endif
/* ****************************************
* Dependency
******************************************/
#include <stddef.h> /* size_t */
/* ****************************************
* Huff0 simple functions
******************************************/
static size_t HUF_decompress(void* dst, size_t dstSize,
const void* cSrc, size_t cSrcSize);
/*!
HUF_decompress():
Decompress Huff0 data from buffer 'cSrc', of size 'cSrcSize',
into already allocated destination buffer 'dst', of size 'dstSize'.
'dstSize' must be the exact size of original (uncompressed) data.
Note : in contrast with FSE, HUF_decompress can regenerate RLE (cSrcSize==1) and uncompressed (cSrcSize==dstSize) data, because it knows size to regenerate.
@return : size of regenerated data (== dstSize)
or an error code, which can be tested using HUF_isError()
*/
/* ****************************************
* Tool functions
******************************************/
/* Error Management */
static unsigned HUF_isError(size_t code); /* tells if a return value is an error code */
#if defined (__cplusplus)
}
#endif
#endif /* HUFF0_H */
/* ******************************************************************
Huff0 : Huffman coder, part of New Generation Entropy library
header file for static linking (only)
Copyright (C) 2013-2015, Yann Collet
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above
copyright notice, this list of conditions and the following disclaimer
in the documentation and/or other materials provided with the
distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
You can contact the author at :
- Source repository : https://github.com/Cyan4973/FiniteStateEntropy
- Public forum : https://groups.google.com/forum/#!forum/lz4c
****************************************************************** */
#ifndef HUFF0_STATIC_H
#define HUFF0_STATIC_H
#if defined (__cplusplus)
extern "C" {
#endif
/* ****************************************
* Dependency
******************************************/
#include "huff0.h"
/* ****************************************
* Static allocation macros
******************************************/
/* static allocation of Huff0's DTable */
#define HUF_DTABLE_SIZE(maxTableLog) (1 + (1<<maxTableLog)) /* nb Cells; use unsigned short for X2, unsigned int for X4 */
#define HUF_CREATE_STATIC_DTABLEX2(DTable, maxTableLog) \
unsigned short DTable[HUF_DTABLE_SIZE(maxTableLog)] = { maxTableLog }
#define HUF_CREATE_STATIC_DTABLEX4(DTable, maxTableLog) \
unsigned int DTable[HUF_DTABLE_SIZE(maxTableLog)] = { maxTableLog }
#define HUF_CREATE_STATIC_DTABLEX6(DTable, maxTableLog) \
unsigned int DTable[HUF_DTABLE_SIZE(maxTableLog) * 3 / 2] = { maxTableLog }
/* ****************************************
* Advanced decompression functions
******************************************/
static size_t HUF_decompress4X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /* single-symbol decoder */
static size_t HUF_decompress4X4 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /* double-symbols decoder */
static size_t HUF_decompress4X6 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /* quad-symbols decoder */
/* ****************************************
* Huff0 detailed API
******************************************/
/*!
HUF_decompress() does the following:
1. select the decompression algorithm (X2, X4, X6) based on pre-computed heuristics
2. build Huffman table from save, using HUF_readDTableXn()
3. decode 1 or 4 segments in parallel using HUF_decompressSXn_usingDTable
*/
static size_t HUF_readDTableX2 (unsigned short* DTable, const void* src, size_t srcSize);
static size_t HUF_readDTableX4 (unsigned* DTable, const void* src, size_t srcSize);
static size_t HUF_readDTableX6 (unsigned* DTable, const void* src, size_t srcSize);
static size_t HUF_decompress4X2_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const unsigned short* DTable);
static size_t HUF_decompress4X4_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const unsigned* DTable);
static size_t HUF_decompress4X6_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const unsigned* DTable);
#if defined (__cplusplus)
}
#endif
#endif /* HUFF0_STATIC_H */
/* ******************************************************************
Huff0 : Huffman coder, part of New Generation Entropy library
Copyright (C) 2013-2015, Yann Collet.
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above
copyright notice, this list of conditions and the following disclaimer
in the documentation and/or other materials provided with the
distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
You can contact the author at :
- FSE+Huff0 source repository : https://github.com/Cyan4973/FiniteStateEntropy
****************************************************************** */
/* **************************************************************
* Compiler specifics
****************************************************************/
#if defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */)
/* inline is defined */
#elif defined(_MSC_VER)
# define inline __inline
#else
# define inline /* disable inline */
#endif
#ifdef _MSC_VER /* Visual Studio */
# define FORCE_INLINE static __forceinline
# pragma warning(disable : 4127) /* disable: C4127: conditional expression is constant */
#else
# ifdef __GNUC__
# define GCC_VERSION (__GNUC__ * 100 + __GNUC_MINOR__)
# define FORCE_INLINE static inline __attribute__((always_inline))
# else
# define FORCE_INLINE static inline
# endif
#endif
/* **************************************************************
* Includes
****************************************************************/
#include <stdlib.h> /* malloc, free, qsort */
#include <string.h> /* memcpy, memset */
#include <stdio.h> /* printf (debug) */
/* **************************************************************
* Constants
****************************************************************/
#define HUF_ABSOLUTEMAX_TABLELOG 16 /* absolute limit of HUF_MAX_TABLELOG. Beyond that value, code does not work */
#define HUF_MAX_TABLELOG 12 /* max configured tableLog (for static allocation); can be modified up to HUF_ABSOLUTEMAX_TABLELOG */
#define HUF_DEFAULT_TABLELOG HUF_MAX_TABLELOG /* tableLog by default, when not specified */
#define HUF_MAX_SYMBOL_VALUE 255
#if (HUF_MAX_TABLELOG > HUF_ABSOLUTEMAX_TABLELOG)
# error "HUF_MAX_TABLELOG is too large !"
#endif
/* **************************************************************
* Error Management
****************************************************************/
static unsigned HUF_isError(size_t code) { return ERR_isError(code); }
#define HUF_STATIC_ASSERT(c) { enum { HUF_static_assert = 1/(int)(!!(c)) }; } /* use only *after* variable declarations */
/*-*******************************************************
* Huff0 : Huffman block decompression
*********************************************************/
typedef struct { BYTE byte; BYTE nbBits; } HUF_DEltX2; /* single-symbol decoding */
typedef struct { U16 sequence; BYTE nbBits; BYTE length; } HUF_DEltX4; /* double-symbols decoding */
typedef struct { BYTE symbol; BYTE weight; } sortedSymbol_t;
/*! HUF_readStats
Read compact Huffman tree, saved by HUF_writeCTable
@huffWeight : destination buffer
@return : size read from `src`
*/
static size_t HUF_readStats(BYTE* huffWeight, size_t hwSize, U32* rankStats,
U32* nbSymbolsPtr, U32* tableLogPtr,
const void* src, size_t srcSize)
{
U32 weightTotal;
U32 tableLog;
const BYTE* ip = (const BYTE*) src;
size_t iSize = ip[0];
size_t oSize;
U32 n;
//memset(huffWeight, 0, hwSize); /* is not necessary, even though some analyzer complain ... */
if (iSize >= 128) /* special header */
{
if (iSize >= (242)) /* RLE */
{
static int l[14] = { 1, 2, 3, 4, 7, 8, 15, 16, 31, 32, 63, 64, 127, 128 };
oSize = l[iSize-242];
memset(huffWeight, 1, hwSize);
iSize = 0;
}
else /* Incompressible */
{
oSize = iSize - 127;
iSize = ((oSize+1)/2);
if (iSize+1 > srcSize) return ERROR(srcSize_wrong);
if (oSize >= hwSize) return ERROR(corruption_detected);
ip += 1;
for (n=0; n<oSize; n+=2)
{
huffWeight[n] = ip[n/2] >> 4;
huffWeight[n+1] = ip[n/2] & 15;
}
}
}
else /* header compressed with FSE (normal case) */
{
if (iSize+1 > srcSize) return ERROR(srcSize_wrong);
oSize = FSE_decompress(huffWeight, hwSize-1, ip+1, iSize); /* max (hwSize-1) values decoded, as last one is implied */
if (FSE_isError(oSize)) return oSize;
}
/* collect weight stats */
memset(rankStats, 0, (HUF_ABSOLUTEMAX_TABLELOG + 1) * sizeof(U32));
weightTotal = 0;
for (n=0; n<oSize; n++)
{
if (huffWeight[n] >= HUF_ABSOLUTEMAX_TABLELOG) return ERROR(corruption_detected);
rankStats[huffWeight[n]]++;
weightTotal += (1 << huffWeight[n]) >> 1;
}
/* get last non-null symbol weight (implied, total must be 2^n) */
tableLog = BIT_highbit32(weightTotal) + 1;
if (tableLog > HUF_ABSOLUTEMAX_TABLELOG) return ERROR(corruption_detected);
{
U32 total = 1 << tableLog;
U32 rest = total - weightTotal;
U32 verif = 1 << BIT_highbit32(rest);
U32 lastWeight = BIT_highbit32(rest) + 1;
if (verif != rest) return ERROR(corruption_detected); /* last value must be a clean power of 2 */
huffWeight[oSize] = (BYTE)lastWeight;
rankStats[lastWeight]++;
}
/* check tree construction validity */
if ((rankStats[1] < 2) || (rankStats[1] & 1)) return ERROR(corruption_detected); /* by construction : at least 2 elts of rank 1, must be even */
/* results */
*nbSymbolsPtr = (U32)(oSize+1);
*tableLogPtr = tableLog;
return iSize+1;
}
/**************************/
/* single-symbol decoding */
/**************************/
static size_t HUF_readDTableX2 (U16* DTable, const void* src, size_t srcSize)
{
BYTE huffWeight[HUF_MAX_SYMBOL_VALUE + 1];
U32 rankVal[HUF_ABSOLUTEMAX_TABLELOG + 1]; /* large enough for values from 0 to 16 */
U32 tableLog = 0;
size_t iSize;
U32 nbSymbols = 0;
U32 n;
U32 nextRankStart;
void* const dtPtr = DTable + 1;
HUF_DEltX2* const dt = (HUF_DEltX2*)dtPtr;
HUF_STATIC_ASSERT(sizeof(HUF_DEltX2) == sizeof(U16)); /* if compilation fails here, assertion is false */
//memset(huffWeight, 0, sizeof(huffWeight)); /* is not necessary, even though some analyzer complain ... */
iSize = HUF_readStats(huffWeight, HUF_MAX_SYMBOL_VALUE + 1, rankVal, &nbSymbols, &tableLog, src, srcSize);
if (HUF_isError(iSize)) return iSize;
/* check result */
if (tableLog > DTable[0]) return ERROR(tableLog_tooLarge); /* DTable is too small */
DTable[0] = (U16)tableLog; /* maybe should separate sizeof DTable, as allocated, from used size of DTable, in case of DTable re-use */
/* Prepare ranks */
nextRankStart = 0;
for (n=1; n<=tableLog; n++)
{
U32 current = nextRankStart;
nextRankStart += (rankVal[n] << (n-1));
rankVal[n] = current;
}
/* fill DTable */
for (n=0; n<nbSymbols; n++)
{
const U32 w = huffWeight[n];
const U32 length = (1 << w) >> 1;
U32 i;
HUF_DEltX2 D;
D.byte = (BYTE)n; D.nbBits = (BYTE)(tableLog + 1 - w);
for (i = rankVal[w]; i < rankVal[w] + length; i++)
dt[i] = D;
rankVal[w] += length;
}
return iSize;
}
static BYTE HUF_decodeSymbolX2(BIT_DStream_t* Dstream, const HUF_DEltX2* dt, const U32 dtLog)
{
const size_t val = BIT_lookBitsFast(Dstream, dtLog); /* note : dtLog >= 1 */
const BYTE c = dt[val].byte;
BIT_skipBits(Dstream, dt[val].nbBits);
return c;
}
#define HUF_DECODE_SYMBOLX2_0(ptr, DStreamPtr) \
*ptr++ = HUF_decodeSymbolX2(DStreamPtr, dt, dtLog)
#define HUF_DECODE_SYMBOLX2_1(ptr, DStreamPtr) \
if (MEM_64bits() || (HUF_MAX_TABLELOG<=12)) \
HUF_DECODE_SYMBOLX2_0(ptr, DStreamPtr)
#define HUF_DECODE_SYMBOLX2_2(ptr, DStreamPtr) \
if (MEM_64bits()) \
HUF_DECODE_SYMBOLX2_0(ptr, DStreamPtr)
static inline size_t HUF_decodeStreamX2(BYTE* p, BIT_DStream_t* const bitDPtr, BYTE* const pEnd, const HUF_DEltX2* const dt, const U32 dtLog)
{
BYTE* const pStart = p;
/* up to 4 symbols at a time */
while ((BIT_reloadDStream(bitDPtr) == BIT_DStream_unfinished) && (p <= pEnd-4))
{
HUF_DECODE_SYMBOLX2_2(p, bitDPtr);
HUF_DECODE_SYMBOLX2_1(p, bitDPtr);
HUF_DECODE_SYMBOLX2_2(p, bitDPtr);
HUF_DECODE_SYMBOLX2_0(p, bitDPtr);
}
/* closer to the end */
while ((BIT_reloadDStream(bitDPtr) == BIT_DStream_unfinished) && (p < pEnd))
HUF_DECODE_SYMBOLX2_0(p, bitDPtr);
/* no more data to retrieve from bitstream, hence no need to reload */
while (p < pEnd)
HUF_DECODE_SYMBOLX2_0(p, bitDPtr);
return pEnd-pStart;
}
static size_t HUF_decompress4X2_usingDTable(
void* dst, size_t dstSize,
const void* cSrc, size_t cSrcSize,
const U16* DTable)
{
if (cSrcSize < 10) return ERROR(corruption_detected); /* strict minimum : jump table + 1 byte per stream */
{
const BYTE* const istart = (const BYTE*) cSrc;
BYTE* const ostart = (BYTE*) dst;
BYTE* const oend = ostart + dstSize;
const void* const dtPtr = DTable;
const HUF_DEltX2* const dt = ((const HUF_DEltX2*)dtPtr) +1;
const U32 dtLog = DTable[0];
size_t errorCode;
/* Init */
BIT_DStream_t bitD1;
BIT_DStream_t bitD2;
BIT_DStream_t bitD3;
BIT_DStream_t bitD4;
const size_t length1 = MEM_readLE16(istart);
const size_t length2 = MEM_readLE16(istart+2);
const size_t length3 = MEM_readLE16(istart+4);
size_t length4;
const BYTE* const istart1 = istart + 6; /* jumpTable */
const BYTE* const istart2 = istart1 + length1;
const BYTE* const istart3 = istart2 + length2;
const BYTE* const istart4 = istart3 + length3;
const size_t segmentSize = (dstSize+3) / 4;
BYTE* const opStart2 = ostart + segmentSize;
BYTE* const opStart3 = opStart2 + segmentSize;
BYTE* const opStart4 = opStart3 + segmentSize;
BYTE* op1 = ostart;
BYTE* op2 = opStart2;
BYTE* op3 = opStart3;
BYTE* op4 = opStart4;
U32 endSignal;
length4 = cSrcSize - (length1 + length2 + length3 + 6);
if (length4 > cSrcSize) return ERROR(corruption_detected); /* overflow */
errorCode = BIT_initDStream(&bitD1, istart1, length1);
if (HUF_isError(errorCode)) return errorCode;
errorCode = BIT_initDStream(&bitD2, istart2, length2);
if (HUF_isError(errorCode)) return errorCode;
errorCode = BIT_initDStream(&bitD3, istart3, length3);
if (HUF_isError(errorCode)) return errorCode;
errorCode = BIT_initDStream(&bitD4, istart4, length4);
if (HUF_isError(errorCode)) return errorCode;
/* 16-32 symbols per loop (4-8 symbols per stream) */
endSignal = BIT_reloadDStream(&bitD1) | BIT_reloadDStream(&bitD2) | BIT_reloadDStream(&bitD3) | BIT_reloadDStream(&bitD4);
for ( ; (endSignal==BIT_DStream_unfinished) && (op4<(oend-7)) ; )
{
HUF_DECODE_SYMBOLX2_2(op1, &bitD1);
HUF_DECODE_SYMBOLX2_2(op2, &bitD2);
HUF_DECODE_SYMBOLX2_2(op3, &bitD3);
HUF_DECODE_SYMBOLX2_2(op4, &bitD4);
HUF_DECODE_SYMBOLX2_1(op1, &bitD1);
HUF_DECODE_SYMBOLX2_1(op2, &bitD2);
HUF_DECODE_SYMBOLX2_1(op3, &bitD3);
HUF_DECODE_SYMBOLX2_1(op4, &bitD4);
HUF_DECODE_SYMBOLX2_2(op1, &bitD1);
HUF_DECODE_SYMBOLX2_2(op2, &bitD2);
HUF_DECODE_SYMBOLX2_2(op3, &bitD3);
HUF_DECODE_SYMBOLX2_2(op4, &bitD4);
HUF_DECODE_SYMBOLX2_0(op1, &bitD1);
HUF_DECODE_SYMBOLX2_0(op2, &bitD2);
HUF_DECODE_SYMBOLX2_0(op3, &bitD3);
HUF_DECODE_SYMBOLX2_0(op4, &bitD4);
endSignal = BIT_reloadDStream(&bitD1) | BIT_reloadDStream(&bitD2) | BIT_reloadDStream(&bitD3) | BIT_reloadDStream(&bitD4);
}
/* check corruption */
if (op1 > opStart2) return ERROR(corruption_detected);
if (op2 > opStart3) return ERROR(corruption_detected);
if (op3 > opStart4) return ERROR(corruption_detected);
/* note : op4 supposed already verified within main loop */
/* finish bitStreams one by one */
HUF_decodeStreamX2(op1, &bitD1, opStart2, dt, dtLog);
HUF_decodeStreamX2(op2, &bitD2, opStart3, dt, dtLog);
HUF_decodeStreamX2(op3, &bitD3, opStart4, dt, dtLog);
HUF_decodeStreamX2(op4, &bitD4, oend, dt, dtLog);
/* check */
endSignal = BIT_endOfDStream(&bitD1) & BIT_endOfDStream(&bitD2) & BIT_endOfDStream(&bitD3) & BIT_endOfDStream(&bitD4);
if (!endSignal) return ERROR(corruption_detected);
/* decoded size */
return dstSize;
}
}
static size_t HUF_decompress4X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize)
{
HUF_CREATE_STATIC_DTABLEX2(DTable, HUF_MAX_TABLELOG);
const BYTE* ip = (const BYTE*) cSrc;
size_t errorCode;
errorCode = HUF_readDTableX2 (DTable, cSrc, cSrcSize);
if (HUF_isError(errorCode)) return errorCode;
if (errorCode >= cSrcSize) return ERROR(srcSize_wrong);
ip += errorCode;
cSrcSize -= errorCode;
return HUF_decompress4X2_usingDTable (dst, dstSize, ip, cSrcSize, DTable);
}
/***************************/
/* double-symbols decoding */
/***************************/
static void HUF_fillDTableX4Level2(HUF_DEltX4* DTable, U32 sizeLog, const U32 consumed,
const U32* rankValOrigin, const int minWeight,
const sortedSymbol_t* sortedSymbols, const U32 sortedListSize,
U32 nbBitsBaseline, U16 baseSeq)
{
HUF_DEltX4 DElt;
U32 rankVal[HUF_ABSOLUTEMAX_TABLELOG + 1];
U32 s;
/* get pre-calculated rankVal */
memcpy(rankVal, rankValOrigin, sizeof(rankVal));
/* fill skipped values */
if (minWeight>1)
{
U32 i, skipSize = rankVal[minWeight];
MEM_writeLE16(&(DElt.sequence), baseSeq);
DElt.nbBits = (BYTE)(consumed);
DElt.length = 1;
for (i = 0; i < skipSize; i++)
DTable[i] = DElt;
}
/* fill DTable */
for (s=0; s<sortedListSize; s++) /* note : sortedSymbols already skipped */
{
const U32 symbol = sortedSymbols[s].symbol;
const U32 weight = sortedSymbols[s].weight;
const U32 nbBits = nbBitsBaseline - weight;
const U32 length = 1 << (sizeLog-nbBits);
const U32 start = rankVal[weight];
U32 i = start;
const U32 end = start + length;
MEM_writeLE16(&(DElt.sequence), (U16)(baseSeq + (symbol << 8)));
DElt.nbBits = (BYTE)(nbBits + consumed);
DElt.length = 2;
do { DTable[i++] = DElt; } while (i<end); /* since length >= 1 */
rankVal[weight] += length;
}
}
typedef U32 rankVal_t[HUF_ABSOLUTEMAX_TABLELOG][HUF_ABSOLUTEMAX_TABLELOG + 1];
static void HUF_fillDTableX4(HUF_DEltX4* DTable, const U32 targetLog,
const sortedSymbol_t* sortedList, const U32 sortedListSize,
const U32* rankStart, rankVal_t rankValOrigin, const U32 maxWeight,
const U32 nbBitsBaseline)
{
U32 rankVal[HUF_ABSOLUTEMAX_TABLELOG + 1];
const int scaleLog = nbBitsBaseline - targetLog; /* note : targetLog >= srcLog, hence scaleLog <= 1 */
const U32 minBits = nbBitsBaseline - maxWeight;
U32 s;
memcpy(rankVal, rankValOrigin, sizeof(rankVal));
/* fill DTable */
for (s=0; s<sortedListSize; s++)
{
const U16 symbol = sortedList[s].symbol;
const U32 weight = sortedList[s].weight;
const U32 nbBits = nbBitsBaseline - weight;
const U32 start = rankVal[weight];
const U32 length = 1 << (targetLog-nbBits);
if (targetLog-nbBits >= minBits) /* enough room for a second symbol */
{
U32 sortedRank;
int minWeight = nbBits + scaleLog;
if (minWeight < 1) minWeight = 1;
sortedRank = rankStart[minWeight];
HUF_fillDTableX4Level2(DTable+start, targetLog-nbBits, nbBits,
rankValOrigin[nbBits], minWeight,
sortedList+sortedRank, sortedListSize-sortedRank,
nbBitsBaseline, symbol);
}
else
{
U32 i;
const U32 end = start + length;
HUF_DEltX4 DElt;
MEM_writeLE16(&(DElt.sequence), symbol);
DElt.nbBits = (BYTE)(nbBits);
DElt.length = 1;
for (i = start; i < end; i++)
DTable[i] = DElt;
}
rankVal[weight] += length;
}
}
static size_t HUF_readDTableX4 (U32* DTable, const void* src, size_t srcSize)
{
BYTE weightList[HUF_MAX_SYMBOL_VALUE + 1];
sortedSymbol_t sortedSymbol[HUF_MAX_SYMBOL_VALUE + 1];
U32 rankStats[HUF_ABSOLUTEMAX_TABLELOG + 1] = { 0 };
U32 rankStart0[HUF_ABSOLUTEMAX_TABLELOG + 2] = { 0 };
U32* const rankStart = rankStart0+1;
rankVal_t rankVal;
U32 tableLog, maxW, sizeOfSort, nbSymbols;
const U32 memLog = DTable[0];
size_t iSize;
void* dtPtr = DTable;
HUF_DEltX4* const dt = ((HUF_DEltX4*)dtPtr) + 1;
HUF_STATIC_ASSERT(sizeof(HUF_DEltX4) == sizeof(U32)); /* if compilation fails here, assertion is false */
if (memLog > HUF_ABSOLUTEMAX_TABLELOG) return ERROR(tableLog_tooLarge);
//memset(weightList, 0, sizeof(weightList)); /* is not necessary, even though some analyzer complain ... */
iSize = HUF_readStats(weightList, HUF_MAX_SYMBOL_VALUE + 1, rankStats, &nbSymbols, &tableLog, src, srcSize);
if (HUF_isError(iSize)) return iSize;
/* check result */
if (tableLog > memLog) return ERROR(tableLog_tooLarge); /* DTable can't fit code depth */
/* find maxWeight */
for (maxW = tableLog; rankStats[maxW]==0; maxW--) {} /* necessarily finds a solution before 0 */
/* Get start index of each weight */
{
U32 w, nextRankStart = 0;
for (w=1; w<=maxW; w++)
{
U32 current = nextRankStart;
nextRankStart += rankStats[w];
rankStart[w] = current;
}
rankStart[0] = nextRankStart; /* put all 0w symbols at the end of sorted list*/
sizeOfSort = nextRankStart;
}
/* sort symbols by weight */
{
U32 s;
for (s=0; s<nbSymbols; s++)
{
U32 w = weightList[s];
U32 r = rankStart[w]++;
sortedSymbol[r].symbol = (BYTE)s;
sortedSymbol[r].weight = (BYTE)w;
}
rankStart[0] = 0; /* forget 0w symbols; this is beginning of weight(1) */
}
/* Build rankVal */
{
const U32 minBits = tableLog+1 - maxW;
U32 nextRankVal = 0;
U32 w, consumed;
const int rescale = (memLog-tableLog) - 1; /* tableLog <= memLog */
U32* rankVal0 = rankVal[0];
for (w=1; w<=maxW; w++)
{
U32 current = nextRankVal;
nextRankVal += rankStats[w] << (w+rescale);
rankVal0[w] = current;
}
for (consumed = minBits; consumed <= memLog - minBits; consumed++)
{
U32* rankValPtr = rankVal[consumed];
for (w = 1; w <= maxW; w++)
{
rankValPtr[w] = rankVal0[w] >> consumed;
}
}
}
HUF_fillDTableX4(dt, memLog,
sortedSymbol, sizeOfSort,
rankStart0, rankVal, maxW,
tableLog+1);
return iSize;
}
static U32 HUF_decodeSymbolX4(void* op, BIT_DStream_t* DStream, const HUF_DEltX4* dt, const U32 dtLog)
{
const size_t val = BIT_lookBitsFast(DStream, dtLog); /* note : dtLog >= 1 */
memcpy(op, dt+val, 2);
BIT_skipBits(DStream, dt[val].nbBits);
return dt[val].length;
}
static U32 HUF_decodeLastSymbolX4(void* op, BIT_DStream_t* DStream, const HUF_DEltX4* dt, const U32 dtLog)
{
const size_t val = BIT_lookBitsFast(DStream, dtLog); /* note : dtLog >= 1 */
memcpy(op, dt+val, 1);
if (dt[val].length==1) BIT_skipBits(DStream, dt[val].nbBits);
else
{
if (DStream->bitsConsumed < (sizeof(DStream->bitContainer)*8))
{
BIT_skipBits(DStream, dt[val].nbBits);
if (DStream->bitsConsumed > (sizeof(DStream->bitContainer)*8))
DStream->bitsConsumed = (sizeof(DStream->bitContainer)*8); /* ugly hack; works only because it's the last symbol. Note : can't easily extract nbBits from just this symbol */
}
}
return 1;
}
#define HUF_DECODE_SYMBOLX4_0(ptr, DStreamPtr) \
ptr += HUF_decodeSymbolX4(ptr, DStreamPtr, dt, dtLog)
#define HUF_DECODE_SYMBOLX4_1(ptr, DStreamPtr) \
if (MEM_64bits() || (HUF_MAX_TABLELOG<=12)) \
ptr += HUF_decodeSymbolX4(ptr, DStreamPtr, dt, dtLog)
#define HUF_DECODE_SYMBOLX4_2(ptr, DStreamPtr) \
if (MEM_64bits()) \
ptr += HUF_decodeSymbolX4(ptr, DStreamPtr, dt, dtLog)
static inline size_t HUF_decodeStreamX4(BYTE* p, BIT_DStream_t* bitDPtr, BYTE* const pEnd, const HUF_DEltX4* const dt, const U32 dtLog)
{
BYTE* const pStart = p;
/* up to 8 symbols at a time */
while ((BIT_reloadDStream(bitDPtr) == BIT_DStream_unfinished) && (p < pEnd-7))
{
HUF_DECODE_SYMBOLX4_2(p, bitDPtr);
HUF_DECODE_SYMBOLX4_1(p, bitDPtr);
HUF_DECODE_SYMBOLX4_2(p, bitDPtr);
HUF_DECODE_SYMBOLX4_0(p, bitDPtr);
}
/* closer to the end */
while ((BIT_reloadDStream(bitDPtr) == BIT_DStream_unfinished) && (p <= pEnd-2))
HUF_DECODE_SYMBOLX4_0(p, bitDPtr);
while (p <= pEnd-2)
HUF_DECODE_SYMBOLX4_0(p, bitDPtr); /* no need to reload : reached the end of DStream */
if (p < pEnd)
p += HUF_decodeLastSymbolX4(p, bitDPtr, dt, dtLog);
return p-pStart;
}
static size_t HUF_decompress4X4_usingDTable(
void* dst, size_t dstSize,
const void* cSrc, size_t cSrcSize,
const U32* DTable)
{
if (cSrcSize < 10) return ERROR(corruption_detected); /* strict minimum : jump table + 1 byte per stream */
{
const BYTE* const istart = (const BYTE*) cSrc;
BYTE* const ostart = (BYTE*) dst;
BYTE* const oend = ostart + dstSize;
const void* const dtPtr = DTable;
const HUF_DEltX4* const dt = ((const HUF_DEltX4*)dtPtr) +1;
const U32 dtLog = DTable[0];
size_t errorCode;
/* Init */
BIT_DStream_t bitD1;
BIT_DStream_t bitD2;
BIT_DStream_t bitD3;
BIT_DStream_t bitD4;
const size_t length1 = MEM_readLE16(istart);
const size_t length2 = MEM_readLE16(istart+2);
const size_t length3 = MEM_readLE16(istart+4);
size_t length4;
const BYTE* const istart1 = istart + 6; /* jumpTable */
const BYTE* const istart2 = istart1 + length1;
const BYTE* const istart3 = istart2 + length2;
const BYTE* const istart4 = istart3 + length3;
const size_t segmentSize = (dstSize+3) / 4;
BYTE* const opStart2 = ostart + segmentSize;
BYTE* const opStart3 = opStart2 + segmentSize;
BYTE* const opStart4 = opStart3 + segmentSize;
BYTE* op1 = ostart;
BYTE* op2 = opStart2;
BYTE* op3 = opStart3;
BYTE* op4 = opStart4;
U32 endSignal;
length4 = cSrcSize - (length1 + length2 + length3 + 6);
if (length4 > cSrcSize) return ERROR(corruption_detected); /* overflow */
errorCode = BIT_initDStream(&bitD1, istart1, length1);
if (HUF_isError(errorCode)) return errorCode;
errorCode = BIT_initDStream(&bitD2, istart2, length2);
if (HUF_isError(errorCode)) return errorCode;
errorCode = BIT_initDStream(&bitD3, istart3, length3);
if (HUF_isError(errorCode)) return errorCode;
errorCode = BIT_initDStream(&bitD4, istart4, length4);
if (HUF_isError(errorCode)) return errorCode;
/* 16-32 symbols per loop (4-8 symbols per stream) */
endSignal = BIT_reloadDStream(&bitD1) | BIT_reloadDStream(&bitD2) | BIT_reloadDStream(&bitD3) | BIT_reloadDStream(&bitD4);
for ( ; (endSignal==BIT_DStream_unfinished) && (op4<(oend-7)) ; )
{
HUF_DECODE_SYMBOLX4_2(op1, &bitD1);
HUF_DECODE_SYMBOLX4_2(op2, &bitD2);
HUF_DECODE_SYMBOLX4_2(op3, &bitD3);
HUF_DECODE_SYMBOLX4_2(op4, &bitD4);
HUF_DECODE_SYMBOLX4_1(op1, &bitD1);
HUF_DECODE_SYMBOLX4_1(op2, &bitD2);
HUF_DECODE_SYMBOLX4_1(op3, &bitD3);
HUF_DECODE_SYMBOLX4_1(op4, &bitD4);
HUF_DECODE_SYMBOLX4_2(op1, &bitD1);
HUF_DECODE_SYMBOLX4_2(op2, &bitD2);
HUF_DECODE_SYMBOLX4_2(op3, &bitD3);
HUF_DECODE_SYMBOLX4_2(op4, &bitD4);
HUF_DECODE_SYMBOLX4_0(op1, &bitD1);
HUF_DECODE_SYMBOLX4_0(op2, &bitD2);
HUF_DECODE_SYMBOLX4_0(op3, &bitD3);
HUF_DECODE_SYMBOLX4_0(op4, &bitD4);
endSignal = BIT_reloadDStream(&bitD1) | BIT_reloadDStream(&bitD2) | BIT_reloadDStream(&bitD3) | BIT_reloadDStream(&bitD4);
}
/* check corruption */
if (op1 > opStart2) return ERROR(corruption_detected);
if (op2 > opStart3) return ERROR(corruption_detected);
if (op3 > opStart4) return ERROR(corruption_detected);
/* note : op4 supposed already verified within main loop */
/* finish bitStreams one by one */
HUF_decodeStreamX4(op1, &bitD1, opStart2, dt, dtLog);
HUF_decodeStreamX4(op2, &bitD2, opStart3, dt, dtLog);
HUF_decodeStreamX4(op3, &bitD3, opStart4, dt, dtLog);
HUF_decodeStreamX4(op4, &bitD4, oend, dt, dtLog);
/* check */
endSignal = BIT_endOfDStream(&bitD1) & BIT_endOfDStream(&bitD2) & BIT_endOfDStream(&bitD3) & BIT_endOfDStream(&bitD4);
if (!endSignal) return ERROR(corruption_detected);
/* decoded size */
return dstSize;
}
}
static size_t HUF_decompress4X4 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize)
{
HUF_CREATE_STATIC_DTABLEX4(DTable, HUF_MAX_TABLELOG);
const BYTE* ip = (const BYTE*) cSrc;
size_t hSize = HUF_readDTableX4 (DTable, cSrc, cSrcSize);
if (HUF_isError(hSize)) return hSize;
if (hSize >= cSrcSize) return ERROR(srcSize_wrong);
ip += hSize;
cSrcSize -= hSize;
return HUF_decompress4X4_usingDTable (dst, dstSize, ip, cSrcSize, DTable);
}
/**********************************/
/* quad-symbol decoding */
/**********************************/
typedef struct { BYTE nbBits; BYTE nbBytes; } HUF_DDescX6;
typedef union { BYTE byte[4]; U32 sequence; } HUF_DSeqX6;
/* recursive, up to level 3; may benefit from <template>-like strategy to nest each level inline */
static void HUF_fillDTableX6LevelN(HUF_DDescX6* DDescription, HUF_DSeqX6* DSequence, int sizeLog,
const rankVal_t rankValOrigin, const U32 consumed, const int minWeight, const U32 maxWeight,
const sortedSymbol_t* sortedSymbols, const U32 sortedListSize, const U32* rankStart,
const U32 nbBitsBaseline, HUF_DSeqX6 baseSeq, HUF_DDescX6 DDesc)
{
const int scaleLog = nbBitsBaseline - sizeLog; /* note : targetLog >= (nbBitsBaseline-1), hence scaleLog <= 1 */
const int minBits = nbBitsBaseline - maxWeight;
const U32 level = DDesc.nbBytes;
U32 rankVal[HUF_ABSOLUTEMAX_TABLELOG + 1];
U32 symbolStartPos, s;
/* local rankVal, will be modified */
memcpy(rankVal, rankValOrigin[consumed], sizeof(rankVal));
/* fill skipped values */
if (minWeight>1)
{
U32 i;
const U32 skipSize = rankVal[minWeight];
for (i = 0; i < skipSize; i++)
{
DSequence[i] = baseSeq;
DDescription[i] = DDesc;
}
}
/* fill DTable */
DDesc.nbBytes++;
symbolStartPos = rankStart[minWeight];
for (s=symbolStartPos; s<sortedListSize; s++)
{
const BYTE symbol = sortedSymbols[s].symbol;
const U32 weight = sortedSymbols[s].weight; /* >= 1 (sorted) */
const int nbBits = nbBitsBaseline - weight; /* >= 1 (by construction) */
const int totalBits = consumed+nbBits;
const U32 start = rankVal[weight];
const U32 length = 1 << (sizeLog-nbBits);
baseSeq.byte[level] = symbol;
DDesc.nbBits = (BYTE)totalBits;
if ((level<3) && (sizeLog-totalBits >= minBits)) /* enough room for another symbol */
{
int nextMinWeight = totalBits + scaleLog;
if (nextMinWeight < 1) nextMinWeight = 1;
HUF_fillDTableX6LevelN(DDescription+start, DSequence+start, sizeLog-nbBits,
rankValOrigin, totalBits, nextMinWeight, maxWeight,
sortedSymbols, sortedListSize, rankStart,
nbBitsBaseline, baseSeq, DDesc); /* recursive (max : level 3) */
}
else
{
U32 i;
const U32 end = start + length;
for (i = start; i < end; i++)
{
DDescription[i] = DDesc;
DSequence[i] = baseSeq;
}
}
rankVal[weight] += length;
}
}
/* note : same preparation as X4 */
static size_t HUF_readDTableX6 (U32* DTable, const void* src, size_t srcSize)
{
BYTE weightList[HUF_MAX_SYMBOL_VALUE + 1];
sortedSymbol_t sortedSymbol[HUF_MAX_SYMBOL_VALUE + 1];
U32 rankStats[HUF_ABSOLUTEMAX_TABLELOG + 1] = { 0 };
U32 rankStart0[HUF_ABSOLUTEMAX_TABLELOG + 2] = { 0 };
U32* const rankStart = rankStart0+1;
U32 tableLog, maxW, sizeOfSort, nbSymbols;
rankVal_t rankVal;
const U32 memLog = DTable[0];
size_t iSize;
if (memLog > HUF_ABSOLUTEMAX_TABLELOG) return ERROR(tableLog_tooLarge);
//memset(weightList, 0, sizeof(weightList)); /* is not necessary, even though some analyzer complain ... */
iSize = HUF_readStats(weightList, HUF_MAX_SYMBOL_VALUE + 1, rankStats, &nbSymbols, &tableLog, src, srcSize);
if (HUF_isError(iSize)) return iSize;
/* check result */
if (tableLog > memLog) return ERROR(tableLog_tooLarge); /* DTable is too small */
/* find maxWeight */
for (maxW = tableLog; rankStats[maxW]==0; maxW--) {} /* necessarily finds a solution before 0 */
/* Get start index of each weight */
{
U32 w, nextRankStart = 0;
for (w=1; w<=maxW; w++)
{
U32 current = nextRankStart;
nextRankStart += rankStats[w];
rankStart[w] = current;
}
rankStart[0] = nextRankStart; /* put all 0w symbols at the end of sorted list*/
sizeOfSort = nextRankStart;
}
/* sort symbols by weight */
{
U32 s;
for (s=0; s<nbSymbols; s++)
{
U32 w = weightList[s];
U32 r = rankStart[w]++;
sortedSymbol[r].symbol = (BYTE)s;
sortedSymbol[r].weight = (BYTE)w;
}
rankStart[0] = 0; /* forget 0w symbols; this is beginning of weight(1) */
}
/* Build rankVal */
{
const U32 minBits = tableLog+1 - maxW;
U32 nextRankVal = 0;
U32 w, consumed;
const int rescale = (memLog-tableLog) - 1; /* tableLog <= memLog */
U32* rankVal0 = rankVal[0];
for (w=1; w<=maxW; w++)
{
U32 current = nextRankVal;
nextRankVal += rankStats[w] << (w+rescale);
rankVal0[w] = current;
}
for (consumed = minBits; consumed <= memLog - minBits; consumed++)
{
U32* rankValPtr = rankVal[consumed];
for (w = 1; w <= maxW; w++)
{
rankValPtr[w] = rankVal0[w] >> consumed;
}
}
}
/* fill tables */
{
void* ddPtr = DTable+1;
HUF_DDescX6* DDescription = (HUF_DDescX6*)ddPtr;
void* dsPtr = DTable + 1 + ((size_t)1<<(memLog-1));
HUF_DSeqX6* DSequence = (HUF_DSeqX6*)dsPtr;
HUF_DSeqX6 DSeq;
HUF_DDescX6 DDesc;
DSeq.sequence = 0;
DDesc.nbBits = 0;
DDesc.nbBytes = 0;
HUF_fillDTableX6LevelN(DDescription, DSequence, memLog,
(const U32 (*)[HUF_ABSOLUTEMAX_TABLELOG + 1])rankVal, 0, 1, maxW,
sortedSymbol, sizeOfSort, rankStart0,
tableLog+1, DSeq, DDesc);
}
return iSize;
}
static U32 HUF_decodeSymbolX6(void* op, BIT_DStream_t* DStream, const HUF_DDescX6* dd, const HUF_DSeqX6* ds, const U32 dtLog)
{
const size_t val = BIT_lookBitsFast(DStream, dtLog); /* note : dtLog >= 1 */
memcpy(op, ds+val, sizeof(HUF_DSeqX6));
BIT_skipBits(DStream, dd[val].nbBits);
return dd[val].nbBytes;
}
static U32 HUF_decodeLastSymbolsX6(void* op, const U32 maxL, BIT_DStream_t* DStream,
const HUF_DDescX6* dd, const HUF_DSeqX6* ds, const U32 dtLog)
{
const size_t val = BIT_lookBitsFast(DStream, dtLog); /* note : dtLog >= 1 */
U32 length = dd[val].nbBytes;
if (length <= maxL)
{
memcpy(op, ds+val, length);
BIT_skipBits(DStream, dd[val].nbBits);
return length;
}
memcpy(op, ds+val, maxL);
if (DStream->bitsConsumed < (sizeof(DStream->bitContainer)*8))
{
BIT_skipBits(DStream, dd[val].nbBits);
if (DStream->bitsConsumed > (sizeof(DStream->bitContainer)*8))
DStream->bitsConsumed = (sizeof(DStream->bitContainer)*8); /* ugly hack; works only because it's the last symbol. Note : can't easily extract nbBits from just this symbol */
}
return maxL;
}
#define HUF_DECODE_SYMBOLX6_0(ptr, DStreamPtr) \
ptr += HUF_decodeSymbolX6(ptr, DStreamPtr, dd, ds, dtLog)
#define HUF_DECODE_SYMBOLX6_1(ptr, DStreamPtr) \
if (MEM_64bits() || (HUF_MAX_TABLELOG<=12)) \
HUF_DECODE_SYMBOLX6_0(ptr, DStreamPtr)
#define HUF_DECODE_SYMBOLX6_2(ptr, DStreamPtr) \
if (MEM_64bits()) \
HUF_DECODE_SYMBOLX6_0(ptr, DStreamPtr)
static inline size_t HUF_decodeStreamX6(BYTE* p, BIT_DStream_t* bitDPtr, BYTE* const pEnd, const U32* DTable, const U32 dtLog)
{
const void* const ddPtr = DTable+1;
const HUF_DDescX6* dd = (const HUF_DDescX6*)ddPtr;
const void* const dsPtr = DTable + 1 + ((size_t)1<<(dtLog-1));
const HUF_DSeqX6* ds = (const HUF_DSeqX6*)dsPtr;
BYTE* const pStart = p;
/* up to 16 symbols at a time */
while ((BIT_reloadDStream(bitDPtr) == BIT_DStream_unfinished) && (p <= pEnd-16))
{
HUF_DECODE_SYMBOLX6_2(p, bitDPtr);
HUF_DECODE_SYMBOLX6_1(p, bitDPtr);
HUF_DECODE_SYMBOLX6_2(p, bitDPtr);
HUF_DECODE_SYMBOLX6_0(p, bitDPtr);
}
/* closer to the end, up to 4 symbols at a time */
while ((BIT_reloadDStream(bitDPtr) == BIT_DStream_unfinished) && (p <= pEnd-4))
HUF_DECODE_SYMBOLX6_0(p, bitDPtr);
while ((BIT_reloadDStream(bitDPtr) <= BIT_DStream_endOfBuffer) && (p < pEnd))
p += HUF_decodeLastSymbolsX6(p, (U32)(pEnd-p), bitDPtr, dd, ds, dtLog);
return p-pStart;
}
static size_t HUF_decompress4X6_usingDTable(
void* dst, size_t dstSize,
const void* cSrc, size_t cSrcSize,
const U32* DTable)
{
if (cSrcSize < 10) return ERROR(corruption_detected); /* strict minimum : jump table + 1 byte per stream */
{
const BYTE* const istart = (const BYTE*) cSrc;
BYTE* const ostart = (BYTE*) dst;
BYTE* const oend = ostart + dstSize;
const U32 dtLog = DTable[0];
const void* const ddPtr = DTable+1;
const HUF_DDescX6* dd = (const HUF_DDescX6*)ddPtr;
const void* const dsPtr = DTable + 1 + ((size_t)1<<(dtLog-1));
const HUF_DSeqX6* ds = (const HUF_DSeqX6*)dsPtr;
size_t errorCode;
/* Init */
BIT_DStream_t bitD1;
BIT_DStream_t bitD2;
BIT_DStream_t bitD3;
BIT_DStream_t bitD4;
const size_t length1 = MEM_readLE16(istart);
const size_t length2 = MEM_readLE16(istart+2);
const size_t length3 = MEM_readLE16(istart+4);
size_t length4;
const BYTE* const istart1 = istart + 6; /* jumpTable */
const BYTE* const istart2 = istart1 + length1;
const BYTE* const istart3 = istart2 + length2;
const BYTE* const istart4 = istart3 + length3;
const size_t segmentSize = (dstSize+3) / 4;
BYTE* const opStart2 = ostart + segmentSize;
BYTE* const opStart3 = opStart2 + segmentSize;
BYTE* const opStart4 = opStart3 + segmentSize;
BYTE* op1 = ostart;
BYTE* op2 = opStart2;
BYTE* op3 = opStart3;
BYTE* op4 = opStart4;
U32 endSignal;
length4 = cSrcSize - (length1 + length2 + length3 + 6);
if (length4 > cSrcSize) return ERROR(corruption_detected); /* overflow */
errorCode = BIT_initDStream(&bitD1, istart1, length1);
if (HUF_isError(errorCode)) return errorCode;
errorCode = BIT_initDStream(&bitD2, istart2, length2);
if (HUF_isError(errorCode)) return errorCode;
errorCode = BIT_initDStream(&bitD3, istart3, length3);
if (HUF_isError(errorCode)) return errorCode;
errorCode = BIT_initDStream(&bitD4, istart4, length4);
if (HUF_isError(errorCode)) return errorCode;
/* 16-64 symbols per loop (4-16 symbols per stream) */
endSignal = BIT_reloadDStream(&bitD1) | BIT_reloadDStream(&bitD2) | BIT_reloadDStream(&bitD3) | BIT_reloadDStream(&bitD4);
for ( ; (op3 <= opStart4) && (endSignal==BIT_DStream_unfinished) && (op4<=(oend-16)) ; )
{
HUF_DECODE_SYMBOLX6_2(op1, &bitD1);
HUF_DECODE_SYMBOLX6_2(op2, &bitD2);
HUF_DECODE_SYMBOLX6_2(op3, &bitD3);
HUF_DECODE_SYMBOLX6_2(op4, &bitD4);
HUF_DECODE_SYMBOLX6_1(op1, &bitD1);
HUF_DECODE_SYMBOLX6_1(op2, &bitD2);
HUF_DECODE_SYMBOLX6_1(op3, &bitD3);
HUF_DECODE_SYMBOLX6_1(op4, &bitD4);
HUF_DECODE_SYMBOLX6_2(op1, &bitD1);
HUF_DECODE_SYMBOLX6_2(op2, &bitD2);
HUF_DECODE_SYMBOLX6_2(op3, &bitD3);
HUF_DECODE_SYMBOLX6_2(op4, &bitD4);
HUF_DECODE_SYMBOLX6_0(op1, &bitD1);
HUF_DECODE_SYMBOLX6_0(op2, &bitD2);
HUF_DECODE_SYMBOLX6_0(op3, &bitD3);
HUF_DECODE_SYMBOLX6_0(op4, &bitD4);
endSignal = BIT_reloadDStream(&bitD1) | BIT_reloadDStream(&bitD2) | BIT_reloadDStream(&bitD3) | BIT_reloadDStream(&bitD4);
}
/* check corruption */
if (op1 > opStart2) return ERROR(corruption_detected);
if (op2 > opStart3) return ERROR(corruption_detected);
if (op3 > opStart4) return ERROR(corruption_detected);
/* note : op4 supposed already verified within main loop */
/* finish bitStreams one by one */
HUF_decodeStreamX6(op1, &bitD1, opStart2, DTable, dtLog);
HUF_decodeStreamX6(op2, &bitD2, opStart3, DTable, dtLog);
HUF_decodeStreamX6(op3, &bitD3, opStart4, DTable, dtLog);
HUF_decodeStreamX6(op4, &bitD4, oend, DTable, dtLog);
/* check */
endSignal = BIT_endOfDStream(&bitD1) & BIT_endOfDStream(&bitD2) & BIT_endOfDStream(&bitD3) & BIT_endOfDStream(&bitD4);
if (!endSignal) return ERROR(corruption_detected);
/* decoded size */
return dstSize;
}
}
static size_t HUF_decompress4X6 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize)
{
HUF_CREATE_STATIC_DTABLEX6(DTable, HUF_MAX_TABLELOG);
const BYTE* ip = (const BYTE*) cSrc;
size_t hSize = HUF_readDTableX6 (DTable, cSrc, cSrcSize);
if (HUF_isError(hSize)) return hSize;
if (hSize >= cSrcSize) return ERROR(srcSize_wrong);
ip += hSize;
cSrcSize -= hSize;
return HUF_decompress4X6_usingDTable (dst, dstSize, ip, cSrcSize, DTable);
}
/**********************************/
/* Generic decompression selector */
/**********************************/
typedef struct { U32 tableTime; U32 decode256Time; } algo_time_t;
static const algo_time_t algoTime[16 /* Quantization */][3 /* single, double, quad */] =
{
/* single, double, quad */
{{0,0}, {1,1}, {2,2}}, /* Q==0 : impossible */
{{0,0}, {1,1}, {2,2}}, /* Q==1 : impossible */
{{ 38,130}, {1313, 74}, {2151, 38}}, /* Q == 2 : 12-18% */
{{ 448,128}, {1353, 74}, {2238, 41}}, /* Q == 3 : 18-25% */
{{ 556,128}, {1353, 74}, {2238, 47}}, /* Q == 4 : 25-32% */
{{ 714,128}, {1418, 74}, {2436, 53}}, /* Q == 5 : 32-38% */
{{ 883,128}, {1437, 74}, {2464, 61}}, /* Q == 6 : 38-44% */
{{ 897,128}, {1515, 75}, {2622, 68}}, /* Q == 7 : 44-50% */
{{ 926,128}, {1613, 75}, {2730, 75}}, /* Q == 8 : 50-56% */
{{ 947,128}, {1729, 77}, {3359, 77}}, /* Q == 9 : 56-62% */
{{1107,128}, {2083, 81}, {4006, 84}}, /* Q ==10 : 62-69% */
{{1177,128}, {2379, 87}, {4785, 88}}, /* Q ==11 : 69-75% */
{{1242,128}, {2415, 93}, {5155, 84}}, /* Q ==12 : 75-81% */
{{1349,128}, {2644,106}, {5260,106}}, /* Q ==13 : 81-87% */
{{1455,128}, {2422,124}, {4174,124}}, /* Q ==14 : 87-93% */
{{ 722,128}, {1891,145}, {1936,146}}, /* Q ==15 : 93-99% */
};
typedef size_t (*decompressionAlgo)(void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);
static size_t HUF_decompress (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize)
{
static const decompressionAlgo decompress[3] = { HUF_decompress4X2, HUF_decompress4X4, HUF_decompress4X6 };
/* estimate decompression time */
U32 Q;
const U32 D256 = (U32)(dstSize >> 8);
U32 Dtime[3];
U32 algoNb = 0;
int n;
/* validation checks */
if (dstSize == 0) return ERROR(dstSize_tooSmall);
if (cSrcSize > dstSize) return ERROR(corruption_detected); /* invalid */
if (cSrcSize == dstSize) { memcpy(dst, cSrc, dstSize); return dstSize; } /* not compressed */
if (cSrcSize == 1) { memset(dst, *(const BYTE*)cSrc, dstSize); return dstSize; } /* RLE */
/* decoder timing evaluation */
Q = (U32)(cSrcSize * 16 / dstSize); /* Q < 16 since dstSize > cSrcSize */
for (n=0; n<3; n++)
Dtime[n] = algoTime[Q][n].tableTime + (algoTime[Q][n].decode256Time * D256);
Dtime[1] += Dtime[1] >> 4; Dtime[2] += Dtime[2] >> 3; /* advantage to algorithms using less memory, for cache eviction */
if (Dtime[1] < Dtime[0]) algoNb = 1;
if (Dtime[2] < Dtime[algoNb]) algoNb = 2;
return decompress[algoNb](dst, dstSize, cSrc, cSrcSize);
//return HUF_decompress4X2(dst, dstSize, cSrc, cSrcSize); /* multi-streams single-symbol decoding */
//return HUF_decompress4X4(dst, dstSize, cSrc, cSrcSize); /* multi-streams double-symbols decoding */
//return HUF_decompress4X6(dst, dstSize, cSrc, cSrcSize); /* multi-streams quad-symbols decoding */
}
#endif /* ZSTD_CCOMMON_H_MODULE */
/*
zstd - decompression module fo v0.4 legacy format
Copyright (C) 2015-2016, Yann Collet.
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above
copyright notice, this list of conditions and the following disclaimer
in the documentation and/or other materials provided with the
distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
You can contact the author at :
- zstd source repository : https://github.com/Cyan4973/zstd
- ztsd public forum : https://groups.google.com/forum/#!forum/lz4c
*/
/* ***************************************************************
* Tuning parameters
*****************************************************************/
/*!
* HEAPMODE :
* Select how default decompression function ZSTD_decompress() will allocate memory,
* in memory stack (0), or in memory heap (1, requires malloc())
*/
#ifndef ZSTD_HEAPMODE
# define ZSTD_HEAPMODE 1
#endif
/* *******************************************************
* Includes
*********************************************************/
#include <stdlib.h> /* calloc */
#include <string.h> /* memcpy, memmove */
#include <stdio.h> /* debug : printf */
/* *******************************************************
* Compiler specifics
*********************************************************/
#ifdef _MSC_VER /* Visual Studio */
# define FORCE_INLINE static __forceinline
# include <intrin.h> /* For Visual 2005 */
# pragma warning(disable : 4127) /* disable: C4127: conditional expression is constant */
# pragma warning(disable : 4324) /* disable: C4324: padded structure */
#else
# define GCC_VERSION (__GNUC__ * 100 + __GNUC_MINOR__)
# ifdef __GNUC__
# define FORCE_INLINE static inline __attribute__((always_inline))
# else
# define FORCE_INLINE static inline
# endif
#endif
/* *************************************
* Local types
***************************************/
typedef struct
{
blockType_t blockType;
U32 origSize;
} blockProperties_t;
/* *******************************************************
* Memory operations
**********************************************************/
static void ZSTD_copy4(void* dst, const void* src) { memcpy(dst, src, 4); }
/* *************************************
* Error Management
***************************************/
/*! ZSTD_isError
* tells if a return value is an error code */
static unsigned ZSTD_isError(size_t code) { return ERR_isError(code); }
/* *************************************************************
* Context management
***************************************************************/
typedef enum { ZSTDds_getFrameHeaderSize, ZSTDds_decodeFrameHeader,
ZSTDds_decodeBlockHeader, ZSTDds_decompressBlock } ZSTD_dStage;
struct ZSTDv04_Dctx_s
{
U32 LLTable[FSE_DTABLE_SIZE_U32(LLFSELog)];
U32 OffTable[FSE_DTABLE_SIZE_U32(OffFSELog)];
U32 MLTable[FSE_DTABLE_SIZE_U32(MLFSELog)];
const void* previousDstEnd;
const void* base;
const void* vBase;
const void* dictEnd;
size_t expected;
size_t headerSize;
ZSTD_parameters params;
blockType_t bType;
ZSTD_dStage stage;
const BYTE* litPtr;
size_t litBufSize;
size_t litSize;
BYTE litBuffer[BLOCKSIZE + 8 /* margin for wildcopy */];
BYTE headerBuffer[ZSTD_frameHeaderSize_max];
}; /* typedef'd to ZSTD_DCtx within "zstd_static.h" */
static size_t ZSTD_resetDCtx(ZSTD_DCtx* dctx)
{
dctx->expected = ZSTD_frameHeaderSize_min;
dctx->stage = ZSTDds_getFrameHeaderSize;
dctx->previousDstEnd = NULL;
dctx->base = NULL;
dctx->vBase = NULL;
dctx->dictEnd = NULL;
return 0;
}
static ZSTD_DCtx* ZSTD_createDCtx(void)
{
ZSTD_DCtx* dctx = (ZSTD_DCtx*)malloc(sizeof(ZSTD_DCtx));
if (dctx==NULL) return NULL;
ZSTD_resetDCtx(dctx);
return dctx;
}
static size_t ZSTD_freeDCtx(ZSTD_DCtx* dctx)
{
free(dctx);
return 0;
}
/* *************************************************************
* Decompression section
***************************************************************/
/** ZSTD_decodeFrameHeader_Part1
* decode the 1st part of the Frame Header, which tells Frame Header size.
* srcSize must be == ZSTD_frameHeaderSize_min
* @return : the full size of the Frame Header */
static size_t ZSTD_decodeFrameHeader_Part1(ZSTD_DCtx* zc, const void* src, size_t srcSize)
{
U32 magicNumber;
if (srcSize != ZSTD_frameHeaderSize_min) return ERROR(srcSize_wrong);
magicNumber = MEM_readLE32(src);
if (magicNumber != ZSTD_MAGICNUMBER) return ERROR(prefix_unknown);
zc->headerSize = ZSTD_frameHeaderSize_min;
return zc->headerSize;
}
static size_t ZSTD_getFrameParams(ZSTD_parameters* params, const void* src, size_t srcSize)
{
U32 magicNumber;
if (srcSize < ZSTD_frameHeaderSize_min) return ZSTD_frameHeaderSize_max;
magicNumber = MEM_readLE32(src);
if (magicNumber != ZSTD_MAGICNUMBER) return ERROR(prefix_unknown);
memset(params, 0, sizeof(*params));
params->windowLog = (((const BYTE*)src)[4] & 15) + ZSTD_WINDOWLOG_ABSOLUTEMIN;
if ((((const BYTE*)src)[4] >> 4) != 0) return ERROR(frameParameter_unsupported); /* reserved bits */
return 0;
}
/** ZSTD_decodeFrameHeader_Part2
* decode the full Frame Header
* srcSize must be the size provided by ZSTD_decodeFrameHeader_Part1
* @return : 0, or an error code, which can be tested using ZSTD_isError() */
static size_t ZSTD_decodeFrameHeader_Part2(ZSTD_DCtx* zc, const void* src, size_t srcSize)
{
size_t result;
if (srcSize != zc->headerSize) return ERROR(srcSize_wrong);
result = ZSTD_getFrameParams(&(zc->params), src, srcSize);
if ((MEM_32bits()) && (zc->params.windowLog > 25)) return ERROR(frameParameter_unsupportedBy32bitsImplementation);
return result;
}
static size_t ZSTD_getcBlockSize(const void* src, size_t srcSize, blockProperties_t* bpPtr)
{
const BYTE* const in = (const BYTE* const)src;
BYTE headerFlags;
U32 cSize;
if (srcSize < 3) return ERROR(srcSize_wrong);
headerFlags = *in;
cSize = in[2] + (in[1]<<8) + ((in[0] & 7)<<16);
bpPtr->blockType = (blockType_t)(headerFlags >> 6);
bpPtr->origSize = (bpPtr->blockType == bt_rle) ? cSize : 0;
if (bpPtr->blockType == bt_end) return 0;
if (bpPtr->blockType == bt_rle) return 1;
return cSize;
}
static size_t ZSTD_copyRawBlock(void* dst, size_t maxDstSize, const void* src, size_t srcSize)
{
if (srcSize > maxDstSize) return ERROR(dstSize_tooSmall);
memcpy(dst, src, srcSize);
return srcSize;
}
/** ZSTD_decompressLiterals
@return : nb of bytes read from src, or an error code*/
static size_t ZSTD_decompressLiterals(void* dst, size_t* maxDstSizePtr,
const void* src, size_t srcSize)
{
const BYTE* ip = (const BYTE*)src;
const size_t litSize = (MEM_readLE32(src) & 0x1FFFFF) >> 2; /* no buffer issue : srcSize >= MIN_CBLOCK_SIZE */
const size_t litCSize = (MEM_readLE32(ip+2) & 0xFFFFFF) >> 5; /* no buffer issue : srcSize >= MIN_CBLOCK_SIZE */
if (litSize > *maxDstSizePtr) return ERROR(corruption_detected);
if (litCSize + 5 > srcSize) return ERROR(corruption_detected);
if (HUF_isError(HUF_decompress(dst, litSize, ip+5, litCSize))) return ERROR(corruption_detected);
*maxDstSizePtr = litSize;
return litCSize + 5;
}
/** ZSTD_decodeLiteralsBlock
@return : nb of bytes read from src (< srcSize ) */
static size_t ZSTD_decodeLiteralsBlock(ZSTD_DCtx* dctx,
const void* src, size_t srcSize) /* note : srcSize < BLOCKSIZE */
{
const BYTE* const istart = (const BYTE*) src;
/* any compressed block with literals segment must be at least this size */
if (srcSize < MIN_CBLOCK_SIZE) return ERROR(corruption_detected);
switch(*istart & 3)
{
/* compressed */
case 0:
{
size_t litSize = BLOCKSIZE;
const size_t readSize = ZSTD_decompressLiterals(dctx->litBuffer, &litSize, src, srcSize);
dctx->litPtr = dctx->litBuffer;
dctx->litBufSize = BLOCKSIZE+8;
dctx->litSize = litSize;
return readSize; /* works if it's an error too */
}
case IS_RAW:
{
const size_t litSize = (MEM_readLE32(istart) & 0xFFFFFF) >> 2; /* no buffer issue : srcSize >= MIN_CBLOCK_SIZE */
if (litSize > srcSize-11) /* risk of reading too far with wildcopy */
{
if (litSize > srcSize-3) return ERROR(corruption_detected);
memcpy(dctx->litBuffer, istart, litSize);
dctx->litPtr = dctx->litBuffer;
dctx->litBufSize = BLOCKSIZE+8;
dctx->litSize = litSize;
return litSize+3;
}
/* direct reference into compressed stream */
dctx->litPtr = istart+3;
dctx->litBufSize = srcSize-3;
dctx->litSize = litSize;
return litSize+3; }
case IS_RLE:
{
const size_t litSize = (MEM_readLE32(istart) & 0xFFFFFF) >> 2; /* no buffer issue : srcSize >= MIN_CBLOCK_SIZE */
if (litSize > BLOCKSIZE) return ERROR(corruption_detected);
memset(dctx->litBuffer, istart[3], litSize);
dctx->litPtr = dctx->litBuffer;
dctx->litBufSize = BLOCKSIZE+8;
dctx->litSize = litSize;
return 4;
}
default:
return ERROR(corruption_detected); /* forbidden nominal case */
}
}
static size_t ZSTD_decodeSeqHeaders(int* nbSeq, const BYTE** dumpsPtr, size_t* dumpsLengthPtr,
FSE_DTable* DTableLL, FSE_DTable* DTableML, FSE_DTable* DTableOffb,
const void* src, size_t srcSize)
{
const BYTE* const istart = (const BYTE* const)src;
const BYTE* ip = istart;
const BYTE* const iend = istart + srcSize;
U32 LLtype, Offtype, MLtype;
U32 LLlog, Offlog, MLlog;
size_t dumpsLength;
/* check */
if (srcSize < 5) return ERROR(srcSize_wrong);
/* SeqHead */
*nbSeq = MEM_readLE16(ip); ip+=2;
LLtype = *ip >> 6;
Offtype = (*ip >> 4) & 3;
MLtype = (*ip >> 2) & 3;
if (*ip & 2)
{
dumpsLength = ip[2];
dumpsLength += ip[1] << 8;
ip += 3;
}
else
{
dumpsLength = ip[1];
dumpsLength += (ip[0] & 1) << 8;
ip += 2;
}
*dumpsPtr = ip;
ip += dumpsLength;
*dumpsLengthPtr = dumpsLength;
/* check */
if (ip > iend-3) return ERROR(srcSize_wrong); /* min : all 3 are "raw", hence no header, but at least xxLog bits per type */
/* sequences */
{
S16 norm[MaxML+1]; /* assumption : MaxML >= MaxLL >= MaxOff */
size_t headerSize;
/* Build DTables */
switch(LLtype)
{
U32 max;
case bt_rle :
LLlog = 0;
FSE_buildDTable_rle(DTableLL, *ip++); break;
case bt_raw :
LLlog = LLbits;
FSE_buildDTable_raw(DTableLL, LLbits); break;
default :
max = MaxLL;
headerSize = FSE_readNCount(norm, &max, &LLlog, ip, iend-ip);
if (FSE_isError(headerSize)) return ERROR(GENERIC);
if (LLlog > LLFSELog) return ERROR(corruption_detected);
ip += headerSize;
FSE_buildDTable(DTableLL, norm, max, LLlog);
}
switch(Offtype)
{
U32 max;
case bt_rle :
Offlog = 0;
if (ip > iend-2) return ERROR(srcSize_wrong); /* min : "raw", hence no header, but at least xxLog bits */
FSE_buildDTable_rle(DTableOffb, *ip++ & MaxOff); /* if *ip > MaxOff, data is corrupted */
break;
case bt_raw :
Offlog = Offbits;
FSE_buildDTable_raw(DTableOffb, Offbits); break;
default :
max = MaxOff;
headerSize = FSE_readNCount(norm, &max, &Offlog, ip, iend-ip);
if (FSE_isError(headerSize)) return ERROR(GENERIC);
if (Offlog > OffFSELog) return ERROR(corruption_detected);
ip += headerSize;
FSE_buildDTable(DTableOffb, norm, max, Offlog);
}
switch(MLtype)
{
U32 max;
case bt_rle :
MLlog = 0;
if (ip > iend-2) return ERROR(srcSize_wrong); /* min : "raw", hence no header, but at least xxLog bits */
FSE_buildDTable_rle(DTableML, *ip++); break;
case bt_raw :
MLlog = MLbits;
FSE_buildDTable_raw(DTableML, MLbits); break;
default :
max = MaxML;
headerSize = FSE_readNCount(norm, &max, &MLlog, ip, iend-ip);
if (FSE_isError(headerSize)) return ERROR(GENERIC);
if (MLlog > MLFSELog) return ERROR(corruption_detected);
ip += headerSize;
FSE_buildDTable(DTableML, norm, max, MLlog);
}
}
return ip-istart;
}
typedef struct {
size_t litLength;
size_t offset;
size_t matchLength;
} seq_t;
typedef struct {
BIT_DStream_t DStream;
FSE_DState_t stateLL;
FSE_DState_t stateOffb;
FSE_DState_t stateML;
size_t prevOffset;
const BYTE* dumps;
const BYTE* dumpsEnd;
} seqState_t;
static void ZSTD_decodeSequence(seq_t* seq, seqState_t* seqState)
{
size_t litLength;
size_t prevOffset;
size_t offset;
size_t matchLength;
const BYTE* dumps = seqState->dumps;
const BYTE* const de = seqState->dumpsEnd;
/* Literal length */
litLength = FSE_decodeSymbol(&(seqState->stateLL), &(seqState->DStream));
prevOffset = litLength ? seq->offset : seqState->prevOffset;
if (litLength == MaxLL)
{
U32 add = *dumps++;
if (add < 255) litLength += add;
else
{
litLength = MEM_readLE32(dumps) & 0xFFFFFF; /* no pb : dumps is always followed by seq tables > 1 byte */
dumps += 3;
}
if (dumps >= de) dumps = de-1; /* late correction, to avoid read overflow (data is now corrupted anyway) */
}
/* Offset */
{
static const U32 offsetPrefix[MaxOff+1] = {
1 /*fake*/, 1, 2, 4, 8, 16, 32, 64, 128, 256,
512, 1024, 2048, 4096, 8192, 16384, 32768, 65536, 131072, 262144,
524288, 1048576, 2097152, 4194304, 8388608, 16777216, 33554432, /*fake*/ 1, 1, 1, 1, 1 };
U32 offsetCode, nbBits;
offsetCode = FSE_decodeSymbol(&(seqState->stateOffb), &(seqState->DStream)); /* <= maxOff, by table construction */
if (MEM_32bits()) BIT_reloadDStream(&(seqState->DStream));
nbBits = offsetCode - 1;
if (offsetCode==0) nbBits = 0; /* cmove */
offset = offsetPrefix[offsetCode] + BIT_readBits(&(seqState->DStream), nbBits);
if (MEM_32bits()) BIT_reloadDStream(&(seqState->DStream));
if (offsetCode==0) offset = prevOffset; /* cmove */
if (offsetCode | !litLength) seqState->prevOffset = seq->offset; /* cmove */
}
/* MatchLength */
matchLength = FSE_decodeSymbol(&(seqState->stateML), &(seqState->DStream));
if (matchLength == MaxML)
{
U32 add = *dumps++;
if (add < 255) matchLength += add;
else
{
matchLength = MEM_readLE32(dumps) & 0xFFFFFF; /* no pb : dumps is always followed by seq tables > 1 byte */
dumps += 3;
}
if (dumps >= de) dumps = de-1; /* late correction, to avoid read overflow (data is now corrupted anyway) */
}
matchLength += MINMATCH;
/* save result */
seq->litLength = litLength;
seq->offset = offset;
seq->matchLength = matchLength;
seqState->dumps = dumps;
}
static size_t ZSTD_execSequence(BYTE* op,
BYTE* const oend, seq_t sequence,
const BYTE** litPtr, const BYTE* const litLimit_8,
const BYTE* const base, const BYTE* const vBase, const BYTE* const dictEnd)
{
static const int dec32table[] = { 0, 1, 2, 1, 4, 4, 4, 4 }; /* added */
static const int dec64table[] = { 8, 8, 8, 7, 8, 9,10,11 }; /* substracted */
BYTE* const oLitEnd = op + sequence.litLength;
const size_t sequenceLength = sequence.litLength + sequence.matchLength;
BYTE* const oMatchEnd = op + sequenceLength; /* risk : address space overflow (32-bits) */
BYTE* const oend_8 = oend-8;
const BYTE* const litEnd = *litPtr + sequence.litLength;
const BYTE* match = oLitEnd - sequence.offset;
/* check */
if (oLitEnd > oend_8) return ERROR(dstSize_tooSmall); /* last match must start at a minimum distance of 8 from oend */
if (oMatchEnd > oend) return ERROR(dstSize_tooSmall); /* overwrite beyond dst buffer */
if (litEnd > litLimit_8) return ERROR(corruption_detected); /* risk read beyond lit buffer */
/* copy Literals */
ZSTD_wildcopy(op, *litPtr, sequence.litLength); /* note : oLitEnd <= oend-8 : no risk of overwrite beyond oend */
op = oLitEnd;
*litPtr = litEnd; /* update for next sequence */
/* copy Match */
if (sequence.offset > (size_t)(oLitEnd - base))
{
/* offset beyond prefix */
if (sequence.offset > (size_t)(oLitEnd - vBase))
return ERROR(corruption_detected);
match = dictEnd - (base-match);
if (match + sequence.matchLength <= dictEnd)
{
memmove(oLitEnd, match, sequence.matchLength);
return sequenceLength;
}
/* span extDict & currentPrefixSegment */
{
size_t length1 = dictEnd - match;
memmove(oLitEnd, match, length1);
op = oLitEnd + length1;
sequence.matchLength -= length1;
match = base;
}
}
/* match within prefix */
if (sequence.offset < 8)
{
/* close range match, overlap */
const int sub2 = dec64table[sequence.offset];
op[0] = match[0];
op[1] = match[1];
op[2] = match[2];
op[3] = match[3];
match += dec32table[sequence.offset];
ZSTD_copy4(op+4, match);
match -= sub2;
}
else
{
ZSTD_copy8(op, match);
}
op += 8; match += 8;
if (oMatchEnd > oend-12)
{
if (op < oend_8)
{
ZSTD_wildcopy(op, match, oend_8 - op);
match += oend_8 - op;
op = oend_8;
}
while (op < oMatchEnd) *op++ = *match++;
}
else
{
ZSTD_wildcopy(op, match, sequence.matchLength-8); /* works even if matchLength < 8 */
}
return sequenceLength;
}
static size_t ZSTD_decompressSequences(
ZSTD_DCtx* dctx,
void* dst, size_t maxDstSize,
const void* seqStart, size_t seqSize)
{
const BYTE* ip = (const BYTE*)seqStart;
const BYTE* const iend = ip + seqSize;
BYTE* const ostart = (BYTE* const)dst;
BYTE* op = ostart;
BYTE* const oend = ostart + maxDstSize;
size_t errorCode, dumpsLength;
const BYTE* litPtr = dctx->litPtr;
const BYTE* const litLimit_8 = litPtr + dctx->litBufSize - 8;
const BYTE* const litEnd = litPtr + dctx->litSize;
int nbSeq;
const BYTE* dumps;
U32* DTableLL = dctx->LLTable;
U32* DTableML = dctx->MLTable;
U32* DTableOffb = dctx->OffTable;
const BYTE* const base = (const BYTE*) (dctx->base);
const BYTE* const vBase = (const BYTE*) (dctx->vBase);
const BYTE* const dictEnd = (const BYTE*) (dctx->dictEnd);
/* Build Decoding Tables */
errorCode = ZSTD_decodeSeqHeaders(&nbSeq, &dumps, &dumpsLength,
DTableLL, DTableML, DTableOffb,
ip, iend-ip);
if (ZSTD_isError(errorCode)) return errorCode;
ip += errorCode;
/* Regen sequences */
{
seq_t sequence;
seqState_t seqState;
memset(&sequence, 0, sizeof(sequence));
sequence.offset = 4;
seqState.dumps = dumps;
seqState.dumpsEnd = dumps + dumpsLength;
seqState.prevOffset = 4;
errorCode = BIT_initDStream(&(seqState.DStream), ip, iend-ip);
if (ERR_isError(errorCode)) return ERROR(corruption_detected);
FSE_initDState(&(seqState.stateLL), &(seqState.DStream), DTableLL);
FSE_initDState(&(seqState.stateOffb), &(seqState.DStream), DTableOffb);
FSE_initDState(&(seqState.stateML), &(seqState.DStream), DTableML);
for ( ; (BIT_reloadDStream(&(seqState.DStream)) <= BIT_DStream_completed) && nbSeq ; )
{
size_t oneSeqSize;
nbSeq--;
ZSTD_decodeSequence(&sequence, &seqState);
oneSeqSize = ZSTD_execSequence(op, oend, sequence, &litPtr, litLimit_8, base, vBase, dictEnd);
if (ZSTD_isError(oneSeqSize)) return oneSeqSize;
op += oneSeqSize;
}
/* check if reached exact end */
if ( !BIT_endOfDStream(&(seqState.DStream)) ) return ERROR(corruption_detected); /* DStream should be entirely and exactly consumed; otherwise data is corrupted */
/* last literal segment */
{
size_t lastLLSize = litEnd - litPtr;
if (litPtr > litEnd) return ERROR(corruption_detected);
if (op+lastLLSize > oend) return ERROR(dstSize_tooSmall);
if (op != litPtr) memcpy(op, litPtr, lastLLSize);
op += lastLLSize;
}
}
return op-ostart;
}
static void ZSTD_checkContinuity(ZSTD_DCtx* dctx, const void* dst)
{
if (dst != dctx->previousDstEnd) /* not contiguous */
{
dctx->dictEnd = dctx->previousDstEnd;
dctx->vBase = (const char*)dst - ((const char*)(dctx->previousDstEnd) - (const char*)(dctx->base));
dctx->base = dst;
dctx->previousDstEnd = dst;
}
}
static size_t ZSTD_decompressBlock_internal(ZSTD_DCtx* dctx,
void* dst, size_t maxDstSize,
const void* src, size_t srcSize)
{
/* blockType == blockCompressed */
const BYTE* ip = (const BYTE*)src;
/* Decode literals sub-block */
size_t litCSize = ZSTD_decodeLiteralsBlock(dctx, src, srcSize);
if (ZSTD_isError(litCSize)) return litCSize;
ip += litCSize;
srcSize -= litCSize;
return ZSTD_decompressSequences(dctx, dst, maxDstSize, ip, srcSize);
}
static size_t ZSTD_decompress_usingDict(ZSTD_DCtx* ctx,
void* dst, size_t maxDstSize,
const void* src, size_t srcSize,
const void* dict, size_t dictSize)
{
const BYTE* ip = (const BYTE*)src;
const BYTE* iend = ip + srcSize;
BYTE* const ostart = (BYTE* const)dst;
BYTE* op = ostart;
BYTE* const oend = ostart + maxDstSize;
size_t remainingSize = srcSize;
blockProperties_t blockProperties;
/* init */
ZSTD_resetDCtx(ctx);
if (dict)
{
ZSTD_decompress_insertDictionary(ctx, dict, dictSize);
ctx->dictEnd = ctx->previousDstEnd;
ctx->vBase = (const char*)dst - ((const char*)(ctx->previousDstEnd) - (const char*)(ctx->base));
ctx->base = dst;
}
else
{
ctx->vBase = ctx->base = ctx->dictEnd = dst;
}
/* Frame Header */
{
size_t frameHeaderSize;
if (srcSize < ZSTD_frameHeaderSize_min+ZSTD_blockHeaderSize) return ERROR(srcSize_wrong);
frameHeaderSize = ZSTD_decodeFrameHeader_Part1(ctx, src, ZSTD_frameHeaderSize_min);
if (ZSTD_isError(frameHeaderSize)) return frameHeaderSize;
if (srcSize < frameHeaderSize+ZSTD_blockHeaderSize) return ERROR(srcSize_wrong);
ip += frameHeaderSize; remainingSize -= frameHeaderSize;
frameHeaderSize = ZSTD_decodeFrameHeader_Part2(ctx, src, frameHeaderSize);
if (ZSTD_isError(frameHeaderSize)) return frameHeaderSize;
}
/* Loop on each block */
while (1)
{
size_t decodedSize=0;
size_t cBlockSize = ZSTD_getcBlockSize(ip, iend-ip, &blockProperties);
if (ZSTD_isError(cBlockSize)) return cBlockSize;
ip += ZSTD_blockHeaderSize;
remainingSize -= ZSTD_blockHeaderSize;
if (cBlockSize > remainingSize) return ERROR(srcSize_wrong);
switch(blockProperties.blockType)
{
case bt_compressed:
decodedSize = ZSTD_decompressBlock_internal(ctx, op, oend-op, ip, cBlockSize);
break;
case bt_raw :
decodedSize = ZSTD_copyRawBlock(op, oend-op, ip, cBlockSize);
break;
case bt_rle :
return ERROR(GENERIC); /* not yet supported */
break;
case bt_end :
/* end of frame */
if (remainingSize) return ERROR(srcSize_wrong);
break;
default:
return ERROR(GENERIC); /* impossible */
}
if (cBlockSize == 0) break; /* bt_end */
if (ZSTD_isError(decodedSize)) return decodedSize;
op += decodedSize;
ip += cBlockSize;
remainingSize -= cBlockSize;
}
return op-ostart;
}
/* ******************************
* Streaming Decompression API
********************************/
static size_t ZSTD_nextSrcSizeToDecompress(ZSTD_DCtx* dctx)
{
return dctx->expected;
}
static size_t ZSTD_decompressContinue(ZSTD_DCtx* ctx, void* dst, size_t maxDstSize, const void* src, size_t srcSize)
{
/* Sanity check */
if (srcSize != ctx->expected) return ERROR(srcSize_wrong);
ZSTD_checkContinuity(ctx, dst);
/* Decompress : frame header; part 1 */
switch (ctx->stage)
{
case ZSTDds_getFrameHeaderSize :
{
/* get frame header size */
if (srcSize != ZSTD_frameHeaderSize_min) return ERROR(srcSize_wrong); /* impossible */
ctx->headerSize = ZSTD_decodeFrameHeader_Part1(ctx, src, ZSTD_frameHeaderSize_min);
if (ZSTD_isError(ctx->headerSize)) return ctx->headerSize;
memcpy(ctx->headerBuffer, src, ZSTD_frameHeaderSize_min);
if (ctx->headerSize > ZSTD_frameHeaderSize_min)
{
ctx->expected = ctx->headerSize - ZSTD_frameHeaderSize_min;
ctx->stage = ZSTDds_decodeFrameHeader;
return 0;
}
ctx->expected = 0; /* not necessary to copy more */
}
case ZSTDds_decodeFrameHeader:
{
/* get frame header */
size_t result;
memcpy(ctx->headerBuffer + ZSTD_frameHeaderSize_min, src, ctx->expected);
result = ZSTD_decodeFrameHeader_Part2(ctx, ctx->headerBuffer, ctx->headerSize);
if (ZSTD_isError(result)) return result;
ctx->expected = ZSTD_blockHeaderSize;
ctx->stage = ZSTDds_decodeBlockHeader;
return 0;
}
case ZSTDds_decodeBlockHeader:
{
/* Decode block header */
blockProperties_t bp;
size_t blockSize = ZSTD_getcBlockSize(src, ZSTD_blockHeaderSize, &bp);
if (ZSTD_isError(blockSize)) return blockSize;
if (bp.blockType == bt_end)
{
ctx->expected = 0;
ctx->stage = ZSTDds_getFrameHeaderSize;
}
else
{
ctx->expected = blockSize;
ctx->bType = bp.blockType;
ctx->stage = ZSTDds_decompressBlock;
}
return 0;
}
case ZSTDds_decompressBlock:
{
/* Decompress : block content */
size_t rSize;
switch(ctx->bType)
{
case bt_compressed:
rSize = ZSTD_decompressBlock_internal(ctx, dst, maxDstSize, src, srcSize);
break;
case bt_raw :
rSize = ZSTD_copyRawBlock(dst, maxDstSize, src, srcSize);
break;
case bt_rle :
return ERROR(GENERIC); /* not yet handled */
break;
case bt_end : /* should never happen (filtered at phase 1) */
rSize = 0;
break;
default:
return ERROR(GENERIC);
}
ctx->stage = ZSTDds_decodeBlockHeader;
ctx->expected = ZSTD_blockHeaderSize;
ctx->previousDstEnd = (char*)dst + rSize;
return rSize;
}
default:
return ERROR(GENERIC); /* impossible */
}
}
static void ZSTD_decompress_insertDictionary(ZSTD_DCtx* ctx, const void* dict, size_t dictSize)
{
ctx->dictEnd = ctx->previousDstEnd;
ctx->vBase = (const char*)dict - ((const char*)(ctx->previousDstEnd) - (const char*)(ctx->base));
ctx->base = dict;
ctx->previousDstEnd = (const char*)dict + dictSize;
}
/*
Buffered version of Zstd compression library
Copyright (C) 2015, Yann Collet.
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above
copyright notice, this list of conditions and the following disclaimer
in the documentation and/or other materials provided with the
distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
You can contact the author at :
- zstd source repository : https://github.com/Cyan4973/zstd
- ztsd public forum : https://groups.google.com/forum/#!forum/lz4c
*/
/* The objects defined into this file should be considered experimental.
* They are not labelled stable, as their prototype may change in the future.
* You can use them for tests, provide feedback, or if you can endure risk of future changes.
*/
/* *************************************
* Includes
***************************************/
#include <stdlib.h>
/** ************************************************
* Streaming decompression
*
* A ZBUFF_DCtx object is required to track streaming operation.
* Use ZBUFF_createDCtx() and ZBUFF_freeDCtx() to create/release resources.
* Use ZBUFF_decompressInit() to start a new decompression operation.
* ZBUFF_DCtx objects can be reused multiple times.
*
* Use ZBUFF_decompressContinue() repetitively to consume your input.
* *srcSizePtr and *maxDstSizePtr can be any size.
* The function will report how many bytes were read or written by modifying *srcSizePtr and *maxDstSizePtr.
* Note that it may not consume the entire input, in which case it's up to the caller to call again the function with remaining input.
* The content of dst will be overwritten (up to *maxDstSizePtr) at each function call, so save its content if it matters or change dst .
* @return : a hint to preferred nb of bytes to use as input for next function call (it's only a hint, to improve latency)
* or 0 when a frame is completely decoded
* or an error code, which can be tested using ZBUFF_isError().
*
* Hint : recommended buffer sizes (not compulsory)
* output : 128 KB block size is the internal unit, it ensures it's always possible to write a full block when it's decoded.
* input : just follow indications from ZBUFF_decompressContinue() to minimize latency. It should always be <= 128 KB + 3 .
* **************************************************/
typedef enum { ZBUFFds_init, ZBUFFds_readHeader, ZBUFFds_loadHeader, ZBUFFds_decodeHeader,
ZBUFFds_read, ZBUFFds_load, ZBUFFds_flush } ZBUFF_dStage;
/* *** Resource management *** */
#define ZSTD_frameHeaderSize_max 5 /* too magical, should come from reference */
struct ZBUFFv04_DCtx_s {
ZSTD_DCtx* zc;
ZSTD_parameters params;
char* inBuff;
size_t inBuffSize;
size_t inPos;
char* outBuff;
size_t outBuffSize;
size_t outStart;
size_t outEnd;
size_t hPos;
const char* dict;
size_t dictSize;
ZBUFF_dStage stage;
unsigned char headerBuffer[ZSTD_frameHeaderSize_max];
}; /* typedef'd to ZBUFF_DCtx within "zstd_buffered.h" */
typedef ZBUFFv04_DCtx ZBUFF_DCtx;
static ZBUFF_DCtx* ZBUFF_createDCtx(void)
{
ZBUFF_DCtx* zbc = (ZBUFF_DCtx*)malloc(sizeof(ZBUFF_DCtx));
if (zbc==NULL) return NULL;
memset(zbc, 0, sizeof(*zbc));
zbc->zc = ZSTD_createDCtx();
zbc->stage = ZBUFFds_init;
return zbc;
}
static size_t ZBUFF_freeDCtx(ZBUFF_DCtx* zbc)
{
if (zbc==NULL) return 0; /* support free on null */
ZSTD_freeDCtx(zbc->zc);
free(zbc->inBuff);
free(zbc->outBuff);
free(zbc);
return 0;
}
/* *** Initialization *** */
static size_t ZBUFF_decompressInit(ZBUFF_DCtx* zbc)
{
zbc->stage = ZBUFFds_readHeader;
zbc->hPos = zbc->inPos = zbc->outStart = zbc->outEnd = zbc->dictSize = 0;
return ZSTD_resetDCtx(zbc->zc);
}
static size_t ZBUFF_decompressWithDictionary(ZBUFF_DCtx* zbc, const void* src, size_t srcSize)
{
zbc->dict = (const char*)src;
zbc->dictSize = srcSize;
return 0;
}
static size_t ZBUFF_limitCopy(void* dst, size_t maxDstSize, const void* src, size_t srcSize)
{
size_t length = MIN(maxDstSize, srcSize);
memcpy(dst, src, length);
return length;
}
/* *** Decompression *** */
static size_t ZBUFF_decompressContinue(ZBUFF_DCtx* zbc, void* dst, size_t* maxDstSizePtr, const void* src, size_t* srcSizePtr)
{
const char* const istart = (const char*)src;
const char* ip = istart;
const char* const iend = istart + *srcSizePtr;
char* const ostart = (char*)dst;
char* op = ostart;
char* const oend = ostart + *maxDstSizePtr;
U32 notDone = 1;
while (notDone)
{
switch(zbc->stage)
{
case ZBUFFds_init :
return ERROR(init_missing);
case ZBUFFds_readHeader :
/* read header from src */
{
size_t headerSize = ZSTD_getFrameParams(&(zbc->params), src, *srcSizePtr);
if (ZSTD_isError(headerSize)) return headerSize;
if (headerSize)
{
/* not enough input to decode header : tell how many bytes would be necessary */
memcpy(zbc->headerBuffer+zbc->hPos, src, *srcSizePtr);
zbc->hPos += *srcSizePtr;
*maxDstSizePtr = 0;
zbc->stage = ZBUFFds_loadHeader;
return headerSize - zbc->hPos;
}
zbc->stage = ZBUFFds_decodeHeader;
break;
}
case ZBUFFds_loadHeader:
/* complete header from src */
{
size_t headerSize = ZBUFF_limitCopy(
zbc->headerBuffer + zbc->hPos, ZSTD_frameHeaderSize_max - zbc->hPos,
src, *srcSizePtr);
zbc->hPos += headerSize;
ip += headerSize;
headerSize = ZSTD_getFrameParams(&(zbc->params), zbc->headerBuffer, zbc->hPos);
if (ZSTD_isError(headerSize)) return headerSize;
if (headerSize)
{
/* not enough input to decode header : tell how many bytes would be necessary */
*maxDstSizePtr = 0;
return headerSize - zbc->hPos;
}
// zbc->stage = ZBUFFds_decodeHeader; break; /* useless : stage follows */
}
case ZBUFFds_decodeHeader:
/* apply header to create / resize buffers */
{
size_t neededOutSize = (size_t)1 << zbc->params.windowLog;
size_t neededInSize = BLOCKSIZE; /* a block is never > BLOCKSIZE */
if (zbc->inBuffSize < neededInSize)
{
free(zbc->inBuff);
zbc->inBuffSize = neededInSize;
zbc->inBuff = (char*)malloc(neededInSize);
if (zbc->inBuff == NULL) return ERROR(memory_allocation);
}
if (zbc->outBuffSize < neededOutSize)
{
free(zbc->outBuff);
zbc->outBuffSize = neededOutSize;
zbc->outBuff = (char*)malloc(neededOutSize);
if (zbc->outBuff == NULL) return ERROR(memory_allocation);
}
}
if (zbc->dictSize)
ZSTD_decompress_insertDictionary(zbc->zc, zbc->dict, zbc->dictSize);
if (zbc->hPos)
{
/* some data already loaded into headerBuffer : transfer into inBuff */
memcpy(zbc->inBuff, zbc->headerBuffer, zbc->hPos);
zbc->inPos = zbc->hPos;
zbc->hPos = 0;
zbc->stage = ZBUFFds_load;
break;
}
zbc->stage = ZBUFFds_read;
case ZBUFFds_read:
{
size_t neededInSize = ZSTD_nextSrcSizeToDecompress(zbc->zc);
if (neededInSize==0) /* end of frame */
{
zbc->stage = ZBUFFds_init;
notDone = 0;
break;
}
if ((size_t)(iend-ip) >= neededInSize)
{
/* directly decode from src */
size_t decodedSize = ZSTD_decompressContinue(zbc->zc,
zbc->outBuff + zbc->outStart, zbc->outBuffSize - zbc->outStart,
ip, neededInSize);
if (ZSTD_isError(decodedSize)) return decodedSize;
ip += neededInSize;
if (!decodedSize) break; /* this was just a header */
zbc->outEnd = zbc->outStart + decodedSize;
zbc->stage = ZBUFFds_flush;
break;
}
if (ip==iend) { notDone = 0; break; } /* no more input */
zbc->stage = ZBUFFds_load;
}
case ZBUFFds_load:
{
size_t neededInSize = ZSTD_nextSrcSizeToDecompress(zbc->zc);
size_t toLoad = neededInSize - zbc->inPos; /* should always be <= remaining space within inBuff */
size_t loadedSize;
if (toLoad > zbc->inBuffSize - zbc->inPos) return ERROR(corruption_detected); /* should never happen */
loadedSize = ZBUFF_limitCopy(zbc->inBuff + zbc->inPos, toLoad, ip, iend-ip);
ip += loadedSize;
zbc->inPos += loadedSize;
if (loadedSize < toLoad) { notDone = 0; break; } /* not enough input, wait for more */
{
size_t decodedSize = ZSTD_decompressContinue(zbc->zc,
zbc->outBuff + zbc->outStart, zbc->outBuffSize - zbc->outStart,
zbc->inBuff, neededInSize);
if (ZSTD_isError(decodedSize)) return decodedSize;
zbc->inPos = 0; /* input is consumed */
if (!decodedSize) { zbc->stage = ZBUFFds_read; break; } /* this was just a header */
zbc->outEnd = zbc->outStart + decodedSize;
zbc->stage = ZBUFFds_flush;
// break; /* ZBUFFds_flush follows */
}
}
case ZBUFFds_flush:
{
size_t toFlushSize = zbc->outEnd - zbc->outStart;
size_t flushedSize = ZBUFF_limitCopy(op, oend-op, zbc->outBuff + zbc->outStart, toFlushSize);
op += flushedSize;
zbc->outStart += flushedSize;
if (flushedSize == toFlushSize)
{
zbc->stage = ZBUFFds_read;
if (zbc->outStart + BLOCKSIZE > zbc->outBuffSize)
zbc->outStart = zbc->outEnd = 0;
break;
}
/* cannot flush everything */
notDone = 0;
break;
}
default: return ERROR(GENERIC); /* impossible */
}
}
*srcSizePtr = ip-istart;
*maxDstSizePtr = op-ostart;
{
size_t nextSrcSizeHint = ZSTD_nextSrcSizeToDecompress(zbc->zc);
if (nextSrcSizeHint > 3) nextSrcSizeHint+= 3; /* get the next block header while at it */
nextSrcSizeHint -= zbc->inPos; /* already loaded*/
return nextSrcSizeHint;
}
}
/* *************************************
* Tool functions
***************************************/
unsigned ZBUFFv04_isError(size_t errorCode) { return ERR_isError(errorCode); }
const char* ZBUFFv04_getErrorName(size_t errorCode) { return ERR_getErrorName(errorCode); }
size_t ZBUFFv04_recommendedDInSize() { return BLOCKSIZE + 3; }
size_t ZBUFFv04_recommendedDOutSize() { return BLOCKSIZE; }
/*- ========================================================================= -*/
/* final wrapping stage */
size_t ZSTDv04_decompressDCtx(ZSTD_DCtx* dctx, void* dst, size_t maxDstSize, const void* src, size_t srcSize)
{
return ZSTD_decompress_usingDict(dctx, dst, maxDstSize, src, srcSize, NULL, 0);
}
size_t ZSTDv04_decompress(void* dst, size_t maxDstSize, const void* src, size_t srcSize)
{
#if defined(ZSTD_HEAPMODE) && (ZSTD_HEAPMODE==1)
size_t regenSize;
ZSTD_DCtx* dctx = ZSTD_createDCtx();
if (dctx==NULL) return ERROR(memory_allocation);
regenSize = ZSTDv04_decompressDCtx(dctx, dst, maxDstSize, src, srcSize);
ZSTD_freeDCtx(dctx);
return regenSize;
#else
ZSTD_DCtx dctx;
return ZSTD_decompressDCtx(&dctx, dst, maxDstSize, src, srcSize);
#endif
}
size_t ZSTDv04_resetDCtx(ZSTDv04_Dctx* dctx) { return ZSTD_resetDCtx(dctx); }
size_t ZSTDv04_nextSrcSizeToDecompress(ZSTDv04_Dctx* dctx)
{
return ZSTD_nextSrcSizeToDecompress(dctx);
}
size_t ZSTDv04_decompressContinue(ZSTDv04_Dctx* dctx, void* dst, size_t maxDstSize, const void* src, size_t srcSize)
{
return ZSTD_decompressContinue(dctx, dst, maxDstSize, src, srcSize);
}
ZBUFFv04_DCtx* ZBUFFv04_createDCtx(void) { return ZBUFF_createDCtx(); }
size_t ZBUFFv04_freeDCtx(ZBUFFv04_DCtx* dctx) { return ZBUFF_freeDCtx(dctx); }
size_t ZBUFFv04_decompressInit(ZBUFFv04_DCtx* dctx) { return ZBUFF_decompressInit(dctx); }
size_t ZBUFFv04_decompressWithDictionary(ZBUFFv04_DCtx* dctx, const void* src, size_t srcSize)
{ return ZBUFF_decompressWithDictionary(dctx, src, srcSize); }
size_t ZBUFFv04_decompressContinue(ZBUFFv04_DCtx* dctx, void* dst, size_t* maxDstSizePtr, const void* src, size_t* srcSizePtr)
{
return ZBUFF_decompressContinue(dctx, dst, maxDstSizePtr, src, srcSizePtr);
}
lib/legacy/zstd_v04.h
+148
View File
@@ -0,0 +1,148 @@
/*
zstd_v04 - decoder for 0.4 format
Header File
Copyright (C) 2016, Yann Collet.
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above
copyright notice, this list of conditions and the following disclaimer
in the documentation and/or other materials provided with the
distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
You can contact the author at :
- zstd source repository : https://github.com/Cyan4973/zstd
- ztsd public forum : https://groups.google.com/forum/#!forum/lz4c
*/
#pragma once
#if defined (__cplusplus)
extern "C" {
#endif
/* *************************************
* Includes
***************************************/
#include <stddef.h> /* size_t */
/* *************************************
* Simple one-step function
***************************************/
/**
ZSTDv04_decompress() : decompress ZSTD frames compliant with v0.4.x format
compressedSize : is the exact source size
maxOriginalSize : is the size of the 'dst' buffer, which must be already allocated.
It must be equal or larger than originalSize, otherwise decompression will fail.
return : the number of bytes decompressed into destination buffer (originalSize)
or an errorCode if it fails (which can be tested using ZSTDv01_isError())
*/
size_t ZSTDv04_decompress( void* dst, size_t maxOriginalSize,
const void* src, size_t compressedSize);
/**
ZSTDv04_isError() : tells if the result of ZSTDv04_decompress() is an error
*/
unsigned ZSTDv04_isError(size_t code);
/* *************************************
* Advanced functions
***************************************/
typedef struct ZSTDv04_Dctx_s ZSTDv04_Dctx;
ZSTDv04_Dctx* ZSTDv04_createDCtx(void);
size_t ZSTDv04_freeDCtx(ZSTDv04_Dctx* dctx);
size_t ZSTDv04_decompressDCtx(ZSTDv04_Dctx* dctx,
void* dst, size_t maxOriginalSize,
const void* src, size_t compressedSize);
/* *************************************
* Direct Streaming
***************************************/
size_t ZSTDv04_resetDCtx(ZSTDv04_Dctx* dctx);
size_t ZSTDv04_nextSrcSizeToDecompress(ZSTDv04_Dctx* dctx);
size_t ZSTDv04_decompressContinue(ZSTDv04_Dctx* dctx, void* dst, size_t maxDstSize, const void* src, size_t srcSize);
/**
Use above functions alternatively.
ZSTD_nextSrcSizeToDecompress() tells how much bytes to provide as 'srcSize' to ZSTD_decompressContinue().
ZSTD_decompressContinue() will use previous data blocks to improve compression if they are located prior to current block.
Result is the number of bytes regenerated within 'dst'.
It can be zero, which is not an error; it just means ZSTD_decompressContinue() has decoded some header.
*/
/* *************************************
* Buffered Streaming
***************************************/
typedef struct ZBUFFv04_DCtx_s ZBUFFv04_DCtx;
ZBUFFv04_DCtx* ZBUFFv04_createDCtx(void);
size_t ZBUFFv04_freeDCtx(ZBUFFv04_DCtx* dctx);
size_t ZBUFFv04_decompressInit(ZBUFFv04_DCtx* dctx);
size_t ZBUFFv04_decompressWithDictionary(ZBUFFv04_DCtx* dctx, const void* dict, size_t dictSize);
size_t ZBUFFv04_decompressContinue(ZBUFFv04_DCtx* dctx, void* dst, size_t* maxDstSizePtr, const void* src, size_t* srcSizePtr);
/** ************************************************
* Streaming decompression
*
* A ZBUFF_DCtx object is required to track streaming operation.
* Use ZBUFF_createDCtx() and ZBUFF_freeDCtx() to create/release resources.
* Use ZBUFF_decompressInit() to start a new decompression operation.
* ZBUFF_DCtx objects can be reused multiple times.
*
* Optionally, a reference to a static dictionary can be set, using ZBUFF_decompressWithDictionary()
* It must be the same content as the one set during compression phase.
* Dictionary content must remain accessible during the decompression process.
*
* Use ZBUFF_decompressContinue() repetitively to consume your input.
* *srcSizePtr and *maxDstSizePtr can be any size.
* The function will report how many bytes were read or written by modifying *srcSizePtr and *maxDstSizePtr.
* Note that it may not consume the entire input, in which case it's up to the caller to present remaining input again.
* The content of dst will be overwritten (up to *maxDstSizePtr) at each function call, so save its content if it matters or change dst.
* @return : a hint to preferred nb of bytes to use as input for next function call (it's only a hint, to improve latency)
* or 0 when a frame is completely decoded
* or an error code, which can be tested using ZBUFF_isError().
*
* Hint : recommended buffer sizes (not compulsory) : ZBUFF_recommendedDInSize / ZBUFF_recommendedDOutSize
* output : ZBUFF_recommendedDOutSize==128 KB block size is the internal unit, it ensures it's always possible to write a full block when it's decoded.
* input : ZBUFF_recommendedDInSize==128Kb+3; just follow indications from ZBUFF_decompressContinue() to minimize latency. It should always be <= 128 KB + 3 .
* **************************************************/
unsigned ZBUFFv04_isError(size_t errorCode);
const char* ZBUFFv04_getErrorName(size_t errorCode);
/** The below functions provide recommended buffer sizes for Compression or Decompression operations.
* These sizes are not compulsory, they just tend to offer better latency */
size_t ZBUFFv04_recommendedDInSize(void);
size_t ZBUFFv04_recommendedDOutSize(void);
/* *************************************
* Prefix - version detection
***************************************/
#define ZSTDv04_magicNumber 0xFD2FB524 /* v0.4 */
#if defined (__cplusplus)
}
#endif
lib/mem.h
+18 -30
View File
@@ -39,15 +39,15 @@
extern "C" {
#endif
/******************************************
* Includes
/*-****************************************
* Dependencies
******************************************/
#include <stddef.h> /* size_t, ptrdiff_t */
#include <string.h> /* memcpy */
/******************************************
* Compiler-specific
/*-****************************************
* Compiler specifics
******************************************/
#if defined(__GNUC__)
# define MEM_STATIC static __attribute__((unused))
@@ -60,7 +60,7 @@ extern "C" {
#endif
/****************************************************************
/*-**************************************************************
* Basic Types
*****************************************************************/
#if defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */)
@@ -83,10 +83,10 @@ extern "C" {
#endif
/****************************************************************
/*-**************************************************************
* Memory I/O
*****************************************************************/
/* MEM_FORCE_MEMORY_ACCESS
/*!MEM_FORCE_MEMORY_ACCESS
* By default, access to unaligned memory is controlled by `memcpy()`, which is safe and portable.
* Unfortunately, on some target/compiler combinations, the generated assembly is sub-optimal.
* The below switch allow to select different access method for improved performance.
@@ -94,8 +94,8 @@ extern "C" {
* Method 1 : `__packed` statement. It depends on compiler extension (ie, not portable).
* This method is safe if your compiler supports it, and *generally* as fast or faster than `memcpy`.
* Method 2 : direct access. This method is portable but violate C standard.
* It can generate buggy code on targets generating assembly depending on alignment.
* But in some circumstances, it's the only known way to get the most performance (ie GCC + ARMv6)
* It can generate buggy code on targets depending on alignment.
* In some circumstances, it's the only known way to get the most performance (ie GCC + ARMv6)
* See http://fastcompression.blogspot.fr/2015/08/accessing-unaligned-memory.html for details.
* Prefer these methods in priority order (0 > 1 > 2)
*/
@@ -185,8 +185,7 @@ MEM_STATIC U16 MEM_readLE16(const void* memPtr)
{
if (MEM_isLittleEndian())
return MEM_read16(memPtr);
else
{
else {
const BYTE* p = (const BYTE*)memPtr;
return (U16)(p[0] + (p[1]<<8));
}
@@ -194,12 +193,9 @@ MEM_STATIC U16 MEM_readLE16(const void* memPtr)
MEM_STATIC void MEM_writeLE16(void* memPtr, U16 val)
{
if (MEM_isLittleEndian())
{
if (MEM_isLittleEndian()) {
MEM_write16(memPtr, val);
}
else
{
} else {
BYTE* p = (BYTE*)memPtr;
p[0] = (BYTE)val;
p[1] = (BYTE)(val>>8);
@@ -210,8 +206,7 @@ MEM_STATIC U32 MEM_readLE32(const void* memPtr)
{
if (MEM_isLittleEndian())
return MEM_read32(memPtr);
else
{
else {
const BYTE* p = (const BYTE*)memPtr;
return (U32)((U32)p[0] + ((U32)p[1]<<8) + ((U32)p[2]<<16) + ((U32)p[3]<<24));
}
@@ -219,12 +214,9 @@ MEM_STATIC U32 MEM_readLE32(const void* memPtr)
MEM_STATIC void MEM_writeLE32(void* memPtr, U32 val32)
{
if (MEM_isLittleEndian())
{
if (MEM_isLittleEndian()) {
MEM_write32(memPtr, val32);
}
else
{
} else {
BYTE* p = (BYTE*)memPtr;
p[0] = (BYTE)val32;
p[1] = (BYTE)(val32>>8);
@@ -237,8 +229,7 @@ MEM_STATIC U64 MEM_readLE64(const void* memPtr)
{
if (MEM_isLittleEndian())
return MEM_read64(memPtr);
else
{
else {
const BYTE* p = (const BYTE*)memPtr;
return (U64)((U64)p[0] + ((U64)p[1]<<8) + ((U64)p[2]<<16) + ((U64)p[3]<<24)
+ ((U64)p[4]<<32) + ((U64)p[5]<<40) + ((U64)p[6]<<48) + ((U64)p[7]<<56));
@@ -247,12 +238,9 @@ MEM_STATIC U64 MEM_readLE64(const void* memPtr)
MEM_STATIC void MEM_writeLE64(void* memPtr, U64 val64)
{
if (MEM_isLittleEndian())
{
if (MEM_isLittleEndian()) {
MEM_write64(memPtr, val64);
}
else
{
} else {
BYTE* p = (BYTE*)memPtr;
p[0] = (BYTE)val64;
p[1] = (BYTE)(val64>>8);
lib/zstd.h
+36 -14
View File
@@ -70,27 +70,26 @@ ZSTDLIB_API unsigned ZSTD_versionNumber (void);
/* *************************************
* Simple functions
***************************************/
ZSTDLIB_API size_t ZSTD_compress( void* dst, size_t maxDstSize,
ZSTDLIB_API size_t ZSTD_compress( void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
int compressionLevel);
ZSTDLIB_API size_t ZSTD_decompress( void* dst, size_t maxOriginalSize,
ZSTDLIB_API size_t ZSTD_decompress( void* dst, size_t dstCapacity,
const void* src, size_t compressedSize);
/**
ZSTD_compress() :
Compresses 'srcSize' bytes from buffer 'src' into buffer 'dst', of maximum size 'dstSize'.
Compresses @srcSize bytes from buffer @src into buffer @dst of size @dstCapacity.
Destination buffer must be already allocated.
Compression runs faster if maxDstSize >= ZSTD_compressBound(srcSize).
return : the number of bytes written into buffer 'dst'
or an error code if it fails (which can be tested using ZSTD_isError())
Compression runs faster if @dstCapacity >= ZSTD_compressBound(srcSize).
@return : the number of bytes written into @dst
or an error code if it fails (which can be tested using ZSTD_isError())
ZSTD_decompress() :
compressedSize : is the exact source size
maxOriginalSize : is the size of the 'dst' buffer, which must be already allocated.
It must be equal or larger than originalSize, otherwise decompression will fail.
return : the number of bytes decompressed into destination buffer (<= maxOriginalSize)
or an errorCode if it fails (which can be tested using ZSTD_isError())
@compressedSize : is the _exact_ size of the compressed blob (or decompression will fail)
@dst must be large enough, equal or larger than originalSize.
@return : the number of bytes decompressed into @dst (<= @dstCapacity)
or an errorCode if it fails (which can be tested using ZSTD_isError())
*/
@@ -105,7 +104,7 @@ ZSTDLIB_API const char* ZSTD_getErrorName(size_t code); /** provides error co
/* *************************************
* Advanced functions
* Explicit memory management
***************************************/
/** Compression context management */
typedef struct ZSTD_CCtx_s ZSTD_CCtx; /* incomplete type */
@@ -114,7 +113,7 @@ ZSTDLIB_API size_t ZSTD_freeCCtx(ZSTD_CCtx* cctx);
/** ZSTD_compressCCtx() :
Same as ZSTD_compress(), but requires an already allocated ZSTD_CCtx (see ZSTD_createCCtx()) */
ZSTDLIB_API size_t ZSTD_compressCCtx(ZSTD_CCtx* ctx, void* dst, size_t maxDstSize, const void* src, size_t srcSize, int compressionLevel);
ZSTDLIB_API size_t ZSTD_compressCCtx(ZSTD_CCtx* ctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize, int compressionLevel);
/** Decompression context management */
typedef struct ZSTD_DCtx_s ZSTD_DCtx;
@@ -123,7 +122,30 @@ ZSTDLIB_API size_t ZSTD_freeDCtx(ZSTD_DCtx* dctx);
/** ZSTD_decompressDCtx
* Same as ZSTD_decompress(), but requires an already allocated ZSTD_DCtx (see ZSTD_createDCtx()) */
ZSTDLIB_API size_t ZSTD_decompressDCtx(ZSTD_DCtx* ctx, void* dst, size_t maxDstSize, const void* src, size_t srcSize);
ZSTDLIB_API size_t ZSTD_decompressDCtx(ZSTD_DCtx* ctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize);
/*-***********************
* Dictionary API
*************************/
/*! ZSTD_compress_usingDict
* Compression using a pre-defined Dictionary content (see dictBuilder)
* Note : dict can be NULL, in which case, it's equivalent to ZSTD_compressCCtx() */
ZSTDLIB_API size_t ZSTD_compress_usingDict(ZSTD_CCtx* ctx,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
const void* dict,size_t dictSize,
int compressionLevel);
/*! ZSTD_decompress_usingDict
* Decompression using a pre-defined Dictionary content (see dictBuilder)
* Dictionary must be identical to the one used during compression, otherwise regenerated data will be corrupted
* Note : dict can be NULL, in which case, it's equivalent to ZSTD_decompressDCtx() */
ZSTDLIB_API size_t ZSTD_decompress_usingDict(ZSTD_DCtx* dctx,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
const void* dict,size_t dictSize);
#if defined (__cplusplus)
lib/zstd_buffered.c
+12 -6
View File
@@ -36,7 +36,7 @@
*/
/* *************************************
* Includes
* Dependencies
***************************************/
#include <stdlib.h>
#include "error_private.h"
@@ -44,6 +44,12 @@
#include "zstd_buffered_static.h"
/* *************************************
* Constants
***************************************/
static size_t ZBUFF_blockHeaderSize = 3;
static size_t ZBUFF_endFrameSize = 3;
/** ************************************************
* Streaming compression
*
@@ -522,7 +528,7 @@ size_t ZBUFF_decompressContinue(ZBUFF_DCtx* zbc, void* dst, size_t* maxDstSizePt
{
size_t nextSrcSizeHint = ZSTD_nextSrcSizeToDecompress(zbc->zc);
if (nextSrcSizeHint > 3) nextSrcSizeHint+= 3; /* get the next block header while at it */
if (nextSrcSizeHint > ZBUFF_blockHeaderSize) nextSrcSizeHint+= ZBUFF_blockHeaderSize; /* get next block header too */
nextSrcSizeHint -= zbc->inPos; /* already loaded*/
return nextSrcSizeHint;
}
@@ -536,7 +542,7 @@ size_t ZBUFF_decompressContinue(ZBUFF_DCtx* zbc, void* dst, size_t* maxDstSizePt
unsigned ZBUFF_isError(size_t errorCode) { return ERR_isError(errorCode); }
const char* ZBUFF_getErrorName(size_t errorCode) { return ERR_getErrorName(errorCode); }
size_t ZBUFF_recommendedCInSize() { return BLOCKSIZE; }
size_t ZBUFF_recommendedCOutSize() { return ZSTD_compressBound(BLOCKSIZE) + 6; }
size_t ZBUFF_recommendedDInSize() { return BLOCKSIZE + 3; }
size_t ZBUFF_recommendedDOutSize() { return BLOCKSIZE; }
size_t ZBUFF_recommendedCInSize(void) { return BLOCKSIZE; }
size_t ZBUFF_recommendedCOutSize(void) { return ZSTD_compressBound(BLOCKSIZE) + ZBUFF_blockHeaderSize + ZBUFF_endFrameSize; }
size_t ZBUFF_recommendedDInSize(void) { return BLOCKSIZE + ZBUFF_blockHeaderSize /* block header size*/ ; }
size_t ZBUFF_recommendedDOutSize(void) { return BLOCKSIZE; }
lib/zstd_buffered.h
+17 -17
View File
@@ -71,9 +71,9 @@ ZSTDLIB_API size_t ZBUFF_freeCCtx(ZBUFF_CCtx* cctx);
ZSTDLIB_API size_t ZBUFF_compressInit(ZBUFF_CCtx* cctx, int compressionLevel);
ZSTDLIB_API size_t ZBUFF_compressInitDictionary(ZBUFF_CCtx* cctx, const void* dict, size_t dictSize, int compressionLevel);
ZSTDLIB_API size_t ZBUFF_compressContinue(ZBUFF_CCtx* cctx, void* dst, size_t* maxDstSizePtr, const void* src, size_t* srcSizePtr);
ZSTDLIB_API size_t ZBUFF_compressFlush(ZBUFF_CCtx* cctx, void* dst, size_t* maxDstSizePtr);
ZSTDLIB_API size_t ZBUFF_compressEnd(ZBUFF_CCtx* cctx, void* dst, size_t* maxDstSizePtr);
ZSTDLIB_API size_t ZBUFF_compressContinue(ZBUFF_CCtx* cctx, void* dst, size_t* dstCapacityPtr, const void* src, size_t* srcSizePtr);
ZSTDLIB_API size_t ZBUFF_compressFlush(ZBUFF_CCtx* cctx, void* dst, size_t* dstCapacityPtr);
ZSTDLIB_API size_t ZBUFF_compressEnd(ZBUFF_CCtx* cctx, void* dst, size_t* dstCapacityPtr);
/** ************************************************
* Streaming compression
@@ -87,24 +87,24 @@ ZSTDLIB_API size_t ZBUFF_compressEnd(ZBUFF_CCtx* cctx, void* dst, size_t* maxDst
* Use ZBUFF_compressInitDictionary() for a compression which requires a dictionary.
*
* Use ZBUFF_compressContinue() repetitively to consume input stream.
* *srcSizePtr and *maxDstSizePtr can be any size.
* The function will report how many bytes were read or written within *srcSizePtr and *maxDstSizePtr.
* *srcSizePtr and *dstCapacityPtr can be any size.
* The function will report how many bytes were read or written within *srcSizePtr and *dstCapacityPtr.
* Note that it may not consume the entire input, in which case it's up to the caller to present again remaining data.
* The content of dst will be overwritten (up to *maxDstSizePtr) at each function call, so save its content if it matters or move dst .
* @return : a hint to preferred nb of bytes to use as input for next function call (it's only a hint, to improve latency)
* The content of @dst will be overwritten (up to *dstCapacityPtr) at each call, so save its content if it matters or change @dst .
* @return : a hint to preferred nb of bytes to use as input for next function call (it's just a hint, to improve latency)
* or an error code, which can be tested using ZBUFF_isError().
*
* At any moment, it's possible to flush whatever data remains within buffer, using ZBUFF_compressFlush().
* The nb of bytes written into `dst` will be reported into *maxDstSizePtr.
* Note that the function cannot output more than the size of `dst` buffer (initial value of *maxDstSizePtr).
* Therefore, some content might still be left into internal buffer if dst buffer is too small.
* The nb of bytes written into @dst will be reported into *dstCapacityPtr.
* Note that the function cannot output more than *dstCapacityPtr,
* therefore, some content might still be left into internal buffer if *dstCapacityPtr is too small.
* @return : nb of bytes still present into internal buffer (0 if it's empty)
* or an error code, which can be tested using ZBUFF_isError().
*
* ZBUFF_compressEnd() instructs to finish a frame.
* It will perform a flush and write frame epilogue.
* Note that the epilogue is necessary for decoders to consider a frame completed.
* Similar to ZBUFF_compressFlush(), it may not be able to output the entire internal buffer content if *maxDstSizePtr is too small.
* The epilogue is required for decoders to consider a frame completed.
* Similar to ZBUFF_compressFlush(), it may not be able to output the entire internal buffer content if *dstCapacityPtr is too small.
* In which case, call again ZBUFF_compressFlush() to complete the flush.
* @return : nb of bytes still present into internal buffer (0 if it's empty)
* or an error code, which can be tested using ZBUFF_isError().
@@ -112,7 +112,7 @@ ZSTDLIB_API size_t ZBUFF_compressEnd(ZBUFF_CCtx* cctx, void* dst, size_t* maxDst
* Hint : recommended buffer sizes (not compulsory) : ZBUFF_recommendedCInSize / ZBUFF_recommendedCOutSize
* input : ZBUFF_recommendedCInSize==128 KB block size is the internal unit, it improves latency to use this value (skipped buffering).
* output : ZBUFF_recommendedCOutSize==ZSTD_compressBound(128 KB) + 3 + 3 : ensures it's always possible to write/flush/end a full block. Skip some buffering.
* By using both, you ensure that input will be entirely consumed, and output will always contain the result.
* By using both, it ensures that input will be entirely consumed, and output will always contain the result, reducing intermediate buffering.
* **************************************************/
@@ -123,7 +123,7 @@ ZSTDLIB_API size_t ZBUFF_freeDCtx(ZBUFF_DCtx* dctx);
ZSTDLIB_API size_t ZBUFF_decompressInit(ZBUFF_DCtx* dctx);
ZSTDLIB_API size_t ZBUFF_decompressInitDictionary(ZBUFF_DCtx* dctx, const void* dict, size_t dictSize);
ZSTDLIB_API size_t ZBUFF_decompressContinue(ZBUFF_DCtx* dctx, void* dst, size_t* maxDstSizePtr, const void* src, size_t* srcSizePtr);
ZSTDLIB_API size_t ZBUFF_decompressContinue(ZBUFF_DCtx* dctx, void* dst, size_t* dstCapacityPtr, const void* src, size_t* srcSizePtr);
/** ************************************************
* Streaming decompression
@@ -135,10 +135,10 @@ ZSTDLIB_API size_t ZBUFF_decompressContinue(ZBUFF_DCtx* dctx, void* dst, size_t*
* Note that ZBUFF_DCtx objects can be reused multiple times.
*
* Use ZBUFF_decompressContinue() repetitively to consume your input.
* *srcSizePtr and *maxDstSizePtr can be any size.
* The function will report how many bytes were read or written by modifying *srcSizePtr and *maxDstSizePtr.
* *srcSizePtr and *dstCapacityPtr can be any size.
* The function will report how many bytes were read or written by modifying *srcSizePtr and *dstCapacityPtr.
* Note that it may not consume the entire input, in which case it's up to the caller to present remaining input again.
* The content of dst will be overwritten (up to *maxDstSizePtr) at each function call, so save its content if it matters or change dst.
* The content of @dst will be overwritten (up to *dstCapacityPtr) at each function call, so save its content if it matters or change @dst.
* @return : a hint to preferred nb of bytes to use as input for next function call (it's only a hint, to help latency)
* or 0 when a frame is completely decoded
* or an error code, which can be tested using ZBUFF_isError().
lib/zstd_compress.c
+47 -42
View File
@@ -49,14 +49,13 @@
/* *************************************
* Includes
* Dependencies
***************************************/
#include <stdlib.h> /* malloc */
#include <string.h> /* memset */
#include "mem.h"
#include "fse_static.h"
#include "huff0_static.h"
#include "zstd_static.h"
#include "zstd_internal.h"
@@ -69,8 +68,6 @@ static const U32 g_searchStrength = 8;
/* *************************************
* Helper functions
***************************************/
unsigned ZSTD_maxCLevel(void) { return ZSTD_MAX_CLEVEL; }
size_t ZSTD_compressBound(size_t srcSize) { return FSE_compressBound(srcSize) + 12; }
@@ -114,6 +111,7 @@ struct ZSTD_CCtx_s
U32 dictLimit; /* below that point, need extDict */
U32 lowLimit; /* below that point, no more data */
U32 nextToUpdate; /* index from which to continue dictionary update */
U32 loadedDictEnd;
U32 stage;
ZSTD_parameters params;
void* workSpace;
@@ -217,6 +215,7 @@ static size_t ZSTD_resetCCtx_advanced (ZSTD_CCtx* zc,
zc->seqStore.dumpsStart = zc->seqStore.matchLengthStart + (blockSize>>2);
zc->hbSize = 0;
zc->stage = 0;
zc->loadedDictEnd = 0;
return 0;
}
@@ -249,6 +248,7 @@ size_t ZSTD_copyCCtx(ZSTD_CCtx* dstCCtx, const ZSTD_CCtx* srcCCtx)
dstCCtx->dictBase = srcCCtx->dictBase;
dstCCtx->dictLimit = srcCCtx->dictLimit;
dstCCtx->lowLimit = srcCCtx->lowLimit;
dstCCtx->loadedDictEnd = srcCCtx->loadedDictEnd;
/* copy entropy tables */
dstCCtx->flagStaticTables = srcCCtx->flagStaticTables;
@@ -715,7 +715,7 @@ _check_compressibility:
*/
MEM_STATIC void ZSTD_storeSeq(seqStore_t* seqStorePtr, size_t litLength, const BYTE* literals, size_t offsetCode, size_t matchCode)
{
#if 0
#if 0 /* for debug */
static const BYTE* g_start = NULL;
if (g_start==NULL) g_start = literals;
//if (literals - g_start == 8695)
@@ -734,7 +734,13 @@ MEM_STATIC void ZSTD_storeSeq(seqStore_t* seqStorePtr, size_t litLength, const B
*(seqStorePtr->dumps++) = (BYTE)(litLength - MaxLL);
} else {
*(seqStorePtr->dumps++) = 255;
MEM_writeLE32(seqStorePtr->dumps, (U32)litLength); seqStorePtr->dumps += 3;
if (litLength < (1<<15)) {
MEM_writeLE16(seqStorePtr->dumps, (U16)(litLength<<1));
seqStorePtr->dumps += 2;
} else {
MEM_writeLE32(seqStorePtr->dumps, (U32)((litLength<<1)+1));
seqStorePtr->dumps += 3;
}
} }
else *(seqStorePtr->litLength++) = (BYTE)litLength;
@@ -748,7 +754,13 @@ MEM_STATIC void ZSTD_storeSeq(seqStore_t* seqStorePtr, size_t litLength, const B
*(seqStorePtr->dumps++) = (BYTE)(matchCode - MaxML);
} else {
*(seqStorePtr->dumps++) = 255;
MEM_writeLE32(seqStorePtr->dumps, (U32)matchCode); seqStorePtr->dumps += 3;
if (matchCode < (1<<15)) {
MEM_writeLE16(seqStorePtr->dumps, (U16)(matchCode<<1));
seqStorePtr->dumps += 2;
} else {
MEM_writeLE32(seqStorePtr->dumps, (U32)((matchCode<<1)+1));
seqStorePtr->dumps += 3;
}
} }
else *(seqStorePtr->matchLength++) = (BYTE)matchCode;
}
@@ -906,7 +918,7 @@ static size_t ZSTD_hashPtr(const void* p, U32 hBits, U32 mls)
}
/* *************************************
/*-*************************************
* Fast Scan
***************************************/
#define FILLHASHSTEP 3
@@ -916,7 +928,7 @@ static void ZSTD_fillHashTable (ZSTD_CCtx* zc, const void* end, const U32 mls)
const U32 hBits = zc->params.hashLog;
const BYTE* const base = zc->base;
const BYTE* ip = base + zc->nextToUpdate;
const BYTE* const iend = (const BYTE*) end;
const BYTE* const iend = ((const BYTE*)end) - 8;
while(ip <= iend) {
hashTable[ZSTD_hashPtr(ip, hBits, mls)] = (U32)(ip - base);
@@ -1907,47 +1919,42 @@ static ZSTD_blockCompressor ZSTD_selectBlockCompressor(ZSTD_strategy strat, int
static size_t ZSTD_compressBlock_internal(ZSTD_CCtx* zc, void* dst, size_t maxDstSize, const void* src, size_t srcSize)
{
ZSTD_blockCompressor blockCompressor = ZSTD_selectBlockCompressor(zc->params.strategy, zc->lowLimit < zc->dictLimit);
if (srcSize < MIN_CBLOCK_SIZE+3) return 0; /* don't even attempt compression below a certain srcSize */
if (srcSize < MIN_CBLOCK_SIZE+ZSTD_blockHeaderSize+1) return 0; /* don't even attempt compression below a certain srcSize */
blockCompressor(zc, src, srcSize);
return ZSTD_compressSequences(zc, dst, maxDstSize, srcSize);
}
static size_t ZSTD_compress_generic (ZSTD_CCtx* ctxPtr,
static size_t ZSTD_compress_generic (ZSTD_CCtx* zc,
void* dst, size_t maxDstSize,
const void* src, size_t srcSize)
{
size_t blockSize = ctxPtr->blockSize;
size_t blockSize = zc->blockSize;
size_t remaining = srcSize;
const BYTE* ip = (const BYTE*)src;
BYTE* const ostart = (BYTE*)dst;
BYTE* op = ostart;
const U32 maxDist = 1 << ctxPtr->params.windowLog;
const U32 maxDist = 1 << zc->params.windowLog;
while (remaining)
{
while (remaining) {
size_t cSize;
if (maxDstSize < 3 + MIN_CBLOCK_SIZE) return ERROR(dstSize_tooSmall); /* not enough space to store compressed block */
if (maxDstSize < ZSTD_blockHeaderSize + MIN_CBLOCK_SIZE) return ERROR(dstSize_tooSmall); /* not enough space to store compressed block */
if (remaining < blockSize) blockSize = remaining;
if ((U32)(ip+blockSize - (ctxPtr->base + ctxPtr->lowLimit)) > maxDist)
{
/* respect windowLog contract */
ctxPtr->lowLimit = (U32)(ip+blockSize - ctxPtr->base) - maxDist;
if (ctxPtr->dictLimit < ctxPtr->lowLimit) ctxPtr->dictLimit = ctxPtr->lowLimit;
if ((U32)(ip+blockSize - zc->base) > zc->loadedDictEnd + maxDist) { /* enforce maxDist */
U32 newLowLimit = (U32)(ip+blockSize - zc->base) - maxDist;
if (zc->lowLimit < newLowLimit) zc->lowLimit = newLowLimit;
if (zc->dictLimit < zc->lowLimit) zc->dictLimit = zc->lowLimit;
}
cSize = ZSTD_compressBlock_internal(ctxPtr, op+3, maxDstSize-3, ip, blockSize);
cSize = ZSTD_compressBlock_internal(zc, op+ZSTD_blockHeaderSize, maxDstSize-ZSTD_blockHeaderSize, ip, blockSize);
if (ZSTD_isError(cSize)) return cSize;
if (cSize == 0)
{
cSize = ZSTD_noCompressBlock(op, maxDstSize, ip, blockSize); /* block is not compressible */
if (cSize == 0) { /* block is not compressible */
cSize = ZSTD_noCompressBlock(op, maxDstSize, ip, blockSize);
if (ZSTD_isError(cSize)) return cSize;
}
else
{
} else {
op[0] = (BYTE)(cSize>>16);
op[1] = (BYTE)(cSize>>8);
op[2] = (BYTE)cSize;
@@ -1973,8 +1980,7 @@ static size_t ZSTD_compressContinue_internal (ZSTD_CCtx* zc,
const BYTE* const ip = (const BYTE*) src;
size_t hbSize = 0;
if (frame && (zc->stage==0))
{
if (frame && (zc->stage==0)) {
hbSize = zc->hbSize;
if (dstSize <= hbSize) return ERROR(dstSize_tooSmall);
zc->stage = 1;
@@ -1984,8 +1990,7 @@ static size_t ZSTD_compressContinue_internal (ZSTD_CCtx* zc,
}
/* Check if blocks follow each other */
if (src != zc->nextSrc)
{
if (src != zc->nextSrc) {
/* not contiguous */
size_t delta = zc->nextSrc - ip;
zc->lowLimit = zc->dictLimit;
@@ -1997,8 +2002,7 @@ static size_t ZSTD_compressContinue_internal (ZSTD_CCtx* zc,
}
/* preemptive overflow correction */
if (zc->lowLimit > (1<<30))
{
if (zc->lowLimit > (1<<30)) {
U32 btplus = (zc->params.strategy == ZSTD_btlazy2) || (zc->params.strategy == ZSTD_opt_bt);
U32 contentMask = (1 << (zc->params.contentLog - btplus)) - 1;
U32 newLowLimit = zc->lowLimit & contentMask; /* preserve position % contentSize */
@@ -2013,8 +2017,7 @@ static size_t ZSTD_compressContinue_internal (ZSTD_CCtx* zc,
}
/* if input and dictionary overlap : reduce dictionary (presumed modified by input) */
if ((ip+srcSize > zc->dictBase + zc->lowLimit) && (ip < zc->dictBase + zc->dictLimit))
{
if ((ip+srcSize > zc->dictBase + zc->lowLimit) && (ip < zc->dictBase + zc->dictLimit)) {
zc->lowLimit = (U32)(ip + srcSize - zc->dictBase);
if (zc->lowLimit > zc->dictLimit) zc->lowLimit = zc->dictLimit;
}
@@ -2056,6 +2059,7 @@ static size_t ZSTD_loadDictionaryContent(ZSTD_CCtx* zc, const void* src, size_t
zc->dictBase = zc->base;
zc->base += ip - zc->nextSrc;
zc->nextToUpdate = zc->dictLimit;
zc->loadedDictEnd = (U32)(iend - zc->base);
zc->nextSrc = iend;
if (srcSize <= 8) return 0;
@@ -2063,7 +2067,7 @@ static size_t ZSTD_loadDictionaryContent(ZSTD_CCtx* zc, const void* src, size_t
switch(zc->params.strategy)
{
case ZSTD_fast:
ZSTD_fillHashTable (zc, iend-8, zc->params.searchLength);
ZSTD_fillHashTable (zc, iend, zc->params.searchLength);
break;
case ZSTD_greedy:
@@ -2074,13 +2078,13 @@ static size_t ZSTD_loadDictionaryContent(ZSTD_CCtx* zc, const void* src, size_t
case ZSTD_btlazy2:
ZSTD_updateTree(zc, iend-8, iend, 1 << zc->params.searchLog, zc->params.searchLength);
zc->nextToUpdate = (U32)(iend - zc->base);
break;
default:
return ERROR(GENERIC); /* strategy doesn't exist; impossible */
}
zc->nextToUpdate = zc->loadedDictEnd;
return 0;
}
@@ -2134,8 +2138,7 @@ static size_t ZSTD_loadDictEntropyStats(ZSTD_CCtx* zc, const void* dict, size_t
static size_t ZSTD_compress_insertDictionary(ZSTD_CCtx* zc, const void* dict, size_t dictSize)
{
if (dict && dictSize)
{
if (dict && (dictSize>4)) {
U32 magic = MEM_readLE32(dict);
size_t eSize;
if (magic != ZSTD_DICT_MAGIC)
@@ -2173,7 +2176,7 @@ size_t ZSTD_compressBegin_advanced(ZSTD_CCtx* zc,
size_t ZSTD_compressBegin_usingDict(ZSTD_CCtx* zc, const void* dict, size_t dictSize, int compressionLevel)
{
return ZSTD_compressBegin_advanced(zc, dict, dictSize, ZSTD_getParams(compressionLevel, 0));
return ZSTD_compressBegin_advanced(zc, dict, dictSize, ZSTD_getParams(compressionLevel, 128 KB));
}
size_t ZSTD_compressBegin(ZSTD_CCtx* zc, int compressionLevel)
@@ -2257,7 +2260,7 @@ size_t ZSTD_compress_advanced (ZSTD_CCtx* ctx,
size_t ZSTD_compress_usingDict(ZSTD_CCtx* ctx, void* dst, size_t maxDstSize, const void* src, size_t srcSize, const void* dict, size_t dictSize, int compressionLevel)
{
return ZSTD_compress_advanced(ctx, dst, maxDstSize, src, srcSize, dict, dictSize, ZSTD_getParams(compressionLevel, srcSize+dictSize));
return ZSTD_compress_advanced(ctx, dst, maxDstSize, src, srcSize, dict, dictSize, ZSTD_getParams(compressionLevel, srcSize));
}
size_t ZSTD_compressCCtx (ZSTD_CCtx* ctx, void* dst, size_t maxDstSize, const void* src, size_t srcSize, int compressionLevel)
@@ -2278,6 +2281,8 @@ size_t ZSTD_compress(void* dst, size_t maxDstSize, const void* src, size_t srcSi
/*- Pre-defined compression levels -*/
unsigned ZSTD_maxCLevel(void) { return ZSTD_MAX_CLEVEL; }
static const ZSTD_parameters ZSTD_defaultParameters[4][ZSTD_MAX_CLEVEL+1] = {
{ /* "default" */
/* W, C, H, S, L, strat */
lib/zstd_decompress.c
+23 -17
View File
@@ -1,6 +1,6 @@
/*
zstd - standard compression library
Copyright (C) 2014-2015, Yann Collet.
Copyright (C) 2014-2016, Yann Collet.
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
@@ -27,7 +27,6 @@
You can contact the author at :
- zstd source repository : https://github.com/Cyan4973/zstd
- ztsd public forum : https://groups.google.com/forum/#!forum/lz4c
*/
/* ***************************************************************
@@ -44,7 +43,7 @@
/*!
* LEGACY_SUPPORT :
* ZSTD_decompress() can decode older formats (v0.1+) if set to 1
* if set to 1, ZSTD_decompress() can decode older formats (v0.1+)
*/
#ifndef ZSTD_LEGACY_SUPPORT
# define ZSTD_LEGACY_SUPPORT 0
@@ -58,7 +57,6 @@
#include <string.h> /* memcpy, memmove */
#include <stdio.h> /* debug : printf */
#include "mem.h" /* low level memory routines */
#include "zstd_static.h"
#include "zstd_internal.h"
#include "fse_static.h"
#include "huff0_static.h"
@@ -111,6 +109,10 @@ unsigned ZSTD_versionNumber (void) { return ZSTD_VERSION_NUMBER; }
* tells if a return value is an error code */
unsigned ZSTD_isError(size_t code) { return ERR_isError(code); }
/*! ZSTD_getError
* convert a `size_t` function result into a proper ZSTD_errorCode enum */
ZSTD_errorCode ZSTD_getError(size_t code) { return ERR_getError(code); }
/*! ZSTD_getErrorName
* provides error code string (useful for debugging) */
const char* ZSTD_getErrorName(size_t code) { return ERR_getErrorName(code); }
@@ -304,7 +306,7 @@ static size_t ZSTD_decodeFrameHeader_Part2(ZSTD_DCtx* zc, const void* src, size_
if (srcSize != zc->headerSize)
return ERROR(srcSize_wrong);
result = ZSTD_getFrameParams(&(zc->params), src, srcSize);
if ((MEM_32bits()) && (zc->params.windowLog > 25)) return ERROR(frameParameter_unsupportedBy32bitsImplementation);
if ((MEM_32bits()) && (zc->params.windowLog > 25)) return ERROR(frameParameter_unsupportedBy32bits);
return result;
}
@@ -430,7 +432,7 @@ size_t ZSTD_decodeLiteralsBlock(ZSTD_DCtx* dctx,
}
if (lhSize+litSize+WILDCOPY_OVERLENGTH > srcSize) { /* risk reading beyond src buffer with wildcopy */
if (litSize > srcSize-lhSize) return ERROR(corruption_detected);
if (litSize+lhSize > srcSize) return ERROR(corruption_detected);
memcpy(dctx->litBuffer, istart+lhSize, litSize);
dctx->litPtr = dctx->litBuffer;
dctx->litBufSize = BLOCKSIZE+8;
@@ -628,8 +630,9 @@ static void ZSTD_decodeSequence(seq_t* seq, seqState_t* seqState)
U32 add = *dumps++;
if (add < 255) litLength += add;
else {
litLength = MEM_readLE32(dumps) & 0xFFFFFF; /* no pb : dumps is always followed by seq tables > 1 byte */
dumps += 3;
litLength = MEM_readLE32(dumps) & 0xFFFFFF; /* no risk : dumps is always followed by seq tables > 1 byte */
if (litLength&1) litLength>>=1, dumps += 3;
else litLength = (U16)(litLength)>>1, dumps += 2;
}
if (dumps >= de) dumps = de-1; /* late correction, to avoid read overflow (data is now corrupted anyway) */
}
@@ -666,7 +669,8 @@ static void ZSTD_decodeSequence(seq_t* seq, seqState_t* seqState)
if (add < 255) matchLength += add;
else {
matchLength = MEM_readLE32(dumps) & 0xFFFFFF; /* no pb : dumps is always followed by seq tables > 1 byte */
dumps += 3;
if (matchLength&1) matchLength>>=1, dumps += 3;
else matchLength = (U16)(matchLength)>>1, dumps += 2;
}
if (dumps >= de) dumps = de-1; /* late correction, to avoid read overflow (data is now corrupted anyway) */
}
@@ -678,7 +682,7 @@ static void ZSTD_decodeSequence(seq_t* seq, seqState_t* seqState)
seq->matchLength = matchLength;
seqState->dumps = dumps;
#if 0
#if 0 /* debug */
{
static U64 totalDecoded = 0;
printf("pos %6u : %3u literals & match %3u bytes at distance %6u \n",
@@ -847,28 +851,30 @@ static void ZSTD_checkContinuity(ZSTD_DCtx* dctx, const void* dst)
static size_t ZSTD_decompressBlock_internal(ZSTD_DCtx* dctx,
void* dst, size_t maxDstSize,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize)
{
/* blockType == blockCompressed */
{ /* blockType == blockCompressed */
const BYTE* ip = (const BYTE*)src;
size_t litCSize;
if (srcSize >= BLOCKSIZE) return ERROR(srcSize_wrong);
/* Decode literals sub-block */
size_t litCSize = ZSTD_decodeLiteralsBlock(dctx, src, srcSize);
litCSize = ZSTD_decodeLiteralsBlock(dctx, src, srcSize);
if (ZSTD_isError(litCSize)) return litCSize;
ip += litCSize;
srcSize -= litCSize;
return ZSTD_decompressSequences(dctx, dst, maxDstSize, ip, srcSize);
return ZSTD_decompressSequences(dctx, dst, dstCapacity, ip, srcSize);
}
size_t ZSTD_decompressBlock(ZSTD_DCtx* dctx,
void* dst, size_t maxDstSize,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize)
{
ZSTD_checkContinuity(dctx, dst);
return ZSTD_decompressBlock_internal(dctx, dst, maxDstSize, src, srcSize);
return ZSTD_decompressBlock_internal(dctx, dst, dstCapacity, src, srcSize);
}
lib/zstd_internal.h
+2 -2
View File
@@ -102,8 +102,8 @@ static const size_t ZSTD_frameHeaderSize_min = 5;
#define HufLog 12
#define MIN_SEQUENCES_SIZE 1 /* seqNb */
#define MIN_CBLOCK_SIZE (1 /*litCSize*/ + MIN_SEQUENCES_SIZE)
#define MIN_SEQUENCES_SIZE 1 /* nbSeq==0 */
#define MIN_CBLOCK_SIZE (1 /*litCSize*/ + 1 /* RLE or RAW */ + MIN_SEQUENCES_SIZE /* nbSeq==0 */) /* for a non-null block */
#define WILDCOPY_OVERLENGTH 8
lib/zstd_static.h
+30 -47
View File
@@ -82,6 +82,9 @@ typedef struct
/* *************************************
* Advanced functions
***************************************/
#define ZSTD_MAX_CLEVEL 24
ZSTDLIB_API unsigned ZSTD_maxCLevel (void);
/*! ZSTD_getParams
* @return ZSTD_parameters structure for a selected compression level and srcSize.
* @srcSizeHint value is optional, select 0 if not known */
@@ -91,19 +94,10 @@ ZSTDLIB_API ZSTD_parameters ZSTD_getParams(int compressionLevel, U64 srcSizeHint
* correct params value to remain within authorized range */
ZSTDLIB_API void ZSTD_validateParams(ZSTD_parameters* params);
/*! ZSTD_compress_usingDict
* Same as ZSTD_compressCCtx(), loading a Dictionary content.
* Note : dict can be NULL, in which case, it's equivalent to ZSTD_compressCCtx() */
ZSTDLIB_API size_t ZSTD_compress_usingDict(ZSTD_CCtx* ctx,
void* dst, size_t maxDstSize,
const void* src, size_t srcSize,
const void* dict,size_t dictSize,
int compressionLevel);
/*! ZSTD_compress_advanced
* Same as ZSTD_compress_usingDict(), with fine-tune control of each compression parameter */
ZSTDLIB_API size_t ZSTD_compress_advanced (ZSTD_CCtx* ctx,
void* dst, size_t maxDstSize,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
const void* dict,size_t dictSize,
ZSTD_parameters params);
@@ -115,19 +109,11 @@ ZSTDLIB_API size_t ZSTD_compress_advanced (ZSTD_CCtx* ctx,
* Requires 2 contexts : 1 for reference, which will not be modified, and 1 to run the compression operation */
ZSTDLIB_API size_t ZSTD_compress_usingPreparedCCtx(
ZSTD_CCtx* cctx, const ZSTD_CCtx* preparedCCtx,
void* dst, size_t maxDstSize,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize);
/*- Advanced Decompression functions -*/
/*! ZSTD_decompress_usingDict
* Same as ZSTD_decompressDCtx, using a Dictionary content as prefix
* Note : dict can be NULL, in which case, it's equivalent to ZSTD_decompressDCtx() */
ZSTDLIB_API size_t ZSTD_decompress_usingDict(ZSTD_DCtx* dctx,
void* dst, size_t maxDstSize,
const void* src, size_t srcSize,
const void* dict,size_t dictSize);
/*! ZSTD_decompress_usingPreparedDCtx
* Same as ZSTD_decompress_usingDict, but using a reference context preparedDCtx, where dictionary has been loaded.
* It avoids reloading the dictionary each time.
@@ -135,7 +121,7 @@ ZSTDLIB_API size_t ZSTD_decompress_usingDict(ZSTD_DCtx* dctx,
* Requires 2 contexts : 1 for reference, which will not be modified, and 1 to run the decompression operation */
ZSTDLIB_API size_t ZSTD_decompress_usingPreparedDCtx(
ZSTD_DCtx* dctx, const ZSTD_DCtx* preparedDCtx,
void* dst, size_t maxDstSize,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize);
@@ -147,8 +133,8 @@ ZSTDLIB_API size_t ZSTD_compressBegin_usingDict(ZSTD_CCtx* cctx, const void* dic
ZSTDLIB_API size_t ZSTD_compressBegin_advanced(ZSTD_CCtx* cctx, const void* dict,size_t dictSize, ZSTD_parameters params);
ZSTDLIB_API size_t ZSTD_copyCCtx(ZSTD_CCtx* cctx, const ZSTD_CCtx* preparedCCtx);
ZSTDLIB_API size_t ZSTD_compressContinue(ZSTD_CCtx* cctx, void* dst, size_t maxDstSize, const void* src, size_t srcSize);
ZSTDLIB_API size_t ZSTD_compressEnd(ZSTD_CCtx* cctx, void* dst, size_t maxDstSize);
ZSTDLIB_API size_t ZSTD_compressContinue(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize);
ZSTDLIB_API size_t ZSTD_compressEnd(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity);
/**
Streaming compression, synchronous mode (bufferless)
@@ -181,7 +167,7 @@ ZSTDLIB_API void ZSTD_copyDCtx(ZSTD_DCtx* dctx, const ZSTD_DCtx* preparedDCtx)
ZSTDLIB_API size_t ZSTD_getFrameParams(ZSTD_parameters* params, const void* src, size_t srcSize);
ZSTDLIB_API size_t ZSTD_nextSrcSizeToDecompress(ZSTD_DCtx* dctx);
ZSTDLIB_API size_t ZSTD_decompressContinue(ZSTD_DCtx* dctx, void* dst, size_t maxDstSize, const void* src, size_t srcSize);
ZSTDLIB_API size_t ZSTD_decompressContinue(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize);
/**
Streaming decompression, direct mode (bufferless)
@@ -217,40 +203,37 @@ ZSTDLIB_API size_t ZSTD_decompressContinue(ZSTD_DCtx* dctx, void* dst, size_t ma
/* **************************************
* Block functions
****************************************/
/*!Block functions produce and decode raw zstd blocks, without frame metadata.
Frame headers won't be generated.
User will have to save and regenerate fields required to regenerate data, such as block sizes.
/*! Block functions produce and decode raw zstd blocks, without frame metadata.
User will have to save and regenerate necessary information to regenerate data, such as block sizes.
A few rules to respect :
- Uncompressed block size must be <= 128 KB
- Compressing or decompressing requires a context structure
+ Use ZSTD_createXCtx() to create them
- It is necessary to init context before starting
+ compression : ZSTD_compressBegin()
+ decompression : ZSTD_decompressBegin()
+ variants _usingDict() are also allowed
+ copyXCtx() works too
- When a block is considered not compressible enough, ZSTD_compressBlock() result will be zero.
In which case, nothing is produced into `dst`.
+ User must test for such outcome and deal directly with uncompressed data
+ ZSTD_decompressBlock() doesn't accept uncompressed data as input !!
A few rules to respect :
- Uncompressed block size must be <= 128 KB
- Compressing or decompressing requires a context structure
+ Use ZSTD_createCCtx() and ZSTD_createDCtx()
- It is necessary to init context before starting
+ compression : ZSTD_compressBegin()
+ decompression : ZSTD_decompressBegin()
+ variants _usingDict() are also allowed
+ copyCCtx() and copyDCtx() work too
- When a block is considered not compressible enough, ZSTD_compressBlock() result will be zero.
In which case, nothing is produced into @dst.
+ User must test for such outcome and deal directly with uncompressed data
+ ZSTD_decompressBlock() doesn't accept uncompressed data as input !!
*/
size_t ZSTD_compressBlock (ZSTD_CCtx* cctx, void* dst, size_t maxDstSize, const void* src, size_t srcSize);
size_t ZSTD_decompressBlock(ZSTD_DCtx* dctx, void* dst, size_t maxDstSize, const void* src, size_t srcSize);
/* *************************************
* Pre-defined compression levels
***************************************/
#define ZSTD_MAX_CLEVEL 24
ZSTDLIB_API unsigned ZSTD_maxCLevel (void);
size_t ZSTD_compressBlock (ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize);
size_t ZSTD_decompressBlock(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize);
/* *************************************
* Error management
***************************************/
#include "error_public.h"
/*! ZSTD_getErrorCode
transform a function result using `size_t` into a `ZSTD_error_code` enum type
which can be used to compare directly with enum list within "error_public.h" */
ZSTD_errorCode ZSTD_getError(size_t code);
#if defined (__cplusplus)
programs/Makefile
+1 -1
View File
@@ -53,7 +53,7 @@ MANDIR = $(PREFIX)/share/man/man1
ZSTDDIR = ../lib
ZSTD_FILES := $(ZSTDDIR)/zstd_compress.c $(ZSTDDIR)/zstd_decompress.c $(ZSTDDIR)/fse.c $(ZSTDDIR)/huff0.c
ZSTD_LEGACY:= $(ZSTDDIR)/legacy/zstd_v01.c $(ZSTDDIR)/legacy/zstd_v02.c $(ZSTDDIR)/legacy/zstd_v03.c
ZSTD_LEGACY:= $(ZSTDDIR)/legacy/zstd_v01.c $(ZSTDDIR)/legacy/zstd_v02.c $(ZSTDDIR)/legacy/zstd_v03.c $(ZSTDDIR)/legacy/zstd_v04.c
ifeq ($(ZSTD_LEGACY_SUPPORT), 0)
CPPFLAGS += -DZSTD_LEGACY_SUPPORT=0
programs/bench.c
+2 -1
View File
@@ -212,6 +212,7 @@ typedef struct
} blockParam_t;
#define MIN(a,b) ((a)<(b) ? (a) : (b))
#define MAX(a,b) ((a)>(b) ? (a) : (b))
static int BMK_benchMem(const void* srcBuffer, size_t srcSize,
const char* displayName, int cLevel,
@@ -289,7 +290,7 @@ static int BMK_benchMem(const void* srcBuffer, size_t srcSize,
while (BMK_GetMilliStart() == milliTime);
milliTime = BMK_GetMilliStart();
while (BMK_GetMilliSpan(milliTime) < TIMELOOP) {
ZSTD_compressBegin_advanced(refCtx, dictBuffer, dictBufferSize, ZSTD_getParams(cLevel, dictBufferSize+largestBlockSize));
ZSTD_compressBegin_advanced(refCtx, dictBuffer, dictBufferSize, ZSTD_getParams(cLevel, MAX(dictBufferSize, largestBlockSize)));
for (blockNb=0; blockNb<nbBlocks; blockNb++) {
size_t rSize = ZSTD_compress_usingPreparedCCtx(ctx, refCtx,
blockTable[blockNb].cPtr, blockTable[blockNb].cRoom,
programs/datagen.c
+16 -26
View File
@@ -23,7 +23,7 @@
- Public forum : https://groups.google.com/forum/#!forum/lz4c
*/
/**************************************
/*-************************************
* Includes
**************************************/
#include <stdlib.h> /* malloc */
@@ -31,7 +31,7 @@
#include <string.h> /* memcpy */
/**************************************
/*-************************************
* Basic Types
**************************************/
#if defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */
@@ -50,7 +50,7 @@
#endif
/**************************************
/*-************************************
* OS-specific Includes
**************************************/
#if defined(MSDOS) || defined(OS2) || defined(WIN32) || defined(_WIN32) || defined(__CYGWIN__)
@@ -62,7 +62,7 @@
#endif
/**************************************
/*-************************************
* Constants
**************************************/
#define KB *(1 <<10)
@@ -71,7 +71,7 @@
#define PRIME2 2246822519U
/**************************************
/*-************************************
* Local types
**************************************/
#define LTLOG 13
@@ -81,7 +81,7 @@ typedef BYTE litDistribTable[LTSIZE];
/*********************************************************
/*-*******************************************************
* Local Functions
*********************************************************/
#define RDG_rotl32(x,r) ((x << r) | (x >> (32 - r)))
@@ -103,14 +103,12 @@ static void RDG_fillLiteralDistrib(litDistribTable lt, double ld)
BYTE firstChar = '(';
BYTE lastChar = '}';
if (ld==0.0)
{
if (ld==0.0) {
character = 0;
firstChar = 0;
lastChar =255;
}
while (i<LTSIZE)
{
while (i<LTSIZE) {
U32 weight = (U32)((double)(LTSIZE - i) * ld) + 1;
U32 end;
if (weight + i > LTSIZE) weight = LTSIZE-i;
@@ -140,13 +138,11 @@ void RDG_genBlock(void* buffer, size_t buffSize, size_t prefixSize, double match
U32 prevOffset = 1;
/* special case : sparse content */
while (matchProba >= 1.0)
{
while (matchProba >= 1.0) {
size_t size0 = RDG_rand(seed) & 3;
size0 = (size_t)1 << (16 + size0 * 2);
size0 += RDG_rand(seed) & (size0-1); /* because size0 is power of 2*/
if (buffSize < pos + size0)
{
if (buffSize < pos + size0) {
memset(buffPtr+pos, 0, buffSize-pos);
return;
}
@@ -160,11 +156,9 @@ void RDG_genBlock(void* buffer, size_t buffSize, size_t prefixSize, double match
if (pos==0) buffPtr[0] = RDG_genChar(seed, lt), pos=1;
/* Generate compressible data */
while (pos < buffSize)
{
while (pos < buffSize) {
/* Select : Literal (char) or Match (within 32K) */
if (RDG_RAND15BITS < matchProba32)
{
if (RDG_RAND15BITS < matchProba32) {
/* Copy (within 32K) */
size_t match;
size_t d;
@@ -178,17 +172,14 @@ void RDG_genBlock(void* buffer, size_t buffSize, size_t prefixSize, double match
d = pos + length;
if (d > buffSize) d = buffSize;
while (pos < d) buffPtr[pos++] = buffPtr[match++]; /* correctly manages overlaps */
}
else
{
} else {
/* Literal (noise) */
size_t d;
size_t length = RDG_RANDLENGTH;
d = pos + length;
if (d > buffSize) d = buffSize;
while (pos < d) buffPtr[pos++] = RDG_genChar(seed, lt);
}
}
} }
}
@@ -220,8 +211,7 @@ void RDG_genStdout(unsigned long long size, double matchProba, double litProba,
RDG_genBlock(buff, RDG_DICTSIZE, 0, matchProba, lt, &seed);
/* Generate compressible data */
while (total < size)
{
while (total < size) {
RDG_genBlock(buff, RDG_DICTSIZE+RDG_BLOCKSIZE, RDG_DICTSIZE, matchProba, lt, &seed);
if (size-total < RDG_BLOCKSIZE) genBlockSize = (size_t)(size-total);
total += genBlockSize;
@@ -230,6 +220,6 @@ void RDG_genStdout(unsigned long long size, double matchProba, double litProba,
memcpy(buff, buff + RDG_BLOCKSIZE, RDG_DICTSIZE);
}
// cleanup
/* cleanup */
free(buff);
}
programs/fileio.c
+18 -41
View File
@@ -193,42 +193,30 @@ static U64 FIO_getFileSize(const char* infilename)
static int FIO_getFiles(FILE** fileOutPtr, FILE** fileInPtr,
const char* dstFileName, const char* srcFileName)
{
if (!strcmp (srcFileName, stdinmark))
{
if (!strcmp (srcFileName, stdinmark)) {
DISPLAYLEVEL(4,"Using stdin for input\n");
*fileInPtr = stdin;
SET_BINARY_MODE(stdin);
}
else
{
} else {
*fileInPtr = fopen(srcFileName, "rb");
}
if ( *fileInPtr==0 )
{
if ( *fileInPtr==0 ) {
DISPLAYLEVEL(1, "Unable to access file for processing: %s\n", srcFileName);
return 1;
}
if (!strcmp (dstFileName, stdoutmark))
{
if (!strcmp (dstFileName, stdoutmark)) {
DISPLAYLEVEL(4,"Using stdout for output\n");
*fileOutPtr = stdout;
SET_BINARY_MODE(stdout);
}
else
{
/* Check if destination file already exists */
if (!g_overwrite)
{
} else {
if (!g_overwrite) { /* Check if destination file already exists */
*fileOutPtr = fopen( dstFileName, "rb" );
if (*fileOutPtr != 0)
{
/* prompt for overwrite authorization */
if (*fileOutPtr != 0) { /* dest file exists, prompt for overwrite authorization */
fclose(*fileOutPtr);
DISPLAY("Warning : %s already exists \n", dstFileName);
if ((g_displayLevel <= 1) || (*fileInPtr == stdin))
{
if ((g_displayLevel <= 1) || (*fileInPtr == stdin)) {
/* No interaction possible */
DISPLAY("Operation aborted : %s already exists \n", dstFileName);
return 1;
@@ -236,15 +224,12 @@ static int FIO_getFiles(FILE** fileOutPtr, FILE** fileInPtr,
DISPLAY("Overwrite ? (y/N) : ");
{
int ch = getchar();
if ((ch!='Y') && (ch!='y'))
{
if ((ch!='Y') && (ch!='y')) {
DISPLAY("No. Operation aborted : %s already exists \n", dstFileName);
return 1;
}
while ((ch!=EOF) && (ch!='\n')) ch = getchar(); /* flush rest of input line */
}
}
}
} } }
*fileOutPtr = fopen( dstFileName, "wb" );
}
@@ -265,15 +250,13 @@ static size_t FIO_loadFile(void** bufferPtr, const char* fileName)
U64 fileSize;
*bufferPtr = NULL;
if (fileName == NULL)
return 0;
if (fileName == NULL) return 0;
DISPLAYLEVEL(4,"Loading %s as dictionary \n", fileName);
fileHandle = fopen(fileName, "rb");
if (fileHandle==0) EXM_THROW(31, "Error opening file %s", fileName);
fileSize = FIO_getFileSize(fileName);
if (fileSize > MAX_DICT_SIZE)
{
if (fileSize > MAX_DICT_SIZE) {
int seekResult;
if (fileSize > 1 GB) EXM_THROW(32, "Dictionary file %s is too large", fileName); /* avoid extreme cases */
DISPLAYLEVEL(2,"Dictionary %s is too large : using last %u bytes only \n", fileName, MAX_DICT_SIZE);
@@ -455,8 +438,7 @@ int FIO_compressMultipleFilenames(const char** inFileNamesTable, unsigned nbFile
ress = FIO_createCResources(dictFileName);
/* loop on each file */
for (u=0; u<nbFiles; u++)
{
for (u=0; u<nbFiles; u++) {
size_t ifnSize = strlen(inFileNamesTable[u]);
if (dfnSize <= ifnSize+suffixSize+1) { free(dstFileName); dfnSize = ifnSize + 20; dstFileName = (char*)malloc(dfnSize); }
strcpy(dstFileName, inFileNamesTable[u]);
@@ -525,8 +507,7 @@ unsigned long long FIO_decompressFrame(dRess_t ress,
/* Main decompression Loop */
ZBUFF_decompressInitDictionary(ress.dctx, ress.dictBuffer, ress.dictBufferSize);
while (1)
{
while (1) {
/* Decode */
size_t sizeCheck;
size_t inSize=readSize, decodedSize=ress.dstBufferSize;
@@ -564,8 +545,7 @@ static int FIO_decompressFile_extRess(dRess_t ress,
if (FIO_getFiles(&dstFile, &srcFile, dstFileName, srcFileName)) return 1;
/* for each frame */
for ( ; ; )
{
for ( ; ; ) {
size_t sizeCheck;
/* check magic number -> version */
size_t toRead = 4;
@@ -573,8 +553,7 @@ static int FIO_decompressFile_extRess(dRess_t ress,
if (sizeCheck==0) break; /* no more input */
if (sizeCheck != toRead) EXM_THROW(31, "Read error : cannot read header");
#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT==1)
if (ZSTD_isLegacy(MEM_readLE32(ress.srcBuffer)))
{
if (ZSTD_isLegacy(MEM_readLE32(ress.srcBuffer))) {
filesize += FIO_decompressLegacyFrame(dstFile, srcFile, MEM_readLE32(ress.srcBuffer));
continue;
}
@@ -624,14 +603,12 @@ int FIO_decompressMultipleFilenames(const char** srcNamesTable, unsigned nbFiles
if (dstFileName==NULL) EXM_THROW(70, "not enough memory for dstFileName");
ress = FIO_createDResources(dictFileName);
for (u=0; u<nbFiles; u++)
{
for (u=0; u<nbFiles; u++) {
const char* srcFileName = srcNamesTable[u];
size_t sfnSize = strlen(srcFileName);
const char* suffixPtr = srcFileName + sfnSize - suffixSize;
if (dfnSize <= sfnSize-suffixSize+1) { free(dstFileName); dfnSize = sfnSize + 20; dstFileName = (char*)malloc(dfnSize); if (dstFileName==NULL) EXM_THROW(71, "not enough memory for dstFileName"); }
if (sfnSize <= suffixSize || strcmp(suffixPtr, suffix) != 0)
{
if (sfnSize <= suffixSize || strcmp(suffixPtr, suffix) != 0) {
DISPLAYLEVEL(1, "File extension doesn't match expected extension (%4s); will not process file: %s\n", suffix, srcFileName);
skippedFiles++;
continue;
programs/fuzzer.c
+1 -1
View File
@@ -568,7 +568,7 @@ int fuzzerTests(U32 seed, U32 nbTests, unsigned startTest, double compressibilit
errorCode = ZSTD_compressBegin_usingDict(refCtx, dict, dictSize, (FUZ_rand(&lseed) % (20 - (sampleSizeLog/3))) + 1);
CHECK (ZSTD_isError(errorCode), "ZSTD_compressBegin_usingDict error : %s", ZSTD_getErrorName(errorCode));
errorCode = ZSTD_copyCCtx(ctx, refCtx);
CHECK (ZSTD_isError(errorCode), "context duplication error : %s", ZSTD_getErrorName(errorCode));
CHECK (ZSTD_isError(errorCode), "ZSTD_copyCCtx error : %s", ZSTD_getErrorName(errorCode));
totalTestSize = 0; cSize = 0;
for (n=0; n<nbChunks; n++)
{
programs/legacy/fileio_legacy.c
+82
View File
@@ -329,6 +329,86 @@ unsigned long long FIOv03_decompressFrame(FILE* foutput, FILE* finput)
}
/*- v0.4.x -*/
typedef struct {
void* srcBuffer;
size_t srcBufferSize;
void* dstBuffer;
size_t dstBufferSize;
void* dictBuffer;
size_t dictBufferSize;
ZBUFFv04_DCtx* dctx;
} dRessv04_t;
static dRessv04_t FIOv04_createDResources(void)
{
dRessv04_t ress;
/* init */
ress.dctx = ZBUFFv04_createDCtx();
if (ress.dctx==NULL) EXM_THROW(60, "Can't create ZBUFF decompression context");
ress.dictBuffer = NULL; ress.dictBufferSize=0;
/* Allocate Memory */
ress.srcBufferSize = ZBUFFv04_recommendedDInSize();
ress.srcBuffer = malloc(ress.srcBufferSize);
ress.dstBufferSize = ZBUFFv04_recommendedDOutSize();
ress.dstBuffer = malloc(ress.dstBufferSize);
if (!ress.srcBuffer || !ress.dstBuffer) EXM_THROW(61, "Allocation error : not enough memory");
return ress;
}
static void FIOv04_freeDResources(dRessv04_t ress)
{
size_t errorCode = ZBUFFv04_freeDCtx(ress.dctx);
if (ZBUFFv04_isError(errorCode)) EXM_THROW(69, "Error : can't free ZBUFF context resource : %s", ZBUFFv04_getErrorName(errorCode));
free(ress.srcBuffer);
free(ress.dstBuffer);
free(ress.dictBuffer);
}
unsigned long long FIOv04_decompressFrame(dRessv04_t ress,
FILE* foutput, FILE* finput)
{
U64 frameSize = 0;
size_t readSize = 4;
MEM_writeLE32(ress.srcBuffer, ZSTDv04_magicNumber);
ZBUFFv04_decompressInit(ress.dctx);
ZBUFFv04_decompressWithDictionary(ress.dctx, ress.dictBuffer, ress.dictBufferSize);
while (1)
{
/* Decode */
size_t sizeCheck;
size_t inSize=readSize, decodedSize=ress.dstBufferSize;
size_t toRead = ZBUFFv04_decompressContinue(ress.dctx, ress.dstBuffer, &decodedSize, ress.srcBuffer, &inSize);
if (ZBUFFv04_isError(toRead)) EXM_THROW(36, "Decoding error : %s", ZBUFFv04_getErrorName(toRead));
readSize -= inSize;
/* Write block */
sizeCheck = fwrite(ress.dstBuffer, 1, decodedSize, foutput);
if (sizeCheck != decodedSize) EXM_THROW(37, "Write error : unable to write data block to destination file");
frameSize += decodedSize;
DISPLAYUPDATE(2, "\rDecoded : %u MB... ", (U32)(frameSize>>20) );
if (toRead == 0) break;
if (readSize) EXM_THROW(38, "Decoding error : should consume entire input");
/* Fill input buffer */
if (toRead > ress.srcBufferSize) EXM_THROW(34, "too large block");
readSize = fread(ress.srcBuffer, 1, toRead, finput);
if (readSize != toRead) EXM_THROW(35, "Read error");
}
FIOv04_freeDResources(ress);
return frameSize;
}
unsigned long long FIO_decompressLegacyFrame(FILE* foutput, FILE* finput, U32 magicNumberLE)
{
switch(magicNumberLE)
@@ -339,6 +419,8 @@ unsigned long long FIO_decompressLegacyFrame(FILE* foutput, FILE* finput, U32 ma
return FIOv02_decompressFrame(foutput, finput);
case ZSTDv03_magicNumber :
return FIOv03_decompressFrame(foutput, finput);
case ZSTDv04_magicNumber :
return FIOv04_decompressFrame(FIOv04_createDResources(), foutput, finput);
default :
return ERROR(prefix_unknown);
}
visual/2013/dictBuilder/dictBuilder.vcxproj
+1
View File
@@ -99,6 +99,7 @@
<Link>
<SubSystem>Console</SubSystem>
<GenerateDebugInformation>true</GenerateDebugInformation>
<AdditionalDependencies>setargv.obj;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
</Link>
</ItemDefinitionGroup>
<ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">
visual/2013/zstd/zstd.vcxproj
+2
View File
@@ -24,6 +24,7 @@
<ClCompile Include="..\..\..\lib\legacy\zstd_v01.c" />
<ClCompile Include="..\..\..\lib\legacy\zstd_v02.c" />
<ClCompile Include="..\..\..\lib\legacy\zstd_v03.c" />
<ClCompile Include="..\..\..\lib\legacy\zstd_v04.c" />
<ClCompile Include="..\..\..\lib\zstd_buffered.c" />
<ClCompile Include="..\..\..\lib\zstd_compress.c" />
<ClCompile Include="..\..\..\lib\zstd_decompress.c" />
@@ -43,6 +44,7 @@
<ClInclude Include="..\..\..\lib\legacy\zstd_v01.h" />
<ClInclude Include="..\..\..\lib\legacy\zstd_v02.h" />
<ClInclude Include="..\..\..\lib\legacy\zstd_v03.h" />
<ClInclude Include="..\..\..\lib\legacy\zstd_v04.h" />
<ClInclude Include="..\..\..\lib\zstd.h" />
<ClInclude Include="..\..\..\lib\zstd_buffered.h" />
<ClInclude Include="..\..\..\lib\zstd_buffered_static.h" />
visual/2013/zstd/zstd.vcxproj.filters
+6
View File
@@ -57,6 +57,9 @@
<ClCompile Include="..\..\..\programs\datagen.c">
<Filter>Fichiers sources</Filter>
</ClCompile>
<ClCompile Include="..\..\..\lib\legacy\zstd_v04.c">
<Filter>Fichiers sources</Filter>
</ClCompile>
</ItemGroup>
<ItemGroup>
<ClInclude Include="..\..\..\lib\fse.h">
@@ -113,5 +116,8 @@
<ClInclude Include="..\..\..\programs\datagen.h">
<Filter>Fichiers d%27en-tête</Filter>
</ClInclude>
<ClInclude Include="..\..\..\lib\legacy\zstd_v04.h">
<Filter>Fichiers d%27en-tête</Filter>
</ClInclude>
</ItemGroup>
</Project>