michael@0: /*
michael@0:    LZ4 - Fast LZ compression algorithm
michael@0:    Copyright (C) 2011-2014, Yann Collet.
michael@0:    BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
michael@0: 
michael@0:    Redistribution and use in source and binary forms, with or without
michael@0:    modification, are permitted provided that the following conditions are
michael@0:    met:
michael@0: 
michael@0:        * Redistributions of source code must retain the above copyright
michael@0:    notice, this list of conditions and the following disclaimer.
michael@0:        * Redistributions in binary form must reproduce the above
michael@0:    copyright notice, this list of conditions and the following disclaimer
michael@0:    in the documentation and/or other materials provided with the
michael@0:    distribution.
michael@0: 
michael@0:    THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
michael@0:    "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
michael@0:    LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
michael@0:    A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
michael@0:    OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
michael@0:    SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
michael@0:    LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
michael@0:    DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
michael@0:    THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
michael@0:    (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
michael@0:    OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
michael@0: 
michael@0:    You can contact the author at :
michael@0:    - LZ4 source repository : http://code.google.com/p/lz4/
michael@0:    - LZ4 public forum : https://groups.google.com/forum/#!forum/lz4c
michael@0: */
michael@0: 
michael@0: /**************************************
michael@0:    Tuning parameters
michael@0: **************************************/
michael@0: /*
michael@0:  * HEAPMODE :
michael@0:  * Select how default compression functions will allocate memory for their hash table,
michael@0:  * in memory stack (0:default, fastest), or in memory heap (1:requires memory allocation (malloc)).
michael@0:  */
michael@0: #define HEAPMODE 0
michael@0: 
michael@0: 
michael@0: /**************************************
michael@0:    CPU Feature Detection
michael@0: **************************************/
michael@0: /* 32 or 64 bits ? */
michael@0: #if (defined(__x86_64__) || defined(_M_X64) || defined(_WIN64) \
michael@0:   || defined(__powerpc64__) || defined(__powerpc64le__) \
michael@0:   || defined(__ppc64__) || defined(__ppc64le__) \
michael@0:   || defined(__PPC64__) || defined(__PPC64LE__) \
michael@0:   || defined(__ia64) || defined(__itanium__) || defined(_M_IA64) )   /* Detects 64 bits mode */
michael@0: #  define LZ4_ARCH64 1
michael@0: #else
michael@0: #  define LZ4_ARCH64 0
michael@0: #endif
michael@0: 
michael@0: /*
michael@0:  * Little Endian or Big Endian ?
michael@0:  * Overwrite the #define below if you know your architecture endianess
michael@0:  */
michael@0: #include <stdlib.h>   /* Apparently required to detect endianess */
michael@0: #if defined (__GLIBC__)
michael@0: #  include <endian.h>
michael@0: #  if (__BYTE_ORDER == __BIG_ENDIAN)
michael@0: #     define LZ4_BIG_ENDIAN 1
michael@0: #  endif
michael@0: #elif (defined(__BIG_ENDIAN__) || defined(__BIG_ENDIAN) || defined(_BIG_ENDIAN)) && !(defined(__LITTLE_ENDIAN__) || defined(__LITTLE_ENDIAN) || defined(_LITTLE_ENDIAN))
michael@0: #  define LZ4_BIG_ENDIAN 1
michael@0: #elif defined(__sparc) || defined(__sparc__) \
michael@0:    || defined(__powerpc__) || defined(__ppc__) || defined(__PPC__) \
michael@0:    || defined(__hpux)  || defined(__hppa) \
michael@0:    || defined(_MIPSEB) || defined(__s390__)
michael@0: #  define LZ4_BIG_ENDIAN 1
michael@0: #else
michael@0: /* Little Endian assumed. PDP Endian and other very rare endian format are unsupported. */
michael@0: #endif
michael@0: 
michael@0: /*
michael@0:  * Unaligned memory access is automatically enabled for "common" CPU, such as x86.
michael@0:  * For others CPU, such as ARM, the compiler may be more cautious, inserting unnecessary extra code to ensure aligned access property
michael@0:  * If you know your target CPU supports unaligned memory access, you want to force this option manually to improve performance
michael@0:  */
michael@0: #if defined(__ARM_FEATURE_UNALIGNED)
michael@0: #  define LZ4_FORCE_UNALIGNED_ACCESS 1
michael@0: #endif
michael@0: 
michael@0: /* Define this parameter if your target system or compiler does not support hardware bit count */
michael@0: #if defined(_MSC_VER) && defined(_WIN32_WCE)   /* Visual Studio for Windows CE does not support Hardware bit count */
michael@0: #  define LZ4_FORCE_SW_BITCOUNT
michael@0: #endif
michael@0: 
michael@0: /*
michael@0:  * BIG_ENDIAN_NATIVE_BUT_INCOMPATIBLE :
michael@0:  * This option may provide a small boost to performance for some big endian cpu, although probably modest.
michael@0:  * You may set this option to 1 if data will remain within closed environment.
michael@0:  * This option is useless on Little_Endian CPU (such as x86)
michael@0:  */
michael@0: 
michael@0: /* #define BIG_ENDIAN_NATIVE_BUT_INCOMPATIBLE 1 */
michael@0: 
michael@0: 
michael@0: /**************************************
michael@0:  Compiler Options
michael@0: **************************************/
michael@0: #if defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L)   /* C99 */
michael@0: /* "restrict" is a known keyword */
michael@0: #else
michael@0: #  define restrict /* Disable restrict */
michael@0: #endif
michael@0: 
michael@0: #ifdef _MSC_VER    /* Visual Studio */
michael@0: #  define FORCE_INLINE static __forceinline
michael@0: #  include <intrin.h>                    /* For Visual 2005 */
michael@0: #  if LZ4_ARCH64   /* 64-bits */
michael@0: #    pragma intrinsic(_BitScanForward64) /* For Visual 2005 */
michael@0: #    pragma intrinsic(_BitScanReverse64) /* For Visual 2005 */
michael@0: #  else            /* 32-bits */
michael@0: #    pragma intrinsic(_BitScanForward)   /* For Visual 2005 */
michael@0: #    pragma intrinsic(_BitScanReverse)   /* For Visual 2005 */
michael@0: #  endif
michael@0: #  pragma warning(disable : 4127)        /* disable: C4127: conditional expression is constant */
michael@0: #else
michael@0: #  ifdef __GNUC__
michael@0: #    define FORCE_INLINE static inline __attribute__((always_inline))
michael@0: #  else
michael@0: #    define FORCE_INLINE static inline
michael@0: #  endif
michael@0: #endif
michael@0: 
michael@0: #ifdef _MSC_VER  /* Visual Studio */
michael@0: #  define lz4_bswap16(x) _byteswap_ushort(x)
michael@0: #else
michael@0: #  define lz4_bswap16(x) ((unsigned short int) ((((x) >> 8) & 0xffu) | (((x) & 0xffu) << 8)))
michael@0: #endif
michael@0: 
michael@0: #define GCC_VERSION (__GNUC__ * 100 + __GNUC_MINOR__)
michael@0: 
michael@0: #if (GCC_VERSION >= 302) || (__INTEL_COMPILER >= 800) || defined(__clang__)
michael@0: #  define expect(expr,value)    (__builtin_expect ((expr),(value)) )
michael@0: #else
michael@0: #  define expect(expr,value)    (expr)
michael@0: #endif
michael@0: 
michael@0: #define likely(expr)     expect((expr) != 0, 1)
michael@0: #define unlikely(expr)   expect((expr) != 0, 0)
michael@0: 
michael@0: 
michael@0: /**************************************
michael@0:    Memory routines
michael@0: **************************************/
michael@0: #include <stdlib.h>   /* malloc, calloc, free */
michael@0: #define ALLOCATOR(n,s) calloc(n,s)
michael@0: #define FREEMEM        free
michael@0: #include <string.h>   /* memset, memcpy */
michael@0: #define MEM_INIT       memset
michael@0: 
michael@0: 
michael@0: /**************************************
michael@0:    Includes
michael@0: **************************************/
michael@0: #include "lz4.h"
michael@0: 
michael@0: 
michael@0: /**************************************
michael@0:    Basic Types
michael@0: **************************************/
michael@0: #if defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L)   /* C99 */
michael@0: # include <stdint.h>
michael@0:   typedef  uint8_t BYTE;
michael@0:   typedef uint16_t U16;
michael@0:   typedef uint32_t U32;
michael@0:   typedef  int32_t S32;
michael@0:   typedef uint64_t U64;
michael@0: #else
michael@0:   typedef unsigned char       BYTE;
michael@0:   typedef unsigned short      U16;
michael@0:   typedef unsigned int        U32;
michael@0:   typedef   signed int        S32;
michael@0:   typedef unsigned long long  U64;
michael@0: #endif
michael@0: 
michael@0: #if defined(__GNUC__)  && !defined(LZ4_FORCE_UNALIGNED_ACCESS)
michael@0: #  define _PACKED __attribute__ ((packed))
michael@0: #else
michael@0: #  define _PACKED
michael@0: #endif
michael@0: 
michael@0: #if !defined(LZ4_FORCE_UNALIGNED_ACCESS) && !defined(__GNUC__)
michael@0: #  if defined(__IBMC__) || defined(__SUNPRO_C) || defined(__SUNPRO_CC)
michael@0: #    pragma pack(1)
michael@0: #  else
michael@0: #    pragma pack(push, 1)
michael@0: #  endif
michael@0: #endif
michael@0: 
michael@0: typedef struct { U16 v; }  _PACKED U16_S;
michael@0: typedef struct { U32 v; }  _PACKED U32_S;
michael@0: typedef struct { U64 v; }  _PACKED U64_S;
michael@0: typedef struct {size_t v;} _PACKED size_t_S;
michael@0: 
michael@0: #if !defined(LZ4_FORCE_UNALIGNED_ACCESS) && !defined(__GNUC__)
michael@0: #  if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
michael@0: #    pragma pack(0)
michael@0: #  else
michael@0: #    pragma pack(pop)
michael@0: #  endif
michael@0: #endif
michael@0: 
michael@0: #define A16(x)   (((U16_S *)(x))->v)
michael@0: #define A32(x)   (((U32_S *)(x))->v)
michael@0: #define A64(x)   (((U64_S *)(x))->v)
michael@0: #define AARCH(x) (((size_t_S *)(x))->v)
michael@0: 
michael@0: 
michael@0: /**************************************
michael@0:    Constants
michael@0: **************************************/
michael@0: #define LZ4_HASHLOG   (LZ4_MEMORY_USAGE-2)
michael@0: #define HASHTABLESIZE (1 << LZ4_MEMORY_USAGE)
michael@0: #define HASH_SIZE_U32 (1 << LZ4_HASHLOG)
michael@0: 
michael@0: #define MINMATCH 4
michael@0: 
michael@0: #define COPYLENGTH 8
michael@0: #define LASTLITERALS 5
michael@0: #define MFLIMIT (COPYLENGTH+MINMATCH)
michael@0: static const int LZ4_minLength = (MFLIMIT+1);
michael@0: 
michael@0: #define KB *(1U<<10)
michael@0: #define MB *(1U<<20)
michael@0: #define GB *(1U<<30)
michael@0: 
michael@0: #define LZ4_64KLIMIT ((64 KB) + (MFLIMIT-1))
michael@0: #define SKIPSTRENGTH 6   /* Increasing this value will make the compression run slower on incompressible data */
michael@0: 
michael@0: #define MAXD_LOG 16
michael@0: #define MAX_DISTANCE ((1 << MAXD_LOG) - 1)
michael@0: 
michael@0: #define ML_BITS  4
michael@0: #define ML_MASK  ((1U<<ML_BITS)-1)
michael@0: #define RUN_BITS (8-ML_BITS)
michael@0: #define RUN_MASK ((1U<<RUN_BITS)-1)
michael@0: 
michael@0: 
michael@0: /**************************************
michael@0:    Structures and local types
michael@0: **************************************/
michael@0: typedef struct {
michael@0:     U32  hashTable[HASH_SIZE_U32];
michael@0:     U32  currentOffset;
michael@0:     U32  initCheck;
michael@0:     const BYTE* dictionary;
michael@0:     const BYTE* bufferStart;
michael@0:     U32  dictSize;
michael@0: } LZ4_stream_t_internal;
michael@0: 
michael@0: typedef enum { notLimited = 0, limitedOutput = 1 } limitedOutput_directive;
michael@0: typedef enum { byPtr, byU32, byU16 } tableType_t;
michael@0: 
michael@0: typedef enum { noDict = 0, withPrefix64k, usingExtDict } dict_directive;
michael@0: typedef enum { noDictIssue = 0, dictSmall } dictIssue_directive;
michael@0: 
michael@0: typedef enum { endOnOutputSize = 0, endOnInputSize = 1 } endCondition_directive;
michael@0: typedef enum { full = 0, partial = 1 } earlyEnd_directive;
michael@0: 
michael@0: 
michael@0: /**************************************
michael@0:    Architecture-specific macros
michael@0: **************************************/
michael@0: #define STEPSIZE                  sizeof(size_t)
michael@0: #define LZ4_COPYSTEP(d,s)         { AARCH(d) = AARCH(s); d+=STEPSIZE; s+=STEPSIZE; }
michael@0: #define LZ4_COPY8(d,s)            { LZ4_COPYSTEP(d,s); if (STEPSIZE<8) LZ4_COPYSTEP(d,s); }
michael@0: 
michael@0: #if (defined(LZ4_BIG_ENDIAN) && !defined(BIG_ENDIAN_NATIVE_BUT_INCOMPATIBLE))
michael@0: #  define LZ4_READ_LITTLEENDIAN_16(d,s,p) { U16 v = A16(p); v = lz4_bswap16(v); d = (s) - v; }
michael@0: #  define LZ4_WRITE_LITTLEENDIAN_16(p,i)  { U16 v = (U16)(i); v = lz4_bswap16(v); A16(p) = v; p+=2; }
michael@0: #else      /* Little Endian */
michael@0: #  define LZ4_READ_LITTLEENDIAN_16(d,s,p) { d = (s) - A16(p); }
michael@0: #  define LZ4_WRITE_LITTLEENDIAN_16(p,v)  { A16(p) = v; p+=2; }
michael@0: #endif
michael@0: 
michael@0: 
michael@0: /**************************************
michael@0:    Macros
michael@0: **************************************/
michael@0: #define LZ4_STATIC_ASSERT(c)    { enum { LZ4_static_assert = 1/(!!(c)) }; }   /* use only *after* variable declarations */
michael@0: #if LZ4_ARCH64 || !defined(__GNUC__)
michael@0: #  define LZ4_WILDCOPY(d,s,e)   { do { LZ4_COPY8(d,s) } while (d<e); }        /* at the end, d>=e; */
michael@0: #else
michael@0: #  define LZ4_WILDCOPY(d,s,e)   { if (likely(e-d <= 8)) LZ4_COPY8(d,s) else do { LZ4_COPY8(d,s) } while (d<e); }
michael@0: #endif
michael@0: 
michael@0: 
michael@0: /****************************
michael@0:    Private local functions
michael@0: ****************************/
michael@0: #if LZ4_ARCH64
michael@0: 
michael@0: int LZ4_NbCommonBytes (register U64 val)
michael@0: {
michael@0: # if defined(LZ4_BIG_ENDIAN)
michael@0: #   if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
michael@0:     unsigned long r = 0;
michael@0:     _BitScanReverse64( &r, val );
michael@0:     return (int)(r>>3);
michael@0: #   elif defined(__GNUC__) && (GCC_VERSION >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
michael@0:     return (__builtin_clzll(val) >> 3);
michael@0: #   else
michael@0:     int r;
michael@0:     if (!(val>>32)) { r=4; } else { r=0; val>>=32; }
michael@0:     if (!(val>>16)) { r+=2; val>>=8; } else { val>>=24; }
michael@0:     r += (!val);
michael@0:     return r;
michael@0: #   endif
michael@0: # else
michael@0: #   if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
michael@0:     unsigned long r = 0;
michael@0:     _BitScanForward64( &r, val );
michael@0:     return (int)(r>>3);
michael@0: #   elif defined(__GNUC__) && (GCC_VERSION >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
michael@0:     return (__builtin_ctzll(val) >> 3);
michael@0: #   else
michael@0:     static const int DeBruijnBytePos[64] = { 0, 0, 0, 0, 0, 1, 1, 2, 0, 3, 1, 3, 1, 4, 2, 7, 0, 2, 3, 6, 1, 5, 3, 5, 1, 3, 4, 4, 2, 5, 6, 7, 7, 0, 1, 2, 3, 3, 4, 6, 2, 6, 5, 5, 3, 4, 5, 6, 7, 1, 2, 4, 6, 4, 4, 5, 7, 2, 6, 5, 7, 6, 7, 7 };
michael@0:     return DeBruijnBytePos[((U64)((val & -(long long)val) * 0x0218A392CDABBD3FULL)) >> 58];
michael@0: #   endif
michael@0: # endif
michael@0: }
michael@0: 
michael@0: #else
michael@0: 
michael@0: int LZ4_NbCommonBytes (register U32 val)
michael@0: {
michael@0: # if defined(LZ4_BIG_ENDIAN)
michael@0: #   if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
michael@0:     unsigned long r = 0;
michael@0:     _BitScanReverse( &r, val );
michael@0:     return (int)(r>>3);
michael@0: #   elif defined(__GNUC__) && (GCC_VERSION >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
michael@0:     return (__builtin_clz(val) >> 3);
michael@0: #   else
michael@0:     int r;
michael@0:     if (!(val>>16)) { r=2; val>>=8; } else { r=0; val>>=24; }
michael@0:     r += (!val);
michael@0:     return r;
michael@0: #   endif
michael@0: # else
michael@0: #   if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
michael@0:     unsigned long r;
michael@0:     _BitScanForward( &r, val );
michael@0:     return (int)(r>>3);
michael@0: #   elif defined(__GNUC__) && (GCC_VERSION >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
michael@0:     return (__builtin_ctz(val) >> 3);
michael@0: #   else
michael@0:     static const int DeBruijnBytePos[32] = { 0, 0, 3, 0, 3, 1, 3, 0, 3, 2, 2, 1, 3, 2, 0, 1, 3, 3, 1, 2, 2, 2, 2, 0, 3, 1, 2, 0, 1, 0, 1, 1 };
michael@0:     return DeBruijnBytePos[((U32)((val & -(S32)val) * 0x077CB531U)) >> 27];
michael@0: #   endif
michael@0: # endif
michael@0: }
michael@0: 
michael@0: #endif
michael@0: 
michael@0: 
michael@0: /********************************
michael@0:    Compression functions
michael@0: ********************************/
michael@0: int LZ4_compressBound(int isize)  { return LZ4_COMPRESSBOUND(isize); }
michael@0: 
michael@0: static int LZ4_hashSequence(U32 sequence, tableType_t tableType)
michael@0: {
michael@0:     if (tableType == byU16)
michael@0:         return (((sequence) * 2654435761U) >> ((MINMATCH*8)-(LZ4_HASHLOG+1)));
michael@0:     else
michael@0:         return (((sequence) * 2654435761U) >> ((MINMATCH*8)-LZ4_HASHLOG));
michael@0: }
michael@0: 
michael@0: static int LZ4_hashPosition(const BYTE* p, tableType_t tableType) { return LZ4_hashSequence(A32(p), tableType); }
michael@0: 
michael@0: static void LZ4_putPositionOnHash(const BYTE* p, U32 h, void* tableBase, tableType_t tableType, const BYTE* srcBase)
michael@0: {
michael@0:     switch (tableType)
michael@0:     {
michael@0:     case byPtr: { const BYTE** hashTable = (const BYTE**) tableBase; hashTable[h] = p; break; }
michael@0:     case byU32: { U32* hashTable = (U32*) tableBase; hashTable[h] = (U32)(p-srcBase); break; }
michael@0:     case byU16: { U16* hashTable = (U16*) tableBase; hashTable[h] = (U16)(p-srcBase); break; }
michael@0:     }
michael@0: }
michael@0: 
michael@0: static void LZ4_putPosition(const BYTE* p, void* tableBase, tableType_t tableType, const BYTE* srcBase)
michael@0: {
michael@0:     U32 h = LZ4_hashPosition(p, tableType);
michael@0:     LZ4_putPositionOnHash(p, h, tableBase, tableType, srcBase);
michael@0: }
michael@0: 
michael@0: static const BYTE* LZ4_getPositionOnHash(U32 h, void* tableBase, tableType_t tableType, const BYTE* srcBase)
michael@0: {
michael@0:     if (tableType == byPtr) { const BYTE** hashTable = (const BYTE**) tableBase; return hashTable[h]; }
michael@0:     if (tableType == byU32) { U32* hashTable = (U32*) tableBase; return hashTable[h] + srcBase; }
michael@0:     { U16* hashTable = (U16*) tableBase; return hashTable[h] + srcBase; }   /* default, to ensure a return */
michael@0: }
michael@0: 
michael@0: static const BYTE* LZ4_getPosition(const BYTE* p, void* tableBase, tableType_t tableType, const BYTE* srcBase)
michael@0: {
michael@0:     U32 h = LZ4_hashPosition(p, tableType);
michael@0:     return LZ4_getPositionOnHash(h, tableBase, tableType, srcBase);
michael@0: }
michael@0: 
michael@0: static unsigned LZ4_count(const BYTE* pIn, const BYTE* pRef, const BYTE* pInLimit)
michael@0: {
michael@0:     const BYTE* const pStart = pIn;
michael@0: 
michael@0:     while (likely(pIn<pInLimit-(STEPSIZE-1)))
michael@0:     {
michael@0:         size_t diff = AARCH(pRef) ^ AARCH(pIn);
michael@0:         if (!diff) { pIn+=STEPSIZE; pRef+=STEPSIZE; continue; }
michael@0:         pIn += LZ4_NbCommonBytes(diff);
michael@0:         return (unsigned)(pIn - pStart);
michael@0:     }
michael@0:     if (sizeof(void*)==8) if ((pIn<(pInLimit-3)) && (A32(pRef) == A32(pIn))) { pIn+=4; pRef+=4; }
michael@0:     if ((pIn<(pInLimit-1)) && (A16(pRef) == A16(pIn))) { pIn+=2; pRef+=2; }
michael@0:     if ((pIn<pInLimit) && (*pRef == *pIn)) pIn++;
michael@0: 
michael@0:     return (unsigned)(pIn - pStart);
michael@0: }
michael@0: 
michael@0: 
michael@0: static int LZ4_compress_generic(
michael@0:                  void* ctx,
michael@0:                  const char* source,
michael@0:                  char* dest,
michael@0:                  int inputSize,
michael@0:                  int maxOutputSize,
michael@0: 
michael@0:                  limitedOutput_directive outputLimited,
michael@0:                  tableType_t tableType,
michael@0:                  dict_directive dict,
michael@0:                  dictIssue_directive dictIssue)
michael@0: {
michael@0:     LZ4_stream_t_internal* const dictPtr = (LZ4_stream_t_internal*)ctx;
michael@0: 
michael@0:     const BYTE* ip = (const BYTE*) source;
michael@0:     const BYTE* base;
michael@0:     const BYTE* lowLimit;
michael@0:     const BYTE* const lowRefLimit = ip - dictPtr->dictSize;
michael@0:     const BYTE* const dictionary = dictPtr->dictionary;
michael@0:     const BYTE* const dictEnd = dictionary + dictPtr->dictSize;
michael@0:     const size_t dictDelta = dictEnd - (const BYTE*)source;
michael@0:     const BYTE* anchor = (const BYTE*) source;
michael@0:     const BYTE* const iend = ip + inputSize;
michael@0:     const BYTE* const mflimit = iend - MFLIMIT;
michael@0:     const BYTE* const matchlimit = iend - LASTLITERALS;
michael@0: 
michael@0:     BYTE* op = (BYTE*) dest;
michael@0:     BYTE* const olimit = op + maxOutputSize;
michael@0: 
michael@0:     const int skipStrength = SKIPSTRENGTH;
michael@0:     U32 forwardH;
michael@0:     size_t refDelta=0;
michael@0: 
michael@0:     /* Init conditions */
michael@0:     if ((U32)inputSize > (U32)LZ4_MAX_INPUT_SIZE) return 0;          /* Unsupported input size, too large (or negative) */
michael@0:     switch(dict)
michael@0:     {
michael@0:     case noDict:
michael@0:     default:
michael@0:         base = (const BYTE*)source;
michael@0:         lowLimit = (const BYTE*)source;
michael@0:         break;
michael@0:     case withPrefix64k:
michael@0:         base = (const BYTE*)source - dictPtr->currentOffset;
michael@0:         lowLimit = (const BYTE*)source - dictPtr->dictSize;
michael@0:         break;
michael@0:     case usingExtDict:
michael@0:         base = (const BYTE*)source - dictPtr->currentOffset;
michael@0:         lowLimit = (const BYTE*)source;
michael@0:         break;
michael@0:     }
michael@0:     if ((tableType == byU16) && (inputSize>=(int)LZ4_64KLIMIT)) return 0;   /* Size too large (not within 64K limit) */
michael@0:     if (inputSize<LZ4_minLength) goto _last_literals;                       /* Input too small, no compression (all literals) */
michael@0: 
michael@0:     /* First Byte */
michael@0:     LZ4_putPosition(ip, ctx, tableType, base);
michael@0:     ip++; forwardH = LZ4_hashPosition(ip, tableType);
michael@0: 
michael@0:     /* Main Loop */
michael@0:     for ( ; ; )
michael@0:     {
michael@0:         const BYTE* ref;
michael@0:         BYTE* token;
michael@0:         {
michael@0:             const BYTE* forwardIp = ip;
michael@0:             unsigned step=1;
michael@0:             unsigned searchMatchNb = (1U << skipStrength);
michael@0: 
michael@0:             /* Find a match */
michael@0:             do {
michael@0:                 U32 h = forwardH;
michael@0:                 ip = forwardIp;
michael@0:                 forwardIp += step;
michael@0:                 step = searchMatchNb++ >> skipStrength;
michael@0:                 //if (step>8) step=8;   // required for valid forwardIp ; slows down uncompressible data a bit
michael@0: 
michael@0:                 if (unlikely(forwardIp > mflimit)) goto _last_literals;
michael@0: 
michael@0:                 ref = LZ4_getPositionOnHash(h, ctx, tableType, base);
michael@0:                 if (dict==usingExtDict)
michael@0:                 {
michael@0:                     if (ref<(const BYTE*)source)
michael@0:                     {
michael@0:                         refDelta = dictDelta;
michael@0:                         lowLimit = dictionary;
michael@0:                     }
michael@0:                     else
michael@0:                     {
michael@0:                         refDelta = 0;
michael@0:                         lowLimit = (const BYTE*)source;
michael@0:                     }
michael@0:                 }
michael@0:                 forwardH = LZ4_hashPosition(forwardIp, tableType);
michael@0:                 LZ4_putPositionOnHash(ip, h, ctx, tableType, base);
michael@0: 
michael@0:             } while ( ((dictIssue==dictSmall) ? (ref < lowRefLimit) : 0)
michael@0:                 || ((tableType==byU16) ? 0 : (ref + MAX_DISTANCE < ip))
michael@0:                 || (A32(ref+refDelta) != A32(ip)) );
michael@0:         }
michael@0: 
michael@0:         /* Catch up */
michael@0:         while ((ip>anchor) && (ref+refDelta > lowLimit) && (unlikely(ip[-1]==ref[refDelta-1]))) { ip--; ref--; }
michael@0: 
michael@0:         {
michael@0:             /* Encode Literal length */
michael@0:             unsigned litLength = (unsigned)(ip - anchor);
michael@0:             token = op++;
michael@0:             if ((outputLimited) && (unlikely(op + litLength + (2 + 1 + LASTLITERALS) + (litLength/255) > olimit)))
michael@0:                 return 0;   /* Check output limit */
michael@0:             if (litLength>=RUN_MASK)
michael@0:             {
michael@0:                 int len = (int)litLength-RUN_MASK;
michael@0:                 *token=(RUN_MASK<<ML_BITS);
michael@0:                 for(; len >= 255 ; len-=255) *op++ = 255;
michael@0:                 *op++ = (BYTE)len;
michael@0:             }
michael@0:             else *token = (BYTE)(litLength<<ML_BITS);
michael@0: 
michael@0:             /* Copy Literals */
michael@0:             { BYTE* end = op+litLength; LZ4_WILDCOPY(op,anchor,end); op=end; }
michael@0:         }
michael@0: 
michael@0: _next_match:
michael@0:         /* Encode Offset */
michael@0:         LZ4_WRITE_LITTLEENDIAN_16(op, (U16)(ip-ref));
michael@0: 
michael@0:         /* Encode MatchLength */
michael@0:         {
michael@0:             unsigned matchLength;
michael@0: 
michael@0:             if ((dict==usingExtDict) && (lowLimit==dictionary))
michael@0:             {
michael@0:                 const BYTE* limit;
michael@0:                 ref += refDelta;
michael@0:                 limit = ip + (dictEnd-ref);
michael@0:                 if (limit > matchlimit) limit = matchlimit;
michael@0:                 matchLength = LZ4_count(ip+MINMATCH, ref+MINMATCH, limit);
michael@0:                 ip += MINMATCH + matchLength;
michael@0:                 if (ip==limit)
michael@0:                 {
michael@0:                     unsigned more = LZ4_count(ip, (const BYTE*)source, matchlimit);
michael@0:                     matchLength += more;
michael@0:                     ip += more;
michael@0:                 }
michael@0:             }
michael@0:             else
michael@0:             {
michael@0:                 matchLength = LZ4_count(ip+MINMATCH, ref+MINMATCH, matchlimit);
michael@0:                 ip += MINMATCH + matchLength;
michael@0:             }
michael@0: 
michael@0:             if (matchLength>=ML_MASK)
michael@0:             {
michael@0:                 if ((outputLimited) && (unlikely(op + (1 + LASTLITERALS) + (matchLength>>8) > olimit)))
michael@0:                     return 0;    /* Check output limit */
michael@0:                 *token += ML_MASK;
michael@0:                 matchLength -= ML_MASK;
michael@0:                 for (; matchLength >= 510 ; matchLength-=510) { *op++ = 255; *op++ = 255; }
michael@0:                 if (matchLength >= 255) { matchLength-=255; *op++ = 255; }
michael@0:                 *op++ = (BYTE)matchLength;
michael@0:             }
michael@0:             else *token += (BYTE)(matchLength);
michael@0:         }
michael@0: 
michael@0:         anchor = ip;
michael@0: 
michael@0:         /* Test end of chunk */
michael@0:         if (ip > mflimit) break;
michael@0: 
michael@0:         /* Fill table */
michael@0:         LZ4_putPosition(ip-2, ctx, tableType, base);
michael@0: 
michael@0:         /* Test next position */
michael@0:         ref = LZ4_getPosition(ip, ctx, tableType, base);
michael@0:         if (dict==usingExtDict)
michael@0:         {
michael@0:             if (ref<(const BYTE*)source)
michael@0:             {
michael@0:                 refDelta = dictDelta;
michael@0:                 lowLimit = dictionary;
michael@0:             }
michael@0:             else
michael@0:             {
michael@0:                 refDelta = 0;
michael@0:                 lowLimit = (const BYTE*)source;
michael@0:             }
michael@0:         }
michael@0:         LZ4_putPosition(ip, ctx, tableType, base);
michael@0:         if ( ((dictIssue==dictSmall) ? (ref>=lowRefLimit) : 1)
michael@0:             && (ref+MAX_DISTANCE>=ip)
michael@0:             && (A32(ref+refDelta)==A32(ip)) )
michael@0:         { token=op++; *token=0; goto _next_match; }
michael@0: 
michael@0:         /* Prepare next loop */
michael@0:         forwardH = LZ4_hashPosition(++ip, tableType);
michael@0:     }
michael@0: 
michael@0: _last_literals:
michael@0:     /* Encode Last Literals */
michael@0:     {
michael@0:         int lastRun = (int)(iend - anchor);
michael@0:         if ((outputLimited) && (((char*)op - dest) + lastRun + 1 + ((lastRun+255-RUN_MASK)/255) > (U32)maxOutputSize))
michael@0:             return 0;   /* Check output limit */
michael@0:         if (lastRun>=(int)RUN_MASK) { *op++=(RUN_MASK<<ML_BITS); lastRun-=RUN_MASK; for(; lastRun >= 255 ; lastRun-=255) *op++ = 255; *op++ = (BYTE) lastRun; }
michael@0:         else *op++ = (BYTE)(lastRun<<ML_BITS);
michael@0:         memcpy(op, anchor, iend - anchor);
michael@0:         op += iend-anchor;
michael@0:     }
michael@0: 
michael@0:     /* End */
michael@0:     return (int) (((char*)op)-dest);
michael@0: }
michael@0: 
michael@0: 
michael@0: int LZ4_compress(const char* source, char* dest, int inputSize)
michael@0: {
michael@0: #if (HEAPMODE)
michael@0:     void* ctx = ALLOCATOR(LZ4_STREAMSIZE_U32, 4);   /* Aligned on 4-bytes boundaries */
michael@0: #else
michael@0:     U32 ctx[LZ4_STREAMSIZE_U32] = {0};      /* Ensure data is aligned on 4-bytes boundaries */
michael@0: #endif
michael@0:     int result;
michael@0: 
michael@0:     if (inputSize < (int)LZ4_64KLIMIT)
michael@0:         result = LZ4_compress_generic((void*)ctx, source, dest, inputSize, 0, notLimited, byU16, noDict, noDictIssue);
michael@0:     else
michael@0:         result = LZ4_compress_generic((void*)ctx, source, dest, inputSize, 0, notLimited, (sizeof(void*)==8) ? byU32 : byPtr, noDict, noDictIssue);
michael@0: 
michael@0: #if (HEAPMODE)
michael@0:     FREEMEM(ctx);
michael@0: #endif
michael@0:     return result;
michael@0: }
michael@0: 
michael@0: int LZ4_compress_limitedOutput(const char* source, char* dest, int inputSize, int maxOutputSize)
michael@0: {
michael@0: #if (HEAPMODE)
michael@0:     void* ctx = ALLOCATOR(LZ4_STREAMSIZE_U32, 4);   /* Aligned on 4-bytes boundaries */
michael@0: #else
michael@0:     U32 ctx[LZ4_STREAMSIZE_U32] = {0};      /* Ensure data is aligned on 4-bytes boundaries */
michael@0: #endif
michael@0:     int result;
michael@0: 
michael@0:     if (inputSize < (int)LZ4_64KLIMIT)
michael@0:         result = LZ4_compress_generic((void*)ctx, source, dest, inputSize, maxOutputSize, limitedOutput, byU16, noDict, noDictIssue);
michael@0:     else
michael@0:         result = LZ4_compress_generic((void*)ctx, source, dest, inputSize, maxOutputSize, limitedOutput, (sizeof(void*)==8) ? byU32 : byPtr, noDict, noDictIssue);
michael@0: 
michael@0: #if (HEAPMODE)
michael@0:     FREEMEM(ctx);
michael@0: #endif
michael@0:     return result;
michael@0: }
michael@0: 
michael@0: 
michael@0: /*****************************************
michael@0:    Experimental : Streaming functions
michael@0: *****************************************/
michael@0: 
michael@0: void* LZ4_createStream()
michael@0: {
michael@0:     void* lz4s = ALLOCATOR(4, LZ4_STREAMSIZE_U32);
michael@0:     MEM_INIT(lz4s, 0, LZ4_STREAMSIZE);
michael@0:     return lz4s;
michael@0: }
michael@0: 
michael@0: int LZ4_free (void* LZ4_stream)
michael@0: {
michael@0:     FREEMEM(LZ4_stream);
michael@0:     return (0);
michael@0: }
michael@0: 
michael@0: 
michael@0: int LZ4_loadDict (void* LZ4_dict, const char* dictionary, int dictSize)
michael@0: {
michael@0:     LZ4_stream_t_internal* dict = (LZ4_stream_t_internal*) LZ4_dict;
michael@0:     const BYTE* p = (const BYTE*)dictionary;
michael@0:     const BYTE* const dictEnd = p + dictSize;
michael@0:     const BYTE* base;
michael@0: 
michael@0:     LZ4_STATIC_ASSERT(LZ4_STREAMSIZE >= sizeof(LZ4_stream_t_internal));      /* A compilation error here means LZ4_STREAMSIZE is not large enough */
michael@0:     if (dict->initCheck) MEM_INIT(dict, 0, sizeof(LZ4_stream_t_internal));   /* Uninitialized structure detected */
michael@0: 
michael@0:     if (dictSize < MINMATCH)
michael@0:     {
michael@0:         dict->dictionary = NULL;
michael@0:         dict->dictSize = 0;
michael@0:         return 1;
michael@0:     }
michael@0: 
michael@0:     if (p <= dictEnd - 64 KB) p = dictEnd - 64 KB;
michael@0:     base = p - dict->currentOffset;
michael@0:     dict->dictionary = p;
michael@0:     dict->dictSize = (U32)(dictEnd - p);
michael@0:     dict->currentOffset += dict->dictSize;
michael@0: 
michael@0:     while (p <= dictEnd-MINMATCH)
michael@0:     {
michael@0:         LZ4_putPosition(p, dict, byU32, base);
michael@0:         p+=3;
michael@0:     }
michael@0: 
michael@0:     return 1;
michael@0: }
michael@0: 
michael@0: 
michael@0: void LZ4_renormDictT(LZ4_stream_t_internal* LZ4_dict, const BYTE* src)
michael@0: {
michael@0:     if ((LZ4_dict->currentOffset > 0x80000000) ||
michael@0:         ((size_t)LZ4_dict->currentOffset > (size_t)src))   /* address space overflow */
michael@0:     {
michael@0:         /* rescale hash table */
michael@0:         U32 delta = LZ4_dict->currentOffset - 64 KB;
michael@0:         const BYTE* dictEnd = LZ4_dict->dictionary + LZ4_dict->dictSize;
michael@0:         int i;
michael@0:         for (i=0; i<HASH_SIZE_U32; i++)
michael@0:         {
michael@0:             if (LZ4_dict->hashTable[i] < delta) LZ4_dict->hashTable[i]=0;
michael@0:             else LZ4_dict->hashTable[i] -= delta;
michael@0:         }
michael@0:         LZ4_dict->currentOffset = 64 KB;
michael@0:         if (LZ4_dict->dictSize > 64 KB) LZ4_dict->dictSize = 64 KB;
michael@0:         LZ4_dict->dictionary = dictEnd - LZ4_dict->dictSize;
michael@0:     }
michael@0: }
michael@0: 
michael@0: 
michael@0: FORCE_INLINE int LZ4_compress_continue_generic (void* LZ4_stream, const char* source, char* dest, int inputSize,
michael@0:                                                 int maxOutputSize, limitedOutput_directive limit)
michael@0: {
michael@0:     LZ4_stream_t_internal* streamPtr = (LZ4_stream_t_internal*)LZ4_stream;
michael@0:     const BYTE* const dictEnd = streamPtr->dictionary + streamPtr->dictSize;
michael@0: 
michael@0:     const BYTE* smallest = (const BYTE*) source;
michael@0:     if (streamPtr->initCheck) return 0;   /* Uninitialized structure detected */
michael@0:     if ((streamPtr->dictSize>0) && (smallest>dictEnd)) smallest = dictEnd;
michael@0:     LZ4_renormDictT(streamPtr, smallest);
michael@0: 
michael@0:     /* Check overlapping input/dictionary space */
michael@0:     {
michael@0:         const BYTE* sourceEnd = (const BYTE*) source + inputSize;
michael@0:         if ((sourceEnd > streamPtr->dictionary) && (sourceEnd < dictEnd))
michael@0:         {
michael@0:             streamPtr->dictSize = (U32)(dictEnd - sourceEnd);
michael@0:             if (streamPtr->dictSize > 64 KB) streamPtr->dictSize = 64 KB;
michael@0:             if (streamPtr->dictSize < 4) streamPtr->dictSize = 0;
michael@0:             streamPtr->dictionary = dictEnd - streamPtr->dictSize;
michael@0:         }
michael@0:     }
michael@0: 
michael@0:     /* prefix mode : source data follows dictionary */
michael@0:     if (dictEnd == (const BYTE*)source)
michael@0:     {
michael@0:         int result;
michael@0:         if ((streamPtr->dictSize < 64 KB) && (streamPtr->dictSize < streamPtr->currentOffset))
michael@0:             result = LZ4_compress_generic(LZ4_stream, source, dest, inputSize, maxOutputSize, limit, byU32, withPrefix64k, dictSmall);
michael@0:         else
michael@0:             result = LZ4_compress_generic(LZ4_stream, source, dest, inputSize, maxOutputSize, limit, byU32, withPrefix64k, noDictIssue);
michael@0:         streamPtr->dictSize += (U32)inputSize;
michael@0:         streamPtr->currentOffset += (U32)inputSize;
michael@0:         return result;
michael@0:     }
michael@0: 
michael@0:     /* external dictionary mode */
michael@0:     {
michael@0:         int result;
michael@0:         if ((streamPtr->dictSize < 64 KB) && (streamPtr->dictSize < streamPtr->currentOffset))
michael@0:             result = LZ4_compress_generic(LZ4_stream, source, dest, inputSize, maxOutputSize, limit, byU32, usingExtDict, dictSmall);
michael@0:         else
michael@0:             result = LZ4_compress_generic(LZ4_stream, source, dest, inputSize, maxOutputSize, limit, byU32, usingExtDict, noDictIssue);
michael@0:         streamPtr->dictionary = (const BYTE*)source;
michael@0:         streamPtr->dictSize = (U32)inputSize;
michael@0:         streamPtr->currentOffset += (U32)inputSize;
michael@0:         return result;
michael@0:     }
michael@0: }
michael@0: 
michael@0: 
michael@0: int LZ4_compress_continue (void* LZ4_stream, const char* source, char* dest, int inputSize)
michael@0: {
michael@0:     return LZ4_compress_continue_generic(LZ4_stream, source, dest, inputSize, 0, notLimited);
michael@0: }
michael@0: 
michael@0: int LZ4_compress_limitedOutput_continue (void* LZ4_stream, const char* source, char* dest, int inputSize, int maxOutputSize)
michael@0: {
michael@0:     return LZ4_compress_continue_generic(LZ4_stream, source, dest, inputSize, maxOutputSize, limitedOutput);
michael@0: }
michael@0: 
michael@0: 
michael@0: // Hidden debug function, to force separate dictionary mode
michael@0: int LZ4_compress_forceExtDict (LZ4_stream_t* LZ4_dict, const char* source, char* dest, int inputSize)
michael@0: {
michael@0:     LZ4_stream_t_internal* streamPtr = (LZ4_stream_t_internal*)LZ4_dict;
michael@0:     int result;
michael@0:     const BYTE* const dictEnd = streamPtr->dictionary + streamPtr->dictSize;
michael@0: 
michael@0:     const BYTE* smallest = dictEnd;
michael@0:     if (smallest > (const BYTE*) source) smallest = (const BYTE*) source;
michael@0:     LZ4_renormDictT((LZ4_stream_t_internal*)LZ4_dict, smallest);
michael@0: 
michael@0:     result = LZ4_compress_generic(LZ4_dict, source, dest, inputSize, 0, notLimited, byU32, usingExtDict, noDictIssue);
michael@0: 
michael@0:     streamPtr->dictionary = (const BYTE*)source;
michael@0:     streamPtr->dictSize = (U32)inputSize;
michael@0:     streamPtr->currentOffset += (U32)inputSize;
michael@0: 
michael@0:     return result;
michael@0: }
michael@0: 
michael@0: 
michael@0: int LZ4_saveDict (void* LZ4_dict, char* safeBuffer, int dictSize)
michael@0: {
michael@0:     LZ4_stream_t_internal* dict = (LZ4_stream_t_internal*) LZ4_dict;
michael@0:     const BYTE* previousDictEnd = dict->dictionary + dict->dictSize;
michael@0: 
michael@0:     if ((U32)dictSize > 64 KB) dictSize = 64 KB;   /* useless to define a dictionary > 64 KB */
michael@0:     if ((U32)dictSize > dict->dictSize) dictSize = dict->dictSize;
michael@0: 
michael@0:     memcpy(safeBuffer, previousDictEnd - dictSize, dictSize);
michael@0: 
michael@0:     dict->dictionary = (const BYTE*)safeBuffer;
michael@0:     dict->dictSize = (U32)dictSize;
michael@0: 
michael@0:     return 1;
michael@0: }
michael@0: 
michael@0: 
michael@0: 
michael@0: /****************************
michael@0:    Decompression functions
michael@0: ****************************/
michael@0: /*
michael@0:  * This generic decompression function cover all use cases.
michael@0:  * It shall be instanciated several times, using different sets of directives
michael@0:  * Note that it is essential this generic function is really inlined,
michael@0:  * in order to remove useless branches during compilation optimisation.
michael@0:  */
michael@0: FORCE_INLINE int LZ4_decompress_generic(
michael@0:                  const char* source,
michael@0:                  char* dest,
michael@0:                  int inputSize,
michael@0:                  int outputSize,         /* If endOnInput==endOnInputSize, this value is the max size of Output Buffer. */
michael@0: 
michael@0:                  int endOnInput,         /* endOnOutputSize, endOnInputSize */
michael@0:                  int partialDecoding,    /* full, partial */
michael@0:                  int targetOutputSize,   /* only used if partialDecoding==partial */
michael@0:                  int dict,               /* noDict, withPrefix64k, usingExtDict */
michael@0:                  const char* dictStart,  /* only if dict==usingExtDict */
michael@0:                  int dictSize            /* note : = 0 if noDict */
michael@0:                  )
michael@0: {
michael@0:     /* Local Variables */
michael@0:     const BYTE* restrict ip = (const BYTE*) source;
michael@0:     const BYTE* ref;
michael@0:     const BYTE* const iend = ip + inputSize;
michael@0: 
michael@0:     BYTE* op = (BYTE*) dest;
michael@0:     BYTE* const oend = op + outputSize;
michael@0:     BYTE* cpy;
michael@0:     BYTE* oexit = op + targetOutputSize;
michael@0:     const BYTE* const lowLimit = (const BYTE*)dest - dictSize;
michael@0: 
michael@0:     const BYTE* const dictEnd = (const BYTE*)dictStart + dictSize;
michael@0: //#define OLD
michael@0: #ifdef OLD
michael@0:     const size_t dec32table[] = {0, 3, 2, 3, 0, 0, 0, 0};   /* static reduces speed for LZ4_decompress_safe() on GCC64 */
michael@0: #else
michael@0:     const size_t dec32table[] = {4-0, 4-3, 4-2, 4-3, 4-0, 4-0, 4-0, 4-0};   /* static reduces speed for LZ4_decompress_safe() on GCC64 */
michael@0: #endif
michael@0:     static const size_t dec64table[] = {0, 0, 0, (size_t)-1, 0, 1, 2, 3};
michael@0: 
michael@0:     const int checkOffset = (endOnInput) && (dictSize < (int)(64 KB));
michael@0: 
michael@0: 
michael@0:     /* Special cases */
michael@0:     if ((partialDecoding) && (oexit> oend-MFLIMIT)) oexit = oend-MFLIMIT;                        /* targetOutputSize too high => decode everything */
michael@0:     if ((endOnInput) && (unlikely(outputSize==0))) return ((inputSize==1) && (*ip==0)) ? 0 : -1;   /* Empty output buffer */
michael@0:     if ((!endOnInput) && (unlikely(outputSize==0))) return (*ip==0?1:-1);
michael@0: 
michael@0: 
michael@0:     /* Main Loop */
michael@0:     while (1)
michael@0:     {
michael@0:         unsigned token;
michael@0:         size_t length;
michael@0: 
michael@0:         /* get runlength */
michael@0:         token = *ip++;
michael@0:         if ((length=(token>>ML_BITS)) == RUN_MASK)
michael@0:         {
michael@0:             unsigned s;
michael@0:             do
michael@0:             {
michael@0:                 s = *ip++;
michael@0:                 length += s;
michael@0:             }
michael@0:             while (likely((endOnInput)?ip<iend-RUN_MASK:1) && (s==255));
michael@0:             //if ((sizeof(void*)==4) && unlikely(length>LZ4_MAX_INPUT_SIZE)) goto _output_error;   /* overflow detection */
michael@0:             if ((sizeof(void*)==4) && unlikely((size_t)(op+length)<(size_t)(op))) goto _output_error;   /* quickfix issue 134 */
michael@0:             if ((endOnInput) && (sizeof(void*)==4) && unlikely((size_t)(ip+length)<(size_t)(ip))) goto _output_error;   /* quickfix issue 134 */
michael@0:         }
michael@0: 
michael@0:         /* copy literals */
michael@0:         cpy = op+length;
michael@0:         if (((endOnInput) && ((cpy>(partialDecoding?oexit:oend-MFLIMIT)) || (ip+length>iend-(2+1+LASTLITERALS))) )
michael@0:             || ((!endOnInput) && (cpy>oend-COPYLENGTH)))
michael@0:         {
michael@0:             if (partialDecoding)
michael@0:             {
michael@0:                 if (cpy > oend) goto _output_error;                           /* Error : write attempt beyond end of output buffer */
michael@0:                 if ((endOnInput) && (ip+length > iend)) goto _output_error;   /* Error : read attempt beyond end of input buffer */
michael@0:             }
michael@0:             else
michael@0:             {
michael@0:                 if ((!endOnInput) && (cpy != oend)) goto _output_error;       /* Error : block decoding must stop exactly there */
michael@0:                 if ((endOnInput) && ((ip+length != iend) || (cpy > oend))) goto _output_error;   /* Error : input must be consumed */
michael@0:             }
michael@0:             memcpy(op, ip, length);
michael@0:             ip += length;
michael@0:             op += length;
michael@0:             break;                                       /* Necessarily EOF, due to parsing restrictions */
michael@0:         }
michael@0:         LZ4_WILDCOPY(op, ip, cpy); ip -= (op-cpy); op = cpy;
michael@0: 
michael@0:         /* get offset */
michael@0:         LZ4_READ_LITTLEENDIAN_16(ref,cpy,ip); ip+=2;
michael@0:         if ((checkOffset) && (unlikely(ref < lowLimit))) goto _output_error;   /* Error : offset outside destination buffer */
michael@0: 
michael@0:         /* get matchlength */
michael@0:         if ((length=(token&ML_MASK)) == ML_MASK)
michael@0:         {
michael@0:             unsigned s;
michael@0:             do
michael@0:             {
michael@0:                 if ((endOnInput) && (ip > iend-LASTLITERALS)) goto _output_error;
michael@0:                 s = *ip++;
michael@0:                 length += s;
michael@0:             } while (s==255);
michael@0:             //if ((sizeof(void*)==4) && unlikely(length>LZ4_MAX_INPUT_SIZE)) goto _output_error;   /* overflow detection */
michael@0:             if ((sizeof(void*)==4) && unlikely((size_t)(op+length)<(size_t)op)) goto _output_error;   /* quickfix issue 134 */
michael@0:         }
michael@0: 
michael@0:         /* check external dictionary */
michael@0:         if ((dict==usingExtDict) && (ref < (BYTE* const)dest))
michael@0:         {
michael@0:             if (unlikely(op+length+MINMATCH > oend-LASTLITERALS)) goto _output_error;
michael@0: 
michael@0:             if (length+MINMATCH <= (size_t)(dest-(char*)ref))
michael@0:             {
michael@0:                 ref = dictEnd - (dest-(char*)ref);
michael@0:                 memcpy(op, ref, length+MINMATCH);
michael@0:                 op += length+MINMATCH;
michael@0:             }
michael@0:             else
michael@0:             {
michael@0:                 size_t copySize = (size_t)(dest-(char*)ref);
michael@0:                 memcpy(op, dictEnd - copySize, copySize);
michael@0:                 op += copySize;
michael@0:                 copySize = length+MINMATCH - copySize;
michael@0:                 if (copySize > (size_t)((char*)op-dest))   /* overlap */
michael@0:                 {
michael@0:                     BYTE* const cpy = op + copySize;
michael@0:                     const BYTE* ref = (BYTE*)dest;
michael@0:                     while (op < cpy) *op++ = *ref++;
michael@0:                 }
michael@0:                 else
michael@0:                 {
michael@0:                     memcpy(op, dest, copySize);
michael@0:                     op += copySize;
michael@0:                 }
michael@0:             }
michael@0:             continue;
michael@0:         }
michael@0: 
michael@0:         /* copy repeated sequence */
michael@0:         if (unlikely((op-ref)<(int)STEPSIZE))
michael@0:         {
michael@0:             const size_t dec64 = dec64table[(sizeof(void*)==4) ? 0 : op-ref];
michael@0:             op[0] = ref[0];
michael@0:             op[1] = ref[1];
michael@0:             op[2] = ref[2];
michael@0:             op[3] = ref[3];
michael@0: #ifdef OLD
michael@0:             op += 4, ref += 4; ref -= dec32table[op-ref];
michael@0:             A32(op) = A32(ref);
michael@0:             op += STEPSIZE-4; ref -= dec64;
michael@0: #else
michael@0:             ref += dec32table[op-ref];
michael@0:             A32(op+4) = A32(ref);
michael@0:             op += STEPSIZE; ref -= dec64;
michael@0: #endif
michael@0:         } else { LZ4_COPYSTEP(op,ref); }
michael@0:         cpy = op + length - (STEPSIZE-4);
michael@0: 
michael@0:         if (unlikely(cpy>oend-COPYLENGTH-(STEPSIZE-4)))
michael@0:         {
michael@0:             if (cpy > oend-LASTLITERALS) goto _output_error;    /* Error : last 5 bytes must be literals */
michael@0:             if (op<oend-COPYLENGTH) LZ4_WILDCOPY(op, ref, (oend-COPYLENGTH));
michael@0:             while(op<cpy) *op++=*ref++;
michael@0:             op=cpy;
michael@0:             continue;
michael@0:         }
michael@0:         LZ4_WILDCOPY(op, ref, cpy);
michael@0:         op=cpy;   /* correction */
michael@0:     }
michael@0: 
michael@0:     /* end of decoding */
michael@0:     if (endOnInput)
michael@0:        return (int) (((char*)op)-dest);     /* Nb of output bytes decoded */
michael@0:     else
michael@0:        return (int) (((char*)ip)-source);   /* Nb of input bytes read */
michael@0: 
michael@0:     /* Overflow error detected */
michael@0: _output_error:
michael@0:     return (int) (-(((char*)ip)-source))-1;
michael@0: }
michael@0: 
michael@0: 
michael@0: int LZ4_decompress_safe(const char* source, char* dest, int compressedSize, int maxOutputSize)
michael@0: {
michael@0:     return LZ4_decompress_generic(source, dest, compressedSize, maxOutputSize, endOnInputSize, full, 0, noDict, NULL, 0);
michael@0: }
michael@0: 
michael@0: int LZ4_decompress_safe_partial(const char* source, char* dest, int compressedSize, int targetOutputSize, int maxOutputSize)
michael@0: {
michael@0:     return LZ4_decompress_generic(source, dest, compressedSize, maxOutputSize, endOnInputSize, partial, targetOutputSize, noDict, NULL, 0);
michael@0: }
michael@0: 
michael@0: int LZ4_decompress_fast(const char* source, char* dest, int originalSize)
michael@0: {
michael@0:     return LZ4_decompress_generic(source, dest, 0, originalSize, endOnOutputSize, full, 0, withPrefix64k, NULL, 0);
michael@0: }
michael@0: 
michael@0: /* streaming decompression functions */
michael@0: 
michael@0: //#define LZ4_STREAMDECODESIZE_U32 4
michael@0: //#define LZ4_STREAMDECODESIZE     (LZ4_STREAMDECODESIZE_U32 * sizeof(unsigned int))
michael@0: //typedef struct { unsigned int table[LZ4_STREAMDECODESIZE_U32]; } LZ4_streamDecode_t;
michael@0: typedef struct
michael@0: {
michael@0:     const char* dictionary;
michael@0:     int dictSize;
michael@0: } LZ4_streamDecode_t_internal;
michael@0: 
michael@0: /*
michael@0:  * If you prefer dynamic allocation methods,
michael@0:  * LZ4_createStreamDecode()
michael@0:  * provides a pointer (void*) towards an initialized LZ4_streamDecode_t structure.
michael@0:  */
michael@0: void* LZ4_createStreamDecode()
michael@0: {
michael@0:     void* lz4s = ALLOCATOR(sizeof(U32), LZ4_STREAMDECODESIZE_U32);
michael@0:     MEM_INIT(lz4s, 0, LZ4_STREAMDECODESIZE);
michael@0:     return lz4s;
michael@0: }
michael@0: 
michael@0: /*
michael@0:  * LZ4_setDictDecode
michael@0:  * Use this function to instruct where to find the dictionary
michael@0:  * This function is not necessary if previous data is still available where it was decoded.
michael@0:  * Loading a size of 0 is allowed (same effect as no dictionary).
michael@0:  * Return : 1 if OK, 0 if error
michael@0:  */
michael@0: int LZ4_setDictDecode (void* LZ4_streamDecode, const char* dictionary, int dictSize)
michael@0: {
michael@0:     LZ4_streamDecode_t_internal* lz4sd = (LZ4_streamDecode_t_internal*) LZ4_streamDecode;
michael@0:     lz4sd->dictionary = dictionary;
michael@0:     lz4sd->dictSize = dictSize;
michael@0:     return 1;
michael@0: }
michael@0: 
michael@0: /*
michael@0: *_continue() :
michael@0:     These decoding functions allow decompression of multiple blocks in "streaming" mode.
michael@0:     Previously decoded blocks must still be available at the memory position where they were decoded.
michael@0:     If it's not possible, save the relevant part of decoded data into a safe buffer,
michael@0:     and indicate where it stands using LZ4_setDictDecode()
michael@0: */
michael@0: int LZ4_decompress_safe_continue (void* LZ4_streamDecode, const char* source, char* dest, int compressedSize, int maxOutputSize)
michael@0: {
michael@0:     LZ4_streamDecode_t_internal* lz4sd = (LZ4_streamDecode_t_internal*) LZ4_streamDecode;
michael@0:     int result;
michael@0: 
michael@0:     result = LZ4_decompress_generic(source, dest, compressedSize, maxOutputSize, endOnInputSize, full, 0, usingExtDict, lz4sd->dictionary, lz4sd->dictSize);
michael@0:     if (result <= 0) return result;
michael@0:     if (lz4sd->dictionary + lz4sd->dictSize == dest)
michael@0:     {
michael@0:         lz4sd->dictSize += result;
michael@0:     }
michael@0:     else
michael@0:     {
michael@0:         lz4sd->dictionary = dest;
michael@0:         lz4sd->dictSize = result;
michael@0:     }
michael@0: 
michael@0:     return result;
michael@0: }
michael@0: 
michael@0: int LZ4_decompress_fast_continue (void* LZ4_streamDecode, const char* source, char* dest, int originalSize)
michael@0: {
michael@0:     LZ4_streamDecode_t_internal* lz4sd = (LZ4_streamDecode_t_internal*) LZ4_streamDecode;
michael@0:     int result;
michael@0: 
michael@0:     result = LZ4_decompress_generic(source, dest, 0, originalSize, endOnOutputSize, full, 0, usingExtDict, lz4sd->dictionary, lz4sd->dictSize);
michael@0:     if (result <= 0) return result;
michael@0:     if (lz4sd->dictionary + lz4sd->dictSize == dest)
michael@0:     {
michael@0:         lz4sd->dictSize += result;
michael@0:     }
michael@0:     else
michael@0:     {
michael@0:         lz4sd->dictionary = dest;
michael@0:         lz4sd->dictSize = result;
michael@0:     }
michael@0: 
michael@0:     return result;
michael@0: }
michael@0: 
michael@0: 
michael@0: /*
michael@0: Advanced decoding functions :
michael@0: *_usingDict() :
michael@0:     These decoding functions work the same as "_continue" ones,
michael@0:     the dictionary must be explicitly provided within parameters
michael@0: */
michael@0: 
michael@0: int LZ4_decompress_safe_usingDict(const char* source, char* dest, int compressedSize, int maxOutputSize, const char* dictStart, int dictSize)
michael@0: {
michael@0:     return LZ4_decompress_generic(source, dest, compressedSize, maxOutputSize, endOnInputSize, full, 0, usingExtDict, dictStart, dictSize);
michael@0: }
michael@0: 
michael@0: int LZ4_decompress_fast_usingDict(const char* source, char* dest, int originalSize, const char* dictStart, int dictSize)
michael@0: {
michael@0:     return LZ4_decompress_generic(source, dest, 0, originalSize, endOnOutputSize, full, 0, usingExtDict, dictStart, dictSize);
michael@0: }