LZ4 compressed package file support. Closes #6.

Docs/GettingStarted.dox

@@ -340,6 +340,8 @@ Third-party libraries are used for the following functionality:
 - GLEW: OpenGL extensions handling
 - kNet: UDP networking
 - libcpuid: CPU properties detection
+- Lua, LuaJIT, tolua+: Lua scripting implementation and bindings
+- LZ4: data compression for package files
 - MojoShader: HLSL shader reflection after compiling
 - Open Asset Import Library: reading various 3D file formats
 - pugixml: parsing XML files

Docs/Urho3D.dox

@@ -98,6 +98,7 @@ Urho3D uses the following third-party libraries:
 - libcpuid 0.2.0 (http://libcpuid.sourceforge.net/)
 - Lua 5.1 (http://www.lua.org)
 - LuaJIT 2.0.2 (http://www.luajit.org)
+- LZ4 (http://code.google.com/p/lz4/)
 - MojoShader (http://icculus.org/mojoshader/)
 - Open Asset Import Library (http://assimp.sourceforge.net/)
 - pugixml 1.0 (http://pugixml.org/)

License.txt

@@ -143,7 +143,7 @@ THE SOFTWARE.
 The FreeType Project disclaimer
 -------------------------------
 
-Portions of this software are copyright © 2010 The FreeType Project 
+Portions of this software are copyright © 2010 The FreeType Project
 (www.freetype.org).  All rights reserved.
 
 
@@ -321,6 +321,43 @@ LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
 OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 THE SOFTWARE.
 
+
+LZ4 license
+-----------
+
+LZ4 - Fast LZ compression algorithm
+Header File
+Copyright (C) 2011-2013, 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 :
+- LZ4 homepage : http://fastcompression.blogspot.com/p/lz4.html
+- LZ4 source repository : http://code.google.com/p/lz4/
+
+
 MojoShader license
 ------------------
 

Readme.txt

@@ -68,6 +68,7 @@ Urho3D uses the following third-party libraries:
 - libcpuid 0.2.0 (http://libcpuid.sourceforge.net/)
 - Lua 5.1 (http://www.lua.org)
 - LuaJIT 2.0.2 (http://www.luajit.org)
+- LZ4 (http://code.google.com/p/lz4/)
 - MojoShader (http://icculus.org/mojoshader/)
 - Open Asset Import Library (http://assimp.sourceforge.net/)
 - pugixml 1.0 (http://pugixml.org/)

Source/CMakeLists.txt

@@ -89,13 +89,14 @@ add_subdirectory (ThirdParty/Bullet)
 add_subdirectory (ThirdParty/Civetweb)
 add_subdirectory (ThirdParty/Detour)
 add_subdirectory (ThirdParty/FreeType)
+add_subdirectory (ThirdParty/JO)
 add_subdirectory (ThirdParty/kNet)
+add_subdirectory (ThirdParty/LZ4)
 add_subdirectory (ThirdParty/PugiXml)
 add_subdirectory (ThirdParty/Recast)
 add_subdirectory (ThirdParty/SDL)
 add_subdirectory (ThirdParty/StanHull)
 add_subdirectory (ThirdParty/STB)
-add_subdirectory (ThirdParty/JO)
 
 # Alternative Lua bindings target
 if (ENABLE_LUA)

Source/Engine/IO/CMakeLists.txt

@@ -34,7 +34,7 @@ endif()
 set (SOURCE_FILES ${CPP_FILES} ${C_FILES} ${H_FILES})
 
 # Define dependency libs
-set (LIBS ../Container ../Core ../Math)
+set (LIBS ../Container ../Core ../Math ../../ThirdParty/LZ4)
 
 # Setup target
 enable_pch ()

Source/Engine/IO/File.cpp

@@ -24,10 +24,12 @@
 #include "File.h"
 #include "FileSystem.h"
 #include "Log.h"
+#include "MemoryBuffer.h"
 #include "PackageFile.h"
 #include "Profiler.h"
 
 #include <cstdio>
+#include <lz4.h>
 
 #include "DebugNew.h"
 
@@ -50,9 +52,7 @@ static const char* openMode[] =
 };
 #endif
 
-#ifdef ANDROID
 static const unsigned READ_BUFFER_SIZE = 1024;
-#endif
 
 File::File(Context* context) :
     Object(context),
@@ -60,9 +60,9 @@ File::File(Context* context) :
     handle_(0),
     #ifdef ANDROID
     assetHandle_(0),
+    #endif
     readBufferOffset_(0),
     readBufferSize_(0),
-    #endif
     offset_(0),
     checksum_(0)
 {
@@ -74,9 +74,9 @@ File::File(Context* context, const String& fileName, FileMode mode) :
     handle_(0),
     #ifdef ANDROID
     assetHandle_(0),
+    #endif
     readBufferOffset_(0),
     readBufferSize_(0),
-    #endif
     offset_(0),
     checksum_(0)
 {
@@ -89,9 +89,9 @@ File::File(Context* context, PackageFile* package, const String& fileName) :
     handle_(0),
     #ifdef ANDROID
     assetHandle_(0),
+    #endif
     readBufferOffset_(0),
     readBufferSize_(0),
-    #endif
     offset_(0),
     checksum_(0)
 {
@@ -198,6 +198,13 @@ bool File::Open(PackageFile* package, const String& fileName)
     position_ = 0;
     size_ = entry->size_;
     
+    if (package->IsCompressed())
+    {
+        readBuffer_ = new unsigned char[package->GetBlockSize()];
+        readBufferOffset_ = 0;
+        readBufferSize_ = 0;
+    }
+    
     fseek((FILE*)handle_, offset_, SEEK_SET);
     return true;
 }
@@ -240,6 +247,44 @@ unsigned File::Read(void* dest, unsigned size)
         
         return size;
     }
+    #else
+    // Package compressed mode
+    if (readBuffer_)
+    {
+        unsigned sizeLeft = size;
+        unsigned char* destPtr = (unsigned char*)dest;
+        
+        while (sizeLeft)
+        {
+            if (readBufferOffset_ >= readBufferSize_)
+            {
+                unsigned char blockHeaderBytes[4];
+                fread(blockHeaderBytes, sizeof blockHeaderBytes, 1, (FILE*)handle_);
+                
+                MemoryBuffer blockHeader(&blockHeaderBytes[0], sizeof blockHeaderBytes);
+                unsigned unpackedSize = blockHeader.ReadUShort();
+                unsigned packedSize = blockHeader.ReadUShort();
+                
+                /// \todo Should reuse a block to hold the packed data
+                /// \todo Handle errors
+                SharedArrayPtr<unsigned char> packedBytes(new unsigned char[packedSize]);
+                fread(packedBytes.Get(), packedSize, 1, (FILE*)handle_);
+                LZ4_decompress_fast((const char*)packedBytes.Get(), (char *)readBuffer_.Get(), unpackedSize);
+                
+                readBufferSize_ = unpackedSize;
+                readBufferOffset_ = 0;
+            }
+            
+            unsigned copySize = Min((int)(readBufferSize_ - readBufferOffset_), (int)sizeLeft);
+            memcpy(destPtr, readBuffer_.Get() + readBufferOffset_, copySize);
+            destPtr += copySize;
+            sizeLeft -= copySize;
+            readBufferOffset_ += copySize;
+            position_ += copySize;
+        }
+        
+        return size;
+    }
     #endif
     
     if (!handle_)
@@ -276,6 +321,30 @@ unsigned File::Seek(unsigned position)
         readBufferSize_ = 0;
         return position_;
     }
+    #else
+    // Package compressed mode
+    if (readBuffer_)
+    {
+        // Start over from the beginning
+        if (position == 0)
+        {
+            position_ = 0;
+            readBufferOffset_ = 0;
+            readBufferSize_ = 0;
+            fseek((FILE*)handle_, offset_, SEEK_SET);
+        }
+        // Skip bytes
+        else if (position >= position_)
+        {
+            unsigned char skipBuffer[READ_BUFFER_SIZE];
+            while (position > position_)
+                Read(skipBuffer, Min((int)position - position_, (int)READ_BUFFER_SIZE));
+        }
+        else
+            LOGERROR("Seeking backward in a compressed file is not supported");
+        
+        return position_;
+    }
     #endif
     
     if (!handle_)
@@ -353,10 +422,11 @@ void File::Close()
     {
         SDL_RWclose(assetHandle_);
         assetHandle_ = 0;
-        readBuffer_.Reset();
     }
     #endif
     
+    readBuffer_.Reset();
+    
     if (handle_)
     {
         fclose((FILE*)handle_);

Source/Engine/IO/File.h

@@ -22,13 +22,12 @@
 
 #pragma once
 
+#include "ArrayPtr.h"
 #include "Deserializer.h"
 #include "Serializer.h"
 #include "Object.h"
 
 #ifdef ANDROID
-#include "ArrayPtr.h"
-
 #include <SDL_rwops.h>
 #endif
 
@@ -101,13 +100,13 @@ private:
     #ifdef ANDROID
     /// SDL RWops context for Android asset loading.
     SDL_RWops* assetHandle_;
-    /// Read buffer for Android asset loading.
+    #endif
+    /// Read buffer for Android asset or compressed file loading.
     SharedArrayPtr<unsigned char> readBuffer_;
     /// Read buffer position.
     unsigned readBufferOffset_;
     /// Bytes in the current read buffer.
     unsigned readBufferSize_;
-    #endif
     /// Start position within a package file, 0 for regular files.
     unsigned offset_;
     /// Content checksum.

Source/Engine/IO/PackageFile.cpp

@@ -31,14 +31,18 @@ namespace Urho3D
 PackageFile::PackageFile(Context* context) :
     Object(context),
     totalSize_(0),
-    checksum_(0)
+    checksum_(0),
+    blockSize_(0),
+    compressed_(false)
 {
 }
 
 PackageFile::PackageFile(Context* context, const String& fileName) :
     Object(context),
     totalSize_(0),
-    checksum_(0)
+    checksum_(0),
+    blockSize_(0),
+    compressed_(false)
 {
     Open(fileName);
 }
@@ -54,7 +58,8 @@ bool PackageFile::Open(const String& fileName)
         return false;
     
     // Check ID, then read the directory
-    if (file->ReadFileID() != "UPAK")
+    String id = file->ReadFileID();
+    if (id != "UPAK" && id != "ULZ4")
     {
         LOGERROR(fileName + " is not a valid package file");
         return false;
@@ -63,9 +68,12 @@ bool PackageFile::Open(const String& fileName)
     fileName_ = fileName;
     nameHash_ = fileName_;
     totalSize_ = file->GetSize();
+    compressed_ = id == "ULZ4";
     
     unsigned numFiles = file->ReadUInt();
     checksum_ = file->ReadUInt();
+    if (compressed_)
+        blockSize_ = file->ReadUInt();
     
     for (unsigned i = 0; i < numFiles; ++i)
     {
@@ -74,7 +82,7 @@ bool PackageFile::Open(const String& fileName)
         newEntry.offset_ = file->ReadUInt();
         newEntry.size_ = file->ReadUInt();
         newEntry.checksum_ = file->ReadUInt();
-        if (newEntry.offset_ + newEntry.size_ > totalSize_)
+        if (!compressed_ && newEntry.offset_ + newEntry.size_ > totalSize_)
             LOGERROR("File entry " + entryName + " outside package file");
         else
             entries_[entryName.ToLower()] = newEntry;

Source/Engine/IO/PackageFile.h

@@ -69,6 +69,10 @@ public:
     unsigned GetTotalSize() const { return totalSize_; }
     /// Return checksum of the package file contents.
     unsigned GetChecksum() const { return checksum_; }
+    /// Return compressed mode block size.
+    unsigned GetBlockSize() const { return blockSize_; }
+    /// Return whether the files are compressed.
+    bool IsCompressed() const { return compressed_; }
     
 private:
     /// File entries.
@@ -81,6 +85,10 @@ private:
     unsigned totalSize_;
     /// Package file checksum.
     unsigned checksum_;
+    /// Compressed mode block size.
+    unsigned blockSize_;
+    /// Compressed flag.
+    bool compressed_;
 };
 
 }

Source/ThirdParty/LZ4/CMakeLists.txt

@@ -0,0 +1,10 @@
+# Define target name
+set (TARGET_NAME LZ4)
+
+# Define source files
+file (GLOB C_FILES *.c)
+file (GLOB H_FILES *.h)
+set (SOURCE_FILES ${C_FILES} ${H_FILES})
+
+# Setup target
+setup_library ()

Source/ThirdParty/LZ4/NEWS

@@ -0,0 +1,10 @@
+r106 :
+Makefile : make dist modify text files in the package to respect Unix EoL convention
+lz4cli : corrected small display bug in HC mode
+
+r105 :
+Makefile : New install script and man page, contributed by Prasad Pandit
+lz4cli.c : Minor modifications, for easier extensibility
+COPYING : added license file
+LZ4_Streaming_Format.odt : modified file name to remove white space characters
+exe : .exe suffix now properly added only for Windows target

Source/ThirdParty/LZ4/lz4.c

@@ -0,0 +1,815 @@
+/*
+   LZ4 - Fast LZ compression algorithm
+   Copyright (C) 2011-2013, 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 :
+   - LZ4 source repository : http://code.google.com/p/lz4/
+   - LZ4 public forum : https://groups.google.com/forum/#!forum/lz4c
+*/
+
+//**************************************
+// 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
+// Default value is 14, for 16KB, which nicely fits into Intel x86 L1 cache
+#define MEMORY_USAGE 14
+
+// HEAPMODE :
+// Select how default compression functions will allocate memory for their hash table,
+// in memory stack (0:default, fastest), or in memory heap (1:requires memory allocation (malloc)).
+#define HEAPMODE 0
+
+
+//**************************************
+// CPU Feature Detection
+//**************************************
+// 32 or 64 bits ?
+#if (defined(__x86_64__) || defined(_M_X64) || defined(_WIN64) \
+  || defined(__powerpc64__) || defined(__ppc64__) || defined(__PPC64__) \
+  || defined(__64BIT__) || defined(_LP64) || defined(__LP64__) \
+  || defined(__ia64) || defined(__itanium__) || defined(_M_IA64) )   // Detects 64 bits mode
+#  define LZ4_ARCH64 1
+#else
+#  define LZ4_ARCH64 0
+#endif
+
+// Little Endian or Big Endian ?
+// Overwrite the #define below if you know your architecture endianess
+#if defined (__GLIBC__)
+#  include <endian.h>
+#  if (__BYTE_ORDER == __BIG_ENDIAN)
+#     define LZ4_BIG_ENDIAN 1
+#  endif
+#elif (defined(__BIG_ENDIAN__) || defined(__BIG_ENDIAN) || defined(_BIG_ENDIAN)) && !(defined(__LITTLE_ENDIAN__) || defined(__LITTLE_ENDIAN) || defined(_LITTLE_ENDIAN))
+#  define LZ4_BIG_ENDIAN 1
+#elif defined(__sparc) || defined(__sparc__) \
+   || defined(__powerpc__) || defined(__ppc__) || defined(__PPC__) \
+   || defined(__hpux)  || defined(__hppa) \
+   || defined(_MIPSEB) || defined(__s390__)
+#  define LZ4_BIG_ENDIAN 1
+#else
+// Little Endian assumed. PDP Endian and other very rare endian format are unsupported.
+#endif
+
+// Unaligned memory access is automatically enabled for "common" CPU, such as x86.
+// For others CPU, such as ARM, the compiler may be more cautious, inserting unnecessary extra code to ensure aligned access property
+// If you know your target CPU supports unaligned memory access, you want to force this option manually to improve performance
+#if defined(__ARM_FEATURE_UNALIGNED)
+#  define LZ4_FORCE_UNALIGNED_ACCESS 1
+#endif
+
+// Define this parameter if your target system or compiler does not support hardware bit count
+#if defined(_MSC_VER) && defined(_WIN32_WCE)            // Visual Studio for Windows CE does not support Hardware bit count
+#  define LZ4_FORCE_SW_BITCOUNT
+#endif
+
+// BIG_ENDIAN_NATIVE_BUT_INCOMPATIBLE :
+// This option may provide a small boost to performance for some big endian cpu, although probably modest.
+// You may set this option to 1 if data will remain within closed environment.
+// This option is useless on Little_Endian CPU (such as x86)
+//#define BIG_ENDIAN_NATIVE_BUT_INCOMPATIBLE 1
+
+
+//**************************************
+// Compiler Options
+//**************************************
+#if defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L)   // C99
+/* "restrict" is a known keyword */
+#else
+#  define restrict // Disable restrict
+#endif
+
+#ifdef _MSC_VER    // Visual Studio
+#  define FORCE_INLINE static __forceinline
+#  include <intrin.h>                    // For Visual 2005
+#  if LZ4_ARCH64   // 64-bits
+#    pragma intrinsic(_BitScanForward64) // For Visual 2005
+#    pragma intrinsic(_BitScanReverse64) // For Visual 2005
+#  else            // 32-bits
+#    pragma intrinsic(_BitScanForward)   // For Visual 2005
+#    pragma intrinsic(_BitScanReverse)   // For Visual 2005
+#  endif
+#  pragma warning(disable : 4127)        // disable: C4127: conditional expression is constant
+#else 
+#  ifdef __GNUC__
+#    define FORCE_INLINE static inline __attribute__((always_inline))
+#  else
+#    define FORCE_INLINE static inline
+#  endif
+#endif
+
+#ifdef _MSC_VER
+#  define lz4_bswap16(x) _byteswap_ushort(x)
+#else
+#  define lz4_bswap16(x) ((unsigned short int) ((((x) >> 8) & 0xffu) | (((x) & 0xffu) << 8)))
+#endif
+
+#define GCC_VERSION (__GNUC__ * 100 + __GNUC_MINOR__)
+
+#if (GCC_VERSION >= 302) || (__INTEL_COMPILER >= 800) || defined(__clang__)
+#  define expect(expr,value)    (__builtin_expect ((expr),(value)) )
+#else
+#  define expect(expr,value)    (expr)
+#endif
+
+#define likely(expr)     expect((expr) != 0, 1)
+#define unlikely(expr)   expect((expr) != 0, 0)
+
+
+//**************************************
+// Memory routines
+//**************************************
+#include <stdlib.h>   // malloc, calloc, free
+#define ALLOCATOR(n,s) calloc(n,s)
+#define FREEMEM        free
+#include <string.h>   // memset, memcpy
+#define MEM_INIT       memset
+
+
+//**************************************
+// Includes
+//**************************************
+#include "lz4.h"
+
+
+//**************************************
+// Basic Types
+//**************************************
+#if defined (__STDC_VERSION__) && __STDC_VERSION__ >= 199901L   // C99
+# include <stdint.h>
+  typedef  uint8_t BYTE;
+  typedef uint16_t U16;
+  typedef uint32_t U32;
+  typedef  int32_t S32;
+  typedef uint64_t U64;
+#else
+  typedef unsigned char       BYTE;
+  typedef unsigned short      U16;
+  typedef unsigned int        U32;
+  typedef   signed int        S32;
+  typedef unsigned long long  U64;
+#endif
+
+#if defined(__GNUC__)  && !defined(LZ4_FORCE_UNALIGNED_ACCESS)
+#  define _PACKED __attribute__ ((packed))
+#else
+#  define _PACKED
+#endif
+
+#if !defined(LZ4_FORCE_UNALIGNED_ACCESS) && !defined(__GNUC__)
+#  if defined(__IBMC__) || defined(__SUNPRO_C) || defined(__SUNPRO_CC)
+#    pragma pack(1)
+#  else
+#    pragma pack(push, 1)
+#  endif
+#endif
+
+typedef struct { U16 v; }  _PACKED U16_S;
+typedef struct { U32 v; }  _PACKED U32_S;
+typedef struct { U64 v; }  _PACKED U64_S;
+typedef struct {size_t v;} _PACKED size_t_S;
+
+#if !defined(LZ4_FORCE_UNALIGNED_ACCESS) && !defined(__GNUC__)
+#  if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
+#    pragma pack(0)
+#  else
+#    pragma pack(pop)
+#  endif
+#endif
+
+#define A16(x)   (((U16_S *)(x))->v)
+#define A32(x)   (((U32_S *)(x))->v)
+#define A64(x)   (((U64_S *)(x))->v)
+#define AARCH(x) (((size_t_S *)(x))->v)
+
+
+//**************************************
+// Constants
+//**************************************
+#define LZ4_HASHLOG   (MEMORY_USAGE-2)
+#define HASHTABLESIZE (1 << MEMORY_USAGE)
+#define HASHNBCELLS4  (1 << LZ4_HASHLOG)
+
+#define MINMATCH 4
+
+#define COPYLENGTH 8
+#define LASTLITERALS 5
+#define MFLIMIT (COPYLENGTH+MINMATCH)
+const int LZ4_minLength = (MFLIMIT+1);
+
+#define LZ4_64KLIMIT ((1<<16) + (MFLIMIT-1))
+#define SKIPSTRENGTH 6     // Increasing this value will make the compression run slower on incompressible data
+
+#define MAXD_LOG 16
+#define MAX_DISTANCE ((1 << MAXD_LOG) - 1)
+
+#define ML_BITS  4
+#define ML_MASK  ((1U<<ML_BITS)-1)
+#define RUN_BITS (8-ML_BITS)
+#define RUN_MASK ((1U<<RUN_BITS)-1)
+
+#define KB *(1U<<10)
+#define MB *(1U<<20)
+#define GB *(1U<<30)
+
+
+//**************************************
+// Structures and local types
+//**************************************
+
+typedef struct {
+    U32 hashTable[HASHNBCELLS4];
+    const BYTE* bufferStart;
+    const BYTE* base;
+    const BYTE* nextBlock;
+} LZ4_Data_Structure;
+
+typedef enum { notLimited = 0, limited = 1 } limitedOutput_directive;
+typedef enum { byPtr, byU32, byU16 } tableType_t;
+
+typedef enum { noPrefix = 0, withPrefix = 1 } prefix64k_directive;
+
+typedef enum { endOnOutputSize = 0, endOnInputSize = 1 } endCondition_directive;
+typedef enum { full = 0, partial = 1 } earlyEnd_directive;
+
+
+//**************************************
+// Architecture-specific macros
+//**************************************
+#define STEPSIZE                  sizeof(size_t)
+#define LZ4_COPYSTEP(d,s)         { AARCH(d) = AARCH(s); d+=STEPSIZE; s+=STEPSIZE; }
+#define LZ4_COPY8(d,s)            { LZ4_COPYSTEP(d,s); if (STEPSIZE<8) LZ4_COPYSTEP(d,s); }
+#define LZ4_SECURECOPY(d,s,e)     { if ((STEPSIZE==4)||(d<e)) LZ4_WILDCOPY(d,s,e); }
+
+#if LZ4_ARCH64   // 64-bit
+#  define HTYPE                   U32
+#  define INITBASE(base)          const BYTE* const base = ip
+#else            // 32-bit
+#  define HTYPE                   const BYTE*
+#  define INITBASE(base)          const int base = 0
+#endif
+
+#if (defined(LZ4_BIG_ENDIAN) && !defined(BIG_ENDIAN_NATIVE_BUT_INCOMPATIBLE))
+#  define LZ4_READ_LITTLEENDIAN_16(d,s,p) { U16 v = A16(p); v = lz4_bswap16(v); d = (s) - v; }
+#  define LZ4_WRITE_LITTLEENDIAN_16(p,i)  { U16 v = (U16)(i); v = lz4_bswap16(v); A16(p) = v; p+=2; }
+#else      // Little Endian
+#  define LZ4_READ_LITTLEENDIAN_16(d,s,p) { d = (s) - A16(p); }
+#  define LZ4_WRITE_LITTLEENDIAN_16(p,v)  { A16(p) = v; p+=2; }
+#endif
+
+
+//**************************************
+// Macros
+//**************************************
+#define LZ4_WILDCOPY(d,s,e)     { do { LZ4_COPY8(d,s) } while (d<e); }           // at the end, d>=e;
+
+
+//****************************
+// Private functions
+//****************************
+#if LZ4_ARCH64
+
+FORCE_INLINE int LZ4_NbCommonBytes (register U64 val)
+{
+# if defined(LZ4_BIG_ENDIAN)
+#   if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
+    unsigned long r = 0;
+    _BitScanReverse64( &r, val );
+    return (int)(r>>3);
+#   elif defined(__GNUC__) && (GCC_VERSION >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
+    return (__builtin_clzll(val) >> 3);
+#   else
+    int r;
+    if (!(val>>32)) { r=4; } else { r=0; val>>=32; }
+    if (!(val>>16)) { r+=2; val>>=8; } else { val>>=24; }
+    r += (!val);
+    return r;
+#   endif
+# else
+#   if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
+    unsigned long r = 0;
+    _BitScanForward64( &r, val );
+    return (int)(r>>3);
+#   elif defined(__GNUC__) && (GCC_VERSION >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
+    return (__builtin_ctzll(val) >> 3);
+#   else
+    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 };
+    return DeBruijnBytePos[((U64)((val & -(long long)val) * 0x0218A392CDABBD3FULL)) >> 58];
+#   endif
+# endif
+}
+
+#else
+
+FORCE_INLINE int LZ4_NbCommonBytes (register U32 val)
+{
+# if defined(LZ4_BIG_ENDIAN)
+#   if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
+    unsigned long r = 0;
+    _BitScanReverse( &r, val );
+    return (int)(r>>3);
+#   elif defined(__GNUC__) && (GCC_VERSION >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
+    return (__builtin_clz(val) >> 3);
+#   else
+    int r;
+    if (!(val>>16)) { r=2; val>>=8; } else { r=0; val>>=24; }
+    r += (!val);
+    return r;
+#   endif
+# else
+#   if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
+    unsigned long r;
+    _BitScanForward( &r, val );
+    return (int)(r>>3);
+#   elif defined(__GNUC__) && (GCC_VERSION >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
+    return (__builtin_ctz(val) >> 3);
+#   else
+    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 };
+    return DeBruijnBytePos[((U32)((val & -(S32)val) * 0x077CB531U)) >> 27];
+#   endif
+# endif
+}
+
+#endif
+
+
+//****************************
+// Compression functions
+//****************************
+FORCE_INLINE int LZ4_hashSequence(U32 sequence, tableType_t tableType)
+{
+    if (tableType == byU16)
+        return (((sequence) * 2654435761U) >> ((MINMATCH*8)-(LZ4_HASHLOG+1)));
+    else
+        return (((sequence) * 2654435761U) >> ((MINMATCH*8)-LZ4_HASHLOG));
+}
+
+FORCE_INLINE int LZ4_hashPosition(const BYTE* p, tableType_t tableType) { return LZ4_hashSequence(A32(p), tableType); }
+
+FORCE_INLINE void LZ4_putPositionOnHash(const BYTE* p, U32 h, void* tableBase, tableType_t tableType, const BYTE* srcBase)
+{
+    switch (tableType)
+    {
+    case byPtr: { const BYTE** hashTable = (const BYTE**) tableBase; hashTable[h] = p; break; }
+    case byU32: { U32* hashTable = (U32*) tableBase; hashTable[h] = (U32)(p-srcBase); break; }
+    case byU16: { U16* hashTable = (U16*) tableBase; hashTable[h] = (U16)(p-srcBase); break; }
+    }
+}
+
+FORCE_INLINE void LZ4_putPosition(const BYTE* p, void* tableBase, tableType_t tableType, const BYTE* srcBase)
+{
+    U32 h = LZ4_hashPosition(p, tableType);
+    LZ4_putPositionOnHash(p, h, tableBase, tableType, srcBase);
+}
+
+FORCE_INLINE const BYTE* LZ4_getPositionOnHash(U32 h, void* tableBase, tableType_t tableType, const BYTE* srcBase)
+{
+    if (tableType == byPtr) { const BYTE** hashTable = (const BYTE**) tableBase; return hashTable[h]; }
+    if (tableType == byU32) { U32* hashTable = (U32*) tableBase; return hashTable[h] + srcBase; }
+    { U16* hashTable = (U16*) tableBase; return hashTable[h] + srcBase; }   // default, to ensure a return
+}
+
+FORCE_INLINE const BYTE* LZ4_getPosition(const BYTE* p, void* tableBase, tableType_t tableType, const BYTE* srcBase)
+{
+    U32 h = LZ4_hashPosition(p, tableType);
+    return LZ4_getPositionOnHash(h, tableBase, tableType, srcBase);
+}
+
+
+FORCE_INLINE int LZ4_compress_generic(
+                 void* ctx,
+                 const char* source,
+                 char* dest,
+                 int inputSize,
+                 int maxOutputSize,
+
+                 limitedOutput_directive limitedOutput,
+                 tableType_t tableType,
+                 prefix64k_directive prefix)
+{
+    const BYTE* ip = (const BYTE*) source;
+    const BYTE* const base = (prefix==withPrefix) ? ((LZ4_Data_Structure*)ctx)->base : (const BYTE*) source;
+    const BYTE* const lowLimit = ((prefix==withPrefix) ? ((LZ4_Data_Structure*)ctx)->bufferStart : (const BYTE*)source);
+    const BYTE* anchor = (const BYTE*) source;
+    const BYTE* const iend = ip + inputSize;
+    const BYTE* const mflimit = iend - MFLIMIT;
+    const BYTE* const matchlimit = iend - LASTLITERALS;
+
+    BYTE* op = (BYTE*) dest;
+    BYTE* const oend = op + maxOutputSize;
+
+    int length;
+    const int skipStrength = SKIPSTRENGTH;
+    U32 forwardH;
+
+    // Init conditions
+    if ((prefix==withPrefix) && (ip != ((LZ4_Data_Structure*)ctx)->nextBlock)) return 0;   // must continue from end of previous block
+    if (prefix==withPrefix) ((LZ4_Data_Structure*)ctx)->nextBlock=iend;                    // do it now, due to potential early exit
+    if ((tableType == byU16) && (inputSize>=LZ4_64KLIMIT)) return 0;                       // Size too large (not within 64K limit)
+    if (inputSize<LZ4_minLength) goto _last_literals;                                      // Input too small, no compression (all literals)
+
+    // First Byte
+    LZ4_putPosition(ip, ctx, tableType, base);
+    ip++; forwardH = LZ4_hashPosition(ip, tableType);
+
+    // Main Loop
+    for ( ; ; )
+    {
+        int findMatchAttempts = (1U << skipStrength) + 3;
+        const BYTE* forwardIp = ip;
+        const BYTE* ref;
+        BYTE* token;
+
+        // Find a match
+        do {
+            U32 h = forwardH;
+            int step = findMatchAttempts++ >> skipStrength;
+            ip = forwardIp;
+            forwardIp = ip + step;
+
+            if unlikely(forwardIp > mflimit) { goto _last_literals; }
+
+            forwardH = LZ4_hashPosition(forwardIp, tableType);
+            ref = LZ4_getPositionOnHash(h, ctx, tableType, base);
+            LZ4_putPositionOnHash(ip, h, ctx, tableType, base);
+
+        } while ((ref + MAX_DISTANCE < ip) || (A32(ref) != A32(ip)));
+
+        // Catch up
+        while ((ip>anchor) && (ref > lowLimit) && unlikely(ip[-1]==ref[-1])) { ip--; ref--; }
+
+        // Encode Literal length
+        length = (int)(ip - anchor);
+        token = op++;
+        if ((limitedOutput) && unlikely(op + length + (2 + 1 + LASTLITERALS) + (length>>8) > oend)) return 0;   // Check output limit
+        if (length>=(int)RUN_MASK) 
+        { 
+            int len = length-RUN_MASK; 
+            *token=(RUN_MASK<<ML_BITS); 
+            for(; len >= 255 ; len-=255) *op++ = 255; 
+            *op++ = (BYTE)len; 
+        }
+        else *token = (BYTE)(length<<ML_BITS);
+
+        // Copy Literals
+        { BYTE* end=(op)+(length); LZ4_WILDCOPY(op,anchor,end); op=end; }
+
+_next_match:
+        // Encode Offset
+        LZ4_WRITE_LITTLEENDIAN_16(op,(U16)(ip-ref));
+
+        // Start Counting
+        ip+=MINMATCH; ref+=MINMATCH;    // MinMatch already verified
+        anchor = ip;
+        while likely(ip<matchlimit-(STEPSIZE-1))
+        {
+            size_t diff = AARCH(ref) ^ AARCH(ip);
+            if (!diff) { ip+=STEPSIZE; ref+=STEPSIZE; continue; }
+            ip += LZ4_NbCommonBytes(diff);
+            goto _endCount;
+        }
+        if (LZ4_ARCH64) if ((ip<(matchlimit-3)) && (A32(ref) == A32(ip))) { ip+=4; ref+=4; }
+        if ((ip<(matchlimit-1)) && (A16(ref) == A16(ip))) { ip+=2; ref+=2; }
+        if ((ip<matchlimit) && (*ref == *ip)) ip++;
+_endCount:
+
+        // Encode MatchLength
+        length = (int)(ip - anchor);
+        if ((limitedOutput) && unlikely(op + (1 + LASTLITERALS) + (length>>8) > oend)) return 0;    // Check output limit
+        if (length>=(int)ML_MASK) 
+        { 
+            *token += ML_MASK; 
+            length -= ML_MASK; 
+            for (; length > 509 ; length-=510) { *op++ = 255; *op++ = 255; } 
+            if (length >= 255) { length-=255; *op++ = 255; } 
+            *op++ = (BYTE)length; 
+        }
+        else *token += (BYTE)(length);
+
+        // Test end of chunk
+        if (ip > mflimit) { anchor = ip;  break; }
+
+        // Fill table
+        LZ4_putPosition(ip-2, ctx, tableType, base);
+
+        // Test next position
+        ref = LZ4_getPosition(ip, ctx, tableType, base); 
+        LZ4_putPosition(ip, ctx, tableType, base);
+        if ((ref + MAX_DISTANCE >= ip) && (A32(ref) == A32(ip))) { token = op++; *token=0; goto _next_match; }
+
+        // Prepare next loop
+        anchor = ip++;
+        forwardH = LZ4_hashPosition(ip, tableType);
+    }
+
+_last_literals:
+    // Encode Last Literals
+    {
+        int lastRun = (int)(iend - anchor);
+        if ((limitedOutput) && (((char*)op - dest) + lastRun + 1 + ((lastRun+255-RUN_MASK)/255) > (U32)maxOutputSize)) return 0;  // Check output limit
+        if (lastRun>=(int)RUN_MASK) { *op++=(RUN_MASK<<ML_BITS); lastRun-=RUN_MASK; for(; lastRun >= 255 ; lastRun-=255) *op++ = 255; *op++ = (BYTE) lastRun; }
+        else *op++ = (BYTE)(lastRun<<ML_BITS);
+        memcpy(op, anchor, iend - anchor);
+        op += iend-anchor;
+    }
+
+    // End
+    return (int) (((char*)op)-dest);
+}
+
+
+int LZ4_compress(const char* source, char* dest, int inputSize)
+{
+#if (HEAPMODE)
+    void* ctx = ALLOCATOR(HASHNBCELLS4, 4);   // Aligned on 4-bytes boundaries
+#else
+    U32 ctx[1U<<(MEMORY_USAGE-2)] = {0};           // Ensure data is aligned on 4-bytes boundaries
+#endif
+    int result;
+
+    if (inputSize < (int)LZ4_64KLIMIT)
+        result = LZ4_compress_generic((void*)ctx, source, dest, inputSize, 0, notLimited, byU16, noPrefix);
+    else
+        result = LZ4_compress_generic((void*)ctx, source, dest, inputSize, 0, notLimited, (sizeof(void*)==8) ? byU32 : byPtr, noPrefix);
+
+#if (HEAPMODE)
+    FREEMEM(ctx);
+#endif
+    return result;
+}
+
+int LZ4_compress_continue (void* LZ4_Data, const char* source, char* dest, int inputSize)
+{
+    return LZ4_compress_generic(LZ4_Data, source, dest, inputSize, 0, notLimited, byU32, withPrefix);
+}
+
+
+int LZ4_compress_limitedOutput(const char* source, char* dest, int inputSize, int maxOutputSize)
+{
+#if (HEAPMODE)
+    void* ctx = ALLOCATOR(HASHNBCELLS4, 4);   // Aligned on 4-bytes boundaries
+#else
+    U32 ctx[1U<<(MEMORY_USAGE-2)] = {0};           // Ensure data is aligned on 4-bytes boundaries
+#endif
+    int result;
+
+    if (inputSize < (int)LZ4_64KLIMIT)
+        result = LZ4_compress_generic((void*)ctx, source, dest, inputSize, maxOutputSize, limited, byU16, noPrefix);
+    else
+        result = LZ4_compress_generic((void*)ctx, source, dest, inputSize, maxOutputSize, limited, (sizeof(void*)==8) ? byU32 : byPtr, noPrefix);
+
+#if (HEAPMODE)
+    FREEMEM(ctx);
+#endif
+    return result;
+}
+
+int LZ4_compress_limitedOutput_continue (void* LZ4_Data, const char* source, char* dest, int inputSize, int maxOutputSize)
+{
+    return LZ4_compress_generic(LZ4_Data, source, dest, inputSize, maxOutputSize, limited, byU32, withPrefix);
+}
+
+
+//****************************
+// Stream functions
+//****************************
+
+FORCE_INLINE void LZ4_init(LZ4_Data_Structure* lz4ds, const BYTE* base)
+{
+    MEM_INIT(lz4ds->hashTable, 0, sizeof(lz4ds->hashTable));
+    lz4ds->bufferStart = base;
+    lz4ds->base = base;
+    lz4ds->nextBlock = base;
+}
+
+
+void* LZ4_create (const char* inputBuffer)
+{
+    void* lz4ds = ALLOCATOR(1, sizeof(LZ4_Data_Structure));
+    LZ4_init ((LZ4_Data_Structure*)lz4ds, (const BYTE*)inputBuffer);
+    return lz4ds;
+}
+
+
+int LZ4_free (void* LZ4_Data)
+{
+    FREEMEM(LZ4_Data);
+    return (0);
+}
+
+
+char* LZ4_slideInputBuffer (void* LZ4_Data)
+{
+    LZ4_Data_Structure* lz4ds = (LZ4_Data_Structure*)LZ4_Data;
+    size_t delta = lz4ds->nextBlock - (lz4ds->bufferStart + 64 KB);
+
+    if(lz4ds->base - delta > lz4ds->base)   // underflow control
+    {
+        size_t newBaseDelta = (lz4ds->nextBlock - 64 KB) - lz4ds->base;
+        int nH;
+
+        for (nH=0; nH < HASHNBCELLS4; nH++)
+        {
+            if (lz4ds->hashTable[nH] < (U32)newBaseDelta) lz4ds->hashTable[nH] = 0;
+            else lz4ds->hashTable[nH] -= (U32)newBaseDelta;
+        }
+        lz4ds->base += newBaseDelta;
+    }
+    memcpy((void*)(lz4ds->bufferStart), (const void*)(lz4ds->nextBlock - 64 KB), 64 KB);
+    lz4ds->nextBlock -= delta;
+    lz4ds->base -= delta;
+
+    return (char*)(lz4ds->nextBlock);
+}
+
+
+//****************************
+// Decompression functions
+//****************************
+
+// This generic decompression function cover all use cases.
+// It shall be instanciated several times, using different sets of directives
+// Note that it is essential this generic function is really inlined, 
+// in order to remove useless branches during compilation optimisation.
+FORCE_INLINE int LZ4_decompress_generic(
+                 const char* source,
+                 char* dest,
+                 int inputSize,          //
+                 int outputSize,         // If endOnInput==endOnInputSize, this value is the max size of Output Buffer.
+
+                 int endOnInput,         // endOnOutputSize, endOnInputSize
+                 int prefix64k,          // noPrefix, withPrefix
+                 int partialDecoding,    // full, partial
+                 int targetOutputSize    // only used if partialDecoding==partial
+                 )
+{
+    // Local Variables
+    const BYTE* restrict ip = (const BYTE*) source;
+    const BYTE* ref;
+    const BYTE* const iend = ip + inputSize;
+
+    BYTE* op = (BYTE*) dest;
+    BYTE* const oend = op + outputSize;
+    BYTE* cpy;
+    BYTE* oexit = op + targetOutputSize;
+
+    const size_t dec32table[] = {0, 3, 2, 3, 0, 0, 0, 0};   // static reduces speed for LZ4_decompress_safe() on GCC64
+    static const size_t dec64table[] = {0, 0, 0, (size_t)-1, 0, 1, 2, 3};
+
+
+    // Special cases
+    if ((partialDecoding) && (oexit> oend-MFLIMIT)) oexit = oend-MFLIMIT;                        // targetOutputSize too high => decode everything
+    if ((endOnInput) && unlikely(outputSize==0)) return ((inputSize==1) && (*ip==0)) ? 0 : -1;   // Empty output buffer
+    if ((!endOnInput) && unlikely(outputSize==0)) return (*ip==0?1:-1);
+
+
+    // Main Loop
+    while (1)
+    {
+        unsigned token;
+        size_t length;
+
+        // get runlength
+        token = *ip++;
+        if ((length=(token>>ML_BITS)) == RUN_MASK)
+        { 
+            unsigned s=255; 
+            while (((endOnInput)?ip<iend:1) && (s==255))
+            { 
+                s = *ip++; 
+                length += s; 
+            } 
+        }
+
+        // copy literals
+        cpy = op+length;
+        if (((endOnInput) && ((cpy>(partialDecoding?oexit:oend-MFLIMIT)) || (ip+length>iend-(2+1+LASTLITERALS))) )
+            || ((!endOnInput) && (cpy>oend-COPYLENGTH)))
+        {
+            if (partialDecoding)
+            {
+                if (cpy > oend) goto _output_error;                           // Error : write attempt beyond end of output buffer
+                if ((endOnInput) && (ip+length > iend)) goto _output_error;   // Error : read attempt beyond end of input buffer
+            }
+            else
+            {
+                if ((!endOnInput) && (cpy != oend)) goto _output_error;       // Error : block decoding must stop exactly there
+                if ((endOnInput) && ((ip+length != iend) || (cpy > oend))) goto _output_error;   // Error : input must be consumed
+            }
+            memcpy(op, ip, length);
+            ip += length;
+            op += length;
+            break;                                       // Necessarily EOF, due to parsing restrictions
+        }
+        LZ4_WILDCOPY(op, ip, cpy); ip -= (op-cpy); op = cpy;
+
+        // get offset
+        LZ4_READ_LITTLEENDIAN_16(ref,cpy,ip); ip+=2;
+        if ((prefix64k==noPrefix) && unlikely(ref < (BYTE* const)dest)) goto _output_error;   // Error : offset outside destination buffer
+
+        // get matchlength
+        if ((length=(token&ML_MASK)) == ML_MASK) 
+        { 
+            while ((!endOnInput) || (ip<iend-(LASTLITERALS+1)))   // Ensure enough bytes remain for LASTLITERALS + token
+            {
+                unsigned s = *ip++; 
+                length += s; 
+                if (s==255) continue; 
+                break; 
+            }
+        }
+
+        // copy repeated sequence
+        if unlikely((op-ref)<(int)STEPSIZE)
+        {
+            const size_t dec64 = dec64table[(sizeof(void*)==4) ? 0 : op-ref];
+            op[0] = ref[0];
+            op[1] = ref[1];
+            op[2] = ref[2];
+            op[3] = ref[3];
+            op += 4, ref += 4; ref -= dec32table[op-ref];
+            A32(op) = A32(ref); 
+            op += STEPSIZE-4; ref -= dec64;
+        } else { LZ4_COPYSTEP(op,ref); }
+        cpy = op + length - (STEPSIZE-4);
+
+        if unlikely(cpy>oend-COPYLENGTH-(STEPSIZE-4))
+        {
+            if (cpy > oend-LASTLITERALS) goto _output_error;    // Error : last 5 bytes must be literals
+            LZ4_SECURECOPY(op, ref, (oend-COPYLENGTH));
+            while(op<cpy) *op++=*ref++;
+            op=cpy;
+            continue;
+        }
+        LZ4_WILDCOPY(op, ref, cpy);
+        op=cpy;   // correction
+    }
+
+    // end of decoding
+    if (endOnInput)
+       return (int) (((char*)op)-dest);     // Nb of output bytes decoded
+    else
+       return (int) (((char*)ip)-source);   // Nb of input bytes read
+
+    // Overflow error detected
+_output_error:
+    return (int) (-(((char*)ip)-source))-1;
+}
+
+
+int LZ4_decompress_safe(const char* source, char* dest, int inputSize, int maxOutputSize)
+{
+    return LZ4_decompress_generic(source, dest, inputSize, maxOutputSize, endOnInputSize, noPrefix, full, 0);
+}
+
+int LZ4_decompress_safe_withPrefix64k(const char* source, char* dest, int inputSize, int maxOutputSize)
+{
+    return LZ4_decompress_generic(source, dest, inputSize, maxOutputSize, endOnInputSize, withPrefix, full, 0);
+}
+
+int LZ4_decompress_safe_partial(const char* source, char* dest, int inputSize, int targetOutputSize, int maxOutputSize)
+{
+    return LZ4_decompress_generic(source, dest, inputSize, maxOutputSize, endOnInputSize, noPrefix, partial, targetOutputSize);
+}
+
+int LZ4_decompress_fast_withPrefix64k(const char* source, char* dest, int outputSize)
+{
+    return LZ4_decompress_generic(source, dest, 0, outputSize, endOnOutputSize, withPrefix, full, 0);
+}
+
+int LZ4_decompress_fast(const char* source, char* dest, int outputSize)
+{
+#ifdef _MSC_VER   // This version is faster with Visual
+    return LZ4_decompress_generic(source, dest, 0, outputSize, endOnOutputSize, noPrefix, full, 0);
+#else
+    return LZ4_decompress_generic(source, dest, 0, outputSize, endOnOutputSize, withPrefix, full, 0);
+#endif
+}
+

Source/ThirdParty/LZ4/lz4.h

@@ -0,0 +1,205 @@
+/*
+   LZ4 - Fast LZ compression algorithm
+   Header File
+   Copyright (C) 2011-2013, 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 :
+   - LZ4 homepage : http://fastcompression.blogspot.com/p/lz4.html
+   - LZ4 source repository : http://code.google.com/p/lz4/
+*/
+#pragma once
+
+#if defined (__cplusplus)
+extern "C" {
+#endif
+
+
+//**************************************
+// Compiler Options
+//**************************************
+#if defined(_MSC_VER) && !defined(__cplusplus)   // Visual Studio
+#  define inline __inline           // Visual C is not C99, but supports some kind of inline
+#endif
+
+
+//****************************
+// Simple Functions
+//****************************
+
+int LZ4_compress        (const char* source, char* dest, int inputSize);
+int LZ4_decompress_safe (const char* source, char* dest, int inputSize, int maxOutputSize);
+
+/*
+LZ4_compress() :
+    Compresses 'inputSize' bytes from 'source' into 'dest'.
+    Destination buffer must be already allocated,
+    and must be sized to handle worst cases situations (input data not compressible)
+    Worst case size evaluation is provided by function LZ4_compressBound()
+    inputSize : Max supported value is ~1.9GB
+    return : the number of bytes written in buffer dest
+             or 0 if the compression fails
+
+LZ4_decompress_safe() :
+    maxOutputSize : is the size of the destination buffer (which must be already allocated)
+    return : the number of bytes decoded in the destination buffer (necessarily <= maxOutputSize)
+             If the source stream is detected malformed, the function will stop decoding and return a negative result.
+             This function is protected against buffer overflow exploits (never writes outside of output buffer, and never reads outside of input buffer). Therefore, it is protected against malicious data packets
+*/
+
+
+//****************************
+// Advanced Functions
+//****************************
+
+static inline int LZ4_compressBound(int isize)   { return ((isize) + ((isize)/255) + 16); }
+#define           LZ4_COMPRESSBOUND(    isize)            ((isize) + ((isize)/255) + 16)
+
+/*
+LZ4_compressBound() :
+    Provides the maximum size that LZ4 may output in a "worst case" scenario (input data not compressible)
+    primarily useful for memory allocation of output buffer.
+    inline function is recommended for the general case,
+    macro is also provided when result needs to be evaluated at compilation (such as stack memory allocation).
+
+    isize  : is the input size. Max supported value is ~1.9GB
+    return : maximum output size in a "worst case" scenario
+    note : this function is limited by "int" range (2^31-1)
+*/
+
+
+int LZ4_compress_limitedOutput (const char* source, char* dest, int inputSize, int maxOutputSize);
+
+/*
+LZ4_compress_limitedOutput() :
+    Compress 'inputSize' bytes from 'source' into an output buffer 'dest' of maximum size 'maxOutputSize'.
+    If it cannot achieve it, compression will stop, and result of the function will be zero.
+    This function never writes outside of provided output buffer.
+
+    inputSize  : Max supported value is ~1.9GB
+    maxOutputSize : is the size of the destination buffer (which must be already allocated)
+    return : the number of bytes written in buffer 'dest'
+             or 0 if the compression fails
+*/
+
+
+int LZ4_decompress_fast (const char* source, char* dest, int outputSize);
+
+/*
+LZ4_decompress_fast() :
+    outputSize : is the original (uncompressed) size
+    return : the number of bytes read from the source buffer (in other words, the compressed size)
+             If the source stream is malformed, the function will stop decoding and return a negative result.
+    note : This function is a bit faster than LZ4_decompress_safe()
+           This function never writes outside of output buffers, but may read beyond input buffer in case of malicious data packet.
+           Use this function preferably into a trusted environment (data to decode comes from a trusted source).
+           Destination buffer must be already allocated. Its size must be a minimum of 'outputSize' bytes.
+*/
+
+int LZ4_decompress_safe_partial (const char* source, char* dest, int inputSize, int targetOutputSize, int maxOutputSize);
+
+/*
+LZ4_decompress_safe_partial() :
+    This function decompress a compressed block of size 'inputSize' at position 'source'
+    into output buffer 'dest' of size 'maxOutputSize'.
+    The function stops decompressing operation as soon as 'targetOutputSize' has been reached,
+    reducing decompression time.
+    return : the number of bytes decoded in the destination buffer (necessarily <= maxOutputSize)
+       Note : this number can be < 'targetOutputSize' should the compressed block to decode be smaller.
+             Always control how many bytes were decoded.
+             If the source stream is detected malformed, the function will stop decoding and return a negative result.
+             This function never writes outside of output buffer, and never reads outside of input buffer. It is therefore protected against malicious data packets
+*/
+
+
+//****************************
+// Stream Functions
+//****************************
+
+void* LZ4_create (const char* inputBuffer);
+int   LZ4_compress_continue (void* LZ4_Data, const char* source, char* dest, int inputSize);
+int   LZ4_compress_limitedOutput_continue (void* LZ4_Data, const char* source, char* dest, int inputSize, int maxOutputSize);
+char* LZ4_slideInputBuffer (void* LZ4_Data);
+int   LZ4_free (void* LZ4_Data);
+
+/* 
+These functions allow the compression of dependent blocks, where each block benefits from prior 64 KB within preceding blocks.
+In order to achieve this, it is necessary to start creating the LZ4 Data Structure, thanks to the function :
+
+void* LZ4_create (const char* inputBuffer);
+The result of the function is the (void*) pointer on the LZ4 Data Structure.
+This pointer will be needed in all other functions.
+If the pointer returned is NULL, then the allocation has failed, and compression must be aborted.
+The only parameter 'const char* inputBuffer' must, obviously, point at the beginning of input buffer.
+The input buffer must be already allocated, and size at least 192KB.
+'inputBuffer' will also be the 'const char* source' of the first block.
+
+All blocks are expected to lay next to each other within the input buffer, starting from 'inputBuffer'.
+To compress each block, use either LZ4_compress_continue() or LZ4_compress_limitedOutput_continue().
+Their behavior are identical to LZ4_compress() or LZ4_compress_limitedOutput(), 
+but require the LZ4 Data Structure as their first argument, and check that each block starts right after the previous one.
+If next block does not begin immediately after the previous one, the compression will fail (return 0).
+
+When it's no longer possible to lay the next block after the previous one (not enough space left into input buffer), a call to : 
+char* LZ4_slideInputBuffer(void* LZ4_Data);
+must be performed. It will typically copy the latest 64KB of input at the beginning of input buffer.
+Note that, for this function to work properly, minimum size of an input buffer must be 192KB.
+==> The memory position where the next input data block must start is provided as the result of the function.
+
+Compression can then resume, using LZ4_compress_continue() or LZ4_compress_limitedOutput_continue(), as usual.
+
+When compression is completed, a call to LZ4_free() will release the memory used by the LZ4 Data Structure.
+*/
+
+
+int LZ4_decompress_safe_withPrefix64k (const char* source, char* dest, int inputSize, int maxOutputSize);
+int LZ4_decompress_fast_withPrefix64k (const char* source, char* dest, int outputSize);
+
+/*
+*_withPrefix64k() :
+    These decoding functions work the same as their "normal name" versions,
+    but can use up to 64KB of data in front of 'char* dest'.
+    These functions are necessary to decode inter-dependant blocks.
+*/
+
+
+//****************************
+// Obsolete Functions
+//****************************
+
+static inline int LZ4_uncompress (const char* source, char* dest, int outputSize) { return LZ4_decompress_fast(source, dest, outputSize); }
+static inline int LZ4_uncompress_unknownOutputSize (const char* source, char* dest, int isize, int maxOutputSize) { return LZ4_decompress_safe(source, dest, isize, maxOutputSize); }
+
+/*
+These functions are deprecated and should no longer be used.
+They are provided here for compatibility with existing user programs.
+*/
+
+
+
+#if defined (__cplusplus)
+}
+#endif

Source/ThirdParty/LZ4/lz4_format_description.txt

@@ -0,0 +1,120 @@
+LZ4 Format Description
+Last revised: 2012-02-27
+Author : Y. Collet
+
+
+
+This small specification intents to provide enough information
+to anyone willing to produce LZ4-compatible compressed data blocks
+using any programming language.
+
+LZ4 is an LZ77-type compressor with a fixed, byte-oriented encoding.
+The most important design principle behind LZ4 is simplicity.
+It helps to create an easy to read and maintain source code.
+It also helps later on for optimisations, compactness, and speed.
+There is no entropy encoder backend nor framing layer.
+The latter is assumed to be handled by other parts of the system.
+
+This document only describes the format,
+not how the LZ4 compressor nor decompressor actually work.
+The correctness of the decompressor should not depend
+on implementation details of the compressor, and vice versa.
+
+
+
+-- Compressed block format --
+
+An LZ4 compressed block is composed of sequences.
+Schematically, a sequence is a suite of literals, followed by a match copy.
+
+Each sequence starts with a token.
+The token is a one byte value, separated into two 4-bits fields.
+Therefore each field ranges from 0 to 15.
+
+
+The first field uses the 4 high-bits of the token.
+It provides the length of literals to follow.
+(Note : a literal is a not-compressed byte).
+If the field value is 0, then there is no literal.
+If it is 15, then we need to add some more bytes to indicate the full length.
+Each additionnal byte then represent a value from 0 to 255,
+which is added to the previous value to produce a total length.
+When the byte value is 255, another byte is output.
+There can be any number of bytes following the token. There is no "size limit".
+(Sidenote this is why a not-compressible input block is expanded by 0.4%).
+
+Example 1 : A length of 48 will be represented as :
+- 15 : value for the 4-bits High field
+- 33 : (=48-15) remaining length to reach 48
+
+Example 2 : A length of 280 will be represented as :
+- 15  : value for the 4-bits High field
+- 255 : following byte is maxed, since 280-15 >= 255
+- 10  : (=280 - 15 - 255) ) remaining length to reach 280
+
+Example 3 : A length of 15 will be represented as :
+- 15 : value for the 4-bits High field
+- 0  : (=15-15) yes, the zero must be output
+
+Following the token and optional length bytes, are the literals themselves.
+They are exactly as numerous as previously decoded (length of literals).
+It's possible that there are zero literal.
+
+
+Following the literals is the match copy operation.
+
+It starts by the offset.
+This is a 2 bytes value, in little endian format.
+
+The offset represents the position of the match to be copied from.
+1 means "current position - 1 byte".
+The maximum offset value is 65535, 65536 cannot be coded.
+Note that 0 is an invalid value, not used. 
+
+Then we need to extract the match length.
+For this, we use the second token field, the low 4-bits.
+Value, obviously, ranges from 0 to 15.
+However here, 0 means that the copy operation will be minimal.
+The minimum length of a match, called minmatch, is 4. 
+As a consequence, a 0 value means 4 bytes, and a value of 15 means 19+ bytes.
+Similar to literal length, on reaching the highest possible value (15), 
+we output additional bytes, one at a time, with values ranging from 0 to 255.
+They are added to total to provide the final match length.
+A 255 value means there is another byte to read and add.
+There is no limit to the number of optional bytes that can be output this way.
+(This points towards a maximum achievable compression ratio of ~250).
+
+With the offset and the matchlength,
+the decoder can now proceed to copy the data from the already decoded buffer.
+On decoding the matchlength, we reach the end of the compressed sequence,
+and therefore start another one.
+
+
+-- Parsing restrictions --
+
+There are specific parsing rules to respect in order to remain compatible
+with assumptions made by the decoder :
+1) The last 5 bytes are always literals
+2) The last match must start at least 12 bytes before end of block
+Consequently, a block with less than 13 bytes cannot be compressed.
+These rules are in place to ensure that the decoder
+will never read beyond the input buffer, nor write beyond the output buffer.
+
+Note that the last sequence is also incomplete,
+and stops right after literals.
+
+
+-- Additional notes --
+
+There is no assumption nor limits to the way the compressor
+searches and selects matches within the source data block.
+It could be a fast scan, a multi-probe, a full search using BST,
+standard hash chains or MMC, well whatever.
+
+Advanced parsing strategies can also be implemented, such as lazy match,
+or full optimal parsing.
+
+All these trade-off offer distinctive speed/memory/compression advantages.
+Whatever the method used by the compressor, its result will be decodable
+by any LZ4 decoder if it follows the format specification described above.
+

Source/ThirdParty/LZ4/lz4hc.c

@@ -0,0 +1,817 @@
+/*
+   LZ4 HC - High Compression Mode of LZ4
+   Copyright (C) 2011-2013, 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 :
+   - LZ4 homepage : http://fastcompression.blogspot.com/p/lz4.html
+   - LZ4 source repository : http://code.google.com/p/lz4/
+*/
+
+//**************************************
+// Memory routines
+//**************************************
+#include <stdlib.h>   // calloc, free
+#define ALLOCATOR(s)  calloc(1,s)
+#define FREEMEM       free
+#include <string.h>   // memset, memcpy
+#define MEM_INIT      memset
+
+
+//**************************************
+// CPU Feature Detection
+//**************************************
+// 32 or 64 bits ?
+#if (defined(__x86_64__) || defined(_M_X64) || defined(_WIN64) \
+  || defined(__powerpc64__) || defined(__ppc64__) || defined(__PPC64__) \
+  || defined(__64BIT__) || defined(_LP64) || defined(__LP64__) \
+  || defined(__ia64) || defined(__itanium__) || defined(_M_IA64) )   // Detects 64 bits mode
+#  define LZ4_ARCH64 1
+#else
+#  define LZ4_ARCH64 0
+#endif
+
+// Little Endian or Big Endian ?
+// Overwrite the #define below if you know your architecture endianess
+#if defined (__GLIBC__)
+#  include <endian.h>
+#  if (__BYTE_ORDER == __BIG_ENDIAN)
+#     define LZ4_BIG_ENDIAN 1
+#  endif
+#elif (defined(__BIG_ENDIAN__) || defined(__BIG_ENDIAN) || defined(_BIG_ENDIAN)) && !(defined(__LITTLE_ENDIAN__) || defined(__LITTLE_ENDIAN) || defined(_LITTLE_ENDIAN))
+#  define LZ4_BIG_ENDIAN 1
+#elif defined(__sparc) || defined(__sparc__) \
+   || defined(__powerpc__) || defined(__ppc__) || defined(__PPC__) \
+   || defined(__hpux)  || defined(__hppa) \
+   || defined(_MIPSEB) || defined(__s390__)
+#  define LZ4_BIG_ENDIAN 1
+#else
+// Little Endian assumed. PDP Endian and other very rare endian format are unsupported.
+#endif
+
+// Unaligned memory access is automatically enabled for "common" CPU, such as x86.
+// For others CPU, the compiler will be more cautious, and insert extra code to ensure aligned access is respected
+// If you know your target CPU supports unaligned memory access, you want to force this option manually to improve performance
+#if defined(__ARM_FEATURE_UNALIGNED)
+#  define LZ4_FORCE_UNALIGNED_ACCESS 1
+#endif
+
+// Define this parameter if your target system or compiler does not support hardware bit count
+#if defined(_MSC_VER) && defined(_WIN32_WCE)            // Visual Studio for Windows CE does not support Hardware bit count
+#  define LZ4_FORCE_SW_BITCOUNT
+#endif
+
+
+//**************************************
+// Compiler Options
+//**************************************
+#if defined (__STDC_VERSION__) && __STDC_VERSION__ >= 199901L   // C99
+  /* "restrict" is a known keyword */
+#else
+#  define restrict  // Disable restrict
+#endif
+
+#ifdef _MSC_VER    // Visual Studio
+#  define FORCE_INLINE static __forceinline
+#  include <intrin.h>                    // For Visual 2005
+#  if LZ4_ARCH64   // 64-bits
+#    pragma intrinsic(_BitScanForward64) // For Visual 2005
+#    pragma intrinsic(_BitScanReverse64) // For Visual 2005
+#  else            // 32-bits
+#    pragma intrinsic(_BitScanForward)   // For Visual 2005
+#    pragma intrinsic(_BitScanReverse)   // For Visual 2005
+#  endif
+#  pragma warning(disable : 4127)        // disable: C4127: conditional expression is constant
+#  pragma warning(disable : 4701)        // disable: C4701: potentially uninitialized local variable used
+#else 
+#  ifdef __GNUC__
+#    define FORCE_INLINE static inline __attribute__((always_inline))
+#  else
+#    define FORCE_INLINE static inline
+#  endif
+#endif
+
+#ifdef _MSC_VER  // Visual Studio
+#  define lz4_bswap16(x) _byteswap_ushort(x)
+#else
+#  define lz4_bswap16(x)  ((unsigned short int) ((((x) >> 8) & 0xffu) | (((x) & 0xffu) << 8)))
+#endif
+
+
+//**************************************
+// Includes
+//**************************************
+#include "lz4hc.h"
+#include "lz4.h"
+
+
+//**************************************
+// Basic Types
+//**************************************
+#if defined (__STDC_VERSION__) && __STDC_VERSION__ >= 199901L   // C99
+# include <stdint.h>
+  typedef uint8_t  BYTE;
+  typedef uint16_t U16;
+  typedef uint32_t U32;
+  typedef  int32_t S32;
+  typedef uint64_t U64;
+#else
+  typedef unsigned char       BYTE;
+  typedef unsigned short      U16;
+  typedef unsigned int        U32;
+  typedef   signed int        S32;
+  typedef unsigned long long  U64;
+#endif
+
+#if defined(__GNUC__)  && !defined(LZ4_FORCE_UNALIGNED_ACCESS)
+#  define _PACKED __attribute__ ((packed))
+#else
+#  define _PACKED
+#endif
+
+#if !defined(LZ4_FORCE_UNALIGNED_ACCESS) && !defined(__GNUC__)
+#  ifdef __IBMC__
+#    pragma pack(1)
+#  else
+#    pragma pack(push, 1)
+#  endif
+#endif
+
+typedef struct _U16_S { U16 v; } _PACKED U16_S;
+typedef struct _U32_S { U32 v; } _PACKED U32_S;
+typedef struct _U64_S { U64 v; } _PACKED U64_S;
+
+#if !defined(LZ4_FORCE_UNALIGNED_ACCESS) && !defined(__GNUC__)
+#  pragma pack(pop)
+#endif
+
+#define A64(x) (((U64_S *)(x))->v)
+#define A32(x) (((U32_S *)(x))->v)
+#define A16(x) (((U16_S *)(x))->v)
+
+
+//**************************************
+// Constants
+//**************************************
+#define MINMATCH 4
+
+#define DICTIONARY_LOGSIZE 16
+#define MAXD (1<<DICTIONARY_LOGSIZE)
+#define MAXD_MASK ((U32)(MAXD - 1))
+#define MAX_DISTANCE (MAXD - 1)
+
+#define HASH_LOG (DICTIONARY_LOGSIZE-1)
+#define HASHTABLESIZE (1 << HASH_LOG)
+#define HASH_MASK (HASHTABLESIZE - 1)
+
+#define MAX_NB_ATTEMPTS 256
+
+#define ML_BITS  4
+#define ML_MASK  (size_t)((1U<<ML_BITS)-1)
+#define RUN_BITS (8-ML_BITS)
+#define RUN_MASK ((1U<<RUN_BITS)-1)
+
+#define COPYLENGTH 8
+#define LASTLITERALS 5
+#define MFLIMIT (COPYLENGTH+MINMATCH)
+#define MINLENGTH (MFLIMIT+1)
+#define OPTIMAL_ML (int)((ML_MASK-1)+MINMATCH)
+
+#define KB *(1U<<10)
+#define MB *(1U<<20)
+#define GB *(1U<<30)
+
+
+//**************************************
+// Architecture-specific macros
+//**************************************
+#if LZ4_ARCH64   // 64-bit
+#  define STEPSIZE 8
+#  define LZ4_COPYSTEP(s,d)     A64(d) = A64(s); d+=8; s+=8;
+#  define LZ4_COPYPACKET(s,d)   LZ4_COPYSTEP(s,d)
+#  define UARCH U64
+#  define AARCH A64
+#  define HTYPE                 U32
+#  define INITBASE(b,s)         const BYTE* const b = s
+#else   // 32-bit
+#  define STEPSIZE 4
+#  define LZ4_COPYSTEP(s,d)     A32(d) = A32(s); d+=4; s+=4;
+#  define LZ4_COPYPACKET(s,d)   LZ4_COPYSTEP(s,d); LZ4_COPYSTEP(s,d);
+#  define UARCH U32
+#  define AARCH A32
+//#  define HTYPE                 const BYTE*
+//#  define INITBASE(b,s)         const int b = 0
+#  define HTYPE                 U32
+#  define INITBASE(b,s)         const BYTE* const b = s
+#endif
+
+#if defined(LZ4_BIG_ENDIAN)
+#  define LZ4_READ_LITTLEENDIAN_16(d,s,p) { U16 v = A16(p); v = lz4_bswap16(v); d = (s) - v; }
+#  define LZ4_WRITE_LITTLEENDIAN_16(p,i)  { U16 v = (U16)(i); v = lz4_bswap16(v); A16(p) = v; p+=2; }
+#else   // Little Endian
+#  define LZ4_READ_LITTLEENDIAN_16(d,s,p) { d = (s) - A16(p); }
+#  define LZ4_WRITE_LITTLEENDIAN_16(p,v)  { A16(p) = v; p+=2; }
+#endif
+
+
+//************************************************************
+// Local Types
+//************************************************************
+typedef struct 
+{
+    const BYTE* inputBuffer;
+    const BYTE* base;
+    const BYTE* end;
+    HTYPE hashTable[HASHTABLESIZE];
+    U16 chainTable[MAXD];
+    const BYTE* nextToUpdate;
+} LZ4HC_Data_Structure;
+
+
+//**************************************
+// Macros
+//**************************************
+#define LZ4_WILDCOPY(s,d,e)    do { LZ4_COPYPACKET(s,d) } while (d<e);
+#define LZ4_BLINDCOPY(s,d,l)   { BYTE* e=d+l; LZ4_WILDCOPY(s,d,e); d=e; }
+#define HASH_FUNCTION(i)       (((i) * 2654435761U) >> ((MINMATCH*8)-HASH_LOG))
+#define HASH_VALUE(p)          HASH_FUNCTION(A32(p))
+#define HASH_POINTER(p)        (HashTable[HASH_VALUE(p)] + base)
+#define DELTANEXT(p)           chainTable[(size_t)(p) & MAXD_MASK] 
+#define GETNEXT(p)             ((p) - (size_t)DELTANEXT(p))
+
+
+//**************************************
+// Private functions
+//**************************************
+#if LZ4_ARCH64
+
+FORCE_INLINE int LZ4_NbCommonBytes (register U64 val)
+{
+#if defined(LZ4_BIG_ENDIAN)
+#  if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
+    unsigned long r = 0;
+    _BitScanReverse64( &r, val );
+    return (int)(r>>3);
+#  elif defined(__GNUC__) && ((__GNUC__ * 100 + __GNUC_MINOR__) >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
+    return (__builtin_clzll(val) >> 3); 
+#  else
+    int r;
+    if (!(val>>32)) { r=4; } else { r=0; val>>=32; }
+    if (!(val>>16)) { r+=2; val>>=8; } else { val>>=24; }
+    r += (!val);
+    return r;
+#  endif
+#else
+#  if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
+    unsigned long r = 0;
+    _BitScanForward64( &r, val );
+    return (int)(r>>3);
+#  elif defined(__GNUC__) && ((__GNUC__ * 100 + __GNUC_MINOR__) >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
+    return (__builtin_ctzll(val) >> 3); 
+#  else
+    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 };
+    return DeBruijnBytePos[((U64)((val & -val) * 0x0218A392CDABBD3F)) >> 58];
+#  endif
+#endif
+}
+
+#else
+
+FORCE_INLINE int LZ4_NbCommonBytes (register U32 val)
+{
+#if defined(LZ4_BIG_ENDIAN)
+#  if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
+    unsigned long r;
+    _BitScanReverse( &r, val );
+    return (int)(r>>3);
+#  elif defined(__GNUC__) && ((__GNUC__ * 100 + __GNUC_MINOR__) >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
+    return (__builtin_clz(val) >> 3); 
+#  else
+    int r;
+    if (!(val>>16)) { r=2; val>>=8; } else { r=0; val>>=24; }
+    r += (!val);
+    return r;
+#  endif
+#else
+#  if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
+    unsigned long r;
+    _BitScanForward( &r, val );
+    return (int)(r>>3);
+#  elif defined(__GNUC__) && ((__GNUC__ * 100 + __GNUC_MINOR__) >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
+    return (__builtin_ctz(val) >> 3); 
+#  else
+    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 };
+    return DeBruijnBytePos[((U32)((val & -(S32)val) * 0x077CB531U)) >> 27];
+#  endif
+#endif
+}
+
+#endif
+
+
+FORCE_INLINE void LZ4_initHC (LZ4HC_Data_Structure* hc4, const BYTE* base)
+{
+    MEM_INIT((void*)hc4->hashTable, 0, sizeof(hc4->hashTable));
+    MEM_INIT(hc4->chainTable, 0xFF, sizeof(hc4->chainTable));
+    hc4->nextToUpdate = base + 1;
+    hc4->base = base;
+    hc4->inputBuffer = base;
+    hc4->end = base;
+}
+
+
+void* LZ4_createHC (const char* inputBuffer)
+{
+    void* hc4 = ALLOCATOR(sizeof(LZ4HC_Data_Structure));
+    LZ4_initHC ((LZ4HC_Data_Structure*)hc4, (const BYTE*)inputBuffer);
+    return hc4;
+}
+
+
+int LZ4_freeHC (void* LZ4HC_Data)
+{
+    FREEMEM(LZ4HC_Data);
+    return (0);
+}
+
+
+// Update chains up to ip (excluded)
+FORCE_INLINE void LZ4HC_Insert (LZ4HC_Data_Structure* hc4, const BYTE* ip)
+{
+    U16*   chainTable = hc4->chainTable;
+    HTYPE* HashTable  = hc4->hashTable;
+    INITBASE(base,hc4->base);
+
+    while(hc4->nextToUpdate < ip)
+    {
+        const BYTE* const p = hc4->nextToUpdate;
+        size_t delta = (p) - HASH_POINTER(p); 
+        if (delta>MAX_DISTANCE) delta = MAX_DISTANCE; 
+        DELTANEXT(p) = (U16)delta; 
+        HashTable[HASH_VALUE(p)] = (HTYPE)((p) - base);
+        hc4->nextToUpdate++;
+    }
+}
+
+
+char* LZ4_slideInputBufferHC(void* LZ4HC_Data)
+{
+    LZ4HC_Data_Structure* hc4 = (LZ4HC_Data_Structure*)LZ4HC_Data;
+    U32 distance = (U32)(hc4->end - hc4->inputBuffer) - 64 KB;
+    distance = (distance >> 16) << 16;   // Must be a multiple of 64 KB
+    LZ4HC_Insert(hc4, hc4->end - MINMATCH);
+    memcpy((void*)(hc4->end - 64 KB - distance), (const void*)(hc4->end - 64 KB), 64 KB);
+    hc4->nextToUpdate -= distance;
+    hc4->base -= distance;
+    if ((U32)(hc4->inputBuffer - hc4->base) > 1 GB + 64 KB)   // Avoid overflow
+    {
+        int i;
+        hc4->base += 1 GB;
+        for (i=0; i<HASHTABLESIZE; i++) hc4->hashTable[i] -= 1 GB;
+    }
+    hc4->end -= distance;
+    return (char*)(hc4->end);
+}
+
+
+FORCE_INLINE size_t LZ4HC_CommonLength (const BYTE* p1, const BYTE* p2, const BYTE* const matchlimit)
+{
+    const BYTE* p1t = p1;
+
+    while (p1t<matchlimit-(STEPSIZE-1))
+    {
+        UARCH diff = AARCH(p2) ^ AARCH(p1t);
+        if (!diff) { p1t+=STEPSIZE; p2+=STEPSIZE; continue; }
+        p1t += LZ4_NbCommonBytes(diff);
+        return (p1t - p1);
+    }
+    if (LZ4_ARCH64) if ((p1t<(matchlimit-3)) && (A32(p2) == A32(p1t))) { p1t+=4; p2+=4; }
+    if ((p1t<(matchlimit-1)) && (A16(p2) == A16(p1t))) { p1t+=2; p2+=2; }
+    if ((p1t<matchlimit) && (*p2 == *p1t)) p1t++;
+    return (p1t - p1);
+}
+
+
+FORCE_INLINE int LZ4HC_InsertAndFindBestMatch (LZ4HC_Data_Structure* hc4, const BYTE* ip, const BYTE* const matchlimit, const BYTE** matchpos)
+{
+    U16* const chainTable = hc4->chainTable;
+    HTYPE* const HashTable = hc4->hashTable;
+    const BYTE* ref;
+    INITBASE(base,hc4->base);
+    int nbAttempts=MAX_NB_ATTEMPTS;
+    size_t repl=0, ml=0;
+    U16 delta=0;  // useless assignment, to remove an uninitialization warning
+
+    // HC4 match finder
+    LZ4HC_Insert(hc4, ip);
+    ref = HASH_POINTER(ip);
+
+#define REPEAT_OPTIMIZATION
+#ifdef REPEAT_OPTIMIZATION
+    // Detect repetitive sequences of length <= 4
+    if ((U32)(ip-ref) <= 4)        // potential repetition
+    {
+        if (A32(ref) == A32(ip))   // confirmed
+        {
+            delta = (U16)(ip-ref);
+            repl = ml  = LZ4HC_CommonLength(ip+MINMATCH, ref+MINMATCH, matchlimit) + MINMATCH;
+            *matchpos = ref;
+        }
+        ref = GETNEXT(ref);
+    }
+#endif
+
+    while (((U32)(ip-ref) <= MAX_DISTANCE) && (nbAttempts))
+    {
+        nbAttempts--;
+        if (*(ref+ml) == *(ip+ml))
+        if (A32(ref) == A32(ip))
+        {
+            size_t mlt = LZ4HC_CommonLength(ip+MINMATCH, ref+MINMATCH, matchlimit) + MINMATCH;
+            if (mlt > ml) { ml = mlt; *matchpos = ref; }
+        }
+        ref = GETNEXT(ref);
+    }
+
+#ifdef REPEAT_OPTIMIZATION
+    // Complete table
+    if (repl)
+    {
+        const BYTE* ptr = ip;
+        const BYTE* end;
+
+        end = ip + repl - (MINMATCH-1);
+        while(ptr < end-delta)
+        {
+            DELTANEXT(ptr) = delta;    // Pre-Load
+            ptr++;
+        }
+        do
+        {
+            DELTANEXT(ptr) = delta;    
+            HashTable[HASH_VALUE(ptr)] = (HTYPE)((ptr) - base);     // Head of chain
+            ptr++;
+        } while(ptr < end);
+        hc4->nextToUpdate = end;
+    }
+#endif 
+
+    return (int)ml;
+}
+
+
+FORCE_INLINE int LZ4HC_InsertAndGetWiderMatch (LZ4HC_Data_Structure* hc4, const BYTE* ip, const BYTE* startLimit, const BYTE* matchlimit, int longest, const BYTE** matchpos, const BYTE** startpos)
+{
+    U16* const  chainTable = hc4->chainTable;
+    HTYPE* const HashTable = hc4->hashTable;
+    INITBASE(base,hc4->base);
+    const BYTE*  ref;
+    int nbAttempts = MAX_NB_ATTEMPTS;
+    int delta = (int)(ip-startLimit);
+
+    // First Match
+    LZ4HC_Insert(hc4, ip);
+    ref = HASH_POINTER(ip);
+
+    while (((U32)(ip-ref) <= MAX_DISTANCE) && (nbAttempts))
+    {
+        nbAttempts--;
+        if (*(startLimit + longest) == *(ref - delta + longest))
+        if (A32(ref) == A32(ip))
+        {
+#if 1
+            const BYTE* reft = ref+MINMATCH;
+            const BYTE* ipt = ip+MINMATCH;
+            const BYTE* startt = ip;
+
+            while (ipt<matchlimit-(STEPSIZE-1))
+            {
+                UARCH diff = AARCH(reft) ^ AARCH(ipt);
+                if (!diff) { ipt+=STEPSIZE; reft+=STEPSIZE; continue; }
+                ipt += LZ4_NbCommonBytes(diff);
+                goto _endCount;
+            }
+            if (LZ4_ARCH64) if ((ipt<(matchlimit-3)) && (A32(reft) == A32(ipt))) { ipt+=4; reft+=4; }
+            if ((ipt<(matchlimit-1)) && (A16(reft) == A16(ipt))) { ipt+=2; reft+=2; }
+            if ((ipt<matchlimit) && (*reft == *ipt)) ipt++;
+_endCount:
+            reft = ref;
+#else
+            // Easier for code maintenance, but unfortunately slower too
+            const BYTE* startt = ip;
+            const BYTE* reft = ref;
+            const BYTE* ipt = ip + MINMATCH + LZ4HC_CommonLength(ip+MINMATCH, ref+MINMATCH, matchlimit);
+#endif
+
+            while ((startt>startLimit) && (reft > hc4->inputBuffer) && (startt[-1] == reft[-1])) {startt--; reft--;}
+
+            if ((ipt-startt) > longest)
+            {
+                longest = (int)(ipt-startt);
+                *matchpos = reft;
+                *startpos = startt;
+            }
+        }
+        ref = GETNEXT(ref);
+    }
+
+    return longest;
+}
+
+
+typedef enum { noLimit = 0, limitedOutput = 1 } limitedOutput_directive;
+
+FORCE_INLINE int LZ4HC_encodeSequence (
+                       const BYTE** ip,
+                       BYTE** op,
+                       const BYTE** anchor,
+                       int matchLength,
+                       const BYTE* ref,
+                       limitedOutput_directive limitedOutputBuffer,
+                       BYTE* oend)
+{
+    int length;
+    BYTE* token;
+
+    // Encode Literal length
+    length = (int)(*ip - *anchor);
+    token = (*op)++;
+    if ((limitedOutputBuffer) && ((*op + length + (2 + 1 + LASTLITERALS) + (length>>8)) > oend)) return 1;   // Check output limit
+    if (length>=(int)RUN_MASK) { int len; *token=(RUN_MASK<<ML_BITS); len = length-RUN_MASK; for(; len > 254 ; len-=255) *(*op)++ = 255;  *(*op)++ = (BYTE)len; } 
+    else *token = (BYTE)(length<<ML_BITS);
+
+    // Copy Literals
+    LZ4_BLINDCOPY(*anchor, *op, length);
+
+    // Encode Offset
+    LZ4_WRITE_LITTLEENDIAN_16(*op,(U16)(*ip-ref));
+
+    // Encode MatchLength
+    length = (int)(matchLength-MINMATCH);
+    if ((limitedOutputBuffer) && (*op + (1 + LASTLITERALS) + (length>>8) > oend)) return 1;   // Check output limit
+    if (length>=(int)ML_MASK) { *token+=ML_MASK; length-=ML_MASK; for(; length > 509 ; length-=510) { *(*op)++ = 255; *(*op)++ = 255; } if (length > 254) { length-=255; *(*op)++ = 255; } *(*op)++ = (BYTE)length; } 
+    else *token += (BYTE)(length);	
+
+    // Prepare next loop
+    *ip += matchLength;
+    *anchor = *ip; 
+
+    return 0;
+}
+
+
+static int LZ4HC_compress_generic (
+                 void* ctxvoid,
+                 const char* source, 
+                 char* dest,
+                 int inputSize,
+                 int maxOutputSize,
+                 limitedOutput_directive limit
+                )
+{
+    LZ4HC_Data_Structure* ctx = (LZ4HC_Data_Structure*) ctxvoid;
+    const BYTE* ip = (const BYTE*) source;
+    const BYTE* anchor = ip;
+    const BYTE* const iend = ip + inputSize;
+    const BYTE* const mflimit = iend - MFLIMIT;
+    const BYTE* const matchlimit = (iend - LASTLITERALS);
+
+    BYTE* op = (BYTE*) dest;
+    BYTE* const oend = op + maxOutputSize;
+
+    int   ml, ml2, ml3, ml0;
+    const BYTE* ref=NULL;
+    const BYTE* start2=NULL;
+    const BYTE* ref2=NULL;
+    const BYTE* start3=NULL;
+    const BYTE* ref3=NULL;
+    const BYTE* start0;
+    const BYTE* ref0;
+
+
+    // Ensure blocks follow each other
+    if (ip != ctx->end) return 0;
+    ctx->end += inputSize;
+
+    ip++;
+
+    // Main Loop
+    while (ip < mflimit)
+    {
+        ml = LZ4HC_InsertAndFindBestMatch (ctx, ip, matchlimit, (&ref));
+        if (!ml) { ip++; continue; }
+
+        // saved, in case we would skip too much
+        start0 = ip;
+        ref0 = ref;
+        ml0 = ml;
+
+_Search2:
+        if (ip+ml < mflimit)
+            ml2 = LZ4HC_InsertAndGetWiderMatch(ctx, ip + ml - 2, ip + 1, matchlimit, ml, &ref2, &start2);
+        else ml2 = ml;
+
+        if (ml2 == ml)  // No better match
+        {
+            if (LZ4HC_encodeSequence(&ip, &op, &anchor, ml, ref, limit, oend)) return 0;
+            continue;
+        }
+
+        if (start0 < ip)
+        {
+            if (start2 < ip + ml0)   // empirical
+            {
+                ip = start0;
+                ref = ref0;
+                ml = ml0;
+            }
+        }
+
+        // Here, start0==ip
+        if ((start2 - ip) < 3)   // First Match too small : removed
+        {
+            ml = ml2;
+            ip = start2;
+            ref =ref2;
+            goto _Search2;
+        }
+
+_Search3:
+        // Currently we have :
+        // ml2 > ml1, and
+        // ip1+3 <= ip2 (usually < ip1+ml1)
+        if ((start2 - ip) < OPTIMAL_ML)
+        {
+            int correction;
+            int new_ml = ml;
+            if (new_ml > OPTIMAL_ML) new_ml = OPTIMAL_ML;
+            if (ip+new_ml > start2 + ml2 - MINMATCH) new_ml = (int)(start2 - ip) + ml2 - MINMATCH;
+            correction = new_ml - (int)(start2 - ip);
+            if (correction > 0)
+            {
+                start2 += correction;
+                ref2 += correction;
+                ml2 -= correction;
+            }
+        }
+        // Now, we have start2 = ip+new_ml, with new_ml = min(ml, OPTIMAL_ML=18)
+
+        if (start2 + ml2 < mflimit)
+            ml3 = LZ4HC_InsertAndGetWiderMatch(ctx, start2 + ml2 - 3, start2, matchlimit, ml2, &ref3, &start3);
+        else ml3 = ml2;
+
+        if (ml3 == ml2) // No better match : 2 sequences to encode
+        {
+            // ip & ref are known; Now for ml
+            if (start2 < ip+ml)  ml = (int)(start2 - ip);
+            // Now, encode 2 sequences
+            if (LZ4HC_encodeSequence(&ip, &op, &anchor, ml, ref, limit, oend)) return 0;
+            ip = start2;
+            if (LZ4HC_encodeSequence(&ip, &op, &anchor, ml2, ref2, limit, oend)) return 0;
+            continue;
+        }
+
+        if (start3 < ip+ml+3) // Not enough space for match 2 : remove it
+        {
+            if (start3 >= (ip+ml)) // can write Seq1 immediately ==> Seq2 is removed, so Seq3 becomes Seq1
+            {
+                if (start2 < ip+ml)
+                {
+                    int correction = (int)(ip+ml - start2);
+                    start2 += correction;
+                    ref2 += correction;
+                    ml2 -= correction;
+                    if (ml2 < MINMATCH)
+                    {
+                        start2 = start3;
+                        ref2 = ref3;
+                        ml2 = ml3;
+                    }
+                }
+
+                if (LZ4HC_encodeSequence(&ip, &op, &anchor, ml, ref, limit, oend)) return 0;
+                ip  = start3;
+                ref = ref3;
+                ml  = ml3;
+
+                start0 = start2;
+                ref0 = ref2;
+                ml0 = ml2;
+                goto _Search2;
+            }
+
+            start2 = start3;
+            ref2 = ref3;
+            ml2 = ml3;
+            goto _Search3;
+        }
+
+        // OK, now we have 3 ascending matches; let's write at least the first one
+        // ip & ref are known; Now for ml
+        if (start2 < ip+ml)
+        {
+            if ((start2 - ip) < (int)ML_MASK)
+            {
+                int correction;
+                if (ml > OPTIMAL_ML) ml = OPTIMAL_ML;
+                if (ip + ml > start2 + ml2 - MINMATCH) ml = (int)(start2 - ip) + ml2 - MINMATCH;
+                correction = ml - (int)(start2 - ip);
+                if (correction > 0)
+                {
+                    start2 += correction;
+                    ref2 += correction;
+                    ml2 -= correction;
+                }
+            }
+            else
+            {
+                ml = (int)(start2 - ip);
+            }
+        }
+        if (LZ4HC_encodeSequence(&ip, &op, &anchor, ml, ref, limit, oend)) return 0;
+
+        ip = start2;
+        ref = ref2;
+        ml = ml2;
+
+        start2 = start3;
+        ref2 = ref3;
+        ml2 = ml3;
+
+        goto _Search3;
+
+    }
+
+    // Encode Last Literals
+    {
+        int lastRun = (int)(iend - anchor);
+        if ((limit) && (((char*)op - dest) + lastRun + 1 + ((lastRun+255-RUN_MASK)/255) > (U32)maxOutputSize)) return 0;  // Check output limit
+        if (lastRun>=(int)RUN_MASK) { *op++=(RUN_MASK<<ML_BITS); lastRun-=RUN_MASK; for(; lastRun > 254 ; lastRun-=255) *op++ = 255; *op++ = (BYTE) lastRun; } 
+        else *op++ = (BYTE)(lastRun<<ML_BITS);
+        memcpy(op, anchor, iend - anchor);
+        op += iend-anchor;
+    } 
+
+    // End
+    return (int) (((char*)op)-dest);
+}
+
+
+int LZ4_compressHC(const char* source, char* dest, int inputSize)
+{
+    void* ctx = LZ4_createHC(source);
+    int result;
+    if (ctx==NULL) return 0;
+
+    result = LZ4HC_compress_generic (ctx, source, dest, inputSize, 0, noLimit);
+
+    LZ4_freeHC(ctx);
+    return result;
+}
+
+int LZ4_compressHC_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize)
+{
+    return LZ4HC_compress_generic (LZ4HC_Data, source, dest, inputSize, 0, noLimit);
+}
+
+
+int LZ4_compressHC_limitedOutput(const char* source, char* dest, int inputSize, int maxOutputSize)
+{
+    void* ctx = LZ4_createHC(source);
+    int result;
+    if (ctx==NULL) return 0;
+
+    result = LZ4HC_compress_generic (ctx, source, dest, inputSize, maxOutputSize, limitedOutput);
+
+    LZ4_freeHC(ctx);
+    return result;
+}
+
+int LZ4_compressHC_limitedOutput_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize, int maxOutputSize)
+{
+    return LZ4HC_compress_generic (LZ4HC_Data, source, dest, inputSize, maxOutputSize, limitedOutput);
+}
+

Source/ThirdParty/LZ4/lz4hc.h

@@ -0,0 +1,111 @@
+/*
+   LZ4 HC - High Compression Mode of LZ4
+   Header File
+   Copyright (C) 2011-2013, 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 :
+   - LZ4 homepage : http://fastcompression.blogspot.com/p/lz4.html
+   - LZ4 source repository : http://code.google.com/p/lz4/
+*/
+#pragma once
+
+
+#if defined (__cplusplus)
+extern "C" {
+#endif
+
+
+int LZ4_compressHC (const char* source, char* dest, int inputSize);
+/*
+LZ4_compressHC :
+    return : the number of bytes in compressed buffer dest
+             or 0 if compression fails.
+    note : destination buffer must be already allocated. 
+        To avoid any problem, size it to handle worst cases situations (input data not compressible)
+        Worst case size evaluation is provided by function LZ4_compressBound() (see "lz4.h")
+*/
+
+int LZ4_compressHC_limitedOutput (const char* source, char* dest, int inputSize, int maxOutputSize);
+/*
+LZ4_compress_limitedOutput() :
+    Compress 'inputSize' bytes from 'source' into an output buffer 'dest' of maximum size 'maxOutputSize'.
+    If it cannot achieve it, compression will stop, and result of the function will be zero.
+    This function never writes outside of provided output buffer.
+
+    inputSize  : Max supported value is 1 GB
+    maxOutputSize : is maximum allowed size into the destination buffer (which must be already allocated)
+    return : the number of output bytes written in buffer 'dest'
+             or 0 if compression fails.
+*/
+
+
+/* Note :
+Decompression functions are provided within LZ4 source code (see "lz4.h") (BSD license)
+*/
+
+
+/* Advanced Functions */
+
+void* LZ4_createHC (const char* inputBuffer);
+int   LZ4_compressHC_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize);
+int   LZ4_compressHC_limitedOutput_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize, int maxOutputSize);
+char* LZ4_slideInputBufferHC (void* LZ4HC_Data);
+int   LZ4_freeHC (void* LZ4HC_Data);
+
+/* 
+These functions allow the compression of dependent blocks, where each block benefits from prior 64 KB within preceding blocks.
+In order to achieve this, it is necessary to start creating the LZ4HC Data Structure, thanks to the function :
+
+void* LZ4_createHC (const char* inputBuffer);
+The result of the function is the (void*) pointer on the LZ4HC Data Structure.
+This pointer will be needed in all other functions.
+If the pointer returned is NULL, then the allocation has failed, and compression must be aborted.
+The only parameter 'const char* inputBuffer' must, obviously, point at the beginning of input buffer.
+The input buffer must be already allocated, and size at least 192KB.
+'inputBuffer' will also be the 'const char* source' of the first block.
+
+All blocks are expected to lay next to each other within the input buffer, starting from 'inputBuffer'.
+To compress each block, use either LZ4_compressHC_continue() or LZ4_compressHC_limitedOutput_continue().
+Their behavior are identical to LZ4_compressHC() or LZ4_compressHC_limitedOutput(), 
+but require the LZ4HC Data Structure as their first argument, and check that each block starts right after the previous one.
+If next block does not begin immediately after the previous one, the compression will fail (return 0).
+
+When it's no longer possible to lay the next block after the previous one (not enough space left into input buffer), a call to : 
+char* LZ4_slideInputBufferHC(void* LZ4HC_Data);
+must be performed. It will typically copy the latest 64KB of input at the beginning of input buffer.
+Note that, for this function to work properly, minimum size of an input buffer must be 192KB.
+==> The memory position where the next input data block must start is provided as the result of the function.
+
+Compression can then resume, using LZ4_compressHC_continue() or LZ4_compressHC_limitedOutput_continue(), as usual.
+
+When compression is completed, a call to LZ4_freeHC() will release the memory used by the LZ4HC Data Structure.
+*/
+
+
+#if defined (__cplusplus)
+}
+#endif

Source/Tools/PackageTool/CMakeLists.txt

@@ -27,7 +27,7 @@ set (TARGET_NAME PackageTool)
 set (SOURCE_FILES PackageTool.cpp)
 
 # Define dependency libs
-set (LIBS ../../Engine/Container ../../Engine/Core ../../Engine/IO ../../Engine/Math ../../Engine/Resource)
+set (LIBS ../../Engine/Container ../../Engine/Core ../../Engine/IO ../../Engine/Math ../../Engine/Resource ../../ThirdParty/LZ4)
 
 # Setup target
 setup_executable ()

Source/Tools/PackageTool/PackageTool.cpp

@@ -32,11 +32,15 @@
 
 #include <cstdio>
 #include <cstring>
+#include <lz4.h>
+#include <lz4hc.h>
 
 #include "DebugNew.h"
 
 using namespace Urho3D;
 
+static const unsigned COMPRESSED_BLOCK_SIZE = 32768;
+
 struct FileEntry
 {
     String name_;
@@ -50,6 +54,8 @@ SharedPtr<FileSystem> fileSystem_(new FileSystem(context_));
 String basePath_;
 Vector<FileEntry> entries_;
 unsigned checksum_ = 0;
+bool compress_ = false;
+unsigned blockSize_ = COMPRESSED_BLOCK_SIZE;
 
 String ignoreExtensions_[] = {
     ".bak",
@@ -61,6 +67,7 @@ int main(int argc, char** argv);
 void Run(const Vector<String>& arguments);
 void ProcessFile(const String& fileName, const String& rootDir);
 void WritePackageFile(const String& fileName, const String& rootDir);
+void WriteHeader(File& dest);
 
 int main(int argc, char** argv)
 {
@@ -79,12 +86,30 @@ int main(int argc, char** argv)
 void Run(const Vector<String>& arguments)
 {
     if (arguments.Size() < 2)
-        ErrorExit("Usage: PackageTool <directory to process> <package name> [basepath]\n");
+        ErrorExit("Usage: PackageTool <directory to process> <package name> [basepath] [-c]\n");
     
     const String& dirName = arguments[0];
     const String& packageName = arguments[1];
     if (arguments.Size() > 2)
-        basePath_ = AddTrailingSlash(arguments[2]);
+    {
+        for (unsigned i = 2; i < arguments.Size(); ++i)
+        {
+            if (arguments[i][0] != '-')
+                basePath_ = AddTrailingSlash(arguments[i]);
+            else
+            {
+                if (arguments[i].Length() > 1)
+                {
+                    switch (arguments[i][1])
+                    {
+                    case 'c':
+                        compress_ = true;
+                        break;
+                    }
+                }
+            }
+        }
+    }
     
     PrintLine("Scanning directory " + dirName + " for files");
     
@@ -140,9 +165,7 @@ void WritePackageFile(const String& fileName, const String& rootDir)
         ErrorExit("Could not open output file " + fileName);
     
     // Write ID, number of files & placeholder for checksum
-    dest.WriteFileID("UPAK");
-    dest.WriteUInt(entries_.Size());
-    dest.WriteUInt(checksum_);
+    WriteHeader(dest);
     
     for (unsigned i = 0; i < entries_.Size(); ++i)
     {
@@ -156,8 +179,6 @@ void WritePackageFile(const String& fileName, const String& rootDir)
     // Write file data, calculate checksums & correct offsets
     for (unsigned i = 0; i < entries_.Size(); ++i)
     {
-        PrintLine("Writing file " + entries_[i].name_);
-        
         entries_[i].offset_ = dest.GetSize();
         String fileFullPath = rootDir + "/" + entries_[i].name_;
         
@@ -178,14 +199,43 @@ void WritePackageFile(const String& fileName, const String& rootDir)
             entries_[i].checksum_ = SDBMHash(entries_[i].checksum_, buffer[j]);
         }
         
-        dest.Write(&buffer[0], entries_[i].size_);
+        if (!compress_)
+        {
+            PrintLine(entries_[i].name_ + " size " + String(dataSize));
+            dest.Write(&buffer[0], entries_[i].size_);
+        }
+        else
+        {
+            SharedArrayPtr<unsigned char> compressBuffer(new unsigned char[LZ4_compressBound(blockSize_)]);
+            
+            unsigned pos = 0;
+            unsigned totalPackedBytes = 0;
+            
+            while (pos < dataSize)
+            {
+                unsigned unpackedSize = blockSize_;
+                if (pos + unpackedSize > dataSize)
+                    unpackedSize = dataSize - pos;
+                
+                unsigned packedSize = LZ4_compressHC((const char*)&buffer[pos], (char*)compressBuffer.Get(), unpackedSize);
+                if (!packedSize)
+                    ErrorExit("LZ4 compression failed for file " + entries_[i].name_ + " at offset " + pos);
+                
+                dest.WriteUShort(unpackedSize);
+                dest.WriteUShort(packedSize);
+                dest.Write(compressBuffer.Get(), packedSize);
+                totalPackedBytes += 6 + packedSize;
+                
+                pos += unpackedSize;
+            }
+            
+            PrintLine(entries_[i].name_ + " size " + String(dataSize) + " packed " + String(totalPackedBytes));
+        }
     }
     
     // Write header again with correct offsets & checksums
     dest.Seek(0);
-    dest.WriteFileID("UPAK");
-    dest.WriteUInt(entries_.Size());
-    dest.WriteUInt(checksum_);
+    WriteHeader(dest);
     
     for (unsigned i = 0; i < entries_.Size(); ++i)
     {
@@ -195,5 +245,17 @@ void WritePackageFile(const String& fileName, const String& rootDir)
         dest.WriteUInt(entries_[i].checksum_);
     }
     
-    PrintLine("Package total size " + String(dest.GetSize()) + " bytes");
+    PrintLine("Package total size " + String(dest.GetSize()));
+}
+
+void WriteHeader(File& dest)
+{
+    if (!compress_)
+        dest.WriteFileID("UPAK");
+    else
+        dest.WriteFileID("ULZ4");
+    dest.WriteUInt(entries_.Size());
+    dest.WriteUInt(checksum_);
+    if (compress_)
+        dest.WriteUInt(blockSize_);
 }