sereal 0.0.3 → 0.0.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 802e13eb6a0c49314803629027fd015345366f57
-  data.tar.gz: cbf309f9f930d9943e39851b48f5292832495e66
+  metadata.gz: b17ebc8e9c34db04bd5e21e712925a91c6e03aa6
+  data.tar.gz: f0984c679f1c7a7053f961d1d211ce5160c76375
 SHA512:
-  metadata.gz: 753e34f2e15086c4c4ec707ec8bfda05df122c237ac2bd18b992132a470a9475039753c9383cb0ffcaf43289802a5c9e513c0f4ed5026165ea38f59bb62a19d5
-  data.tar.gz: ba290f161d4c651e5b0a783fff75b22d8c356e8019f9a706ff8a4a1a7ec8797fe8b53ed285b5c306289c1d8ba6ac9d575b5b881d1b6e0bf19371a164f030b296
+  metadata.gz: d7d044c11e0c06e24d3d5ab7fe131761bc8031a85030e23bd693e633d2e8ab842e52d4f76665314b1ae155eafa812e74abf26c13d91582ed24c939c476342fc9
+  data.tar.gz: 71ab51d02f317100e0d9fe613227e4e1d26856d6b1c0374ff8af223867d8ceefae71712be8f4d04ce9dc7b3023e7809850511a85b26a61b802f1a13aa02a3996

data/ext/sereal/buffer.h

@@ -16,7 +16,7 @@ static inline sereal_t * s_create(void) {
 }
 
 static inline void s_destroy(sereal_t *s) {
-        if (s->data)
+        if (s->data && (s->flags & FLAG_NOT_MINE) == 0)
                 free(s->data);
 
         free(s);
@@ -76,14 +76,21 @@ static inline u32 s_shift_position_bang(sereal_t *s, u32 len) {
         s->pos += len;
         return len;
 }
-static inline void s_dump(sereal_t *s) {
-        u32 i;
-        for (i = 0; i < s->size; i++) {
-                if (i == s->pos) 
-                        fprintf(stderr," [%c %d] ",s->data[i],s->data[i]);
+
+static void b_dump(u8 *p, u32 len, u32 pos) {
+        int i;
+
+        fprintf(stderr,"\n-----------\n");
+        for (i = 0; i < len; i++) {
+                if (i == pos) 
+                        fprintf(stderr," [%c %d] ",p[i],p[i]);
                 else
-                        fprintf(stderr," (%c %d) ",s->data[i],s->data[i]);
+                        fprintf(stderr," (%c %d) ",p[i],p[i]);
         }
-        fprintf(stderr,"\n");
+        fprintf(stderr,"\n-----------\n");
+}
+
+static void s_dump(sereal_t *s) {
+        b_dump(s->data,s->size,s->pos);
 }
 

data/ext/sereal/decode.c

@@ -5,7 +5,7 @@
 
 static VALUE s_default_reader(sereal_t *s, u8 tag) {
         // s_dump(s);
-        rb_raise(rb_eTypeError,"unsupported tag %d",tag);
+        s_raise(s,rb_eTypeError,"unsupported tag %d",tag);
         return Qnil;
 }
 
@@ -16,8 +16,8 @@ static u64 s_get_varint_bang(sereal_t *s) {
         while (s_get_u8(s) & 0x80) {
                 uv |= ((u64)(s_get_u8_bang(s) & 0x7F) << lshift);
                 lshift += 7;
-                if (lshift > (sizeof(uv) * 8)) 
-                        rb_raise(rb_eTypeError, "varint too big");
+                if (lshift > (sizeof(uv) * 8))
+                        s_raise(s,rb_eTypeError, "varint too big");
         }
         uv |= ((u64)(s_get_u8_bang(s)) << lshift);
         return uv;
@@ -128,7 +128,7 @@ static VALUE s_read_rb_string_bang(sereal_t *s,u8 t) {
                               rb_str_new(s_get_p(s), len));
         }
         #undef RETURN_STRING
-        rb_raise(rb_eTypeError, "undefined string type %d",t);
+        s_raise(s,rb_eTypeError, "undefined string type %d",t);
 }
 
 static VALUE s_read_next_rb_string_bang(sereal_t *s) {
@@ -172,7 +172,7 @@ static VALUE s_read_pad(sereal_t *s, u8 tag) {
         return sereal_to_rb_object(s);
 }
 static VALUE s_read_extend(sereal_t *s, u8 tag) {
-        rb_raise(rb_eArgError,"extend tags are not supported");
+        s_raise(s,rb_eArgError,"extend tags are not supported");
 }
 
 VALUE sereal_to_rb_object(sereal_t *s) {
@@ -181,7 +181,7 @@ VALUE sereal_to_rb_object(sereal_t *s) {
         while (s->pos < s->size) {
                 t = s_get_u8_bang(s);
                 if (t & SRL_HDR_TRACK_FLAG)
-                        rb_raise(rb_eArgError, "trackable objects are not supported");
+                        s_raise(s,rb_eArgError, "trackable objects are not supported");
                 S_RECURSE_DEC(s);
                 return (*READERS[t])(s,t);
         }
@@ -193,46 +193,90 @@ VALUE method_sereal_decode(VALUE self, VALUE args) {
         if (argc < 1)
                 rb_raise(rb_eArgError,"need at least 1 argument (object)");
         VALUE payload = rb_ary_shift(args);
-        VALUE safe = Qtrue;
         if (TYPE(payload) != T_STRING) 
                 rb_raise(rb_eTypeError,"can not decode objects of type %s",rb_obj_classname(payload));
 
         sereal_t *s = s_create();
-        s->data = RSTRING_PTR(payload);
-        s->size = RSTRING_LEN(payload);
-
-        if (s->size < 6 || s_get_u32_bang(s) != 0x6c72733d) 
-                rb_raise(rb_eTypeError,"invalid header"); 
+        u64 offset = 0;
+again: 
+        if (offset >= RSTRING_LEN(payload)) {
+            free(s);
+            return Qnil;
+        }
+        s->flags |= FLAG_NOT_MINE;
+        s->data = RSTRING_PTR(payload) + offset;
+        s->size = RSTRING_LEN(payload) - offset;
+        s->pos = 0;
+        if (s->size < 6 || s_get_u32_bang(s) != SRL_MAGIC_STRING_LILIPUTIAN) 
+                s_raise(s,rb_eTypeError,"invalid header"); 
 
         u8 version = s_get_u8_bang(s);
         u8 suffix = s_get_varint_bang(s);
-        int is_compressed = (version & SRL_PROTOCOL_ENCODING_MASK) == SRL_PROTOCOL_ENCODING_SNAPPY ? TRUE : FALSE;
+        u8 is_compressed;
 
+        if ((version & SRL_PROTOCOL_ENCODING_MASK) == SRL_PROTOCOL_ENCODING_SNAPPY)
+                is_compressed = __SNAPPY;
+        else if ((version & SRL_PROTOCOL_ENCODING_MASK) == SRL_PROTOCOL_ENCODING_SNAPPY_INCR)
+                is_compressed = __SNAPPY_INCR;
+        else if ((version & SRL_PROTOCOL_ENCODING_MASK) == SRL_PROTOCOL_ENCODING_LZ4_INCR)
+                is_compressed = __LZ4_INCR;
+        else
+                is_compressed = __RAW;
         if (is_compressed) {
                 u32 uncompressed_len;
-                int snappy_header_len = csnappy_get_uncompressed_length(s_get_p(s),
-                                                                        (s->size - s->pos),
-                                                                        &uncompressed_len);
-                if (snappy_header_len == CSNAPPY_E_HEADER_BAD) 
-                        rb_raise(rb_eTypeError,"invalid snappy header");
+                u32 compressed_len;
+
+                if (is_compressed == __LZ4_INCR) {
+                        uncompressed_len = s_get_varint_bang(s);
+                        if (!uncompressed_len)
+                                rb_raise(rb_eTypeError,"LZ4 compression requires <varint uncompressed size><varint compressed size><blob>, unable to get uncompressed_len");
+
+                        compressed_len = s_get_varint_bang(s);
+                } else {
+                        if (is_compressed == __SNAPPY_INCR) {
+                                compressed_len = s_get_varint_bang(s);
+                        } else {
+                                compressed_len = s->size - s->pos;
+                        }
 
+                        int snappy_header_len = csnappy_get_uncompressed_length(s_get_p(s),
+                                                                                compressed_len,
+                                                                                &uncompressed_len);
+                        if (snappy_header_len == CSNAPPY_E_HEADER_BAD) 
+                                rb_raise(rb_eTypeError,"invalid snappy header");
+                }
                 u8 *uncompressed = alloc_or_raise(uncompressed_len);
-                int done = csnappy_decompress(s_get_p(s),
-                                              (s->size - s->pos),
+                int done;
+                if (is_compressed == __LZ4_INCR) {
+                        done = LZ4_decompress_safe(s_get_p(s),
+                                                   uncompressed,
+                                                   compressed_len,
+                                                   uncompressed_len) == uncompressed_len ? 1 : 0;
+                } else {
+                        done = csnappy_decompress(s_get_p(s),
+                                              compressed_len,
                                               uncompressed,
-                                              uncompressed_len);
-                if (done)
-                        rb_raise(rb_eTypeError, "snappy decompression failed error: %d",done);
+                                              uncompressed_len) == CSNAPPY_E_OK ? 1 : 0;
+                }
+                if (!done)
+                        s_raise(s,rb_eTypeError, "decompression failed error: %d type: %d, unompressed size: %d compressed size: %d",done,is_compressed,uncompressed_len,compressed_len);
+
+                offset += s->pos + compressed_len;
                 s->data = uncompressed;
                 s->size = uncompressed_len;
                 s->pos = 0;
+                s->flags &= ~FLAG_NOT_MINE;
         }
 
-        VALUE result = sereal_to_rb_object(s); 
-        if (is_compressed)
-                s_destroy(s);
-        else
-                free(s); // we do not destroy because it will free s->data which is RSTRING_PTR(payload) 
+        VALUE result = sereal_to_rb_object(s);
+        if (is_compressed && rb_block_given_p()) {
+                free(s->data);
+                s->data = NULL;
+                rb_yield(result);
+                goto again;
+        }
+        s_destroy(s);
+
         return result;
 }
 

data/ext/sereal/encode.c

@@ -2,7 +2,8 @@
 #include "buffer.h"
 #include "encode.h"
 #include "snappy/csnappy_compress.c"
-
+#include "lz4/lz4.h"
+#include "lz4/lz4hc.h"
 #define W_SIZE 32
 #if T_FIXNUM > W_SIZE
 	#define W_SIZE T_FIXNUM
@@ -70,7 +71,7 @@ void s_init_writers(void) {
 }
 
 static void s_default_writer(sereal_t *s, VALUE object) {
-        rb_raise(rb_eTypeError, "invalid type for input %s",rb_obj_classname(object));
+        s_raise(s,rb_eTypeError, "invalid type for input %s",rb_obj_classname(object));
 }
 
 
@@ -222,46 +223,134 @@ static void rb_object_to_sereal(sereal_t *s, VALUE object) {
         S_RECURSE_DEC(s);
 }
 
+// https://github.com/Sereal/Sereal/blob/master/Perl/Encoder/srl_encoder.c#L623
+void fixup_varint_from_to(u8 *varint_start, u8 *varint_end, u32 number) {
+    while (number >= 0x80) {
+        *varint_start++ = (number & 0x7f) | 0x80;
+        number = number >> 7;
+    }
+    if ( varint_start == varint_end ) {
+        *varint_start = number;
+    } else {
+        *varint_start++ = (number & 0x7f) | 0x80;
+        while ( varint_start < varint_end )
+            *varint_start++ = 0x80;
+        *varint_start= 0;
+    }
+}
+
 VALUE method_sereal_encode(VALUE self, VALUE args) {
-        sereal_t *s = s_create();
         u32 argc = RARRAY_LEN(args);
         if (argc < 1)
                 rb_raise(rb_eArgError,"need at least 1 argument (object)");
+
+        sereal_t *s = s_create();
         VALUE payload = rb_ary_shift(args);
         VALUE compress = Qfalse;
         if (argc == 2) 
                 compress = rb_ary_shift(args);
-        int do_compress = (compress == Qtrue ? TRUE : FALSE);
+        u8 do_compress;
+        u8 version = SRL_PROTOCOL_VERSION;
 
+        if (TYPE(compress) == T_FIXNUM) {
+            do_compress = (u8) FIX2LONG(compress);
+        } else {
+            do_compress = (compress == Qtrue ? 1 : 0);
+        }
+
+        switch(do_compress) {
+            case __SNAPPY:
+                version |= SRL_PROTOCOL_ENCODING_SNAPPY;
+                break;
+            case __SNAPPY_INCR:
+                version |= SRL_PROTOCOL_ENCODING_SNAPPY_INCR;
+                break;
+            case __LZ4HC_INCR:
+            case __LZ4_INCR:
+                version |= SRL_PROTOCOL_ENCODING_LZ4_INCR;
+                break;
+            case __RAW:
+            default:
+                version |= SRL_PROTOCOL_ENCODING_RAW;
+        }
         // setup header
         s_append_u32(s,SRL_MAGIC_STRING_LILIPUTIAN);
-        s_append_u8(s,SRL_PROTOCOL_VERSION | (do_compress ? SRL_PROTOCOL_ENCODING_SNAPPY : SRL_PROTOCOL_ENCODING_RAW));
+        s_append_u8(s,version);
         s_append_u8(s,0x0);
         u32 s_header_len = s->pos;
         
         // serialize
         rb_object_to_sereal(s,payload);
-
         // compress
         if (do_compress) {
+                u8 *start_compressed_varint = NULL, *start_un_compressed_varint= NULL, *compressed, *end;
+                u32 compressed_len = 0;
+                u32 compressed_len_varint = 0;
+                u32 un_compressed_len_varint = 0;
                 u32 s_body_len = s->size - s_header_len;
-                u32 compressed_len = csnappy_max_compressed_length(s_body_len);
+                // snappy <compressed blob>
+                // snappy incr <varint blob len><compressed blob>
+                // lz4 <varint uncompressed len><varint blob len><compressed blob>
+                if (do_compress == __SNAPPY || do_compress == __SNAPPY_INCR) {
+                        compressed_len = csnappy_max_compressed_length(s_body_len);
+                } else {
+                        compressed_len = LZ4_compressBound(s_body_len);
+                        start_un_compressed_varint = s_get_p(s);
+                        s_append_varint(s,s_body_len);
+                        end = s_get_p(s);
+                        un_compressed_len_varint = end - start_un_compressed_varint;
+                }
 
-                u8 *working_buf = alloc_or_raise(CSNAPPY_WORKMEM_BYTES);
-                u8 *compressed = alloc_or_raise(compressed_len + s_header_len);
+                if (do_compress == __SNAPPY_INCR || do_compress == __LZ4_INCR || do_compress == __LZ4HC_INCR) {
+                        start_compressed_varint = s_get_p(s);
+                        s_append_varint(s,compressed_len);
+                        end = s_get_p(s);
+                        compressed_len_varint = end - start_compressed_varint;
+                }
 
+                compressed = alloc_or_raise(compressed_len + s_header_len + compressed_len_varint + un_compressed_len_varint);
                 COPY(s_get_p_at_pos(s,0,0),compressed,s_header_len);
-                
-                csnappy_compress(s_get_p_at_pos(s,s_header_len,1), 
-                                 s_body_len,
-                                 (compressed + s_header_len),
-                                 &compressed_len,
-                                 working_buf, 
-                                 CSNAPPY_WORKMEM_BYTES_POWER_OF_TWO);
+
+                if (start_un_compressed_varint)
+                        COPY(start_un_compressed_varint,
+                            compressed + s_header_len,
+                            un_compressed_len_varint);
+                if (start_compressed_varint)
+                        COPY(start_compressed_varint,
+                             compressed + s_header_len + un_compressed_len_varint,
+                             compressed_len_varint);
+
+                u8 *start = s_get_p_at_pos(s,s_header_len,1);
+                if (do_compress == __LZ4_INCR) {
+                        compressed_len = LZ4_compress(start, 
+                                     (compressed + s_header_len + compressed_len_varint + un_compressed_len_varint),
+                                     s_body_len);
+                } else if (do_compress == __LZ4HC_INCR) {
+                        compressed_len = LZ4_compressHC(start, 
+                                     (compressed + s_header_len + compressed_len_varint + un_compressed_len_varint),
+                                     s_body_len);
+
+                } else {
+                        u8 *working_buf = alloc_or_raise(CSNAPPY_WORKMEM_BYTES);
+                        csnappy_compress(start, 
+                                         s_body_len,
+                                         (compressed + s_header_len + compressed_len_varint + un_compressed_len_varint),
+                                         &compressed_len,
+                                         working_buf, 
+                                         CSNAPPY_WORKMEM_BYTES_POWER_OF_TWO);
+
+                        free(working_buf);
+                }
+                if (compressed_len == 0)
+                    s_raise(s,rb_eTypeError,"failed to compress");
+                if (start_compressed_varint)
+                        fixup_varint_from_to(compressed + s_header_len + un_compressed_len_varint, 
+                                             compressed + s_header_len + un_compressed_len_varint + compressed_len_varint - 1,
+                                             compressed_len);
                 free(s->data);
-                free(working_buf);
                 s->data = compressed;
-                s->size = compressed_len + s_header_len;
+                s->size = compressed_len + s_header_len + compressed_len_varint + un_compressed_len_varint;
+                s->pos = s->size;
         }
 
         VALUE result = rb_str_new(s->data,s->size);

data/ext/sereal/lz4.c

@@ -0,0 +1,822 @@
+/*
+   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 ((U32)inputSize > (U32)LZ4_MAX_INPUT_SIZE) return 0;                                // Unsupported input size, too large (or negative)
+    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)(lz4ds->nextBlock - lz4ds->base) > 0xE0000000) )   // close to 32-bits limit
+    {
+        size_t deltaLimit = (lz4ds->nextBlock - 64 KB) - lz4ds->base;
+        int nH;
+
+        for (nH=0; nH < HASHNBCELLS4; nH++)
+        {
+            if ((size_t)(lz4ds->hashTable[nH]) < deltaLimit) lz4ds->hashTable[nH] = 0;
+            else lz4ds->hashTable[nH] -= (U32)deltaLimit;
+        }
+        memcpy((void*)(lz4ds->bufferStart), (const void*)(lz4ds->nextBlock - 64 KB), 64 KB);
+        lz4ds->base = lz4ds->bufferStart;
+        lz4ds->nextBlock = lz4ds->base + 64 KB;
+    }
+    else
+    {
+		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
+}
+