zstd-ruby 1.4.4.0 → 1.5.1.0

data/ext/zstdruby/libzstd/dictBuilder/fastcover.c

@@ -1,3 +1,13 @@
+/*
+ * Copyright (c) Facebook, Inc.
+ * All rights reserved.
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+ */
+
 /*-*************************************
 *  Dependencies
 ***************************************/
@@ -6,24 +16,33 @@
 #include <string.h> /* memset */
 #include <time.h>   /* clock */
 
-#include "mem.h" /* read */
-#include "pool.h"
-#include "threading.h"
-#include "cover.h"
-#include "zstd_internal.h" /* includes zstd.h */
 #ifndef ZDICT_STATIC_LINKING_ONLY
-#define ZDICT_STATIC_LINKING_ONLY
+#  define ZDICT_STATIC_LINKING_ONLY
 #endif
-#include "zdict.h"
+
+#include "../common/mem.h" /* read */
+#include "../common/pool.h"
+#include "../common/threading.h"
+#include "../common/zstd_internal.h" /* includes zstd.h */
+#include "../compress/zstd_compress_internal.h" /* ZSTD_hash*() */
+#include "../zdict.h"
+#include "cover.h"
 
 
 /*-*************************************
 *  Constants
 ***************************************/
+/**
+* There are 32bit indexes used to ref samples, so limit samples size to 4GB
+* on 64bit builds.
+* For 32bit builds we choose 1 GB.
+* Most 32bit platforms have 2GB user-mode addressable space and we allocate a large
+* contiguous buffer, so 1GB is already a high limit.
+*/
 #define FASTCOVER_MAX_SAMPLES_SIZE (sizeof(size_t) == 8 ? ((unsigned)-1) : ((unsigned)1 GB))
 #define FASTCOVER_MAX_F 31
 #define FASTCOVER_MAX_ACCEL 10
-#define DEFAULT_SPLITPOINT 0.75
+#define FASTCOVER_DEFAULT_SPLITPOINT 0.75
 #define DEFAULT_F 20
 #define DEFAULT_ACCEL 1
 
@@ -31,50 +50,50 @@
 /*-*************************************
 *  Console display
 ***************************************/
-static int g_displayLevel = 2;
+#ifndef LOCALDISPLAYLEVEL
+static int g_displayLevel = 0;
+#endif
+#undef  DISPLAY
 #define DISPLAY(...)                                                           \
   {                                                                            \
     fprintf(stderr, __VA_ARGS__);                                              \
     fflush(stderr);                                                            \
   }
+#undef  LOCALDISPLAYLEVEL
 #define LOCALDISPLAYLEVEL(displayLevel, l, ...)                                \
   if (displayLevel >= l) {                                                     \
     DISPLAY(__VA_ARGS__);                                                      \
   } /* 0 : no display;   1: errors;   2: default;  3: details;  4: debug */
+#undef  DISPLAYLEVEL
 #define DISPLAYLEVEL(l, ...) LOCALDISPLAYLEVEL(g_displayLevel, l, __VA_ARGS__)
 
+#ifndef LOCALDISPLAYUPDATE
+static const clock_t g_refreshRate = CLOCKS_PER_SEC * 15 / 100;
+static clock_t g_time = 0;
+#endif
+#undef  LOCALDISPLAYUPDATE
 #define LOCALDISPLAYUPDATE(displayLevel, l, ...)                               \
   if (displayLevel >= l) {                                                     \
-    if ((clock() - g_time > refreshRate) || (displayLevel >= 4)) {             \
+    if ((clock() - g_time > g_refreshRate) || (displayLevel >= 4)) {             \
       g_time = clock();                                                        \
       DISPLAY(__VA_ARGS__);                                                    \
     }                                                                          \
   }
+#undef  DISPLAYUPDATE
 #define DISPLAYUPDATE(l, ...) LOCALDISPLAYUPDATE(g_displayLevel, l, __VA_ARGS__)
-static const clock_t refreshRate = CLOCKS_PER_SEC * 15 / 100;
-static clock_t g_time = 0;
 
 
 /*-*************************************
 * Hash Functions
 ***************************************/
-static const U64 prime6bytes = 227718039650203ULL;
-static size_t ZSTD_hash6(U64 u, U32 h) { return (size_t)(((u  << (64-48)) * prime6bytes) >> (64-h)) ; }
-static size_t ZSTD_hash6Ptr(const void* p, U32 h) { return ZSTD_hash6(MEM_readLE64(p), h); }
-
-static const U64 prime8bytes = 0xCF1BBCDCB7A56463ULL;
-static size_t ZSTD_hash8(U64 u, U32 h) { return (size_t)(((u) * prime8bytes) >> (64-h)) ; }
-static size_t ZSTD_hash8Ptr(const void* p, U32 h) { return ZSTD_hash8(MEM_readLE64(p), h); }
-
-
 /**
- * Hash the d-byte value pointed to by p and mod 2^f
+ * Hash the d-byte value pointed to by p and mod 2^f into the frequency vector
  */
-static size_t FASTCOVER_hashPtrToIndex(const void* p, U32 h, unsigned d) {
+static size_t FASTCOVER_hashPtrToIndex(const void* p, U32 f, unsigned d) {
   if (d == 6) {
-    return ZSTD_hash6Ptr(p, h) & ((1 << h) - 1);
+    return ZSTD_hash6Ptr(p, f);
   }
-  return ZSTD_hash8Ptr(p, h) & ((1 << h) - 1);
+  return ZSTD_hash8Ptr(p, f);
 }
 
 
@@ -451,20 +470,20 @@ typedef struct FASTCOVER_tryParameters_data_s {
  * This function is thread safe if zstd is compiled with multithreaded support.
  * It takes its parameters as an *OWNING* opaque pointer to support threading.
  */
-static void FASTCOVER_tryParameters(void *opaque)
+static void FASTCOVER_tryParameters(void* opaque)
 {
   /* Save parameters as local variables */
-  FASTCOVER_tryParameters_data_t *const data = (FASTCOVER_tryParameters_data_t *)opaque;
+  FASTCOVER_tryParameters_data_t *const data = (FASTCOVER_tryParameters_data_t*)opaque;
   const FASTCOVER_ctx_t *const ctx = data->ctx;
   const ZDICT_cover_params_t parameters = data->parameters;
   size_t dictBufferCapacity = data->dictBufferCapacity;
   size_t totalCompressedSize = ERROR(GENERIC);
   /* Initialize array to keep track of frequency of dmer within activeSegment */
-  U16* segmentFreqs = (U16 *)calloc(((U64)1 << ctx->f), sizeof(U16));
+  U16* segmentFreqs = (U16*)calloc(((U64)1 << ctx->f), sizeof(U16));
   /* Allocate space for hash table, dict, and freqs */
-  BYTE *const dict = (BYTE * const)malloc(dictBufferCapacity);
+  BYTE *const dict = (BYTE*)malloc(dictBufferCapacity);
   COVER_dictSelection_t selection = COVER_dictSelectionError(ERROR(GENERIC));
-  U32 *freqs = (U32*) malloc(((U64)1 << ctx->f) * sizeof(U32));
+  U32* freqs = (U32*) malloc(((U64)1 << ctx->f) * sizeof(U32));
   if (!segmentFreqs || !dict || !freqs) {
     DISPLAYLEVEL(1, "Failed to allocate buffers: out of memory\n");
     goto _cleanup;
@@ -476,7 +495,7 @@ static void FASTCOVER_tryParameters(void *opaque)
                                                     parameters, segmentFreqs);
 
     const unsigned nbFinalizeSamples = (unsigned)(ctx->nbTrainSamples * ctx->accelParams.finalize / 100);
-    selection = COVER_selectDict(dict + tail, dictBufferCapacity - tail,
+    selection = COVER_selectDict(dict + tail, dictBufferCapacity, dictBufferCapacity - tail,
          ctx->samples, ctx->samplesSizes, nbFinalizeSamples, ctx->nbTrainSamples, ctx->nbSamples, parameters, ctx->offsets,
          totalCompressedSize);
 
@@ -537,7 +556,7 @@ ZDICT_trainFromBuffer_fastCover(void* dictBuffer, size_t dictBufferCapacity,
     ZDICT_cover_params_t coverParams;
     FASTCOVER_accel_t accelParams;
     /* Initialize global data */
-    g_displayLevel = parameters.zParams.notificationLevel;
+    g_displayLevel = (int)parameters.zParams.notificationLevel;
     /* Assign splitPoint and f if not provided */
     parameters.splitPoint = 1.0;
     parameters.f = parameters.f == 0 ? DEFAULT_F : parameters.f;
@@ -607,7 +626,7 @@ ZDICT_optimizeTrainFromBuffer_fastCover(
     /* constants */
     const unsigned nbThreads = parameters->nbThreads;
     const double splitPoint =
-        parameters->splitPoint <= 0.0 ? DEFAULT_SPLITPOINT : parameters->splitPoint;
+        parameters->splitPoint <= 0.0 ? FASTCOVER_DEFAULT_SPLITPOINT : parameters->splitPoint;
     const unsigned kMinD = parameters->d == 0 ? 6 : parameters->d;
     const unsigned kMaxD = parameters->d == 0 ? 8 : parameters->d;
     const unsigned kMinK = parameters->k == 0 ? 50 : parameters->k;
@@ -620,7 +639,7 @@ ZDICT_optimizeTrainFromBuffer_fastCover(
     const unsigned accel = parameters->accel == 0 ? DEFAULT_ACCEL : parameters->accel;
     const unsigned shrinkDict = 0;
     /* Local variables */
-    const int displayLevel = parameters->zParams.notificationLevel;
+    const int displayLevel = (int)parameters->zParams.notificationLevel;
     unsigned iteration = 1;
     unsigned d;
     unsigned k;
@@ -704,7 +723,7 @@ ZDICT_optimizeTrainFromBuffer_fastCover(
         data->parameters.splitPoint = splitPoint;
         data->parameters.steps = kSteps;
         data->parameters.shrinkDict = shrinkDict;
-        data->parameters.zParams.notificationLevel = g_displayLevel;
+        data->parameters.zParams.notificationLevel = (unsigned)g_displayLevel;
         /* Check the parameters */
         if (!FASTCOVER_checkParameters(data->parameters, dictBufferCapacity,
                                        data->ctx->f, accel)) {

data/ext/zstdruby/libzstd/dictBuilder/zdict.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
+ * Copyright (c) Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -23,9 +23,13 @@
 /* Unix Large Files support (>4GB) */
 #define _FILE_OFFSET_BITS 64
 #if (defined(__sun__) && (!defined(__LP64__)))   /* Sun Solaris 32-bits requires specific definitions */
+#  ifndef _LARGEFILE_SOURCE
 #  define _LARGEFILE_SOURCE
+#  endif
 #elif ! defined(__LP64__)                        /* No point defining Large file for 64 bit */
+#  ifndef _LARGEFILE64_SOURCE
 #  define _LARGEFILE64_SOURCE
+#  endif
 #endif
 
 
@@ -37,17 +41,19 @@
 #include <stdio.h>         /* fprintf, fopen, ftello64 */
 #include <time.h>          /* clock */
 
-#include "mem.h"           /* read */
-#include "fse.h"           /* FSE_normalizeCount, FSE_writeNCount */
-#define HUF_STATIC_LINKING_ONLY
-#include "huf.h"           /* HUF_buildCTable, HUF_writeCTable */
-#include "zstd_internal.h" /* includes zstd.h */
-#include "xxhash.h"        /* XXH64 */
-#include "divsufsort.h"
 #ifndef ZDICT_STATIC_LINKING_ONLY
 #  define ZDICT_STATIC_LINKING_ONLY
 #endif
-#include "zdict.h"
+#define HUF_STATIC_LINKING_ONLY
+
+#include "../common/mem.h"           /* read */
+#include "../common/fse.h"           /* FSE_normalizeCount, FSE_writeNCount */
+#include "../common/huf.h"           /* HUF_buildCTable, HUF_writeCTable */
+#include "../common/zstd_internal.h" /* includes zstd.h */
+#include "../common/xxhash.h"        /* XXH64 */
+#include "../compress/zstd_compress_internal.h" /* ZSTD_loadCEntropy() */
+#include "../zdict.h"
+#include "divsufsort.h"
 
 
 /*-*************************************
@@ -61,14 +67,15 @@
 
 #define NOISELENGTH 32
 
-static const int g_compressionLevel_default = 3;
 static const U32 g_selectivity_default = 9;
 
 
 /*-*************************************
 *  Console display
 ***************************************/
+#undef  DISPLAY
 #define DISPLAY(...)         { fprintf(stderr, __VA_ARGS__); fflush( stderr ); }
+#undef  DISPLAYLEVEL
 #define DISPLAYLEVEL(l, ...) if (notificationLevel>=l) { DISPLAY(__VA_ARGS__); }    /* 0 : no display;   1: errors;   2: default;  3: details;  4: debug */
 
 static clock_t ZDICT_clockSpan(clock_t nPrevious) { return clock() - nPrevious; }
@@ -99,6 +106,26 @@ unsigned ZDICT_getDictID(const void* dictBuffer, size_t dictSize)
     return MEM_readLE32((const char*)dictBuffer + 4);
 }
 
+size_t ZDICT_getDictHeaderSize(const void* dictBuffer, size_t dictSize)
+{
+    size_t headerSize;
+    if (dictSize <= 8 || MEM_readLE32(dictBuffer) != ZSTD_MAGIC_DICTIONARY) return ERROR(dictionary_corrupted);
+
+    {   ZSTD_compressedBlockState_t* bs = (ZSTD_compressedBlockState_t*)malloc(sizeof(ZSTD_compressedBlockState_t));
+        U32* wksp = (U32*)malloc(HUF_WORKSPACE_SIZE);
+        if (!bs || !wksp) {
+            headerSize = ERROR(memory_allocation);
+        } else {
+            ZSTD_reset_compressedBlockState(bs);
+            headerSize = ZSTD_loadCEntropy(bs, wksp, dictBuffer, dictSize);
+        }
+
+        free(bs);
+        free(wksp);
+    }
+
+    return headerSize;
+}
 
 /*-********************************************************
 *  Dictionary training functions
@@ -108,22 +135,32 @@ static unsigned ZDICT_NbCommonBytes (size_t val)
     if (MEM_isLittleEndian()) {
         if (MEM_64bits()) {
 #       if defined(_MSC_VER) && defined(_WIN64)
-            unsigned long r = 0;
-            _BitScanForward64( &r, (U64)val );
-            return (unsigned)(r>>3);
+            if (val != 0) {
+                unsigned long r;
+                _BitScanForward64(&r, (U64)val);
+                return (unsigned)(r >> 3);
+            } else {
+                /* Should not reach this code path */
+                __assume(0);
+            }
 #       elif defined(__GNUC__) && (__GNUC__ >= 3)
-            return (__builtin_ctzll((U64)val) >> 3);
+            return (unsigned)(__builtin_ctzll((U64)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
         } else { /* 32 bits */
 #       if defined(_MSC_VER)
-            unsigned long r=0;
-            _BitScanForward( &r, (U32)val );
-            return (unsigned)(r>>3);
+            if (val != 0) {
+                unsigned long r;
+                _BitScanForward(&r, (U32)val);
+                return (unsigned)(r >> 3);
+            } else {
+                /* Should not reach this code path */
+                __assume(0);
+            }
 #       elif defined(__GNUC__) && (__GNUC__ >= 3)
-            return (__builtin_ctz((U32)val) >> 3);
+            return (unsigned)(__builtin_ctz((U32)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];
@@ -132,11 +169,16 @@ static unsigned ZDICT_NbCommonBytes (size_t val)
     } else {  /* Big Endian CPU */
         if (MEM_64bits()) {
 #       if defined(_MSC_VER) && defined(_WIN64)
-            unsigned long r = 0;
-            _BitScanReverse64( &r, val );
-            return (unsigned)(r>>3);
+            if (val != 0) {
+                unsigned long r;
+                _BitScanReverse64(&r, val);
+                return (unsigned)(r >> 3);
+            } else {
+                /* Should not reach this code path */
+                __assume(0);
+            }
 #       elif defined(__GNUC__) && (__GNUC__ >= 3)
-            return (__builtin_clzll(val) >> 3);
+            return (unsigned)(__builtin_clzll(val) >> 3);
 #       else
             unsigned r;
             const unsigned n32 = sizeof(size_t)*4;   /* calculate this way due to compiler complaining in 32-bits mode */
@@ -147,11 +189,16 @@ static unsigned ZDICT_NbCommonBytes (size_t val)
 #       endif
         } else { /* 32 bits */
 #       if defined(_MSC_VER)
-            unsigned long r = 0;
-            _BitScanReverse( &r, (unsigned long)val );
-            return (unsigned)(r>>3);
+            if (val != 0) {
+                unsigned long r;
+                _BitScanReverse(&r, (unsigned long)val);
+                return (unsigned)(r >> 3);
+            } else {
+                /* Should not reach this code path */
+                __assume(0);
+            }
 #       elif defined(__GNUC__) && (__GNUC__ >= 3)
-            return (__builtin_clz((U32)val) >> 3);
+            return (unsigned)(__builtin_clz((U32)val) >> 3);
 #       else
             unsigned r;
             if (!(val>>16)) { r=2; val>>=8; } else { r=0; val>>=24; }
@@ -208,7 +255,7 @@ static dictItem ZDICT_analyzePos(
     U32 savings[LLIMIT] = {0};
     const BYTE* b = (const BYTE*)buffer;
     size_t maxLength = LLIMIT;
-    size_t pos = suffix[start];
+    size_t pos = (size_t)suffix[start];
     U32 end = start;
     dictItem solution;
 
@@ -342,7 +389,7 @@ static dictItem ZDICT_analyzePos(
             savings[i] = savings[i-1] + (lengthList[i] * (i-3));
 
         DISPLAYLEVEL(4, "Selected dict at position %u, of length %u : saves %u (ratio: %.2f)  \n",
-                     (unsigned)pos, (unsigned)maxLength, (unsigned)savings[maxLength], (double)savings[maxLength] / maxLength);
+                     (unsigned)pos, (unsigned)maxLength, (unsigned)savings[maxLength], (double)savings[maxLength] / (double)maxLength);
 
         solution.pos = (U32)pos;
         solution.length = (U32)maxLength;
@@ -352,7 +399,7 @@ static dictItem ZDICT_analyzePos(
         {   U32 id;
             for (id=start; id<end; id++) {
                 U32 p, pEnd, length;
-                U32 const testedPos = suffix[id];
+                U32 const testedPos = (U32)suffix[id];
                 if (testedPos == pos)
                     length = solution.length;
                 else {
@@ -415,7 +462,7 @@ static U32 ZDICT_tryMerge(dictItem* table, dictItem elt, U32 eltNbToSkip, const
 
         if ((table[u].pos + table[u].length >= elt.pos) && (table[u].pos < elt.pos)) {  /* overlap, existing < new */
             /* append */
-            int const addedLength = (int)eltEnd - (table[u].pos + table[u].length);
+            int const addedLength = (int)eltEnd - (int)(table[u].pos + table[u].length);
             table[u].savings += elt.length / 8;    /* rough approx bonus */
             if (addedLength > 0) {   /* otherwise, elt fully included into existing */
                 table[u].length += addedLength;
@@ -508,6 +555,7 @@ static size_t ZDICT_trainBuffer_legacy(dictItem* dictList, U32 dictListSize,
     clock_t displayClock = 0;
     clock_t const refreshRate = CLOCKS_PER_SEC * 3 / 10;
 
+#   undef  DISPLAYUPDATE
 #   define DISPLAYUPDATE(l, ...) if (notificationLevel>=l) { \
             if (ZDICT_clockSpan(displayClock) > refreshRate)  \
             { displayClock = clock(); DISPLAY(__VA_ARGS__); \
@@ -588,12 +636,12 @@ typedef struct
 
 #define MAXREPOFFSET 1024
 
-static void ZDICT_countEStats(EStats_ress_t esr, ZSTD_parameters params,
+static void ZDICT_countEStats(EStats_ress_t esr, const ZSTD_parameters* params,
                               unsigned* countLit, unsigned* offsetcodeCount, unsigned* matchlengthCount, unsigned* litlengthCount, U32* repOffsets,
                               const void* src, size_t srcSize,
                               U32 notificationLevel)
 {
-    size_t const blockSizeMax = MIN (ZSTD_BLOCKSIZE_MAX, 1 << params.cParams.windowLog);
+    size_t const blockSizeMax = MIN (ZSTD_BLOCKSIZE_MAX, 1 << params->cParams.windowLog);
     size_t cSize;
 
     if (srcSize > blockSizeMax) srcSize = blockSizeMax;   /* protection vs large samples */
@@ -682,7 +730,7 @@ static void ZDICT_flatLit(unsigned* countLit)
 
 #define OFFCODE_MAX 30  /* only applicable to first block */
 static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
-                                   unsigned compressionLevel,
+                                   int compressionLevel,
                              const void*  srcBuffer, const size_t* fileSizes, unsigned nbFiles,
                              const void* dictBuffer, size_t  dictBufferSize,
                                    unsigned notificationLevel)
@@ -717,7 +765,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
     memset(repOffset, 0, sizeof(repOffset));
     repOffset[1] = repOffset[4] = repOffset[8] = 1;
     memset(bestRepOffset, 0, sizeof(bestRepOffset));
-    if (compressionLevel==0) compressionLevel = g_compressionLevel_default;
+    if (compressionLevel==0) compressionLevel = ZSTD_CLEVEL_DEFAULT;
     params = ZSTD_getParams(compressionLevel, averageSampleSize, dictBufferSize);
 
     esr.dict = ZSTD_createCDict_advanced(dictBuffer, dictBufferSize, ZSTD_dlm_byRef, ZSTD_dct_rawContent, params.cParams, ZSTD_defaultCMem);
@@ -731,13 +779,20 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
 
     /* collect stats on all samples */
     for (u=0; u<nbFiles; u++) {
-        ZDICT_countEStats(esr, params,
+        ZDICT_countEStats(esr, &params,
                           countLit, offcodeCount, matchLengthCount, litLengthCount, repOffset,
                          (const char*)srcBuffer + pos, fileSizes[u],
                           notificationLevel);
         pos += fileSizes[u];
     }
 
+    if (notificationLevel >= 4) {
+        /* writeStats */
+        DISPLAYLEVEL(4, "Offset Code Frequencies : \n");
+        for (u=0; u<=offcodeMax; u++) {
+            DISPLAYLEVEL(4, "%2u :%7u \n", u, offcodeCount[u]);
+    }   }
+
     /* analyze, build stats, starting with literals */
     {   size_t maxNbBits = HUF_buildCTable (hufTable, countLit, 255, huffLog);
         if (HUF_isError(maxNbBits)) {
@@ -762,7 +817,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
     /* note : the result of this phase should be used to better appreciate the impact on statistics */
 
     total=0; for (u=0; u<=offcodeMax; u++) total+=offcodeCount[u];
-    errorCode = FSE_normalizeCount(offcodeNCount, Offlog, offcodeCount, total, offcodeMax);
+    errorCode = FSE_normalizeCount(offcodeNCount, Offlog, offcodeCount, total, offcodeMax, /* useLowProbCount */ 1);
     if (FSE_isError(errorCode)) {
         eSize = errorCode;
         DISPLAYLEVEL(1, "FSE_normalizeCount error with offcodeCount \n");
@@ -771,7 +826,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
     Offlog = (U32)errorCode;
 
     total=0; for (u=0; u<=MaxML; u++) total+=matchLengthCount[u];
-    errorCode = FSE_normalizeCount(matchLengthNCount, mlLog, matchLengthCount, total, MaxML);
+    errorCode = FSE_normalizeCount(matchLengthNCount, mlLog, matchLengthCount, total, MaxML, /* useLowProbCount */ 1);
     if (FSE_isError(errorCode)) {
         eSize = errorCode;
         DISPLAYLEVEL(1, "FSE_normalizeCount error with matchLengthCount \n");
@@ -780,7 +835,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
     mlLog = (U32)errorCode;
 
     total=0; for (u=0; u<=MaxLL; u++) total+=litLengthCount[u];
-    errorCode = FSE_normalizeCount(litLengthNCount, llLog, litLengthCount, total, MaxLL);
+    errorCode = FSE_normalizeCount(litLengthNCount, llLog, litLengthCount, total, MaxLL, /* useLowProbCount */ 1);
     if (FSE_isError(errorCode)) {
         eSize = errorCode;
         DISPLAYLEVEL(1, "FSE_normalizeCount error with litLengthCount \n");
@@ -844,7 +899,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
     MEM_writeLE32(dstPtr+8, bestRepOffset[2].offset);
 #else
     /* at this stage, we don't use the result of "most common first offset",
-       as the impact of statistics is not properly evaluated */
+     * as the impact of statistics is not properly evaluated */
     MEM_writeLE32(dstPtr+0, repStartValue[0]);
     MEM_writeLE32(dstPtr+4, repStartValue[1]);
     MEM_writeLE32(dstPtr+8, repStartValue[2]);
@@ -860,6 +915,17 @@ _cleanup:
 }
 
 
+/**
+ * @returns the maximum repcode value
+ */
+static U32 ZDICT_maxRep(U32 const reps[ZSTD_REP_NUM])
+{
+    U32 maxRep = reps[0];
+    int r;
+    for (r = 1; r < ZSTD_REP_NUM; ++r)
+        maxRep = MAX(maxRep, reps[r]);
+    return maxRep;
+}
 
 size_t ZDICT_finalizeDictionary(void* dictBuffer, size_t dictBufferCapacity,
                           const void* customDictContent, size_t dictContentSize,
@@ -869,13 +935,15 @@ size_t ZDICT_finalizeDictionary(void* dictBuffer, size_t dictBufferCapacity,
     size_t hSize;
 #define HBUFFSIZE 256   /* should prove large enough for all entropy headers */
     BYTE header[HBUFFSIZE];
-    int const compressionLevel = (params.compressionLevel == 0) ? g_compressionLevel_default : params.compressionLevel;
+    int const compressionLevel = (params.compressionLevel == 0) ? ZSTD_CLEVEL_DEFAULT : params.compressionLevel;
     U32 const notificationLevel = params.notificationLevel;
+    /* The final dictionary content must be at least as large as the largest repcode */
+    size_t const minContentSize = (size_t)ZDICT_maxRep(repStartValue);
+    size_t paddingSize;
 
     /* check conditions */
     DEBUGLOG(4, "ZDICT_finalizeDictionary");
     if (dictBufferCapacity < dictContentSize) return ERROR(dstSize_tooSmall);
-    if (dictContentSize < ZDICT_CONTENTSIZE_MIN) return ERROR(srcSize_wrong);
     if (dictBufferCapacity < ZDICT_DICTSIZE_MIN) return ERROR(dstSize_tooSmall);
 
     /* dictionary header */
@@ -899,12 +967,43 @@ size_t ZDICT_finalizeDictionary(void* dictBuffer, size_t dictBufferCapacity,
         hSize += eSize;
     }
 
-    /* copy elements in final buffer ; note : src and dst buffer can overlap */
-    if (hSize + dictContentSize > dictBufferCapacity) dictContentSize = dictBufferCapacity - hSize;
-    {   size_t const dictSize = hSize + dictContentSize;
-        char* dictEnd = (char*)dictBuffer + dictSize;
-        memmove(dictEnd - dictContentSize, customDictContent, dictContentSize);
-        memcpy(dictBuffer, header, hSize);
+    /* Shrink the content size if it doesn't fit in the buffer */
+    if (hSize + dictContentSize > dictBufferCapacity) {
+        dictContentSize = dictBufferCapacity - hSize;
+    }
+
+    /* Pad the dictionary content with zeros if it is too small */
+    if (dictContentSize < minContentSize) {
+        RETURN_ERROR_IF(hSize + minContentSize > dictBufferCapacity, dstSize_tooSmall,
+                        "dictBufferCapacity too small to fit max repcode");
+        paddingSize = minContentSize - dictContentSize;
+    } else {
+        paddingSize = 0;
+    }
+
+    {
+        size_t const dictSize = hSize + paddingSize + dictContentSize;
+
+        /* The dictionary consists of the header, optional padding, and the content.
+         * The padding comes before the content because the "best" position in the
+         * dictionary is the last byte.
+         */
+        BYTE* const outDictHeader = (BYTE*)dictBuffer;
+        BYTE* const outDictPadding = outDictHeader + hSize;
+        BYTE* const outDictContent = outDictPadding + paddingSize;
+
+        assert(dictSize <= dictBufferCapacity);
+        assert(outDictContent + dictContentSize == (BYTE*)dictBuffer + dictSize);
+
+        /* First copy the customDictContent into its final location.
+         * `customDictContent` and `dictBuffer` may overlap, so we must
+         * do this before any other writes into the output buffer.
+         * Then copy the header & padding into the output buffer.
+         */
+        memmove(outDictContent, customDictContent, dictContentSize);
+        memcpy(outDictHeader, header, hSize);
+        memset(outDictPadding, 0, paddingSize);
+
         return dictSize;
     }
 }
@@ -915,7 +1014,7 @@ static size_t ZDICT_addEntropyTablesFromBuffer_advanced(
         const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples,
         ZDICT_params_t params)
 {
-    int const compressionLevel = (params.compressionLevel == 0) ? g_compressionLevel_default : params.compressionLevel;
+    int const compressionLevel = (params.compressionLevel == 0) ? ZSTD_CLEVEL_DEFAULT : params.compressionLevel;
     U32 const notificationLevel = params.notificationLevel;
     size_t hSize = 8;
 
@@ -944,16 +1043,11 @@ static size_t ZDICT_addEntropyTablesFromBuffer_advanced(
     return MIN(dictBufferCapacity, hSize+dictContentSize);
 }
 
-/* Hidden declaration for dbio.c */
-size_t ZDICT_trainFromBuffer_unsafe_legacy(
-                            void* dictBuffer, size_t maxDictSize,
-                            const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples,
-                            ZDICT_legacy_params_t params);
 /*! ZDICT_trainFromBuffer_unsafe_legacy() :
-*   Warning : `samplesBuffer` must be followed by noisy guard band.
+*   Warning : `samplesBuffer` must be followed by noisy guard band !!!
 *   @return : size of dictionary, or an error code which can be tested with ZDICT_isError()
 */
-size_t ZDICT_trainFromBuffer_unsafe_legacy(
+static size_t ZDICT_trainFromBuffer_unsafe_legacy(
                             void* dictBuffer, size_t maxDictSize,
                             const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples,
                             ZDICT_legacy_params_t params)
@@ -1090,8 +1184,8 @@ size_t ZDICT_trainFromBuffer(void* dictBuffer, size_t dictBufferCapacity,
     memset(&params, 0, sizeof(params));
     params.d = 8;
     params.steps = 4;
-    /* Default to level 6 since no compression level information is available */
-    params.zParams.compressionLevel = 3;
+    /* Use default level since no compression level information is available */
+    params.zParams.compressionLevel = ZSTD_CLEVEL_DEFAULT;
 #if defined(DEBUGLEVEL) && (DEBUGLEVEL>=1)
     params.zParams.notificationLevel = DEBUGLEVEL;
 #endif

data/ext/zstdruby/libzstd/dll/example/Makefile

@@ -1,10 +1,11 @@
 # ################################################################
-# Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
+# Copyright (c) Yann Collet, Facebook, Inc.
 # All rights reserved.
 #
 # This source code is licensed under both the BSD-style license (found in the
 # LICENSE file in the root directory of this source tree) and the GPLv2 (found
 # in the COPYING file in the root directory of this source tree).
+# You may select, at your option, one of the above-listed licenses.
 # ################################################################
 
 VOID    := /dev/null