From 2cf6d662a8ff11f947e135b052514a7fb64bbb1c Mon Sep 17 00:00:00 2001
From: Yann Collet <yann.collet.73@gmail.com>
Date: Tue, 23 Feb 2016 21:21:29 +0100
Subject: [PATCH 1/3] fixed #131

---
 programs/.gitignore | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/programs/.gitignore b/programs/.gitignore
index 021e89371..2fc85021a 100644
--- a/programs/.gitignore
+++ b/programs/.gitignore
@@ -29,3 +29,6 @@ datagen
 # Visual solution files
 *.suo
 *.user
+
+# Default dictionary name
+dictionary

From 3152a8c87a5261d03eb4b4a3c359523f331d0c46 Mon Sep 17 00:00:00 2001
From: Yann Collet <yann.collet.73@gmail.com>
Date: Tue, 23 Feb 2016 21:28:59 +0100
Subject: [PATCH 2/3] finer dictionary completion

---
 lib/zdict.c | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/lib/zdict.c b/lib/zdict.c
index d3d5784dd..2b3d3ae8a 100644
--- a/lib/zdict.c
+++ b/lib/zdict.c
@@ -618,6 +618,14 @@ static void ZDICT_countEStats(EStats_ress_t esr,
         litlengthCount[*bytePtr]++;
 }
 
+static size_t ZDICT_maxSampleSize(const size_t* fileSizes, unsigned nbFiles)
+{
+    unsigned u;
+    size_t max=0;
+    for (u=0; u<nbFiles; u++)
+        if (max < fileSizes[u]) max = fileSizes[u];
+    return max;
+}
 
 #define OFFCODE_MAX 18  /* only applicable to first block */
 static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
@@ -653,7 +661,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
             goto _cleanup;
     }
     if (compressionLevel==0) compressionLevel=g_compressionLevel_default;
-    params = ZSTD_getParams(compressionLevel, dictBufferSize + 15 KB);
+    params = ZSTD_getParams(compressionLevel, MAX(dictBufferSize, ZDICT_maxSampleSize(fileSizes, nbFiles)));
     params.strategy = ZSTD_greedy;
     ZSTD_compressBegin_advanced(esr.ref, dictBuffer, dictBufferSize, params);
 
@@ -800,7 +808,7 @@ size_t ZDICT_trainFromBuffer_unsafe(
     dictItem* dictList = (dictItem*)malloc(dictListSize * sizeof(*dictList));
     unsigned selectivity = params.selectivityLevel;
     unsigned compressionLevel = params.compressionLevel;
-    size_t targetDictSize = maxDictSize - g_provision_entropySize;
+    size_t targetDictSize = maxDictSize;
     size_t sBuffSize;
     size_t dictSize = 0;
 
@@ -859,8 +867,8 @@ size_t ZDICT_trainFromBuffer_unsafe(
         if (selectivity==1) {  /* note could also be used to complete a dictionary, but not necessarily better */
             DISPLAYLEVEL(3, "\r%70s\r", "");   /* clean display line */
             DISPLAYLEVEL(3, "Adding %u KB with fast sampling \n", (U32)(targetDictSize>>10));
-            dictContentSize = (U32)ZDICT_fastSampling((char*)dictBuffer + g_provision_entropySize,
-                                               targetDictSize, samplesBuffer, sBuffSize);
+            dictContentSize = (U32)ZDICT_fastSampling(dictBuffer, targetDictSize,
+                                                      samplesBuffer, sBuffSize);
         }
 
        /* dictionary header */

From 7de4f9fd81c997077b85d0d8bf76590bcfe258ee Mon Sep 17 00:00:00 2001
From: Yann Collet <yann.collet.73@gmail.com>
Date: Tue, 23 Feb 2016 21:34:18 +0100
Subject: [PATCH 3/3] minor cosmetic

---
 programs/dibio.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/programs/dibio.c b/programs/dibio.c
index 646fe2c60..17f89586b 100644
--- a/programs/dibio.c
+++ b/programs/dibio.c
@@ -259,7 +259,7 @@ int DiB_trainFromFiles(const char* dictFileName, unsigned maxDictSize,
                         srcBuffer, fileSizes, nbFiles,
                         params);
     if (ZDICT_isError(dictSize)) {
-        DISPLAYLEVEL(1, "dictionary training failed : %s", ZDICT_getErrorName(dictSize));   /* should not happen */
+        DISPLAYLEVEL(1, "dictionary training failed : %s \n", ZDICT_getErrorName(dictSize));   /* should not happen */
         result = 1;
         goto _cleanup;
     }