2 * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
5 * This source code is licensed under the BSD-style license found in the
6 * LICENSE file in the root directory of this source tree. An additional grant
7 * of patent rights can be found in the PATENTS file in the same directory.
10 #ifndef DICTBUILDER_H_001
11 #define DICTBUILDER_H_001
13 #if defined (__cplusplus)
18 /*====== Dependencies ======*/
19 #include <stddef.h> /* size_t */
22 /*====== Export for Windows ======*/
25 * Enable exporting of functions when building a Windows DLL
27 #if defined(_WIN32) && defined(ZSTD_DLL_EXPORT) && (ZSTD_DLL_EXPORT==1)
28 # define ZDICTLIB_API __declspec(dllexport)
34 /*! ZDICT_trainFromBuffer() :
35 Train a dictionary from an array of samples.
36 Samples must be stored concatenated in a single flat buffer `samplesBuffer`,
37 supplied with an array of sizes `samplesSizes`, providing the size of each sample, in order.
38 The resulting dictionary will be saved into `dictBuffer`.
39 @return : size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`)
40 or an error code, which can be tested with ZDICT_isError().
41 Tips : In general, a reasonable dictionary has a size of ~ 100 KB.
42 It's obviously possible to target smaller or larger ones, just by specifying different `dictBufferCapacity`.
43 In general, it's recommended to provide a few thousands samples, but this can vary a lot.
44 It's recommended that total size of all samples be about ~x100 times the target size of dictionary.
46 ZDICTLIB_API
size_t ZDICT_trainFromBuffer(void* dictBuffer
, size_t dictBufferCapacity
,
47 const void* samplesBuffer
, const size_t* samplesSizes
, unsigned nbSamples
);
50 /*====== Helper functions ======*/
51 ZDICTLIB_API
unsigned ZDICT_getDictID(const void* dictBuffer
, size_t dictSize
); /**< extracts dictID; @return zero if error (not a valid dictionary) */
52 ZDICTLIB_API
unsigned ZDICT_isError(size_t errorCode
);
53 ZDICTLIB_API
const char* ZDICT_getErrorName(size_t errorCode
);
57 #ifdef ZDICT_STATIC_LINKING_ONLY
59 /* ====================================================================================
60 * The definitions in this section are considered experimental.
61 * They should never be used with a dynamic library, as they may change in the future.
62 * They are provided for advanced usages.
63 * Use them only in association with static linking.
64 * ==================================================================================== */
67 unsigned selectivityLevel
; /* 0 means default; larger => select more => larger dictionary */
68 int compressionLevel
; /* 0 means default; target a specific zstd compression level */
69 unsigned notificationLevel
; /* Write to stderr; 0 = none (default); 1 = errors; 2 = progression; 3 = details; 4 = debug; */
70 unsigned dictID
; /* 0 means auto mode (32-bits random value); other : force dictID value */
71 unsigned reserved
[2]; /* reserved space for future parameters */
75 /*! ZDICT_trainFromBuffer_advanced() :
76 Same as ZDICT_trainFromBuffer() with control over more parameters.
77 `parameters` is optional and can be provided with values set to 0 to mean "default".
78 @return : size of dictionary stored into `dictBuffer` (<= `dictBufferSize`),
79 or an error code, which can be tested by ZDICT_isError().
80 note : ZDICT_trainFromBuffer_advanced() will send notifications into stderr if instructed to, using notificationLevel>0.
82 size_t ZDICT_trainFromBuffer_advanced(void* dictBuffer
, size_t dictBufferCapacity
,
83 const void* samplesBuffer
, const size_t* samplesSizes
, unsigned nbSamples
,
84 ZDICT_params_t parameters
);
87 /*! ZDICT_addEntropyTablesFromBuffer() :
89 Given a content-only dictionary (built using any 3rd party algorithm),
90 add entropy tables computed from an array of samples.
91 Samples must be stored concatenated in a flat buffer `samplesBuffer`,
92 supplied with an array of sizes `samplesSizes`, providing the size of each sample in order.
94 The input dictionary content must be stored *at the end* of `dictBuffer`.
95 Its size is `dictContentSize`.
96 The resulting dictionary with added entropy tables will be *written back to `dictBuffer`*,
97 starting from its beginning.
98 @return : size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`).
100 size_t ZDICT_addEntropyTablesFromBuffer(void* dictBuffer
, size_t dictContentSize
, size_t dictBufferCapacity
,
101 const void* samplesBuffer
, const size_t* samplesSizes
, unsigned nbSamples
);
105 #endif /* ZDICT_STATIC_LINKING_ONLY */
107 #if defined (__cplusplus)
111 #endif /* DICTBUILDER_H_001 */