aboutsummaryrefslogtreecommitdiffstats
path: root/contrib/libs/zstd/lib/zdict.h
blob: f1e139a40ddb000f4dcc72f53b64e6a206304f79 (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
/*
 * 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.
 */

#ifndef DICTBUILDER_H_001
#define DICTBUILDER_H_001

#if defined (__cplusplus)
extern "C" {
#endif


/*======  Dependencies  ======*/
#include <stddef.h>  /* size_t */


/* =====   ZDICTLIB_API : control library symbols visibility   ===== */
#ifndef ZDICTLIB_VISIBILITY
#  if defined(__GNUC__) && (__GNUC__ >= 4)
#    define ZDICTLIB_VISIBILITY __attribute__ ((visibility ("default")))
#  else
#    define ZDICTLIB_VISIBILITY
#  endif
#endif
#if defined(ZSTD_DLL_EXPORT) && (ZSTD_DLL_EXPORT==1)
#  define ZDICTLIB_API __declspec(dllexport) ZDICTLIB_VISIBILITY
#elif defined(ZSTD_DLL_IMPORT) && (ZSTD_DLL_IMPORT==1)
#  define ZDICTLIB_API __declspec(dllimport) ZDICTLIB_VISIBILITY /* It isn't required but allows to generate better code, saving a function pointer load from the IAT and an indirect jump.*/
#else
#  define ZDICTLIB_API ZDICTLIB_VISIBILITY
#endif

/*******************************************************************************
 * Zstd dictionary builder
 *
 * FAQ
 * ===
 * Why should I use a dictionary?
 * ------------------------------
 *
 * Zstd can use dictionaries to improve compression ratio of small data.
 * Traditionally small files don't compress well because there is very little
 * repetition in a single sample, since it is small. But, if you are compressing
 * many similar files, like a bunch of JSON records that share the same
 * structure, you can train a dictionary on ahead of time on some samples of
 * these files. Then, zstd can use the dictionary to find repetitions that are
 * present across samples. This can vastly improve compression ratio.
 *
 * When is a dictionary useful?
 * ----------------------------
 *
 * Dictionaries are useful when compressing many small files that are similar.
 * The larger a file is, the less benefit a dictionary will have. Generally,
 * we don't expect dictionary compression to be effective past 100KB. And the
 * smaller a file is, the more we would expect the dictionary to help.
 *
 * How do I use a dictionary?
 * --------------------------
 *
 * Simply pass the dictionary to the zstd compressor with
 * `ZSTD_CCtx_loadDictionary()`. The same dictionary must then be passed to
 * the decompressor, using `ZSTD_DCtx_loadDictionary()`. There are other
 * more advanced functions that allow selecting some options, see zstd.h for
 * complete documentation.
 *
 * What is a zstd dictionary?
 * --------------------------
 *
 * A zstd dictionary has two pieces: Its header, and its content. The header
 * contains a magic number, the dictionary ID, and entropy tables. These
 * entropy tables allow zstd to save on header costs in the compressed file,
 * which really matters for small data. The content is just bytes, which are
 * repeated content that is common across many samples.
 *
 * What is a raw content dictionary?
 * ---------------------------------
 *
 * A raw content dictionary is just bytes. It doesn't have a zstd dictionary
 * header, a dictionary ID, or entropy tables. Any buffer is a valid raw
 * content dictionary.
 *
 * How do I train a dictionary?
 * ----------------------------
 *
 * Gather samples from your use case. These samples should be similar to each
 * other. If you have several use cases, you could try to train one dictionary
 * per use case.
 *
 * Pass those samples to `ZDICT_trainFromBuffer()` and that will train your
 * dictionary. There are a few advanced versions of this function, but this
 * is a great starting point. If you want to further tune your dictionary
 * you could try `ZDICT_optimizeTrainFromBuffer_cover()`. If that is too slow
 * you can try `ZDICT_optimizeTrainFromBuffer_fastCover()`.
 *
 * If the dictionary training function fails, that is likely because you
 * either passed too few samples, or a dictionary would not be effective
 * for your data. Look at the messages that the dictionary trainer printed,
 * if it doesn't say too few samples, then a dictionary would not be effective.
 *
 * How large should my dictionary be?
 * ----------------------------------
 *
 * A reasonable dictionary size, the `dictBufferCapacity`, is about 100KB.
 * The zstd CLI defaults to a 110KB dictionary. You likely don't need a
 * dictionary larger than that. But, most use cases can get away with a
 * smaller dictionary. The advanced dictionary builders can automatically
 * shrink the dictionary for you, and select a the smallest size that
 * doesn't hurt compression ratio too much. See the `shrinkDict` parameter.
 * A smaller dictionary can save memory, and potentially speed up
 * compression.
 *
 * How many samples should I provide to the dictionary builder?
 * ------------------------------------------------------------
 *
 * We generally recommend passing ~100x the size of the dictionary
 * in samples. A few thousand should suffice. Having too few samples
 * can hurt the dictionaries effectiveness. Having more samples will
 * only improve the dictionaries effectiveness. But having too many
 * samples can slow down the dictionary builder.
 *
 * How do I determine if a dictionary will be effective?
 * -----------------------------------------------------
 *
 * Simply train a dictionary and try it out. You can use zstd's built in
 * benchmarking tool to test the dictionary effectiveness.
 *
 *   # Benchmark levels 1-3 without a dictionary
 *   zstd -b1e3 -r /path/to/my/files
 *   # Benchmark levels 1-3 with a dictionary
 *   zstd -b1e3 -r /path/to/my/files -D /path/to/my/dictionary
 *
 * When should I retrain a dictionary?
 * -----------------------------------
 *
 * You should retrain a dictionary when its effectiveness drops. Dictionary
 * effectiveness drops as the data you are compressing changes. Generally, we do
 * expect dictionaries to "decay" over time, as your data changes, but the rate
 * at which they decay depends on your use case. Internally, we regularly
 * retrain dictionaries, and if the new dictionary performs significantly
 * better than the old dictionary, we will ship the new dictionary.
 *
 * I have a raw content dictionary, how do I turn it into a zstd dictionary?
 * -------------------------------------------------------------------------
 *
 * If you have a raw content dictionary, e.g. by manually constructing it, or
 * using a third-party dictionary builder, you can turn it into a zstd
 * dictionary by using `ZDICT_finalizeDictionary()`. You'll also have to
 * provide some samples of the data. It will add the zstd header to the
 * raw content, which contains a dictionary ID and entropy tables, which
 * will improve compression ratio, and allow zstd to write the dictionary ID
 * into the frame, if you so choose.
 *
 * Do I have to use zstd's dictionary builder?
 * -------------------------------------------
 *
 * No! You can construct dictionary content however you please, it is just
 * bytes. It will always be valid as a raw content dictionary. If you want
 * a zstd dictionary, which can improve compression ratio, use
 * `ZDICT_finalizeDictionary()`.
 *
 * What is the attack surface of a zstd dictionary?
 * ------------------------------------------------
 *
 * Zstd is heavily fuzz tested, including loading fuzzed dictionaries, so
 * zstd should never crash, or access out-of-bounds memory no matter what
 * the dictionary is. However, if an attacker can control the dictionary
 * during decompression, they can cause zstd to generate arbitrary bytes,
 * just like if they controlled the compressed data.
 *
 ******************************************************************************/


/*! ZDICT_trainFromBuffer():
 *  Train a dictionary from an array of samples.
 *  Redirect towards ZDICT_optimizeTrainFromBuffer_fastCover() single-threaded, with d=8, steps=4,
 *  f=20, and accel=1.
 *  Samples must be stored concatenated in a single flat buffer `samplesBuffer`,
 *  supplied with an array of sizes `samplesSizes`, providing the size of each sample, in order.
 *  The resulting dictionary will be saved into `dictBuffer`.
 * @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`)
 *          or an error code, which can be tested with ZDICT_isError().
 *  Note:  Dictionary training will fail if there are not enough samples to construct a
 *         dictionary, or if most of the samples are too small (< 8 bytes being the lower limit).
 *         If dictionary training fails, you should use zstd without a dictionary, as the dictionary
 *         would've been ineffective anyways. If you believe your samples would benefit from a dictionary
 *         please open an issue with details, and we can look into it.
 *  Note: ZDICT_trainFromBuffer()'s memory usage is about 6 MB.
 *  Tips: In general, a reasonable dictionary has a size of ~ 100 KB.
 *        It's possible to select smaller or larger size, just by specifying `dictBufferCapacity`.
 *        In general, it's recommended to provide a few thousands samples, though this can vary a lot.
 *        It's recommended that total size of all samples be about ~x100 times the target size of dictionary.
 */
ZDICTLIB_API size_t ZDICT_trainFromBuffer(void* dictBuffer, size_t dictBufferCapacity,
                                    const void* samplesBuffer,
                                    const size_t* samplesSizes, unsigned nbSamples);

typedef struct {
    int      compressionLevel;   /*< optimize for a specific zstd compression level; 0 means default */
    unsigned notificationLevel;  /*< Write log to stderr; 0 = none (default); 1 = errors; 2 = progression; 3 = details; 4 = debug; */
    unsigned dictID;             /*< force dictID value; 0 means auto mode (32-bits random value)
                                  *   NOTE: The zstd format reserves some dictionary IDs for future use.
                                  *         You may use them in private settings, but be warned that they
                                  *         may be used by zstd in a public dictionary registry in the future.
                                  *         These dictionary IDs are:
                                  *           - low range  : <= 32767
                                  *           - high range : >= (2^31)
                                  */
} ZDICT_params_t;

/*! ZDICT_finalizeDictionary():
 * Given a custom content as a basis for dictionary, and a set of samples,
 * finalize dictionary by adding headers and statistics according to the zstd
 * dictionary format.
 *
 * Samples must be stored concatenated in a flat buffer `samplesBuffer`,
 * supplied with an array of sizes `samplesSizes`, providing the size of each
 * sample in order. The samples are used to construct the statistics, so they
 * should be representative of what you will compress with this dictionary.
 *
 * The compression level can be set in `parameters`. You should pass the
 * compression level you expect to use in production. The statistics for each
 * compression level differ, so tuning the dictionary for the compression level
 * can help quite a bit.
 *
 * You can set an explicit dictionary ID in `parameters`, or allow us to pick
 * a random dictionary ID for you, but we can't guarantee no collisions.
 *
 * The dstDictBuffer and the dictContent may overlap, and the content will be
 * appended to the end of the header. If the header + the content doesn't fit in
 * maxDictSize the beginning of the content is truncated to make room, since it
 * is presumed that the most profitable content is at the end of the dictionary,
 * since that is the cheapest to reference.
 *
 * `maxDictSize` must be >= max(dictContentSize, ZSTD_DICTSIZE_MIN).
 *
 * @return: size of dictionary stored into `dstDictBuffer` (<= `maxDictSize`),
 *          or an error code, which can be tested by ZDICT_isError().
 * Note: ZDICT_finalizeDictionary() will push notifications into stderr if
 *       instructed to, using notificationLevel>0.
 * NOTE: This function currently may fail in several edge cases including:
 *         * Not enough samples
 *         * Samples are uncompressible
 *         * Samples are all exactly the same
 */
ZDICTLIB_API size_t ZDICT_finalizeDictionary(void* dstDictBuffer, size_t maxDictSize,
                                const void* dictContent, size_t dictContentSize,
                                const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples,
                                ZDICT_params_t parameters);


/*======   Helper functions   ======*/
ZDICTLIB_API unsigned ZDICT_getDictID(const void* dictBuffer, size_t dictSize);  /**< extracts dictID; @return zero if error (not a valid dictionary) */
ZDICTLIB_API size_t ZDICT_getDictHeaderSize(const void* dictBuffer, size_t dictSize);  /* returns dict header size; returns a ZSTD error code on failure */
ZDICTLIB_API unsigned ZDICT_isError(size_t errorCode);
ZDICTLIB_API const char* ZDICT_getErrorName(size_t errorCode);



#ifdef ZDICT_STATIC_LINKING_ONLY

/* ====================================================================================
 * The definitions in this section are considered experimental.
 * They should never be used with a dynamic library, as they may change in the future.
 * They are provided for advanced usages.
 * Use them only in association with static linking.
 * ==================================================================================== */

#define ZDICT_DICTSIZE_MIN    256
/* Deprecated: Remove in v1.6.0 */
#define ZDICT_CONTENTSIZE_MIN 128

/*! ZDICT_cover_params_t:
 *  k and d are the only required parameters.
 *  For others, value 0 means default.
 */
typedef struct {
    unsigned k;                  /* Segment size : constraint: 0 < k : Reasonable range [16, 2048+] */
    unsigned d;                  /* dmer size : constraint: 0 < d <= k : Reasonable range [6, 16] */
    unsigned steps;              /* Number of steps : Only used for optimization : 0 means default (40) : Higher means more parameters checked */
    unsigned nbThreads;          /* Number of threads : constraint: 0 < nbThreads : 1 means single-threaded : Only used for optimization : Ignored if ZSTD_MULTITHREAD is not defined */
    double splitPoint;           /* Percentage of samples used for training: Only used for optimization : the first nbSamples * splitPoint samples will be used to training, the last nbSamples * (1 - splitPoint) samples will be used for testing, 0 means default (1.0), 1.0 when all samples are used for both training and testing */
    unsigned shrinkDict;         /* Train dictionaries to shrink in size starting from the minimum size and selects the smallest dictionary that is shrinkDictMaxRegression% worse than the largest dictionary. 0 means no shrinking and 1 means shrinking  */
    unsigned shrinkDictMaxRegression; /* Sets shrinkDictMaxRegression so that a smaller dictionary can be at worse shrinkDictMaxRegression% worse than the max dict size dictionary. */
    ZDICT_params_t zParams;
} ZDICT_cover_params_t;

typedef struct {
    unsigned k;                  /* Segment size : constraint: 0 < k : Reasonable range [16, 2048+] */
    unsigned d;                  /* dmer size : constraint: 0 < d <= k : Reasonable range [6, 16] */
    unsigned f;                  /* log of size of frequency array : constraint: 0 < f <= 31 : 1 means default(20)*/
    unsigned steps;              /* Number of steps : Only used for optimization : 0 means default (40) : Higher means more parameters checked */
    unsigned nbThreads;          /* Number of threads : constraint: 0 < nbThreads : 1 means single-threaded : Only used for optimization : Ignored if ZSTD_MULTITHREAD is not defined */
    double splitPoint;           /* Percentage of samples used for training: Only used for optimization : the first nbSamples * splitPoint samples will be used to training, the last nbSamples * (1 - splitPoint) samples will be used for testing, 0 means default (0.75), 1.0 when all samples are used for both training and testing */
    unsigned accel;              /* Acceleration level: constraint: 0 < accel <= 10, higher means faster and less accurate, 0 means default(1) */
    unsigned shrinkDict;         /* Train dictionaries to shrink in size starting from the minimum size and selects the smallest dictionary that is shrinkDictMaxRegression% worse than the largest dictionary. 0 means no shrinking and 1 means shrinking  */
    unsigned shrinkDictMaxRegression; /* Sets shrinkDictMaxRegression so that a smaller dictionary can be at worse shrinkDictMaxRegression% worse than the max dict size dictionary. */

    ZDICT_params_t zParams;
} ZDICT_fastCover_params_t;

/*! ZDICT_trainFromBuffer_cover():
 *  Train a dictionary from an array of samples using the COVER algorithm.
 *  Samples must be stored concatenated in a single flat buffer `samplesBuffer`,
 *  supplied with an array of sizes `samplesSizes`, providing the size of each sample, in order.
 *  The resulting dictionary will be saved into `dictBuffer`.
 * @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`)
 *          or an error code, which can be tested with ZDICT_isError().
 *          See ZDICT_trainFromBuffer() for details on failure modes.
 *  Note: ZDICT_trainFromBuffer_cover() requires about 9 bytes of memory for each input byte.
 *  Tips: In general, a reasonable dictionary has a size of ~ 100 KB.
 *        It's possible to select smaller or larger size, just by specifying `dictBufferCapacity`.
 *        In general, it's recommended to provide a few thousands samples, though this can vary a lot.
 *        It's recommended that total size of all samples be about ~x100 times the target size of dictionary.
 */
ZDICTLIB_API size_t ZDICT_trainFromBuffer_cover(
          void *dictBuffer, size_t dictBufferCapacity,
    const void *samplesBuffer, const size_t *samplesSizes, unsigned nbSamples,
          ZDICT_cover_params_t parameters);

/*! ZDICT_optimizeTrainFromBuffer_cover():
 * The same requirements as above hold for all the parameters except `parameters`.
 * This function tries many parameter combinations and picks the best parameters.
 * `*parameters` is filled with the best parameters found,
 * dictionary constructed with those parameters is stored in `dictBuffer`.
 *
 * All of the parameters d, k, steps are optional.
 * If d is non-zero then we don't check multiple values of d, otherwise we check d = {6, 8}.
 * if steps is zero it defaults to its default value.
 * If k is non-zero then we don't check multiple values of k, otherwise we check steps values in [50, 2000].
 *
 * @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`)
 *          or an error code, which can be tested with ZDICT_isError().
 *          On success `*parameters` contains the parameters selected.
 *          See ZDICT_trainFromBuffer() for details on failure modes.
 * Note: ZDICT_optimizeTrainFromBuffer_cover() requires about 8 bytes of memory for each input byte and additionally another 5 bytes of memory for each byte of memory for each thread.
 */
ZDICTLIB_API size_t ZDICT_optimizeTrainFromBuffer_cover(
          void* dictBuffer, size_t dictBufferCapacity,
    const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples,
          ZDICT_cover_params_t* parameters);

/*! ZDICT_trainFromBuffer_fastCover():
 *  Train a dictionary from an array of samples using a modified version of COVER algorithm.
 *  Samples must be stored concatenated in a single flat buffer `samplesBuffer`,
 *  supplied with an array of sizes `samplesSizes`, providing the size of each sample, in order.
 *  d and k are required.
 *  All other parameters are optional, will use default values if not provided
 *  The resulting dictionary will be saved into `dictBuffer`.
 * @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`)
 *          or an error code, which can be tested with ZDICT_isError().
 *          See ZDICT_trainFromBuffer() for details on failure modes.
 *  Note: ZDICT_trainFromBuffer_fastCover() requires 6 * 2^f bytes of memory.
 *  Tips: In general, a reasonable dictionary has a size of ~ 100 KB.
 *        It's possible to select smaller or larger size, just by specifying `dictBufferCapacity`.
 *        In general, it's recommended to provide a few thousands samples, though this can vary a lot.
 *        It's recommended that total size of all samples be about ~x100 times the target size of dictionary.
 */
ZDICTLIB_API size_t ZDICT_trainFromBuffer_fastCover(void *dictBuffer,
                    size_t dictBufferCapacity, const void *samplesBuffer,
                    const size_t *samplesSizes, unsigned nbSamples,
                    ZDICT_fastCover_params_t parameters);

/*! ZDICT_optimizeTrainFromBuffer_fastCover():
 * The same requirements as above hold for all the parameters except `parameters`.
 * This function tries many parameter combinations (specifically, k and d combinations)
 * and picks the best parameters. `*parameters` is filled with the best parameters found,
 * dictionary constructed with those parameters is stored in `dictBuffer`.
 * All of the parameters d, k, steps, f, and accel are optional.
 * If d is non-zero then we don't check multiple values of d, otherwise we check d = {6, 8}.
 * if steps is zero it defaults to its default value.
 * If k is non-zero then we don't check multiple values of k, otherwise we check steps values in [50, 2000].
 * If f is zero, default value of 20 is used.
 * If accel is zero, default value of 1 is used.
 *
 * @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`)
 *          or an error code, which can be tested with ZDICT_isError().
 *          On success `*parameters` contains the parameters selected.
 *          See ZDICT_trainFromBuffer() for details on failure modes.
 * Note: ZDICT_optimizeTrainFromBuffer_fastCover() requires about 6 * 2^f bytes of memory for each thread.
 */
ZDICTLIB_API size_t ZDICT_optimizeTrainFromBuffer_fastCover(void* dictBuffer,
                    size_t dictBufferCapacity, const void* samplesBuffer,
                    const size_t* samplesSizes, unsigned nbSamples,
                    ZDICT_fastCover_params_t* parameters);

typedef struct {
    unsigned selectivityLevel;   /* 0 means default; larger => select more => larger dictionary */
    ZDICT_params_t zParams;
} ZDICT_legacy_params_t;

/*! ZDICT_trainFromBuffer_legacy():
 *  Train a dictionary from an array of samples.
 *  Samples must be stored concatenated in a single flat buffer `samplesBuffer`,
 *  supplied with an array of sizes `samplesSizes`, providing the size of each sample, in order.
 *  The resulting dictionary will be saved into `dictBuffer`.
 * `parameters` is optional and can be provided with values set to 0 to mean "default".
 * @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`)
 *          or an error code, which can be tested with ZDICT_isError().
 *          See ZDICT_trainFromBuffer() for details on failure modes.
 *  Tips: In general, a reasonable dictionary has a size of ~ 100 KB.
 *        It's possible to select smaller or larger size, just by specifying `dictBufferCapacity`.
 *        In general, it's recommended to provide a few thousands samples, though this can vary a lot.
 *        It's recommended that total size of all samples be about ~x100 times the target size of dictionary.
 *  Note: ZDICT_trainFromBuffer_legacy() will send notifications into stderr if instructed to, using notificationLevel>0.
 */
ZDICTLIB_API size_t ZDICT_trainFromBuffer_legacy(
    void* dictBuffer, size_t dictBufferCapacity,
    const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples,
    ZDICT_legacy_params_t parameters);


/* Deprecation warnings */
/* It is generally possible to disable deprecation warnings from compiler,
   for example with -Wno-deprecated-declarations for gcc
   or _CRT_SECURE_NO_WARNINGS in Visual.
   Otherwise, it's also possible to manually define ZDICT_DISABLE_DEPRECATE_WARNINGS */
#ifdef ZDICT_DISABLE_DEPRECATE_WARNINGS
#  define ZDICT_DEPRECATED(message) ZDICTLIB_API   /* disable deprecation warnings */
#else
#  define ZDICT_GCC_VERSION (__GNUC__ * 100 + __GNUC_MINOR__)
#  if defined (__cplusplus) && (__cplusplus >= 201402) /* C++14 or greater */
#    define ZDICT_DEPRECATED(message) [[deprecated(message)]] ZDICTLIB_API
#  elif defined(__clang__) || (ZDICT_GCC_VERSION >= 405)
#    define ZDICT_DEPRECATED(message) ZDICTLIB_API __attribute__((deprecated(message)))
#  elif (ZDICT_GCC_VERSION >= 301)
#    define ZDICT_DEPRECATED(message) ZDICTLIB_API __attribute__((deprecated))
#  elif defined(_MSC_VER)
#    define ZDICT_DEPRECATED(message) ZDICTLIB_API __declspec(deprecated(message))
#  else
#    pragma message("WARNING: You need to implement ZDICT_DEPRECATED for this compiler")
#    define ZDICT_DEPRECATED(message) ZDICTLIB_API
#  endif
#endif /* ZDICT_DISABLE_DEPRECATE_WARNINGS */

ZDICT_DEPRECATED("use ZDICT_finalizeDictionary() instead")
size_t ZDICT_addEntropyTablesFromBuffer(void* dictBuffer, size_t dictContentSize, size_t dictBufferCapacity,
                                  const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples);


#endif   /* ZDICT_STATIC_LINKING_ONLY */

#if defined (__cplusplus)
}
#endif

#endif   /* DICTBUILDER_H_001 */