1 | /* |
2 | * Copyright (c) 2016-present, Yann Collet, Facebook, Inc. |
3 | * All rights reserved. |
4 | * |
5 | * This source code is licensed under both the BSD-style license (found in the |
6 | * LICENSE file in the root directory of this source tree) and the GPLv2 (found |
7 | * in the COPYING file in the root directory of this source tree). |
8 | * You may select, at your option, one of the above-listed licenses. |
9 | */ |
10 | #if defined (__cplusplus) |
11 | extern "C" { |
12 | #endif |
13 | |
14 | #ifndef ZSTD_H_235446 |
15 | #define ZSTD_H_235446 |
16 | |
17 | /* ====== Dependency ======*/ |
18 | #include <stddef.h> /* size_t */ |
19 | |
20 | |
21 | /* ===== ZSTDLIB_API : control library symbols visibility ===== */ |
22 | #ifndef ZSTDLIB_VISIBILITY |
23 | # if defined(__GNUC__) && (__GNUC__ >= 4) |
24 | # define ZSTDLIB_VISIBILITY __attribute__ ((visibility ("default"))) |
25 | # else |
26 | # define ZSTDLIB_VISIBILITY |
27 | # endif |
28 | #endif |
29 | #if defined(ZSTD_DLL_EXPORT) && (ZSTD_DLL_EXPORT==1) |
30 | # define ZSTDLIB_API __declspec(dllexport) ZSTDLIB_VISIBILITY |
31 | #elif defined(ZSTD_DLL_IMPORT) && (ZSTD_DLL_IMPORT==1) |
32 | # define ZSTDLIB_API __declspec(dllimport) ZSTDLIB_VISIBILITY /* It isn't required but allows to generate better code, saving a function pointer load from the IAT and an indirect jump.*/ |
33 | #else |
34 | # define ZSTDLIB_API ZSTDLIB_VISIBILITY |
35 | #endif |
36 | |
37 | |
38 | /******************************************************************************* |
39 | Introduction |
40 | |
41 | zstd, short for Zstandard, is a fast lossless compression algorithm, targeting |
42 | real-time compression scenarios at zlib-level and better compression ratios. |
43 | The zstd compression library provides in-memory compression and decompression |
44 | functions. |
45 | |
46 | The library supports regular compression levels from 1 up to ZSTD_maxCLevel(), |
47 | which is currently 22. Levels >= 20, labeled `--ultra`, should be used with |
48 | caution, as they require more memory. The library also offers negative |
49 | compression levels, which extend the range of speed vs. ratio preferences. |
50 | The lower the level, the faster the speed (at the cost of compression). |
51 | |
52 | Compression can be done in: |
53 | - a single step (described as Simple API) |
54 | - a single step, reusing a context (described as Explicit context) |
55 | - unbounded multiple steps (described as Streaming compression) |
56 | |
57 | The compression ratio achievable on small data can be highly improved using |
58 | a dictionary. Dictionary compression can be performed in: |
59 | - a single step (described as Simple dictionary API) |
60 | - a single step, reusing a dictionary (described as Bulk-processing |
61 | dictionary API) |
62 | |
63 | Advanced experimental functions can be accessed using |
64 | `#define ZSTD_STATIC_LINKING_ONLY` before including zstd.h. |
65 | |
66 | Advanced experimental APIs should never be used with a dynamically-linked |
67 | library. They are not "stable"; their definitions or signatures may change in |
68 | the future. Only static linking is allowed. |
69 | *******************************************************************************/ |
70 | |
71 | /*------ Version ------*/ |
72 | #define ZSTD_VERSION_MAJOR 1 |
73 | #define ZSTD_VERSION_MINOR 3 |
74 | #define ZSTD_VERSION_RELEASE 8 |
75 | |
76 | #define ZSTD_VERSION_NUMBER (ZSTD_VERSION_MAJOR *100*100 + ZSTD_VERSION_MINOR *100 + ZSTD_VERSION_RELEASE) |
77 | ZSTDLIB_API unsigned ZSTD_versionNumber(void); /**< to check runtime library version */ |
78 | |
79 | #define ZSTD_LIB_VERSION ZSTD_VERSION_MAJOR.ZSTD_VERSION_MINOR.ZSTD_VERSION_RELEASE |
80 | #define ZSTD_QUOTE(str) #str |
81 | #define ZSTD_EXPAND_AND_QUOTE(str) ZSTD_QUOTE(str) |
82 | #define ZSTD_VERSION_STRING ZSTD_EXPAND_AND_QUOTE(ZSTD_LIB_VERSION) |
83 | ZSTDLIB_API const char* ZSTD_versionString(void); /* requires v1.3.0+ */ |
84 | |
85 | /*************************************** |
86 | * Default constant |
87 | ***************************************/ |
88 | #ifndef ZSTD_CLEVEL_DEFAULT |
89 | # define ZSTD_CLEVEL_DEFAULT 3 |
90 | #endif |
91 | |
92 | /*************************************** |
93 | * Simple API |
94 | ***************************************/ |
95 | /*! ZSTD_compress() : |
96 | * Compresses `src` content as a single zstd compressed frame into already allocated `dst`. |
97 | * Hint : compression runs faster if `dstCapacity` >= `ZSTD_compressBound(srcSize)`. |
98 | * @return : compressed size written into `dst` (<= `dstCapacity), |
99 | * or an error code if it fails (which can be tested using ZSTD_isError()). */ |
100 | ZSTDLIB_API size_t ZSTD_compress( void* dst, size_t dstCapacity, |
101 | const void* src, size_t srcSize, |
102 | int compressionLevel); |
103 | |
104 | /*! ZSTD_decompress() : |
105 | * `compressedSize` : must be the _exact_ size of some number of compressed and/or skippable frames. |
106 | * `dstCapacity` is an upper bound of originalSize to regenerate. |
107 | * If user cannot imply a maximum upper bound, it's better to use streaming mode to decompress data. |
108 | * @return : the number of bytes decompressed into `dst` (<= `dstCapacity`), |
109 | * or an errorCode if it fails (which can be tested using ZSTD_isError()). */ |
110 | ZSTDLIB_API size_t ZSTD_decompress( void* dst, size_t dstCapacity, |
111 | const void* src, size_t compressedSize); |
112 | |
113 | /*! ZSTD_getFrameContentSize() : requires v1.3.0+ |
114 | * `src` should point to the start of a ZSTD encoded frame. |
115 | * `srcSize` must be at least as large as the frame header. |
116 | * hint : any size >= `ZSTD_frameHeaderSize_max` is large enough. |
117 | * @return : - decompressed size of `src` frame content, if known |
118 | * - ZSTD_CONTENTSIZE_UNKNOWN if the size cannot be determined |
119 | * - ZSTD_CONTENTSIZE_ERROR if an error occurred (e.g. invalid magic number, srcSize too small) |
120 | * note 1 : a 0 return value means the frame is valid but "empty". |
121 | * note 2 : decompressed size is an optional field, it may not be present, typically in streaming mode. |
122 | * When `return==ZSTD_CONTENTSIZE_UNKNOWN`, data to decompress could be any size. |
123 | * In which case, it's necessary to use streaming mode to decompress data. |
124 | * Optionally, application can rely on some implicit limit, |
125 | * as ZSTD_decompress() only needs an upper bound of decompressed size. |
126 | * (For example, data could be necessarily cut into blocks <= 16 KB). |
127 | * note 3 : decompressed size is always present when compression is completed using single-pass functions, |
128 | * such as ZSTD_compress(), ZSTD_compressCCtx() ZSTD_compress_usingDict() or ZSTD_compress_usingCDict(). |
129 | * note 4 : decompressed size can be very large (64-bits value), |
130 | * potentially larger than what local system can handle as a single memory segment. |
131 | * In which case, it's necessary to use streaming mode to decompress data. |
132 | * note 5 : If source is untrusted, decompressed size could be wrong or intentionally modified. |
133 | * Always ensure return value fits within application's authorized limits. |
134 | * Each application can set its own limits. |
135 | * note 6 : This function replaces ZSTD_getDecompressedSize() */ |
136 | #define ZSTD_CONTENTSIZE_UNKNOWN (0ULL - 1) |
137 | #define ZSTD_CONTENTSIZE_ERROR (0ULL - 2) |
138 | ZSTDLIB_API unsigned long long ZSTD_getFrameContentSize(const void *src, size_t srcSize); |
139 | |
140 | /*! ZSTD_getDecompressedSize() : |
141 | * NOTE: This function is now obsolete, in favor of ZSTD_getFrameContentSize(). |
142 | * Both functions work the same way, but ZSTD_getDecompressedSize() blends |
143 | * "empty", "unknown" and "error" results to the same return value (0), |
144 | * while ZSTD_getFrameContentSize() gives them separate return values. |
145 | * @return : decompressed size of `src` frame content _if known and not empty_, 0 otherwise. */ |
146 | ZSTDLIB_API unsigned long long ZSTD_getDecompressedSize(const void* src, size_t srcSize); |
147 | |
148 | |
149 | /*====== Helper functions ======*/ |
150 | #define ZSTD_COMPRESSBOUND(srcSize) ((srcSize) + ((srcSize)>>8) + (((srcSize) < (128<<10)) ? (((128<<10) - (srcSize)) >> 11) /* margin, from 64 to 0 */ : 0)) /* this formula ensures that bound(A) + bound(B) <= bound(A+B) as long as A and B >= 128 KB */ |
151 | ZSTDLIB_API size_t ZSTD_compressBound(size_t srcSize); /*!< maximum compressed size in worst case single-pass scenario */ |
152 | ZSTDLIB_API unsigned ZSTD_isError(size_t code); /*!< tells if a `size_t` function result is an error code */ |
153 | ZSTDLIB_API const char* ZSTD_getErrorName(size_t code); /*!< provides readable string from an error code */ |
154 | ZSTDLIB_API int ZSTD_maxCLevel(void); /*!< maximum compression level available */ |
155 | |
156 | |
157 | /*************************************** |
158 | * Explicit context |
159 | ***************************************/ |
160 | /*= Compression context |
161 | * When compressing many times, |
162 | * it is recommended to allocate a context just once, and re-use it for each successive compression operation. |
163 | * This will make workload friendlier for system's memory. |
164 | * Use one context per thread for parallel execution in multi-threaded environments. */ |
165 | typedef struct ZSTD_CCtx_s ZSTD_CCtx; |
166 | ZSTDLIB_API ZSTD_CCtx* ZSTD_createCCtx(void); |
167 | ZSTDLIB_API size_t ZSTD_freeCCtx(ZSTD_CCtx* cctx); |
168 | |
169 | /*! ZSTD_compressCCtx() : |
170 | * Same as ZSTD_compress(), using an explicit ZSTD_CCtx |
171 | * The function will compress at requested compression level, |
172 | * ignoring any other parameter */ |
173 | ZSTDLIB_API size_t ZSTD_compressCCtx(ZSTD_CCtx* cctx, |
174 | void* dst, size_t dstCapacity, |
175 | const void* src, size_t srcSize, |
176 | int compressionLevel); |
177 | |
178 | /*= Decompression context |
179 | * When decompressing many times, |
180 | * it is recommended to allocate a context only once, |
181 | * and re-use it for each successive compression operation. |
182 | * This will make workload friendlier for system's memory. |
183 | * Use one context per thread for parallel execution. */ |
184 | typedef struct ZSTD_DCtx_s ZSTD_DCtx; |
185 | ZSTDLIB_API ZSTD_DCtx* ZSTD_createDCtx(void); |
186 | ZSTDLIB_API size_t ZSTD_freeDCtx(ZSTD_DCtx* dctx); |
187 | |
188 | /*! ZSTD_decompressDCtx() : |
189 | * Same as ZSTD_decompress(), |
190 | * requires an allocated ZSTD_DCtx. |
191 | * Compatible with sticky parameters. |
192 | */ |
193 | ZSTDLIB_API size_t ZSTD_decompressDCtx(ZSTD_DCtx* dctx, |
194 | void* dst, size_t dstCapacity, |
195 | const void* src, size_t srcSize); |
196 | |
197 | |
198 | /************************** |
199 | * Simple dictionary API |
200 | ***************************/ |
201 | /*! ZSTD_compress_usingDict() : |
202 | * Compression at an explicit compression level using a Dictionary. |
203 | * A dictionary can be any arbitrary data segment (also called a prefix), |
204 | * or a buffer with specified information (see dictBuilder/zdict.h). |
205 | * Note : This function loads the dictionary, resulting in significant startup delay. |
206 | * It's intended for a dictionary used only once. |
207 | * Note 2 : When `dict == NULL || dictSize < 8` no dictionary is used. */ |
208 | ZSTDLIB_API size_t ZSTD_compress_usingDict(ZSTD_CCtx* ctx, |
209 | void* dst, size_t dstCapacity, |
210 | const void* src, size_t srcSize, |
211 | const void* dict,size_t dictSize, |
212 | int compressionLevel); |
213 | |
214 | /*! ZSTD_decompress_usingDict() : |
215 | * Decompression using a known Dictionary. |
216 | * Dictionary must be identical to the one used during compression. |
217 | * Note : This function loads the dictionary, resulting in significant startup delay. |
218 | * It's intended for a dictionary used only once. |
219 | * Note : When `dict == NULL || dictSize < 8` no dictionary is used. */ |
220 | ZSTDLIB_API size_t ZSTD_decompress_usingDict(ZSTD_DCtx* dctx, |
221 | void* dst, size_t dstCapacity, |
222 | const void* src, size_t srcSize, |
223 | const void* dict,size_t dictSize); |
224 | |
225 | |
226 | /*********************************** |
227 | * Bulk processing dictionary API |
228 | **********************************/ |
229 | typedef struct ZSTD_CDict_s ZSTD_CDict; |
230 | |
231 | /*! ZSTD_createCDict() : |
232 | * When compressing multiple messages / blocks using the same dictionary, it's recommended to load it only once. |
233 | * ZSTD_createCDict() will create a digested dictionary, ready to start future compression operations without startup cost. |
234 | * ZSTD_CDict can be created once and shared by multiple threads concurrently, since its usage is read-only. |
235 | * `dictBuffer` can be released after ZSTD_CDict creation, because its content is copied within CDict. |
236 | * Consider experimental function `ZSTD_createCDict_byReference()` if you prefer to not duplicate `dictBuffer` content. |
237 | * Note : A ZSTD_CDict can be created from an empty dictBuffer, but it is inefficient when used to compress small data. */ |
238 | ZSTDLIB_API ZSTD_CDict* ZSTD_createCDict(const void* dictBuffer, size_t dictSize, |
239 | int compressionLevel); |
240 | |
241 | /*! ZSTD_freeCDict() : |
242 | * Function frees memory allocated by ZSTD_createCDict(). */ |
243 | ZSTDLIB_API size_t ZSTD_freeCDict(ZSTD_CDict* CDict); |
244 | |
245 | /*! ZSTD_compress_usingCDict() : |
246 | * Compression using a digested Dictionary. |
247 | * Recommended when same dictionary is used multiple times. |
248 | * Note : compression level is _decided at dictionary creation time_, |
249 | * and frame parameters are hardcoded (dictID=yes, contentSize=yes, checksum=no) */ |
250 | ZSTDLIB_API size_t ZSTD_compress_usingCDict(ZSTD_CCtx* cctx, |
251 | void* dst, size_t dstCapacity, |
252 | const void* src, size_t srcSize, |
253 | const ZSTD_CDict* cdict); |
254 | |
255 | |
256 | typedef struct ZSTD_DDict_s ZSTD_DDict; |
257 | |
258 | /*! ZSTD_createDDict() : |
259 | * Create a digested dictionary, ready to start decompression operation without startup delay. |
260 | * dictBuffer can be released after DDict creation, as its content is copied inside DDict. */ |
261 | ZSTDLIB_API ZSTD_DDict* ZSTD_createDDict(const void* dictBuffer, size_t dictSize); |
262 | |
263 | /*! ZSTD_freeDDict() : |
264 | * Function frees memory allocated with ZSTD_createDDict() */ |
265 | ZSTDLIB_API size_t ZSTD_freeDDict(ZSTD_DDict* ddict); |
266 | |
267 | /*! ZSTD_decompress_usingDDict() : |
268 | * Decompression using a digested Dictionary. |
269 | * Recommended when same dictionary is used multiple times. */ |
270 | ZSTDLIB_API size_t ZSTD_decompress_usingDDict(ZSTD_DCtx* dctx, |
271 | void* dst, size_t dstCapacity, |
272 | const void* src, size_t srcSize, |
273 | const ZSTD_DDict* ddict); |
274 | |
275 | |
276 | /**************************** |
277 | * Streaming |
278 | ****************************/ |
279 | |
280 | typedef struct ZSTD_inBuffer_s { |
281 | const void* src; /**< start of input buffer */ |
282 | size_t size; /**< size of input buffer */ |
283 | size_t pos; /**< position where reading stopped. Will be updated. Necessarily 0 <= pos <= size */ |
284 | } ZSTD_inBuffer; |
285 | |
286 | typedef struct ZSTD_outBuffer_s { |
287 | void* dst; /**< start of output buffer */ |
288 | size_t size; /**< size of output buffer */ |
289 | size_t pos; /**< position where writing stopped. Will be updated. Necessarily 0 <= pos <= size */ |
290 | } ZSTD_outBuffer; |
291 | |
292 | |
293 | |
294 | /*-*********************************************************************** |
295 | * Streaming compression - HowTo |
296 | * |
297 | * A ZSTD_CStream object is required to track streaming operation. |
298 | * Use ZSTD_createCStream() and ZSTD_freeCStream() to create/release resources. |
299 | * ZSTD_CStream objects can be reused multiple times on consecutive compression operations. |
300 | * It is recommended to re-use ZSTD_CStream since it will play nicer with system's memory, by re-using already allocated memory. |
301 | * |
302 | * For parallel execution, use one separate ZSTD_CStream per thread. |
303 | * |
304 | * note : since v1.3.0, ZSTD_CStream and ZSTD_CCtx are the same thing. |
305 | * |
306 | * Parameters are sticky : when starting a new compression on the same context, |
307 | * it will re-use the same sticky parameters as previous compression session. |
308 | * When in doubt, it's recommended to fully initialize the context before usage. |
309 | * Use ZSTD_initCStream() to set the parameter to a selected compression level. |
310 | * Use advanced API (ZSTD_CCtx_setParameter(), etc.) to set more specific parameters. |
311 | * |
312 | * Use ZSTD_compressStream() as many times as necessary to consume input stream. |
313 | * The function will automatically update both `pos` fields within `input` and `output`. |
314 | * Note that the function may not consume the entire input, |
315 | * for example, because the output buffer is already full, |
316 | * in which case `input.pos < input.size`. |
317 | * The caller must check if input has been entirely consumed. |
318 | * If not, the caller must make some room to receive more compressed data, |
319 | * and then present again remaining input data. |
320 | * @return : a size hint, preferred nb of bytes to use as input for next function call |
321 | * or an error code, which can be tested using ZSTD_isError(). |
322 | * Note 1 : it's just a hint, to help latency a little, any value will work fine. |
323 | * Note 2 : size hint is guaranteed to be <= ZSTD_CStreamInSize() |
324 | * |
325 | * At any moment, it's possible to flush whatever data might remain stuck within internal buffer, |
326 | * using ZSTD_flushStream(). `output->pos` will be updated. |
327 | * Note that, if `output->size` is too small, a single invocation of ZSTD_flushStream() might not be enough (return code > 0). |
328 | * In which case, make some room to receive more compressed data, and call again ZSTD_flushStream(). |
329 | * @return : 0 if internal buffers are entirely flushed, |
330 | * >0 if some data still present within internal buffer (the value is minimal estimation of remaining size), |
331 | * or an error code, which can be tested using ZSTD_isError(). |
332 | * |
333 | * ZSTD_endStream() instructs to finish a frame. |
334 | * It will perform a flush and write frame epilogue. |
335 | * The epilogue is required for decoders to consider a frame completed. |
336 | * flush() operation is the same, and follows same rules as ZSTD_flushStream(). |
337 | * @return : 0 if frame fully completed and fully flushed, |
338 | * >0 if some data still present within internal buffer (the value is minimal estimation of remaining size), |
339 | * or an error code, which can be tested using ZSTD_isError(). |
340 | * |
341 | * *******************************************************************/ |
342 | |
343 | typedef ZSTD_CCtx ZSTD_CStream; /**< CCtx and CStream are now effectively same object (>= v1.3.0) */ |
344 | /* Continue to distinguish them for compatibility with older versions <= v1.2.0 */ |
345 | /*===== ZSTD_CStream management functions =====*/ |
346 | ZSTDLIB_API ZSTD_CStream* ZSTD_createCStream(void); |
347 | ZSTDLIB_API size_t ZSTD_freeCStream(ZSTD_CStream* zcs); |
348 | |
349 | /*===== Streaming compression functions =====*/ |
350 | ZSTDLIB_API size_t ZSTD_initCStream(ZSTD_CStream* zcs, int compressionLevel); |
351 | ZSTDLIB_API size_t ZSTD_compressStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output, ZSTD_inBuffer* input); |
352 | ZSTDLIB_API size_t ZSTD_flushStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output); |
353 | ZSTDLIB_API size_t ZSTD_endStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output); |
354 | |
355 | ZSTDLIB_API size_t ZSTD_CStreamInSize(void); /**< recommended size for input buffer */ |
356 | ZSTDLIB_API size_t ZSTD_CStreamOutSize(void); /**< recommended size for output buffer. Guarantee to successfully flush at least one complete compressed block in all circumstances. */ |
357 | |
358 | |
359 | |
360 | /*-*************************************************************************** |
361 | * Streaming decompression - HowTo |
362 | * |
363 | * A ZSTD_DStream object is required to track streaming operations. |
364 | * Use ZSTD_createDStream() and ZSTD_freeDStream() to create/release resources. |
365 | * ZSTD_DStream objects can be re-used multiple times. |
366 | * |
367 | * Use ZSTD_initDStream() to start a new decompression operation. |
368 | * @return : recommended first input size |
369 | * Alternatively, use advanced API to set specific properties. |
370 | * |
371 | * Use ZSTD_decompressStream() repetitively to consume your input. |
372 | * The function will update both `pos` fields. |
373 | * If `input.pos < input.size`, some input has not been consumed. |
374 | * It's up to the caller to present again remaining data. |
375 | * The function tries to flush all data decoded immediately, respecting output buffer size. |
376 | * If `output.pos < output.size`, decoder has flushed everything it could. |
377 | * But if `output.pos == output.size`, there might be some data left within internal buffers., |
378 | * In which case, call ZSTD_decompressStream() again to flush whatever remains in the buffer. |
379 | * Note : with no additional input provided, amount of data flushed is necessarily <= ZSTD_BLOCKSIZE_MAX. |
380 | * @return : 0 when a frame is completely decoded and fully flushed, |
381 | * or an error code, which can be tested using ZSTD_isError(), |
382 | * or any other value > 0, which means there is still some decoding or flushing to do to complete current frame : |
383 | * the return value is a suggested next input size (just a hint for better latency) |
384 | * that will never request more than the remaining frame size. |
385 | * *******************************************************************************/ |
386 | |
387 | typedef ZSTD_DCtx ZSTD_DStream; /**< DCtx and DStream are now effectively same object (>= v1.3.0) */ |
388 | /* For compatibility with versions <= v1.2.0, prefer differentiating them. */ |
389 | /*===== ZSTD_DStream management functions =====*/ |
390 | ZSTDLIB_API ZSTD_DStream* ZSTD_createDStream(void); |
391 | ZSTDLIB_API size_t ZSTD_freeDStream(ZSTD_DStream* zds); |
392 | |
393 | /*===== Streaming decompression functions =====*/ |
394 | ZSTDLIB_API size_t ZSTD_initDStream(ZSTD_DStream* zds); |
395 | ZSTDLIB_API size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inBuffer* input); |
396 | |
397 | ZSTDLIB_API size_t ZSTD_DStreamInSize(void); /*!< recommended size for input buffer */ |
398 | ZSTDLIB_API size_t ZSTD_DStreamOutSize(void); /*!< recommended size for output buffer. Guarantee to successfully flush at least one complete block in all circumstances. */ |
399 | |
400 | #endif /* ZSTD_H_235446 */ |
401 | |
402 | |
403 | |
404 | |
405 | /**************************************************************************************** |
406 | * ADVANCED AND EXPERIMENTAL FUNCTIONS |
407 | **************************************************************************************** |
408 | * The definitions in the following section are considered experimental. |
409 | * They are provided for advanced scenarios. |
410 | * They should never be used with a dynamic library, as prototypes may change in the future. |
411 | * Use them only in association with static linking. |
412 | * ***************************************************************************************/ |
413 | |
414 | #if defined(ZSTD_STATIC_LINKING_ONLY) && !defined(ZSTD_H_ZSTD_STATIC_LINKING_ONLY) |
415 | #define ZSTD_H_ZSTD_STATIC_LINKING_ONLY |
416 | |
417 | |
418 | /**************************************************************************************** |
419 | * Candidate API for promotion to stable status |
420 | **************************************************************************************** |
421 | * The following symbols and constants form the "staging area" : |
422 | * they are considered to join "stable API" by v1.4.0. |
423 | * The proposal is written so that it can be made stable "as is", |
424 | * though it's still possible to suggest improvements. |
425 | * Staging is in fact last chance for changes, |
426 | * the API is locked once reaching "stable" status. |
427 | * ***************************************************************************************/ |
428 | |
429 | |
430 | /* === Constants === */ |
431 | |
432 | /* all magic numbers are supposed read/written to/from files/memory using little-endian convention */ |
433 | #define ZSTD_MAGICNUMBER 0xFD2FB528 /* valid since v0.8.0 */ |
434 | #define ZSTD_MAGIC_DICTIONARY 0xEC30A437 /* valid since v0.7.0 */ |
435 | #define ZSTD_MAGIC_SKIPPABLE_START 0x184D2A50 /* all 16 values, from 0x184D2A50 to 0x184D2A5F, signal the beginning of a skippable frame */ |
436 | #define ZSTD_MAGIC_SKIPPABLE_MASK 0xFFFFFFF0 |
437 | |
438 | #define ZSTD_BLOCKSIZELOG_MAX 17 |
439 | #define ZSTD_BLOCKSIZE_MAX (1<<ZSTD_BLOCKSIZELOG_MAX) |
440 | |
441 | |
442 | /* === query limits === */ |
443 | |
444 | ZSTDLIB_API int ZSTD_minCLevel(void); /*!< minimum negative compression level allowed */ |
445 | |
446 | |
447 | /* === frame size === */ |
448 | |
449 | /*! ZSTD_findFrameCompressedSize() : |
450 | * `src` should point to the start of a ZSTD frame or skippable frame. |
451 | * `srcSize` must be >= first frame size |
452 | * @return : the compressed size of the first frame starting at `src`, |
453 | * suitable to pass as `srcSize` to `ZSTD_decompress` or similar, |
454 | * or an error code if input is invalid */ |
455 | ZSTDLIB_API size_t ZSTD_findFrameCompressedSize(const void* src, size_t srcSize); |
456 | |
457 | |
458 | /* === Memory management === */ |
459 | |
460 | /*! ZSTD_sizeof_*() : |
461 | * These functions give the _current_ memory usage of selected object. |
462 | * Note that object memory usage can evolve (increase or decrease) over time. */ |
463 | ZSTDLIB_API size_t ZSTD_sizeof_CCtx(const ZSTD_CCtx* cctx); |
464 | ZSTDLIB_API size_t ZSTD_sizeof_DCtx(const ZSTD_DCtx* dctx); |
465 | ZSTDLIB_API size_t ZSTD_sizeof_CStream(const ZSTD_CStream* zcs); |
466 | ZSTDLIB_API size_t ZSTD_sizeof_DStream(const ZSTD_DStream* zds); |
467 | ZSTDLIB_API size_t ZSTD_sizeof_CDict(const ZSTD_CDict* cdict); |
468 | ZSTDLIB_API size_t ZSTD_sizeof_DDict(const ZSTD_DDict* ddict); |
469 | |
470 | |
471 | /*************************************** |
472 | * Advanced compression API |
473 | ***************************************/ |
474 | |
475 | /* API design : |
476 | * Parameters are pushed one by one into an existing context, |
477 | * using ZSTD_CCtx_set*() functions. |
478 | * Pushed parameters are sticky : they are valid for next compressed frame, and any subsequent frame. |
479 | * "sticky" parameters are applicable to `ZSTD_compress2()` and `ZSTD_compressStream*()` ! |
480 | * They do not apply to "simple" one-shot variants such as ZSTD_compressCCtx() |
481 | * |
482 | * It's possible to reset all parameters to "default" using ZSTD_CCtx_reset(). |
483 | * |
484 | * This API supercedes all other "advanced" API entry points in the experimental section. |
485 | * In the future, we expect to remove from experimental API entry points which are redundant with this API. |
486 | */ |
487 | |
488 | |
489 | /* Compression strategies, listed from fastest to strongest */ |
490 | typedef enum { ZSTD_fast=1, |
491 | ZSTD_dfast=2, |
492 | ZSTD_greedy=3, |
493 | ZSTD_lazy=4, |
494 | ZSTD_lazy2=5, |
495 | ZSTD_btlazy2=6, |
496 | ZSTD_btopt=7, |
497 | ZSTD_btultra=8, |
498 | ZSTD_btultra2=9 |
499 | /* note : new strategies _might_ be added in the future. |
500 | Only the order (from fast to strong) is guaranteed */ |
501 | } ZSTD_strategy; |
502 | |
503 | |
504 | typedef enum { |
505 | |
506 | /* compression parameters */ |
507 | ZSTD_c_compressionLevel=100, /* Update all compression parameters according to pre-defined cLevel table |
508 | * Default level is ZSTD_CLEVEL_DEFAULT==3. |
509 | * Special: value 0 means default, which is controlled by ZSTD_CLEVEL_DEFAULT. |
510 | * Note 1 : it's possible to pass a negative compression level. |
511 | * Note 2 : setting a level sets all default values of other compression parameters */ |
512 | ZSTD_c_windowLog=101, /* Maximum allowed back-reference distance, expressed as power of 2. |
513 | * Must be clamped between ZSTD_WINDOWLOG_MIN and ZSTD_WINDOWLOG_MAX. |
514 | * Special: value 0 means "use default windowLog". |
515 | * Note: Using a windowLog greater than ZSTD_WINDOWLOG_LIMIT_DEFAULT |
516 | * requires explicitly allowing such window size at decompression stage if using streaming. */ |
517 | ZSTD_c_hashLog=102, /* Size of the initial probe table, as a power of 2. |
518 | * Resulting memory usage is (1 << (hashLog+2)). |
519 | * Must be clamped between ZSTD_HASHLOG_MIN and ZSTD_HASHLOG_MAX. |
520 | * Larger tables improve compression ratio of strategies <= dFast, |
521 | * and improve speed of strategies > dFast. |
522 | * Special: value 0 means "use default hashLog". */ |
523 | ZSTD_c_chainLog=103, /* Size of the multi-probe search table, as a power of 2. |
524 | * Resulting memory usage is (1 << (chainLog+2)). |
525 | * Must be clamped between ZSTD_CHAINLOG_MIN and ZSTD_CHAINLOG_MAX. |
526 | * Larger tables result in better and slower compression. |
527 | * This parameter is useless when using "fast" strategy. |
528 | * It's still useful when using "dfast" strategy, |
529 | * in which case it defines a secondary probe table. |
530 | * Special: value 0 means "use default chainLog". */ |
531 | ZSTD_c_searchLog=104, /* Number of search attempts, as a power of 2. |
532 | * More attempts result in better and slower compression. |
533 | * This parameter is useless when using "fast" and "dFast" strategies. |
534 | * Special: value 0 means "use default searchLog". */ |
535 | ZSTD_c_minMatch=105, /* Minimum size of searched matches. |
536 | * Note that Zstandard can still find matches of smaller size, |
537 | * it just tweaks its search algorithm to look for this size and larger. |
538 | * Larger values increase compression and decompression speed, but decrease ratio. |
539 | * Must be clamped between ZSTD_MINMATCH_MIN and ZSTD_MINMATCH_MAX. |
540 | * Note that currently, for all strategies < btopt, effective minimum is 4. |
541 | * , for all strategies > fast, effective maximum is 6. |
542 | * Special: value 0 means "use default minMatchLength". */ |
543 | ZSTD_c_targetLength=106, /* Impact of this field depends on strategy. |
544 | * For strategies btopt, btultra & btultra2: |
545 | * Length of Match considered "good enough" to stop search. |
546 | * Larger values make compression stronger, and slower. |
547 | * For strategy fast: |
548 | * Distance between match sampling. |
549 | * Larger values make compression faster, and weaker. |
550 | * Special: value 0 means "use default targetLength". */ |
551 | ZSTD_c_strategy=107, /* See ZSTD_strategy enum definition. |
552 | * The higher the value of selected strategy, the more complex it is, |
553 | * resulting in stronger and slower compression. |
554 | * Special: value 0 means "use default strategy". */ |
555 | |
556 | /* LDM mode parameters */ |
557 | ZSTD_c_enableLongDistanceMatching=160, /* Enable long distance matching. |
558 | * This parameter is designed to improve compression ratio |
559 | * for large inputs, by finding large matches at long distance. |
560 | * It increases memory usage and window size. |
561 | * Note: enabling this parameter increases default ZSTD_c_windowLog to 128 MB |
562 | * except when expressly set to a different value. */ |
563 | ZSTD_c_ldmHashLog=161, /* Size of the table for long distance matching, as a power of 2. |
564 | * Larger values increase memory usage and compression ratio, |
565 | * but decrease compression speed. |
566 | * Must be clamped between ZSTD_HASHLOG_MIN and ZSTD_HASHLOG_MAX |
567 | * default: windowlog - 7. |
568 | * Special: value 0 means "automatically determine hashlog". */ |
569 | ZSTD_c_ldmMinMatch=162, /* Minimum match size for long distance matcher. |
570 | * Larger/too small values usually decrease compression ratio. |
571 | * Must be clamped between ZSTD_LDM_MINMATCH_MIN and ZSTD_LDM_MINMATCH_MAX. |
572 | * Special: value 0 means "use default value" (default: 64). */ |
573 | ZSTD_c_ldmBucketSizeLog=163, /* Log size of each bucket in the LDM hash table for collision resolution. |
574 | * Larger values improve collision resolution but decrease compression speed. |
575 | * The maximum value is ZSTD_LDM_BUCKETSIZELOG_MAX. |
576 | * Special: value 0 means "use default value" (default: 3). */ |
577 | ZSTD_c_ldmHashRateLog=164, /* Frequency of inserting/looking up entries into the LDM hash table. |
578 | * Must be clamped between 0 and (ZSTD_WINDOWLOG_MAX - ZSTD_HASHLOG_MIN). |
579 | * Default is MAX(0, (windowLog - ldmHashLog)), optimizing hash table usage. |
580 | * Larger values improve compression speed. |
581 | * Deviating far from default value will likely result in a compression ratio decrease. |
582 | * Special: value 0 means "automatically determine hashRateLog". */ |
583 | |
584 | /* frame parameters */ |
585 | ZSTD_c_contentSizeFlag=200, /* Content size will be written into frame header _whenever known_ (default:1) |
586 | * Content size must be known at the beginning of compression. |
587 | * This is automatically the case when using ZSTD_compress2(), |
588 | * For streaming variants, content size must be provided with ZSTD_CCtx_setPledgedSrcSize() */ |
589 | ZSTD_c_checksumFlag=201, /* A 32-bits checksum of content is written at end of frame (default:0) */ |
590 | ZSTD_c_dictIDFlag=202, /* When applicable, dictionary's ID is written into frame header (default:1) */ |
591 | |
592 | /* multi-threading parameters */ |
593 | /* These parameters are only useful if multi-threading is enabled (compiled with build macro ZSTD_MULTITHREAD). |
594 | * They return an error otherwise. */ |
595 | ZSTD_c_nbWorkers=400, /* Select how many threads will be spawned to compress in parallel. |
596 | * When nbWorkers >= 1, triggers asynchronous mode when used with ZSTD_compressStream*() : |
597 | * ZSTD_compressStream*() consumes input and flush output if possible, but immediately gives back control to caller, |
598 | * while compression work is performed in parallel, within worker threads. |
599 | * (note : a strong exception to this rule is when first invocation of ZSTD_compressStream2() sets ZSTD_e_end : |
600 | * in which case, ZSTD_compressStream2() delegates to ZSTD_compress2(), which is always a blocking call). |
601 | * More workers improve speed, but also increase memory usage. |
602 | * Default value is `0`, aka "single-threaded mode" : no worker is spawned, compression is performed inside Caller's thread, all invocations are blocking */ |
603 | ZSTD_c_jobSize=401, /* Size of a compression job. This value is enforced only when nbWorkers >= 1. |
604 | * Each compression job is completed in parallel, so this value can indirectly impact the nb of active threads. |
605 | * 0 means default, which is dynamically determined based on compression parameters. |
606 | * Job size must be a minimum of overlap size, or 1 MB, whichever is largest. |
607 | * The minimum size is automatically and transparently enforced */ |
608 | ZSTD_c_overlapLog=402, /* Control the overlap size, as a fraction of window size. |
609 | * The overlap size is an amount of data reloaded from previous job at the beginning of a new job. |
610 | * It helps preserve compression ratio, while each job is compressed in parallel. |
611 | * This value is enforced only when nbWorkers >= 1. |
612 | * Larger values increase compression ratio, but decrease speed. |
613 | * Possible values range from 0 to 9 : |
614 | * - 0 means "default" : value will be determined by the library, depending on strategy |
615 | * - 1 means "no overlap" |
616 | * - 9 means "full overlap", using a full window size. |
617 | * Each intermediate rank increases/decreases load size by a factor 2 : |
618 | * 9: full window; 8: w/2; 7: w/4; 6: w/8; 5:w/16; 4: w/32; 3:w/64; 2:w/128; 1:no overlap; 0:default |
619 | * default value varies between 6 and 9, depending on strategy */ |
620 | |
621 | /* note : additional experimental parameters are also available |
622 | * within the experimental section of the API. |
623 | * At the time of this writing, they include : |
624 | * ZSTD_c_rsyncable |
625 | * ZSTD_c_format |
626 | * ZSTD_c_forceMaxWindow |
627 | * ZSTD_c_forceAttachDict |
628 | * Because they are not stable, it's necessary to define ZSTD_STATIC_LINKING_ONLY to access them. |
629 | * note : never ever use experimentalParam? names directly; |
630 | * also, the enums values themselves are unstable and can still change. |
631 | */ |
632 | ZSTD_c_experimentalParam1=500, |
633 | ZSTD_c_experimentalParam2=10, |
634 | ZSTD_c_experimentalParam3=1000, |
635 | ZSTD_c_experimentalParam4=1001 |
636 | } ZSTD_cParameter; |
637 | |
638 | |
639 | typedef struct { |
640 | size_t error; |
641 | int lowerBound; |
642 | int upperBound; |
643 | } ZSTD_bounds; |
644 | |
645 | /*! ZSTD_cParam_getBounds() : |
646 | * All parameters must belong to an interval with lower and upper bounds, |
647 | * otherwise they will either trigger an error or be automatically clamped. |
648 | * @return : a structure, ZSTD_bounds, which contains |
649 | * - an error status field, which must be tested using ZSTD_isError() |
650 | * - lower and upper bounds, both inclusive |
651 | */ |
652 | ZSTDLIB_API ZSTD_bounds ZSTD_cParam_getBounds(ZSTD_cParameter cParam); |
653 | |
654 | /*! ZSTD_CCtx_setParameter() : |
655 | * Set one compression parameter, selected by enum ZSTD_cParameter. |
656 | * All parameters have valid bounds. Bounds can be queried using ZSTD_cParam_getBounds(). |
657 | * Providing a value beyond bound will either clamp it, or trigger an error (depending on parameter). |
658 | * Setting a parameter is generally only possible during frame initialization (before starting compression). |
659 | * Exception : when using multi-threading mode (nbWorkers >= 1), |
660 | * the following parameters can be updated _during_ compression (within same frame): |
661 | * => compressionLevel, hashLog, chainLog, searchLog, minMatch, targetLength and strategy. |
662 | * new parameters will be active for next job only (after a flush()). |
663 | * @return : an error code (which can be tested using ZSTD_isError()). |
664 | */ |
665 | ZSTDLIB_API size_t ZSTD_CCtx_setParameter(ZSTD_CCtx* cctx, ZSTD_cParameter param, int value); |
666 | |
667 | /*! ZSTD_CCtx_setPledgedSrcSize() : |
668 | * Total input data size to be compressed as a single frame. |
669 | * Value will be written in frame header, unless if explicitly forbidden using ZSTD_c_contentSizeFlag. |
670 | * This value will also be controlled at end of frame, and trigger an error if not respected. |
671 | * @result : 0, or an error code (which can be tested with ZSTD_isError()). |
672 | * Note 1 : pledgedSrcSize==0 actually means zero, aka an empty frame. |
673 | * In order to mean "unknown content size", pass constant ZSTD_CONTENTSIZE_UNKNOWN. |
674 | * ZSTD_CONTENTSIZE_UNKNOWN is default value for any new frame. |
675 | * Note 2 : pledgedSrcSize is only valid once, for the next frame. |
676 | * It's discarded at the end of the frame, and replaced by ZSTD_CONTENTSIZE_UNKNOWN. |
677 | * Note 3 : Whenever all input data is provided and consumed in a single round, |
678 | * for example with ZSTD_compress2(), |
679 | * or invoking immediately ZSTD_compressStream2(,,,ZSTD_e_end), |
680 | * this value is automatically overriden by srcSize instead. |
681 | */ |
682 | ZSTDLIB_API size_t ZSTD_CCtx_setPledgedSrcSize(ZSTD_CCtx* cctx, unsigned long long pledgedSrcSize); |
683 | |
684 | /*! ZSTD_CCtx_loadDictionary() : |
685 | * Create an internal CDict from `dict` buffer. |
686 | * Decompression will have to use same dictionary. |
687 | * @result : 0, or an error code (which can be tested with ZSTD_isError()). |
688 | * Special: Loading a NULL (or 0-size) dictionary invalidates previous dictionary, |
689 | * meaning "return to no-dictionary mode". |
690 | * Note 1 : Dictionary is sticky, it will be used for all future compressed frames. |
691 | * To return to "no-dictionary" situation, load a NULL dictionary (or reset parameters). |
692 | * Note 2 : Loading a dictionary involves building tables. |
693 | * It's also a CPU consuming operation, with non-negligible impact on latency. |
694 | * Tables are dependent on compression parameters, and for this reason, |
695 | * compression parameters can no longer be changed after loading a dictionary. |
696 | * Note 3 :`dict` content will be copied internally. |
697 | * Use experimental ZSTD_CCtx_loadDictionary_byReference() to reference content instead. |
698 | * In such a case, dictionary buffer must outlive its users. |
699 | * Note 4 : Use ZSTD_CCtx_loadDictionary_advanced() |
700 | * to precisely select how dictionary content must be interpreted. */ |
701 | ZSTDLIB_API size_t ZSTD_CCtx_loadDictionary(ZSTD_CCtx* cctx, const void* dict, size_t dictSize); |
702 | |
703 | /*! ZSTD_CCtx_refCDict() : |
704 | * Reference a prepared dictionary, to be used for all next compressed frames. |
705 | * Note that compression parameters are enforced from within CDict, |
706 | * and supercede any compression parameter previously set within CCtx. |
707 | * The dictionary will remain valid for future compressed frames using same CCtx. |
708 | * @result : 0, or an error code (which can be tested with ZSTD_isError()). |
709 | * Special : Referencing a NULL CDict means "return to no-dictionary mode". |
710 | * Note 1 : Currently, only one dictionary can be managed. |
711 | * Referencing a new dictionary effectively "discards" any previous one. |
712 | * Note 2 : CDict is just referenced, its lifetime must outlive its usage within CCtx. */ |
713 | ZSTDLIB_API size_t ZSTD_CCtx_refCDict(ZSTD_CCtx* cctx, const ZSTD_CDict* cdict); |
714 | |
715 | /*! ZSTD_CCtx_refPrefix() : |
716 | * Reference a prefix (single-usage dictionary) for next compressed frame. |
717 | * A prefix is **only used once**. Tables are discarded at end of frame (ZSTD_e_end). |
718 | * Decompression will need same prefix to properly regenerate data. |
719 | * Compressing with a prefix is similar in outcome as performing a diff and compressing it, |
720 | * but performs much faster, especially during decompression (compression speed is tunable with compression level). |
721 | * @result : 0, or an error code (which can be tested with ZSTD_isError()). |
722 | * Special: Adding any prefix (including NULL) invalidates any previous prefix or dictionary |
723 | * Note 1 : Prefix buffer is referenced. It **must** outlive compression. |
724 | * Its content must remain unmodified during compression. |
725 | * Note 2 : If the intention is to diff some large src data blob with some prior version of itself, |
726 | * ensure that the window size is large enough to contain the entire source. |
727 | * See ZSTD_c_windowLog. |
728 | * Note 3 : Referencing a prefix involves building tables, which are dependent on compression parameters. |
729 | * It's a CPU consuming operation, with non-negligible impact on latency. |
730 | * If there is a need to use the same prefix multiple times, consider loadDictionary instead. |
731 | * Note 4 : By default, the prefix is interpreted as raw content (ZSTD_dm_rawContent). |
732 | * Use experimental ZSTD_CCtx_refPrefix_advanced() to alter dictionary interpretation. */ |
733 | ZSTDLIB_API size_t ZSTD_CCtx_refPrefix(ZSTD_CCtx* cctx, |
734 | const void* prefix, size_t prefixSize); |
735 | |
736 | |
737 | typedef enum { |
738 | ZSTD_reset_session_only = 1, |
739 | ZSTD_reset_parameters = 2, |
740 | ZSTD_reset_session_and_parameters = 3 |
741 | } ZSTD_ResetDirective; |
742 | |
743 | /*! ZSTD_CCtx_reset() : |
744 | * There are 2 different things that can be reset, independently or jointly : |
745 | * - The session : will stop compressing current frame, and make CCtx ready to start a new one. |
746 | * Useful after an error, or to interrupt any ongoing compression. |
747 | * Any internal data not yet flushed is cancelled. |
748 | * Compression parameters and dictionary remain unchanged. |
749 | * They will be used to compress next frame. |
750 | * Resetting session never fails. |
751 | * - The parameters : changes all parameters back to "default". |
752 | * This removes any reference to any dictionary too. |
753 | * Parameters can only be changed between 2 sessions (i.e. no compression is currently ongoing) |
754 | * otherwise the reset fails, and function returns an error value (which can be tested using ZSTD_isError()) |
755 | * - Both : similar to resetting the session, followed by resetting parameters. |
756 | */ |
757 | ZSTDLIB_API size_t ZSTD_CCtx_reset(ZSTD_CCtx* cctx, ZSTD_ResetDirective reset); |
758 | |
759 | |
760 | |
761 | /*! ZSTD_compress2() : |
762 | * Behave the same as ZSTD_compressCCtx(), but compression parameters are set using the advanced API. |
763 | * ZSTD_compress2() always starts a new frame. |
764 | * Should cctx hold data from a previously unfinished frame, everything about it is forgotten. |
765 | * - Compression parameters are pushed into CCtx before starting compression, using ZSTD_CCtx_set*() |
766 | * - The function is always blocking, returns when compression is completed. |
767 | * Hint : compression runs faster if `dstCapacity` >= `ZSTD_compressBound(srcSize)`. |
768 | * @return : compressed size written into `dst` (<= `dstCapacity), |
769 | * or an error code if it fails (which can be tested using ZSTD_isError()). |
770 | */ |
771 | ZSTDLIB_API size_t ZSTD_compress2( ZSTD_CCtx* cctx, |
772 | void* dst, size_t dstCapacity, |
773 | const void* src, size_t srcSize); |
774 | |
775 | typedef enum { |
776 | ZSTD_e_continue=0, /* collect more data, encoder decides when to output compressed result, for optimal compression ratio */ |
777 | ZSTD_e_flush=1, /* flush any data provided so far, |
778 | * it creates (at least) one new block, that can be decoded immediately on reception; |
779 | * frame will continue: any future data can still reference previously compressed data, improving compression. */ |
780 | ZSTD_e_end=2 /* flush any remaining data _and_ close current frame. |
781 | * note that frame is only closed after compressed data is fully flushed (return value == 0). |
782 | * After that point, any additional data starts a new frame. |
783 | * note : each frame is independent (does not reference any content from previous frame). */ |
784 | } ZSTD_EndDirective; |
785 | |
786 | /*! ZSTD_compressStream2() : |
787 | * Behaves about the same as ZSTD_compressStream, with additional control on end directive. |
788 | * - Compression parameters are pushed into CCtx before starting compression, using ZSTD_CCtx_set*() |
789 | * - Compression parameters cannot be changed once compression is started (save a list of exceptions in multi-threading mode) |
790 | * - outpot->pos must be <= dstCapacity, input->pos must be <= srcSize |
791 | * - outpot->pos and input->pos will be updated. They are guaranteed to remain below their respective limit. |
792 | * - When nbWorkers==0 (default), function is blocking : it completes its job before returning to caller. |
793 | * - When nbWorkers>=1, function is non-blocking : it just acquires a copy of input, and distributes jobs to internal worker threads, flush whatever is available, |
794 | * and then immediately returns, just indicating that there is some data remaining to be flushed. |
795 | * The function nonetheless guarantees forward progress : it will return only after it reads or write at least 1+ byte. |
796 | * - Exception : if the first call requests a ZSTD_e_end directive and provides enough dstCapacity, the function delegates to ZSTD_compress2() which is always blocking. |
797 | * - @return provides a minimum amount of data remaining to be flushed from internal buffers |
798 | * or an error code, which can be tested using ZSTD_isError(). |
799 | * if @return != 0, flush is not fully completed, there is still some data left within internal buffers. |
800 | * This is useful for ZSTD_e_flush, since in this case more flushes are necessary to empty all buffers. |
801 | * For ZSTD_e_end, @return == 0 when internal buffers are fully flushed and frame is completed. |
802 | * - after a ZSTD_e_end directive, if internal buffer is not fully flushed (@return != 0), |
803 | * only ZSTD_e_end or ZSTD_e_flush operations are allowed. |
804 | * Before starting a new compression job, or changing compression parameters, |
805 | * it is required to fully flush internal buffers. |
806 | */ |
807 | ZSTDLIB_API size_t ZSTD_compressStream2( ZSTD_CCtx* cctx, |
808 | ZSTD_outBuffer* output, |
809 | ZSTD_inBuffer* input, |
810 | ZSTD_EndDirective endOp); |
811 | |
812 | |
813 | |
814 | /* ============================== */ |
815 | /* Advanced decompression API */ |
816 | /* ============================== */ |
817 | |
818 | /* The advanced API pushes parameters one by one into an existing DCtx context. |
819 | * Parameters are sticky, and remain valid for all following frames |
820 | * using the same DCtx context. |
821 | * It's possible to reset parameters to default values using ZSTD_DCtx_reset(). |
822 | * Note : This API is compatible with existing ZSTD_decompressDCtx() and ZSTD_decompressStream(). |
823 | * Therefore, no new decompression function is necessary. |
824 | */ |
825 | |
826 | |
827 | typedef enum { |
828 | |
829 | ZSTD_d_windowLogMax=100, /* Select a size limit (in power of 2) beyond which |
830 | * the streaming API will refuse to allocate memory buffer |
831 | * in order to protect the host from unreasonable memory requirements. |
832 | * This parameter is only useful in streaming mode, since no internal buffer is allocated in single-pass mode. |
833 | * By default, a decompression context accepts window sizes <= (1 << ZSTD_WINDOWLOG_LIMIT_DEFAULT) */ |
834 | |
835 | /* note : additional experimental parameters are also available |
836 | * within the experimental section of the API. |
837 | * At the time of this writing, they include : |
838 | * ZSTD_c_format |
839 | * Because they are not stable, it's necessary to define ZSTD_STATIC_LINKING_ONLY to access them. |
840 | * note : never ever use experimentalParam? names directly |
841 | */ |
842 | ZSTD_d_experimentalParam1=1000 |
843 | |
844 | } ZSTD_dParameter; |
845 | |
846 | |
847 | /*! ZSTD_dParam_getBounds() : |
848 | * All parameters must belong to an interval with lower and upper bounds, |
849 | * otherwise they will either trigger an error or be automatically clamped. |
850 | * @return : a structure, ZSTD_bounds, which contains |
851 | * - an error status field, which must be tested using ZSTD_isError() |
852 | * - both lower and upper bounds, inclusive |
853 | */ |
854 | ZSTDLIB_API ZSTD_bounds ZSTD_dParam_getBounds(ZSTD_dParameter dParam); |
855 | |
856 | /*! ZSTD_DCtx_setParameter() : |
857 | * Set one compression parameter, selected by enum ZSTD_dParameter. |
858 | * All parameters have valid bounds. Bounds can be queried using ZSTD_dParam_getBounds(). |
859 | * Providing a value beyond bound will either clamp it, or trigger an error (depending on parameter). |
860 | * Setting a parameter is only possible during frame initialization (before starting decompression). |
861 | * @return : 0, or an error code (which can be tested using ZSTD_isError()). |
862 | */ |
863 | ZSTDLIB_API size_t ZSTD_DCtx_setParameter(ZSTD_DCtx* dctx, ZSTD_dParameter param, int value); |
864 | |
865 | |
866 | /*! ZSTD_DCtx_loadDictionary() : |
867 | * Create an internal DDict from dict buffer, |
868 | * to be used to decompress next frames. |
869 | * The dictionary remains valid for all future frames, until explicitly invalidated. |
870 | * @result : 0, or an error code (which can be tested with ZSTD_isError()). |
871 | * Special : Adding a NULL (or 0-size) dictionary invalidates any previous dictionary, |
872 | * meaning "return to no-dictionary mode". |
873 | * Note 1 : Loading a dictionary involves building tables, |
874 | * which has a non-negligible impact on CPU usage and latency. |
875 | * It's recommended to "load once, use many times", to amortize the cost |
876 | * Note 2 :`dict` content will be copied internally, so `dict` can be released after loading. |
877 | * Use ZSTD_DCtx_loadDictionary_byReference() to reference dictionary content instead. |
878 | * Note 3 : Use ZSTD_DCtx_loadDictionary_advanced() to take control of |
879 | * how dictionary content is loaded and interpreted. |
880 | */ |
881 | ZSTDLIB_API size_t ZSTD_DCtx_loadDictionary(ZSTD_DCtx* dctx, const void* dict, size_t dictSize); |
882 | |
883 | /*! ZSTD_DCtx_refDDict() : |
884 | * Reference a prepared dictionary, to be used to decompress next frames. |
885 | * The dictionary remains active for decompression of future frames using same DCtx. |
886 | * @result : 0, or an error code (which can be tested with ZSTD_isError()). |
887 | * Note 1 : Currently, only one dictionary can be managed. |
888 | * Referencing a new dictionary effectively "discards" any previous one. |
889 | * Special: referencing a NULL DDict means "return to no-dictionary mode". |
890 | * Note 2 : DDict is just referenced, its lifetime must outlive its usage from DCtx. |
891 | */ |
892 | ZSTDLIB_API size_t ZSTD_DCtx_refDDict(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict); |
893 | |
894 | /*! ZSTD_DCtx_refPrefix() : |
895 | * Reference a prefix (single-usage dictionary) to decompress next frame. |
896 | * This is the reverse operation of ZSTD_CCtx_refPrefix(), |
897 | * and must use the same prefix as the one used during compression. |
898 | * Prefix is **only used once**. Reference is discarded at end of frame. |
899 | * End of frame is reached when ZSTD_decompressStream() returns 0. |
900 | * @result : 0, or an error code (which can be tested with ZSTD_isError()). |
901 | * Note 1 : Adding any prefix (including NULL) invalidates any previously set prefix or dictionary |
902 | * Note 2 : Prefix buffer is referenced. It **must** outlive decompression. |
903 | * Prefix buffer must remain unmodified up to the end of frame, |
904 | * reached when ZSTD_decompressStream() returns 0. |
905 | * Note 3 : By default, the prefix is treated as raw content (ZSTD_dm_rawContent). |
906 | * Use ZSTD_CCtx_refPrefix_advanced() to alter dictMode (Experimental section) |
907 | * Note 4 : Referencing a raw content prefix has almost no cpu nor memory cost. |
908 | * A full dictionary is more costly, as it requires building tables. |
909 | */ |
910 | ZSTDLIB_API size_t ZSTD_DCtx_refPrefix(ZSTD_DCtx* dctx, |
911 | const void* prefix, size_t prefixSize); |
912 | |
913 | /*! ZSTD_DCtx_reset() : |
914 | * Return a DCtx to clean state. |
915 | * Session and parameters can be reset jointly or separately. |
916 | * Parameters can only be reset when no active frame is being decompressed. |
917 | * @return : 0, or an error code, which can be tested with ZSTD_isError() |
918 | */ |
919 | ZSTDLIB_API size_t ZSTD_DCtx_reset(ZSTD_DCtx* dctx, ZSTD_ResetDirective reset); |
920 | |
921 | |
922 | |
923 | /**************************************************************************************** |
924 | * experimental API (static linking only) |
925 | **************************************************************************************** |
926 | * The following symbols and constants |
927 | * are not planned to join "stable API" status in the near future. |
928 | * They can still change in future versions. |
929 | * Some of them are planned to remain in the static_only section indefinitely. |
930 | * Some of them might be removed in the future (especially when redundant with existing stable functions) |
931 | * ***************************************************************************************/ |
932 | |
933 | #define ZSTD_FRAMEHEADERSIZE_PREFIX 5 /* minimum input size required to query frame header size */ |
934 | #define ZSTD_FRAMEHEADERSIZE_MIN 6 |
935 | #define ZSTD_FRAMEHEADERSIZE_MAX 18 /* can be useful for static allocation */ |
936 | #define ZSTD_SKIPPABLEHEADERSIZE 8 |
937 | |
938 | /* compression parameter bounds */ |
939 | #define ZSTD_WINDOWLOG_MAX_32 30 |
940 | #define ZSTD_WINDOWLOG_MAX_64 31 |
941 | #define ZSTD_WINDOWLOG_MAX ((int)(sizeof(size_t) == 4 ? ZSTD_WINDOWLOG_MAX_32 : ZSTD_WINDOWLOG_MAX_64)) |
942 | #define ZSTD_WINDOWLOG_MIN 10 |
943 | #define ZSTD_HASHLOG_MAX ((ZSTD_WINDOWLOG_MAX < 30) ? ZSTD_WINDOWLOG_MAX : 30) |
944 | #define ZSTD_HASHLOG_MIN 6 |
945 | #define ZSTD_CHAINLOG_MAX_32 29 |
946 | #define ZSTD_CHAINLOG_MAX_64 30 |
947 | #define ZSTD_CHAINLOG_MAX ((int)(sizeof(size_t) == 4 ? ZSTD_CHAINLOG_MAX_32 : ZSTD_CHAINLOG_MAX_64)) |
948 | #define ZSTD_CHAINLOG_MIN ZSTD_HASHLOG_MIN |
949 | #define ZSTD_SEARCHLOG_MAX (ZSTD_WINDOWLOG_MAX-1) |
950 | #define ZSTD_SEARCHLOG_MIN 1 |
951 | #define ZSTD_MINMATCH_MAX 7 /* only for ZSTD_fast, other strategies are limited to 6 */ |
952 | #define ZSTD_MINMATCH_MIN 3 /* only for ZSTD_btopt+, faster strategies are limited to 4 */ |
953 | #define ZSTD_TARGETLENGTH_MAX ZSTD_BLOCKSIZE_MAX |
954 | #define ZSTD_TARGETLENGTH_MIN 0 /* note : comparing this constant to an unsigned results in a tautological test */ |
955 | #define ZSTD_STRATEGY_MIN ZSTD_fast |
956 | #define ZSTD_STRATEGY_MAX ZSTD_btultra2 |
957 | |
958 | |
959 | #define ZSTD_OVERLAPLOG_MIN 0 |
960 | #define ZSTD_OVERLAPLOG_MAX 9 |
961 | |
962 | #define ZSTD_WINDOWLOG_LIMIT_DEFAULT 27 /* by default, the streaming decoder will refuse any frame |
963 | * requiring larger than (1<<ZSTD_WINDOWLOG_LIMIT_DEFAULT) window size, |
964 | * to preserve host's memory from unreasonable requirements. |
965 | * This limit can be overriden using ZSTD_DCtx_setParameter(,ZSTD_d_windowLogMax,). |
966 | * The limit does not apply for one-pass decoders (such as ZSTD_decompress()), since no additional memory is allocated */ |
967 | |
968 | |
969 | /* LDM parameter bounds */ |
970 | #define ZSTD_LDM_HASHLOG_MIN ZSTD_HASHLOG_MIN |
971 | #define ZSTD_LDM_HASHLOG_MAX ZSTD_HASHLOG_MAX |
972 | #define ZSTD_LDM_MINMATCH_MIN 4 |
973 | #define ZSTD_LDM_MINMATCH_MAX 4096 |
974 | #define ZSTD_LDM_BUCKETSIZELOG_MIN 1 |
975 | #define ZSTD_LDM_BUCKETSIZELOG_MAX 8 |
976 | #define ZSTD_LDM_HASHRATELOG_MIN 0 |
977 | #define ZSTD_LDM_HASHRATELOG_MAX (ZSTD_WINDOWLOG_MAX - ZSTD_HASHLOG_MIN) |
978 | |
979 | /* internal */ |
980 | #define ZSTD_HASHLOG3_MAX 17 |
981 | |
982 | |
983 | /* --- Advanced types --- */ |
984 | |
985 | typedef struct ZSTD_CCtx_params_s ZSTD_CCtx_params; |
986 | |
987 | typedef struct { |
988 | unsigned windowLog; /**< largest match distance : larger == more compression, more memory needed during decompression */ |
989 | unsigned chainLog; /**< fully searched segment : larger == more compression, slower, more memory (useless for fast) */ |
990 | unsigned hashLog; /**< dispatch table : larger == faster, more memory */ |
991 | unsigned searchLog; /**< nb of searches : larger == more compression, slower */ |
992 | unsigned minMatch; /**< match length searched : larger == faster decompression, sometimes less compression */ |
993 | unsigned targetLength; /**< acceptable match size for optimal parser (only) : larger == more compression, slower */ |
994 | ZSTD_strategy strategy; /**< see ZSTD_strategy definition above */ |
995 | } ZSTD_compressionParameters; |
996 | |
997 | typedef struct { |
998 | int contentSizeFlag; /**< 1: content size will be in frame header (when known) */ |
999 | int checksumFlag; /**< 1: generate a 32-bits checksum using XXH64 algorithm at end of frame, for error detection */ |
1000 | int noDictIDFlag; /**< 1: no dictID will be saved into frame header (dictID is only useful for dictionary compression) */ |
1001 | } ZSTD_frameParameters; |
1002 | |
1003 | typedef struct { |
1004 | ZSTD_compressionParameters cParams; |
1005 | ZSTD_frameParameters fParams; |
1006 | } ZSTD_parameters; |
1007 | |
1008 | typedef enum { |
1009 | ZSTD_dct_auto = 0, /* dictionary is "full" when starting with ZSTD_MAGIC_DICTIONARY, otherwise it is "rawContent" */ |
1010 | ZSTD_dct_rawContent = 1, /* ensures dictionary is always loaded as rawContent, even if it starts with ZSTD_MAGIC_DICTIONARY */ |
1011 | ZSTD_dct_fullDict = 2 /* refuses to load a dictionary if it does not respect Zstandard's specification, starting with ZSTD_MAGIC_DICTIONARY */ |
1012 | } ZSTD_dictContentType_e; |
1013 | |
1014 | typedef enum { |
1015 | ZSTD_dlm_byCopy = 0, /**< Copy dictionary content internally */ |
1016 | ZSTD_dlm_byRef = 1, /**< Reference dictionary content -- the dictionary buffer must outlive its users. */ |
1017 | } ZSTD_dictLoadMethod_e; |
1018 | |
1019 | typedef enum { |
1020 | /* Opened question : should we have a format ZSTD_f_auto ? |
1021 | * Today, it would mean exactly the same as ZSTD_f_zstd1. |
1022 | * But, in the future, should several formats become supported, |
1023 | * on the compression side, it would mean "default format". |
1024 | * On the decompression side, it would mean "automatic format detection", |
1025 | * so that ZSTD_f_zstd1 would mean "accept *only* zstd frames". |
1026 | * Since meaning is a little different, another option could be to define different enums for compression and decompression. |
1027 | * This question could be kept for later, when there are actually multiple formats to support, |
1028 | * but there is also the question of pinning enum values, and pinning value `0` is especially important */ |
1029 | ZSTD_f_zstd1 = 0, /* zstd frame format, specified in zstd_compression_format.md (default) */ |
1030 | ZSTD_f_zstd1_magicless = 1, /* Variant of zstd frame format, without initial 4-bytes magic number. |
1031 | * Useful to save 4 bytes per generated frame. |
1032 | * Decoder cannot recognise automatically this format, requiring this instruction. */ |
1033 | } ZSTD_format_e; |
1034 | |
1035 | typedef enum { |
1036 | /* Note: this enum and the behavior it controls are effectively internal |
1037 | * implementation details of the compressor. They are expected to continue |
1038 | * to evolve and should be considered only in the context of extremely |
1039 | * advanced performance tuning. |
1040 | * |
1041 | * Zstd currently supports the use of a CDict in two ways: |
1042 | * |
1043 | * - The contents of the CDict can be copied into the working context. This |
1044 | * means that the compression can search both the dictionary and input |
1045 | * while operating on a single set of internal tables. This makes |
1046 | * the compression faster per-byte of input. However, the initial copy of |
1047 | * the CDict's tables incurs a fixed cost at the beginning of the |
1048 | * compression. For small compressions (< 8 KB), that copy can dominate |
1049 | * the cost of the compression. |
1050 | * |
1051 | * - The CDict's tables can be used in-place. In this model, compression is |
1052 | * slower per input byte, because the compressor has to search two sets of |
1053 | * tables. However, this model incurs no start-up cost (as long as the |
1054 | * working context's tables can be reused). For small inputs, this can be |
1055 | * faster than copying the CDict's tables. |
1056 | * |
1057 | * Zstd has a simple internal heuristic that selects which strategy to use |
1058 | * at the beginning of a compression. However, if experimentation shows that |
1059 | * Zstd is making poor choices, it is possible to override that choice with |
1060 | * this enum. |
1061 | */ |
1062 | ZSTD_dictDefaultAttach = 0, /* Use the default heuristic. */ |
1063 | ZSTD_dictForceAttach = 1, /* Never copy the dictionary. */ |
1064 | ZSTD_dictForceCopy = 2, /* Always copy the dictionary. */ |
1065 | } ZSTD_dictAttachPref_e; |
1066 | |
1067 | |
1068 | /*************************************** |
1069 | * Frame size functions |
1070 | ***************************************/ |
1071 | |
1072 | /*! ZSTD_findDecompressedSize() : |
1073 | * `src` should point the start of a series of ZSTD encoded and/or skippable frames |
1074 | * `srcSize` must be the _exact_ size of this series |
1075 | * (i.e. there should be a frame boundary exactly at `srcSize` bytes after `src`) |
1076 | * @return : - decompressed size of all data in all successive frames |
1077 | * - if the decompressed size cannot be determined: ZSTD_CONTENTSIZE_UNKNOWN |
1078 | * - if an error occurred: ZSTD_CONTENTSIZE_ERROR |
1079 | * |
1080 | * note 1 : decompressed size is an optional field, that may not be present, especially in streaming mode. |
1081 | * When `return==ZSTD_CONTENTSIZE_UNKNOWN`, data to decompress could be any size. |
1082 | * In which case, it's necessary to use streaming mode to decompress data. |
1083 | * note 2 : decompressed size is always present when compression is done with ZSTD_compress() |
1084 | * note 3 : decompressed size can be very large (64-bits value), |
1085 | * potentially larger than what local system can handle as a single memory segment. |
1086 | * In which case, it's necessary to use streaming mode to decompress data. |
1087 | * note 4 : If source is untrusted, decompressed size could be wrong or intentionally modified. |
1088 | * Always ensure result fits within application's authorized limits. |
1089 | * Each application can set its own limits. |
1090 | * note 5 : ZSTD_findDecompressedSize handles multiple frames, and so it must traverse the input to |
1091 | * read each contained frame header. This is fast as most of the data is skipped, |
1092 | * however it does mean that all frame data must be present and valid. */ |
1093 | ZSTDLIB_API unsigned long long ZSTD_findDecompressedSize(const void* src, size_t srcSize); |
1094 | |
1095 | /*! ZSTD_frameHeaderSize() : |
1096 | * srcSize must be >= ZSTD_FRAMEHEADERSIZE_PREFIX. |
1097 | * @return : size of the Frame Header, |
1098 | * or an error code (if srcSize is too small) */ |
1099 | ZSTDLIB_API size_t ZSTD_frameHeaderSize(const void* src, size_t srcSize); |
1100 | |
1101 | |
1102 | /*************************************** |
1103 | * Memory management |
1104 | ***************************************/ |
1105 | |
1106 | /*! ZSTD_estimate*() : |
1107 | * These functions make it possible to estimate memory usage |
1108 | * of a future {D,C}Ctx, before its creation. |
1109 | * ZSTD_estimateCCtxSize() will provide a budget large enough for any compression level up to selected one. |
1110 | * It will also consider src size to be arbitrarily "large", which is worst case. |
1111 | * If srcSize is known to always be small, ZSTD_estimateCCtxSize_usingCParams() can provide a tighter estimation. |
1112 | * ZSTD_estimateCCtxSize_usingCParams() can be used in tandem with ZSTD_getCParams() to create cParams from compressionLevel. |
1113 | * ZSTD_estimateCCtxSize_usingCCtxParams() can be used in tandem with ZSTD_CCtxParam_setParameter(). Only single-threaded compression is supported. This function will return an error code if ZSTD_c_nbWorkers is >= 1. |
1114 | * Note : CCtx size estimation is only correct for single-threaded compression. */ |
1115 | ZSTDLIB_API size_t ZSTD_estimateCCtxSize(int compressionLevel); |
1116 | ZSTDLIB_API size_t ZSTD_estimateCCtxSize_usingCParams(ZSTD_compressionParameters cParams); |
1117 | ZSTDLIB_API size_t ZSTD_estimateCCtxSize_usingCCtxParams(const ZSTD_CCtx_params* params); |
1118 | ZSTDLIB_API size_t ZSTD_estimateDCtxSize(void); |
1119 | |
1120 | /*! ZSTD_estimateCStreamSize() : |
1121 | * ZSTD_estimateCStreamSize() will provide a budget large enough for any compression level up to selected one. |
1122 | * It will also consider src size to be arbitrarily "large", which is worst case. |
1123 | * If srcSize is known to always be small, ZSTD_estimateCStreamSize_usingCParams() can provide a tighter estimation. |
1124 | * ZSTD_estimateCStreamSize_usingCParams() can be used in tandem with ZSTD_getCParams() to create cParams from compressionLevel. |
1125 | * ZSTD_estimateCStreamSize_usingCCtxParams() can be used in tandem with ZSTD_CCtxParam_setParameter(). Only single-threaded compression is supported. This function will return an error code if ZSTD_c_nbWorkers is >= 1. |
1126 | * Note : CStream size estimation is only correct for single-threaded compression. |
1127 | * ZSTD_DStream memory budget depends on window Size. |
1128 | * This information can be passed manually, using ZSTD_estimateDStreamSize, |
1129 | * or deducted from a valid frame Header, using ZSTD_estimateDStreamSize_fromFrame(); |
1130 | * Note : if streaming is init with function ZSTD_init?Stream_usingDict(), |
1131 | * an internal ?Dict will be created, which additional size is not estimated here. |
1132 | * In this case, get total size by adding ZSTD_estimate?DictSize */ |
1133 | ZSTDLIB_API size_t ZSTD_estimateCStreamSize(int compressionLevel); |
1134 | ZSTDLIB_API size_t ZSTD_estimateCStreamSize_usingCParams(ZSTD_compressionParameters cParams); |
1135 | ZSTDLIB_API size_t ZSTD_estimateCStreamSize_usingCCtxParams(const ZSTD_CCtx_params* params); |
1136 | ZSTDLIB_API size_t ZSTD_estimateDStreamSize(size_t windowSize); |
1137 | ZSTDLIB_API size_t ZSTD_estimateDStreamSize_fromFrame(const void* src, size_t srcSize); |
1138 | |
1139 | /*! ZSTD_estimate?DictSize() : |
1140 | * ZSTD_estimateCDictSize() will bet that src size is relatively "small", and content is copied, like ZSTD_createCDict(). |
1141 | * ZSTD_estimateCDictSize_advanced() makes it possible to control compression parameters precisely, like ZSTD_createCDict_advanced(). |
1142 | * Note : dictionaries created by reference (`ZSTD_dlm_byRef`) are logically smaller. |
1143 | */ |
1144 | ZSTDLIB_API size_t ZSTD_estimateCDictSize(size_t dictSize, int compressionLevel); |
1145 | ZSTDLIB_API size_t ZSTD_estimateCDictSize_advanced(size_t dictSize, ZSTD_compressionParameters cParams, ZSTD_dictLoadMethod_e dictLoadMethod); |
1146 | ZSTDLIB_API size_t ZSTD_estimateDDictSize(size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod); |
1147 | |
1148 | /*! ZSTD_initStatic*() : |
1149 | * Initialize an object using a pre-allocated fixed-size buffer. |
1150 | * workspace: The memory area to emplace the object into. |
1151 | * Provided pointer *must be 8-bytes aligned*. |
1152 | * Buffer must outlive object. |
1153 | * workspaceSize: Use ZSTD_estimate*Size() to determine |
1154 | * how large workspace must be to support target scenario. |
1155 | * @return : pointer to object (same address as workspace, just different type), |
1156 | * or NULL if error (size too small, incorrect alignment, etc.) |
1157 | * Note : zstd will never resize nor malloc() when using a static buffer. |
1158 | * If the object requires more memory than available, |
1159 | * zstd will just error out (typically ZSTD_error_memory_allocation). |
1160 | * Note 2 : there is no corresponding "free" function. |
1161 | * Since workspace is allocated externally, it must be freed externally too. |
1162 | * Note 3 : cParams : use ZSTD_getCParams() to convert a compression level |
1163 | * into its associated cParams. |
1164 | * Limitation 1 : currently not compatible with internal dictionary creation, triggered by |
1165 | * ZSTD_CCtx_loadDictionary(), ZSTD_initCStream_usingDict() or ZSTD_initDStream_usingDict(). |
1166 | * Limitation 2 : static cctx currently not compatible with multi-threading. |
1167 | * Limitation 3 : static dctx is incompatible with legacy support. |
1168 | */ |
1169 | ZSTDLIB_API ZSTD_CCtx* ZSTD_initStaticCCtx(void* workspace, size_t workspaceSize); |
1170 | ZSTDLIB_API ZSTD_CStream* ZSTD_initStaticCStream(void* workspace, size_t workspaceSize); /**< same as ZSTD_initStaticCCtx() */ |
1171 | |
1172 | ZSTDLIB_API ZSTD_DCtx* ZSTD_initStaticDCtx(void* workspace, size_t workspaceSize); |
1173 | ZSTDLIB_API ZSTD_DStream* ZSTD_initStaticDStream(void* workspace, size_t workspaceSize); /**< same as ZSTD_initStaticDCtx() */ |
1174 | |
1175 | ZSTDLIB_API const ZSTD_CDict* ZSTD_initStaticCDict( |
1176 | void* workspace, size_t workspaceSize, |
1177 | const void* dict, size_t dictSize, |
1178 | ZSTD_dictLoadMethod_e dictLoadMethod, |
1179 | ZSTD_dictContentType_e dictContentType, |
1180 | ZSTD_compressionParameters cParams); |
1181 | |
1182 | ZSTDLIB_API const ZSTD_DDict* ZSTD_initStaticDDict( |
1183 | void* workspace, size_t workspaceSize, |
1184 | const void* dict, size_t dictSize, |
1185 | ZSTD_dictLoadMethod_e dictLoadMethod, |
1186 | ZSTD_dictContentType_e dictContentType); |
1187 | |
1188 | |
1189 | /*! Custom memory allocation : |
1190 | * These prototypes make it possible to pass your own allocation/free functions. |
1191 | * ZSTD_customMem is provided at creation time, using ZSTD_create*_advanced() variants listed below. |
1192 | * All allocation/free operations will be completed using these custom variants instead of regular <stdlib.h> ones. |
1193 | */ |
1194 | typedef void* (*ZSTD_allocFunction) (void* opaque, size_t size); |
1195 | typedef void (*ZSTD_freeFunction) (void* opaque, void* address); |
1196 | typedef struct { ZSTD_allocFunction customAlloc; ZSTD_freeFunction customFree; void* opaque; } ZSTD_customMem; |
1197 | static ZSTD_customMem const ZSTD_defaultCMem = { NULL, NULL, NULL }; /**< this constant defers to stdlib's functions */ |
1198 | |
1199 | ZSTDLIB_API ZSTD_CCtx* ZSTD_createCCtx_advanced(ZSTD_customMem customMem); |
1200 | ZSTDLIB_API ZSTD_CStream* ZSTD_createCStream_advanced(ZSTD_customMem customMem); |
1201 | ZSTDLIB_API ZSTD_DCtx* ZSTD_createDCtx_advanced(ZSTD_customMem customMem); |
1202 | ZSTDLIB_API ZSTD_DStream* ZSTD_createDStream_advanced(ZSTD_customMem customMem); |
1203 | |
1204 | ZSTDLIB_API ZSTD_CDict* ZSTD_createCDict_advanced(const void* dict, size_t dictSize, |
1205 | ZSTD_dictLoadMethod_e dictLoadMethod, |
1206 | ZSTD_dictContentType_e dictContentType, |
1207 | ZSTD_compressionParameters cParams, |
1208 | ZSTD_customMem customMem); |
1209 | |
1210 | ZSTDLIB_API ZSTD_DDict* ZSTD_createDDict_advanced(const void* dict, size_t dictSize, |
1211 | ZSTD_dictLoadMethod_e dictLoadMethod, |
1212 | ZSTD_dictContentType_e dictContentType, |
1213 | ZSTD_customMem customMem); |
1214 | |
1215 | |
1216 | |
1217 | /*************************************** |
1218 | * Advanced compression functions |
1219 | ***************************************/ |
1220 | |
1221 | /*! ZSTD_createCDict_byReference() : |
1222 | * Create a digested dictionary for compression |
1223 | * Dictionary content is just referenced, not duplicated. |
1224 | * As a consequence, `dictBuffer` **must** outlive CDict, |
1225 | * and its content must remain unmodified throughout the lifetime of CDict. */ |
1226 | ZSTDLIB_API ZSTD_CDict* ZSTD_createCDict_byReference(const void* dictBuffer, size_t dictSize, int compressionLevel); |
1227 | |
1228 | /*! ZSTD_getCParams() : |
1229 | * @return ZSTD_compressionParameters structure for a selected compression level and estimated srcSize. |
1230 | * `estimatedSrcSize` value is optional, select 0 if not known */ |
1231 | ZSTDLIB_API ZSTD_compressionParameters ZSTD_getCParams(int compressionLevel, unsigned long long estimatedSrcSize, size_t dictSize); |
1232 | |
1233 | /*! ZSTD_getParams() : |
1234 | * same as ZSTD_getCParams(), but @return a full `ZSTD_parameters` object instead of sub-component `ZSTD_compressionParameters`. |
1235 | * All fields of `ZSTD_frameParameters` are set to default : contentSize=1, checksum=0, noDictID=0 */ |
1236 | ZSTDLIB_API ZSTD_parameters ZSTD_getParams(int compressionLevel, unsigned long long estimatedSrcSize, size_t dictSize); |
1237 | |
1238 | /*! ZSTD_checkCParams() : |
1239 | * Ensure param values remain within authorized range */ |
1240 | ZSTDLIB_API size_t ZSTD_checkCParams(ZSTD_compressionParameters params); |
1241 | |
1242 | /*! ZSTD_adjustCParams() : |
1243 | * optimize params for a given `srcSize` and `dictSize`. |
1244 | * both values are optional, select `0` if unknown. */ |
1245 | ZSTDLIB_API ZSTD_compressionParameters ZSTD_adjustCParams(ZSTD_compressionParameters cPar, unsigned long long srcSize, size_t dictSize); |
1246 | |
1247 | /*! ZSTD_compress_advanced() : |
1248 | * Same as ZSTD_compress_usingDict(), with fine-tune control over compression parameters (by structure) */ |
1249 | ZSTDLIB_API size_t ZSTD_compress_advanced(ZSTD_CCtx* cctx, |
1250 | void* dst, size_t dstCapacity, |
1251 | const void* src, size_t srcSize, |
1252 | const void* dict,size_t dictSize, |
1253 | ZSTD_parameters params); |
1254 | |
1255 | /*! ZSTD_compress_usingCDict_advanced() : |
1256 | * Same as ZSTD_compress_usingCDict(), with fine-tune control over frame parameters */ |
1257 | ZSTDLIB_API size_t ZSTD_compress_usingCDict_advanced(ZSTD_CCtx* cctx, |
1258 | void* dst, size_t dstCapacity, |
1259 | const void* src, size_t srcSize, |
1260 | const ZSTD_CDict* cdict, |
1261 | ZSTD_frameParameters fParams); |
1262 | |
1263 | |
1264 | /*! ZSTD_CCtx_loadDictionary_byReference() : |
1265 | * Same as ZSTD_CCtx_loadDictionary(), but dictionary content is referenced, instead of being copied into CCtx. |
1266 | * It saves some memory, but also requires that `dict` outlives its usage within `cctx` */ |
1267 | ZSTDLIB_API size_t ZSTD_CCtx_loadDictionary_byReference(ZSTD_CCtx* cctx, const void* dict, size_t dictSize); |
1268 | |
1269 | /*! ZSTD_CCtx_loadDictionary_advanced() : |
1270 | * Same as ZSTD_CCtx_loadDictionary(), but gives finer control over |
1271 | * how to load the dictionary (by copy ? by reference ?) |
1272 | * and how to interpret it (automatic ? force raw mode ? full mode only ?) */ |
1273 | ZSTDLIB_API size_t ZSTD_CCtx_loadDictionary_advanced(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod, ZSTD_dictContentType_e dictContentType); |
1274 | |
1275 | /*! ZSTD_CCtx_refPrefix_advanced() : |
1276 | * Same as ZSTD_CCtx_refPrefix(), but gives finer control over |
1277 | * how to interpret prefix content (automatic ? force raw mode (default) ? full mode only ?) */ |
1278 | ZSTDLIB_API size_t ZSTD_CCtx_refPrefix_advanced(ZSTD_CCtx* cctx, const void* prefix, size_t prefixSize, ZSTD_dictContentType_e dictContentType); |
1279 | |
1280 | /* === experimental parameters === */ |
1281 | /* these parameters can be used with ZSTD_setParameter() |
1282 | * they are not guaranteed to remain supported in the future */ |
1283 | |
1284 | /* Enables rsyncable mode, |
1285 | * which makes compressed files more rsync friendly |
1286 | * by adding periodic synchronization points to the compressed data. |
1287 | * The target average block size is ZSTD_c_jobSize / 2. |
1288 | * It's possible to modify the job size to increase or decrease |
1289 | * the granularity of the synchronization point. |
1290 | * Once the jobSize is smaller than the window size, |
1291 | * it will result in compression ratio degradation. |
1292 | * NOTE 1: rsyncable mode only works when multithreading is enabled. |
1293 | * NOTE 2: rsyncable performs poorly in combination with long range mode, |
1294 | * since it will decrease the effectiveness of synchronization points, |
1295 | * though mileage may vary. |
1296 | * NOTE 3: Rsyncable mode limits maximum compression speed to ~400 MB/s. |
1297 | * If the selected compression level is already running significantly slower, |
1298 | * the overall speed won't be significantly impacted. |
1299 | */ |
1300 | #define ZSTD_c_rsyncable ZSTD_c_experimentalParam1 |
1301 | |
1302 | /* Select a compression format. |
1303 | * The value must be of type ZSTD_format_e. |
1304 | * See ZSTD_format_e enum definition for details */ |
1305 | #define ZSTD_c_format ZSTD_c_experimentalParam2 |
1306 | |
1307 | /* Force back-reference distances to remain < windowSize, |
1308 | * even when referencing into Dictionary content (default:0) */ |
1309 | #define ZSTD_c_forceMaxWindow ZSTD_c_experimentalParam3 |
1310 | |
1311 | /* Controls whether the contents of a CDict |
1312 | * are used in place, or copied into the working context. |
1313 | * Accepts values from the ZSTD_dictAttachPref_e enum. |
1314 | * See the comments on that enum for an explanation of the feature. */ |
1315 | #define ZSTD_c_forceAttachDict ZSTD_c_experimentalParam4 |
1316 | |
1317 | /*! ZSTD_CCtx_getParameter() : |
1318 | * Get the requested compression parameter value, selected by enum ZSTD_cParameter, |
1319 | * and store it into int* value. |
1320 | * @return : 0, or an error code (which can be tested with ZSTD_isError()). |
1321 | */ |
1322 | ZSTDLIB_API size_t ZSTD_CCtx_getParameter(ZSTD_CCtx* cctx, ZSTD_cParameter param, int* value); |
1323 | |
1324 | |
1325 | /*! ZSTD_CCtx_params : |
1326 | * Quick howto : |
1327 | * - ZSTD_createCCtxParams() : Create a ZSTD_CCtx_params structure |
1328 | * - ZSTD_CCtxParam_setParameter() : Push parameters one by one into |
1329 | * an existing ZSTD_CCtx_params structure. |
1330 | * This is similar to |
1331 | * ZSTD_CCtx_setParameter(). |
1332 | * - ZSTD_CCtx_setParametersUsingCCtxParams() : Apply parameters to |
1333 | * an existing CCtx. |
1334 | * These parameters will be applied to |
1335 | * all subsequent frames. |
1336 | * - ZSTD_compressStream2() : Do compression using the CCtx. |
1337 | * - ZSTD_freeCCtxParams() : Free the memory. |
1338 | * |
1339 | * This can be used with ZSTD_estimateCCtxSize_advanced_usingCCtxParams() |
1340 | * for static allocation of CCtx for single-threaded compression. |
1341 | */ |
1342 | ZSTDLIB_API ZSTD_CCtx_params* ZSTD_createCCtxParams(void); |
1343 | ZSTDLIB_API size_t ZSTD_freeCCtxParams(ZSTD_CCtx_params* params); |
1344 | |
1345 | /*! ZSTD_CCtxParams_reset() : |
1346 | * Reset params to default values. |
1347 | */ |
1348 | ZSTDLIB_API size_t ZSTD_CCtxParams_reset(ZSTD_CCtx_params* params); |
1349 | |
1350 | /*! ZSTD_CCtxParams_init() : |
1351 | * Initializes the compression parameters of cctxParams according to |
1352 | * compression level. All other parameters are reset to their default values. |
1353 | */ |
1354 | ZSTDLIB_API size_t ZSTD_CCtxParams_init(ZSTD_CCtx_params* cctxParams, int compressionLevel); |
1355 | |
1356 | /*! ZSTD_CCtxParams_init_advanced() : |
1357 | * Initializes the compression and frame parameters of cctxParams according to |
1358 | * params. All other parameters are reset to their default values. |
1359 | */ |
1360 | ZSTDLIB_API size_t ZSTD_CCtxParams_init_advanced(ZSTD_CCtx_params* cctxParams, ZSTD_parameters params); |
1361 | |
1362 | /*! ZSTD_CCtxParam_setParameter() : |
1363 | * Similar to ZSTD_CCtx_setParameter. |
1364 | * Set one compression parameter, selected by enum ZSTD_cParameter. |
1365 | * Parameters must be applied to a ZSTD_CCtx using ZSTD_CCtx_setParametersUsingCCtxParams(). |
1366 | * @result : 0, or an error code (which can be tested with ZSTD_isError()). |
1367 | */ |
1368 | ZSTDLIB_API size_t ZSTD_CCtxParam_setParameter(ZSTD_CCtx_params* params, ZSTD_cParameter param, int value); |
1369 | |
1370 | /*! ZSTD_CCtxParam_getParameter() : |
1371 | * Similar to ZSTD_CCtx_getParameter. |
1372 | * Get the requested value of one compression parameter, selected by enum ZSTD_cParameter. |
1373 | * @result : 0, or an error code (which can be tested with ZSTD_isError()). |
1374 | */ |
1375 | ZSTDLIB_API size_t ZSTD_CCtxParam_getParameter(ZSTD_CCtx_params* params, ZSTD_cParameter param, int* value); |
1376 | |
1377 | /*! ZSTD_CCtx_setParametersUsingCCtxParams() : |
1378 | * Apply a set of ZSTD_CCtx_params to the compression context. |
1379 | * This can be done even after compression is started, |
1380 | * if nbWorkers==0, this will have no impact until a new compression is started. |
1381 | * if nbWorkers>=1, new parameters will be picked up at next job, |
1382 | * with a few restrictions (windowLog, pledgedSrcSize, nbWorkers, jobSize, and overlapLog are not updated). |
1383 | */ |
1384 | ZSTDLIB_API size_t ZSTD_CCtx_setParametersUsingCCtxParams( |
1385 | ZSTD_CCtx* cctx, const ZSTD_CCtx_params* params); |
1386 | |
1387 | /*! ZSTD_compressStream2_simpleArgs() : |
1388 | * Same as ZSTD_compressStream2(), |
1389 | * but using only integral types as arguments. |
1390 | * This variant might be helpful for binders from dynamic languages |
1391 | * which have troubles handling structures containing memory pointers. |
1392 | */ |
1393 | ZSTDLIB_API size_t ZSTD_compressStream2_simpleArgs ( |
1394 | ZSTD_CCtx* cctx, |
1395 | void* dst, size_t dstCapacity, size_t* dstPos, |
1396 | const void* src, size_t srcSize, size_t* srcPos, |
1397 | ZSTD_EndDirective endOp); |
1398 | |
1399 | |
1400 | /*************************************** |
1401 | * Advanced decompression functions |
1402 | ***************************************/ |
1403 | |
1404 | /*! ZSTD_isFrame() : |
1405 | * Tells if the content of `buffer` starts with a valid Frame Identifier. |
1406 | * Note : Frame Identifier is 4 bytes. If `size < 4`, @return will always be 0. |
1407 | * Note 2 : Legacy Frame Identifiers are considered valid only if Legacy Support is enabled. |
1408 | * Note 3 : Skippable Frame Identifiers are considered valid. */ |
1409 | ZSTDLIB_API unsigned ZSTD_isFrame(const void* buffer, size_t size); |
1410 | |
1411 | /*! ZSTD_createDDict_byReference() : |
1412 | * Create a digested dictionary, ready to start decompression operation without startup delay. |
1413 | * Dictionary content is referenced, and therefore stays in dictBuffer. |
1414 | * It is important that dictBuffer outlives DDict, |
1415 | * it must remain read accessible throughout the lifetime of DDict */ |
1416 | ZSTDLIB_API ZSTD_DDict* ZSTD_createDDict_byReference(const void* dictBuffer, size_t dictSize); |
1417 | |
1418 | |
1419 | /*! ZSTD_getDictID_fromDict() : |
1420 | * Provides the dictID stored within dictionary. |
1421 | * if @return == 0, the dictionary is not conformant with Zstandard specification. |
1422 | * It can still be loaded, but as a content-only dictionary. */ |
1423 | ZSTDLIB_API unsigned ZSTD_getDictID_fromDict(const void* dict, size_t dictSize); |
1424 | |
1425 | /*! ZSTD_getDictID_fromDDict() : |
1426 | * Provides the dictID of the dictionary loaded into `ddict`. |
1427 | * If @return == 0, the dictionary is not conformant to Zstandard specification, or empty. |
1428 | * Non-conformant dictionaries can still be loaded, but as content-only dictionaries. */ |
1429 | ZSTDLIB_API unsigned ZSTD_getDictID_fromDDict(const ZSTD_DDict* ddict); |
1430 | |
1431 | /*! ZSTD_getDictID_fromFrame() : |
1432 | * Provides the dictID required to decompressed the frame stored within `src`. |
1433 | * If @return == 0, the dictID could not be decoded. |
1434 | * This could for one of the following reasons : |
1435 | * - The frame does not require a dictionary to be decoded (most common case). |
1436 | * - The frame was built with dictID intentionally removed. Whatever dictionary is necessary is a hidden information. |
1437 | * Note : this use case also happens when using a non-conformant dictionary. |
1438 | * - `srcSize` is too small, and as a result, the frame header could not be decoded (only possible if `srcSize < ZSTD_FRAMEHEADERSIZE_MAX`). |
1439 | * - This is not a Zstandard frame. |
1440 | * When identifying the exact failure cause, it's possible to use ZSTD_getFrameHeader(), which will provide a more precise error code. */ |
1441 | ZSTDLIB_API unsigned ZSTD_getDictID_fromFrame(const void* src, size_t srcSize); |
1442 | |
1443 | /*! ZSTD_DCtx_loadDictionary_byReference() : |
1444 | * Same as ZSTD_DCtx_loadDictionary(), |
1445 | * but references `dict` content instead of copying it into `dctx`. |
1446 | * This saves memory if `dict` remains around., |
1447 | * However, it's imperative that `dict` remains accessible (and unmodified) while being used, so it must outlive decompression. */ |
1448 | ZSTDLIB_API size_t ZSTD_DCtx_loadDictionary_byReference(ZSTD_DCtx* dctx, const void* dict, size_t dictSize); |
1449 | |
1450 | /*! ZSTD_DCtx_loadDictionary_advanced() : |
1451 | * Same as ZSTD_DCtx_loadDictionary(), |
1452 | * but gives direct control over |
1453 | * how to load the dictionary (by copy ? by reference ?) |
1454 | * and how to interpret it (automatic ? force raw mode ? full mode only ?). */ |
1455 | ZSTDLIB_API size_t ZSTD_DCtx_loadDictionary_advanced(ZSTD_DCtx* dctx, const void* dict, size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod, ZSTD_dictContentType_e dictContentType); |
1456 | |
1457 | /*! ZSTD_DCtx_refPrefix_advanced() : |
1458 | * Same as ZSTD_DCtx_refPrefix(), but gives finer control over |
1459 | * how to interpret prefix content (automatic ? force raw mode (default) ? full mode only ?) */ |
1460 | ZSTDLIB_API size_t ZSTD_DCtx_refPrefix_advanced(ZSTD_DCtx* dctx, const void* prefix, size_t prefixSize, ZSTD_dictContentType_e dictContentType); |
1461 | |
1462 | /*! ZSTD_DCtx_setMaxWindowSize() : |
1463 | * Refuses allocating internal buffers for frames requiring a window size larger than provided limit. |
1464 | * This protects a decoder context from reserving too much memory for itself (potential attack scenario). |
1465 | * This parameter is only useful in streaming mode, since no internal buffer is allocated in single-pass mode. |
1466 | * By default, a decompression context accepts all window sizes <= (1 << ZSTD_WINDOWLOG_LIMIT_DEFAULT) |
1467 | * @return : 0, or an error code (which can be tested using ZSTD_isError()). |
1468 | */ |
1469 | ZSTDLIB_API size_t ZSTD_DCtx_setMaxWindowSize(ZSTD_DCtx* dctx, size_t maxWindowSize); |
1470 | |
1471 | /* ZSTD_d_format |
1472 | * experimental parameter, |
1473 | * allowing selection between ZSTD_format_e input compression formats |
1474 | */ |
1475 | #define ZSTD_d_format ZSTD_d_experimentalParam1 |
1476 | |
1477 | /*! ZSTD_DCtx_setFormat() : |
1478 | * Instruct the decoder context about what kind of data to decode next. |
1479 | * This instruction is mandatory to decode data without a fully-formed header, |
1480 | * such ZSTD_f_zstd1_magicless for example. |
1481 | * @return : 0, or an error code (which can be tested using ZSTD_isError()). */ |
1482 | ZSTDLIB_API size_t ZSTD_DCtx_setFormat(ZSTD_DCtx* dctx, ZSTD_format_e format); |
1483 | |
1484 | /*! ZSTD_decompressStream_simpleArgs() : |
1485 | * Same as ZSTD_decompressStream(), |
1486 | * but using only integral types as arguments. |
1487 | * This can be helpful for binders from dynamic languages |
1488 | * which have troubles handling structures containing memory pointers. |
1489 | */ |
1490 | ZSTDLIB_API size_t ZSTD_decompressStream_simpleArgs ( |
1491 | ZSTD_DCtx* dctx, |
1492 | void* dst, size_t dstCapacity, size_t* dstPos, |
1493 | const void* src, size_t srcSize, size_t* srcPos); |
1494 | |
1495 | |
1496 | /******************************************************************** |
1497 | * Advanced streaming functions |
1498 | * Warning : most of these functions are now redundant with the Advanced API. |
1499 | * Once Advanced API reaches "stable" status, |
1500 | * redundant functions will be deprecated, and then at some point removed. |
1501 | ********************************************************************/ |
1502 | |
1503 | /*===== Advanced Streaming compression functions =====*/ |
1504 | ZSTDLIB_API size_t ZSTD_initCStream_srcSize(ZSTD_CStream* zcs, int compressionLevel, unsigned long long pledgedSrcSize); /**< pledgedSrcSize must be correct. If it is not known at init time, use ZSTD_CONTENTSIZE_UNKNOWN. Note that, for compatibility with older programs, "0" also disables frame content size field. It may be enabled in the future. */ |
1505 | ZSTDLIB_API size_t ZSTD_initCStream_usingDict(ZSTD_CStream* zcs, const void* dict, size_t dictSize, int compressionLevel); /**< creates of an internal CDict (incompatible with static CCtx), except if dict == NULL or dictSize < 8, in which case no dict is used. Note: dict is loaded with ZSTD_dm_auto (treated as a full zstd dictionary if it begins with ZSTD_MAGIC_DICTIONARY, else as raw content) and ZSTD_dlm_byCopy.*/ |
1506 | ZSTDLIB_API size_t ZSTD_initCStream_advanced(ZSTD_CStream* zcs, const void* dict, size_t dictSize, |
1507 | ZSTD_parameters params, unsigned long long pledgedSrcSize); /**< pledgedSrcSize must be correct. If srcSize is not known at init time, use value ZSTD_CONTENTSIZE_UNKNOWN. dict is loaded with ZSTD_dm_auto and ZSTD_dlm_byCopy. */ |
1508 | ZSTDLIB_API size_t ZSTD_initCStream_usingCDict(ZSTD_CStream* zcs, const ZSTD_CDict* cdict); /**< note : cdict will just be referenced, and must outlive compression session */ |
1509 | ZSTDLIB_API size_t ZSTD_initCStream_usingCDict_advanced(ZSTD_CStream* zcs, const ZSTD_CDict* cdict, ZSTD_frameParameters fParams, unsigned long long pledgedSrcSize); /**< same as ZSTD_initCStream_usingCDict(), with control over frame parameters. pledgedSrcSize must be correct. If srcSize is not known at init time, use value ZSTD_CONTENTSIZE_UNKNOWN. */ |
1510 | |
1511 | /*! ZSTD_resetCStream() : |
1512 | * start a new frame, using same parameters from previous frame. |
1513 | * This is typically useful to skip dictionary loading stage, since it will re-use it in-place. |
1514 | * Note that zcs must be init at least once before using ZSTD_resetCStream(). |
1515 | * If pledgedSrcSize is not known at reset time, use macro ZSTD_CONTENTSIZE_UNKNOWN. |
1516 | * If pledgedSrcSize > 0, its value must be correct, as it will be written in header, and controlled at the end. |
1517 | * For the time being, pledgedSrcSize==0 is interpreted as "srcSize unknown" for compatibility with older programs, |
1518 | * but it will change to mean "empty" in future version, so use macro ZSTD_CONTENTSIZE_UNKNOWN instead. |
1519 | * @return : 0, or an error code (which can be tested using ZSTD_isError()) |
1520 | */ |
1521 | ZSTDLIB_API size_t ZSTD_resetCStream(ZSTD_CStream* zcs, unsigned long long pledgedSrcSize); |
1522 | |
1523 | |
1524 | typedef struct { |
1525 | unsigned long long ingested; /* nb input bytes read and buffered */ |
1526 | unsigned long long consumed; /* nb input bytes actually compressed */ |
1527 | unsigned long long produced; /* nb of compressed bytes generated and buffered */ |
1528 | unsigned long long flushed; /* nb of compressed bytes flushed : not provided; can be tracked from caller side */ |
1529 | unsigned currentJobID; /* MT only : latest started job nb */ |
1530 | unsigned nbActiveWorkers; /* MT only : nb of workers actively compressing at probe time */ |
1531 | } ZSTD_frameProgression; |
1532 | |
1533 | /* ZSTD_getFrameProgression() : |
1534 | * tells how much data has been ingested (read from input) |
1535 | * consumed (input actually compressed) and produced (output) for current frame. |
1536 | * Note : (ingested - consumed) is amount of input data buffered internally, not yet compressed. |
1537 | * Aggregates progression inside active worker threads. |
1538 | */ |
1539 | ZSTDLIB_API ZSTD_frameProgression ZSTD_getFrameProgression(const ZSTD_CCtx* cctx); |
1540 | |
1541 | /*! ZSTD_toFlushNow() : |
1542 | * Tell how many bytes are ready to be flushed immediately. |
1543 | * Useful for multithreading scenarios (nbWorkers >= 1). |
1544 | * Probe the oldest active job, defined as oldest job not yet entirely flushed, |
1545 | * and check its output buffer. |
1546 | * @return : amount of data stored in oldest job and ready to be flushed immediately. |
1547 | * if @return == 0, it means either : |
1548 | * + there is no active job (could be checked with ZSTD_frameProgression()), or |
1549 | * + oldest job is still actively compressing data, |
1550 | * but everything it has produced has also been flushed so far, |
1551 | * therefore flush speed is limited by production speed of oldest job |
1552 | * irrespective of the speed of concurrent (and newer) jobs. |
1553 | */ |
1554 | ZSTDLIB_API size_t ZSTD_toFlushNow(ZSTD_CCtx* cctx); |
1555 | |
1556 | |
1557 | /*===== Advanced Streaming decompression functions =====*/ |
1558 | ZSTDLIB_API size_t ZSTD_initDStream_usingDict(ZSTD_DStream* zds, const void* dict, size_t dictSize); /**< note: no dictionary will be used if dict == NULL or dictSize < 8 */ |
1559 | ZSTDLIB_API size_t ZSTD_initDStream_usingDDict(ZSTD_DStream* zds, const ZSTD_DDict* ddict); /**< note : ddict is referenced, it must outlive decompression session */ |
1560 | ZSTDLIB_API size_t ZSTD_resetDStream(ZSTD_DStream* zds); /**< re-use decompression parameters from previous init; saves dictionary loading */ |
1561 | |
1562 | |
1563 | /********************************************************************* |
1564 | * Buffer-less and synchronous inner streaming functions |
1565 | * |
1566 | * This is an advanced API, giving full control over buffer management, for users which need direct control over memory. |
1567 | * But it's also a complex one, with several restrictions, documented below. |
1568 | * Prefer normal streaming API for an easier experience. |
1569 | ********************************************************************* */ |
1570 | |
1571 | /** |
1572 | Buffer-less streaming compression (synchronous mode) |
1573 | |
1574 | A ZSTD_CCtx object is required to track streaming operations. |
1575 | Use ZSTD_createCCtx() / ZSTD_freeCCtx() to manage resource. |
1576 | ZSTD_CCtx object can be re-used multiple times within successive compression operations. |
1577 | |
1578 | Start by initializing a context. |
1579 | Use ZSTD_compressBegin(), or ZSTD_compressBegin_usingDict() for dictionary compression, |
1580 | or ZSTD_compressBegin_advanced(), for finer parameter control. |
1581 | It's also possible to duplicate a reference context which has already been initialized, using ZSTD_copyCCtx() |
1582 | |
1583 | Then, consume your input using ZSTD_compressContinue(). |
1584 | There are some important considerations to keep in mind when using this advanced function : |
1585 | - ZSTD_compressContinue() has no internal buffer. It uses externally provided buffers only. |
1586 | - Interface is synchronous : input is consumed entirely and produces 1+ compressed blocks. |
1587 | - Caller must ensure there is enough space in `dst` to store compressed data under worst case scenario. |
1588 | Worst case evaluation is provided by ZSTD_compressBound(). |
1589 | ZSTD_compressContinue() doesn't guarantee recover after a failed compression. |
1590 | - ZSTD_compressContinue() presumes prior input ***is still accessible and unmodified*** (up to maximum distance size, see WindowLog). |
1591 | It remembers all previous contiguous blocks, plus one separated memory segment (which can itself consists of multiple contiguous blocks) |
1592 | - ZSTD_compressContinue() detects that prior input has been overwritten when `src` buffer overlaps. |
1593 | In which case, it will "discard" the relevant memory section from its history. |
1594 | |
1595 | Finish a frame with ZSTD_compressEnd(), which will write the last block(s) and optional checksum. |
1596 | It's possible to use srcSize==0, in which case, it will write a final empty block to end the frame. |
1597 | Without last block mark, frames are considered unfinished (hence corrupted) by compliant decoders. |
1598 | |
1599 | `ZSTD_CCtx` object can be re-used (ZSTD_compressBegin()) to compress again. |
1600 | */ |
1601 | |
1602 | /*===== Buffer-less streaming compression functions =====*/ |
1603 | ZSTDLIB_API size_t ZSTD_compressBegin(ZSTD_CCtx* cctx, int compressionLevel); |
1604 | ZSTDLIB_API size_t ZSTD_compressBegin_usingDict(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, int compressionLevel); |
1605 | ZSTDLIB_API size_t ZSTD_compressBegin_advanced(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, ZSTD_parameters params, unsigned long long pledgedSrcSize); /**< pledgedSrcSize : If srcSize is not known at init time, use ZSTD_CONTENTSIZE_UNKNOWN */ |
1606 | ZSTDLIB_API size_t ZSTD_compressBegin_usingCDict(ZSTD_CCtx* cctx, const ZSTD_CDict* cdict); /**< note: fails if cdict==NULL */ |
1607 | ZSTDLIB_API size_t ZSTD_compressBegin_usingCDict_advanced(ZSTD_CCtx* const cctx, const ZSTD_CDict* const cdict, ZSTD_frameParameters const fParams, unsigned long long const pledgedSrcSize); /* compression parameters are already set within cdict. pledgedSrcSize must be correct. If srcSize is not known, use macro ZSTD_CONTENTSIZE_UNKNOWN */ |
1608 | ZSTDLIB_API size_t ZSTD_copyCCtx(ZSTD_CCtx* cctx, const ZSTD_CCtx* preparedCCtx, unsigned long long pledgedSrcSize); /**< note: if pledgedSrcSize is not known, use ZSTD_CONTENTSIZE_UNKNOWN */ |
1609 | |
1610 | ZSTDLIB_API size_t ZSTD_compressContinue(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize); |
1611 | ZSTDLIB_API size_t ZSTD_compressEnd(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize); |
1612 | |
1613 | |
1614 | /*- |
1615 | Buffer-less streaming decompression (synchronous mode) |
1616 | |
1617 | A ZSTD_DCtx object is required to track streaming operations. |
1618 | Use ZSTD_createDCtx() / ZSTD_freeDCtx() to manage it. |
1619 | A ZSTD_DCtx object can be re-used multiple times. |
1620 | |
1621 | First typical operation is to retrieve frame parameters, using ZSTD_getFrameHeader(). |
1622 | Frame header is extracted from the beginning of compressed frame, so providing only the frame's beginning is enough. |
1623 | Data fragment must be large enough to ensure successful decoding. |
1624 | `ZSTD_frameHeaderSize_max` bytes is guaranteed to always be large enough. |
1625 | @result : 0 : successful decoding, the `ZSTD_frameHeader` structure is correctly filled. |
1626 | >0 : `srcSize` is too small, please provide at least @result bytes on next attempt. |
1627 | errorCode, which can be tested using ZSTD_isError(). |
1628 | |
1629 | It fills a ZSTD_frameHeader structure with important information to correctly decode the frame, |
1630 | such as the dictionary ID, content size, or maximum back-reference distance (`windowSize`). |
1631 | Note that these values could be wrong, either because of data corruption, or because a 3rd party deliberately spoofs false information. |
1632 | As a consequence, check that values remain within valid application range. |
1633 | For example, do not allocate memory blindly, check that `windowSize` is within expectation. |
1634 | Each application can set its own limits, depending on local restrictions. |
1635 | For extended interoperability, it is recommended to support `windowSize` of at least 8 MB. |
1636 | |
1637 | ZSTD_decompressContinue() needs previous data blocks during decompression, up to `windowSize` bytes. |
1638 | ZSTD_decompressContinue() is very sensitive to contiguity, |
1639 | if 2 blocks don't follow each other, make sure that either the compressor breaks contiguity at the same place, |
1640 | or that previous contiguous segment is large enough to properly handle maximum back-reference distance. |
1641 | There are multiple ways to guarantee this condition. |
1642 | |
1643 | The most memory efficient way is to use a round buffer of sufficient size. |
1644 | Sufficient size is determined by invoking ZSTD_decodingBufferSize_min(), |
1645 | which can @return an error code if required value is too large for current system (in 32-bits mode). |
1646 | In a round buffer methodology, ZSTD_decompressContinue() decompresses each block next to previous one, |
1647 | up to the moment there is not enough room left in the buffer to guarantee decoding another full block, |
1648 | which maximum size is provided in `ZSTD_frameHeader` structure, field `blockSizeMax`. |
1649 | At which point, decoding can resume from the beginning of the buffer. |
1650 | Note that already decoded data stored in the buffer should be flushed before being overwritten. |
1651 | |
1652 | There are alternatives possible, for example using two or more buffers of size `windowSize` each, though they consume more memory. |
1653 | |
1654 | Finally, if you control the compression process, you can also ignore all buffer size rules, |
1655 | as long as the encoder and decoder progress in "lock-step", |
1656 | aka use exactly the same buffer sizes, break contiguity at the same place, etc. |
1657 | |
1658 | Once buffers are setup, start decompression, with ZSTD_decompressBegin(). |
1659 | If decompression requires a dictionary, use ZSTD_decompressBegin_usingDict() or ZSTD_decompressBegin_usingDDict(). |
1660 | |
1661 | Then use ZSTD_nextSrcSizeToDecompress() and ZSTD_decompressContinue() alternatively. |
1662 | ZSTD_nextSrcSizeToDecompress() tells how many bytes to provide as 'srcSize' to ZSTD_decompressContinue(). |
1663 | ZSTD_decompressContinue() requires this _exact_ amount of bytes, or it will fail. |
1664 | |
1665 | @result of ZSTD_decompressContinue() is the number of bytes regenerated within 'dst' (necessarily <= dstCapacity). |
1666 | It can be zero : it just means ZSTD_decompressContinue() has decoded some metadata item. |
1667 | It can also be an error code, which can be tested with ZSTD_isError(). |
1668 | |
1669 | A frame is fully decoded when ZSTD_nextSrcSizeToDecompress() returns zero. |
1670 | Context can then be reset to start a new decompression. |
1671 | |
1672 | Note : it's possible to know if next input to present is a header or a block, using ZSTD_nextInputType(). |
1673 | This information is not required to properly decode a frame. |
1674 | |
1675 | == Special case : skippable frames == |
1676 | |
1677 | Skippable frames allow integration of user-defined data into a flow of concatenated frames. |
1678 | Skippable frames will be ignored (skipped) by decompressor. |
1679 | The format of skippable frames is as follows : |
1680 | a) Skippable frame ID - 4 Bytes, Little endian format, any value from 0x184D2A50 to 0x184D2A5F |
1681 | b) Frame Size - 4 Bytes, Little endian format, unsigned 32-bits |
1682 | c) Frame Content - any content (User Data) of length equal to Frame Size |
1683 | For skippable frames ZSTD_getFrameHeader() returns zfhPtr->frameType==ZSTD_skippableFrame. |
1684 | For skippable frames ZSTD_decompressContinue() always returns 0 : it only skips the content. |
1685 | */ |
1686 | |
1687 | /*===== Buffer-less streaming decompression functions =====*/ |
1688 | typedef enum { ZSTD_frame, ZSTD_skippableFrame } ZSTD_frameType_e; |
1689 | typedef struct { |
1690 | unsigned long long frameContentSize; /* if == ZSTD_CONTENTSIZE_UNKNOWN, it means this field is not available. 0 means "empty" */ |
1691 | unsigned long long windowSize; /* can be very large, up to <= frameContentSize */ |
1692 | unsigned blockSizeMax; |
1693 | ZSTD_frameType_e frameType; /* if == ZSTD_skippableFrame, frameContentSize is the size of skippable content */ |
1694 | unsigned headerSize; |
1695 | unsigned dictID; |
1696 | unsigned checksumFlag; |
1697 | } ZSTD_frameHeader; |
1698 | |
1699 | /** ZSTD_getFrameHeader() : |
1700 | * decode Frame Header, or requires larger `srcSize`. |
1701 | * @return : 0, `zfhPtr` is correctly filled, |
1702 | * >0, `srcSize` is too small, value is wanted `srcSize` amount, |
1703 | * or an error code, which can be tested using ZSTD_isError() */ |
1704 | ZSTDLIB_API size_t ZSTD_getFrameHeader(ZSTD_frameHeader* zfhPtr, const void* src, size_t srcSize); /**< doesn't consume input */ |
1705 | /*! ZSTD_getFrameHeader_advanced() : |
1706 | * same as ZSTD_getFrameHeader(), |
1707 | * with added capability to select a format (like ZSTD_f_zstd1_magicless) */ |
1708 | ZSTDLIB_API size_t ZSTD_getFrameHeader_advanced(ZSTD_frameHeader* zfhPtr, const void* src, size_t srcSize, ZSTD_format_e format); |
1709 | ZSTDLIB_API size_t ZSTD_decodingBufferSize_min(unsigned long long windowSize, unsigned long long frameContentSize); /**< when frame content size is not known, pass in frameContentSize == ZSTD_CONTENTSIZE_UNKNOWN */ |
1710 | |
1711 | ZSTDLIB_API size_t ZSTD_decompressBegin(ZSTD_DCtx* dctx); |
1712 | ZSTDLIB_API size_t ZSTD_decompressBegin_usingDict(ZSTD_DCtx* dctx, const void* dict, size_t dictSize); |
1713 | ZSTDLIB_API size_t ZSTD_decompressBegin_usingDDict(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict); |
1714 | |
1715 | ZSTDLIB_API size_t ZSTD_nextSrcSizeToDecompress(ZSTD_DCtx* dctx); |
1716 | ZSTDLIB_API size_t ZSTD_decompressContinue(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize); |
1717 | |
1718 | /* misc */ |
1719 | ZSTDLIB_API void ZSTD_copyDCtx(ZSTD_DCtx* dctx, const ZSTD_DCtx* preparedDCtx); |
1720 | typedef enum { ZSTDnit_frameHeader, ZSTDnit_blockHeader, ZSTDnit_block, ZSTDnit_lastBlock, ZSTDnit_checksum, ZSTDnit_skippableFrame } ZSTD_nextInputType_e; |
1721 | ZSTDLIB_API ZSTD_nextInputType_e ZSTD_nextInputType(ZSTD_DCtx* dctx); |
1722 | |
1723 | |
1724 | |
1725 | |
1726 | /* ============================ */ |
1727 | /** Block level API */ |
1728 | /* ============================ */ |
1729 | |
1730 | /*! |
1731 | Block functions produce and decode raw zstd blocks, without frame metadata. |
1732 | Frame metadata cost is typically ~18 bytes, which can be non-negligible for very small blocks (< 100 bytes). |
1733 | User will have to take in charge required information to regenerate data, such as compressed and content sizes. |
1734 | |
1735 | A few rules to respect : |
1736 | - Compressing and decompressing require a context structure |
1737 | + Use ZSTD_createCCtx() and ZSTD_createDCtx() |
1738 | - It is necessary to init context before starting |
1739 | + compression : any ZSTD_compressBegin*() variant, including with dictionary |
1740 | + decompression : any ZSTD_decompressBegin*() variant, including with dictionary |
1741 | + copyCCtx() and copyDCtx() can be used too |
1742 | - Block size is limited, it must be <= ZSTD_getBlockSize() <= ZSTD_BLOCKSIZE_MAX == 128 KB |
1743 | + If input is larger than a block size, it's necessary to split input data into multiple blocks |
1744 | + For inputs larger than a single block, really consider using regular ZSTD_compress() instead. |
1745 | Frame metadata is not that costly, and quickly becomes negligible as source size grows larger. |
1746 | - When a block is considered not compressible enough, ZSTD_compressBlock() result will be zero. |
1747 | In which case, nothing is produced into `dst` ! |
1748 | + User must test for such outcome and deal directly with uncompressed data |
1749 | + ZSTD_decompressBlock() doesn't accept uncompressed data as input !!! |
1750 | + In case of multiple successive blocks, should some of them be uncompressed, |
1751 | decoder must be informed of their existence in order to follow proper history. |
1752 | Use ZSTD_insertBlock() for such a case. |
1753 | */ |
1754 | |
1755 | /*===== Raw zstd block functions =====*/ |
1756 | ZSTDLIB_API size_t ZSTD_getBlockSize (const ZSTD_CCtx* cctx); |
1757 | ZSTDLIB_API size_t ZSTD_compressBlock (ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize); |
1758 | ZSTDLIB_API size_t ZSTD_decompressBlock(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize); |
1759 | ZSTDLIB_API size_t ZSTD_insertBlock (ZSTD_DCtx* dctx, const void* blockStart, size_t blockSize); /**< insert uncompressed block into `dctx` history. Useful for multi-blocks decompression. */ |
1760 | |
1761 | |
1762 | #endif /* ZSTD_H_ZSTD_STATIC_LINKING_ONLY */ |
1763 | |
1764 | #if defined (__cplusplus) |
1765 | } |
1766 | #endif |
1767 | |