Nick Terrell | a26589d | 2017-08-09 19:35:53 -0700 | [diff] [blame] | 1 | /* |
| 2 | * Copyright (c) 2016-present, Yann Collet, Facebook, Inc. |
| 3 | * All rights reserved. |
| 4 | * |
| 5 | * This source code is licensed under the BSD-style license found in the |
| 6 | * LICENSE file in the root directory of https://github.com/facebook/zstd. |
| 7 | * An additional grant of patent rights can be found in the PATENTS file in the |
| 8 | * same directory. |
| 9 | * |
| 10 | * This program is free software; you can redistribute it and/or modify it under |
| 11 | * the terms of the GNU General Public License version 2 as published by the |
| 12 | * Free Software Foundation. This program is dual-licensed; you may select |
| 13 | * either version 2 of the GNU General Public License ("GPL") or BSD license |
| 14 | * ("BSD"). |
| 15 | */ |
| 16 | |
| 17 | #ifndef ZSTD_H |
| 18 | #define ZSTD_H |
| 19 | |
| 20 | /* ====== Dependency ======*/ |
| 21 | #include <linux/types.h> /* size_t */ |
| 22 | |
| 23 | |
| 24 | /*-***************************************************************************** |
| 25 | * Introduction |
| 26 | * |
| 27 | * zstd, short for Zstandard, is a fast lossless compression algorithm, |
| 28 | * targeting real-time compression scenarios at zlib-level and better |
| 29 | * compression ratios. The zstd compression library provides in-memory |
| 30 | * compression and decompression functions. The library supports compression |
| 31 | * levels from 1 up to ZSTD_maxCLevel() which is 22. Levels >= 20, labeled |
| 32 | * ultra, should be used with caution, as they require more memory. |
| 33 | * Compression can be done in: |
| 34 | * - a single step, reusing a context (described as Explicit memory management) |
| 35 | * - unbounded multiple steps (described as Streaming compression) |
| 36 | * The compression ratio achievable on small data can be highly improved using |
| 37 | * compression with a dictionary in: |
| 38 | * - a single step (described as Simple dictionary API) |
| 39 | * - a single step, reusing a dictionary (described as Fast dictionary API) |
| 40 | ******************************************************************************/ |
| 41 | |
| 42 | /*====== Helper functions ======*/ |
| 43 | |
| 44 | /** |
| 45 | * enum ZSTD_ErrorCode - zstd error codes |
| 46 | * |
| 47 | * Functions that return size_t can be checked for errors using ZSTD_isError() |
| 48 | * and the ZSTD_ErrorCode can be extracted using ZSTD_getErrorCode(). |
| 49 | */ |
| 50 | typedef enum { |
| 51 | ZSTD_error_no_error, |
| 52 | ZSTD_error_GENERIC, |
| 53 | ZSTD_error_prefix_unknown, |
| 54 | ZSTD_error_version_unsupported, |
| 55 | ZSTD_error_parameter_unknown, |
| 56 | ZSTD_error_frameParameter_unsupported, |
| 57 | ZSTD_error_frameParameter_unsupportedBy32bits, |
| 58 | ZSTD_error_frameParameter_windowTooLarge, |
| 59 | ZSTD_error_compressionParameter_unsupported, |
| 60 | ZSTD_error_init_missing, |
| 61 | ZSTD_error_memory_allocation, |
| 62 | ZSTD_error_stage_wrong, |
| 63 | ZSTD_error_dstSize_tooSmall, |
| 64 | ZSTD_error_srcSize_wrong, |
| 65 | ZSTD_error_corruption_detected, |
| 66 | ZSTD_error_checksum_wrong, |
| 67 | ZSTD_error_tableLog_tooLarge, |
| 68 | ZSTD_error_maxSymbolValue_tooLarge, |
| 69 | ZSTD_error_maxSymbolValue_tooSmall, |
| 70 | ZSTD_error_dictionary_corrupted, |
| 71 | ZSTD_error_dictionary_wrong, |
| 72 | ZSTD_error_dictionaryCreation_failed, |
| 73 | ZSTD_error_maxCode |
| 74 | } ZSTD_ErrorCode; |
| 75 | |
| 76 | /** |
| 77 | * ZSTD_maxCLevel() - maximum compression level available |
| 78 | * |
| 79 | * Return: Maximum compression level available. |
| 80 | */ |
| 81 | int ZSTD_maxCLevel(void); |
| 82 | /** |
| 83 | * ZSTD_compressBound() - maximum compressed size in worst case scenario |
| 84 | * @srcSize: The size of the data to compress. |
| 85 | * |
| 86 | * Return: The maximum compressed size in the worst case scenario. |
| 87 | */ |
| 88 | size_t ZSTD_compressBound(size_t srcSize); |
| 89 | /** |
| 90 | * ZSTD_isError() - tells if a size_t function result is an error code |
| 91 | * @code: The function result to check for error. |
| 92 | * |
| 93 | * Return: Non-zero iff the code is an error. |
| 94 | */ |
| 95 | static __attribute__((unused)) unsigned int ZSTD_isError(size_t code) |
| 96 | { |
| 97 | return code > (size_t)-ZSTD_error_maxCode; |
| 98 | } |
| 99 | /** |
| 100 | * ZSTD_getErrorCode() - translates an error function result to a ZSTD_ErrorCode |
| 101 | * @functionResult: The result of a function for which ZSTD_isError() is true. |
| 102 | * |
| 103 | * Return: The ZSTD_ErrorCode corresponding to the functionResult or 0 |
| 104 | * if the functionResult isn't an error. |
| 105 | */ |
| 106 | static __attribute__((unused)) ZSTD_ErrorCode ZSTD_getErrorCode( |
| 107 | size_t functionResult) |
| 108 | { |
| 109 | if (!ZSTD_isError(functionResult)) |
| 110 | return (ZSTD_ErrorCode)0; |
| 111 | return (ZSTD_ErrorCode)(0 - functionResult); |
| 112 | } |
| 113 | |
| 114 | /** |
| 115 | * enum ZSTD_strategy - zstd compression search strategy |
| 116 | * |
| 117 | * From faster to stronger. |
| 118 | */ |
| 119 | typedef enum { |
| 120 | ZSTD_fast, |
| 121 | ZSTD_dfast, |
| 122 | ZSTD_greedy, |
| 123 | ZSTD_lazy, |
| 124 | ZSTD_lazy2, |
| 125 | ZSTD_btlazy2, |
| 126 | ZSTD_btopt, |
| 127 | ZSTD_btopt2 |
| 128 | } ZSTD_strategy; |
| 129 | |
| 130 | /** |
| 131 | * struct ZSTD_compressionParameters - zstd compression parameters |
| 132 | * @windowLog: Log of the largest match distance. Larger means more |
| 133 | * compression, and more memory needed during decompression. |
| 134 | * @chainLog: Fully searched segment. Larger means more compression, slower, |
| 135 | * and more memory (useless for fast). |
| 136 | * @hashLog: Dispatch table. Larger means more compression, |
| 137 | * slower, and more memory. |
| 138 | * @searchLog: Number of searches. Larger means more compression and slower. |
| 139 | * @searchLength: Match length searched. Larger means faster decompression, |
| 140 | * sometimes less compression. |
| 141 | * @targetLength: Acceptable match size for optimal parser (only). Larger means |
| 142 | * more compression, and slower. |
| 143 | * @strategy: The zstd compression strategy. |
| 144 | */ |
| 145 | typedef struct { |
| 146 | unsigned int windowLog; |
| 147 | unsigned int chainLog; |
| 148 | unsigned int hashLog; |
| 149 | unsigned int searchLog; |
| 150 | unsigned int searchLength; |
| 151 | unsigned int targetLength; |
| 152 | ZSTD_strategy strategy; |
| 153 | } ZSTD_compressionParameters; |
| 154 | |
| 155 | /** |
| 156 | * struct ZSTD_frameParameters - zstd frame parameters |
| 157 | * @contentSizeFlag: Controls whether content size will be present in the frame |
| 158 | * header (when known). |
| 159 | * @checksumFlag: Controls whether a 32-bit checksum is generated at the end |
| 160 | * of the frame for error detection. |
| 161 | * @noDictIDFlag: Controls whether dictID will be saved into the frame header |
| 162 | * when using dictionary compression. |
| 163 | * |
| 164 | * The default value is all fields set to 0. |
| 165 | */ |
| 166 | typedef struct { |
| 167 | unsigned int contentSizeFlag; |
| 168 | unsigned int checksumFlag; |
| 169 | unsigned int noDictIDFlag; |
| 170 | } ZSTD_frameParameters; |
| 171 | |
| 172 | /** |
| 173 | * struct ZSTD_parameters - zstd parameters |
| 174 | * @cParams: The compression parameters. |
| 175 | * @fParams: The frame parameters. |
| 176 | */ |
| 177 | typedef struct { |
| 178 | ZSTD_compressionParameters cParams; |
| 179 | ZSTD_frameParameters fParams; |
| 180 | } ZSTD_parameters; |
| 181 | |
| 182 | /** |
| 183 | * ZSTD_getCParams() - returns ZSTD_compressionParameters for selected level |
| 184 | * @compressionLevel: The compression level from 1 to ZSTD_maxCLevel(). |
| 185 | * @estimatedSrcSize: The estimated source size to compress or 0 if unknown. |
| 186 | * @dictSize: The dictionary size or 0 if a dictionary isn't being used. |
| 187 | * |
| 188 | * Return: The selected ZSTD_compressionParameters. |
| 189 | */ |
| 190 | ZSTD_compressionParameters ZSTD_getCParams(int compressionLevel, |
| 191 | unsigned long long estimatedSrcSize, size_t dictSize); |
| 192 | |
| 193 | /** |
| 194 | * ZSTD_getParams() - returns ZSTD_parameters for selected level |
| 195 | * @compressionLevel: The compression level from 1 to ZSTD_maxCLevel(). |
| 196 | * @estimatedSrcSize: The estimated source size to compress or 0 if unknown. |
| 197 | * @dictSize: The dictionary size or 0 if a dictionary isn't being used. |
| 198 | * |
| 199 | * The same as ZSTD_getCParams() except also selects the default frame |
| 200 | * parameters (all zero). |
| 201 | * |
| 202 | * Return: The selected ZSTD_parameters. |
| 203 | */ |
| 204 | ZSTD_parameters ZSTD_getParams(int compressionLevel, |
| 205 | unsigned long long estimatedSrcSize, size_t dictSize); |
| 206 | |
| 207 | /*-************************************* |
| 208 | * Explicit memory management |
| 209 | **************************************/ |
| 210 | |
| 211 | /** |
| 212 | * ZSTD_CCtxWorkspaceBound() - amount of memory needed to initialize a ZSTD_CCtx |
| 213 | * @cParams: The compression parameters to be used for compression. |
| 214 | * |
| 215 | * If multiple compression parameters might be used, the caller must call |
| 216 | * ZSTD_CCtxWorkspaceBound() for each set of parameters and use the maximum |
| 217 | * size. |
| 218 | * |
| 219 | * Return: A lower bound on the size of the workspace that is passed to |
| 220 | * ZSTD_initCCtx(). |
| 221 | */ |
| 222 | size_t ZSTD_CCtxWorkspaceBound(ZSTD_compressionParameters cParams); |
| 223 | |
| 224 | /** |
| 225 | * struct ZSTD_CCtx - the zstd compression context |
| 226 | * |
| 227 | * When compressing many times it is recommended to allocate a context just once |
| 228 | * and reuse it for each successive compression operation. |
| 229 | */ |
| 230 | typedef struct ZSTD_CCtx_s ZSTD_CCtx; |
| 231 | /** |
| 232 | * ZSTD_initCCtx() - initialize a zstd compression context |
| 233 | * @workspace: The workspace to emplace the context into. It must outlive |
| 234 | * the returned context. |
| 235 | * @workspaceSize: The size of workspace. Use ZSTD_CCtxWorkspaceBound() to |
| 236 | * determine how large the workspace must be. |
| 237 | * |
| 238 | * Return: A compression context emplaced into workspace. |
| 239 | */ |
| 240 | ZSTD_CCtx *ZSTD_initCCtx(void *workspace, size_t workspaceSize); |
| 241 | |
| 242 | /** |
| 243 | * ZSTD_compressCCtx() - compress src into dst |
| 244 | * @ctx: The context. Must have been initialized with a workspace at |
| 245 | * least as large as ZSTD_CCtxWorkspaceBound(params.cParams). |
| 246 | * @dst: The buffer to compress src into. |
| 247 | * @dstCapacity: The size of the destination buffer. May be any size, but |
| 248 | * ZSTD_compressBound(srcSize) is guaranteed to be large enough. |
| 249 | * @src: The data to compress. |
| 250 | * @srcSize: The size of the data to compress. |
| 251 | * @params: The parameters to use for compression. See ZSTD_getParams(). |
| 252 | * |
| 253 | * Return: The compressed size or an error, which can be checked using |
| 254 | * ZSTD_isError(). |
| 255 | */ |
| 256 | size_t ZSTD_compressCCtx(ZSTD_CCtx *ctx, void *dst, size_t dstCapacity, |
| 257 | const void *src, size_t srcSize, ZSTD_parameters params); |
| 258 | |
| 259 | /** |
| 260 | * ZSTD_DCtxWorkspaceBound() - amount of memory needed to initialize a ZSTD_DCtx |
| 261 | * |
| 262 | * Return: A lower bound on the size of the workspace that is passed to |
| 263 | * ZSTD_initDCtx(). |
| 264 | */ |
| 265 | size_t ZSTD_DCtxWorkspaceBound(void); |
| 266 | |
| 267 | /** |
| 268 | * struct ZSTD_DCtx - the zstd decompression context |
| 269 | * |
| 270 | * When decompressing many times it is recommended to allocate a context just |
| 271 | * once and reuse it for each successive decompression operation. |
| 272 | */ |
| 273 | typedef struct ZSTD_DCtx_s ZSTD_DCtx; |
| 274 | /** |
| 275 | * ZSTD_initDCtx() - initialize a zstd decompression context |
| 276 | * @workspace: The workspace to emplace the context into. It must outlive |
| 277 | * the returned context. |
| 278 | * @workspaceSize: The size of workspace. Use ZSTD_DCtxWorkspaceBound() to |
| 279 | * determine how large the workspace must be. |
| 280 | * |
| 281 | * Return: A decompression context emplaced into workspace. |
| 282 | */ |
| 283 | ZSTD_DCtx *ZSTD_initDCtx(void *workspace, size_t workspaceSize); |
| 284 | |
| 285 | /** |
| 286 | * ZSTD_decompressDCtx() - decompress zstd compressed src into dst |
| 287 | * @ctx: The decompression context. |
| 288 | * @dst: The buffer to decompress src into. |
| 289 | * @dstCapacity: The size of the destination buffer. Must be at least as large |
| 290 | * as the decompressed size. If the caller cannot upper bound the |
| 291 | * decompressed size, then it's better to use the streaming API. |
| 292 | * @src: The zstd compressed data to decompress. Multiple concatenated |
| 293 | * frames and skippable frames are allowed. |
| 294 | * @srcSize: The exact size of the data to decompress. |
| 295 | * |
| 296 | * Return: The decompressed size or an error, which can be checked using |
| 297 | * ZSTD_isError(). |
| 298 | */ |
| 299 | size_t ZSTD_decompressDCtx(ZSTD_DCtx *ctx, void *dst, size_t dstCapacity, |
| 300 | const void *src, size_t srcSize); |
| 301 | |
| 302 | /*-************************ |
| 303 | * Simple dictionary API |
| 304 | **************************/ |
| 305 | |
| 306 | /** |
| 307 | * ZSTD_compress_usingDict() - compress src into dst using a dictionary |
| 308 | * @ctx: The context. Must have been initialized with a workspace at |
| 309 | * least as large as ZSTD_CCtxWorkspaceBound(params.cParams). |
| 310 | * @dst: The buffer to compress src into. |
| 311 | * @dstCapacity: The size of the destination buffer. May be any size, but |
| 312 | * ZSTD_compressBound(srcSize) is guaranteed to be large enough. |
| 313 | * @src: The data to compress. |
| 314 | * @srcSize: The size of the data to compress. |
| 315 | * @dict: The dictionary to use for compression. |
| 316 | * @dictSize: The size of the dictionary. |
| 317 | * @params: The parameters to use for compression. See ZSTD_getParams(). |
| 318 | * |
| 319 | * Compression using a predefined dictionary. The same dictionary must be used |
| 320 | * during decompression. |
| 321 | * |
| 322 | * Return: The compressed size or an error, which can be checked using |
| 323 | * ZSTD_isError(). |
| 324 | */ |
| 325 | size_t ZSTD_compress_usingDict(ZSTD_CCtx *ctx, void *dst, size_t dstCapacity, |
| 326 | const void *src, size_t srcSize, const void *dict, size_t dictSize, |
| 327 | ZSTD_parameters params); |
| 328 | |
| 329 | /** |
| 330 | * ZSTD_decompress_usingDict() - decompress src into dst using a dictionary |
| 331 | * @ctx: The decompression context. |
| 332 | * @dst: The buffer to decompress src into. |
| 333 | * @dstCapacity: The size of the destination buffer. Must be at least as large |
| 334 | * as the decompressed size. If the caller cannot upper bound the |
| 335 | * decompressed size, then it's better to use the streaming API. |
| 336 | * @src: The zstd compressed data to decompress. Multiple concatenated |
| 337 | * frames and skippable frames are allowed. |
| 338 | * @srcSize: The exact size of the data to decompress. |
| 339 | * @dict: The dictionary to use for decompression. The same dictionary |
| 340 | * must've been used to compress the data. |
| 341 | * @dictSize: The size of the dictionary. |
| 342 | * |
| 343 | * Return: The decompressed size or an error, which can be checked using |
| 344 | * ZSTD_isError(). |
| 345 | */ |
| 346 | size_t ZSTD_decompress_usingDict(ZSTD_DCtx *ctx, void *dst, size_t dstCapacity, |
| 347 | const void *src, size_t srcSize, const void *dict, size_t dictSize); |
| 348 | |
| 349 | /*-************************** |
| 350 | * Fast dictionary API |
| 351 | ***************************/ |
| 352 | |
| 353 | /** |
| 354 | * ZSTD_CDictWorkspaceBound() - memory needed to initialize a ZSTD_CDict |
| 355 | * @cParams: The compression parameters to be used for compression. |
| 356 | * |
| 357 | * Return: A lower bound on the size of the workspace that is passed to |
| 358 | * ZSTD_initCDict(). |
| 359 | */ |
| 360 | size_t ZSTD_CDictWorkspaceBound(ZSTD_compressionParameters cParams); |
| 361 | |
| 362 | /** |
| 363 | * struct ZSTD_CDict - a digested dictionary to be used for compression |
| 364 | */ |
| 365 | typedef struct ZSTD_CDict_s ZSTD_CDict; |
| 366 | |
| 367 | /** |
| 368 | * ZSTD_initCDict() - initialize a digested dictionary for compression |
| 369 | * @dictBuffer: The dictionary to digest. The buffer is referenced by the |
| 370 | * ZSTD_CDict so it must outlive the returned ZSTD_CDict. |
| 371 | * @dictSize: The size of the dictionary. |
| 372 | * @params: The parameters to use for compression. See ZSTD_getParams(). |
| 373 | * @workspace: The workspace. It must outlive the returned ZSTD_CDict. |
| 374 | * @workspaceSize: The workspace size. Must be at least |
| 375 | * ZSTD_CDictWorkspaceBound(params.cParams). |
| 376 | * |
| 377 | * When compressing multiple messages / blocks with the same dictionary it is |
| 378 | * recommended to load it just once. The ZSTD_CDict merely references the |
| 379 | * dictBuffer, so it must outlive the returned ZSTD_CDict. |
| 380 | * |
| 381 | * Return: The digested dictionary emplaced into workspace. |
| 382 | */ |
| 383 | ZSTD_CDict *ZSTD_initCDict(const void *dictBuffer, size_t dictSize, |
| 384 | ZSTD_parameters params, void *workspace, size_t workspaceSize); |
| 385 | |
| 386 | /** |
| 387 | * ZSTD_compress_usingCDict() - compress src into dst using a ZSTD_CDict |
| 388 | * @ctx: The context. Must have been initialized with a workspace at |
| 389 | * least as large as ZSTD_CCtxWorkspaceBound(cParams) where |
| 390 | * cParams are the compression parameters used to initialize the |
| 391 | * cdict. |
| 392 | * @dst: The buffer to compress src into. |
| 393 | * @dstCapacity: The size of the destination buffer. May be any size, but |
| 394 | * ZSTD_compressBound(srcSize) is guaranteed to be large enough. |
| 395 | * @src: The data to compress. |
| 396 | * @srcSize: The size of the data to compress. |
| 397 | * @cdict: The digested dictionary to use for compression. |
| 398 | * @params: The parameters to use for compression. See ZSTD_getParams(). |
| 399 | * |
| 400 | * Compression using a digested dictionary. The same dictionary must be used |
| 401 | * during decompression. |
| 402 | * |
| 403 | * Return: The compressed size or an error, which can be checked using |
| 404 | * ZSTD_isError(). |
| 405 | */ |
| 406 | size_t ZSTD_compress_usingCDict(ZSTD_CCtx *cctx, void *dst, size_t dstCapacity, |
| 407 | const void *src, size_t srcSize, const ZSTD_CDict *cdict); |
| 408 | |
| 409 | |
| 410 | /** |
| 411 | * ZSTD_DDictWorkspaceBound() - memory needed to initialize a ZSTD_DDict |
| 412 | * |
| 413 | * Return: A lower bound on the size of the workspace that is passed to |
| 414 | * ZSTD_initDDict(). |
| 415 | */ |
| 416 | size_t ZSTD_DDictWorkspaceBound(void); |
| 417 | |
| 418 | /** |
| 419 | * struct ZSTD_DDict - a digested dictionary to be used for decompression |
| 420 | */ |
| 421 | typedef struct ZSTD_DDict_s ZSTD_DDict; |
| 422 | |
| 423 | /** |
| 424 | * ZSTD_initDDict() - initialize a digested dictionary for decompression |
| 425 | * @dictBuffer: The dictionary to digest. The buffer is referenced by the |
| 426 | * ZSTD_DDict so it must outlive the returned ZSTD_DDict. |
| 427 | * @dictSize: The size of the dictionary. |
| 428 | * @workspace: The workspace. It must outlive the returned ZSTD_DDict. |
| 429 | * @workspaceSize: The workspace size. Must be at least |
| 430 | * ZSTD_DDictWorkspaceBound(). |
| 431 | * |
| 432 | * When decompressing multiple messages / blocks with the same dictionary it is |
| 433 | * recommended to load it just once. The ZSTD_DDict merely references the |
| 434 | * dictBuffer, so it must outlive the returned ZSTD_DDict. |
| 435 | * |
| 436 | * Return: The digested dictionary emplaced into workspace. |
| 437 | */ |
| 438 | ZSTD_DDict *ZSTD_initDDict(const void *dictBuffer, size_t dictSize, |
| 439 | void *workspace, size_t workspaceSize); |
| 440 | |
| 441 | /** |
| 442 | * ZSTD_decompress_usingDDict() - decompress src into dst using a ZSTD_DDict |
| 443 | * @ctx: The decompression context. |
| 444 | * @dst: The buffer to decompress src into. |
| 445 | * @dstCapacity: The size of the destination buffer. Must be at least as large |
| 446 | * as the decompressed size. If the caller cannot upper bound the |
| 447 | * decompressed size, then it's better to use the streaming API. |
| 448 | * @src: The zstd compressed data to decompress. Multiple concatenated |
| 449 | * frames and skippable frames are allowed. |
| 450 | * @srcSize: The exact size of the data to decompress. |
| 451 | * @ddict: The digested dictionary to use for decompression. The same |
| 452 | * dictionary must've been used to compress the data. |
| 453 | * |
| 454 | * Return: The decompressed size or an error, which can be checked using |
| 455 | * ZSTD_isError(). |
| 456 | */ |
| 457 | size_t ZSTD_decompress_usingDDict(ZSTD_DCtx *dctx, void *dst, |
| 458 | size_t dstCapacity, const void *src, size_t srcSize, |
| 459 | const ZSTD_DDict *ddict); |
| 460 | |
| 461 | |
| 462 | /*-************************** |
| 463 | * Streaming |
| 464 | ***************************/ |
| 465 | |
| 466 | /** |
| 467 | * struct ZSTD_inBuffer - input buffer for streaming |
| 468 | * @src: Start of the input buffer. |
| 469 | * @size: Size of the input buffer. |
| 470 | * @pos: Position where reading stopped. Will be updated. |
| 471 | * Necessarily 0 <= pos <= size. |
| 472 | */ |
| 473 | typedef struct ZSTD_inBuffer_s { |
| 474 | const void *src; |
| 475 | size_t size; |
| 476 | size_t pos; |
| 477 | } ZSTD_inBuffer; |
| 478 | |
| 479 | /** |
| 480 | * struct ZSTD_outBuffer - output buffer for streaming |
| 481 | * @dst: Start of the output buffer. |
| 482 | * @size: Size of the output buffer. |
| 483 | * @pos: Position where writing stopped. Will be updated. |
| 484 | * Necessarily 0 <= pos <= size. |
| 485 | */ |
| 486 | typedef struct ZSTD_outBuffer_s { |
| 487 | void *dst; |
| 488 | size_t size; |
| 489 | size_t pos; |
| 490 | } ZSTD_outBuffer; |
| 491 | |
| 492 | |
| 493 | |
| 494 | /*-***************************************************************************** |
| 495 | * Streaming compression - HowTo |
| 496 | * |
| 497 | * A ZSTD_CStream object is required to track streaming operation. |
| 498 | * Use ZSTD_initCStream() to initialize a ZSTD_CStream object. |
| 499 | * ZSTD_CStream objects can be reused multiple times on consecutive compression |
| 500 | * operations. It is recommended to re-use ZSTD_CStream in situations where many |
| 501 | * streaming operations will be achieved consecutively. Use one separate |
| 502 | * ZSTD_CStream per thread for parallel execution. |
| 503 | * |
| 504 | * Use ZSTD_compressStream() repetitively to consume input stream. |
| 505 | * The function will automatically update both `pos` fields. |
| 506 | * Note that it may not consume the entire input, in which case `pos < size`, |
| 507 | * and it's up to the caller to present again remaining data. |
| 508 | * It returns a hint for the preferred number of bytes to use as an input for |
| 509 | * the next function call. |
| 510 | * |
| 511 | * At any moment, it's possible to flush whatever data remains within internal |
| 512 | * buffer, using ZSTD_flushStream(). `output->pos` will be updated. There might |
| 513 | * still be some content left within the internal buffer if `output->size` is |
| 514 | * too small. It returns the number of bytes left in the internal buffer and |
| 515 | * must be called until it returns 0. |
| 516 | * |
| 517 | * ZSTD_endStream() instructs to finish a frame. It will perform a flush and |
| 518 | * write frame epilogue. The epilogue is required for decoders to consider a |
| 519 | * frame completed. Similar to ZSTD_flushStream(), it may not be able to flush |
| 520 | * the full content if `output->size` is too small. In which case, call again |
| 521 | * ZSTD_endStream() to complete the flush. It returns the number of bytes left |
| 522 | * in the internal buffer and must be called until it returns 0. |
| 523 | ******************************************************************************/ |
| 524 | |
| 525 | /** |
| 526 | * ZSTD_CStreamWorkspaceBound() - memory needed to initialize a ZSTD_CStream |
| 527 | * @cParams: The compression parameters to be used for compression. |
| 528 | * |
| 529 | * Return: A lower bound on the size of the workspace that is passed to |
| 530 | * ZSTD_initCStream() and ZSTD_initCStream_usingCDict(). |
| 531 | */ |
| 532 | size_t ZSTD_CStreamWorkspaceBound(ZSTD_compressionParameters cParams); |
| 533 | |
| 534 | /** |
| 535 | * struct ZSTD_CStream - the zstd streaming compression context |
| 536 | */ |
| 537 | typedef struct ZSTD_CStream_s ZSTD_CStream; |
| 538 | |
| 539 | /*===== ZSTD_CStream management functions =====*/ |
| 540 | /** |
| 541 | * ZSTD_initCStream() - initialize a zstd streaming compression context |
| 542 | * @params: The zstd compression parameters. |
| 543 | * @pledgedSrcSize: If params.fParams.contentSizeFlag == 1 then the caller must |
| 544 | * pass the source size (zero means empty source). Otherwise, |
| 545 | * the caller may optionally pass the source size, or zero if |
| 546 | * unknown. |
| 547 | * @workspace: The workspace to emplace the context into. It must outlive |
| 548 | * the returned context. |
| 549 | * @workspaceSize: The size of workspace. |
| 550 | * Use ZSTD_CStreamWorkspaceBound(params.cParams) to determine |
| 551 | * how large the workspace must be. |
| 552 | * |
| 553 | * Return: The zstd streaming compression context. |
| 554 | */ |
| 555 | ZSTD_CStream *ZSTD_initCStream(ZSTD_parameters params, |
| 556 | unsigned long long pledgedSrcSize, void *workspace, |
| 557 | size_t workspaceSize); |
| 558 | |
| 559 | /** |
| 560 | * ZSTD_initCStream_usingCDict() - initialize a streaming compression context |
| 561 | * @cdict: The digested dictionary to use for compression. |
| 562 | * @pledgedSrcSize: Optionally the source size, or zero if unknown. |
| 563 | * @workspace: The workspace to emplace the context into. It must outlive |
| 564 | * the returned context. |
| 565 | * @workspaceSize: The size of workspace. Call ZSTD_CStreamWorkspaceBound() |
| 566 | * with the cParams used to initialize the cdict to determine |
| 567 | * how large the workspace must be. |
| 568 | * |
| 569 | * Return: The zstd streaming compression context. |
| 570 | */ |
| 571 | ZSTD_CStream *ZSTD_initCStream_usingCDict(const ZSTD_CDict *cdict, |
| 572 | unsigned long long pledgedSrcSize, void *workspace, |
| 573 | size_t workspaceSize); |
| 574 | |
| 575 | /*===== Streaming compression functions =====*/ |
| 576 | /** |
| 577 | * ZSTD_resetCStream() - reset the context using parameters from creation |
| 578 | * @zcs: The zstd streaming compression context to reset. |
| 579 | * @pledgedSrcSize: Optionally the source size, or zero if unknown. |
| 580 | * |
| 581 | * Resets the context using the parameters from creation. Skips dictionary |
| 582 | * loading, since it can be reused. If `pledgedSrcSize` is non-zero the frame |
| 583 | * content size is always written into the frame header. |
| 584 | * |
| 585 | * Return: Zero or an error, which can be checked using ZSTD_isError(). |
| 586 | */ |
| 587 | size_t ZSTD_resetCStream(ZSTD_CStream *zcs, unsigned long long pledgedSrcSize); |
| 588 | /** |
| 589 | * ZSTD_compressStream() - streaming compress some of input into output |
| 590 | * @zcs: The zstd streaming compression context. |
| 591 | * @output: Destination buffer. `output->pos` is updated to indicate how much |
| 592 | * compressed data was written. |
| 593 | * @input: Source buffer. `input->pos` is updated to indicate how much data was |
| 594 | * read. Note that it may not consume the entire input, in which case |
| 595 | * `input->pos < input->size`, and it's up to the caller to present |
| 596 | * remaining data again. |
| 597 | * |
| 598 | * The `input` and `output` buffers may be any size. Guaranteed to make some |
| 599 | * forward progress if `input` and `output` are not empty. |
| 600 | * |
| 601 | * Return: A hint for the number of bytes to use as the input for the next |
| 602 | * function call or an error, which can be checked using |
| 603 | * ZSTD_isError(). |
| 604 | */ |
| 605 | size_t ZSTD_compressStream(ZSTD_CStream *zcs, ZSTD_outBuffer *output, |
| 606 | ZSTD_inBuffer *input); |
| 607 | /** |
| 608 | * ZSTD_flushStream() - flush internal buffers into output |
| 609 | * @zcs: The zstd streaming compression context. |
| 610 | * @output: Destination buffer. `output->pos` is updated to indicate how much |
| 611 | * compressed data was written. |
| 612 | * |
| 613 | * ZSTD_flushStream() must be called until it returns 0, meaning all the data |
| 614 | * has been flushed. Since ZSTD_flushStream() causes a block to be ended, |
| 615 | * calling it too often will degrade the compression ratio. |
| 616 | * |
| 617 | * Return: The number of bytes still present within internal buffers or an |
| 618 | * error, which can be checked using ZSTD_isError(). |
| 619 | */ |
| 620 | size_t ZSTD_flushStream(ZSTD_CStream *zcs, ZSTD_outBuffer *output); |
| 621 | /** |
| 622 | * ZSTD_endStream() - flush internal buffers into output and end the frame |
| 623 | * @zcs: The zstd streaming compression context. |
| 624 | * @output: Destination buffer. `output->pos` is updated to indicate how much |
| 625 | * compressed data was written. |
| 626 | * |
| 627 | * ZSTD_endStream() must be called until it returns 0, meaning all the data has |
| 628 | * been flushed and the frame epilogue has been written. |
| 629 | * |
| 630 | * Return: The number of bytes still present within internal buffers or an |
| 631 | * error, which can be checked using ZSTD_isError(). |
| 632 | */ |
| 633 | size_t ZSTD_endStream(ZSTD_CStream *zcs, ZSTD_outBuffer *output); |
| 634 | |
| 635 | /** |
| 636 | * ZSTD_CStreamInSize() - recommended size for the input buffer |
| 637 | * |
| 638 | * Return: The recommended size for the input buffer. |
| 639 | */ |
| 640 | size_t ZSTD_CStreamInSize(void); |
| 641 | /** |
| 642 | * ZSTD_CStreamOutSize() - recommended size for the output buffer |
| 643 | * |
| 644 | * When the output buffer is at least this large, it is guaranteed to be large |
| 645 | * enough to flush at least one complete compressed block. |
| 646 | * |
| 647 | * Return: The recommended size for the output buffer. |
| 648 | */ |
| 649 | size_t ZSTD_CStreamOutSize(void); |
| 650 | |
| 651 | |
| 652 | |
| 653 | /*-***************************************************************************** |
| 654 | * Streaming decompression - HowTo |
| 655 | * |
| 656 | * A ZSTD_DStream object is required to track streaming operations. |
| 657 | * Use ZSTD_initDStream() to initialize a ZSTD_DStream object. |
| 658 | * ZSTD_DStream objects can be re-used multiple times. |
| 659 | * |
| 660 | * Use ZSTD_decompressStream() repetitively to consume your input. |
| 661 | * The function will update both `pos` fields. |
| 662 | * If `input->pos < input->size`, some input has not been consumed. |
| 663 | * It's up to the caller to present again remaining data. |
| 664 | * If `output->pos < output->size`, decoder has flushed everything it could. |
| 665 | * Returns 0 iff a frame is completely decoded and fully flushed. |
| 666 | * Otherwise it returns a suggested next input size that will never load more |
| 667 | * than the current frame. |
| 668 | ******************************************************************************/ |
| 669 | |
| 670 | /** |
| 671 | * ZSTD_DStreamWorkspaceBound() - memory needed to initialize a ZSTD_DStream |
| 672 | * @maxWindowSize: The maximum window size allowed for compressed frames. |
| 673 | * |
| 674 | * Return: A lower bound on the size of the workspace that is passed to |
| 675 | * ZSTD_initDStream() and ZSTD_initDStream_usingDDict(). |
| 676 | */ |
| 677 | size_t ZSTD_DStreamWorkspaceBound(size_t maxWindowSize); |
| 678 | |
| 679 | /** |
| 680 | * struct ZSTD_DStream - the zstd streaming decompression context |
| 681 | */ |
| 682 | typedef struct ZSTD_DStream_s ZSTD_DStream; |
| 683 | /*===== ZSTD_DStream management functions =====*/ |
| 684 | /** |
| 685 | * ZSTD_initDStream() - initialize a zstd streaming decompression context |
| 686 | * @maxWindowSize: The maximum window size allowed for compressed frames. |
| 687 | * @workspace: The workspace to emplace the context into. It must outlive |
| 688 | * the returned context. |
| 689 | * @workspaceSize: The size of workspace. |
| 690 | * Use ZSTD_DStreamWorkspaceBound(maxWindowSize) to determine |
| 691 | * how large the workspace must be. |
| 692 | * |
| 693 | * Return: The zstd streaming decompression context. |
| 694 | */ |
| 695 | ZSTD_DStream *ZSTD_initDStream(size_t maxWindowSize, void *workspace, |
| 696 | size_t workspaceSize); |
| 697 | /** |
| 698 | * ZSTD_initDStream_usingDDict() - initialize streaming decompression context |
| 699 | * @maxWindowSize: The maximum window size allowed for compressed frames. |
| 700 | * @ddict: The digested dictionary to use for decompression. |
| 701 | * @workspace: The workspace to emplace the context into. It must outlive |
| 702 | * the returned context. |
| 703 | * @workspaceSize: The size of workspace. |
| 704 | * Use ZSTD_DStreamWorkspaceBound(maxWindowSize) to determine |
| 705 | * how large the workspace must be. |
| 706 | * |
| 707 | * Return: The zstd streaming decompression context. |
| 708 | */ |
| 709 | ZSTD_DStream *ZSTD_initDStream_usingDDict(size_t maxWindowSize, |
| 710 | const ZSTD_DDict *ddict, void *workspace, size_t workspaceSize); |
| 711 | |
| 712 | /*===== Streaming decompression functions =====*/ |
| 713 | /** |
| 714 | * ZSTD_resetDStream() - reset the context using parameters from creation |
| 715 | * @zds: The zstd streaming decompression context to reset. |
| 716 | * |
| 717 | * Resets the context using the parameters from creation. Skips dictionary |
| 718 | * loading, since it can be reused. |
| 719 | * |
| 720 | * Return: Zero or an error, which can be checked using ZSTD_isError(). |
| 721 | */ |
| 722 | size_t ZSTD_resetDStream(ZSTD_DStream *zds); |
| 723 | /** |
| 724 | * ZSTD_decompressStream() - streaming decompress some of input into output |
| 725 | * @zds: The zstd streaming decompression context. |
| 726 | * @output: Destination buffer. `output.pos` is updated to indicate how much |
| 727 | * decompressed data was written. |
| 728 | * @input: Source buffer. `input.pos` is updated to indicate how much data was |
| 729 | * read. Note that it may not consume the entire input, in which case |
| 730 | * `input.pos < input.size`, and it's up to the caller to present |
| 731 | * remaining data again. |
| 732 | * |
| 733 | * The `input` and `output` buffers may be any size. Guaranteed to make some |
| 734 | * forward progress if `input` and `output` are not empty. |
| 735 | * ZSTD_decompressStream() will not consume the last byte of the frame until |
| 736 | * the entire frame is flushed. |
| 737 | * |
| 738 | * Return: Returns 0 iff a frame is completely decoded and fully flushed. |
| 739 | * Otherwise returns a hint for the number of bytes to use as the input |
| 740 | * for the next function call or an error, which can be checked using |
| 741 | * ZSTD_isError(). The size hint will never load more than the frame. |
| 742 | */ |
| 743 | size_t ZSTD_decompressStream(ZSTD_DStream *zds, ZSTD_outBuffer *output, |
| 744 | ZSTD_inBuffer *input); |
| 745 | |
| 746 | /** |
| 747 | * ZSTD_DStreamInSize() - recommended size for the input buffer |
| 748 | * |
| 749 | * Return: The recommended size for the input buffer. |
| 750 | */ |
| 751 | size_t ZSTD_DStreamInSize(void); |
| 752 | /** |
| 753 | * ZSTD_DStreamOutSize() - recommended size for the output buffer |
| 754 | * |
| 755 | * When the output buffer is at least this large, it is guaranteed to be large |
| 756 | * enough to flush at least one complete decompressed block. |
| 757 | * |
| 758 | * Return: The recommended size for the output buffer. |
| 759 | */ |
| 760 | size_t ZSTD_DStreamOutSize(void); |
| 761 | |
| 762 | |
| 763 | /* --- Constants ---*/ |
| 764 | #define ZSTD_MAGICNUMBER 0xFD2FB528 /* >= v0.8.0 */ |
| 765 | #define ZSTD_MAGIC_SKIPPABLE_START 0x184D2A50U |
| 766 | |
| 767 | #define ZSTD_CONTENTSIZE_UNKNOWN (0ULL - 1) |
| 768 | #define ZSTD_CONTENTSIZE_ERROR (0ULL - 2) |
| 769 | |
| 770 | #define ZSTD_WINDOWLOG_MAX_32 27 |
| 771 | #define ZSTD_WINDOWLOG_MAX_64 27 |
| 772 | #define ZSTD_WINDOWLOG_MAX \ |
| 773 | ((unsigned int)(sizeof(size_t) == 4 \ |
| 774 | ? ZSTD_WINDOWLOG_MAX_32 \ |
| 775 | : ZSTD_WINDOWLOG_MAX_64)) |
| 776 | #define ZSTD_WINDOWLOG_MIN 10 |
| 777 | #define ZSTD_HASHLOG_MAX ZSTD_WINDOWLOG_MAX |
| 778 | #define ZSTD_HASHLOG_MIN 6 |
| 779 | #define ZSTD_CHAINLOG_MAX (ZSTD_WINDOWLOG_MAX+1) |
| 780 | #define ZSTD_CHAINLOG_MIN ZSTD_HASHLOG_MIN |
| 781 | #define ZSTD_HASHLOG3_MAX 17 |
| 782 | #define ZSTD_SEARCHLOG_MAX (ZSTD_WINDOWLOG_MAX-1) |
| 783 | #define ZSTD_SEARCHLOG_MIN 1 |
| 784 | /* only for ZSTD_fast, other strategies are limited to 6 */ |
| 785 | #define ZSTD_SEARCHLENGTH_MAX 7 |
| 786 | /* only for ZSTD_btopt, other strategies are limited to 4 */ |
| 787 | #define ZSTD_SEARCHLENGTH_MIN 3 |
| 788 | #define ZSTD_TARGETLENGTH_MIN 4 |
| 789 | #define ZSTD_TARGETLENGTH_MAX 999 |
| 790 | |
| 791 | /* for static allocation */ |
| 792 | #define ZSTD_FRAMEHEADERSIZE_MAX 18 |
| 793 | #define ZSTD_FRAMEHEADERSIZE_MIN 6 |
| 794 | static const size_t ZSTD_frameHeaderSize_prefix = 5; |
| 795 | static const size_t ZSTD_frameHeaderSize_min = ZSTD_FRAMEHEADERSIZE_MIN; |
| 796 | static const size_t ZSTD_frameHeaderSize_max = ZSTD_FRAMEHEADERSIZE_MAX; |
| 797 | /* magic number + skippable frame length */ |
| 798 | static const size_t ZSTD_skippableHeaderSize = 8; |
| 799 | |
| 800 | |
| 801 | /*-************************************* |
| 802 | * Compressed size functions |
| 803 | **************************************/ |
| 804 | |
| 805 | /** |
| 806 | * ZSTD_findFrameCompressedSize() - returns the size of a compressed frame |
| 807 | * @src: Source buffer. It should point to the start of a zstd encoded frame |
| 808 | * or a skippable frame. |
| 809 | * @srcSize: The size of the source buffer. It must be at least as large as the |
| 810 | * size of the frame. |
| 811 | * |
| 812 | * Return: The compressed size of the frame pointed to by `src` or an error, |
| 813 | * which can be check with ZSTD_isError(). |
| 814 | * Suitable to pass to ZSTD_decompress() or similar functions. |
| 815 | */ |
| 816 | size_t ZSTD_findFrameCompressedSize(const void *src, size_t srcSize); |
| 817 | |
| 818 | /*-************************************* |
| 819 | * Decompressed size functions |
| 820 | **************************************/ |
| 821 | /** |
| 822 | * ZSTD_getFrameContentSize() - returns the content size in a zstd frame header |
| 823 | * @src: It should point to the start of a zstd encoded frame. |
| 824 | * @srcSize: The size of the source buffer. It must be at least as large as the |
| 825 | * frame header. `ZSTD_frameHeaderSize_max` is always large enough. |
| 826 | * |
| 827 | * Return: The frame content size stored in the frame header if known. |
| 828 | * `ZSTD_CONTENTSIZE_UNKNOWN` if the content size isn't stored in the |
| 829 | * frame header. `ZSTD_CONTENTSIZE_ERROR` on invalid input. |
| 830 | */ |
| 831 | unsigned long long ZSTD_getFrameContentSize(const void *src, size_t srcSize); |
| 832 | |
| 833 | /** |
| 834 | * ZSTD_findDecompressedSize() - returns decompressed size of a series of frames |
| 835 | * @src: It should point to the start of a series of zstd encoded and/or |
| 836 | * skippable frames. |
| 837 | * @srcSize: The exact size of the series of frames. |
| 838 | * |
| 839 | * If any zstd encoded frame in the series doesn't have the frame content size |
| 840 | * set, `ZSTD_CONTENTSIZE_UNKNOWN` is returned. But frame content size is always |
| 841 | * set when using ZSTD_compress(). The decompressed size can be very large. |
| 842 | * If the source is untrusted, the decompressed size could be wrong or |
| 843 | * intentionally modified. Always ensure the result fits within the |
| 844 | * application's authorized limits. ZSTD_findDecompressedSize() handles multiple |
| 845 | * frames, and so it must traverse the input to read each frame header. This is |
| 846 | * efficient as most of the data is skipped, however it does mean that all frame |
| 847 | * data must be present and valid. |
| 848 | * |
| 849 | * Return: Decompressed size of all the data contained in the frames if known. |
| 850 | * `ZSTD_CONTENTSIZE_UNKNOWN` if the decompressed size is unknown. |
| 851 | * `ZSTD_CONTENTSIZE_ERROR` if an error occurred. |
| 852 | */ |
| 853 | unsigned long long ZSTD_findDecompressedSize(const void *src, size_t srcSize); |
| 854 | |
| 855 | /*-************************************* |
| 856 | * Advanced compression functions |
| 857 | **************************************/ |
| 858 | /** |
| 859 | * ZSTD_checkCParams() - ensure parameter values remain within authorized range |
| 860 | * @cParams: The zstd compression parameters. |
| 861 | * |
| 862 | * Return: Zero or an error, which can be checked using ZSTD_isError(). |
| 863 | */ |
| 864 | size_t ZSTD_checkCParams(ZSTD_compressionParameters cParams); |
| 865 | |
| 866 | /** |
| 867 | * ZSTD_adjustCParams() - optimize parameters for a given srcSize and dictSize |
| 868 | * @srcSize: Optionally the estimated source size, or zero if unknown. |
| 869 | * @dictSize: Optionally the estimated dictionary size, or zero if unknown. |
| 870 | * |
| 871 | * Return: The optimized parameters. |
| 872 | */ |
| 873 | ZSTD_compressionParameters ZSTD_adjustCParams( |
| 874 | ZSTD_compressionParameters cParams, unsigned long long srcSize, |
| 875 | size_t dictSize); |
| 876 | |
| 877 | /*--- Advanced decompression functions ---*/ |
| 878 | |
| 879 | /** |
| 880 | * ZSTD_isFrame() - returns true iff the buffer starts with a valid frame |
| 881 | * @buffer: The source buffer to check. |
| 882 | * @size: The size of the source buffer, must be at least 4 bytes. |
| 883 | * |
| 884 | * Return: True iff the buffer starts with a zstd or skippable frame identifier. |
| 885 | */ |
| 886 | unsigned int ZSTD_isFrame(const void *buffer, size_t size); |
| 887 | |
| 888 | /** |
| 889 | * ZSTD_getDictID_fromDict() - returns the dictionary id stored in a dictionary |
| 890 | * @dict: The dictionary buffer. |
| 891 | * @dictSize: The size of the dictionary buffer. |
| 892 | * |
| 893 | * Return: The dictionary id stored within the dictionary or 0 if the |
| 894 | * dictionary is not a zstd dictionary. If it returns 0 the |
| 895 | * dictionary can still be loaded as a content-only dictionary. |
| 896 | */ |
| 897 | unsigned int ZSTD_getDictID_fromDict(const void *dict, size_t dictSize); |
| 898 | |
| 899 | /** |
| 900 | * ZSTD_getDictID_fromDDict() - returns the dictionary id stored in a ZSTD_DDict |
| 901 | * @ddict: The ddict to find the id of. |
| 902 | * |
| 903 | * Return: The dictionary id stored within `ddict` or 0 if the dictionary is not |
| 904 | * a zstd dictionary. If it returns 0 `ddict` will be loaded as a |
| 905 | * content-only dictionary. |
| 906 | */ |
| 907 | unsigned int ZSTD_getDictID_fromDDict(const ZSTD_DDict *ddict); |
| 908 | |
| 909 | /** |
| 910 | * ZSTD_getDictID_fromFrame() - returns the dictionary id stored in a zstd frame |
| 911 | * @src: Source buffer. It must be a zstd encoded frame. |
| 912 | * @srcSize: The size of the source buffer. It must be at least as large as the |
| 913 | * frame header. `ZSTD_frameHeaderSize_max` is always large enough. |
| 914 | * |
| 915 | * Return: The dictionary id required to decompress the frame stored within |
| 916 | * `src` or 0 if the dictionary id could not be decoded. It can return |
| 917 | * 0 if the frame does not require a dictionary, the dictionary id |
| 918 | * wasn't stored in the frame, `src` is not a zstd frame, or `srcSize` |
| 919 | * is too small. |
| 920 | */ |
| 921 | unsigned int ZSTD_getDictID_fromFrame(const void *src, size_t srcSize); |
| 922 | |
| 923 | /** |
| 924 | * struct ZSTD_frameParams - zstd frame parameters stored in the frame header |
| 925 | * @frameContentSize: The frame content size, or 0 if not present. |
| 926 | * @windowSize: The window size, or 0 if the frame is a skippable frame. |
| 927 | * @dictID: The dictionary id, or 0 if not present. |
| 928 | * @checksumFlag: Whether a checksum was used. |
| 929 | */ |
| 930 | typedef struct { |
| 931 | unsigned long long frameContentSize; |
| 932 | unsigned int windowSize; |
| 933 | unsigned int dictID; |
| 934 | unsigned int checksumFlag; |
| 935 | } ZSTD_frameParams; |
| 936 | |
| 937 | /** |
| 938 | * ZSTD_getFrameParams() - extracts parameters from a zstd or skippable frame |
| 939 | * @fparamsPtr: On success the frame parameters are written here. |
| 940 | * @src: The source buffer. It must point to a zstd or skippable frame. |
| 941 | * @srcSize: The size of the source buffer. `ZSTD_frameHeaderSize_max` is |
| 942 | * always large enough to succeed. |
| 943 | * |
| 944 | * Return: 0 on success. If more data is required it returns how many bytes |
| 945 | * must be provided to make forward progress. Otherwise it returns |
| 946 | * an error, which can be checked using ZSTD_isError(). |
| 947 | */ |
| 948 | size_t ZSTD_getFrameParams(ZSTD_frameParams *fparamsPtr, const void *src, |
| 949 | size_t srcSize); |
| 950 | |
| 951 | /*-***************************************************************************** |
| 952 | * Buffer-less and synchronous inner streaming functions |
| 953 | * |
| 954 | * This is an advanced API, giving full control over buffer management, for |
| 955 | * users which need direct control over memory. |
| 956 | * But it's also a complex one, with many restrictions (documented below). |
| 957 | * Prefer using normal streaming API for an easier experience |
| 958 | ******************************************************************************/ |
| 959 | |
| 960 | /*-***************************************************************************** |
| 961 | * Buffer-less streaming compression (synchronous mode) |
| 962 | * |
| 963 | * A ZSTD_CCtx object is required to track streaming operations. |
| 964 | * Use ZSTD_initCCtx() to initialize a context. |
| 965 | * ZSTD_CCtx object can be re-used multiple times within successive compression |
| 966 | * operations. |
| 967 | * |
| 968 | * Start by initializing a context. |
| 969 | * Use ZSTD_compressBegin(), or ZSTD_compressBegin_usingDict() for dictionary |
| 970 | * compression, |
| 971 | * or ZSTD_compressBegin_advanced(), for finer parameter control. |
| 972 | * It's also possible to duplicate a reference context which has already been |
| 973 | * initialized, using ZSTD_copyCCtx() |
| 974 | * |
| 975 | * Then, consume your input using ZSTD_compressContinue(). |
| 976 | * There are some important considerations to keep in mind when using this |
| 977 | * advanced function : |
| 978 | * - ZSTD_compressContinue() has no internal buffer. It uses externally provided |
| 979 | * buffer only. |
| 980 | * - Interface is synchronous : input is consumed entirely and produce 1+ |
| 981 | * (or more) compressed blocks. |
| 982 | * - Caller must ensure there is enough space in `dst` to store compressed data |
| 983 | * under worst case scenario. Worst case evaluation is provided by |
| 984 | * ZSTD_compressBound(). |
| 985 | * ZSTD_compressContinue() doesn't guarantee recover after a failed |
| 986 | * compression. |
| 987 | * - ZSTD_compressContinue() presumes prior input ***is still accessible and |
| 988 | * unmodified*** (up to maximum distance size, see WindowLog). |
| 989 | * It remembers all previous contiguous blocks, plus one separated memory |
| 990 | * segment (which can itself consists of multiple contiguous blocks) |
| 991 | * - ZSTD_compressContinue() detects that prior input has been overwritten when |
| 992 | * `src` buffer overlaps. In which case, it will "discard" the relevant memory |
| 993 | * section from its history. |
| 994 | * |
| 995 | * Finish a frame with ZSTD_compressEnd(), which will write the last block(s) |
| 996 | * and optional checksum. It's possible to use srcSize==0, in which case, it |
| 997 | * will write a final empty block to end the frame. Without last block mark, |
| 998 | * frames will be considered unfinished (corrupted) by decoders. |
| 999 | * |
| 1000 | * `ZSTD_CCtx` object can be re-used (ZSTD_compressBegin()) to compress some new |
| 1001 | * frame. |
| 1002 | ******************************************************************************/ |
| 1003 | |
| 1004 | /*===== Buffer-less streaming compression functions =====*/ |
| 1005 | size_t ZSTD_compressBegin(ZSTD_CCtx *cctx, int compressionLevel); |
| 1006 | size_t ZSTD_compressBegin_usingDict(ZSTD_CCtx *cctx, const void *dict, |
| 1007 | size_t dictSize, int compressionLevel); |
| 1008 | size_t ZSTD_compressBegin_advanced(ZSTD_CCtx *cctx, const void *dict, |
| 1009 | size_t dictSize, ZSTD_parameters params, |
| 1010 | unsigned long long pledgedSrcSize); |
| 1011 | size_t ZSTD_copyCCtx(ZSTD_CCtx *cctx, const ZSTD_CCtx *preparedCCtx, |
| 1012 | unsigned long long pledgedSrcSize); |
| 1013 | size_t ZSTD_compressBegin_usingCDict(ZSTD_CCtx *cctx, const ZSTD_CDict *cdict, |
| 1014 | unsigned long long pledgedSrcSize); |
| 1015 | size_t ZSTD_compressContinue(ZSTD_CCtx *cctx, void *dst, size_t dstCapacity, |
| 1016 | const void *src, size_t srcSize); |
| 1017 | size_t ZSTD_compressEnd(ZSTD_CCtx *cctx, void *dst, size_t dstCapacity, |
| 1018 | const void *src, size_t srcSize); |
| 1019 | |
| 1020 | |
| 1021 | |
| 1022 | /*-***************************************************************************** |
| 1023 | * Buffer-less streaming decompression (synchronous mode) |
| 1024 | * |
| 1025 | * A ZSTD_DCtx object is required to track streaming operations. |
| 1026 | * Use ZSTD_initDCtx() to initialize a context. |
| 1027 | * A ZSTD_DCtx object can be re-used multiple times. |
| 1028 | * |
| 1029 | * First typical operation is to retrieve frame parameters, using |
| 1030 | * ZSTD_getFrameParams(). It fills a ZSTD_frameParams structure which provide |
| 1031 | * important information to correctly decode the frame, such as the minimum |
| 1032 | * rolling buffer size to allocate to decompress data (`windowSize`), and the |
| 1033 | * dictionary ID used. |
| 1034 | * Note: content size is optional, it may not be present. 0 means unknown. |
| 1035 | * Note that these values could be wrong, either because of data malformation, |
| 1036 | * or because an attacker is spoofing deliberate false information. As a |
| 1037 | * consequence, check that values remain within valid application range, |
| 1038 | * especially `windowSize`, before allocation. Each application can set its own |
| 1039 | * limit, depending on local restrictions. For extended interoperability, it is |
| 1040 | * recommended to support at least 8 MB. |
| 1041 | * Frame parameters are extracted from the beginning of the compressed frame. |
| 1042 | * Data fragment must be large enough to ensure successful decoding, typically |
| 1043 | * `ZSTD_frameHeaderSize_max` bytes. |
| 1044 | * Result: 0: successful decoding, the `ZSTD_frameParams` structure is filled. |
| 1045 | * >0: `srcSize` is too small, provide at least this many bytes. |
| 1046 | * errorCode, which can be tested using ZSTD_isError(). |
| 1047 | * |
| 1048 | * Start decompression, with ZSTD_decompressBegin() or |
| 1049 | * ZSTD_decompressBegin_usingDict(). Alternatively, you can copy a prepared |
| 1050 | * context, using ZSTD_copyDCtx(). |
| 1051 | * |
| 1052 | * Then use ZSTD_nextSrcSizeToDecompress() and ZSTD_decompressContinue() |
| 1053 | * alternatively. |
| 1054 | * ZSTD_nextSrcSizeToDecompress() tells how many bytes to provide as 'srcSize' |
| 1055 | * to ZSTD_decompressContinue(). |
| 1056 | * ZSTD_decompressContinue() requires this _exact_ amount of bytes, or it will |
| 1057 | * fail. |
| 1058 | * |
| 1059 | * The result of ZSTD_decompressContinue() is the number of bytes regenerated |
| 1060 | * within 'dst' (necessarily <= dstCapacity). It can be zero, which is not an |
| 1061 | * error; it just means ZSTD_decompressContinue() has decoded some metadata |
| 1062 | * item. It can also be an error code, which can be tested with ZSTD_isError(). |
| 1063 | * |
| 1064 | * ZSTD_decompressContinue() needs previous data blocks during decompression, up |
| 1065 | * to `windowSize`. They should preferably be located contiguously, prior to |
| 1066 | * current block. Alternatively, a round buffer of sufficient size is also |
| 1067 | * possible. Sufficient size is determined by frame parameters. |
| 1068 | * ZSTD_decompressContinue() is very sensitive to contiguity, if 2 blocks don't |
| 1069 | * follow each other, make sure that either the compressor breaks contiguity at |
| 1070 | * the same place, or that previous contiguous segment is large enough to |
| 1071 | * properly handle maximum back-reference. |
| 1072 | * |
| 1073 | * A frame is fully decoded when ZSTD_nextSrcSizeToDecompress() returns zero. |
| 1074 | * Context can then be reset to start a new decompression. |
| 1075 | * |
| 1076 | * Note: it's possible to know if next input to present is a header or a block, |
| 1077 | * using ZSTD_nextInputType(). This information is not required to properly |
| 1078 | * decode a frame. |
| 1079 | * |
| 1080 | * == Special case: skippable frames == |
| 1081 | * |
| 1082 | * Skippable frames allow integration of user-defined data into a flow of |
| 1083 | * concatenated frames. Skippable frames will be ignored (skipped) by a |
| 1084 | * decompressor. The format of skippable frames is as follows: |
| 1085 | * a) Skippable frame ID - 4 Bytes, Little endian format, any value from |
| 1086 | * 0x184D2A50 to 0x184D2A5F |
| 1087 | * b) Frame Size - 4 Bytes, Little endian format, unsigned 32-bits |
| 1088 | * c) Frame Content - any content (User Data) of length equal to Frame Size |
| 1089 | * For skippable frames ZSTD_decompressContinue() always returns 0. |
| 1090 | * For skippable frames ZSTD_getFrameParams() returns fparamsPtr->windowLog==0 |
| 1091 | * what means that a frame is skippable. |
| 1092 | * Note: If fparamsPtr->frameContentSize==0, it is ambiguous: the frame might |
| 1093 | * actually be a zstd encoded frame with no content. For purposes of |
| 1094 | * decompression, it is valid in both cases to skip the frame using |
| 1095 | * ZSTD_findFrameCompressedSize() to find its size in bytes. |
| 1096 | * It also returns frame size as fparamsPtr->frameContentSize. |
| 1097 | ******************************************************************************/ |
| 1098 | |
| 1099 | /*===== Buffer-less streaming decompression functions =====*/ |
| 1100 | size_t ZSTD_decompressBegin(ZSTD_DCtx *dctx); |
| 1101 | size_t ZSTD_decompressBegin_usingDict(ZSTD_DCtx *dctx, const void *dict, |
| 1102 | size_t dictSize); |
| 1103 | void ZSTD_copyDCtx(ZSTD_DCtx *dctx, const ZSTD_DCtx *preparedDCtx); |
| 1104 | size_t ZSTD_nextSrcSizeToDecompress(ZSTD_DCtx *dctx); |
| 1105 | size_t ZSTD_decompressContinue(ZSTD_DCtx *dctx, void *dst, size_t dstCapacity, |
| 1106 | const void *src, size_t srcSize); |
| 1107 | typedef enum { |
| 1108 | ZSTDnit_frameHeader, |
| 1109 | ZSTDnit_blockHeader, |
| 1110 | ZSTDnit_block, |
| 1111 | ZSTDnit_lastBlock, |
| 1112 | ZSTDnit_checksum, |
| 1113 | ZSTDnit_skippableFrame |
| 1114 | } ZSTD_nextInputType_e; |
| 1115 | ZSTD_nextInputType_e ZSTD_nextInputType(ZSTD_DCtx *dctx); |
| 1116 | |
| 1117 | /*-***************************************************************************** |
| 1118 | * Block functions |
| 1119 | * |
| 1120 | * Block functions produce and decode raw zstd blocks, without frame metadata. |
| 1121 | * Frame metadata cost is typically ~18 bytes, which can be non-negligible for |
| 1122 | * very small blocks (< 100 bytes). User will have to take in charge required |
| 1123 | * information to regenerate data, such as compressed and content sizes. |
| 1124 | * |
| 1125 | * A few rules to respect: |
| 1126 | * - Compressing and decompressing require a context structure |
| 1127 | * + Use ZSTD_initCCtx() and ZSTD_initDCtx() |
| 1128 | * - It is necessary to init context before starting |
| 1129 | * + compression : ZSTD_compressBegin() |
| 1130 | * + decompression : ZSTD_decompressBegin() |
| 1131 | * + variants _usingDict() are also allowed |
| 1132 | * + copyCCtx() and copyDCtx() work too |
| 1133 | * - Block size is limited, it must be <= ZSTD_getBlockSizeMax() |
| 1134 | * + If you need to compress more, cut data into multiple blocks |
| 1135 | * + Consider using the regular ZSTD_compress() instead, as frame metadata |
| 1136 | * costs become negligible when source size is large. |
| 1137 | * - When a block is considered not compressible enough, ZSTD_compressBlock() |
| 1138 | * result will be zero. In which case, nothing is produced into `dst`. |
| 1139 | * + User must test for such outcome and deal directly with uncompressed data |
| 1140 | * + ZSTD_decompressBlock() doesn't accept uncompressed data as input!!! |
| 1141 | * + In case of multiple successive blocks, decoder must be informed of |
| 1142 | * uncompressed block existence to follow proper history. Use |
| 1143 | * ZSTD_insertBlock() in such a case. |
| 1144 | ******************************************************************************/ |
| 1145 | |
| 1146 | /* Define for static allocation */ |
| 1147 | #define ZSTD_BLOCKSIZE_ABSOLUTEMAX (128 * 1024) |
| 1148 | /*===== Raw zstd block functions =====*/ |
| 1149 | size_t ZSTD_getBlockSizeMax(ZSTD_CCtx *cctx); |
| 1150 | size_t ZSTD_compressBlock(ZSTD_CCtx *cctx, void *dst, size_t dstCapacity, |
| 1151 | const void *src, size_t srcSize); |
| 1152 | size_t ZSTD_decompressBlock(ZSTD_DCtx *dctx, void *dst, size_t dstCapacity, |
| 1153 | const void *src, size_t srcSize); |
| 1154 | size_t ZSTD_insertBlock(ZSTD_DCtx *dctx, const void *blockStart, |
| 1155 | size_t blockSize); |
| 1156 | |
| 1157 | #endif /* ZSTD_H */ |