Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -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 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 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 38 | /******************************************************************************* |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 39 | Introduction |
| 40 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 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 | |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 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) |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 56 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 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 | *******************************************************************************/ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 70 | |
| 71 | /*------ Version ------*/ |
| 72 | #define ZSTD_VERSION_MAJOR 1 |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 73 | #define ZSTD_VERSION_MINOR 4 |
| 74 | #define ZSTD_VERSION_RELEASE 1 |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 75 | |
| 76 | #define ZSTD_VERSION_NUMBER (ZSTD_VERSION_MAJOR *100*100 + ZSTD_VERSION_MINOR *100 + ZSTD_VERSION_RELEASE) |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 77 | ZSTDLIB_API unsigned ZSTD_versionNumber(void); /**< to check runtime library version */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 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) |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 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 | * Constants |
| 94 | ***************************************/ |
| 95 | |
| 96 | /* All magic numbers are supposed read/written to/from files/memory using little-endian convention */ |
| 97 | #define ZSTD_MAGICNUMBER 0xFD2FB528 /* valid since v0.8.0 */ |
| 98 | #define ZSTD_MAGIC_DICTIONARY 0xEC30A437 /* valid since v0.7.0 */ |
| 99 | #define ZSTD_MAGIC_SKIPPABLE_START 0x184D2A50 /* all 16 values, from 0x184D2A50 to 0x184D2A5F, signal the beginning of a skippable frame */ |
| 100 | #define ZSTD_MAGIC_SKIPPABLE_MASK 0xFFFFFFF0 |
| 101 | |
| 102 | #define ZSTD_BLOCKSIZELOG_MAX 17 |
| 103 | #define ZSTD_BLOCKSIZE_MAX (1<<ZSTD_BLOCKSIZELOG_MAX) |
| 104 | |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 105 | |
| 106 | |
| 107 | /*************************************** |
| 108 | * Simple API |
| 109 | ***************************************/ |
| 110 | /*! ZSTD_compress() : |
| 111 | * Compresses `src` content as a single zstd compressed frame into already allocated `dst`. |
| 112 | * Hint : compression runs faster if `dstCapacity` >= `ZSTD_compressBound(srcSize)`. |
| 113 | * @return : compressed size written into `dst` (<= `dstCapacity), |
| 114 | * or an error code if it fails (which can be tested using ZSTD_isError()). */ |
| 115 | ZSTDLIB_API size_t ZSTD_compress( void* dst, size_t dstCapacity, |
| 116 | const void* src, size_t srcSize, |
| 117 | int compressionLevel); |
| 118 | |
| 119 | /*! ZSTD_decompress() : |
| 120 | * `compressedSize` : must be the _exact_ size of some number of compressed and/or skippable frames. |
| 121 | * `dstCapacity` is an upper bound of originalSize to regenerate. |
| 122 | * If user cannot imply a maximum upper bound, it's better to use streaming mode to decompress data. |
| 123 | * @return : the number of bytes decompressed into `dst` (<= `dstCapacity`), |
| 124 | * or an errorCode if it fails (which can be tested using ZSTD_isError()). */ |
| 125 | ZSTDLIB_API size_t ZSTD_decompress( void* dst, size_t dstCapacity, |
| 126 | const void* src, size_t compressedSize); |
| 127 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 128 | /*! ZSTD_getFrameContentSize() : requires v1.3.0+ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 129 | * `src` should point to the start of a ZSTD encoded frame. |
| 130 | * `srcSize` must be at least as large as the frame header. |
| 131 | * hint : any size >= `ZSTD_frameHeaderSize_max` is large enough. |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 132 | * @return : - decompressed size of `src` frame content, if known |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 133 | * - ZSTD_CONTENTSIZE_UNKNOWN if the size cannot be determined |
| 134 | * - ZSTD_CONTENTSIZE_ERROR if an error occurred (e.g. invalid magic number, srcSize too small) |
| 135 | * note 1 : a 0 return value means the frame is valid but "empty". |
| 136 | * note 2 : decompressed size is an optional field, it may not be present, typically in streaming mode. |
| 137 | * When `return==ZSTD_CONTENTSIZE_UNKNOWN`, data to decompress could be any size. |
| 138 | * In which case, it's necessary to use streaming mode to decompress data. |
| 139 | * Optionally, application can rely on some implicit limit, |
| 140 | * as ZSTD_decompress() only needs an upper bound of decompressed size. |
| 141 | * (For example, data could be necessarily cut into blocks <= 16 KB). |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 142 | * note 3 : decompressed size is always present when compression is completed using single-pass functions, |
| 143 | * such as ZSTD_compress(), ZSTD_compressCCtx() ZSTD_compress_usingDict() or ZSTD_compress_usingCDict(). |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 144 | * note 4 : decompressed size can be very large (64-bits value), |
| 145 | * potentially larger than what local system can handle as a single memory segment. |
| 146 | * In which case, it's necessary to use streaming mode to decompress data. |
| 147 | * note 5 : If source is untrusted, decompressed size could be wrong or intentionally modified. |
| 148 | * Always ensure return value fits within application's authorized limits. |
| 149 | * Each application can set its own limits. |
| 150 | * note 6 : This function replaces ZSTD_getDecompressedSize() */ |
| 151 | #define ZSTD_CONTENTSIZE_UNKNOWN (0ULL - 1) |
| 152 | #define ZSTD_CONTENTSIZE_ERROR (0ULL - 2) |
| 153 | ZSTDLIB_API unsigned long long ZSTD_getFrameContentSize(const void *src, size_t srcSize); |
| 154 | |
| 155 | /*! ZSTD_getDecompressedSize() : |
| 156 | * NOTE: This function is now obsolete, in favor of ZSTD_getFrameContentSize(). |
| 157 | * Both functions work the same way, but ZSTD_getDecompressedSize() blends |
| 158 | * "empty", "unknown" and "error" results to the same return value (0), |
| 159 | * while ZSTD_getFrameContentSize() gives them separate return values. |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 160 | * @return : decompressed size of `src` frame content _if known and not empty_, 0 otherwise. */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 161 | ZSTDLIB_API unsigned long long ZSTD_getDecompressedSize(const void* src, size_t srcSize); |
| 162 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 163 | /*! ZSTD_findFrameCompressedSize() : |
| 164 | * `src` should point to the start of a ZSTD frame or skippable frame. |
| 165 | * `srcSize` must be >= first frame size |
| 166 | * @return : the compressed size of the first frame starting at `src`, |
| 167 | * suitable to pass as `srcSize` to `ZSTD_decompress` or similar, |
| 168 | * or an error code if input is invalid */ |
| 169 | ZSTDLIB_API size_t ZSTD_findFrameCompressedSize(const void* src, size_t srcSize); |
| 170 | |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 171 | |
| 172 | /*====== Helper functions ======*/ |
| 173 | #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 */ |
| 174 | ZSTDLIB_API size_t ZSTD_compressBound(size_t srcSize); /*!< maximum compressed size in worst case single-pass scenario */ |
| 175 | ZSTDLIB_API unsigned ZSTD_isError(size_t code); /*!< tells if a `size_t` function result is an error code */ |
| 176 | ZSTDLIB_API const char* ZSTD_getErrorName(size_t code); /*!< provides readable string from an error code */ |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 177 | ZSTDLIB_API int ZSTD_minCLevel(void); /*!< minimum negative compression level allowed */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 178 | ZSTDLIB_API int ZSTD_maxCLevel(void); /*!< maximum compression level available */ |
| 179 | |
| 180 | |
| 181 | /*************************************** |
| 182 | * Explicit context |
| 183 | ***************************************/ |
| 184 | /*= Compression context |
| 185 | * When compressing many times, |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 186 | * it is recommended to allocate a context just once, |
| 187 | * and re-use it for each successive compression operation. |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 188 | * This will make workload friendlier for system's memory. |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 189 | * Note : re-using context is just a speed / resource optimization. |
| 190 | * It doesn't change the compression ratio, which remains identical. |
| 191 | * Note 2 : In multi-threaded environments, |
| 192 | * use one different context per thread for parallel execution. |
| 193 | */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 194 | typedef struct ZSTD_CCtx_s ZSTD_CCtx; |
| 195 | ZSTDLIB_API ZSTD_CCtx* ZSTD_createCCtx(void); |
| 196 | ZSTDLIB_API size_t ZSTD_freeCCtx(ZSTD_CCtx* cctx); |
| 197 | |
| 198 | /*! ZSTD_compressCCtx() : |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 199 | * Same as ZSTD_compress(), using an explicit ZSTD_CCtx |
| 200 | * The function will compress at requested compression level, |
| 201 | * ignoring any other parameter */ |
| 202 | ZSTDLIB_API size_t ZSTD_compressCCtx(ZSTD_CCtx* cctx, |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 203 | void* dst, size_t dstCapacity, |
| 204 | const void* src, size_t srcSize, |
| 205 | int compressionLevel); |
| 206 | |
| 207 | /*= Decompression context |
| 208 | * When decompressing many times, |
| 209 | * it is recommended to allocate a context only once, |
| 210 | * and re-use it for each successive compression operation. |
| 211 | * This will make workload friendlier for system's memory. |
| 212 | * Use one context per thread for parallel execution. */ |
| 213 | typedef struct ZSTD_DCtx_s ZSTD_DCtx; |
| 214 | ZSTDLIB_API ZSTD_DCtx* ZSTD_createDCtx(void); |
| 215 | ZSTDLIB_API size_t ZSTD_freeDCtx(ZSTD_DCtx* dctx); |
| 216 | |
| 217 | /*! ZSTD_decompressDCtx() : |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 218 | * Same as ZSTD_decompress(), |
| 219 | * requires an allocated ZSTD_DCtx. |
| 220 | * Compatible with sticky parameters. |
| 221 | */ |
| 222 | ZSTDLIB_API size_t ZSTD_decompressDCtx(ZSTD_DCtx* dctx, |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 223 | void* dst, size_t dstCapacity, |
| 224 | const void* src, size_t srcSize); |
| 225 | |
| 226 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 227 | /*************************************** |
| 228 | * Advanced compression API |
| 229 | ***************************************/ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 230 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 231 | /* API design : |
| 232 | * Parameters are pushed one by one into an existing context, |
| 233 | * using ZSTD_CCtx_set*() functions. |
| 234 | * Pushed parameters are sticky : they are valid for next compressed frame, and any subsequent frame. |
| 235 | * "sticky" parameters are applicable to `ZSTD_compress2()` and `ZSTD_compressStream*()` ! |
| 236 | * They do not apply to "simple" one-shot variants such as ZSTD_compressCCtx() |
| 237 | * |
| 238 | * It's possible to reset all parameters to "default" using ZSTD_CCtx_reset(). |
| 239 | * |
| 240 | * This API supercedes all other "advanced" API entry points in the experimental section. |
| 241 | * In the future, we expect to remove from experimental API entry points which are redundant with this API. |
| 242 | */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 243 | |
| 244 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 245 | /* Compression strategies, listed from fastest to strongest */ |
| 246 | typedef enum { ZSTD_fast=1, |
| 247 | ZSTD_dfast=2, |
| 248 | ZSTD_greedy=3, |
| 249 | ZSTD_lazy=4, |
| 250 | ZSTD_lazy2=5, |
| 251 | ZSTD_btlazy2=6, |
| 252 | ZSTD_btopt=7, |
| 253 | ZSTD_btultra=8, |
| 254 | ZSTD_btultra2=9 |
| 255 | /* note : new strategies _might_ be added in the future. |
| 256 | Only the order (from fast to strong) is guaranteed */ |
| 257 | } ZSTD_strategy; |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 258 | |
| 259 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 260 | typedef enum { |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 261 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 262 | /* compression parameters |
| 263 | * Note: When compressing with a ZSTD_CDict these parameters are superseded |
| 264 | * by the parameters used to construct the ZSTD_CDict. See ZSTD_CCtx_refCDict() |
| 265 | * for more info (superseded-by-cdict). */ |
| 266 | ZSTD_c_compressionLevel=100, /* Update all compression parameters according to pre-defined cLevel table |
| 267 | * Default level is ZSTD_CLEVEL_DEFAULT==3. |
| 268 | * Special: value 0 means default, which is controlled by ZSTD_CLEVEL_DEFAULT. |
| 269 | * Note 1 : it's possible to pass a negative compression level. |
| 270 | * Note 2 : setting a level sets all default values of other compression parameters */ |
| 271 | ZSTD_c_windowLog=101, /* Maximum allowed back-reference distance, expressed as power of 2. |
| 272 | * Must be clamped between ZSTD_WINDOWLOG_MIN and ZSTD_WINDOWLOG_MAX. |
| 273 | * Special: value 0 means "use default windowLog". |
| 274 | * Note: Using a windowLog greater than ZSTD_WINDOWLOG_LIMIT_DEFAULT |
| 275 | * requires explicitly allowing such window size at decompression stage if using streaming. */ |
| 276 | ZSTD_c_hashLog=102, /* Size of the initial probe table, as a power of 2. |
| 277 | * Resulting memory usage is (1 << (hashLog+2)). |
| 278 | * Must be clamped between ZSTD_HASHLOG_MIN and ZSTD_HASHLOG_MAX. |
| 279 | * Larger tables improve compression ratio of strategies <= dFast, |
| 280 | * and improve speed of strategies > dFast. |
| 281 | * Special: value 0 means "use default hashLog". */ |
| 282 | ZSTD_c_chainLog=103, /* Size of the multi-probe search table, as a power of 2. |
| 283 | * Resulting memory usage is (1 << (chainLog+2)). |
| 284 | * Must be clamped between ZSTD_CHAINLOG_MIN and ZSTD_CHAINLOG_MAX. |
| 285 | * Larger tables result in better and slower compression. |
| 286 | * This parameter is useless when using "fast" strategy. |
| 287 | * It's still useful when using "dfast" strategy, |
| 288 | * in which case it defines a secondary probe table. |
| 289 | * Special: value 0 means "use default chainLog". */ |
| 290 | ZSTD_c_searchLog=104, /* Number of search attempts, as a power of 2. |
| 291 | * More attempts result in better and slower compression. |
| 292 | * This parameter is useless when using "fast" and "dFast" strategies. |
| 293 | * Special: value 0 means "use default searchLog". */ |
| 294 | ZSTD_c_minMatch=105, /* Minimum size of searched matches. |
| 295 | * Note that Zstandard can still find matches of smaller size, |
| 296 | * it just tweaks its search algorithm to look for this size and larger. |
| 297 | * Larger values increase compression and decompression speed, but decrease ratio. |
| 298 | * Must be clamped between ZSTD_MINMATCH_MIN and ZSTD_MINMATCH_MAX. |
| 299 | * Note that currently, for all strategies < btopt, effective minimum is 4. |
| 300 | * , for all strategies > fast, effective maximum is 6. |
| 301 | * Special: value 0 means "use default minMatchLength". */ |
| 302 | ZSTD_c_targetLength=106, /* Impact of this field depends on strategy. |
| 303 | * For strategies btopt, btultra & btultra2: |
| 304 | * Length of Match considered "good enough" to stop search. |
| 305 | * Larger values make compression stronger, and slower. |
| 306 | * For strategy fast: |
| 307 | * Distance between match sampling. |
| 308 | * Larger values make compression faster, and weaker. |
| 309 | * Special: value 0 means "use default targetLength". */ |
| 310 | ZSTD_c_strategy=107, /* See ZSTD_strategy enum definition. |
| 311 | * The higher the value of selected strategy, the more complex it is, |
| 312 | * resulting in stronger and slower compression. |
| 313 | * Special: value 0 means "use default strategy". */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 314 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 315 | /* LDM mode parameters */ |
| 316 | ZSTD_c_enableLongDistanceMatching=160, /* Enable long distance matching. |
| 317 | * This parameter is designed to improve compression ratio |
| 318 | * for large inputs, by finding large matches at long distance. |
| 319 | * It increases memory usage and window size. |
| 320 | * Note: enabling this parameter increases default ZSTD_c_windowLog to 128 MB |
| 321 | * except when expressly set to a different value. */ |
| 322 | ZSTD_c_ldmHashLog=161, /* Size of the table for long distance matching, as a power of 2. |
| 323 | * Larger values increase memory usage and compression ratio, |
| 324 | * but decrease compression speed. |
| 325 | * Must be clamped between ZSTD_HASHLOG_MIN and ZSTD_HASHLOG_MAX |
| 326 | * default: windowlog - 7. |
| 327 | * Special: value 0 means "automatically determine hashlog". */ |
| 328 | ZSTD_c_ldmMinMatch=162, /* Minimum match size for long distance matcher. |
| 329 | * Larger/too small values usually decrease compression ratio. |
| 330 | * Must be clamped between ZSTD_LDM_MINMATCH_MIN and ZSTD_LDM_MINMATCH_MAX. |
| 331 | * Special: value 0 means "use default value" (default: 64). */ |
| 332 | ZSTD_c_ldmBucketSizeLog=163, /* Log size of each bucket in the LDM hash table for collision resolution. |
| 333 | * Larger values improve collision resolution but decrease compression speed. |
| 334 | * The maximum value is ZSTD_LDM_BUCKETSIZELOG_MAX. |
| 335 | * Special: value 0 means "use default value" (default: 3). */ |
| 336 | ZSTD_c_ldmHashRateLog=164, /* Frequency of inserting/looking up entries into the LDM hash table. |
| 337 | * Must be clamped between 0 and (ZSTD_WINDOWLOG_MAX - ZSTD_HASHLOG_MIN). |
| 338 | * Default is MAX(0, (windowLog - ldmHashLog)), optimizing hash table usage. |
| 339 | * Larger values improve compression speed. |
| 340 | * Deviating far from default value will likely result in a compression ratio decrease. |
| 341 | * Special: value 0 means "automatically determine hashRateLog". */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 342 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 343 | /* frame parameters */ |
| 344 | ZSTD_c_contentSizeFlag=200, /* Content size will be written into frame header _whenever known_ (default:1) |
| 345 | * Content size must be known at the beginning of compression. |
| 346 | * This is automatically the case when using ZSTD_compress2(), |
| 347 | * For streaming variants, content size must be provided with ZSTD_CCtx_setPledgedSrcSize() */ |
| 348 | ZSTD_c_checksumFlag=201, /* A 32-bits checksum of content is written at end of frame (default:0) */ |
| 349 | ZSTD_c_dictIDFlag=202, /* When applicable, dictionary's ID is written into frame header (default:1) */ |
| 350 | |
| 351 | /* multi-threading parameters */ |
| 352 | /* These parameters are only useful if multi-threading is enabled (compiled with build macro ZSTD_MULTITHREAD). |
| 353 | * They return an error otherwise. */ |
| 354 | ZSTD_c_nbWorkers=400, /* Select how many threads will be spawned to compress in parallel. |
| 355 | * When nbWorkers >= 1, triggers asynchronous mode when used with ZSTD_compressStream*() : |
| 356 | * ZSTD_compressStream*() consumes input and flush output if possible, but immediately gives back control to caller, |
| 357 | * while compression work is performed in parallel, within worker threads. |
| 358 | * (note : a strong exception to this rule is when first invocation of ZSTD_compressStream2() sets ZSTD_e_end : |
| 359 | * in which case, ZSTD_compressStream2() delegates to ZSTD_compress2(), which is always a blocking call). |
| 360 | * More workers improve speed, but also increase memory usage. |
| 361 | * Default value is `0`, aka "single-threaded mode" : no worker is spawned, compression is performed inside Caller's thread, all invocations are blocking */ |
| 362 | ZSTD_c_jobSize=401, /* Size of a compression job. This value is enforced only when nbWorkers >= 1. |
| 363 | * Each compression job is completed in parallel, so this value can indirectly impact the nb of active threads. |
| 364 | * 0 means default, which is dynamically determined based on compression parameters. |
| 365 | * Job size must be a minimum of overlap size, or 1 MB, whichever is largest. |
| 366 | * The minimum size is automatically and transparently enforced */ |
| 367 | ZSTD_c_overlapLog=402, /* Control the overlap size, as a fraction of window size. |
| 368 | * The overlap size is an amount of data reloaded from previous job at the beginning of a new job. |
| 369 | * It helps preserve compression ratio, while each job is compressed in parallel. |
| 370 | * This value is enforced only when nbWorkers >= 1. |
| 371 | * Larger values increase compression ratio, but decrease speed. |
| 372 | * Possible values range from 0 to 9 : |
| 373 | * - 0 means "default" : value will be determined by the library, depending on strategy |
| 374 | * - 1 means "no overlap" |
| 375 | * - 9 means "full overlap", using a full window size. |
| 376 | * Each intermediate rank increases/decreases load size by a factor 2 : |
| 377 | * 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 |
| 378 | * default value varies between 6 and 9, depending on strategy */ |
| 379 | |
| 380 | /* note : additional experimental parameters are also available |
| 381 | * within the experimental section of the API. |
| 382 | * At the time of this writing, they include : |
| 383 | * ZSTD_c_rsyncable |
| 384 | * ZSTD_c_format |
| 385 | * ZSTD_c_forceMaxWindow |
| 386 | * ZSTD_c_forceAttachDict |
| 387 | * ZSTD_c_literalCompressionMode |
| 388 | * ZSTD_c_targetCBlockSize |
| 389 | * Because they are not stable, it's necessary to define ZSTD_STATIC_LINKING_ONLY to access them. |
| 390 | * note : never ever use experimentalParam? names directly; |
| 391 | * also, the enums values themselves are unstable and can still change. |
| 392 | */ |
| 393 | ZSTD_c_experimentalParam1=500, |
| 394 | ZSTD_c_experimentalParam2=10, |
| 395 | ZSTD_c_experimentalParam3=1000, |
| 396 | ZSTD_c_experimentalParam4=1001, |
| 397 | ZSTD_c_experimentalParam5=1002, |
| 398 | ZSTD_c_experimentalParam6=1003, |
| 399 | } ZSTD_cParameter; |
| 400 | |
| 401 | typedef struct { |
| 402 | size_t error; |
| 403 | int lowerBound; |
| 404 | int upperBound; |
| 405 | } ZSTD_bounds; |
| 406 | |
| 407 | /*! ZSTD_cParam_getBounds() : |
| 408 | * All parameters must belong to an interval with lower and upper bounds, |
| 409 | * otherwise they will either trigger an error or be automatically clamped. |
| 410 | * @return : a structure, ZSTD_bounds, which contains |
| 411 | * - an error status field, which must be tested using ZSTD_isError() |
| 412 | * - lower and upper bounds, both inclusive |
| 413 | */ |
| 414 | ZSTDLIB_API ZSTD_bounds ZSTD_cParam_getBounds(ZSTD_cParameter cParam); |
| 415 | |
| 416 | /*! ZSTD_CCtx_setParameter() : |
| 417 | * Set one compression parameter, selected by enum ZSTD_cParameter. |
| 418 | * All parameters have valid bounds. Bounds can be queried using ZSTD_cParam_getBounds(). |
| 419 | * Providing a value beyond bound will either clamp it, or trigger an error (depending on parameter). |
| 420 | * Setting a parameter is generally only possible during frame initialization (before starting compression). |
| 421 | * Exception : when using multi-threading mode (nbWorkers >= 1), |
| 422 | * the following parameters can be updated _during_ compression (within same frame): |
| 423 | * => compressionLevel, hashLog, chainLog, searchLog, minMatch, targetLength and strategy. |
| 424 | * new parameters will be active for next job only (after a flush()). |
| 425 | * @return : an error code (which can be tested using ZSTD_isError()). |
| 426 | */ |
| 427 | ZSTDLIB_API size_t ZSTD_CCtx_setParameter(ZSTD_CCtx* cctx, ZSTD_cParameter param, int value); |
| 428 | |
| 429 | /*! ZSTD_CCtx_setPledgedSrcSize() : |
| 430 | * Total input data size to be compressed as a single frame. |
| 431 | * Value will be written in frame header, unless if explicitly forbidden using ZSTD_c_contentSizeFlag. |
| 432 | * This value will also be controlled at end of frame, and trigger an error if not respected. |
| 433 | * @result : 0, or an error code (which can be tested with ZSTD_isError()). |
| 434 | * Note 1 : pledgedSrcSize==0 actually means zero, aka an empty frame. |
| 435 | * In order to mean "unknown content size", pass constant ZSTD_CONTENTSIZE_UNKNOWN. |
| 436 | * ZSTD_CONTENTSIZE_UNKNOWN is default value for any new frame. |
| 437 | * Note 2 : pledgedSrcSize is only valid once, for the next frame. |
| 438 | * It's discarded at the end of the frame, and replaced by ZSTD_CONTENTSIZE_UNKNOWN. |
| 439 | * Note 3 : Whenever all input data is provided and consumed in a single round, |
| 440 | * for example with ZSTD_compress2(), |
| 441 | * or invoking immediately ZSTD_compressStream2(,,,ZSTD_e_end), |
| 442 | * this value is automatically overridden by srcSize instead. |
| 443 | */ |
| 444 | ZSTDLIB_API size_t ZSTD_CCtx_setPledgedSrcSize(ZSTD_CCtx* cctx, unsigned long long pledgedSrcSize); |
| 445 | |
| 446 | typedef enum { |
| 447 | ZSTD_reset_session_only = 1, |
| 448 | ZSTD_reset_parameters = 2, |
| 449 | ZSTD_reset_session_and_parameters = 3 |
| 450 | } ZSTD_ResetDirective; |
| 451 | |
| 452 | /*! ZSTD_CCtx_reset() : |
| 453 | * There are 2 different things that can be reset, independently or jointly : |
| 454 | * - The session : will stop compressing current frame, and make CCtx ready to start a new one. |
| 455 | * Useful after an error, or to interrupt any ongoing compression. |
| 456 | * Any internal data not yet flushed is cancelled. |
| 457 | * Compression parameters and dictionary remain unchanged. |
| 458 | * They will be used to compress next frame. |
| 459 | * Resetting session never fails. |
| 460 | * - The parameters : changes all parameters back to "default". |
| 461 | * This removes any reference to any dictionary too. |
| 462 | * Parameters can only be changed between 2 sessions (i.e. no compression is currently ongoing) |
| 463 | * otherwise the reset fails, and function returns an error value (which can be tested using ZSTD_isError()) |
| 464 | * - Both : similar to resetting the session, followed by resetting parameters. |
| 465 | */ |
| 466 | ZSTDLIB_API size_t ZSTD_CCtx_reset(ZSTD_CCtx* cctx, ZSTD_ResetDirective reset); |
| 467 | |
| 468 | /*! ZSTD_compress2() : |
| 469 | * Behave the same as ZSTD_compressCCtx(), but compression parameters are set using the advanced API. |
| 470 | * ZSTD_compress2() always starts a new frame. |
| 471 | * Should cctx hold data from a previously unfinished frame, everything about it is forgotten. |
| 472 | * - Compression parameters are pushed into CCtx before starting compression, using ZSTD_CCtx_set*() |
| 473 | * - The function is always blocking, returns when compression is completed. |
| 474 | * Hint : compression runs faster if `dstCapacity` >= `ZSTD_compressBound(srcSize)`. |
| 475 | * @return : compressed size written into `dst` (<= `dstCapacity), |
| 476 | * or an error code if it fails (which can be tested using ZSTD_isError()). |
| 477 | */ |
| 478 | ZSTDLIB_API size_t ZSTD_compress2( ZSTD_CCtx* cctx, |
| 479 | void* dst, size_t dstCapacity, |
| 480 | const void* src, size_t srcSize); |
| 481 | |
| 482 | |
| 483 | /*************************************** |
| 484 | * Advanced decompression API |
| 485 | ***************************************/ |
| 486 | |
| 487 | /* The advanced API pushes parameters one by one into an existing DCtx context. |
| 488 | * Parameters are sticky, and remain valid for all following frames |
| 489 | * using the same DCtx context. |
| 490 | * It's possible to reset parameters to default values using ZSTD_DCtx_reset(). |
| 491 | * Note : This API is compatible with existing ZSTD_decompressDCtx() and ZSTD_decompressStream(). |
| 492 | * Therefore, no new decompression function is necessary. |
| 493 | */ |
| 494 | |
| 495 | typedef enum { |
| 496 | |
| 497 | ZSTD_d_windowLogMax=100, /* Select a size limit (in power of 2) beyond which |
| 498 | * the streaming API will refuse to allocate memory buffer |
| 499 | * in order to protect the host from unreasonable memory requirements. |
| 500 | * This parameter is only useful in streaming mode, since no internal buffer is allocated in single-pass mode. |
| 501 | * By default, a decompression context accepts window sizes <= (1 << ZSTD_WINDOWLOG_LIMIT_DEFAULT). |
| 502 | * Special: value 0 means "use default maximum windowLog". */ |
| 503 | |
| 504 | /* note : additional experimental parameters are also available |
| 505 | * within the experimental section of the API. |
| 506 | * At the time of this writing, they include : |
| 507 | * ZSTD_c_format |
| 508 | * Because they are not stable, it's necessary to define ZSTD_STATIC_LINKING_ONLY to access them. |
| 509 | * note : never ever use experimentalParam? names directly |
| 510 | */ |
| 511 | ZSTD_d_experimentalParam1=1000 |
| 512 | |
| 513 | } ZSTD_dParameter; |
| 514 | |
| 515 | /*! ZSTD_dParam_getBounds() : |
| 516 | * All parameters must belong to an interval with lower and upper bounds, |
| 517 | * otherwise they will either trigger an error or be automatically clamped. |
| 518 | * @return : a structure, ZSTD_bounds, which contains |
| 519 | * - an error status field, which must be tested using ZSTD_isError() |
| 520 | * - both lower and upper bounds, inclusive |
| 521 | */ |
| 522 | ZSTDLIB_API ZSTD_bounds ZSTD_dParam_getBounds(ZSTD_dParameter dParam); |
| 523 | |
| 524 | /*! ZSTD_DCtx_setParameter() : |
| 525 | * Set one compression parameter, selected by enum ZSTD_dParameter. |
| 526 | * All parameters have valid bounds. Bounds can be queried using ZSTD_dParam_getBounds(). |
| 527 | * Providing a value beyond bound will either clamp it, or trigger an error (depending on parameter). |
| 528 | * Setting a parameter is only possible during frame initialization (before starting decompression). |
| 529 | * @return : 0, or an error code (which can be tested using ZSTD_isError()). |
| 530 | */ |
| 531 | ZSTDLIB_API size_t ZSTD_DCtx_setParameter(ZSTD_DCtx* dctx, ZSTD_dParameter param, int value); |
| 532 | |
| 533 | /*! ZSTD_DCtx_reset() : |
| 534 | * Return a DCtx to clean state. |
| 535 | * Session and parameters can be reset jointly or separately. |
| 536 | * Parameters can only be reset when no active frame is being decompressed. |
| 537 | * @return : 0, or an error code, which can be tested with ZSTD_isError() |
| 538 | */ |
| 539 | ZSTDLIB_API size_t ZSTD_DCtx_reset(ZSTD_DCtx* dctx, ZSTD_ResetDirective reset); |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 540 | |
| 541 | |
| 542 | /**************************** |
| 543 | * Streaming |
| 544 | ****************************/ |
| 545 | |
| 546 | typedef struct ZSTD_inBuffer_s { |
| 547 | const void* src; /**< start of input buffer */ |
| 548 | size_t size; /**< size of input buffer */ |
| 549 | size_t pos; /**< position where reading stopped. Will be updated. Necessarily 0 <= pos <= size */ |
| 550 | } ZSTD_inBuffer; |
| 551 | |
| 552 | typedef struct ZSTD_outBuffer_s { |
| 553 | void* dst; /**< start of output buffer */ |
| 554 | size_t size; /**< size of output buffer */ |
| 555 | size_t pos; /**< position where writing stopped. Will be updated. Necessarily 0 <= pos <= size */ |
| 556 | } ZSTD_outBuffer; |
| 557 | |
| 558 | |
| 559 | |
| 560 | /*-*********************************************************************** |
| 561 | * Streaming compression - HowTo |
| 562 | * |
| 563 | * A ZSTD_CStream object is required to track streaming operation. |
| 564 | * Use ZSTD_createCStream() and ZSTD_freeCStream() to create/release resources. |
| 565 | * ZSTD_CStream objects can be reused multiple times on consecutive compression operations. |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 566 | * It is recommended to re-use ZSTD_CStream since it will play nicer with system's memory, by re-using already allocated memory. |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 567 | * |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 568 | * For parallel execution, use one separate ZSTD_CStream per thread. |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 569 | * |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 570 | * note : since v1.3.0, ZSTD_CStream and ZSTD_CCtx are the same thing. |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 571 | * |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 572 | * Parameters are sticky : when starting a new compression on the same context, |
| 573 | * it will re-use the same sticky parameters as previous compression session. |
| 574 | * When in doubt, it's recommended to fully initialize the context before usage. |
| 575 | * Use ZSTD_CCtx_reset() to reset the context and ZSTD_CCtx_setParameter(), |
| 576 | * ZSTD_CCtx_setPledgedSrcSize(), or ZSTD_CCtx_loadDictionary() and friends to |
| 577 | * set more specific parameters, the pledged source size, or load a dictionary. |
| 578 | * |
| 579 | * Use ZSTD_compressStream2() with ZSTD_e_continue as many times as necessary to |
| 580 | * consume input stream. The function will automatically update both `pos` |
| 581 | * fields within `input` and `output`. |
| 582 | * Note that the function may not consume the entire input, for example, because |
| 583 | * the output buffer is already full, in which case `input.pos < input.size`. |
| 584 | * The caller must check if input has been entirely consumed. |
| 585 | * If not, the caller must make some room to receive more compressed data, |
| 586 | * and then present again remaining input data. |
| 587 | * note: ZSTD_e_continue is guaranteed to make some forward progress when called, |
| 588 | * but doesn't guarantee maximal forward progress. This is especially relevant |
| 589 | * when compressing with multiple threads. The call won't block if it can |
| 590 | * consume some input, but if it can't it will wait for some, but not all, |
| 591 | * output to be flushed. |
| 592 | * @return : provides a minimum amount of data remaining to be flushed from internal buffers |
| 593 | * or an error code, which can be tested using ZSTD_isError(). |
| 594 | * |
| 595 | * At any moment, it's possible to flush whatever data might remain stuck within internal buffer, |
| 596 | * using ZSTD_compressStream2() with ZSTD_e_flush. `output->pos` will be updated. |
| 597 | * Note that, if `output->size` is too small, a single invocation with ZSTD_e_flush might not be enough (return code > 0). |
| 598 | * In which case, make some room to receive more compressed data, and call again ZSTD_compressStream2() with ZSTD_e_flush. |
| 599 | * You must continue calling ZSTD_compressStream2() with ZSTD_e_flush until it returns 0, at which point you can change the |
| 600 | * operation. |
| 601 | * note: ZSTD_e_flush will flush as much output as possible, meaning when compressing with multiple threads, it will |
| 602 | * block until the flush is complete or the output buffer is full. |
| 603 | * @return : 0 if internal buffers are entirely flushed, |
| 604 | * >0 if some data still present within internal buffer (the value is minimal estimation of remaining size), |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 605 | * or an error code, which can be tested using ZSTD_isError(). |
| 606 | * |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 607 | * Calling ZSTD_compressStream2() with ZSTD_e_end instructs to finish a frame. |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 608 | * It will perform a flush and write frame epilogue. |
| 609 | * The epilogue is required for decoders to consider a frame completed. |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 610 | * flush operation is the same, and follows same rules as calling ZSTD_compressStream2() with ZSTD_e_flush. |
| 611 | * You must continue calling ZSTD_compressStream2() with ZSTD_e_end until it returns 0, at which point you are free to |
| 612 | * start a new frame. |
| 613 | * note: ZSTD_e_end will flush as much output as possible, meaning when compressing with multiple threads, it will |
| 614 | * block until the flush is complete or the output buffer is full. |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 615 | * @return : 0 if frame fully completed and fully flushed, |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 616 | * >0 if some data still present within internal buffer (the value is minimal estimation of remaining size), |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 617 | * or an error code, which can be tested using ZSTD_isError(). |
| 618 | * |
| 619 | * *******************************************************************/ |
| 620 | |
| 621 | typedef ZSTD_CCtx ZSTD_CStream; /**< CCtx and CStream are now effectively same object (>= v1.3.0) */ |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 622 | /* Continue to distinguish them for compatibility with older versions <= v1.2.0 */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 623 | /*===== ZSTD_CStream management functions =====*/ |
| 624 | ZSTDLIB_API ZSTD_CStream* ZSTD_createCStream(void); |
| 625 | ZSTDLIB_API size_t ZSTD_freeCStream(ZSTD_CStream* zcs); |
| 626 | |
| 627 | /*===== Streaming compression functions =====*/ |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 628 | typedef enum { |
| 629 | ZSTD_e_continue=0, /* collect more data, encoder decides when to output compressed result, for optimal compression ratio */ |
| 630 | ZSTD_e_flush=1, /* flush any data provided so far, |
| 631 | * it creates (at least) one new block, that can be decoded immediately on reception; |
| 632 | * frame will continue: any future data can still reference previously compressed data, improving compression. |
| 633 | * note : multithreaded compression will block to flush as much output as possible. */ |
| 634 | ZSTD_e_end=2 /* flush any remaining data _and_ close current frame. |
| 635 | * note that frame is only closed after compressed data is fully flushed (return value == 0). |
| 636 | * After that point, any additional data starts a new frame. |
| 637 | * note : each frame is independent (does not reference any content from previous frame). |
| 638 | : note : multithreaded compression will block to flush as much output as possible. */ |
| 639 | } ZSTD_EndDirective; |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 640 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 641 | /*! ZSTD_compressStream2() : |
| 642 | * Behaves about the same as ZSTD_compressStream, with additional control on end directive. |
| 643 | * - Compression parameters are pushed into CCtx before starting compression, using ZSTD_CCtx_set*() |
| 644 | * - Compression parameters cannot be changed once compression is started (save a list of exceptions in multi-threading mode) |
| 645 | * - output->pos must be <= dstCapacity, input->pos must be <= srcSize |
| 646 | * - output->pos and input->pos will be updated. They are guaranteed to remain below their respective limit. |
| 647 | * - When nbWorkers==0 (default), function is blocking : it completes its job before returning to caller. |
| 648 | * - 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, |
| 649 | * and then immediately returns, just indicating that there is some data remaining to be flushed. |
| 650 | * The function nonetheless guarantees forward progress : it will return only after it reads or write at least 1+ byte. |
| 651 | * - 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. |
| 652 | * - @return provides a minimum amount of data remaining to be flushed from internal buffers |
| 653 | * or an error code, which can be tested using ZSTD_isError(). |
| 654 | * if @return != 0, flush is not fully completed, there is still some data left within internal buffers. |
| 655 | * This is useful for ZSTD_e_flush, since in this case more flushes are necessary to empty all buffers. |
| 656 | * For ZSTD_e_end, @return == 0 when internal buffers are fully flushed and frame is completed. |
| 657 | * - after a ZSTD_e_end directive, if internal buffer is not fully flushed (@return != 0), |
| 658 | * only ZSTD_e_end or ZSTD_e_flush operations are allowed. |
| 659 | * Before starting a new compression job, or changing compression parameters, |
| 660 | * it is required to fully flush internal buffers. |
| 661 | */ |
| 662 | ZSTDLIB_API size_t ZSTD_compressStream2( ZSTD_CCtx* cctx, |
| 663 | ZSTD_outBuffer* output, |
| 664 | ZSTD_inBuffer* input, |
| 665 | ZSTD_EndDirective endOp); |
| 666 | |
| 667 | |
| 668 | /* These buffer sizes are softly recommended. |
| 669 | * They are not required : ZSTD_compressStream*() happily accepts any buffer size, for both input and output. |
| 670 | * Respecting the recommended size just makes it a bit easier for ZSTD_compressStream*(), |
| 671 | * reducing the amount of memory shuffling and buffering, resulting in minor performance savings. |
| 672 | * |
| 673 | * However, note that these recommendations are from the perspective of a C caller program. |
| 674 | * If the streaming interface is invoked from some other language, |
| 675 | * especially managed ones such as Java or Go, through a foreign function interface such as jni or cgo, |
| 676 | * a major performance rule is to reduce crossing such interface to an absolute minimum. |
| 677 | * It's not rare that performance ends being spent more into the interface, rather than compression itself. |
| 678 | * In which cases, prefer using large buffers, as large as practical, |
| 679 | * for both input and output, to reduce the nb of roundtrips. |
| 680 | */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 681 | ZSTDLIB_API size_t ZSTD_CStreamInSize(void); /**< recommended size for input buffer */ |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 682 | ZSTDLIB_API size_t ZSTD_CStreamOutSize(void); /**< recommended size for output buffer. Guarantee to successfully flush at least one complete compressed block. */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 683 | |
| 684 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 685 | /* ***************************************************************************** |
| 686 | * This following is a legacy streaming API. |
| 687 | * It can be replaced by ZSTD_CCtx_reset() and ZSTD_compressStream2(). |
| 688 | * It is redundant, but remains fully supported. |
| 689 | * Advanced parameters and dictionary compression can only be used through the |
| 690 | * new API. |
| 691 | ******************************************************************************/ |
| 692 | |
| 693 | /*! |
| 694 | * Equivalent to: |
| 695 | * |
| 696 | * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); |
| 697 | * ZSTD_CCtx_refCDict(zcs, NULL); // clear the dictionary (if any) |
| 698 | * ZSTD_CCtx_setParameter(zcs, ZSTD_c_compressionLevel, compressionLevel); |
| 699 | */ |
| 700 | ZSTDLIB_API size_t ZSTD_initCStream(ZSTD_CStream* zcs, int compressionLevel); |
| 701 | /*! |
| 702 | * Alternative for ZSTD_compressStream2(zcs, output, input, ZSTD_e_continue). |
| 703 | * NOTE: The return value is different. ZSTD_compressStream() returns a hint for |
| 704 | * the next read size (if non-zero and not an error). ZSTD_compressStream2() |
| 705 | * returns the minimum nb of bytes left to flush (if non-zero and not an error). |
| 706 | */ |
| 707 | ZSTDLIB_API size_t ZSTD_compressStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output, ZSTD_inBuffer* input); |
| 708 | /*! Equivalent to ZSTD_compressStream2(zcs, output, &emptyInput, ZSTD_e_flush). */ |
| 709 | ZSTDLIB_API size_t ZSTD_flushStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output); |
| 710 | /*! Equivalent to ZSTD_compressStream2(zcs, output, &emptyInput, ZSTD_e_end). */ |
| 711 | ZSTDLIB_API size_t ZSTD_endStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output); |
| 712 | |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 713 | |
| 714 | /*-*************************************************************************** |
| 715 | * Streaming decompression - HowTo |
| 716 | * |
| 717 | * A ZSTD_DStream object is required to track streaming operations. |
| 718 | * Use ZSTD_createDStream() and ZSTD_freeDStream() to create/release resources. |
| 719 | * ZSTD_DStream objects can be re-used multiple times. |
| 720 | * |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 721 | * Use ZSTD_initDStream() to start a new decompression operation. |
| 722 | * @return : recommended first input size |
| 723 | * Alternatively, use advanced API to set specific properties. |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 724 | * |
| 725 | * Use ZSTD_decompressStream() repetitively to consume your input. |
| 726 | * The function will update both `pos` fields. |
| 727 | * If `input.pos < input.size`, some input has not been consumed. |
| 728 | * It's up to the caller to present again remaining data. |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 729 | * The function tries to flush all data decoded immediately, respecting output buffer size. |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 730 | * If `output.pos < output.size`, decoder has flushed everything it could. |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 731 | * But if `output.pos == output.size`, there might be some data left within internal buffers., |
| 732 | * In which case, call ZSTD_decompressStream() again to flush whatever remains in the buffer. |
| 733 | * Note : with no additional input provided, amount of data flushed is necessarily <= ZSTD_BLOCKSIZE_MAX. |
| 734 | * @return : 0 when a frame is completely decoded and fully flushed, |
| 735 | * or an error code, which can be tested using ZSTD_isError(), |
| 736 | * or any other value > 0, which means there is still some decoding or flushing to do to complete current frame : |
| 737 | * the return value is a suggested next input size (just a hint for better latency) |
| 738 | * that will never request more than the remaining frame size. |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 739 | * *******************************************************************************/ |
| 740 | |
| 741 | typedef ZSTD_DCtx ZSTD_DStream; /**< DCtx and DStream are now effectively same object (>= v1.3.0) */ |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 742 | /* For compatibility with versions <= v1.2.0, prefer differentiating them. */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 743 | /*===== ZSTD_DStream management functions =====*/ |
| 744 | ZSTDLIB_API ZSTD_DStream* ZSTD_createDStream(void); |
| 745 | ZSTDLIB_API size_t ZSTD_freeDStream(ZSTD_DStream* zds); |
| 746 | |
| 747 | /*===== Streaming decompression functions =====*/ |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 748 | |
| 749 | /* This function is redundant with the advanced API and equivalent to: |
| 750 | * |
| 751 | * ZSTD_DCtx_reset(zds); |
| 752 | * ZSTD_DCtx_refDDict(zds, NULL); |
| 753 | */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 754 | ZSTDLIB_API size_t ZSTD_initDStream(ZSTD_DStream* zds); |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 755 | |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 756 | ZSTDLIB_API size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inBuffer* input); |
| 757 | |
| 758 | ZSTDLIB_API size_t ZSTD_DStreamInSize(void); /*!< recommended size for input buffer */ |
| 759 | ZSTDLIB_API size_t ZSTD_DStreamOutSize(void); /*!< recommended size for output buffer. Guarantee to successfully flush at least one complete block in all circumstances. */ |
| 760 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 761 | |
| 762 | /************************** |
| 763 | * Simple dictionary API |
| 764 | ***************************/ |
| 765 | /*! ZSTD_compress_usingDict() : |
| 766 | * Compression at an explicit compression level using a Dictionary. |
| 767 | * A dictionary can be any arbitrary data segment (also called a prefix), |
| 768 | * or a buffer with specified information (see dictBuilder/zdict.h). |
| 769 | * Note : This function loads the dictionary, resulting in significant startup delay. |
| 770 | * It's intended for a dictionary used only once. |
| 771 | * Note 2 : When `dict == NULL || dictSize < 8` no dictionary is used. */ |
| 772 | ZSTDLIB_API size_t ZSTD_compress_usingDict(ZSTD_CCtx* ctx, |
| 773 | void* dst, size_t dstCapacity, |
| 774 | const void* src, size_t srcSize, |
| 775 | const void* dict,size_t dictSize, |
| 776 | int compressionLevel); |
| 777 | |
| 778 | /*! ZSTD_decompress_usingDict() : |
| 779 | * Decompression using a known Dictionary. |
| 780 | * Dictionary must be identical to the one used during compression. |
| 781 | * Note : This function loads the dictionary, resulting in significant startup delay. |
| 782 | * It's intended for a dictionary used only once. |
| 783 | * Note : When `dict == NULL || dictSize < 8` no dictionary is used. */ |
| 784 | ZSTDLIB_API size_t ZSTD_decompress_usingDict(ZSTD_DCtx* dctx, |
| 785 | void* dst, size_t dstCapacity, |
| 786 | const void* src, size_t srcSize, |
| 787 | const void* dict,size_t dictSize); |
| 788 | |
| 789 | |
| 790 | /*********************************** |
| 791 | * Bulk processing dictionary API |
| 792 | **********************************/ |
| 793 | typedef struct ZSTD_CDict_s ZSTD_CDict; |
| 794 | |
| 795 | /*! ZSTD_createCDict() : |
| 796 | * When compressing multiple messages / blocks using the same dictionary, it's recommended to load it only once. |
| 797 | * ZSTD_createCDict() will create a digested dictionary, ready to start future compression operations without startup cost. |
| 798 | * ZSTD_CDict can be created once and shared by multiple threads concurrently, since its usage is read-only. |
| 799 | * `dictBuffer` can be released after ZSTD_CDict creation, because its content is copied within CDict. |
| 800 | * Consider experimental function `ZSTD_createCDict_byReference()` if you prefer to not duplicate `dictBuffer` content. |
| 801 | * Note : A ZSTD_CDict can be created from an empty dictBuffer, but it is inefficient when used to compress small data. */ |
| 802 | ZSTDLIB_API ZSTD_CDict* ZSTD_createCDict(const void* dictBuffer, size_t dictSize, |
| 803 | int compressionLevel); |
| 804 | |
| 805 | /*! ZSTD_freeCDict() : |
| 806 | * Function frees memory allocated by ZSTD_createCDict(). */ |
| 807 | ZSTDLIB_API size_t ZSTD_freeCDict(ZSTD_CDict* CDict); |
| 808 | |
| 809 | /*! ZSTD_compress_usingCDict() : |
| 810 | * Compression using a digested Dictionary. |
| 811 | * Recommended when same dictionary is used multiple times. |
| 812 | * Note : compression level is _decided at dictionary creation time_, |
| 813 | * and frame parameters are hardcoded (dictID=yes, contentSize=yes, checksum=no) */ |
| 814 | ZSTDLIB_API size_t ZSTD_compress_usingCDict(ZSTD_CCtx* cctx, |
| 815 | void* dst, size_t dstCapacity, |
| 816 | const void* src, size_t srcSize, |
| 817 | const ZSTD_CDict* cdict); |
| 818 | |
| 819 | |
| 820 | typedef struct ZSTD_DDict_s ZSTD_DDict; |
| 821 | |
| 822 | /*! ZSTD_createDDict() : |
| 823 | * Create a digested dictionary, ready to start decompression operation without startup delay. |
| 824 | * dictBuffer can be released after DDict creation, as its content is copied inside DDict. */ |
| 825 | ZSTDLIB_API ZSTD_DDict* ZSTD_createDDict(const void* dictBuffer, size_t dictSize); |
| 826 | |
| 827 | /*! ZSTD_freeDDict() : |
| 828 | * Function frees memory allocated with ZSTD_createDDict() */ |
| 829 | ZSTDLIB_API size_t ZSTD_freeDDict(ZSTD_DDict* ddict); |
| 830 | |
| 831 | /*! ZSTD_decompress_usingDDict() : |
| 832 | * Decompression using a digested Dictionary. |
| 833 | * Recommended when same dictionary is used multiple times. */ |
| 834 | ZSTDLIB_API size_t ZSTD_decompress_usingDDict(ZSTD_DCtx* dctx, |
| 835 | void* dst, size_t dstCapacity, |
| 836 | const void* src, size_t srcSize, |
| 837 | const ZSTD_DDict* ddict); |
| 838 | |
| 839 | |
| 840 | /******************************** |
| 841 | * Dictionary helper functions |
| 842 | *******************************/ |
| 843 | |
| 844 | /*! ZSTD_getDictID_fromDict() : |
| 845 | * Provides the dictID stored within dictionary. |
| 846 | * if @return == 0, the dictionary is not conformant with Zstandard specification. |
| 847 | * It can still be loaded, but as a content-only dictionary. */ |
| 848 | ZSTDLIB_API unsigned ZSTD_getDictID_fromDict(const void* dict, size_t dictSize); |
| 849 | |
| 850 | /*! ZSTD_getDictID_fromDDict() : |
| 851 | * Provides the dictID of the dictionary loaded into `ddict`. |
| 852 | * If @return == 0, the dictionary is not conformant to Zstandard specification, or empty. |
| 853 | * Non-conformant dictionaries can still be loaded, but as content-only dictionaries. */ |
| 854 | ZSTDLIB_API unsigned ZSTD_getDictID_fromDDict(const ZSTD_DDict* ddict); |
| 855 | |
| 856 | /*! ZSTD_getDictID_fromFrame() : |
| 857 | * Provides the dictID required to decompressed the frame stored within `src`. |
| 858 | * If @return == 0, the dictID could not be decoded. |
| 859 | * This could for one of the following reasons : |
| 860 | * - The frame does not require a dictionary to be decoded (most common case). |
| 861 | * - The frame was built with dictID intentionally removed. Whatever dictionary is necessary is a hidden information. |
| 862 | * Note : this use case also happens when using a non-conformant dictionary. |
| 863 | * - `srcSize` is too small, and as a result, the frame header could not be decoded (only possible if `srcSize < ZSTD_FRAMEHEADERSIZE_MAX`). |
| 864 | * - This is not a Zstandard frame. |
| 865 | * When identifying the exact failure cause, it's possible to use ZSTD_getFrameHeader(), which will provide a more precise error code. */ |
| 866 | ZSTDLIB_API unsigned ZSTD_getDictID_fromFrame(const void* src, size_t srcSize); |
| 867 | |
| 868 | |
| 869 | /******************************************************************************* |
| 870 | * Advanced dictionary and prefix API |
| 871 | * |
| 872 | * This API allows dictionaries to be used with ZSTD_compress2(), |
| 873 | * ZSTD_compressStream2(), and ZSTD_decompress(). Dictionaries are sticky, and |
| 874 | * only reset with the context is reset with ZSTD_reset_parameters or |
| 875 | * ZSTD_reset_session_and_parameters. Prefixes are single-use. |
| 876 | ******************************************************************************/ |
| 877 | |
| 878 | |
| 879 | /*! ZSTD_CCtx_loadDictionary() : |
| 880 | * Create an internal CDict from `dict` buffer. |
| 881 | * Decompression will have to use same dictionary. |
| 882 | * @result : 0, or an error code (which can be tested with ZSTD_isError()). |
| 883 | * Special: Loading a NULL (or 0-size) dictionary invalidates previous dictionary, |
| 884 | * meaning "return to no-dictionary mode". |
| 885 | * Note 1 : Dictionary is sticky, it will be used for all future compressed frames. |
| 886 | * To return to "no-dictionary" situation, load a NULL dictionary (or reset parameters). |
| 887 | * Note 2 : Loading a dictionary involves building tables. |
| 888 | * It's also a CPU consuming operation, with non-negligible impact on latency. |
| 889 | * Tables are dependent on compression parameters, and for this reason, |
| 890 | * compression parameters can no longer be changed after loading a dictionary. |
| 891 | * Note 3 :`dict` content will be copied internally. |
| 892 | * Use experimental ZSTD_CCtx_loadDictionary_byReference() to reference content instead. |
| 893 | * In such a case, dictionary buffer must outlive its users. |
| 894 | * Note 4 : Use ZSTD_CCtx_loadDictionary_advanced() |
| 895 | * to precisely select how dictionary content must be interpreted. */ |
| 896 | ZSTDLIB_API size_t ZSTD_CCtx_loadDictionary(ZSTD_CCtx* cctx, const void* dict, size_t dictSize); |
| 897 | |
| 898 | /*! ZSTD_CCtx_refCDict() : |
| 899 | * Reference a prepared dictionary, to be used for all next compressed frames. |
| 900 | * Note that compression parameters are enforced from within CDict, |
| 901 | * and supersede any compression parameter previously set within CCtx. |
| 902 | * The parameters ignored are labled as "superseded-by-cdict" in the ZSTD_cParameter enum docs. |
| 903 | * The ignored parameters will be used again if the CCtx is returned to no-dictionary mode. |
| 904 | * The dictionary will remain valid for future compressed frames using same CCtx. |
| 905 | * @result : 0, or an error code (which can be tested with ZSTD_isError()). |
| 906 | * Special : Referencing a NULL CDict means "return to no-dictionary mode". |
| 907 | * Note 1 : Currently, only one dictionary can be managed. |
| 908 | * Referencing a new dictionary effectively "discards" any previous one. |
| 909 | * Note 2 : CDict is just referenced, its lifetime must outlive its usage within CCtx. */ |
| 910 | ZSTDLIB_API size_t ZSTD_CCtx_refCDict(ZSTD_CCtx* cctx, const ZSTD_CDict* cdict); |
| 911 | |
| 912 | /*! ZSTD_CCtx_refPrefix() : |
| 913 | * Reference a prefix (single-usage dictionary) for next compressed frame. |
| 914 | * A prefix is **only used once**. Tables are discarded at end of frame (ZSTD_e_end). |
| 915 | * Decompression will need same prefix to properly regenerate data. |
| 916 | * Compressing with a prefix is similar in outcome as performing a diff and compressing it, |
| 917 | * but performs much faster, especially during decompression (compression speed is tunable with compression level). |
| 918 | * @result : 0, or an error code (which can be tested with ZSTD_isError()). |
| 919 | * Special: Adding any prefix (including NULL) invalidates any previous prefix or dictionary |
| 920 | * Note 1 : Prefix buffer is referenced. It **must** outlive compression. |
| 921 | * Its content must remain unmodified during compression. |
| 922 | * Note 2 : If the intention is to diff some large src data blob with some prior version of itself, |
| 923 | * ensure that the window size is large enough to contain the entire source. |
| 924 | * See ZSTD_c_windowLog. |
| 925 | * Note 3 : Referencing a prefix involves building tables, which are dependent on compression parameters. |
| 926 | * It's a CPU consuming operation, with non-negligible impact on latency. |
| 927 | * If there is a need to use the same prefix multiple times, consider loadDictionary instead. |
| 928 | * Note 4 : By default, the prefix is interpreted as raw content (ZSTD_dm_rawContent). |
| 929 | * Use experimental ZSTD_CCtx_refPrefix_advanced() to alter dictionary interpretation. */ |
| 930 | ZSTDLIB_API size_t ZSTD_CCtx_refPrefix(ZSTD_CCtx* cctx, |
| 931 | const void* prefix, size_t prefixSize); |
| 932 | |
| 933 | /*! ZSTD_DCtx_loadDictionary() : |
| 934 | * Create an internal DDict from dict buffer, |
| 935 | * to be used to decompress next frames. |
| 936 | * The dictionary remains valid for all future frames, until explicitly invalidated. |
| 937 | * @result : 0, or an error code (which can be tested with ZSTD_isError()). |
| 938 | * Special : Adding a NULL (or 0-size) dictionary invalidates any previous dictionary, |
| 939 | * meaning "return to no-dictionary mode". |
| 940 | * Note 1 : Loading a dictionary involves building tables, |
| 941 | * which has a non-negligible impact on CPU usage and latency. |
| 942 | * It's recommended to "load once, use many times", to amortize the cost |
| 943 | * Note 2 :`dict` content will be copied internally, so `dict` can be released after loading. |
| 944 | * Use ZSTD_DCtx_loadDictionary_byReference() to reference dictionary content instead. |
| 945 | * Note 3 : Use ZSTD_DCtx_loadDictionary_advanced() to take control of |
| 946 | * how dictionary content is loaded and interpreted. |
| 947 | */ |
| 948 | ZSTDLIB_API size_t ZSTD_DCtx_loadDictionary(ZSTD_DCtx* dctx, const void* dict, size_t dictSize); |
| 949 | |
| 950 | /*! ZSTD_DCtx_refDDict() : |
| 951 | * Reference a prepared dictionary, to be used to decompress next frames. |
| 952 | * The dictionary remains active for decompression of future frames using same DCtx. |
| 953 | * @result : 0, or an error code (which can be tested with ZSTD_isError()). |
| 954 | * Note 1 : Currently, only one dictionary can be managed. |
| 955 | * Referencing a new dictionary effectively "discards" any previous one. |
| 956 | * Special: referencing a NULL DDict means "return to no-dictionary mode". |
| 957 | * Note 2 : DDict is just referenced, its lifetime must outlive its usage from DCtx. |
| 958 | */ |
| 959 | ZSTDLIB_API size_t ZSTD_DCtx_refDDict(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict); |
| 960 | |
| 961 | /*! ZSTD_DCtx_refPrefix() : |
| 962 | * Reference a prefix (single-usage dictionary) to decompress next frame. |
| 963 | * This is the reverse operation of ZSTD_CCtx_refPrefix(), |
| 964 | * and must use the same prefix as the one used during compression. |
| 965 | * Prefix is **only used once**. Reference is discarded at end of frame. |
| 966 | * End of frame is reached when ZSTD_decompressStream() returns 0. |
| 967 | * @result : 0, or an error code (which can be tested with ZSTD_isError()). |
| 968 | * Note 1 : Adding any prefix (including NULL) invalidates any previously set prefix or dictionary |
| 969 | * Note 2 : Prefix buffer is referenced. It **must** outlive decompression. |
| 970 | * Prefix buffer must remain unmodified up to the end of frame, |
| 971 | * reached when ZSTD_decompressStream() returns 0. |
| 972 | * Note 3 : By default, the prefix is treated as raw content (ZSTD_dm_rawContent). |
| 973 | * Use ZSTD_CCtx_refPrefix_advanced() to alter dictMode (Experimental section) |
| 974 | * Note 4 : Referencing a raw content prefix has almost no cpu nor memory cost. |
| 975 | * A full dictionary is more costly, as it requires building tables. |
| 976 | */ |
| 977 | ZSTDLIB_API size_t ZSTD_DCtx_refPrefix(ZSTD_DCtx* dctx, |
| 978 | const void* prefix, size_t prefixSize); |
| 979 | |
| 980 | /* === Memory management === */ |
| 981 | |
| 982 | /*! ZSTD_sizeof_*() : |
| 983 | * These functions give the _current_ memory usage of selected object. |
| 984 | * Note that object memory usage can evolve (increase or decrease) over time. */ |
| 985 | ZSTDLIB_API size_t ZSTD_sizeof_CCtx(const ZSTD_CCtx* cctx); |
| 986 | ZSTDLIB_API size_t ZSTD_sizeof_DCtx(const ZSTD_DCtx* dctx); |
| 987 | ZSTDLIB_API size_t ZSTD_sizeof_CStream(const ZSTD_CStream* zcs); |
| 988 | ZSTDLIB_API size_t ZSTD_sizeof_DStream(const ZSTD_DStream* zds); |
| 989 | ZSTDLIB_API size_t ZSTD_sizeof_CDict(const ZSTD_CDict* cdict); |
| 990 | ZSTDLIB_API size_t ZSTD_sizeof_DDict(const ZSTD_DDict* ddict); |
| 991 | |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 992 | #endif /* ZSTD_H_235446 */ |
| 993 | |
| 994 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 995 | /* ************************************************************************************** |
| 996 | * ADVANCED AND EXPERIMENTAL FUNCTIONS |
| 997 | **************************************************************************************** |
| 998 | * The definitions in the following section are considered experimental. |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 999 | * They are provided for advanced scenarios. |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1000 | * They should never be used with a dynamic library, as prototypes may change in the future. |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1001 | * Use them only in association with static linking. |
| 1002 | * ***************************************************************************************/ |
| 1003 | |
| 1004 | #if defined(ZSTD_STATIC_LINKING_ONLY) && !defined(ZSTD_H_ZSTD_STATIC_LINKING_ONLY) |
| 1005 | #define ZSTD_H_ZSTD_STATIC_LINKING_ONLY |
| 1006 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1007 | /**************************************************************************************** |
| 1008 | * experimental API (static linking only) |
| 1009 | **************************************************************************************** |
| 1010 | * The following symbols and constants |
| 1011 | * are not planned to join "stable API" status in the near future. |
| 1012 | * They can still change in future versions. |
| 1013 | * Some of them are planned to remain in the static_only section indefinitely. |
| 1014 | * Some of them might be removed in the future (especially when redundant with existing stable functions) |
| 1015 | * ***************************************************************************************/ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1016 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1017 | #define ZSTD_FRAMEHEADERSIZE_PREFIX 5 /* minimum input size required to query frame header size */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1018 | #define ZSTD_FRAMEHEADERSIZE_MIN 6 |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1019 | #define ZSTD_FRAMEHEADERSIZE_MAX 18 /* can be useful for static allocation */ |
| 1020 | #define ZSTD_SKIPPABLEHEADERSIZE 8 |
| 1021 | |
| 1022 | /* compression parameter bounds */ |
| 1023 | #define ZSTD_WINDOWLOG_MAX_32 30 |
| 1024 | #define ZSTD_WINDOWLOG_MAX_64 31 |
| 1025 | #define ZSTD_WINDOWLOG_MAX ((int)(sizeof(size_t) == 4 ? ZSTD_WINDOWLOG_MAX_32 : ZSTD_WINDOWLOG_MAX_64)) |
| 1026 | #define ZSTD_WINDOWLOG_MIN 10 |
| 1027 | #define ZSTD_HASHLOG_MAX ((ZSTD_WINDOWLOG_MAX < 30) ? ZSTD_WINDOWLOG_MAX : 30) |
| 1028 | #define ZSTD_HASHLOG_MIN 6 |
| 1029 | #define ZSTD_CHAINLOG_MAX_32 29 |
| 1030 | #define ZSTD_CHAINLOG_MAX_64 30 |
| 1031 | #define ZSTD_CHAINLOG_MAX ((int)(sizeof(size_t) == 4 ? ZSTD_CHAINLOG_MAX_32 : ZSTD_CHAINLOG_MAX_64)) |
| 1032 | #define ZSTD_CHAINLOG_MIN ZSTD_HASHLOG_MIN |
| 1033 | #define ZSTD_SEARCHLOG_MAX (ZSTD_WINDOWLOG_MAX-1) |
| 1034 | #define ZSTD_SEARCHLOG_MIN 1 |
| 1035 | #define ZSTD_MINMATCH_MAX 7 /* only for ZSTD_fast, other strategies are limited to 6 */ |
| 1036 | #define ZSTD_MINMATCH_MIN 3 /* only for ZSTD_btopt+, faster strategies are limited to 4 */ |
| 1037 | #define ZSTD_TARGETLENGTH_MAX ZSTD_BLOCKSIZE_MAX |
| 1038 | #define ZSTD_TARGETLENGTH_MIN 0 /* note : comparing this constant to an unsigned results in a tautological test */ |
| 1039 | #define ZSTD_STRATEGY_MIN ZSTD_fast |
| 1040 | #define ZSTD_STRATEGY_MAX ZSTD_btultra2 |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1041 | |
| 1042 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1043 | #define ZSTD_OVERLAPLOG_MIN 0 |
| 1044 | #define ZSTD_OVERLAPLOG_MAX 9 |
| 1045 | |
| 1046 | #define ZSTD_WINDOWLOG_LIMIT_DEFAULT 27 /* by default, the streaming decoder will refuse any frame |
| 1047 | * requiring larger than (1<<ZSTD_WINDOWLOG_LIMIT_DEFAULT) window size, |
| 1048 | * to preserve host's memory from unreasonable requirements. |
| 1049 | * This limit can be overridden using ZSTD_DCtx_setParameter(,ZSTD_d_windowLogMax,). |
| 1050 | * The limit does not apply for one-pass decoders (such as ZSTD_decompress()), since no additional memory is allocated */ |
| 1051 | |
| 1052 | |
| 1053 | /* LDM parameter bounds */ |
| 1054 | #define ZSTD_LDM_HASHLOG_MIN ZSTD_HASHLOG_MIN |
| 1055 | #define ZSTD_LDM_HASHLOG_MAX ZSTD_HASHLOG_MAX |
| 1056 | #define ZSTD_LDM_MINMATCH_MIN 4 |
| 1057 | #define ZSTD_LDM_MINMATCH_MAX 4096 |
| 1058 | #define ZSTD_LDM_BUCKETSIZELOG_MIN 1 |
| 1059 | #define ZSTD_LDM_BUCKETSIZELOG_MAX 8 |
| 1060 | #define ZSTD_LDM_HASHRATELOG_MIN 0 |
| 1061 | #define ZSTD_LDM_HASHRATELOG_MAX (ZSTD_WINDOWLOG_MAX - ZSTD_HASHLOG_MIN) |
| 1062 | |
| 1063 | /* Advanced parameter bounds */ |
| 1064 | #define ZSTD_TARGETCBLOCKSIZE_MIN 64 |
| 1065 | #define ZSTD_TARGETCBLOCKSIZE_MAX ZSTD_BLOCKSIZE_MAX |
| 1066 | |
| 1067 | /* internal */ |
| 1068 | #define ZSTD_HASHLOG3_MAX 17 |
| 1069 | |
| 1070 | |
| 1071 | /* --- Advanced types --- */ |
| 1072 | |
| 1073 | typedef struct ZSTD_CCtx_params_s ZSTD_CCtx_params; |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1074 | |
| 1075 | typedef struct { |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1076 | unsigned windowLog; /**< largest match distance : larger == more compression, more memory needed during decompression */ |
| 1077 | unsigned chainLog; /**< fully searched segment : larger == more compression, slower, more memory (useless for fast) */ |
| 1078 | unsigned hashLog; /**< dispatch table : larger == faster, more memory */ |
| 1079 | unsigned searchLog; /**< nb of searches : larger == more compression, slower */ |
| 1080 | unsigned minMatch; /**< match length searched : larger == faster decompression, sometimes less compression */ |
| 1081 | unsigned targetLength; /**< acceptable match size for optimal parser (only) : larger == more compression, slower */ |
| 1082 | ZSTD_strategy strategy; /**< see ZSTD_strategy definition above */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1083 | } ZSTD_compressionParameters; |
| 1084 | |
| 1085 | typedef struct { |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1086 | int contentSizeFlag; /**< 1: content size will be in frame header (when known) */ |
| 1087 | int checksumFlag; /**< 1: generate a 32-bits checksum using XXH64 algorithm at end of frame, for error detection */ |
| 1088 | int noDictIDFlag; /**< 1: no dictID will be saved into frame header (dictID is only useful for dictionary compression) */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1089 | } ZSTD_frameParameters; |
| 1090 | |
| 1091 | typedef struct { |
| 1092 | ZSTD_compressionParameters cParams; |
| 1093 | ZSTD_frameParameters fParams; |
| 1094 | } ZSTD_parameters; |
| 1095 | |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1096 | typedef enum { |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1097 | ZSTD_dct_auto = 0, /* dictionary is "full" when starting with ZSTD_MAGIC_DICTIONARY, otherwise it is "rawContent" */ |
| 1098 | ZSTD_dct_rawContent = 1, /* ensures dictionary is always loaded as rawContent, even if it starts with ZSTD_MAGIC_DICTIONARY */ |
| 1099 | ZSTD_dct_fullDict = 2 /* refuses to load a dictionary if it does not respect Zstandard's specification, starting with ZSTD_MAGIC_DICTIONARY */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1100 | } ZSTD_dictContentType_e; |
| 1101 | |
| 1102 | typedef enum { |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1103 | ZSTD_dlm_byCopy = 0, /**< Copy dictionary content internally */ |
| 1104 | ZSTD_dlm_byRef = 1, /**< Reference dictionary content -- the dictionary buffer must outlive its users. */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1105 | } ZSTD_dictLoadMethod_e; |
| 1106 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1107 | typedef enum { |
| 1108 | /* Opened question : should we have a format ZSTD_f_auto ? |
| 1109 | * Today, it would mean exactly the same as ZSTD_f_zstd1. |
| 1110 | * But, in the future, should several formats become supported, |
| 1111 | * on the compression side, it would mean "default format". |
| 1112 | * On the decompression side, it would mean "automatic format detection", |
| 1113 | * so that ZSTD_f_zstd1 would mean "accept *only* zstd frames". |
| 1114 | * Since meaning is a little different, another option could be to define different enums for compression and decompression. |
| 1115 | * This question could be kept for later, when there are actually multiple formats to support, |
| 1116 | * but there is also the question of pinning enum values, and pinning value `0` is especially important */ |
| 1117 | ZSTD_f_zstd1 = 0, /* zstd frame format, specified in zstd_compression_format.md (default) */ |
| 1118 | ZSTD_f_zstd1_magicless = 1, /* Variant of zstd frame format, without initial 4-bytes magic number. |
| 1119 | * Useful to save 4 bytes per generated frame. |
| 1120 | * Decoder cannot recognise automatically this format, requiring this instruction. */ |
| 1121 | } ZSTD_format_e; |
| 1122 | |
| 1123 | typedef enum { |
| 1124 | /* Note: this enum and the behavior it controls are effectively internal |
| 1125 | * implementation details of the compressor. They are expected to continue |
| 1126 | * to evolve and should be considered only in the context of extremely |
| 1127 | * advanced performance tuning. |
| 1128 | * |
| 1129 | * Zstd currently supports the use of a CDict in two ways: |
| 1130 | * |
| 1131 | * - The contents of the CDict can be copied into the working context. This |
| 1132 | * means that the compression can search both the dictionary and input |
| 1133 | * while operating on a single set of internal tables. This makes |
| 1134 | * the compression faster per-byte of input. However, the initial copy of |
| 1135 | * the CDict's tables incurs a fixed cost at the beginning of the |
| 1136 | * compression. For small compressions (< 8 KB), that copy can dominate |
| 1137 | * the cost of the compression. |
| 1138 | * |
| 1139 | * - The CDict's tables can be used in-place. In this model, compression is |
| 1140 | * slower per input byte, because the compressor has to search two sets of |
| 1141 | * tables. However, this model incurs no start-up cost (as long as the |
| 1142 | * working context's tables can be reused). For small inputs, this can be |
| 1143 | * faster than copying the CDict's tables. |
| 1144 | * |
| 1145 | * Zstd has a simple internal heuristic that selects which strategy to use |
| 1146 | * at the beginning of a compression. However, if experimentation shows that |
| 1147 | * Zstd is making poor choices, it is possible to override that choice with |
| 1148 | * this enum. |
| 1149 | */ |
| 1150 | ZSTD_dictDefaultAttach = 0, /* Use the default heuristic. */ |
| 1151 | ZSTD_dictForceAttach = 1, /* Never copy the dictionary. */ |
| 1152 | ZSTD_dictForceCopy = 2, /* Always copy the dictionary. */ |
| 1153 | } ZSTD_dictAttachPref_e; |
| 1154 | |
| 1155 | typedef enum { |
| 1156 | ZSTD_lcm_auto = 0, /**< Automatically determine the compression mode based on the compression level. |
| 1157 | * Negative compression levels will be uncompressed, and positive compression |
| 1158 | * levels will be compressed. */ |
| 1159 | ZSTD_lcm_huffman = 1, /**< Always attempt Huffman compression. Uncompressed literals will still be |
| 1160 | * emitted if Huffman compression is not profitable. */ |
| 1161 | ZSTD_lcm_uncompressed = 2, /**< Always emit uncompressed literals. */ |
| 1162 | } ZSTD_literalCompressionMode_e; |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1163 | |
| 1164 | |
| 1165 | /*************************************** |
| 1166 | * Frame size functions |
| 1167 | ***************************************/ |
| 1168 | |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1169 | /*! ZSTD_findDecompressedSize() : |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1170 | * `src` should point to the start of a series of ZSTD encoded and/or skippable frames |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1171 | * `srcSize` must be the _exact_ size of this series |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1172 | * (i.e. there should be a frame boundary at `src + srcSize`) |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1173 | * @return : - decompressed size of all data in all successive frames |
| 1174 | * - if the decompressed size cannot be determined: ZSTD_CONTENTSIZE_UNKNOWN |
| 1175 | * - if an error occurred: ZSTD_CONTENTSIZE_ERROR |
| 1176 | * |
| 1177 | * note 1 : decompressed size is an optional field, that may not be present, especially in streaming mode. |
| 1178 | * When `return==ZSTD_CONTENTSIZE_UNKNOWN`, data to decompress could be any size. |
| 1179 | * In which case, it's necessary to use streaming mode to decompress data. |
| 1180 | * note 2 : decompressed size is always present when compression is done with ZSTD_compress() |
| 1181 | * note 3 : decompressed size can be very large (64-bits value), |
| 1182 | * potentially larger than what local system can handle as a single memory segment. |
| 1183 | * In which case, it's necessary to use streaming mode to decompress data. |
| 1184 | * note 4 : If source is untrusted, decompressed size could be wrong or intentionally modified. |
| 1185 | * Always ensure result fits within application's authorized limits. |
| 1186 | * Each application can set its own limits. |
| 1187 | * note 5 : ZSTD_findDecompressedSize handles multiple frames, and so it must traverse the input to |
| 1188 | * read each contained frame header. This is fast as most of the data is skipped, |
| 1189 | * however it does mean that all frame data must be present and valid. */ |
| 1190 | ZSTDLIB_API unsigned long long ZSTD_findDecompressedSize(const void* src, size_t srcSize); |
| 1191 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1192 | /*! ZSTD_decompressBound() : |
| 1193 | * `src` should point to the start of a series of ZSTD encoded and/or skippable frames |
| 1194 | * `srcSize` must be the _exact_ size of this series |
| 1195 | * (i.e. there should be a frame boundary at `src + srcSize`) |
| 1196 | * @return : - upper-bound for the decompressed size of all data in all successive frames |
| 1197 | * - if an error occured: ZSTD_CONTENTSIZE_ERROR |
| 1198 | * |
| 1199 | * note 1 : an error can occur if `src` contains an invalid or incorrectly formatted frame. |
| 1200 | * note 2 : the upper-bound is exact when the decompressed size field is available in every ZSTD encoded frame of `src`. |
| 1201 | * in this case, `ZSTD_findDecompressedSize` and `ZSTD_decompressBound` return the same value. |
| 1202 | * note 3 : when the decompressed size field isn't available, the upper-bound for that frame is calculated by: |
| 1203 | * upper-bound = # blocks * min(128 KB, Window_Size) |
| 1204 | */ |
| 1205 | ZSTDLIB_API unsigned long long ZSTD_decompressBound(const void* src, size_t srcSize); |
| 1206 | |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1207 | /*! ZSTD_frameHeaderSize() : |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1208 | * srcSize must be >= ZSTD_FRAMEHEADERSIZE_PREFIX. |
| 1209 | * @return : size of the Frame Header, |
| 1210 | * or an error code (if srcSize is too small) */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1211 | ZSTDLIB_API size_t ZSTD_frameHeaderSize(const void* src, size_t srcSize); |
| 1212 | |
| 1213 | |
| 1214 | /*************************************** |
| 1215 | * Memory management |
| 1216 | ***************************************/ |
| 1217 | |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1218 | /*! ZSTD_estimate*() : |
| 1219 | * These functions make it possible to estimate memory usage |
| 1220 | * of a future {D,C}Ctx, before its creation. |
| 1221 | * ZSTD_estimateCCtxSize() will provide a budget large enough for any compression level up to selected one. |
| 1222 | * It will also consider src size to be arbitrarily "large", which is worst case. |
| 1223 | * If srcSize is known to always be small, ZSTD_estimateCCtxSize_usingCParams() can provide a tighter estimation. |
| 1224 | * ZSTD_estimateCCtxSize_usingCParams() can be used in tandem with ZSTD_getCParams() to create cParams from compressionLevel. |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1225 | * ZSTD_estimateCCtxSize_usingCCtxParams() can be used in tandem with ZSTD_CCtxParams_setParameter(). Only single-threaded compression is supported. This function will return an error code if ZSTD_c_nbWorkers is >= 1. |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1226 | * Note : CCtx size estimation is only correct for single-threaded compression. */ |
| 1227 | ZSTDLIB_API size_t ZSTD_estimateCCtxSize(int compressionLevel); |
| 1228 | ZSTDLIB_API size_t ZSTD_estimateCCtxSize_usingCParams(ZSTD_compressionParameters cParams); |
| 1229 | ZSTDLIB_API size_t ZSTD_estimateCCtxSize_usingCCtxParams(const ZSTD_CCtx_params* params); |
| 1230 | ZSTDLIB_API size_t ZSTD_estimateDCtxSize(void); |
| 1231 | |
| 1232 | /*! ZSTD_estimateCStreamSize() : |
| 1233 | * ZSTD_estimateCStreamSize() will provide a budget large enough for any compression level up to selected one. |
| 1234 | * It will also consider src size to be arbitrarily "large", which is worst case. |
| 1235 | * If srcSize is known to always be small, ZSTD_estimateCStreamSize_usingCParams() can provide a tighter estimation. |
| 1236 | * ZSTD_estimateCStreamSize_usingCParams() can be used in tandem with ZSTD_getCParams() to create cParams from compressionLevel. |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1237 | * ZSTD_estimateCStreamSize_usingCCtxParams() can be used in tandem with ZSTD_CCtxParams_setParameter(). Only single-threaded compression is supported. This function will return an error code if ZSTD_c_nbWorkers is >= 1. |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1238 | * Note : CStream size estimation is only correct for single-threaded compression. |
| 1239 | * ZSTD_DStream memory budget depends on window Size. |
| 1240 | * This information can be passed manually, using ZSTD_estimateDStreamSize, |
| 1241 | * or deducted from a valid frame Header, using ZSTD_estimateDStreamSize_fromFrame(); |
| 1242 | * Note : if streaming is init with function ZSTD_init?Stream_usingDict(), |
| 1243 | * an internal ?Dict will be created, which additional size is not estimated here. |
| 1244 | * In this case, get total size by adding ZSTD_estimate?DictSize */ |
| 1245 | ZSTDLIB_API size_t ZSTD_estimateCStreamSize(int compressionLevel); |
| 1246 | ZSTDLIB_API size_t ZSTD_estimateCStreamSize_usingCParams(ZSTD_compressionParameters cParams); |
| 1247 | ZSTDLIB_API size_t ZSTD_estimateCStreamSize_usingCCtxParams(const ZSTD_CCtx_params* params); |
| 1248 | ZSTDLIB_API size_t ZSTD_estimateDStreamSize(size_t windowSize); |
| 1249 | ZSTDLIB_API size_t ZSTD_estimateDStreamSize_fromFrame(const void* src, size_t srcSize); |
| 1250 | |
| 1251 | /*! ZSTD_estimate?DictSize() : |
| 1252 | * ZSTD_estimateCDictSize() will bet that src size is relatively "small", and content is copied, like ZSTD_createCDict(). |
| 1253 | * ZSTD_estimateCDictSize_advanced() makes it possible to control compression parameters precisely, like ZSTD_createCDict_advanced(). |
| 1254 | * Note : dictionaries created by reference (`ZSTD_dlm_byRef`) are logically smaller. |
| 1255 | */ |
| 1256 | ZSTDLIB_API size_t ZSTD_estimateCDictSize(size_t dictSize, int compressionLevel); |
| 1257 | ZSTDLIB_API size_t ZSTD_estimateCDictSize_advanced(size_t dictSize, ZSTD_compressionParameters cParams, ZSTD_dictLoadMethod_e dictLoadMethod); |
| 1258 | ZSTDLIB_API size_t ZSTD_estimateDDictSize(size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod); |
| 1259 | |
| 1260 | /*! ZSTD_initStatic*() : |
| 1261 | * Initialize an object using a pre-allocated fixed-size buffer. |
| 1262 | * workspace: The memory area to emplace the object into. |
| 1263 | * Provided pointer *must be 8-bytes aligned*. |
| 1264 | * Buffer must outlive object. |
| 1265 | * workspaceSize: Use ZSTD_estimate*Size() to determine |
| 1266 | * how large workspace must be to support target scenario. |
| 1267 | * @return : pointer to object (same address as workspace, just different type), |
| 1268 | * or NULL if error (size too small, incorrect alignment, etc.) |
| 1269 | * Note : zstd will never resize nor malloc() when using a static buffer. |
| 1270 | * If the object requires more memory than available, |
| 1271 | * zstd will just error out (typically ZSTD_error_memory_allocation). |
| 1272 | * Note 2 : there is no corresponding "free" function. |
| 1273 | * Since workspace is allocated externally, it must be freed externally too. |
| 1274 | * Note 3 : cParams : use ZSTD_getCParams() to convert a compression level |
| 1275 | * into its associated cParams. |
| 1276 | * Limitation 1 : currently not compatible with internal dictionary creation, triggered by |
| 1277 | * ZSTD_CCtx_loadDictionary(), ZSTD_initCStream_usingDict() or ZSTD_initDStream_usingDict(). |
| 1278 | * Limitation 2 : static cctx currently not compatible with multi-threading. |
| 1279 | * Limitation 3 : static dctx is incompatible with legacy support. |
| 1280 | */ |
| 1281 | ZSTDLIB_API ZSTD_CCtx* ZSTD_initStaticCCtx(void* workspace, size_t workspaceSize); |
| 1282 | ZSTDLIB_API ZSTD_CStream* ZSTD_initStaticCStream(void* workspace, size_t workspaceSize); /**< same as ZSTD_initStaticCCtx() */ |
| 1283 | |
| 1284 | ZSTDLIB_API ZSTD_DCtx* ZSTD_initStaticDCtx(void* workspace, size_t workspaceSize); |
| 1285 | ZSTDLIB_API ZSTD_DStream* ZSTD_initStaticDStream(void* workspace, size_t workspaceSize); /**< same as ZSTD_initStaticDCtx() */ |
| 1286 | |
| 1287 | ZSTDLIB_API const ZSTD_CDict* ZSTD_initStaticCDict( |
| 1288 | void* workspace, size_t workspaceSize, |
| 1289 | const void* dict, size_t dictSize, |
| 1290 | ZSTD_dictLoadMethod_e dictLoadMethod, |
| 1291 | ZSTD_dictContentType_e dictContentType, |
| 1292 | ZSTD_compressionParameters cParams); |
| 1293 | |
| 1294 | ZSTDLIB_API const ZSTD_DDict* ZSTD_initStaticDDict( |
| 1295 | void* workspace, size_t workspaceSize, |
| 1296 | const void* dict, size_t dictSize, |
| 1297 | ZSTD_dictLoadMethod_e dictLoadMethod, |
| 1298 | ZSTD_dictContentType_e dictContentType); |
| 1299 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1300 | |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1301 | /*! Custom memory allocation : |
| 1302 | * These prototypes make it possible to pass your own allocation/free functions. |
| 1303 | * ZSTD_customMem is provided at creation time, using ZSTD_create*_advanced() variants listed below. |
| 1304 | * All allocation/free operations will be completed using these custom variants instead of regular <stdlib.h> ones. |
| 1305 | */ |
| 1306 | typedef void* (*ZSTD_allocFunction) (void* opaque, size_t size); |
| 1307 | typedef void (*ZSTD_freeFunction) (void* opaque, void* address); |
| 1308 | typedef struct { ZSTD_allocFunction customAlloc; ZSTD_freeFunction customFree; void* opaque; } ZSTD_customMem; |
| 1309 | static ZSTD_customMem const ZSTD_defaultCMem = { NULL, NULL, NULL }; /**< this constant defers to stdlib's functions */ |
| 1310 | |
| 1311 | ZSTDLIB_API ZSTD_CCtx* ZSTD_createCCtx_advanced(ZSTD_customMem customMem); |
| 1312 | ZSTDLIB_API ZSTD_CStream* ZSTD_createCStream_advanced(ZSTD_customMem customMem); |
| 1313 | ZSTDLIB_API ZSTD_DCtx* ZSTD_createDCtx_advanced(ZSTD_customMem customMem); |
| 1314 | ZSTDLIB_API ZSTD_DStream* ZSTD_createDStream_advanced(ZSTD_customMem customMem); |
| 1315 | |
| 1316 | ZSTDLIB_API ZSTD_CDict* ZSTD_createCDict_advanced(const void* dict, size_t dictSize, |
| 1317 | ZSTD_dictLoadMethod_e dictLoadMethod, |
| 1318 | ZSTD_dictContentType_e dictContentType, |
| 1319 | ZSTD_compressionParameters cParams, |
| 1320 | ZSTD_customMem customMem); |
| 1321 | |
| 1322 | ZSTDLIB_API ZSTD_DDict* ZSTD_createDDict_advanced(const void* dict, size_t dictSize, |
| 1323 | ZSTD_dictLoadMethod_e dictLoadMethod, |
| 1324 | ZSTD_dictContentType_e dictContentType, |
| 1325 | ZSTD_customMem customMem); |
| 1326 | |
| 1327 | |
| 1328 | |
| 1329 | /*************************************** |
| 1330 | * Advanced compression functions |
| 1331 | ***************************************/ |
| 1332 | |
| 1333 | /*! ZSTD_createCDict_byReference() : |
| 1334 | * Create a digested dictionary for compression |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1335 | * Dictionary content is just referenced, not duplicated. |
| 1336 | * As a consequence, `dictBuffer` **must** outlive CDict, |
| 1337 | * and its content must remain unmodified throughout the lifetime of CDict. */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1338 | ZSTDLIB_API ZSTD_CDict* ZSTD_createCDict_byReference(const void* dictBuffer, size_t dictSize, int compressionLevel); |
| 1339 | |
| 1340 | /*! ZSTD_getCParams() : |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1341 | * @return ZSTD_compressionParameters structure for a selected compression level and estimated srcSize. |
| 1342 | * `estimatedSrcSize` value is optional, select 0 if not known */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1343 | ZSTDLIB_API ZSTD_compressionParameters ZSTD_getCParams(int compressionLevel, unsigned long long estimatedSrcSize, size_t dictSize); |
| 1344 | |
| 1345 | /*! ZSTD_getParams() : |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1346 | * same as ZSTD_getCParams(), but @return a full `ZSTD_parameters` object instead of sub-component `ZSTD_compressionParameters`. |
| 1347 | * All fields of `ZSTD_frameParameters` are set to default : contentSize=1, checksum=0, noDictID=0 */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1348 | ZSTDLIB_API ZSTD_parameters ZSTD_getParams(int compressionLevel, unsigned long long estimatedSrcSize, size_t dictSize); |
| 1349 | |
| 1350 | /*! ZSTD_checkCParams() : |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1351 | * Ensure param values remain within authorized range. |
| 1352 | * @return 0 on success, or an error code (can be checked with ZSTD_isError()) */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1353 | ZSTDLIB_API size_t ZSTD_checkCParams(ZSTD_compressionParameters params); |
| 1354 | |
| 1355 | /*! ZSTD_adjustCParams() : |
| 1356 | * optimize params for a given `srcSize` and `dictSize`. |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1357 | * `srcSize` can be unknown, in which case use ZSTD_CONTENTSIZE_UNKNOWN. |
| 1358 | * `dictSize` must be `0` when there is no dictionary. |
| 1359 | * cPar can be invalid : all parameters will be clamped within valid range in the @return struct. |
| 1360 | * This function never fails (wide contract) */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1361 | ZSTDLIB_API ZSTD_compressionParameters ZSTD_adjustCParams(ZSTD_compressionParameters cPar, unsigned long long srcSize, size_t dictSize); |
| 1362 | |
| 1363 | /*! ZSTD_compress_advanced() : |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1364 | * Same as ZSTD_compress_usingDict(), with fine-tune control over compression parameters (by structure) */ |
| 1365 | ZSTDLIB_API size_t ZSTD_compress_advanced(ZSTD_CCtx* cctx, |
| 1366 | void* dst, size_t dstCapacity, |
| 1367 | const void* src, size_t srcSize, |
| 1368 | const void* dict,size_t dictSize, |
| 1369 | ZSTD_parameters params); |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1370 | |
| 1371 | /*! ZSTD_compress_usingCDict_advanced() : |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1372 | * Same as ZSTD_compress_usingCDict(), with fine-tune control over frame parameters */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1373 | ZSTDLIB_API size_t ZSTD_compress_usingCDict_advanced(ZSTD_CCtx* cctx, |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1374 | void* dst, size_t dstCapacity, |
| 1375 | const void* src, size_t srcSize, |
| 1376 | const ZSTD_CDict* cdict, |
| 1377 | ZSTD_frameParameters fParams); |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1378 | |
| 1379 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1380 | /*! ZSTD_CCtx_loadDictionary_byReference() : |
| 1381 | * Same as ZSTD_CCtx_loadDictionary(), but dictionary content is referenced, instead of being copied into CCtx. |
| 1382 | * It saves some memory, but also requires that `dict` outlives its usage within `cctx` */ |
| 1383 | ZSTDLIB_API size_t ZSTD_CCtx_loadDictionary_byReference(ZSTD_CCtx* cctx, const void* dict, size_t dictSize); |
| 1384 | |
| 1385 | /*! ZSTD_CCtx_loadDictionary_advanced() : |
| 1386 | * Same as ZSTD_CCtx_loadDictionary(), but gives finer control over |
| 1387 | * how to load the dictionary (by copy ? by reference ?) |
| 1388 | * and how to interpret it (automatic ? force raw mode ? full mode only ?) */ |
| 1389 | 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); |
| 1390 | |
| 1391 | /*! ZSTD_CCtx_refPrefix_advanced() : |
| 1392 | * Same as ZSTD_CCtx_refPrefix(), but gives finer control over |
| 1393 | * how to interpret prefix content (automatic ? force raw mode (default) ? full mode only ?) */ |
| 1394 | ZSTDLIB_API size_t ZSTD_CCtx_refPrefix_advanced(ZSTD_CCtx* cctx, const void* prefix, size_t prefixSize, ZSTD_dictContentType_e dictContentType); |
| 1395 | |
| 1396 | /* === experimental parameters === */ |
| 1397 | /* these parameters can be used with ZSTD_setParameter() |
| 1398 | * they are not guaranteed to remain supported in the future */ |
| 1399 | |
| 1400 | /* Enables rsyncable mode, |
| 1401 | * which makes compressed files more rsync friendly |
| 1402 | * by adding periodic synchronization points to the compressed data. |
| 1403 | * The target average block size is ZSTD_c_jobSize / 2. |
| 1404 | * It's possible to modify the job size to increase or decrease |
| 1405 | * the granularity of the synchronization point. |
| 1406 | * Once the jobSize is smaller than the window size, |
| 1407 | * it will result in compression ratio degradation. |
| 1408 | * NOTE 1: rsyncable mode only works when multithreading is enabled. |
| 1409 | * NOTE 2: rsyncable performs poorly in combination with long range mode, |
| 1410 | * since it will decrease the effectiveness of synchronization points, |
| 1411 | * though mileage may vary. |
| 1412 | * NOTE 3: Rsyncable mode limits maximum compression speed to ~400 MB/s. |
| 1413 | * If the selected compression level is already running significantly slower, |
| 1414 | * the overall speed won't be significantly impacted. |
| 1415 | */ |
| 1416 | #define ZSTD_c_rsyncable ZSTD_c_experimentalParam1 |
| 1417 | |
| 1418 | /* Select a compression format. |
| 1419 | * The value must be of type ZSTD_format_e. |
| 1420 | * See ZSTD_format_e enum definition for details */ |
| 1421 | #define ZSTD_c_format ZSTD_c_experimentalParam2 |
| 1422 | |
| 1423 | /* Force back-reference distances to remain < windowSize, |
| 1424 | * even when referencing into Dictionary content (default:0) */ |
| 1425 | #define ZSTD_c_forceMaxWindow ZSTD_c_experimentalParam3 |
| 1426 | |
| 1427 | /* Controls whether the contents of a CDict |
| 1428 | * are used in place, or copied into the working context. |
| 1429 | * Accepts values from the ZSTD_dictAttachPref_e enum. |
| 1430 | * See the comments on that enum for an explanation of the feature. */ |
| 1431 | #define ZSTD_c_forceAttachDict ZSTD_c_experimentalParam4 |
| 1432 | |
| 1433 | /* Controls how the literals are compressed (default is auto). |
| 1434 | * The value must be of type ZSTD_literalCompressionMode_e. |
| 1435 | * See ZSTD_literalCompressionMode_t enum definition for details. |
| 1436 | */ |
| 1437 | #define ZSTD_c_literalCompressionMode ZSTD_c_experimentalParam5 |
| 1438 | |
| 1439 | /* Tries to fit compressed block size to be around targetCBlockSize. |
| 1440 | * No target when targetCBlockSize == 0. |
| 1441 | * There is no guarantee on compressed block size (default:0) */ |
| 1442 | #define ZSTD_c_targetCBlockSize ZSTD_c_experimentalParam6 |
| 1443 | |
| 1444 | /*! ZSTD_CCtx_getParameter() : |
| 1445 | * Get the requested compression parameter value, selected by enum ZSTD_cParameter, |
| 1446 | * and store it into int* value. |
| 1447 | * @return : 0, or an error code (which can be tested with ZSTD_isError()). |
| 1448 | */ |
| 1449 | ZSTDLIB_API size_t ZSTD_CCtx_getParameter(ZSTD_CCtx* cctx, ZSTD_cParameter param, int* value); |
| 1450 | |
| 1451 | |
| 1452 | /*! ZSTD_CCtx_params : |
| 1453 | * Quick howto : |
| 1454 | * - ZSTD_createCCtxParams() : Create a ZSTD_CCtx_params structure |
| 1455 | * - ZSTD_CCtxParams_setParameter() : Push parameters one by one into |
| 1456 | * an existing ZSTD_CCtx_params structure. |
| 1457 | * This is similar to |
| 1458 | * ZSTD_CCtx_setParameter(). |
| 1459 | * - ZSTD_CCtx_setParametersUsingCCtxParams() : Apply parameters to |
| 1460 | * an existing CCtx. |
| 1461 | * These parameters will be applied to |
| 1462 | * all subsequent frames. |
| 1463 | * - ZSTD_compressStream2() : Do compression using the CCtx. |
| 1464 | * - ZSTD_freeCCtxParams() : Free the memory. |
| 1465 | * |
| 1466 | * This can be used with ZSTD_estimateCCtxSize_advanced_usingCCtxParams() |
| 1467 | * for static allocation of CCtx for single-threaded compression. |
| 1468 | */ |
| 1469 | ZSTDLIB_API ZSTD_CCtx_params* ZSTD_createCCtxParams(void); |
| 1470 | ZSTDLIB_API size_t ZSTD_freeCCtxParams(ZSTD_CCtx_params* params); |
| 1471 | |
| 1472 | /*! ZSTD_CCtxParams_reset() : |
| 1473 | * Reset params to default values. |
| 1474 | */ |
| 1475 | ZSTDLIB_API size_t ZSTD_CCtxParams_reset(ZSTD_CCtx_params* params); |
| 1476 | |
| 1477 | /*! ZSTD_CCtxParams_init() : |
| 1478 | * Initializes the compression parameters of cctxParams according to |
| 1479 | * compression level. All other parameters are reset to their default values. |
| 1480 | */ |
| 1481 | ZSTDLIB_API size_t ZSTD_CCtxParams_init(ZSTD_CCtx_params* cctxParams, int compressionLevel); |
| 1482 | |
| 1483 | /*! ZSTD_CCtxParams_init_advanced() : |
| 1484 | * Initializes the compression and frame parameters of cctxParams according to |
| 1485 | * params. All other parameters are reset to their default values. |
| 1486 | */ |
| 1487 | ZSTDLIB_API size_t ZSTD_CCtxParams_init_advanced(ZSTD_CCtx_params* cctxParams, ZSTD_parameters params); |
| 1488 | |
| 1489 | /*! ZSTD_CCtxParams_setParameter() : |
| 1490 | * Similar to ZSTD_CCtx_setParameter. |
| 1491 | * Set one compression parameter, selected by enum ZSTD_cParameter. |
| 1492 | * Parameters must be applied to a ZSTD_CCtx using ZSTD_CCtx_setParametersUsingCCtxParams(). |
| 1493 | * @result : 0, or an error code (which can be tested with ZSTD_isError()). |
| 1494 | */ |
| 1495 | ZSTDLIB_API size_t ZSTD_CCtxParams_setParameter(ZSTD_CCtx_params* params, ZSTD_cParameter param, int value); |
| 1496 | |
| 1497 | /*! ZSTD_CCtxParams_getParameter() : |
| 1498 | * Similar to ZSTD_CCtx_getParameter. |
| 1499 | * Get the requested value of one compression parameter, selected by enum ZSTD_cParameter. |
| 1500 | * @result : 0, or an error code (which can be tested with ZSTD_isError()). |
| 1501 | */ |
| 1502 | ZSTDLIB_API size_t ZSTD_CCtxParams_getParameter(ZSTD_CCtx_params* params, ZSTD_cParameter param, int* value); |
| 1503 | |
| 1504 | /*! ZSTD_CCtx_setParametersUsingCCtxParams() : |
| 1505 | * Apply a set of ZSTD_CCtx_params to the compression context. |
| 1506 | * This can be done even after compression is started, |
| 1507 | * if nbWorkers==0, this will have no impact until a new compression is started. |
| 1508 | * if nbWorkers>=1, new parameters will be picked up at next job, |
| 1509 | * with a few restrictions (windowLog, pledgedSrcSize, nbWorkers, jobSize, and overlapLog are not updated). |
| 1510 | */ |
| 1511 | ZSTDLIB_API size_t ZSTD_CCtx_setParametersUsingCCtxParams( |
| 1512 | ZSTD_CCtx* cctx, const ZSTD_CCtx_params* params); |
| 1513 | |
| 1514 | /*! ZSTD_compressStream2_simpleArgs() : |
| 1515 | * Same as ZSTD_compressStream2(), |
| 1516 | * but using only integral types as arguments. |
| 1517 | * This variant might be helpful for binders from dynamic languages |
| 1518 | * which have troubles handling structures containing memory pointers. |
| 1519 | */ |
| 1520 | ZSTDLIB_API size_t ZSTD_compressStream2_simpleArgs ( |
| 1521 | ZSTD_CCtx* cctx, |
| 1522 | void* dst, size_t dstCapacity, size_t* dstPos, |
| 1523 | const void* src, size_t srcSize, size_t* srcPos, |
| 1524 | ZSTD_EndDirective endOp); |
| 1525 | |
| 1526 | |
| 1527 | /*************************************** |
| 1528 | * Advanced decompression functions |
| 1529 | ***************************************/ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1530 | |
| 1531 | /*! ZSTD_isFrame() : |
| 1532 | * Tells if the content of `buffer` starts with a valid Frame Identifier. |
| 1533 | * Note : Frame Identifier is 4 bytes. If `size < 4`, @return will always be 0. |
| 1534 | * Note 2 : Legacy Frame Identifiers are considered valid only if Legacy Support is enabled. |
| 1535 | * Note 3 : Skippable Frame Identifiers are considered valid. */ |
| 1536 | ZSTDLIB_API unsigned ZSTD_isFrame(const void* buffer, size_t size); |
| 1537 | |
| 1538 | /*! ZSTD_createDDict_byReference() : |
| 1539 | * Create a digested dictionary, ready to start decompression operation without startup delay. |
| 1540 | * Dictionary content is referenced, and therefore stays in dictBuffer. |
| 1541 | * It is important that dictBuffer outlives DDict, |
| 1542 | * it must remain read accessible throughout the lifetime of DDict */ |
| 1543 | ZSTDLIB_API ZSTD_DDict* ZSTD_createDDict_byReference(const void* dictBuffer, size_t dictSize); |
| 1544 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1545 | /*! ZSTD_DCtx_loadDictionary_byReference() : |
| 1546 | * Same as ZSTD_DCtx_loadDictionary(), |
| 1547 | * but references `dict` content instead of copying it into `dctx`. |
| 1548 | * This saves memory if `dict` remains around., |
| 1549 | * However, it's imperative that `dict` remains accessible (and unmodified) while being used, so it must outlive decompression. */ |
| 1550 | ZSTDLIB_API size_t ZSTD_DCtx_loadDictionary_byReference(ZSTD_DCtx* dctx, const void* dict, size_t dictSize); |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1551 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1552 | /*! ZSTD_DCtx_loadDictionary_advanced() : |
| 1553 | * Same as ZSTD_DCtx_loadDictionary(), |
| 1554 | * but gives direct control over |
| 1555 | * how to load the dictionary (by copy ? by reference ?) |
| 1556 | * and how to interpret it (automatic ? force raw mode ? full mode only ?). */ |
| 1557 | 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); |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1558 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1559 | /*! ZSTD_DCtx_refPrefix_advanced() : |
| 1560 | * Same as ZSTD_DCtx_refPrefix(), but gives finer control over |
| 1561 | * how to interpret prefix content (automatic ? force raw mode (default) ? full mode only ?) */ |
| 1562 | ZSTDLIB_API size_t ZSTD_DCtx_refPrefix_advanced(ZSTD_DCtx* dctx, const void* prefix, size_t prefixSize, ZSTD_dictContentType_e dictContentType); |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1563 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1564 | /*! ZSTD_DCtx_setMaxWindowSize() : |
| 1565 | * Refuses allocating internal buffers for frames requiring a window size larger than provided limit. |
| 1566 | * This protects a decoder context from reserving too much memory for itself (potential attack scenario). |
| 1567 | * This parameter is only useful in streaming mode, since no internal buffer is allocated in single-pass mode. |
| 1568 | * By default, a decompression context accepts all window sizes <= (1 << ZSTD_WINDOWLOG_LIMIT_DEFAULT) |
| 1569 | * @return : 0, or an error code (which can be tested using ZSTD_isError()). |
| 1570 | */ |
| 1571 | ZSTDLIB_API size_t ZSTD_DCtx_setMaxWindowSize(ZSTD_DCtx* dctx, size_t maxWindowSize); |
| 1572 | |
| 1573 | /* ZSTD_d_format |
| 1574 | * experimental parameter, |
| 1575 | * allowing selection between ZSTD_format_e input compression formats |
| 1576 | */ |
| 1577 | #define ZSTD_d_format ZSTD_d_experimentalParam1 |
| 1578 | |
| 1579 | /*! ZSTD_DCtx_setFormat() : |
| 1580 | * Instruct the decoder context about what kind of data to decode next. |
| 1581 | * This instruction is mandatory to decode data without a fully-formed header, |
| 1582 | * such ZSTD_f_zstd1_magicless for example. |
| 1583 | * @return : 0, or an error code (which can be tested using ZSTD_isError()). */ |
| 1584 | ZSTDLIB_API size_t ZSTD_DCtx_setFormat(ZSTD_DCtx* dctx, ZSTD_format_e format); |
| 1585 | |
| 1586 | /*! ZSTD_decompressStream_simpleArgs() : |
| 1587 | * Same as ZSTD_decompressStream(), |
| 1588 | * but using only integral types as arguments. |
| 1589 | * This can be helpful for binders from dynamic languages |
| 1590 | * which have troubles handling structures containing memory pointers. |
| 1591 | */ |
| 1592 | ZSTDLIB_API size_t ZSTD_decompressStream_simpleArgs ( |
| 1593 | ZSTD_DCtx* dctx, |
| 1594 | void* dst, size_t dstCapacity, size_t* dstPos, |
| 1595 | const void* src, size_t srcSize, size_t* srcPos); |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1596 | |
| 1597 | |
| 1598 | /******************************************************************** |
| 1599 | * Advanced streaming functions |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1600 | * Warning : most of these functions are now redundant with the Advanced API. |
| 1601 | * Once Advanced API reaches "stable" status, |
| 1602 | * redundant functions will be deprecated, and then at some point removed. |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1603 | ********************************************************************/ |
| 1604 | |
| 1605 | /*===== Advanced Streaming compression functions =====*/ |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1606 | /**! ZSTD_initCStream_srcSize() : |
| 1607 | * This function is deprecated, and equivalent to: |
| 1608 | * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); |
| 1609 | * ZSTD_CCtx_refCDict(zcs, NULL); // clear the dictionary (if any) |
| 1610 | * ZSTD_CCtx_setParameter(zcs, ZSTD_c_compressionLevel, compressionLevel); |
| 1611 | * ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize); |
| 1612 | * |
| 1613 | * pledgedSrcSize must be correct. If it is not known at init time, use |
| 1614 | * ZSTD_CONTENTSIZE_UNKNOWN. Note that, for compatibility with older programs, |
| 1615 | * "0" also disables frame content size field. It may be enabled in the future. |
| 1616 | */ |
| 1617 | ZSTDLIB_API size_t ZSTD_initCStream_srcSize(ZSTD_CStream* zcs, int compressionLevel, unsigned long long pledgedSrcSize); |
| 1618 | /**! ZSTD_initCStream_usingDict() : |
| 1619 | * This function is deprecated, and is equivalent to: |
| 1620 | * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); |
| 1621 | * ZSTD_CCtx_setParameter(zcs, ZSTD_c_compressionLevel, compressionLevel); |
| 1622 | * ZSTD_CCtx_loadDictionary(zcs, dict, dictSize); |
| 1623 | * |
| 1624 | * Creates of an internal CDict (incompatible with static CCtx), except if |
| 1625 | * dict == NULL or dictSize < 8, in which case no dict is used. |
| 1626 | * Note: dict is loaded with ZSTD_dm_auto (treated as a full zstd dictionary if |
| 1627 | * it begins with ZSTD_MAGIC_DICTIONARY, else as raw content) and ZSTD_dlm_byCopy. |
| 1628 | */ |
| 1629 | ZSTDLIB_API size_t ZSTD_initCStream_usingDict(ZSTD_CStream* zcs, const void* dict, size_t dictSize, int compressionLevel); |
| 1630 | /**! ZSTD_initCStream_advanced() : |
| 1631 | * This function is deprecated, and is approximately equivalent to: |
| 1632 | * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); |
| 1633 | * ZSTD_CCtx_setZstdParams(zcs, params); // Set the zstd params and leave the rest as-is |
| 1634 | * ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize); |
| 1635 | * ZSTD_CCtx_loadDictionary(zcs, dict, dictSize); |
| 1636 | * |
| 1637 | * pledgedSrcSize must be correct. If srcSize is not known at init time, use |
| 1638 | * value ZSTD_CONTENTSIZE_UNKNOWN. dict is loaded with ZSTD_dm_auto and ZSTD_dlm_byCopy. |
| 1639 | */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1640 | ZSTDLIB_API size_t ZSTD_initCStream_advanced(ZSTD_CStream* zcs, const void* dict, size_t dictSize, |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1641 | ZSTD_parameters params, unsigned long long pledgedSrcSize); |
| 1642 | /**! ZSTD_initCStream_usingCDict() : |
| 1643 | * This function is deprecated, and equivalent to: |
| 1644 | * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); |
| 1645 | * ZSTD_CCtx_refCDict(zcs, cdict); |
| 1646 | * |
| 1647 | * note : cdict will just be referenced, and must outlive compression session |
| 1648 | */ |
| 1649 | ZSTDLIB_API size_t ZSTD_initCStream_usingCDict(ZSTD_CStream* zcs, const ZSTD_CDict* cdict); |
| 1650 | /**! ZSTD_initCStream_usingCDict_advanced() : |
| 1651 | * This function is deprecated, and is approximately equivalent to: |
| 1652 | * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); |
| 1653 | * ZSTD_CCtx_setZstdFrameParams(zcs, fParams); // Set the zstd frame params and leave the rest as-is |
| 1654 | * ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize); |
| 1655 | * ZSTD_CCtx_refCDict(zcs, cdict); |
| 1656 | * |
| 1657 | * same as ZSTD_initCStream_usingCDict(), with control over frame parameters. |
| 1658 | * pledgedSrcSize must be correct. If srcSize is not known at init time, use |
| 1659 | * value ZSTD_CONTENTSIZE_UNKNOWN. |
| 1660 | */ |
| 1661 | ZSTDLIB_API size_t ZSTD_initCStream_usingCDict_advanced(ZSTD_CStream* zcs, const ZSTD_CDict* cdict, ZSTD_frameParameters fParams, unsigned long long pledgedSrcSize); |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1662 | |
| 1663 | /*! ZSTD_resetCStream() : |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1664 | * This function is deprecated, and is equivalent to: |
| 1665 | * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); |
| 1666 | * ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize); |
| 1667 | * |
| 1668 | * start a new frame, using same parameters from previous frame. |
| 1669 | * This is typically useful to skip dictionary loading stage, since it will re-use it in-place. |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1670 | * Note that zcs must be init at least once before using ZSTD_resetCStream(). |
| 1671 | * If pledgedSrcSize is not known at reset time, use macro ZSTD_CONTENTSIZE_UNKNOWN. |
| 1672 | * If pledgedSrcSize > 0, its value must be correct, as it will be written in header, and controlled at the end. |
| 1673 | * For the time being, pledgedSrcSize==0 is interpreted as "srcSize unknown" for compatibility with older programs, |
| 1674 | * but it will change to mean "empty" in future version, so use macro ZSTD_CONTENTSIZE_UNKNOWN instead. |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1675 | * @return : 0, or an error code (which can be tested using ZSTD_isError()) |
| 1676 | */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1677 | ZSTDLIB_API size_t ZSTD_resetCStream(ZSTD_CStream* zcs, unsigned long long pledgedSrcSize); |
| 1678 | |
| 1679 | |
| 1680 | typedef struct { |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1681 | unsigned long long ingested; /* nb input bytes read and buffered */ |
| 1682 | unsigned long long consumed; /* nb input bytes actually compressed */ |
| 1683 | unsigned long long produced; /* nb of compressed bytes generated and buffered */ |
| 1684 | unsigned long long flushed; /* nb of compressed bytes flushed : not provided; can be tracked from caller side */ |
| 1685 | unsigned currentJobID; /* MT only : latest started job nb */ |
| 1686 | unsigned nbActiveWorkers; /* MT only : nb of workers actively compressing at probe time */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1687 | } ZSTD_frameProgression; |
| 1688 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1689 | /* ZSTD_getFrameProgression() : |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1690 | * tells how much data has been ingested (read from input) |
| 1691 | * consumed (input actually compressed) and produced (output) for current frame. |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1692 | * Note : (ingested - consumed) is amount of input data buffered internally, not yet compressed. |
| 1693 | * Aggregates progression inside active worker threads. |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1694 | */ |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1695 | ZSTDLIB_API ZSTD_frameProgression ZSTD_getFrameProgression(const ZSTD_CCtx* cctx); |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1696 | |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1697 | /*! ZSTD_toFlushNow() : |
| 1698 | * Tell how many bytes are ready to be flushed immediately. |
| 1699 | * Useful for multithreading scenarios (nbWorkers >= 1). |
| 1700 | * Probe the oldest active job, defined as oldest job not yet entirely flushed, |
| 1701 | * and check its output buffer. |
| 1702 | * @return : amount of data stored in oldest job and ready to be flushed immediately. |
| 1703 | * if @return == 0, it means either : |
| 1704 | * + there is no active job (could be checked with ZSTD_frameProgression()), or |
| 1705 | * + oldest job is still actively compressing data, |
| 1706 | * but everything it has produced has also been flushed so far, |
| 1707 | * therefore flush speed is limited by production speed of oldest job |
| 1708 | * irrespective of the speed of concurrent (and newer) jobs. |
| 1709 | */ |
| 1710 | ZSTDLIB_API size_t ZSTD_toFlushNow(ZSTD_CCtx* cctx); |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1711 | |
| 1712 | |
| 1713 | /*===== Advanced Streaming decompression functions =====*/ |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1714 | /** |
| 1715 | * This function is deprecated, and is equivalent to: |
| 1716 | * |
| 1717 | * ZSTD_DCtx_reset(zds, ZSTD_reset_session_only); |
| 1718 | * ZSTD_DCtx_loadDictionary(zds, dict, dictSize); |
| 1719 | * |
| 1720 | * note: no dictionary will be used if dict == NULL or dictSize < 8 |
| 1721 | */ |
| 1722 | ZSTDLIB_API size_t ZSTD_initDStream_usingDict(ZSTD_DStream* zds, const void* dict, size_t dictSize); |
| 1723 | /** |
| 1724 | * This function is deprecated, and is equivalent to: |
| 1725 | * |
| 1726 | * ZSTD_DCtx_reset(zds, ZSTD_reset_session_only); |
| 1727 | * ZSTD_DCtx_refDDict(zds, ddict); |
| 1728 | * |
| 1729 | * note : ddict is referenced, it must outlive decompression session |
| 1730 | */ |
| 1731 | ZSTDLIB_API size_t ZSTD_initDStream_usingDDict(ZSTD_DStream* zds, const ZSTD_DDict* ddict); |
| 1732 | /** |
| 1733 | * This function is deprecated, and is equivalent to: |
| 1734 | * |
| 1735 | * ZSTD_DCtx_reset(zds, ZSTD_reset_session_only); |
| 1736 | * |
| 1737 | * re-use decompression parameters from previous init; saves dictionary loading |
| 1738 | */ |
| 1739 | ZSTDLIB_API size_t ZSTD_resetDStream(ZSTD_DStream* zds); |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1740 | |
| 1741 | |
| 1742 | /********************************************************************* |
| 1743 | * Buffer-less and synchronous inner streaming functions |
| 1744 | * |
| 1745 | * This is an advanced API, giving full control over buffer management, for users which need direct control over memory. |
| 1746 | * But it's also a complex one, with several restrictions, documented below. |
| 1747 | * Prefer normal streaming API for an easier experience. |
| 1748 | ********************************************************************* */ |
| 1749 | |
| 1750 | /** |
| 1751 | Buffer-less streaming compression (synchronous mode) |
| 1752 | |
| 1753 | A ZSTD_CCtx object is required to track streaming operations. |
| 1754 | Use ZSTD_createCCtx() / ZSTD_freeCCtx() to manage resource. |
| 1755 | ZSTD_CCtx object can be re-used multiple times within successive compression operations. |
| 1756 | |
| 1757 | Start by initializing a context. |
| 1758 | Use ZSTD_compressBegin(), or ZSTD_compressBegin_usingDict() for dictionary compression, |
| 1759 | or ZSTD_compressBegin_advanced(), for finer parameter control. |
| 1760 | It's also possible to duplicate a reference context which has already been initialized, using ZSTD_copyCCtx() |
| 1761 | |
| 1762 | Then, consume your input using ZSTD_compressContinue(). |
| 1763 | There are some important considerations to keep in mind when using this advanced function : |
| 1764 | - ZSTD_compressContinue() has no internal buffer. It uses externally provided buffers only. |
| 1765 | - Interface is synchronous : input is consumed entirely and produces 1+ compressed blocks. |
| 1766 | - Caller must ensure there is enough space in `dst` to store compressed data under worst case scenario. |
| 1767 | Worst case evaluation is provided by ZSTD_compressBound(). |
| 1768 | ZSTD_compressContinue() doesn't guarantee recover after a failed compression. |
| 1769 | - ZSTD_compressContinue() presumes prior input ***is still accessible and unmodified*** (up to maximum distance size, see WindowLog). |
| 1770 | It remembers all previous contiguous blocks, plus one separated memory segment (which can itself consists of multiple contiguous blocks) |
| 1771 | - ZSTD_compressContinue() detects that prior input has been overwritten when `src` buffer overlaps. |
| 1772 | In which case, it will "discard" the relevant memory section from its history. |
| 1773 | |
| 1774 | Finish a frame with ZSTD_compressEnd(), which will write the last block(s) and optional checksum. |
| 1775 | It's possible to use srcSize==0, in which case, it will write a final empty block to end the frame. |
| 1776 | Without last block mark, frames are considered unfinished (hence corrupted) by compliant decoders. |
| 1777 | |
| 1778 | `ZSTD_CCtx` object can be re-used (ZSTD_compressBegin()) to compress again. |
| 1779 | */ |
| 1780 | |
| 1781 | /*===== Buffer-less streaming compression functions =====*/ |
| 1782 | ZSTDLIB_API size_t ZSTD_compressBegin(ZSTD_CCtx* cctx, int compressionLevel); |
| 1783 | ZSTDLIB_API size_t ZSTD_compressBegin_usingDict(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, int compressionLevel); |
| 1784 | 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 */ |
| 1785 | ZSTDLIB_API size_t ZSTD_compressBegin_usingCDict(ZSTD_CCtx* cctx, const ZSTD_CDict* cdict); /**< note: fails if cdict==NULL */ |
| 1786 | 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 */ |
| 1787 | 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 */ |
| 1788 | |
| 1789 | ZSTDLIB_API size_t ZSTD_compressContinue(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize); |
| 1790 | ZSTDLIB_API size_t ZSTD_compressEnd(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize); |
| 1791 | |
| 1792 | |
| 1793 | /*- |
| 1794 | Buffer-less streaming decompression (synchronous mode) |
| 1795 | |
| 1796 | A ZSTD_DCtx object is required to track streaming operations. |
| 1797 | Use ZSTD_createDCtx() / ZSTD_freeDCtx() to manage it. |
| 1798 | A ZSTD_DCtx object can be re-used multiple times. |
| 1799 | |
| 1800 | First typical operation is to retrieve frame parameters, using ZSTD_getFrameHeader(). |
| 1801 | Frame header is extracted from the beginning of compressed frame, so providing only the frame's beginning is enough. |
| 1802 | Data fragment must be large enough to ensure successful decoding. |
| 1803 | `ZSTD_frameHeaderSize_max` bytes is guaranteed to always be large enough. |
| 1804 | @result : 0 : successful decoding, the `ZSTD_frameHeader` structure is correctly filled. |
| 1805 | >0 : `srcSize` is too small, please provide at least @result bytes on next attempt. |
| 1806 | errorCode, which can be tested using ZSTD_isError(). |
| 1807 | |
| 1808 | It fills a ZSTD_frameHeader structure with important information to correctly decode the frame, |
| 1809 | such as the dictionary ID, content size, or maximum back-reference distance (`windowSize`). |
| 1810 | Note that these values could be wrong, either because of data corruption, or because a 3rd party deliberately spoofs false information. |
| 1811 | As a consequence, check that values remain within valid application range. |
| 1812 | For example, do not allocate memory blindly, check that `windowSize` is within expectation. |
| 1813 | Each application can set its own limits, depending on local restrictions. |
| 1814 | For extended interoperability, it is recommended to support `windowSize` of at least 8 MB. |
| 1815 | |
| 1816 | ZSTD_decompressContinue() needs previous data blocks during decompression, up to `windowSize` bytes. |
| 1817 | ZSTD_decompressContinue() is very sensitive to contiguity, |
| 1818 | if 2 blocks don't follow each other, make sure that either the compressor breaks contiguity at the same place, |
| 1819 | or that previous contiguous segment is large enough to properly handle maximum back-reference distance. |
| 1820 | There are multiple ways to guarantee this condition. |
| 1821 | |
| 1822 | The most memory efficient way is to use a round buffer of sufficient size. |
| 1823 | Sufficient size is determined by invoking ZSTD_decodingBufferSize_min(), |
| 1824 | which can @return an error code if required value is too large for current system (in 32-bits mode). |
| 1825 | In a round buffer methodology, ZSTD_decompressContinue() decompresses each block next to previous one, |
| 1826 | up to the moment there is not enough room left in the buffer to guarantee decoding another full block, |
| 1827 | which maximum size is provided in `ZSTD_frameHeader` structure, field `blockSizeMax`. |
| 1828 | At which point, decoding can resume from the beginning of the buffer. |
| 1829 | Note that already decoded data stored in the buffer should be flushed before being overwritten. |
| 1830 | |
| 1831 | There are alternatives possible, for example using two or more buffers of size `windowSize` each, though they consume more memory. |
| 1832 | |
| 1833 | Finally, if you control the compression process, you can also ignore all buffer size rules, |
| 1834 | as long as the encoder and decoder progress in "lock-step", |
| 1835 | aka use exactly the same buffer sizes, break contiguity at the same place, etc. |
| 1836 | |
| 1837 | Once buffers are setup, start decompression, with ZSTD_decompressBegin(). |
| 1838 | If decompression requires a dictionary, use ZSTD_decompressBegin_usingDict() or ZSTD_decompressBegin_usingDDict(). |
| 1839 | |
| 1840 | Then use ZSTD_nextSrcSizeToDecompress() and ZSTD_decompressContinue() alternatively. |
| 1841 | ZSTD_nextSrcSizeToDecompress() tells how many bytes to provide as 'srcSize' to ZSTD_decompressContinue(). |
| 1842 | ZSTD_decompressContinue() requires this _exact_ amount of bytes, or it will fail. |
| 1843 | |
| 1844 | @result of ZSTD_decompressContinue() is the number of bytes regenerated within 'dst' (necessarily <= dstCapacity). |
| 1845 | It can be zero : it just means ZSTD_decompressContinue() has decoded some metadata item. |
| 1846 | It can also be an error code, which can be tested with ZSTD_isError(). |
| 1847 | |
| 1848 | A frame is fully decoded when ZSTD_nextSrcSizeToDecompress() returns zero. |
| 1849 | Context can then be reset to start a new decompression. |
| 1850 | |
| 1851 | Note : it's possible to know if next input to present is a header or a block, using ZSTD_nextInputType(). |
| 1852 | This information is not required to properly decode a frame. |
| 1853 | |
| 1854 | == Special case : skippable frames == |
| 1855 | |
| 1856 | Skippable frames allow integration of user-defined data into a flow of concatenated frames. |
| 1857 | Skippable frames will be ignored (skipped) by decompressor. |
| 1858 | The format of skippable frames is as follows : |
| 1859 | a) Skippable frame ID - 4 Bytes, Little endian format, any value from 0x184D2A50 to 0x184D2A5F |
| 1860 | b) Frame Size - 4 Bytes, Little endian format, unsigned 32-bits |
| 1861 | c) Frame Content - any content (User Data) of length equal to Frame Size |
| 1862 | For skippable frames ZSTD_getFrameHeader() returns zfhPtr->frameType==ZSTD_skippableFrame. |
| 1863 | For skippable frames ZSTD_decompressContinue() always returns 0 : it only skips the content. |
| 1864 | */ |
| 1865 | |
| 1866 | /*===== Buffer-less streaming decompression functions =====*/ |
| 1867 | typedef enum { ZSTD_frame, ZSTD_skippableFrame } ZSTD_frameType_e; |
| 1868 | typedef struct { |
| 1869 | unsigned long long frameContentSize; /* if == ZSTD_CONTENTSIZE_UNKNOWN, it means this field is not available. 0 means "empty" */ |
| 1870 | unsigned long long windowSize; /* can be very large, up to <= frameContentSize */ |
| 1871 | unsigned blockSizeMax; |
| 1872 | ZSTD_frameType_e frameType; /* if == ZSTD_skippableFrame, frameContentSize is the size of skippable content */ |
| 1873 | unsigned headerSize; |
| 1874 | unsigned dictID; |
| 1875 | unsigned checksumFlag; |
| 1876 | } ZSTD_frameHeader; |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1877 | |
| 1878 | /*! ZSTD_getFrameHeader() : |
| 1879 | * decode Frame Header, or requires larger `srcSize`. |
| 1880 | * @return : 0, `zfhPtr` is correctly filled, |
| 1881 | * >0, `srcSize` is too small, value is wanted `srcSize` amount, |
| 1882 | * or an error code, which can be tested using ZSTD_isError() */ |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1883 | ZSTDLIB_API size_t ZSTD_getFrameHeader(ZSTD_frameHeader* zfhPtr, const void* src, size_t srcSize); /**< doesn't consume input */ |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1884 | /*! ZSTD_getFrameHeader_advanced() : |
| 1885 | * same as ZSTD_getFrameHeader(), |
| 1886 | * with added capability to select a format (like ZSTD_f_zstd1_magicless) */ |
| 1887 | ZSTDLIB_API size_t ZSTD_getFrameHeader_advanced(ZSTD_frameHeader* zfhPtr, const void* src, size_t srcSize, ZSTD_format_e format); |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1888 | 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 */ |
| 1889 | |
| 1890 | ZSTDLIB_API size_t ZSTD_decompressBegin(ZSTD_DCtx* dctx); |
| 1891 | ZSTDLIB_API size_t ZSTD_decompressBegin_usingDict(ZSTD_DCtx* dctx, const void* dict, size_t dictSize); |
| 1892 | ZSTDLIB_API size_t ZSTD_decompressBegin_usingDDict(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict); |
| 1893 | |
| 1894 | ZSTDLIB_API size_t ZSTD_nextSrcSizeToDecompress(ZSTD_DCtx* dctx); |
| 1895 | ZSTDLIB_API size_t ZSTD_decompressContinue(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize); |
| 1896 | |
| 1897 | /* misc */ |
| 1898 | ZSTDLIB_API void ZSTD_copyDCtx(ZSTD_DCtx* dctx, const ZSTD_DCtx* preparedDCtx); |
| 1899 | typedef enum { ZSTDnit_frameHeader, ZSTDnit_blockHeader, ZSTDnit_block, ZSTDnit_lastBlock, ZSTDnit_checksum, ZSTDnit_skippableFrame } ZSTD_nextInputType_e; |
| 1900 | ZSTDLIB_API ZSTD_nextInputType_e ZSTD_nextInputType(ZSTD_DCtx* dctx); |
| 1901 | |
| 1902 | |
| 1903 | |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1904 | |
| 1905 | /* ============================ */ |
| 1906 | /** Block level API */ |
| 1907 | /* ============================ */ |
| 1908 | |
| 1909 | /*! |
| 1910 | Block functions produce and decode raw zstd blocks, without frame metadata. |
| 1911 | Frame metadata cost is typically ~18 bytes, which can be non-negligible for very small blocks (< 100 bytes). |
| 1912 | User will have to take in charge required information to regenerate data, such as compressed and content sizes. |
| 1913 | |
| 1914 | A few rules to respect : |
| 1915 | - Compressing and decompressing require a context structure |
| 1916 | + Use ZSTD_createCCtx() and ZSTD_createDCtx() |
| 1917 | - It is necessary to init context before starting |
| 1918 | + compression : any ZSTD_compressBegin*() variant, including with dictionary |
| 1919 | + decompression : any ZSTD_decompressBegin*() variant, including with dictionary |
| 1920 | + copyCCtx() and copyDCtx() can be used too |
| 1921 | - Block size is limited, it must be <= ZSTD_getBlockSize() <= ZSTD_BLOCKSIZE_MAX == 128 KB |
| 1922 | + If input is larger than a block size, it's necessary to split input data into multiple blocks |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1923 | + For inputs larger than a single block, really consider using regular ZSTD_compress() instead. |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1924 | Frame metadata is not that costly, and quickly becomes negligible as source size grows larger. |
| 1925 | - When a block is considered not compressible enough, ZSTD_compressBlock() result will be zero. |
Scott Baker | 611f6bd | 2019-10-18 13:45:19 -0700 | [diff] [blame^] | 1926 | In which case, nothing is produced into `dst` ! |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1927 | + User must test for such outcome and deal directly with uncompressed data |
| 1928 | + ZSTD_decompressBlock() doesn't accept uncompressed data as input !!! |
| 1929 | + In case of multiple successive blocks, should some of them be uncompressed, |
| 1930 | decoder must be informed of their existence in order to follow proper history. |
| 1931 | Use ZSTD_insertBlock() for such a case. |
| 1932 | */ |
| 1933 | |
Scott Baker | eee8dd8 | 2019-09-24 12:52:34 -0700 | [diff] [blame] | 1934 | /*===== Raw zstd block functions =====*/ |
| 1935 | ZSTDLIB_API size_t ZSTD_getBlockSize (const ZSTD_CCtx* cctx); |
| 1936 | ZSTDLIB_API size_t ZSTD_compressBlock (ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize); |
| 1937 | ZSTDLIB_API size_t ZSTD_decompressBlock(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize); |
| 1938 | 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. */ |
| 1939 | |
| 1940 | |
| 1941 | #endif /* ZSTD_H_ZSTD_STATIC_LINKING_ONLY */ |
| 1942 | |
| 1943 | #if defined (__cplusplus) |
| 1944 | } |
| 1945 | #endif |