You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

huf.h 19KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340
  1. /* ******************************************************************
  2. * huff0 huffman codec,
  3. * part of Finite State Entropy library
  4. * Copyright (c) 2013-2020, Yann Collet, Facebook, Inc.
  5. *
  6. * You can contact the author at :
  7. * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
  8. *
  9. * This source code is licensed under both the BSD-style license (found in the
  10. * LICENSE file in the root directory of this source tree) and the GPLv2 (found
  11. * in the COPYING file in the root directory of this source tree).
  12. * You may select, at your option, one of the above-listed licenses.
  13. ****************************************************************** */
  14. #if defined (__cplusplus)
  15. extern "C" {
  16. #endif
  17. #ifndef HUF_H_298734234
  18. #define HUF_H_298734234
  19. /* *** Dependencies *** */
  20. #include <stddef.h> /* size_t */
  21. /* *** library symbols visibility *** */
  22. /* Note : when linking with -fvisibility=hidden on gcc, or by default on Visual,
  23. * HUF symbols remain "private" (internal symbols for library only).
  24. * Set macro FSE_DLL_EXPORT to 1 if you want HUF symbols visible on DLL interface */
  25. #if defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) && defined(__GNUC__) && (__GNUC__ >= 4)
  26. # define HUF_PUBLIC_API __attribute__ ((visibility ("default")))
  27. #elif defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) /* Visual expected */
  28. # define HUF_PUBLIC_API __declspec(dllexport)
  29. #elif defined(FSE_DLL_IMPORT) && (FSE_DLL_IMPORT==1)
  30. # define HUF_PUBLIC_API __declspec(dllimport) /* not required, just to generate faster code (saves a function pointer load from IAT and an indirect jump) */
  31. #else
  32. # define HUF_PUBLIC_API
  33. #endif
  34. /* ========================== */
  35. /* *** simple functions *** */
  36. /* ========================== */
  37. /** HUF_compress() :
  38. * Compress content from buffer 'src', of size 'srcSize', into buffer 'dst'.
  39. * 'dst' buffer must be already allocated.
  40. * Compression runs faster if `dstCapacity` >= HUF_compressBound(srcSize).
  41. * `srcSize` must be <= `HUF_BLOCKSIZE_MAX` == 128 KB.
  42. * @return : size of compressed data (<= `dstCapacity`).
  43. * Special values : if return == 0, srcData is not compressible => Nothing is stored within dst !!!
  44. * if HUF_isError(return), compression failed (more details using HUF_getErrorName())
  45. */
  46. HUF_PUBLIC_API size_t HUF_compress(void* dst, size_t dstCapacity,
  47. const void* src, size_t srcSize);
  48. /** HUF_decompress() :
  49. * Decompress HUF data from buffer 'cSrc', of size 'cSrcSize',
  50. * into already allocated buffer 'dst', of minimum size 'dstSize'.
  51. * `originalSize` : **must** be the ***exact*** size of original (uncompressed) data.
  52. * Note : in contrast with FSE, HUF_decompress can regenerate
  53. * RLE (cSrcSize==1) and uncompressed (cSrcSize==dstSize) data,
  54. * because it knows size to regenerate (originalSize).
  55. * @return : size of regenerated data (== originalSize),
  56. * or an error code, which can be tested using HUF_isError()
  57. */
  58. HUF_PUBLIC_API size_t HUF_decompress(void* dst, size_t originalSize,
  59. const void* cSrc, size_t cSrcSize);
  60. /* *** Tool functions *** */
  61. #define HUF_BLOCKSIZE_MAX (128 * 1024) /**< maximum input size for a single block compressed with HUF_compress */
  62. HUF_PUBLIC_API size_t HUF_compressBound(size_t size); /**< maximum compressed size (worst case) */
  63. /* Error Management */
  64. HUF_PUBLIC_API unsigned HUF_isError(size_t code); /**< tells if a return value is an error code */
  65. HUF_PUBLIC_API const char* HUF_getErrorName(size_t code); /**< provides error code string (useful for debugging) */
  66. /* *** Advanced function *** */
  67. /** HUF_compress2() :
  68. * Same as HUF_compress(), but offers control over `maxSymbolValue` and `tableLog`.
  69. * `maxSymbolValue` must be <= HUF_SYMBOLVALUE_MAX .
  70. * `tableLog` must be `<= HUF_TABLELOG_MAX` . */
  71. HUF_PUBLIC_API size_t HUF_compress2 (void* dst, size_t dstCapacity,
  72. const void* src, size_t srcSize,
  73. unsigned maxSymbolValue, unsigned tableLog);
  74. /** HUF_compress4X_wksp() :
  75. * Same as HUF_compress2(), but uses externally allocated `workSpace`.
  76. * `workspace` must have minimum alignment of 4, and be at least as large as HUF_WORKSPACE_SIZE */
  77. #define HUF_WORKSPACE_SIZE ((6 << 10) + 256)
  78. #define HUF_WORKSPACE_SIZE_U32 (HUF_WORKSPACE_SIZE / sizeof(U32))
  79. HUF_PUBLIC_API size_t HUF_compress4X_wksp (void* dst, size_t dstCapacity,
  80. const void* src, size_t srcSize,
  81. unsigned maxSymbolValue, unsigned tableLog,
  82. void* workSpace, size_t wkspSize);
  83. #endif /* HUF_H_298734234 */
  84. /* ******************************************************************
  85. * WARNING !!
  86. * The following section contains advanced and experimental definitions
  87. * which shall never be used in the context of a dynamic library,
  88. * because they are not guaranteed to remain stable in the future.
  89. * Only consider them in association with static linking.
  90. * *****************************************************************/
  91. #if defined(HUF_STATIC_LINKING_ONLY) && !defined(HUF_H_HUF_STATIC_LINKING_ONLY)
  92. #define HUF_H_HUF_STATIC_LINKING_ONLY
  93. /* *** Dependencies *** */
  94. #include "mem.h" /* U32 */
  95. /* *** Constants *** */
  96. #define HUF_TABLELOG_MAX 12 /* max runtime value of tableLog (due to static allocation); can be modified up to HUF_ABSOLUTEMAX_TABLELOG */
  97. #define HUF_TABLELOG_DEFAULT 11 /* default tableLog value when none specified */
  98. #define HUF_SYMBOLVALUE_MAX 255
  99. #define HUF_TABLELOG_ABSOLUTEMAX 15 /* absolute limit of HUF_MAX_TABLELOG. Beyond that value, code does not work */
  100. #if (HUF_TABLELOG_MAX > HUF_TABLELOG_ABSOLUTEMAX)
  101. # error "HUF_TABLELOG_MAX is too large !"
  102. #endif
  103. /* ****************************************
  104. * Static allocation
  105. ******************************************/
  106. /* HUF buffer bounds */
  107. #define HUF_CTABLEBOUND 129
  108. #define HUF_BLOCKBOUND(size) (size + (size>>8) + 8) /* only true when incompressible is pre-filtered with fast heuristic */
  109. #define HUF_COMPRESSBOUND(size) (HUF_CTABLEBOUND + HUF_BLOCKBOUND(size)) /* Macro version, useful for static allocation */
  110. /* static allocation of HUF's Compression Table */
  111. #define HUF_CTABLE_SIZE_U32(maxSymbolValue) ((maxSymbolValue)+1) /* Use tables of U32, for proper alignment */
  112. #define HUF_CTABLE_SIZE(maxSymbolValue) (HUF_CTABLE_SIZE_U32(maxSymbolValue) * sizeof(U32))
  113. #define HUF_CREATE_STATIC_CTABLE(name, maxSymbolValue) \
  114. U32 name##hb[HUF_CTABLE_SIZE_U32(maxSymbolValue)]; \
  115. void* name##hv = &(name##hb); \
  116. HUF_CElt* name = (HUF_CElt*)(name##hv) /* no final ; */
  117. /* static allocation of HUF's DTable */
  118. typedef U32 HUF_DTable;
  119. #define HUF_DTABLE_SIZE(maxTableLog) (1 + (1<<(maxTableLog)))
  120. #define HUF_CREATE_STATIC_DTABLEX1(DTable, maxTableLog) \
  121. HUF_DTable DTable[HUF_DTABLE_SIZE((maxTableLog)-1)] = { ((U32)((maxTableLog)-1) * 0x01000001) }
  122. #define HUF_CREATE_STATIC_DTABLEX2(DTable, maxTableLog) \
  123. HUF_DTable DTable[HUF_DTABLE_SIZE(maxTableLog)] = { ((U32)(maxTableLog) * 0x01000001) }
  124. /* ****************************************
  125. * Advanced decompression functions
  126. ******************************************/
  127. size_t HUF_decompress4X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< single-symbol decoder */
  128. #ifndef HUF_FORCE_DECOMPRESS_X1
  129. size_t HUF_decompress4X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< double-symbols decoder */
  130. #endif
  131. size_t HUF_decompress4X_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< decodes RLE and uncompressed */
  132. size_t HUF_decompress4X_hufOnly(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< considers RLE and uncompressed as errors */
  133. size_t HUF_decompress4X_hufOnly_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /**< considers RLE and uncompressed as errors */
  134. size_t HUF_decompress4X1_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< single-symbol decoder */
  135. size_t HUF_decompress4X1_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /**< single-symbol decoder */
  136. #ifndef HUF_FORCE_DECOMPRESS_X1
  137. size_t HUF_decompress4X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< double-symbols decoder */
  138. size_t HUF_decompress4X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /**< double-symbols decoder */
  139. #endif
  140. /* ****************************************
  141. * HUF detailed API
  142. * ****************************************/
  143. /*! HUF_compress() does the following:
  144. * 1. count symbol occurrence from source[] into table count[] using FSE_count() (exposed within "fse.h")
  145. * 2. (optional) refine tableLog using HUF_optimalTableLog()
  146. * 3. build Huffman table from count using HUF_buildCTable()
  147. * 4. save Huffman table to memory buffer using HUF_writeCTable()
  148. * 5. encode the data stream using HUF_compress4X_usingCTable()
  149. *
  150. * The following API allows targeting specific sub-functions for advanced tasks.
  151. * For example, it's possible to compress several blocks using the same 'CTable',
  152. * or to save and regenerate 'CTable' using external methods.
  153. */
  154. unsigned HUF_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue);
  155. typedef struct HUF_CElt_s HUF_CElt; /* incomplete type */
  156. size_t HUF_buildCTable (HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue, unsigned maxNbBits); /* @return : maxNbBits; CTable and count can overlap. In which case, CTable will overwrite count content */
  157. size_t HUF_writeCTable (void* dst, size_t maxDstSize, const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog);
  158. size_t HUF_compress4X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable);
  159. size_t HUF_estimateCompressedSize(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue);
  160. int HUF_validateCTable(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue);
  161. typedef enum {
  162. HUF_repeat_none, /**< Cannot use the previous table */
  163. HUF_repeat_check, /**< Can use the previous table but it must be checked. Note : The previous table must have been constructed by HUF_compress{1, 4}X_repeat */
  164. HUF_repeat_valid /**< Can use the previous table and it is assumed to be valid */
  165. } HUF_repeat;
  166. /** HUF_compress4X_repeat() :
  167. * Same as HUF_compress4X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none.
  168. * If it uses hufTable it does not modify hufTable or repeat.
  169. * If it doesn't, it sets *repeat = HUF_repeat_none, and it sets hufTable to the table used.
  170. * If preferRepeat then the old table will always be used if valid. */
  171. size_t HUF_compress4X_repeat(void* dst, size_t dstSize,
  172. const void* src, size_t srcSize,
  173. unsigned maxSymbolValue, unsigned tableLog,
  174. void* workSpace, size_t wkspSize, /**< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */
  175. HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, int bmi2);
  176. /** HUF_buildCTable_wksp() :
  177. * Same as HUF_buildCTable(), but using externally allocated scratch buffer.
  178. * `workSpace` must be aligned on 4-bytes boundaries, and its size must be >= HUF_CTABLE_WORKSPACE_SIZE.
  179. */
  180. #define HUF_CTABLE_WORKSPACE_SIZE_U32 (2*HUF_SYMBOLVALUE_MAX +1 +1)
  181. #define HUF_CTABLE_WORKSPACE_SIZE (HUF_CTABLE_WORKSPACE_SIZE_U32 * sizeof(unsigned))
  182. size_t HUF_buildCTable_wksp (HUF_CElt* tree,
  183. const unsigned* count, U32 maxSymbolValue, U32 maxNbBits,
  184. void* workSpace, size_t wkspSize);
  185. /*! HUF_readStats() :
  186. * Read compact Huffman tree, saved by HUF_writeCTable().
  187. * `huffWeight` is destination buffer.
  188. * @return : size read from `src` , or an error Code .
  189. * Note : Needed by HUF_readCTable() and HUF_readDTableXn() . */
  190. size_t HUF_readStats(BYTE* huffWeight, size_t hwSize,
  191. U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr,
  192. const void* src, size_t srcSize);
  193. /** HUF_readCTable() :
  194. * Loading a CTable saved with HUF_writeCTable() */
  195. size_t HUF_readCTable (HUF_CElt* CTable, unsigned* maxSymbolValuePtr, const void* src, size_t srcSize, unsigned *hasZeroWeights);
  196. /** HUF_getNbBits() :
  197. * Read nbBits from CTable symbolTable, for symbol `symbolValue` presumed <= HUF_SYMBOLVALUE_MAX
  198. * Note 1 : is not inlined, as HUF_CElt definition is private
  199. * Note 2 : const void* used, so that it can provide a statically allocated table as argument (which uses type U32) */
  200. U32 HUF_getNbBits(const void* symbolTable, U32 symbolValue);
  201. /*
  202. * HUF_decompress() does the following:
  203. * 1. select the decompression algorithm (X1, X2) based on pre-computed heuristics
  204. * 2. build Huffman table from save, using HUF_readDTableX?()
  205. * 3. decode 1 or 4 segments in parallel using HUF_decompress?X?_usingDTable()
  206. */
  207. /** HUF_selectDecoder() :
  208. * Tells which decoder is likely to decode faster,
  209. * based on a set of pre-computed metrics.
  210. * @return : 0==HUF_decompress4X1, 1==HUF_decompress4X2 .
  211. * Assumption : 0 < dstSize <= 128 KB */
  212. U32 HUF_selectDecoder (size_t dstSize, size_t cSrcSize);
  213. /**
  214. * The minimum workspace size for the `workSpace` used in
  215. * HUF_readDTableX1_wksp() and HUF_readDTableX2_wksp().
  216. *
  217. * The space used depends on HUF_TABLELOG_MAX, ranging from ~1500 bytes when
  218. * HUF_TABLE_LOG_MAX=12 to ~1850 bytes when HUF_TABLE_LOG_MAX=15.
  219. * Buffer overflow errors may potentially occur if code modifications result in
  220. * a required workspace size greater than that specified in the following
  221. * macro.
  222. */
  223. #define HUF_DECOMPRESS_WORKSPACE_SIZE (2 << 10)
  224. #define HUF_DECOMPRESS_WORKSPACE_SIZE_U32 (HUF_DECOMPRESS_WORKSPACE_SIZE / sizeof(U32))
  225. #ifndef HUF_FORCE_DECOMPRESS_X2
  226. size_t HUF_readDTableX1 (HUF_DTable* DTable, const void* src, size_t srcSize);
  227. size_t HUF_readDTableX1_wksp (HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize);
  228. #endif
  229. #ifndef HUF_FORCE_DECOMPRESS_X1
  230. size_t HUF_readDTableX2 (HUF_DTable* DTable, const void* src, size_t srcSize);
  231. size_t HUF_readDTableX2_wksp (HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize);
  232. #endif
  233. size_t HUF_decompress4X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
  234. #ifndef HUF_FORCE_DECOMPRESS_X2
  235. size_t HUF_decompress4X1_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
  236. #endif
  237. #ifndef HUF_FORCE_DECOMPRESS_X1
  238. size_t HUF_decompress4X2_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
  239. #endif
  240. /* ====================== */
  241. /* single stream variants */
  242. /* ====================== */
  243. size_t HUF_compress1X (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog);
  244. size_t HUF_compress1X_wksp (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize); /**< `workSpace` must be a table of at least HUF_WORKSPACE_SIZE_U32 unsigned */
  245. size_t HUF_compress1X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable);
  246. /** HUF_compress1X_repeat() :
  247. * Same as HUF_compress1X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none.
  248. * If it uses hufTable it does not modify hufTable or repeat.
  249. * If it doesn't, it sets *repeat = HUF_repeat_none, and it sets hufTable to the table used.
  250. * If preferRepeat then the old table will always be used if valid. */
  251. size_t HUF_compress1X_repeat(void* dst, size_t dstSize,
  252. const void* src, size_t srcSize,
  253. unsigned maxSymbolValue, unsigned tableLog,
  254. void* workSpace, size_t wkspSize, /**< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */
  255. HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, int bmi2);
  256. size_t HUF_decompress1X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /* single-symbol decoder */
  257. #ifndef HUF_FORCE_DECOMPRESS_X1
  258. size_t HUF_decompress1X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /* double-symbol decoder */
  259. #endif
  260. size_t HUF_decompress1X_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);
  261. size_t HUF_decompress1X_DCtx_wksp (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize);
  262. #ifndef HUF_FORCE_DECOMPRESS_X2
  263. size_t HUF_decompress1X1_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< single-symbol decoder */
  264. size_t HUF_decompress1X1_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /**< single-symbol decoder */
  265. #endif
  266. #ifndef HUF_FORCE_DECOMPRESS_X1
  267. size_t HUF_decompress1X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< double-symbols decoder */
  268. size_t HUF_decompress1X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /**< double-symbols decoder */
  269. #endif
  270. size_t HUF_decompress1X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable); /**< automatic selection of sing or double symbol decoder, based on DTable */
  271. #ifndef HUF_FORCE_DECOMPRESS_X2
  272. size_t HUF_decompress1X1_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
  273. #endif
  274. #ifndef HUF_FORCE_DECOMPRESS_X1
  275. size_t HUF_decompress1X2_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
  276. #endif
  277. /* BMI2 variants.
  278. * If the CPU has BMI2 support, pass bmi2=1, otherwise pass bmi2=0.
  279. */
  280. size_t HUF_decompress1X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2);
  281. #ifndef HUF_FORCE_DECOMPRESS_X2
  282. size_t HUF_decompress1X1_DCtx_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int bmi2);
  283. #endif
  284. size_t HUF_decompress4X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2);
  285. size_t HUF_decompress4X_hufOnly_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int bmi2);
  286. #endif /* HUF_STATIC_LINKING_ONLY */
  287. #if defined (__cplusplus)
  288. }
  289. #endif