huf.h File Reference

#include "zstd_deps.h"
#include "mem.h"
#include "fse.h"

Go to the source code of this file.

Macros
#define	HUF_H_298734234
#define	FSE_STATIC_LINKING_ONLY
#define	HUF_BLOCKSIZE_MAX   (128 * 1024)
#define	HUF_WORKSPACE_SIZE   ((8 << 10) + 512 /* sorting scratch space */)
#define	HUF_WORKSPACE_SIZE_U64   (HUF_WORKSPACE_SIZE / sizeof(U64))
#define	HUF_TABLELOG_MAX   12 /* max runtime value of tableLog (due to static allocation); can be modified up to HUF_TABLELOG_ABSOLUTEMAX */
#define	HUF_TABLELOG_DEFAULT   11 /* default tableLog value when none specified */
#define	HUF_SYMBOLVALUE_MAX   255
#define	HUF_TABLELOG_ABSOLUTEMAX   12 /* absolute limit of HUF_MAX_TABLELOG. Beyond that value, code does not work */
#define	HUF_CTABLEBOUND   129
#define	HUF_BLOCKBOUND(size)
#define	HUF_COMPRESSBOUND(size)
#define	HUF_CTABLE_SIZE_ST(maxSymbolValue)
#define	HUF_CTABLE_SIZE(maxSymbolValue)
#define	HUF_CREATE_STATIC_CTABLE(name, maxSymbolValue)
#define	HUF_DTABLE_SIZE(maxTableLog)
#define	HUF_CREATE_STATIC_DTABLEX1(DTable, maxTableLog)
#define	HUF_CREATE_STATIC_DTABLEX2(DTable, maxTableLog)
#define	HUF_OPTIMAL_DEPTH_THRESHOLD   ZSTD_btultra
#define	HUF_CTABLE_WORKSPACE_SIZE_U32   ((4 * (HUF_SYMBOLVALUE_MAX + 1)) + 192)
#define	HUF_CTABLE_WORKSPACE_SIZE   (HUF_CTABLE_WORKSPACE_SIZE_U32 * sizeof(unsigned))
#define	HUF_READ_STATS_WORKSPACE_SIZE_U32   FSE_DECOMPRESS_WKSP_SIZE_U32(6, HUF_TABLELOG_MAX-1)
#define	HUF_READ_STATS_WORKSPACE_SIZE   (HUF_READ_STATS_WORKSPACE_SIZE_U32 * sizeof(unsigned))
#define	HUF_DECOMPRESS_WORKSPACE_SIZE   ((2 << 10) + (1 << 9))
#define	HUF_DECOMPRESS_WORKSPACE_SIZE_U32   (HUF_DECOMPRESS_WORKSPACE_SIZE / sizeof(U32))

Typedefs
typedef size_t	HUF_CElt
typedef U32	HUF_DTable

Enumerations
enum	HUF_flags_e {   HUF_flags_bmi2 = (1 << 0) , HUF_flags_optimalDepth = (1 << 1) , HUF_flags_preferRepeat = (1 << 2) , HUF_flags_suspectUncompressible = (1 << 3) ,   HUF_flags_disableAsm = (1 << 4) , HUF_flags_disableFast = (1 << 5) }
enum	HUF_repeat { HUF_repeat_none , HUF_repeat_check , HUF_repeat_valid }

Functions
size_t	HUF_compressBound (size_t size)
unsigned	HUF_isError (size_t code)
const char *	HUF_getErrorName (size_t code)
unsigned	HUF_minTableLog (unsigned symbolCardinality)
unsigned	HUF_cardinality (const unsigned *count, unsigned maxSymbolValue)
unsigned	HUF_optimalTableLog (unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue, void *workSpace, size_t wkspSize, HUF_CElt *table, const unsigned *count, int flags)
size_t	HUF_writeCTable_wksp (void *dst, size_t maxDstSize, const HUF_CElt *CTable, unsigned maxSymbolValue, unsigned huffLog, void *workspace, size_t workspaceSize)
size_t	HUF_compress4X_usingCTable (void *dst, size_t dstSize, const void *src, size_t srcSize, const HUF_CElt *CTable, int flags)
size_t	HUF_estimateCompressedSize (const HUF_CElt *CTable, const unsigned *count, unsigned maxSymbolValue)
int	HUF_validateCTable (const HUF_CElt *CTable, const unsigned *count, unsigned maxSymbolValue)
size_t	HUF_compress4X_repeat (void *dst, size_t dstSize, const void *src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog, void *workSpace, size_t wkspSize, HUF_CElt *hufTable, HUF_repeat *repeat, int flags)
size_t	HUF_buildCTable_wksp (HUF_CElt *tree, const unsigned *count, U32 maxSymbolValue, U32 maxNbBits, void *workSpace, size_t wkspSize)
size_t	HUF_readStats (BYTE *huffWeight, size_t hwSize, U32 *rankStats, U32 *nbSymbolsPtr, U32 *tableLogPtr, const void *src, size_t srcSize)
size_t	HUF_readStats_wksp (BYTE *huffWeight, size_t hwSize, U32 *rankStats, U32 *nbSymbolsPtr, U32 *tableLogPtr, const void *src, size_t srcSize, void *workspace, size_t wkspSize, int flags)
size_t	HUF_readCTable (HUF_CElt *CTable, unsigned *maxSymbolValuePtr, const void *src, size_t srcSize, unsigned *hasZeroWeights)
U32	HUF_getNbBitsFromCTable (const HUF_CElt *symbolTable, U32 symbolValue)
U32	HUF_selectDecoder (size_t dstSize, size_t cSrcSize)
size_t	HUF_compress1X_usingCTable (void *dst, size_t dstSize, const void *src, size_t srcSize, const HUF_CElt *CTable, int flags)
size_t	HUF_compress1X_repeat (void *dst, size_t dstSize, const void *src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog, void *workSpace, size_t wkspSize, HUF_CElt *hufTable, HUF_repeat *repeat, int flags)
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, int flags)
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, int flags)
size_t	HUF_decompress1X_usingDTable (void *dst, size_t maxDstSize, const void *cSrc, size_t cSrcSize, const HUF_DTable *DTable, int flags)
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, int flags)
size_t	HUF_decompress4X_usingDTable (void *dst, size_t maxDstSize, const void *cSrc, size_t cSrcSize, const HUF_DTable *DTable, int flags)
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, int flags)
size_t	HUF_readDTableX1_wksp (HUF_DTable *DTable, const void *src, size_t srcSize, void *workSpace, size_t wkspSize, int flags)
size_t	HUF_readDTableX2_wksp (HUF_DTable *DTable, const void *src, size_t srcSize, void *workSpace, size_t wkspSize, int flags)

Macro Definition Documentation

FSE_STATIC_LINKING_ONLY

#define FSE_STATIC_LINKING_ONLY

HUF_BLOCKBOUND

#define HUF_BLOCKBOUND

(

size

)

Value:

(size + (size>>8) + 8)   /* only true when incompressible is pre-filtered with fast heuristic */

HUF_BLOCKSIZE_MAX

#define HUF_BLOCKSIZE_MAX   (128 * 1024)

maximum input size for a single block compressed with HUF_compress

HUF_COMPRESSBOUND

#define HUF_COMPRESSBOUND

(

size

)

Value:

(HUF_CTABLEBOUND + HUF_BLOCKBOUND(size))   /* Macro version, useful for static allocation */

HUF_CREATE_STATIC_CTABLE

#define HUF_CREATE_STATIC_CTABLE	(		name,
			maxSymbolValue )

Value:

    HUF_CElt name[HUF_CTABLE_SIZE_ST(maxSymbolValue)] /* no final ; */

HUF_CREATE_STATIC_DTABLEX1

#define HUF_CREATE_STATIC_DTABLEX1	(		DTable,
			maxTableLog )

Value:

        HUF_DTable DTable[HUF_DTABLE_SIZE((maxTableLog)-1)] = { ((U32)((maxTableLog)-1) * 0x01000001) }

HUF_CREATE_STATIC_DTABLEX2

#define HUF_CREATE_STATIC_DTABLEX2	(		DTable,
			maxTableLog )

Value:

        HUF_DTable DTable[HUF_DTABLE_SIZE(maxTableLog)] = { ((U32)(maxTableLog) * 0x01000001) }

HUF_CTABLE_SIZE

#define HUF_CTABLE_SIZE

(

maxSymbolValue

)

Value:

(HUF_CTABLE_SIZE_ST(maxSymbolValue) * sizeof(size_t))

HUF_CTABLE_SIZE_ST

#define HUF_CTABLE_SIZE_ST

(

maxSymbolValue

)

Value:

((maxSymbolValue)+2)   /* Use tables of size_t, for proper alignment */

HUF_CTABLE_WORKSPACE_SIZE

#define HUF_CTABLE_WORKSPACE_SIZE   (HUF_CTABLE_WORKSPACE_SIZE_U32 * sizeof(unsigned))

HUF_CTABLE_WORKSPACE_SIZE_U32

#define HUF_CTABLE_WORKSPACE_SIZE_U32   ((4 * (HUF_SYMBOLVALUE_MAX + 1)) + 192)

HUF_buildCTable_wksp() : Same as HUF_buildCTable(), but using externally allocated scratch buffer. workSpace must be aligned on 4-bytes boundaries, and its size must be >= HUF_CTABLE_WORKSPACE_SIZE.

HUF_CTABLEBOUND

#define HUF_CTABLEBOUND   129

HUF_DECOMPRESS_WORKSPACE_SIZE

#define HUF_DECOMPRESS_WORKSPACE_SIZE   ((2 << 10) + (1 << 9))

The minimum workspace size for the workSpace used in HUF_readDTableX1_wksp() and HUF_readDTableX2_wksp().

The space used depends on HUF_TABLELOG_MAX, ranging from ~1500 bytes when HUF_TABLE_LOG_MAX=12 to ~1850 bytes when HUF_TABLE_LOG_MAX=15. Buffer overflow errors may potentially occur if code modifications result in a required workspace size greater than that specified in the following macro.

HUF_DECOMPRESS_WORKSPACE_SIZE_U32

#define HUF_DECOMPRESS_WORKSPACE_SIZE_U32   (HUF_DECOMPRESS_WORKSPACE_SIZE / sizeof(U32))

HUF_DTABLE_SIZE

#define HUF_DTABLE_SIZE

(

maxTableLog

)

Value:

(1 + (1<<(maxTableLog)))

HUF_H_298734234

#define HUF_H_298734234

HUF_OPTIMAL_DEPTH_THRESHOLD

#define HUF_OPTIMAL_DEPTH_THRESHOLD   ZSTD_btultra

HUF_READ_STATS_WORKSPACE_SIZE

#define HUF_READ_STATS_WORKSPACE_SIZE   (HUF_READ_STATS_WORKSPACE_SIZE_U32 * sizeof(unsigned))

HUF_READ_STATS_WORKSPACE_SIZE_U32

#define HUF_READ_STATS_WORKSPACE_SIZE_U32   FSE_DECOMPRESS_WKSP_SIZE_U32(6, HUF_TABLELOG_MAX-1)

HUF_readStats_wksp() : Same as HUF_readStats() but takes an external workspace which must be 4-byte aligned and its size must be >= HUF_READ_STATS_WORKSPACE_SIZE. If the CPU has BMI2 support, pass bmi2=1, otherwise pass bmi2=0.

HUF_SYMBOLVALUE_MAX

#define HUF_SYMBOLVALUE_MAX   255

HUF_TABLELOG_ABSOLUTEMAX

#define HUF_TABLELOG_ABSOLUTEMAX   12 /* absolute limit of HUF_MAX_TABLELOG. Beyond that value, code does not work */

HUF_TABLELOG_DEFAULT

#define HUF_TABLELOG_DEFAULT   11 /* default tableLog value when none specified */

HUF_TABLELOG_MAX

#define HUF_TABLELOG_MAX   12 /* max runtime value of tableLog (due to static allocation); can be modified up to HUF_TABLELOG_ABSOLUTEMAX */

HUF_WORKSPACE_SIZE

#define HUF_WORKSPACE_SIZE   ((8 << 10) + 512 /* sorting scratch space */)

HUF_WORKSPACE_SIZE_U64

#define HUF_WORKSPACE_SIZE_U64   (HUF_WORKSPACE_SIZE / sizeof(U64))

Typedef Documentation

HUF_CElt

typedef size_t HUF_CElt

HUF_DTable

typedef U32 HUF_DTable

Enumeration Type Documentation

HUF_flags_e

enum HUF_flags_e

Huffman flags bitset. For all flags, 0 is the default value.

Enumerator
HUF_flags_bmi2	If compiled with DYNAMIC_BMI2: Set flag only if the CPU supports BMI2 at runtime. Otherwise: Ignored.
HUF_flags_optimalDepth	If set: Test possible table depths to find the one that produces the smallest header + encoded size. If unset: Use heuristic to find the table depth.
HUF_flags_preferRepeat	If set: If the previous table can encode the input, always reuse the previous table. If unset: If the previous table can encode the input, reuse the previous table if it results in a smaller output.
HUF_flags_suspectUncompressible	If set: Sample the input and check if the sample is uncompressible, if it is then don't attempt to compress. If unset: Always histogram the entire input.
HUF_flags_disableAsm	If set: Don't use assembly implementations If unset: Allow using assembly implementations
HUF_flags_disableFast	If set: Don't use the fast decoding loop, always use the fallback decoding loop. If unset: Use the fast decoding loop when possible.

HUF_repeat

enum HUF_repeat

Enumerator
HUF_repeat_none	Cannot use the previous table
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
HUF_repeat_valid	Can use the previous table and it is assumed to be valid

Function Documentation

HUF_buildCTable_wksp()

size_t HUF_buildCTable_wksp	(	HUF_CElt *	tree,
		const unsigned *	count,
		U32	maxSymbolValue,
		U32	maxNbBits,
		void *	workSpace,
		size_t	wkspSize )

HUF_cardinality()

unsigned HUF_cardinality	(	const unsigned *	count,
		unsigned	maxSymbolValue )

HUF_compress1X_repeat()

size_t HUF_compress1X_repeat	(	void *	dst,
		size_t	dstSize,
		const void *	src,
		size_t	srcSize,
		unsigned	maxSymbolValue,
		unsigned	tableLog,
		void *	workSpace,
		size_t	wkspSize,
		HUF_CElt *	hufTable,
		HUF_repeat *	repeat,
		int	flags )

HUF_compress1X_repeat() : Same as HUF_compress1X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none. If it uses hufTable it does not modify hufTable or repeat. If it doesn't, it sets *repeat = HUF_repeat_none, and it sets hufTable to the table used. If preferRepeat then the old table will always be used if valid. If suspectUncompressible then some sampling checks will be run to potentially skip huffman coding

Parameters

wkspSize

workSpace must be aligned on 4-bytes boundaries, wkspSize must be >= HUF_WORKSPACE_SIZE

HUF_compress1X_usingCTable()

size_t HUF_compress1X_usingCTable	(	void *	dst,
		size_t	dstSize,
		const void *	src,
		size_t	srcSize,
		const HUF_CElt *	CTable,
		int	flags )

HUF_compress4X_repeat()

size_t HUF_compress4X_repeat	(	void *	dst,
		size_t	dstSize,
		const void *	src,
		size_t	srcSize,
		unsigned	maxSymbolValue,
		unsigned	tableLog,
		void *	workSpace,
		size_t	wkspSize,
		HUF_CElt *	hufTable,
		HUF_repeat *	repeat,
		int	flags )

HUF_compress4X_repeat() : Same as HUF_compress4X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none. If it uses hufTable it does not modify hufTable or repeat. If it doesn't, it sets *repeat = HUF_repeat_none, and it sets hufTable to the table used. If preferRepeat then the old table will always be used if valid. If suspectUncompressible then some sampling checks will be run to potentially skip huffman coding

Parameters

wkspSize

workSpace must be aligned on 4-bytes boundaries, wkspSize must be >= HUF_WORKSPACE_SIZE

HUF_compress4X_usingCTable()

size_t HUF_compress4X_usingCTable	(	void *	dst,
		size_t	dstSize,
		const void *	src,
		size_t	srcSize,
		const HUF_CElt *	CTable,
		int	flags )

HUF_compressBound()

size_t HUF_compressBound

(

size_t

size

)

maximum compressed size (worst case)

HUF_decompress1X1_DCtx_wksp()

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,
		int	flags )

HUF_decompress1X2_DCtx_wksp()

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,
		int	flags )

double-symbols decoder

HUF_decompress1X_DCtx_wksp()

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,
		int	flags )

HUF_decompress1X_usingDTable()

size_t HUF_decompress1X_usingDTable	(	void *	dst,
		size_t	maxDstSize,
		const void *	cSrc,
		size_t	cSrcSize,
		const HUF_DTable *	DTable,
		int	flags )

HUF_decompress4X_hufOnly_wksp()

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,
		int	flags )

HUF_decompress4X_usingDTable()

size_t HUF_decompress4X_usingDTable	(	void *	dst,
		size_t	maxDstSize,
		const void *	cSrc,
		size_t	cSrcSize,
		const HUF_DTable *	DTable,
		int	flags )

HUF_estimateCompressedSize()

size_t HUF_estimateCompressedSize	(	const HUF_CElt *	CTable,
		const unsigned *	count,
		unsigned	maxSymbolValue )

HUF_getErrorName()

const char * HUF_getErrorName

(

size_t

code

)

provides error code string (useful for debugging)

HUF_getNbBitsFromCTable()

U32 HUF_getNbBitsFromCTable	(	const HUF_CElt *	symbolTable,
		U32	symbolValue )

HUF_getNbBitsFromCTable() : Read nbBits from CTable symbolTable, for symbol symbolValue presumed <= HUF_SYMBOLVALUE_MAX Note 1 : is not inlined, as HUF_CElt definition is private

HUF_isError()

unsigned HUF_isError

(

size_t

code

)

tells if a return value is an error code

HUF_minTableLog()

unsigned HUF_minTableLog

(

unsigned

symbolCardinality

)

HUF_compress() does the following:

1. count symbol occurrence from source[] into table count[] using FSE_count() (exposed within "fse.h")
2. (optional) refine tableLog using HUF_optimalTableLog()
3. build Huffman table from count using HUF_buildCTable()
4. save Huffman table to memory buffer using HUF_writeCTable()
5. encode the data stream using HUF_compress4X_usingCTable()

The following API allows targeting specific sub-functions for advanced tasks. For example, it's possible to compress several blocks using the same 'CTable', or to save and regenerate 'CTable' using external methods.

HUF_optimalTableLog()

unsigned HUF_optimalTableLog	(	unsigned	maxTableLog,
		size_t	srcSize,
		unsigned	maxSymbolValue,
		void *	workSpace,
		size_t	wkspSize,
		HUF_CElt *	table,
		const unsigned *	count,
		int	flags )

HUF_readCTable()

size_t HUF_readCTable	(	HUF_CElt *	CTable,
		unsigned *	maxSymbolValuePtr,
		const void *	src,
		size_t	srcSize,
		unsigned *	hasZeroWeights )

HUF_readCTable() : Loading a CTable saved with HUF_writeCTable()

HUF_readDTableX1_wksp()

size_t HUF_readDTableX1_wksp	(	HUF_DTable *	DTable,
		const void *	src,
		size_t	srcSize,
		void *	workSpace,
		size_t	wkspSize,
		int	flags )

HUF_readDTableX2_wksp()

size_t HUF_readDTableX2_wksp	(	HUF_DTable *	DTable,
		const void *	src,
		size_t	srcSize,
		void *	workSpace,
		size_t	wkspSize,
		int	flags )

HUF_readStats()

size_t HUF_readStats	(	BYTE *	huffWeight,
		size_t	hwSize,
		U32 *	rankStats,
		U32 *	nbSymbolsPtr,
		U32 *	tableLogPtr,
		const void *	src,
		size_t	srcSize )

HUF_readStats() : Read compact Huffman tree, saved by HUF_writeCTable(). huffWeight is destination buffer.

Returns

: size read from src , or an error Code . Note : Needed by HUF_readCTable() and HUF_readDTableXn() .

HUF_readStats() : Read compact Huffman tree, saved by HUF_writeCTable(). huffWeight is destination buffer. rankStats is assumed to be a table of at least HUF_TABLELOG_MAX U32.

Returns

: size read from src , or an error Code . Note : Needed by HUF_readCTable() and HUF_readDTableX?() .

HUF_readStats_wksp()

size_t HUF_readStats_wksp	(	BYTE *	huffWeight,
		size_t	hwSize,
		U32 *	rankStats,
		U32 *	nbSymbolsPtr,
		U32 *	tableLogPtr,
		const void *	src,
		size_t	srcSize,
		void *	workspace,
		size_t	wkspSize,
		int	flags )

HUF_selectDecoder()

U32 HUF_selectDecoder	(	size_t	dstSize,
		size_t	cSrcSize )

HUF_selectDecoder() : Tells which decoder is likely to decode faster, based on a set of pre-computed metrics.

Returns

: 0==HUF_decompress4X1, 1==HUF_decompress4X2 . Assumption : 0 < dstSize <= 128 KB

HUF_validateCTable()

int HUF_validateCTable	(	const HUF_CElt *	CTable,
		const unsigned *	count,
		unsigned	maxSymbolValue )

HUF_writeCTable_wksp()

size_t HUF_writeCTable_wksp	(	void *	dst,
		size_t	maxDstSize,
		const HUF_CElt *	CTable,
		unsigned	maxSymbolValue,
		unsigned	huffLog,
		void *	workspace,
		size_t	workspaceSize )