xref: /freebsd/sys/contrib/zstd/doc/zstd_manual.html (revision cddbc3b40812213ff00041f79174cac0be360a2a)
1<html>
2<head>
3<meta http-equiv="Content-Type" content="text/html; charset=ISO-8859-1">
4<title>zstd 1.4.0 Manual</title>
5</head>
6<body>
7<h1>zstd 1.4.0 Manual</h1>
8<hr>
9<a name="Contents"></a><h2>Contents</h2>
10<ol>
11<li><a href="#Chapter1">Introduction</a></li>
12<li><a href="#Chapter2">Version</a></li>
13<li><a href="#Chapter3">Default constant</a></li>
14<li><a href="#Chapter4">Constants</a></li>
15<li><a href="#Chapter5">Simple API</a></li>
16<li><a href="#Chapter6">Explicit context</a></li>
17<li><a href="#Chapter7">Advanced compression API</a></li>
18<li><a href="#Chapter8">Advanced decompression API</a></li>
19<li><a href="#Chapter9">Streaming</a></li>
20<li><a href="#Chapter10">Streaming compression - HowTo</a></li>
21<li><a href="#Chapter11">This is a legacy streaming API, and can be replaced by ZSTD_CCtx_reset() and</a></li>
22<li><a href="#Chapter12">Equivalent to:</a></li>
23<li><a href="#Chapter13">Alternative for ZSTD_compressStream2(zcs, output, input, ZSTD_e_continue).</a></li>
24<li><a href="#Chapter14">Equivalent to ZSTD_compressStream2(zcs, output, &emptyInput, ZSTD_e_flush).</a></li>
25<li><a href="#Chapter15">Equivalent to ZSTD_compressStream2(zcs, output, &emptyInput, ZSTD_e_end).</a></li>
26<li><a href="#Chapter16">Streaming decompression - HowTo</a></li>
27<li><a href="#Chapter17">Simple dictionary API</a></li>
28<li><a href="#Chapter18">Bulk processing dictionary API</a></li>
29<li><a href="#Chapter19">Dictionary helper functions</a></li>
30<li><a href="#Chapter20">Advanced dictionary and prefix API</a></li>
31<li><a href="#Chapter21">ADVANCED AND EXPERIMENTAL FUNCTIONS</a></li>
32<li><a href="#Chapter22">experimental API (static linking only)</a></li>
33<li><a href="#Chapter23">Frame size functions</a></li>
34<li><a href="#Chapter24">ZSTD_decompressBound() :</a></li>
35<li><a href="#Chapter25">Memory management</a></li>
36<li><a href="#Chapter26">Advanced compression functions</a></li>
37<li><a href="#Chapter27">Advanced decompression functions</a></li>
38<li><a href="#Chapter28">Advanced streaming functions</a></li>
39<li><a href="#Chapter29">Buffer-less and synchronous inner streaming functions</a></li>
40<li><a href="#Chapter30">Buffer-less streaming compression (synchronous mode)</a></li>
41<li><a href="#Chapter31">Buffer-less streaming decompression (synchronous mode)</a></li>
42<li><a href="#Chapter32">ZSTD_getFrameHeader() :</a></li>
43<li><a href="#Chapter33">Block level API</a></li>
44</ol>
45<hr>
46<a name="Chapter1"></a><h2>Introduction</h2><pre>
47  zstd, short for Zstandard, is a fast lossless compression algorithm, targeting
48  real-time compression scenarios at zlib-level and better compression ratios.
49  The zstd compression library provides in-memory compression and decompression
50  functions.
51
52  The library supports regular compression levels from 1 up to ZSTD_maxCLevel(),
53  which is currently 22. Levels >= 20, labeled `--ultra`, should be used with
54  caution, as they require more memory. The library also offers negative
55  compression levels, which extend the range of speed vs. ratio preferences.
56  The lower the level, the faster the speed (at the cost of compression).
57
58  Compression can be done in:
59    - a single step (described as Simple API)
60    - a single step, reusing a context (described as Explicit context)
61    - unbounded multiple steps (described as Streaming compression)
62
63  The compression ratio achievable on small data can be highly improved using
64  a dictionary. Dictionary compression can be performed in:
65    - a single step (described as Simple dictionary API)
66    - a single step, reusing a dictionary (described as Bulk-processing
67      dictionary API)
68
69  Advanced experimental functions can be accessed using
70  `#define ZSTD_STATIC_LINKING_ONLY` before including zstd.h.
71
72  Advanced experimental APIs should never be used with a dynamically-linked
73  library. They are not "stable"; their definitions or signatures may change in
74  the future. Only static linking is allowed.
75<BR></pre>
76
77<a name="Chapter2"></a><h2>Version</h2><pre></pre>
78
79<pre><b>unsigned ZSTD_versionNumber(void);   </b>/**< to check runtime library version */<b>
80</b></pre><BR>
81<a name="Chapter3"></a><h2>Default constant</h2><pre></pre>
82
83<a name="Chapter4"></a><h2>Constants</h2><pre></pre>
84
85<a name="Chapter5"></a><h2>Simple API</h2><pre></pre>
86
87<pre><b>size_t ZSTD_compress( void* dst, size_t dstCapacity,
88                const void* src, size_t srcSize,
89                      int compressionLevel);
90</b><p>  Compresses `src` content as a single zstd compressed frame into already allocated `dst`.
91  Hint : compression runs faster if `dstCapacity` >=  `ZSTD_compressBound(srcSize)`.
92  @return : compressed size written into `dst` (<= `dstCapacity),
93            or an error code if it fails (which can be tested using ZSTD_isError()).
94</p></pre><BR>
95
96<pre><b>size_t ZSTD_decompress( void* dst, size_t dstCapacity,
97                  const void* src, size_t compressedSize);
98</b><p>  `compressedSize` : must be the _exact_ size of some number of compressed and/or skippable frames.
99  `dstCapacity` is an upper bound of originalSize to regenerate.
100  If user cannot imply a maximum upper bound, it's better to use streaming mode to decompress data.
101  @return : the number of bytes decompressed into `dst` (<= `dstCapacity`),
102            or an errorCode if it fails (which can be tested using ZSTD_isError()).
103</p></pre><BR>
104
105<pre><b>#define ZSTD_CONTENTSIZE_UNKNOWN (0ULL - 1)
106#define ZSTD_CONTENTSIZE_ERROR   (0ULL - 2)
107unsigned long long ZSTD_getFrameContentSize(const void *src, size_t srcSize);
108</b><p>  `src` should point to the start of a ZSTD encoded frame.
109  `srcSize` must be at least as large as the frame header.
110            hint : any size >= `ZSTD_frameHeaderSize_max` is large enough.
111  @return : - decompressed size of `src` frame content, if known
112            - ZSTD_CONTENTSIZE_UNKNOWN if the size cannot be determined
113            - ZSTD_CONTENTSIZE_ERROR if an error occurred (e.g. invalid magic number, srcSize too small)
114   note 1 : a 0 return value means the frame is valid but "empty".
115   note 2 : decompressed size is an optional field, it may not be present, typically in streaming mode.
116            When `return==ZSTD_CONTENTSIZE_UNKNOWN`, data to decompress could be any size.
117            In which case, it's necessary to use streaming mode to decompress data.
118            Optionally, application can rely on some implicit limit,
119            as ZSTD_decompress() only needs an upper bound of decompressed size.
120            (For example, data could be necessarily cut into blocks <= 16 KB).
121   note 3 : decompressed size is always present when compression is completed using single-pass functions,
122            such as ZSTD_compress(), ZSTD_compressCCtx() ZSTD_compress_usingDict() or ZSTD_compress_usingCDict().
123   note 4 : decompressed size can be very large (64-bits value),
124            potentially larger than what local system can handle as a single memory segment.
125            In which case, it's necessary to use streaming mode to decompress data.
126   note 5 : If source is untrusted, decompressed size could be wrong or intentionally modified.
127            Always ensure return value fits within application's authorized limits.
128            Each application can set its own limits.
129   note 6 : This function replaces ZSTD_getDecompressedSize()
130</p></pre><BR>
131
132<pre><b>unsigned long long ZSTD_getDecompressedSize(const void* src, size_t srcSize);
133</b><p>  NOTE: This function is now obsolete, in favor of ZSTD_getFrameContentSize().
134  Both functions work the same way, but ZSTD_getDecompressedSize() blends
135  "empty", "unknown" and "error" results to the same return value (0),
136  while ZSTD_getFrameContentSize() gives them separate return values.
137 @return : decompressed size of `src` frame content _if known and not empty_, 0 otherwise.
138</p></pre><BR>
139
140<pre><b>size_t ZSTD_findFrameCompressedSize(const void* src, size_t srcSize);
141</b><p> `src` should point to the start of a ZSTD frame or skippable frame.
142 `srcSize` must be >= first frame size
143 @return : the compressed size of the first frame starting at `src`,
144           suitable to pass as `srcSize` to `ZSTD_decompress` or similar,
145        or an error code if input is invalid
146</p></pre><BR>
147
148<h3>Helper functions</h3><pre></pre><b><pre>#define ZSTD_COMPRESSBOUND(srcSize)   ((srcSize) + ((srcSize)>>8) + (((srcSize) < (128<<10)) ? (((128<<10) - (srcSize)) >> 11) </b>/* 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 */<b>
149size_t      ZSTD_compressBound(size_t srcSize); </b>/*!< maximum compressed size in worst case single-pass scenario */<b>
150unsigned    ZSTD_isError(size_t code);          </b>/*!< tells if a `size_t` function result is an error code */<b>
151const char* ZSTD_getErrorName(size_t code);     </b>/*!< provides readable string from an error code */<b>
152int         ZSTD_minCLevel(void);               </b>/*!< minimum negative compression level allowed */<b>
153int         ZSTD_maxCLevel(void);               </b>/*!< maximum compression level available */<b>
154</pre></b><BR>
155<a name="Chapter6"></a><h2>Explicit context</h2><pre></pre>
156
157<h3>Compression context</h3><pre>  When compressing many times,
158  it is recommended to allocate a context just once, and re-use it for each successive compression operation.
159  This will make workload friendlier for system's memory.
160  Use one context per thread for parallel execution in multi-threaded environments.
161</pre><b><pre>typedef struct ZSTD_CCtx_s ZSTD_CCtx;
162ZSTD_CCtx* ZSTD_createCCtx(void);
163size_t     ZSTD_freeCCtx(ZSTD_CCtx* cctx);
164</pre></b><BR>
165<pre><b>size_t ZSTD_compressCCtx(ZSTD_CCtx* cctx,
166                         void* dst, size_t dstCapacity,
167                   const void* src, size_t srcSize,
168                         int compressionLevel);
169</b><p>  Same as ZSTD_compress(), using an explicit ZSTD_CCtx
170  The function will compress at requested compression level,
171  ignoring any other parameter
172</p></pre><BR>
173
174<h3>Decompression context</h3><pre>  When decompressing many times,
175  it is recommended to allocate a context only once,
176  and re-use it for each successive compression operation.
177  This will make workload friendlier for system's memory.
178  Use one context per thread for parallel execution.
179</pre><b><pre>typedef struct ZSTD_DCtx_s ZSTD_DCtx;
180ZSTD_DCtx* ZSTD_createDCtx(void);
181size_t     ZSTD_freeDCtx(ZSTD_DCtx* dctx);
182</pre></b><BR>
183<pre><b>size_t ZSTD_decompressDCtx(ZSTD_DCtx* dctx,
184                           void* dst, size_t dstCapacity,
185                     const void* src, size_t srcSize);
186</b><p>  Same as ZSTD_decompress(),
187  requires an allocated ZSTD_DCtx.
188  Compatible with sticky parameters.
189
190</p></pre><BR>
191
192<a name="Chapter7"></a><h2>Advanced compression API</h2><pre></pre>
193
194<pre><b>typedef enum { ZSTD_fast=1,
195               ZSTD_dfast=2,
196               ZSTD_greedy=3,
197               ZSTD_lazy=4,
198               ZSTD_lazy2=5,
199               ZSTD_btlazy2=6,
200               ZSTD_btopt=7,
201               ZSTD_btultra=8,
202               ZSTD_btultra2=9
203               </b>/* note : new strategies _might_ be added in the future.<b>
204                         Only the order (from fast to strong) is guaranteed */
205} ZSTD_strategy;
206</b></pre><BR>
207<pre><b>typedef enum {
208
209    </b>/* compression parameters<b>
210     * Note: When compressing with a ZSTD_CDict these parameters are superseded
211     * by the parameters used to construct the ZSTD_CDict. See ZSTD_CCtx_refCDict()
212     * for more info (superseded-by-cdict). */
213    ZSTD_c_compressionLevel=100, </b>/* Update all compression parameters according to pre-defined cLevel table<b>
214                              * Default level is ZSTD_CLEVEL_DEFAULT==3.
215                              * Special: value 0 means default, which is controlled by ZSTD_CLEVEL_DEFAULT.
216                              * Note 1 : it's possible to pass a negative compression level.
217                              * Note 2 : setting a level sets all default values of other compression parameters */
218    ZSTD_c_windowLog=101,    </b>/* Maximum allowed back-reference distance, expressed as power of 2.<b>
219                              * Must be clamped between ZSTD_WINDOWLOG_MIN and ZSTD_WINDOWLOG_MAX.
220                              * Special: value 0 means "use default windowLog".
221                              * Note: Using a windowLog greater than ZSTD_WINDOWLOG_LIMIT_DEFAULT
222                              *       requires explicitly allowing such window size at decompression stage if using streaming. */
223    ZSTD_c_hashLog=102,      </b>/* Size of the initial probe table, as a power of 2.<b>
224                              * Resulting memory usage is (1 << (hashLog+2)).
225                              * Must be clamped between ZSTD_HASHLOG_MIN and ZSTD_HASHLOG_MAX.
226                              * Larger tables improve compression ratio of strategies <= dFast,
227                              * and improve speed of strategies > dFast.
228                              * Special: value 0 means "use default hashLog". */
229    ZSTD_c_chainLog=103,     </b>/* Size of the multi-probe search table, as a power of 2.<b>
230                              * Resulting memory usage is (1 << (chainLog+2)).
231                              * Must be clamped between ZSTD_CHAINLOG_MIN and ZSTD_CHAINLOG_MAX.
232                              * Larger tables result in better and slower compression.
233                              * This parameter is useless when using "fast" strategy.
234                              * It's still useful when using "dfast" strategy,
235                              * in which case it defines a secondary probe table.
236                              * Special: value 0 means "use default chainLog". */
237    ZSTD_c_searchLog=104,    </b>/* Number of search attempts, as a power of 2.<b>
238                              * More attempts result in better and slower compression.
239                              * This parameter is useless when using "fast" and "dFast" strategies.
240                              * Special: value 0 means "use default searchLog". */
241    ZSTD_c_minMatch=105,     </b>/* Minimum size of searched matches.<b>
242                              * Note that Zstandard can still find matches of smaller size,
243                              * it just tweaks its search algorithm to look for this size and larger.
244                              * Larger values increase compression and decompression speed, but decrease ratio.
245                              * Must be clamped between ZSTD_MINMATCH_MIN and ZSTD_MINMATCH_MAX.
246                              * Note that currently, for all strategies < btopt, effective minimum is 4.
247                              *                    , for all strategies > fast, effective maximum is 6.
248                              * Special: value 0 means "use default minMatchLength". */
249    ZSTD_c_targetLength=106, </b>/* Impact of this field depends on strategy.<b>
250                              * For strategies btopt, btultra & btultra2:
251                              *     Length of Match considered "good enough" to stop search.
252                              *     Larger values make compression stronger, and slower.
253                              * For strategy fast:
254                              *     Distance between match sampling.
255                              *     Larger values make compression faster, and weaker.
256                              * Special: value 0 means "use default targetLength". */
257    ZSTD_c_strategy=107,     </b>/* See ZSTD_strategy enum definition.<b>
258                              * The higher the value of selected strategy, the more complex it is,
259                              * resulting in stronger and slower compression.
260                              * Special: value 0 means "use default strategy". */
261
262    </b>/* LDM mode parameters */<b>
263    ZSTD_c_enableLongDistanceMatching=160, </b>/* Enable long distance matching.<b>
264                                     * This parameter is designed to improve compression ratio
265                                     * for large inputs, by finding large matches at long distance.
266                                     * It increases memory usage and window size.
267                                     * Note: enabling this parameter increases default ZSTD_c_windowLog to 128 MB
268                                     * except when expressly set to a different value. */
269    ZSTD_c_ldmHashLog=161,   </b>/* Size of the table for long distance matching, as a power of 2.<b>
270                              * Larger values increase memory usage and compression ratio,
271                              * but decrease compression speed.
272                              * Must be clamped between ZSTD_HASHLOG_MIN and ZSTD_HASHLOG_MAX
273                              * default: windowlog - 7.
274                              * Special: value 0 means "automatically determine hashlog". */
275    ZSTD_c_ldmMinMatch=162,  </b>/* Minimum match size for long distance matcher.<b>
276                              * Larger/too small values usually decrease compression ratio.
277                              * Must be clamped between ZSTD_LDM_MINMATCH_MIN and ZSTD_LDM_MINMATCH_MAX.
278                              * Special: value 0 means "use default value" (default: 64). */
279    ZSTD_c_ldmBucketSizeLog=163, </b>/* Log size of each bucket in the LDM hash table for collision resolution.<b>
280                              * Larger values improve collision resolution but decrease compression speed.
281                              * The maximum value is ZSTD_LDM_BUCKETSIZELOG_MAX.
282                              * Special: value 0 means "use default value" (default: 3). */
283    ZSTD_c_ldmHashRateLog=164, </b>/* Frequency of inserting/looking up entries into the LDM hash table.<b>
284                              * Must be clamped between 0 and (ZSTD_WINDOWLOG_MAX - ZSTD_HASHLOG_MIN).
285                              * Default is MAX(0, (windowLog - ldmHashLog)), optimizing hash table usage.
286                              * Larger values improve compression speed.
287                              * Deviating far from default value will likely result in a compression ratio decrease.
288                              * Special: value 0 means "automatically determine hashRateLog". */
289
290    </b>/* frame parameters */<b>
291    ZSTD_c_contentSizeFlag=200, </b>/* Content size will be written into frame header _whenever known_ (default:1)<b>
292                              * Content size must be known at the beginning of compression.
293                              * This is automatically the case when using ZSTD_compress2(),
294                              * For streaming variants, content size must be provided with ZSTD_CCtx_setPledgedSrcSize() */
295    ZSTD_c_checksumFlag=201, </b>/* A 32-bits checksum of content is written at end of frame (default:0) */<b>
296    ZSTD_c_dictIDFlag=202,   </b>/* When applicable, dictionary's ID is written into frame header (default:1) */<b>
297
298    </b>/* multi-threading parameters */<b>
299    </b>/* These parameters are only useful if multi-threading is enabled (compiled with build macro ZSTD_MULTITHREAD).<b>
300     * They return an error otherwise. */
301    ZSTD_c_nbWorkers=400,    </b>/* Select how many threads will be spawned to compress in parallel.<b>
302                              * When nbWorkers >= 1, triggers asynchronous mode when used with ZSTD_compressStream*() :
303                              * ZSTD_compressStream*() consumes input and flush output if possible, but immediately gives back control to caller,
304                              * while compression work is performed in parallel, within worker threads.
305                              * (note : a strong exception to this rule is when first invocation of ZSTD_compressStream2() sets ZSTD_e_end :
306                              *  in which case, ZSTD_compressStream2() delegates to ZSTD_compress2(), which is always a blocking call).
307                              * More workers improve speed, but also increase memory usage.
308                              * Default value is `0`, aka "single-threaded mode" : no worker is spawned, compression is performed inside Caller's thread, all invocations are blocking */
309    ZSTD_c_jobSize=401,      </b>/* Size of a compression job. This value is enforced only when nbWorkers >= 1.<b>
310                              * Each compression job is completed in parallel, so this value can indirectly impact the nb of active threads.
311                              * 0 means default, which is dynamically determined based on compression parameters.
312                              * Job size must be a minimum of overlap size, or 1 MB, whichever is largest.
313                              * The minimum size is automatically and transparently enforced */
314    ZSTD_c_overlapLog=402,   </b>/* Control the overlap size, as a fraction of window size.<b>
315                              * The overlap size is an amount of data reloaded from previous job at the beginning of a new job.
316                              * It helps preserve compression ratio, while each job is compressed in parallel.
317                              * This value is enforced only when nbWorkers >= 1.
318                              * Larger values increase compression ratio, but decrease speed.
319                              * Possible values range from 0 to 9 :
320                              * - 0 means "default" : value will be determined by the library, depending on strategy
321                              * - 1 means "no overlap"
322                              * - 9 means "full overlap", using a full window size.
323                              * Each intermediate rank increases/decreases load size by a factor 2 :
324                              * 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
325                              * default value varies between 6 and 9, depending on strategy */
326
327    </b>/* note : additional experimental parameters are also available<b>
328     * within the experimental section of the API.
329     * At the time of this writing, they include :
330     * ZSTD_c_rsyncable
331     * ZSTD_c_format
332     * ZSTD_c_forceMaxWindow
333     * ZSTD_c_forceAttachDict
334     * ZSTD_c_literalCompressionMode
335     * Because they are not stable, it's necessary to define ZSTD_STATIC_LINKING_ONLY to access them.
336     * note : never ever use experimentalParam? names directly;
337     *        also, the enums values themselves are unstable and can still change.
338     */
339     ZSTD_c_experimentalParam1=500,
340     ZSTD_c_experimentalParam2=10,
341     ZSTD_c_experimentalParam3=1000,
342     ZSTD_c_experimentalParam4=1001,
343     ZSTD_c_experimentalParam5=1002,
344} ZSTD_cParameter;
345</b></pre><BR>
346<pre><b>typedef struct {
347    size_t error;
348    int lowerBound;
349    int upperBound;
350} ZSTD_bounds;
351</b></pre><BR>
352<pre><b>ZSTD_bounds ZSTD_cParam_getBounds(ZSTD_cParameter cParam);
353</b><p>  All parameters must belong to an interval with lower and upper bounds,
354  otherwise they will either trigger an error or be automatically clamped.
355 @return : a structure, ZSTD_bounds, which contains
356         - an error status field, which must be tested using ZSTD_isError()
357         - lower and upper bounds, both inclusive
358
359</p></pre><BR>
360
361<pre><b>size_t ZSTD_CCtx_setParameter(ZSTD_CCtx* cctx, ZSTD_cParameter param, int value);
362</b><p>  Set one compression parameter, selected by enum ZSTD_cParameter.
363  All parameters have valid bounds. Bounds can be queried using ZSTD_cParam_getBounds().
364  Providing a value beyond bound will either clamp it, or trigger an error (depending on parameter).
365  Setting a parameter is generally only possible during frame initialization (before starting compression).
366  Exception : when using multi-threading mode (nbWorkers >= 1),
367              the following parameters can be updated _during_ compression (within same frame):
368              => compressionLevel, hashLog, chainLog, searchLog, minMatch, targetLength and strategy.
369              new parameters will be active for next job only (after a flush()).
370 @return : an error code (which can be tested using ZSTD_isError()).
371
372</p></pre><BR>
373
374<pre><b>size_t ZSTD_CCtx_setPledgedSrcSize(ZSTD_CCtx* cctx, unsigned long long pledgedSrcSize);
375</b><p>  Total input data size to be compressed as a single frame.
376  Value will be written in frame header, unless if explicitly forbidden using ZSTD_c_contentSizeFlag.
377  This value will also be controlled at end of frame, and trigger an error if not respected.
378 @result : 0, or an error code (which can be tested with ZSTD_isError()).
379  Note 1 : pledgedSrcSize==0 actually means zero, aka an empty frame.
380           In order to mean "unknown content size", pass constant ZSTD_CONTENTSIZE_UNKNOWN.
381           ZSTD_CONTENTSIZE_UNKNOWN is default value for any new frame.
382  Note 2 : pledgedSrcSize is only valid once, for the next frame.
383           It's discarded at the end of the frame, and replaced by ZSTD_CONTENTSIZE_UNKNOWN.
384  Note 3 : Whenever all input data is provided and consumed in a single round,
385           for example with ZSTD_compress2(),
386           or invoking immediately ZSTD_compressStream2(,,,ZSTD_e_end),
387           this value is automatically overridden by srcSize instead.
388
389</p></pre><BR>
390
391<pre><b>typedef enum {
392    ZSTD_reset_session_only = 1,
393    ZSTD_reset_parameters = 2,
394    ZSTD_reset_session_and_parameters = 3
395} ZSTD_ResetDirective;
396</b></pre><BR>
397<pre><b>size_t ZSTD_CCtx_reset(ZSTD_CCtx* cctx, ZSTD_ResetDirective reset);
398</b><p>  There are 2 different things that can be reset, independently or jointly :
399  - The session : will stop compressing current frame, and make CCtx ready to start a new one.
400                  Useful after an error, or to interrupt any ongoing compression.
401                  Any internal data not yet flushed is cancelled.
402                  Compression parameters and dictionary remain unchanged.
403                  They will be used to compress next frame.
404                  Resetting session never fails.
405  - The parameters : changes all parameters back to "default".
406                  This removes any reference to any dictionary too.
407                  Parameters can only be changed between 2 sessions (i.e. no compression is currently ongoing)
408                  otherwise the reset fails, and function returns an error value (which can be tested using ZSTD_isError())
409  - Both : similar to resetting the session, followed by resetting parameters.
410
411</p></pre><BR>
412
413<pre><b>size_t ZSTD_compress2( ZSTD_CCtx* cctx,
414                       void* dst, size_t dstCapacity,
415                 const void* src, size_t srcSize);
416</b><p>  Behave the same as ZSTD_compressCCtx(), but compression parameters are set using the advanced API.
417  ZSTD_compress2() always starts a new frame.
418  Should cctx hold data from a previously unfinished frame, everything about it is forgotten.
419  - Compression parameters are pushed into CCtx before starting compression, using ZSTD_CCtx_set*()
420  - The function is always blocking, returns when compression is completed.
421  Hint : compression runs faster if `dstCapacity` >=  `ZSTD_compressBound(srcSize)`.
422 @return : compressed size written into `dst` (<= `dstCapacity),
423           or an error code if it fails (which can be tested using ZSTD_isError()).
424
425</p></pre><BR>
426
427<a name="Chapter8"></a><h2>Advanced decompression API</h2><pre></pre>
428
429<pre><b>typedef enum {
430
431    ZSTD_d_windowLogMax=100, </b>/* Select a size limit (in power of 2) beyond which<b>
432                              * the streaming API will refuse to allocate memory buffer
433                              * in order to protect the host from unreasonable memory requirements.
434                              * This parameter is only useful in streaming mode, since no internal buffer is allocated in single-pass mode.
435                              * By default, a decompression context accepts window sizes <= (1 << ZSTD_WINDOWLOG_LIMIT_DEFAULT).
436                              * Special: value 0 means "use default maximum windowLog". */
437
438    </b>/* note : additional experimental parameters are also available<b>
439     * within the experimental section of the API.
440     * At the time of this writing, they include :
441     * ZSTD_c_format
442     * Because they are not stable, it's necessary to define ZSTD_STATIC_LINKING_ONLY to access them.
443     * note : never ever use experimentalParam? names directly
444     */
445     ZSTD_d_experimentalParam1=1000
446
447} ZSTD_dParameter;
448</b></pre><BR>
449<pre><b>ZSTD_bounds ZSTD_dParam_getBounds(ZSTD_dParameter dParam);
450</b><p>  All parameters must belong to an interval with lower and upper bounds,
451  otherwise they will either trigger an error or be automatically clamped.
452 @return : a structure, ZSTD_bounds, which contains
453         - an error status field, which must be tested using ZSTD_isError()
454         - both lower and upper bounds, inclusive
455
456</p></pre><BR>
457
458<pre><b>size_t ZSTD_DCtx_setParameter(ZSTD_DCtx* dctx, ZSTD_dParameter param, int value);
459</b><p>  Set one compression parameter, selected by enum ZSTD_dParameter.
460  All parameters have valid bounds. Bounds can be queried using ZSTD_dParam_getBounds().
461  Providing a value beyond bound will either clamp it, or trigger an error (depending on parameter).
462  Setting a parameter is only possible during frame initialization (before starting decompression).
463 @return : 0, or an error code (which can be tested using ZSTD_isError()).
464
465</p></pre><BR>
466
467<pre><b>size_t ZSTD_DCtx_reset(ZSTD_DCtx* dctx, ZSTD_ResetDirective reset);
468</b><p>  Return a DCtx to clean state.
469  Session and parameters can be reset jointly or separately.
470  Parameters can only be reset when no active frame is being decompressed.
471 @return : 0, or an error code, which can be tested with ZSTD_isError()
472
473</p></pre><BR>
474
475<a name="Chapter9"></a><h2>Streaming</h2><pre></pre>
476
477<pre><b>typedef struct ZSTD_inBuffer_s {
478  const void* src;    </b>/**< start of input buffer */<b>
479  size_t size;        </b>/**< size of input buffer */<b>
480  size_t pos;         </b>/**< position where reading stopped. Will be updated. Necessarily 0 <= pos <= size */<b>
481} ZSTD_inBuffer;
482</b></pre><BR>
483<pre><b>typedef struct ZSTD_outBuffer_s {
484  void*  dst;         </b>/**< start of output buffer */<b>
485  size_t size;        </b>/**< size of output buffer */<b>
486  size_t pos;         </b>/**< position where writing stopped. Will be updated. Necessarily 0 <= pos <= size */<b>
487} ZSTD_outBuffer;
488</b></pre><BR>
489<a name="Chapter10"></a><h2>Streaming compression - HowTo</h2><pre>
490  A ZSTD_CStream object is required to track streaming operation.
491  Use ZSTD_createCStream() and ZSTD_freeCStream() to create/release resources.
492  ZSTD_CStream objects can be reused multiple times on consecutive compression operations.
493  It is recommended to re-use ZSTD_CStream since it will play nicer with system's memory, by re-using already allocated memory.
494
495  For parallel execution, use one separate ZSTD_CStream per thread.
496
497  note : since v1.3.0, ZSTD_CStream and ZSTD_CCtx are the same thing.
498
499  Parameters are sticky : when starting a new compression on the same context,
500  it will re-use the same sticky parameters as previous compression session.
501  When in doubt, it's recommended to fully initialize the context before usage.
502  Use ZSTD_CCtx_reset() to reset the context and ZSTD_CCtx_setParameter(),
503  ZSTD_CCtx_setPledgedSrcSize(), or ZSTD_CCtx_loadDictionary() and friends to
504  set more specific parameters, the pledged source size, or load a dictionary.
505
506  Use ZSTD_compressStream2() with ZSTD_e_continue as many times as necessary to
507  consume input stream. The function will automatically update both `pos`
508  fields within `input` and `output`.
509  Note that the function may not consume the entire input, for example, because
510  the output buffer is already full, in which case `input.pos < input.size`.
511  The caller must check if input has been entirely consumed.
512  If not, the caller must make some room to receive more compressed data,
513  and then present again remaining input data.
514  note: ZSTD_e_continue is guaranteed to make some forward progress when called,
515        but doesn't guarantee maximal forward progress. This is especially relevant
516        when compressing with multiple threads. The call won't block if it can
517        consume some input, but if it can't it will wait for some, but not all,
518        output to be flushed.
519 @return : provides a minimum amount of data remaining to be flushed from internal buffers
520           or an error code, which can be tested using ZSTD_isError().
521
522  At any moment, it's possible to flush whatever data might remain stuck within internal buffer,
523  using ZSTD_compressStream2() with ZSTD_e_flush. `output->pos` will be updated.
524  Note that, if `output->size` is too small, a single invocation with ZSTD_e_flush might not be enough (return code > 0).
525  In which case, make some room to receive more compressed data, and call again ZSTD_compressStream2() with ZSTD_e_flush.
526  You must continue calling ZSTD_compressStream2() with ZSTD_e_flush until it returns 0, at which point you can change the
527  operation.
528  note: ZSTD_e_flush will flush as much output as possible, meaning when compressing with multiple threads, it will
529        block until the flush is complete or the output buffer is full.
530  @return : 0 if internal buffers are entirely flushed,
531            >0 if some data still present within internal buffer (the value is minimal estimation of remaining size),
532            or an error code, which can be tested using ZSTD_isError().
533
534  Calling ZSTD_compressStream2() with ZSTD_e_end instructs to finish a frame.
535  It will perform a flush and write frame epilogue.
536  The epilogue is required for decoders to consider a frame completed.
537  flush operation is the same, and follows same rules as calling ZSTD_compressStream2() with ZSTD_e_flush.
538  You must continue calling ZSTD_compressStream2() with ZSTD_e_end until it returns 0, at which point you are free to
539  start a new frame.
540  note: ZSTD_e_end will flush as much output as possible, meaning when compressing with multiple threads, it will
541        block until the flush is complete or the output buffer is full.
542  @return : 0 if frame fully completed and fully flushed,
543            >0 if some data still present within internal buffer (the value is minimal estimation of remaining size),
544            or an error code, which can be tested using ZSTD_isError().
545
546
547<BR></pre>
548
549<pre><b>typedef ZSTD_CCtx ZSTD_CStream;  </b>/**< CCtx and CStream are now effectively same object (>= v1.3.0) */<b>
550</b></pre><BR>
551<h3>ZSTD_CStream management functions</h3><pre></pre><b><pre>ZSTD_CStream* ZSTD_createCStream(void);
552size_t ZSTD_freeCStream(ZSTD_CStream* zcs);
553</pre></b><BR>
554<h3>Streaming compression functions</h3><pre></pre><b><pre>typedef enum {
555    ZSTD_e_continue=0, </b>/* collect more data, encoder decides when to output compressed result, for optimal compression ratio */<b>
556    ZSTD_e_flush=1,    </b>/* flush any data provided so far,<b>
557                        * it creates (at least) one new block, that can be decoded immediately on reception;
558                        * frame will continue: any future data can still reference previously compressed data, improving compression.
559                        * note : multithreaded compression will block to flush as much output as possible. */
560    ZSTD_e_end=2       </b>/* flush any remaining data _and_ close current frame.<b>
561                        * note that frame is only closed after compressed data is fully flushed (return value == 0).
562                        * After that point, any additional data starts a new frame.
563                        * note : each frame is independent (does not reference any content from previous frame).
564                        : note : multithreaded compression will block to flush as much output as possible. */
565} ZSTD_EndDirective;
566</pre></b><BR>
567<pre><b>size_t ZSTD_compressStream2( ZSTD_CCtx* cctx,
568                             ZSTD_outBuffer* output,
569                             ZSTD_inBuffer* input,
570                             ZSTD_EndDirective endOp);
571</b><p>  Behaves about the same as ZSTD_compressStream, with additional control on end directive.
572  - Compression parameters are pushed into CCtx before starting compression, using ZSTD_CCtx_set*()
573  - Compression parameters cannot be changed once compression is started (save a list of exceptions in multi-threading mode)
574  - output->pos must be <= dstCapacity, input->pos must be <= srcSize
575  - output->pos and input->pos will be updated. They are guaranteed to remain below their respective limit.
576  - When nbWorkers==0 (default), function is blocking : it completes its job before returning to caller.
577  - 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,
578                                                  and then immediately returns, just indicating that there is some data remaining to be flushed.
579                                                  The function nonetheless guarantees forward progress : it will return only after it reads or write at least 1+ byte.
580  - 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.
581  - @return provides a minimum amount of data remaining to be flushed from internal buffers
582            or an error code, which can be tested using ZSTD_isError().
583            if @return != 0, flush is not fully completed, there is still some data left within internal buffers.
584            This is useful for ZSTD_e_flush, since in this case more flushes are necessary to empty all buffers.
585            For ZSTD_e_end, @return == 0 when internal buffers are fully flushed and frame is completed.
586  - after a ZSTD_e_end directive, if internal buffer is not fully flushed (@return != 0),
587            only ZSTD_e_end or ZSTD_e_flush operations are allowed.
588            Before starting a new compression job, or changing compression parameters,
589            it is required to fully flush internal buffers.
590
591</p></pre><BR>
592
593<pre><b>size_t ZSTD_CStreamInSize(void);    </b>/**< recommended size for input buffer */<b>
594</b></pre><BR>
595<pre><b>size_t ZSTD_CStreamOutSize(void);   </b>/**< recommended size for output buffer. Guarantee to successfully flush at least one complete compressed block in all circumstances. */<b>
596</b></pre><BR>
597<a name="Chapter11"></a><h2>This is a legacy streaming API, and can be replaced by ZSTD_CCtx_reset() and</h2><pre> ZSTD_compressStream2(). It is redundent, but is still fully supported.
598 Advanced parameters and dictionary compression can only be used through the
599 new API.
600<BR></pre>
601
602<a name="Chapter12"></a><h2>Equivalent to:</h2><pre>
603     ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only);
604     ZSTD_CCtx_refCDict(zcs, NULL); // clear the dictionary (if any)
605     ZSTD_CCtx_setParameter(zcs, ZSTD_c_compressionLevel, compressionLevel);
606
607<BR></pre>
608
609<a name="Chapter13"></a><h2>Alternative for ZSTD_compressStream2(zcs, output, input, ZSTD_e_continue).</h2><pre> NOTE: The return value is different. ZSTD_compressStream() returns a hint for
610 the next read size (if non-zero and not an error). ZSTD_compressStream2()
611 returns the number of bytes left to flush (if non-zero and not an error).
612
613<BR></pre>
614
615<a name="Chapter14"></a><h2>Equivalent to ZSTD_compressStream2(zcs, output, &emptyInput, ZSTD_e_flush).</h2><pre></pre>
616
617<a name="Chapter15"></a><h2>Equivalent to ZSTD_compressStream2(zcs, output, &emptyInput, ZSTD_e_end).</h2><pre></pre>
618
619<a name="Chapter16"></a><h2>Streaming decompression - HowTo</h2><pre>
620  A ZSTD_DStream object is required to track streaming operations.
621  Use ZSTD_createDStream() and ZSTD_freeDStream() to create/release resources.
622  ZSTD_DStream objects can be re-used multiple times.
623
624  Use ZSTD_initDStream() to start a new decompression operation.
625 @return : recommended first input size
626  Alternatively, use advanced API to set specific properties.
627
628  Use ZSTD_decompressStream() repetitively to consume your input.
629  The function will update both `pos` fields.
630  If `input.pos < input.size`, some input has not been consumed.
631  It's up to the caller to present again remaining data.
632  The function tries to flush all data decoded immediately, respecting output buffer size.
633  If `output.pos < output.size`, decoder has flushed everything it could.
634  But if `output.pos == output.size`, there might be some data left within internal buffers.,
635  In which case, call ZSTD_decompressStream() again to flush whatever remains in the buffer.
636  Note : with no additional input provided, amount of data flushed is necessarily <= ZSTD_BLOCKSIZE_MAX.
637 @return : 0 when a frame is completely decoded and fully flushed,
638        or an error code, which can be tested using ZSTD_isError(),
639        or any other value > 0, which means there is still some decoding or flushing to do to complete current frame :
640                                the return value is a suggested next input size (just a hint for better latency)
641                                that will never request more than the remaining frame size.
642
643<BR></pre>
644
645<pre><b>typedef ZSTD_DCtx ZSTD_DStream;  </b>/**< DCtx and DStream are now effectively same object (>= v1.3.0) */<b>
646</b></pre><BR>
647<h3>ZSTD_DStream management functions</h3><pre></pre><b><pre>ZSTD_DStream* ZSTD_createDStream(void);
648size_t ZSTD_freeDStream(ZSTD_DStream* zds);
649</pre></b><BR>
650<h3>Streaming decompression functions</h3><pre></pre><b><pre>size_t ZSTD_initDStream(ZSTD_DStream* zds);
651size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inBuffer* input);
652</pre></b><BR>
653<pre><b>size_t ZSTD_DStreamInSize(void);    </b>/*!< recommended size for input buffer */<b>
654</b></pre><BR>
655<pre><b>size_t ZSTD_DStreamOutSize(void);   </b>/*!< recommended size for output buffer. Guarantee to successfully flush at least one complete block in all circumstances. */<b>
656</b></pre><BR>
657<a name="Chapter17"></a><h2>Simple dictionary API</h2><pre></pre>
658
659<pre><b>size_t ZSTD_compress_usingDict(ZSTD_CCtx* ctx,
660                               void* dst, size_t dstCapacity,
661                         const void* src, size_t srcSize,
662                         const void* dict,size_t dictSize,
663                               int compressionLevel);
664</b><p>  Compression at an explicit compression level using a Dictionary.
665  A dictionary can be any arbitrary data segment (also called a prefix),
666  or a buffer with specified information (see dictBuilder/zdict.h).
667  Note : This function loads the dictionary, resulting in significant startup delay.
668         It's intended for a dictionary used only once.
669  Note 2 : When `dict == NULL || dictSize < 8` no dictionary is used.
670</p></pre><BR>
671
672<pre><b>size_t ZSTD_decompress_usingDict(ZSTD_DCtx* dctx,
673                                 void* dst, size_t dstCapacity,
674                           const void* src, size_t srcSize,
675                           const void* dict,size_t dictSize);
676</b><p>  Decompression using a known Dictionary.
677  Dictionary must be identical to the one used during compression.
678  Note : This function loads the dictionary, resulting in significant startup delay.
679         It's intended for a dictionary used only once.
680  Note : When `dict == NULL || dictSize < 8` no dictionary is used.
681</p></pre><BR>
682
683<a name="Chapter18"></a><h2>Bulk processing dictionary API</h2><pre></pre>
684
685<pre><b>ZSTD_CDict* ZSTD_createCDict(const void* dictBuffer, size_t dictSize,
686                             int compressionLevel);
687</b><p>  When compressing multiple messages / blocks using the same dictionary, it's recommended to load it only once.
688  ZSTD_createCDict() will create a digested dictionary, ready to start future compression operations without startup cost.
689  ZSTD_CDict can be created once and shared by multiple threads concurrently, since its usage is read-only.
690 `dictBuffer` can be released after ZSTD_CDict creation, because its content is copied within CDict.
691  Consider experimental function `ZSTD_createCDict_byReference()` if you prefer to not duplicate `dictBuffer` content.
692  Note : A ZSTD_CDict can be created from an empty dictBuffer, but it is inefficient when used to compress small data.
693</p></pre><BR>
694
695<pre><b>size_t      ZSTD_freeCDict(ZSTD_CDict* CDict);
696</b><p>  Function frees memory allocated by ZSTD_createCDict().
697</p></pre><BR>
698
699<pre><b>size_t ZSTD_compress_usingCDict(ZSTD_CCtx* cctx,
700                                void* dst, size_t dstCapacity,
701                          const void* src, size_t srcSize,
702                          const ZSTD_CDict* cdict);
703</b><p>  Compression using a digested Dictionary.
704  Recommended when same dictionary is used multiple times.
705  Note : compression level is _decided at dictionary creation time_,
706     and frame parameters are hardcoded (dictID=yes, contentSize=yes, checksum=no)
707</p></pre><BR>
708
709<pre><b>ZSTD_DDict* ZSTD_createDDict(const void* dictBuffer, size_t dictSize);
710</b><p>  Create a digested dictionary, ready to start decompression operation without startup delay.
711  dictBuffer can be released after DDict creation, as its content is copied inside DDict.
712</p></pre><BR>
713
714<pre><b>size_t      ZSTD_freeDDict(ZSTD_DDict* ddict);
715</b><p>  Function frees memory allocated with ZSTD_createDDict()
716</p></pre><BR>
717
718<pre><b>size_t ZSTD_decompress_usingDDict(ZSTD_DCtx* dctx,
719                                  void* dst, size_t dstCapacity,
720                            const void* src, size_t srcSize,
721                            const ZSTD_DDict* ddict);
722</b><p>  Decompression using a digested Dictionary.
723  Recommended when same dictionary is used multiple times.
724</p></pre><BR>
725
726<a name="Chapter19"></a><h2>Dictionary helper functions</h2><pre></pre>
727
728<pre><b>unsigned ZSTD_getDictID_fromDict(const void* dict, size_t dictSize);
729</b><p>  Provides the dictID stored within dictionary.
730  if @return == 0, the dictionary is not conformant with Zstandard specification.
731  It can still be loaded, but as a content-only dictionary.
732</p></pre><BR>
733
734<pre><b>unsigned ZSTD_getDictID_fromDDict(const ZSTD_DDict* ddict);
735</b><p>  Provides the dictID of the dictionary loaded into `ddict`.
736  If @return == 0, the dictionary is not conformant to Zstandard specification, or empty.
737  Non-conformant dictionaries can still be loaded, but as content-only dictionaries.
738</p></pre><BR>
739
740<pre><b>unsigned ZSTD_getDictID_fromFrame(const void* src, size_t srcSize);
741</b><p>  Provides the dictID required to decompressed the frame stored within `src`.
742  If @return == 0, the dictID could not be decoded.
743  This could for one of the following reasons :
744  - The frame does not require a dictionary to be decoded (most common case).
745  - The frame was built with dictID intentionally removed. Whatever dictionary is necessary is a hidden information.
746    Note : this use case also happens when using a non-conformant dictionary.
747  - `srcSize` is too small, and as a result, the frame header could not be decoded (only possible if `srcSize < ZSTD_FRAMEHEADERSIZE_MAX`).
748  - This is not a Zstandard frame.
749  When identifying the exact failure cause, it's possible to use ZSTD_getFrameHeader(), which will provide a more precise error code.
750</p></pre><BR>
751
752<a name="Chapter20"></a><h2>Advanced dictionary and prefix API</h2><pre>
753 This API allows dictionaries to be used with ZSTD_compress2(),
754 ZSTD_compressStream2(), and ZSTD_decompress(). Dictionaries are sticky, and
755 only reset with the context is reset with ZSTD_reset_parameters or
756 ZSTD_reset_session_and_parameters. Prefixes are single-use.
757<BR></pre>
758
759<pre><b>size_t ZSTD_CCtx_loadDictionary(ZSTD_CCtx* cctx, const void* dict, size_t dictSize);
760</b><p>  Create an internal CDict from `dict` buffer.
761  Decompression will have to use same dictionary.
762 @result : 0, or an error code (which can be tested with ZSTD_isError()).
763  Special: Loading a NULL (or 0-size) dictionary invalidates previous dictionary,
764           meaning "return to no-dictionary mode".
765  Note 1 : Dictionary is sticky, it will be used for all future compressed frames.
766           To return to "no-dictionary" situation, load a NULL dictionary (or reset parameters).
767  Note 2 : Loading a dictionary involves building tables.
768           It's also a CPU consuming operation, with non-negligible impact on latency.
769           Tables are dependent on compression parameters, and for this reason,
770           compression parameters can no longer be changed after loading a dictionary.
771  Note 3 :`dict` content will be copied internally.
772           Use experimental ZSTD_CCtx_loadDictionary_byReference() to reference content instead.
773           In such a case, dictionary buffer must outlive its users.
774  Note 4 : Use ZSTD_CCtx_loadDictionary_advanced()
775           to precisely select how dictionary content must be interpreted.
776</p></pre><BR>
777
778<pre><b>size_t ZSTD_CCtx_refCDict(ZSTD_CCtx* cctx, const ZSTD_CDict* cdict);
779</b><p>  Reference a prepared dictionary, to be used for all next compressed frames.
780  Note that compression parameters are enforced from within CDict,
781  and supersede any compression parameter previously set within CCtx.
782  The parameters ignored are labled as "superseded-by-cdict" in the ZSTD_cParameter enum docs.
783  The ignored parameters will be used again if the CCtx is returned to no-dictionary mode.
784  The dictionary will remain valid for future compressed frames using same CCtx.
785 @result : 0, or an error code (which can be tested with ZSTD_isError()).
786  Special : Referencing a NULL CDict means "return to no-dictionary mode".
787  Note 1 : Currently, only one dictionary can be managed.
788           Referencing a new dictionary effectively "discards" any previous one.
789  Note 2 : CDict is just referenced, its lifetime must outlive its usage within CCtx.
790</p></pre><BR>
791
792<pre><b>size_t ZSTD_CCtx_refPrefix(ZSTD_CCtx* cctx,
793                     const void* prefix, size_t prefixSize);
794</b><p>  Reference a prefix (single-usage dictionary) for next compressed frame.
795  A prefix is **only used once**. Tables are discarded at end of frame (ZSTD_e_end).
796  Decompression will need same prefix to properly regenerate data.
797  Compressing with a prefix is similar in outcome as performing a diff and compressing it,
798  but performs much faster, especially during decompression (compression speed is tunable with compression level).
799 @result : 0, or an error code (which can be tested with ZSTD_isError()).
800  Special: Adding any prefix (including NULL) invalidates any previous prefix or dictionary
801  Note 1 : Prefix buffer is referenced. It **must** outlive compression.
802           Its content must remain unmodified during compression.
803  Note 2 : If the intention is to diff some large src data blob with some prior version of itself,
804           ensure that the window size is large enough to contain the entire source.
805           See ZSTD_c_windowLog.
806  Note 3 : Referencing a prefix involves building tables, which are dependent on compression parameters.
807           It's a CPU consuming operation, with non-negligible impact on latency.
808           If there is a need to use the same prefix multiple times, consider loadDictionary instead.
809  Note 4 : By default, the prefix is interpreted as raw content (ZSTD_dm_rawContent).
810           Use experimental ZSTD_CCtx_refPrefix_advanced() to alter dictionary interpretation.
811</p></pre><BR>
812
813<pre><b>size_t ZSTD_DCtx_loadDictionary(ZSTD_DCtx* dctx, const void* dict, size_t dictSize);
814</b><p>  Create an internal DDict from dict buffer,
815  to be used to decompress next frames.
816  The dictionary remains valid for all future frames, until explicitly invalidated.
817 @result : 0, or an error code (which can be tested with ZSTD_isError()).
818  Special : Adding a NULL (or 0-size) dictionary invalidates any previous dictionary,
819            meaning "return to no-dictionary mode".
820  Note 1 : Loading a dictionary involves building tables,
821           which has a non-negligible impact on CPU usage and latency.
822           It's recommended to "load once, use many times", to amortize the cost
823  Note 2 :`dict` content will be copied internally, so `dict` can be released after loading.
824           Use ZSTD_DCtx_loadDictionary_byReference() to reference dictionary content instead.
825  Note 3 : Use ZSTD_DCtx_loadDictionary_advanced() to take control of
826           how dictionary content is loaded and interpreted.
827
828</p></pre><BR>
829
830<pre><b>size_t ZSTD_DCtx_refDDict(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict);
831</b><p>  Reference a prepared dictionary, to be used to decompress next frames.
832  The dictionary remains active for decompression of future frames using same DCtx.
833 @result : 0, or an error code (which can be tested with ZSTD_isError()).
834  Note 1 : Currently, only one dictionary can be managed.
835           Referencing a new dictionary effectively "discards" any previous one.
836  Special: referencing a NULL DDict means "return to no-dictionary mode".
837  Note 2 : DDict is just referenced, its lifetime must outlive its usage from DCtx.
838
839</p></pre><BR>
840
841<pre><b>size_t ZSTD_DCtx_refPrefix(ZSTD_DCtx* dctx,
842                     const void* prefix, size_t prefixSize);
843</b><p>  Reference a prefix (single-usage dictionary) to decompress next frame.
844  This is the reverse operation of ZSTD_CCtx_refPrefix(),
845  and must use the same prefix as the one used during compression.
846  Prefix is **only used once**. Reference is discarded at end of frame.
847  End of frame is reached when ZSTD_decompressStream() returns 0.
848 @result : 0, or an error code (which can be tested with ZSTD_isError()).
849  Note 1 : Adding any prefix (including NULL) invalidates any previously set prefix or dictionary
850  Note 2 : Prefix buffer is referenced. It **must** outlive decompression.
851           Prefix buffer must remain unmodified up to the end of frame,
852           reached when ZSTD_decompressStream() returns 0.
853  Note 3 : By default, the prefix is treated as raw content (ZSTD_dm_rawContent).
854           Use ZSTD_CCtx_refPrefix_advanced() to alter dictMode (Experimental section)
855  Note 4 : Referencing a raw content prefix has almost no cpu nor memory cost.
856           A full dictionary is more costly, as it requires building tables.
857
858</p></pre><BR>
859
860<pre><b>size_t ZSTD_sizeof_CCtx(const ZSTD_CCtx* cctx);
861size_t ZSTD_sizeof_DCtx(const ZSTD_DCtx* dctx);
862size_t ZSTD_sizeof_CStream(const ZSTD_CStream* zcs);
863size_t ZSTD_sizeof_DStream(const ZSTD_DStream* zds);
864size_t ZSTD_sizeof_CDict(const ZSTD_CDict* cdict);
865size_t ZSTD_sizeof_DDict(const ZSTD_DDict* ddict);
866</b><p>  These functions give the _current_ memory usage of selected object.
867  Note that object memory usage can evolve (increase or decrease) over time.
868</p></pre><BR>
869
870<a name="Chapter21"></a><h2>ADVANCED AND EXPERIMENTAL FUNCTIONS</h2><pre>
871 The definitions in the following section are considered experimental.
872 They are provided for advanced scenarios.
873 They should never be used with a dynamic library, as prototypes may change in the future.
874 Use them only in association with static linking.
875
876<BR></pre>
877
878<a name="Chapter22"></a><h2>experimental API (static linking only)</h2><pre>
879 The following symbols and constants
880 are not planned to join "stable API" status in the near future.
881 They can still change in future versions.
882 Some of them are planned to remain in the static_only section indefinitely.
883 Some of them might be removed in the future (especially when redundant with existing stable functions)
884
885<BR></pre>
886
887<pre><b>typedef struct {
888    unsigned windowLog;       </b>/**< largest match distance : larger == more compression, more memory needed during decompression */<b>
889    unsigned chainLog;        </b>/**< fully searched segment : larger == more compression, slower, more memory (useless for fast) */<b>
890    unsigned hashLog;         </b>/**< dispatch table : larger == faster, more memory */<b>
891    unsigned searchLog;       </b>/**< nb of searches : larger == more compression, slower */<b>
892    unsigned minMatch;        </b>/**< match length searched : larger == faster decompression, sometimes less compression */<b>
893    unsigned targetLength;    </b>/**< acceptable match size for optimal parser (only) : larger == more compression, slower */<b>
894    ZSTD_strategy strategy;   </b>/**< see ZSTD_strategy definition above */<b>
895} ZSTD_compressionParameters;
896</b></pre><BR>
897<pre><b>typedef struct {
898    int contentSizeFlag; </b>/**< 1: content size will be in frame header (when known) */<b>
899    int checksumFlag;    </b>/**< 1: generate a 32-bits checksum using XXH64 algorithm at end of frame, for error detection */<b>
900    int noDictIDFlag;    </b>/**< 1: no dictID will be saved into frame header (dictID is only useful for dictionary compression) */<b>
901} ZSTD_frameParameters;
902</b></pre><BR>
903<pre><b>typedef struct {
904    ZSTD_compressionParameters cParams;
905    ZSTD_frameParameters fParams;
906} ZSTD_parameters;
907</b></pre><BR>
908<pre><b>typedef enum {
909    ZSTD_dct_auto = 0,       </b>/* dictionary is "full" when starting with ZSTD_MAGIC_DICTIONARY, otherwise it is "rawContent" */<b>
910    ZSTD_dct_rawContent = 1, </b>/* ensures dictionary is always loaded as rawContent, even if it starts with ZSTD_MAGIC_DICTIONARY */<b>
911    ZSTD_dct_fullDict = 2    </b>/* refuses to load a dictionary if it does not respect Zstandard's specification, starting with ZSTD_MAGIC_DICTIONARY */<b>
912} ZSTD_dictContentType_e;
913</b></pre><BR>
914<pre><b>typedef enum {
915    ZSTD_dlm_byCopy = 0,  </b>/**< Copy dictionary content internally */<b>
916    ZSTD_dlm_byRef = 1,   </b>/**< Reference dictionary content -- the dictionary buffer must outlive its users. */<b>
917} ZSTD_dictLoadMethod_e;
918</b></pre><BR>
919<pre><b>typedef enum {
920    </b>/* Opened question : should we have a format ZSTD_f_auto ?<b>
921     * Today, it would mean exactly the same as ZSTD_f_zstd1.
922     * But, in the future, should several formats become supported,
923     * on the compression side, it would mean "default format".
924     * On the decompression side, it would mean "automatic format detection",
925     * so that ZSTD_f_zstd1 would mean "accept *only* zstd frames".
926     * Since meaning is a little different, another option could be to define different enums for compression and decompression.
927     * This question could be kept for later, when there are actually multiple formats to support,
928     * but there is also the question of pinning enum values, and pinning value `0` is especially important */
929    ZSTD_f_zstd1 = 0,           </b>/* zstd frame format, specified in zstd_compression_format.md (default) */<b>
930    ZSTD_f_zstd1_magicless = 1, </b>/* Variant of zstd frame format, without initial 4-bytes magic number.<b>
931                                 * Useful to save 4 bytes per generated frame.
932                                 * Decoder cannot recognise automatically this format, requiring this instruction. */
933} ZSTD_format_e;
934</b></pre><BR>
935<pre><b>typedef enum {
936    </b>/* Note: this enum and the behavior it controls are effectively internal<b>
937     * implementation details of the compressor. They are expected to continue
938     * to evolve and should be considered only in the context of extremely
939     * advanced performance tuning.
940     *
941     * Zstd currently supports the use of a CDict in two ways:
942     *
943     * - The contents of the CDict can be copied into the working context. This
944     *   means that the compression can search both the dictionary and input
945     *   while operating on a single set of internal tables. This makes
946     *   the compression faster per-byte of input. However, the initial copy of
947     *   the CDict's tables incurs a fixed cost at the beginning of the
948     *   compression. For small compressions (< 8 KB), that copy can dominate
949     *   the cost of the compression.
950     *
951     * - The CDict's tables can be used in-place. In this model, compression is
952     *   slower per input byte, because the compressor has to search two sets of
953     *   tables. However, this model incurs no start-up cost (as long as the
954     *   working context's tables can be reused). For small inputs, this can be
955     *   faster than copying the CDict's tables.
956     *
957     * Zstd has a simple internal heuristic that selects which strategy to use
958     * at the beginning of a compression. However, if experimentation shows that
959     * Zstd is making poor choices, it is possible to override that choice with
960     * this enum.
961     */
962    ZSTD_dictDefaultAttach = 0, </b>/* Use the default heuristic. */<b>
963    ZSTD_dictForceAttach   = 1, </b>/* Never copy the dictionary. */<b>
964    ZSTD_dictForceCopy     = 2, </b>/* Always copy the dictionary. */<b>
965} ZSTD_dictAttachPref_e;
966</b></pre><BR>
967<pre><b>typedef enum {
968  ZSTD_lcm_auto = 0,          </b>/**< Automatically determine the compression mode based on the compression level.<b>
969                               *   Negative compression levels will be uncompressed, and positive compression
970                               *   levels will be compressed. */
971  ZSTD_lcm_huffman = 1,       </b>/**< Always attempt Huffman compression. Uncompressed literals will still be<b>
972                               *   emitted if Huffman compression is not profitable. */
973  ZSTD_lcm_uncompressed = 2,  </b>/**< Always emit uncompressed literals. */<b>
974} ZSTD_literalCompressionMode_e;
975</b></pre><BR>
976<a name="Chapter23"></a><h2>Frame size functions</h2><pre></pre>
977
978<pre><b>unsigned long long ZSTD_findDecompressedSize(const void* src, size_t srcSize);
979</b><p>  `src` should point to the start of a series of ZSTD encoded and/or skippable frames
980  `srcSize` must be the _exact_ size of this series
981       (i.e. there should be a frame boundary at `src + srcSize`)
982  @return : - decompressed size of all data in all successive frames
983            - if the decompressed size cannot be determined: ZSTD_CONTENTSIZE_UNKNOWN
984            - if an error occurred: ZSTD_CONTENTSIZE_ERROR
985
986   note 1 : decompressed size is an optional field, that may not be present, especially in streaming mode.
987            When `return==ZSTD_CONTENTSIZE_UNKNOWN`, data to decompress could be any size.
988            In which case, it's necessary to use streaming mode to decompress data.
989   note 2 : decompressed size is always present when compression is done with ZSTD_compress()
990   note 3 : decompressed size can be very large (64-bits value),
991            potentially larger than what local system can handle as a single memory segment.
992            In which case, it's necessary to use streaming mode to decompress data.
993   note 4 : If source is untrusted, decompressed size could be wrong or intentionally modified.
994            Always ensure result fits within application's authorized limits.
995            Each application can set its own limits.
996   note 5 : ZSTD_findDecompressedSize handles multiple frames, and so it must traverse the input to
997            read each contained frame header.  This is fast as most of the data is skipped,
998            however it does mean that all frame data must be present and valid.
999</p></pre><BR>
1000
1001<a name="Chapter24"></a><h2>ZSTD_decompressBound() :</h2><pre>  `src` should point to the start of a series of ZSTD encoded and/or skippable frames
1002  `srcSize` must be the _exact_ size of this series
1003       (i.e. there should be a frame boundary at `src + srcSize`)
1004  @return : - upper-bound for the decompressed size of all data in all successive frames
1005            - if an error occured: ZSTD_CONTENTSIZE_ERROR
1006
1007  note 1  : an error can occur if `src` contains an invalid or incorrectly formatted frame.
1008  note 2  : the upper-bound is exact when the decompressed size field is available in every ZSTD encoded frame of `src`.
1009            in this case, `ZSTD_findDecompressedSize` and `ZSTD_decompressBound` return the same value.
1010  note 3  : when the decompressed size field isn't available, the upper-bound for that frame is calculated by:
1011              upper-bound = # blocks * min(128 KB, Window_Size)
1012
1013<BR></pre>
1014
1015<pre><b>size_t ZSTD_frameHeaderSize(const void* src, size_t srcSize);
1016</b><p>  srcSize must be >= ZSTD_FRAMEHEADERSIZE_PREFIX.
1017 @return : size of the Frame Header,
1018           or an error code (if srcSize is too small)
1019</p></pre><BR>
1020
1021<a name="Chapter25"></a><h2>Memory management</h2><pre></pre>
1022
1023<pre><b>size_t ZSTD_estimateCCtxSize(int compressionLevel);
1024size_t ZSTD_estimateCCtxSize_usingCParams(ZSTD_compressionParameters cParams);
1025size_t ZSTD_estimateCCtxSize_usingCCtxParams(const ZSTD_CCtx_params* params);
1026size_t ZSTD_estimateDCtxSize(void);
1027</b><p>  These functions make it possible to estimate memory usage
1028  of a future {D,C}Ctx, before its creation.
1029  ZSTD_estimateCCtxSize() will provide a budget large enough for any compression level up to selected one.
1030  It will also consider src size to be arbitrarily "large", which is worst case.
1031  If srcSize is known to always be small, ZSTD_estimateCCtxSize_usingCParams() can provide a tighter estimation.
1032  ZSTD_estimateCCtxSize_usingCParams() can be used in tandem with ZSTD_getCParams() to create cParams from compressionLevel.
1033  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.
1034  Note : CCtx size estimation is only correct for single-threaded compression.
1035</p></pre><BR>
1036
1037<pre><b>size_t ZSTD_estimateCStreamSize(int compressionLevel);
1038size_t ZSTD_estimateCStreamSize_usingCParams(ZSTD_compressionParameters cParams);
1039size_t ZSTD_estimateCStreamSize_usingCCtxParams(const ZSTD_CCtx_params* params);
1040size_t ZSTD_estimateDStreamSize(size_t windowSize);
1041size_t ZSTD_estimateDStreamSize_fromFrame(const void* src, size_t srcSize);
1042</b><p>  ZSTD_estimateCStreamSize() will provide a budget large enough for any compression level up to selected one.
1043  It will also consider src size to be arbitrarily "large", which is worst case.
1044  If srcSize is known to always be small, ZSTD_estimateCStreamSize_usingCParams() can provide a tighter estimation.
1045  ZSTD_estimateCStreamSize_usingCParams() can be used in tandem with ZSTD_getCParams() to create cParams from compressionLevel.
1046  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.
1047  Note : CStream size estimation is only correct for single-threaded compression.
1048  ZSTD_DStream memory budget depends on window Size.
1049  This information can be passed manually, using ZSTD_estimateDStreamSize,
1050  or deducted from a valid frame Header, using ZSTD_estimateDStreamSize_fromFrame();
1051  Note : if streaming is init with function ZSTD_init?Stream_usingDict(),
1052         an internal ?Dict will be created, which additional size is not estimated here.
1053         In this case, get total size by adding ZSTD_estimate?DictSize
1054</p></pre><BR>
1055
1056<pre><b>size_t ZSTD_estimateCDictSize(size_t dictSize, int compressionLevel);
1057size_t ZSTD_estimateCDictSize_advanced(size_t dictSize, ZSTD_compressionParameters cParams, ZSTD_dictLoadMethod_e dictLoadMethod);
1058size_t ZSTD_estimateDDictSize(size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod);
1059</b><p>  ZSTD_estimateCDictSize() will bet that src size is relatively "small", and content is copied, like ZSTD_createCDict().
1060  ZSTD_estimateCDictSize_advanced() makes it possible to control compression parameters precisely, like ZSTD_createCDict_advanced().
1061  Note : dictionaries created by reference (`ZSTD_dlm_byRef`) are logically smaller.
1062
1063</p></pre><BR>
1064
1065<pre><b>ZSTD_CCtx*    ZSTD_initStaticCCtx(void* workspace, size_t workspaceSize);
1066ZSTD_CStream* ZSTD_initStaticCStream(void* workspace, size_t workspaceSize);    </b>/**< same as ZSTD_initStaticCCtx() */<b>
1067</b><p>  Initialize an object using a pre-allocated fixed-size buffer.
1068  workspace: The memory area to emplace the object into.
1069             Provided pointer *must be 8-bytes aligned*.
1070             Buffer must outlive object.
1071  workspaceSize: Use ZSTD_estimate*Size() to determine
1072                 how large workspace must be to support target scenario.
1073 @return : pointer to object (same address as workspace, just different type),
1074           or NULL if error (size too small, incorrect alignment, etc.)
1075  Note : zstd will never resize nor malloc() when using a static buffer.
1076         If the object requires more memory than available,
1077         zstd will just error out (typically ZSTD_error_memory_allocation).
1078  Note 2 : there is no corresponding "free" function.
1079           Since workspace is allocated externally, it must be freed externally too.
1080  Note 3 : cParams : use ZSTD_getCParams() to convert a compression level
1081           into its associated cParams.
1082  Limitation 1 : currently not compatible with internal dictionary creation, triggered by
1083                 ZSTD_CCtx_loadDictionary(), ZSTD_initCStream_usingDict() or ZSTD_initDStream_usingDict().
1084  Limitation 2 : static cctx currently not compatible with multi-threading.
1085  Limitation 3 : static dctx is incompatible with legacy support.
1086
1087</p></pre><BR>
1088
1089<pre><b>ZSTD_DStream* ZSTD_initStaticDStream(void* workspace, size_t workspaceSize);    </b>/**< same as ZSTD_initStaticDCtx() */<b>
1090</b></pre><BR>
1091<pre><b>typedef void* (*ZSTD_allocFunction) (void* opaque, size_t size);
1092typedef void  (*ZSTD_freeFunction) (void* opaque, void* address);
1093typedef struct { ZSTD_allocFunction customAlloc; ZSTD_freeFunction customFree; void* opaque; } ZSTD_customMem;
1094static ZSTD_customMem const ZSTD_defaultCMem = { NULL, NULL, NULL };  </b>/**< this constant defers to stdlib's functions */<b>
1095</b><p>  These prototypes make it possible to pass your own allocation/free functions.
1096  ZSTD_customMem is provided at creation time, using ZSTD_create*_advanced() variants listed below.
1097  All allocation/free operations will be completed using these custom variants instead of regular <stdlib.h> ones.
1098
1099</p></pre><BR>
1100
1101<a name="Chapter26"></a><h2>Advanced compression functions</h2><pre></pre>
1102
1103<pre><b>ZSTD_CDict* ZSTD_createCDict_byReference(const void* dictBuffer, size_t dictSize, int compressionLevel);
1104</b><p>  Create a digested dictionary for compression
1105  Dictionary content is just referenced, not duplicated.
1106  As a consequence, `dictBuffer` **must** outlive CDict,
1107  and its content must remain unmodified throughout the lifetime of CDict.
1108</p></pre><BR>
1109
1110<pre><b>ZSTD_compressionParameters ZSTD_getCParams(int compressionLevel, unsigned long long estimatedSrcSize, size_t dictSize);
1111</b><p> @return ZSTD_compressionParameters structure for a selected compression level and estimated srcSize.
1112 `estimatedSrcSize` value is optional, select 0 if not known
1113</p></pre><BR>
1114
1115<pre><b>ZSTD_parameters ZSTD_getParams(int compressionLevel, unsigned long long estimatedSrcSize, size_t dictSize);
1116</b><p>  same as ZSTD_getCParams(), but @return a full `ZSTD_parameters` object instead of sub-component `ZSTD_compressionParameters`.
1117  All fields of `ZSTD_frameParameters` are set to default : contentSize=1, checksum=0, noDictID=0
1118</p></pre><BR>
1119
1120<pre><b>size_t ZSTD_checkCParams(ZSTD_compressionParameters params);
1121</b><p>  Ensure param values remain within authorized range.
1122 @return 0 on success, or an error code (can be checked with ZSTD_isError())
1123</p></pre><BR>
1124
1125<pre><b>ZSTD_compressionParameters ZSTD_adjustCParams(ZSTD_compressionParameters cPar, unsigned long long srcSize, size_t dictSize);
1126</b><p>  optimize params for a given `srcSize` and `dictSize`.
1127 `srcSize` can be unknown, in which case use ZSTD_CONTENTSIZE_UNKNOWN.
1128 `dictSize` must be `0` when there is no dictionary.
1129  cPar can be invalid : all parameters will be clamped within valid range in the @return struct.
1130  This function never fails (wide contract)
1131</p></pre><BR>
1132
1133<pre><b>size_t ZSTD_compress_advanced(ZSTD_CCtx* cctx,
1134                              void* dst, size_t dstCapacity,
1135                        const void* src, size_t srcSize,
1136                        const void* dict,size_t dictSize,
1137                              ZSTD_parameters params);
1138</b><p>  Same as ZSTD_compress_usingDict(), with fine-tune control over compression parameters (by structure)
1139</p></pre><BR>
1140
1141<pre><b>size_t ZSTD_compress_usingCDict_advanced(ZSTD_CCtx* cctx,
1142                                  void* dst, size_t dstCapacity,
1143                            const void* src, size_t srcSize,
1144                            const ZSTD_CDict* cdict,
1145                                  ZSTD_frameParameters fParams);
1146</b><p>  Same as ZSTD_compress_usingCDict(), with fine-tune control over frame parameters
1147</p></pre><BR>
1148
1149<pre><b>size_t ZSTD_CCtx_loadDictionary_byReference(ZSTD_CCtx* cctx, const void* dict, size_t dictSize);
1150</b><p>  Same as ZSTD_CCtx_loadDictionary(), but dictionary content is referenced, instead of being copied into CCtx.
1151  It saves some memory, but also requires that `dict` outlives its usage within `cctx`
1152</p></pre><BR>
1153
1154<pre><b>size_t ZSTD_CCtx_loadDictionary_advanced(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod, ZSTD_dictContentType_e dictContentType);
1155</b><p>  Same as ZSTD_CCtx_loadDictionary(), but gives finer control over
1156  how to load the dictionary (by copy ? by reference ?)
1157  and how to interpret it (automatic ? force raw mode ? full mode only ?)
1158</p></pre><BR>
1159
1160<pre><b>size_t ZSTD_CCtx_refPrefix_advanced(ZSTD_CCtx* cctx, const void* prefix, size_t prefixSize, ZSTD_dictContentType_e dictContentType);
1161</b><p>  Same as ZSTD_CCtx_refPrefix(), but gives finer control over
1162  how to interpret prefix content (automatic ? force raw mode (default) ? full mode only ?)
1163</p></pre><BR>
1164
1165<pre><b>size_t ZSTD_CCtx_getParameter(ZSTD_CCtx* cctx, ZSTD_cParameter param, int* value);
1166</b><p>  Get the requested compression parameter value, selected by enum ZSTD_cParameter,
1167  and store it into int* value.
1168 @return : 0, or an error code (which can be tested with ZSTD_isError()).
1169
1170</p></pre><BR>
1171
1172<pre><b>ZSTD_CCtx_params* ZSTD_createCCtxParams(void);
1173size_t ZSTD_freeCCtxParams(ZSTD_CCtx_params* params);
1174</b><p>  Quick howto :
1175  - ZSTD_createCCtxParams() : Create a ZSTD_CCtx_params structure
1176  - ZSTD_CCtxParams_setParameter() : Push parameters one by one into
1177                                     an existing ZSTD_CCtx_params structure.
1178                                     This is similar to
1179                                     ZSTD_CCtx_setParameter().
1180  - ZSTD_CCtx_setParametersUsingCCtxParams() : Apply parameters to
1181                                    an existing CCtx.
1182                                    These parameters will be applied to
1183                                    all subsequent frames.
1184  - ZSTD_compressStream2() : Do compression using the CCtx.
1185  - ZSTD_freeCCtxParams() : Free the memory.
1186
1187  This can be used with ZSTD_estimateCCtxSize_advanced_usingCCtxParams()
1188  for static allocation of CCtx for single-threaded compression.
1189
1190</p></pre><BR>
1191
1192<pre><b>size_t ZSTD_CCtxParams_reset(ZSTD_CCtx_params* params);
1193</b><p>  Reset params to default values.
1194
1195</p></pre><BR>
1196
1197<pre><b>size_t ZSTD_CCtxParams_init(ZSTD_CCtx_params* cctxParams, int compressionLevel);
1198</b><p>  Initializes the compression parameters of cctxParams according to
1199  compression level. All other parameters are reset to their default values.
1200
1201</p></pre><BR>
1202
1203<pre><b>size_t ZSTD_CCtxParams_init_advanced(ZSTD_CCtx_params* cctxParams, ZSTD_parameters params);
1204</b><p>  Initializes the compression and frame parameters of cctxParams according to
1205  params. All other parameters are reset to their default values.
1206
1207</p></pre><BR>
1208
1209<pre><b>size_t ZSTD_CCtxParams_setParameter(ZSTD_CCtx_params* params, ZSTD_cParameter param, int value);
1210</b><p>  Similar to ZSTD_CCtx_setParameter.
1211  Set one compression parameter, selected by enum ZSTD_cParameter.
1212  Parameters must be applied to a ZSTD_CCtx using ZSTD_CCtx_setParametersUsingCCtxParams().
1213 @result : 0, or an error code (which can be tested with ZSTD_isError()).
1214
1215</p></pre><BR>
1216
1217<pre><b>size_t ZSTD_CCtxParams_getParameter(ZSTD_CCtx_params* params, ZSTD_cParameter param, int* value);
1218</b><p> Similar to ZSTD_CCtx_getParameter.
1219 Get the requested value of one compression parameter, selected by enum ZSTD_cParameter.
1220 @result : 0, or an error code (which can be tested with ZSTD_isError()).
1221
1222</p></pre><BR>
1223
1224<pre><b>size_t ZSTD_CCtx_setParametersUsingCCtxParams(
1225        ZSTD_CCtx* cctx, const ZSTD_CCtx_params* params);
1226</b><p>  Apply a set of ZSTD_CCtx_params to the compression context.
1227  This can be done even after compression is started,
1228    if nbWorkers==0, this will have no impact until a new compression is started.
1229    if nbWorkers>=1, new parameters will be picked up at next job,
1230       with a few restrictions (windowLog, pledgedSrcSize, nbWorkers, jobSize, and overlapLog are not updated).
1231
1232</p></pre><BR>
1233
1234<pre><b>size_t ZSTD_compressStream2_simpleArgs (
1235                ZSTD_CCtx* cctx,
1236                void* dst, size_t dstCapacity, size_t* dstPos,
1237          const void* src, size_t srcSize, size_t* srcPos,
1238                ZSTD_EndDirective endOp);
1239</b><p>  Same as ZSTD_compressStream2(),
1240  but using only integral types as arguments.
1241  This variant might be helpful for binders from dynamic languages
1242  which have troubles handling structures containing memory pointers.
1243
1244</p></pre><BR>
1245
1246<a name="Chapter27"></a><h2>Advanced decompression functions</h2><pre></pre>
1247
1248<pre><b>unsigned ZSTD_isFrame(const void* buffer, size_t size);
1249</b><p>  Tells if the content of `buffer` starts with a valid Frame Identifier.
1250  Note : Frame Identifier is 4 bytes. If `size < 4`, @return will always be 0.
1251  Note 2 : Legacy Frame Identifiers are considered valid only if Legacy Support is enabled.
1252  Note 3 : Skippable Frame Identifiers are considered valid.
1253</p></pre><BR>
1254
1255<pre><b>ZSTD_DDict* ZSTD_createDDict_byReference(const void* dictBuffer, size_t dictSize);
1256</b><p>  Create a digested dictionary, ready to start decompression operation without startup delay.
1257  Dictionary content is referenced, and therefore stays in dictBuffer.
1258  It is important that dictBuffer outlives DDict,
1259  it must remain read accessible throughout the lifetime of DDict
1260</p></pre><BR>
1261
1262<pre><b>size_t ZSTD_DCtx_loadDictionary_byReference(ZSTD_DCtx* dctx, const void* dict, size_t dictSize);
1263</b><p>  Same as ZSTD_DCtx_loadDictionary(),
1264  but references `dict` content instead of copying it into `dctx`.
1265  This saves memory if `dict` remains around.,
1266  However, it's imperative that `dict` remains accessible (and unmodified) while being used, so it must outlive decompression.
1267</p></pre><BR>
1268
1269<pre><b>size_t ZSTD_DCtx_loadDictionary_advanced(ZSTD_DCtx* dctx, const void* dict, size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod, ZSTD_dictContentType_e dictContentType);
1270</b><p>  Same as ZSTD_DCtx_loadDictionary(),
1271  but gives direct control over
1272  how to load the dictionary (by copy ? by reference ?)
1273  and how to interpret it (automatic ? force raw mode ? full mode only ?).
1274</p></pre><BR>
1275
1276<pre><b>size_t ZSTD_DCtx_refPrefix_advanced(ZSTD_DCtx* dctx, const void* prefix, size_t prefixSize, ZSTD_dictContentType_e dictContentType);
1277</b><p>  Same as ZSTD_DCtx_refPrefix(), but gives finer control over
1278  how to interpret prefix content (automatic ? force raw mode (default) ? full mode only ?)
1279</p></pre><BR>
1280
1281<pre><b>size_t ZSTD_DCtx_setMaxWindowSize(ZSTD_DCtx* dctx, size_t maxWindowSize);
1282</b><p>  Refuses allocating internal buffers for frames requiring a window size larger than provided limit.
1283  This protects a decoder context from reserving too much memory for itself (potential attack scenario).
1284  This parameter is only useful in streaming mode, since no internal buffer is allocated in single-pass mode.
1285  By default, a decompression context accepts all window sizes <= (1 << ZSTD_WINDOWLOG_LIMIT_DEFAULT)
1286 @return : 0, or an error code (which can be tested using ZSTD_isError()).
1287
1288</p></pre><BR>
1289
1290<pre><b>size_t ZSTD_DCtx_setFormat(ZSTD_DCtx* dctx, ZSTD_format_e format);
1291</b><p>  Instruct the decoder context about what kind of data to decode next.
1292  This instruction is mandatory to decode data without a fully-formed header,
1293  such ZSTD_f_zstd1_magicless for example.
1294 @return : 0, or an error code (which can be tested using ZSTD_isError()).
1295</p></pre><BR>
1296
1297<pre><b>size_t ZSTD_decompressStream_simpleArgs (
1298                ZSTD_DCtx* dctx,
1299                void* dst, size_t dstCapacity, size_t* dstPos,
1300          const void* src, size_t srcSize, size_t* srcPos);
1301</b><p>  Same as ZSTD_decompressStream(),
1302  but using only integral types as arguments.
1303  This can be helpful for binders from dynamic languages
1304  which have troubles handling structures containing memory pointers.
1305
1306</p></pre><BR>
1307
1308<a name="Chapter28"></a><h2>Advanced streaming functions</h2><pre>  Warning : most of these functions are now redundant with the Advanced API.
1309  Once Advanced API reaches "stable" status,
1310  redundant functions will be deprecated, and then at some point removed.
1311<BR></pre>
1312
1313<h3>Advanced Streaming compression functions</h3><pre></pre><b><pre></b>/**! ZSTD_initCStream_srcSize() :<b>
1314 * This function is deprecated, and equivalent to:
1315 *     ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only);
1316 *     ZSTD_CCtx_refCDict(zcs, NULL); // clear the dictionary (if any)
1317 *     ZSTD_CCtx_setParameter(zcs, ZSTD_c_compressionLevel, compressionLevel);
1318 *     ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize);
1319 *
1320 * pledgedSrcSize must be correct. If it is not known at init time, use
1321 * ZSTD_CONTENTSIZE_UNKNOWN. Note that, for compatibility with older programs,
1322 * "0" also disables frame content size field. It may be enabled in the future.
1323 */
1324size_t ZSTD_initCStream_srcSize(ZSTD_CStream* zcs, int compressionLevel, unsigned long long pledgedSrcSize);
1325</b>/**! ZSTD_initCStream_usingDict() :<b>
1326 * This function is deprecated, and is equivalent to:
1327 *     ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only);
1328 *     ZSTD_CCtx_setParameter(zcs, ZSTD_c_compressionLevel, compressionLevel);
1329 *     ZSTD_CCtx_loadDictionary(zcs, dict, dictSize);
1330 *
1331 * Creates of an internal CDict (incompatible with static CCtx), except if
1332 * dict == NULL or dictSize < 8, in which case no dict is used.
1333 * Note: dict is loaded with ZSTD_dm_auto (treated as a full zstd dictionary if
1334 * it begins with ZSTD_MAGIC_DICTIONARY, else as raw content) and ZSTD_dlm_byCopy.
1335 */
1336size_t ZSTD_initCStream_usingDict(ZSTD_CStream* zcs, const void* dict, size_t dictSize, int compressionLevel);
1337</b>/**! ZSTD_initCStream_advanced() :<b>
1338 * This function is deprecated, and is approximately equivalent to:
1339 *     ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only);
1340 *     ZSTD_CCtx_setZstdParams(zcs, params); // Set the zstd params and leave the rest as-is
1341 *     ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize);
1342 *     ZSTD_CCtx_loadDictionary(zcs, dict, dictSize);
1343 *
1344 * pledgedSrcSize must be correct. If srcSize is not known at init time, use
1345 * value ZSTD_CONTENTSIZE_UNKNOWN. dict is loaded with ZSTD_dm_auto and ZSTD_dlm_byCopy.
1346 */
1347size_t ZSTD_initCStream_advanced(ZSTD_CStream* zcs, const void* dict, size_t dictSize,
1348                                             ZSTD_parameters params, unsigned long long pledgedSrcSize);
1349</b>/**! ZSTD_initCStream_usingCDict() :<b>
1350 * This function is deprecated, and equivalent to:
1351 *     ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only);
1352 *     ZSTD_CCtx_refCDict(zcs, cdict);
1353 *
1354 * note : cdict will just be referenced, and must outlive compression session
1355 */
1356size_t ZSTD_initCStream_usingCDict(ZSTD_CStream* zcs, const ZSTD_CDict* cdict);
1357</b>/**! ZSTD_initCStream_usingCDict_advanced() :<b>
1358 * This function is deprecated, and is approximately equivalent to:
1359 *     ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only);
1360 *     ZSTD_CCtx_setZstdFrameParams(zcs, fParams); // Set the zstd frame params and leave the rest as-is
1361 *     ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize);
1362 *     ZSTD_CCtx_refCDict(zcs, cdict);
1363 *
1364 * same as ZSTD_initCStream_usingCDict(), with control over frame parameters.
1365 * pledgedSrcSize must be correct. If srcSize is not known at init time, use
1366 * value ZSTD_CONTENTSIZE_UNKNOWN.
1367 */
1368size_t ZSTD_initCStream_usingCDict_advanced(ZSTD_CStream* zcs, const ZSTD_CDict* cdict, ZSTD_frameParameters fParams, unsigned long long pledgedSrcSize);
1369</pre></b><BR>
1370<pre><b>size_t ZSTD_resetCStream(ZSTD_CStream* zcs, unsigned long long pledgedSrcSize);
1371</b><p> This function is deprecated, and is equivalent to:
1372     ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only);
1373     ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize);
1374
1375  start a new frame, using same parameters from previous frame.
1376  This is typically useful to skip dictionary loading stage, since it will re-use it in-place.
1377  Note that zcs must be init at least once before using ZSTD_resetCStream().
1378  If pledgedSrcSize is not known at reset time, use macro ZSTD_CONTENTSIZE_UNKNOWN.
1379  If pledgedSrcSize > 0, its value must be correct, as it will be written in header, and controlled at the end.
1380  For the time being, pledgedSrcSize==0 is interpreted as "srcSize unknown" for compatibility with older programs,
1381  but it will change to mean "empty" in future version, so use macro ZSTD_CONTENTSIZE_UNKNOWN instead.
1382 @return : 0, or an error code (which can be tested using ZSTD_isError())
1383
1384</p></pre><BR>
1385
1386<pre><b>typedef struct {
1387    unsigned long long ingested;   </b>/* nb input bytes read and buffered */<b>
1388    unsigned long long consumed;   </b>/* nb input bytes actually compressed */<b>
1389    unsigned long long produced;   </b>/* nb of compressed bytes generated and buffered */<b>
1390    unsigned long long flushed;    </b>/* nb of compressed bytes flushed : not provided; can be tracked from caller side */<b>
1391    unsigned currentJobID;         </b>/* MT only : latest started job nb */<b>
1392    unsigned nbActiveWorkers;      </b>/* MT only : nb of workers actively compressing at probe time */<b>
1393} ZSTD_frameProgression;
1394</b></pre><BR>
1395<pre><b>size_t ZSTD_toFlushNow(ZSTD_CCtx* cctx);
1396</b><p>  Tell how many bytes are ready to be flushed immediately.
1397  Useful for multithreading scenarios (nbWorkers >= 1).
1398  Probe the oldest active job, defined as oldest job not yet entirely flushed,
1399  and check its output buffer.
1400 @return : amount of data stored in oldest job and ready to be flushed immediately.
1401  if @return == 0, it means either :
1402  + there is no active job (could be checked with ZSTD_frameProgression()), or
1403  + oldest job is still actively compressing data,
1404    but everything it has produced has also been flushed so far,
1405    therefore flush speed is limited by production speed of oldest job
1406    irrespective of the speed of concurrent (and newer) jobs.
1407
1408</p></pre><BR>
1409
1410<h3>Advanced Streaming decompression functions</h3><pre></pre><b><pre>size_t ZSTD_initDStream_usingDict(ZSTD_DStream* zds, const void* dict, size_t dictSize); </b>/**< note: no dictionary will be used if dict == NULL or dictSize < 8 */<b>
1411size_t ZSTD_initDStream_usingDDict(ZSTD_DStream* zds, const ZSTD_DDict* ddict);  </b>/**< note : ddict is referenced, it must outlive decompression session */<b>
1412size_t ZSTD_resetDStream(ZSTD_DStream* zds);  </b>/**< re-use decompression parameters from previous init; saves dictionary loading */<b>
1413</pre></b><BR>
1414<a name="Chapter29"></a><h2>Buffer-less and synchronous inner streaming functions</h2><pre>
1415  This is an advanced API, giving full control over buffer management, for users which need direct control over memory.
1416  But it's also a complex one, with several restrictions, documented below.
1417  Prefer normal streaming API for an easier experience.
1418
1419<BR></pre>
1420
1421<a name="Chapter30"></a><h2>Buffer-less streaming compression (synchronous mode)</h2><pre>
1422  A ZSTD_CCtx object is required to track streaming operations.
1423  Use ZSTD_createCCtx() / ZSTD_freeCCtx() to manage resource.
1424  ZSTD_CCtx object can be re-used multiple times within successive compression operations.
1425
1426  Start by initializing a context.
1427  Use ZSTD_compressBegin(), or ZSTD_compressBegin_usingDict() for dictionary compression,
1428  or ZSTD_compressBegin_advanced(), for finer parameter control.
1429  It's also possible to duplicate a reference context which has already been initialized, using ZSTD_copyCCtx()
1430
1431  Then, consume your input using ZSTD_compressContinue().
1432  There are some important considerations to keep in mind when using this advanced function :
1433  - ZSTD_compressContinue() has no internal buffer. It uses externally provided buffers only.
1434  - Interface is synchronous : input is consumed entirely and produces 1+ compressed blocks.
1435  - Caller must ensure there is enough space in `dst` to store compressed data under worst case scenario.
1436    Worst case evaluation is provided by ZSTD_compressBound().
1437    ZSTD_compressContinue() doesn't guarantee recover after a failed compression.
1438  - ZSTD_compressContinue() presumes prior input ***is still accessible and unmodified*** (up to maximum distance size, see WindowLog).
1439    It remembers all previous contiguous blocks, plus one separated memory segment (which can itself consists of multiple contiguous blocks)
1440  - ZSTD_compressContinue() detects that prior input has been overwritten when `src` buffer overlaps.
1441    In which case, it will "discard" the relevant memory section from its history.
1442
1443  Finish a frame with ZSTD_compressEnd(), which will write the last block(s) and optional checksum.
1444  It's possible to use srcSize==0, in which case, it will write a final empty block to end the frame.
1445  Without last block mark, frames are considered unfinished (hence corrupted) by compliant decoders.
1446
1447  `ZSTD_CCtx` object can be re-used (ZSTD_compressBegin()) to compress again.
1448<BR></pre>
1449
1450<h3>Buffer-less streaming compression functions</h3><pre></pre><b><pre>size_t ZSTD_compressBegin(ZSTD_CCtx* cctx, int compressionLevel);
1451size_t ZSTD_compressBegin_usingDict(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, int compressionLevel);
1452size_t ZSTD_compressBegin_advanced(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, ZSTD_parameters params, unsigned long long pledgedSrcSize); </b>/**< pledgedSrcSize : If srcSize is not known at init time, use ZSTD_CONTENTSIZE_UNKNOWN */<b>
1453size_t ZSTD_compressBegin_usingCDict(ZSTD_CCtx* cctx, const ZSTD_CDict* cdict); </b>/**< note: fails if cdict==NULL */<b>
1454size_t ZSTD_compressBegin_usingCDict_advanced(ZSTD_CCtx* const cctx, const ZSTD_CDict* const cdict, ZSTD_frameParameters const fParams, unsigned long long const pledgedSrcSize);   </b>/* compression parameters are already set within cdict. pledgedSrcSize must be correct. If srcSize is not known, use macro ZSTD_CONTENTSIZE_UNKNOWN */<b>
1455size_t ZSTD_copyCCtx(ZSTD_CCtx* cctx, const ZSTD_CCtx* preparedCCtx, unsigned long long pledgedSrcSize); </b>/**<  note: if pledgedSrcSize is not known, use ZSTD_CONTENTSIZE_UNKNOWN */<b>
1456</pre></b><BR>
1457<a name="Chapter31"></a><h2>Buffer-less streaming decompression (synchronous mode)</h2><pre>
1458  A ZSTD_DCtx object is required to track streaming operations.
1459  Use ZSTD_createDCtx() / ZSTD_freeDCtx() to manage it.
1460  A ZSTD_DCtx object can be re-used multiple times.
1461
1462  First typical operation is to retrieve frame parameters, using ZSTD_getFrameHeader().
1463  Frame header is extracted from the beginning of compressed frame, so providing only the frame's beginning is enough.
1464  Data fragment must be large enough to ensure successful decoding.
1465 `ZSTD_frameHeaderSize_max` bytes is guaranteed to always be large enough.
1466  @result : 0 : successful decoding, the `ZSTD_frameHeader` structure is correctly filled.
1467           >0 : `srcSize` is too small, please provide at least @result bytes on next attempt.
1468           errorCode, which can be tested using ZSTD_isError().
1469
1470  It fills a ZSTD_frameHeader structure with important information to correctly decode the frame,
1471  such as the dictionary ID, content size, or maximum back-reference distance (`windowSize`).
1472  Note that these values could be wrong, either because of data corruption, or because a 3rd party deliberately spoofs false information.
1473  As a consequence, check that values remain within valid application range.
1474  For example, do not allocate memory blindly, check that `windowSize` is within expectation.
1475  Each application can set its own limits, depending on local restrictions.
1476  For extended interoperability, it is recommended to support `windowSize` of at least 8 MB.
1477
1478  ZSTD_decompressContinue() needs previous data blocks during decompression, up to `windowSize` bytes.
1479  ZSTD_decompressContinue() is very sensitive to contiguity,
1480  if 2 blocks don't follow each other, make sure that either the compressor breaks contiguity at the same place,
1481  or that previous contiguous segment is large enough to properly handle maximum back-reference distance.
1482  There are multiple ways to guarantee this condition.
1483
1484  The most memory efficient way is to use a round buffer of sufficient size.
1485  Sufficient size is determined by invoking ZSTD_decodingBufferSize_min(),
1486  which can @return an error code if required value is too large for current system (in 32-bits mode).
1487  In a round buffer methodology, ZSTD_decompressContinue() decompresses each block next to previous one,
1488  up to the moment there is not enough room left in the buffer to guarantee decoding another full block,
1489  which maximum size is provided in `ZSTD_frameHeader` structure, field `blockSizeMax`.
1490  At which point, decoding can resume from the beginning of the buffer.
1491  Note that already decoded data stored in the buffer should be flushed before being overwritten.
1492
1493  There are alternatives possible, for example using two or more buffers of size `windowSize` each, though they consume more memory.
1494
1495  Finally, if you control the compression process, you can also ignore all buffer size rules,
1496  as long as the encoder and decoder progress in "lock-step",
1497  aka use exactly the same buffer sizes, break contiguity at the same place, etc.
1498
1499  Once buffers are setup, start decompression, with ZSTD_decompressBegin().
1500  If decompression requires a dictionary, use ZSTD_decompressBegin_usingDict() or ZSTD_decompressBegin_usingDDict().
1501
1502  Then use ZSTD_nextSrcSizeToDecompress() and ZSTD_decompressContinue() alternatively.
1503  ZSTD_nextSrcSizeToDecompress() tells how many bytes to provide as 'srcSize' to ZSTD_decompressContinue().
1504  ZSTD_decompressContinue() requires this _exact_ amount of bytes, or it will fail.
1505
1506 @result of ZSTD_decompressContinue() is the number of bytes regenerated within 'dst' (necessarily <= dstCapacity).
1507  It can be zero : it just means ZSTD_decompressContinue() has decoded some metadata item.
1508  It can also be an error code, which can be tested with ZSTD_isError().
1509
1510  A frame is fully decoded when ZSTD_nextSrcSizeToDecompress() returns zero.
1511  Context can then be reset to start a new decompression.
1512
1513  Note : it's possible to know if next input to present is a header or a block, using ZSTD_nextInputType().
1514  This information is not required to properly decode a frame.
1515
1516  == Special case : skippable frames
1517
1518  Skippable frames allow integration of user-defined data into a flow of concatenated frames.
1519  Skippable frames will be ignored (skipped) by decompressor.
1520  The format of skippable frames is as follows :
1521  a) Skippable frame ID - 4 Bytes, Little endian format, any value from 0x184D2A50 to 0x184D2A5F
1522  b) Frame Size - 4 Bytes, Little endian format, unsigned 32-bits
1523  c) Frame Content - any content (User Data) of length equal to Frame Size
1524  For skippable frames ZSTD_getFrameHeader() returns zfhPtr->frameType==ZSTD_skippableFrame.
1525  For skippable frames ZSTD_decompressContinue() always returns 0 : it only skips the content.
1526<BR></pre>
1527
1528<h3>Buffer-less streaming decompression functions</h3><pre></pre><b><pre>typedef enum { ZSTD_frame, ZSTD_skippableFrame } ZSTD_frameType_e;
1529typedef struct {
1530    unsigned long long frameContentSize; </b>/* if == ZSTD_CONTENTSIZE_UNKNOWN, it means this field is not available. 0 means "empty" */<b>
1531    unsigned long long windowSize;       </b>/* can be very large, up to <= frameContentSize */<b>
1532    unsigned blockSizeMax;
1533    ZSTD_frameType_e frameType;          </b>/* if == ZSTD_skippableFrame, frameContentSize is the size of skippable content */<b>
1534    unsigned headerSize;
1535    unsigned dictID;
1536    unsigned checksumFlag;
1537} ZSTD_frameHeader;
1538</pre></b><BR>
1539<a name="Chapter32"></a><h2>ZSTD_getFrameHeader() :</h2><pre>  decode Frame Header, or requires larger `srcSize`.
1540 @return : 0, `zfhPtr` is correctly filled,
1541          >0, `srcSize` is too small, value is wanted `srcSize` amount,
1542           or an error code, which can be tested using ZSTD_isError()
1543<BR></pre>
1544
1545<pre><b>size_t ZSTD_getFrameHeader(ZSTD_frameHeader* zfhPtr, const void* src, size_t srcSize);   </b>/**< doesn't consume input */<b>
1546</b></pre><BR>
1547<pre><b>size_t ZSTD_getFrameHeader_advanced(ZSTD_frameHeader* zfhPtr, const void* src, size_t srcSize, ZSTD_format_e format);
1548size_t ZSTD_decodingBufferSize_min(unsigned long long windowSize, unsigned long long frameContentSize);  </b>/**< when frame content size is not known, pass in frameContentSize == ZSTD_CONTENTSIZE_UNKNOWN */<b>
1549</b><p>  same as ZSTD_getFrameHeader(),
1550  with added capability to select a format (like ZSTD_f_zstd1_magicless)
1551</p></pre><BR>
1552
1553<pre><b>typedef enum { ZSTDnit_frameHeader, ZSTDnit_blockHeader, ZSTDnit_block, ZSTDnit_lastBlock, ZSTDnit_checksum, ZSTDnit_skippableFrame } ZSTD_nextInputType_e;
1554</b></pre><BR>
1555<a name="Chapter33"></a><h2>Block level API</h2><pre></pre>
1556
1557<pre><b></b><p>    Frame metadata cost is typically ~18 bytes, which can be non-negligible for very small blocks (< 100 bytes).
1558    User will have to take in charge required information to regenerate data, such as compressed and content sizes.
1559
1560    A few rules to respect :
1561    - Compressing and decompressing require a context structure
1562      + Use ZSTD_createCCtx() and ZSTD_createDCtx()
1563    - It is necessary to init context before starting
1564      + compression : any ZSTD_compressBegin*() variant, including with dictionary
1565      + decompression : any ZSTD_decompressBegin*() variant, including with dictionary
1566      + copyCCtx() and copyDCtx() can be used too
1567    - Block size is limited, it must be <= ZSTD_getBlockSize() <= ZSTD_BLOCKSIZE_MAX == 128 KB
1568      + If input is larger than a block size, it's necessary to split input data into multiple blocks
1569      + For inputs larger than a single block, really consider using regular ZSTD_compress() instead.
1570        Frame metadata is not that costly, and quickly becomes negligible as source size grows larger.
1571    - When a block is considered not compressible enough, ZSTD_compressBlock() result will be zero.
1572      In which case, nothing is produced into `dst` !
1573      + User must test for such outcome and deal directly with uncompressed data
1574      + ZSTD_decompressBlock() doesn't accept uncompressed data as input !!!
1575      + In case of multiple successive blocks, should some of them be uncompressed,
1576        decoder must be informed of their existence in order to follow proper history.
1577        Use ZSTD_insertBlock() for such a case.
1578</p></pre><BR>
1579
1580<h3>Raw zstd block functions</h3><pre></pre><b><pre>size_t ZSTD_getBlockSize   (const ZSTD_CCtx* cctx);
1581size_t ZSTD_compressBlock  (ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize);
1582size_t ZSTD_decompressBlock(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize);
1583size_t ZSTD_insertBlock    (ZSTD_DCtx* dctx, const void* blockStart, size_t blockSize);  </b>/**< insert uncompressed block into `dctx` history. Useful for multi-blocks decompression. */<b>
1584</pre></b><BR>
1585</html>
1586</body>
1587