• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1<html>
2<head>
3<meta http-equiv="Content-Type" content="text/html; charset=ISO-8859-1">
4<title>1.9.4 Manual</title>
5</head>
6<body>
7<h1>1.9.4 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">Tuning parameter</a></li>
14<li><a href="#Chapter4">Simple Functions</a></li>
15<li><a href="#Chapter5">Advanced Functions</a></li>
16<li><a href="#Chapter6">Streaming Compression Functions</a></li>
17<li><a href="#Chapter7">Streaming Decompression Functions</a></li>
18<li><a href="#Chapter8">Experimental section</a></li>
19<li><a href="#Chapter9">Private Definitions</a></li>
20<li><a href="#Chapter10">Obsolete Functions</a></li>
21</ol>
22<hr>
23<a name="Chapter1"></a><h2>Introduction</h2><pre>
24  LZ4 is lossless compression algorithm, providing compression speed >500 MB/s per core,
25  scalable with multi-cores CPU. It features an extremely fast decoder, with speed in
26  multiple GB/s per core, typically reaching RAM speed limits on multi-core systems.
27
28  The LZ4 compression library provides in-memory compression and decompression functions.
29  It gives full buffer control to user.
30  Compression can be done in:
31    - a single step (described as Simple Functions)
32    - a single step, reusing a context (described in Advanced Functions)
33    - unbounded multiple steps (described as Streaming compression)
34
35  lz4.h generates and decodes LZ4-compressed blocks (doc/lz4_Block_format.md).
36  Decompressing such a compressed block requires additional metadata.
37  Exact metadata depends on exact decompression function.
38  For the typical case of LZ4_decompress_safe(),
39  metadata includes block's compressed size, and maximum bound of decompressed size.
40  Each application is free to encode and pass such metadata in whichever way it wants.
41
42  lz4.h only handle blocks, it can not generate Frames.
43
44  Blocks are different from Frames (doc/lz4_Frame_format.md).
45  Frames bundle both blocks and metadata in a specified manner.
46  Embedding metadata is required for compressed data to be self-contained and portable.
47  Frame format is delivered through a companion API, declared in lz4frame.h.
48  The `lz4` CLI can only manage frames.
49<BR></pre>
50
51<pre><b>#if defined(LZ4_FREESTANDING) && (LZ4_FREESTANDING == 1)
52#  define LZ4_HEAPMODE 0
53#  define LZ4HC_HEAPMODE 0
54#  define LZ4_STATIC_LINKING_ONLY_DISABLE_MEMORY_ALLOCATION 1
55#  if !defined(LZ4_memcpy)
56#    error "LZ4_FREESTANDING requires macro 'LZ4_memcpy'."
57#  endif
58#  if !defined(LZ4_memset)
59#    error "LZ4_FREESTANDING requires macro 'LZ4_memset'."
60#  endif
61#  if !defined(LZ4_memmove)
62#    error "LZ4_FREESTANDING requires macro 'LZ4_memmove'."
63#  endif
64#elif ! defined(LZ4_FREESTANDING)
65#  define LZ4_FREESTANDING 0
66#endif
67</b><p>  When this macro is set to 1, it enables "freestanding mode" that is
68  suitable for typical freestanding environment which doesn't support
69  standard C library.
70
71  - LZ4_FREESTANDING is a compile-time switch.
72  - It requires the following macros to be defined:
73    LZ4_memcpy, LZ4_memmove, LZ4_memset.
74  - It only enables LZ4/HC functions which don't use heap.
75    All LZ4F_* functions are not supported.
76  - See tests/freestanding.c to check its basic setup.
77
78</p></pre><BR>
79
80<a name="Chapter2"></a><h2>Version</h2><pre></pre>
81
82<pre><b>int LZ4_versionNumber (void);  </b>/**< library version number; useful to check dll version; requires v1.3.0+ */<b>
83</b></pre><BR>
84<pre><b>const char* LZ4_versionString (void);   </b>/**< library version string; useful to check dll version; requires v1.7.5+ */<b>
85</b></pre><BR>
86<a name="Chapter3"></a><h2>Tuning parameter</h2><pre></pre>
87
88<pre><b>#ifndef LZ4_MEMORY_USAGE
89# define LZ4_MEMORY_USAGE LZ4_MEMORY_USAGE_DEFAULT
90#endif
91</b><p> Memory usage formula : N->2^N Bytes (examples : 10 -> 1KB; 12 -> 4KB ; 16 -> 64KB; 20 -> 1MB; )
92 Increasing memory usage improves compression ratio, at the cost of speed.
93 Reduced memory usage may improve speed at the cost of ratio, thanks to better cache locality.
94 Default value is 14, for 16KB, which nicely fits into Intel x86 L1 cache
95
96</p></pre><BR>
97
98<a name="Chapter4"></a><h2>Simple Functions</h2><pre></pre>
99
100<pre><b>int LZ4_compress_default(const char* src, char* dst, int srcSize, int dstCapacity);
101</b><p>  Compresses 'srcSize' bytes from buffer 'src'
102  into already allocated 'dst' buffer of size 'dstCapacity'.
103  Compression is guaranteed to succeed if 'dstCapacity' >= LZ4_compressBound(srcSize).
104  It also runs faster, so it's a recommended setting.
105  If the function cannot compress 'src' into a more limited 'dst' budget,
106  compression stops *immediately*, and the function result is zero.
107  In which case, 'dst' content is undefined (invalid).
108      srcSize : max supported value is LZ4_MAX_INPUT_SIZE.
109      dstCapacity : size of buffer 'dst' (which must be already allocated)
110     @return  : the number of bytes written into buffer 'dst' (necessarily <= dstCapacity)
111                or 0 if compression fails
112 Note : This function is protected against buffer overflow scenarios (never writes outside 'dst' buffer, nor read outside 'source' buffer).
113
114</p></pre><BR>
115
116<pre><b>int LZ4_decompress_safe (const char* src, char* dst, int compressedSize, int dstCapacity);
117</b><p>  compressedSize : is the exact complete size of the compressed block.
118  dstCapacity : is the size of destination buffer (which must be already allocated), presumed an upper bound of decompressed size.
119 @return : the number of bytes decompressed into destination buffer (necessarily <= dstCapacity)
120           If destination buffer is not large enough, decoding will stop and output an error code (negative value).
121           If the source stream is detected malformed, the function will stop decoding and return a negative result.
122 Note 1 : This function is protected against malicious data packets :
123          it will never writes outside 'dst' buffer, nor read outside 'source' buffer,
124          even if the compressed block is maliciously modified to order the decoder to do these actions.
125          In such case, the decoder stops immediately, and considers the compressed block malformed.
126 Note 2 : compressedSize and dstCapacity must be provided to the function, the compressed block does not contain them.
127          The implementation is free to send / store / derive this information in whichever way is most beneficial.
128          If there is a need for a different format which bundles together both compressed data and its metadata, consider looking at lz4frame.h instead.
129
130</p></pre><BR>
131
132<a name="Chapter5"></a><h2>Advanced Functions</h2><pre></pre>
133
134<pre><b>int LZ4_compressBound(int inputSize);
135</b><p>    Provides the maximum size that LZ4 compression may output in a "worst case" scenario (input data not compressible)
136    This function is primarily useful for memory allocation purposes (destination buffer size).
137    Macro LZ4_COMPRESSBOUND() is also provided for compilation-time evaluation (stack memory allocation for example).
138    Note that LZ4_compress_default() compresses faster when dstCapacity is >= LZ4_compressBound(srcSize)
139        inputSize  : max supported value is LZ4_MAX_INPUT_SIZE
140        return : maximum output size in a "worst case" scenario
141              or 0, if input size is incorrect (too large or negative)
142</p></pre><BR>
143
144<pre><b>int LZ4_compress_fast (const char* src, char* dst, int srcSize, int dstCapacity, int acceleration);
145</b><p>    Same as LZ4_compress_default(), but allows selection of "acceleration" factor.
146    The larger the acceleration value, the faster the algorithm, but also the lesser the compression.
147    It's a trade-off. It can be fine tuned, with each successive value providing roughly +~3% to speed.
148    An acceleration value of "1" is the same as regular LZ4_compress_default()
149    Values <= 0 will be replaced by LZ4_ACCELERATION_DEFAULT (currently == 1, see lz4.c).
150    Values > LZ4_ACCELERATION_MAX will be replaced by LZ4_ACCELERATION_MAX (currently == 65537, see lz4.c).
151</p></pre><BR>
152
153<pre><b>int LZ4_sizeofState(void);
154int LZ4_compress_fast_extState (void* state, const char* src, char* dst, int srcSize, int dstCapacity, int acceleration);
155</b><p>  Same as LZ4_compress_fast(), using an externally allocated memory space for its state.
156  Use LZ4_sizeofState() to know how much memory must be allocated,
157  and allocate it on 8-bytes boundaries (using `malloc()` typically).
158  Then, provide this buffer as `void* state` to compression function.
159
160</p></pre><BR>
161
162<pre><b>int LZ4_compress_destSize (const char* src, char* dst, int* srcSizePtr, int targetDstSize);
163</b><p>  Reverse the logic : compresses as much data as possible from 'src' buffer
164  into already allocated buffer 'dst', of size >= 'targetDestSize'.
165  This function either compresses the entire 'src' content into 'dst' if it's large enough,
166  or fill 'dst' buffer completely with as much data as possible from 'src'.
167  note: acceleration parameter is fixed to "default".
168
169 *srcSizePtr : will be modified to indicate how many bytes where read from 'src' to fill 'dst'.
170               New value is necessarily <= input value.
171 @return : Nb bytes written into 'dst' (necessarily <= targetDestSize)
172           or 0 if compression fails.
173
174 Note : from v1.8.2 to v1.9.1, this function had a bug (fixed un v1.9.2+):
175        the produced compressed content could, in specific circumstances,
176        require to be decompressed into a destination buffer larger
177        by at least 1 byte than the content to decompress.
178        If an application uses `LZ4_compress_destSize()`,
179        it's highly recommended to update liblz4 to v1.9.2 or better.
180        If this can't be done or ensured,
181        the receiving decompression function should provide
182        a dstCapacity which is > decompressedSize, by at least 1 byte.
183        See https://github.com/lz4/lz4/issues/859 for details
184
185</p></pre><BR>
186
187<pre><b>int LZ4_decompress_safe_partial (const char* src, char* dst, int srcSize, int targetOutputSize, int dstCapacity);
188</b><p>  Decompress an LZ4 compressed block, of size 'srcSize' at position 'src',
189  into destination buffer 'dst' of size 'dstCapacity'.
190  Up to 'targetOutputSize' bytes will be decoded.
191  The function stops decoding on reaching this objective.
192  This can be useful to boost performance
193  whenever only the beginning of a block is required.
194
195 @return : the number of bytes decoded in `dst` (necessarily <= targetOutputSize)
196           If source stream is detected malformed, function returns a negative result.
197
198  Note 1 : @return can be < targetOutputSize, if compressed block contains less data.
199
200  Note 2 : targetOutputSize must be <= dstCapacity
201
202  Note 3 : this function effectively stops decoding on reaching targetOutputSize,
203           so dstCapacity is kind of redundant.
204           This is because in older versions of this function,
205           decoding operation would still write complete sequences.
206           Therefore, there was no guarantee that it would stop writing at exactly targetOutputSize,
207           it could write more bytes, though only up to dstCapacity.
208           Some "margin" used to be required for this operation to work properly.
209           Thankfully, this is no longer necessary.
210           The function nonetheless keeps the same signature, in an effort to preserve API compatibility.
211
212  Note 4 : If srcSize is the exact size of the block,
213           then targetOutputSize can be any value,
214           including larger than the block's decompressed size.
215           The function will, at most, generate block's decompressed size.
216
217  Note 5 : If srcSize is _larger_ than block's compressed size,
218           then targetOutputSize **MUST** be <= block's decompressed size.
219           Otherwise, *silent corruption will occur*.
220
221</p></pre><BR>
222
223<a name="Chapter6"></a><h2>Streaming Compression Functions</h2><pre></pre>
224
225<pre><b>void LZ4_resetStream_fast (LZ4_stream_t* streamPtr);
226</b><p>  Use this to prepare an LZ4_stream_t for a new chain of dependent blocks
227  (e.g., LZ4_compress_fast_continue()).
228
229  An LZ4_stream_t must be initialized once before usage.
230  This is automatically done when created by LZ4_createStream().
231  However, should the LZ4_stream_t be simply declared on stack (for example),
232  it's necessary to initialize it first, using LZ4_initStream().
233
234  After init, start any new stream with LZ4_resetStream_fast().
235  A same LZ4_stream_t can be re-used multiple times consecutively
236  and compress multiple streams,
237  provided that it starts each new stream with LZ4_resetStream_fast().
238
239  LZ4_resetStream_fast() is much faster than LZ4_initStream(),
240  but is not compatible with memory regions containing garbage data.
241
242  Note: it's only useful to call LZ4_resetStream_fast()
243        in the context of streaming compression.
244        The *extState* functions perform their own resets.
245        Invoking LZ4_resetStream_fast() before is redundant, and even counterproductive.
246
247</p></pre><BR>
248
249<pre><b>int LZ4_loadDict (LZ4_stream_t* streamPtr, const char* dictionary, int dictSize);
250</b><p>  Use this function to reference a static dictionary into LZ4_stream_t.
251  The dictionary must remain available during compression.
252  LZ4_loadDict() triggers a reset, so any previous data will be forgotten.
253  The same dictionary will have to be loaded on decompression side for successful decoding.
254  Dictionary are useful for better compression of small data (KB range).
255  While LZ4 accept any input as dictionary,
256  results are generally better when using Zstandard's Dictionary Builder.
257  Loading a size of 0 is allowed, and is the same as reset.
258 @return : loaded dictionary size, in bytes (necessarily <= 64 KB)
259
260</p></pre><BR>
261
262<pre><b>int LZ4_compress_fast_continue (LZ4_stream_t* streamPtr, const char* src, char* dst, int srcSize, int dstCapacity, int acceleration);
263</b><p>  Compress 'src' content using data from previously compressed blocks, for better compression ratio.
264 'dst' buffer must be already allocated.
265  If dstCapacity >= LZ4_compressBound(srcSize), compression is guaranteed to succeed, and runs faster.
266
267 @return : size of compressed block
268           or 0 if there is an error (typically, cannot fit into 'dst').
269
270  Note 1 : Each invocation to LZ4_compress_fast_continue() generates a new block.
271           Each block has precise boundaries.
272           Each block must be decompressed separately, calling LZ4_decompress_*() with relevant metadata.
273           It's not possible to append blocks together and expect a single invocation of LZ4_decompress_*() to decompress them together.
274
275  Note 2 : The previous 64KB of source data is __assumed__ to remain present, unmodified, at same address in memory !
276
277  Note 3 : When input is structured as a double-buffer, each buffer can have any size, including < 64 KB.
278           Make sure that buffers are separated, by at least one byte.
279           This construction ensures that each block only depends on previous block.
280
281  Note 4 : If input buffer is a ring-buffer, it can have any size, including < 64 KB.
282
283  Note 5 : After an error, the stream status is undefined (invalid), it can only be reset or freed.
284
285</p></pre><BR>
286
287<pre><b>int LZ4_saveDict (LZ4_stream_t* streamPtr, char* safeBuffer, int maxDictSize);
288</b><p>  If last 64KB data cannot be guaranteed to remain available at its current memory location,
289  save it into a safer place (char* safeBuffer).
290  This is schematically equivalent to a memcpy() followed by LZ4_loadDict(),
291  but is much faster, because LZ4_saveDict() doesn't need to rebuild tables.
292 @return : saved dictionary size in bytes (necessarily <= maxDictSize), or 0 if error.
293
294</p></pre><BR>
295
296<a name="Chapter7"></a><h2>Streaming Decompression Functions</h2><pre>  Bufferless synchronous API
297<BR></pre>
298
299<pre><b>#if !defined(LZ4_STATIC_LINKING_ONLY_DISABLE_MEMORY_ALLOCATION)
300LZ4_streamDecode_t* LZ4_createStreamDecode(void);
301int                 LZ4_freeStreamDecode (LZ4_streamDecode_t* LZ4_stream);
302#endif </b>/* !defined(LZ4_STATIC_LINKING_ONLY_DISABLE_MEMORY_ALLOCATION) */<b>
303</b><p>  creation / destruction of streaming decompression tracking context.
304  A tracking context can be re-used multiple times.
305
306</p></pre><BR>
307
308<pre><b>int LZ4_setStreamDecode (LZ4_streamDecode_t* LZ4_streamDecode, const char* dictionary, int dictSize);
309</b><p>  An LZ4_streamDecode_t context can be allocated once and re-used multiple times.
310  Use this function to start decompression of a new stream of blocks.
311  A dictionary can optionally be set. Use NULL or size 0 for a reset order.
312  Dictionary is presumed stable : it must remain accessible and unmodified during next decompression.
313 @return : 1 if OK, 0 if error
314
315</p></pre><BR>
316
317<pre><b>int LZ4_decoderRingBufferSize(int maxBlockSize);
318#define LZ4_DECODER_RING_BUFFER_SIZE(maxBlockSize) (65536 + 14 + (maxBlockSize))  </b>/* for static allocation; maxBlockSize presumed valid */<b>
319</b><p>  Note : in a ring buffer scenario (optional),
320  blocks are presumed decompressed next to each other
321  up to the moment there is not enough remaining space for next block (remainingSize < maxBlockSize),
322  at which stage it resumes from beginning of ring buffer.
323  When setting such a ring buffer for streaming decompression,
324  provides the minimum size of this ring buffer
325  to be compatible with any source respecting maxBlockSize condition.
326 @return : minimum ring buffer size,
327           or 0 if there is an error (invalid maxBlockSize).
328
329</p></pre><BR>
330
331<pre><b>int
332LZ4_decompress_safe_continue (LZ4_streamDecode_t* LZ4_streamDecode,
333                        const char* src, char* dst,
334                        int srcSize, int dstCapacity);
335</b><p>  These decoding functions allow decompression of consecutive blocks in "streaming" mode.
336  A block is an unsplittable entity, it must be presented entirely to a decompression function.
337  Decompression functions only accepts one block at a time.
338  The last 64KB of previously decoded data *must* remain available and unmodified at the memory position where they were decoded.
339  If less than 64KB of data has been decoded, all the data must be present.
340
341  Special : if decompression side sets a ring buffer, it must respect one of the following conditions :
342  - Decompression buffer size is _at least_ LZ4_decoderRingBufferSize(maxBlockSize).
343    maxBlockSize is the maximum size of any single block. It can have any value > 16 bytes.
344    In which case, encoding and decoding buffers do not need to be synchronized.
345    Actually, data can be produced by any source compliant with LZ4 format specification, and respecting maxBlockSize.
346  - Synchronized mode :
347    Decompression buffer size is _exactly_ the same as compression buffer size,
348    and follows exactly same update rule (block boundaries at same positions),
349    and decoding function is provided with exact decompressed size of each block (exception for last block of the stream),
350    _then_ decoding & encoding ring buffer can have any size, including small ones ( < 64 KB).
351  - Decompression buffer is larger than encoding buffer, by a minimum of maxBlockSize more bytes.
352    In which case, encoding and decoding buffers do not need to be synchronized,
353    and encoding ring buffer can have any size, including small ones ( < 64 KB).
354
355  Whenever these conditions are not possible,
356  save the last 64KB of decoded data into a safe buffer where it can't be modified during decompression,
357  then indicate where this data is saved using LZ4_setStreamDecode(), before decompressing next block.
358</p></pre><BR>
359
360<pre><b>int
361LZ4_decompress_safe_usingDict(const char* src, char* dst,
362                              int srcSize, int dstCapacity,
363                              const char* dictStart, int dictSize);
364</b><p>  These decoding functions work the same as
365  a combination of LZ4_setStreamDecode() followed by LZ4_decompress_*_continue()
366  They are stand-alone, and don't need an LZ4_streamDecode_t structure.
367  Dictionary is presumed stable : it must remain accessible and unmodified during decompression.
368  Performance tip : Decompression speed can be substantially increased
369                    when dst == dictStart + dictSize.
370
371</p></pre><BR>
372
373<a name="Chapter8"></a><h2>Experimental section</h2><pre>
374 Symbols declared in this section must be considered unstable. Their
375 signatures or semantics may change, or they may be removed altogether in the
376 future. They are therefore only safe to depend on when the caller is
377 statically linked against the library.
378
379 To protect against unsafe usage, not only are the declarations guarded,
380 the definitions are hidden by default
381 when building LZ4 as a shared/dynamic library.
382
383 In order to access these declarations,
384 define LZ4_STATIC_LINKING_ONLY in your application
385 before including LZ4's headers.
386
387 In order to make their implementations accessible dynamically, you must
388 define LZ4_PUBLISH_STATIC_FUNCTIONS when building the LZ4 library.
389<BR></pre>
390
391<pre><b>LZ4LIB_STATIC_API int LZ4_compress_fast_extState_fastReset (void* state, const char* src, char* dst, int srcSize, int dstCapacity, int acceleration);
392</b><p>  A variant of LZ4_compress_fast_extState().
393
394  Using this variant avoids an expensive initialization step.
395  It is only safe to call if the state buffer is known to be correctly initialized already
396  (see above comment on LZ4_resetStream_fast() for a definition of "correctly initialized").
397  From a high level, the difference is that
398  this function initializes the provided state with a call to something like LZ4_resetStream_fast()
399  while LZ4_compress_fast_extState() starts with a call to LZ4_resetStream().
400
401</p></pre><BR>
402
403<pre><b>LZ4LIB_STATIC_API void
404LZ4_attach_dictionary(LZ4_stream_t* workingStream,
405                const LZ4_stream_t* dictionaryStream);
406</b><p>  This is an experimental API that allows
407  efficient use of a static dictionary many times.
408
409  Rather than re-loading the dictionary buffer into a working context before
410  each compression, or copying a pre-loaded dictionary's LZ4_stream_t into a
411  working LZ4_stream_t, this function introduces a no-copy setup mechanism,
412  in which the working stream references the dictionary stream in-place.
413
414  Several assumptions are made about the state of the dictionary stream.
415  Currently, only streams which have been prepared by LZ4_loadDict() should
416  be expected to work.
417
418  Alternatively, the provided dictionaryStream may be NULL,
419  in which case any existing dictionary stream is unset.
420
421  If a dictionary is provided, it replaces any pre-existing stream history.
422  The dictionary contents are the only history that can be referenced and
423  logically immediately precede the data compressed in the first subsequent
424  compression call.
425
426  The dictionary will only remain attached to the working stream through the
427  first compression call, at the end of which it is cleared. The dictionary
428  stream (and source buffer) must remain in-place / accessible / unchanged
429  through the completion of the first compression call on the stream.
430
431</p></pre><BR>
432
433<pre><b></b><p>
434 It's possible to have input and output sharing the same buffer,
435 for highly constrained memory environments.
436 In both cases, it requires input to lay at the end of the buffer,
437 and decompression to start at beginning of the buffer.
438 Buffer size must feature some margin, hence be larger than final size.
439
440 |<------------------------buffer--------------------------------->|
441                             |<-----------compressed data--------->|
442 |<-----------decompressed size------------------>|
443                                                  |<----margin---->|
444
445 This technique is more useful for decompression,
446 since decompressed size is typically larger,
447 and margin is short.
448
449 In-place decompression will work inside any buffer
450 which size is >= LZ4_DECOMPRESS_INPLACE_BUFFER_SIZE(decompressedSize).
451 This presumes that decompressedSize > compressedSize.
452 Otherwise, it means compression actually expanded data,
453 and it would be more efficient to store such data with a flag indicating it's not compressed.
454 This can happen when data is not compressible (already compressed, or encrypted).
455
456 For in-place compression, margin is larger, as it must be able to cope with both
457 history preservation, requiring input data to remain unmodified up to LZ4_DISTANCE_MAX,
458 and data expansion, which can happen when input is not compressible.
459 As a consequence, buffer size requirements are much higher,
460 and memory savings offered by in-place compression are more limited.
461
462 There are ways to limit this cost for compression :
463 - Reduce history size, by modifying LZ4_DISTANCE_MAX.
464   Note that it is a compile-time constant, so all compressions will apply this limit.
465   Lower values will reduce compression ratio, except when input_size < LZ4_DISTANCE_MAX,
466   so it's a reasonable trick when inputs are known to be small.
467 - Require the compressor to deliver a "maximum compressed size".
468   This is the `dstCapacity` parameter in `LZ4_compress*()`.
469   When this size is < LZ4_COMPRESSBOUND(inputSize), then compression can fail,
470   in which case, the return code will be 0 (zero).
471   The caller must be ready for these cases to happen,
472   and typically design a backup scheme to send data uncompressed.
473 The combination of both techniques can significantly reduce
474 the amount of margin required for in-place compression.
475
476 In-place compression can work in any buffer
477 which size is >= (maxCompressedSize)
478 with maxCompressedSize == LZ4_COMPRESSBOUND(srcSize) for guaranteed compression success.
479 LZ4_COMPRESS_INPLACE_BUFFER_SIZE() depends on both maxCompressedSize and LZ4_DISTANCE_MAX,
480 so it's possible to reduce memory requirements by playing with them.
481
482</p></pre><BR>
483
484<pre><b>#define LZ4_DECOMPRESS_INPLACE_BUFFER_SIZE(decompressedSize)   ((decompressedSize) + LZ4_DECOMPRESS_INPLACE_MARGIN(decompressedSize))  </b>/**< note: presumes that compressedSize < decompressedSize. note2: margin is overestimated a bit, since it could use compressedSize instead */<b>
485</b></pre><BR>
486<pre><b>#define LZ4_COMPRESS_INPLACE_BUFFER_SIZE(maxCompressedSize)   ((maxCompressedSize) + LZ4_COMPRESS_INPLACE_MARGIN)  </b>/**< maxCompressedSize is generally LZ4_COMPRESSBOUND(inputSize), but can be set to any lower value, with the risk that compression can fail (return code 0(zero)) */<b>
487</b></pre><BR>
488<a name="Chapter9"></a><h2>Private Definitions</h2><pre>
489 Do not use these definitions directly.
490 They are only exposed to allow static allocation of `LZ4_stream_t` and `LZ4_streamDecode_t`.
491 Accessing members will expose user code to API and/or ABI break in future versions of the library.
492<BR></pre>
493
494<pre><b></b><p>  Never ever use below internal definitions directly !
495  These definitions are not API/ABI safe, and may change in future versions.
496  If you need static allocation, declare or allocate an LZ4_stream_t object.
497</p></pre><BR>
498
499<pre><b>LZ4_stream_t* LZ4_initStream (void* buffer, size_t size);
500</b><p>  An LZ4_stream_t structure must be initialized at least once.
501  This is automatically done when invoking LZ4_createStream(),
502  but it's not when the structure is simply declared on stack (for example).
503
504  Use LZ4_initStream() to properly initialize a newly declared LZ4_stream_t.
505  It can also initialize any arbitrary buffer of sufficient size,
506  and will @return a pointer of proper type upon initialization.
507
508  Note : initialization fails if size and alignment conditions are not respected.
509         In which case, the function will @return NULL.
510  Note2: An LZ4_stream_t structure guarantees correct alignment and size.
511  Note3: Before v1.9.0, use LZ4_resetStream() instead
512</p></pre><BR>
513
514<pre><b>typedef struct {
515    const LZ4_byte* externalDict;
516    const LZ4_byte* prefixEnd;
517    size_t extDictSize;
518    size_t prefixSize;
519} LZ4_streamDecode_t_internal;
520</b><p>  Never ever use below internal definitions directly !
521  These definitions are not API/ABI safe, and may change in future versions.
522  If you need static allocation, declare or allocate an LZ4_streamDecode_t object.
523</p></pre><BR>
524
525<a name="Chapter10"></a><h2>Obsolete Functions</h2><pre></pre>
526
527<pre><b>#ifdef LZ4_DISABLE_DEPRECATE_WARNINGS
528#  define LZ4_DEPRECATED(message)   </b>/* disable deprecation warnings */<b>
529#else
530#  if defined (__cplusplus) && (__cplusplus >= 201402) </b>/* C++14 or greater */<b>
531#    define LZ4_DEPRECATED(message) [[deprecated(message)]]
532#  elif defined(_MSC_VER)
533#    define LZ4_DEPRECATED(message) __declspec(deprecated(message))
534#  elif defined(__clang__) || (defined(__GNUC__) && (__GNUC__ * 10 + __GNUC_MINOR__ >= 45))
535#    define LZ4_DEPRECATED(message) __attribute__((deprecated(message)))
536#  elif defined(__GNUC__) && (__GNUC__ * 10 + __GNUC_MINOR__ >= 31)
537#    define LZ4_DEPRECATED(message) __attribute__((deprecated))
538#  else
539#    pragma message("WARNING: LZ4_DEPRECATED needs custom implementation for this compiler")
540#    define LZ4_DEPRECATED(message)   </b>/* disabled */<b>
541#  endif
542#endif </b>/* LZ4_DISABLE_DEPRECATE_WARNINGS */<b>
543</b><p>
544  Deprecated functions make the compiler generate a warning when invoked.
545  This is meant to invite users to update their source code.
546  Should deprecation warnings be a problem, it is generally possible to disable them,
547  typically with -Wno-deprecated-declarations for gcc
548  or _CRT_SECURE_NO_WARNINGS in Visual.
549
550  Another method is to define LZ4_DISABLE_DEPRECATE_WARNINGS
551  before including the header file.
552
553</p></pre><BR>
554
555<pre><b>LZ4_DEPRECATED("use LZ4_compress_default() instead")       LZ4LIB_API int LZ4_compress               (const char* src, char* dest, int srcSize);
556LZ4_DEPRECATED("use LZ4_compress_default() instead")       LZ4LIB_API int LZ4_compress_limitedOutput (const char* src, char* dest, int srcSize, int maxOutputSize);
557LZ4_DEPRECATED("use LZ4_compress_fast_extState() instead") LZ4LIB_API int LZ4_compress_withState               (void* state, const char* source, char* dest, int inputSize);
558LZ4_DEPRECATED("use LZ4_compress_fast_extState() instead") LZ4LIB_API int LZ4_compress_limitedOutput_withState (void* state, const char* source, char* dest, int inputSize, int maxOutputSize);
559LZ4_DEPRECATED("use LZ4_compress_fast_continue() instead") LZ4LIB_API int LZ4_compress_continue                (LZ4_stream_t* LZ4_streamPtr, const char* source, char* dest, int inputSize);
560LZ4_DEPRECATED("use LZ4_compress_fast_continue() instead") LZ4LIB_API int LZ4_compress_limitedOutput_continue  (LZ4_stream_t* LZ4_streamPtr, const char* source, char* dest, int inputSize, int maxOutputSize);
561</b><p></p></pre><BR>
562
563<pre><b>LZ4_DEPRECATED("use LZ4_decompress_fast() instead") LZ4LIB_API int LZ4_uncompress (const char* source, char* dest, int outputSize);
564LZ4_DEPRECATED("use LZ4_decompress_safe() instead") LZ4LIB_API int LZ4_uncompress_unknownOutputSize (const char* source, char* dest, int isize, int maxOutputSize);
565</b><p></p></pre><BR>
566
567<pre><b>LZ4_DEPRECATED("use LZ4_decompress_safe_usingDict() instead") LZ4LIB_API int LZ4_decompress_safe_withPrefix64k (const char* src, char* dst, int compressedSize, int maxDstSize);
568LZ4_DEPRECATED("use LZ4_decompress_fast_usingDict() instead") LZ4LIB_API int LZ4_decompress_fast_withPrefix64k (const char* src, char* dst, int originalSize);
569</b><p></p></pre><BR>
570
571<pre><b>LZ4_DEPRECATED("This function is deprecated and unsafe. Consider using LZ4_decompress_safe() instead")
572int LZ4_decompress_fast (const char* src, char* dst, int originalSize);
573LZ4_DEPRECATED("This function is deprecated and unsafe. Consider using LZ4_decompress_safe_continue() instead")
574int LZ4_decompress_fast_continue (LZ4_streamDecode_t* LZ4_streamDecode, const char* src, char* dst, int originalSize);
575LZ4_DEPRECATED("This function is deprecated and unsafe. Consider using LZ4_decompress_safe_usingDict() instead")
576int LZ4_decompress_fast_usingDict (const char* src, char* dst, int originalSize, const char* dictStart, int dictSize);
577</b><p>  These functions used to be faster than LZ4_decompress_safe(),
578  but this is no longer the case. They are now slower.
579  This is because LZ4_decompress_fast() doesn't know the input size,
580  and therefore must progress more cautiously into the input buffer to not read beyond the end of block.
581  On top of that `LZ4_decompress_fast()` is not protected vs malformed or malicious inputs, making it a security liability.
582  As a consequence, LZ4_decompress_fast() is strongly discouraged, and deprecated.
583
584  The last remaining LZ4_decompress_fast() specificity is that
585  it can decompress a block without knowing its compressed size.
586  Such functionality can be achieved in a more secure manner
587  by employing LZ4_decompress_safe_partial().
588
589  Parameters:
590  originalSize : is the uncompressed size to regenerate.
591                 `dst` must be already allocated, its size must be >= 'originalSize' bytes.
592 @return : number of bytes read from source buffer (== compressed size).
593           The function expects to finish at block's end exactly.
594           If the source stream is detected malformed, the function stops decoding and returns a negative result.
595  note : LZ4_decompress_fast*() requires originalSize. Thanks to this information, it never writes past the output buffer.
596         However, since it doesn't know its 'src' size, it may read an unknown amount of input, past input buffer bounds.
597         Also, since match offsets are not validated, match reads from 'src' may underflow too.
598         These issues never happen if input (compressed) data is correct.
599         But they may happen if input data is invalid (error or intentional tampering).
600         As a consequence, use these functions in trusted environments with trusted data **only**.
601
602</p></pre><BR>
603
604<pre><b>void LZ4_resetStream (LZ4_stream_t* streamPtr);
605</b><p>  An LZ4_stream_t structure must be initialized at least once.
606  This is done with LZ4_initStream(), or LZ4_resetStream().
607  Consider switching to LZ4_initStream(),
608  invoking LZ4_resetStream() will trigger deprecation warnings in the future.
609
610</p></pre><BR>
611
612</html>
613</body>
614