... | ... |
@@ -1,72 +1,7 @@ |
1 |
-Original unmodified files from the LZMA SDK are included under libclamav/lzma. |
|
2 | 1 |
The original lzma.txt license file, a small portion of which is reproduced below, |
3 |
-is available under libclamav/lzma/lzma.txt. |
|
2 |
+is available under libclamav/7z/lzma.txt. |
|
4 | 3 |
|
5 | 4 |
LICENSE |
6 | 5 |
------- |
7 | 6 |
|
8 |
-LZMA SDK is available under any of the following licenses: |
|
9 |
- |
|
10 |
-1) GNU Lesser General Public License (GNU LGPL) |
|
11 |
-2) Common Public License (CPL) |
|
12 |
-3) Simplified license for unmodified code (read SPECIAL EXCEPTION) |
|
13 |
-4) Proprietary license |
|
14 |
- |
|
15 |
-It means that you can select one of these four options and follow rules of that license. |
|
16 |
- |
|
17 |
- |
|
18 |
-1,2) GNU LGPL and CPL licenses are pretty similar and both these |
|
19 |
-licenses are classified as |
|
20 |
- - "Free software licenses" at http://www.gnu.org/ |
|
21 |
- - "OSI-approved" at http://www.opensource.org/ |
|
22 |
- |
|
23 |
- |
|
24 |
-3) SPECIAL EXCEPTION |
|
25 |
- |
|
26 |
-Igor Pavlov, as the author of this code, expressly permits you |
|
27 |
-to statically or dynamically link your code (or bind by name) |
|
28 |
-to the files from LZMA SDK without subjecting your linked |
|
29 |
-code to the terms of the CPL or GNU LGPL. |
|
30 |
-Any modifications or additions to files from LZMA SDK, however, |
|
31 |
-are subject to the GNU LGPL or CPL terms. |
|
32 |
- |
|
33 |
-SPECIAL EXCEPTION allows you to use LZMA SDK in applications with closed code, |
|
34 |
-while you keep LZMA SDK code unmodified. |
|
35 |
- |
|
36 |
- |
|
37 |
-SPECIAL EXCEPTION #2: Igor Pavlov, as the author of this code, expressly permits |
|
38 |
-you to use this code under the same terms and conditions contained in the License |
|
39 |
-Agreement you have for any previous version of LZMA SDK developed by Igor Pavlov. |
|
40 |
- |
|
41 |
-SPECIAL EXCEPTION #2 allows owners of proprietary licenses to use latest version |
|
42 |
-of LZMA SDK as update for previous versions. |
|
43 |
- |
|
44 |
- |
|
45 |
-SPECIAL EXCEPTION #3: Igor Pavlov, as the author of this code, expressly permits |
|
46 |
-you to use code of the following files: |
|
47 |
-BranchTypes.h, LzmaTypes.h, LzmaTest.c, LzmaStateTest.c, LzmaAlone.cpp, |
|
48 |
-LzmaAlone.cs, LzmaAlone.java |
|
49 |
-as public domain code. |
|
50 |
- |
|
51 |
- |
|
52 |
-4) Proprietary license |
|
53 |
- |
|
54 |
-LZMA SDK also can be available under a proprietary license which |
|
55 |
-can include: |
|
56 |
- |
|
57 |
-1) Right to modify code without subjecting modified code to the |
|
58 |
-terms of the CPL or GNU LGPL |
|
59 |
-2) Technical support for code |
|
60 |
- |
|
61 |
-To request such proprietary license or any additional consultations, |
|
62 |
-send email message from that page: |
|
63 |
-http://www.7-zip.org/support.html |
|
64 |
- |
|
65 |
- |
|
66 |
-You should have received a copy of the GNU Lesser General Public |
|
67 |
-License along with this library; if not, write to the Free Software |
|
68 |
-Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA |
|
69 |
- |
|
70 |
-You should have received a copy of the Common Public License |
|
71 |
-along with this library. |
|
72 |
- |
|
7 |
+LZMA SDK is written and placed in the public domain by Igor Pavlov. |
73 | 8 |
new file mode 100644 |
... | ... |
@@ -0,0 +1,594 @@ |
0 |
+LZMA SDK 4.65 |
|
1 |
+------------- |
|
2 |
+ |
|
3 |
+LZMA SDK provides the documentation, samples, header files, libraries, |
|
4 |
+and tools you need to develop applications that use LZMA compression. |
|
5 |
+ |
|
6 |
+LZMA is default and general compression method of 7z format |
|
7 |
+in 7-Zip compression program (www.7-zip.org). LZMA provides high |
|
8 |
+compression ratio and very fast decompression. |
|
9 |
+ |
|
10 |
+LZMA is an improved version of famous LZ77 compression algorithm. |
|
11 |
+It was improved in way of maximum increasing of compression ratio, |
|
12 |
+keeping high decompression speed and low memory requirements for |
|
13 |
+decompressing. |
|
14 |
+ |
|
15 |
+ |
|
16 |
+ |
|
17 |
+LICENSE |
|
18 |
+------- |
|
19 |
+ |
|
20 |
+LZMA SDK is written and placed in the public domain by Igor Pavlov. |
|
21 |
+ |
|
22 |
+ |
|
23 |
+LZMA SDK Contents |
|
24 |
+----------------- |
|
25 |
+ |
|
26 |
+LZMA SDK includes: |
|
27 |
+ |
|
28 |
+ - ANSI-C/C++/C#/Java source code for LZMA compressing and decompressing |
|
29 |
+ - Compiled file->file LZMA compressing/decompressing program for Windows system |
|
30 |
+ |
|
31 |
+ |
|
32 |
+UNIX/Linux version |
|
33 |
+------------------ |
|
34 |
+To compile C++ version of file->file LZMA encoding, go to directory |
|
35 |
+C++/7zip/Compress/LZMA_Alone |
|
36 |
+and call make to recompile it: |
|
37 |
+ make -f makefile.gcc clean all |
|
38 |
+ |
|
39 |
+In some UNIX/Linux versions you must compile LZMA with static libraries. |
|
40 |
+To compile with static libraries, you can use |
|
41 |
+LIB = -lm -static |
|
42 |
+ |
|
43 |
+ |
|
44 |
+Files |
|
45 |
+--------------------- |
|
46 |
+lzma.txt - LZMA SDK description (this file) |
|
47 |
+7zFormat.txt - 7z Format description |
|
48 |
+7zC.txt - 7z ANSI-C Decoder description |
|
49 |
+methods.txt - Compression method IDs for .7z |
|
50 |
+lzma.exe - Compiled file->file LZMA encoder/decoder for Windows |
|
51 |
+history.txt - history of the LZMA SDK |
|
52 |
+ |
|
53 |
+ |
|
54 |
+Source code structure |
|
55 |
+--------------------- |
|
56 |
+ |
|
57 |
+C/ - C files |
|
58 |
+ 7zCrc*.* - CRC code |
|
59 |
+ Alloc.* - Memory allocation functions |
|
60 |
+ Bra*.* - Filters for x86, IA-64, ARM, ARM-Thumb, PowerPC and SPARC code |
|
61 |
+ LzFind.* - Match finder for LZ (LZMA) encoders |
|
62 |
+ LzFindMt.* - Match finder for LZ (LZMA) encoders for multithreading encoding |
|
63 |
+ LzHash.h - Additional file for LZ match finder |
|
64 |
+ LzmaDec.* - LZMA decoding |
|
65 |
+ LzmaEnc.* - LZMA encoding |
|
66 |
+ LzmaLib.* - LZMA Library for DLL calling |
|
67 |
+ Types.h - Basic types for another .c files |
|
68 |
+ Threads.* - The code for multithreading. |
|
69 |
+ |
|
70 |
+ LzmaLib - LZMA Library (.DLL for Windows) |
|
71 |
+ |
|
72 |
+ LzmaUtil - LZMA Utility (file->file LZMA encoder/decoder). |
|
73 |
+ |
|
74 |
+ Archive - files related to archiving |
|
75 |
+ 7z - 7z ANSI-C Decoder |
|
76 |
+ |
|
77 |
+CPP/ -- CPP files |
|
78 |
+ |
|
79 |
+ Common - common files for C++ projects |
|
80 |
+ Windows - common files for Windows related code |
|
81 |
+ |
|
82 |
+ 7zip - files related to 7-Zip Project |
|
83 |
+ |
|
84 |
+ Common - common files for 7-Zip |
|
85 |
+ |
|
86 |
+ Compress - files related to compression/decompression |
|
87 |
+ |
|
88 |
+ Copy - Copy coder |
|
89 |
+ RangeCoder - Range Coder (special code of compression/decompression) |
|
90 |
+ LZMA - LZMA compression/decompression on C++ |
|
91 |
+ LZMA_Alone - file->file LZMA compression/decompression |
|
92 |
+ Branch - Filters for x86, IA-64, ARM, ARM-Thumb, PowerPC and SPARC code |
|
93 |
+ |
|
94 |
+ Archive - files related to archiving |
|
95 |
+ |
|
96 |
+ Common - common files for archive handling |
|
97 |
+ 7z - 7z C++ Encoder/Decoder |
|
98 |
+ |
|
99 |
+ Bundles - Modules that are bundles of other modules |
|
100 |
+ |
|
101 |
+ Alone7z - 7zr.exe: Standalone version of 7z.exe that supports only 7z/LZMA/BCJ/BCJ2 |
|
102 |
+ Format7zR - 7zr.dll: Reduced version of 7za.dll: extracting/compressing to 7z/LZMA/BCJ/BCJ2 |
|
103 |
+ Format7zExtractR - 7zxr.dll: Reduced version of 7zxa.dll: extracting from 7z/LZMA/BCJ/BCJ2. |
|
104 |
+ |
|
105 |
+ UI - User Interface files |
|
106 |
+ |
|
107 |
+ Client7z - Test application for 7za.dll, 7zr.dll, 7zxr.dll |
|
108 |
+ Common - Common UI files |
|
109 |
+ Console - Code for console archiver |
|
110 |
+ |
|
111 |
+ |
|
112 |
+ |
|
113 |
+CS/ - C# files |
|
114 |
+ 7zip |
|
115 |
+ Common - some common files for 7-Zip |
|
116 |
+ Compress - files related to compression/decompression |
|
117 |
+ LZ - files related to LZ (Lempel-Ziv) compression algorithm |
|
118 |
+ LZMA - LZMA compression/decompression |
|
119 |
+ LzmaAlone - file->file LZMA compression/decompression |
|
120 |
+ RangeCoder - Range Coder (special code of compression/decompression) |
|
121 |
+ |
|
122 |
+Java/ - Java files |
|
123 |
+ SevenZip |
|
124 |
+ Compression - files related to compression/decompression |
|
125 |
+ LZ - files related to LZ (Lempel-Ziv) compression algorithm |
|
126 |
+ LZMA - LZMA compression/decompression |
|
127 |
+ RangeCoder - Range Coder (special code of compression/decompression) |
|
128 |
+ |
|
129 |
+ |
|
130 |
+C/C++ source code of LZMA SDK is part of 7-Zip project. |
|
131 |
+7-Zip source code can be downloaded from 7-Zip's SourceForge page: |
|
132 |
+ |
|
133 |
+ http://sourceforge.net/projects/sevenzip/ |
|
134 |
+ |
|
135 |
+ |
|
136 |
+ |
|
137 |
+LZMA features |
|
138 |
+------------- |
|
139 |
+ - Variable dictionary size (up to 1 GB) |
|
140 |
+ - Estimated compressing speed: about 2 MB/s on 2 GHz CPU |
|
141 |
+ - Estimated decompressing speed: |
|
142 |
+ - 20-30 MB/s on 2 GHz Core 2 or AMD Athlon 64 |
|
143 |
+ - 1-2 MB/s on 200 MHz ARM, MIPS, PowerPC or other simple RISC |
|
144 |
+ - Small memory requirements for decompressing (16 KB + DictionarySize) |
|
145 |
+ - Small code size for decompressing: 5-8 KB |
|
146 |
+ |
|
147 |
+LZMA decoder uses only integer operations and can be |
|
148 |
+implemented in any modern 32-bit CPU (or on 16-bit CPU with some conditions). |
|
149 |
+ |
|
150 |
+Some critical operations that affect the speed of LZMA decompression: |
|
151 |
+ 1) 32*16 bit integer multiply |
|
152 |
+ 2) Misspredicted branches (penalty mostly depends from pipeline length) |
|
153 |
+ 3) 32-bit shift and arithmetic operations |
|
154 |
+ |
|
155 |
+The speed of LZMA decompressing mostly depends from CPU speed. |
|
156 |
+Memory speed has no big meaning. But if your CPU has small data cache, |
|
157 |
+overall weight of memory speed will slightly increase. |
|
158 |
+ |
|
159 |
+ |
|
160 |
+How To Use |
|
161 |
+---------- |
|
162 |
+ |
|
163 |
+Using LZMA encoder/decoder executable |
|
164 |
+-------------------------------------- |
|
165 |
+ |
|
166 |
+Usage: LZMA <e|d> inputFile outputFile [<switches>...] |
|
167 |
+ |
|
168 |
+ e: encode file |
|
169 |
+ |
|
170 |
+ d: decode file |
|
171 |
+ |
|
172 |
+ b: Benchmark. There are two tests: compressing and decompressing |
|
173 |
+ with LZMA method. Benchmark shows rating in MIPS (million |
|
174 |
+ instructions per second). Rating value is calculated from |
|
175 |
+ measured speed and it is normalized with Intel's Core 2 results. |
|
176 |
+ Also Benchmark checks possible hardware errors (RAM |
|
177 |
+ errors in most cases). Benchmark uses these settings: |
|
178 |
+ (-a1, -d21, -fb32, -mfbt4). You can change only -d parameter. |
|
179 |
+ Also you can change the number of iterations. Example for 30 iterations: |
|
180 |
+ LZMA b 30 |
|
181 |
+ Default number of iterations is 10. |
|
182 |
+ |
|
183 |
+<Switches> |
|
184 |
+ |
|
185 |
+ |
|
186 |
+ -a{N}: set compression mode 0 = fast, 1 = normal |
|
187 |
+ default: 1 (normal) |
|
188 |
+ |
|
189 |
+ d{N}: Sets Dictionary size - [0, 30], default: 23 (8MB) |
|
190 |
+ The maximum value for dictionary size is 1 GB = 2^30 bytes. |
|
191 |
+ Dictionary size is calculated as DictionarySize = 2^N bytes. |
|
192 |
+ For decompressing file compressed by LZMA method with dictionary |
|
193 |
+ size D = 2^N you need about D bytes of memory (RAM). |
|
194 |
+ |
|
195 |
+ -fb{N}: set number of fast bytes - [5, 273], default: 128 |
|
196 |
+ Usually big number gives a little bit better compression ratio |
|
197 |
+ and slower compression process. |
|
198 |
+ |
|
199 |
+ -lc{N}: set number of literal context bits - [0, 8], default: 3 |
|
200 |
+ Sometimes lc=4 gives gain for big files. |
|
201 |
+ |
|
202 |
+ -lp{N}: set number of literal pos bits - [0, 4], default: 0 |
|
203 |
+ lp switch is intended for periodical data when period is |
|
204 |
+ equal 2^N. For example, for 32-bit (4 bytes) |
|
205 |
+ periodical data you can use lp=2. Often it's better to set lc0, |
|
206 |
+ if you change lp switch. |
|
207 |
+ |
|
208 |
+ -pb{N}: set number of pos bits - [0, 4], default: 2 |
|
209 |
+ pb switch is intended for periodical data |
|
210 |
+ when period is equal 2^N. |
|
211 |
+ |
|
212 |
+ -mf{MF_ID}: set Match Finder. Default: bt4. |
|
213 |
+ Algorithms from hc* group doesn't provide good compression |
|
214 |
+ ratio, but they often works pretty fast in combination with |
|
215 |
+ fast mode (-a0). |
|
216 |
+ |
|
217 |
+ Memory requirements depend from dictionary size |
|
218 |
+ (parameter "d" in table below). |
|
219 |
+ |
|
220 |
+ MF_ID Memory Description |
|
221 |
+ |
|
222 |
+ bt2 d * 9.5 + 4MB Binary Tree with 2 bytes hashing. |
|
223 |
+ bt3 d * 11.5 + 4MB Binary Tree with 3 bytes hashing. |
|
224 |
+ bt4 d * 11.5 + 4MB Binary Tree with 4 bytes hashing. |
|
225 |
+ hc4 d * 7.5 + 4MB Hash Chain with 4 bytes hashing. |
|
226 |
+ |
|
227 |
+ -eos: write End Of Stream marker. By default LZMA doesn't write |
|
228 |
+ eos marker, since LZMA decoder knows uncompressed size |
|
229 |
+ stored in .lzma file header. |
|
230 |
+ |
|
231 |
+ -si: Read data from stdin (it will write End Of Stream marker). |
|
232 |
+ -so: Write data to stdout |
|
233 |
+ |
|
234 |
+ |
|
235 |
+Examples: |
|
236 |
+ |
|
237 |
+1) LZMA e file.bin file.lzma -d16 -lc0 |
|
238 |
+ |
|
239 |
+compresses file.bin to file.lzma with 64 KB dictionary (2^16=64K) |
|
240 |
+and 0 literal context bits. -lc0 allows to reduce memory requirements |
|
241 |
+for decompression. |
|
242 |
+ |
|
243 |
+ |
|
244 |
+2) LZMA e file.bin file.lzma -lc0 -lp2 |
|
245 |
+ |
|
246 |
+compresses file.bin to file.lzma with settings suitable |
|
247 |
+for 32-bit periodical data (for example, ARM or MIPS code). |
|
248 |
+ |
|
249 |
+3) LZMA d file.lzma file.bin |
|
250 |
+ |
|
251 |
+decompresses file.lzma to file.bin. |
|
252 |
+ |
|
253 |
+ |
|
254 |
+Compression ratio hints |
|
255 |
+----------------------- |
|
256 |
+ |
|
257 |
+Recommendations |
|
258 |
+--------------- |
|
259 |
+ |
|
260 |
+To increase the compression ratio for LZMA compressing it's desirable |
|
261 |
+to have aligned data (if it's possible) and also it's desirable to locate |
|
262 |
+data in such order, where code is grouped in one place and data is |
|
263 |
+grouped in other place (it's better than such mixing: code, data, code, |
|
264 |
+data, ...). |
|
265 |
+ |
|
266 |
+ |
|
267 |
+Filters |
|
268 |
+------- |
|
269 |
+You can increase the compression ratio for some data types, using |
|
270 |
+special filters before compressing. For example, it's possible to |
|
271 |
+increase the compression ratio on 5-10% for code for those CPU ISAs: |
|
272 |
+x86, IA-64, ARM, ARM-Thumb, PowerPC, SPARC. |
|
273 |
+ |
|
274 |
+You can find C source code of such filters in C/Bra*.* files |
|
275 |
+ |
|
276 |
+You can check the compression ratio gain of these filters with such |
|
277 |
+7-Zip commands (example for ARM code): |
|
278 |
+No filter: |
|
279 |
+ 7z a a1.7z a.bin -m0=lzma |
|
280 |
+ |
|
281 |
+With filter for little-endian ARM code: |
|
282 |
+ 7z a a2.7z a.bin -m0=arm -m1=lzma |
|
283 |
+ |
|
284 |
+It works in such manner: |
|
285 |
+Compressing = Filter_encoding + LZMA_encoding |
|
286 |
+Decompressing = LZMA_decoding + Filter_decoding |
|
287 |
+ |
|
288 |
+Compressing and decompressing speed of such filters is very high, |
|
289 |
+so it will not increase decompressing time too much. |
|
290 |
+Moreover, it reduces decompression time for LZMA_decoding, |
|
291 |
+since compression ratio with filtering is higher. |
|
292 |
+ |
|
293 |
+These filters convert CALL (calling procedure) instructions |
|
294 |
+from relative offsets to absolute addresses, so such data becomes more |
|
295 |
+compressible. |
|
296 |
+ |
|
297 |
+For some ISAs (for example, for MIPS) it's impossible to get gain from such filter. |
|
298 |
+ |
|
299 |
+ |
|
300 |
+LZMA compressed file format |
|
301 |
+--------------------------- |
|
302 |
+Offset Size Description |
|
303 |
+ 0 1 Special LZMA properties (lc,lp, pb in encoded form) |
|
304 |
+ 1 4 Dictionary size (little endian) |
|
305 |
+ 5 8 Uncompressed size (little endian). -1 means unknown size |
|
306 |
+ 13 Compressed data |
|
307 |
+ |
|
308 |
+ |
|
309 |
+ANSI-C LZMA Decoder |
|
310 |
+~~~~~~~~~~~~~~~~~~~ |
|
311 |
+ |
|
312 |
+Please note that interfaces for ANSI-C code were changed in LZMA SDK 4.58. |
|
313 |
+If you want to use old interfaces you can download previous version of LZMA SDK |
|
314 |
+from sourceforge.net site. |
|
315 |
+ |
|
316 |
+To use ANSI-C LZMA Decoder you need the following files: |
|
317 |
+1) LzmaDec.h + LzmaDec.c + Types.h |
|
318 |
+LzmaUtil/LzmaUtil.c is example application that uses these files. |
|
319 |
+ |
|
320 |
+ |
|
321 |
+Memory requirements for LZMA decoding |
|
322 |
+------------------------------------- |
|
323 |
+ |
|
324 |
+Stack usage of LZMA decoding function for local variables is not |
|
325 |
+larger than 200-400 bytes. |
|
326 |
+ |
|
327 |
+LZMA Decoder uses dictionary buffer and internal state structure. |
|
328 |
+Internal state structure consumes |
|
329 |
+ state_size = (4 + (1.5 << (lc + lp))) KB |
|
330 |
+by default (lc=3, lp=0), state_size = 16 KB. |
|
331 |
+ |
|
332 |
+ |
|
333 |
+How To decompress data |
|
334 |
+---------------------- |
|
335 |
+ |
|
336 |
+LZMA Decoder (ANSI-C version) now supports 2 interfaces: |
|
337 |
+1) Single-call Decompressing |
|
338 |
+2) Multi-call State Decompressing (zlib-like interface) |
|
339 |
+ |
|
340 |
+You must use external allocator: |
|
341 |
+Example: |
|
342 |
+void *SzAlloc(void *p, size_t size) { p = p; return malloc(size); } |
|
343 |
+void SzFree(void *p, void *address) { p = p; free(address); } |
|
344 |
+ISzAlloc alloc = { SzAlloc, SzFree }; |
|
345 |
+ |
|
346 |
+You can use p = p; operator to disable compiler warnings. |
|
347 |
+ |
|
348 |
+ |
|
349 |
+Single-call Decompressing |
|
350 |
+------------------------- |
|
351 |
+When to use: RAM->RAM decompressing |
|
352 |
+Compile files: LzmaDec.h + LzmaDec.c + Types.h |
|
353 |
+Compile defines: no defines |
|
354 |
+Memory Requirements: |
|
355 |
+ - Input buffer: compressed size |
|
356 |
+ - Output buffer: uncompressed size |
|
357 |
+ - LZMA Internal Structures: state_size (16 KB for default settings) |
|
358 |
+ |
|
359 |
+Interface: |
|
360 |
+ int LzmaDecode(Byte *dest, SizeT *destLen, const Byte *src, SizeT *srcLen, |
|
361 |
+ const Byte *propData, unsigned propSize, ELzmaFinishMode finishMode, |
|
362 |
+ ELzmaStatus *status, ISzAlloc *alloc); |
|
363 |
+ In: |
|
364 |
+ dest - output data |
|
365 |
+ destLen - output data size |
|
366 |
+ src - input data |
|
367 |
+ srcLen - input data size |
|
368 |
+ propData - LZMA properties (5 bytes) |
|
369 |
+ propSize - size of propData buffer (5 bytes) |
|
370 |
+ finishMode - It has meaning only if the decoding reaches output limit (*destLen). |
|
371 |
+ LZMA_FINISH_ANY - Decode just destLen bytes. |
|
372 |
+ LZMA_FINISH_END - Stream must be finished after (*destLen). |
|
373 |
+ You can use LZMA_FINISH_END, when you know that |
|
374 |
+ current output buffer covers last bytes of stream. |
|
375 |
+ alloc - Memory allocator. |
|
376 |
+ |
|
377 |
+ Out: |
|
378 |
+ destLen - processed output size |
|
379 |
+ srcLen - processed input size |
|
380 |
+ |
|
381 |
+ Output: |
|
382 |
+ SZ_OK |
|
383 |
+ status: |
|
384 |
+ LZMA_STATUS_FINISHED_WITH_MARK |
|
385 |
+ LZMA_STATUS_NOT_FINISHED |
|
386 |
+ LZMA_STATUS_MAYBE_FINISHED_WITHOUT_MARK |
|
387 |
+ SZ_ERROR_DATA - Data error |
|
388 |
+ SZ_ERROR_MEM - Memory allocation error |
|
389 |
+ SZ_ERROR_UNSUPPORTED - Unsupported properties |
|
390 |
+ SZ_ERROR_INPUT_EOF - It needs more bytes in input buffer (src). |
|
391 |
+ |
|
392 |
+ If LZMA decoder sees end_marker before reaching output limit, it returns OK result, |
|
393 |
+ and output value of destLen will be less than output buffer size limit. |
|
394 |
+ |
|
395 |
+ You can use multiple checks to test data integrity after full decompression: |
|
396 |
+ 1) Check Result and "status" variable. |
|
397 |
+ 2) Check that output(destLen) = uncompressedSize, if you know real uncompressedSize. |
|
398 |
+ 3) Check that output(srcLen) = compressedSize, if you know real compressedSize. |
|
399 |
+ You must use correct finish mode in that case. */ |
|
400 |
+ |
|
401 |
+ |
|
402 |
+Multi-call State Decompressing (zlib-like interface) |
|
403 |
+---------------------------------------------------- |
|
404 |
+ |
|
405 |
+When to use: file->file decompressing |
|
406 |
+Compile files: LzmaDec.h + LzmaDec.c + Types.h |
|
407 |
+ |
|
408 |
+Memory Requirements: |
|
409 |
+ - Buffer for input stream: any size (for example, 16 KB) |
|
410 |
+ - Buffer for output stream: any size (for example, 16 KB) |
|
411 |
+ - LZMA Internal Structures: state_size (16 KB for default settings) |
|
412 |
+ - LZMA dictionary (dictionary size is encoded in LZMA properties header) |
|
413 |
+ |
|
414 |
+1) read LZMA properties (5 bytes) and uncompressed size (8 bytes, little-endian) to header: |
|
415 |
+ unsigned char header[LZMA_PROPS_SIZE + 8]; |
|
416 |
+ ReadFile(inFile, header, sizeof(header) |
|
417 |
+ |
|
418 |
+2) Allocate CLzmaDec structures (state + dictionary) using LZMA properties |
|
419 |
+ |
|
420 |
+ CLzmaDec state; |
|
421 |
+ LzmaDec_Constr(&state); |
|
422 |
+ res = LzmaDec_Allocate(&state, header, LZMA_PROPS_SIZE, &g_Alloc); |
|
423 |
+ if (res != SZ_OK) |
|
424 |
+ return res; |
|
425 |
+ |
|
426 |
+3) Init LzmaDec structure before any new LZMA stream. And call LzmaDec_DecodeToBuf in loop |
|
427 |
+ |
|
428 |
+ LzmaDec_Init(&state); |
|
429 |
+ for (;;) |
|
430 |
+ { |
|
431 |
+ ... |
|
432 |
+ int res = LzmaDec_DecodeToBuf(CLzmaDec *p, Byte *dest, SizeT *destLen, |
|
433 |
+ const Byte *src, SizeT *srcLen, ELzmaFinishMode finishMode); |
|
434 |
+ ... |
|
435 |
+ } |
|
436 |
+ |
|
437 |
+ |
|
438 |
+4) Free all allocated structures |
|
439 |
+ LzmaDec_Free(&state, &g_Alloc); |
|
440 |
+ |
|
441 |
+For full code example, look at C/LzmaUtil/LzmaUtil.c code. |
|
442 |
+ |
|
443 |
+ |
|
444 |
+How To compress data |
|
445 |
+-------------------- |
|
446 |
+ |
|
447 |
+Compile files: LzmaEnc.h + LzmaEnc.c + Types.h + |
|
448 |
+LzFind.c + LzFind.h + LzFindMt.c + LzFindMt.h + LzHash.h |
|
449 |
+ |
|
450 |
+Memory Requirements: |
|
451 |
+ - (dictSize * 11.5 + 6 MB) + state_size |
|
452 |
+ |
|
453 |
+Lzma Encoder can use two memory allocators: |
|
454 |
+1) alloc - for small arrays. |
|
455 |
+2) allocBig - for big arrays. |
|
456 |
+ |
|
457 |
+For example, you can use Large RAM Pages (2 MB) in allocBig allocator for |
|
458 |
+better compression speed. Note that Windows has bad implementation for |
|
459 |
+Large RAM Pages. |
|
460 |
+It's OK to use same allocator for alloc and allocBig. |
|
461 |
+ |
|
462 |
+ |
|
463 |
+Single-call Compression with callbacks |
|
464 |
+-------------------------------------- |
|
465 |
+ |
|
466 |
+Check C/LzmaUtil/LzmaUtil.c as example, |
|
467 |
+ |
|
468 |
+When to use: file->file decompressing |
|
469 |
+ |
|
470 |
+1) you must implement callback structures for interfaces: |
|
471 |
+ISeqInStream |
|
472 |
+ISeqOutStream |
|
473 |
+ICompressProgress |
|
474 |
+ISzAlloc |
|
475 |
+ |
|
476 |
+static void *SzAlloc(void *p, size_t size) { p = p; return MyAlloc(size); } |
|
477 |
+static void SzFree(void *p, void *address) { p = p; MyFree(address); } |
|
478 |
+static ISzAlloc g_Alloc = { SzAlloc, SzFree }; |
|
479 |
+ |
|
480 |
+ CFileSeqInStream inStream; |
|
481 |
+ CFileSeqOutStream outStream; |
|
482 |
+ |
|
483 |
+ inStream.funcTable.Read = MyRead; |
|
484 |
+ inStream.file = inFile; |
|
485 |
+ outStream.funcTable.Write = MyWrite; |
|
486 |
+ outStream.file = outFile; |
|
487 |
+ |
|
488 |
+ |
|
489 |
+2) Create CLzmaEncHandle object; |
|
490 |
+ |
|
491 |
+ CLzmaEncHandle enc; |
|
492 |
+ |
|
493 |
+ enc = LzmaEnc_Create(&g_Alloc); |
|
494 |
+ if (enc == 0) |
|
495 |
+ return SZ_ERROR_MEM; |
|
496 |
+ |
|
497 |
+ |
|
498 |
+3) initialize CLzmaEncProps properties; |
|
499 |
+ |
|
500 |
+ LzmaEncProps_Init(&props); |
|
501 |
+ |
|
502 |
+ Then you can change some properties in that structure. |
|
503 |
+ |
|
504 |
+4) Send LZMA properties to LZMA Encoder |
|
505 |
+ |
|
506 |
+ res = LzmaEnc_SetProps(enc, &props); |
|
507 |
+ |
|
508 |
+5) Write encoded properties to header |
|
509 |
+ |
|
510 |
+ Byte header[LZMA_PROPS_SIZE + 8]; |
|
511 |
+ size_t headerSize = LZMA_PROPS_SIZE; |
|
512 |
+ UInt64 fileSize; |
|
513 |
+ int i; |
|
514 |
+ |
|
515 |
+ res = LzmaEnc_WriteProperties(enc, header, &headerSize); |
|
516 |
+ fileSize = MyGetFileLength(inFile); |
|
517 |
+ for (i = 0; i < 8; i++) |
|
518 |
+ header[headerSize++] = (Byte)(fileSize >> (8 * i)); |
|
519 |
+ MyWriteFileAndCheck(outFile, header, headerSize) |
|
520 |
+ |
|
521 |
+6) Call encoding function: |
|
522 |
+ res = LzmaEnc_Encode(enc, &outStream.funcTable, &inStream.funcTable, |
|
523 |
+ NULL, &g_Alloc, &g_Alloc); |
|
524 |
+ |
|
525 |
+7) Destroy LZMA Encoder Object |
|
526 |
+ LzmaEnc_Destroy(enc, &g_Alloc, &g_Alloc); |
|
527 |
+ |
|
528 |
+ |
|
529 |
+If callback function return some error code, LzmaEnc_Encode also returns that code. |
|
530 |
+ |
|
531 |
+ |
|
532 |
+Single-call RAM->RAM Compression |
|
533 |
+-------------------------------- |
|
534 |
+ |
|
535 |
+Single-call RAM->RAM Compression is similar to Compression with callbacks, |
|
536 |
+but you provide pointers to buffers instead of pointers to stream callbacks: |
|
537 |
+ |
|
538 |
+HRes LzmaEncode(Byte *dest, SizeT *destLen, const Byte *src, SizeT srcLen, |
|
539 |
+ CLzmaEncProps *props, Byte *propsEncoded, SizeT *propsSize, int writeEndMark, |
|
540 |
+ ICompressProgress *progress, ISzAlloc *alloc, ISzAlloc *allocBig); |
|
541 |
+ |
|
542 |
+Return code: |
|
543 |
+ SZ_OK - OK |
|
544 |
+ SZ_ERROR_MEM - Memory allocation error |
|
545 |
+ SZ_ERROR_PARAM - Incorrect paramater |
|
546 |
+ SZ_ERROR_OUTPUT_EOF - output buffer overflow |
|
547 |
+ SZ_ERROR_THREAD - errors in multithreading functions (only for Mt version) |
|
548 |
+ |
|
549 |
+ |
|
550 |
+ |
|
551 |
+LZMA Defines |
|
552 |
+------------ |
|
553 |
+ |
|
554 |
+_LZMA_SIZE_OPT - Enable some optimizations in LZMA Decoder to get smaller executable code. |
|
555 |
+ |
|
556 |
+_LZMA_PROB32 - It can increase the speed on some 32-bit CPUs, but memory usage for |
|
557 |
+ some structures will be doubled in that case. |
|
558 |
+ |
|
559 |
+_LZMA_UINT32_IS_ULONG - Define it if int is 16-bit on your compiler and long is 32-bit. |
|
560 |
+ |
|
561 |
+_LZMA_NO_SYSTEM_SIZE_T - Define it if you don't want to use size_t type. |
|
562 |
+ |
|
563 |
+ |
|
564 |
+C++ LZMA Encoder/Decoder |
|
565 |
+~~~~~~~~~~~~~~~~~~~~~~~~ |
|
566 |
+C++ LZMA code use COM-like interfaces. So if you want to use it, |
|
567 |
+you can study basics of COM/OLE. |
|
568 |
+C++ LZMA code is just wrapper over ANSI-C code. |
|
569 |
+ |
|
570 |
+ |
|
571 |
+C++ Notes |
|
572 |
+~~~~~~~~~~~~~~~~~~~~~~~~ |
|
573 |
+If you use some C++ code folders in 7-Zip (for example, C++ code for .7z handling), |
|
574 |
+you must check that you correctly work with "new" operator. |
|
575 |
+7-Zip can be compiled with MSVC 6.0 that doesn't throw "exception" from "new" operator. |
|
576 |
+So 7-Zip uses "CPP\Common\NewHandler.cpp" that redefines "new" operator: |
|
577 |
+operator new(size_t size) |
|
578 |
+{ |
|
579 |
+ void *p = ::malloc(size); |
|
580 |
+ if (p == 0) |
|
581 |
+ throw CNewException(); |
|
582 |
+ return p; |
|
583 |
+} |
|
584 |
+If you use MSCV that throws exception for "new" operator, you can compile without |
|
585 |
+"NewHandler.cpp". So standard exception will be used. Actually some code of |
|
586 |
+7-Zip catches any exception in internal code and converts it to HRESULT code. |
|
587 |
+So you don't need to catch CNewException, if you call COM interfaces of 7-Zip. |
|
588 |
+ |
|
589 |
+--- |
|
590 |
+ |
|
591 |
+http://www.7-zip.org |
|
592 |
+http://www.7-zip.org/sdk.html |
|
593 |
+http://www.7-zip.org/support.html |