annotate src/bzip2-1.0.6/bzip2.1.preformatted @ 83:ae30d91d2ffe

Replace these with versions built using an older toolset (so as to avoid ABI compatibilities when linking on Ubuntu 14.04 for packaging purposes)
author Chris Cannam
date Fri, 07 Feb 2020 11:51:13 +0000
parents e13257ea84a4
children
rev   line source
Chris@4 1 bzip2(1) bzip2(1)
Chris@4 2
Chris@4 3
Chris@4 4
Chris@4 5 NNAAMMEE
Chris@4 6 bzip2, bunzip2 − a block‐sorting file compressor, v1.0.6
Chris@4 7 bzcat − decompresses files to stdout
Chris@4 8 bzip2recover − recovers data from damaged bzip2 files
Chris@4 9
Chris@4 10
Chris@4 11 SSYYNNOOPPSSIISS
Chris@4 12 bbzziipp22 [ −−ccddffkkqqssttvvzzVVLL112233445566778899 ] [ _f_i_l_e_n_a_m_e_s _._._. ]
Chris@4 13 bbuunnzziipp22 [ −−ffkkvvssVVLL ] [ _f_i_l_e_n_a_m_e_s _._._. ]
Chris@4 14 bbzzccaatt [ −−ss ] [ _f_i_l_e_n_a_m_e_s _._._. ]
Chris@4 15 bbzziipp22rreeccoovveerr _f_i_l_e_n_a_m_e
Chris@4 16
Chris@4 17
Chris@4 18 DDEESSCCRRIIPPTTIIOONN
Chris@4 19 _b_z_i_p_2 compresses files using the Burrows‐Wheeler block
Chris@4 20 sorting text compression algorithm, and Huffman coding.
Chris@4 21 Compression is generally considerably better than that
Chris@4 22 achieved by more conventional LZ77/LZ78‐based compressors,
Chris@4 23 and approaches the performance of the PPM family of sta­
Chris@4 24 tistical compressors.
Chris@4 25
Chris@4 26 The command‐line options are deliberately very similar to
Chris@4 27 those of _G_N_U _g_z_i_p_, but they are not identical.
Chris@4 28
Chris@4 29 _b_z_i_p_2 expects a list of file names to accompany the com­
Chris@4 30 mand‐line flags. Each file is replaced by a compressed
Chris@4 31 version of itself, with the name "original_name.bz2".
Chris@4 32 Each compressed file has the same modification date, per­
Chris@4 33 missions, and, when possible, ownership as the correspond­
Chris@4 34 ing original, so that these properties can be correctly
Chris@4 35 restored at decompression time. File name handling is
Chris@4 36 naive in the sense that there is no mechanism for preserv­
Chris@4 37 ing original file names, permissions, ownerships or dates
Chris@4 38 in filesystems which lack these concepts, or have serious
Chris@4 39 file name length restrictions, such as MS‐DOS.
Chris@4 40
Chris@4 41 _b_z_i_p_2 and _b_u_n_z_i_p_2 will by default not overwrite existing
Chris@4 42 files. If you want this to happen, specify the −f flag.
Chris@4 43
Chris@4 44 If no file names are specified, _b_z_i_p_2 compresses from
Chris@4 45 standard input to standard output. In this case, _b_z_i_p_2
Chris@4 46 will decline to write compressed output to a terminal, as
Chris@4 47 this would be entirely incomprehensible and therefore
Chris@4 48 pointless.
Chris@4 49
Chris@4 50 _b_u_n_z_i_p_2 (or _b_z_i_p_2 _−_d_) decompresses all specified files.
Chris@4 51 Files which were not created by _b_z_i_p_2 will be detected and
Chris@4 52 ignored, and a warning issued. _b_z_i_p_2 attempts to guess
Chris@4 53 the filename for the decompressed file from that of the
Chris@4 54 compressed file as follows:
Chris@4 55
Chris@4 56 filename.bz2 becomes filename
Chris@4 57 filename.bz becomes filename
Chris@4 58 filename.tbz2 becomes filename.tar
Chris@4 59 filename.tbz becomes filename.tar
Chris@4 60 anyothername becomes anyothername.out
Chris@4 61
Chris@4 62 If the file does not end in one of the recognised endings,
Chris@4 63 _._b_z_2_, _._b_z_, _._t_b_z_2 or _._t_b_z_, _b_z_i_p_2 complains that it cannot
Chris@4 64 guess the name of the original file, and uses the original
Chris@4 65 name with _._o_u_t appended.
Chris@4 66
Chris@4 67 As with compression, supplying no filenames causes decom­
Chris@4 68 pression from standard input to standard output.
Chris@4 69
Chris@4 70 _b_u_n_z_i_p_2 will correctly decompress a file which is the con­
Chris@4 71 catenation of two or more compressed files. The result is
Chris@4 72 the concatenation of the corresponding uncompressed files.
Chris@4 73 Integrity testing (−t) of concatenated compressed files is
Chris@4 74 also supported.
Chris@4 75
Chris@4 76 You can also compress or decompress files to the standard
Chris@4 77 output by giving the −c flag. Multiple files may be com­
Chris@4 78 pressed and decompressed like this. The resulting outputs
Chris@4 79 are fed sequentially to stdout. Compression of multiple
Chris@4 80 files in this manner generates a stream containing multi­
Chris@4 81 ple compressed file representations. Such a stream can be
Chris@4 82 decompressed correctly only by _b_z_i_p_2 version 0.9.0 or
Chris@4 83 later. Earlier versions of _b_z_i_p_2 will stop after decom­
Chris@4 84 pressing the first file in the stream.
Chris@4 85
Chris@4 86 _b_z_c_a_t (or _b_z_i_p_2 _‐_d_c_) decompresses all specified files to
Chris@4 87 the standard output.
Chris@4 88
Chris@4 89 _b_z_i_p_2 will read arguments from the environment variables
Chris@4 90 _B_Z_I_P_2 and _B_Z_I_P_, in that order, and will process them
Chris@4 91 before any arguments read from the command line. This
Chris@4 92 gives a convenient way to supply default arguments.
Chris@4 93
Chris@4 94 Compression is always performed, even if the compressed
Chris@4 95 file is slightly larger than the original. Files of less
Chris@4 96 than about one hundred bytes tend to get larger, since the
Chris@4 97 compression mechanism has a constant overhead in the
Chris@4 98 region of 50 bytes. Random data (including the output of
Chris@4 99 most file compressors) is coded at about 8.05 bits per
Chris@4 100 byte, giving an expansion of around 0.5%.
Chris@4 101
Chris@4 102 As a self‐check for your protection, _b_z_i_p_2 uses 32‐bit
Chris@4 103 CRCs to make sure that the decompressed version of a file
Chris@4 104 is identical to the original. This guards against corrup­
Chris@4 105 tion of the compressed data, and against undetected bugs
Chris@4 106 in _b_z_i_p_2 (hopefully very unlikely). The chances of data
Chris@4 107 corruption going undetected is microscopic, about one
Chris@4 108 chance in four billion for each file processed. Be aware,
Chris@4 109 though, that the check occurs upon decompression, so it
Chris@4 110 can only tell you that something is wrong. It can’t help
Chris@4 111 you recover the original uncompressed data. You can use
Chris@4 112 _b_z_i_p_2_r_e_c_o_v_e_r to try to recover data from damaged files.
Chris@4 113
Chris@4 114 Return values: 0 for a normal exit, 1 for environmental
Chris@4 115 problems (file not found, invalid flags, I/O errors, &c),
Chris@4 116 2 to indicate a corrupt compressed file, 3 for an internal
Chris@4 117 consistency error (eg, bug) which caused _b_z_i_p_2 to panic.
Chris@4 118
Chris@4 119
Chris@4 120 OOPPTTIIOONNSS
Chris@4 121 −−cc ‐‐‐‐ssttddoouutt
Chris@4 122 Compress or decompress to standard output.
Chris@4 123
Chris@4 124 −−dd ‐‐‐‐ddeeccoommpprreessss
Chris@4 125 Force decompression. _b_z_i_p_2_, _b_u_n_z_i_p_2 and _b_z_c_a_t are
Chris@4 126 really the same program, and the decision about
Chris@4 127 what actions to take is done on the basis of which
Chris@4 128 name is used. This flag overrides that mechanism,
Chris@4 129 and forces _b_z_i_p_2 to decompress.
Chris@4 130
Chris@4 131 −−zz ‐‐‐‐ccoommpprreessss
Chris@4 132 The complement to −d: forces compression,
Chris@4 133 regardless of the invocation name.
Chris@4 134
Chris@4 135 −−tt ‐‐‐‐tteesstt
Chris@4 136 Check integrity of the specified file(s), but don’t
Chris@4 137 decompress them. This really performs a trial
Chris@4 138 decompression and throws away the result.
Chris@4 139
Chris@4 140 −−ff ‐‐‐‐ffoorrccee
Chris@4 141 Force overwrite of output files. Normally, _b_z_i_p_2
Chris@4 142 will not overwrite existing output files. Also
Chris@4 143 forces _b_z_i_p_2 to break hard links to files, which it
Chris@4 144 otherwise wouldn’t do.
Chris@4 145
Chris@4 146 bzip2 normally declines to decompress files which
Chris@4 147 don’t have the correct magic header bytes. If
Chris@4 148 forced (‐f), however, it will pass such files
Chris@4 149 through unmodified. This is how GNU gzip behaves.
Chris@4 150
Chris@4 151 −−kk ‐‐‐‐kkeeeepp
Chris@4 152 Keep (don’t delete) input files during compression
Chris@4 153 or decompression.
Chris@4 154
Chris@4 155 −−ss ‐‐‐‐ssmmaallll
Chris@4 156 Reduce memory usage, for compression, decompression
Chris@4 157 and testing. Files are decompressed and tested
Chris@4 158 using a modified algorithm which only requires 2.5
Chris@4 159 bytes per block byte. This means any file can be
Chris@4 160 decompressed in 2300k of memory, albeit at about
Chris@4 161 half the normal speed.
Chris@4 162
Chris@4 163 During compression, −s selects a block size of
Chris@4 164 200k, which limits memory use to around the same
Chris@4 165 figure, at the expense of your compression ratio.
Chris@4 166 In short, if your machine is low on memory (8
Chris@4 167 megabytes or less), use −s for everything. See
Chris@4 168 MEMORY MANAGEMENT below.
Chris@4 169
Chris@4 170 −−qq ‐‐‐‐qquuiieett
Chris@4 171 Suppress non‐essential warning messages. Messages
Chris@4 172 pertaining to I/O errors and other critical events
Chris@4 173 will not be suppressed.
Chris@4 174
Chris@4 175 −−vv ‐‐‐‐vveerrbboossee
Chris@4 176 Verbose mode ‐‐ show the compression ratio for each
Chris@4 177 file processed. Further −v’s increase the ver­
Chris@4 178 bosity level, spewing out lots of information which
Chris@4 179 is primarily of interest for diagnostic purposes.
Chris@4 180
Chris@4 181 −−LL ‐‐‐‐lliicceennssee ‐‐VV ‐‐‐‐vveerrssiioonn
Chris@4 182 Display the software version, license terms and
Chris@4 183 conditions.
Chris@4 184
Chris@4 185 −−11 ((oorr −−−−ffaasstt)) ttoo −−99 ((oorr −−−−bbeesstt))
Chris@4 186 Set the block size to 100 k, 200 k .. 900 k when
Chris@4 187 compressing. Has no effect when decompressing.
Chris@4 188 See MEMORY MANAGEMENT below. The −−fast and −−best
Chris@4 189 aliases are primarily for GNU gzip compatibility.
Chris@4 190 In particular, −−fast doesn’t make things signifi­
Chris@4 191 cantly faster. And −−best merely selects the
Chris@4 192 default behaviour.
Chris@4 193
Chris@4 194 −−‐‐ Treats all subsequent arguments as file names, even
Chris@4 195 if they start with a dash. This is so you can han­
Chris@4 196 dle files with names beginning with a dash, for
Chris@4 197 example: bzip2 −‐ −myfilename.
Chris@4 198
Chris@4 199 −−‐‐rreeppeettiittiivvee‐‐ffaasstt ‐‐‐‐rreeppeettiittiivvee‐‐bbeesstt
Chris@4 200 These flags are redundant in versions 0.9.5 and
Chris@4 201 above. They provided some coarse control over the
Chris@4 202 behaviour of the sorting algorithm in earlier ver­
Chris@4 203 sions, which was sometimes useful. 0.9.5 and above
Chris@4 204 have an improved algorithm which renders these
Chris@4 205 flags irrelevant.
Chris@4 206
Chris@4 207
Chris@4 208 MMEEMMOORRYY MMAANNAAGGEEMMEENNTT
Chris@4 209 _b_z_i_p_2 compresses large files in blocks. The block size
Chris@4 210 affects both the compression ratio achieved, and the
Chris@4 211 amount of memory needed for compression and decompression.
Chris@4 212 The flags −1 through −9 specify the block size to be
Chris@4 213 100,000 bytes through 900,000 bytes (the default) respec­
Chris@4 214 tively. At decompression time, the block size used for
Chris@4 215 compression is read from the header of the compressed
Chris@4 216 file, and _b_u_n_z_i_p_2 then allocates itself just enough memory
Chris@4 217 to decompress the file. Since block sizes are stored in
Chris@4 218 compressed files, it follows that the flags −1 to −9 are
Chris@4 219 irrelevant to and so ignored during decompression.
Chris@4 220
Chris@4 221 Compression and decompression requirements, in bytes, can
Chris@4 222 be estimated as:
Chris@4 223
Chris@4 224 Compression: 400k + ( 8 x block size )
Chris@4 225
Chris@4 226 Decompression: 100k + ( 4 x block size ), or
Chris@4 227 100k + ( 2.5 x block size )
Chris@4 228
Chris@4 229 Larger block sizes give rapidly diminishing marginal
Chris@4 230 returns. Most of the compression comes from the first two
Chris@4 231 or three hundred k of block size, a fact worth bearing in
Chris@4 232 mind when using _b_z_i_p_2 on small machines. It is also
Chris@4 233 important to appreciate that the decompression memory
Chris@4 234 requirement is set at compression time by the choice of
Chris@4 235 block size.
Chris@4 236
Chris@4 237 For files compressed with the default 900k block size,
Chris@4 238 _b_u_n_z_i_p_2 will require about 3700 kbytes to decompress. To
Chris@4 239 support decompression of any file on a 4 megabyte machine,
Chris@4 240 _b_u_n_z_i_p_2 has an option to decompress using approximately
Chris@4 241 half this amount of memory, about 2300 kbytes. Decompres­
Chris@4 242 sion speed is also halved, so you should use this option
Chris@4 243 only where necessary. The relevant flag is ‐s.
Chris@4 244
Chris@4 245 In general, try and use the largest block size memory con­
Chris@4 246 straints allow, since that maximises the compression
Chris@4 247 achieved. Compression and decompression speed are virtu­
Chris@4 248 ally unaffected by block size.
Chris@4 249
Chris@4 250 Another significant point applies to files which fit in a
Chris@4 251 single block ‐‐ that means most files you’d encounter
Chris@4 252 using a large block size. The amount of real memory
Chris@4 253 touched is proportional to the size of the file, since the
Chris@4 254 file is smaller than a block. For example, compressing a
Chris@4 255 file 20,000 bytes long with the flag ‐9 will cause the
Chris@4 256 compressor to allocate around 7600k of memory, but only
Chris@4 257 touch 400k + 20000 * 8 = 560 kbytes of it. Similarly, the
Chris@4 258 decompressor will allocate 3700k but only touch 100k +
Chris@4 259 20000 * 4 = 180 kbytes.
Chris@4 260
Chris@4 261 Here is a table which summarises the maximum memory usage
Chris@4 262 for different block sizes. Also recorded is the total
Chris@4 263 compressed size for 14 files of the Calgary Text Compres­
Chris@4 264 sion Corpus totalling 3,141,622 bytes. This column gives
Chris@4 265 some feel for how compression varies with block size.
Chris@4 266 These figures tend to understate the advantage of larger
Chris@4 267 block sizes for larger files, since the Corpus is domi­
Chris@4 268 nated by smaller files.
Chris@4 269
Chris@4 270 Compress Decompress Decompress Corpus
Chris@4 271 Flag usage usage ‐s usage Size
Chris@4 272
Chris@4 273 ‐1 1200k 500k 350k 914704
Chris@4 274 ‐2 2000k 900k 600k 877703
Chris@4 275 ‐3 2800k 1300k 850k 860338
Chris@4 276 ‐4 3600k 1700k 1100k 846899
Chris@4 277 ‐5 4400k 2100k 1350k 845160
Chris@4 278 ‐6 5200k 2500k 1600k 838626
Chris@4 279 ‐7 6100k 2900k 1850k 834096
Chris@4 280 ‐8 6800k 3300k 2100k 828642
Chris@4 281 ‐9 7600k 3700k 2350k 828642
Chris@4 282
Chris@4 283
Chris@4 284 RREECCOOVVEERRIINNGG DDAATTAA FFRROOMM DDAAMMAAGGEEDD FFIILLEESS
Chris@4 285 _b_z_i_p_2 compresses files in blocks, usually 900kbytes long.
Chris@4 286 Each block is handled independently. If a media or trans­
Chris@4 287 mission error causes a multi‐block .bz2 file to become
Chris@4 288 damaged, it may be possible to recover data from the
Chris@4 289 undamaged blocks in the file.
Chris@4 290
Chris@4 291 The compressed representation of each block is delimited
Chris@4 292 by a 48‐bit pattern, which makes it possible to find the
Chris@4 293 block boundaries with reasonable certainty. Each block
Chris@4 294 also carries its own 32‐bit CRC, so damaged blocks can be
Chris@4 295 distinguished from undamaged ones.
Chris@4 296
Chris@4 297 _b_z_i_p_2_r_e_c_o_v_e_r is a simple program whose purpose is to
Chris@4 298 search for blocks in .bz2 files, and write each block out
Chris@4 299 into its own .bz2 file. You can then use _b_z_i_p_2 −t to test
Chris@4 300 the integrity of the resulting files, and decompress those
Chris@4 301 which are undamaged.
Chris@4 302
Chris@4 303 _b_z_i_p_2_r_e_c_o_v_e_r takes a single argument, the name of the dam­
Chris@4 304 aged file, and writes a number of files
Chris@4 305 "rec00001file.bz2", "rec00002file.bz2", etc, containing
Chris@4 306 the extracted blocks. The output filenames are
Chris@4 307 designed so that the use of wildcards in subsequent pro­
Chris@4 308 cessing ‐‐ for example, "bzip2 ‐dc rec*file.bz2 > recov­
Chris@4 309 ered_data" ‐‐ processes the files in the correct order.
Chris@4 310
Chris@4 311 _b_z_i_p_2_r_e_c_o_v_e_r should be of most use dealing with large .bz2
Chris@4 312 files, as these will contain many blocks. It is clearly
Chris@4 313 futile to use it on damaged single‐block files, since a
Chris@4 314 damaged block cannot be recovered. If you wish to min­
Chris@4 315 imise any potential data loss through media or transmis­
Chris@4 316 sion errors, you might consider compressing with a smaller
Chris@4 317 block size.
Chris@4 318
Chris@4 319
Chris@4 320 PPEERRFFOORRMMAANNCCEE NNOOTTEESS
Chris@4 321 The sorting phase of compression gathers together similar
Chris@4 322 strings in the file. Because of this, files containing
Chris@4 323 very long runs of repeated symbols, like "aabaabaabaab
Chris@4 324 ..." (repeated several hundred times) may compress more
Chris@4 325 slowly than normal. Versions 0.9.5 and above fare much
Chris@4 326 better than previous versions in this respect. The ratio
Chris@4 327 between worst‐case and average‐case compression time is in
Chris@4 328 the region of 10:1. For previous versions, this figure
Chris@4 329 was more like 100:1. You can use the −vvvv option to mon­
Chris@4 330 itor progress in great detail, if you want.
Chris@4 331
Chris@4 332 Decompression speed is unaffected by these phenomena.
Chris@4 333
Chris@4 334 _b_z_i_p_2 usually allocates several megabytes of memory to
Chris@4 335 operate in, and then charges all over it in a fairly ran­
Chris@4 336 dom fashion. This means that performance, both for com­
Chris@4 337 pressing and decompressing, is largely determined by the
Chris@4 338 speed at which your machine can service cache misses.
Chris@4 339 Because of this, small changes to the code to reduce the
Chris@4 340 miss rate have been observed to give disproportionately
Chris@4 341 large performance improvements. I imagine _b_z_i_p_2 will per­
Chris@4 342 form best on machines with very large caches.
Chris@4 343
Chris@4 344
Chris@4 345 CCAAVVEEAATTSS
Chris@4 346 I/O error messages are not as helpful as they could be.
Chris@4 347 _b_z_i_p_2 tries hard to detect I/O errors and exit cleanly,
Chris@4 348 but the details of what the problem is sometimes seem
Chris@4 349 rather misleading.
Chris@4 350
Chris@4 351 This manual page pertains to version 1.0.6 of _b_z_i_p_2_. Com­
Chris@4 352 pressed data created by this version is entirely forwards
Chris@4 353 and backwards compatible with the previous public
Chris@4 354 releases, versions 0.1pl2, 0.9.0, 0.9.5, 1.0.0, 1.0.1,
Chris@4 355 1.0.2 and above, but with the following exception: 0.9.0
Chris@4 356 and above can correctly decompress multiple concatenated
Chris@4 357 compressed files. 0.1pl2 cannot do this; it will stop
Chris@4 358 after decompressing just the first file in the stream.
Chris@4 359
Chris@4 360 _b_z_i_p_2_r_e_c_o_v_e_r versions prior to 1.0.2 used 32‐bit integers
Chris@4 361 to represent bit positions in compressed files, so they
Chris@4 362 could not handle compressed files more than 512 megabytes
Chris@4 363 long. Versions 1.0.2 and above use 64‐bit ints on some
Chris@4 364 platforms which support them (GNU supported targets, and
Chris@4 365 Windows). To establish whether or not bzip2recover was
Chris@4 366 built with such a limitation, run it without arguments.
Chris@4 367 In any event you can build yourself an unlimited version
Chris@4 368 if you can recompile it with MaybeUInt64 set to be an
Chris@4 369 unsigned 64‐bit integer.
Chris@4 370
Chris@4 371
Chris@4 372
Chris@4 373
Chris@4 374 AAUUTTHHOORR
Chris@4 375 Julian Seward, jsewardbzip.org.
Chris@4 376
Chris@4 377 http://www.bzip.org
Chris@4 378
Chris@4 379 The ideas embodied in _b_z_i_p_2 are due to (at least) the fol­
Chris@4 380 lowing people: Michael Burrows and David Wheeler (for the
Chris@4 381 block sorting transformation), David Wheeler (again, for
Chris@4 382 the Huffman coder), Peter Fenwick (for the structured cod­
Chris@4 383 ing model in the original _b_z_i_p_, and many refinements), and
Chris@4 384 Alistair Moffat, Radford Neal and Ian Witten (for the
Chris@4 385 arithmetic coder in the original _b_z_i_p_)_. I am much
Chris@4 386 indebted for their help, support and advice. See the man­
Chris@4 387 ual in the source distribution for pointers to sources of
Chris@4 388 documentation. Christian von Roques encouraged me to look
Chris@4 389 for faster sorting algorithms, so as to speed up compres­
Chris@4 390 sion. Bela Lubkin encouraged me to improve the worst‐case
Chris@4 391 compression performance. Donna Robinson XMLised the docu­
Chris@4 392 mentation. The bz* scripts are derived from those of GNU
Chris@4 393 gzip. Many people sent patches, helped with portability
Chris@4 394 problems, lent machines, gave advice and were generally
Chris@4 395 helpful.
Chris@4 396
Chris@4 397
Chris@4 398
Chris@4 399 bzip2(1)