mirrors/bzip2
0
0
Fork 0
mirror of git://sourceware.org/git/bzip2.git synced 2024-11-23 03:33:27 +08:00
Code Issues Packages Projects Releases Wiki Activity

bzip2-0.9.5d

Browse Source
This commit is contained in:
Julian Seward 1999-09-04 22:13:13 +02:00
parent 977101ad5f
commit f93cd82a9a
27 changed files with 32907 additions and 2147 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

55
CHANGES
View File

@ -43,3 +43,58 @@ In compress.c:
do a bit better on small files. This _does_ effect
bzip2.c.
0.9.5a
~~~~~~
Major change: add a fallback sorting algorithm (blocksort.c)
to give reasonable behaviour even for very repetitive inputs.
Nuked --repetitive-best and --repetitive-fast since they are
no longer useful.
Minor changes: mostly a whole bunch of small changes/
bugfixes in the driver (bzip2.c). Changes pertaining to the
user interface are:
allow decompression of symlink'd files to stdout
decompress/test files even without .bz2 extension
give more accurate error messages for I/O errors
when compressing/decompressing to stdout, don't catch control-C
read flags from BZIP2 and BZIP environment variables
decline to break hard links to a file unless forced with -f
allow -c flag even with no filenames
preserve file ownerships as far as possible
make -s -1 give the expected block size (100k)
add a flag -q --quiet to suppress nonessential warnings
stop decoding flags after --, so files beginning in - can be handled
resolved inconsistent naming: bzcat or bz2cat ?
bzip2 --help now returns 0
Programming-level changes are:
fixed syntax error in GET_LL4 for Borland C++ 5.02
let bzBuffToBuffDecompress return BZ_DATA_ERROR{_MAGIC}
fix overshoot of mode-string end in bzopen_or_bzdopen
wrapped bzlib.h in #ifdef __cplusplus ... extern "C" { ... }
close file handles under all error conditions
added minor mods so it compiles with DJGPP out of the box
fixed Makefile so it doesn't give problems with BSD make
fix uninitialised memory reads in dlltest.c
0.9.5b
~~~~~~
Open stdin/stdout in binary mode for DJGPP.
0.9.5c
~~~~~~
Changed BZ_N_OVERSHOOT to be ... + 2 instead of ... + 1. The + 1
version could cause the sorted order to be wrong in some extremely
obscure cases. Also changed setting of quadrant in blocksort.c.
0.9.5d
~~~~~~
The only functional change is to make bzlibVersion() in the library
return the correct string. This has no effect whatsoever on the
functioning of the bzip2 program or library. Added a couple of casts
so the library compiles without warnings at level 3 in MS Visual
Studio 6.0. Included a Y2K statement in the file Y2K_INFO. All other
changes are minor documentation changes.

6
LICENSE
View File

@ -1,6 +1,6 @@
This program, "bzip2" and associated library "libbzip2", are
copyright (C) 1996-1998 Julian R Seward. All rights reserved.
copyright (C) 1996-1999 Julian R Seward. All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
@ -33,7 +33,7 @@ WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
Julian Seward, Guildford, Surrey, UK.
Julian Seward, Cambridge, UK.
jseward@acm.org
bzip2/libbzip2 version 0.9.0 of 28 June 1998
bzip2/libbzip2 version 0.9.5 of 24 May 1999

89
Makefile
View File

@ -1,6 +1,7 @@
SHELL=/bin/sh
CC=gcc
CFLAGS=-Wall -O2 -fomit-frame-pointer -fno-strength-reduce
CFLAGS=-Wall -Winline -O2 -fomit-frame-pointer -fno-strength-reduce
OBJS= blocksort.o \
huffman.o \
@ -10,37 +11,93 @@ OBJS= blocksort.o \
decompress.o \
bzlib.o
all: lib bzip2 test
all: libbz2.a bzip2 bzip2recover test
bzip2: lib
$(CC) $(CFLAGS) -c bzip2.c
bzip2: libbz2.a bzip2.o
$(CC) $(CFLAGS) -o bzip2 bzip2.o -L. -lbz2
$(CC) $(CFLAGS) -o bzip2recover bzip2recover.c
lib: $(OBJS)
bzip2recover: bzip2recover.o
$(CC) $(CFLAGS) -o bzip2recover bzip2recover.o
libbz2.a: $(OBJS)
rm -f libbz2.a
ar clq libbz2.a $(OBJS)
ar cq libbz2.a $(OBJS)
@if ( test -f /usr/bin/ranlib -o -f /bin/ranlib -o \
-f /usr/ccs/bin/ranlib ) ; then \
echo ranlib libbz2.a ; \
ranlib libbz2.a ; \
fi
test: bzip2
@cat words1
./bzip2 -1 < sample1.ref > sample1.rb2
./bzip2 -2 < sample2.ref > sample2.rb2
./bzip2 -d < sample1.bz2 > sample1.tst
./bzip2 -d < sample2.bz2 > sample2.tst
@cat words2
./bzip2 -1 < sample1.ref > sample1.rb2
./bzip2 -2 < sample2.ref > sample2.rb2
./bzip2 -3 < sample3.ref > sample3.rb2
./bzip2 -d < sample1.bz2 > sample1.tst
./bzip2 -d < sample2.bz2 > sample2.tst
./bzip2 -ds < sample3.bz2 > sample3.tst
cmp sample1.bz2 sample1.rb2
cmp sample2.bz2 sample2.rb2
cmp sample3.bz2 sample3.rb2
cmp sample1.tst sample1.ref
cmp sample2.tst sample2.ref
cmp sample3.tst sample3.ref
@cat words3
PREFIX=/usr
install: bzip2 bzip2recover
if ( test ! -d $(PREFIX)/bin ) ; then mkdir $(PREFIX)/bin ; fi
if ( test ! -d $(PREFIX)/lib ) ; then mkdir $(PREFIX)/lib ; fi
if ( test ! -d $(PREFIX)/man ) ; then mkdir $(PREFIX)/man ; fi
if ( test ! -d $(PREFIX)/man/man1 ) ; then mkdir $(PREFIX)/man/man1 ; fi
if ( test ! -d $(PREFIX)/include ) ; then mkdir $(PREFIX)/include ; fi
cp -f bzip2 $(PREFIX)/bin/bzip2
cp -f bzip2 $(PREFIX)/bin/bunzip2
cp -f bzip2 $(PREFIX)/bin/bzcat
cp -f bzip2recover $(PREFIX)/bin/bzip2recover
chmod a+x $(PREFIX)/bin/bzip2
chmod a+x $(PREFIX)/bin/bunzip2
chmod a+x $(PREFIX)/bin/bzcat
chmod a+x $(PREFIX)/bin/bzip2recover
cp -f bzip2.1 $(PREFIX)/man/man1
chmod a+r $(PREFIX)/man/man1/bzip2.1
cp -f bzlib.h $(PREFIX)/include
chmod a+r $(PREFIX)/include/bzlib.h
cp -f libbz2.a $(PREFIX)/lib
chmod a+r $(PREFIX)/lib/libbz2.a
clean:
rm -f *.o libbz2.a bzip2 bzip2recover sample1.rb2 sample2.rb2 sample1.tst sample2.tst
rm -f *.o libbz2.a bzip2 bzip2recover \
sample1.rb2 sample2.rb2 sample3.rb2 \
sample1.tst sample2.tst sample3.tst
.c.o: $*.o bzlib.h bzlib_private.h
$(CC) $(CFLAGS) -c $*.c -o $*.o
blocksort.o: blocksort.c
$(CC) $(CFLAGS) -c blocksort.c
huffman.o: huffman.c
$(CC) $(CFLAGS) -c huffman.c
crctable.o: crctable.c
$(CC) $(CFLAGS) -c crctable.c
randtable.o: randtable.c
$(CC) $(CFLAGS) -c randtable.c
compress.o: compress.c
$(CC) $(CFLAGS) -c compress.c
decompress.o: decompress.c
$(CC) $(CFLAGS) -c decompress.c
bzlib.o: bzlib.c
$(CC) $(CFLAGS) -c bzlib.c
bzip2.o: bzip2.c
$(CC) $(CFLAGS) -c bzip2.c
bzip2recover.o: bzip2recover.c
$(CC) $(CFLAGS) -c bzip2recover.c
tarfile:
tar cvf interim.tar *.c *.h Makefile manual.texi manual.ps LICENSE bzip2.1 bzip2.1.preformatted bzip2.txt words1 words2 words3 sample1.ref sample2.ref sample1.bz2 sample2.bz2 *.html README CHANGES libbz2.def libbz2.dsp dlltest.dsp
tar cvf interim.tar blocksort.c huffman.c crctable.c \
randtable.c compress.c decompress.c bzlib.c bzip2.c \
bzip2recover.c bzlib.h bzlib_private.h Makefile manual.texi \
manual.ps LICENSE bzip2.1 bzip2.1.preformatted bzip2.txt \
words1 words2 words3 sample1.ref sample2.ref sample3.ref \
sample1.bz2 sample2.bz2 sample3.bz2 dlltest.c \
*.html README CHANGES libbz2.def libbz2.dsp \
dlltest.dsp makefile.msc Y2K_INFO

78
README
View File

@ -1,48 +1,44 @@
This is the README for bzip2, a block-sorting file compressor, version
0.9.0. This version is fully compatible with the previous public
release, bzip2-0.1pl2.
0.9.5d. This version is fully compatible with the previous public
releases, bzip2-0.1pl2 and bzip2-0.9.0.
bzip2-0.9.0 is distributed under a BSD-style license. For details,
bzip2-0.9.5 is distributed under a BSD-style license. For details,
see the file LICENSE.
Complete documentation is available in Postscript form (manual.ps)
or html (manual_toc.html). A plain-text version of the manual page is
available as bzip2.txt.
Complete documentation is available in Postscript form (manual.ps) or
html (manual_toc.html). A plain-text version of the manual page is
available as bzip2.txt. A statement about Y2K issues is now included
in the file Y2K_INFO.
HOW TO BUILD -- UNIX
Type `make'.
Type `make'. This builds the library libbz2.a and then the
programs bzip2 and bzip2recover. Six self-tests are run.
If the self-tests complete ok, carry on to installation:
This creates binaries "bzip2" and "bzip2recover".
To install in /usr/bin, /usr/lib, /usr/man and /usr/include, type
make install
To install somewhere else, eg, /xxx/yyy/{bin,lib,man,include}, type
make install PREFIX=/xxx/yyy
If you are (justifiably) paranoid and want to see what 'make install'
is going to do, you can first do
make -n install or
make -n install PREFIX=/xxx/yyy respectively.
The -n instructs make to show the commands it would execute, but
not actually execute them.
It also runs four compress-decompress tests to make sure things are
working properly. If all goes well, you should be up & running.
Please be sure to read the output from `make' just to be sure that the
tests went ok.
To install bzip2 properly:
* Copy the binaries "bzip2" and "bzip2recover" to a publically visible
place, possibly /usr/bin or /usr/local/bin.
* In that directory, make "bunzip2" and "bzcat" be symbolic links
to "bzip2".
* Copy the manual page, bzip2.1, to the relevant place.
Probably the right place is /usr/man/man1/.
If you want to program with the library, you'll need to copy libbz2.a
and bzlib.h to /usr/lib and /usr/include respectively.
HOW TO BUILD -- Windows 95, NT, DOS, Mac, etc.
It's difficult for me to support compilation on all these platforms.
My approach is to collect binaries for these platforms, and put them
on my web page (http://www.muraroa.demon.co.uk). Look there.
on my web page (http://www.muraroa.demon.co.uk). Look there. However
(FWIW), bzip2-0.9.5 is very standard ANSI C and should compile
unmodified with MS Visual C. For Win32, there is one important
caveat: in bzip2.c, you must set BZ_UNIX to 0 and BZ_LCCWIN32 to 1
before building.
VALIDATION
@ -112,26 +108,32 @@ WHAT'S NEW IN 0.9.0 (as compared to 0.1pl2) ?
* Much more documentation, i.e., a proper user manual
* Hopefully, improved portability (at least of the library)
WHAT'S NEW IN 0.9.5 ?
* Compression speed is much less sensitive to the input
data than in previous versions. Specifically, the very
slow performance caused by repetitive data is fixed.
* Many small improvements in file and flag handling.
* A Y2K statement.
I hope you find bzip2 useful. Feel free to contact me at
jseward@acm.org
if you have any suggestions or queries. Many people mailed me with
comments, suggestions and patches after the releases of bzip-0.15,
bzip-0.21 and bzip2-0.1pl2, and the changes in bzip2 are largely a
result of this feedback. I thank you for your comments.
bzip-0.21, bzip2-0.1pl2 and bzip2-0.9.0, and the changes in bzip2 are
largely a result of this feedback. I thank you for your comments.
At least for the time being, bzip2's "home" is
At least for the time being, bzip2's "home" is (or can be reached via)
http://www.muraroa.demon.co.uk.
Julian Seward
jseward@acm.org
Manchester, UK
18 July 1996 (version 0.15)
Cambridge, UK
18 July 1996 (version 0.15)
25 August 1996 (version 0.21)
Guildford, Surrey, UK
7 August 1997 (bzip2, version 0.1)
7 August 1997 (bzip2, version 0.1)
29 August 1997 (bzip2, version 0.1pl2)
23 August 1998 (bzip2, version 0.9.0)
8 June 1999 (bzip2, version 0.9.5)
4 Sept 1999 (bzip2, version 0.9.5d)

34
Y2K_INFO Normal file
View File

@ -0,0 +1,34 @@
Y2K status of bzip2 and libbzip2, versions 0.1, 0.9.0 and 0.9.5
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Informally speaking:
bzip2 is a compression program built on top of libbzip2,
a library which does the real work of compression and
decompression. As far as I am aware, libbzip2 does not have
any date-related code at all.
bzip2 itself copies dates from source to destination files
when compressing or decompressing, using the 'stat' and 'utime'
UNIX system calls. It doesn't examine, manipulate or store the
dates in any way. So as far as I can see, there shouldn't be any
problem with bzip2 providing 'stat' and 'utime' work correctly
on your system.
On non-unix platforms (those for which BZ_UNIX in bzip2.c is
not set to 1), bzip2 doesn't even do the date copying.
Overall, informally speaking, I don't think bzip2 or libbzip2
have a Y2K problem.
Formally speaking:
I am not prepared to offer you any assurance whatsoever
regarding Y2K issues in my software. You alone assume the
entire risk of using the software. The disclaimer of liability
in the LICENSE file in the bzip2 source distribution continues
to apply on this issue as with every other issue pertaining
to the software.
Julian Seward
Cambridge, UK
25 August 1999

1248
blocksort.c
View File

File diff suppressed because it is too large Load Diff

604
bzip2.1
View File

@ -1,7 +1,7 @@
.PU
.TH bzip2 1
.SH NAME
bzip2, bunzip2 \- a block-sorting file compressor, v0.9.0
bzip2, bunzip2 \- a block-sorting file compressor, v0.9.5
.br
bzcat \- decompresses files to stdout
.br
@ -10,7 +10,7 @@ bzip2recover \- recovers data from damaged bzip2 files
.SH SYNOPSIS
.ll +8
.B bzip2
.RB [ " \-cdfkstvzVL123456789 " ]
.RB [ " \-cdfkqstvzVL123456789 " ]
[
.I "filenames \&..."
]
@ -18,13 +18,13 @@ bzip2recover \- recovers data from damaged bzip2 files
.br
.B bunzip2
.RB [ " \-fkvsVL " ]
[
[
.I "filenames \&..."
]
.br
.br
.B bzcat
.RB [ " \-s " ]
[
[
.I "filenames \&..."
]
.br
@ -33,211 +33,171 @@ bzip2recover \- recovers data from damaged bzip2 files
.SH DESCRIPTION
.I bzip2
compresses files using the Burrows-Wheeler block-sorting
text compression algorithm, and Huffman coding.
Compression is generally considerably
better than that
achieved by more conventional LZ77/LZ78-based compressors,
and approaches the performance of the PPM family of statistical
compressors.
compresses files using the Burrows-Wheeler block sorting
text compression algorithm, and Huffman coding. Compression is
generally considerably better than that achieved by more conventional
LZ77/LZ78-based compressors, and approaches the performance of the PPM
family of statistical compressors.
The command-line options are deliberately very similar to
those of
.I GNU Gzip,
.I GNU gzip,
but they are not identical.
.I bzip2
expects a list of file names to accompany the command-line flags.
Each file is replaced by a compressed version of itself,
with the name "original_name.bz2".
Each compressed file has the same modification date and permissions
as the corresponding original, so that these properties can be
correctly restored at decompression time. File name handling is
naive in the sense that there is no mechanism for preserving
original file names, permissions and dates in filesystems
which lack these concepts, or have serious file name length
restrictions, such as MS-DOS.
.I bzip2
expects a list of file names to accompany the
command-line flags. Each file is replaced by a compressed version of
itself, with the name "original_name.bz2".
Each compressed file
has the same modification date, permissions, and, when possible,
ownership as the corresponding original, so that these properties can
be correctly restored at decompression time. File name handling is
naive in the sense that there is no mechanism for preserving original
file names, permissions, ownerships or dates in filesystems which lack
these concepts, or have serious file name length restrictions, such as
MS-DOS.
.I bzip2
and
.I bunzip2
will by default not overwrite existing files;
if you want this to happen, specify the \-f flag.
will by default not overwrite existing
files. If you want this to happen, specify the \-f flag.
If no file names are specified,
.I bzip2
compresses from standard input to standard output.
In this case,
compresses from standard
input to standard output. In this case,
.I bzip2
will decline to write compressed output to a terminal, as
this would be entirely incomprehensible and therefore pointless.
will decline to
write compressed output to a terminal, as this would be entirely
incomprehensible and therefore pointless.
.I bunzip2
(or
.I bzip2 \-d
) decompresses and restores all specified files whose names
end in ".bz2".
Files without this suffix are ignored.
Again, supplying no filenames
causes decompression from standard input to standard output.
.I bunzip2
will correctly decompress a file which is the concatenation
of two or more compressed files. The result is the concatenation
of the corresponding uncompressed files. Integrity testing
(\-t) of concatenated compressed files is also supported.
You can also compress or decompress files to
the standard output by giving the \-c flag.
Multiple files may be compressed and decompressed like this.
The resulting outputs are fed sequentially to stdout.
Compression of multiple files in this manner generates
a stream containing multiple compressed file representations.
Such a stream can be decompressed correctly only by
.I bzip2 \-d)
decompresses all
specified files. Files which were not created by
.I bzip2
version 0.9.0 or later. Earlier versions of
will be detected and ignored, and a warning issued.
.I bzip2
will stop after decompressing the first file in the stream.
attempts to guess the filename for the decompressed file
from that of the compressed file as follows:
filename.bz2 becomes filename
filename.bz becomes filename
filename.tbz2 becomes filename.tar
filename.tbz becomes filename.tar
anyothername becomes anyothername.out
If the file does not end in one of the recognised endings,
.I .bz2,
.I .bz,
.I .tbz2
or
.I .tbz,
.I bzip2
complains that it cannot
guess the name of the original file, and uses the original name
with
.I .out
appended.
As with compression, supplying no
filenames causes decompression from
standard input to standard output.
.I bunzip2
will correctly decompress a file which is the
concatenation of two or more compressed files. The result is the
concatenation of the corresponding uncompressed files. Integrity
testing (\-t)
of concatenated
compressed files is also supported.
You can also compress or decompress files to the standard output by
giving the \-c flag. Multiple files may be compressed and
decompressed like this. The resulting outputs are fed sequentially to
stdout. Compression of multiple files
in this manner generates a stream
containing multiple compressed file representations. Such a stream
can be decompressed correctly only by
.I bzip2
version 0.9.0 or
later. Earlier versions of
.I bzip2
will stop after decompressing
the first file in the stream.
.I bzcat
(or
.I bzip2 \-dc
) decompresses all specified files to the standard output.
.I bzip2 -dc)
decompresses all specified files to
the standard output.
Compression is always performed, even if the compressed file is
slightly larger than the original. Files of less than about
one hundred bytes tend to get larger, since the compression
mechanism has a constant overhead in the region of 50 bytes.
Random data (including the output of most file compressors)
is coded at about 8.05 bits per byte, giving an expansion of
around 0.5%.
.I bzip2
will read arguments from the environment variables
.I BZIP2
and
.I BZIP,
in that order, and will process them
before any arguments read from the command line. This gives a
convenient way to supply default arguments.
As a self-check for your protection,
Compression is always performed, even if the compressed
file is slightly
larger than the original. Files of less than about one hundred bytes
tend to get larger, since the compression mechanism has a constant
overhead in the region of 50 bytes. Random data (including the output
of most file compressors) is coded at about 8.05 bits per byte, giving
an expansion of around 0.5%.
As a self-check for your protection,
.I
bzip2
uses 32-bit CRCs to
make sure that the decompressed version of a file is identical to the
original. This guards against corruption of the compressed data, and
against undetected bugs in
.I bzip2
uses 32-bit CRCs to make sure that the decompressed
version of a file is identical to the original.
This guards against corruption of the compressed data,
and against undetected bugs in
.I bzip2
(hopefully very unlikely).
The chances of data corruption going undetected is
microscopic, about one chance in four billion
for each file processed. Be aware, though, that the check
occurs upon decompression, so it can only tell you that
that something is wrong. It can't help you recover the
original uncompressed data.
You can use
(hopefully very unlikely). The
chances of data corruption going undetected is microscopic, about one
chance in four billion for each file processed. Be aware, though, that
the check occurs upon decompression, so it can only tell you that
something is wrong. It can't help you
recover the original uncompressed
data. You can use
.I bzip2recover
to try to recover data from damaged files.
to try to recover data from
damaged files.
Return values:
0 for a normal exit,
1 for environmental
problems (file not found, invalid flags, I/O errors, &c),
2 to indicate a corrupt compressed file,
3 for an internal consistency error (eg, bug) which caused
.I bzip2
to panic.
.SH MEMORY MANAGEMENT
.I Bzip2
compresses large files in blocks. The block size affects both the
compression ratio achieved, and the amount of memory needed both for
compression and decompression. The flags \-1 through \-9
specify the block size to be 100,000 bytes through 900,000 bytes
(the default) respectively. At decompression-time, the block size used for
compression is read from the header of the compressed file, and
.I bunzip2
then allocates itself just enough memory to decompress the file.
Since block sizes are stored in compressed files, it follows that the flags
\-1 to \-9
are irrelevant to and so ignored during decompression.
Compression and decompression requirements, in bytes, can be estimated as:
Compression: 400k + ( 7 x block size )
Decompression: 100k + ( 4 x block size ), or
.br
100k + ( 2.5 x block size )
Larger block sizes give rapidly diminishing marginal returns; most
of the
compression comes from the first two or three hundred k of block size,
a fact worth bearing in mind when using
Return values: 0 for a normal exit, 1 for environmental problems (file
not found, invalid flags, I/O errors, &c), 2 to indicate a corrupt
compressed file, 3 for an internal consistency error (eg, bug) which
caused
.I bzip2
on small machines. It is also important to appreciate that the
decompression memory requirement is set at compression-time by the
choice of block size.
For files compressed with the default 900k block size,
.I bunzip2
will require about 3700 kbytes to decompress.
To support decompression of any file on a 4 megabyte machine,
.I bunzip2
has an option to decompress using approximately half this
amount of memory, about 2300 kbytes. Decompression speed is
also halved, so you should use this option only where necessary.
The relevant flag is \-s.
In general, try and use the largest block size
memory constraints allow, since that maximises the compression
achieved. Compression and decompression
speed are virtually unaffected by block size.
Another significant point applies to files which fit in a single
block -- that means most files you'd encounter using a large
block size. The amount of real memory touched is proportional
to the size of the file, since the file is smaller than a block.
For example, compressing a file 20,000 bytes long with the flag
\-9
will cause the compressor to allocate around
6700k of memory, but only touch 400k + 20000 * 7 = 540
kbytes of it. Similarly, the decompressor will allocate 3700k but
only touch 100k + 20000 * 4 = 180 kbytes.
Here is a table which summarises the maximum memory usage for
different block sizes. Also recorded is the total compressed
size for 14 files of the Calgary Text Compression Corpus
totalling 3,141,622 bytes. This column gives some feel for how
compression varies with block size. These figures tend to understate
the advantage of larger block sizes for larger files, since the
Corpus is dominated by smaller files.
Compress Decompress Decompress Corpus
Flag usage usage -s usage Size
-1 1100k 500k 350k 914704
-2 1800k 900k 600k 877703
-3 2500k 1300k 850k 860338
-4 3200k 1700k 1100k 846899
-5 3900k 2100k 1350k 845160
-6 4600k 2500k 1600k 838626
-7 5400k 2900k 1850k 834096
-8 6000k 3300k 2100k 828642
-9 6700k 3700k 2350k 828642
to panic.
.SH OPTIONS
.TP
.B \-c --stdout
Compress or decompress to standard output. \-c will decompress
multiple files to stdout, but will only compress a single file to
stdout.
Compress or decompress to standard output.
.TP
.B \-d --decompress
Force decompression.
.I bzip2,
.I bunzip2
Force decompression.
.I bzip2,
.I bunzip2
and
.I bzcat
are really the same program, and the decision about what actions
to take is done on the basis of which name is
used. This flag overrides that mechanism, and forces
.I bzcat
are
really the same program, and the decision about what actions to take is
done on the basis of which name is used. This flag overrides that
mechanism, and forces
.I bzip2
to decompress.
.TP
.TP
.B \-z --compress
The complement to \-d: forces compression, regardless of the invokation
name.
The complement to \-d: forces compression, regardless of the
invokation name.
.TP
.B \-t --test
Check integrity of the specified file(s), but don't decompress them.
@ -245,25 +205,31 @@ This really performs a trial decompression and throws away the result.
.TP
.B \-f --force
Force overwrite of output files. Normally,
.I bzip2
will not overwrite existing output files.
.I bzip2
will not overwrite
existing output files. Also forces
.I bzip2
to break hard links
to files, which it otherwise wouldn't do.
.TP
.B \-k --keep
Keep (don't delete) input files during compression or decompression.
Keep (don't delete) input files during compression
or decompression.
.TP
.B \-s --small
Reduce memory usage, for compression, decompression and
testing.
Files are decompressed and tested using a modified algorithm which only
Reduce memory usage, for compression, decompression and testing. Files
are decompressed and tested using a modified algorithm which only
requires 2.5 bytes per block byte. This means any file can be
decompressed in 2300k of memory, albeit at about half the normal
speed.
decompressed in 2300k of memory, albeit at about half the normal speed.
During compression, -s selects a block size of 200k, which limits
memory use to around the same figure, at the expense of your
compression ratio. In short, if your machine is low on memory
(8 megabytes or less), use -s for everything. See
MEMORY MANAGEMENT above.
During compression, \-s selects a block size of 200k, which limits
memory use to around the same figure, at the expense of your compression
ratio. In short, if your machine is low on memory (8 megabytes or
less), use \-s for everything. See MEMORY MANAGEMENT below.
.TP
.B \-q --quiet
Suppress non-essential warning messages. Messages pertaining to
I/O errors and other critical events will not be suppressed.
.TP
.B \-v --verbose
Verbose mode -- show the compression ratio for each file processed.
@ -273,147 +239,199 @@ information which is primarily of interest for diagnostic purposes.
.B \-L --license -V --version
Display the software version, license terms and conditions.
.TP
.B \-1 to \-9
Set the block size to 100 k, 200 k .. 900 k when
compressing. Has no effect when decompressing.
See MEMORY MANAGEMENT above.
.B \-1 to \-9
Set the block size to 100 k, 200 k .. 900 k when compressing. Has no
effect when decompressing. See MEMORY MANAGEMENT below.
.TP
.B \--repetitive-fast
.I bzip2
injects some small pseudo-random variations
into very repetitive blocks to limit
worst-case performance during compression.
If sorting runs into difficulties, the block
is randomised, and sorting is restarted.
Very roughly,
.I bzip2
persists for three times as long as a well-behaved input
would take before resorting to randomisation.
This flag makes it give up much sooner.
.B \--
Treats all subsequent arguments as file names, even if they start
with a dash. This is so you can handle files with names beginning
with a dash, for example: bzip2 \-- \-myfilename.
.TP
.B \--repetitive-fast --repetitive-best
These flags are redundant in versions 0.9.5 and above. They provided
some coarse control over the behaviour of the sorting algorithm in
earlier versions, which was sometimes useful. 0.9.5 and above have an
improved algorithm which renders these flags irrelevant.
.TP
.B \--repetitive-best
Opposite of \--repetitive-fast; try a lot harder before
resorting to randomisation.
.SH MEMORY MANAGEMENT
.I bzip2
compresses large files in blocks. The block size affects
both the compression ratio achieved, and the amount of memory needed for
compression and decompression. The flags \-1 through \-9
specify the block size to be 100,000 bytes through 900,000 bytes (the
default) respectively. At decompression time, the block size used for
compression is read from the header of the compressed file, and
.I bunzip2
then allocates itself just enough memory to decompress
the file. Since block sizes are stored in compressed files, it follows
that the flags \-1 to \-9 are irrelevant to and so ignored
during decompression.
Compression and decompression requirements,
in bytes, can be estimated as:
Compression: 400k + ( 8 x block size )
Decompression: 100k + ( 4 x block size ), or
100k + ( 2.5 x block size )
Larger block sizes give rapidly diminishing marginal returns. Most of
the compression comes from the first two or three hundred k of block
size, a fact worth bearing in mind when using
.I bzip2
on small machines.
It is also important to appreciate that the decompression memory
requirement is set at compression time by the choice of block size.
For files compressed with the default 900k block size,
.I bunzip2
will require about 3700 kbytes to decompress. To support decompression
of any file on a 4 megabyte machine,
.I bunzip2
has an option to
decompress using approximately half this amount of memory, about 2300
kbytes. Decompression speed is also halved, so you should use this
option only where necessary. The relevant flag is -s.
In general, try and use the largest block size memory constraints allow,
since that maximises the compression achieved. Compression and
decompression speed are virtually unaffected by block size.
Another significant point applies to files which fit in a single block
-- that means most files you'd encounter using a large block size. The
amount of real memory touched is proportional to the size of the file,
since the file is smaller than a block. For example, compressing a file
20,000 bytes long with the flag -9 will cause the compressor to
allocate around 7600k of memory, but only touch 400k + 20000 * 8 = 560
kbytes of it. Similarly, the decompressor will allocate 3700k but only
touch 100k + 20000 * 4 = 180 kbytes.
Here is a table which summarises the maximum memory usage for different
block sizes. Also recorded is the total compressed size for 14 files of
the Calgary Text Compression Corpus totalling 3,141,622 bytes. This
column gives some feel for how compression varies with block size.
These figures tend to understate the advantage of larger block sizes for
larger files, since the Corpus is dominated by smaller files.
Compress Decompress Decompress Corpus
Flag usage usage -s usage Size
-1 1200k 500k 350k 914704
-2 2000k 900k 600k 877703
-3 2800k 1300k 850k 860338
-4 3600k 1700k 1100k 846899
-5 4400k 2100k 1350k 845160
-6 5200k 2500k 1600k 838626
-7 6100k 2900k 1850k 834096
-8 6800k 3300k 2100k 828642
-9 7600k 3700k 2350k 828642
.SH RECOVERING DATA FROM DAMAGED FILES
.I bzip2
compresses files in blocks, usually 900kbytes long.
Each block is handled independently. If a media or
transmission error causes a multi-block .bz2
file to become damaged,
it may be possible to recover data from the undamaged blocks
in the file.
compresses files in blocks, usually 900kbytes long. Each
block is handled independently. If a media or transmission error causes
a multi-block .bz2
file to become damaged, it may be possible to
recover data from the undamaged blocks in the file.
The compressed representation of each block is delimited by
a 48-bit pattern, which makes it possible to find the block
boundaries with reasonable certainty. Each block also carries
its own 32-bit CRC, so damaged blocks can be
distinguished from undamaged ones.
The compressed representation of each block is delimited by a 48-bit
pattern, which makes it possible to find the block boundaries with
reasonable certainty. Each block also carries its own 32-bit CRC, so
damaged blocks can be distinguished from undamaged ones.
.I bzip2recover
is a simple program whose purpose is to search for
blocks in .bz2 files, and write each block out into
its own .bz2 file. You can then use
.I bzip2 -t
to test the integrity of the resulting files,
and decompress those which are undamaged.
is a simple program whose purpose is to search for
blocks in .bz2 files, and write each block out into its own .bz2
file. You can then use
.I bzip2
\-t
to test the
integrity of the resulting files, and decompress those which are
undamaged.
.I bzip2recover
takes a single argument, the name of the damaged file,
and writes a number of files "rec0001file.bz2", "rec0002file.bz2",
etc, containing the extracted blocks. The output filenames
are designed so that the use of wildcards in subsequent processing
-- for example, "bzip2 -dc rec*file.bz2 > recovered_data" --
lists the files in the "right" order.
takes a single argument, the name of the damaged file,
and writes a number of files "rec0001file.bz2",
"rec0002file.bz2", etc, containing the extracted blocks.
The output filenames are designed so that the use of
wildcards in subsequent processing -- for example,
"bzip2 -dc rec*file.bz2 > recovered_data" -- lists the files in
the correct order.
.I bzip2recover
should be of most use dealing with large .bz2 files, as
these will contain many blocks. It is clearly futile to
use it on damaged single-block files, since a damaged
block cannot be recovered. If you wish to minimise
any potential data loss through media or transmission
errors, you might consider compressing with a smaller
should be of most use dealing with large .bz2
files, as these will contain many blocks. It is clearly
futile to use it on damaged single-block files, since a
damaged block cannot be recovered. If you wish to minimise
any potential data loss through media or transmission errors,
you might consider compressing with a smaller
block size.
.SH PERFORMANCE NOTES
The sorting phase of compression gathers together similar strings
in the file. Because of this, files containing very long
runs of repeated symbols, like "aabaabaabaab ..." (repeated
several hundred times) may compress extraordinarily slowly.
You can use the
\-vvvvv
option to monitor progress in great detail, if you want.
Decompression speed is unaffected.
The sorting phase of compression gathers together similar strings in the
file. Because of this, files containing very long runs of repeated
symbols, like "aabaabaabaab ..." (repeated several hundred times) may
compress more slowly than normal. Versions 0.9.5 and above fare much
better than previous versions in this respect. The ratio between
worst-case and average-case compression time is in the region of 10:1.
For previous versions, this figure was more like 100:1. You can use the
\-vvvv option to monitor progress in great detail, if you want.
Such pathological cases
seem rare in practice, appearing mostly in artificially-constructed
test files, and in low-level disk images. It may be inadvisable to
use
.I bzip2
to compress the latter.
If you do get a file which causes severe slowness in compression,
try making the block size as small as possible, with flag \-1.
Decompression speed is unaffected by these phenomena.
.I bzip2
usually allocates several megabytes of memory to operate in,
and then charges all over it in a fairly random fashion. This
means that performance, both for compressing and decompressing,
is largely determined by the speed
at which your machine can service cache misses.
Because of this, small changes
to the code to reduce the miss rate have been observed to give
disproportionately large performance improvements.
usually allocates several megabytes of memory to operate
in, and then charges all over it in a fairly random fashion. This means
that performance, both for compressing and decompressing, is largely
determined by the speed at which your machine can service cache misses.
Because of this, small changes to the code to reduce the miss rate have
been observed to give disproportionately large performance improvements.
I imagine
.I bzip2
will perform best on machines with very large caches.
.SH CAVEATS
I/O error messages are not as helpful as they could be.
.I Bzip2
tries hard to detect I/O errors and exit cleanly, but the
details of what the problem is sometimes seem rather misleading.
.I bzip2
tries hard to detect I/O errors and exit cleanly, but the details of
what the problem is sometimes seem rather misleading.
This manual page pertains to version 0.9.0 of
This manual page pertains to version 0.9.5 of
.I bzip2.
Compressed data created by this version is entirely forwards and
backwards compatible with the previous public release, version 0.1pl2,
but with the following exception: 0.9.0 can correctly decompress
multiple concatenated compressed files. 0.1pl2 cannot do this; it
will stop after decompressing just the first file in the stream.
Wildcard expansion for Windows 95 and NT
is flaky.
Compressed
data created by this version is entirely forwards and backwards
compatible with the previous public releases, versions 0.1pl2 and 0.9.0,
but with the following exception: 0.9.0 and above can correctly
decompress multiple concatenated compressed files. 0.1pl2 cannot do
this; it will stop after decompressing just the first file in the
stream.
.I bzip2recover
uses 32-bit integers to represent bit positions in
compressed files, so it cannot handle compressed files
more than 512 megabytes long. This could easily be fixed.
compressed files, so it cannot handle compressed files more than 512
megabytes long. This could easily be fixed.
.SH AUTHOR
Julian Seward, jseward@acm.org.
http://www.muraroa.demon.co.uk
The ideas embodied in
The ideas embodied in
.I bzip2
are due to (at least) the following people:
Michael Burrows and David Wheeler (for the block sorting
transformation), David Wheeler (again, for the Huffman coder),
Peter Fenwick (for the structured coding model in the original
.I bzip,
and many refinements),
and
Alistair Moffat, Radford Neal and Ian Witten (for the arithmetic
coder in the original
are due to (at least) the following
people: Michael Burrows and David Wheeler (for the block sorting
transformation), David Wheeler (again, for the Huffman coder), Peter
Fenwick (for the structured coding model in the original
.I bzip,
and many refinements), and Alistair Moffat, Radford Neal and Ian Witten
(for the arithmetic coder in the original
.I bzip).
I am much indebted for their help, support and advice.
See the manual in the source distribution for pointers to
sources of documentation.
Christian von Roques encouraged me to look for faster
sorting algorithms, so as to speed up compression.
Bela Lubkin encouraged me to improve the worst-case
compression performance.
Many people sent patches, helped with portability problems,
lent machines, gave advice and were generally helpful.
I am much
indebted for their help, support and advice. See the manual in the
source distribution for pointers to sources of documentation. Christian
von Roques encouraged me to look for faster sorting algorithms, so as to
speed up compression. Bela Lubkin encouraged me to improve the
worst-case compression performance. Many people sent patches, helped
with portability problems, lent machines, gave advice and were generally
helpful.

436
bzip2.1.preformatted
View File

@ -1,24 +1,20 @@
bzip2(1) bzip2(1)
NNAAMMEE
bzip2, bunzip2 - a block-sorting file compressor, v0.9.0
bzip2, bunzip2 - a block-sorting file compressor, v0.9.5
bzcat - decompresses files to stdout
bzip2recover - recovers data from damaged bzip2 files
SSYYNNOOPPSSIISS
bbzziipp22 [ --ccddffkkssttvvzzVVLL112233445566778899 ] [ _f_i_l_e_n_a_m_e_s _._._. ]
bbzziipp22 [ --ccddffkkqqssttvvzzVVLL112233445566778899 ] [ _f_i_l_e_n_a_m_e_s _._._. ]
bbuunnzziipp22 [ --ffkkvvssVVLL ] [ _f_i_l_e_n_a_m_e_s _._._. ]
bbzzccaatt [ --ss ] [ _f_i_l_e_n_a_m_e_s _._._. ]
bbzziipp22rreeccoovveerr _f_i_l_e_n_a_m_e
DDEESSCCRRIIPPTTIIOONN
_b_z_i_p_2 compresses files using the Burrows-Wheeler block-
_b_z_i_p_2 compresses files using the Burrows-Wheeler block
sorting text compression algorithm, and Huffman coding.
Compression is generally considerably better than that
achieved by more conventional LZ77/LZ78-based compressors,
@ -26,22 +22,22 @@ DDEESSCCRRIIPPTTIIOONN
tistical compressors.
The command-line options are deliberately very similar to
those of _G_N_U _G_z_i_p_, but they are not identical.
those of _G_N_U _g_z_i_p_, but they are not identical.
_b_z_i_p_2 expects a list of file names to accompany the com-
mand-line flags. Each file is replaced by a compressed
version of itself, with the name "original_name.bz2".
Each compressed file has the same modification date and
permissions as the corresponding original, so that these
properties can be correctly restored at decompression
time. File name handling is naive in the sense that there
is no mechanism for preserving original file names, per-
missions and dates in filesystems which lack these con-
cepts, or have serious file name length restrictions, such
as MS-DOS.
Each compressed file has the same modification date, per-
missions, and, when possible, ownership as the correspond-
ing original, so that these properties can be correctly
restored at decompression time. File name handling is
naive in the sense that there is no mechanism for preserv-
ing original file names, permissions, ownerships or dates
in filesystems which lack these concepts, or have serious
file name length restrictions, such as MS-DOS.
_b_z_i_p_2 and _b_u_n_z_i_p_2 will by default not overwrite existing
files; if you want this to happen, specify the -f flag.
files. If you want this to happen, specify the -f flag.
If no file names are specified, _b_z_i_p_2 compresses from
standard input to standard output. In this case, _b_z_i_p_2
@ -49,42 +45,50 @@ DDEESSCCRRIIPPTTIIOONN
this would be entirely incomprehensible and therefore
pointless.
_b_u_n_z_i_p_2 (or _b_z_i_p_2 _-_d ) decompresses and restores all spec-
ified files whose names end in ".bz2". Files without this
suffix are ignored. Again, supplying no filenames causes
decompression from standard input to standard output.
_b_u_n_z_i_p_2 (or _b_z_i_p_2 _-_d_) decompresses all specified files.
Files which were not created by _b_z_i_p_2 will be detected and
ignored, and a warning issued. _b_z_i_p_2 attempts to guess
the filename for the decompressed file from that of the
compressed file as follows:
filename.bz2 becomes filename
filename.bz becomes filename
filename.tbz2 becomes filename.tar
filename.tbz becomes filename.tar
anyothername becomes anyothername.out
If the file does not end in one of the recognised endings,
_._b_z_2_, _._b_z_, _._t_b_z_2 or _._t_b_z_, _b_z_i_p_2 complains that it cannot
guess the name of the original file, and uses the original
name with _._o_u_t appended.
As with compression, supplying no filenames causes decom-
pression from standard input to standard output.
_b_u_n_z_i_p_2 will correctly decompress a file which is the con-
catenation of two or more compressed files. The result is
the concatenation of the corresponding uncompressed files.
Integrity testing (-t) of concatenated compressed files is
1
bzip2(1) bzip2(1)
also supported.
You can also compress or decompress files to the standard
output by giving the -c flag. Multiple files may be com-
You can also compress or decompress files to the standard
output by giving the -c flag. Multiple files may be com-
pressed and decompressed like this. The resulting outputs
are fed sequentially to stdout. Compression of multiple
files in this manner generates a stream containing multi-
are fed sequentially to stdout. Compression of multiple
files in this manner generates a stream containing multi-
ple compressed file representations. Such a stream can be
decompressed correctly only by _b_z_i_p_2 version 0.9.0 or
later. Earlier versions of _b_z_i_p_2 will stop after decom-
decompressed correctly only by _b_z_i_p_2 version 0.9.0 or
later. Earlier versions of _b_z_i_p_2 will stop after decom-
pressing the first file in the stream.
_b_z_c_a_t (or _b_z_i_p_2 _-_d_c ) decompresses all specified files to
_b_z_c_a_t (or _b_z_i_p_2 _-_d_c_) decompresses all specified files to
the standard output.
_b_z_i_p_2 will read arguments from the environment variables
_B_Z_I_P_2 and _B_Z_I_P_, in that order, and will process them
before any arguments read from the command line. This
gives a convenient way to supply default arguments.
Compression is always performed, even if the compressed
file is slightly larger than the original. Files of less
than about one hundred bytes tend to get larger, since the
@ -101,121 +105,19 @@ bzip2(1) bzip2(1)
corruption going undetected is microscopic, about one
chance in four billion for each file processed. Be aware,
though, that the check occurs upon decompression, so it
can only tell you that that something is wrong. It can't
help you recover the original uncompressed data. You can
use _b_z_i_p_2_r_e_c_o_v_e_r to try to recover data from damaged
files.
can only tell you that something is wrong. It can't help
you recover the original uncompressed data. You can use
_b_z_i_p_2_r_e_c_o_v_e_r to try to recover data from damaged files.
Return values: 0 for a normal exit, 1 for environmental
problems (file not found, invalid flags, I/O errors, &c),
Return values: 0 for a normal exit, 1 for environmental
problems (file not found, invalid flags, I/O errors, &c),
2 to indicate a corrupt compressed file, 3 for an internal
consistency error (eg, bug) which caused _b_z_i_p_2 to panic.
MMEEMMOORRYY MMAANNAAGGEEMMEENNTT
_B_z_i_p_2 compresses large files in blocks. The block size
affects both the compression ratio achieved, and the
amount of memory needed both for compression and decom-
pression. The flags -1 through -9 specify the block size
to be 100,000 bytes through 900,000 bytes (the default)
respectively. At decompression-time, the block size used
for compression is read from the header of the compressed
file, and _b_u_n_z_i_p_2 then allocates itself just enough memory
to decompress the file. Since block sizes are stored in
compressed files, it follows that the flags -1 to -9 are
irrelevant to and so ignored during decompression.
2
bzip2(1) bzip2(1)
Compression and decompression requirements, in bytes, can
be estimated as:
Compression: 400k + ( 7 x block size )
Decompression: 100k + ( 4 x block size ), or
100k + ( 2.5 x block size )
Larger block sizes give rapidly diminishing marginal
returns; most of the compression comes from the first two
or three hundred k of block size, a fact worth bearing in
mind when using _b_z_i_p_2 on small machines. It is also
important to appreciate that the decompression memory
requirement is set at compression-time by the choice of
block size.
For files compressed with the default 900k block size,
_b_u_n_z_i_p_2 will require about 3700 kbytes to decompress. To
support decompression of any file on a 4 megabyte machine,
_b_u_n_z_i_p_2 has an option to decompress using approximately
half this amount of memory, about 2300 kbytes. Decompres-
sion speed is also halved, so you should use this option
only where necessary. The relevant flag is -s.
In general, try and use the largest block size memory con-
straints allow, since that maximises the compression
achieved. Compression and decompression speed are virtu-
ally unaffected by block size.
Another significant point applies to files which fit in a
single block -- that means most files you'd encounter
using a large block size. The amount of real memory
touched is proportional to the size of the file, since the
file is smaller than a block. For example, compressing a
file 20,000 bytes long with the flag -9 will cause the
compressor to allocate around 6700k of memory, but only
touch 400k + 20000 * 7 = 540 kbytes of it. Similarly, the
decompressor will allocate 3700k but only touch 100k +
20000 * 4 = 180 kbytes.
Here is a table which summarises the maximum memory usage
for different block sizes. Also recorded is the total
compressed size for 14 files of the Calgary Text Compres-
sion Corpus totalling 3,141,622 bytes. This column gives
some feel for how compression varies with block size.
These figures tend to understate the advantage of larger
block sizes for larger files, since the Corpus is domi-
nated by smaller files.
Compress Decompress Decompress Corpus
Flag usage usage -s usage Size
-1 1100k 500k 350k 914704
-2 1800k 900k 600k 877703
3
bzip2(1) bzip2(1)
-3 2500k 1300k 850k 860338
-4 3200k 1700k 1100k 846899
-5 3900k 2100k 1350k 845160
-6 4600k 2500k 1600k 838626
-7 5400k 2900k 1850k 834096
-8 6000k 3300k 2100k 828642
-9 6700k 3700k 2350k 828642
OOPPTTIIOONNSS
--cc ----ssttddoouutt
Compress or decompress to standard output. -c will
decompress multiple files to stdout, but will only
compress a single file to stdout.
Compress or decompress to standard output.
--dd ----ddeeccoommpprreessss
Force decompression. _b_z_i_p_2_, _b_u_n_z_i_p_2 and _b_z_c_a_t are
@ -235,7 +137,9 @@ OOPPTTIIOONNSS
--ff ----ffoorrccee
Force overwrite of output files. Normally, _b_z_i_p_2
will not overwrite existing output files.
will not overwrite existing output files. Also
forces _b_z_i_p_2 to break hard links to files, which it
otherwise wouldn't do.
--kk ----kkeeeepp
Keep (don't delete) input files during compression
@ -254,19 +158,12 @@ OOPPTTIIOONNSS
figure, at the expense of your compression ratio.
In short, if your machine is low on memory (8
megabytes or less), use -s for everything. See
MEMORY MANAGEMENT above.
4
bzip2(1) bzip2(1)
MEMORY MANAGEMENT below.
--qq ----qquuiieett
Suppress non-essential warning messages. Messages
pertaining to I/O errors and other critical events
will not be suppressed.
--vv ----vveerrbboossee
Verbose mode -- show the compression ratio for each
@ -281,22 +178,96 @@ bzip2(1) bzip2(1)
--11 ttoo --99
Set the block size to 100 k, 200 k .. 900 k when
compressing. Has no effect when decompressing.
See MEMORY MANAGEMENT above.
See MEMORY MANAGEMENT below.
----rreeppeettiittiivvee--ffaasstt
_b_z_i_p_2 injects some small pseudo-random variations
into very repetitive blocks to limit worst-case
performance during compression. If sorting runs
into difficulties, the block is randomised, and
sorting is restarted. Very roughly, _b_z_i_p_2 persists
for three times as long as a well-behaved input
would take before resorting to randomisation. This
flag makes it give up much sooner.
---- Treats all subsequent arguments as file names, even
if they start with a dash. This is so you can han-
dle files with names beginning with a dash, for
example: bzip2 -- -myfilename.
----rreeppeettiittiivvee--ffaasstt ----rreeppeettiittiivvee--bbeesstt
These flags are redundant in versions 0.9.5 and
above. They provided some coarse control over the
behaviour of the sorting algorithm in earlier ver-
sions, which was sometimes useful. 0.9.5 and above
have an improved algorithm which renders these
flags irrelevant.
----rreeppeettiittiivvee--bbeesstt
Opposite of --repetitive-fast; try a lot harder
before resorting to randomisation.
MMEEMMOORRYY MMAANNAAGGEEMMEENNTT
_b_z_i_p_2 compresses large files in blocks. The block size
affects both the compression ratio achieved, and the
amount of memory needed for compression and decompression.
The flags -1 through -9 specify the block size to be
100,000 bytes through 900,000 bytes (the default) respec-
tively. At decompression time, the block size used for
compression is read from the header of the compressed
file, and _b_u_n_z_i_p_2 then allocates itself just enough memory
to decompress the file. Since block sizes are stored in
compressed files, it follows that the flags -1 to -9 are
irrelevant to and so ignored during decompression.
Compression and decompression requirements, in bytes, can
be estimated as:
Compression: 400k + ( 8 x block size )
Decompression: 100k + ( 4 x block size ), or
100k + ( 2.5 x block size )
Larger block sizes give rapidly diminishing marginal
returns. Most of the compression comes from the first two
or three hundred k of block size, a fact worth bearing in
mind when using _b_z_i_p_2 on small machines. It is also
important to appreciate that the decompression memory
requirement is set at compression time by the choice of
block size.
For files compressed with the default 900k block size,
_b_u_n_z_i_p_2 will require about 3700 kbytes to decompress. To
support decompression of any file on a 4 megabyte machine,
_b_u_n_z_i_p_2 has an option to decompress using approximately
half this amount of memory, about 2300 kbytes. Decompres-
sion speed is also halved, so you should use this option
only where necessary. The relevant flag is -s.
In general, try and use the largest block size memory con-
straints allow, since that maximises the compression
achieved. Compression and decompression speed are virtu-
ally unaffected by block size.
Another significant point applies to files which fit in a
single block -- that means most files you'd encounter
using a large block size. The amount of real memory
touched is proportional to the size of the file, since the
file is smaller than a block. For example, compressing a
file 20,000 bytes long with the flag -9 will cause the
compressor to allocate around 7600k of memory, but only
touch 400k + 20000 * 8 = 560 kbytes of it. Similarly, the
decompressor will allocate 3700k but only touch 100k +
20000 * 4 = 180 kbytes.
Here is a table which summarises the maximum memory usage
for different block sizes. Also recorded is the total
compressed size for 14 files of the Calgary Text Compres-
sion Corpus totalling 3,141,622 bytes. This column gives
some feel for how compression varies with block size.
These figures tend to understate the advantage of larger
block sizes for larger files, since the Corpus is domi-
nated by smaller files.
Compress Decompress Decompress Corpus
Flag usage usage -s usage Size
-1 1200k 500k 350k 914704
-2 2000k 900k 600k 877703
-3 2800k 1300k 850k 860338
-4 3600k 1700k 1100k 846899
-5 4400k 2100k 1350k 845160
-6 5200k 2500k 1600k 838626
-7 6100k 2900k 1850k 834096
-8 6800k 3300k 2100k 828642
-9 7600k 3700k 2350k 828642
RREECCOOVVEERRIINNGG DDAATTAA FFRROOMM DDAAMMAAGGEEDD FFIILLEESS
@ -314,7 +285,7 @@ RREECCOOVVEERRIINNGG DDAATTAA FFRROOMM DDAAMMAAGGEEDD F
_b_z_i_p_2_r_e_c_o_v_e_r is a simple program whose purpose is to
search for blocks in .bz2 files, and write each block out
into its own .bz2 file. You can then use _b_z_i_p_2 _-_t to test
into its own .bz2 file. You can then use _b_z_i_p_2 -t to test
the integrity of the resulting files, and decompress those
which are undamaged.
@ -322,21 +293,9 @@ RREECCOOVVEERRIINNGG DDAATTAA FFRROOMM DDAAMMAAGGEEDD F
aged file, and writes a number of files "rec0001file.bz2",
"rec0002file.bz2", etc, containing the extracted blocks.
The output filenames are designed so that the use of
5
bzip2(1) bzip2(1)
wildcards in subsequent processing -- for example, "bzip2
-dc rec*file.bz2 > recovered_data" -- lists the files in
the "right" order.
-dc rec*file.bz2 > recovered_data" -- lists the files in
the correct order.
_b_z_i_p_2_r_e_c_o_v_e_r should be of most use dealing with large .bz2
files, as these will contain many blocks. It is clearly
@ -351,17 +310,15 @@ PPEERRFFOORRMMAANNCCEE NNOOTTEESS
The sorting phase of compression gathers together similar
strings in the file. Because of this, files containing
very long runs of repeated symbols, like "aabaabaabaab
..." (repeated several hundred times) may compress
extraordinarily slowly. You can use the -vvvvv option to
monitor progress in great detail, if you want. Decompres-
sion speed is unaffected.
..." (repeated several hundred times) may compress more
slowly than normal. Versions 0.9.5 and above fare much
better than previous versions in this respect. The ratio
between worst-case and average-case compression time is in
the region of 10:1. For previous versions, this figure
was more like 100:1. You can use the -vvvv option to mon-
itor progress in great detail, if you want.
Such pathological cases seem rare in practice, appearing
mostly in artificially-constructed test files, and in low-
level disk images. It may be inadvisable to use _b_z_i_p_2 to
compress the latter. If you do get a file which causes
severe slowness in compression, try making the block size
as small as possible, with flag -1.
Decompression speed is unaffected by these phenomena.
_b_z_i_p_2 usually allocates several megabytes of memory to
operate in, and then charges all over it in a fairly ran-
@ -376,88 +333,43 @@ PPEERRFFOORRMMAANNCCEE NNOOTTEESS
CCAAVVEEAATTSS
I/O error messages are not as helpful as they could be.
_B_z_i_p_2 tries hard to detect I/O errors and exit cleanly,
_b_z_i_p_2 tries hard to detect I/O errors and exit cleanly,
but the details of what the problem is sometimes seem
rather misleading.
This manual page pertains to version 0.9.0 of _b_z_i_p_2_. Com-
This manual page pertains to version 0.9.5 of _b_z_i_p_2_. Com-
pressed data created by this version is entirely forwards
and backwards compatible with the previous public release,
version 0.1pl2, but with the following exception: 0.9.0
can correctly decompress multiple concatenated compressed
files. 0.1pl2 cannot do this; it will stop after decom-
pressing just the first file in the stream.
and backwards compatible with the previous public
releases, versions 0.1pl2 and 0.9.0, but with the follow-
ing exception: 0.9.0 and above can correctly decompress
multiple concatenated compressed files. 0.1pl2 cannot do
this; it will stop after decompressing just the first file
in the stream.
6
bzip2(1) bzip2(1)
Wildcard expansion for Windows 95 and NT is flaky.
_b_z_i_p_2_r_e_c_o_v_e_r uses 32-bit integers to represent bit posi-
tions in compressed files, so it cannot handle compressed
files more than 512 megabytes long. This could easily be
_b_z_i_p_2_r_e_c_o_v_e_r uses 32-bit integers to represent bit posi-
tions in compressed files, so it cannot handle compressed
files more than 512 megabytes long. This could easily be
fixed.
AAUUTTHHOORR
Julian Seward, jseward@acm.org.
http://www.muraroa.demon.co.uk
The ideas embodied in _b_z_i_p_2 are due to (at least) the fol-
lowing people: Michael Burrows and David Wheeler (for the
block sorting transformation), David Wheeler (again, for
lowing people: Michael Burrows and David Wheeler (for the
block sorting transformation), David Wheeler (again, for
the Huffman coder), Peter Fenwick (for the structured cod-
ing model in the original _b_z_i_p_, and many refinements), and
Alistair Moffat, Radford Neal and Ian Witten (for the
Alistair Moffat, Radford Neal and Ian Witten (for the
arithmetic coder in the original _b_z_i_p_)_. I am much
indebted for their help, support and advice. See the man-
ual in the source distribution for pointers to sources of
ual in the source distribution for pointers to sources of
documentation. Christian von Roques encouraged me to look
for faster sorting algorithms, so as to speed up compres-
for faster sorting algorithms, so as to speed up compres-
sion. Bela Lubkin encouraged me to improve the worst-case
compression performance. Many people sent patches, helped
with portability problems, lent machines, gave advice and
with portability problems, lent machines, gave advice and
were generally helpful.
7

549
bzip2.c
View File

@ -7,7 +7,7 @@
This file is a part of bzip2 and/or libbzip2, a program and
library for lossless, block-sorting data compression.
Copyright (C) 1996-1998 Julian R Seward. All rights reserved.
Copyright (C) 1996-1999 Julian R Seward. All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
@ -40,9 +40,9 @@
NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
Julian Seward, Guildford, Surrey, UK.
Julian Seward, Cambridge, UK.
jseward@acm.org
bzip2/libbzip2 version 0.9.0c of 18 October 1998
bzip2/libbzip2 version 0.9.5 of 24 May 1999
This program is based on (at least) the work of:
Mike Burrows
@ -123,8 +123,10 @@
--*/
#define BZ_LCCWIN32 0
#ifdef _WIN32
#if defined(_WIN32) && !defined(__CYGWIN32__)
#undef BZ_LCCWIN32
#define BZ_LCCWIN32 1
#undef BZ_UNIX
#define BZ_UNIX 0
#endif
@ -139,6 +141,8 @@
#include <string.h>
#include <signal.h>
#include <math.h>
#include <errno.h>
#include <ctype.h>
#include "bzlib.h"
#define ERROR_IF_EOF(i) { if ((i) == EOF) ioError(); }
@ -166,6 +170,9 @@
# define APPEND_FILESPEC(root, name) \
root=snocString((root), (name))
# define APPEND_FLAG(root, name) \
root=snocString((root), (name))
# define SET_BINARY_MODE(fd) /**/
# ifdef __GNUC__
@ -173,6 +180,19 @@
# else
# define NORETURN /**/
# endif
# ifdef __DJGPP__
# include <io.h>
# include <fcntl.h>
# undef MY_LSTAT
# define MY_LSTAT stat
# undef SET_BINARY_MODE
# define SET_BINARY_MODE(fd) \
do { \
int retVal = setmode ( fileno ( fd ), \
O_BINARY ); \
ERROR_IF_MINUS_ONE ( retVal ); \
} while ( 0 )
# endif
#endif
@ -188,6 +208,9 @@
# define MY_STAT _stat
# define MY_S_IFREG(x) ((x) & _S_IFREG)
# define APPEND_FLAG(root, name) \
root=snocString((root), (name))
# if 0
/*-- lcc-win32 seems to expand wildcards itself --*/
# define APPEND_FILESPEC(root, spec) \
@ -254,7 +277,7 @@ typedef int IntNative;
Int32 verbosity;
Bool keepInputFiles, smallMode;
Bool forceOverwrite, testFailsExist;
Bool forceOverwrite, testFailsExist, noisy;
Int32 numFileNames, numFilesProcessed, blockSize100k;
@ -274,8 +297,9 @@ Int32 srcMode;
#define FILE_NAME_LEN 1034
Int32 longestFileName;
Char inName[FILE_NAME_LEN];
Char inName [FILE_NAME_LEN];
Char outName[FILE_NAME_LEN];
Char tmpName[FILE_NAME_LEN];
Char *progName;
Char progNameReally[FILE_NAME_LEN];
FILE *outputHandleJustInCase;
@ -467,6 +491,7 @@ Bool uncompressStream ( FILE *zStream, FILE *stream )
if (streamNo == 1) {
return False;
} else {
if (noisy)
fprintf ( stderr,
"\n%s: %s: trailing garbage after EOF ignored\n",
progName, inName );
@ -532,32 +557,31 @@ Bool testStream ( FILE *zStream )
errhandler:
bzReadClose ( &bzerr_dummy, bzf );
if (verbosity == 0)
fprintf ( stderr, "%s: %s: ", progName, inName );
switch (bzerr) {
case BZ_IO_ERROR:
errhandler_io:
ioError(); break;
case BZ_DATA_ERROR:
fprintf ( stderr,
"\n%s: data integrity (CRC) error in data\n",
inName );
"data integrity (CRC) error in data\n" );
return False;
case BZ_MEM_ERROR:
outOfMemory();
case BZ_UNEXPECTED_EOF:
fprintf ( stderr,
"\n%s: file ends unexpectedly\n",
inName );
"file ends unexpectedly\n" );
return False;
case BZ_DATA_ERROR_MAGIC:
if (streamNo == 1) {
fprintf ( stderr,
"\n%s: bad magic number (ie, not created by bzip2)\n",
inName );
"bad magic number (file not created by bzip2)\n" );
return False;
} else {
if (noisy)
fprintf ( stderr,
"\n%s: %s: trailing garbage after EOF ignored\n",
progName, inName );
"trailing garbage after EOF ignored\n" );
return True;
}
default:
@ -576,6 +600,7 @@ Bool testStream ( FILE *zStream )
/*---------------------------------------------*/
void cadvise ( void )
{
if (noisy)
fprintf (
stderr,
"\nIt is possible that the compressed file(s) have become corrupted.\n"
@ -589,6 +614,7 @@ void cadvise ( void )
/*---------------------------------------------*/
void showFileNames ( void )
{
if (noisy)
fprintf (
stderr,
"\tInput file = %s, output file = %s\n",
@ -603,6 +629,7 @@ void cleanUpAndFail ( Int32 ec )
IntNative retVal;
if ( srcMode == SM_F2F && opMode != OM_TEST ) {
if (noisy)
fprintf ( stderr, "%s: Deleting output file %s, if it exists.\n",
progName, outName );
if (outputHandleJustInCase != NULL)
@ -613,7 +640,7 @@ void cleanUpAndFail ( Int32 ec )
"%s: WARNING: deletion of output file (apparently) failed.\n",
progName );
}
if (numFileNames > 0 && numFilesProcessed < numFileNames) {
if (noisy && numFileNames > 0 && numFilesProcessed < numFileNames) {
fprintf ( stderr,
"%s: WARNING: some files have not been processed:\n"
"\t%d specified on command line, %d not processed yet.\n\n",
@ -639,7 +666,7 @@ void panic ( Char* s )
/*---------------------------------------------*/
void crcError ()
void crcError ( void )
{
fprintf ( stderr,
"\n%s: Data integrity error when decompressing.\n",
@ -665,7 +692,7 @@ void compressedStreamEOF ( void )
/*---------------------------------------------*/
void ioError ( )
void ioError ( void )
{
fprintf ( stderr,
"\n%s: I/O or other error, bailing out. Possible reason follows.\n",
@ -680,7 +707,7 @@ void ioError ( )
void mySignalCatcher ( IntNative n )
{
fprintf ( stderr,
"\n%s: Control-C (or similar) caught, quitting.\n",
"\n%s: Control-C or similar caught, quitting.\n",
progName );
cleanUpAndFail(1);
}
@ -740,9 +767,10 @@ void copyFileName ( Char* to, Char* from )
if ( strlen(from) > FILE_NAME_LEN-10 ) {
fprintf (
stderr,
"bzip2: file name\n`%s'\nis suspiciously (> 1024 chars) long.\n"
"Try using a reasonable file name instead. Sorry! :)\n",
from
"bzip2: file name\n`%s'\n"
"is suspiciously (more than %d chars) long.\n"
"Try using a reasonable file name instead. Sorry! :-)\n",
from, FILE_NAME_LEN-10
);
exit(1);
}
@ -778,6 +806,21 @@ Bool notAStandardFile ( Char* name )
}
/*---------------------------------------------*/
/*--
rac 11/21/98 see if file has hard links to it
--*/
Int32 countHardLinks ( Char* name )
{
IntNative i;
struct MY_STAT statBuf;
i = MY_LSTAT ( name, &statBuf );
if (i != 0) return 0;
return (statBuf.st_nlink - 1);
}
/*---------------------------------------------*/
void copyDatePermissionsAndOwner ( Char *srcName, Char *dstName )
{
@ -793,17 +836,14 @@ void copyDatePermissionsAndOwner ( Char *srcName, Char *dstName )
retVal = chmod ( dstName, statBuf.st_mode );
ERROR_IF_NOT_ZERO ( retVal );
/* Not sure if this is really portable or not. Causes
problems on my x86-Linux Redhat 5.0 box. Decided
to omit it from 0.9.0. JRS, 27 June 98. If you
understand Unix file semantics and portability issues
well enough to fix this properly, drop me a line
at jseward@acm.org.
retVal = chown ( dstName, statBuf.st_uid, statBuf.st_gid );
ERROR_IF_NOT_ZERO ( retVal );
*/
retVal = utime ( dstName, &uTimBuf );
ERROR_IF_NOT_ZERO ( retVal );
retVal = chown ( dstName, statBuf.st_uid, statBuf.st_gid );
/* chown() will in many cases return with EPERM, which can
be safely ignored.
*/
#endif
}
@ -819,20 +859,6 @@ void setInterimPermissions ( Char *dstName )
}
/*---------------------------------------------*/
Bool endsInBz2 ( Char* name )
{
Int32 n = strlen ( name );
if (n <= 4) return False;
return
(name[n-4] == '.' &&
name[n-3] == 'b' &&
name[n-2] == 'z' &&
name[n-1] == '2');
}
/*---------------------------------------------*/
Bool containsDubiousChars ( Char* name )
{
@ -843,50 +869,95 @@ Bool containsDubiousChars ( Char* name )
}
/*---------------------------------------------*/
#define BZ_N_SUFFIX_PAIRS 4
Char* zSuffix[BZ_N_SUFFIX_PAIRS]
= { ".bz2", ".bz", ".tbz2", ".tbz" };
Char* unzSuffix[BZ_N_SUFFIX_PAIRS]
= { "", "", ".tar", ".tar" };
Bool hasSuffix ( Char* s, Char* suffix )
{
Int32 ns = strlen(s);
Int32 nx = strlen(suffix);
if (ns < nx) return False;
if (strcmp(s + ns - nx, suffix) == 0) return True;
return False;
}
Bool mapSuffix ( Char* name,
Char* oldSuffix, Char* newSuffix )
{
if (!hasSuffix(name,oldSuffix)) return False;
name[strlen(name)-strlen(oldSuffix)] = 0;
strcat ( name, newSuffix );
return True;
}
/*---------------------------------------------*/
void compress ( Char *name )
{
FILE *inStr;
FILE *outStr;
FILE *inStr;
FILE *outStr;
Int32 n, i;
if (name == NULL && srcMode != SM_I2O)
panic ( "compress: bad modes\n" );
switch (srcMode) {
case SM_I2O: copyFileName ( inName, "(stdin)" );
copyFileName ( outName, "(stdout)" ); break;
case SM_F2F: copyFileName ( inName, name );
copyFileName ( outName, name );
strcat ( outName, ".bz2" ); break;
case SM_F2O: copyFileName ( inName, name );
copyFileName ( outName, "(stdout)" ); break;
case SM_I2O:
copyFileName ( inName, "(stdin)" );
copyFileName ( outName, "(stdout)" );
break;
case SM_F2F:
copyFileName ( inName, name );
copyFileName ( outName, name );
strcat ( outName, ".bz2" );
break;
case SM_F2O:
copyFileName ( inName, name );
copyFileName ( outName, "(stdout)" );
break;
}
if ( srcMode != SM_I2O && containsDubiousChars ( inName ) ) {
if (noisy)
fprintf ( stderr, "%s: There are no files matching `%s'.\n",
progName, inName );
return;
}
if ( srcMode != SM_I2O && !fileExists ( inName ) ) {
fprintf ( stderr, "%s: Input file %s doesn't exist, skipping.\n",
progName, inName );
fprintf ( stderr, "%s: Can't open input file %s: %s.\n",
progName, inName, strerror(errno) );
return;
}
if ( srcMode != SM_I2O && endsInBz2 ( inName )) {
fprintf ( stderr, "%s: Input file name %s ends in `.bz2', skipping.\n",
progName, inName );
return;
for (i = 0; i < BZ_N_SUFFIX_PAIRS; i++) {
if (hasSuffix(inName, zSuffix[i])) {
if (noisy)
fprintf ( stderr,
"%s: Input file %s already has %s suffix.\n",
progName, inName, zSuffix[i] );
return;
}
}
if ( srcMode != SM_I2O && notAStandardFile ( inName )) {
fprintf ( stderr, "%s: Input file %s is not a normal file, skipping.\n",
if ( srcMode == SM_F2F && !forceOverwrite && notAStandardFile ( inName )) {
if (noisy)
fprintf ( stderr, "%s: Input file %s is not a normal file.\n",
progName, inName );
return;
}
if ( srcMode == SM_F2F && !forceOverwrite && fileExists ( outName ) ) {
fprintf ( stderr, "%s: Output file %s already exists, skipping.\n",
fprintf ( stderr, "%s: Output file %s already exists.\n",
progName, outName );
return;
}
if ( srcMode == SM_F2F && !forceOverwrite &&
(n=countHardLinks ( inName )) > 0) {
fprintf ( stderr, "%s: Input file %s has %d other link%s.\n",
progName, inName, n, n > 1 ? "s" : "" );
return;
}
switch ( srcMode ) {
@ -912,11 +983,12 @@ void compress ( Char *name )
progName );
fprintf ( stderr, "%s: For help, type: `%s --help'.\n",
progName, progName );
if ( inStr != NULL ) fclose ( inStr );
return;
};
if ( inStr == NULL ) {
fprintf ( stderr, "%s: Can't open input file %s, skipping.\n",
progName, inName );
fprintf ( stderr, "%s: Can't open input file %s: %s.\n",
progName, inName, strerror(errno) );
return;
};
break;
@ -925,13 +997,15 @@ void compress ( Char *name )
inStr = fopen ( inName, "rb" );
outStr = fopen ( outName, "wb" );
if ( outStr == NULL) {
fprintf ( stderr, "%s: Can't create output file %s, skipping.\n",
progName, outName );
fprintf ( stderr, "%s: Can't create output file %s: %s.\n",
progName, outName, strerror(errno) );
if ( inStr != NULL ) fclose ( inStr );
return;
}
if ( inStr == NULL ) {
fprintf ( stderr, "%s: Can't open input file %s, skipping.\n",
progName, inName );
fprintf ( stderr, "%s: Can't open input file %s: %s.\n",
progName, inName, strerror(errno) );
if ( outStr != NULL ) fclose ( outStr );
return;
};
setInterimPermissions ( outName );
@ -967,51 +1041,72 @@ void compress ( Char *name )
/*---------------------------------------------*/
void uncompress ( Char *name )
{
FILE *inStr;
FILE *outStr;
Bool magicNumberOK;
FILE *inStr;
FILE *outStr;
Int32 n, i;
Bool magicNumberOK;
Bool cantGuess;
if (name == NULL && srcMode != SM_I2O)
panic ( "uncompress: bad modes\n" );
cantGuess = False;
switch (srcMode) {
case SM_I2O: copyFileName ( inName, "(stdin)" );
copyFileName ( outName, "(stdout)" ); break;
case SM_F2F: copyFileName ( inName, name );
copyFileName ( outName, name );
if (endsInBz2 ( outName ))
outName [ strlen ( outName ) - 4 ] = '\0';
break;
case SM_F2O: copyFileName ( inName, name );
copyFileName ( outName, "(stdout)" ); break;
case SM_I2O:
copyFileName ( inName, "(stdin)" );
copyFileName ( outName, "(stdout)" );
break;
case SM_F2F:
copyFileName ( inName, name );
copyFileName ( outName, name );
for (i = 0; i < BZ_N_SUFFIX_PAIRS; i++)
if (mapSuffix(outName,zSuffix[i],unzSuffix[i]))
goto zzz;
cantGuess = True;
strcat ( outName, ".out" );
break;
case SM_F2O:
copyFileName ( inName, name );
copyFileName ( outName, "(stdout)" );
break;
}
zzz:
if ( srcMode != SM_I2O && containsDubiousChars ( inName ) ) {
if (noisy)
fprintf ( stderr, "%s: There are no files matching `%s'.\n",
progName, inName );
return;
}
if ( srcMode != SM_I2O && !fileExists ( inName ) ) {
fprintf ( stderr, "%s: Input file %s doesn't exist, skipping.\n",
progName, inName );
return;
}
if ( srcMode != SM_I2O && !endsInBz2 ( inName )) {
fprintf ( stderr,
"%s: Input file name %s doesn't end in `.bz2', skipping.\n",
progName, inName );
return;
}
if ( srcMode != SM_I2O && notAStandardFile ( inName )) {
fprintf ( stderr, "%s: Input file %s is not a normal file, skipping.\n",
fprintf ( stderr, "%s: Can't open input file %s: %s.\n",
progName, inName, strerror(errno) );
return;
}
if ( srcMode == SM_F2F && !forceOverwrite && notAStandardFile ( inName )) {
if (noisy)
fprintf ( stderr, "%s: Input file %s is not a normal file.\n",
progName, inName );
return;
}
if ( /* srcMode == SM_F2F implied && */ cantGuess ) {
if (noisy)
fprintf ( stderr,
"%s: Can't guess original name for %s -- using %s\n",
progName, inName, outName );
/* just a warning, no return */
}
if ( srcMode == SM_F2F && !forceOverwrite && fileExists ( outName ) ) {
fprintf ( stderr, "%s: Output file %s already exists, skipping.\n",
fprintf ( stderr, "%s: Output file %s already exists.\n",
progName, outName );
return;
}
if ( srcMode == SM_F2F && !forceOverwrite &&
(n=countHardLinks ( inName ) ) > 0) {
fprintf ( stderr, "%s: Input file %s has %d other link%s.\n",
progName, inName, n, n > 1 ? "s" : "" );
return;
}
switch ( srcMode ) {
@ -1032,8 +1127,9 @@ void uncompress ( Char *name )
inStr = fopen ( inName, "rb" );
outStr = stdout;
if ( inStr == NULL ) {
fprintf ( stderr, "%s: Can't open input file %s, skipping.\n",
progName, inName );
fprintf ( stderr, "%s: Can't open input file %s:%s.\n",
progName, inName, strerror(errno) );
if ( inStr != NULL ) fclose ( inStr );
return;
};
break;
@ -1042,13 +1138,15 @@ void uncompress ( Char *name )
inStr = fopen ( inName, "rb" );
outStr = fopen ( outName, "wb" );
if ( outStr == NULL) {
fprintf ( stderr, "%s: Can't create output file %s, skipping.\n",
progName, outName );
fprintf ( stderr, "%s: Can't create output file %s: %s.\n",
progName, outName, strerror(errno) );
if ( inStr != NULL ) fclose ( inStr );
return;
}
if ( inStr == NULL ) {
fprintf ( stderr, "%s: Can't open input file %s, skipping.\n",
progName, inName );
fprintf ( stderr, "%s: Can't open input file %s: %s.\n",
progName, inName, strerror(errno) );
if ( outStr != NULL ) fclose ( outStr );
return;
};
setInterimPermissions ( outName );
@ -1091,9 +1189,9 @@ void uncompress ( Char *name )
fprintf ( stderr, "done\n" );
} else {
if (verbosity >= 1)
fprintf ( stderr, "not a bzip2 file, skipping.\n" ); else
fprintf ( stderr, "not a bzip2 file.\n" ); else
fprintf ( stderr,
"%s: %s is not a bzip2 file, skipping.\n",
"%s: %s is not a bzip2 file.\n",
progName, inName );
}
@ -1117,24 +1215,14 @@ void testf ( Char *name )
}
if ( srcMode != SM_I2O && containsDubiousChars ( inName ) ) {
if (noisy)
fprintf ( stderr, "%s: There are no files matching `%s'.\n",
progName, inName );
return;
}
if ( srcMode != SM_I2O && !fileExists ( inName ) ) {
fprintf ( stderr, "%s: Input file %s doesn't exist, skipping.\n",
progName, inName );
return;
}
if ( srcMode != SM_I2O && !endsInBz2 ( inName )) {
fprintf ( stderr,
"%s: Input file name %s doesn't end in `.bz2', skipping.\n",
progName, inName );
return;
}
if ( srcMode != SM_I2O && notAStandardFile ( inName )) {
fprintf ( stderr, "%s: Input file %s is not a normal file, skipping.\n",
progName, inName );
fprintf ( stderr, "%s: Can't open input %s: %s.\n",
progName, inName, strerror(errno) );
return;
}
@ -1155,8 +1243,8 @@ void testf ( Char *name )
case SM_F2O: case SM_F2F:
inStr = fopen ( inName, "rb" );
if ( inStr == NULL ) {
fprintf ( stderr, "%s: Can't open input file %s, skipping.\n",
progName, inName );
fprintf ( stderr, "%s: Can't open input file %s:%s.\n",
progName, inName, strerror(errno) );
return;
};
break;
@ -1186,13 +1274,13 @@ void license ( void )
fprintf ( stderr,
"bzip2, a block-sorting file compressor. "
"Version 0.9.0c, 18-Oct-98.\n"
"Version 0.9.5d, 4-Sept-99.\n"
" \n"
" Copyright (C) 1996, 1997, 1998 by Julian Seward.\n"
" Copyright (C) 1996, 1997, 1998, 1999 by Julian Seward.\n"
" \n"
" This program is free software; you can redistribute it and/or modify\n"
" it under the terms set out in the LICENSE file, which is included\n"
" in the bzip2-0.9.0c source distribution.\n"
" in the bzip2-0.9.5 source distribution.\n"
" \n"
" This program is distributed in the hope that it will be useful,\n"
" but WITHOUT ANY WARRANTY; without even the implied warranty of\n"
@ -1209,27 +1297,26 @@ void usage ( Char *fullProgName )
fprintf (
stderr,
"bzip2, a block-sorting file compressor. "
"Version 0.9.0c, 18-Oct-98.\n"
"Version 0.9.5d, 4-Sept-99.\n"
"\n usage: %s [flags and input files in any order]\n"
"\n"
" -h --help print this message\n"
" -d --decompress force decompression\n"
" -z --compress force compression\n"
" -k --keep keep (don't delete) input files\n"
" -f --force overwrite existing output filess\n"
" -f --force overwrite existing output files\n"
" -t --test test compressed file integrity\n"
" -c --stdout output to standard out\n"
" -q --quiet suppress noncritical error messages\n"
" -v --verbose be verbose (a 2nd -v gives more)\n"
" -L --license display software version & license\n"
" -V --version display software version & license\n"
" -s --small use less memory (at most 2500k)\n"
" -1 .. -9 set block size to 100k .. 900k\n"
" --repetitive-fast compress repetitive blocks faster\n"
" --repetitive-best compress repetitive blocks better\n"
"\n"
" If invoked as `bzip2', default action is to compress.\n"
" as `bunzip2', default action is to decompress.\n"
" as `bz2cat', default action is to decompress to stdout.\n"
" as `bzcat', default action is to decompress to stdout.\n"
"\n"
" If no file names are given, bzip2 compresses or decompresses\n"
" from standard input to standard output. You can combine\n"
@ -1244,19 +1331,29 @@ void usage ( Char *fullProgName )
}
/*---------------------------------------------*/
void redundant ( Char* flag )
{
fprintf (
stderr,
"%s: %s is redundant in versions 0.9.5 and above\n",
progName, flag );
}
/*---------------------------------------------*/
/*--
All the garbage from here to main() is purely to
implement a linked list of command-line arguments,
into which main() copies argv[1 .. argc-1].
The purpose of this ridiculous exercise is to
facilitate the expansion of wildcard characters
* and ? in filenames for halfwitted OSs like
MSDOS, Windows 95 and NT.
The purpose of this exercise is to facilitate
the expansion of wildcard characters * and ? in
filenames for OSs which don't know how to do it
themselves, like MSDOS, Windows 95 and NT.
The actual Dirty Work is done by the platform-specific
macro APPEND_FILESPEC.
The actual Dirty Work is done by the platform-
specific macro APPEND_FILESPEC.
--*/
typedef
@ -1308,8 +1405,34 @@ Cell *snocString ( Cell *root, Char *name )
/*---------------------------------------------*/
#define ISFLAG(s) (strcmp(aa->name, (s))==0)
void addFlagsFromEnvVar ( Cell** argList, Char* varName )
{
Int32 i, j, k;
Char *envbase, *p;
envbase = getenv(varName);
if (envbase != NULL) {
p = envbase;
i = 0;
while (True) {
if (p[i] == 0) break;
p += i;
i = 0;
while (isspace((Int32)(p[0]))) p++;
while (p[i] != 0 && !isspace((Int32)(p[i]))) i++;
if (i > 0) {
k = i; if (k > FILE_NAME_LEN-10) k = FILE_NAME_LEN-10;
for (j = 0; j < k; j++) tmpName[j] = p[j];
tmpName[k] = 0;
APPEND_FLAG(*argList, tmpName);
}
}
}
}
/*---------------------------------------------*/
#define ISFLAG(s) (strcmp(aa->name, (s))==0)
IntNative main ( IntNative argc, Char *argv[] )
{
@ -1317,6 +1440,7 @@ IntNative main ( IntNative argc, Char *argv[] )
Char *tmp;
Cell *argList;
Cell *aa;
Bool decode;
/*-- Be really really really paranoid :-) --*/
if (sizeof(Int32) != 4 || sizeof(UInt32) != 4 ||
@ -1332,27 +1456,27 @@ IntNative main ( IntNative argc, Char *argv[] )
}
/*-- Set up signal handlers --*/
signal (SIGINT, mySignalCatcher);
signal (SIGTERM, mySignalCatcher);
signal (SIGSEGV, mySIGSEGVorSIGBUScatcher);
#if BZ_UNIX
signal (SIGHUP, mySignalCatcher);
signal (SIGBUS, mySIGSEGVorSIGBUScatcher);
#endif
/*-- Initialise --*/
outputHandleJustInCase = NULL;
smallMode = False;
keepInputFiles = False;
forceOverwrite = False;
noisy = True;
verbosity = 0;
blockSize100k = 9;
testFailsExist = False;
numFileNames = 0;
numFilesProcessed = 0;
workFactor = 30;
i = j = 0; /* avoid bogus warning from egcs-1.1.X */
/*-- Set up signal handlers for mem access errors --*/
signal (SIGSEGV, mySIGSEGVorSIGBUScatcher);
#if BZ_UNIX
#ifndef __DJGPP__
signal (SIGBUS, mySIGSEGVorSIGBUScatcher);
#endif
#endif
copyFileName ( inName, "(none)" );
copyFileName ( outName, "(none)" );
@ -1363,8 +1487,12 @@ IntNative main ( IntNative argc, Char *argv[] )
if (*tmp == PATH_SEP) progName = tmp + 1;
/*-- Expand filename wildcards in arg list --*/
/*-- Copy flags from env var BZIP2, and
expand filename wildcards in arg list.
--*/
argList = NULL;
addFlagsFromEnvVar ( &argList, "BZIP2" );
addFlagsFromEnvVar ( &argList, "BZIP" );
for (i = 1; i <= argc-1; i++)
APPEND_FILESPEC(argList, argv[i]);
@ -1372,12 +1500,14 @@ IntNative main ( IntNative argc, Char *argv[] )
/*-- Find the length of the longest filename --*/
longestFileName = 7;
numFileNames = 0;
for (aa = argList; aa != NULL; aa = aa->link)
if (aa->name[0] != '-') {
numFileNames++;
if (longestFileName < (Int32)strlen(aa->name) )
longestFileName = (Int32)strlen(aa->name);
}
decode = True;
for (aa = argList; aa != NULL; aa = aa->link) {
if (ISFLAG("--")) { decode = False; continue; }
if (aa->name[0] == '-' && decode) continue;
numFileNames++;
if (longestFileName < (Int32)strlen(aa->name) )
longestFileName = (Int32)strlen(aa->name);
}
/*-- Determine source modes; flag handling may change this too. --*/
@ -1403,9 +1533,10 @@ IntNative main ( IntNative argc, Char *argv[] )
/*-- Look at the flags. --*/
for (aa = argList; aa != NULL; aa = aa->link)
if (aa->name[0] == '-' && aa->name[1] != '-')
for (j = 1; aa->name[j] != '\0'; j++)
for (aa = argList; aa != NULL; aa = aa->link) {
if (ISFLAG("--")) break;
if (aa->name[0] == '-' && aa->name[1] != '-') {
for (j = 1; aa->name[j] != '\0'; j++) {
switch (aa->name[j]) {
case 'c': srcMode = SM_F2O; break;
case 'd': opMode = OM_UNZ; break;
@ -1414,6 +1545,7 @@ IntNative main ( IntNative argc, Char *argv[] )
case 't': opMode = OM_TEST; break;
case 'k': keepInputFiles = True; break;
case 's': smallMode = True; break;
case 'q': noisy = False; break;
case '1': blockSize100k = 1; break;
case '2': blockSize100k = 2; break;
case '3': blockSize100k = 3; break;
@ -1427,17 +1559,21 @@ IntNative main ( IntNative argc, Char *argv[] )
case 'L': license(); break;
case 'v': verbosity++; break;
case 'h': usage ( progName );
exit ( 1 );
exit ( 0 );
break;
default: fprintf ( stderr, "%s: Bad flag `%s'\n",
progName, aa->name );
usage ( progName );
exit ( 1 );
break;
}
}
}
}
/*-- And again ... --*/
for (aa = argList; aa != NULL; aa = aa->link) {
if (ISFLAG("--")) break;
if (ISFLAG("--stdout")) srcMode = SM_F2O; else
if (ISFLAG("--decompress")) opMode = OM_UNZ; else
if (ISFLAG("--compress")) opMode = OM_Z; else
@ -1445,12 +1581,14 @@ IntNative main ( IntNative argc, Char *argv[] )
if (ISFLAG("--test")) opMode = OM_TEST; else
if (ISFLAG("--keep")) keepInputFiles = True; else
if (ISFLAG("--small")) smallMode = True; else
if (ISFLAG("--quiet")) noisy = False; else
if (ISFLAG("--version")) license(); else
if (ISFLAG("--license")) license(); else
if (ISFLAG("--repetitive-fast")) workFactor = 5; else
if (ISFLAG("--repetitive-best")) workFactor = 150; else
if (ISFLAG("--exponential")) workFactor = 1; else
if (ISFLAG("--repetitive-best")) redundant(aa->name); else
if (ISFLAG("--repetitive-fast")) redundant(aa->name); else
if (ISFLAG("--verbose")) verbosity++; else
if (ISFLAG("--help")) { usage ( progName ); exit ( 1 ); }
if (ISFLAG("--help")) { usage ( progName ); exit ( 0 ); }
else
if (strncmp ( aa->name, "--", 2) == 0) {
fprintf ( stderr, "%s: Bad flag `%s'\n", progName, aa->name );
@ -1460,13 +1598,8 @@ IntNative main ( IntNative argc, Char *argv[] )
}
if (verbosity > 4) verbosity = 4;
if (opMode == OM_Z && smallMode) blockSize100k = 2;
if (srcMode == SM_F2O && numFileNames == 0) {
fprintf ( stderr, "%s: -c expects at least one filename.\n",
progName );
exit ( 1 );
}
if (opMode == OM_Z && smallMode && blockSize100k > 2)
blockSize100k = 2;
if (opMode == OM_TEST && srcMode == SM_F2O) {
fprintf ( stderr, "%s: -c and -t cannot be used together.\n",
@ -1474,46 +1607,82 @@ IntNative main ( IntNative argc, Char *argv[] )
exit ( 1 );
}
if (srcMode == SM_F2O && numFileNames == 0)
srcMode = SM_I2O;
if (opMode != OM_Z) blockSize100k = 0;
if (srcMode == SM_F2F) {
signal (SIGINT, mySignalCatcher);
signal (SIGTERM, mySignalCatcher);
# if BZ_UNIX
signal (SIGHUP, mySignalCatcher);
# endif
}
if (opMode == OM_Z) {
if (srcMode == SM_I2O)
compress ( NULL );
else
for (aa = argList; aa != NULL; aa = aa->link)
if (aa->name[0] != '-') {
numFilesProcessed++;
compress ( aa->name );
}
} else
if (srcMode == SM_I2O) {
compress ( NULL );
} else {
decode = True;
for (aa = argList; aa != NULL; aa = aa->link) {
if (ISFLAG("--")) { decode = False; continue; }
if (aa->name[0] == '-' && decode) continue;
numFilesProcessed++;
compress ( aa->name );
}
}
}
else
if (opMode == OM_UNZ) {
if (srcMode == SM_I2O)
if (srcMode == SM_I2O) {
uncompress ( NULL );
else
for (aa = argList; aa != NULL; aa = aa->link)
if (aa->name[0] != '-') {
numFilesProcessed++;
uncompress ( aa->name );
}
} else {
} else {
decode = True;
for (aa = argList; aa != NULL; aa = aa->link) {
if (ISFLAG("--")) { decode = False; continue; }
if (aa->name[0] == '-' && decode) continue;
numFilesProcessed++;
uncompress ( aa->name );
}
}
}
else {
testFailsExist = False;
if (srcMode == SM_I2O)
if (srcMode == SM_I2O) {
testf ( NULL );
else
for (aa = argList; aa != NULL; aa = aa->link)
if (aa->name[0] != '-') {
numFilesProcessed++;
testf ( aa->name );
}
if (testFailsExist) {
} else {
decode = True;
for (aa = argList; aa != NULL; aa = aa->link) {
if (ISFLAG("--")) { decode = False; continue; }
if (aa->name[0] == '-' && decode) continue;
numFilesProcessed++;
testf ( aa->name );
}
}
if (testFailsExist && noisy) {
fprintf ( stderr,
"\n"
"You can use the `bzip2recover' program to *attempt* to recover\n"
"You can use the `bzip2recover' program to attempt to recover\n"
"data from undamaged sections of corrupted files.\n\n"
);
exit(2);
}
}
/* Free the argument list memory to mollify leak detectors
(eg) Purify, Checker. Serves no other useful purpose.
*/
aa = argList;
while (aa != NULL) {
Cell* aa2 = aa->link;
if (aa->name) free(aa->name);
free(aa);
aa = aa2;
}
return 0;
}

332
bzip2.txt
View File

@ -1,22 +1,20 @@
bzip2(1) bzip2(1)
NAME
bzip2, bunzip2 - a block-sorting file compressor, v0.9.0
bzip2, bunzip2 - a block-sorting file compressor, v0.9.5
bzcat - decompresses files to stdout
bzip2recover - recovers data from damaged bzip2 files
SYNOPSIS
bzip2 [ -cdfkstvzVL123456789 ] [ filenames ... ]
bzip2 [ -cdfkqstvzVL123456789 ] [ filenames ... ]
bunzip2 [ -fkvsVL ] [ filenames ... ]
bzcat [ -s ] [ filenames ... ]
bzip2recover filename
DESCRIPTION
bzip2 compresses files using the Burrows-Wheeler block-
bzip2 compresses files using the Burrows-Wheeler block
sorting text compression algorithm, and Huffman coding.
Compression is generally considerably better than that
achieved by more conventional LZ77/LZ78-based compressors,
@ -24,22 +22,22 @@ DESCRIPTION
tistical compressors.
The command-line options are deliberately very similar to
those of GNU Gzip, but they are not identical.
those of GNU gzip, but they are not identical.
bzip2 expects a list of file names to accompany the com-
mand-line flags. Each file is replaced by a compressed
version of itself, with the name "original_name.bz2".
Each compressed file has the same modification date and
permissions as the corresponding original, so that these
properties can be correctly restored at decompression
time. File name handling is naive in the sense that there
is no mechanism for preserving original file names, per-
missions and dates in filesystems which lack these con-
cepts, or have serious file name length restrictions, such
as MS-DOS.
Each compressed file has the same modification date, per-
missions, and, when possible, ownership as the correspond-
ing original, so that these properties can be correctly
restored at decompression time. File name handling is
naive in the sense that there is no mechanism for preserv-
ing original file names, permissions, ownerships or dates
in filesystems which lack these concepts, or have serious
file name length restrictions, such as MS-DOS.
bzip2 and bunzip2 will by default not overwrite existing
files; if you want this to happen, specify the -f flag.
files. If you want this to happen, specify the -f flag.
If no file names are specified, bzip2 compresses from
standard input to standard output. In this case, bzip2
@ -47,10 +45,25 @@ DESCRIPTION
this would be entirely incomprehensible and therefore
pointless.
bunzip2 (or bzip2 -d ) decompresses and restores all spec-
ified files whose names end in ".bz2". Files without this
suffix are ignored. Again, supplying no filenames causes
decompression from standard input to standard output.
bunzip2 (or bzip2 -d) decompresses all specified files.
Files which were not created by bzip2 will be detected and
ignored, and a warning issued. bzip2 attempts to guess
the filename for the decompressed file from that of the
compressed file as follows:
filename.bz2 becomes filename
filename.bz becomes filename
filename.tbz2 becomes filename.tar
filename.tbz becomes filename.tar
anyothername becomes anyothername.out
If the file does not end in one of the recognised endings,
.bz2, .bz, .tbz2 or .tbz, bzip2 complains that it cannot
guess the name of the original file, and uses the original
name with .out appended.
As with compression, supplying no filenames causes decom-
pression from standard input to standard output.
bunzip2 will correctly decompress a file which is the con-
catenation of two or more compressed files. The result is
@ -58,19 +71,24 @@ DESCRIPTION
Integrity testing (-t) of concatenated compressed files is
also supported.
You can also compress or decompress files to the standard
output by giving the -c flag. Multiple files may be com-
You can also compress or decompress files to the standard
output by giving the -c flag. Multiple files may be com-
pressed and decompressed like this. The resulting outputs
are fed sequentially to stdout. Compression of multiple
files in this manner generates a stream containing multi-
are fed sequentially to stdout. Compression of multiple
files in this manner generates a stream containing multi-
ple compressed file representations. Such a stream can be
decompressed correctly only by bzip2 version 0.9.0 or
later. Earlier versions of bzip2 will stop after decom-
decompressed correctly only by bzip2 version 0.9.0 or
later. Earlier versions of bzip2 will stop after decom-
pressing the first file in the stream.
bzcat (or bzip2 -dc ) decompresses all specified files to
bzcat (or bzip2 -dc) decompresses all specified files to
the standard output.
bzip2 will read arguments from the environment variables
BZIP2 and BZIP, in that order, and will process them
before any arguments read from the command line. This
gives a convenient way to supply default arguments.
Compression is always performed, even if the compressed
file is slightly larger than the original. Files of less
than about one hundred bytes tend to get larger, since the
@ -87,98 +105,19 @@ DESCRIPTION
corruption going undetected is microscopic, about one
chance in four billion for each file processed. Be aware,
though, that the check occurs upon decompression, so it
can only tell you that that something is wrong. It can't
help you recover the original uncompressed data. You can
use bzip2recover to try to recover data from damaged
files.
can only tell you that something is wrong. It can't help
you recover the original uncompressed data. You can use
bzip2recover to try to recover data from damaged files.
Return values: 0 for a normal exit, 1 for environmental
problems (file not found, invalid flags, I/O errors, &c),
Return values: 0 for a normal exit, 1 for environmental
problems (file not found, invalid flags, I/O errors, &c),
2 to indicate a corrupt compressed file, 3 for an internal
consistency error (eg, bug) which caused bzip2 to panic.
MEMORY MANAGEMENT
Bzip2 compresses large files in blocks. The block size
affects both the compression ratio achieved, and the
amount of memory needed both for compression and decom-
pression. The flags -1 through -9 specify the block size
to be 100,000 bytes through 900,000 bytes (the default)
respectively. At decompression-time, the block size used
for compression is read from the header of the compressed
file, and bunzip2 then allocates itself just enough memory
to decompress the file. Since block sizes are stored in
compressed files, it follows that the flags -1 to -9 are
irrelevant to and so ignored during decompression.
Compression and decompression requirements, in bytes, can
be estimated as:
Compression: 400k + ( 7 x block size )
Decompression: 100k + ( 4 x block size ), or
100k + ( 2.5 x block size )
Larger block sizes give rapidly diminishing marginal
returns; most of the compression comes from the first two
or three hundred k of block size, a fact worth bearing in
mind when using bzip2 on small machines. It is also
important to appreciate that the decompression memory
requirement is set at compression-time by the choice of
block size.
For files compressed with the default 900k block size,
bunzip2 will require about 3700 kbytes to decompress. To
support decompression of any file on a 4 megabyte machine,
bunzip2 has an option to decompress using approximately
half this amount of memory, about 2300 kbytes. Decompres-
sion speed is also halved, so you should use this option
only where necessary. The relevant flag is -s.
In general, try and use the largest block size memory con-
straints allow, since that maximises the compression
achieved. Compression and decompression speed are virtu-
ally unaffected by block size.
Another significant point applies to files which fit in a
single block -- that means most files you'd encounter
using a large block size. The amount of real memory
touched is proportional to the size of the file, since the
file is smaller than a block. For example, compressing a
file 20,000 bytes long with the flag -9 will cause the
compressor to allocate around 6700k of memory, but only
touch 400k + 20000 * 7 = 540 kbytes of it. Similarly, the
decompressor will allocate 3700k but only touch 100k +
20000 * 4 = 180 kbytes.
Here is a table which summarises the maximum memory usage
for different block sizes. Also recorded is the total
compressed size for 14 files of the Calgary Text Compres-
sion Corpus totalling 3,141,622 bytes. This column gives
some feel for how compression varies with block size.
These figures tend to understate the advantage of larger
block sizes for larger files, since the Corpus is domi-
nated by smaller files.
Compress Decompress Decompress Corpus
Flag usage usage -s usage Size
-1 1100k 500k 350k 914704
-2 1800k 900k 600k 877703
-3 2500k 1300k 850k 860338
-4 3200k 1700k 1100k 846899
-5 3900k 2100k 1350k 845160
-6 4600k 2500k 1600k 838626
-7 5400k 2900k 1850k 834096
-8 6000k 3300k 2100k 828642
-9 6700k 3700k 2350k 828642
OPTIONS
-c --stdout
Compress or decompress to standard output. -c will
decompress multiple files to stdout, but will only
compress a single file to stdout.
Compress or decompress to standard output.
-d --decompress
Force decompression. bzip2, bunzip2 and bzcat are
@ -198,7 +137,9 @@ OPTIONS
-f --force
Force overwrite of output files. Normally, bzip2
will not overwrite existing output files.
will not overwrite existing output files. Also
forces bzip2 to break hard links to files, which it
otherwise wouldn't do.
-k --keep
Keep (don't delete) input files during compression
@ -217,7 +158,12 @@ OPTIONS
figure, at the expense of your compression ratio.
In short, if your machine is low on memory (8
megabytes or less), use -s for everything. See
MEMORY MANAGEMENT above.
MEMORY MANAGEMENT below.
-q --quiet
Suppress non-essential warning messages. Messages
pertaining to I/O errors and other critical events
will not be suppressed.
-v --verbose
Verbose mode -- show the compression ratio for each
@ -232,21 +178,96 @@ OPTIONS
-1 to -9
Set the block size to 100 k, 200 k .. 900 k when
compressing. Has no effect when decompressing.
See MEMORY MANAGEMENT above.
See MEMORY MANAGEMENT below.
--repetitive-fast
bzip2 injects some small pseudo-random variations
into very repetitive blocks to limit worst-case
performance during compression. If sorting runs
into difficulties, the block is randomised, and
sorting is restarted. Very roughly, bzip2 persists
for three times as long as a well-behaved input
would take before resorting to randomisation. This
flag makes it give up much sooner.
-- Treats all subsequent arguments as file names, even
if they start with a dash. This is so you can han-
dle files with names beginning with a dash, for
example: bzip2 -- -myfilename.
--repetitive-best
Opposite of --repetitive-fast; try a lot harder
before resorting to randomisation.
--repetitive-fast --repetitive-best
These flags are redundant in versions 0.9.5 and
above. They provided some coarse control over the
behaviour of the sorting algorithm in earlier ver-
sions, which was sometimes useful. 0.9.5 and above
have an improved algorithm which renders these
flags irrelevant.
MEMORY MANAGEMENT
bzip2 compresses large files in blocks. The block size
affects both the compression ratio achieved, and the
amount of memory needed for compression and decompression.
The flags -1 through -9 specify the block size to be
100,000 bytes through 900,000 bytes (the default) respec-
tively. At decompression time, the block size used for
compression is read from the header of the compressed
file, and bunzip2 then allocates itself just enough memory
to decompress the file. Since block sizes are stored in
compressed files, it follows that the flags -1 to -9 are
irrelevant to and so ignored during decompression.
Compression and decompression requirements, in bytes, can
be estimated as:
Compression: 400k + ( 8 x block size )
Decompression: 100k + ( 4 x block size ), or
100k + ( 2.5 x block size )
Larger block sizes give rapidly diminishing marginal
returns. Most of the compression comes from the first two
or three hundred k of block size, a fact worth bearing in
mind when using bzip2 on small machines. It is also
important to appreciate that the decompression memory
requirement is set at compression time by the choice of
block size.
For files compressed with the default 900k block size,
bunzip2 will require about 3700 kbytes to decompress. To
support decompression of any file on a 4 megabyte machine,
bunzip2 has an option to decompress using approximately
half this amount of memory, about 2300 kbytes. Decompres-
sion speed is also halved, so you should use this option
only where necessary. The relevant flag is -s.
In general, try and use the largest block size memory con-
straints allow, since that maximises the compression
achieved. Compression and decompression speed are virtu-
ally unaffected by block size.
Another significant point applies to files which fit in a
single block -- that means most files you'd encounter
using a large block size. The amount of real memory
touched is proportional to the size of the file, since the
file is smaller than a block. For example, compressing a
file 20,000 bytes long with the flag -9 will cause the
compressor to allocate around 7600k of memory, but only
touch 400k + 20000 * 8 = 560 kbytes of it. Similarly, the
decompressor will allocate 3700k but only touch 100k +
20000 * 4 = 180 kbytes.
Here is a table which summarises the maximum memory usage
for different block sizes. Also recorded is the total
compressed size for 14 files of the Calgary Text Compres-
sion Corpus totalling 3,141,622 bytes. This column gives
some feel for how compression varies with block size.
These figures tend to understate the advantage of larger
block sizes for larger files, since the Corpus is domi-
nated by smaller files.
Compress Decompress Decompress Corpus
Flag usage usage -s usage Size
-1 1200k 500k 350k 914704
-2 2000k 900k 600k 877703
-3 2800k 1300k 850k 860338
-4 3600k 1700k 1100k 846899
-5 4400k 2100k 1350k 845160
-6 5200k 2500k 1600k 838626
-7 6100k 2900k 1850k 834096
-8 6800k 3300k 2100k 828642
-9 7600k 3700k 2350k 828642
RECOVERING DATA FROM DAMAGED FILES
@ -273,8 +294,8 @@ RECOVERING DATA FROM DAMAGED FILES
"rec0002file.bz2", etc, containing the extracted blocks.
The output filenames are designed so that the use of
wildcards in subsequent processing -- for example, "bzip2
-dc rec*file.bz2 > recovered_data" -- lists the files in
the "right" order.
-dc rec*file.bz2 > recovered_data" -- lists the files in
the correct order.
bzip2recover should be of most use dealing with large .bz2
files, as these will contain many blocks. It is clearly
@ -289,17 +310,15 @@ PERFORMANCE NOTES
The sorting phase of compression gathers together similar
strings in the file. Because of this, files containing
very long runs of repeated symbols, like "aabaabaabaab
..." (repeated several hundred times) may compress
extraordinarily slowly. You can use the -vvvvv option to
monitor progress in great detail, if you want. Decompres-
sion speed is unaffected.
..." (repeated several hundred times) may compress more
slowly than normal. Versions 0.9.5 and above fare much
better than previous versions in this respect. The ratio
between worst-case and average-case compression time is in
the region of 10:1. For previous versions, this figure
was more like 100:1. You can use the -vvvv option to mon-
itor progress in great detail, if you want.
Such pathological cases seem rare in practice, appearing
mostly in artificially-constructed test files, and in low-
level disk images. It may be inadvisable to use bzip2 to
compress the latter. If you do get a file which causes
severe slowness in compression, try making the block size
as small as possible, with flag -1.
Decompression speed is unaffected by these phenomena.
bzip2 usually allocates several megabytes of memory to
operate in, and then charges all over it in a fairly ran-
@ -314,42 +333,43 @@ PERFORMANCE NOTES
CAVEATS
I/O error messages are not as helpful as they could be.
Bzip2 tries hard to detect I/O errors and exit cleanly,
bzip2 tries hard to detect I/O errors and exit cleanly,
but the details of what the problem is sometimes seem
rather misleading.
This manual page pertains to version 0.9.0 of bzip2. Com-
This manual page pertains to version 0.9.5 of bzip2. Com-
pressed data created by this version is entirely forwards
and backwards compatible with the previous public release,
version 0.1pl2, but with the following exception: 0.9.0
can correctly decompress multiple concatenated compressed
files. 0.1pl2 cannot do this; it will stop after decom-
pressing just the first file in the stream.
and backwards compatible with the previous public
releases, versions 0.1pl2 and 0.9.0, but with the follow-
ing exception: 0.9.0 and above can correctly decompress
multiple concatenated compressed files. 0.1pl2 cannot do
this; it will stop after decompressing just the first file
in the stream.
Wildcard expansion for Windows 95 and NT is flaky.
bzip2recover uses 32-bit integers to represent bit posi-
tions in compressed files, so it cannot handle compressed
files more than 512 megabytes long. This could easily be
bzip2recover uses 32-bit integers to represent bit posi-
tions in compressed files, so it cannot handle compressed
files more than 512 megabytes long. This could easily be
fixed.
AUTHOR
Julian Seward, jseward@acm.org.
http://www.muraroa.demon.co.uk
The ideas embodied in bzip2 are due to (at least) the fol-
lowing people: Michael Burrows and David Wheeler (for the
block sorting transformation), David Wheeler (again, for
lowing people: Michael Burrows and David Wheeler (for the
block sorting transformation), David Wheeler (again, for
the Huffman coder), Peter Fenwick (for the structured cod-
ing model in the original bzip, and many refinements), and
Alistair Moffat, Radford Neal and Ian Witten (for the
Alistair Moffat, Radford Neal and Ian Witten (for the
arithmetic coder in the original bzip). I am much
indebted for their help, support and advice. See the man-
ual in the source distribution for pointers to sources of
ual in the source distribution for pointers to sources of
documentation. Christian von Roques encouraged me to look
for faster sorting algorithms, so as to speed up compres-
for faster sorting algorithms, so as to speed up compres-
sion. Bela Lubkin encouraged me to improve the worst-case
compression performance. Many people sent patches, helped
with portability problems, lent machines, gave advice and
with portability problems, lent machines, gave advice and
were generally helpful.

21
bzip2recover.c
View File

@ -7,9 +7,9 @@
/*--
This program is bzip2recover, a program to attempt data
salvage from damaged files created by the accompanying
bzip2-0.9.0c program.
bzip2-0.9.5 program.
Copyright (C) 1996-1998 Julian R Seward. All rights reserved.
Copyright (C) 1996-1999 Julian R Seward. All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
@ -42,9 +42,9 @@
NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
Julian Seward, Guildford, Surrey, UK.
Julian Seward, Cambridge, UK.
jseward@acm.org
bzip2/libbzip2 version 0.9.0c of 18 October 1998
bzip2/libbzip2 version 0.9.5 of 24 May 1999
--*/
/*--
@ -260,6 +260,12 @@ Bool endsInBz2 ( Char* name )
#define BLOCK_ENDMARK_HI 0x00001772UL
#define BLOCK_ENDMARK_LO 0x45385090UL
UInt32 bStart[20000];
UInt32 bEnd[20000];
UInt32 rbStart[20000];
UInt32 rbEnd[20000];
Int32 main ( Int32 argc, Char** argv )
{
FILE* inFile;
@ -267,11 +273,6 @@ Int32 main ( Int32 argc, Char** argv )
BitStream* bsIn, *bsWr;
Int32 currBlock, b, wrBlock;
UInt32 bitsRead;
UInt32 bStart[20000];
UInt32 bEnd[20000];
UInt32 rbStart[20000];
UInt32 rbEnd[20000];
Int32 rbCtr;
@ -281,7 +282,7 @@ Int32 main ( Int32 argc, Char** argv )
strcpy ( progName, argv[0] );
inFileName[0] = outFileName[0] = 0;
fprintf ( stderr, "bzip2recover v0.9.0c: extracts blocks from damaged .bz2 files.\n" );
fprintf ( stderr, "bzip2recover 0.9.5d: extracts blocks from damaged .bz2 files.\n" );
if (argc != 2) {
fprintf ( stderr, "%s: usage is `%s damaged_file_name'.\n",

134
bzlib.c
View File

@ -8,7 +8,7 @@
This file is a part of bzip2 and/or libbzip2, a program and
library for lossless, block-sorting data compression.
Copyright (C) 1996-1998 Julian R Seward. All rights reserved.
Copyright (C) 1996-1999 Julian R Seward. All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
@ -41,9 +41,9 @@
NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
Julian Seward, Guildford, Surrey, UK.
Julian Seward, Cambridge, UK.
jseward@acm.org
bzip2/libbzip2 version 0.9.0c of 18 October 1998
bzip2/libbzip2 version 0.9.5 of 24 May 1999
This program is based on (at least) the work of:
Mike Burrows
@ -86,14 +86,14 @@
void bz__AssertH__fail ( int errcode )
{
fprintf(stderr,
"\n\nbzip2/libbzip2, v0.9.0c: internal error number %d.\n"
"This is a bug in bzip2/libbzip2, v0.9.0c. Please report\n"
"\n\nbzip2/libbzip2, v0.9.5d: internal error number %d.\n"
"This is a bug in bzip2/libbzip2, v0.9.5d. Please report\n"
"it to me at: jseward@acm.org. If this happened when\n"
"you were using some program which uses libbzip2 as a\n"
"component, you should also report this bug to the author(s)\n"
"of that program. Please make an effort to report this bug;\n"
"timely and accurate bug reports eventually lead to higher\n"
"quality software. Thx. Julian Seward, 18 October 1998.\n\n",
"quality software. Thanks. Julian Seward, 4 Sept 1999.\n\n",
errcode
);
exit(3);
@ -171,29 +171,23 @@ int BZ_API(bzCompressInit)
if (s == NULL) return BZ_MEM_ERROR;
s->strm = strm;
s->block = NULL;
s->quadrant = NULL;
s->zptr = NULL;
s->ftab = NULL;
s->arr1 = NULL;
s->arr2 = NULL;
s->ftab = NULL;
n = 100000 * blockSize100k;
s->block = BZALLOC( (n + BZ_NUM_OVERSHOOT_BYTES) * sizeof(UChar) );
s->quadrant = BZALLOC( (n + BZ_NUM_OVERSHOOT_BYTES) * sizeof(Int16) );
s->zptr = BZALLOC( n * sizeof(Int32) );
s->ftab = BZALLOC( 65537 * sizeof(Int32) );
n = 100000 * blockSize100k;
s->arr1 = BZALLOC( n * sizeof(UInt32) );
s->arr2 = BZALLOC( (n+BZ_N_OVERSHOOT) * sizeof(UInt32) );
s->ftab = BZALLOC( 65537 * sizeof(UInt32) );
if (s->block == NULL || s->quadrant == NULL ||
s->zptr == NULL || s->ftab == NULL) {
if (s->block != NULL) BZFREE(s->block);
if (s->quadrant != NULL) BZFREE(s->quadrant);
if (s->zptr != NULL) BZFREE(s->zptr);
if (s->ftab != NULL) BZFREE(s->ftab);
if (s != NULL) BZFREE(s);
if (s->arr1 == NULL || s->arr2 == NULL || s->ftab == NULL) {
if (s->arr1 != NULL) BZFREE(s->arr1);
if (s->arr2 != NULL) BZFREE(s->arr2);
if (s->ftab != NULL) BZFREE(s->ftab);
if (s != NULL) BZFREE(s);
return BZ_MEM_ERROR;
}
s->szptr = (UInt16*)(s->zptr);
s->blockNo = 0;
s->state = BZ_S_INPUT;
s->mode = BZ_M_RUNNING;
@ -202,7 +196,12 @@ int BZ_API(bzCompressInit)
s->nblockMAX = 100000 * blockSize100k - 19;
s->verbosity = verbosity;
s->workFactor = workFactor;
s->nBlocksRandomised = 0;
s->block = (UInt16*)s->arr2;
s->mtfv = (UInt16*)s->arr1;
s->zbits = NULL;
s->ptr = (UInt32*)s->arr1;
strm->state = s;
strm->total_in = 0;
strm->total_out = 0;
@ -224,24 +223,24 @@ void add_pair_to_block ( EState* s )
s->inUse[s->state_in_ch] = True;
switch (s->state_in_len) {
case 1:
s->block[s->nblock] = (UChar)ch; s->nblock++;
s->block[s->nblock] = (UInt16)ch; s->nblock++;
break;
case 2:
s->block[s->nblock] = (UChar)ch; s->nblock++;
s->block[s->nblock] = (UChar)ch; s->nblock++;
s->block[s->nblock] = (UInt16)ch; s->nblock++;
s->block[s->nblock] = (UInt16)ch; s->nblock++;
break;
case 3:
s->block[s->nblock] = (UChar)ch; s->nblock++;
s->block[s->nblock] = (UChar)ch; s->nblock++;
s->block[s->nblock] = (UChar)ch; s->nblock++;
s->block[s->nblock] = (UInt16)ch; s->nblock++;
s->block[s->nblock] = (UInt16)ch; s->nblock++;
s->block[s->nblock] = (UInt16)ch; s->nblock++;
break;
default:
s->inUse[s->state_in_len-4] = True;
s->block[s->nblock] = (UChar)ch; s->nblock++;
s->block[s->nblock] = (UChar)ch; s->nblock++;
s->block[s->nblock] = (UChar)ch; s->nblock++;
s->block[s->nblock] = (UChar)ch; s->nblock++;
s->block[s->nblock] = (UChar)(s->state_in_len-4);
s->block[s->nblock] = (UInt16)ch; s->nblock++;
s->block[s->nblock] = (UInt16)ch; s->nblock++;
s->block[s->nblock] = (UInt16)ch; s->nblock++;
s->block[s->nblock] = (UInt16)ch; s->nblock++;
s->block[s->nblock] = ((UInt16)(s->state_in_len-4));
s->nblock++;
break;
}
@ -267,7 +266,7 @@ void flush_RL ( EState* s )
UChar ch = (UChar)(zs->state_in_ch); \
BZ_UPDATE_CRC( zs->blockCRC, ch ); \
zs->inUse[zs->state_in_ch] = True; \
zs->block[zs->nblock] = (UChar)ch; \
zs->block[zs->nblock] = (UInt16)ch; \
zs->nblock++; \
zs->state_in_ch = zchh; \
} \
@ -343,7 +342,7 @@ Bool copy_output_until_stop ( EState* s )
if (s->state_out_pos >= s->numZ) break;
progress_out = True;
*(s->strm->next_out) = ((UChar*)(s->quadrant))[s->state_out_pos];
*(s->strm->next_out) = s->zbits[s->state_out_pos];
s->state_out_pos++;
s->strm->avail_out--;
s->strm->next_out++;
@ -382,7 +381,7 @@ Bool handle_compress ( bz_stream* strm )
progress_in |= copy_input_until_stop ( s );
if (s->mode != BZ_M_RUNNING && s->avail_in_expect == 0) {
flush_RL ( s );
compressBlock ( s, s->mode == BZ_M_FINISHING );
compressBlock ( s, (Bool)(s->mode == BZ_M_FINISHING) );
s->state = BZ_S_OUTPUT;
}
else
@ -470,10 +469,9 @@ int BZ_API(bzCompressEnd) ( bz_stream *strm )
if (s == NULL) return BZ_PARAM_ERROR;
if (s->strm != strm) return BZ_PARAM_ERROR;
if (s->block != NULL) BZFREE(s->block);
if (s->quadrant != NULL) BZFREE(s->quadrant);
if (s->zptr != NULL) BZFREE(s->zptr);
if (s->ftab != NULL) BZFREE(s->ftab);
if (s->arr1 != NULL) BZFREE(s->arr1);
if (s->arr2 != NULL) BZFREE(s->arr2);
if (s->ftab != NULL) BZFREE(s->ftab);
BZFREE(strm->state);
strm->state = NULL;
@ -816,7 +814,8 @@ int BZ_API(bzDecompress) ( bz_stream *strm )
}
AssertH ( 0, 6001 );
/*notreached*/
return 0; /*NOTREACHED*/
}
@ -1284,7 +1283,7 @@ int BZ_API(bzBuffToBuffDecompress)
errhandler:
bzDecompressEnd ( &strm );
return BZ_SEQUENCE_ERROR;
return ret;
}
@ -1339,24 +1338,18 @@ BZFILE * bzopen_or_bzdopen
int smallMode = 0;
int nUnused = 0;
if(mode==NULL){return NULL;}
while(*mode){
switch(*mode){
if (mode == NULL) return NULL;
while (*mode) {
switch (*mode) {
case 'r':
writing = 0;break;
writing = 0; break;
case 'w':
writing = 1;break;
writing = 1; break;
case 's':
smallMode = 1;break;
smallMode = 1; break;
default:
if(isdigit(*mode)){
blockSize100k = 0;
while(isdigit(*mode)){
blockSize100k = blockSize100k*10 + *mode-'0';
mode++;
}
}else{
/* ignore */
if (isdigit((int)(*mode))) {
blockSize100k = *mode-'0';
}
}
mode++;
@ -1364,29 +1357,32 @@ BZFILE * bzopen_or_bzdopen
strcat(mode2, writing ? "w" : "r" );
strcat(mode2,"b"); /* binary mode */
if(open_mode==0){
if(path==NULL || strcmp(path,"")==0){
if (open_mode==0) {
if (path==NULL || strcmp(path,"")==0) {
fp = (writing ? stdout : stdin);
SET_BINARY_MODE(fp);
}else{
} else {
fp = fopen(path,mode2);
}
}else{
} else {
#ifdef BZ_STRICT_ANSI
fp = NULL;
#else
fp = fdopen(fd,mode2);
#endif
}
if(fp==NULL){return NULL;}
if (fp == NULL) return NULL;
if(writing){
if (writing) {
/* Guard against total chaos and anarchy -- JRS */
if (blockSize100k < 1) blockSize100k = 1;
if (blockSize100k > 9) blockSize100k = 9;
bzfp = bzWriteOpen(&bzerr,fp,blockSize100k,verbosity,workFactor);
}else{
} else {
bzfp = bzReadOpen(&bzerr,fp,verbosity,smallMode,unused,nUnused);
}
if(bzfp==NULL){
if(fp!=stdin && fp!=stdout) fclose(fp);
if (bzfp == NULL) {
if (fp != stdin && fp != stdout) fclose(fp);
return NULL;
}
return bzfp;
@ -1458,7 +1454,7 @@ void BZ_API(bzclose) (BZFILE* b)
int bzerr;
FILE *fp = ((bzFile *)b)->handle;
if(b==NULL){return;}
if (b==NULL) {return;}
if(((bzFile*)b)->writing){
bzWriteClose(&bzerr,b,0,NULL,NULL);
if(bzerr != BZ_OK){

13
bzlib.h
View File

@ -8,7 +8,7 @@
This file is a part of bzip2 and/or libbzip2, a program and
library for lossless, block-sorting data compression.
Copyright (C) 1996-1998 Julian R Seward. All rights reserved.
Copyright (C) 1996-1999 Julian R Seward. All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
@ -41,9 +41,9 @@
NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
Julian Seward, Guildford, Surrey, UK.
Julian Seward, Cambridge, UK.
jseward@acm.org
bzip2/libbzip2 version 0.9.0c of 18 October 1998
bzip2/libbzip2 version 0.9.5 of 24 May 1999
This program is based on (at least) the work of:
Mike Burrows
@ -62,6 +62,10 @@
#ifndef _BZLIB_H
#define _BZLIB_H
#ifdef __cplusplus
extern "C" {
#endif
#define BZ_RUN 0
#define BZ_FLUSH 1
#define BZ_FINISH 2
@ -291,6 +295,9 @@ BZ_EXTERN const char * BZ_API(bzerror) (
);
#endif
#ifdef __cplusplus
}
#endif
#endif

47
bzlib_private.h
View File

@ -8,7 +8,7 @@
This file is a part of bzip2 and/or libbzip2, a program and
library for lossless, block-sorting data compression.
Copyright (C) 1996-1998 Julian R Seward. All rights reserved.
Copyright (C) 1996-1999 Julian R Seward. All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
@ -41,9 +41,9 @@
NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
Julian Seward, Guildford, Surrey, UK.
Julian Seward, Cambridge, UK.
jseward@acm.org
bzip2/libbzip2 version 0.9.0c of 18 October 1998
bzip2/libbzip2 version 0.9.5 of 24 May 1999
This program is based on (at least) the work of:
Mike Burrows
@ -76,7 +76,7 @@
/*-- General stuff. --*/
#define BZ_VERSION "0.9.0c"
#define BZ_VERSION "0.9.5d"
typedef char Char;
typedef unsigned char Bool;
@ -85,7 +85,7 @@ typedef int Int32;
typedef unsigned int UInt32;
typedef short Int16;
typedef unsigned short UInt16;
#define True ((Bool)1)
#define False ((Bool)0)
@ -210,7 +210,11 @@ extern UInt32 crc32Table[256];
#define BZ_S_OUTPUT 1
#define BZ_S_INPUT 2
#define BZ_NUM_OVERSHOOT_BYTES 20
#define BZ_N_RADIX 2
#define BZ_N_QSORT 12
#define BZ_N_SHELL 18
#define BZ_N_OVERSHOOT (BZ_N_RADIX + BZ_N_QSORT + BZ_N_SHELL + 2)
@ -230,18 +234,20 @@ typedef
UInt32 avail_in_expect;
/* for doing the block sorting */
UChar* block;
UInt16* quadrant;
UInt32* zptr;
UInt16* szptr;
Int32* ftab;
Int32 workDone;
Int32 workLimit;
Int32 workFactor;
Bool firstAttempt;
Bool blockRandomised;
UInt32* arr1;
UInt32* arr2;
UInt32* ftab;
Int32 origPtr;
/* aliases for arr1 and arr2 */
UInt32* ptr;
UInt16* block;
UInt16* mtfv;
UChar* zbits;
/* for deciding when to use the fallback sorting algorithm */
Int32 workFactor;
/* run-length-encoding of the input */
UInt32 state_in_ch;
Int32 state_in_len;
@ -269,7 +275,6 @@ typedef
/* misc administratium */
Int32 verbosity;
Int32 blockNo;
Int32 nBlocksRandomised;
Int32 blockSize100k;
/* stuff for coding the MTF values */
@ -478,17 +483,17 @@ typedef
}
#define GET_LL4(i) \
(((UInt32)(s->ll4[(i) >> 1])) >> (((i) << 2) & 0x4) & 0xF)
((((UInt32)(s->ll4[(i) >> 1])) >> (((i) << 2) & 0x4)) & 0xF)
#define SET_LL(i,n) \
#define SET_LL(i,n) \
{ s->ll16[i] = (UInt16)(n & 0x0000ffff); \
SET_LL4(i, n >> 16); \
SET_LL4(i, n >> 16); \
}
#define GET_LL(i) \
(((UInt32)s->ll16[i]) | (GET_LL4(i) << 16))
#define BZ_GET_SMALL(cccc) \
#define BZ_GET_SMALL(cccc) \
cccc = indexIntoF ( s->tPos, s->cftab ); \
s->tPos = GET_LL(s->tPos);

141
compress.c
View File

@ -8,7 +8,7 @@
This file is a part of bzip2 and/or libbzip2, a program and
library for lossless, block-sorting data compression.
Copyright (C) 1996-1998 Julian R Seward. All rights reserved.
Copyright (C) 1996-1999 Julian R Seward. All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
@ -41,9 +41,9 @@
NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
Julian Seward, Guildford, Surrey, UK.
Julian Seward, Cambridge, UK.
jseward@acm.org
bzip2/libbzip2 version 0.9.0 of 28 June 1998
bzip2/libbzip2 version 0.9.5 of 24 May 1999
This program is based on (at least) the work of:
Mike Burrows
@ -90,7 +90,7 @@ static
void bsFinishWrite ( EState* s )
{
while (s->bsLive > 0) {
((UChar*)(s->quadrant))[s->numZ] = (UChar)(s->bsBuff >> 24);
s->zbits[s->numZ] = (UChar)(s->bsBuff >> 24);
s->numZ++;
s->bsBuff <<= 8;
s->bsLive -= 8;
@ -102,7 +102,7 @@ void bsFinishWrite ( EState* s )
#define bsNEEDW(nz) \
{ \
while (s->bsLive >= 8) { \
((UChar*)(s->quadrant))[s->numZ] \
s->zbits[s->numZ] \
= (UChar)(s->bsBuff >> 24); \
s->numZ++; \
s->bsBuff <<= 8; \
@ -162,13 +162,39 @@ void makeMaps_e ( EState* s )
static
void generateMTFValues ( EState* s )
{
UChar yy[256];
Int32 i, j;
UChar tmp;
UChar tmp2;
Int32 zPend;
Int32 wr;
Int32 EOB;
UChar yy[256];
Int32 i, j;
UChar tmp;
UChar tmp2;
Int32 zPend;
Int32 wr;
Int32 EOB;
/*
After sorting (eg, here),
s->arr1 [ 0 .. s->nblock-1 ] holds sorted order,
and
((UInt16*)s->arr2) [ 0 .. s->nblock-1 ] [15:8]
holds the original block data.
The first thing to do is generate the MTF values,
and put them in
((UInt16*)s->arr1) [ 0 .. s->nblock-1 ].
Because there are strictly fewer or equal MTF values
than block values, ptr values in this area are overwritten
with MTF values only when they are no longer needed.
The final compressed bitstream is generated into the
area starting at
(UChar*) (&((UInt16)s->arr2)[s->nblock])
These storage aliases are set up in bzCompressInit(),
except for the last one, which is arranged in
compressBlock().
*/
UInt32* ptr = s->ptr;
UInt16* block = s->block;
UInt16* mtfv = s->mtfv;
makeMaps_e ( s );
EOB = s->nInUse+1;
@ -183,52 +209,61 @@ void generateMTFValues ( EState* s )
UChar ll_i;
AssertD ( wr <= i, "generateMTFValues(1)" );
j = s->zptr[i]-1; if (j < 0) j += s->nblock;
ll_i = s->unseqToSeq[s->block[j]];
j = ptr[i]-1; if (j < 0) j += s->nblock;
ll_i = s->unseqToSeq[block[j] >> 8];
AssertD ( ll_i < s->nInUse, "generateMTFValues(2a)" );
j = 0;
tmp = yy[j];
while ( ll_i != tmp ) {
j++;
tmp2 = tmp;
tmp = yy[j];
yy[j] = tmp2;
};
yy[0] = tmp;
if (j == 0) {
tmp = yy[0];
if (tmp == ll_i) {
zPend++;
} else {
tmp2 = tmp;
tmp = yy[1];
yy[1] = tmp2;
j = 1;
while ( ll_i != tmp ) {
j++;
tmp2 = tmp;
tmp = yy[j];
yy[j] = tmp2;
};
yy[0] = tmp;
if (zPend > 0) {
zPend--;
while (True) {
switch (zPend % 2) {
case 0: s->szptr[wr] = BZ_RUNA; wr++; s->mtfFreq[BZ_RUNA]++; break;
case 1: s->szptr[wr] = BZ_RUNB; wr++; s->mtfFreq[BZ_RUNB]++; break;
};
if (zPend & 1) {
mtfv[wr] = BZ_RUNB; wr++;
s->mtfFreq[BZ_RUNB]++;
} else {
mtfv[wr] = BZ_RUNA; wr++;
s->mtfFreq[BZ_RUNA]++;
}
if (zPend < 2) break;
zPend = (zPend - 2) / 2;
};
zPend = 0;
}
s->szptr[wr] = j+1; wr++; s->mtfFreq[j+1]++;
mtfv[wr] = j+1; wr++; s->mtfFreq[j+1]++;
}
}
if (zPend > 0) {
zPend--;
while (True) {
switch (zPend % 2) {
case 0: s->szptr[wr] = BZ_RUNA; wr++; s->mtfFreq[BZ_RUNA]++; break;
case 1: s->szptr[wr] = BZ_RUNB; wr++; s->mtfFreq[BZ_RUNB]++; break;
};
if (zPend & 1) {
mtfv[wr] = BZ_RUNB; wr++;
s->mtfFreq[BZ_RUNB]++;
} else {
mtfv[wr] = BZ_RUNA; wr++;
s->mtfFreq[BZ_RUNA]++;
}
if (zPend < 2) break;
zPend = (zPend - 2) / 2;
};
}
s->szptr[wr] = EOB; wr++; s->mtfFreq[EOB]++;
mtfv[wr] = EOB; wr++; s->mtfFreq[EOB]++;
s->nMTF = wr;
}
@ -259,6 +294,8 @@ void sendMTFValues ( EState* s )
UInt16 cost[BZ_N_GROUPS];
Int32 fave[BZ_N_GROUPS];
UInt16* mtfv = s->mtfv;
if (s->verbosity >= 3)
VPrintf3( " %d in block, %d after MTF & 1-2 coding, "
"%d+2 syms in use\n",
@ -348,7 +385,7 @@ void sendMTFValues ( EState* s )
register UInt16 cost0, cost1, cost2, cost3, cost4, cost5;
cost0 = cost1 = cost2 = cost3 = cost4 = cost5 = 0;
for (i = gs; i <= ge; i++) {
UInt16 icv = s->szptr[i];
UInt16 icv = mtfv[i];
cost0 += s->len[0][icv];
cost1 += s->len[1][icv];
cost2 += s->len[2][icv];
@ -360,7 +397,7 @@ void sendMTFValues ( EState* s )
cost[3] = cost3; cost[4] = cost4; cost[5] = cost5;
} else {
for (i = gs; i <= ge; i++) {
UInt16 icv = s->szptr[i];
UInt16 icv = mtfv[i];
for (t = 0; t < nGroups; t++) cost[t] += s->len[t][icv];
}
}
@ -381,7 +418,7 @@ void sendMTFValues ( EState* s )
Increment the symbol frequencies for the selected table.
--*/
for (i = gs; i <= ge; i++)
s->rfreq[bt][ s->szptr[i] ]++;
s->rfreq[bt][ mtfv[i] ]++;
gs = ge+1;
}
@ -502,8 +539,8 @@ void sendMTFValues ( EState* s )
for (i = gs; i <= ge; i++) {
AssertH ( s->selector[selCtr] < nGroups, 3006 );
bsW ( s,
s->len [s->selector[selCtr]] [s->szptr[i]],
s->code [s->selector[selCtr]] [s->szptr[i]] );
s->len [s->selector[selCtr]] [mtfv[i]],
s->code [s->selector[selCtr]] [mtfv[i]] );
}
gs = ge+1;
@ -534,13 +571,15 @@ void compressBlock ( EState* s, Bool is_last_block )
blockSort ( s );
}
s->zbits = (UChar*) (&((UInt16*)s->arr2)[s->nblock]);
/*-- If this is the first block, create the stream header. --*/
if (s->blockNo == 1) {
bsInitWrite ( s );
bsPutUChar ( s, 'B' );
bsPutUChar ( s, 'Z' );
bsPutUChar ( s, 'h' );
bsPutUChar ( s, '0' + s->blockSize100k );
bsPutUChar ( s, (UChar)('0' + s->blockSize100k) );
}
if (s->nblock > 0) {
@ -552,11 +591,16 @@ void compressBlock ( EState* s, Bool is_last_block )
/*-- Now the block's CRC, so it is in a known place. --*/
bsPutUInt32 ( s, s->blockCRC );
/*-- Now a single bit indicating randomisation. --*/
if (s->blockRandomised) {
bsW(s,1,1); s->nBlocksRandomised++;
} else
bsW(s,1,0);
/*--
Now a single bit indicating (non-)randomisation.
As of version 0.9.5, we use a better sorting algorithm
which makes randomisation unnecessary. So always set
the randomised bit to 'no'. Of course, the decoder
still needs to be able to handle randomised blocks
so as to maintain backwards compatibility with
older versions of bzip2.
--*/
bsW(s,1,0);
bsW ( s, 24, s->origPtr );
generateMTFValues ( s );
@ -567,11 +611,6 @@ void compressBlock ( EState* s, Bool is_last_block )
/*-- If this is the last block, add the stream trailer. --*/
if (is_last_block) {
if (s->verbosity >= 2 && s->nBlocksRandomised > 0)
VPrintf2 ( " %d block%s needed randomisation\n",
s->nBlocksRandomised,
s->nBlocksRandomised == 1 ? "" : "s" );
bsPutUChar ( s, 0x17 ); bsPutUChar ( s, 0x72 );
bsPutUChar ( s, 0x45 ); bsPutUChar ( s, 0x38 );
bsPutUChar ( s, 0x50 ); bsPutUChar ( s, 0x90 );

6
crctable.c
View File

@ -8,7 +8,7 @@
This file is a part of bzip2 and/or libbzip2, a program and
library for lossless, block-sorting data compression.
Copyright (C) 1996-1998 Julian R Seward. All rights reserved.
Copyright (C) 1996-1999 Julian R Seward. All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
@ -41,9 +41,9 @@
NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
Julian Seward, Guildford, Surrey, UK.
Julian Seward, Cambridge, UK.
jseward@acm.org
bzip2/libbzip2 version 0.9.0c of 18 October 1998
bzip2/libbzip2 version 0.9.5 of 24 May 1999
This program is based on (at least) the work of:
Mike Burrows

8
decompress.c
View File

@ -8,7 +8,7 @@
This file is a part of bzip2 and/or libbzip2, a program and
library for lossless, block-sorting data compression.
Copyright (C) 1996-1998 Julian R Seward. All rights reserved.
Copyright (C) 1996-1999 Julian R Seward. All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
@ -41,9 +41,9 @@
NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
Julian Seward, Guildford, Surrey, UK.
Julian Seward, Cambridge, UK.
jseward@acm.org
bzip2/libbzip2 version 0.9.0c of 18 October 1998
bzip2/libbzip2 version 0.9.5 of 24 May 1999
This program is based on (at least) the work of:
Mike Burrows
@ -141,7 +141,7 @@ Int32 decompress ( DState* s )
bz_stream* strm = s->strm;
/* stuff that needs to be saved/restored */
Int32 i ;
Int32 i;
Int32 j;
Int32 t;
Int32 alphaSize;

242
dlltest.c
View File

@ -1,17 +1,17 @@
/*
minibz2
libbz2.dll test program.
by Yoshioka Tsuneo(QWF00133@nifty.ne.jp/tsuneo-y@is.aist-nara.ac.jp)
This file is Public Domain.
welcome any email to me.
minibz2
libbz2.dll test program.
by Yoshioka Tsuneo(QWF00133@nifty.ne.jp/tsuneo-y@is.aist-nara.ac.jp)
This file is Public Domain.
welcome any email to me.
usage: minibz2 [-d] [-{1,2,..9}] [[srcfilename] destfilename]
usage: minibz2 [-d] [-{1,2,..9}] [[srcfilename] destfilename]
*/
#define BZ_IMPORT
#include "bzlib.h"
#include <stdio.h>
#include <stdlib.h>
#include "bzlib.h"
#ifdef _WIN32
#include <io.h>
#endif
@ -24,140 +24,142 @@ static int BZ2DLLLoaded = 0;
static HINSTANCE BZ2DLLhLib;
int BZ2DLLLoadLibrary(void)
{
HINSTANCE hLib;
HINSTANCE hLib;
if(BZ2DLLLoaded==1){return 0;}
hLib=LoadLibrary("libbz2.dll");
if(hLib == NULL){
puts("Can't load libbz2.dll");
return -1;
}
BZ2DLLLoaded=1;
BZ2DLLhLib=hLib;
bzlibVersion=GetProcAddress(hLib,"bzlibVersion");
bzopen=GetProcAddress(hLib,"bzopen");
bzdopen=GetProcAddress(hLib,"bzdopen");
bzread=GetProcAddress(hLib,"bzread");
bzwrite=GetProcAddress(hLib,"bzwrite");
bzflush=GetProcAddress(hLib,"bzflush");
bzclose=GetProcAddress(hLib,"bzclose");
bzerror=GetProcAddress(hLib,"bzerror");
return 0;
if(BZ2DLLLoaded==1){return 0;}
hLib=LoadLibrary("libbz2.dll");
if(hLib == NULL){
puts("Can't load libbz2.dll");
return -1;
}
BZ2DLLLoaded=1;
BZ2DLLhLib=hLib;
bzlibVersion=GetProcAddress(hLib,"bzlibVersion");
bzopen=GetProcAddress(hLib,"bzopen");
bzdopen=GetProcAddress(hLib,"bzdopen");
bzread=GetProcAddress(hLib,"bzread");
bzwrite=GetProcAddress(hLib,"bzwrite");
bzflush=GetProcAddress(hLib,"bzflush");
bzclose=GetProcAddress(hLib,"bzclose");
bzerror=GetProcAddress(hLib,"bzerror");
return 0;
}
int BZ2DLLFreeLibrary(void)
{
if(BZ2DLLLoaded==0){return 0;}
FreeLibrary(BZ2DLLhLib);
BZ2DLLLoaded=0;
if(BZ2DLLLoaded==0){return 0;}
FreeLibrary(BZ2DLLhLib);
BZ2DLLLoaded=0;
}
#endif /* WIN32 */
void usage(void)
{
puts("usage: minibz2 [-d] [-{1,2,..9}] [[srcfilename] destfilename]");
puts("usage: minibz2 [-d] [-{1,2,..9}] [[srcfilename] destfilename]");
}
void main(int argc,char *argv[])
int main(int argc,char *argv[])
{
int decompress = 0;
int level = 9;
char *fn_r,*fn_w;
int decompress = 0;
int level = 9;
char *fn_r = NULL;
char *fn_w = NULL;
#ifdef _WIN32
if(BZ2DLLLoadLibrary()<0){
puts("can't load dll");
exit(1);
}
if(BZ2DLLLoadLibrary()<0){
puts("can't load dll");
exit(1);
}
#endif
while(++argv,--argc){
if(**argv =='-' || **argv=='/'){
char *p;
while(++argv,--argc){
if(**argv =='-' || **argv=='/'){
char *p;
for(p=*argv+1;*p;p++){
if(*p=='d'){
decompress = 1;
}else if('1'<=*p && *p<='9'){
level = *p - '0';
}else{
usage();
exit(1);
}
}
}else{
break;
}
}
if(argc>=1){
fn_r = *argv;
argc--;argv++;
}else{
fn_r = NULL;
}
if(argc>=1){
fn_w = *argv;
argc--;argv++;
}else{
fn_w = NULL;
}
{
int len;
char buff[0x1000];
char mode[10];
for(p=*argv+1;*p;p++){
if(*p=='d'){
decompress = 1;
}else if('1'<=*p && *p<='9'){
level = *p - '0';
}else{
usage();
exit(1);
}
}
}else{
break;
}
}
if(argc>=1){
fn_r = *argv;
argc--;argv++;
}else{
fn_r = NULL;
}
if(argc>=1){
fn_w = *argv;
argc--;argv++;
}else{
fn_w = NULL;
}
{
int len;
char buff[0x1000];
char mode[10];
if(decompress){
BZFILE *BZ2fp_r;
FILE *fp_w;
if(decompress){
BZFILE *BZ2fp_r = NULL;
FILE *fp_w = NULL;
if(fn_w){
if((fp_w = fopen(fn_w,"wb"))==NULL){
printf("can't open [%s]\n",fn_w);
perror("reason:");
exit(1);
}
}else{
fp_w = stdout;
}
if((BZ2fp_r == NULL && (BZ2fp_r = bzdopen(fileno(stdin),"rb"))==NULL)
|| (BZ2fp_r != NULL && (BZ2fp_r = bzopen(fn_r,"rb"))==NULL)){
printf("can't bz2openstream\n");
exit(1);
}
while((len=bzread(BZ2fp_r,buff,0x1000))>0){
fwrite(buff,1,len,fp_w);
}
bzclose(BZ2fp_r);
if(fp_w != stdout) fclose(fp_w);
}else{
BZFILE *BZ2fp_w;
FILE *fp_r;
if(fn_w){
if((fp_w = fopen(fn_w,"wb"))==NULL){
printf("can't open [%s]\n",fn_w);
perror("reason:");
exit(1);
}
}else{
fp_w = stdout;
}
if((BZ2fp_r == NULL && (BZ2fp_r = bzdopen(fileno(stdin),"rb"))==NULL)
|| (BZ2fp_r != NULL && (BZ2fp_r = bzopen(fn_r,"rb"))==NULL)){
printf("can't bz2openstream\n");
exit(1);
}
while((len=bzread(BZ2fp_r,buff,0x1000))>0){
fwrite(buff,1,len,fp_w);
}
bzclose(BZ2fp_r);
if(fp_w != stdout) fclose(fp_w);
}else{
BZFILE *BZ2fp_w = NULL;
FILE *fp_r = NULL;
if(fn_r){
if((fp_r = fopen(fn_r,"rb"))==NULL){
printf("can't open [%s]\n",fn_r);
perror("reason:");
exit(1);
}
}else{
fp_r = stdin;
}
mode[0]='w';
mode[1] = '0' + level;
mode[2] = '\0';
if(fn_r){
if((fp_r = fopen(fn_r,"rb"))==NULL){
printf("can't open [%s]\n",fn_r);
perror("reason:");
exit(1);
}
}else{
fp_r = stdin;
}
mode[0]='w';
mode[1] = '0' + level;
mode[2] = '\0';
if((fn_w == NULL && (BZ2fp_w = bzdopen(fileno(stdout),mode))==NULL)
|| (fn_w !=NULL && (BZ2fp_w = bzopen(fn_w,mode))==NULL)){
printf("can't bz2openstream\n");
exit(1);
}
while((len=fread(buff,1,0x1000,fp_r))>0){
bzwrite(BZ2fp_w,buff,len);
}
bzclose(BZ2fp_w);
if(fp_r!=stdin)fclose(fp_r);
}
}
if((fn_w == NULL && (BZ2fp_w = bzdopen(fileno(stdout),mode))==NULL)
|| (fn_w !=NULL && (BZ2fp_w = bzopen(fn_w,mode))==NULL)){
printf("can't bz2openstream\n");
exit(1);
}
while((len=fread(buff,1,0x1000,fp_r))>0){
bzwrite(BZ2fp_w,buff,len);
}
bzclose(BZ2fp_w);
if(fp_r!=stdin)fclose(fp_r);
}
}
#ifdef _WIN32
BZ2DLLFreeLibrary();
BZ2DLLFreeLibrary();
#endif
return 0;
}

37
howbig.c
View File

@ -1,37 +0,0 @@
#include <stdio.h>
#include <assert.h>
#include "bzlib.h"
unsigned char ibuff[1000000];
unsigned char obuff[1000000];
void doone ( int n )
{
int i, j, k, q, nobuff;
q = 0;
for (k = 0; k < 1; k++) {
for (i = 0; i < n; i++)
ibuff[i] = ((unsigned long)(random())) & 0xff;
nobuff = 1000000;
j = bzBuffToBuffCompress ( obuff, &nobuff, ibuff, n, 9,0,0 );
assert (j == BZ_OK);
if (nobuff > q) q = nobuff;
}
printf ( "%d %d(%d)\n", n, q, (int)((float)n * 1.01 - (float)q) );
}
int main ( int argc, char** argv )
{
int i;
i = 0;
while (1) {
if (i >= 900000) break;
doone(i);
if ( (int)(1.10 * i) > i )
i = (int)(1.10 * i); else i++;
}
return 0;
}

6
huffman.c
View File

@ -8,7 +8,7 @@
This file is a part of bzip2 and/or libbzip2, a program and
library for lossless, block-sorting data compression.
Copyright (C) 1996-1998 Julian R Seward. All rights reserved.
Copyright (C) 1996-1999 Julian R Seward. All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
@ -41,9 +41,9 @@
NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
Julian Seward, Guildford, Surrey, UK.
Julian Seward, Cambridge, UK.
jseward@acm.org
bzip2/libbzip2 version 0.9.0c of 18 October 1998
bzip2/libbzip2 version 0.9.5 of 24 May 1999
This program is based on (at least) the work of:
Mike Burrows

65
makefile.msc Normal file
View File

@ -0,0 +1,65 @@
# Makefile for Microsoft Visual C++ 6.0
# usage: nmake -f makefile.msc
# K.M. Syring (syring@gsf.de)
# Fixed up by JRS for bzip2-0.9.5d release.
CC=cl
CFLAGS= -DWIN32 -MD -Ox
OBJS= blocksort.obj \
huffman.obj \
crctable.obj \
randtable.obj \
compress.obj \
decompress.obj \
bzlib.obj
all: lib bzip2 test
bzip2: lib
$(CC) $(CFLAGS) -o bzip2 bzip2.c libbz2.lib setargv.obj
$(CC) $(CFLAGS) -o bzip2recover bzip2recover.c
lib: $(OBJS)
del libbz2.lib
lib /out:libbz2.lib $(OBJS)
test: bzip2
type words1
.\\bzip2 -1 < sample1.ref > sample1.rb2
.\\bzip2 -2 < sample2.ref > sample2.rb2
.\\bzip2 -3 < sample3.ref > sample3.rb2
.\\bzip2 -d < sample1.bz2 > sample1.tst
.\\bzip2 -d < sample2.bz2 > sample2.tst
.\\bzip2 -ds < sample3.bz2 > sample3.tst
fc sample1.bz2 sample1.rb2
fc sample2.bz2 sample2.rb2
fc sample3.bz2 sample3.rb2
fc sample1.tst sample1.ref
fc sample2.tst sample2.ref
fc sample3.tst sample3.ref
@echo All six of the fc's should find no differences.
@echo If fc finds an error on sample3.tst, this could be
@echo because WinZips 'TAR file smart CR/LF conversion'
@echo is too clever for its own good. Disable this option.
@echo The correct size for sample3.ref is 120,244. If it
@echo is around 150k, WinZip has stuffed it up.
@echo Also remember to set BZ_UNIX to 0 and BZ_LCCWIN32
@echo to 1 in bzip2.c.
clean:
del *.obj
del libbz2.lib
del bzip2.exe
del bzip2recover.exe
del sample1.rb2
del sample2.rb2
del sample3.rb2
del sample1.tst
del sample2.tst
del sample3.tst
.c.obj:
$(CC) $(CFLAGS) -c $*.c -o $*.obj

867
manual.texi
View File

File diff suppressed because it is too large Load Diff

6
randtable.c
View File

@ -8,7 +8,7 @@
This file is a part of bzip2 and/or libbzip2, a program and
library for lossless, block-sorting data compression.
Copyright (C) 1996-1998 Julian R Seward. All rights reserved.
Copyright (C) 1996-1999 Julian R Seward. All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
@ -41,9 +41,9 @@
NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
Julian Seward, Guildford, Surrey, UK.
Julian Seward, Cambridge, UK.
jseward@acm.org
bzip2/libbzip2 version 0.9.0c of 18 October 1998
bzip2/libbzip2 version 0.9.5 of 24 May 1999
This program is based on (at least) the work of:
Mike Burrows

BIN
sample3.bz2 Normal file
View File

Binary file not shown.

30007
sample3.ref Normal file
View File

File diff suppressed because it is too large Load Diff

2
words1
View File

@ -1,4 +1,4 @@
Doing 4 tests (2 compress, 2 uncompress) ...
Doing 6 tests (3 compress, 3 uncompress) ...
If there's a problem, things might stop at this point.

21
words3
View File

@ -1,12 +1,17 @@
If you got this far and the "cmp"s didn't find anything amiss, looks
like you're in business. You should install bzip2, bunzip2 and bzcat:
If you got this far and the "cmp"s didn't complain, it looks
like you're in business.
Copy bzip2 and bzip2recover to a public place, maybe /usr/bin.
In that public place, make bunzip2 and bzcat be
symbolic links to the bzip2 you just copied there.
Put the manual page, bzip2.1, somewhere appropriate;
perhaps in /usr/man/man1.
To install in /usr/bin, /usr/lib, /usr/man and /usr/include, type
make install
To install somewhere else, eg, /xxx/yyy/{bin,lib,man,include}, type
make install PREFIX=/xxx/yyy
If you are (justifiably) paranoid and want to see what 'make install'
is going to do, you can first do
make -n install or
make -n install PREFIX=/xxx/yyy respectively.
The -n instructs make to show the commands it would execute, but
not actually execute them.
Instructions for use are in the preformatted manual page, in the file
bzip2.txt. For more detailed documentation, read the full manual.
@ -16,5 +21,3 @@ It is available in Postscript form (manual.ps) and HTML form
You can also do "bzip2 --help" to see some helpful information.
"bzip2 -L" displays the software license.
Happy compressing. -- JRS, 30 August 1998.