summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorJulian Seward <jseward@acm.org>2005-02-15 22:13:13 +0100
committerJulian Seward <jseward@acm.org>2005-02-15 22:13:13 +0100
commit4d540bfc95a4b0eefc1d1f388ec33534aaeb3a2f (patch)
tree3b7e9c650b4c61d114e1716c4698e40d5c8d7ef7
parentbzip2-1.0.2 (diff)
downloadbzip2-4d540bfc95a4b0eefc1d1f388ec33534aaeb3a2f.tar.gz
bzip2-4d540bfc95a4b0eefc1d1f388ec33534aaeb3a2f.tar.bz2
bzip2-4d540bfc95a4b0eefc1d1f388ec33534aaeb3a2f.tar.xz
bzip2-1.0.3bzip2-1.0.3
-rw-r--r--CHANGES22
-rw-r--r--LICENSE7
-rw-r--r--Makefile51
-rw-r--r--Makefile-libbz2_so12
-rw-r--r--README40
-rw-r--r--README.COMPILATION.PROBLEMS103
-rw-r--r--README.XML.STUFF31
-rw-r--r--blocksort.c4
-rw-r--r--bz-common.xsl39
-rw-r--r--bz-fo.xsl257
-rw-r--r--bz-html.xsl20
-rw-r--r--bzip.css74
-rw-r--r--bzip2.117
-rw-r--r--bzip2.1.preformatted247
-rw-r--r--bzip2.c20
-rw-r--r--bzip2.txt119
-rw-r--r--bzip2recover.c12
-rw-r--r--bzlib.c59
-rw-r--r--bzlib.h6
-rw-r--r--bzlib_private.h6
-rw-r--r--compress.c16
-rw-r--r--crctable.c4
-rw-r--r--decompress.c20
-rw-r--r--entities.xml9
-rwxr-xr-xformat.pl53
-rw-r--r--huffman.c23
-rw-r--r--manual.texi2243
-rw-r--r--manual.xml2966
-rw-r--r--randtable.c4
-rwxr-xr-xxmlproc.sh99
30 files changed, 3947 insertions, 2636 deletions
diff --git a/CHANGES b/CHANGES
index d984395..e31b03a 100644
--- a/CHANGES
+++ b/CHANGES
@@ -251,3 +251,25 @@ of bzip2:
251 251
252* added --fast and --best aliases for -1 -9 for gzip compatibility. 252* added --fast and --best aliases for -1 -9 for gzip compatibility.
253 253
254
2551.0.3 (15 Feb 05)
256~~~~~~~~~~~~~~~~~
257Fixes some minor bugs since the last version, 1.0.2.
258
259* Further robustification against corrupted compressed data.
260 There are currently no known bitstreams which can cause the
261 decompressor to crash, loop or access memory which does not
262 belong to it. If you are using bzip2 or the library to
263 decompress bitstreams from untrusted sources, an upgrade
264 to 1.0.3 is recommended.
265
266* The documentation has been converted to XML, from which html
267 and pdf can be derived.
268
269* Various minor bugs in the documentation have been fixed.
270
271* Fixes for various compilation warnings with newer versions of
272 gcc, and on 64-bit platforms.
273
274* The BZ_NO_STDIO cpp symbol was not properly observed in 1.0.2.
275 This has been fixed.
diff --git a/LICENSE b/LICENSE
index 9d4fa43..e60845b 100644
--- a/LICENSE
+++ b/LICENSE
@@ -1,6 +1,7 @@
1 1
2This program, "bzip2" and associated library "libbzip2", are 2This program, "bzip2", the associated library "libbzip2", and all
3copyright (C) 1996-2002 Julian R Seward. All rights reserved. 3documentation, are copyright (C) 1996-2005 Julian R Seward. All
4rights reserved.
4 5
5Redistribution and use in source and binary forms, with or without 6Redistribution and use in source and binary forms, with or without
6modification, are permitted provided that the following conditions 7modification, are permitted provided that the following conditions
@@ -35,5 +36,5 @@ SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
35 36
36Julian Seward, Cambridge, UK. 37Julian Seward, Cambridge, UK.
37jseward@acm.org 38jseward@acm.org
38bzip2/libbzip2 version 1.0.2 of 30 December 2001 39bzip2/libbzip2 version 1.0.3 of 15 February 2005
39 40
diff --git a/Makefile b/Makefile
index 8305235..eea329a 100644
--- a/Makefile
+++ b/Makefile
@@ -7,9 +7,8 @@ AR=ar
7RANLIB=ranlib 7RANLIB=ranlib
8LDFLAGS= 8LDFLAGS=
9 9
10# Suitably paranoid flags to avoid bugs in gcc-2.7
11BIGFILES=-D_FILE_OFFSET_BITS=64 10BIGFILES=-D_FILE_OFFSET_BITS=64
12CFLAGS=-Wall -Winline -O2 -fomit-frame-pointer -fno-strength-reduce $(BIGFILES) 11CFLAGS=-Wall -Winline -O -g $(BIGFILES)
13 12
14# Where you want it installed when you do 'make install' 13# Where you want it installed when you do 'make install'
15PREFIX=/usr 14PREFIX=/usr
@@ -96,7 +95,6 @@ install: bzip2 bzip2recover
96 echo ".so man1/bzmore.1" > $(PREFIX)/man/man1/bzless.1 95 echo ".so man1/bzmore.1" > $(PREFIX)/man/man1/bzless.1
97 echo ".so man1/bzdiff.1" > $(PREFIX)/man/man1/bzcmp.1 96 echo ".so man1/bzdiff.1" > $(PREFIX)/man/man1/bzcmp.1
98 97
99distclean: clean
100clean: 98clean:
101 rm -f *.o libbz2.a bzip2 bzip2recover \ 99 rm -f *.o libbz2.a bzip2 bzip2recover \
102 sample1.rb2 sample2.rb2 sample3.rb2 \ 100 sample1.rb2 sample2.rb2 sample3.rb2 \
@@ -122,8 +120,12 @@ bzip2.o: bzip2.c
122bzip2recover.o: bzip2recover.c 120bzip2recover.o: bzip2recover.c
123 $(CC) $(CFLAGS) -c bzip2recover.c 121 $(CC) $(CFLAGS) -c bzip2recover.c
124 122
125DISTNAME=bzip2-1.0.2 123
126tarfile: 124distclean: clean
125 rm -f manual.ps manual.html manual.pdf
126
127DISTNAME=bzip2-1.0.3
128dist: check manual
127 rm -f $(DISTNAME) 129 rm -f $(DISTNAME)
128 ln -sf . $(DISTNAME) 130 ln -sf . $(DISTNAME)
129 tar cvf $(DISTNAME).tar \ 131 tar cvf $(DISTNAME).tar \
@@ -139,9 +141,6 @@ tarfile:
139 $(DISTNAME)/bzlib.h \ 141 $(DISTNAME)/bzlib.h \
140 $(DISTNAME)/bzlib_private.h \ 142 $(DISTNAME)/bzlib_private.h \
141 $(DISTNAME)/Makefile \ 143 $(DISTNAME)/Makefile \
142 $(DISTNAME)/manual.texi \
143 $(DISTNAME)/manual.ps \
144 $(DISTNAME)/manual.pdf \
145 $(DISTNAME)/LICENSE \ 144 $(DISTNAME)/LICENSE \
146 $(DISTNAME)/bzip2.1 \ 145 $(DISTNAME)/bzip2.1 \
147 $(DISTNAME)/bzip2.1.preformatted \ 146 $(DISTNAME)/bzip2.1.preformatted \
@@ -157,9 +156,12 @@ tarfile:
157 $(DISTNAME)/sample2.bz2 \ 156 $(DISTNAME)/sample2.bz2 \
158 $(DISTNAME)/sample3.bz2 \ 157 $(DISTNAME)/sample3.bz2 \
159 $(DISTNAME)/dlltest.c \ 158 $(DISTNAME)/dlltest.c \
160 $(DISTNAME)/*.html \ 159 $(DISTNAME)/manual.html \
160 $(DISTNAME)/manual.pdf \
161 $(DISTNAME)/manual.ps \
161 $(DISTNAME)/README \ 162 $(DISTNAME)/README \
162 $(DISTNAME)/README.COMPILATION.PROBLEMS \ 163 $(DISTNAME)/README.COMPILATION.PROBLEMS \
164 $(DISTNAME)/README.XML.STUFF \
163 $(DISTNAME)/CHANGES \ 165 $(DISTNAME)/CHANGES \
164 $(DISTNAME)/libbz2.def \ 166 $(DISTNAME)/libbz2.def \
165 $(DISTNAME)/libbz2.dsp \ 167 $(DISTNAME)/libbz2.dsp \
@@ -175,18 +177,29 @@ tarfile:
175 $(DISTNAME)/bzmore.1 \ 177 $(DISTNAME)/bzmore.1 \
176 $(DISTNAME)/bzgrep \ 178 $(DISTNAME)/bzgrep \
177 $(DISTNAME)/bzgrep.1 \ 179 $(DISTNAME)/bzgrep.1 \
178 $(DISTNAME)/Makefile-libbz2_so 180 $(DISTNAME)/Makefile-libbz2_so \
181 $(DISTNAME)/bz-common.xsl \
182 $(DISTNAME)/bz-fo.xsl \
183 $(DISTNAME)/bz-html.xsl \
184 $(DISTNAME)/bzip.css \
185 $(DISTNAME)/entities.xml \
186 $(DISTNAME)/manual.xml \
187 $(DISTNAME)/format.pl \
188 $(DISTNAME)/xmlproc.sh
179 gzip -v $(DISTNAME).tar 189 gzip -v $(DISTNAME).tar
180 190
181# For rebuilding the manual from sources on my RedHat 7.2 box 191# For rebuilding the manual from sources on my SuSE 9.1 box
182manual: manual.ps manual.pdf manual.html 192
193MANUAL_SRCS= bz-common.xsl bz-fo.xsl bz-html.xsl bzip.css \
194 entities.xml manual.xml
195
196manual: manual.html manual.ps manual.pdf
183 197
184manual.ps: manual.texi 198manual.ps: $(MANUAL_SRCS)
185 tex manual.texi 199 ./xmlproc.sh -ps manual.xml
186 dvips -o manual.ps manual.dvi
187 200
188manual.pdf: manual.ps 201manual.pdf: $(MANUAL_SRCS)
189 ps2pdf manual.ps 202 ./xmlproc.sh -pdf manual.xml
190 203
191manual.html: manual.texi 204manual.html: $(MANUAL_SRCS)
192 texi2html -split_chapter manual.texi 205 ./xmlproc.sh -html manual.xml
diff --git a/Makefile-libbz2_so b/Makefile-libbz2_so
index 4986fe2..458c5a1 100644
--- a/Makefile-libbz2_so
+++ b/Makefile-libbz2_so
@@ -1,6 +1,6 @@
1 1
2# This Makefile builds a shared version of the library, 2# This Makefile builds a shared version of the library,
3# libbz2.so.1.0.2, with soname libbz2.so.1.0, 3# libbz2.so.1.0.3, with soname libbz2.so.1.0,
4# at least on x86-Linux (RedHat 7.2), 4# at least on x86-Linux (RedHat 7.2),
5# with gcc-2.96 20000731 (Red Hat Linux 7.1 2.96-98). 5# with gcc-2.96 20000731 (Red Hat Linux 7.1 2.96-98).
6# Please see the README file for some 6# Please see the README file for some
@@ -9,7 +9,7 @@
9SHELL=/bin/sh 9SHELL=/bin/sh
10CC=gcc 10CC=gcc
11BIGFILES=-D_FILE_OFFSET_BITS=64 11BIGFILES=-D_FILE_OFFSET_BITS=64
12CFLAGS=-fpic -fPIC -Wall -Winline -O2 -fomit-frame-pointer -fno-strength-reduce $(BIGFILES) 12CFLAGS=-fpic -fPIC -Wall -Winline -O -g
13 13
14OBJS= blocksort.o \ 14OBJS= blocksort.o \
15 huffman.o \ 15 huffman.o \
@@ -20,13 +20,13 @@ OBJS= blocksort.o \
20 bzlib.o 20 bzlib.o
21 21
22all: $(OBJS) 22all: $(OBJS)
23 $(CC) -shared -Wl,-soname -Wl,libbz2.so.1.0 -o libbz2.so.1.0.2 $(OBJS) 23 $(CC) -shared -Wl,-soname -Wl,libbz2.so.1.0 -o libbz2.so.1.0.3 $(OBJS)
24 $(CC) $(CFLAGS) -o bzip2-shared bzip2.c libbz2.so.1.0.2 24 $(CC) $(CFLAGS) -o bzip2-shared bzip2.c libbz2.so.1.0.3
25 rm -f libbz2.so.1.0 25 rm -f libbz2.so.1.0
26 ln -s libbz2.so.1.0.2 libbz2.so.1.0 26 ln -s libbz2.so.1.0.3 libbz2.so.1.0
27 27
28clean: 28clean:
29 rm -f $(OBJS) bzip2.o libbz2.so.1.0.2 libbz2.so.1.0 bzip2-shared 29 rm -f $(OBJS) bzip2.o libbz2.so.1.0.3 libbz2.so.1.0 bzip2-shared
30 30
31blocksort.o: blocksort.c 31blocksort.o: blocksort.c
32 $(CC) $(CFLAGS) -c blocksort.c 32 $(CC) $(CFLAGS) -c blocksort.c
diff --git a/README b/README
index 07505d8..1aff448 100644
--- a/README
+++ b/README
@@ -1,15 +1,15 @@
1 1
2This is the README for bzip2, a block-sorting file compressor, version 2This is the README for bzip2, a block-sorting file compressor, version
31.0.2. This version is fully compatible with the previous public 31.0.3. This version is fully compatible with the previous public
4releases, versions 0.1pl2, 0.9.0, 0.9.5, 1.0.0 and 1.0.1. 4releases, versions 0.1pl2, 0.9.0, 0.9.5, 1.0.0, 1.0.1 and 1.0.2.
5 5
6bzip2-1.0.2 is distributed under a BSD-style license. For details, 6bzip2-1.0.3 is distributed under a BSD-style license. For details,
7see the file LICENSE. 7see the file LICENSE.
8 8
9Complete documentation is available in Postscript form (manual.ps), 9Complete documentation is available in Postscript form (manual.ps),
10PDF (manual.pdf, amazingly enough) or html (manual_toc.html). A 10PDF (manual.pdf) or html (manual.html). A plain-text version of the
11plain-text version of the manual page is available as bzip2.txt. 11manual page is available as bzip2.txt. A statement about Y2K issues
12A statement about Y2K issues is now included in the file Y2K_INFO. 12is now included in the file Y2K_INFO.
13 13
14 14
15HOW TO BUILD -- UNIX 15HOW TO BUILD -- UNIX
@@ -78,8 +78,7 @@ importance. To validate bzip2, I used a modified version of Mark
78Nelson's churn program. Churn is an automated test driver which 78Nelson's churn program. Churn is an automated test driver which
79recursively traverses a directory structure, using bzip2 to compress 79recursively traverses a directory structure, using bzip2 to compress
80and then decompress each file it encounters, and checking that the 80and then decompress each file it encounters, and checking that the
81decompressed data is the same as the original. There are more details 81decompressed data is the same as the original.
82in Section 4 of the user guide.
83 82
84 83
85 84
@@ -119,9 +118,9 @@ DISCLAIMER:
119PATENTS: 118PATENTS:
120 119
121 To the best of my knowledge, bzip2 does not use any patented 120 To the best of my knowledge, bzip2 does not use any patented
122 algorithms. However, I do not have the resources available to 121 algorithms. However, I do not have the resources to carry out
123 carry out a full patent search. Therefore I cannot give any 122 a patent search. Therefore I cannot give any guarantee of the
124 guarantee of the above statement. 123 above statement.
125 124
126End of legalities. 125End of legalities.
127 126
@@ -153,22 +152,26 @@ WHAT'S NEW IN 1.0.2 ?
153 152
154 See the CHANGES file. 153 See the CHANGES file.
155 154
155WHAT'S NEW IN 1.0.3 ?
156
157 See the CHANGES file.
158
156 159
157I hope you find bzip2 useful. Feel free to contact me at 160I hope you find bzip2 useful. Feel free to contact me at
158 jseward@acm.org 161 jseward@bzip.org
159if you have any suggestions or queries. Many people mailed me with 162if you have any suggestions or queries. Many people mailed me with
160comments, suggestions and patches after the releases of bzip-0.15, 163comments, suggestions and patches after the releases of bzip-0.15,
161bzip-0.21, and bzip2 versions 0.1pl2, 0.9.0, 0.9.5, 1.0.0 and 1.0.1, 164bzip-0.21, and bzip2 versions 0.1pl2, 0.9.0, 0.9.5, 1.0.0, 1.0.1 and
162and the changes in bzip2 are largely a result of this feedback. 1651.0.2, and the changes in bzip2 are largely a result of this feedback.
163I thank you for your comments. 166I thank you for your comments.
164 167
165At least for the time being, bzip2's "home" is (or can be reached via) 168At least for the time being, bzip2's "home" is (or can be reached via)
166http://sources.redhat.com/bzip2. 169http://www.bzip.org
167 170
168Julian Seward 171Julian Seward
169jseward@acm.org 172jseward@bzip.org
170 173
171Cambridge, UK (and what a great town this is!) 174Cambridge, UK.
172 175
17318 July 1996 (version 0.15) 17618 July 1996 (version 0.15)
17425 August 1996 (version 0.21) 17725 August 1996 (version 0.21)
@@ -178,4 +181,5 @@ Cambridge, UK (and what a great town this is!)
178 8 June 1999 (bzip2, version 0.9.5) 181 8 June 1999 (bzip2, version 0.9.5)
179 4 Sept 1999 (bzip2, version 0.9.5d) 182 4 Sept 1999 (bzip2, version 0.9.5d)
180 5 May 2000 (bzip2, version 1.0pre8) 183 5 May 2000 (bzip2, version 1.0pre8)
18130 December 2001 (bzip2, version 1.0.2pre1) \ No newline at end of file 18430 December 2001 (bzip2, version 1.0.2pre1)
18515 February 2005 (bzip2, version 1.0.3)
diff --git a/README.COMPILATION.PROBLEMS b/README.COMPILATION.PROBLEMS
index bd1822d..f1bc396 100644
--- a/README.COMPILATION.PROBLEMS
+++ b/README.COMPILATION.PROBLEMS
@@ -1,11 +1,10 @@
1 1
2bzip2-1.0 should compile without problems on the vast majority of 2bzip2-1.0.3 should compile without problems on the vast majority of
3platforms. Using the supplied Makefile, I've built and tested it 3platforms. Using the supplied Makefile, I've built and tested it
4myself for x86-linux, sparc-solaris, alpha-linux, x86-cygwin32 and 4myself for x86-linux and x86_64-linux. With makefile.msc, Visual C++
5alpha-tru64unix. With makefile.msc, Visual C++ 6.0 and nmake, you can 56.0 and nmake, you can build a native Win32 version too. Large file
6build a native Win32 version too. Large file support seems to work 6support seems to work correctly on at least alpha-tru64unix and
7correctly on at least alpha-tru64unix and x86-cygwin32 (on Windows 7x86-cygwin32 (on Windows 2000).
82000).
9 8
10When I say "large file" I mean a file of size 2,147,483,648 (2^31) 9When I say "large file" I mean a file of size 2,147,483,648 (2^31)
11bytes or above. Many older OSs can't handle files above this size, 10bytes or above. Many older OSs can't handle files above this size,
@@ -22,7 +21,7 @@ The technique of adding -D_FILE_OFFSET_BITS=64 to get large file
22support is, as far as I know, the Recommended Way to get correct large 21support is, as far as I know, the Recommended Way to get correct large
23file support. For more details, see the Large File Support 22file support. For more details, see the Large File Support
24Specification, published by the Large File Summit, at 23Specification, published by the Large File Summit, at
25 http://www.sas.com/standard/large.file/ 24 http://ftp.sas.com/standards/large.file
26 25
27As a general comment, if you get compilation errors which you think 26As a general comment, if you get compilation errors which you think
28are related to large file support, try removing the above define from 27are related to large file support, try removing the above define from
@@ -38,93 +37,3 @@ You can use the spewG.c program to generate huge files to test bzip2's
38large file support, if you are feeling paranoid. Be aware though that 37large file support, if you are feeling paranoid. Be aware though that
39any compilation problems which affect bzip2 will also affect spewG.c, 38any compilation problems which affect bzip2 will also affect spewG.c,
40alas. 39alas.
41
42
43Known problems as of 1.0pre8:
44~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
45
46* HP/UX 10.20 and 11.00, using gcc (2.7.2.3 and 2.95.2): A large
47 number of warnings appear, including the following:
48
49 /usr/include/sys/resource.h: In function `getrlimit':
50 /usr/include/sys/resource.h:168:
51 warning: implicit declaration of function `__getrlimit64'
52 /usr/include/sys/resource.h: In function `setrlimit':
53 /usr/include/sys/resource.h:170:
54 warning: implicit declaration of function `__setrlimit64'
55
56 This would appear to be a problem with large file support, header
57 files and gcc. gcc may or may not give up at this point. If it
58 fails, you might be able to improve matters by adding
59 -D__STDC_EXT__=1
60 to the BIGFILES variable in the Makefile (ie, change its definition
61 to
62 BIGFILES=-D_FILE_OFFSET_BITS=64 -D__STDC_EXT__=1
63
64 Even if gcc does produce a binary which appears to work (ie passes
65 its self-tests), you might want to test it to see if it works properly
66 on large files.
67
68
69* HP/UX 10.20 and 11.00, using HP's cc compiler.
70
71 No specific problems for this combination, except that you'll need to
72 specify the -Ae flag, and zap the gcc-specific stuff
73 -Wall -Winline -O2 -fomit-frame-pointer -fno-strength-reduce.
74 You should retain -D_FILE_OFFSET_BITS=64 in order to get large
75 file support -- which is reported to work ok for this HP/UX + cc
76 combination.
77
78
79* SunOS 4.1.X.
80
81 Amazingly, there are still people out there using this venerable old
82 banger. I shouldn't be too rude -- I started life on SunOS, and
83 it was a pretty darn good OS, way back then. Anyway:
84
85 SunOS doesn't seem to have strerror(), so you'll have to use
86 perror(), perhaps by doing adding this (warning: UNTESTED CODE):
87
88 char* strerror ( int errnum )
89 {
90 if (errnum < 0 || errnum >= sys_nerr)
91 return "Unknown error";
92 else
93 return sys_errlist[errnum];
94 }
95
96 Or you could comment out the relevant calls to strerror; they're
97 not mission-critical. Or you could upgrade to Solaris. Ha ha ha!
98 (what?? you think I've got Bad Attitude?)
99
100
101* Making a shared library on Solaris. (Not really a compilation
102 problem, but many people ask ...)
103
104 Firstly, if you have Solaris 8, either you have libbz2.so already
105 on your system, or you can install it from the Solaris CD.
106
107 Secondly, be aware that there are potential naming conflicts
108 between the .so file supplied with Solaris 8, and the .so file
109 which Makefile-libbz2_so will make. Makefile-libbz2_so creates
110 a .so which has the names which I intend to be "official" as
111 of version 1.0.0 and onwards. Unfortunately, the .so in
112 Solaris 8 appeared before I decided on the final names, so
113 the two libraries are incompatible. We have since communicated
114 and I hope that the problems will have been solved in the next
115 version of Solaris, whenever that might appear.
116
117 All that said: you might be able to get somewhere
118 by finding the line in Makefile-libbz2_so which says
119
120 $(CC) -shared -Wl,-soname -Wl,libbz2.so.1.0 -o libbz2.so.1.0.2 $(OBJS)
121
122 and replacing with
123
124 $(CC) -G -shared -o libbz2.so.1.0.2 -h libbz2.so.1.0 $(OBJS)
125
126 If gcc objects to the combination -fpic -fPIC, get rid of
127 the second one, leaving just "-fpic".
128
129
130That's the end of the currently known compilation problems.
diff --git a/README.XML.STUFF b/README.XML.STUFF
new file mode 100644
index 0000000..0ff209f
--- /dev/null
+++ b/README.XML.STUFF
@@ -0,0 +1,31 @@
1The script xmlproc.sh takes an xml file as input,
2and processes it to create .pdf, .html or .ps output.
3It uses format.pl, a perl script to format <pre> blocks nicely,
4 and add CDATA tags so writers do not have to use eg. &lt;
5
6The file "entities.xml" must be edited to reflect current
7version, year, etc.
8
9
10Usage:
11
12 xmlproc.sh -v manual.xml
13 Validates an xml file to ensure no dtd-compliance errors
14
15 xmlproc.sh -html manual.xml
16 Output: manual.html
17
18 xmlproc.sh -pdf manual.xml
19 Output: manual.pdf
20
21 xmlproc.sh -ps manual.xml
22 Output: manual.ps
23
24
25Notum bene:
26- pdfxmltex barfs if given a filename with an underscore in it
27
28- xmltex won't work yet - there's a bug in passivetex
29 which we are all waiting for Sebastian to fix.
30 So we are going the xml -> pdf -> ps route for the time being,
31 using pdfxmltex.
diff --git a/blocksort.c b/blocksort.c
index aba3efc..33ec9f5 100644
--- a/blocksort.c
+++ b/blocksort.c
@@ -8,7 +8,7 @@
8 This file is a part of bzip2 and/or libbzip2, a program and 8 This file is a part of bzip2 and/or libbzip2, a program and
9 library for lossless, block-sorting data compression. 9 library for lossless, block-sorting data compression.
10 10
11 Copyright (C) 1996-2002 Julian R Seward. All rights reserved. 11 Copyright (C) 1996-2005 Julian R Seward. All rights reserved.
12 12
13 Redistribution and use in source and binary forms, with or without 13 Redistribution and use in source and binary forms, with or without
14 modification, are permitted provided that the following conditions 14 modification, are permitted provided that the following conditions
@@ -42,7 +42,7 @@
42 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 42 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
43 43
44 Julian Seward, Cambridge, UK. 44 Julian Seward, Cambridge, UK.
45 jseward@acm.org 45 jseward@bzip.org
46 bzip2/libbzip2 version 1.0 of 21 March 2000 46 bzip2/libbzip2 version 1.0 of 21 March 2000
47 47
48 This program is based on (at least) the work of: 48 This program is based on (at least) the work of:
diff --git a/bz-common.xsl b/bz-common.xsl
new file mode 100644
index 0000000..66fcd6f
--- /dev/null
+++ b/bz-common.xsl
@@ -0,0 +1,39 @@
1<?xml version="1.0"?> <!-- -*- sgml -*- -->
2<xsl:stylesheet
3 xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">
4
5<!-- we like '1.2 Title' -->
6<xsl:param name="section.autolabel" select="'1'"/>
7<xsl:param name="section.label.includes.component.label" select="'1'"/>
8
9<!-- Do not put 'Chapter' at the start of eg 'Chapter 1. Doing This' -->
10<xsl:param name="local.l10n.xml" select="document('')"/>
11<l:i18n xmlns:l="http://docbook.sourceforge.net/xmlns/l10n/1.0">
12 <l:l10n language="en">
13 <l:context name="title-numbered">
14 <l:template name="chapter" text="%n.&#160;%t"/>
15 </l:context>
16 </l:l10n>
17</l:i18n>
18
19<!-- don't generate sub-tocs for qanda sets -->
20<xsl:param name="generate.toc">
21set toc,title
22book toc,title,figure,table,example,equation
23chapter toc,title
24section toc
25sect1 toc
26sect2 toc
27sect3 toc
28sect4 nop
29sect5 nop
30qandaset toc
31qandadiv nop
32appendix toc,title
33article/appendix nop
34article toc,title
35preface toc,title
36reference toc,title
37</xsl:param>
38
39</xsl:stylesheet>
diff --git a/bz-fo.xsl b/bz-fo.xsl
new file mode 100644
index 0000000..7f2a767
--- /dev/null
+++ b/bz-fo.xsl
@@ -0,0 +1,257 @@
1<?xml version="1.0" encoding="UTF-8"?> <!-- -*- sgml -*- -->
2<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
3 xmlns:fo="http://www.w3.org/1999/XSL/Format" version="1.0">
4
5<xsl:import href="http://docbook.sourceforge.net/release/xsl/current/fo/docbook.xsl"/>
6<xsl:import href="bz-common.xsl"/>
7
8<!-- set indent = yes while debugging, then change to NO -->
9<xsl:output method="xml" indent="yes"/>
10
11<!-- ensure only passivetex extensions are on -->
12<xsl:param name="stylesheet.result.type" select="'fo'"/>
13<!-- fo extensions: PDF bookmarks and index terms -->
14<xsl:param name="use.extensions" select="'1'"/>
15<xsl:param name="xep.extensions" select="0"/>
16<xsl:param name="fop.extensions" select="0"/>
17<xsl:param name="saxon.extensions" select="0"/>
18<xsl:param name="passivetex.extensions" select="1"/>
19<xsl:param name="tablecolumns.extension" select="'1'"/>
20
21<!-- ensure we are using single sided -->
22<xsl:param name="double.sided" select="'0'"/>
23
24<!-- insert cross references to page numbers -->
25<xsl:param name="insert.xref.page.number" select="1"/>
26
27<!-- <?custom-pagebreak?> inserts a page break at this point -->
28<xsl:template match="processing-instruction('custom-pagebreak')">
29 <fo:block break-before='page'/>
30</xsl:template>
31
32<!-- show links in color -->
33<xsl:attribute-set name="xref.properties">
34 <xsl:attribute name="color">blue</xsl:attribute>
35</xsl:attribute-set>
36
37<!-- make pre listings indented a bit + a bg colour -->
38<xsl:template match="programlisting | screen">
39 <fo:block start-indent="0.25in" wrap-option="no-wrap"
40 white-space-collapse="false" text-align="start"
41 font-family="monospace" background-color="#f2f2f9"
42 linefeed-treatment="preserve"
43 xsl:use-attribute-sets="normal.para.spacing">
44 <xsl:apply-templates/>
45 </fo:block>
46</xsl:template>
47<!-- make verbatim output prettier -->
48<xsl:template match="literallayout">
49 <fo:block start-indent="0.25in" wrap-option="no-wrap"
50 white-space-collapse="false" text-align="start"
51 font-family="monospace" background-color="#edf7f4"
52 linefeed-treatment="preserve"
53 space-before="0em" space-after="0em">
54 <xsl:apply-templates/>
55 </fo:block>
56</xsl:template>
57
58<!-- workaround bug in passivetex fo output for itemizedlist -->
59<xsl:template match="itemizedlist/listitem">
60 <xsl:variable name="id">
61 <xsl:call-template name="object.id"/></xsl:variable>
62 <xsl:variable name="itemsymbol">
63 <xsl:call-template name="list.itemsymbol">
64 <xsl:with-param name="node" select="parent::itemizedlist"/>
65 </xsl:call-template>
66 </xsl:variable>
67 <xsl:variable name="item.contents">
68 <fo:list-item-label end-indent="label-end()">
69 <fo:block>
70 <xsl:choose>
71 <xsl:when test="$itemsymbol='disc'">&#x2022;</xsl:when>
72 <xsl:when test="$itemsymbol='bullet'">&#x2022;</xsl:when>
73 <xsl:otherwise>&#x2022;</xsl:otherwise>
74 </xsl:choose>
75 </fo:block>
76 </fo:list-item-label>
77 <fo:list-item-body start-indent="body-start()">
78 <xsl:apply-templates/> <!-- removed extra block wrapper -->
79 </fo:list-item-body>
80 </xsl:variable>
81 <xsl:choose>
82 <xsl:when test="parent::*/@spacing = 'compact'">
83 <fo:list-item id="{$id}"
84 xsl:use-attribute-sets="compact.list.item.spacing">
85 <xsl:copy-of select="$item.contents"/>
86 </fo:list-item>
87 </xsl:when>
88 <xsl:otherwise>
89 <fo:list-item id="{$id}" xsl:use-attribute-sets="list.item.spacing">
90 <xsl:copy-of select="$item.contents"/>
91 </fo:list-item>
92 </xsl:otherwise>
93 </xsl:choose>
94</xsl:template>
95
96<!-- workaround bug in passivetex fo output for orderedlist -->
97<xsl:template match="orderedlist/listitem">
98 <xsl:variable name="id">
99 <xsl:call-template name="object.id"/></xsl:variable>
100 <xsl:variable name="item.contents">
101 <fo:list-item-label end-indent="label-end()">
102 <fo:block>
103 <xsl:apply-templates select="." mode="item-number"/>
104 </fo:block>
105 </fo:list-item-label>
106 <fo:list-item-body start-indent="body-start()">
107 <xsl:apply-templates/> <!-- removed extra block wrapper -->
108 </fo:list-item-body>
109 </xsl:variable>
110 <xsl:choose>
111 <xsl:when test="parent::*/@spacing = 'compact'">
112 <fo:list-item id="{$id}"
113 xsl:use-attribute-sets="compact.list.item.spacing">
114 <xsl:copy-of select="$item.contents"/>
115 </fo:list-item>
116 </xsl:when>
117 <xsl:otherwise>
118 <fo:list-item id="{$id}" xsl:use-attribute-sets="list.item.spacing">
119 <xsl:copy-of select="$item.contents"/>
120 </fo:list-item>
121 </xsl:otherwise>
122 </xsl:choose>
123</xsl:template>
124
125<!-- workaround bug in passivetex fo output for variablelist -->
126<xsl:param name="variablelist.as.blocks" select="1"/>
127<xsl:template match="varlistentry" mode="vl.as.blocks">
128 <xsl:variable name="id">
129 <xsl:call-template name="object.id"/></xsl:variable>
130 <fo:block id="{$id}" xsl:use-attribute-sets="list.item.spacing"
131 keep-together.within-column="always"
132 keep-with-next.within-column="always">
133 <xsl:apply-templates select="term"/>
134 </fo:block>
135 <fo:block start-indent="0.5in" end-indent="0in"
136 space-after.minimum="0.2em"
137 space-after.optimum="0.4em"
138 space-after.maximum="0.6em">
139 <fo:block>
140 <xsl:apply-templates select="listitem"/>
141 </fo:block>
142 </fo:block>
143</xsl:template>
144
145
146<!-- workaround bug in footers: force right-align w/two 80|30 cols -->
147<xsl:template name="footer.table">
148 <xsl:param name="pageclass" select="''"/>
149 <xsl:param name="sequence" select="''"/>
150 <xsl:param name="gentext-key" select="''"/>
151 <xsl:choose>
152 <xsl:when test="$pageclass = 'index'">
153 <xsl:attribute name="margin-left">0pt</xsl:attribute>
154 </xsl:when>
155 </xsl:choose>
156 <xsl:variable name="candidate">
157 <fo:table table-layout="fixed" width="100%">
158 <fo:table-column column-number="1" column-width="80%"/>
159 <fo:table-column column-number="2" column-width="20%"/>
160 <fo:table-body>
161 <fo:table-row height="14pt">
162 <fo:table-cell text-align="left" display-align="after">
163 <xsl:attribute name="relative-align">baseline</xsl:attribute>
164 <fo:block>
165 <fo:block> </fo:block><!-- empty cell -->
166 </fo:block>
167 </fo:table-cell>
168 <fo:table-cell text-align="center" display-align="after">
169 <xsl:attribute name="relative-align">baseline</xsl:attribute>
170 <fo:block>
171 <xsl:call-template name="footer.content">
172 <xsl:with-param name="pageclass" select="$pageclass"/>
173 <xsl:with-param name="sequence" select="$sequence"/>
174 <xsl:with-param name="position" select="'center'"/>
175 <xsl:with-param name="gentext-key" select="$gentext-key"/>
176 </xsl:call-template>
177 </fo:block>
178 </fo:table-cell>
179 </fo:table-row>
180 </fo:table-body>
181 </fo:table>
182 </xsl:variable>
183 <!-- Really output a footer? -->
184 <xsl:choose>
185 <xsl:when test="$pageclass='titlepage' and $gentext-key='book'
186 and $sequence='first'">
187 <!-- no, book titlepages have no footers at all -->
188 </xsl:when>
189 <xsl:when test="$sequence = 'blank' and $footers.on.blank.pages = 0">
190 <!-- no output -->
191 </xsl:when>
192 <xsl:otherwise>
193 <xsl:copy-of select="$candidate"/>
194 </xsl:otherwise>
195 </xsl:choose>
196</xsl:template>
197
198
199<!-- fix bug in headers: force right-align w/two 40|60 cols -->
200<xsl:template name="header.table">
201 <xsl:param name="pageclass" select="''"/>
202 <xsl:param name="sequence" select="''"/>
203 <xsl:param name="gentext-key" select="''"/>
204 <xsl:choose>
205 <xsl:when test="$pageclass = 'index'">
206 <xsl:attribute name="margin-left">0pt</xsl:attribute>
207 </xsl:when>
208 </xsl:choose>
209 <xsl:variable name="candidate">
210 <fo:table table-layout="fixed" width="100%">
211 <xsl:call-template name="head.sep.rule">
212 <xsl:with-param name="pageclass" select="$pageclass"/>
213 <xsl:with-param name="sequence" select="$sequence"/>
214 <xsl:with-param name="gentext-key" select="$gentext-key"/>
215 </xsl:call-template>
216 <fo:table-column column-number="1" column-width="40%"/>
217 <fo:table-column column-number="2" column-width="60%"/>
218 <fo:table-body>
219 <fo:table-row height="14pt">
220 <fo:table-cell text-align="left" display-align="before">
221 <xsl:attribute name="relative-align">baseline</xsl:attribute>
222 <fo:block>
223 <fo:block> </fo:block><!-- empty cell -->
224 </fo:block>
225 </fo:table-cell>
226 <fo:table-cell text-align="center" display-align="before">
227 <xsl:attribute name="relative-align">baseline</xsl:attribute>
228 <fo:block>
229 <xsl:call-template name="header.content">
230 <xsl:with-param name="pageclass" select="$pageclass"/>
231 <xsl:with-param name="sequence" select="$sequence"/>
232 <xsl:with-param name="position" select="'center'"/>
233 <xsl:with-param name="gentext-key" select="$gentext-key"/>
234 </xsl:call-template>
235 </fo:block>
236 </fo:table-cell>
237 </fo:table-row>
238 </fo:table-body>
239 </fo:table>
240 </xsl:variable>
241 <!-- Really output a header? -->
242 <xsl:choose>
243 <xsl:when test="$pageclass = 'titlepage' and $gentext-key = 'book'
244 and $sequence='first'">
245 <!-- no, book titlepages have no headers at all -->
246 </xsl:when>
247 <xsl:when test="$sequence = 'blank' and $headers.on.blank.pages = 0">
248 <!-- no output -->
249 </xsl:when>
250 <xsl:otherwise>
251 <xsl:copy-of select="$candidate"/>
252 </xsl:otherwise>
253 </xsl:choose>
254</xsl:template>
255
256
257</xsl:stylesheet>
diff --git a/bz-html.xsl b/bz-html.xsl
new file mode 100644
index 0000000..1785fff
--- /dev/null
+++ b/bz-html.xsl
@@ -0,0 +1,20 @@
1<?xml version="1.0"?> <!-- -*- sgml -*- -->
2<!DOCTYPE xsl:stylesheet [ <!ENTITY bz-css SYSTEM "./bzip.css"> ]>
3
4<xsl:stylesheet
5 xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">
6
7<xsl:import href="http://docbook.sourceforge.net/release/xsl/current/html/docbook.xsl"/>
8<xsl:import href="bz-common.xsl"/>
9
10<!-- use 8859-1 encoding -->
11<xsl:output method="html" encoding="ISO-8859-1" indent="yes"/>
12
13<!-- we include the css directly when generating one large file -->
14<xsl:template name="user.head.content">
15 <style type="text/css" media="screen">
16 <xsl:text>&bz-css;</xsl:text>
17 </style>
18</xsl:template>
19
20</xsl:stylesheet>
diff --git a/bzip.css b/bzip.css
new file mode 100644
index 0000000..43193d8
--- /dev/null
+++ b/bzip.css
@@ -0,0 +1,74 @@
1/* Colours:
2#74240f dark brown h1, h2, h3, h4
3#336699 medium blue links
4#339999 turquoise link hover colour
5#202020 almost black general text
6#761596 purple md5sum text
7#626262 dark gray pre border
8#eeeeee very light gray pre background
9#f2f2f9 very light blue nav table background
10#3366cc medium blue nav table border
11*/
12
13a, a:link, a:visited, a:active { color: #336699; }
14a:hover { color: #339999; }
15
16body { font: 80%/126% sans-serif; }
17h1, h2, h3, h4 { color: #74240f; }
18
19dt { color: #336699; font-weight: bold }
20dd {
21 margin-left: 1.5em;
22 padding-bottom: 0.8em;
23}
24
25/* -- ruler -- */
26div.hr_blue {
27 height: 3px;
28 background:#ffffff url("/images/hr_blue.png") repeat-x; }
29div.hr_blue hr { display:none; }
30
31/* release styles */
32#release p { margin-top: 0.4em; }
33#release .md5sum { color: #761596; }
34
35
36/* ------ styles for docs|manuals|howto ------ */
37/* -- lists -- */
38ul {
39 margin: 0px 4px 16px 16px;
40 padding: 0px;
41 list-style: url("/images/li-blue.png");
42}
43ul li {
44 margin-bottom: 10px;
45}
46ul ul {
47 list-style-type: none;
48 list-style-image: none;
49 margin-left: 0px;
50}
51
52/* header / footer nav tables */
53table.nav {
54 border: solid 1px #3366cc;
55 background: #f2f2f9;
56 background-color: #f2f2f9;
57 margin-bottom: 0.5em;
58}
59/* don't have underlined links in chunked nav menus */
60table.nav a { text-decoration: none; }
61table.nav a:hover { text-decoration: underline; }
62table.nav td { font-size: 85%; }
63
64code, tt, pre { font-size: 120%; }
65code, tt { color: #761596; }
66
67div.literallayout, pre.programlisting, pre.screen {
68 color: #000000;
69 padding: 0.5em;
70 background: #eeeeee;
71 border: 1px solid #626262;
72 background-color: #eeeeee;
73 margin: 4px 0px 4px 0px;
74}
diff --git a/bzip2.1 b/bzip2.1
index 623435c..d2c0661 100644
--- a/bzip2.1
+++ b/bzip2.1
@@ -1,7 +1,7 @@
1.PU 1.PU
2.TH bzip2 1 2.TH bzip2 1
3.SH NAME 3.SH NAME
4bzip2, bunzip2 \- a block-sorting file compressor, v1.0.2 4bzip2, bunzip2 \- a block-sorting file compressor, v1.0.3
5.br 5.br
6bzcat \- decompresses files to stdout 6bzcat \- decompresses files to stdout
7.br 7.br
@@ -405,19 +405,19 @@ I/O error messages are not as helpful as they could be.
405tries hard to detect I/O errors and exit cleanly, but the details of 405tries hard to detect I/O errors and exit cleanly, but the details of
406what the problem is sometimes seem rather misleading. 406what the problem is sometimes seem rather misleading.
407 407
408This manual page pertains to version 1.0.2 of 408This manual page pertains to version 1.0.3 of
409.I bzip2. 409.I bzip2.
410Compressed data created by this version is entirely forwards and 410Compressed data created by this version is entirely forwards and
411backwards compatible with the previous public releases, versions 411backwards compatible with the previous public releases, versions
4120.1pl2, 0.9.0, 0.9.5, 1.0.0 and 1.0.1, but with the following 4120.1pl2, 0.9.0, 0.9.5, 1.0.0, 1.0.1 and 1.0.2, but with the following
413exception: 0.9.0 and above can correctly decompress multiple 413exception: 0.9.0 and above can correctly decompress multiple
414concatenated compressed files. 0.1pl2 cannot do this; it will stop 414concatenated compressed files. 0.1pl2 cannot do this; it will stop
415after decompressing just the first file in the stream. 415after decompressing just the first file in the stream.
416 416
417.I bzip2recover 417.I bzip2recover
418versions prior to this one, 1.0.2, used 32-bit integers to represent 418versions prior to 1.0.2 used 32-bit integers to represent
419bit positions in compressed files, so it could not handle compressed 419bit positions in compressed files, so they could not handle compressed
420files more than 512 megabytes long. Version 1.0.2 and above uses 420files more than 512 megabytes long. Versions 1.0.2 and above use
42164-bit ints on some platforms which support them (GNU supported 42164-bit ints on some platforms which support them (GNU supported
422targets, and Windows). To establish whether or not bzip2recover was 422targets, and Windows). To establish whether or not bzip2recover was
423built with such a limitation, run it without arguments. In any event 423built with such a limitation, run it without arguments. In any event
@@ -427,9 +427,9 @@ with MaybeUInt64 set to be an unsigned 64-bit integer.
427 427
428 428
429.SH AUTHOR 429.SH AUTHOR
430Julian Seward, jseward@acm.org. 430Julian Seward, jsewardbzip.org.
431 431
432http://sources.redhat.com/bzip2 432http://www.bzip.org
433 433
434The ideas embodied in 434The ideas embodied in
435.I bzip2 435.I bzip2
@@ -447,6 +447,7 @@ source distribution for pointers to sources of documentation. Christian
447von Roques encouraged me to look for faster sorting algorithms, so as to 447von Roques encouraged me to look for faster sorting algorithms, so as to
448speed up compression. Bela Lubkin encouraged me to improve the 448speed up compression. Bela Lubkin encouraged me to improve the
449worst-case compression performance. 449worst-case compression performance.
450Donna Robinson XMLised the documentation.
450The bz* scripts are derived from those of GNU gzip. 451The bz* scripts are derived from those of GNU gzip.
451Many people sent patches, helped 452Many people sent patches, helped
452with portability problems, lent machines, gave advice and were generally 453with portability problems, lent machines, gave advice and were generally
diff --git a/bzip2.1.preformatted b/bzip2.1.preformatted
index 0f20cb5..129ca83 100644
--- a/bzip2.1.preformatted
+++ b/bzip2.1.preformatted
@@ -3,43 +3,43 @@ bzip2(1) bzip2(1)
3 3
4 4
5NNAAMMEE 5NNAAMMEE
6 bzip2, bunzip2 - a block-sorting file compressor, v1.0.2 6 bzip2, bunzip2 a blocksorting file compressor, v1.0.3
7 bzcat - decompresses files to stdout 7 bzcat decompresses files to stdout
8 bzip2recover - recovers data from damaged bzip2 files 8 bzip2recover recovers data from damaged bzip2 files
9 9
10 10
11SSYYNNOOPPSSIISS 11SSYYNNOOPPSSIISS
12 bbzziipp22 [ --ccddffkkqqssttvvzzVVLL112233445566778899 ] [ _f_i_l_e_n_a_m_e_s _._._. ] 12 bbzziipp22 [ −−ccddffkkqqssttvvzzVVLL112233445566778899 ] [ _f_i_l_e_n_a_m_e_s _._._. ]
13 bbuunnzziipp22 [ --ffkkvvssVVLL ] [ _f_i_l_e_n_a_m_e_s _._._. ] 13 bbuunnzziipp22 [ ffkkvvssVVLL ] [ _f_i_l_e_n_a_m_e_s _._._. ]
14 bbzzccaatt [ --ss ] [ _f_i_l_e_n_a_m_e_s _._._. ] 14 bbzzccaatt [ ss ] [ _f_i_l_e_n_a_m_e_s _._._. ]
15 bbzziipp22rreeccoovveerr _f_i_l_e_n_a_m_e 15 bbzziipp22rreeccoovveerr _f_i_l_e_n_a_m_e
16 16
17 17
18DDEESSCCRRIIPPTTIIOONN 18DDEESSCCRRIIPPTTIIOONN
19 _b_z_i_p_2 compresses files using the Burrows-Wheeler block 19 _b_z_i_p_2 compresses files using the BurrowsWheeler block
20 sorting text compression algorithm, and Huffman coding. 20 sorting text compression algorithm, and Huffman coding.
21 Compression is generally considerably better than that 21 Compression is generally considerably better than that
22 achieved by more conventional LZ77/LZ78-based compressors, 22 achieved by more conventional LZ77/LZ78based compressors,
23 and approaches the performance of the PPM family of sta 23 and approaches the performance of the PPM family of sta
24 tistical compressors. 24 tistical compressors.
25 25
26 The command-line options are deliberately very similar to 26 The commandline options are deliberately very similar to
27 those of _G_N_U _g_z_i_p_, but they are not identical. 27 those of _G_N_U _g_z_i_p_, but they are not identical.
28 28
29 _b_z_i_p_2 expects a list of file names to accompany the com 29 _b_z_i_p_2 expects a list of file names to accompany the com
30 mand-line flags. Each file is replaced by a compressed 30 mandline flags. Each file is replaced by a compressed
31 version of itself, with the name "original_name.bz2". 31 version of itself, with the name "original_name.bz2".
32 Each compressed file has the same modification date, per 32 Each compressed file has the same modification date, per
33 missions, and, when possible, ownership as the correspond 33 missions, and, when possible, ownership as the correspond
34 ing original, so that these properties can be correctly 34 ing original, so that these properties can be correctly
35 restored at decompression time. File name handling is 35 restored at decompression time. File name handling is
36 naive in the sense that there is no mechanism for preserv 36 naive in the sense that there is no mechanism for preserv
37 ing original file names, permissions, ownerships or dates 37 ing original file names, permissions, ownerships or dates
38 in filesystems which lack these concepts, or have serious 38 in filesystems which lack these concepts, or have serious
39 file name length restrictions, such as MS-DOS. 39 file name length restrictions, such as MSDOS.
40 40
41 _b_z_i_p_2 and _b_u_n_z_i_p_2 will by default not overwrite existing 41 _b_z_i_p_2 and _b_u_n_z_i_p_2 will by default not overwrite existing
42 files. If you want this to happen, specify the -f flag. 42 files. If you want this to happen, specify the f flag.
43 43
44 If no file names are specified, _b_z_i_p_2 compresses from 44 If no file names are specified, _b_z_i_p_2 compresses from
45 standard input to standard output. In this case, _b_z_i_p_2 45 standard input to standard output. In this case, _b_z_i_p_2
@@ -47,7 +47,7 @@ DDEESSCCRRIIPPTTIIOONN
47 this would be entirely incomprehensible and therefore 47 this would be entirely incomprehensible and therefore
48 pointless. 48 pointless.
49 49
50 _b_u_n_z_i_p_2 (or _b_z_i_p_2 _-_d_) decompresses all specified files. 50 _b_u_n_z_i_p_2 (or _b_z_i_p_2 __d_) decompresses all specified files.
51 Files which were not created by _b_z_i_p_2 will be detected and 51 Files which were not created by _b_z_i_p_2 will be detected and
52 ignored, and a warning issued. _b_z_i_p_2 attempts to guess 52 ignored, and a warning issued. _b_z_i_p_2 attempts to guess
53 the filename for the decompressed file from that of the 53 the filename for the decompressed file from that of the
@@ -64,26 +64,26 @@ DDEESSCCRRIIPPTTIIOONN
64 guess the name of the original file, and uses the original 64 guess the name of the original file, and uses the original
65 name with _._o_u_t appended. 65 name with _._o_u_t appended.
66 66
67 As with compression, supplying no filenames causes decom 67 As with compression, supplying no filenames causes decom
68 pression from standard input to standard output. 68 pression from standard input to standard output.
69 69
70 _b_u_n_z_i_p_2 will correctly decompress a file which is the con 70 _b_u_n_z_i_p_2 will correctly decompress a file which is the con
71 catenation of two or more compressed files. The result is 71 catenation of two or more compressed files. The result is
72 the concatenation of the corresponding uncompressed files. 72 the concatenation of the corresponding uncompressed files.
73 Integrity testing (-t) of concatenated compressed files is 73 Integrity testing (t) of concatenated compressed files is
74 also supported. 74 also supported.
75 75
76 You can also compress or decompress files to the standard 76 You can also compress or decompress files to the standard
77 output by giving the -c flag. Multiple files may be com 77 output by giving the c flag. Multiple files may be com
78 pressed and decompressed like this. The resulting outputs 78 pressed and decompressed like this. The resulting outputs
79 are fed sequentially to stdout. Compression of multiple 79 are fed sequentially to stdout. Compression of multiple
80 files in this manner generates a stream containing multi 80 files in this manner generates a stream containing multi
81 ple compressed file representations. Such a stream can be 81 ple compressed file representations. Such a stream can be
82 decompressed correctly only by _b_z_i_p_2 version 0.9.0 or 82 decompressed correctly only by _b_z_i_p_2 version 0.9.0 or
83 later. Earlier versions of _b_z_i_p_2 will stop after decom 83 later. Earlier versions of _b_z_i_p_2 will stop after decom
84 pressing the first file in the stream. 84 pressing the first file in the stream.
85 85
86 _b_z_c_a_t (or _b_z_i_p_2 _-_d_c_) decompresses all specified files to 86 _b_z_c_a_t (or _b_z_i_p_2 __d_c_) decompresses all specified files to
87 the standard output. 87 the standard output.
88 88
89 _b_z_i_p_2 will read arguments from the environment variables 89 _b_z_i_p_2 will read arguments from the environment variables
@@ -99,15 +99,15 @@ DDEESSCCRRIIPPTTIIOONN
99 most file compressors) is coded at about 8.05 bits per 99 most file compressors) is coded at about 8.05 bits per
100 byte, giving an expansion of around 0.5%. 100 byte, giving an expansion of around 0.5%.
101 101
102 As a self-check for your protection, _b_z_i_p_2 uses 32-bit 102 As a selfcheck for your protection, _b_z_i_p_2 uses 32bit
103 CRCs to make sure that the decompressed version of a file 103 CRCs to make sure that the decompressed version of a file
104 is identical to the original. This guards against corrup 104 is identical to the original. This guards against corrup
105 tion of the compressed data, and against undetected bugs 105 tion of the compressed data, and against undetected bugs
106 in _b_z_i_p_2 (hopefully very unlikely). The chances of data 106 in _b_z_i_p_2 (hopefully very unlikely). The chances of data
107 corruption going undetected is microscopic, about one 107 corruption going undetected is microscopic, about one
108 chance in four billion for each file processed. Be aware, 108 chance in four billion for each file processed. Be aware,
109 though, that the check occurs upon decompression, so it 109 though, that the check occurs upon decompression, so it
110 can only tell you that something is wrong. It can't help 110 can only tell you that something is wrong. It cant help
111 you recover the original uncompressed data. You can use 111 you recover the original uncompressed data. You can use
112 _b_z_i_p_2_r_e_c_o_v_e_r to try to recover data from damaged files. 112 _b_z_i_p_2_r_e_c_o_v_e_r to try to recover data from damaged files.
113 113
@@ -118,41 +118,41 @@ DDEESSCCRRIIPPTTIIOONN
118 118
119 119
120OOPPTTIIOONNSS 120OOPPTTIIOONNSS
121 --cc ----ssttddoouutt 121 cc ‐‐ssttddoouutt
122 Compress or decompress to standard output. 122 Compress or decompress to standard output.
123 123
124 --dd ----ddeeccoommpprreessss 124 dd ‐‐ddeeccoommpprreessss
125 Force decompression. _b_z_i_p_2_, _b_u_n_z_i_p_2 and _b_z_c_a_t are 125 Force decompression. _b_z_i_p_2_, _b_u_n_z_i_p_2 and _b_z_c_a_t are
126 really the same program, and the decision about 126 really the same program, and the decision about
127 what actions to take is done on the basis of which 127 what actions to take is done on the basis of which
128 name is used. This flag overrides that mechanism, 128 name is used. This flag overrides that mechanism,
129 and forces _b_z_i_p_2 to decompress. 129 and forces _b_z_i_p_2 to decompress.
130 130
131 --zz ----ccoommpprreessss 131 zz ‐‐ccoommpprreessss
132 The complement to -d: forces compression, 132 The complement to d: forces compression,
133 regardless of the invocation name. 133 regardless of the invocation name.
134 134
135 --tt ----tteesstt 135 tt ‐‐tteesstt
136 Check integrity of the specified file(s), but don't 136 Check integrity of the specified file(s), but dont
137 decompress them. This really performs a trial 137 decompress them. This really performs a trial
138 decompression and throws away the result. 138 decompression and throws away the result.
139 139
140 --ff ----ffoorrccee 140 ff ‐‐ffoorrccee
141 Force overwrite of output files. Normally, _b_z_i_p_2 141 Force overwrite of output files. Normally, _b_z_i_p_2
142 will not overwrite existing output files. Also 142 will not overwrite existing output files. Also
143 forces _b_z_i_p_2 to break hard links to files, which it 143 forces _b_z_i_p_2 to break hard links to files, which it
144 otherwise wouldn't do. 144 otherwise wouldnt do.
145 145
146 bzip2 normally declines to decompress files which 146 bzip2 normally declines to decompress files which
147 don't have the correct magic header bytes. If 147 dont have the correct magic header bytes. If
148 forced (-f), however, it will pass such files 148 forced (f), however, it will pass such files
149 through unmodified. This is how GNU gzip behaves. 149 through unmodified. This is how GNU gzip behaves.
150 150
151 --kk ----kkeeeepp 151 kk ‐‐kkeeeepp
152 Keep (don't delete) input files during compression 152 Keep (dont delete) input files during compression
153 or decompression. 153 or decompression.
154 154
155 --ss ----ssmmaallll 155 ss ‐‐ssmmaallll
156 Reduce memory usage, for compression, decompression 156 Reduce memory usage, for compression, decompression
157 and testing. Files are decompressed and tested 157 and testing. Files are decompressed and tested
158 using a modified algorithm which only requires 2.5 158 using a modified algorithm which only requires 2.5
@@ -160,46 +160,46 @@ OOPPTTIIOONNSS
160 decompressed in 2300k of memory, albeit at about 160 decompressed in 2300k of memory, albeit at about
161 half the normal speed. 161 half the normal speed.
162 162
163 During compression, -s selects a block size of 163 During compression, s selects a block size of
164 200k, which limits memory use to around the same 164 200k, which limits memory use to around the same
165 figure, at the expense of your compression ratio. 165 figure, at the expense of your compression ratio.
166 In short, if your machine is low on memory (8 166 In short, if your machine is low on memory (8
167 megabytes or less), use -s for everything. See 167 megabytes or less), use s for everything. See
168 MEMORY MANAGEMENT below. 168 MEMORY MANAGEMENT below.
169 169
170 --qq ----qquuiieett 170 qq ‐‐qquuiieett
171 Suppress non-essential warning messages. Messages 171 Suppress nonessential warning messages. Messages
172 pertaining to I/O errors and other critical events 172 pertaining to I/O errors and other critical events
173 will not be suppressed. 173 will not be suppressed.
174 174
175 --vv ----vveerrbboossee 175 vv ‐‐vveerrbboossee
176 Verbose mode -- show the compression ratio for each 176 Verbose mode ‐‐ show the compression ratio for each
177 file processed. Further -v's increase the ver 177 file processed. Further vs increase the ver
178 bosity level, spewing out lots of information which 178 bosity level, spewing out lots of information which
179 is primarily of interest for diagnostic purposes. 179 is primarily of interest for diagnostic purposes.
180 180
181 --LL ----lliicceennssee --VV ----vveerrssiioonn 181 LL ‐‐lliicceennssee VV ‐‐vveerrssiioonn
182 Display the software version, license terms and 182 Display the software version, license terms and
183 conditions. 183 conditions.
184 184
185 --11 ((oorr ----ffaasstt)) ttoo --99 ((oorr ----bbeesstt)) 185 11 ((oorr −−ffaasstt)) ttoo 99 ((oorr −−bbeesstt))
186 Set the block size to 100 k, 200 k .. 900 k when 186 Set the block size to 100 k, 200 k .. 900 k when
187 compressing. Has no effect when decompressing. 187 compressing. Has no effect when decompressing.
188 See MEMORY MANAGEMENT below. The --fast and --best 188 See MEMORY MANAGEMENT below. The −−fast and −−best
189 aliases are primarily for GNU gzip compatibility. 189 aliases are primarily for GNU gzip compatibility.
190 In particular, --fast doesn't make things signifi 190 In particular, −−fast doesnt make things signifi
191 cantly faster. And --best merely selects the 191 cantly faster. And −−best merely selects the
192 default behaviour. 192 default behaviour.
193 193
194 ---- Treats all subsequent arguments as file names, even 194 −‐ Treats all subsequent arguments as file names, even
195 if they start with a dash. This is so you can han 195 if they start with a dash. This is so you can han
196 dle files with names beginning with a dash, for 196 dle files with names beginning with a dash, for
197 example: bzip2 -- -myfilename. 197 example: bzip2 −‐ myfilename.
198 198
199 ----rreeppeettiittiivvee--ffaasstt ----rreeppeettiittiivvee--bbeesstt 199 −−‐‐rreeppeettiittiivvee‐‐ffaasstt ‐‐‐‐rreeppeettiittiivvee‐‐bbeesstt
200 These flags are redundant in versions 0.9.5 and 200 These flags are redundant in versions 0.9.5 and
201 above. They provided some coarse control over the 201 above. They provided some coarse control over the
202 behaviour of the sorting algorithm in earlier ver 202 behaviour of the sorting algorithm in earlier ver
203 sions, which was sometimes useful. 0.9.5 and above 203 sions, which was sometimes useful. 0.9.5 and above
204 have an improved algorithm which renders these 204 have an improved algorithm which renders these
205 flags irrelevant. 205 flags irrelevant.
@@ -209,13 +209,13 @@ MMEEMMOORRYY MMAANNAAGGEEMMEENNTT
209 _b_z_i_p_2 compresses large files in blocks. The block size 209 _b_z_i_p_2 compresses large files in blocks. The block size
210 affects both the compression ratio achieved, and the 210 affects both the compression ratio achieved, and the
211 amount of memory needed for compression and decompression. 211 amount of memory needed for compression and decompression.
212 The flags -1 through -9 specify the block size to be 212 The flags 1 through 9 specify the block size to be
213 100,000 bytes through 900,000 bytes (the default) respec 213 100,000 bytes through 900,000 bytes (the default) respec
214 tively. At decompression time, the block size used for 214 tively. At decompression time, the block size used for
215 compression is read from the header of the compressed 215 compression is read from the header of the compressed
216 file, and _b_u_n_z_i_p_2 then allocates itself just enough memory 216 file, and _b_u_n_z_i_p_2 then allocates itself just enough memory
217 to decompress the file. Since block sizes are stored in 217 to decompress the file. Since block sizes are stored in
218 compressed files, it follows that the flags -1 to -9 are 218 compressed files, it follows that the flags 1 to 9 are
219 irrelevant to and so ignored during decompression. 219 irrelevant to and so ignored during decompression.
220 220
221 Compression and decompression requirements, in bytes, can 221 Compression and decompression requirements, in bytes, can
@@ -238,21 +238,21 @@ MMEEMMOORRYY MMAANNAAGGEEMMEENNTT
238 _b_u_n_z_i_p_2 will require about 3700 kbytes to decompress. To 238 _b_u_n_z_i_p_2 will require about 3700 kbytes to decompress. To
239 support decompression of any file on a 4 megabyte machine, 239 support decompression of any file on a 4 megabyte machine,
240 _b_u_n_z_i_p_2 has an option to decompress using approximately 240 _b_u_n_z_i_p_2 has an option to decompress using approximately
241 half this amount of memory, about 2300 kbytes. Decompres 241 half this amount of memory, about 2300 kbytes. Decompres
242 sion speed is also halved, so you should use this option 242 sion speed is also halved, so you should use this option
243 only where necessary. The relevant flag is -s. 243 only where necessary. The relevant flag is s.
244 244
245 In general, try and use the largest block size memory con 245 In general, try and use the largest block size memory con
246 straints allow, since that maximises the compression 246 straints allow, since that maximises the compression
247 achieved. Compression and decompression speed are virtu 247 achieved. Compression and decompression speed are virtu
248 ally unaffected by block size. 248 ally unaffected by block size.
249 249
250 Another significant point applies to files which fit in a 250 Another significant point applies to files which fit in a
251 single block -- that means most files you'd encounter 251 single block ‐‐ that means most files youd encounter
252 using a large block size. The amount of real memory 252 using a large block size. The amount of real memory
253 touched is proportional to the size of the file, since the 253 touched is proportional to the size of the file, since the
254 file is smaller than a block. For example, compressing a 254 file is smaller than a block. For example, compressing a
255 file 20,000 bytes long with the flag -9 will cause the 255 file 20,000 bytes long with the flag 9 will cause the
256 compressor to allocate around 7600k of memory, but only 256 compressor to allocate around 7600k of memory, but only
257 touch 400k + 20000 * 8 = 560 kbytes of it. Similarly, the 257 touch 400k + 20000 * 8 = 560 kbytes of it. Similarly, the
258 decompressor will allocate 3700k but only touch 100k + 258 decompressor will allocate 3700k but only touch 100k +
@@ -260,59 +260,59 @@ MMEEMMOORRYY MMAANNAAGGEEMMEENNTT
260 260
261 Here is a table which summarises the maximum memory usage 261 Here is a table which summarises the maximum memory usage
262 for different block sizes. Also recorded is the total 262 for different block sizes. Also recorded is the total
263 compressed size for 14 files of the Calgary Text Compres 263 compressed size for 14 files of the Calgary Text Compres
264 sion Corpus totalling 3,141,622 bytes. This column gives 264 sion Corpus totalling 3,141,622 bytes. This column gives
265 some feel for how compression varies with block size. 265 some feel for how compression varies with block size.
266 These figures tend to understate the advantage of larger 266 These figures tend to understate the advantage of larger
267 block sizes for larger files, since the Corpus is domi 267 block sizes for larger files, since the Corpus is domi
268 nated by smaller files. 268 nated by smaller files.
269 269
270 Compress Decompress Decompress Corpus 270 Compress Decompress Decompress Corpus
271 Flag usage usage -s usage Size 271 Flag usage usage s usage Size
272 272
273 -1 1200k 500k 350k 914704 273 1 1200k 500k 350k 914704
274 -2 2000k 900k 600k 877703 274 2 2000k 900k 600k 877703
275 -3 2800k 1300k 850k 860338 275 3 2800k 1300k 850k 860338
276 -4 3600k 1700k 1100k 846899 276 4 3600k 1700k 1100k 846899
277 -5 4400k 2100k 1350k 845160 277 5 4400k 2100k 1350k 845160
278 -6 5200k 2500k 1600k 838626 278 6 5200k 2500k 1600k 838626
279 -7 6100k 2900k 1850k 834096 279 7 6100k 2900k 1850k 834096
280 -8 6800k 3300k 2100k 828642 280 8 6800k 3300k 2100k 828642
281 -9 7600k 3700k 2350k 828642 281 9 7600k 3700k 2350k 828642
282 282
283 283
284RREECCOOVVEERRIINNGG DDAATTAA FFRROOMM DDAAMMAAGGEEDD FFIILLEESS 284RREECCOOVVEERRIINNGG DDAATTAA FFRROOMM DDAAMMAAGGEEDD FFIILLEESS
285 _b_z_i_p_2 compresses files in blocks, usually 900kbytes long. 285 _b_z_i_p_2 compresses files in blocks, usually 900kbytes long.
286 Each block is handled independently. If a media or trans 286 Each block is handled independently. If a media or trans
287 mission error causes a multi-block .bz2 file to become 287 mission error causes a multiblock .bz2 file to become
288 damaged, it may be possible to recover data from the 288 damaged, it may be possible to recover data from the
289 undamaged blocks in the file. 289 undamaged blocks in the file.
290 290
291 The compressed representation of each block is delimited 291 The compressed representation of each block is delimited
292 by a 48-bit pattern, which makes it possible to find the 292 by a 48bit pattern, which makes it possible to find the
293 block boundaries with reasonable certainty. Each block 293 block boundaries with reasonable certainty. Each block
294 also carries its own 32-bit CRC, so damaged blocks can be 294 also carries its own 32bit CRC, so damaged blocks can be
295 distinguished from undamaged ones. 295 distinguished from undamaged ones.
296 296
297 _b_z_i_p_2_r_e_c_o_v_e_r is a simple program whose purpose is to 297 _b_z_i_p_2_r_e_c_o_v_e_r is a simple program whose purpose is to
298 search for blocks in .bz2 files, and write each block out 298 search for blocks in .bz2 files, and write each block out
299 into its own .bz2 file. You can then use _b_z_i_p_2 -t to test 299 into its own .bz2 file. You can then use _b_z_i_p_2 t to test
300 the integrity of the resulting files, and decompress those 300 the integrity of the resulting files, and decompress those
301 which are undamaged. 301 which are undamaged.
302 302
303 _b_z_i_p_2_r_e_c_o_v_e_r takes a single argument, the name of the dam 303 _b_z_i_p_2_r_e_c_o_v_e_r takes a single argument, the name of the dam
304 aged file, and writes a number of files 304 aged file, and writes a number of files
305 "rec00001file.bz2", "rec00002file.bz2", etc, containing 305 "rec00001file.bz2", "rec00002file.bz2", etc, containing
306 the extracted blocks. The output filenames are 306 the extracted blocks. The output filenames are
307 designed so that the use of wildcards in subsequent pro 307 designed so that the use of wildcards in subsequent pro
308 cessing -- for example, "bzip2 -dc rec*file.bz2 > recov 308 cessing ‐‐ for example, "bzip2 dc rec*file.bz2 > recov
309 ered_data" -- processes the files in the correct order. 309 ered_data" ‐‐ processes the files in the correct order.
310 310
311 _b_z_i_p_2_r_e_c_o_v_e_r should be of most use dealing with large .bz2 311 _b_z_i_p_2_r_e_c_o_v_e_r should be of most use dealing with large .bz2
312 files, as these will contain many blocks. It is clearly 312 files, as these will contain many blocks. It is clearly
313 futile to use it on damaged single-block files, since a 313 futile to use it on damaged singleblock files, since a
314 damaged block cannot be recovered. If you wish to min 314 damaged block cannot be recovered. If you wish to min
315 imise any potential data loss through media or transmis 315 imise any potential data loss through media or transmis
316 sion errors, you might consider compressing with a smaller 316 sion errors, you might consider compressing with a smaller
317 block size. 317 block size.
318 318
@@ -324,21 +324,21 @@ PPEERRFFOORRMMAANNCCEE NNOOTTEESS
324 ..." (repeated several hundred times) may compress more 324 ..." (repeated several hundred times) may compress more
325 slowly than normal. Versions 0.9.5 and above fare much 325 slowly than normal. Versions 0.9.5 and above fare much
326 better than previous versions in this respect. The ratio 326 better than previous versions in this respect. The ratio
327 between worst-case and average-case compression time is in 327 between worstcase and averagecase compression time is in
328 the region of 10:1. For previous versions, this figure 328 the region of 10:1. For previous versions, this figure
329 was more like 100:1. You can use the -vvvv option to mon 329 was more like 100:1. You can use the vvvv option to mon
330 itor progress in great detail, if you want. 330 itor progress in great detail, if you want.
331 331
332 Decompression speed is unaffected by these phenomena. 332 Decompression speed is unaffected by these phenomena.
333 333
334 _b_z_i_p_2 usually allocates several megabytes of memory to 334 _b_z_i_p_2 usually allocates several megabytes of memory to
335 operate in, and then charges all over it in a fairly ran 335 operate in, and then charges all over it in a fairly ran
336 dom fashion. This means that performance, both for com 336 dom fashion. This means that performance, both for com
337 pressing and decompressing, is largely determined by the 337 pressing and decompressing, is largely determined by the
338 speed at which your machine can service cache misses. 338 speed at which your machine can service cache misses.
339 Because of this, small changes to the code to reduce the 339 Because of this, small changes to the code to reduce the
340 miss rate have been observed to give disproportionately 340 miss rate have been observed to give disproportionately
341 large performance improvements. I imagine _b_z_i_p_2 will per 341 large performance improvements. I imagine _b_z_i_p_2 will per
342 form best on machines with very large caches. 342 form best on machines with very large caches.
343 343
344 344
@@ -348,50 +348,51 @@ CCAAVVEEAATTSS
348 but the details of what the problem is sometimes seem 348 but the details of what the problem is sometimes seem
349 rather misleading. 349 rather misleading.
350 350
351 This manual page pertains to version 1.0.2 of _b_z_i_p_2_. Com 351 This manual page pertains to version 1.0.3 of _b_z_i_p_2_. Com
352 pressed data created by this version is entirely forwards 352 pressed data created by this version is entirely forwards
353 and backwards compatible with the previous public 353 and backwards compatible with the previous public
354 releases, versions 0.1pl2, 0.9.0, 0.9.5, 1.0.0 and 1.0.1, 354 releases, versions 0.1pl2, 0.9.0, 0.9.5, 1.0.0, 1.0.1 and
355 but with the following exception: 0.9.0 and above can cor 355 1.0.2, but with the following exception: 0.9.0 and above
356 rectly decompress multiple concatenated compressed files. 356 can correctly decompress multiple concatenated compressed
357 0.1pl2 cannot do this; it will stop after decompressing 357 files. 0.1pl2 cannot do this; it will stop after decom­
358 just the first file in the stream. 358 pressing just the first file in the stream.
359 359
360 _b_z_i_p_2_r_e_c_o_v_e_r versions prior to this one, 1.0.2, used 360 _b_z_i_p_2_r_e_c_o_v_e_r versions prior to 1.0.2 used 32‐bit integers
361 32-bit integers to represent bit positions in compressed 361 to represent bit positions in compressed files, so they
362 files, so it could not handle compressed files more than 362 could not handle compressed files more than 512 megabytes
363 512 megabytes long. Version 1.0.2 and above uses 64-bit 363 long. Versions 1.0.2 and above use 64bit ints on some
364 ints on some platforms which support them (GNU supported 364 platforms which support them (GNU supported targets, and
365 targets, and Windows). To establish whether or not 365 Windows). To establish whether or not bzip2recover was
366 bzip2recover was built with such a limitation, run it 366 built with such a limitation, run it without arguments.
367 without arguments. In any event you can build yourself an 367 In any event you can build yourself an unlimited version
368 unlimited version if you can recompile it with MaybeUInt64 368 if you can recompile it with MaybeUInt64 set to be an
369 set to be an unsigned 64-bit integer. 369 unsigned 64bit integer.
370 370
371 371
372 372
373 373
374AAUUTTHHOORR 374AAUUTTHHOORR
375 Julian Seward, jseward@acm.org. 375 Julian Seward, jsewardbzip.org.
376 376
377 http://sources.redhat.com/bzip2 377 http://www.bzip.org
378 378
379 The ideas embodied in _b_z_i_p_2 are due to (at least) the fol 379 The ideas embodied in _b_z_i_p_2 are due to (at least) the fol
380 lowing people: Michael Burrows and David Wheeler (for the 380 lowing people: Michael Burrows and David Wheeler (for the
381 block sorting transformation), David Wheeler (again, for 381 block sorting transformation), David Wheeler (again, for
382 the Huffman coder), Peter Fenwick (for the structured cod 382 the Huffman coder), Peter Fenwick (for the structured cod
383 ing model in the original _b_z_i_p_, and many refinements), and 383 ing model in the original _b_z_i_p_, and many refinements), and
384 Alistair Moffat, Radford Neal and Ian Witten (for the 384 Alistair Moffat, Radford Neal and Ian Witten (for the
385 arithmetic coder in the original _b_z_i_p_)_. I am much 385 arithmetic coder in the original _b_z_i_p_)_. I am much
386 indebted for their help, support and advice. See the man 386 indebted for their help, support and advice. See the man
387 ual in the source distribution for pointers to sources of 387 ual in the source distribution for pointers to sources of
388 documentation. Christian von Roques encouraged me to look 388 documentation. Christian von Roques encouraged me to look
389 for faster sorting algorithms, so as to speed up compres 389 for faster sorting algorithms, so as to speed up compres­
390 sion. Bela Lubkin encouraged me to improve the worst-case 390 sion. Bela Lubkin encouraged me to improve the worst‐case
391 compression performance. The bz* scripts are derived from 391 compression performance. Donna Robinson XMLised the docu­
392 those of GNU gzip. Many people sent patches, helped with 392 mentation. The bz* scripts are derived from those of GNU
393 portability problems, lent machines, gave advice and were 393 gzip. Many people sent patches, helped with portability
394 generally helpful. 394 problems, lent machines, gave advice and were generally
395 helpful.
395 396
396 397
397 398
diff --git a/bzip2.c b/bzip2.c
index 807f420..79f87a5 100644
--- a/bzip2.c
+++ b/bzip2.c
@@ -7,7 +7,7 @@
7 This file is a part of bzip2 and/or libbzip2, a program and 7 This file is a part of bzip2 and/or libbzip2, a program and
8 library for lossless, block-sorting data compression. 8 library for lossless, block-sorting data compression.
9 9
10 Copyright (C) 1996-2002 Julian R Seward. All rights reserved. 10 Copyright (C) 1996-2005 Julian R Seward. All rights reserved.
11 11
12 Redistribution and use in source and binary forms, with or without 12 Redistribution and use in source and binary forms, with or without
13 modification, are permitted provided that the following conditions 13 modification, are permitted provided that the following conditions
@@ -41,7 +41,7 @@
41 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 41 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
42 42
43 Julian Seward, Cambridge, UK. 43 Julian Seward, Cambridge, UK.
44 jseward@acm.org 44 jseward@bzip.org
45 bzip2/libbzip2 version 1.0 of 21 March 2000 45 bzip2/libbzip2 version 1.0 of 21 March 2000
46 46
47 This program is based on (at least) the work of: 47 This program is based on (at least) the work of:
@@ -525,6 +525,7 @@ Bool uncompressStream ( FILE *zStream, FILE *stream )
525 UChar obuf[5000]; 525 UChar obuf[5000];
526 UChar unused[BZ_MAX_UNUSED]; 526 UChar unused[BZ_MAX_UNUSED];
527 Int32 nUnused; 527 Int32 nUnused;
528 void* unusedTmpV;
528 UChar* unusedTmp; 529 UChar* unusedTmp;
529 530
530 nUnused = 0; 531 nUnused = 0;
@@ -554,9 +555,10 @@ Bool uncompressStream ( FILE *zStream, FILE *stream )
554 } 555 }
555 if (bzerr != BZ_STREAM_END) goto errhandler; 556 if (bzerr != BZ_STREAM_END) goto errhandler;
556 557
557 BZ2_bzReadGetUnused ( &bzerr, bzf, (void**)(&unusedTmp), &nUnused ); 558 BZ2_bzReadGetUnused ( &bzerr, bzf, &unusedTmpV, &nUnused );
558 if (bzerr != BZ_OK) panic ( "decompress:bzReadGetUnused" ); 559 if (bzerr != BZ_OK) panic ( "decompress:bzReadGetUnused" );
559 560
561 unusedTmp = (UChar*)unusedTmpV;
560 for (i = 0; i < nUnused; i++) unused[i] = unusedTmp[i]; 562 for (i = 0; i < nUnused; i++) unused[i] = unusedTmp[i];
561 563
562 BZ2_bzReadClose ( &bzerr, bzf ); 564 BZ2_bzReadClose ( &bzerr, bzf );
@@ -639,6 +641,7 @@ Bool testStream ( FILE *zStream )
639 UChar obuf[5000]; 641 UChar obuf[5000];
640 UChar unused[BZ_MAX_UNUSED]; 642 UChar unused[BZ_MAX_UNUSED];
641 Int32 nUnused; 643 Int32 nUnused;
644 void* unusedTmpV;
642 UChar* unusedTmp; 645 UChar* unusedTmp;
643 646
644 nUnused = 0; 647 nUnused = 0;
@@ -662,9 +665,10 @@ Bool testStream ( FILE *zStream )
662 } 665 }
663 if (bzerr != BZ_STREAM_END) goto errhandler; 666 if (bzerr != BZ_STREAM_END) goto errhandler;
664 667
665 BZ2_bzReadGetUnused ( &bzerr, bzf, (void**)(&unusedTmp), &nUnused ); 668 BZ2_bzReadGetUnused ( &bzerr, bzf, &unusedTmpV, &nUnused );
666 if (bzerr != BZ_OK) panic ( "test:bzReadGetUnused" ); 669 if (bzerr != BZ_OK) panic ( "test:bzReadGetUnused" );
667 670
671 unusedTmp = (UChar*)unusedTmpV;
668 for (i = 0; i < nUnused; i++) unused[i] = unusedTmp[i]; 672 for (i = 0; i < nUnused; i++) unused[i] = unusedTmp[i];
669 673
670 BZ2_bzReadClose ( &bzerr, bzf ); 674 BZ2_bzReadClose ( &bzerr, bzf );
@@ -828,7 +832,7 @@ void panic ( Char* s )
828 "\n%s: PANIC -- internal consistency error:\n" 832 "\n%s: PANIC -- internal consistency error:\n"
829 "\t%s\n" 833 "\t%s\n"
830 "\tThis is a BUG. Please report it to me at:\n" 834 "\tThis is a BUG. Please report it to me at:\n"
831 "\tjseward@acm.org\n", 835 "\tjseward@bzip.org\n",
832 progName, s ); 836 progName, s );
833 showFileNames(); 837 showFileNames();
834 cleanUpAndFail( 3 ); 838 cleanUpAndFail( 3 );
@@ -908,7 +912,7 @@ void mySIGSEGVorSIGBUScatcher ( IntNative n )
908 " The user's manual, Section 4.3, has more info on (1) and (2).\n" 912 " The user's manual, Section 4.3, has more info on (1) and (2).\n"
909 " \n" 913 " \n"
910 " If you suspect this is a bug in bzip2, or are unsure about (1)\n" 914 " If you suspect this is a bug in bzip2, or are unsure about (1)\n"
911 " or (2), feel free to report it to me at: jseward@acm.org.\n" 915 " or (2), feel free to report it to me at: jseward@bzip.org.\n"
912 " Section 4.3 of the user's manual describes the info a useful\n" 916 " Section 4.3 of the user's manual describes the info a useful\n"
913 " bug report should have. If the manual is available on your\n" 917 " bug report should have. If the manual is available on your\n"
914 " system, please try and read it before mailing me. If you don't\n" 918 " system, please try and read it before mailing me. If you don't\n"
@@ -931,7 +935,7 @@ void mySIGSEGVorSIGBUScatcher ( IntNative n )
931 " The user's manual, Section 4.3, has more info on (2) and (3).\n" 935 " The user's manual, Section 4.3, has more info on (2) and (3).\n"
932 " \n" 936 " \n"
933 " If you suspect this is a bug in bzip2, or are unsure about (2)\n" 937 " If you suspect this is a bug in bzip2, or are unsure about (2)\n"
934 " or (3), feel free to report it to me at: jseward@acm.org.\n" 938 " or (3), feel free to report it to me at: jseward@bzip.org.\n"
935 " Section 4.3 of the user's manual describes the info a useful\n" 939 " Section 4.3 of the user's manual describes the info a useful\n"
936 " bug report should have. If the manual is available on your\n" 940 " bug report should have. If the manual is available on your\n"
937 " system, please try and read it before mailing me. If you don't\n" 941 " system, please try and read it before mailing me. If you don't\n"
@@ -1674,7 +1678,7 @@ void license ( void )
1674 "bzip2, a block-sorting file compressor. " 1678 "bzip2, a block-sorting file compressor. "
1675 "Version %s.\n" 1679 "Version %s.\n"
1676 " \n" 1680 " \n"
1677 " Copyright (C) 1996-2002 by Julian Seward.\n" 1681 " Copyright (C) 1996-2005 by Julian Seward.\n"
1678 " \n" 1682 " \n"
1679 " This program is free software; you can redistribute it and/or modify\n" 1683 " This program is free software; you can redistribute it and/or modify\n"
1680 " it under the terms set out in the LICENSE file, which is included\n" 1684 " it under the terms set out in the LICENSE file, which is included\n"
diff --git a/bzip2.txt b/bzip2.txt
index 6afe358..bf895b6 100644
--- a/bzip2.txt
+++ b/bzip2.txt
@@ -1,6 +1,6 @@
1 1
2NAME 2NAME
3 bzip2, bunzip2 - a block-sorting file compressor, v1.0.2 3 bzip2, bunzip2 - a block-sorting file compressor, v1.0.3
4 bzcat - decompresses files to stdout 4 bzcat - decompresses files to stdout
5 bzip2recover - recovers data from damaged bzip2 files 5 bzip2recover - recovers data from damaged bzip2 files
6 6
@@ -17,20 +17,20 @@ DESCRIPTION
17 sorting text compression algorithm, and Huffman coding. 17 sorting text compression algorithm, and Huffman coding.
18 Compression is generally considerably better than that 18 Compression is generally considerably better than that
19 achieved by more conventional LZ77/LZ78-based compressors, 19 achieved by more conventional LZ77/LZ78-based compressors,
20 and approaches the performance of the PPM family of sta 20 and approaches the performance of the PPM family of sta-
21 tistical compressors. 21 tistical compressors.
22 22
23 The command-line options are deliberately very similar to 23 The command-line options are deliberately very similar to
24 those of GNU gzip, but they are not identical. 24 those of GNU gzip, but they are not identical.
25 25
26 bzip2 expects a list of file names to accompany the com 26 bzip2 expects a list of file names to accompany the com-
27 mand-line flags. Each file is replaced by a compressed 27 mand-line flags. Each file is replaced by a compressed
28 version of itself, with the name "original_name.bz2". 28 version of itself, with the name "original_name.bz2".
29 Each compressed file has the same modification date, per 29 Each compressed file has the same modification date, per-
30 missions, and, when possible, ownership as the correspond 30 missions, and, when possible, ownership as the correspond-
31 ing original, so that these properties can be correctly 31 ing original, so that these properties can be correctly
32 restored at decompression time. File name handling is 32 restored at decompression time. File name handling is
33 naive in the sense that there is no mechanism for preserv 33 naive in the sense that there is no mechanism for preserv-
34 ing original file names, permissions, ownerships or dates 34 ing original file names, permissions, ownerships or dates
35 in filesystems which lack these concepts, or have serious 35 in filesystems which lack these concepts, or have serious
36 file name length restrictions, such as MS-DOS. 36 file name length restrictions, such as MS-DOS.
@@ -61,23 +61,23 @@ DESCRIPTION
61 guess the name of the original file, and uses the original 61 guess the name of the original file, and uses the original
62 name with .out appended. 62 name with .out appended.
63 63
64 As with compression, supplying no filenames causes decom 64 As with compression, supplying no filenames causes decom-
65 pression from standard input to standard output. 65 pression from standard input to standard output.
66 66
67 bunzip2 will correctly decompress a file which is the con 67 bunzip2 will correctly decompress a file which is the con-
68 catenation of two or more compressed files. The result is 68 catenation of two or more compressed files. The result is
69 the concatenation of the corresponding uncompressed files. 69 the concatenation of the corresponding uncompressed files.
70 Integrity testing (-t) of concatenated compressed files is 70 Integrity testing (-t) of concatenated compressed files is
71 also supported. 71 also supported.
72 72
73 You can also compress or decompress files to the standard 73 You can also compress or decompress files to the standard
74 output by giving the -c flag. Multiple files may be com 74 output by giving the -c flag. Multiple files may be com-
75 pressed and decompressed like this. The resulting outputs 75 pressed and decompressed like this. The resulting outputs
76 are fed sequentially to stdout. Compression of multiple 76 are fed sequentially to stdout. Compression of multiple
77 files in this manner generates a stream containing multi 77 files in this manner generates a stream containing multi-
78 ple compressed file representations. Such a stream can be 78 ple compressed file representations. Such a stream can be
79 decompressed correctly only by bzip2 version 0.9.0 or 79 decompressed correctly only by bzip2 version 0.9.0 or
80 later. Earlier versions of bzip2 will stop after decom 80 later. Earlier versions of bzip2 will stop after decom-
81 pressing the first file in the stream. 81 pressing the first file in the stream.
82 82
83 bzcat (or bzip2 -dc) decompresses all specified files to 83 bzcat (or bzip2 -dc) decompresses all specified files to
@@ -98,7 +98,7 @@ DESCRIPTION
98 98
99 As a self-check for your protection, bzip2 uses 32-bit 99 As a self-check for your protection, bzip2 uses 32-bit
100 CRCs to make sure that the decompressed version of a file 100 CRCs to make sure that the decompressed version of a file
101 is identical to the original. This guards against corrup 101 is identical to the original. This guards against corrup-
102 tion of the compressed data, and against undetected bugs 102 tion of the compressed data, and against undetected bugs
103 in bzip2 (hopefully very unlikely). The chances of data 103 in bzip2 (hopefully very unlikely). The chances of data
104 corruption going undetected is microscopic, about one 104 corruption going undetected is microscopic, about one
@@ -171,7 +171,7 @@ OPTIONS
171 171
172 -v --verbose 172 -v --verbose
173 Verbose mode -- show the compression ratio for each 173 Verbose mode -- show the compression ratio for each
174 file processed. Further -v's increase the ver 174 file processed. Further -v's increase the ver-
175 bosity level, spewing out lots of information which 175 bosity level, spewing out lots of information which
176 is primarily of interest for diagnostic purposes. 176 is primarily of interest for diagnostic purposes.
177 177
@@ -184,19 +184,19 @@ OPTIONS
184 compressing. Has no effect when decompressing. 184 compressing. Has no effect when decompressing.
185 See MEMORY MANAGEMENT below. The --fast and --best 185 See MEMORY MANAGEMENT below. The --fast and --best
186 aliases are primarily for GNU gzip compatibility. 186 aliases are primarily for GNU gzip compatibility.
187 In particular, --fast doesn't make things signifi 187 In particular, --fast doesn't make things signifi-
188 cantly faster. And --best merely selects the 188 cantly faster. And --best merely selects the
189 default behaviour. 189 default behaviour.
190 190
191 -- Treats all subsequent arguments as file names, even 191 -- Treats all subsequent arguments as file names, even
192 if they start with a dash. This is so you can han 192 if they start with a dash. This is so you can han-
193 dle files with names beginning with a dash, for 193 dle files with names beginning with a dash, for
194 example: bzip2 -- -myfilename. 194 example: bzip2 -- -myfilename.
195 195
196 --repetitive-fast --repetitive-best 196 --repetitive-fast --repetitive-best
197 These flags are redundant in versions 0.9.5 and 197 These flags are redundant in versions 0.9.5 and
198 above. They provided some coarse control over the 198 above. They provided some coarse control over the
199 behaviour of the sorting algorithm in earlier ver 199 behaviour of the sorting algorithm in earlier ver-
200 sions, which was sometimes useful. 0.9.5 and above 200 sions, which was sometimes useful. 0.9.5 and above
201 have an improved algorithm which renders these 201 have an improved algorithm which renders these
202 flags irrelevant. 202 flags irrelevant.
@@ -207,7 +207,7 @@ MEMORY MANAGEMENT
207 affects both the compression ratio achieved, and the 207 affects both the compression ratio achieved, and the
208 amount of memory needed for compression and decompression. 208 amount of memory needed for compression and decompression.
209 The flags -1 through -9 specify the block size to be 209 The flags -1 through -9 specify the block size to be
210 100,000 bytes through 900,000 bytes (the default) respec 210 100,000 bytes through 900,000 bytes (the default) respec-
211 tively. At decompression time, the block size used for 211 tively. At decompression time, the block size used for
212 compression is read from the header of the compressed 212 compression is read from the header of the compressed
213 file, and bunzip2 then allocates itself just enough memory 213 file, and bunzip2 then allocates itself just enough memory
@@ -235,13 +235,13 @@ MEMORY MANAGEMENT
235 bunzip2 will require about 3700 kbytes to decompress. To 235 bunzip2 will require about 3700 kbytes to decompress. To
236 support decompression of any file on a 4 megabyte machine, 236 support decompression of any file on a 4 megabyte machine,
237 bunzip2 has an option to decompress using approximately 237 bunzip2 has an option to decompress using approximately
238 half this amount of memory, about 2300 kbytes. Decompres 238 half this amount of memory, about 2300 kbytes. Decompres-
239 sion speed is also halved, so you should use this option 239 sion speed is also halved, so you should use this option
240 only where necessary. The relevant flag is -s. 240 only where necessary. The relevant flag is -s.
241 241
242 In general, try and use the largest block size memory con 242 In general, try and use the largest block size memory con-
243 straints allow, since that maximises the compression 243 straints allow, since that maximises the compression
244 achieved. Compression and decompression speed are virtu 244 achieved. Compression and decompression speed are virtu-
245 ally unaffected by block size. 245 ally unaffected by block size.
246 246
247 Another significant point applies to files which fit in a 247 Another significant point applies to files which fit in a
@@ -257,11 +257,11 @@ MEMORY MANAGEMENT
257 257
258 Here is a table which summarises the maximum memory usage 258 Here is a table which summarises the maximum memory usage
259 for different block sizes. Also recorded is the total 259 for different block sizes. Also recorded is the total
260 compressed size for 14 files of the Calgary Text Compres 260 compressed size for 14 files of the Calgary Text Compres-
261 sion Corpus totalling 3,141,622 bytes. This column gives 261 sion Corpus totalling 3,141,622 bytes. This column gives
262 some feel for how compression varies with block size. 262 some feel for how compression varies with block size.
263 These figures tend to understate the advantage of larger 263 These figures tend to understate the advantage of larger
264 block sizes for larger files, since the Corpus is domi 264 block sizes for larger files, since the Corpus is domi-
265 nated by smaller files. 265 nated by smaller files.
266 266
267 Compress Decompress Decompress Corpus 267 Compress Decompress Decompress Corpus
@@ -280,7 +280,7 @@ MEMORY MANAGEMENT
280 280
281RECOVERING DATA FROM DAMAGED FILES 281RECOVERING DATA FROM DAMAGED FILES
282 bzip2 compresses files in blocks, usually 900kbytes long. 282 bzip2 compresses files in blocks, usually 900kbytes long.
283 Each block is handled independently. If a media or trans 283 Each block is handled independently. If a media or trans-
284 mission error causes a multi-block .bz2 file to become 284 mission error causes a multi-block .bz2 file to become
285 damaged, it may be possible to recover data from the 285 damaged, it may be possible to recover data from the
286 undamaged blocks in the file. 286 undamaged blocks in the file.
@@ -297,19 +297,19 @@ RECOVERING DATA FROM DAMAGED FILES
297 the integrity of the resulting files, and decompress those 297 the integrity of the resulting files, and decompress those
298 which are undamaged. 298 which are undamaged.
299 299
300 bzip2recover takes a single argument, the name of the dam 300 bzip2recover takes a single argument, the name of the dam-
301 aged file, and writes a number of files 301 aged file, and writes a number of files
302 "rec00001file.bz2", "rec00002file.bz2", etc, containing 302 "rec00001file.bz2", "rec00002file.bz2", etc, containing
303 the extracted blocks. The output filenames are 303 the extracted blocks. The output filenames are
304 designed so that the use of wildcards in subsequent pro 304 designed so that the use of wildcards in subsequent pro-
305 cessing -- for example, "bzip2 -dc rec*file.bz2 > recov 305 cessing -- for example, "bzip2 -dc rec*file.bz2 > recov-
306 ered_data" -- processes the files in the correct order. 306 ered_data" -- processes the files in the correct order.
307 307
308 bzip2recover should be of most use dealing with large .bz2 308 bzip2recover should be of most use dealing with large .bz2
309 files, as these will contain many blocks. It is clearly 309 files, as these will contain many blocks. It is clearly
310 futile to use it on damaged single-block files, since a 310 futile to use it on damaged single-block files, since a
311 damaged block cannot be recovered. If you wish to min 311 damaged block cannot be recovered. If you wish to min-
312 imise any potential data loss through media or transmis 312 imise any potential data loss through media or transmis-
313 sion errors, you might consider compressing with a smaller 313 sion errors, you might consider compressing with a smaller
314 block size. 314 block size.
315 315
@@ -323,19 +323,19 @@ PERFORMANCE NOTES
323 better than previous versions in this respect. The ratio 323 better than previous versions in this respect. The ratio
324 between worst-case and average-case compression time is in 324 between worst-case and average-case compression time is in
325 the region of 10:1. For previous versions, this figure 325 the region of 10:1. For previous versions, this figure
326 was more like 100:1. You can use the -vvvv option to mon 326 was more like 100:1. You can use the -vvvv option to mon-
327 itor progress in great detail, if you want. 327 itor progress in great detail, if you want.
328 328
329 Decompression speed is unaffected by these phenomena. 329 Decompression speed is unaffected by these phenomena.
330 330
331 bzip2 usually allocates several megabytes of memory to 331 bzip2 usually allocates several megabytes of memory to
332 operate in, and then charges all over it in a fairly ran 332 operate in, and then charges all over it in a fairly ran-
333 dom fashion. This means that performance, both for com 333 dom fashion. This means that performance, both for com-
334 pressing and decompressing, is largely determined by the 334 pressing and decompressing, is largely determined by the
335 speed at which your machine can service cache misses. 335 speed at which your machine can service cache misses.
336 Because of this, small changes to the code to reduce the 336 Because of this, small changes to the code to reduce the
337 miss rate have been observed to give disproportionately 337 miss rate have been observed to give disproportionately
338 large performance improvements. I imagine bzip2 will per 338 large performance improvements. I imagine bzip2 will per-
339 form best on machines with very large caches. 339 form best on machines with very large caches.
340 340
341 341
@@ -345,46 +345,47 @@ CAVEATS
345 but the details of what the problem is sometimes seem 345 but the details of what the problem is sometimes seem
346 rather misleading. 346 rather misleading.
347 347
348 This manual page pertains to version 1.0.2 of bzip2. Com 348 This manual page pertains to version 1.0.3 of bzip2. Com-
349 pressed data created by this version is entirely forwards 349 pressed data created by this version is entirely forwards
350 and backwards compatible with the previous public 350 and backwards compatible with the previous public
351 releases, versions 0.1pl2, 0.9.0, 0.9.5, 1.0.0 and 1.0.1, 351 releases, versions 0.1pl2, 0.9.0, 0.9.5, 1.0.0, 1.0.1 and
352 but with the following exception: 0.9.0 and above can cor 352 1.0.2, but with the following exception: 0.9.0 and above
353 rectly decompress multiple concatenated compressed files. 353 can correctly decompress multiple concatenated compressed
354 0.1pl2 cannot do this; it will stop after decompressing 354 files. 0.1pl2 cannot do this; it will stop after decom-
355 just the first file in the stream. 355 pressing just the first file in the stream.
356 356
357 bzip2recover versions prior to this one, 1.0.2, used 357 bzip2recover versions prior to 1.0.2 used 32-bit integers
358 32-bit integers to represent bit positions in compressed 358 to represent bit positions in compressed files, so they
359 files, so it could not handle compressed files more than 359 could not handle compressed files more than 512 megabytes
360 512 megabytes long. Version 1.0.2 and above uses 64-bit 360 long. Versions 1.0.2 and above use 64-bit ints on some
361 ints on some platforms which support them (GNU supported 361 platforms which support them (GNU supported targets, and
362 targets, and Windows). To establish whether or not 362 Windows). To establish whether or not bzip2recover was
363 bzip2recover was built with such a limitation, run it 363 built with such a limitation, run it without arguments.
364 without arguments. In any event you can build yourself an 364 In any event you can build yourself an unlimited version
365 unlimited version if you can recompile it with MaybeUInt64 365 if you can recompile it with MaybeUInt64 set to be an
366 set to be an unsigned 64-bit integer. 366 unsigned 64-bit integer.
367 367
368 368
369AUTHOR 369AUTHOR
370 Julian Seward, jseward@acm.org. 370 Julian Seward, jsewardbzip.org.
371 371
372 http://sources.redhat.com/bzip2 372 http://www.bzip.org
373 373
374 The ideas embodied in bzip2 are due to (at least) the fol 374 The ideas embodied in bzip2 are due to (at least) the fol-
375 lowing people: Michael Burrows and David Wheeler (for the 375 lowing people: Michael Burrows and David Wheeler (for the
376 block sorting transformation), David Wheeler (again, for 376 block sorting transformation), David Wheeler (again, for
377 the Huffman coder), Peter Fenwick (for the structured cod 377 the Huffman coder), Peter Fenwick (for the structured cod-
378 ing model in the original bzip, and many refinements), and 378 ing model in the original bzip, and many refinements), and
379 Alistair Moffat, Radford Neal and Ian Witten (for the 379 Alistair Moffat, Radford Neal and Ian Witten (for the
380 arithmetic coder in the original bzip). I am much 380 arithmetic coder in the original bzip). I am much
381 indebted for their help, support and advice. See the man 381 indebted for their help, support and advice. See the man-
382 ual in the source distribution for pointers to sources of 382 ual in the source distribution for pointers to sources of
383 documentation. Christian von Roques encouraged me to look 383 documentation. Christian von Roques encouraged me to look
384 for faster sorting algorithms, so as to speed up compres 384 for faster sorting algorithms, so as to speed up compres-
385 sion. Bela Lubkin encouraged me to improve the worst-case 385 sion. Bela Lubkin encouraged me to improve the worst-case
386 compression performance. The bz* scripts are derived from 386 compression performance. Donna Robinson XMLised the docu-
387 those of GNU gzip. Many people sent patches, helped with 387 mentation. The bz* scripts are derived from those of GNU
388 portability problems, lent machines, gave advice and were 388 gzip. Many people sent patches, helped with portability
389 generally helpful. 389 problems, lent machines, gave advice and were generally
390 helpful.
390 391
diff --git a/bzip2recover.c b/bzip2recover.c
index 286873b..5cd405d 100644
--- a/bzip2recover.c
+++ b/bzip2recover.c
@@ -7,9 +7,9 @@
7/*-- 7/*--
8 This program is bzip2recover, a program to attempt data 8 This program is bzip2recover, a program to attempt data
9 salvage from damaged files created by the accompanying 9 salvage from damaged files created by the accompanying
10 bzip2-1.0 program. 10 bzip2-1.0.3 program.
11 11
12 Copyright (C) 1996-2002 Julian R Seward. All rights reserved. 12 Copyright (C) 1996-2005 Julian R Seward. All rights reserved.
13 13
14 Redistribution and use in source and binary forms, with or without 14 Redistribution and use in source and binary forms, with or without
15 modification, are permitted provided that the following conditions 15 modification, are permitted provided that the following conditions
@@ -43,8 +43,8 @@
43 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 43 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
44 44
45 Julian Seward, Cambridge, UK. 45 Julian Seward, Cambridge, UK.
46 jseward@acm.org 46 jseward@bzip.org
47 bzip2/libbzip2 version 1.0 of 21 March 2000 47 bzip2/libbzip2 version 1.0.3 of 15 February 2005
48--*/ 48--*/
49 49
50/*-- 50/*--
@@ -345,7 +345,7 @@ Int32 main ( Int32 argc, Char** argv )
345 inFileName[0] = outFileName[0] = 0; 345 inFileName[0] = outFileName[0] = 0;
346 346
347 fprintf ( stderr, 347 fprintf ( stderr,
348 "bzip2recover 1.0.2: extracts blocks from damaged .bz2 files.\n" ); 348 "bzip2recover 1.0.3: extracts blocks from damaged .bz2 files.\n" );
349 349
350 if (argc != 2) { 350 if (argc != 2) {
351 fprintf ( stderr, "%s: usage is `%s damaged_file_name'.\n", 351 fprintf ( stderr, "%s: usage is `%s damaged_file_name'.\n",
@@ -374,7 +374,7 @@ Int32 main ( Int32 argc, Char** argv )
374 if (strlen(argv[1]) >= BZ_MAX_FILENAME-20) { 374 if (strlen(argv[1]) >= BZ_MAX_FILENAME-20) {
375 fprintf ( stderr, 375 fprintf ( stderr,
376 "%s: supplied filename is suspiciously (>= %d chars) long. Bye!\n", 376 "%s: supplied filename is suspiciously (>= %d chars) long. Bye!\n",
377 progName, strlen(argv[1]) ); 377 progName, (int)strlen(argv[1]) );
378 exit(1); 378 exit(1);
379 } 379 }
380 380
diff --git a/bzlib.c b/bzlib.c
index 7d1cb27..e9c1e87 100644
--- a/bzlib.c
+++ b/bzlib.c
@@ -8,7 +8,7 @@
8 This file is a part of bzip2 and/or libbzip2, a program and 8 This file is a part of bzip2 and/or libbzip2, a program and
9 library for lossless, block-sorting data compression. 9 library for lossless, block-sorting data compression.
10 10
11 Copyright (C) 1996-2002 Julian R Seward. All rights reserved. 11 Copyright (C) 1996-2005 Julian R Seward. All rights reserved.
12 12
13 Redistribution and use in source and binary forms, with or without 13 Redistribution and use in source and binary forms, with or without
14 modification, are permitted provided that the following conditions 14 modification, are permitted provided that the following conditions
@@ -42,7 +42,7 @@
42 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 42 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
43 43
44 Julian Seward, Cambridge, UK. 44 Julian Seward, Cambridge, UK.
45 jseward@acm.org 45 jseward@bzip.org
46 bzip2/libbzip2 version 1.0 of 21 March 2000 46 bzip2/libbzip2 version 1.0 of 21 March 2000
47 47
48 This program is based on (at least) the work of: 48 This program is based on (at least) the work of:
@@ -88,12 +88,12 @@ void BZ2_bz__AssertH__fail ( int errcode )
88 fprintf(stderr, 88 fprintf(stderr,
89 "\n\nbzip2/libbzip2: internal error number %d.\n" 89 "\n\nbzip2/libbzip2: internal error number %d.\n"
90 "This is a bug in bzip2/libbzip2, %s.\n" 90 "This is a bug in bzip2/libbzip2, %s.\n"
91 "Please report it to me at: jseward@acm.org. If this happened\n" 91 "Please report it to me at: jseward@bzip.org. If this happened\n"
92 "when you were using some program which uses libbzip2 as a\n" 92 "when you were using some program which uses libbzip2 as a\n"
93 "component, you should also report this bug to the author(s)\n" 93 "component, you should also report this bug to the author(s)\n"
94 "of that program. Please make an effort to report this bug;\n" 94 "of that program. Please make an effort to report this bug;\n"
95 "timely and accurate bug reports eventually lead to higher\n" 95 "timely and accurate bug reports eventually lead to higher\n"
96 "quality software. Thanks. Julian Seward, 30 December 2001.\n\n", 96 "quality software. Thanks. Julian Seward, 15 February 2005.\n\n",
97 errcode, 97 errcode,
98 BZ2_bzlibVersion() 98 BZ2_bzlibVersion()
99 ); 99 );
@@ -574,8 +574,11 @@ int BZ_API(BZ2_bzDecompressInit)
574 574
575 575
576/*---------------------------------------------------*/ 576/*---------------------------------------------------*/
577/* Return True iff data corruption is discovered.
578 Returns False if there is no problem.
579*/
577static 580static
578void unRLE_obuf_to_output_FAST ( DState* s ) 581Bool unRLE_obuf_to_output_FAST ( DState* s )
579{ 582{
580 UChar k1; 583 UChar k1;
581 584
@@ -584,7 +587,7 @@ void unRLE_obuf_to_output_FAST ( DState* s )
584 while (True) { 587 while (True) {
585 /* try to finish existing run */ 588 /* try to finish existing run */
586 while (True) { 589 while (True) {
587 if (s->strm->avail_out == 0) return; 590 if (s->strm->avail_out == 0) return False;
588 if (s->state_out_len == 0) break; 591 if (s->state_out_len == 0) break;
589 *( (UChar*)(s->strm->next_out) ) = s->state_out_ch; 592 *( (UChar*)(s->strm->next_out) ) = s->state_out_ch;
590 BZ_UPDATE_CRC ( s->calculatedBlockCRC, s->state_out_ch ); 593 BZ_UPDATE_CRC ( s->calculatedBlockCRC, s->state_out_ch );
@@ -594,10 +597,13 @@ void unRLE_obuf_to_output_FAST ( DState* s )
594 s->strm->total_out_lo32++; 597 s->strm->total_out_lo32++;
595 if (s->strm->total_out_lo32 == 0) s->strm->total_out_hi32++; 598 if (s->strm->total_out_lo32 == 0) s->strm->total_out_hi32++;
596 } 599 }
597 600
598 /* can a new run be started? */ 601 /* can a new run be started? */
599 if (s->nblock_used == s->save_nblock+1) return; 602 if (s->nblock_used == s->save_nblock+1) return False;
600 603
604 /* Only caused by corrupt data stream? */
605 if (s->nblock_used > s->save_nblock+1)
606 return True;
601 607
602 s->state_out_len = 1; 608 s->state_out_len = 1;
603 s->state_out_ch = s->k0; 609 s->state_out_ch = s->k0;
@@ -667,6 +673,10 @@ void unRLE_obuf_to_output_FAST ( DState* s )
667 cs_avail_out--; 673 cs_avail_out--;
668 } 674 }
669 } 675 }
676 /* Only caused by corrupt data stream? */
677 if (c_nblock_used > s_save_nblockPP)
678 return True;
679
670 /* can a new run be started? */ 680 /* can a new run be started? */
671 if (c_nblock_used == s_save_nblockPP) { 681 if (c_nblock_used == s_save_nblockPP) {
672 c_state_out_len = 0; goto return_notr; 682 c_state_out_len = 0; goto return_notr;
@@ -712,6 +722,7 @@ void unRLE_obuf_to_output_FAST ( DState* s )
712 s->strm->avail_out = cs_avail_out; 722 s->strm->avail_out = cs_avail_out;
713 /* end save */ 723 /* end save */
714 } 724 }
725 return False;
715} 726}
716 727
717 728
@@ -732,8 +743,11 @@ __inline__ Int32 BZ2_indexIntoF ( Int32 indx, Int32 *cftab )
732 743
733 744
734/*---------------------------------------------------*/ 745/*---------------------------------------------------*/
746/* Return True iff data corruption is discovered.
747 Returns False if there is no problem.
748*/
735static 749static
736void unRLE_obuf_to_output_SMALL ( DState* s ) 750Bool unRLE_obuf_to_output_SMALL ( DState* s )
737{ 751{
738 UChar k1; 752 UChar k1;
739 753
@@ -742,7 +756,7 @@ void unRLE_obuf_to_output_SMALL ( DState* s )
742 while (True) { 756 while (True) {
743 /* try to finish existing run */ 757 /* try to finish existing run */
744 while (True) { 758 while (True) {
745 if (s->strm->avail_out == 0) return; 759 if (s->strm->avail_out == 0) return False;
746 if (s->state_out_len == 0) break; 760 if (s->state_out_len == 0) break;
747 *( (UChar*)(s->strm->next_out) ) = s->state_out_ch; 761 *( (UChar*)(s->strm->next_out) ) = s->state_out_ch;
748 BZ_UPDATE_CRC ( s->calculatedBlockCRC, s->state_out_ch ); 762 BZ_UPDATE_CRC ( s->calculatedBlockCRC, s->state_out_ch );
@@ -754,8 +768,11 @@ void unRLE_obuf_to_output_SMALL ( DState* s )
754 } 768 }
755 769
756 /* can a new run be started? */ 770 /* can a new run be started? */
757 if (s->nblock_used == s->save_nblock+1) return; 771 if (s->nblock_used == s->save_nblock+1) return False;
758 772
773 /* Only caused by corrupt data stream? */
774 if (s->nblock_used > s->save_nblock+1)
775 return True;
759 776
760 s->state_out_len = 1; 777 s->state_out_len = 1;
761 s->state_out_ch = s->k0; 778 s->state_out_ch = s->k0;
@@ -788,7 +805,7 @@ void unRLE_obuf_to_output_SMALL ( DState* s )
788 while (True) { 805 while (True) {
789 /* try to finish existing run */ 806 /* try to finish existing run */
790 while (True) { 807 while (True) {
791 if (s->strm->avail_out == 0) return; 808 if (s->strm->avail_out == 0) return False;
792 if (s->state_out_len == 0) break; 809 if (s->state_out_len == 0) break;
793 *( (UChar*)(s->strm->next_out) ) = s->state_out_ch; 810 *( (UChar*)(s->strm->next_out) ) = s->state_out_ch;
794 BZ_UPDATE_CRC ( s->calculatedBlockCRC, s->state_out_ch ); 811 BZ_UPDATE_CRC ( s->calculatedBlockCRC, s->state_out_ch );
@@ -800,7 +817,11 @@ void unRLE_obuf_to_output_SMALL ( DState* s )
800 } 817 }
801 818
802 /* can a new run be started? */ 819 /* can a new run be started? */
803 if (s->nblock_used == s->save_nblock+1) return; 820 if (s->nblock_used == s->save_nblock+1) return False;
821
822 /* Only caused by corrupt data stream? */
823 if (s->nblock_used > s->save_nblock+1)
824 return True;
804 825
805 s->state_out_len = 1; 826 s->state_out_len = 1;
806 s->state_out_ch = s->k0; 827 s->state_out_ch = s->k0;
@@ -830,6 +851,7 @@ void unRLE_obuf_to_output_SMALL ( DState* s )
830/*---------------------------------------------------*/ 851/*---------------------------------------------------*/
831int BZ_API(BZ2_bzDecompress) ( bz_stream *strm ) 852int BZ_API(BZ2_bzDecompress) ( bz_stream *strm )
832{ 853{
854 Bool corrupt;
833 DState* s; 855 DState* s;
834 if (strm == NULL) return BZ_PARAM_ERROR; 856 if (strm == NULL) return BZ_PARAM_ERROR;
835 s = strm->state; 857 s = strm->state;
@@ -840,12 +862,13 @@ int BZ_API(BZ2_bzDecompress) ( bz_stream *strm )
840 if (s->state == BZ_X_IDLE) return BZ_SEQUENCE_ERROR; 862 if (s->state == BZ_X_IDLE) return BZ_SEQUENCE_ERROR;
841 if (s->state == BZ_X_OUTPUT) { 863 if (s->state == BZ_X_OUTPUT) {
842 if (s->smallDecompress) 864 if (s->smallDecompress)
843 unRLE_obuf_to_output_SMALL ( s ); else 865 corrupt = unRLE_obuf_to_output_SMALL ( s ); else
844 unRLE_obuf_to_output_FAST ( s ); 866 corrupt = unRLE_obuf_to_output_FAST ( s );
867 if (corrupt) return BZ_DATA_ERROR;
845 if (s->nblock_used == s->save_nblock+1 && s->state_out_len == 0) { 868 if (s->nblock_used == s->save_nblock+1 && s->state_out_len == 0) {
846 BZ_FINALISE_CRC ( s->calculatedBlockCRC ); 869 BZ_FINALISE_CRC ( s->calculatedBlockCRC );
847 if (s->verbosity >= 3) 870 if (s->verbosity >= 3)
848 VPrintf2 ( " {0x%x, 0x%x}", s->storedBlockCRC, 871 VPrintf2 ( " {0x%08x, 0x%08x}", s->storedBlockCRC,
849 s->calculatedBlockCRC ); 872 s->calculatedBlockCRC );
850 if (s->verbosity >= 2) VPrintf0 ( "]" ); 873 if (s->verbosity >= 2) VPrintf0 ( "]" );
851 if (s->calculatedBlockCRC != s->storedBlockCRC) 874 if (s->calculatedBlockCRC != s->storedBlockCRC)
@@ -863,7 +886,7 @@ int BZ_API(BZ2_bzDecompress) ( bz_stream *strm )
863 Int32 r = BZ2_decompress ( s ); 886 Int32 r = BZ2_decompress ( s );
864 if (r == BZ_STREAM_END) { 887 if (r == BZ_STREAM_END) {
865 if (s->verbosity >= 3) 888 if (s->verbosity >= 3)
866 VPrintf2 ( "\n combined CRCs: stored = 0x%x, computed = 0x%x", 889 VPrintf2 ( "\n combined CRCs: stored = 0x%08x, computed = 0x%08x",
867 s->storedCombinedCRC, s->calculatedCombinedCRC ); 890 s->storedCombinedCRC, s->calculatedCombinedCRC );
868 if (s->calculatedCombinedCRC != s->storedCombinedCRC) 891 if (s->calculatedCombinedCRC != s->storedCombinedCRC)
869 return BZ_DATA_ERROR; 892 return BZ_DATA_ERROR;
diff --git a/bzlib.h b/bzlib.h
index 9ac43a1..3237243 100644
--- a/bzlib.h
+++ b/bzlib.h
@@ -8,7 +8,7 @@
8 This file is a part of bzip2 and/or libbzip2, a program and 8 This file is a part of bzip2 and/or libbzip2, a program and
9 library for lossless, block-sorting data compression. 9 library for lossless, block-sorting data compression.
10 10
11 Copyright (C) 1996-2002 Julian R Seward. All rights reserved. 11 Copyright (C) 1996-2005 Julian R Seward. All rights reserved.
12 12
13 Redistribution and use in source and binary forms, with or without 13 Redistribution and use in source and binary forms, with or without
14 modification, are permitted provided that the following conditions 14 modification, are permitted provided that the following conditions
@@ -42,7 +42,7 @@
42 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 42 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
43 43
44 Julian Seward, Cambridge, UK. 44 Julian Seward, Cambridge, UK.
45 jseward@acm.org 45 jseward@bzip.org
46 bzip2/libbzip2 version 1.0 of 21 March 2000 46 bzip2/libbzip2 version 1.0 of 21 March 2000
47 47
48 This program is based on (at least) the work of: 48 This program is based on (at least) the work of:
@@ -110,8 +110,10 @@ typedef
110#define BZ_EXPORT 110#define BZ_EXPORT
111#endif 111#endif
112 112
113#ifndef BZ_NO_STDIO
113/* Need a definitition for FILE */ 114/* Need a definitition for FILE */
114#include <stdio.h> 115#include <stdio.h>
116#endif
115 117
116#ifdef _WIN32 118#ifdef _WIN32
117# include <windows.h> 119# include <windows.h>
diff --git a/bzlib_private.h b/bzlib_private.h
index ff973c3..ca76fe6 100644
--- a/bzlib_private.h
+++ b/bzlib_private.h
@@ -8,7 +8,7 @@
8 This file is a part of bzip2 and/or libbzip2, a program and 8 This file is a part of bzip2 and/or libbzip2, a program and
9 library for lossless, block-sorting data compression. 9 library for lossless, block-sorting data compression.
10 10
11 Copyright (C) 1996-2002 Julian R Seward. All rights reserved. 11 Copyright (C) 1996-2005 Julian R Seward. All rights reserved.
12 12
13 Redistribution and use in source and binary forms, with or without 13 Redistribution and use in source and binary forms, with or without
14 modification, are permitted provided that the following conditions 14 modification, are permitted provided that the following conditions
@@ -42,7 +42,7 @@
42 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 42 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
43 43
44 Julian Seward, Cambridge, UK. 44 Julian Seward, Cambridge, UK.
45 jseward@acm.org 45 jseward@bzip.org
46 bzip2/libbzip2 version 1.0 of 21 March 2000 46 bzip2/libbzip2 version 1.0 of 21 March 2000
47 47
48 This program is based on (at least) the work of: 48 This program is based on (at least) the work of:
@@ -76,7 +76,7 @@
76 76
77/*-- General stuff. --*/ 77/*-- General stuff. --*/
78 78
79#define BZ_VERSION "1.0.2, 30-Dec-2001" 79#define BZ_VERSION "1.0.3, 15-Feb-2005"
80 80
81typedef char Char; 81typedef char Char;
82typedef unsigned char Bool; 82typedef unsigned char Bool;
diff --git a/compress.c b/compress.c
index 56501c1..7e0c291 100644
--- a/compress.c
+++ b/compress.c
@@ -8,7 +8,7 @@
8 This file is a part of bzip2 and/or libbzip2, a program and 8 This file is a part of bzip2 and/or libbzip2, a program and
9 library for lossless, block-sorting data compression. 9 library for lossless, block-sorting data compression.
10 10
11 Copyright (C) 1996-2002 Julian R Seward. All rights reserved. 11 Copyright (C) 1996-2005 Julian R Seward. All rights reserved.
12 12
13 Redistribution and use in source and binary forms, with or without 13 Redistribution and use in source and binary forms, with or without
14 modification, are permitted provided that the following conditions 14 modification, are permitted provided that the following conditions
@@ -42,7 +42,7 @@
42 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 42 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
43 43
44 Julian Seward, Cambridge, UK. 44 Julian Seward, Cambridge, UK.
45 jseward@acm.org 45 jseward@bzip.org
46 bzip2/libbzip2 version 1.0 of 21 March 2000 46 bzip2/libbzip2 version 1.0 of 21 March 2000
47 47
48 This program is based on (at least) the work of: 48 This program is based on (at least) the work of:
@@ -488,9 +488,11 @@ void sendMTFValues ( EState* s )
488 /*-- 488 /*--
489 Recompute the tables based on the accumulated frequencies. 489 Recompute the tables based on the accumulated frequencies.
490 --*/ 490 --*/
491 /* maxLen was changed from 20 to 17 in bzip2-1.0.3. See
492 comment in huffman.c for details. */
491 for (t = 0; t < nGroups; t++) 493 for (t = 0; t < nGroups; t++)
492 BZ2_hbMakeCodeLengths ( &(s->len[t][0]), &(s->rfreq[t][0]), 494 BZ2_hbMakeCodeLengths ( &(s->len[t][0]), &(s->rfreq[t][0]),
493 alphaSize, 20 ); 495 alphaSize, 17 /*20*/ );
494 } 496 }
495 497
496 498
@@ -527,7 +529,7 @@ void sendMTFValues ( EState* s )
527 if (s->len[t][i] > maxLen) maxLen = s->len[t][i]; 529 if (s->len[t][i] > maxLen) maxLen = s->len[t][i];
528 if (s->len[t][i] < minLen) minLen = s->len[t][i]; 530 if (s->len[t][i] < minLen) minLen = s->len[t][i];
529 } 531 }
530 AssertH ( !(maxLen > 20), 3004 ); 532 AssertH ( !(maxLen > 17 /*20*/ ), 3004 );
531 AssertH ( !(minLen < 1), 3005 ); 533 AssertH ( !(minLen < 1), 3005 );
532 BZ2_hbAssignCodes ( &(s->code[t][0]), &(s->len[t][0]), 534 BZ2_hbAssignCodes ( &(s->code[t][0]), &(s->len[t][0]),
533 minLen, maxLen, alphaSize ); 535 minLen, maxLen, alphaSize );
@@ -651,8 +653,8 @@ void BZ2_compressBlock ( EState* s, Bool is_last_block )
651 if (s->blockNo > 1) s->numZ = 0; 653 if (s->blockNo > 1) s->numZ = 0;
652 654
653 if (s->verbosity >= 2) 655 if (s->verbosity >= 2)
654 VPrintf4( " block %d: crc = 0x%8x, " 656 VPrintf4( " block %d: crc = 0x%08x, "
655 "combined CRC = 0x%8x, size = %d\n", 657 "combined CRC = 0x%08x, size = %d\n",
656 s->blockNo, s->blockCRC, s->combinedCRC, s->nblock ); 658 s->blockNo, s->blockCRC, s->combinedCRC, s->nblock );
657 659
658 BZ2_blockSort ( s ); 660 BZ2_blockSort ( s );
@@ -703,7 +705,7 @@ void BZ2_compressBlock ( EState* s, Bool is_last_block )
703 bsPutUChar ( s, 0x50 ); bsPutUChar ( s, 0x90 ); 705 bsPutUChar ( s, 0x50 ); bsPutUChar ( s, 0x90 );
704 bsPutUInt32 ( s, s->combinedCRC ); 706 bsPutUInt32 ( s, s->combinedCRC );
705 if (s->verbosity >= 2) 707 if (s->verbosity >= 2)
706 VPrintf1( " final combined CRC = 0x%x\n ", s->combinedCRC ); 708 VPrintf1( " final combined CRC = 0x%08x\n ", s->combinedCRC );
707 bsFinishWrite ( s ); 709 bsFinishWrite ( s );
708 } 710 }
709} 711}
diff --git a/crctable.c b/crctable.c
index b16746a..b6dadfc 100644
--- a/crctable.c
+++ b/crctable.c
@@ -8,7 +8,7 @@
8 This file is a part of bzip2 and/or libbzip2, a program and 8 This file is a part of bzip2 and/or libbzip2, a program and
9 library for lossless, block-sorting data compression. 9 library for lossless, block-sorting data compression.
10 10
11 Copyright (C) 1996-2002 Julian R Seward. All rights reserved. 11 Copyright (C) 1996-2005 Julian R Seward. All rights reserved.
12 12
13 Redistribution and use in source and binary forms, with or without 13 Redistribution and use in source and binary forms, with or without
14 modification, are permitted provided that the following conditions 14 modification, are permitted provided that the following conditions
@@ -42,7 +42,7 @@
42 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 42 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
43 43
44 Julian Seward, Cambridge, UK. 44 Julian Seward, Cambridge, UK.
45 jseward@acm.org 45 jseward@bzip.org
46 bzip2/libbzip2 version 1.0 of 21 March 2000 46 bzip2/libbzip2 version 1.0 of 21 March 2000
47 47
48 This program is based on (at least) the work of: 48 This program is based on (at least) the work of:
diff --git a/decompress.c b/decompress.c
index e921347..81c3d2c 100644
--- a/decompress.c
+++ b/decompress.c
@@ -8,7 +8,7 @@
8 This file is a part of bzip2 and/or libbzip2, a program and 8 This file is a part of bzip2 and/or libbzip2, a program and
9 library for lossless, block-sorting data compression. 9 library for lossless, block-sorting data compression.
10 10
11 Copyright (C) 1996-2002 Julian R Seward. All rights reserved. 11 Copyright (C) 1996-2005 Julian R Seward. All rights reserved.
12 12
13 Redistribution and use in source and binary forms, with or without 13 Redistribution and use in source and binary forms, with or without
14 modification, are permitted provided that the following conditions 14 modification, are permitted provided that the following conditions
@@ -42,7 +42,7 @@
42 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 42 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
43 43
44 Julian Seward, Cambridge, UK. 44 Julian Seward, Cambridge, UK.
45 jseward@acm.org 45 jseward@bzip.org
46 bzip2/libbzip2 version 1.0 of 21 March 2000 46 bzip2/libbzip2 version 1.0 of 21 March 2000
47 47
48 This program is based on (at least) the work of: 48 This program is based on (at least) the work of:
@@ -524,17 +524,23 @@ Int32 BZ2_decompress ( DState* s )
524 if (s->origPtr < 0 || s->origPtr >= nblock) 524 if (s->origPtr < 0 || s->origPtr >= nblock)
525 RETURN(BZ_DATA_ERROR); 525 RETURN(BZ_DATA_ERROR);
526 526
527 /*-- Set up cftab to facilitate generation of T^(-1) --*/
528 s->cftab[0] = 0;
529 for (i = 1; i <= 256; i++) s->cftab[i] = s->unzftab[i-1];
530 for (i = 1; i <= 256; i++) s->cftab[i] += s->cftab[i-1];
531 for (i = 0; i <= 256; i++) {
532 if (s->cftab[i] < 0 || s->cftab[i] > nblock) {
533 /* s->cftab[i] can legitimately be == nblock */
534 RETURN(BZ_DATA_ERROR);
535 }
536 }
537
527 s->state_out_len = 0; 538 s->state_out_len = 0;
528 s->state_out_ch = 0; 539 s->state_out_ch = 0;
529 BZ_INITIALISE_CRC ( s->calculatedBlockCRC ); 540 BZ_INITIALISE_CRC ( s->calculatedBlockCRC );
530 s->state = BZ_X_OUTPUT; 541 s->state = BZ_X_OUTPUT;
531 if (s->verbosity >= 2) VPrintf0 ( "rt+rld" ); 542 if (s->verbosity >= 2) VPrintf0 ( "rt+rld" );
532 543
533 /*-- Set up cftab to facilitate generation of T^(-1) --*/
534 s->cftab[0] = 0;
535 for (i = 1; i <= 256; i++) s->cftab[i] = s->unzftab[i-1];
536 for (i = 1; i <= 256; i++) s->cftab[i] += s->cftab[i-1];
537
538 if (s->smallDecompress) { 544 if (s->smallDecompress) {
539 545
540 /*-- Make a copy of cftab, used in generation of T --*/ 546 /*-- Make a copy of cftab, used in generation of T --*/
diff --git a/entities.xml b/entities.xml
new file mode 100644
index 0000000..6d0975f
--- /dev/null
+++ b/entities.xml
@@ -0,0 +1,9 @@
1<!-- misc. strings -->
2<!ENTITY bz-url "http://www.bzip.org">
3<!ENTITY bz-email "jseward@bzip.org">
4<!ENTITY bz-lifespan "1996-2005">
5
6<!ENTITY bz-version "1.0.3">
7<!ENTITY bz-date "15 February 2005">
8
9<!ENTITY manual-title "bzip2 Manual">
diff --git a/format.pl b/format.pl
new file mode 100755
index 0000000..8ab47ac
--- /dev/null
+++ b/format.pl
@@ -0,0 +1,53 @@
1#!/usr/bin/perl -w
2use strict;
3
4# get command line values:
5if ( $#ARGV !=1 ) {
6 die "Usage: $0 xml_infile xml_outfile\n";
7}
8
9my $infile = shift;
10# check infile exists
11die "Can't find file \"$infile\""
12 unless -f $infile;
13# check we can read infile
14if (! -r $infile) {
15 die "Can't read input $infile\n";
16}
17# check we can open infile
18open( INFILE,"<$infile" ) or
19 die "Can't input $infile $!";
20
21#my $outfile = 'fmt-manual.xml';
22my $outfile = shift;
23#print "Infile: $infile, Outfile: $outfile\n";
24# check we can write to outfile
25open( OUTFILE,">$outfile" ) or
26 die "Can't output $outfile $! for writing";
27
28my ($prev, $curr, $str);
29$prev = ''; $curr = '';
30while ( <INFILE> ) {
31
32 print OUTFILE $prev;
33 $prev = $curr;
34 $curr = $_;
35 $str = '';
36
37 if ( $prev =~ /<programlisting>$|<screen>$/ ) {
38 chomp $prev;
39 $curr = join( '', $prev, "<![CDATA[", $curr );
40 $prev = '';
41 next;
42 }
43 elsif ( $curr =~ /<\/programlisting>|<\/screen>/ ) {
44 chomp $prev;
45 $curr = join( '', $prev, "]]>", $curr );
46 $prev = '';
47 next;
48 }
49}
50print OUTFILE $curr;
51close INFILE;
52close OUTFILE;
53exit;
diff --git a/huffman.c b/huffman.c
index 293095c..5bf190b 100644
--- a/huffman.c
+++ b/huffman.c
@@ -8,7 +8,7 @@
8 This file is a part of bzip2 and/or libbzip2, a program and 8 This file is a part of bzip2 and/or libbzip2, a program and
9 library for lossless, block-sorting data compression. 9 library for lossless, block-sorting data compression.
10 10
11 Copyright (C) 1996-2002 Julian R Seward. All rights reserved. 11 Copyright (C) 1996-2005 Julian R Seward. All rights reserved.
12 12
13 Redistribution and use in source and binary forms, with or without 13 Redistribution and use in source and binary forms, with or without
14 modification, are permitted provided that the following conditions 14 modification, are permitted provided that the following conditions
@@ -42,7 +42,7 @@
42 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 42 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
43 43
44 Julian Seward, Cambridge, UK. 44 Julian Seward, Cambridge, UK.
45 jseward@acm.org 45 jseward@bzip.org
46 bzip2/libbzip2 version 1.0 of 21 March 2000 46 bzip2/libbzip2 version 1.0 of 21 March 2000
47 47
48 This program is based on (at least) the work of: 48 This program is based on (at least) the work of:
@@ -162,7 +162,24 @@ void BZ2_hbMakeCodeLengths ( UChar *len,
162 162
163 if (! tooLong) break; 163 if (! tooLong) break;
164 164
165 for (i = 1; i < alphaSize; i++) { 165 /* 17 Oct 04: keep-going condition for the following loop used
166 to be 'i < alphaSize', which missed the last element,
167 theoretically leading to the possibility of the compressor
168 looping. However, this count-scaling step is only needed if
169 one of the generated Huffman code words is longer than
170 maxLen, which up to and including version 1.0.2 was 20 bits,
171 which is extremely unlikely. In version 1.0.3 maxLen was
172 changed to 17 bits, which has minimal effect on compression
173 ratio, but does mean this scaling step is used from time to
174 time, enough to verify that it works.
175
176 This means that bzip2-1.0.3 and later will only produce
177 Huffman codes with a maximum length of 17 bits. However, in
178 order to preserve backwards compatibility with bitstreams
179 produced by versions pre-1.0.3, the decompressor must still
180 handle lengths of up to 20. */
181
182 for (i = 1; i <= alphaSize; i++) {
166 j = weight[i] >> 8; 183 j = weight[i] >> 8;
167 j = 1 + (j / 2); 184 j = 1 + (j / 2);
168 weight[i] = j << 8; 185 weight[i] = j << 8;
diff --git a/manual.texi b/manual.texi
deleted file mode 100644
index 5bc27d5..0000000
--- a/manual.texi
+++ /dev/null
@@ -1,2243 +0,0 @@
1\input texinfo @c -*- Texinfo -*-
2@setfilename bzip2.info
3
4@ignore
5This file documents bzip2 version 1.0.2, and associated library
6libbzip2, written by Julian Seward (jseward@acm.org).
7
8Copyright (C) 1996-2002 Julian R Seward
9
10Permission is granted to make and distribute verbatim copies of
11this manual provided the copyright notice and this permission notice
12are preserved on all copies.
13
14Permission is granted to copy and distribute translations of this manual
15into another language, under the above conditions for verbatim copies.
16@end ignore
17
18@ifinfo
19@format
20START-INFO-DIR-ENTRY
21* Bzip2: (bzip2). A program and library for data compression.
22END-INFO-DIR-ENTRY
23@end format
24
25@end ifinfo
26
27@iftex
28@c @finalout
29@settitle bzip2 and libbzip2
30@titlepage
31@title bzip2 and libbzip2
32@subtitle a program and library for data compression
33@subtitle copyright (C) 1996-2002 Julian Seward
34@subtitle version 1.0.2 of 30 December 2001
35@author Julian Seward
36
37@end titlepage
38
39@parindent 0mm
40@parskip 2mm
41
42@end iftex
43@node Top,,, (dir)
44
45The following text is the License for this software. You should
46find it identical to that contained in the file LICENSE in the
47source distribution.
48
49@bf{------------------ START OF THE LICENSE ------------------}
50
51This program, @code{bzip2},
52and associated library @code{libbzip2}, are
53Copyright (C) 1996-2002 Julian R Seward. All rights reserved.
54
55Redistribution and use in source and binary forms, with or without
56modification, are permitted provided that the following conditions
57are met:
58@itemize @bullet
59@item
60 Redistributions of source code must retain the above copyright
61 notice, this list of conditions and the following disclaimer.
62@item
63 The origin of this software must not be misrepresented; you must
64 not claim that you wrote the original software. If you use this
65 software in a product, an acknowledgment in the product
66 documentation would be appreciated but is not required.
67@item
68 Altered source versions must be plainly marked as such, and must
69 not be misrepresented as being the original software.
70@item
71 The name of the author may not be used to endorse or promote
72 products derived from this software without specific prior written
73 permission.
74@end itemize
75THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS
76OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
77WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
78ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY
79DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
80DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE
81GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
82INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
83WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
84NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
85SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
86
87Julian Seward, Cambridge, UK.
88
89@code{jseward@@acm.org}
90
91@code{bzip2}/@code{libbzip2} version 1.0.2 of 30 December 2001.
92
93@bf{------------------ END OF THE LICENSE ------------------}
94
95Web sites:
96
97@code{http://sources.redhat.com/bzip2}
98
99@code{http://www.cacheprof.org}
100
101PATENTS: To the best of my knowledge, @code{bzip2} does not use any patented
102algorithms. However, I do not have the resources available to carry out
103a full patent search. Therefore I cannot give any guarantee of the
104above statement.
105
106
107
108
109
110
111
112@chapter Introduction
113
114@code{bzip2} compresses files using the Burrows-Wheeler
115block-sorting text compression algorithm, and Huffman coding.
116Compression is generally considerably better than that
117achieved by more conventional LZ77/LZ78-based compressors,
118and approaches the performance of the PPM family of statistical compressors.
119
120@code{bzip2} is built on top of @code{libbzip2}, a flexible library
121for handling compressed data in the @code{bzip2} format. This manual
122describes both how to use the program and
123how to work with the library interface. Most of the
124manual is devoted to this library, not the program,
125which is good news if your interest is only in the program.
126
127Chapter 2 describes how to use @code{bzip2}; this is the only part
128you need to read if you just want to know how to operate the program.
129Chapter 3 describes the programming interfaces in detail, and
130Chapter 4 records some miscellaneous notes which I thought
131ought to be recorded somewhere.
132
133
134@chapter How to use @code{bzip2}
135
136This chapter contains a copy of the @code{bzip2} man page,
137and nothing else.
138
139@quotation
140
141@unnumberedsubsubsec NAME
142@itemize
143@item @code{bzip2}, @code{bunzip2}
144- a block-sorting file compressor, v1.0.2
145@item @code{bzcat}
146- decompresses files to stdout
147@item @code{bzip2recover}
148- recovers data from damaged bzip2 files
149@end itemize
150
151@unnumberedsubsubsec SYNOPSIS
152@itemize
153@item @code{bzip2} [ -cdfkqstvzVL123456789 ] [ filenames ... ]
154@item @code{bunzip2} [ -fkvsVL ] [ filenames ... ]
155@item @code{bzcat} [ -s ] [ filenames ... ]
156@item @code{bzip2recover} filename
157@end itemize
158
159@unnumberedsubsubsec DESCRIPTION
160
161@code{bzip2} compresses files using the Burrows-Wheeler block sorting
162text compression algorithm, and Huffman coding. Compression is
163generally considerably better than that achieved by more conventional
164LZ77/LZ78-based compressors, and approaches the performance of the PPM
165family of statistical compressors.
166
167The command-line options are deliberately very similar to those of GNU
168@code{gzip}, but they are not identical.
169
170@code{bzip2} expects a list of file names to accompany the command-line
171flags. Each file is replaced by a compressed version of itself, with
172the name @code{original_name.bz2}. Each compressed file has the same
173modification date, permissions, and, when possible, ownership as the
174corresponding original, so that these properties can be correctly
175restored at decompression time. File name handling is naive in the
176sense that there is no mechanism for preserving original file names,
177permissions, ownerships or dates in filesystems which lack these
178concepts, or have serious file name length restrictions, such as MS-DOS.
179
180@code{bzip2} and @code{bunzip2} will by default not overwrite existing
181files. If you want this to happen, specify the @code{-f} flag.
182
183If no file names are specified, @code{bzip2} compresses from standard
184input to standard output. In this case, @code{bzip2} will decline to
185write compressed output to a terminal, as this would be entirely
186incomprehensible and therefore pointless.
187
188@code{bunzip2} (or @code{bzip2 -d}) decompresses all
189specified files. Files which were not created by @code{bzip2}
190will be detected and ignored, and a warning issued.
191@code{bzip2} attempts to guess the filename for the decompressed file
192from that of the compressed file as follows:
193@itemize
194@item @code{filename.bz2 } becomes @code{filename}
195@item @code{filename.bz } becomes @code{filename}
196@item @code{filename.tbz2} becomes @code{filename.tar}
197@item @code{filename.tbz } becomes @code{filename.tar}
198@item @code{anyothername } becomes @code{anyothername.out}
199@end itemize
200If the file does not end in one of the recognised endings,
201@code{.bz2}, @code{.bz},
202@code{.tbz2} or @code{.tbz}, @code{bzip2} complains that it cannot
203guess the name of the original file, and uses the original name
204with @code{.out} appended.
205
206As with compression, supplying no
207filenames causes decompression from standard input to standard output.
208
209@code{bunzip2} will correctly decompress a file which is the
210concatenation of two or more compressed files. The result is the
211concatenation of the corresponding uncompressed files. Integrity
212testing (@code{-t}) of concatenated compressed files is also supported.
213
214You can also compress or decompress files to the standard output by
215giving the @code{-c} flag. Multiple files may be compressed and
216decompressed like this. The resulting outputs are fed sequentially to
217stdout. Compression of multiple files in this manner generates a stream
218containing multiple compressed file representations. Such a stream
219can be decompressed correctly only by @code{bzip2} version 0.9.0 or
220later. Earlier versions of @code{bzip2} will stop after decompressing
221the first file in the stream.
222
223@code{bzcat} (or @code{bzip2 -dc}) decompresses all specified files to
224the standard output.
225
226@code{bzip2} will read arguments from the environment variables
227@code{BZIP2} and @code{BZIP}, in that order, and will process them
228before any arguments read from the command line. This gives a
229convenient way to supply default arguments.
230
231Compression is always performed, even if the compressed file is slightly
232larger than the original. Files of less than about one hundred bytes
233tend to get larger, since the compression mechanism has a constant
234overhead in the region of 50 bytes. Random data (including the output
235of most file compressors) is coded at about 8.05 bits per byte, giving
236an expansion of around 0.5%.
237
238As a self-check for your protection, @code{bzip2} uses 32-bit CRCs to
239make sure that the decompressed version of a file is identical to the
240original. This guards against corruption of the compressed data, and
241against undetected bugs in @code{bzip2} (hopefully very unlikely). The
242chances of data corruption going undetected is microscopic, about one
243chance in four billion for each file processed. Be aware, though, that
244the check occurs upon decompression, so it can only tell you that
245something is wrong. It can't help you recover the original uncompressed
246data. You can use @code{bzip2recover} to try to recover data from
247damaged files.
248
249Return values: 0 for a normal exit, 1 for environmental problems (file
250not found, invalid flags, I/O errors, &c), 2 to indicate a corrupt
251compressed file, 3 for an internal consistency error (eg, bug) which
252caused @code{bzip2} to panic.
253
254
255@unnumberedsubsubsec OPTIONS
256@table @code
257@item -c --stdout
258Compress or decompress to standard output.
259@item -d --decompress
260Force decompression. @code{bzip2}, @code{bunzip2} and @code{bzcat} are
261really the same program, and the decision about what actions to take is
262done on the basis of which name is used. This flag overrides that
263mechanism, and forces bzip2 to decompress.
264@item -z --compress
265The complement to @code{-d}: forces compression, regardless of the
266invokation name.
267@item -t --test
268Check integrity of the specified file(s), but don't decompress them.
269This really performs a trial decompression and throws away the result.
270@item -f --force
271Force overwrite of output files. Normally, @code{bzip2} will not overwrite
272existing output files. Also forces @code{bzip2} to break hard links
273to files, which it otherwise wouldn't do.
274
275@code{bzip2} normally declines to decompress files which don't have the
276correct magic header bytes. If forced (@code{-f}), however, it will
277pass such files through unmodified. This is how GNU @code{gzip}
278behaves.
279@item -k --keep
280Keep (don't delete) input files during compression
281or decompression.
282@item -s --small
283Reduce memory usage, for compression, decompression and testing. Files
284are decompressed and tested using a modified algorithm which only
285requires 2.5 bytes per block byte. This means any file can be
286decompressed in 2300k of memory, albeit at about half the normal speed.
287
288During compression, @code{-s} selects a block size of 200k, which limits
289memory use to around the same figure, at the expense of your compression
290ratio. In short, if your machine is low on memory (8 megabytes or
291less), use -s for everything. See MEMORY MANAGEMENT below.
292@item -q --quiet
293Suppress non-essential warning messages. Messages pertaining to
294I/O errors and other critical events will not be suppressed.
295@item -v --verbose
296Verbose mode -- show the compression ratio for each file processed.
297Further @code{-v}'s increase the verbosity level, spewing out lots of
298information which is primarily of interest for diagnostic purposes.
299@item -L --license -V --version
300Display the software version, license terms and conditions.
301@item -1 (or --fast) to -9 (or --best)
302Set the block size to 100 k, 200 k .. 900 k when compressing. Has no
303effect when decompressing. See MEMORY MANAGEMENT below.
304The @code{--fast} and @code{--best} aliases are primarily for GNU
305@code{gzip} compatibility. In particular, @code{--fast} doesn't make
306things significantly faster. And @code{--best} merely selects the
307default behaviour.
308@item --
309Treats all subsequent arguments as file names, even if they start
310with a dash. This is so you can handle files with names beginning
311with a dash, for example: @code{bzip2 -- -myfilename}.
312@item --repetitive-fast
313@item --repetitive-best
314These flags are redundant in versions 0.9.5 and above. They provided
315some coarse control over the behaviour of the sorting algorithm in
316earlier versions, which was sometimes useful. 0.9.5 and above have an
317improved algorithm which renders these flags irrelevant.
318@end table
319
320
321@unnumberedsubsubsec MEMORY MANAGEMENT
322
323@code{bzip2} compresses large files in blocks. The block size affects
324both the compression ratio achieved, and the amount of memory needed for
325compression and decompression. The flags @code{-1} through @code{-9}
326specify the block size to be 100,000 bytes through 900,000 bytes (the
327default) respectively. At decompression time, the block size used for
328compression is read from the header of the compressed file, and
329@code{bunzip2} then allocates itself just enough memory to decompress
330the file. Since block sizes are stored in compressed files, it follows
331that the flags @code{-1} to @code{-9} are irrelevant to and so ignored
332during decompression.
333
334Compression and decompression requirements, in bytes, can be estimated
335as:
336@example
337 Compression: 400k + ( 8 x block size )
338
339 Decompression: 100k + ( 4 x block size ), or
340 100k + ( 2.5 x block size )
341@end example
342Larger block sizes give rapidly diminishing marginal returns. Most of
343the compression comes from the first two or three hundred k of block
344size, a fact worth bearing in mind when using @code{bzip2} on small machines.
345It is also important to appreciate that the decompression memory
346requirement is set at compression time by the choice of block size.
347
348For files compressed with the default 900k block size, @code{bunzip2}
349will require about 3700 kbytes to decompress. To support decompression
350of any file on a 4 megabyte machine, @code{bunzip2} has an option to
351decompress using approximately half this amount of memory, about 2300
352kbytes. Decompression speed is also halved, so you should use this
353option only where necessary. The relevant flag is @code{-s}.
354
355In general, try and use the largest block size memory constraints allow,
356since that maximises the compression achieved. Compression and
357decompression speed are virtually unaffected by block size.
358
359Another significant point applies to files which fit in a single block
360-- that means most files you'd encounter using a large block size. The
361amount of real memory touched is proportional to the size of the file,
362since the file is smaller than a block. For example, compressing a file
36320,000 bytes long with the flag @code{-9} will cause the compressor to
364allocate around 7600k of memory, but only touch 400k + 20000 * 8 = 560
365kbytes of it. Similarly, the decompressor will allocate 3700k but only
366touch 100k + 20000 * 4 = 180 kbytes.
367
368Here is a table which summarises the maximum memory usage for different
369block sizes. Also recorded is the total compressed size for 14 files of
370the Calgary Text Compression Corpus totalling 3,141,622 bytes. This
371column gives some feel for how compression varies with block size.
372These figures tend to understate the advantage of larger block sizes for
373larger files, since the Corpus is dominated by smaller files.
374@example
375 Compress Decompress Decompress Corpus
376 Flag usage usage -s usage Size
377
378 -1 1200k 500k 350k 914704
379 -2 2000k 900k 600k 877703
380 -3 2800k 1300k 850k 860338
381 -4 3600k 1700k 1100k 846899
382 -5 4400k 2100k 1350k 845160
383 -6 5200k 2500k 1600k 838626
384 -7 6100k 2900k 1850k 834096
385 -8 6800k 3300k 2100k 828642
386 -9 7600k 3700k 2350k 828642
387@end example
388
389@unnumberedsubsubsec RECOVERING DATA FROM DAMAGED FILES
390
391@code{bzip2} compresses files in blocks, usually 900kbytes long. Each
392block is handled independently. If a media or transmission error causes
393a multi-block @code{.bz2} file to become damaged, it may be possible to
394recover data from the undamaged blocks in the file.
395
396The compressed representation of each block is delimited by a 48-bit
397pattern, which makes it possible to find the block boundaries with
398reasonable certainty. Each block also carries its own 32-bit CRC, so
399damaged blocks can be distinguished from undamaged ones.
400
401@code{bzip2recover} is a simple program whose purpose is to search for
402blocks in @code{.bz2} files, and write each block out into its own
403@code{.bz2} file. You can then use @code{bzip2 -t} to test the
404integrity of the resulting files, and decompress those which are
405undamaged.
406
407@code{bzip2recover}
408takes a single argument, the name of the damaged file, and writes a
409number of files @code{rec00001file.bz2}, @code{rec00002file.bz2}, etc,
410containing the extracted blocks. The output filenames are designed so
411that the use of wildcards in subsequent processing -- for example,
412@code{bzip2 -dc rec*file.bz2 > recovered_data} -- processes the files in
413the correct order.
414
415@code{bzip2recover} should be of most use dealing with large @code{.bz2}
416files, as these will contain many blocks. It is clearly futile to use
417it on damaged single-block files, since a damaged block cannot be
418recovered. If you wish to minimise any potential data loss through
419media or transmission errors, you might consider compressing with a
420smaller block size.
421
422
423@unnumberedsubsubsec PERFORMANCE NOTES
424
425The sorting phase of compression gathers together similar strings in the
426file. Because of this, files containing very long runs of repeated
427symbols, like "aabaabaabaab ..." (repeated several hundred times) may
428compress more slowly than normal. Versions 0.9.5 and above fare much
429better than previous versions in this respect. The ratio between
430worst-case and average-case compression time is in the region of 10:1.
431For previous versions, this figure was more like 100:1. You can use the
432@code{-vvvv} option to monitor progress in great detail, if you want.
433
434Decompression speed is unaffected by these phenomena.
435
436@code{bzip2} usually allocates several megabytes of memory to operate
437in, and then charges all over it in a fairly random fashion. This means
438that performance, both for compressing and decompressing, is largely
439determined by the speed at which your machine can service cache misses.
440Because of this, small changes to the code to reduce the miss rate have
441been observed to give disproportionately large performance improvements.
442I imagine @code{bzip2} will perform best on machines with very large
443caches.
444
445
446@unnumberedsubsubsec CAVEATS
447
448I/O error messages are not as helpful as they could be. @code{bzip2}
449tries hard to detect I/O errors and exit cleanly, but the details of
450what the problem is sometimes seem rather misleading.
451
452This manual page pertains to version 1.0.2 of @code{bzip2}. Compressed
453data created by this version is entirely forwards and backwards
454compatible with the previous public releases, versions 0.1pl2, 0.9.0,
4550.9.5, 1.0.0 and 1.0.1, but with the following exception: 0.9.0 and
456above can correctly decompress multiple concatenated compressed files.
4570.1pl2 cannot do this; it will stop after decompressing just the first
458file in the stream.
459
460@code{bzip2recover} versions prior to this one, 1.0.2, used 32-bit
461integers to represent bit positions in compressed files, so it could not
462handle compressed files more than 512 megabytes long. Version 1.0.2 and
463above uses 64-bit ints on some platforms which support them (GNU
464supported targets, and Windows). To establish whether or not
465@code{bzip2recover} was built with such a limitation, run it without
466arguments. In any event you can build yourself an unlimited version if
467you can recompile it with @code{MaybeUInt64} set to be an unsigned
46864-bit integer.
469
470
471
472@unnumberedsubsubsec AUTHOR
473Julian Seward, @code{jseward@@acm.org}.
474
475@code{http://sources.redhat.com/bzip2}
476
477The ideas embodied in @code{bzip2} are due to (at least) the following
478people: Michael Burrows and David Wheeler (for the block sorting
479transformation), David Wheeler (again, for the Huffman coder), Peter
480Fenwick (for the structured coding model in the original @code{bzip},
481and many refinements), and Alistair Moffat, Radford Neal and Ian Witten
482(for the arithmetic coder in the original @code{bzip}). I am much
483indebted for their help, support and advice. See the manual in the
484source distribution for pointers to sources of documentation. Christian
485von Roques encouraged me to look for faster sorting algorithms, so as to
486speed up compression. Bela Lubkin encouraged me to improve the
487worst-case compression performance. The @code{bz*} scripts are derived
488from those of GNU @code{gzip}. Many people sent patches, helped with
489portability problems, lent machines, gave advice and were generally
490helpful.
491
492@end quotation
493
494
495
496
497@chapter Programming with @code{libbzip2}
498
499This chapter describes the programming interface to @code{libbzip2}.
500
501For general background information, particularly about memory
502use and performance aspects, you'd be well advised to read Chapter 2
503as well.
504
505@section Top-level structure
506
507@code{libbzip2} is a flexible library for compressing and decompressing
508data in the @code{bzip2} data format. Although packaged as a single
509entity, it helps to regard the library as three separate parts: the low
510level interface, and the high level interface, and some utility
511functions.
512
513The structure of @code{libbzip2}'s interfaces is similar to
514that of Jean-loup Gailly's and Mark Adler's excellent @code{zlib}
515library.
516
517All externally visible symbols have names beginning @code{BZ2_}.
518This is new in version 1.0. The intention is to minimise pollution
519of the namespaces of library clients.
520
521@subsection Low-level summary
522
523This interface provides services for compressing and decompressing
524data in memory. There's no provision for dealing with files, streams
525or any other I/O mechanisms, just straight memory-to-memory work.
526In fact, this part of the library can be compiled without inclusion
527of @code{stdio.h}, which may be helpful for embedded applications.
528
529The low-level part of the library has no global variables and
530is therefore thread-safe.
531
532Six routines make up the low level interface:
533@code{BZ2_bzCompressInit}, @code{BZ2_bzCompress}, and @* @code{BZ2_bzCompressEnd}
534for compression,
535and a corresponding trio @code{BZ2_bzDecompressInit}, @* @code{BZ2_bzDecompress}
536and @code{BZ2_bzDecompressEnd} for decompression.
537The @code{*Init} functions allocate
538memory for compression/decompression and do other
539initialisations, whilst the @code{*End} functions close down operations
540and release memory.
541
542The real work is done by @code{BZ2_bzCompress} and @code{BZ2_bzDecompress}.
543These compress and decompress data from a user-supplied input buffer
544to a user-supplied output buffer. These buffers can be any size;
545arbitrary quantities of data are handled by making repeated calls
546to these functions. This is a flexible mechanism allowing a
547consumer-pull style of activity, or producer-push, or a mixture of
548both.
549
550
551
552@subsection High-level summary
553
554This interface provides some handy wrappers around the low-level
555interface to facilitate reading and writing @code{bzip2} format
556files (@code{.bz2} files). The routines provide hooks to facilitate
557reading files in which the @code{bzip2} data stream is embedded
558within some larger-scale file structure, or where there are
559multiple @code{bzip2} data streams concatenated end-to-end.
560
561For reading files, @code{BZ2_bzReadOpen}, @code{BZ2_bzRead},
562@code{BZ2_bzReadClose} and @* @code{BZ2_bzReadGetUnused} are supplied. For
563writing files, @code{BZ2_bzWriteOpen}, @code{BZ2_bzWrite} and
564@code{BZ2_bzWriteFinish} are available.
565
566As with the low-level library, no global variables are used
567so the library is per se thread-safe. However, if I/O errors
568occur whilst reading or writing the underlying compressed files,
569you may have to consult @code{errno} to determine the cause of
570the error. In that case, you'd need a C library which correctly
571supports @code{errno} in a multithreaded environment.
572
573To make the library a little simpler and more portable,
574@code{BZ2_bzReadOpen} and @code{BZ2_bzWriteOpen} require you to pass them file
575handles (@code{FILE*}s) which have previously been opened for reading or
576writing respectively. That avoids portability problems associated with
577file operations and file attributes, whilst not being much of an
578imposition on the programmer.
579
580
581
582@subsection Utility functions summary
583For very simple needs, @code{BZ2_bzBuffToBuffCompress} and
584@code{BZ2_bzBuffToBuffDecompress} are provided. These compress
585data in memory from one buffer to another buffer in a single
586function call. You should assess whether these functions
587fulfill your memory-to-memory compression/decompression
588requirements before investing effort in understanding the more
589general but more complex low-level interface.
590
591Yoshioka Tsuneo (@code{QWF00133@@niftyserve.or.jp} /
592@code{tsuneo-y@@is.aist-nara.ac.jp}) has contributed some functions to
593give better @code{zlib} compatibility. These functions are
594@code{BZ2_bzopen}, @code{BZ2_bzread}, @code{BZ2_bzwrite}, @code{BZ2_bzflush},
595@code{BZ2_bzclose},
596@code{BZ2_bzerror} and @code{BZ2_bzlibVersion}. You may find these functions
597more convenient for simple file reading and writing, than those in the
598high-level interface. These functions are not (yet) officially part of
599the library, and are minimally documented here. If they break, you
600get to keep all the pieces. I hope to document them properly when time
601permits.
602
603Yoshioka also contributed modifications to allow the library to be
604built as a Windows DLL.
605
606
607@section Error handling
608
609The library is designed to recover cleanly in all situations, including
610the worst-case situation of decompressing random data. I'm not
611100% sure that it can always do this, so you might want to add
612a signal handler to catch segmentation violations during decompression
613if you are feeling especially paranoid. I would be interested in
614hearing more about the robustness of the library to corrupted
615compressed data.
616
617Version 1.0 is much more robust in this respect than
6180.9.0 or 0.9.5. Investigations with Checker (a tool for
619detecting problems with memory management, similar to Purify)
620indicate that, at least for the few files I tested, all single-bit
621errors in the decompressed data are caught properly, with no
622segmentation faults, no reads of uninitialised data and no
623out of range reads or writes. So it's certainly much improved,
624although I wouldn't claim it to be totally bombproof.
625
626The file @code{bzlib.h} contains all definitions needed to use
627the library. In particular, you should definitely not include
628@code{bzlib_private.h}.
629
630In @code{bzlib.h}, the various return values are defined. The following
631list is not intended as an exhaustive description of the circumstances
632in which a given value may be returned -- those descriptions are given
633later. Rather, it is intended to convey the rough meaning of each
634return value. The first five actions are normal and not intended to
635denote an error situation.
636@table @code
637@item BZ_OK
638The requested action was completed successfully.
639@item BZ_RUN_OK
640@itemx BZ_FLUSH_OK
641@itemx BZ_FINISH_OK
642In @code{BZ2_bzCompress}, the requested flush/finish/nothing-special action
643was completed successfully.
644@item BZ_STREAM_END
645Compression of data was completed, or the logical stream end was
646detected during decompression.
647@end table
648
649The following return values indicate an error of some kind.
650@table @code
651@item BZ_CONFIG_ERROR
652Indicates that the library has been improperly compiled on your
653platform -- a major configuration error. Specifically, it means
654that @code{sizeof(char)}, @code{sizeof(short)} and @code{sizeof(int)}
655are not 1, 2 and 4 respectively, as they should be. Note that the
656library should still work properly on 64-bit platforms which follow
657the LP64 programming model -- that is, where @code{sizeof(long)}
658and @code{sizeof(void*)} are 8. Under LP64, @code{sizeof(int)} is
659still 4, so @code{libbzip2}, which doesn't use the @code{long} type,
660is OK.
661@item BZ_SEQUENCE_ERROR
662When using the library, it is important to call the functions in the
663correct sequence and with data structures (buffers etc) in the correct
664states. @code{libbzip2} checks as much as it can to ensure this is
665happening, and returns @code{BZ_SEQUENCE_ERROR} if not. Code which
666complies precisely with the function semantics, as detailed below,
667should never receive this value; such an event denotes buggy code
668which you should investigate.
669@item BZ_PARAM_ERROR
670Returned when a parameter to a function call is out of range
671or otherwise manifestly incorrect. As with @code{BZ_SEQUENCE_ERROR},
672this denotes a bug in the client code. The distinction between
673@code{BZ_PARAM_ERROR} and @code{BZ_SEQUENCE_ERROR} is a bit hazy, but still worth
674making.
675@item BZ_MEM_ERROR
676Returned when a request to allocate memory failed. Note that the
677quantity of memory needed to decompress a stream cannot be determined
678until the stream's header has been read. So @code{BZ2_bzDecompress} and
679@code{BZ2_bzRead} may return @code{BZ_MEM_ERROR} even though some of
680the compressed data has been read. The same is not true for
681compression; once @code{BZ2_bzCompressInit} or @code{BZ2_bzWriteOpen} have
682successfully completed, @code{BZ_MEM_ERROR} cannot occur.
683@item BZ_DATA_ERROR
684Returned when a data integrity error is detected during decompression.
685Most importantly, this means when stored and computed CRCs for the
686data do not match. This value is also returned upon detection of any
687other anomaly in the compressed data.
688@item BZ_DATA_ERROR_MAGIC
689As a special case of @code{BZ_DATA_ERROR}, it is sometimes useful to
690know when the compressed stream does not start with the correct
691magic bytes (@code{'B' 'Z' 'h'}).
692@item BZ_IO_ERROR
693Returned by @code{BZ2_bzRead} and @code{BZ2_bzWrite} when there is an error
694reading or writing in the compressed file, and by @code{BZ2_bzReadOpen}
695and @code{BZ2_bzWriteOpen} for attempts to use a file for which the
696error indicator (viz, @code{ferror(f)}) is set.
697On receipt of @code{BZ_IO_ERROR}, the caller should consult
698@code{errno} and/or @code{perror} to acquire operating-system
699specific information about the problem.
700@item BZ_UNEXPECTED_EOF
701Returned by @code{BZ2_bzRead} when the compressed file finishes
702before the logical end of stream is detected.
703@item BZ_OUTBUFF_FULL
704Returned by @code{BZ2_bzBuffToBuffCompress} and
705@code{BZ2_bzBuffToBuffDecompress} to indicate that the output data
706will not fit into the output buffer provided.
707@end table
708
709
710
711@section Low-level interface
712
713@subsection @code{BZ2_bzCompressInit}
714@example
715typedef
716 struct @{
717 char *next_in;
718 unsigned int avail_in;
719 unsigned int total_in_lo32;
720 unsigned int total_in_hi32;
721
722 char *next_out;
723 unsigned int avail_out;
724 unsigned int total_out_lo32;
725 unsigned int total_out_hi32;
726
727 void *state;
728
729 void *(*bzalloc)(void *,int,int);
730 void (*bzfree)(void *,void *);
731 void *opaque;
732 @}
733 bz_stream;
734
735int BZ2_bzCompressInit ( bz_stream *strm,
736 int blockSize100k,
737 int verbosity,
738 int workFactor );
739
740@end example
741
742Prepares for compression. The @code{bz_stream} structure
743holds all data pertaining to the compression activity.
744A @code{bz_stream} structure should be allocated and initialised
745prior to the call.
746The fields of @code{bz_stream}
747comprise the entirety of the user-visible data. @code{state}
748is a pointer to the private data structures required for compression.
749
750Custom memory allocators are supported, via fields @code{bzalloc},
751@code{bzfree},
752and @code{opaque}. The value
753@code{opaque} is passed to as the first argument to
754all calls to @code{bzalloc} and @code{bzfree}, but is
755otherwise ignored by the library.
756The call @code{bzalloc ( opaque, n, m )} is expected to return a
757pointer @code{p} to
758@code{n * m} bytes of memory, and @code{bzfree ( opaque, p )}
759should free
760that memory.
761
762If you don't want to use a custom memory allocator, set @code{bzalloc},
763@code{bzfree} and
764@code{opaque} to @code{NULL},
765and the library will then use the standard @code{malloc}/@code{free}
766routines.
767
768Before calling @code{BZ2_bzCompressInit}, fields @code{bzalloc},
769@code{bzfree} and @code{opaque} should
770be filled appropriately, as just described. Upon return, the internal
771state will have been allocated and initialised, and @code{total_in_lo32},
772@code{total_in_hi32}, @code{total_out_lo32} and
773@code{total_out_hi32} will have been set to zero.
774These four fields are used by the library
775to inform the caller of the total amount of data passed into and out of
776the library, respectively. You should not try to change them.
777As of version 1.0, 64-bit counts are maintained, even on 32-bit
778platforms, using the @code{_hi32} fields to store the upper 32 bits
779of the count. So, for example, the total amount of data in
780is @code{(total_in_hi32 << 32) + total_in_lo32}.
781
782Parameter @code{blockSize100k} specifies the block size to be used for
783compression. It should be a value between 1 and 9 inclusive, and the
784actual block size used is 100000 x this figure. 9 gives the best
785compression but takes most memory.
786
787Parameter @code{verbosity} should be set to a number between 0 and 4
788inclusive. 0 is silent, and greater numbers give increasingly verbose
789monitoring/debugging output. If the library has been compiled with
790@code{-DBZ_NO_STDIO}, no such output will appear for any verbosity
791setting.
792
793Parameter @code{workFactor} controls how the compression phase behaves
794when presented with worst case, highly repetitive, input data. If
795compression runs into difficulties caused by repetitive data, the
796library switches from the standard sorting algorithm to a fallback
797algorithm. The fallback is slower than the standard algorithm by
798perhaps a factor of three, but always behaves reasonably, no matter how
799bad the input.
800
801Lower values of @code{workFactor} reduce the amount of effort the
802standard algorithm will expend before resorting to the fallback. You
803should set this parameter carefully; too low, and many inputs will be
804handled by the fallback algorithm and so compress rather slowly, too
805high, and your average-to-worst case compression times can become very
806large. The default value of 30 gives reasonable behaviour over a wide
807range of circumstances.
808
809Allowable values range from 0 to 250 inclusive. 0 is a special case,
810equivalent to using the default value of 30.
811
812Note that the compressed output generated is the same regardless of
813whether or not the fallback algorithm is used.
814
815Be aware also that this parameter may disappear entirely in future
816versions of the library. In principle it should be possible to devise a
817good way to automatically choose which algorithm to use. Such a
818mechanism would render the parameter obsolete.
819
820Possible return values:
821@display
822 @code{BZ_CONFIG_ERROR}
823 if the library has been mis-compiled
824 @code{BZ_PARAM_ERROR}
825 if @code{strm} is @code{NULL}
826 or @code{blockSize} < 1 or @code{blockSize} > 9
827 or @code{verbosity} < 0 or @code{verbosity} > 4
828 or @code{workFactor} < 0 or @code{workFactor} > 250
829 @code{BZ_MEM_ERROR}
830 if not enough memory is available
831 @code{BZ_OK}
832 otherwise
833@end display
834Allowable next actions:
835@display
836 @code{BZ2_bzCompress}
837 if @code{BZ_OK} is returned
838 no specific action needed in case of error
839@end display
840
841@subsection @code{BZ2_bzCompress}
842@example
843 int BZ2_bzCompress ( bz_stream *strm, int action );
844@end example
845Provides more input and/or output buffer space for the library. The
846caller maintains input and output buffers, and calls @code{BZ2_bzCompress} to
847transfer data between them.
848
849Before each call to @code{BZ2_bzCompress}, @code{next_in} should point at
850the data to be compressed, and @code{avail_in} should indicate how many
851bytes the library may read. @code{BZ2_bzCompress} updates @code{next_in},
852@code{avail_in} and @code{total_in} to reflect the number of bytes it
853has read.
854
855Similarly, @code{next_out} should point to a buffer in which the
856compressed data is to be placed, with @code{avail_out} indicating how
857much output space is available. @code{BZ2_bzCompress} updates
858@code{next_out}, @code{avail_out} and @code{total_out} to reflect the
859number of bytes output.
860
861You may provide and remove as little or as much data as you like on each
862call of @code{BZ2_bzCompress}. In the limit, it is acceptable to supply and
863remove data one byte at a time, although this would be terribly
864inefficient. You should always ensure that at least one byte of output
865space is available at each call.
866
867A second purpose of @code{BZ2_bzCompress} is to request a change of mode of the
868compressed stream.
869
870Conceptually, a compressed stream can be in one of four states: IDLE,
871RUNNING, FLUSHING and FINISHING. Before initialisation
872(@code{BZ2_bzCompressInit}) and after termination (@code{BZ2_bzCompressEnd}), a
873stream is regarded as IDLE.
874
875Upon initialisation (@code{BZ2_bzCompressInit}), the stream is placed in the
876RUNNING state. Subsequent calls to @code{BZ2_bzCompress} should pass
877@code{BZ_RUN} as the requested action; other actions are illegal and
878will result in @code{BZ_SEQUENCE_ERROR}.
879
880At some point, the calling program will have provided all the input data
881it wants to. It will then want to finish up -- in effect, asking the
882library to process any data it might have buffered internally. In this
883state, @code{BZ2_bzCompress} will no longer attempt to read data from
884@code{next_in}, but it will want to write data to @code{next_out}.
885Because the output buffer supplied by the user can be arbitrarily small,
886the finishing-up operation cannot necessarily be done with a single call
887of @code{BZ2_bzCompress}.
888
889Instead, the calling program passes @code{BZ_FINISH} as an action to
890@code{BZ2_bzCompress}. This changes the stream's state to FINISHING. Any
891remaining input (ie, @code{next_in[0 .. avail_in-1]}) is compressed and
892transferred to the output buffer. To do this, @code{BZ2_bzCompress} must be
893called repeatedly until all the output has been consumed. At that
894point, @code{BZ2_bzCompress} returns @code{BZ_STREAM_END}, and the stream's
895state is set back to IDLE. @code{BZ2_bzCompressEnd} should then be
896called.
897
898Just to make sure the calling program does not cheat, the library makes
899a note of @code{avail_in} at the time of the first call to
900@code{BZ2_bzCompress} which has @code{BZ_FINISH} as an action (ie, at the
901time the program has announced its intention to not supply any more
902input). By comparing this value with that of @code{avail_in} over
903subsequent calls to @code{BZ2_bzCompress}, the library can detect any
904attempts to slip in more data to compress. Any calls for which this is
905detected will return @code{BZ_SEQUENCE_ERROR}. This indicates a
906programming mistake which should be corrected.
907
908Instead of asking to finish, the calling program may ask
909@code{BZ2_bzCompress} to take all the remaining input, compress it and
910terminate the current (Burrows-Wheeler) compression block. This could
911be useful for error control purposes. The mechanism is analogous to
912that for finishing: call @code{BZ2_bzCompress} with an action of
913@code{BZ_FLUSH}, remove output data, and persist with the
914@code{BZ_FLUSH} action until the value @code{BZ_RUN} is returned. As
915with finishing, @code{BZ2_bzCompress} detects any attempt to provide more
916input data once the flush has begun.
917
918Once the flush is complete, the stream returns to the normal RUNNING
919state.
920
921This all sounds pretty complex, but isn't really. Here's a table
922which shows which actions are allowable in each state, what action
923will be taken, what the next state is, and what the non-error return
924values are. Note that you can't explicitly ask what state the
925stream is in, but nor do you need to -- it can be inferred from the
926values returned by @code{BZ2_bzCompress}.
927@display
928IDLE/@code{any}
929 Illegal. IDLE state only exists after @code{BZ2_bzCompressEnd} or
930 before @code{BZ2_bzCompressInit}.
931 Return value = @code{BZ_SEQUENCE_ERROR}
932
933RUNNING/@code{BZ_RUN}
934 Compress from @code{next_in} to @code{next_out} as much as possible.
935 Next state = RUNNING
936 Return value = @code{BZ_RUN_OK}
937
938RUNNING/@code{BZ_FLUSH}
939 Remember current value of @code{next_in}. Compress from @code{next_in}
940 to @code{next_out} as much as possible, but do not accept any more input.
941 Next state = FLUSHING
942 Return value = @code{BZ_FLUSH_OK}
943
944RUNNING/@code{BZ_FINISH}
945 Remember current value of @code{next_in}. Compress from @code{next_in}
946 to @code{next_out} as much as possible, but do not accept any more input.
947 Next state = FINISHING
948 Return value = @code{BZ_FINISH_OK}
949
950FLUSHING/@code{BZ_FLUSH}
951 Compress from @code{next_in} to @code{next_out} as much as possible,
952 but do not accept any more input.
953 If all the existing input has been used up and all compressed
954 output has been removed
955 Next state = RUNNING; Return value = @code{BZ_RUN_OK}
956 else
957 Next state = FLUSHING; Return value = @code{BZ_FLUSH_OK}
958
959FLUSHING/other
960 Illegal.
961 Return value = @code{BZ_SEQUENCE_ERROR}
962
963FINISHING/@code{BZ_FINISH}
964 Compress from @code{next_in} to @code{next_out} as much as possible,
965 but to not accept any more input.
966 If all the existing input has been used up and all compressed
967 output has been removed
968 Next state = IDLE; Return value = @code{BZ_STREAM_END}
969 else
970 Next state = FINISHING; Return value = @code{BZ_FINISHING}
971
972FINISHING/other
973 Illegal.
974 Return value = @code{BZ_SEQUENCE_ERROR}
975@end display
976
977That still looks complicated? Well, fair enough. The usual sequence
978of calls for compressing a load of data is:
979@itemize @bullet
980@item Get started with @code{BZ2_bzCompressInit}.
981@item Shovel data in and shlurp out its compressed form using zero or more
982calls of @code{BZ2_bzCompress} with action = @code{BZ_RUN}.
983@item Finish up.
984Repeatedly call @code{BZ2_bzCompress} with action = @code{BZ_FINISH},
985copying out the compressed output, until @code{BZ_STREAM_END} is returned.
986@item Close up and go home. Call @code{BZ2_bzCompressEnd}.
987@end itemize
988If the data you want to compress fits into your input buffer all
989at once, you can skip the calls of @code{BZ2_bzCompress ( ..., BZ_RUN )} and
990just do the @code{BZ2_bzCompress ( ..., BZ_FINISH )} calls.
991
992All required memory is allocated by @code{BZ2_bzCompressInit}. The
993compression library can accept any data at all (obviously). So you
994shouldn't get any error return values from the @code{BZ2_bzCompress} calls.
995If you do, they will be @code{BZ_SEQUENCE_ERROR}, and indicate a bug in
996your programming.
997
998Trivial other possible return values:
999@display
1000 @code{BZ_PARAM_ERROR}
1001 if @code{strm} is @code{NULL}, or @code{strm->s} is @code{NULL}
1002@end display
1003
1004@subsection @code{BZ2_bzCompressEnd}
1005@example
1006int BZ2_bzCompressEnd ( bz_stream *strm );
1007@end example
1008Releases all memory associated with a compression stream.
1009
1010Possible return values:
1011@display
1012 @code{BZ_PARAM_ERROR} if @code{strm} is @code{NULL} or @code{strm->s} is @code{NULL}
1013 @code{BZ_OK} otherwise
1014@end display
1015
1016
1017@subsection @code{BZ2_bzDecompressInit}
1018@example
1019int BZ2_bzDecompressInit ( bz_stream *strm, int verbosity, int small );
1020@end example
1021Prepares for decompression. As with @code{BZ2_bzCompressInit}, a
1022@code{bz_stream} record should be allocated and initialised before the
1023call. Fields @code{bzalloc}, @code{bzfree} and @code{opaque} should be
1024set if a custom memory allocator is required, or made @code{NULL} for
1025the normal @code{malloc}/@code{free} routines. Upon return, the internal
1026state will have been initialised, and @code{total_in} and
1027@code{total_out} will be zero.
1028
1029For the meaning of parameter @code{verbosity}, see @code{BZ2_bzCompressInit}.
1030
1031If @code{small} is nonzero, the library will use an alternative
1032decompression algorithm which uses less memory but at the cost of
1033decompressing more slowly (roughly speaking, half the speed, but the
1034maximum memory requirement drops to around 2300k). See Chapter 2 for
1035more information on memory management.
1036
1037Note that the amount of memory needed to decompress
1038a stream cannot be determined until the stream's header has been read,
1039so even if @code{BZ2_bzDecompressInit} succeeds, a subsequent
1040@code{BZ2_bzDecompress} could fail with @code{BZ_MEM_ERROR}.
1041
1042Possible return values:
1043@display
1044 @code{BZ_CONFIG_ERROR}
1045 if the library has been mis-compiled
1046 @code{BZ_PARAM_ERROR}
1047 if @code{(small != 0 && small != 1)}
1048 or @code{(verbosity < 0 || verbosity > 4)}
1049 @code{BZ_MEM_ERROR}
1050 if insufficient memory is available
1051@end display
1052
1053Allowable next actions:
1054@display
1055 @code{BZ2_bzDecompress}
1056 if @code{BZ_OK} was returned
1057 no specific action required in case of error
1058@end display
1059
1060
1061
1062@subsection @code{BZ2_bzDecompress}
1063@example
1064int BZ2_bzDecompress ( bz_stream *strm );
1065@end example
1066Provides more input and/out output buffer space for the library. The
1067caller maintains input and output buffers, and uses @code{BZ2_bzDecompress}
1068to transfer data between them.
1069
1070Before each call to @code{BZ2_bzDecompress}, @code{next_in}
1071should point at the compressed data,
1072and @code{avail_in} should indicate how many bytes the library
1073may read. @code{BZ2_bzDecompress} updates @code{next_in}, @code{avail_in}
1074and @code{total_in}
1075to reflect the number of bytes it has read.
1076
1077Similarly, @code{next_out} should point to a buffer in which the uncompressed
1078output is to be placed, with @code{avail_out} indicating how much output space
1079is available. @code{BZ2_bzCompress} updates @code{next_out},
1080@code{avail_out} and @code{total_out} to reflect
1081the number of bytes output.
1082
1083You may provide and remove as little or as much data as you like on
1084each call of @code{BZ2_bzDecompress}.
1085In the limit, it is acceptable to
1086supply and remove data one byte at a time, although this would be
1087terribly inefficient. You should always ensure that at least one
1088byte of output space is available at each call.
1089
1090Use of @code{BZ2_bzDecompress} is simpler than @code{BZ2_bzCompress}.
1091
1092You should provide input and remove output as described above, and
1093repeatedly call @code{BZ2_bzDecompress} until @code{BZ_STREAM_END} is
1094returned. Appearance of @code{BZ_STREAM_END} denotes that
1095@code{BZ2_bzDecompress} has detected the logical end of the compressed
1096stream. @code{BZ2_bzDecompress} will not produce @code{BZ_STREAM_END} until
1097all output data has been placed into the output buffer, so once
1098@code{BZ_STREAM_END} appears, you are guaranteed to have available all
1099the decompressed output, and @code{BZ2_bzDecompressEnd} can safely be
1100called.
1101
1102If case of an error return value, you should call @code{BZ2_bzDecompressEnd}
1103to clean up and release memory.
1104
1105Possible return values:
1106@display
1107 @code{BZ_PARAM_ERROR}
1108 if @code{strm} is @code{NULL} or @code{strm->s} is @code{NULL}
1109 or @code{strm->avail_out < 1}
1110 @code{BZ_DATA_ERROR}
1111 if a data integrity error is detected in the compressed stream
1112 @code{BZ_DATA_ERROR_MAGIC}
1113 if the compressed stream doesn't begin with the right magic bytes
1114 @code{BZ_MEM_ERROR}
1115 if there wasn't enough memory available
1116 @code{BZ_STREAM_END}
1117 if the logical end of the data stream was detected and all
1118 output in has been consumed, eg @code{s->avail_out > 0}
1119 @code{BZ_OK}
1120 otherwise
1121@end display
1122Allowable next actions:
1123@display
1124 @code{BZ2_bzDecompress}
1125 if @code{BZ_OK} was returned
1126 @code{BZ2_bzDecompressEnd}
1127 otherwise
1128@end display
1129
1130
1131@subsection @code{BZ2_bzDecompressEnd}
1132@example
1133int BZ2_bzDecompressEnd ( bz_stream *strm );
1134@end example
1135Releases all memory associated with a decompression stream.
1136
1137Possible return values:
1138@display
1139 @code{BZ_PARAM_ERROR}
1140 if @code{strm} is @code{NULL} or @code{strm->s} is @code{NULL}
1141 @code{BZ_OK}
1142 otherwise
1143@end display
1144
1145Allowable next actions:
1146@display
1147 None.
1148@end display
1149
1150
1151@section High-level interface
1152
1153This interface provides functions for reading and writing
1154@code{bzip2} format files. First, some general points.
1155
1156@itemize @bullet
1157@item All of the functions take an @code{int*} first argument,
1158 @code{bzerror}.
1159 After each call, @code{bzerror} should be consulted first to determine
1160 the outcome of the call. If @code{bzerror} is @code{BZ_OK},
1161 the call completed
1162 successfully, and only then should the return value of the function
1163 (if any) be consulted. If @code{bzerror} is @code{BZ_IO_ERROR},
1164 there was an error
1165 reading/writing the underlying compressed file, and you should
1166 then consult @code{errno}/@code{perror} to determine the
1167 cause of the difficulty.
1168 @code{bzerror} may also be set to various other values; precise details are
1169 given on a per-function basis below.
1170@item If @code{bzerror} indicates an error
1171 (ie, anything except @code{BZ_OK} and @code{BZ_STREAM_END}),
1172 you should immediately call @code{BZ2_bzReadClose} (or @code{BZ2_bzWriteClose},
1173 depending on whether you are attempting to read or to write)
1174 to free up all resources associated
1175 with the stream. Once an error has been indicated, behaviour of all calls
1176 except @code{BZ2_bzReadClose} (@code{BZ2_bzWriteClose}) is undefined.
1177 The implication is that (1) @code{bzerror} should
1178 be checked after each call, and (2) if @code{bzerror} indicates an error,
1179 @code{BZ2_bzReadClose} (@code{BZ2_bzWriteClose}) should then be called to clean up.
1180@item The @code{FILE*} arguments passed to
1181 @code{BZ2_bzReadOpen}/@code{BZ2_bzWriteOpen}
1182 should be set to binary mode.
1183 Most Unix systems will do this by default, but other platforms,
1184 including Windows and Mac, will not. If you omit this, you may
1185 encounter problems when moving code to new platforms.
1186@item Memory allocation requests are handled by
1187 @code{malloc}/@code{free}.
1188 At present
1189 there is no facility for user-defined memory allocators in the file I/O
1190 functions (could easily be added, though).
1191@end itemize
1192
1193
1194
1195@subsection @code{BZ2_bzReadOpen}
1196@example
1197 typedef void BZFILE;
1198
1199 BZFILE *BZ2_bzReadOpen ( int *bzerror, FILE *f,
1200 int small, int verbosity,
1201 void *unused, int nUnused );
1202@end example
1203Prepare to read compressed data from file handle @code{f}. @code{f}
1204should refer to a file which has been opened for reading, and for which
1205the error indicator (@code{ferror(f)})is not set. If @code{small} is 1,
1206the library will try to decompress using less memory, at the expense of
1207speed.
1208
1209For reasons explained below, @code{BZ2_bzRead} will decompress the
1210@code{nUnused} bytes starting at @code{unused}, before starting to read
1211from the file @code{f}. At most @code{BZ_MAX_UNUSED} bytes may be
1212supplied like this. If this facility is not required, you should pass
1213@code{NULL} and @code{0} for @code{unused} and n@code{Unused}
1214respectively.
1215
1216For the meaning of parameters @code{small} and @code{verbosity},
1217see @code{BZ2_bzDecompressInit}.
1218
1219The amount of memory needed to decompress a file cannot be determined
1220until the file's header has been read. So it is possible that
1221@code{BZ2_bzReadOpen} returns @code{BZ_OK} but a subsequent call of
1222@code{BZ2_bzRead} will return @code{BZ_MEM_ERROR}.
1223
1224Possible assignments to @code{bzerror}:
1225@display
1226 @code{BZ_CONFIG_ERROR}
1227 if the library has been mis-compiled
1228 @code{BZ_PARAM_ERROR}
1229 if @code{f} is @code{NULL}
1230 or @code{small} is neither @code{0} nor @code{1}
1231 or @code{(unused == NULL && nUnused != 0)}
1232 or @code{(unused != NULL && !(0 <= nUnused <= BZ_MAX_UNUSED))}
1233 @code{BZ_IO_ERROR}
1234 if @code{ferror(f)} is nonzero
1235 @code{BZ_MEM_ERROR}
1236 if insufficient memory is available
1237 @code{BZ_OK}
1238 otherwise.
1239@end display
1240
1241Possible return values:
1242@display
1243 Pointer to an abstract @code{BZFILE}
1244 if @code{bzerror} is @code{BZ_OK}
1245 @code{NULL}
1246 otherwise
1247@end display
1248
1249Allowable next actions:
1250@display
1251 @code{BZ2_bzRead}
1252 if @code{bzerror} is @code{BZ_OK}
1253 @code{BZ2_bzClose}
1254 otherwise
1255@end display
1256
1257
1258@subsection @code{BZ2_bzRead}
1259@example
1260 int BZ2_bzRead ( int *bzerror, BZFILE *b, void *buf, int len );
1261@end example
1262Reads up to @code{len} (uncompressed) bytes from the compressed file
1263@code{b} into
1264the buffer @code{buf}. If the read was successful,
1265@code{bzerror} is set to @code{BZ_OK}
1266and the number of bytes read is returned. If the logical end-of-stream
1267was detected, @code{bzerror} will be set to @code{BZ_STREAM_END},
1268and the number
1269of bytes read is returned. All other @code{bzerror} values denote an error.
1270
1271@code{BZ2_bzRead} will supply @code{len} bytes,
1272unless the logical stream end is detected
1273or an error occurs. Because of this, it is possible to detect the
1274stream end by observing when the number of bytes returned is
1275less than the number
1276requested. Nevertheless, this is regarded as inadvisable; you should
1277instead check @code{bzerror} after every call and watch out for
1278@code{BZ_STREAM_END}.
1279
1280Internally, @code{BZ2_bzRead} copies data from the compressed file in chunks
1281of size @code{BZ_MAX_UNUSED} bytes
1282before decompressing it. If the file contains more bytes than strictly
1283needed to reach the logical end-of-stream, @code{BZ2_bzRead} will almost certainly
1284read some of the trailing data before signalling @code{BZ_SEQUENCE_END}.
1285To collect the read but unused data once @code{BZ_SEQUENCE_END} has
1286appeared, call @code{BZ2_bzReadGetUnused} immediately before @code{BZ2_bzReadClose}.
1287
1288Possible assignments to @code{bzerror}:
1289@display
1290 @code{BZ_PARAM_ERROR}
1291 if @code{b} is @code{NULL} or @code{buf} is @code{NULL} or @code{len < 0}
1292 @code{BZ_SEQUENCE_ERROR}
1293 if @code{b} was opened with @code{BZ2_bzWriteOpen}
1294 @code{BZ_IO_ERROR}
1295 if there is an error reading from the compressed file
1296 @code{BZ_UNEXPECTED_EOF}
1297 if the compressed file ended before the logical end-of-stream was detected
1298 @code{BZ_DATA_ERROR}
1299 if a data integrity error was detected in the compressed stream
1300 @code{BZ_DATA_ERROR_MAGIC}
1301 if the stream does not begin with the requisite header bytes (ie, is not
1302 a @code{bzip2} data file). This is really a special case of @code{BZ_DATA_ERROR}.
1303 @code{BZ_MEM_ERROR}
1304 if insufficient memory was available
1305 @code{BZ_STREAM_END}
1306 if the logical end of stream was detected.
1307 @code{BZ_OK}
1308 otherwise.
1309@end display
1310
1311Possible return values:
1312@display
1313 number of bytes read
1314 if @code{bzerror} is @code{BZ_OK} or @code{BZ_STREAM_END}
1315 undefined
1316 otherwise
1317@end display
1318
1319Allowable next actions:
1320@display
1321 collect data from @code{buf}, then @code{BZ2_bzRead} or @code{BZ2_bzReadClose}
1322 if @code{bzerror} is @code{BZ_OK}
1323 collect data from @code{buf}, then @code{BZ2_bzReadClose} or @code{BZ2_bzReadGetUnused}
1324 if @code{bzerror} is @code{BZ_SEQUENCE_END}
1325 @code{BZ2_bzReadClose}
1326 otherwise
1327@end display
1328
1329
1330
1331@subsection @code{BZ2_bzReadGetUnused}
1332@example
1333 void BZ2_bzReadGetUnused ( int* bzerror, BZFILE *b,