bzip2-1.0.3bzip2-1.0.3

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:
 
 * added --fast and --best aliases for -1 -9 for gzip compatibility.
 
+
+1.0.3 (15 Feb 05)
+~~~~~~~~~~~~~~~~~
+Fixes some minor bugs since the last version, 1.0.2.
+
+* Further robustification against corrupted compressed data.
+  There are currently no known bitstreams which can cause the
+  decompressor to crash, loop or access memory which does not
+  belong to it.  If you are using bzip2 or the library to 
+  decompress bitstreams from untrusted sources, an upgrade
+  to 1.0.3 is recommended.
+
+* The documentation has been converted to XML, from which html
+  and pdf can be derived.
+
+* Various minor bugs in the documentation have been fixed.
+
+* Fixes for various compilation warnings with newer versions of
+  gcc, and on 64-bit platforms.
+
+* The BZ_NO_STDIO cpp symbol was not properly observed in 1.0.2.
+  This has been fixed.
diff --git a/LICENSE b/LICENSE
index 9d4fa43..e60845b 100644
--- a/LICENSE
+++ b/LICENSE
@@ -1,6 +1,7 @@
 
-This program, "bzip2" and associated library "libbzip2", are
-copyright (C) 1996-2002 Julian R Seward.  All rights reserved.
+This program, "bzip2", the associated library "libbzip2", and all
+documentation, are copyright (C) 1996-2005 Julian R Seward.  All
+rights reserved.
 
 Redistribution and use in source and binary forms, with or without
 modification, are permitted provided that the following conditions
@@ -35,5 +36,5 @@ SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
 Julian Seward, Cambridge, UK.
 jseward@acm.org
-bzip2/libbzip2 version 1.0.2 of 30 December 2001
+bzip2/libbzip2 version 1.0.3 of 15 February 2005
 
diff --git a/Makefile b/Makefile
index 8305235..eea329a 100644
--- a/Makefile
+++ b/Makefile
@@ -7,9 +7,8 @@ AR=ar
 RANLIB=ranlib
 LDFLAGS=
 
-# Suitably paranoid flags to avoid bugs in gcc-2.7
 BIGFILES=-D_FILE_OFFSET_BITS=64
-CFLAGS=-Wall -Winline -O2 -fomit-frame-pointer -fno-strength-reduce $(BIGFILES)
+CFLAGS=-Wall -Winline -O -g $(BIGFILES)
 
 # Where you want it installed when you do 'make install'
 PREFIX=/usr
@@ -96,7 +95,6 @@ install: bzip2 bzip2recover
         echo ".so man1/bzmore.1" > $(PREFIX)/man/man1/bzless.1
         echo ".so man1/bzdiff.1" > $(PREFIX)/man/man1/bzcmp.1
 
-distclean: clean
 clean: 
         rm -f *.o libbz2.a bzip2 bzip2recover \
         sample1.rb2 sample2.rb2 sample3.rb2 \
@@ -122,8 +120,12 @@ bzip2.o: bzip2.c
 bzip2recover.o: bzip2recover.c
         $(CC) $(CFLAGS) -c bzip2recover.c
 
-DISTNAME=bzip2-1.0.2
-tarfile:
+
+distclean: clean
+        rm -f manual.ps manual.html manual.pdf
+
+DISTNAME=bzip2-1.0.3
+dist: check manual
         rm -f $(DISTNAME)
         ln -sf . $(DISTNAME)
         tar cvf $(DISTNAME).tar \
@@ -139,9 +141,6 @@ tarfile:
            $(DISTNAME)/bzlib.h \
            $(DISTNAME)/bzlib_private.h \
            $(DISTNAME)/Makefile \
-           $(DISTNAME)/manual.texi \
-           $(DISTNAME)/manual.ps \
-           $(DISTNAME)/manual.pdf \
            $(DISTNAME)/LICENSE \
            $(DISTNAME)/bzip2.1 \
            $(DISTNAME)/bzip2.1.preformatted \
@@ -157,9 +156,12 @@ tarfile:
            $(DISTNAME)/sample2.bz2 \
            $(DISTNAME)/sample3.bz2 \
            $(DISTNAME)/dlltest.c \
-           $(DISTNAME)/*.html \
+           $(DISTNAME)/manual.html \
+           $(DISTNAME)/manual.pdf \
+           $(DISTNAME)/manual.ps \
            $(DISTNAME)/README \
            $(DISTNAME)/README.COMPILATION.PROBLEMS \
+           $(DISTNAME)/README.XML.STUFF \
            $(DISTNAME)/CHANGES \
            $(DISTNAME)/libbz2.def \
            $(DISTNAME)/libbz2.dsp \
@@ -175,18 +177,29 @@ tarfile:
            $(DISTNAME)/bzmore.1 \
            $(DISTNAME)/bzgrep \
            $(DISTNAME)/bzgrep.1 \
-           $(DISTNAME)/Makefile-libbz2_so
+           $(DISTNAME)/Makefile-libbz2_so \
+           $(DISTNAME)/bz-common.xsl \
+           $(DISTNAME)/bz-fo.xsl \
+           $(DISTNAME)/bz-html.xsl \
+           $(DISTNAME)/bzip.css \
+           $(DISTNAME)/entities.xml \
+           $(DISTNAME)/manual.xml \
+           $(DISTNAME)/format.pl \
+           $(DISTNAME)/xmlproc.sh
         gzip -v $(DISTNAME).tar
 
-# For rebuilding the manual from sources on my RedHat 7.2 box
-manual: manual.ps manual.pdf manual.html
+# For rebuilding the manual from sources on my SuSE 9.1 box
+
+MANUAL_SRCS=    bz-common.xsl bz-fo.xsl bz-html.xsl bzip.css \
+                entities.xml manual.xml 
+
+manual: manual.html manual.ps manual.pdf
 
-manual.ps: manual.texi
-        tex manual.texi
-        dvips -o manual.ps manual.dvi
+manual.ps: $(MANUAL_SRCS)
+        ./xmlproc.sh -ps manual.xml
 
-manual.pdf: manual.ps
-        ps2pdf manual.ps
+manual.pdf: $(MANUAL_SRCS)
+        ./xmlproc.sh -pdf manual.xml
 
-manual.html: manual.texi
-        texi2html -split_chapter manual.texi
+manual.html: $(MANUAL_SRCS)
+        ./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 @@
 
 # This Makefile builds a shared version of the library, 
-# libbz2.so.1.0.2, with soname libbz2.so.1.0,
+# libbz2.so.1.0.3, with soname libbz2.so.1.0,
 # at least on x86-Linux (RedHat 7.2), 
 # with gcc-2.96 20000731 (Red Hat Linux 7.1 2.96-98).  
 # Please see the README file for some 
@@ -9,7 +9,7 @@
 SHELL=/bin/sh
 CC=gcc
 BIGFILES=-D_FILE_OFFSET_BITS=64
-CFLAGS=-fpic -fPIC -Wall -Winline -O2 -fomit-frame-pointer -fno-strength-reduce $(BIGFILES)
+CFLAGS=-fpic -fPIC -Wall -Winline -O -g
 
 OBJS= blocksort.o  \
       huffman.o    \
@@ -20,13 +20,13 @@ OBJS= blocksort.o  \
       bzlib.o
 
 all: $(OBJS)
-        $(CC) -shared -Wl,-soname -Wl,libbz2.so.1.0 -o libbz2.so.1.0.2 $(OBJS)
-        $(CC) $(CFLAGS) -o bzip2-shared bzip2.c libbz2.so.1.0.2
+        $(CC) -shared -Wl,-soname -Wl,libbz2.so.1.0 -o libbz2.so.1.0.3 $(OBJS)
+        $(CC) $(CFLAGS) -o bzip2-shared bzip2.c libbz2.so.1.0.3
         rm -f libbz2.so.1.0
-        ln -s libbz2.so.1.0.2 libbz2.so.1.0
+        ln -s libbz2.so.1.0.3 libbz2.so.1.0
 
 clean: 
-        rm -f $(OBJS) bzip2.o libbz2.so.1.0.2 libbz2.so.1.0 bzip2-shared
+        rm -f $(OBJS) bzip2.o libbz2.so.1.0.3 libbz2.so.1.0 bzip2-shared
 
 blocksort.o: blocksort.c
         $(CC) $(CFLAGS) -c blocksort.c
diff --git a/README b/README
index 07505d8..1aff448 100644
--- a/README
+++ b/README
@@ -1,15 +1,15 @@
 
 This is the README for bzip2, a block-sorting file compressor, version
-1.0.2.  This version is fully compatible with the previous public
-releases, versions 0.1pl2, 0.9.0, 0.9.5, 1.0.0 and 1.0.1.
+1.0.3.  This version is fully compatible with the previous public
+releases, versions 0.1pl2, 0.9.0, 0.9.5, 1.0.0, 1.0.1 and 1.0.2.
 
-bzip2-1.0.2 is distributed under a BSD-style license.  For details,
+bzip2-1.0.3 is distributed under a BSD-style license.  For details,
 see the file LICENSE.
 
 Complete documentation is available in Postscript form (manual.ps),
-PDF (manual.pdf, amazingly enough) or html (manual_toc.html).  A
-plain-text version of the manual page is available as bzip2.txt.  
-A statement about Y2K issues is now included in the file Y2K_INFO.
+PDF (manual.pdf) or html (manual.html).  A plain-text version of the
+manual page is available as bzip2.txt.  A statement about Y2K issues
+is now included in the file Y2K_INFO.
 
 
 HOW TO BUILD -- UNIX
@@ -78,8 +78,7 @@ importance.  To validate bzip2, I used a modified version of Mark
 Nelson's churn program.  Churn is an automated test driver which
 recursively traverses a directory structure, using bzip2 to compress
 and then decompress each file it encounters, and checking that the
-decompressed data is the same as the original.  There are more details
-in Section 4 of the user guide.
+decompressed data is the same as the original.
 
 
 
@@ -119,9 +118,9 @@ DISCLAIMER:
 PATENTS:
 
    To the best of my knowledge, bzip2 does not use any patented
-   algorithms.  However, I do not have the resources available to
-   carry out a full patent search.  Therefore I cannot give any
-   guarantee of the above statement.
+   algorithms.  However, I do not have the resources to carry out
+   a patent search.  Therefore I cannot give any guarantee of the
+   above statement.
 
 End of legalities.
 
@@ -153,22 +152,26 @@ WHAT'S NEW IN 1.0.2 ?
 
    See the CHANGES file.
 
+WHAT'S NEW IN 1.0.3 ?
+
+   See the CHANGES file.
+
 
 I hope you find bzip2 useful.  Feel free to contact me at
-   jseward@acm.org
+   jseward@bzip.org
 if you have any suggestions or queries.  Many people mailed me with
 comments, suggestions and patches after the releases of bzip-0.15,
-bzip-0.21, and bzip2 versions 0.1pl2, 0.9.0, 0.9.5, 1.0.0 and 1.0.1,
-and the changes in bzip2 are largely a result of this feedback.
+bzip-0.21, and bzip2 versions 0.1pl2, 0.9.0, 0.9.5, 1.0.0, 1.0.1 and
+1.0.2, and the changes in bzip2 are largely a result of this feedback.
 I thank you for your comments.
 
 At least for the time being, bzip2's "home" is (or can be reached via)
-http://sources.redhat.com/bzip2.
+http://www.bzip.org
 
 Julian Seward
-jseward@acm.org
+jseward@bzip.org
 
-Cambridge, UK (and what a great town this is!)
+Cambridge, UK.
 
 18     July 1996 (version 0.15)
 25   August 1996 (version 0.21)
@@ -178,4 +181,5 @@ Cambridge, UK (and what a great town this is!)
  8     June 1999 (bzip2, version 0.9.5)
  4     Sept 1999 (bzip2, version 0.9.5d)
  5      May 2000 (bzip2, version 1.0pre8)
-30 December 2001 (bzip2, version 1.0.2pre1)
\ No newline at end of file
+30 December 2001 (bzip2, version 1.0.2pre1)
+15 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 @@
 
-bzip2-1.0 should compile without problems on the vast majority of
+bzip2-1.0.3 should compile without problems on the vast majority of
 platforms.  Using the supplied Makefile, I've built and tested it
-myself for x86-linux, sparc-solaris, alpha-linux, x86-cygwin32 and
-alpha-tru64unix.  With makefile.msc, Visual C++ 6.0 and nmake, you can
-build a native Win32 version too.  Large file support seems to work
-correctly on at least alpha-tru64unix and x86-cygwin32 (on Windows
-2000).
+myself for x86-linux and x86_64-linux.  With makefile.msc, Visual C++
+6.0 and nmake, you can build a native Win32 version too.  Large file
+support seems to work correctly on at least alpha-tru64unix and
+x86-cygwin32 (on Windows 2000).
 
 When I say "large file" I mean a file of size 2,147,483,648 (2^31)
 bytes 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
 support is, as far as I know, the Recommended Way to get correct large
 file support.  For more details, see the Large File Support
 Specification, published by the Large File Summit, at
-   http://www.sas.com/standard/large.file/
+   http://ftp.sas.com/standards/large.file
 
 As a general comment, if you get compilation errors which you think
 are 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
 large file support, if you are feeling paranoid.  Be aware though that
 any compilation problems which affect bzip2 will also affect spewG.c,
 alas.
-
-
-Known problems as of 1.0pre8:
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-* HP/UX 10.20 and 11.00, using gcc (2.7.2.3 and 2.95.2):  A large
-  number of warnings appear, including the following:
-
-     /usr/include/sys/resource.h: In function `getrlimit':
-     /usr/include/sys/resource.h:168: 
-        warning: implicit declaration of function `__getrlimit64'
-     /usr/include/sys/resource.h: In function `setrlimit':
-     /usr/include/sys/resource.h:170: 
-        warning: implicit declaration of function `__setrlimit64'
-
-  This would appear to be a problem with large file support, header
-  files and gcc.  gcc may or may not give up at this point.  If it
-  fails, you might be able to improve matters by adding 
-     -D__STDC_EXT__=1
-  to the BIGFILES variable in the Makefile (ie, change its definition
-  to
-     BIGFILES=-D_FILE_OFFSET_BITS=64 -D__STDC_EXT__=1
-
-  Even if gcc does produce a binary which appears to work (ie passes
-  its self-tests), you might want to test it to see if it works properly
-  on large files.
-
-
-* HP/UX 10.20 and 11.00, using HP's cc compiler.
-
-  No specific problems for this combination, except that you'll need to
-  specify the -Ae flag, and zap the gcc-specific stuff
-  -Wall -Winline -O2 -fomit-frame-pointer -fno-strength-reduce.
-  You should retain -D_FILE_OFFSET_BITS=64 in order to get large
-  file support -- which is reported to work ok for this HP/UX + cc
-  combination.
-
-
-* SunOS 4.1.X.
-
-  Amazingly, there are still people out there using this venerable old
-  banger.  I shouldn't be too rude -- I started life on SunOS, and
-  it was a pretty darn good OS, way back then.  Anyway:
-
-     SunOS doesn't seem to have strerror(), so you'll have to use
-     perror(), perhaps by doing adding this (warning: UNTESTED CODE):
-
-     char* strerror ( int errnum )
-     {
-        if (errnum < 0 || errnum >= sys_nerr)
-           return "Unknown error"; 
-        else
-           return sys_errlist[errnum];
-     }
-
-   Or you could comment out the relevant calls to strerror; they're
-   not mission-critical.  Or you could upgrade to Solaris.  Ha ha ha!
-   (what??  you think I've got Bad Attitude?) 
-
-
-* Making a shared library on Solaris.  (Not really a compilation
-  problem, but many people ask ...)  
-
-  Firstly, if you have Solaris 8, either you have libbz2.so already
-  on your system, or you can install it from the Solaris CD.  
-
-  Secondly, be aware that there are potential naming conflicts
-  between the .so file supplied with Solaris 8, and the .so file
-  which Makefile-libbz2_so will make.  Makefile-libbz2_so creates
-  a .so which has the names which I intend to be "official" as
-  of version 1.0.0 and onwards.  Unfortunately, the .so in
-  Solaris 8 appeared before I decided on the final names, so
-  the two libraries are incompatible.  We have since communicated
-  and I hope that the problems will have been solved in the next
-  version of Solaris, whenever that might appear.
-
-  All that said: you might be able to get somewhere
-  by finding the line in Makefile-libbz2_so which says
-
-  $(CC) -shared -Wl,-soname -Wl,libbz2.so.1.0 -o libbz2.so.1.0.2 $(OBJS)
-
-  and replacing with 
-
-  $(CC) -G -shared -o libbz2.so.1.0.2 -h libbz2.so.1.0 $(OBJS)
-  
-  If gcc objects to the combination -fpic -fPIC, get rid of
-  the second one, leaving just "-fpic".
-
-
-That'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 @@
+The script xmlproc.sh takes an xml file as input,
+and processes it to create .pdf, .html or .ps output.
+It uses format.pl, a perl script to format <pre> blocks nicely,
+ and add CDATA tags so writers do not have to use eg. &lt; 
+
+The file "entities.xml" must be edited to reflect current
+version, year, etc.
+
+
+Usage:
+
+  xmlproc.sh -v manual.xml
+  Validates an xml file to ensure no dtd-compliance errors
+
+  xmlproc.sh -html manual.xml
+  Output: manual.html
+
+  xmlproc.sh -pdf manual.xml
+  Output: manual.pdf
+
+  xmlproc.sh -ps manual.xml
+  Output: manual.ps
+
+
+Notum bene: 
+- pdfxmltex barfs if given a filename with an underscore in it
+
+- xmltex won't work yet - there's a bug in passivetex
+    which we are all waiting for Sebastian to fix.
+  So we are going the xml -> pdf -> ps route for the time being,
+    using pdfxmltex.
diff --git a/blocksort.c b/blocksort.c
index aba3efc..33ec9f5 100644
--- a/blocksort.c
+++ b/blocksort.c
@@ -8,7 +8,7 @@
   This file is a part of bzip2 and/or libbzip2, a program and
   library for lossless, block-sorting data compression.
 
-  Copyright (C) 1996-2002 Julian R Seward.  All rights reserved.
+  Copyright (C) 1996-2005 Julian R Seward.  All rights reserved.
 
   Redistribution and use in source and binary forms, with or without
   modification, are permitted provided that the following conditions
@@ -42,7 +42,7 @@
   SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
   Julian Seward, Cambridge, UK.
-  jseward@acm.org
+  jseward@bzip.org
   bzip2/libbzip2 version 1.0 of 21 March 2000
 
   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 @@
+<?xml version="1.0"?> <!-- -*- sgml -*- -->
+<xsl:stylesheet 
+     xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">
+
+<!-- we like '1.2 Title' -->
+<xsl:param name="section.autolabel" select="'1'"/> 
+<xsl:param name="section.label.includes.component.label" select="'1'"/>
+
+<!-- Do not put 'Chapter' at the start of eg 'Chapter 1. Doing This' -->
+<xsl:param name="local.l10n.xml" select="document('')"/> 
+<l:i18n xmlns:l="http://docbook.sourceforge.net/xmlns/l10n/1.0"> 
+  <l:l10n language="en"> 
+    <l:context name="title-numbered">
+      <l:template name="chapter" text="%n.&#160;%t"/>
+    </l:context> 
+  </l:l10n>
+</l:i18n>
+
+<!-- don't generate sub-tocs for qanda sets -->
+<xsl:param name="generate.toc">
+set       toc,title
+book      toc,title,figure,table,example,equation
+chapter   toc,title
+section   toc
+sect1     toc
+sect2     toc
+sect3     toc
+sect4     nop
+sect5     nop
+qandaset  toc
+qandadiv  nop
+appendix  toc,title
+article/appendix  nop
+article   toc,title
+preface   toc,title
+reference toc,title
+</xsl:param>
+
+</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 @@
+<?xml version="1.0" encoding="UTF-8"?> <!-- -*- sgml -*- -->
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" 
+     xmlns:fo="http://www.w3.org/1999/XSL/Format" version="1.0">
+
+<xsl:import href="http://docbook.sourceforge.net/release/xsl/current/fo/docbook.xsl"/>
+<xsl:import href="bz-common.xsl"/>
+
+<!-- set indent = yes while debugging, then change to NO -->
+<xsl:output method="xml" indent="yes"/>
+
+<!-- ensure only passivetex extensions are on -->
+<xsl:param name="stylesheet.result.type" select="'fo'"/>
+<!-- fo extensions: PDF bookmarks and index terms -->
+<xsl:param name="use.extensions" select="'1'"/>
+<xsl:param name="xep.extensions" select="0"/>      
+<xsl:param name="fop.extensions" select="0"/>     
+<xsl:param name="saxon.extensions" select="0"/>   
+<xsl:param name="passivetex.extensions" select="1"/>
+<xsl:param name="tablecolumns.extension" select="'1'"/>
+
+<!-- ensure we are using single sided -->
+<xsl:param name="double.sided" select="'0'"/> 
+
+<!-- insert cross references to page numbers -->
+<xsl:param name="insert.xref.page.number" select="1"/>
+
+<!-- <?custom-pagebreak?> inserts a page break at this point -->
+<xsl:template match="processing-instruction('custom-pagebreak')">
+  <fo:block break-before='page'/>
+</xsl:template>
+
+<!-- show links in color -->
+<xsl:attribute-set name="xref.properties">
+  <xsl:attribute name="color">blue</xsl:attribute>
+</xsl:attribute-set>
+
+<!-- make pre listings indented a bit + a bg colour -->
+<xsl:template match="programlisting | screen">
+  <fo:block start-indent="0.25in" wrap-option="no-wrap" 
+            white-space-collapse="false" text-align="start" 
+            font-family="monospace" background-color="#f2f2f9"
+            linefeed-treatment="preserve" 
+            xsl:use-attribute-sets="normal.para.spacing">
+    <xsl:apply-templates/>
+  </fo:block>
+</xsl:template>
+<!-- make verbatim output prettier -->
+<xsl:template match="literallayout">
+  <fo:block start-indent="0.25in" wrap-option="no-wrap" 
+            white-space-collapse="false" text-align="start" 
+            font-family="monospace" background-color="#edf7f4"
+            linefeed-treatment="preserve" 
+            space-before="0em" space-after="0em">
+    <xsl:apply-templates/>
+  </fo:block>
+</xsl:template>
+
+<!-- workaround bug in passivetex fo output for itemizedlist -->
+<xsl:template match="itemizedlist/listitem">
+  <xsl:variable name="id">
+  <xsl:call-template name="object.id"/></xsl:variable>
+  <xsl:variable name="itemsymbol">
+    <xsl:call-template name="list.itemsymbol">
+      <xsl:with-param name="node" select="parent::itemizedlist"/>
+    </xsl:call-template>
+  </xsl:variable>
+  <xsl:variable name="item.contents">
+    <fo:list-item-label end-indent="label-end()">
+      <fo:block>
+        <xsl:choose>
+          <xsl:when test="$itemsymbol='disc'">&#x2022;</xsl:when>
+          <xsl:when test="$itemsymbol='bullet'">&#x2022;</xsl:when>
+          <xsl:otherwise>&#x2022;</xsl:otherwise>
+        </xsl:choose>
+      </fo:block>
+    </fo:list-item-label>
+    <fo:list-item-body start-indent="body-start()">
+      <xsl:apply-templates/>    <!-- removed extra block wrapper -->
+    </fo:list-item-body>
+  </xsl:variable>
+  <xsl:choose>
+    <xsl:when test="parent::*/@spacing = 'compact'">
+      <fo:list-item id="{$id}" 
+          xsl:use-attribute-sets="compact.list.item.spacing">
+        <xsl:copy-of select="$item.contents"/>
+      </fo:list-item>
+    </xsl:when>
+    <xsl:otherwise>
+      <fo:list-item id="{$id}" xsl:use-attribute-sets="list.item.spacing">
+        <xsl:copy-of select="$item.contents"/>
+      </fo:list-item>
+    </xsl:otherwise>
+  </xsl:choose>
+</xsl:template>
+
+<!-- workaround bug in passivetex fo output for orderedlist -->
+<xsl:template match="orderedlist/listitem">
+  <xsl:variable name="id">
+  <xsl:call-template name="object.id"/></xsl:variable>
+  <xsl:variable name="item.contents">
+    <fo:list-item-label end-indent="label-end()">
+      <fo:block>
+        <xsl:apply-templates select="." mode="item-number"/>
+      </fo:block>
+    </fo:list-item-label>
+    <fo:list-item-body start-indent="body-start()">
+      <xsl:apply-templates/>    <!-- removed extra block wrapper -->
+    </fo:list-item-body>
+  </xsl:variable>
+  <xsl:choose>
+    <xsl:when test="parent::*/@spacing = 'compact'">
+      <fo:list-item id="{$id}" 
+          xsl:use-attribute-sets="compact.list.item.spacing">
+        <xsl:copy-of select="$item.contents"/>
+      </fo:list-item>
+    </xsl:when>
+    <xsl:otherwise>
+      <fo:list-item id="{$id}" xsl:use-attribute-sets="list.item.spacing">
+        <xsl:copy-of select="$item.contents"/>
+      </fo:list-item>
+    </xsl:otherwise>
+  </xsl:choose>
+</xsl:template>
+
+<!-- workaround bug in passivetex fo output for variablelist -->
+<xsl:param name="variablelist.as.blocks" select="1"/>
+<xsl:template match="varlistentry" mode="vl.as.blocks">
+  <xsl:variable name="id">
+    <xsl:call-template name="object.id"/></xsl:variable>
+  <fo:block id="{$id}" xsl:use-attribute-sets="list.item.spacing"  
+      keep-together.within-column="always" 
+      keep-with-next.within-column="always">
+    <xsl:apply-templates select="term"/>
+  </fo:block>
+  <fo:block start-indent="0.5in" end-indent="0in" 
+            space-after.minimum="0.2em" 
+            space-after.optimum="0.4em" 
+            space-after.maximum="0.6em">
+    <fo:block>
+      <xsl:apply-templates select="listitem"/>
+    </fo:block>
+  </fo:block>
+</xsl:template>
+
+
+<!-- workaround bug in footers: force right-align w/two 80|30 cols -->
+<xsl:template name="footer.table">
+  <xsl:param name="pageclass" select="''"/>
+  <xsl:param name="sequence" select="''"/>
+  <xsl:param name="gentext-key" select="''"/>
+  <xsl:choose>
+    <xsl:when test="$pageclass = 'index'">
+      <xsl:attribute name="margin-left">0pt</xsl:attribute>
+    </xsl:when>
+  </xsl:choose>
+  <xsl:variable name="candidate">
+    <fo:table table-layout="fixed" width="100%">
+      <fo:table-column column-number="1" column-width="80%"/>
+      <fo:table-column column-number="2" column-width="20%"/>
+      <fo:table-body>
+        <fo:table-row height="14pt">
+          <fo:table-cell text-align="left" display-align="after">
+            <xsl:attribute name="relative-align">baseline</xsl:attribute>
+            <fo:block> 
+              <fo:block> </fo:block><!-- empty cell -->
+            </fo:block>
+          </fo:table-cell>
+          <fo:table-cell text-align="center" display-align="after">
+            <xsl:attribute name="relative-align">baseline</xsl:attribute>
+            <fo:block>
+              <xsl:call-template name="footer.content">
+                <xsl:with-param name="pageclass" select="$pageclass"/>
+                <xsl:with-param name="sequence" select="$sequence"/>
+                <xsl:with-param name="position" select="'center'"/>
+                <xsl:with-param name="gentext-key" select="$gentext-key"/>
+              </xsl:call-template>
+            </fo:block>
+          </fo:table-cell>
+        </fo:table-row>
+      </fo:table-body>
+    </fo:table>
+  </xsl:variable>
+  <!-- Really output a footer? -->
+  <xsl:choose>
+    <xsl:when test="$pageclass='titlepage' and $gentext-key='book'
+                    and $sequence='first'">
+      <!-- no, book titlepages have no footers at all -->
+    </xsl:when>
+    <xsl:when test="$sequence = 'blank' and $footers.on.blank.pages = 0">
+      <!-- no output -->
+    </xsl:when>
+    <xsl:otherwise>
+      <xsl:copy-of select="$candidate"/>
+    </xsl:otherwise>
+  </xsl:choose>
+</xsl:template>
+
+
+<!-- fix bug in headers: force right-align w/two 40|60 cols -->
+<xsl:template name="header.table">
+  <xsl:param name="pageclass" select="''"/>
+  <xsl:param name="sequence" select="''"/>
+  <xsl:param name="gentext-key" select="''"/>
+  <xsl:choose>
+    <xsl:when test="$pageclass = 'index'">
+      <xsl:attribute name="margin-left">0pt</xsl:attribute>
+    </xsl:when>
+  </xsl:choose>
+  <xsl:variable name="candidate">
+    <fo:table table-layout="fixed" width="100%">
+      <xsl:call-template name="head.sep.rule">
+        <xsl:with-param name="pageclass" select="$pageclass"/>
+        <xsl:with-param name="sequence" select="$sequence"/>
+        <xsl:with-param name="gentext-key" select="$gentext-key"/>
+      </xsl:call-template>
+      <fo:table-column column-number="1" column-width="40%"/>
+      <fo:table-column column-number="2" column-width="60%"/>
+      <fo:table-body>
+        <fo:table-row height="14pt">
+          <fo:table-cell text-align="left" display-align="before">
+            <xsl:attribute name="relative-align">baseline</xsl:attribute>
+            <fo:block>
+              <fo:block> </fo:block><!-- empty cell -->
+            </fo:block>
+          </fo:table-cell>
+          <fo:table-cell text-align="center" display-align="before">
+            <xsl:attribute name="relative-align">baseline</xsl:attribute>
+            <fo:block>
+              <xsl:call-template name="header.content">
+                <xsl:with-param name="pageclass" select="$pageclass"/>
+                <xsl:with-param name="sequence" select="$sequence"/>
+                <xsl:with-param name="position" select="'center'"/>
+                <xsl:with-param name="gentext-key" select="$gentext-key"/>
+              </xsl:call-template>
+            </fo:block>
+          </fo:table-cell>
+        </fo:table-row>
+      </fo:table-body>
+    </fo:table>
+  </xsl:variable>
+  <!-- Really output a header? -->
+  <xsl:choose>
+    <xsl:when test="$pageclass = 'titlepage' and $gentext-key = 'book'
+                    and $sequence='first'">
+      <!-- no, book titlepages have no headers at all -->
+    </xsl:when>
+    <xsl:when test="$sequence = 'blank' and $headers.on.blank.pages = 0">
+      <!-- no output -->
+    </xsl:when>
+    <xsl:otherwise>
+      <xsl:copy-of select="$candidate"/>
+    </xsl:otherwise>
+  </xsl:choose>
+</xsl:template>
+
+
+</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 @@
+<?xml version="1.0"?> <!-- -*- sgml -*- -->
+<!DOCTYPE xsl:stylesheet [ <!ENTITY bz-css SYSTEM "./bzip.css"> ]>
+
+<xsl:stylesheet 
+   xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">
+
+<xsl:import href="http://docbook.sourceforge.net/release/xsl/current/html/docbook.xsl"/>
+<xsl:import href="bz-common.xsl"/>
+
+<!-- use 8859-1 encoding -->
+<xsl:output method="html" encoding="ISO-8859-1" indent="yes"/>
+
+<!-- we include the css directly when generating one large file -->
+<xsl:template name="user.head.content">  
+  <style type="text/css" media="screen">
+    <xsl:text>&bz-css;</xsl:text>
+  </style>
+</xsl:template>
+
+</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 @@
+/* Colours:
+#74240f  dark brown      h1, h2, h3, h4
+#336699  medium blue     links
+#339999  turquoise       link hover colour
+#202020  almost black    general text
+#761596  purple          md5sum text
+#626262  dark gray       pre border
+#eeeeee  very light gray pre background
+#f2f2f9  very light blue nav table background
+#3366cc  medium blue     nav table border
+*/
+
+a, a:link, a:visited, a:active { color: #336699; }
+a:hover { color: #339999; }
+
+body { font: 80%/126% sans-serif; }
+h1, h2, h3, h4 { color: #74240f; }
+
+dt { color: #336699; font-weight: bold }
+dd { 
+ margin-left: 1.5em; 
+ padding-bottom: 0.8em;
+}
+
+/* -- ruler -- */
+div.hr_blue { 
+  height:  3px; 
+  background:#ffffff url("/images/hr_blue.png") repeat-x; }
+div.hr_blue hr { display:none; }
+
+/* release styles */
+#release p { margin-top: 0.4em; }
+#release .md5sum { color: #761596; }
+
+
+/* ------ styles for docs|manuals|howto ------ */
+/* -- lists -- */
+ul  { 
+ margin:     0px 4px 16px 16px;
+ padding:    0px;
+ list-style: url("/images/li-blue.png"); 
+}
+ul li { 
+ margin-bottom: 10px;
+}
+ul ul   { 
+ list-style-type:  none; 
+ list-style-image: none; 
+ margin-left:      0px; 
+}
+
+/* header / footer nav tables */
+table.nav {
+ border:     solid 1px #3366cc;
+ background: #f2f2f9;
+ background-color: #f2f2f9;
+ margin-bottom: 0.5em;
+}
+/* don't have underlined links in chunked nav menus */
+table.nav a { text-decoration: none; }
+table.nav a:hover { text-decoration: underline; }
+table.nav td { font-size: 85%; }
+
+code, tt, pre { font-size: 120%; }
+code, tt { color: #761596; }
+
+div.literallayout, pre.programlisting, pre.screen {
+ color:      #000000;
+ padding:    0.5em;
+ background: #eeeeee;
+ border:     1px solid #626262;
+ background-color: #eeeeee;
+ margin: 4px 0px 4px 0px; 
+}
diff --git a/bzip2.1 b/bzip2.1
index 623435c..d2c0661 100644
--- a/bzip2.1
+++ b/bzip2.1
@@ -1,7 +1,7 @@
 .PU
 .TH bzip2 1
 .SH NAME
-bzip2, bunzip2 \- a block-sorting file compressor, v1.0.2
+bzip2, bunzip2 \- a block-sorting file compressor, v1.0.3
 .br
 bzcat \- decompresses files to stdout
 .br
@@ -405,19 +405,19 @@ I/O error messages are not as helpful as they could be.
 tries hard to detect I/O errors and exit cleanly, but the details of
 what the problem is sometimes seem rather misleading.
 
-This manual page pertains to version 1.0.2 of
+This manual page pertains to version 1.0.3 of
 .I bzip2.  
 Compressed data created by this version is entirely forwards and
 backwards compatible with the previous public releases, versions
-0.1pl2, 0.9.0, 0.9.5, 1.0.0 and 1.0.1, but with the following
+0.1pl2, 0.9.0, 0.9.5, 1.0.0, 1.0.1 and 1.0.2, but with the following
 exception: 0.9.0 and above can correctly decompress multiple
 concatenated compressed files.  0.1pl2 cannot do this; it will stop
 after decompressing just the first file in the stream.
 
 .I bzip2recover
-versions prior to this one, 1.0.2, used 32-bit integers to represent
-bit positions in compressed files, so it could not handle compressed
-files more than 512 megabytes long.  Version 1.0.2 and above uses
+versions prior to 1.0.2 used 32-bit integers to represent
+bit positions in compressed files, so they could not handle compressed
+files more than 512 megabytes long.  Versions 1.0.2 and above use
 64-bit ints on some platforms which support them (GNU supported
 targets, and Windows).  To establish whether or not bzip2recover was
 built 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.
 
 
 .SH AUTHOR
-Julian Seward, jseward@acm.org.
+Julian Seward, jsewardbzip.org.
 
-http://sources.redhat.com/bzip2
+http://www.bzip.org
 
 The ideas embodied in
 .I bzip2
@@ -447,6 +447,7 @@ source distribution for pointers to sources of documentation.  Christian
 von Roques encouraged me to look for faster sorting algorithms, so as to
 speed up compression.  Bela Lubkin encouraged me to improve the
 worst-case compression performance.  
+Donna Robinson XMLised the documentation.
 The bz* scripts are derived from those of GNU gzip.
 Many people sent patches, helped
 with 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)
 
 
 NNAAMMEE
-       bzip2, bunzip2 - a block-sorting file compressor, v1.0.2
-       bzcat - decompresses files to stdout
-       bzip2recover - recovers data from damaged bzip2 files
+       bzip2, bunzip2 − a block�sorting file compressor, v1.0.3
+       bzcat − decompresses files to stdout
+       bzip2recover − recovers data from damaged bzip2 files
 
 
 SSYYNNOOPPSSIISS
-       bbzziipp22 [ --ccddffkkqqssttvvzzVVLL112233445566778899 ] [ _f_i_l_e_n_a_m_e_s _._._.  ]
-       bbuunnzziipp22 [ --ffkkvvssVVLL ] [ _f_i_l_e_n_a_m_e_s _._._.  ]
-       bbzzccaatt [ --ss ] [ _f_i_l_e_n_a_m_e_s _._._.  ]
+       bbzziipp22 [ −−ccddffkkqqssttvvzzVVLL112233445566778899 ] [ _f_i_l_e_n_a_m_e_s _._._.  ]
+       bbuunnzziipp22 [ −−ffkkvvssVVLL ] [ _f_i_l_e_n_a_m_e_s _._._.  ]
+       bbzzccaatt [ −−ss ] [ _f_i_l_e_n_a_m_e_s _._._.  ]
        bbzziipp22rreeccoovveerr _f_i_l_e_n_a_m_e
 
 
 DDEESSCCRRIIPPTTIIOONN
-       _b_z_i_p_2  compresses  files  using  the Burrows-Wheeler block
+       _b_z_i_p_2  compresses  files  using  the Burrows�Wheeler block
        sorting text compression algorithm,  and  Huffman  coding.
        Compression  is  generally  considerably  better than that
-       achieved by more conventional LZ77/LZ78-based compressors,
-       and  approaches  the performance of the PPM family of sta­
+       achieved by more conventional LZ77/LZ78�based compressors,
+       and  approaches  the performance of the PPM family of sta­
        tistical compressors.
 
-       The command-line options are deliberately very similar  to
+       The command�line options are deliberately very similar  to
        those of _G_N_U _g_z_i_p_, but they are not identical.
 
-       _b_z_i_p_2  expects  a list of file names to accompany the com­
-       mand-line flags.  Each file is replaced  by  a  compressed
+       _b_z_i_p_2  expects  a list of file names to accompany the com­
+       mand�line flags.  Each file is replaced  by  a  compressed
        version  of  itself,  with  the  name "original_name.bz2".
-       Each compressed file has the same modification date,  per­
-       missions, and, when possible, ownership as the correspond­
+       Each compressed file has the same modification date,  per­
+       missions, and, when possible, ownership as the correspond­
        ing original, so that these properties  can  be  correctly
        restored  at  decompression  time.   File name handling is
-       naive in the sense that there is no mechanism for preserv­
+       naive in the sense that there is no mechanism for preserv­
        ing  original file names, permissions, ownerships or dates
        in filesystems which lack these concepts, or have  serious
-       file name length restrictions, such as MS-DOS.
+       file name length restrictions, such as MS�DOS.
 
        _b_z_i_p_2  and  _b_u_n_z_i_p_2 will by default not overwrite existing
-       files.  If you want this to happen, specify the -f flag.
+       files.  If you want this to happen, specify the −f flag.
 
        If no file names  are  specified,  _b_z_i_p_2  compresses  from
        standard  input  to  standard output.  In this case, _b_z_i_p_2
@@ -47,7 +47,7 @@ DDEESSCCRRIIPPTTIIOONN
        this  would  be  entirely  incomprehensible  and therefore
        pointless.
 
-       _b_u_n_z_i_p_2 (or _b_z_i_p_2 _-_d_) decompresses  all  specified  files.
+       _b_u_n_z_i_p_2 (or _b_z_i_p_2 _−_d_) decompresses  all  specified  files.
        Files which were not created by _b_z_i_p_2 will be detected and
        ignored, and a warning issued.  _b_z_i_p_2  attempts  to  guess
        the  filename  for  the decompressed file from that of the
@@ -64,26 +64,26 @@ DDEESSCCRRIIPPTTIIOONN
        guess the name of the original file, and uses the original
        name with _._o_u_t appended.
 
-       As  with compression, supplying no filenames causes decom­
+       As  with compression, supplying no filenames causes decom­
        pression from standard input to standard output.
 
-       _b_u_n_z_i_p_2 will correctly decompress a file which is the con­
+       _b_u_n_z_i_p_2 will correctly decompress a file which is the con­
        catenation of two or more compressed files.  The result is
        the concatenation of the corresponding uncompressed files.
-       Integrity testing (-t) of concatenated compressed files is
+       Integrity testing (−t) of concatenated compressed files is
        also supported.
 
        You can also compress or decompress files to the  standard
-       output  by giving the -c flag.  Multiple files may be com­
+       output  by giving the −c flag.  Multiple files may be com­
        pressed and decompressed like this.  The resulting outputs
        are  fed  sequentially to stdout.  Compression of multiple
-       files in this manner generates a stream containing  multi­
+       files in this manner generates a stream containing  multi­
        ple compressed file representations.  Such a stream can be
        decompressed correctly only  by  _b_z_i_p_2  version  0.9.0  or
-       later.   Earlier  versions of _b_z_i_p_2 will stop after decom­
+       later.   Earlier  versions of _b_z_i_p_2 will stop after decom­
        pressing the first file in the stream.
 
-       _b_z_c_a_t (or _b_z_i_p_2 _-_d_c_) decompresses all specified  files  to
+       _b_z_c_a_t (or _b_z_i_p_2 _�_d_c_) decompresses all specified  files  to
        the standard output.
 
        _b_z_i_p_2  will  read arguments from the environment variables
@@ -99,15 +99,15 @@ DDEESSCCRRIIPPTTIIOONN
        most  file  compressors)  is  coded at about 8.05 bits per
        byte, giving an expansion of around 0.5%.
 
-       As a self-check for your  protection,  _b_z_i_p_2  uses  32-bit
+       As a self�check for your  protection,  _b_z_i_p_2  uses  32�bit
        CRCs  to make sure that the decompressed version of a file
-       is identical to the original.  This guards against corrup­
+       is identical to the original.  This guards against corrup­
        tion  of  the compressed data, and against undetected bugs
        in _b_z_i_p_2 (hopefully very unlikely).  The chances  of  data
        corruption  going  undetected  is  microscopic,  about one
        chance in four billion for each file processed.  Be aware,
        though,  that  the  check occurs upon decompression, so it
-       can only tell you that something is wrong.  It can't  help
+       can only tell you that something is wrong.  It can’t  help
        you  recover  the original uncompressed data.  You can use
        _b_z_i_p_2_r_e_c_o_v_e_r to try to recover data from damaged files.
 
@@ -118,41 +118,41 @@ DDEESSCCRRIIPPTTIIOONN
 
 
 OOPPTTIIOONNSS
-       --cc ----ssttddoouutt
+       −−cc ����ssttddoouutt
               Compress or decompress to standard output.
 
-       --dd ----ddeeccoommpprreessss
+       −−dd ����ddeeccoommpprreessss
               Force  decompression.  _b_z_i_p_2_, _b_u_n_z_i_p_2 and _b_z_c_a_t are
               really the same program,  and  the  decision  about
               what  actions to take is done on the basis of which
               name is used.  This flag overrides that  mechanism,
               and forces _b_z_i_p_2 to decompress.
 
-       --zz ----ccoommpprreessss
-              The   complement   to   -d:   forces   compression,
+       −−zz ����ccoommpprreessss
+              The   complement   to   −d:   forces   compression,
               regardless of the invocation name.
 
-       --tt ----tteesstt
-              Check integrity of the specified file(s), but don't
+       −−tt ����tteesstt
+              Check integrity of the specified file(s), but don’t
               decompress  them.   This  really  performs  a trial
               decompression and throws away the result.
 
-       --ff ----ffoorrccee
+       −−ff ����ffoorrccee
               Force overwrite of output files.   Normally,  _b_z_i_p_2
               will  not  overwrite  existing  output files.  Also
               forces _b_z_i_p_2 to break hard links to files, which it
-              otherwise wouldn't do.
+              otherwise wouldn’t do.
 
               bzip2  normally  declines to decompress files which
-              don't have the  correct  magic  header  bytes.   If
-              forced  (-f),  however,  it  will  pass  such files
+              don’t have the  correct  magic  header  bytes.   If
+              forced  (�f),  however,  it  will  pass  such files
               through unmodified.  This is how GNU gzip  behaves.
 
-       --kk ----kkeeeepp
-              Keep  (don't delete) input files during compression
+       −−kk ����kkeeeepp
+              Keep  (don’t delete) input files during compression
               or decompression.
 
-       --ss ----ssmmaallll
+       −−ss ����ssmmaallll
               Reduce memory usage, for compression, decompression
               and  testing.   Files  are  decompressed and tested
               using a modified algorithm which only requires  2.5
@@ -160,46 +160,46 @@ OOPPTTIIOONNSS
               decompressed in 2300k of memory,  albeit  at  about
               half the normal speed.
 
-              During  compression,  -s  selects  a  block size of
+              During  compression,  −s  selects  a  block size of
               200k, which limits memory use to  around  the  same
               figure,  at  the expense of your compression ratio.
               In short, if your  machine  is  low  on  memory  (8
-              megabytes  or  less),  use  -s for everything.  See
+              megabytes  or  less),  use  −s for everything.  See
               MEMORY MANAGEMENT below.
 
-       --qq ----qquuiieett
-              Suppress non-essential warning messages.   Messages
+       −−qq ����qquuiieett
+              Suppress non�essential warning messages.   Messages
               pertaining  to I/O errors and other critical events
               will not be suppressed.
 
-       --vv ----vveerrbboossee
-              Verbose mode -- show the compression ratio for each
-              file  processed.   Further  -v's  increase the ver­
+       −−vv ����vveerrbboossee
+              Verbose mode �� show the compression ratio for each
+              file  processed.   Further  −v’s  increase the ver­
               bosity level, spewing out lots of information which
               is primarily of interest for diagnostic purposes.
 
-       --LL ----lliicceennssee --VV ----vveerrssiioonn
+       −−LL ����lliicceennssee ��VV ����vveerrssiioonn
               Display  the  software  version,  license terms and
               conditions.
 
-       --11 ((oorr ----ffaasstt)) ttoo --99 ((oorr ----bbeesstt))
+       −−11 ((oorr −−−−ffaasstt)) ttoo −−99 ((oorr −−−−bbeesstt))
               Set the block size to 100 k, 200 k ..  900  k  when
               compressing.   Has  no  effect  when decompressing.
-              See MEMORY MANAGEMENT below.  The --fast and --best
+              See MEMORY MANAGEMENT below.  The −−fast and −−best
               aliases  are  primarily for GNU gzip compatibility.
-              In particular, --fast doesn't make things  signifi­
-              cantly  faster.   And  --best  merely  selects  the
+              In particular, −−fast doesn’t make things  signifi­
+              cantly  faster.   And  −−best  merely  selects  the
               default behaviour.
 
-       ----     Treats all subsequent arguments as file names, even
-              if they start with a dash.  This is so you can han­
+       −−��     Treats all subsequent arguments as file names, even
+              if they start with a dash.  This is so you can han­
               dle files with names beginning  with  a  dash,  for
-              example: bzip2 -- -myfilename.
+              example: bzip2 −� −myfilename.
 
-       ----rreeppeettiittiivvee--ffaasstt ----rreeppeettiittiivvee--bbeesstt
+       −−��rreeppeettiittiivvee��ffaasstt ����rreeppeettiittiivvee��bbeesstt
               These  flags  are  redundant  in versions 0.9.5 and
               above.  They provided some coarse control over  the
-              behaviour  of the sorting algorithm in earlier ver­
+              behaviour  of the sorting algorithm in earlier ver­
               sions, which was sometimes useful.  0.9.5 and above
               have  an  improved  algorithm  which  renders these
               flags irrelevant.
@@ -209,13 +209,13 @@ MMEEMMOORRYY MMAANNAAGGEEMMEENNTT
        _b_z_i_p_2 compresses large files in blocks.   The  block  size
        affects  both  the  compression  ratio  achieved,  and the
        amount of memory needed for compression and decompression.
-       The  flags  -1  through  -9  specify  the block size to be
-       100,000 bytes through 900,000 bytes (the default)  respec­
+       The  flags  −1  through  −9  specify  the block size to be
+       100,000 bytes through 900,000 bytes (the default)  respec­
        tively.   At  decompression  time, the block size used for
        compression is read from  the  header  of  the  compressed
        file, and _b_u_n_z_i_p_2 then allocates itself just enough memory
        to decompress the file.  Since block sizes are  stored  in
-       compressed  files,  it follows that the flags -1 to -9 are
+       compressed  files,  it follows that the flags −1 to −9 are
        irrelevant to and so ignored during decompression.
 
        Compression and decompression requirements, in bytes,  can
@@ -238,21 +238,21 @@ MMEEMMOORRYY MMAANNAAGGEEMMEENNTT
        _b_u_n_z_i_p_2 will require about 3700 kbytes to decompress.   To
        support decompression of any file on a 4 megabyte machine,
        _b_u_n_z_i_p_2 has an option to  decompress  using  approximately
-       half this amount of memory, about 2300 kbytes.  Decompres­
+       half this amount of memory, about 2300 kbytes.  Decompres­
        sion speed is also halved, so you should use  this  option
-       only where necessary.  The relevant flag is -s.
+       only where necessary.  The relevant flag is �s.
 
-       In general, try and use the largest block size memory con­
+       In general, try and use the largest block size memory con­
        straints  allow,  since  that  maximises  the  compression
-       achieved.   Compression and decompression speed are virtu­
+       achieved.   Compression and decompression speed are virtu­
        ally unaffected by block size.
 
        Another significant point applies to files which fit in  a
-       single  block  --  that  means  most files you'd encounter
+       single  block  ��  that  means  most files you’d encounter
        using a large block  size.   The  amount  of  real  memory
        touched is proportional to the size of the file, since the
        file is smaller than a block.  For example, compressing  a
-       file  20,000  bytes  long  with the flag -9 will cause the
+       file  20,000  bytes  long  with the flag �9 will cause the
        compressor to allocate around 7600k of  memory,  but  only
        touch 400k + 20000 * 8 = 560 kbytes of it.  Similarly, the
        decompressor will allocate 3700k but  only  touch  100k  +
@@ -260,59 +260,59 @@ MMEEMMOORRYY MMAANNAAGGEEMMEENNTT
 
        Here  is a table which summarises the maximum memory usage
        for different block sizes.  Also  recorded  is  the  total
-       compressed  size for 14 files of the Calgary Text Compres­
+       compressed  size for 14 files of the Calgary Text Compres­
        sion Corpus totalling 3,141,622 bytes.  This column  gives
        some  feel  for  how  compression  varies with block size.
        These figures tend to understate the advantage  of  larger
-       block  sizes  for  larger files, since the Corpus is domi­
+       block  sizes  for  larger files, since the Corpus is domi­
        nated by smaller files.
 
                   Compress   Decompress   Decompress   Corpus
-           Flag     usage      usage       -s usage     Size
+           Flag     usage      usage       �s usage     Size
 
-            -1      1200k       500k         350k      914704
-            -2      2000k       900k         600k      877703
-            -3      2800k      1300k         850k      860338
-            -4      3600k      1700k        1100k      846899
-            -5      4400k      2100k        1350k      845160
-            -6      5200k      2500k        1600k      838626
-            -7      6100k      2900k        1850k      834096
-            -8      6800k      3300k        2100k      828642
-            -9      7600k      3700k        2350k      828642
+            �1      1200k       500k         350k      914704
+            �2      2000k       900k         600k      877703
+            �3      2800k      1300k         850k      860338
+            �4      3600k      1700k        1100k      846899
+            �5      4400k      2100k        1350k      845160
+            �6      5200k      2500k        1600k      838626
+            �7      6100k      2900k        1850k      834096
+            �8      6800k      3300k        2100k      828642
+            �9      7600k      3700k        2350k      828642
 
 
 RREECCOOVVEERRIINNGG DDAATTAA FFRROOMM DDAAMMAAGGEEDD FFIILLEESS
        _b_z_i_p_2 compresses files in blocks, usually 900kbytes  long.
-       Each block is handled independently.  If a media or trans­
-       mission error causes a multi-block  .bz2  file  to  become
+       Each block is handled independently.  If a media or trans­
+       mission error causes a multi�block  .bz2  file  to  become
        damaged,  it  may  be  possible  to  recover data from the
        undamaged blocks in the file.
 
        The compressed representation of each block  is  delimited
-       by  a  48-bit pattern, which makes it possible to find the
+       by  a  48�bit pattern, which makes it possible to find the
        block boundaries with reasonable  certainty.   Each  block
-       also  carries its own 32-bit CRC, so damaged blocks can be
+       also  carries its own 32�bit CRC, so damaged blocks can be
        distinguished from undamaged ones.
 
        _b_z_i_p_2_r_e_c_o_v_e_r is a  simple  program  whose  purpose  is  to
        search  for blocks in .bz2 files, and write each block out
-       into its own .bz2 file.  You can then use _b_z_i_p_2 -t to test
+       into its own .bz2 file.  You can then use _b_z_i_p_2 −t to test
        the integrity of the resulting files, and decompress those
        which are undamaged.
 
-       _b_z_i_p_2_r_e_c_o_v_e_r takes a single argument, the name of the dam­
+       _b_z_i_p_2_r_e_c_o_v_e_r takes a single argument, the name of the dam­
        aged    file,    and    writes    a    number   of   files
        "rec00001file.bz2",  "rec00002file.bz2",  etc,  containing
        the   extracted   blocks.   The   output   filenames   are
-       designed  so  that the use of wildcards in subsequent pro­
-       cessing  -- for example, "bzip2 -dc  rec*file.bz2 > recov­
-       ered_data" -- processes the files in the correct order.
+       designed  so  that the use of wildcards in subsequent pro­
+       cessing  �� for example, "bzip2 �dc  rec*file.bz2 > recov­
+       ered_data" �� processes the files in the correct order.
 
        _b_z_i_p_2_r_e_c_o_v_e_r should be of most use dealing with large .bz2
        files,  as  these will contain many blocks.  It is clearly
-       futile to use it on damaged single-block  files,  since  a
-       damaged  block  cannot  be recovered.  If you wish to min­
-       imise any potential data loss through media  or  transmis­
+       futile to use it on damaged single�block  files,  since  a
+       damaged  block  cannot  be recovered.  If you wish to min­
+       imise any potential data loss through media  or  transmis­
        sion errors, you might consider compressing with a smaller
        block size.
 
@@ -324,21 +324,21 @@ PPEERRFFOORRMMAANNCCEE NNOOTTEESS
        ..."   (repeated  several hundred times) may compress more
        slowly than normal.  Versions 0.9.5 and  above  fare  much
        better  than previous versions in this respect.  The ratio
-       between worst-case and average-case compression time is in
+       between worst�case and average�case compression time is in
        the  region  of  10:1.  For previous versions, this figure
-       was more like 100:1.  You can use the -vvvv option to mon­
+       was more like 100:1.  You can use the −vvvv option to mon­
        itor progress in great detail, if you want.
 
        Decompression speed is unaffected by these phenomena.
 
        _b_z_i_p_2  usually  allocates  several  megabytes of memory to
-       operate in, and then charges all over it in a fairly  ran­
-       dom  fashion.   This means that performance, both for com­
+       operate in, and then charges all over it in a fairly  ran­
+       dom  fashion.   This means that performance, both for com­
        pressing and decompressing, is largely determined  by  the
        speed  at  which  your  machine  can service cache misses.
        Because of this, small changes to the code to  reduce  the
        miss  rate  have  been observed to give disproportionately
-       large performance improvements.  I imagine _b_z_i_p_2 will per­
+       large performance improvements.  I imagine _b_z_i_p_2 will per­
        form best on machines with very large caches.
 
 
@@ -348,50 +348,51 @@ CCAAVVEEAATTSS
        but  the  details  of  what  the problem is sometimes seem
        rather misleading.
 
-       This manual page pertains to version 1.0.2 of _b_z_i_p_2_.  Com­
+       This manual page pertains to version 1.0.3 of _b_z_i_p_2_.  Com­
        pressed  data created by this version is entirely forwards
        and  backwards  compatible  with   the   previous   public
-       releases,  versions 0.1pl2, 0.9.0, 0.9.5, 1.0.0 and 1.0.1,
-       but with the following exception: 0.9.0 and above can cor­
-       rectly  decompress multiple concatenated compressed files.
-       0.1pl2 cannot do this; it will  stop  after  decompressing
-       just the first file in the stream.
+       releases,  versions 0.1pl2, 0.9.0, 0.9.5, 1.0.0, 1.0.1 and
+       1.0.2, but with the following exception: 0.9.0  and  above
+       can  correctly decompress multiple concatenated compressed
+       files.  0.1pl2 cannot do this; it will stop  after  decom­
+       pressing just the first file in the stream.
 
-       _b_z_i_p_2_r_e_c_o_v_e_r  versions  prior  to  this  one,  1.0.2, used
-       32-bit integers to represent bit positions  in  compressed
-       files,  so  it could not handle compressed files more than
-       512 megabytes long.  Version 1.0.2 and above  uses  64-bit
-       ints  on  some platforms which support them (GNU supported
-       targets,  and  Windows).   To  establish  whether  or  not
-       bzip2recover  was  built  with  such  a limitation, run it
-       without arguments.  In any event you can build yourself an
-       unlimited version if you can recompile it with MaybeUInt64
-       set to be an unsigned 64-bit integer.
+       _b_z_i_p_2_r_e_c_o_v_e_r  versions prior to 1.0.2 used 32�bit integers
+       to represent bit positions in compressed  files,  so  they
+       could  not handle compressed files more than 512 megabytes
+       long.  Versions 1.0.2 and above use 64�bit  ints  on  some
+       platforms  which  support them (GNU supported targets, and
+       Windows).  To establish whether or  not  bzip2recover  was
+       built  with  such  a limitation, run it without arguments.
+       In any event you can build yourself an  unlimited  version
+       if  you  can  recompile  it  with MaybeUInt64 set to be an
+       unsigned 64�bit integer.
 
 
 
 
 AAUUTTHHOORR
-       Julian Seward, jseward@acm.org.
+       Julian Seward, jsewardbzip.org.
 
-       http://sources.redhat.com/bzip2
+       http://www.bzip.org
 
-       The ideas embodied in _b_z_i_p_2 are due to (at least) the fol­
+       The ideas embodied in _b_z_i_p_2 are due to (at least) the fol­
        lowing  people: Michael Burrows and David Wheeler (for the
        block sorting transformation), David Wheeler  (again,  for
-       the Huffman coder), Peter Fenwick (for the structured cod­
+       the Huffman coder), Peter Fenwick (for the structured cod­
        ing model in the original _b_z_i_p_, and many refinements), and
        Alistair  Moffat,  Radford  Neal  and  Ian Witten (for the
        arithmetic  coder  in  the  original  _b_z_i_p_)_.   I  am  much
-       indebted for their help, support and advice.  See the man­
+       indebted for their help, support and advice.  See the man­
        ual in the source distribution for pointers to sources  of
        documentation.  Christian von Roques encouraged me to look
-       for faster sorting algorithms, so as to speed up  compres­
-       sion.  Bela Lubkin encouraged me to improve the worst-case
-       compression performance.  The bz* scripts are derived from
-       those  of GNU gzip.  Many people sent patches, helped with
-       portability problems, lent machines, gave advice and  were
-       generally helpful.
+       for faster sorting algorithms, so as to speed up  compres­
+       sion.  Bela Lubkin encouraged me to improve the worst�case
+       compression performance.  Donna Robinson XMLised the docu­
+       mentation.   The bz* scripts are derived from those of GNU
+       gzip.  Many people sent patches, helped  with  portability
+       problems,  lent  machines,  gave advice and were generally
+       helpful.
 
 
 
diff --git a/bzip2.c b/bzip2.c
index 807f420..79f87a5 100644
--- a/bzip2.c
+++ b/bzip2.c
@@ -7,7 +7,7 @@
   This file is a part of bzip2 and/or libbzip2, a program and
   library for lossless, block-sorting data compression.
 
-  Copyright (C) 1996-2002 Julian R Seward.  All rights reserved.
+  Copyright (C) 1996-2005 Julian R Seward.  All rights reserved.
 
   Redistribution and use in source and binary forms, with or without
   modification, are permitted provided that the following conditions
@@ -41,7 +41,7 @@
   SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
   Julian Seward, Cambridge, UK.
-  jseward@acm.org
+  jseward@bzip.org
   bzip2/libbzip2 version 1.0 of 21 March 2000
 
   This program is based on (at least) the work of:
@@ -525,6 +525,7 @@ Bool uncompressStream ( FILE *zStream, FILE *stream )
    UChar   obuf[5000];
    UChar   unused[BZ_MAX_UNUSED];
    Int32   nUnused;
+   void*   unusedTmpV;
    UChar*  unusedTmp;
 
    nUnused = 0;
@@ -554,9 +555,10 @@ Bool uncompressStream ( FILE *zStream, FILE *stream )
       }
       if (bzerr != BZ_STREAM_END) goto errhandler;
 
-      BZ2_bzReadGetUnused ( &bzerr, bzf, (void**)(&unusedTmp), &nUnused );
+      BZ2_bzReadGetUnused ( &bzerr, bzf, &unusedTmpV, &nUnused );
       if (bzerr != BZ_OK) panic ( "decompress:bzReadGetUnused" );
 
+      unusedTmp = (UChar*)unusedTmpV;
       for (i = 0; i < nUnused; i++) unused[i] = unusedTmp[i];
 
       BZ2_bzReadClose ( &bzerr, bzf );
@@ -639,6 +641,7 @@ Bool testStream ( FILE *zStream )
    UChar   obuf[5000];
    UChar   unused[BZ_MAX_UNUSED];
    Int32   nUnused;
+   void*   unusedTmpV;
    UChar*  unusedTmp;
 
    nUnused = 0;
@@ -662,9 +665,10 @@ Bool testStream ( FILE *zStream )
       }
       if (bzerr != BZ_STREAM_END) goto errhandler;
 
-      BZ2_bzReadGetUnused ( &bzerr, bzf, (void**)(&unusedTmp), &nUnused );
+      BZ2_bzReadGetUnused ( &bzerr, bzf, &unusedTmpV, &nUnused );
       if (bzerr != BZ_OK) panic ( "test:bzReadGetUnused" );
 
+      unusedTmp = (UChar*)unusedTmpV;
       for (i = 0; i < nUnused; i++) unused[i] = unusedTmp[i];
 
       BZ2_bzReadClose ( &bzerr, bzf );
@@ -828,7 +832,7 @@ void panic ( Char* s )
              "\n%s: PANIC -- internal consistency error:\n"
              "\t%s\n"
              "\tThis is a BUG.  Please report it to me at:\n"
-             "\tjseward@acm.org\n",
+             "\tjseward@bzip.org\n",
              progName, s );
    showFileNames();
    cleanUpAndFail( 3 );
@@ -908,7 +912,7 @@ void mySIGSEGVorSIGBUScatcher ( IntNative n )
       "   The user's manual, Section 4.3, has more info on (1) and (2).\n"
       "   \n"
       "   If you suspect this is a bug in bzip2, or are unsure about (1)\n"
-      "   or (2), feel free to report it to me at: jseward@acm.org.\n"
+      "   or (2), feel free to report it to me at: jseward@bzip.org.\n"
       "   Section 4.3 of the user's manual describes the info a useful\n"
       "   bug report should have.  If the manual is available on your\n"
       "   system, please try and read it before mailing me.  If you don't\n"
@@ -931,7 +935,7 @@ void mySIGSEGVorSIGBUScatcher ( IntNative n )
       "   The user's manual, Section 4.3, has more info on (2) and (3).\n"
       "   \n"
       "   If you suspect this is a bug in bzip2, or are unsure about (2)\n"
-      "   or (3), feel free to report it to me at: jseward@acm.org.\n"
+      "   or (3), feel free to report it to me at: jseward@bzip.org.\n"
       "   Section 4.3 of the user's manual describes the info a useful\n"
       "   bug report should have.  If the manual is available on your\n"
       "   system, please try and read it before mailing me.  If you don't\n"
@@ -1674,7 +1678,7 @@ void license ( void )
     "bzip2, a block-sorting file compressor.  "
     "Version %s.\n"
     "   \n"
-    "   Copyright (C) 1996-2002 by Julian Seward.\n"
+    "   Copyright (C) 1996-2005 by Julian Seward.\n"
     "   \n"
     "   This program is free software; you can redistribute it and/or modify\n"
     "   it under the terms set out in the LICENSE file, which is included\n"
diff --git a/bzip2.txt b/bzip2.txt
index 6afe358..bf895b6 100644
--- a/bzip2.txt
+++ b/bzip2.txt
@@ -1,6 +1,6 @@
 
 NAME
-       bzip2, bunzip2 - a block-sorting file compressor, v1.0.2
+       bzip2, bunzip2 - a block-sorting file compressor, v1.0.3
        bzcat - decompresses files to stdout
        bzip2recover - recovers data from damaged bzip2 files
 
@@ -17,20 +17,20 @@ DESCRIPTION
        sorting text compression algorithm,  and  Huffman  coding.
        Compression  is  generally  considerably  better than that
        achieved by more conventional LZ77/LZ78-based compressors,
-       and  approaches  the performance of the PPM family of sta­
+       and  approaches  the performance of the PPM family of sta-
        tistical compressors.
 
        The command-line options are deliberately very similar  to
        those of GNU gzip, but they are not identical.
 
-       bzip2  expects  a list of file names to accompany the com­
+       bzip2  expects  a list of file names to accompany the com-
        mand-line flags.  Each file is replaced  by  a  compressed
        version  of  itself,  with  the  name "original_name.bz2".
-       Each compressed file has the same modification date,  per­
-       missions, and, when possible, ownership as the correspond­
+       Each compressed file has the same modification date,  per-
+       missions, and, when possible, ownership as the correspond-
        ing original, so that these properties  can  be  correctly
        restored  at  decompression  time.   File name handling is
-       naive in the sense that there is no mechanism for preserv­
+       naive in the sense that there is no mechanism for preserv-
        ing  original file names, permissions, ownerships or dates
        in filesystems which lack these concepts, or have  serious
        file name length restrictions, such as MS-DOS.
@@ -61,23 +61,23 @@ DESCRIPTION
        guess the name of the original file, and uses the original
        name with .out appended.
 
-       As  with compression, supplying no filenames causes decom­
+       As  with compression, supplying no filenames causes decom-
        pression from standard input to standard output.
 
-       bunzip2 will correctly decompress a file which is the con­
+       bunzip2 will correctly decompress a file which is the con-
        catenation of two or more compressed files.  The result is
        the concatenation of the corresponding uncompressed files.
        Integrity testing (-t) of concatenated compressed files is
        also supported.
 
        You can also compress or decompress files to the  standard
-       output  by giving the -c flag.  Multiple files may be com­
+       output  by giving the -c flag.  Multiple files may be com-
        pressed and decompressed like this.  The resulting outputs
        are  fed  sequentially to stdout.  Compression of multiple
-       files in this manner generates a stream containing  multi­
+       files in this manner generates a stream containing  multi-
        ple compressed file representations.  Such a stream can be
        decompressed correctly only  by  bzip2  version  0.9.0  or
-       later.   Earlier  versions of bzip2 will stop after decom­
+       later.   Earlier  versions of bzip2 will stop after decom-
        pressing the first file in the stream.
 
        bzcat (or bzip2 -dc) decompresses all specified  files  to
@@ -98,7 +98,7 @@ DESCRIPTION
 
        As a self-check for your  protection,  bzip2  uses  32-bit
        CRCs  to make sure that the decompressed version of a file
-       is identical to the original.  This guards against corrup­
+       is identical to the original.  This guards against corrup-
        tion  of  the compressed data, and against undetected bugs
        in bzip2 (hopefully very unlikely).  The chances  of  data
        corruption  going  undetected  is  microscopic,  about one
@@ -171,7 +171,7 @@ OPTIONS
 
        -v --verbose
               Verbose mode -- show the compression ratio for each
-              file  processed.   Further  -v's  increase the ver­
+              file  processed.   Further  -v's  increase the ver-
               bosity level, spewing out lots of information which
               is primarily of interest for diagnostic purposes.
 
@@ -184,19 +184,19 @@ OPTIONS
               compressing.   Has  no  effect  when decompressing.
               See MEMORY MANAGEMENT below.  The --fast and --best
               aliases  are  primarily for GNU gzip compatibility.
-              In particular, --fast doesn't make things  signifi­
+              In particular, --fast doesn't make things  signifi-
               cantly  faster.   And  --best  merely  selects  the
               default behaviour.
 
        --     Treats all subsequent arguments as file names, even
-              if they start with a dash.  This is so you can han­
+              if they start with a dash.  This is so you can han-
               dle files with names beginning  with  a  dash,  for
               example: bzip2 -- -myfilename.
 
        --repetitive-fast --repetitive-best
               These  flags  are  redundant  in versions 0.9.5 and
               above.  They provided some coarse control over  the
-              behaviour  of the sorting algorithm in earlier ver­
+              behaviour  of the sorting algorithm in earlier ver-
               sions, which was sometimes useful.  0.9.5 and above
               have  an  improved  algorithm  which  renders these
               flags irrelevant.
@@ -207,7 +207,7 @@ MEMORY MANAGEMENT
        affects  both  the  compression  ratio  achieved,  and the
        amount of memory needed for compression and decompression.
        The  flags  -1  through  -9  specify  the block size to be
-       100,000 bytes through 900,000 bytes (the default)  respec­
+       100,000 bytes through 900,000 bytes (the default)  respec-
        tively.   At  decompression  time, the block size used for
        compression is read from  the  header  of  the  compressed
        file, and bunzip2 then allocates itself just enough memory
@@ -235,13 +235,13 @@ MEMORY MANAGEMENT
        bunzip2 will require about 3700 kbytes to decompress.   To
        support decompression of any file on a 4 megabyte machine,
        bunzip2 has an option to  decompress  using  approximately
-       half this amount of memory, about 2300 kbytes.  Decompres­
+       half this amount of memory, about 2300 kbytes.  Decompres-
        sion speed is also halved, so you should use  this  option
        only where necessary.  The relevant flag is -s.
 
-       In general, try and use the largest block size memory con­
+       In general, try and use the largest block size memory con-
        straints  allow,  since  that  maximises  the  compression
-       achieved.   Compression and decompression speed are virtu­
+       achieved.   Compression and decompression speed are virtu-
        ally unaffected by block size.
 
        Another significant point applies to files which fit in  a
@@ -257,11 +257,11 @@ MEMORY MANAGEMENT
 
        Here  is a table which summarises the maximum memory usage
        for different block sizes.  Also  recorded  is  the  total
-       compressed  size for 14 files of the Calgary Text Compres­
+       compressed  size for 14 files of the Calgary Text Compres-
        sion Corpus totalling 3,141,622 bytes.  This column  gives
        some  feel  for  how  compression  varies with block size.
        These figures tend to understate the advantage  of  larger
-       block  sizes  for  larger files, since the Corpus is domi­
+       block  sizes  for  larger files, since the Corpus is domi-
        nated by smaller files.
 
                   Compress   Decompress   Decompress   Corpus
@@ -280,7 +280,7 @@ MEMORY MANAGEMENT
 
 RECOVERING DATA FROM DAMAGED FILES
        bzip2 compresses files in blocks, usually 900kbytes  long.
-       Each block is handled independently.  If a media or trans­
+       Each block is handled independently.  If a media or trans-
        mission error causes a multi-block  .bz2  file  to  become
        damaged,  it  may  be  possible  to  recover data from the
        undamaged blocks in the file.
@@ -297,19 +297,19 @@ RECOVERING DATA FROM DAMAGED FILES
        the integrity of the resulting files, and decompress those
        which are undamaged.
 
-       bzip2recover takes a single argument, the name of the dam­
+       bzip2recover takes a single argument, the name of the dam-
        aged    file,    and    writes    a    number   of   files
        "rec00001file.bz2",  "rec00002file.bz2",  etc,  containing
        the   extracted   blocks.   The   output   filenames   are
-       designed  so  that the use of wildcards in subsequent pro­
-       cessing  -- for example, "bzip2 -dc  rec*file.bz2 > recov­
+       designed  so  that the use of wildcards in subsequent pro-
+       cessing  -- for example, "bzip2 -dc  rec*file.bz2 > recov-
        ered_data" -- processes the files in the correct order.
 
        bzip2recover should be of most use dealing with large .bz2
        files,  as  these will contain many blocks.  It is clearly
        futile to use it on damaged single-block  files,  since  a
-       damaged  block  cannot  be recovered.  If you wish to min­
-       imise any potential data loss through media  or  transmis­
+       damaged  block  cannot  be recovered.  If you wish to min-
+       imise any potential data loss through media  or  transmis-
        sion errors, you might consider compressing with a smaller
        block size.
 
@@ -323,19 +323,19 @@ PERFORMANCE NOTES
        better  than previous versions in this respect.  The ratio
        between worst-case and average-case compression time is in
        the  region  of  10:1.  For previous versions, this figure
-       was more like 100:1.  You can use the -vvvv option to mon­
+       was more like 100:1.  You can use the -vvvv option to mon-
        itor progress in great detail, if you want.
 
        Decompression speed is unaffected by these phenomena.
 
        bzip2  usually  allocates  several  megabytes of memory to
-       operate in, and then charges all over it in a fairly  ran­
-       dom  fashion.   This means that performance, both for com­
+       operate in, and then charges all over it in a fairly  ran-
+       dom  fashion.   This means that performance, both for com-
        pressing and decompressing, is largely determined  by  the
        speed  at  which  your  machine  can service cache misses.
        Because of this, small changes to the code to  reduce  the
        miss  rate  have  been observed to give disproportionately
-       large performance improvements.  I imagine bzip2 will per­
+       large performance improvements.  I imagine bzip2 will per-
        form best on machines with very large caches.
 
 
@@ -345,46 +345,47 @@ CAVEATS
        but  the  details  of  what  the problem is sometimes seem
        rather misleading.
 
-       This manual page pertains to version 1.0.2 of bzip2.  Com­
+       This manual page pertains to version 1.0.3 of bzip2.  Com-
        pressed  data created by this version is entirely forwards
        and  backwards  compatible  with   the   previous   public
-       releases,  versions 0.1pl2, 0.9.0, 0.9.5, 1.0.0 and 1.0.1,
-       but with the following exception: 0.9.0 and above can cor­
-       rectly  decompress multiple concatenated compressed files.
-       0.1pl2 cannot do this; it will  stop  after  decompressing
-       just the first file in the stream.
-
-       bzip2recover  versions  prior  to  this  one,  1.0.2, used
-       32-bit integers to represent bit positions  in  compressed
-       files,  so  it could not handle compressed files more than
-       512 megabytes long.  Version 1.0.2 and above  uses  64-bit
-       ints  on  some platforms which support them (GNU supported
-       targets,  and  Windows).   To  establish  whether  or  not
-       bzip2recover  was  built  with  such  a limitation, run it
-       without arguments.  In any event you can build yourself an
-       unlimited version if you can recompile it with MaybeUInt64
-       set to be an unsigned 64-bit integer.
+       releases,  versions 0.1pl2, 0.9.0, 0.9.5, 1.0.0, 1.0.1 and
+       1.0.2, but with the following exception: 0.9.0  and  above
+       can  correctly decompress multiple concatenated compressed
+       files.  0.1pl2 cannot do this; it will stop  after  decom-
+       pressing just the first file in the stream.
+
+       bzip2recover  versions prior to 1.0.2 used 32-bit integers
+       to represent bit positions in compressed  files,  so  they
+       could  not handle compressed files more than 512 megabytes
+       long.  Versions 1.0.2 and above use 64-bit  ints  on  some
+       platforms  which  support them (GNU supported targets, and
+       Windows).  To establish whether or  not  bzip2recover  was
+       built  with  such  a limitation, run it without arguments.
+       In any event you can build yourself an  unlimited  version
+       if  you  can  recompile  it  with MaybeUInt64 set to be an
+       unsigned 64-bit integer.
 
 
 AUTHOR
-       Julian Seward, jseward@acm.org.
+       Julian Seward, jsewardbzip.org.
 
-       http://sources.redhat.com/bzip2
+       http://www.bzip.org
 
-       The ideas embodied in bzip2 are due to (at least) the fol­
+       The ideas embodied in bzip2 are due to (at least) the fol-
        lowing  people: Michael Burrows and David Wheeler (for the
        block sorting transformation), David Wheeler  (again,  for
-       the Huffman coder), Peter Fenwick (for the structured cod­
+       the Huffman coder), Peter Fenwick (for the structured cod-
        ing model in the original bzip, and many refinements), and
        Alistair  Moffat,  Radford  Neal  and  Ian Witten (for the
        arithmetic  coder  in  the  original  bzip).   I  am  much
-       indebted for their help, support and advice.  See the man­
+       indebted for their help, support and advice.  See the man-
        ual in the source distribution for pointers to sources  of
        documentation.  Christian von Roques encouraged me to look
-       for faster sorting algorithms, so as to speed up  compres­
+       for faster sorting algorithms, so as to speed up  compres-
        sion.  Bela Lubkin encouraged me to improve the worst-case
-       compression performance.  The bz* scripts are derived from
-       those  of GNU gzip.  Many people sent patches, helped with
-       portability problems, lent machines, gave advice and  were
-       generally helpful.
+       compression performance.  Donna Robinson XMLised the docu-
+       mentation.   The bz* scripts are derived from those of GNU
+       gzip.  Many people sent patches, helped  with  portability
+       problems,  lent  machines,  gave advice and were generally
+       helpful.
 
diff --git a/bzip2recover.c b/bzip2recover.c
index 286873b..5cd405d 100644
--- a/bzip2recover.c
+++ b/bzip2recover.c
@@ -7,9 +7,9 @@
 /*--
   This program is bzip2recover, a program to attempt data 
   salvage from damaged files created by the accompanying
-  bzip2-1.0 program.
+  bzip2-1.0.3 program.
 
-  Copyright (C) 1996-2002 Julian R Seward.  All rights reserved.
+  Copyright (C) 1996-2005 Julian R Seward.  All rights reserved.
 
   Redistribution and use in source and binary forms, with or without
   modification, are permitted provided that the following conditions
@@ -43,8 +43,8 @@
   SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
   Julian Seward, Cambridge, UK.
-  jseward@acm.org
-  bzip2/libbzip2 version 1.0 of 21 March 2000
+  jseward@bzip.org
+  bzip2/libbzip2 version 1.0.3 of 15 February 2005
 --*/
 
 /*--
@@ -345,7 +345,7 @@ Int32 main ( Int32 argc, Char** argv )
    inFileName[0] = outFileName[0] = 0;
 
    fprintf ( stderr, 
-             "bzip2recover 1.0.2: extracts blocks from damaged .bz2 files.\n" );
+             "bzip2recover 1.0.3: extracts blocks from damaged .bz2 files.\n" );
 
    if (argc != 2) {
       fprintf ( stderr, "%s: usage is `%s damaged_file_name'.\n",
@@ -374,7 +374,7 @@ Int32 main ( Int32 argc, Char** argv )
    if (strlen(argv[1]) >= BZ_MAX_FILENAME-20) {
       fprintf ( stderr, 
                 "%s: supplied filename is suspiciously (>= %d chars) long.  Bye!\n",
-                progName, strlen(argv[1]) );
+                progName, (int)strlen(argv[1]) );
       exit(1);
    }
 
diff --git a/bzlib.c b/bzlib.c
index 7d1cb27..e9c1e87 100644
--- a/bzlib.c
+++ b/bzlib.c
@@ -8,7 +8,7 @@
   This file is a part of bzip2 and/or libbzip2, a program and
   library for lossless, block-sorting data compression.
 
-  Copyright (C) 1996-2002 Julian R Seward.  All rights reserved.
+  Copyright (C) 1996-2005 Julian R Seward.  All rights reserved.
 
   Redistribution and use in source and binary forms, with or without
   modification, are permitted provided that the following conditions
@@ -42,7 +42,7 @@
   SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
   Julian Seward, Cambridge, UK.
-  jseward@acm.org
+  jseward@bzip.org
   bzip2/libbzip2 version 1.0 of 21 March 2000
 
   This program is based on (at least) the work of:
@@ -88,12 +88,12 @@ void BZ2_bz__AssertH__fail ( int errcode )
    fprintf(stderr, 
       "\n\nbzip2/libbzip2: internal error number %d.\n"
       "This is a bug in bzip2/libbzip2, %s.\n"
-      "Please report it to me at: jseward@acm.org.  If this happened\n"
+      "Please report it to me at: jseward@bzip.org.  If this happened\n"
       "when you were using some program which uses libbzip2 as a\n"
       "component, you should also report this bug to the author(s)\n"
       "of that program.  Please make an effort to report this bug;\n"
       "timely and accurate bug reports eventually lead to higher\n"
-      "quality software.  Thanks.  Julian Seward, 30 December 2001.\n\n",
+      "quality software.  Thanks.  Julian Seward, 15 February 2005.\n\n",
       errcode,
       BZ2_bzlibVersion()
    );
@@ -574,8 +574,11 @@ int BZ_API(BZ2_bzDecompressInit)
 
 
 /*---------------------------------------------------*/
+/* Return  True iff data corruption is discovered.
+   Returns False if there is no problem.
+*/
 static
-void unRLE_obuf_to_output_FAST ( DState* s )
+Bool unRLE_obuf_to_output_FAST ( DState* s )
 {
    UChar k1;
 
@@ -584,7 +587,7 @@ void unRLE_obuf_to_output_FAST ( DState* s )
       while (True) {
          /* try to finish existing run */
          while (True) {
-            if (s->strm->avail_out == 0) return;
+            if (s->strm->avail_out == 0) return False;
             if (s->state_out_len == 0) break;
             *( (UChar*)(s->strm->next_out) ) = s->state_out_ch;
             BZ_UPDATE_CRC ( s->calculatedBlockCRC, s->state_out_ch );
@@ -594,10 +597,13 @@ void unRLE_obuf_to_output_FAST ( DState* s )
             s->strm->total_out_lo32++;
             if (s->strm->total_out_lo32 == 0) s->strm->total_out_hi32++;
          }
-   
+
          /* can a new run be started? */
-         if (s->nblock_used == s->save_nblock+1) return;
+         if (s->nblock_used == s->save_nblock+1) return False;
                
+         /* Only caused by corrupt data stream? */
+         if (s->nblock_used > s->save_nblock+1)
+            return True;
    
          s->state_out_len = 1;
          s->state_out_ch = s->k0;
@@ -667,6 +673,10 @@ void unRLE_obuf_to_output_FAST ( DState* s )
                cs_avail_out--;
             }
          }   
+         /* Only caused by corrupt data stream? */
+         if (c_nblock_used > s_save_nblockPP)
+            return True;
+
          /* can a new run be started? */
          if (c_nblock_used == s_save_nblockPP) {
             c_state_out_len = 0; goto return_notr;
@@ -712,6 +722,7 @@ void unRLE_obuf_to_output_FAST ( DState* s )
       s->strm->avail_out    = cs_avail_out;
       /* end save */
    }
+   return False;
 }
 
 
@@ -732,8 +743,11 @@ __inline__ Int32 BZ2_indexIntoF ( Int32 indx, Int32 *cftab )
 
 
 /*---------------------------------------------------*/
+/* Return  True iff data corruption is discovered.
+   Returns False if there is no problem.
+*/
 static
-void unRLE_obuf_to_output_SMALL ( DState* s )
+Bool unRLE_obuf_to_output_SMALL ( DState* s )
 {
    UChar k1;
 
@@ -742,7 +756,7 @@ void unRLE_obuf_to_output_SMALL ( DState* s )
       while (True) {
          /* try to finish existing run */
          while (True) {
-            if (s->strm->avail_out == 0) return;
+            if (s->strm->avail_out == 0) return False;
             if (s->state_out_len == 0) break;
             *( (UChar*)(s->strm->next_out) ) = s->state_out_ch;
             BZ_UPDATE_CRC ( s->calculatedBlockCRC, s->state_out_ch );
@@ -754,8 +768,11 @@ void unRLE_obuf_to_output_SMALL ( DState* s )
          }
    
          /* can a new run be started? */
-         if (s->nblock_used == s->save_nblock+1) return;
-               
+         if (s->nblock_used == s->save_nblock+1) return False;
+
+         /* Only caused by corrupt data stream? */
+         if (s->nblock_used > s->save_nblock+1)
+            return True;
    
          s->state_out_len = 1;
          s->state_out_ch = s->k0;
@@ -788,7 +805,7 @@ void unRLE_obuf_to_output_SMALL ( DState* s )
       while (True) {
          /* try to finish existing run */
          while (True) {
-            if (s->strm->avail_out == 0) return;
+            if (s->strm->avail_out == 0) return False;
             if (s->state_out_len == 0) break;
             *( (UChar*)(s->strm->next_out) ) = s->state_out_ch;
             BZ_UPDATE_CRC ( s->calculatedBlockCRC, s->state_out_ch );
@@ -800,7 +817,11 @@ void unRLE_obuf_to_output_SMALL ( DState* s )
          }
    
          /* can a new run be started? */
-         if (s->nblock_used == s->save_nblock+1) return;
+         if (s->nblock_used == s->save_nblock+1) return False;
+
+         /* Only caused by corrupt data stream? */
+         if (s->nblock_used > s->save_nblock+1)
+            return True;
    
          s->state_out_len = 1;
          s->state_out_ch = s->k0;
@@ -830,6 +851,7 @@ void unRLE_obuf_to_output_SMALL ( DState* s )
 /*---------------------------------------------------*/
 int BZ_API(BZ2_bzDecompress) ( bz_stream *strm )
 {
+   Bool    corrupt;
    DState* s;
    if (strm == NULL) return BZ_PARAM_ERROR;
    s = strm->state;
@@ -840,12 +862,13 @@ int BZ_API(BZ2_bzDecompress) ( bz_stream *strm )
       if (s->state == BZ_X_IDLE) return BZ_SEQUENCE_ERROR;
       if (s->state == BZ_X_OUTPUT) {
          if (s->smallDecompress)
-            unRLE_obuf_to_output_SMALL ( s ); else
-            unRLE_obuf_to_output_FAST  ( s );
+            corrupt = unRLE_obuf_to_output_SMALL ( s ); else
+            corrupt = unRLE_obuf_to_output_FAST  ( s );
+         if (corrupt) return BZ_DATA_ERROR;
          if (s->nblock_used == s->save_nblock+1 && s->state_out_len == 0) {
             BZ_FINALISE_CRC ( s->calculatedBlockCRC );
             if (s->verbosity >= 3) 
-               VPrintf2 ( " {0x%x, 0x%x}", s->storedBlockCRC, 
+               VPrintf2 ( " {0x%08x, 0x%08x}", s->storedBlockCRC, 
                           s->calculatedBlockCRC );
             if (s->verbosity >= 2) VPrintf0 ( "]" );
             if (s->calculatedBlockCRC != s->storedBlockCRC)
@@ -863,7 +886,7 @@ int BZ_API(BZ2_bzDecompress) ( bz_stream *strm )
          Int32 r = BZ2_decompress ( s );
          if (r == BZ_STREAM_END) {
             if (s->verbosity >= 3)
-               VPrintf2 ( "\n    combined CRCs: stored = 0x%x, computed = 0x%x", 
+               VPrintf2 ( "\n    combined CRCs: stored = 0x%08x, computed = 0x%08x", 
                           s->storedCombinedCRC, s->calculatedCombinedCRC );
             if (s->calculatedCombinedCRC != s->storedCombinedCRC)
                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 @@
   This file is a part of bzip2 and/or libbzip2, a program and
   library for lossless, block-sorting data compression.
 
-  Copyright (C) 1996-2002 Julian R Seward.  All rights reserved.
+  Copyright (C) 1996-2005 Julian R Seward.  All rights reserved.
 
   Redistribution and use in source and binary forms, with or without
   modification, are permitted provided that the following conditions
@@ -42,7 +42,7 @@
   SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
   Julian Seward, Cambridge, UK.
-  jseward@acm.org
+  jseward@bzip.org
   bzip2/libbzip2 version 1.0 of 21 March 2000
 
   This program is based on (at least) the work of:
@@ -110,8 +110,10 @@ typedef
 #define BZ_EXPORT
 #endif
 
+#ifndef BZ_NO_STDIO
 /* Need a definitition for FILE */
 #include <stdio.h>
+#endif
 
 #ifdef _WIN32
 #   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 @@
   This file is a part of bzip2 and/or libbzip2, a program and
   library for lossless, block-sorting data compression.
 
-  Copyright (C) 1996-2002 Julian R Seward.  All rights reserved.
+  Copyright (C) 1996-2005 Julian R Seward.  All rights reserved.
 
   Redistribution and use in source and binary forms, with or without
   modification, are permitted provided that the following conditions
@@ -42,7 +42,7 @@
   SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
   Julian Seward, Cambridge, UK.
-  jseward@acm.org
+  jseward@bzip.org
   bzip2/libbzip2 version 1.0 of 21 March 2000
 
   This program is based on (at least) the work of:
@@ -76,7 +76,7 @@
 
 /*-- General stuff. --*/
 
-#define BZ_VERSION  "1.0.2, 30-Dec-2001"
+#define BZ_VERSION  "1.0.3, 15-Feb-2005"
 
 typedef char            Char;
 typedef 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 @@
   This file is a part of bzip2 and/or libbzip2, a program and
   library for lossless, block-sorting data compression.
 
-  Copyright (C) 1996-2002 Julian R Seward.  All rights reserved.
+  Copyright (C) 1996-2005 Julian R Seward.  All rights reserved.
 
   Redistribution and use in source and binary forms, with or without
   modification, are permitted provided that the following conditions
@@ -42,7 +42,7 @@
   SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
   Julian Seward, Cambridge, UK.
-  jseward@acm.org
+  jseward@bzip.org
   bzip2/libbzip2 version 1.0 of 21 March 2000
 
   This program is based on (at least) the work of:
@@ -488,9 +488,11 @@ void sendMTFValues ( EState* s )
       /*--
         Recompute the tables based on the accumulated frequencies.
       --*/
+      /* maxLen was changed from 20 to 17 in bzip2-1.0.3.  See 
+         comment in huffman.c for details. */
       for (t = 0; t < nGroups; t++)
          BZ2_hbMakeCodeLengths ( &(s->len[t][0]), &(s->rfreq[t][0]), 
-                                 alphaSize, 20 );
+                                 alphaSize, 17 /*20*/ );
    }
 
 
@@ -527,7 +529,7 @@ void sendMTFValues ( EState* s )
          if (s->len[t][i] > maxLen) maxLen = s->len[t][i];
          if (s->len[t][i] < minLen) minLen = s->len[t][i];
       }
-      AssertH ( !(maxLen > 20), 3004 );
+      AssertH ( !(maxLen > 17 /*20*/ ), 3004 );
       AssertH ( !(minLen < 1),  3005 );
       BZ2_hbAssignCodes ( &(s->code[t][0]), &(s->len[t][0]), 
                           minLen, maxLen, alphaSize );
@@ -651,8 +653,8 @@ void BZ2_compressBlock ( EState* s, Bool is_last_block )
       if (s->blockNo > 1) s->numZ = 0;
 
       if (s->verbosity >= 2)
-         VPrintf4( "    block %d: crc = 0x%8x, "
-                   "combined CRC = 0x%8x, size = %d\n",
+         VPrintf4( "    block %d: crc = 0x%08x, "
+                   "combined CRC = 0x%08x, size = %d\n",
                    s->blockNo, s->blockCRC, s->combinedCRC, s->nblock );
 
       BZ2_blockSort ( s );
@@ -703,7 +705,7 @@ void BZ2_compressBlock ( EState* s, Bool is_last_block )
       bsPutUChar ( s, 0x50 ); bsPutUChar ( s, 0x90 );
       bsPutUInt32 ( s, s->combinedCRC );
       if (s->verbosity >= 2)
-         VPrintf1( "    final combined CRC = 0x%x\n   ", s->combinedCRC );
+         VPrintf1( "    final combined CRC = 0x%08x\n   ", s->combinedCRC );
       bsFinishWrite ( s );
    }
 }
diff --git a/crctable.c b/crctable.c
index b16746a..b6dadfc 100644
--- a/crctable.c
+++ b/crctable.c
@@ -8,7 +8,7 @@
   This file is a part of bzip2 and/or libbzip2, a program and
   library for lossless, block-sorting data compression.
 
-  Copyright (C) 1996-2002 Julian R Seward.  All rights reserved.
+  Copyright (C) 1996-2005 Julian R Seward.  All rights reserved.
 
   Redistribution and use in source and binary forms, with or without
   modification, are permitted provided that the following conditions
@@ -42,7 +42,7 @@
   SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
   Julian Seward, Cambridge, UK.
-  jseward@acm.org
+  jseward@bzip.org
   bzip2/libbzip2 version 1.0 of 21 March 2000
 
   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 @@
   This file is a part of bzip2 and/or libbzip2, a program and
   library for lossless, block-sorting data compression.
 
-  Copyright (C) 1996-2002 Julian R Seward.  All rights reserved.
+  Copyright (C) 1996-2005 Julian R Seward.  All rights reserved.
 
   Redistribution and use in source and binary forms, with or without
   modification, are permitted provided that the following conditions
@@ -42,7 +42,7 @@
   SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
   Julian Seward, Cambridge, UK.
-  jseward@acm.org
+  jseward@bzip.org
   bzip2/libbzip2 version 1.0 of 21 March 2000
 
   This program is based on (at least) the work of:
@@ -524,17 +524,23 @@ Int32 BZ2_decompress ( DState* s )
       if (s->origPtr < 0 || s->origPtr >= nblock)
          RETURN(BZ_DATA_ERROR);
 
+      /*-- Set up cftab to facilitate generation of T^(-1) --*/
+      s->cftab[0] = 0;
+      for (i = 1; i <= 256; i++) s->cftab[i] = s->unzftab[i-1];
+      for (i = 1; i <= 256; i++) s->cftab[i] += s->cftab[i-1];
+      for (i = 0; i <= 256; i++) {
+         if (s->cftab[i] < 0 || s->cftab[i] > nblock) {
+            /* s->cftab[i] can legitimately be == nblock */
+            RETURN(BZ_DATA_ERROR);
+         }
+      }
+
       s->state_out_len = 0;
       s->state_out_ch  = 0;
       BZ_INITIALISE_CRC ( s->calculatedBlockCRC );
       s->state = BZ_X_OUTPUT;
       if (s->verbosity >= 2) VPrintf0 ( "rt+rld" );
 
-      /*-- Set up cftab to facilitate generation of T^(-1) --*/
-      s->cftab[0] = 0;
-      for (i = 1; i <= 256; i++) s->cftab[i] = s->unzftab[i-1];
-      for (i = 1; i <= 256; i++) s->cftab[i] += s->cftab[i-1];
-
       if (s->smallDecompress) {
 
          /*-- 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 @@
+<!-- misc. strings -->
+<!ENTITY bz-url "http://www.bzip.org">
+<!ENTITY bz-email "jseward@bzip.org">
+<!ENTITY bz-lifespan "1996-2005">
+
+<!ENTITY bz-version "1.0.3">
+<!ENTITY bz-date "15 February 2005">
+
+<!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 @@
+#!/usr/bin/perl -w
+use strict;
+
+# get command line values:
+if ( $#ARGV !=1 ) {
+    die "Usage:  $0 xml_infile xml_outfile\n";
+}
+
+my $infile = shift;
+# check infile exists
+die "Can't find file \"$infile\""
+  unless -f $infile;
+# check we can read infile
+if (! -r $infile) {
+    die "Can't read input $infile\n";
+}
+# check we can open infile
+open( INFILE,"<$infile" ) or 
+    die "Can't input $infile $!";
+
+#my $outfile = 'fmt-manual.xml';
+my $outfile = shift;
+#print "Infile: $infile, Outfile: $outfile\n";
+# check we can write to outfile
+open( OUTFILE,">$outfile" ) or 
+    die "Can't output $outfile $! for writing";
+
+my ($prev, $curr, $str);
+$prev = ''; $curr = '';
+while ( <INFILE> ) {
+
+                print OUTFILE $prev;
+    $prev = $curr;
+    $curr = $_;
+    $str = '';
+
+    if ( $prev =~ /<programlisting>$|<screen>$/ ) {
+        chomp $prev;
+        $curr = join( '', $prev, "<![CDATA[", $curr );
+                                $prev = '';
+        next;
+    }
+    elsif ( $curr =~ /<\/programlisting>|<\/screen>/ ) {
+        chomp $prev;
+        $curr = join( '', $prev, "]]>", $curr );
+                                $prev = '';
+        next;
+    }
+}
+print OUTFILE $curr;
+close INFILE;
+close OUTFILE;
+exit;
diff --git a/huffman.c b/huffman.c
index 293095c..5bf190b 100644
--- a/huffman.c
+++ b/huffman.c
@@ -8,7 +8,7 @@
   This file is a part of bzip2 and/or libbzip2, a program and
   library for lossless, block-sorting data compression.
 
-  Copyright (C) 1996-2002 Julian R Seward.  All rights reserved.
+  Copyright (C) 1996-2005 Julian R Seward.  All rights reserved.
 
   Redistribution and use in source and binary forms, with or without
   modification, are permitted provided that the following conditions
@@ -42,7 +42,7 @@
   SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
   Julian Seward, Cambridge, UK.
-  jseward@acm.org
+  jseward@bzip.org
   bzip2/libbzip2 version 1.0 of 21 March 2000
 
   This program is based on (at least) the work of:
@@ -162,7 +162,24 @@ void BZ2_hbMakeCodeLengths ( UChar *len,
       
       if (! tooLong) break;
 
-      for (i = 1; i < alphaSize; i++) {
+      /* 17 Oct 04: keep-going condition for the following loop used
+         to be 'i < alphaSize', which missed the last element,
+         theoretically leading to the possibility of the compressor
+         looping.  However, this count-scaling step is only needed if
+         one of the generated Huffman code words is longer than
+         maxLen, which up to and including version 1.0.2 was 20 bits,
+         which is extremely unlikely.  In version 1.0.3 maxLen was
+         changed to 17 bits, which has minimal effect on compression
+         ratio, but does mean this scaling step is used from time to
+         time, enough to verify that it works.
+
+         This means that bzip2-1.0.3 and later will only produce
+         Huffman codes with a maximum length of 17 bits.  However, in
+         order to preserve backwards compatibility with bitstreams
+         produced by versions pre-1.0.3, the decompressor must still
+         handle lengths of up to 20. */
+
+      for (i = 1; i <= alphaSize; i++) {
          j = weight[i] >> 8;
          j = 1 + (j / 2);
          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 @@
-\input texinfo  @c                                  -*- Texinfo -*-
-@setfilename bzip2.info
-
-@ignore
-This file documents bzip2 version 1.0.2, and associated library
-libbzip2, written by Julian Seward (jseward@acm.org).
-
-Copyright (C) 1996-2002 Julian R Seward
-
-Permission is granted to make and distribute verbatim copies of
-this manual provided the copyright notice and this permission notice
-are preserved on all copies.
-
-Permission is granted to copy and distribute translations of this manual
-into another language, under the above conditions for verbatim copies.
-@end ignore
-
-@ifinfo
-@format
-START-INFO-DIR-ENTRY
-* Bzip2: (bzip2).               A program and library for data compression.
-END-INFO-DIR-ENTRY
-@end format
-
-@end ifinfo
-
-@iftex
-@c @finalout
-@settitle bzip2 and libbzip2
-@titlepage
-@title bzip2 and libbzip2
-@subtitle a program and library for data compression
-@subtitle copyright (C) 1996-2002 Julian Seward
-@subtitle version 1.0.2 of 30 December 2001
-@author Julian Seward
-
-@end titlepage
-
-@parindent 0mm
-@parskip 2mm
-
-@end iftex
-@node Top,,, (dir)
-
-The following text is the License for this software.  You should
-find it identical to that contained in the file LICENSE in the 
-source distribution.
-
-@bf{------------------ START OF THE LICENSE ------------------}
-
-This program, @code{bzip2}, 
-and associated library @code{libbzip2}, are
-Copyright (C) 1996-2002 Julian R Seward.  All rights reserved.
-
-Redistribution and use in source and binary forms, with or without
-modification, are permitted provided that the following conditions
-are met:
-@itemize @bullet
-@item
-   Redistributions of source code must retain the above copyright
-   notice, this list of conditions and the following disclaimer.
-@item
-   The origin of this software must not be misrepresented; you must 
-   not claim that you wrote the original software.  If you use this 
-   software in a product, an acknowledgment in the product 
-   documentation would be appreciated but is not required.
-@item
-   Altered source versions must be plainly marked as such, and must
-   not be misrepresented as being the original software.
-@item
-   The name of the author may not be used to endorse or promote 
-   products derived from this software without specific prior written 
-   permission.
-@end itemize
-THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS
-OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
-WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
-ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY
-DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
-DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE
-GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
-INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
-WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
-NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
-SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-
-Julian Seward, Cambridge, UK.
-
-@code{jseward@@acm.org}
-
-@code{bzip2}/@code{libbzip2} version 1.0.2 of 30 December 2001.
-
-@bf{------------------ END OF THE LICENSE ------------------}
-
-Web sites:
-
-@code{http://sources.redhat.com/bzip2}
-
-@code{http://www.cacheprof.org}
-
-PATENTS: To the best of my knowledge, @code{bzip2} does not use any patented
-algorithms.  However, I do not have the resources available to carry out
-a full patent search.  Therefore I cannot give any guarantee of the
-above statement.
-
-
-
-
-
-
-
-@chapter Introduction
-
-@code{bzip2}  compresses  files  using the Burrows-Wheeler 
-block-sorting text compression algorithm,  and  Huffman  coding.
-Compression  is  generally  considerably  better than that
-achieved by more conventional LZ77/LZ78-based compressors,
-and  approaches  the performance of the PPM family of statistical compressors.
-
-@code{bzip2} is built on top of @code{libbzip2}, a flexible library
-for handling compressed data in the @code{bzip2} format.  This manual
-describes both how to use the program and 
-how to work with the library interface.  Most of the
-manual is devoted to this library, not the program, 
-which is good news if your interest is only in the program.
-
-Chapter 2 describes how to use @code{bzip2}; this is the only part 
-you need to read if you just want to know how to operate the program.
-Chapter 3 describes the programming interfaces in detail, and
-Chapter 4 records some miscellaneous notes which I thought
-ought to be recorded somewhere.
-
-
-@chapter How to use @code{bzip2}
-
-This chapter contains a copy of the @code{bzip2} man page,
-and nothing else.
-
-@quotation
-
-@unnumberedsubsubsec NAME
-@itemize
-@item @code{bzip2}, @code{bunzip2}
-- a block-sorting file compressor, v1.0.2
-@item @code{bzcat} 
-- decompresses files to stdout
-@item @code{bzip2recover}
-- recovers data from damaged bzip2 files
-@end itemize
-
-@unnumberedsubsubsec SYNOPSIS
-@itemize
-@item @code{bzip2} [ -cdfkqstvzVL123456789 ] [ filenames ...  ]
-@item @code{bunzip2} [ -fkvsVL ] [ filenames ...  ]
-@item @code{bzcat} [ -s ] [ filenames ...  ]
-@item @code{bzip2recover} filename
-@end itemize
-
-@unnumberedsubsubsec DESCRIPTION
-
-@code{bzip2} compresses files using the Burrows-Wheeler block sorting
-text compression algorithm, and Huffman coding.  Compression is
-generally considerably better than that achieved by more conventional
-LZ77/LZ78-based compressors, and approaches the performance of the PPM
-family of statistical compressors.
-
-The command-line options are deliberately very similar to those of GNU
-@code{gzip}, but they are not identical.
-
-@code{bzip2} expects a list of file names to accompany the command-line
-flags.  Each file is replaced by a compressed version of itself, with
-the name @code{original_name.bz2}.  Each compressed file has the same
-modification date, permissions, and, when possible, ownership as the
-corresponding original, so that these properties can be correctly
-restored at decompression time.  File name handling is naive in the
-sense that there is no mechanism for preserving original file names,
-permissions, ownerships or dates in filesystems which lack these
-concepts, or have serious file name length restrictions, such as MS-DOS.
-
-@code{bzip2} and @code{bunzip2} will by default not overwrite existing
-files.  If you want this to happen, specify the @code{-f} flag.
-
-If no file names are specified, @code{bzip2} compresses from standard
-input to standard output.  In this case, @code{bzip2} will decline to
-write compressed output to a terminal, as this would be entirely
-incomprehensible and therefore pointless.
-
-@code{bunzip2} (or @code{bzip2 -d}) decompresses all
-specified files.  Files which were not created by @code{bzip2}
-will be detected and ignored, and a warning issued.  
-@code{bzip2} attempts to guess the filename for the decompressed file 
-from that of the compressed file as follows:
-@itemize
-@item @code{filename.bz2 } becomes @code{filename}
-@item @code{filename.bz  } becomes @code{filename}
-@item @code{filename.tbz2} becomes @code{filename.tar}
-@item @code{filename.tbz } becomes @code{filename.tar}
-@item @code{anyothername } becomes @code{anyothername.out}
-@end itemize
-If the file does not end in one of the recognised endings, 
-@code{.bz2}, @code{.bz}, 
-@code{.tbz2} or @code{.tbz}, @code{bzip2} complains that it cannot
-guess the name of the original file, and uses the original name
-with @code{.out} appended.
-
-As with compression, supplying no
-filenames causes decompression from standard input to standard output.
-
-@code{bunzip2} will correctly decompress a file which is the
-concatenation of two or more compressed files.  The result is the
-concatenation of the corresponding uncompressed files.  Integrity
-testing (@code{-t}) of concatenated compressed files is also supported.
-
-You can also compress or decompress files to the standard output by
-giving the @code{-c} flag.  Multiple files may be compressed and
-decompressed like this.  The resulting outputs are fed sequentially to
-stdout.  Compression of multiple files in this manner generates a stream
-containing multiple compressed file representations.  Such a stream
-can be decompressed correctly only by @code{bzip2} version 0.9.0 or
-later.  Earlier versions of @code{bzip2} will stop after decompressing
-the first file in the stream.
-
-@code{bzcat} (or @code{bzip2 -dc}) decompresses all specified files to
-the standard output.
-
-@code{bzip2} will read arguments from the environment variables
-@code{BZIP2} and @code{BZIP}, in that order, and will process them
-before any arguments read from the command line.  This gives a 
-convenient way to supply default arguments.
-
-Compression is always performed, even if the compressed file is slightly
-larger than the original.  Files of less than about one hundred bytes
-tend to get larger, since the compression mechanism has a constant
-overhead in the region of 50 bytes.  Random data (including the output
-of most file compressors) is coded at about 8.05 bits per byte, giving
-an expansion of around 0.5%.
-
-As a self-check for your protection, @code{bzip2} uses 32-bit CRCs to
-make sure that the decompressed version of a file is identical to the
-original.  This guards against corruption of the compressed data, and
-against undetected bugs in @code{bzip2} (hopefully very unlikely).  The
-chances of data corruption going undetected is microscopic, about one
-chance in four billion for each file processed.  Be aware, though, that
-the check occurs upon decompression, so it can only tell you that
-something is wrong.  It can't help you recover the original uncompressed
-data.  You can use @code{bzip2recover} to try to recover data from
-damaged files.
-
-Return values: 0 for a normal exit, 1 for environmental problems (file
-not found, invalid flags, I/O errors, &c), 2 to indicate a corrupt
-compressed file, 3 for an internal consistency error (eg, bug) which
-caused @code{bzip2} to panic.
-
-
-@unnumberedsubsubsec OPTIONS
-@table @code
-@item -c  --stdout
-Compress or decompress to standard output.
-@item -d  --decompress
-Force decompression.  @code{bzip2}, @code{bunzip2} and @code{bzcat} are
-really the same program, and the decision about what actions to take is
-done on the basis of which name is used.  This flag overrides that
-mechanism, and forces bzip2 to decompress.
-@item -z --compress
-The complement to @code{-d}: forces compression, regardless of the
-invokation name.
-@item -t --test
-Check integrity of the specified file(s), but don't decompress them.
-This really performs a trial decompression and throws away the result.
-@item -f --force
-Force overwrite of output files.  Normally, @code{bzip2} will not overwrite
-existing output files.  Also forces @code{bzip2} to break hard links
-to files, which it otherwise wouldn't do.
-
-@code{bzip2} normally declines to decompress files which don't have the
-correct magic header bytes.  If forced (@code{-f}), however, it will
-pass such files through unmodified.  This is how GNU @code{gzip}
-behaves.
-@item -k --keep
-Keep (don't delete) input files during compression
-or decompression.
-@item -s --small
-Reduce memory usage, for compression, decompression and testing.  Files
-are decompressed and tested using a modified algorithm which only
-requires 2.5 bytes per block byte.  This means any file can be
-decompressed in 2300k of memory, albeit at about half the normal speed.
-
-During compression, @code{-s} selects a block size of 200k, which limits
-memory use to around the same figure, at the expense of your compression
-ratio.  In short, if your machine is low on memory (8 megabytes or
-less), use -s for everything.  See MEMORY MANAGEMENT below.
-@item -q --quiet
-Suppress non-essential warning messages.  Messages pertaining to
-I/O errors and other critical events will not be suppressed.
-@item -v --verbose
-Verbose mode -- show the compression ratio for each file processed.
-Further @code{-v}'s increase the verbosity level, spewing out lots of
-information which is primarily of interest for diagnostic purposes.
-@item -L --license -V --version
-Display the software version, license terms and conditions.
-@item -1 (or --fast) to -9 (or --best)
-Set the block size to 100 k, 200 k ..  900 k when compressing.  Has no
-effect when decompressing.  See MEMORY MANAGEMENT below.
-The @code{--fast} and @code{--best} aliases are primarily for GNU
-@code{gzip} compatibility.  In particular, @code{--fast} doesn't make
-things significantly faster.  And @code{--best} merely selects the
-default behaviour.
-@item --
-Treats all subsequent arguments as file names, even if they start
-with a dash.  This is so you can handle files with names beginning
-with a dash, for example: @code{bzip2 -- -myfilename}.
-@item --repetitive-fast 
-@item --repetitive-best
-These flags are redundant in versions 0.9.5 and above.  They provided
-some coarse control over the behaviour of the sorting algorithm in
-earlier versions, which was sometimes useful.  0.9.5 and above have an
-improved algorithm which renders these flags irrelevant.
-@end table
-
-
-@unnumberedsubsubsec MEMORY MANAGEMENT
-
-@code{bzip2} compresses large files in blocks.  The block size affects
-both the compression ratio achieved, and the amount of memory needed for
-compression and decompression.  The flags @code{-1} through @code{-9}
-specify the block size to be 100,000 bytes through 900,000 bytes (the
-default) respectively.  At decompression time, the block size used for
-compression is read from the header of the compressed file, and
-@code{bunzip2} then allocates itself just enough memory to decompress
-the file.  Since block sizes are stored in compressed files, it follows
-that the flags @code{-1} to @code{-9} are irrelevant to and so ignored
-during decompression.
-
-Compression and decompression requirements, in bytes, can be estimated
-as:
-@example
-     Compression:   400k + ( 8 x block size )
-
-     Decompression: 100k + ( 4 x block size ), or
-                    100k + ( 2.5 x block size )
-@end example
-Larger block sizes give rapidly diminishing marginal returns.  Most of
-the compression comes from the first two or three hundred k of block
-size, a fact worth bearing in mind when using @code{bzip2} on small machines.
-It is also important to appreciate that the decompression memory
-requirement is set at compression time by the choice of block size.
-
-For files compressed with the default 900k block size, @code{bunzip2}
-will require about 3700 kbytes to decompress.  To support decompression
-of any file on a 4 megabyte machine, @code{bunzip2} has an option to
-decompress using approximately half this amount of memory, about 2300
-kbytes.  Decompression speed is also halved, so you should use this
-option only where necessary.  The relevant flag is @code{-s}.
-
-In general, try and use the largest block size memory constraints allow,
-since that maximises the compression achieved.  Compression and
-decompression speed are virtually unaffected by block size.
-
-Another significant point applies to files which fit in a single block
--- that means most files you'd encounter using a large block size.  The
-amount of real memory touched is proportional to the size of the file,
-since the file is smaller than a block.  For example, compressing a file
-20,000 bytes long with the flag @code{-9} will cause the compressor to
-allocate around 7600k of memory, but only touch 400k + 20000 * 8 = 560
-kbytes of it.  Similarly, the decompressor will allocate 3700k but only
-touch 100k + 20000 * 4 = 180 kbytes.
-
-Here is a table which summarises the maximum memory usage for different
-block sizes.  Also recorded is the total compressed size for 14 files of
-the Calgary Text Compression Corpus totalling 3,141,622 bytes.  This
-column gives some feel for how compression varies with block size.
-These figures tend to understate the advantage of larger block sizes for
-larger files, since the Corpus is dominated by smaller files.
-@example
-          Compress   Decompress   Decompress   Corpus
-   Flag     usage      usage       -s usage     Size
-
-    -1      1200k       500k         350k      914704
-    -2      2000k       900k         600k      877703
-    -3      2800k      1300k         850k      860338
-    -4      3600k      1700k        1100k      846899
-    -5      4400k      2100k        1350k      845160
-    -6      5200k      2500k        1600k      838626
-    -7      6100k      2900k        1850k      834096
-    -8      6800k      3300k        2100k      828642
-    -9      7600k      3700k        2350k      828642
-@end example
-
-@unnumberedsubsubsec RECOVERING DATA FROM DAMAGED FILES
-
-@code{bzip2} compresses files in blocks, usually 900kbytes long.  Each
-block is handled independently.  If a media or transmission error causes
-a multi-block @code{.bz2} file to become damaged, it may be possible to
-recover data from the undamaged blocks in the file.
-
-The compressed representation of each block is delimited by a 48-bit
-pattern, which makes it possible to find the block boundaries with
-reasonable certainty.  Each block also carries its own 32-bit CRC, so
-damaged blocks can be distinguished from undamaged ones.
-
-@code{bzip2recover} is a simple program whose purpose is to search for
-blocks in @code{.bz2} files, and write each block out into its own
-@code{.bz2} file.  You can then use @code{bzip2 -t} to test the
-integrity of the resulting files, and decompress those which are
-undamaged.
-
-@code{bzip2recover} 
-takes a single argument, the name of the damaged file, and writes a
-number of files @code{rec00001file.bz2}, @code{rec00002file.bz2}, etc,
-containing the extracted blocks.  The output filenames are designed so
-that the use of wildcards in subsequent processing -- for example,
-@code{bzip2 -dc rec*file.bz2 > recovered_data} -- processes the files in
-the correct order.
-
-@code{bzip2recover} should be of most use dealing with large @code{.bz2}
-files, as these will contain many blocks.  It is clearly futile to use
-it on damaged single-block files, since a damaged block cannot be
-recovered.  If you wish to minimise any potential data loss through
-media or transmission errors, you might consider compressing with a
-smaller block size.
-
-
-@unnumberedsubsubsec PERFORMANCE NOTES
-
-The sorting phase of compression gathers together similar strings in the
-file.  Because of this, files containing very long runs of repeated
-symbols, like "aabaabaabaab ..."  (repeated several hundred times) may
-compress more slowly than normal.  Versions 0.9.5 and above fare much
-better than previous versions in this respect.  The ratio between
-worst-case and average-case compression time is in the region of 10:1.
-For previous versions, this figure was more like 100:1.  You can use the
-@code{-vvvv} option to monitor progress in great detail, if you want.
-
-Decompression speed is unaffected by these phenomena.
-
-@code{bzip2} usually allocates several megabytes of memory to operate
-in, and then charges all over it in a fairly random fashion.  This means
-that performance, both for compressing and decompressing, is largely
-determined by the speed at which your machine can service cache misses.
-Because of this, small changes to the code to reduce the miss rate have
-been observed to give disproportionately large performance improvements.
-I imagine @code{bzip2} will perform best on machines with very large
-caches.
-
-
-@unnumberedsubsubsec CAVEATS
-
-I/O error messages are not as helpful as they could be.  @code{bzip2}
-tries hard to detect I/O errors and exit cleanly, but the details of
-what the problem is sometimes seem rather misleading.
-
-This manual page pertains to version 1.0.2 of @code{bzip2}.  Compressed
-data created by this version is entirely forwards and backwards
-compatible with the previous public releases, versions 0.1pl2, 0.9.0,
-0.9.5, 1.0.0 and 1.0.1, but with the following exception: 0.9.0 and
-above can correctly decompress multiple concatenated compressed files.
-0.1pl2 cannot do this; it will stop after decompressing just the first
-file in the stream.
-
-@code{bzip2recover} versions prior to this one, 1.0.2, used 32-bit
-integers to represent bit positions in compressed files, so it could not
-handle compressed files more than 512 megabytes long.  Version 1.0.2 and
-above uses 64-bit ints on some platforms which support them (GNU
-supported targets, and Windows).  To establish whether or not
-@code{bzip2recover} was built with such a limitation, run it without
-arguments.  In any event you can build yourself an unlimited version if
-you can recompile it with @code{MaybeUInt64} set to be an unsigned
-64-bit integer.
-
-
-
-@unnumberedsubsubsec AUTHOR
-Julian Seward, @code{jseward@@acm.org}.
-
-@code{http://sources.redhat.com/bzip2}
-
-The ideas embodied in @code{bzip2} are due to (at least) the following
-people: Michael Burrows and David Wheeler (for the block sorting
-transformation), David Wheeler (again, for the Huffman coder), Peter
-Fenwick (for the structured coding model in the original @code{bzip},
-and many refinements), and Alistair Moffat, Radford Neal and Ian Witten
-(for the arithmetic coder in the original @code{bzip}).  I am much
-indebted for their help, support and advice.  See the manual in the
-source distribution for pointers to sources of documentation.  Christian
-von Roques encouraged me to look for faster sorting algorithms, so as to
-speed up compression.  Bela Lubkin encouraged me to improve the
-worst-case compression performance.  The @code{bz*} scripts are derived
-from those of GNU @code{gzip}.  Many people sent patches, helped with
-portability problems, lent machines, gave advice and were generally
-helpful.
-
-@end quotation
-
-
-
-
-@chapter Programming with @code{libbzip2}
-
-This chapter describes the programming interface to @code{libbzip2}.
-
-For general background information, particularly about memory
-use and performance aspects, you'd be well advised to read Chapter 2
-as well.
-
-@section Top-level structure
-
-@code{libbzip2} is a flexible library for compressing and decompressing
-data in the @code{bzip2} data format.  Although packaged as a single
-entity, it helps to regard the library as three separate parts: the low
-level interface, and the high level interface, and some utility
-functions.
-
-The structure of @code{libbzip2}'s interfaces is similar to
-that of Jean-loup Gailly's and Mark Adler's excellent @code{zlib} 
-library.
-
-All externally visible symbols have names beginning @code{BZ2_}.
-This is new in version 1.0.  The intention is to minimise pollution
-of the namespaces of library clients.
-
-@subsection Low-level summary
-
-This interface provides services for compressing and decompressing
-data in memory.  There's no provision for dealing with files, streams
-or any other I/O mechanisms, just straight memory-to-memory work.
-In fact, this part of the library can be compiled without inclusion
-of @code{stdio.h}, which may be helpful for embedded applications.
-
-The low-level part of the library has no global variables and
-is therefore thread-safe.
-
-Six routines make up the low level interface: 
-@code{BZ2_bzCompressInit}, @code{BZ2_bzCompress}, and @* @code{BZ2_bzCompressEnd}
-for compression,
-and a corresponding trio @code{BZ2_bzDecompressInit}, @* @code{BZ2_bzDecompress}
-and @code{BZ2_bzDecompressEnd} for decompression.  
-The @code{*Init} functions allocate
-memory for compression/decompression and do other
-initialisations, whilst the @code{*End} functions close down operations
-and release memory.
-
-The real work is done by @code{BZ2_bzCompress} and @code{BZ2_bzDecompress}.  
-These compress and decompress data from a user-supplied input buffer
-to a user-supplied output buffer.  These buffers can be any size;
-arbitrary quantities of data are handled by making repeated calls
-to these functions.  This is a flexible mechanism allowing a 
-consumer-pull style of activity, or producer-push, or a mixture of
-both.
-
-
-
-@subsection High-level summary
-
-This interface provides some handy wrappers around the low-level
-interface to facilitate reading and writing @code{bzip2} format
-files (@code{.bz2} files).  The routines provide hooks to facilitate
-reading files in which the @code{bzip2} data stream is embedded 
-within some larger-scale file structure, or where there are
-multiple @code{bzip2} data streams concatenated end-to-end.
-
-For reading files, @code{BZ2_bzReadOpen}, @code{BZ2_bzRead},
-@code{BZ2_bzReadClose} and @* @code{BZ2_bzReadGetUnused} are supplied.  For
-writing files, @code{BZ2_bzWriteOpen}, @code{BZ2_bzWrite} and
-@code{BZ2_bzWriteFinish} are available.
-
-As with the low-level library, no global variables are used
-so the library is per se thread-safe.  However, if I/O errors
-occur whilst reading or writing the underlying compressed files,
-you may have to consult @code{errno} to determine the cause of
-the error.  In that case, you'd need a C library which correctly
-supports @code{errno} in a multithreaded environment.
-
-To make the library a little simpler and more portable,
-@code{BZ2_bzReadOpen} and @code{BZ2_bzWriteOpen} require you to pass them file
-handles (@code{FILE*}s) which have previously been opened for reading or
-writing respectively.  That avoids portability problems associated with
-file operations and file attributes, whilst not being much of an
-imposition on the programmer.
-
-
-
-@subsection Utility functions summary
-For very simple needs, @code{BZ2_bzBuffToBuffCompress} and
-@code{BZ2_bzBuffToBuffDecompress} are provided.  These compress
-data in memory from one buffer to another buffer in a single
-function call.  You should assess whether these functions
-fulfill your memory-to-memory compression/decompression
-requirements before investing effort in understanding the more
-general but more complex low-level interface.
-
-Yoshioka Tsuneo (@code{QWF00133@@niftyserve.or.jp} /
-@code{tsuneo-y@@is.aist-nara.ac.jp}) has contributed some functions to
-give better @code{zlib} compatibility.  These functions are
-@code{BZ2_bzopen}, @code{BZ2_bzread}, @code{BZ2_bzwrite}, @code{BZ2_bzflush},
-@code{BZ2_bzclose},
-@code{BZ2_bzerror} and @code{BZ2_bzlibVersion}.  You may find these functions
-more convenient for simple file reading and writing, than those in the
-high-level interface.  These functions are not (yet) officially part of
-the library, and are minimally documented here.  If they break, you
-get to keep all the pieces.  I hope to document them properly when time
-permits.
-
-Yoshioka also contributed modifications to allow the library to be
-built as a Windows DLL.
-
-
-@section Error handling
-
-The library is designed to recover cleanly in all situations, including
-the worst-case situation of decompressing random data.  I'm not 
-100% sure that it can always do this, so you might want to add
-a signal handler to catch segmentation violations during decompression
-if you are feeling especially paranoid.  I would be interested in
-hearing more about the robustness of the library to corrupted
-compressed data.
-
-Version 1.0 is much more robust in this respect than
-0.9.0 or 0.9.5.  Investigations with Checker (a tool for 
-detecting problems with memory management, similar to Purify)
-indicate that, at least for the few files I tested, all single-bit
-errors in the decompressed data are caught properly, with no
-segmentation faults, no reads of uninitialised data and no 
-out of range reads or writes.  So it's certainly much improved,
-although I wouldn't claim it to be totally bombproof.
-
-The file @code{bzlib.h} contains all definitions needed to use
-the library.  In particular, you should definitely not include
-@code{bzlib_private.h}.
-
-In @code{bzlib.h}, the various return values are defined.  The following
-list is not intended as an exhaustive description of the circumstances 
-in which a given value may be returned -- those descriptions are given
-later.  Rather, it is intended to convey the rough meaning of each
-return value.  The first five actions are normal and not intended to 
-denote an error situation.
-@table @code
-@item BZ_OK
-The requested action was completed successfully.
-@item BZ_RUN_OK
-@itemx BZ_FLUSH_OK
-@itemx BZ_FINISH_OK
-In @code{BZ2_bzCompress}, the requested flush/finish/nothing-special action
-was completed successfully.
-@item BZ_STREAM_END
-Compression of data was completed, or the logical stream end was
-detected during decompression.
-@end table
-
-The following return values indicate an error of some kind.
-@table @code
-@item BZ_CONFIG_ERROR
-Indicates that the library has been improperly compiled on your
-platform -- a major configuration error.  Specifically, it means
-that @code{sizeof(char)}, @code{sizeof(short)} and @code{sizeof(int)}
-are not 1, 2 and 4 respectively, as they should be.  Note that the 
-library should still work properly on 64-bit platforms which follow
-the LP64 programming model -- that is, where @code{sizeof(long)}
-and @code{sizeof(void*)} are 8.  Under LP64, @code{sizeof(int)} is
-still 4, so @code{libbzip2}, which doesn't use the @code{long} type,
-is OK.
-@item BZ_SEQUENCE_ERROR
-When using the library, it is important to call the functions in the
-correct sequence and with data structures (buffers etc) in the correct
-states.  @code{libbzip2} checks as much as it can to ensure this is
-happening, and returns @code{BZ_SEQUENCE_ERROR} if not.  Code which
-complies precisely with the function semantics, as detailed below,
-should never receive this value; such an event denotes buggy code
-which you should investigate.
-@item BZ_PARAM_ERROR
-Returned when a parameter to a function call is out of range 
-or otherwise manifestly incorrect.  As with @code{BZ_SEQUENCE_ERROR},
-this denotes a bug in the client code.  The distinction between
-@code{BZ_PARAM_ERROR} and @code{BZ_SEQUENCE_ERROR} is a bit hazy, but still worth
-making.
-@item BZ_MEM_ERROR
-Returned when a request to allocate memory failed.  Note that the
-quantity of memory needed to decompress a stream cannot be determined
-until the stream's header has been read.  So @code{BZ2_bzDecompress} and
-@code{BZ2_bzRead} may return @code{BZ_MEM_ERROR} even though some of
-the compressed data has been read.  The same is not true for
-compression; once @code{BZ2_bzCompressInit} or @code{BZ2_bzWriteOpen} have
-successfully completed, @code{BZ_MEM_ERROR} cannot occur.
-@item BZ_DATA_ERROR
-Returned when a data integrity error is detected during decompression.
-Most importantly, this means when stored and computed CRCs for the
-data do not match.  This value is also returned upon detection of any
-other anomaly in the compressed data.
-@item BZ_DATA_ERROR_MAGIC
-As a special case of @code{BZ_DATA_ERROR}, it is sometimes useful to
-know when the compressed stream does not start with the correct
-magic bytes (@code{'B' 'Z' 'h'}).  
-@item BZ_IO_ERROR
-Returned by @code{BZ2_bzRead} and @code{BZ2_bzWrite} when there is an error
-reading or writing in the compressed file, and by @code{BZ2_bzReadOpen}
-and @code{BZ2_bzWriteOpen} for attempts to use a file for which the
-error indicator (viz, @code{ferror(f)}) is set.
-On receipt of @code{BZ_IO_ERROR}, the caller should consult
-@code{errno} and/or @code{perror} to acquire operating-system
-specific information about the problem.
-@item BZ_UNEXPECTED_EOF
-Returned by @code{BZ2_bzRead} when the compressed file finishes
-before the logical end of stream is detected.
-@item BZ_OUTBUFF_FULL
-Returned by @code{BZ2_bzBuffToBuffCompress} and
-@code{BZ2_bzBuffToBuffDecompress} to indicate that the output data
-will not fit into the output buffer provided.
-@end table
-
-
-
-@section Low-level interface
-
-@subsection @code{BZ2_bzCompressInit}
-@example
-typedef 
-   struct @{
-      char *next_in;
-      unsigned int avail_in;
-      unsigned int total_in_lo32;
-      unsigned int total_in_hi32;
-
-      char *next_out;
-      unsigned int avail_out;
-      unsigned int total_out_lo32;
-      unsigned int total_out_hi32;
-
-      void *state;
-
-      void *(*bzalloc)(void *,int,int);
-      void (*bzfree)(void *,void *);
-      void *opaque;
-   @} 
-   bz_stream;
-
-int BZ2_bzCompressInit ( bz_stream *strm, 
-                         int blockSize100k, 
-                         int verbosity,
-                         int workFactor );
-
-@end example
-
-Prepares for compression.  The @code{bz_stream} structure
-holds all data pertaining to the compression activity.  
-A @code{bz_stream} structure should be allocated and initialised
-prior to the call.
-The fields of @code{bz_stream}
-comprise the entirety of the user-visible data.  @code{state}
-is a pointer to the private data structures required for compression.
-
-Custom memory allocators are supported, via fields @code{bzalloc}, 
-@code{bzfree},
-and @code{opaque}.  The value 
-@code{opaque} is passed to as the first argument to
-all calls to @code{bzalloc} and @code{bzfree}, but is 
-otherwise ignored by the library.
-The call @code{bzalloc ( opaque, n, m )} is expected to return a 
-pointer @code{p} to
-@code{n * m} bytes of memory, and @code{bzfree ( opaque, p )} 
-should free
-that memory.
-
-If you don't want to use a custom memory allocator, set @code{bzalloc}, 
-@code{bzfree} and
-@code{opaque} to @code{NULL}, 
-and the library will then use the standard @code{malloc}/@code{free}
-routines.
-
-Before calling @code{BZ2_bzCompressInit}, fields @code{bzalloc}, 
-@code{bzfree} and @code{opaque} should
-be filled appropriately, as just described.  Upon return, the internal
-state will have been allocated and initialised, and @code{total_in_lo32}, 
-@code{total_in_hi32}, @code{total_out_lo32} and 
-@code{total_out_hi32} will have been set to zero.  
-These four fields are used by the library
-to inform the caller of the total amount of data passed into and out of
-the library, respectively.  You should not try to change them.
-As of version 1.0, 64-bit counts are maintained, even on 32-bit
-platforms, using the @code{_hi32} fields to store the upper 32 bits
-of the count.  So, for example, the total amount of data in
-is @code{(total_in_hi32 << 32) + total_in_lo32}.
-
-Parameter @code{blockSize100k} specifies the block size to be used for
-compression.  It should be a value between 1 and 9 inclusive, and the
-actual block size used is 100000 x this figure.  9 gives the best
-compression but takes most memory.
-
-Parameter @code{verbosity} should be set to a number between 0 and 4
-inclusive.  0 is silent, and greater numbers give increasingly verbose
-monitoring/debugging output.  If the library has been compiled with
-@code{-DBZ_NO_STDIO}, no such output will appear for any verbosity
-setting.
-
-Parameter @code{workFactor} controls how the compression phase behaves
-when presented with worst case, highly repetitive, input data.  If
-compression runs into difficulties caused by repetitive data, the
-library switches from the standard sorting algorithm to a fallback
-algorithm.  The fallback is slower than the standard algorithm by
-perhaps a factor of three, but always behaves reasonably, no matter how
-bad the input.
-
-Lower values of @code{workFactor} reduce the amount of effort the
-standard algorithm will expend before resorting to the fallback.  You
-should set this parameter carefully; too low, and many inputs will be
-handled by the fallback algorithm and so compress rather slowly, too
-high, and your average-to-worst case compression times can become very
-large.  The default value of 30 gives reasonable behaviour over a wide
-range of circumstances.
-
-Allowable values range from 0 to 250 inclusive.  0 is a special case,
-equivalent to using the default value of 30.
-
-Note that the compressed output generated is the same regardless of
-whether or not the fallback algorithm is used.
-
-Be aware also that this parameter may disappear entirely in future
-versions of the library.  In principle it should be possible to devise a
-good way to automatically choose which algorithm to use.  Such a
-mechanism would render the parameter obsolete.
-
-Possible return values:
-@display
-      @code{BZ_CONFIG_ERROR}
-         if the library has been mis-compiled
-      @code{BZ_PARAM_ERROR} 
-         if @code{strm} is @code{NULL} 
-         or @code{blockSize} < 1 or @code{blockSize} > 9
-         or @code{verbosity} < 0 or @code{verbosity} > 4
-         or @code{workFactor} < 0 or @code{workFactor} > 250
-      @code{BZ_MEM_ERROR} 
-         if not enough memory is available
-      @code{BZ_OK} 
-         otherwise
-@end display
-Allowable next actions:
-@display
-      @code{BZ2_bzCompress} 
-         if @code{BZ_OK} is returned
-      no specific action needed in case of error
-@end display
-
-@subsection @code{BZ2_bzCompress}
-@example
-   int BZ2_bzCompress ( bz_stream *strm, int action );
-@end example
-Provides more input and/or output buffer space for the library.  The
-caller maintains input and output buffers, and calls @code{BZ2_bzCompress} to
-transfer data between them.
-
-Before each call to @code{BZ2_bzCompress}, @code{next_in} should point at
-the data to be compressed, and @code{avail_in} should indicate how many
-bytes the library may read.  @code{BZ2_bzCompress} updates @code{next_in},
-@code{avail_in} and @code{total_in} to reflect the number of bytes it
-has read.
-
-Similarly, @code{next_out} should point to a buffer in which the
-compressed data is to be placed, with @code{avail_out} indicating how
-much output space is available.  @code{BZ2_bzCompress} updates
-@code{next_out}, @code{avail_out} and @code{total_out} to reflect the
-number of bytes output.
-
-You may provide and remove as little or as much data as you like on each
-call of @code{BZ2_bzCompress}.  In the limit, it is acceptable to supply and
-remove data one byte at a time, although this would be terribly
-inefficient.  You should always ensure that at least one byte of output
-space is available at each call.
-
-A second purpose of @code{BZ2_bzCompress} is to request a change of mode of the
-compressed stream.  
-
-Conceptually, a compressed stream can be in one of four states: IDLE,
-RUNNING, FLUSHING and FINISHING.  Before initialisation
-(@code{BZ2_bzCompressInit}) and after termination (@code{BZ2_bzCompressEnd}), a
-stream is regarded as IDLE.
-
-Upon initialisation (@code{BZ2_bzCompressInit}), the stream is placed in the
-RUNNING state.  Subsequent calls to @code{BZ2_bzCompress} should pass
-@code{BZ_RUN} as the requested action; other actions are illegal and
-will result in @code{BZ_SEQUENCE_ERROR}.
-
-At some point, the calling program will have provided all the input data
-it wants to.  It will then want to finish up -- in effect, asking the
-library to process any data it might have buffered internally.  In this
-state, @code{BZ2_bzCompress} will no longer attempt to read data from
-@code{next_in}, but it will want to write data to @code{next_out}.
-Because the output buffer supplied by the user can be arbitrarily small,
-the finishing-up operation cannot necessarily be done with a single call
-of @code{BZ2_bzCompress}.
-
-Instead, the calling program passes @code{BZ_FINISH} as an action to
-@code{BZ2_bzCompress}.  This changes the stream's state to FINISHING.  Any
-remaining input (ie, @code{next_in[0 .. avail_in-1]}) is compressed and
-transferred to the output buffer.  To do this, @code{BZ2_bzCompress} must be
-called repeatedly until all the output has been consumed.  At that
-point, @code{BZ2_bzCompress} returns @code{BZ_STREAM_END}, and the stream's
-state is set back to IDLE.  @code{BZ2_bzCompressEnd} should then be
-called.
-
-Just to make sure the calling program does not cheat, the library makes
-a note of @code{avail_in} at the time of the first call to
-@code{BZ2_bzCompress} which has @code{BZ_FINISH} as an action (ie, at the
-time the program has announced its intention to not supply any more
-input).  By comparing this value with that of @code{avail_in} over
-subsequent calls to @code{BZ2_bzCompress}, the library can detect any
-attempts to slip in more data to compress.  Any calls for which this is
-detected will return @code{BZ_SEQUENCE_ERROR}.  This indicates a
-programming mistake which should be corrected.
-
-Instead of asking to finish, the calling program may ask
-@code{BZ2_bzCompress} to take all the remaining input, compress it and
-terminate the current (Burrows-Wheeler) compression block.  This could
-be useful for error control purposes.  The mechanism is analogous to
-that for finishing: call @code{BZ2_bzCompress} with an action of
-@code{BZ_FLUSH}, remove output data, and persist with the
-@code{BZ_FLUSH} action until the value @code{BZ_RUN} is returned.  As
-with finishing, @code{BZ2_bzCompress} detects any attempt to provide more
-input data once the flush has begun.
-
-Once the flush is complete, the stream returns to the normal RUNNING
-state.
-
-This all sounds pretty complex, but isn't really.  Here's a table
-which shows which actions are allowable in each state, what action
-will be taken, what the next state is, and what the non-error return
-values are.  Note that you can't explicitly ask what state the
-stream is in, but nor do you need to -- it can be inferred from the
-values returned by @code{BZ2_bzCompress}.
-@display
-IDLE/@code{any}           
-      Illegal.  IDLE state only exists after @code{BZ2_bzCompressEnd} or
-      before @code{BZ2_bzCompressInit}.
-      Return value = @code{BZ_SEQUENCE_ERROR}
-
-RUNNING/@code{BZ_RUN}     
-      Compress from @code{next_in} to @code{next_out} as much as possible.
-      Next state = RUNNING
-      Return value = @code{BZ_RUN_OK}
-
-RUNNING/@code{BZ_FLUSH}   
-      Remember current value of @code{next_in}.  Compress from @code{next_in}
-      to @code{next_out} as much as possible, but do not accept any more input.  
-      Next state = FLUSHING
-      Return value = @code{BZ_FLUSH_OK}
-
-RUNNING/@code{BZ_FINISH}  
-      Remember current value of @code{next_in}.  Compress from @code{next_in}
-      to @code{next_out} as much as possible, but do not accept any more input.
-      Next state = FINISHING
-      Return value = @code{BZ_FINISH_OK}
-
-FLUSHING/@code{BZ_FLUSH}  
-      Compress from @code{next_in} to @code{next_out} as much as possible, 
-      but do not accept any more input.  
-      If all the existing input has been used up and all compressed
-      output has been removed
-         Next state = RUNNING; Return value = @code{BZ_RUN_OK}
-      else
-         Next state = FLUSHING; Return value = @code{BZ_FLUSH_OK}
-
-FLUSHING/other     
-      Illegal.
-      Return value = @code{BZ_SEQUENCE_ERROR}
-
-FINISHING/@code{BZ_FINISH}  
-      Compress from @code{next_in} to @code{next_out} as much as possible,
-      but to not accept any more input.  
-      If all the existing input has been used up and all compressed
-      output has been removed
-         Next state = IDLE; Return value = @code{BZ_STREAM_END}
-      else
-         Next state = FINISHING; Return value = @code{BZ_FINISHING}
-
-FINISHING/other
-      Illegal.
-      Return value = @code{BZ_SEQUENCE_ERROR}
-@end display
-
-That still looks complicated?  Well, fair enough.  The usual sequence
-of calls for compressing a load of data is:
-@itemize @bullet
-@item Get started with @code{BZ2_bzCompressInit}.
-@item Shovel data in and shlurp out its compressed form using zero or more
-calls of @code{BZ2_bzCompress} with action = @code{BZ_RUN}.
-@item Finish up.  
-Repeatedly call @code{BZ2_bzCompress} with action = @code{BZ_FINISH}, 
-copying out the compressed output, until @code{BZ_STREAM_END} is returned.
-@item Close up and go home.  Call @code{BZ2_bzCompressEnd}.
-@end itemize
-If the data you want to compress fits into your input buffer all
-at once, you can skip the calls of @code{BZ2_bzCompress ( ..., BZ_RUN )} and 
-just do the @code{BZ2_bzCompress ( ..., BZ_FINISH )} calls.
-
-All required memory is allocated by @code{BZ2_bzCompressInit}.  The
-compression library can accept any data at all (obviously).  So you
-shouldn't get any error return values from the @code{BZ2_bzCompress} calls.
-If you do, they will be @code{BZ_SEQUENCE_ERROR}, and indicate a bug in
-your programming.
-
-Trivial other possible return values:
-@display
-      @code{BZ_PARAM_ERROR}   
-         if @code{strm} is @code{NULL}, or @code{strm->s} is @code{NULL}
-@end display
-
-@subsection @code{BZ2_bzCompressEnd}
-@example
-int BZ2_bzCompressEnd ( bz_stream *strm );
-@end example
-Releases all memory associated with a compression stream.
-
-Possible return values:
-@display
-   @code{BZ_PARAM_ERROR}    if @code{strm} is @code{NULL} or @code{strm->s} is @code{NULL}
-   @code{BZ_OK}    otherwise
-@end display
-
-
-@subsection @code{BZ2_bzDecompressInit}
-@example
-int BZ2_bzDecompressInit ( bz_stream *strm, int verbosity, int small );
-@end example
-Prepares for decompression.  As with @code{BZ2_bzCompressInit}, a
-@code{bz_stream} record should be allocated and initialised before the
-call.  Fields @code{bzalloc}, @code{bzfree} and @code{opaque} should be
-set if a custom memory allocator is required, or made @code{NULL} for
-the normal @code{malloc}/@code{free} routines.  Upon return, the internal
-state will have been initialised, and @code{total_in} and
-@code{total_out} will be zero.
-
-For the meaning of parameter @code{verbosity}, see @code{BZ2_bzCompressInit}.
-
-If @code{small} is nonzero, the library will use an alternative
-decompression algorithm which uses less memory but at the cost of
-decompressing more slowly (roughly speaking, half the speed, but the
-maximum memory requirement drops to around 2300k).  See Chapter 2 for
-more information on memory management.
-
-Note that the amount of memory needed to decompress
-a stream cannot be determined until the stream's header has been read,
-so even if @code{BZ2_bzDecompressInit} succeeds, a subsequent
-@code{BZ2_bzDecompress} could fail with @code{BZ_MEM_ERROR}.
-
-Possible return values:
-@display
-      @code{BZ_CONFIG_ERROR}
-         if the library has been mis-compiled
-      @code{BZ_PARAM_ERROR}
-         if @code{(small != 0 && small != 1)}
-         or @code{(verbosity < 0 || verbosity > 4)}
-      @code{BZ_MEM_ERROR}
-         if insufficient memory is available
-@end display
-
-Allowable next actions:
-@display
-      @code{BZ2_bzDecompress}
-         if @code{BZ_OK} was returned
-      no specific action required in case of error
-@end display
-
- 
-
-@subsection @code{BZ2_bzDecompress}
-@example
-int BZ2_bzDecompress ( bz_stream *strm );
-@end example
-Provides more input and/out output buffer space for the library.  The
-caller maintains input and output buffers, and uses @code{BZ2_bzDecompress}
-to transfer data between them.
-
-Before each call to @code{BZ2_bzDecompress}, @code{next_in} 
-should point at the compressed data,
-and @code{avail_in} should indicate how many bytes the library
-may read.  @code{BZ2_bzDecompress} updates @code{next_in}, @code{avail_in} 
-and @code{total_in}
-to reflect the number of bytes it has read.
-
-Similarly, @code{next_out} should point to a buffer in which the uncompressed
-output is to be placed, with @code{avail_out} indicating how much output space
-is available.  @code{BZ2_bzCompress} updates @code{next_out},
-@code{avail_out} and @code{total_out} to reflect
-the number of bytes output.
-
-You may provide and remove as little or as much data as you like on
-each call of @code{BZ2_bzDecompress}.  
-In the limit, it is acceptable to
-supply and remove data one byte at a time, although this would be
-terribly inefficient.  You should always ensure that at least one
-byte of output space is available at each call.
-
-Use of @code{BZ2_bzDecompress} is simpler than @code{BZ2_bzCompress}.
-
-You should provide input and remove output as described above, and
-repeatedly call @code{BZ2_bzDecompress} until @code{BZ_STREAM_END} is
-returned.  Appearance of @code{BZ_STREAM_END} denotes that
-@code{BZ2_bzDecompress} has detected the logical end of the compressed
-stream.  @code{BZ2_bzDecompress} will not produce @code{BZ_STREAM_END} until
-all output data has been placed into the output buffer, so once
-@code{BZ_STREAM_END} appears, you are guaranteed to have available all
-the decompressed output, and @code{BZ2_bzDecompressEnd} can safely be
-called.
-
-If case of an error return value, you should call @code{BZ2_bzDecompressEnd}
-to clean up and release memory.
-
-Possible return values:
-@display
-      @code{BZ_PARAM_ERROR}
-         if @code{strm} is @code{NULL} or @code{strm->s} is @code{NULL}
-         or @code{strm->avail_out < 1}
-      @code{BZ_DATA_ERROR}
-         if a data integrity error is detected in the compressed stream
-      @code{BZ_DATA_ERROR_MAGIC}
-         if the compressed stream doesn't begin with the right magic bytes
-      @code{BZ_MEM_ERROR}
-         if there wasn't enough memory available
-      @code{BZ_STREAM_END}
-         if the logical end of the data stream was detected and all
-         output in has been consumed, eg @code{s->avail_out > 0}
-      @code{BZ_OK}
-         otherwise
-@end display
-Allowable next actions:
-@display
-      @code{BZ2_bzDecompress}
-         if @code{BZ_OK} was returned
-      @code{BZ2_bzDecompressEnd}
-         otherwise
-@end display
-
-
-@subsection @code{BZ2_bzDecompressEnd}
-@example
-int BZ2_bzDecompressEnd ( bz_stream *strm );
-@end example
-Releases all memory associated with a decompression stream.
-
-Possible return values:
-@display
-      @code{BZ_PARAM_ERROR}
-         if @code{strm} is @code{NULL} or @code{strm->s} is @code{NULL}
-      @code{BZ_OK}
-         otherwise
-@end display
-
-Allowable next actions:
-@display
-      None.
-@end display
-
-
-@section High-level interface
-
-This interface provides functions for reading and writing 
-@code{bzip2} format files.  First, some general points.
-
-@itemize @bullet
-@item All of the functions take an @code{int*} first argument,
-  @code{bzerror}.
-  After each call, @code{bzerror} should be consulted first to determine
-  the outcome of the call.  If @code{bzerror} is @code{BZ_OK}, 
-  the call completed
-  successfully, and only then should the return value of the function
-  (if any) be consulted.  If @code{bzerror} is @code{BZ_IO_ERROR}, 
-  there was an error
-  reading/writing the underlying compressed file, and you should
-  then consult @code{errno}/@code{perror} to determine the 
-  cause of the difficulty.
-  @code{bzerror} may also be set to various other values; precise details are
-  given on a per-function basis below.
-@item If @code{bzerror} indicates an error 
-  (ie, anything except @code{BZ_OK} and @code{BZ_STREAM_END}),
-  you should immediately call @code{BZ2_bzReadClose} (or @code{BZ2_bzWriteClose},
-  depending on whether you are attempting to read or to write)
-  to free up all resources associated
-  with the stream.  Once an error has been indicated, behaviour of all calls
-  except @code{BZ2_bzReadClose} (@code{BZ2_bzWriteClose}) is undefined.  
-  The implication is that (1) @code{bzerror} should
-  be checked after each call, and (2) if @code{bzerror} indicates an error, 
-  @code{BZ2_bzReadClose} (@code{BZ2_bzWriteClose}) should then be called to clean up.
-@item The @code{FILE*} arguments passed to
-   @code{BZ2_bzReadOpen}/@code{BZ2_bzWriteOpen}  
-  should be set to binary mode.
-  Most Unix systems will do this by default, but other platforms,
-  including Windows and Mac, will not.  If you omit this, you may
-  encounter problems when moving code to new platforms.
-@item Memory allocation requests are handled by
-  @code{malloc}/@code{free}.  
-  At present
-  there is no facility for user-defined memory allocators in the file I/O
-  functions (could easily be added, though).
-@end itemize
-
-
-
-@subsection @code{BZ2_bzReadOpen}
-@example
-   typedef void BZFILE;
-
-   BZFILE *BZ2_bzReadOpen ( int *bzerror, FILE *f, 
-                            int small, int verbosity,
-                            void *unused, int nUnused );
-@end example
-Prepare to read compressed data from file handle @code{f}.  @code{f}
-should refer to a file which has been opened for reading, and for which
-the error indicator (@code{ferror(f)})is not set.  If @code{small} is 1,
-the library will try to decompress using less memory, at the expense of
-speed.
-
-For reasons explained below, @code{BZ2_bzRead} will decompress the
-@code{nUnused} bytes starting at @code{unused}, before starting to read
-from the file @code{f}.  At most @code{BZ_MAX_UNUSED} bytes may be
-supplied like this.  If this facility is not required, you should pass
-@code{NULL} and @code{0} for @code{unused} and n@code{Unused}
-respectively.
-
-For the meaning of parameters @code{small} and @code{verbosity},
-see @code{BZ2_bzDecompressInit}.
-
-The amount of memory needed to decompress a file cannot be determined
-until the file's header has been read.  So it is possible that
-@code{BZ2_bzReadOpen} returns @code{BZ_OK} but a subsequent call of
-@code{BZ2_bzRead} will return @code{BZ_MEM_ERROR}.
-
-Possible assignments to @code{bzerror}:
-@display
-      @code{BZ_CONFIG_ERROR}
-         if the library has been mis-compiled
-      @code{BZ_PARAM_ERROR}
-         if @code{f} is @code{NULL} 
-         or @code{small} is neither @code{0} nor @code{1}                 
-         or @code{(unused == NULL && nUnused != 0)}
-         or @code{(unused != NULL && !(0 <= nUnused <= BZ_MAX_UNUSED))}
-      @code{BZ_IO_ERROR}    
-         if @code{ferror(f)} is nonzero
-      @code{BZ_MEM_ERROR}   
-         if insufficient memory is available
-      @code{BZ_OK}
-         otherwise.
-@end display
-
-Possible return values:
-@display
-      Pointer to an abstract @code{BZFILE}        
-         if @code{bzerror} is @code{BZ_OK}   
-      @code{NULL}
-         otherwise
-@end display
-
-Allowable next actions:
-@display
-      @code{BZ2_bzRead}
-         if @code{bzerror} is @code{BZ_OK}   
-      @code{BZ2_bzClose} 
-         otherwise
-@end display
-
-
-@subsection @code{BZ2_bzRead}
-@example
-   int BZ2_bzRead ( int *bzerror, BZFILE *b, void *buf, int len );
-@end example
-Reads up to @code{len} (uncompressed) bytes from the compressed file 
-@code{b} into
-the buffer @code{buf}.  If the read was successful, 
-@code{bzerror} is set to @code{BZ_OK}
-and the number of bytes read is returned.  If the logical end-of-stream
-was detected, @code{bzerror} will be set to @code{BZ_STREAM_END}, 
-and the number
-of bytes read is returned.  All other @code{bzerror} values denote an error.
-
-@code{BZ2_bzRead} will supply @code{len} bytes,
-unless the logical stream end is detected
-or an error occurs.  Because of this, it is possible to detect the 
-stream end by observing when the number of bytes returned is 
-less than the number
-requested.  Nevertheless, this is regarded as inadvisable; you should
-instead check @code{bzerror} after every call and watch out for
-@code{BZ_STREAM_END}.
-
-Internally, @code{BZ2_bzRead} copies data from the compressed file in chunks
-of size @code{BZ_MAX_UNUSED} bytes
-before decompressing it.  If the file contains more bytes than strictly
-needed to reach the logical end-of-stream, @code{BZ2_bzRead} will almost certainly
-read some of the trailing data before signalling @code{BZ_SEQUENCE_END}.
-To collect the read but unused data once @code{BZ_SEQUENCE_END} has 
-appeared, call @code{BZ2_bzReadGetUnused} immediately before @code{BZ2_bzReadClose}.
-
-Possible assignments to @code{bzerror}:
-@display
-      @code{BZ_PARAM_ERROR}
-         if @code{b} is @code{NULL} or @code{buf} is @code{NULL} or @code{len < 0}
-      @code{BZ_SEQUENCE_ERROR} 
-         if @code{b} was opened with @code{BZ2_bzWriteOpen}
-      @code{BZ_IO_ERROR} 
-         if there is an error reading from the compressed file
-      @code{BZ_UNEXPECTED_EOF} 
-         if the compressed file ended before the logical end-of-stream was detected
-      @code{BZ_DATA_ERROR} 
-         if a data integrity error was detected in the compressed stream
-      @code{BZ_DATA_ERROR_MAGIC}
-         if the stream does not begin with the requisite header bytes (ie, is not 
-         a @code{bzip2} data file).  This is really a special case of @code{BZ_DATA_ERROR}.
-      @code{BZ_MEM_ERROR} 
-         if insufficient memory was available
-      @code{BZ_STREAM_END} 
-         if the logical end of stream was detected.
-      @code{BZ_OK}
-         otherwise.
-@end display
-
-Possible return values:
-@display
-      number of bytes read
-         if @code{bzerror} is @code{BZ_OK} or @code{BZ_STREAM_END}
-      undefined
-         otherwise
-@end display
-
-Allowable next actions:
-@display
-      collect data from @code{buf}, then @code{BZ2_bzRead} or @code{BZ2_bzReadClose}
-         if @code{bzerror} is @code{BZ_OK} 
-      collect data from @code{buf}, then @code{BZ2_bzReadClose} or @code{BZ2_bzReadGetUnused} 
-         if @code{bzerror} is @code{BZ_SEQUENCE_END}   
-      @code{BZ2_bzReadClose} 
-         otherwise
-@end display
-
-
-
-@subsection @code{BZ2_bzReadGetUnused}
-@example
-   void BZ2_bzReadGetUnused ( int* bzerror, BZFILE *b, 
-                              void** unused, int* nUnused );
-@end example
-Returns data which was read from the compressed file but was not needed
-to get to the logical end-of-stream.  @code{*unused} is set to the address
-of the data, and @code{*nUnused} to the number of bytes.  @code{*nUnused} will
-be set to a value between @code{0} and @code{BZ_MAX_UNUSED} inclusive.
-
-This function may only be called once @code{BZ2_bzRead} has signalled 
-@code{BZ_STREAM_END} but before @code{BZ2_bzReadClose}.
-
-Possible assignments to @code{bzerror}:
-@display
-      @code{BZ_PARAM_ERROR} 
-         if @code{b} is @code{NULL} 
-         or @code{unused} is @code{NULL} or @code{nUnused} is @code{NULL}
-      @code{BZ_SEQUENCE_ERROR} 
-         if @code{BZ_STREAM_END} has not been signalled
-         or if @code{b} was opened with @code{BZ2_bzWriteOpen}
-     @code{BZ_OK}
-         otherwise
-@end display
-
-Allowable next actions:
-@display 
-      @code{BZ2_bzReadClose}
-@end display
-
-
-@subsection @code{BZ2_bzReadClose}
-@example
-   void BZ2_bzReadClose ( int *bzerror, BZFILE *b );
-@end example
-Releases all memory pertaining to the compressed file @code{b}.  
-@code{BZ2_bzReadClose} does not call @code{fclose} on the underlying file
-handle, so you should do that yourself if appropriate.
-@code{BZ2_bzReadClose} should be called to clean up after all error
-situations.
-
-Possible assignments to @code{bzerror}:
-@display
-      @code{BZ_SEQUENCE_ERROR} 
-         if @code{b} was opened with @code{BZ2_bzOpenWrite} 
-      @code{BZ_OK} 
-         otherwise
-@end display
-
-Allowable next actions:
-@display
-      none
-@end display
-
-
-
-@subsection @code{BZ2_bzWriteOpen}
-@example
-   BZFILE *BZ2_bzWriteOpen ( int *bzerror, FILE *f, 
-                             int blockSize100k, int verbosity,
-                             int workFactor );
-@end example
-Prepare to write compressed data to file handle @code{f}.  
-@code{f} should refer to
-a file which has been opened for writing, and for which the error
-indicator (@code{ferror(f)})is not set.  
-
-For the meaning of parameters @code{blockSize100k},
-@code{verbosity} and @code{workFactor}, see
-@* @code{BZ2_bzCompressInit}.
-
-All required memory is allocated at this stage, so if the call
-completes successfully, @code{BZ_MEM_ERROR} cannot be signalled by a
-subsequent call to @code{BZ2_bzWrite}.
-
-Possible assignments to @code{bzerror}:
-@display 
-      @code{BZ_CONFIG_ERROR}
-         if the library has been mis-compiled
-      @code{BZ_PARAM_ERROR} 
-         if @code{f} is @code{NULL} 
-         or @code{blockSize100k < 1} or @code{blockSize100k > 9}
-      @code{BZ_IO_ERROR} 
-         if @code{ferror(f)} is nonzero
-      @code{BZ_MEM_ERROR} 
-         if insufficient memory is available
-      @code{BZ_OK} 
-         otherwise
-@end display
-
-Possible return values:
-@display
-      Pointer to an abstract @code{BZFILE}  
-         if @code{bzerror} is @code{BZ_OK}   
-      @code{NULL} 
-         otherwise
-@end display
-
-Allowable next actions:
-@display
-      @code{BZ2_bzWrite} 
-         if @code{bzerror} is @code{BZ_OK} 
-         (you could go directly to @code{BZ2_bzWriteClose}, but this would be pretty pointless)
-      @code{BZ2_bzWriteClose} 
-         otherwise
-@end display
-
-
-
-@subsection @code{BZ2_bzWrite}
-@example
-   void BZ2_bzWrite ( int *bzerror, BZFILE *b, void *buf, int len );
-@end example
-Absorbs @code{len} bytes from the buffer @code{buf}, eventually to be
-compressed and written to the file.
-
-Possible assignments to @code{bzerror}:
-@display
-      @code{BZ_PARAM_ERROR} 
-         if @code{b} is @code{NULL} or @code{buf} is @code{NULL} or @code{len < 0}
-      @code{BZ_SEQUENCE_ERROR} 
-         if b was opened with @code{BZ2_bzReadOpen}
-      @code{BZ_IO_ERROR} 
-         if there is an error writing the compressed file.
-      @code{BZ_OK} 
-         otherwise
-@end display
-
-
-
-
-@subsection @code{BZ2_bzWriteClose}
-@example
-   void BZ2_bzWriteClose ( int *bzerror, BZFILE* f,
-                           int abandon,
-                           unsigned int* nbytes_in,
-                           unsigned int* nbytes_out );
-
-   void BZ2_bzWriteClose64 ( int *bzerror, BZFILE* f,
-                             int abandon,
-                             unsigned int* nbytes_in_lo32,
-                             unsigned int* nbytes_in_hi32,
-                             unsigned int* nbytes_out_lo32,
-                             unsigned int* nbytes_out_hi32 );
-@end example
-
-Compresses and flushes to the compressed file all data so far supplied
-by @code{BZ2_bzWrite}.  The logical end-of-stream markers are also written, so
-subsequent calls to @code{BZ2_bzWrite} are illegal.  All memory associated 
-with the compressed file @code{b} is released.  
-@code{fflush} is called on the
-compressed file, but it is not @code{fclose}'d.
-
-If @code{BZ2_bzWriteClose} is called to clean up after an error, the only
-action is to release the memory.  The library records the error codes
-issued by previous calls, so this situation will be detected
-automatically.  There is no attempt to complete the compression
-operation, nor to @code{fflush} the compressed file.  You can force this
-behaviour to happen even in the case of no error, by passing a nonzero
-value to @code{abandon}.
-
-If @code{nbytes_in} is non-null, @code{*nbytes_in} will be set to be the
-total volume of uncompressed data handled.  Similarly, @code{nbytes_out}
-will be set to the total volume of compressed data written.  For 
-compatibility with older versions of the library, @code{BZ2_bzWriteClose}
-only yields the lower 32 bits of these counts.  Use
-@code{BZ2_bzWriteClose64} if you want the full 64 bit counts.  These
-two functions are otherwise absolutely identical.
-
-
-Possible assignments to @code{bzerror}:
-@display
-      @code{BZ_SEQUENCE_ERROR} 
-         if @code{b} was opened with @code{BZ2_bzReadOpen}
-      @code{BZ_IO_ERROR} 
-         if there is an error writing the compressed file
-      @code{BZ_OK} 
-         otherwise
-@end display
-
-@subsection Handling embedded compressed data streams
-
-The high-level library facilitates use of
-@code{bzip2} data streams which form some part of a surrounding, larger
-data stream.
-@itemize @bullet
-@item For writing, the library takes an open file handle, writes
-compressed data to it, @code{fflush}es it but does not @code{fclose} it.
-The calling application can write its own data before and after the
-compressed data stream, using that same file handle.
-@item Reading is more complex, and the facilities are not as general
-as they could be since generality is hard to reconcile with efficiency.
-@code{BZ2_bzRead} reads from the compressed file in blocks of size
-@code{BZ_MAX_UNUSED} bytes, and in doing so probably will overshoot
-the logical end of compressed stream.
-To recover this data once decompression has
-ended, call @code{BZ2_bzReadGetUnused} after the last call of @code{BZ2_bzRead}
-(the one returning @code{BZ_STREAM_END}) but before calling
-@code{BZ2_bzReadClose}.
-@end itemize
-
-This mechanism makes it easy to decompress multiple @code{bzip2}
-streams placed end-to-end.  As the end of one stream, when @code{BZ2_bzRead}
-returns @code{BZ_STREAM_END}, call @code{BZ2_bzReadGetUnused} to collect the
-unused data (copy it into your own buffer somewhere).  
-That data forms the start of the next compressed stream.
-To start uncompressing that next stream, call @code{BZ2_bzReadOpen} again,
-feeding in the unused data via the @code{unused}/@code{nUnused}
-parameters.
-Keep doing this until @code{BZ_STREAM_END} return coincides with the
-physical end of file (@code{feof(f)}).  In this situation
-@code{BZ2_bzReadGetUnused}
-will of course return no data.
-
-This should give some feel for how the high-level interface can be used.
-If you require extra flexibility, you'll have to bite the bullet and get
-to grips with the low-level interface.
-
-@subsection Standard file-reading/writing code
-Here's how you'd write data to a compressed file:
-@example @code
-FILE*   f;
-BZFILE* b;
-int     nBuf;
-char    buf[ /* whatever size you like */ ];
-int     bzerror;
-int     nWritten;
-
-f = fopen ( "myfile.bz2", "w" );
-if (!f) @{
-   /* handle error */
-@}
-b = BZ2_bzWriteOpen ( &bzerror, f, 9 );
-if (bzerror != BZ_OK) @{
-   BZ2_bzWriteClose ( b );
-   /* handle error */
-@}
-
-while ( /* condition */ ) @{
-   /* get data to write into buf, and set nBuf appropriately */
-   nWritten = BZ2_bzWrite ( &bzerror, b, buf, nBuf );
-   if (bzerror == BZ_IO_ERROR) @{ 
-      BZ2_bzWriteClose ( &bzerror, b );
-      /* handle error */
-   @}
-@}
-
-BZ2_bzWriteClose ( &bzerror, b );
-if (bzerror == BZ_IO_ERROR) @{
-   /* handle error */
-@}
-@end example
-And to read from a compressed file:
-@example
-FILE*   f;
-BZFILE* b;
-int     nBuf;
-char    buf[ /* whatever size you like */ ];
-int     bzerror;
-int     nWritten;
-
-f = fopen ( "myfile.bz2", "r" );
-if (!f) @{
-   /* handle error */
-@}
-b = BZ2_bzReadOpen ( &bzerror, f, 0, NULL, 0 );
-if (bzerror != BZ_OK) @{
-   BZ2_bzReadClose ( &bzerror, b );
-   /* handle error */
-@}
-
-bzerror = BZ_OK;
-while (bzerror == BZ_OK && /* arbitrary other conditions */) @{
-   nBuf = BZ2_bzRead ( &bzerror, b, buf, /* size of buf */ );
-   if (bzerror == BZ_OK) @{
-      /* do something with buf[0 .. nBuf-1] */
-   @}
-@}
-if (bzerror != BZ_STREAM_END) @{
-   BZ2_bzReadClose ( &bzerror, b );
-   /* handle error */
-@} else @{
-   BZ2_bzReadClose ( &bzerror );
-@}
-@end example
-
-
-
-@section Utility functions
-@subsection @code{BZ2_bzBuffToBuffCompress}
-@example
-   int BZ2_bzBuffToBuffCompress( char*         dest,
-                                 unsigned int* destLen,
-                                 char*         source,
-                                 unsigned int  sourceLen,
-                                 int           blockSize100k,
-                                 int           verbosity,
-                                 int           workFactor );
-@end example
-Attempts to compress the data in @code{source[0 .. sourceLen-1]}
-into the destination buffer, @code{dest[0 .. *destLen-1]}.
-If the destination buffer is big enough, @code{*destLen} is
-set to the size of the compressed data, and @code{BZ_OK} is
-returned.  If the compressed data won't fit, @code{*destLen}
-is unchanged, and @code{BZ_OUTBUFF_FULL} is returned.
-
-Compression in this manner is a one-shot event, done with a single call
-to this function.  The resulting compressed data is a complete
-@code{bzip2} format data stream.  There is no mechanism for making
-additional calls to provide extra input data.  If you want that kind of
-mechanism, use the low-level interface.
-
-For the meaning of parameters @code{blockSize100k}, @code{verbosity}
-and @code{workFactor}, @* see @code{BZ2_bzCompressInit}.
-
-To guarantee that the compressed data will fit in its buffer, allocate
-an output buffer of size 1% larger than the uncompressed data, plus
-six hundred extra bytes.
-
-@code{BZ2_bzBuffToBuffDecompress} will not write data at or
-beyond @code{dest[*destLen]}, even in case of buffer overflow.
-
-Possible return values:
-@display
-      @code{BZ_CONFIG_ERROR}
-         if the library has been mis-compiled
-      @code{BZ_PARAM_ERROR} 
-         if @code{dest} is @code{NULL} or @code{destLen} is @code{NULL}
-         or @code{blockSize100k < 1} or @code{blockSize100k > 9}
-         or @code{verbosity < 0} or @code{verbosity > 4} 
-         or @code{workFactor < 0} or @code{workFactor > 250}
-      @code{BZ_MEM_ERROR}
-         if insufficient memory is available 
-      @code{BZ_OUTBUFF_FULL}
-         if the size of the compressed data exceeds @code{*destLen}
-      @code{BZ_OK} 
-         otherwise
-@end display
-
-
-
-@subsection @code{BZ2_bzBuffToBuffDecompress}
-@example
-   int BZ2_bzBuffToBuffDecompress ( char*         dest,
-                                    unsigned int* destLen,
-                                    char*         source,
-                                    unsigned int  sourceLen,
-                                    int           small,
-                                    int           verbosity );
-@end example
-Attempts to decompress the data in @code{source[0 .. sourceLen-1]}
-into the destination buffer, @code{dest[0 .. *destLen-1]}.
-If the destination buffer is big enough, @code{*destLen} is
-set to the size of the uncompressed data, and @code{BZ_OK} is
-returned.  If the compressed data won't fit, @code{*destLen}
-is unchanged, and @code{BZ_OUTBUFF_FULL} is returned.
-
-@code{source} is assumed to hold a complete @code{bzip2} format
-data stream.  @* @code{BZ2_bzBuffToBuffDecompress} tries to decompress
-the entirety of the stream into the output buffer.
-
-For the meaning of parameters @code{small} and @code{verbosity},
-see @code{BZ2_bzDecompressInit}.
-
-Because the compression ratio of the compressed data cannot be known in
-advance, there is no easy way to guarantee that the output buffer will
-be big enough.  You may of course make arrangements in your code to
-record the size of the uncompressed data, but such a mechanism is beyond
-the scope of this library.
-
-@code{BZ2_bzBuffToBuffDecompress} will not write data at or
-beyond @code{dest[*destLen]}, even in case of buffer overflow.
-
-Possible return values:
-@display
-      @code{BZ_CONFIG_ERROR}
-         if the library has been mis-compiled
-      @code{BZ_PARAM_ERROR} 
-         if @code{dest} is @code{NULL} or @code{destLen} is @code{NULL}
-         or @code{small != 0 && small != 1}
-         or @code{verbosity < 0} or @code{verbosity > 4} 
-      @code{BZ_MEM_ERROR}
-         if insufficient memory is available 
-      @code{BZ_OUTBUFF_FULL}
-         if the size of the compressed data exceeds @code{*destLen}
-      @code{BZ_DATA_ERROR}
-         if a data integrity error was detected in the compressed data
-      @code{BZ_DATA_ERROR_MAGIC}
-         if the compressed data doesn't begin with the right magic bytes
-      @code{BZ_UNEXPECTED_EOF}
-         if the compressed data ends unexpectedly
-      @code{BZ_OK} 
-         otherwise
-@end display
-
-
-
-@section @code{zlib} compatibility functions
-Yoshioka Tsuneo has contributed some functions to
-give better @code{zlib} compatibility.  These functions are
-@code{BZ2_bzopen}, @code{BZ2_bzread}, @code{BZ2_bzwrite}, @code{BZ2_bzflush},
-@code{BZ2_bzclose},
-@code{BZ2_bzerror} and @code{BZ2_bzlibVersion}.
-These functions are not (yet) officially part of
-the library.  If they break, you get to keep all the pieces.
-Nevertheless, I think they work ok.
-@example
-typedef void BZFILE;
-
-const char * BZ2_bzlibVersion ( void );
-@end example
-Returns a string indicating the library version.
-@example
-BZFILE * BZ2_bzopen  ( const char *path, const char *mode );
-BZFILE * BZ2_bzdopen ( int        fd,    const char *mode );
-@end example
-Opens a @code{.bz2} file for reading or writing, using either its name
-or a pre-existing file descriptor. 
-Analogous to @code{fopen} and @code{fdopen}.
-@example         
-int BZ2_bzread  ( BZFILE* b, void* buf, int len );
-int BZ2_bzwrite ( BZFILE* b, void* buf, int len );
-@end example
-Reads/writes data from/to a previously opened @code{BZFILE}.
-Analogous to @code{fread} and @code{fwrite}.
-@example
-int  BZ2_bzflush ( BZFILE* b );
-void BZ2_bzclose ( BZFILE* b );
-@end example
-Flushes/closes a @code{BZFILE}.  @code{BZ2_bzflush} doesn't actually do
-anything.  Analogous to @code{fflush} and @code{fclose}.
-
-@example 
-const char * BZ2_bzerror ( BZFILE *b, int *errnum )
-@end example
-Returns a string describing the more recent error status of
-@code{b}, and also sets @code{*errnum} to its numerical value.
-
-
-@section Using the library in a @code{stdio}-free environment
-
-@subsection Getting rid of @code{stdio}
-
-In a deeply embedded application, you might want to use just
-the memory-to-memory functions.  You can do this conveniently
-by compiling the library with preprocessor symbol @code{BZ_NO_STDIO}
-defined.  Doing this gives you a library containing only the following
-eight functions:
-
-@code{BZ2_bzCompressInit}, @code{BZ2_bzCompress}, @code{BZ2_bzCompressEnd} @*
-@code{BZ2_bzDecompressInit}, @code{BZ2_bzDecompress}, @code{BZ2_bzDecompressEnd} @*
-@code{BZ2_bzBuffToBuffCompress}, @code{BZ2_bzBuffToBuffDecompress}
-
-When compiled like this, all functions will ignore @code{verbosity}
-settings.
-
-@subsection Critical error handling
-@code{libbzip2} contains a number of internal assertion checks which
-should, needless to say, never be activated.  Nevertheless, if an
-assertion should fail, behaviour depends on whether or not the library
-was compiled with @code{BZ_NO_STDIO} set.
-
-For a normal compile, an assertion failure yields the message
-@example
-   bzip2/libbzip2: internal error number N.
-   This is a bug in bzip2/libbzip2, 1.0.2, 30-Dec-2001.
-   Please report it to me at: jseward@@acm.org.  If this happened
-   when you were using some program which uses libbzip2 as a
-   component, you should also report this bug to the author(s)
-   of that program.  Please make an effort to report this bug;
-   timely and accurate bug reports eventually lead to higher
-   quality software.  Thanks.  Julian Seward, 30 December 2001.
-@end example
-where @code{N} is some error code number.  If @code{N == 1007}, it also
-prints some extra text advising the reader that unreliable memory is
-often associated with internal error 1007.  (This is a
-frequently-observed-phenomenon with versions 1.0.0/1.0.1).
-
-@code{exit(3)} is then called.
-
-For a @code{stdio}-free library, assertion failures result
-in a call to a function declared as:
-@example
-   extern void bz_internal_error ( int errcode );
-@end example
-The relevant code is passed as a parameter.  You should supply
-such a function.
-
-In either case, once an assertion failure has occurred, any 
-@code{bz_stream} records involved can be regarded as invalid.
-You should not attempt to resume normal operation with them.
-
-You may, of course, change critical error handling to suit
-your needs.  As I said above, critical errors indicate bugs
-in the library and should not occur.  All "normal" error
-situations are indicated via error return codes from functions,
-and can be recovered from.
-
-
-@section Making a Windows DLL
-Everything related to Windows has been contributed by Yoshioka Tsuneo
-@* (@code{QWF00133@@niftyserve.or.jp} /
-@code{tsuneo-y@@is.aist-nara.ac.jp}), so you should send your queries to
-him (but perhaps Cc: me, @code{jseward@@acm.org}).
-
-My vague understanding of what to do is: using Visual C++ 5.0,
-open the project file @code{libbz2.dsp}, and build.  That's all.
-
-If you can't
-open the project file for some reason, make a new one, naming these files:
-@code{blocksort.c}, @code{bzlib.c}, @code{compress.c}, 
-@code{crctable.c}, @code{decompress.c}, @code{huffman.c}, @*
-@code{randtable.c} and @code{libbz2.def}.  You will also need
-to name the header files @code{bzlib.h} and @code{bzlib_private.h}.
-
-If you don't use VC++, you may need to define the proprocessor symbol
-@code{_WIN32}. 
-
-Finally, @code{dlltest.c} is a sample program using the DLL.  It has a
-project file, @code{dlltest.dsp}.
-
-If you just want a makefile for Visual C, have a look at
-@code{makefile.msc}.
-
-Be aware that if you compile @code{bzip2} itself on Win32, you must set
-@code{BZ_UNIX} to 0 and @code{BZ_LCCWIN32} to 1, in the file
-@code{bzip2.c}, before compiling.  Otherwise the resulting binary won't
-work correctly.
-
-I haven't tried any of this stuff myself, but it all looks plausible.
-
-
-
-@chapter Miscellanea
-
-These are just some random thoughts of mine.  Your mileage may
-vary.
-
-@section Limitations of the compressed file format
-@code{bzip2-1.0}, @code{0.9.5} and @code{0.9.0}
-use exactly the same file format as the previous
-version, @code{bzip2-0.1}.  This decision was made in the interests of
-stability.  Creating yet another incompatible compressed file format
-would create further confusion and disruption for users.
-
-Nevertheless, this is not a painless decision.  Development
-work since the release of @code{bzip2-0.1} in August 1997
-has shown complexities in the file format which slow down
-decompression and, in retrospect, are unnecessary.  These are:
-@itemize @bullet
-@item The run-length encoder, which is the first of the 
-      compression transformations, is entirely irrelevant.
-      The original purpose was to protect the sorting algorithm
-      from the very worst case input: a string of repeated
-      symbols.  But algorithm steps Q6a and Q6b in the original
-      Burrows-Wheeler technical report (SRC-124) show how
-      repeats can be handled without difficulty in block
-      sorting.
-@item The randomisation mechanism doesn't really need to be
-      there.  Udi Manber and Gene Myers published a suffix
-      array construction algorithm a few years back, which
-      can be employed to sort any block, no matter how 
-      repetitive, in O(N log N) time.  Subsequent work by
-      Kunihiko Sadakane has produced a derivative O(N (log N)^2) 
-      algorithm which usually outperforms the Manber-Myers
-      algorithm.
-
-      I could have changed to Sadakane's algorithm, but I find
-      it to be slower than @code{bzip2}'s existing algorithm for
-      most inputs, and the randomisation mechanism protects
-      adequately against bad cases.  I didn't think it was
-      a good tradeoff to make.  Partly this is due to the fact
-      that I was not flooded with email complaints about
-      @code{bzip2-0.1}'s performance on repetitive data, so
-      perhaps it isn't a problem for real inputs.
-
-      Probably the best long-term solution,
-      and the one I have incorporated into 0.9.5 and above,
-      is to use the existing sorting
-      algorithm initially, and fall back to a O(N (log N)^2)
-      algorithm if the standard algorithm gets into difficulties.
-@item The compressed file format was never designed to be
-      handled by a library, and I have had to jump though
-      some hoops to produce an efficient implementation of
-      decompression.  It's a bit hairy.  Try passing
-      @code{decompress.c} through the C preprocessor 
-      and you'll see what I mean.  Much of this complexity
-      could have been avoided if the compressed size of
-      each block of data was recorded in the data stream.
-@item An Adler-32 checksum, rather than a CRC32 checksum,
-      would be faster to compute.
-@end itemize
-It would be fair to say that the @code{bzip2} format was frozen
-before I properly and fully understood the performance
-consequences of doing so.
-
-Improvements which I was able to incorporate into
-0.9.0, despite using the same file format, are:
-@itemize @bullet
-@item Single array implementation of the inverse BWT.  This
-      significantly speeds up decompression, presumably
-      because it reduces the number of cache misses.
-@item Faster inverse MTF transform for large MTF values.  The
-      new implementation is based on the notion of sliding blocks
-      of values.
-@item @code{bzip2-0.9.0} now reads and writes files with @code{fread}
-      and @code{fwrite}; version 0.1 used @code{putc} and @code{getc}.
-      Duh!  Well, you live and learn.
-
-@end itemize
-Further ahead, it would be nice 
-to be able to do random access into files.  This will 
-require some careful design of compressed file formats.
-
-
-
-@section Portability issues
-After some consideration, I have decided not to use
-GNU @code{autoconf} to configure 0.9.5 or 1.0.
-
-@code{autoconf}, admirable and wonderful though it is, 
-mainly assists with portability problems between Unix-like
-platforms.  But @code{bzip2} doesn't have much in the way
-of portability problems on Unix; most of the difficulties appear
-when porting to the Mac, or to Microsoft's operating systems.
-@code{autoconf} doesn't help in those cases, and brings in a 
-whole load of new complexity.
-
-Most people should be able to compile the library and program
-under Unix straight out-of-the-box, so to speak, especially 
-if you have a version of GNU C available.
-
-There are a couple of @code{__inline__} directives in the code.  GNU C
-(@code{gcc}) should be able to handle them.  If you're not using
-GNU C, your C compiler shouldn't see them at all.
-If your compiler does, for some reason, see them and doesn't
-like them, just @code{#define} @code{__inline__} to be @code{/* */}.  One
-easy way to do this is to compile with the flag @code{-D__inline__=}, 
-which should be understood by most Unix compilers.
-
-If you still have difficulties, try compiling with the macro
-@code{BZ_STRICT_ANSI} defined.  This should enable you to build the
-library in a strictly ANSI compliant environment.  Building the program
-itself like this is dangerous and not supported, since you remove
-@code{bzip2}'s checks against compressing directories, symbolic links,
-devices, and other not-really-a-file entities.  This could cause
-filesystem corruption!
-
-One other thing: if you create a @code{bzip2} binary for public
-distribution, please try and link it statically (@code{gcc -s}).  This
-avoids all sorts of library-version issues that others may encounter
-later on.
-
-If you build @code{bzip2} on Win32, you must set @code{BZ_UNIX} to 0 and
-@code{BZ_LCCWIN32} to 1, in the file @code{bzip2.c}, before compiling.
-Otherwise the resulting binary won't work correctly.
-
-
-
-@section Reporting bugs
-I tried pretty hard to make sure @code{bzip2} is
-bug free, both by design and by testing.  Hopefully
-you'll never need to read this section for real.
-
-Nevertheless, if @code{bzip2} dies with a segmentation
-fault, a bus error or an internal assertion failure, it
-will ask you to email me a bug report.  Experience with
-version 0.1 shows that almost all these problems can
-be traced to either compiler bugs or hardware problems.
-@itemize @bullet
-@item
-Recompile the program with no optimisation, and see if it
-works.  And/or try a different compiler.
-I heard all sorts of stories about various flavours
-of GNU C (and other compilers) generating bad code for
-@code{bzip2}, and I've run across two such examples myself.
-
-2.7.X versions of GNU C are known to generate bad code from
-time to time, at high optimisation levels.  
-If you get problems, try using the flags
-@code{-O2} @code{-fomit-frame-pointer} @code{-fno-strength-reduce}.
-You should specifically @emph{not} use @code{-funroll-loops}.
-
-You may notice that the Makefile runs six tests as part of
-the build process.  If the program passes all of these, it's
-a pretty good (but not 100%) indication that the compiler has
-done its job correctly.
-@item
-If @code{bzip2} crashes randomly, and the crashes are not
-repeatable, you may have a flaky memory subsystem.  @code{bzip2}
-really hammers your memory hierarchy, and if it's a bit marginal,
-you may get these problems.  Ditto if your disk or I/O subsystem
-is slowly failing.  Yup, this really does happen.
-
-Try using a different machine of the same type, and see if
-you can repeat the problem.
-@item This isn't really a bug, but ... If @code{bzip2} tells
-you your file is corrupted on decompression, and you
-obtained the file via FTP, there is a possibility that you
-forgot to tell FTP to do a binary mode transfer.  That absolutely
-will cause the file to be non-decompressible.  You'll have to transfer
-it again.
-@end itemize
-
-If you've incorporated @code{libbzip2} into your own program
-and are getting problems, please, please, please, check that the 
-parameters you are passing in calls to the library, are
-correct, and in accordance with what the documentation says
-is allowable.  I have tried to make the library robust against
-such problems, but I'm sure I haven't succeeded.
-
-Finally, if the above comments don't help, you'll have to send
-me a bug report.  Now, it's just amazing how many people will 
-send me a bug report saying something like
-@display
-   bzip2 crashed with segmentation fault on my machine
-@end display
-and absolutely nothing else.  Needless to say, a such a report
-is @emph{totally, utterly, completely and comprehensively 100% useless; 
-a waste of your time, my time, and net bandwidth}.
-With no details at all, there's no way I can possibly begin
-to figure out what the problem is.
-
-The rules of the game are: facts, facts, facts.  Don't omit
-them because "oh, they won't be relevant".  At the bare 
-minimum:
-@display
-   Machine type.  Operating system version.  
-   Exact version of @code{bzip2} (do @code{bzip2 -V}).  
-   Exact version of the compiler used.  
-   Flags passed to the compiler.
-@end display
-However, the most important single thing that will help me is
-the file that you were trying to compress or decompress at the
-time the problem happened.  Without that, my ability to do anything
-more than speculate about the cause, is limited.
-
-Please remember that I connect to the Internet with a modem, so
-you should contact me before mailing me huge files.
-
-
-@section Did you get the right package?
-
-@code{bzip2} is a resource hog.  It soaks up large amounts of CPU cycles
-and memory.  Also, it gives very large latencies.  In the worst case, you
-can feed many megabytes of uncompressed data into the library before
-getting any compressed output, so this probably rules out applications
-requiring interactive behaviour.
-
-These aren't faults of my implementation, I hope, but more
-an intrinsic property of the Burrows-Wheeler transform (unfortunately).  
-Maybe this isn't what you want.
-
-If you want a compressor and/or library which is faster, uses less
-memory but gets pretty good compression, and has minimal latency,
-consider Jean-loup
-Gailly's and Mark Adler's work, @code{zlib-1.1.3} and
-@code{gzip-1.2.4}.  Look for them at
-
-@code{http://www.zlib.org} and
-@code{http://www.gzip.org} respectively.
-
-For something faster and lighter still, you might try Markus F X J
-Oberhumer's @code{LZO} real-time compression/decompression library, at
-@* @code{http://wildsau.idv.uni-linz.ac.at/mfx/lzo.html}.
-
-If you want to use the @code{bzip2} algorithms to compress small blocks
-of data, 64k bytes or smaller, for example on an on-the-fly disk
-compressor, you'd be well advised not to use this library.  Instead,
-I've made a special library tuned for that kind of use.  It's part of
-@code{e2compr-0.40}, an on-the-fly disk compressor for the Linux
-@code{ext2} filesystem.  Look at
-@code{http://www.netspace.net.au/~reiter/e2compr}.
-
-
-
-@section Testing
-
-A record of the tests I've done.
-
-First, some data sets:
-@itemize @bullet
-@item B: a directory containing 6001 files, one for every length in the
-      range 0 to 6000 bytes.  The files contain random lowercase
-      letters.  18.7 megabytes.
-@item H: my home directory tree.  Documents, source code, mail files,
-      compressed data.  H contains B, and also a directory of 
-      files designed as boundary cases for the sorting; mostly very
-      repetitive, nasty files.  565 megabytes.
-@item A: directory tree holding various applications built from source:
-      @code{egcs}, @code{gcc-2.8.1}, KDE, GTK, Octave, etc.
-      2200 megabytes.
-@end itemize
-The tests conducted are as follows.  Each test means compressing 
-(a copy of) each file in the data set, decompressing it and
-comparing it against the original.
-
-First, a bunch of tests with block sizes and internal buffer
-sizes set very small, 
-to detect any problems with the
-blocking and buffering mechanisms.  
-This required modifying the source code so as to try to 
-break it.
-@enumerate
-@item Data set H, with
-      buffer size of 1 byte, and block size of 23 bytes.
-@item Data set B, buffer sizes 1 byte, block size 1 byte.
-@item As (2) but small-mode decompression.
-@item As (2) with block size 2 bytes.
-@item As (2) with block size 3 bytes.
-@item As (2) with block size 4 bytes.
-@item As (2) with block size 5 bytes.
-@item As (2) with block size 6 bytes and small-mode decompression.
-@item H with buffer size of 1 byte, but normal block
-      size (up to 900000 bytes).
-@end enumerate
-Then some tests with unmodified source code.
-@enumerate
-@item H, all settings normal.
-@item As (1), with small-mode decompress.
-@item H, compress with flag @code{-1}.
-@item H, compress with flag @code{-s}, decompress with flag @code{-s}.
-@item Forwards compatibility: H, @code{bzip2-0.1pl2} compressing,
-      @code{bzip2-0.9.5} decompressing, all settings normal.
-@item Backwards compatibility:  H, @code{bzip2-0.9.5} compressing,
-      @code{bzip2-0.1pl2} decompressing, all settings normal.
-@item Bigger tests: A, all settings normal.
-@item As (7), using the fallback (Sadakane-like) sorting algorithm.
-@item As (8), compress with flag @code{-1}, decompress with flag
-      @code{-s}.
-@item H, using the fallback sorting algorithm.
-@item Forwards compatibility: A, @code{bzip2-0.1pl2} compressing,
-      @code{bzip2-0.9.5} decompressing, all settings normal.
-@item Backwards compatibility:  A, @code{bzip2-0.9.5} compressing,
-      @code{bzip2-0.1pl2} decompressing, all settings normal.
-@item Misc test: about 400 megabytes of @code{.tar} files with
-      @code{bzip2} compiled with Checker (a memory access error
-       detector, like Purify).
-@item Misc tests to make sure it builds and runs ok on non-Linux/x86
-      platforms.
-@end enumerate
-These tests were conducted on a 225 MHz IDT WinChip machine, running
-Linux 2.0.36.  They represent nearly a week of continuous computation.
-All tests completed successfully.
-
-
-@section Further reading
-@code{bzip2} is not research work, in the sense that it doesn't present
-any new ideas.  Rather, it's an engineering exercise based on existing
-ideas.
-
-Four documents describe essentially all the ideas behind @code{bzip2}:
-@example
-Michael Burrows and D. J. Wheeler:
-  "A block-sorting lossless data compression algorithm"
-   10th May 1994. 
-   Digital SRC Research Report 124.
-   ftp://ftp.digital.com/pub/DEC/SRC/research-reports/SRC-124.ps.gz
-   If you have trouble finding it, try searching at the
-   New Zealand Digital Library, http://www.nzdl.org.
-
-Daniel S. Hirschberg and Debra A. LeLewer
-  "Efficient Decoding of Prefix Codes"
-   Communications of the ACM, April 1990, Vol 33, Number 4.
-   You might be able to get an electronic copy of this
-      from the ACM Digital Library.
-
-David J. Wheeler
-   Program bred3.c and accompanying document bred3.ps.
-   This contains the idea behind the multi-table Huffman
-   coding scheme.
-   ftp://ftp.cl.cam.ac.uk/users/djw3/
-
-Jon L. Bentley and Robert Sedgewick
-  "Fast Algorithms for Sorting and Searching Strings"
-   Available from Sedgewick's web page,
-   www.cs.princeton.edu/~rs
-@end example
-The following paper gives valuable additional insights into the
-algorithm, but is not immediately the basis of any code
-used in bzip2.
-@example
-Peter Fenwick:
-   Block Sorting Text Compression
-   Proceedings of the 19th Australasian Computer Science Conference,
-     Melbourne, Australia.  Jan 31 - Feb 2, 1996.
-   ftp://ftp.cs.auckland.ac.nz/pub/peter-f/ACSC96paper.ps
-@end example
-Kunihiko Sadakane's sorting algorithm, mentioned above,
-is available from:
-@example
-http://naomi.is.s.u-tokyo.ac.jp/~sada/papers/Sada98b.ps.gz
-@end example
-The Manber-Myers suffix array construction
-algorithm is described in a paper
-available from:
-@example
-http://www.cs.arizona.edu/people/gene/PAPERS/suffix.ps
-@end example
-Finally, the following paper documents some recent investigations
-I made into the performance of sorting algorithms:
-@example
-Julian Seward:
-   On the Performance of BWT Sorting Algorithms
-   Proceedings of the IEEE Data Compression Conference 2000
-     Snowbird, Utah.  28-30 March 2000.
-@end example
-
-
-@contents
-
-@bye
-
diff --git a/manual.xml b/manual.xml
new file mode 100644
index 0000000..1ab5bd7
--- /dev/null
+++ b/manual.xml
@@ -0,0 +1,2966 @@
+<?xml version="1.0"?> <!-- -*- sgml -*- -->
+<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN"
+  "http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd"[
+
+<!-- various strings, dates etc. common to all docs -->
+<!ENTITY % common-ents SYSTEM "entities.xml"> %common-ents;
+]>
+
+<book lang="en" id="userman" xreflabel="bzip2 Manual">
+
+ <bookinfo>
+  <title>bzip2 and libbzip2, version 1.0.3</title>
+  <subtitle>A program and library for data compression</subtitle>
+  <copyright>
+   <year>&bz-lifespan;</year>
+   <holder>Julian Seward</holder>
+  </copyright>
+  <releaseinfo>Version &bz-version; of &bz-date;</releaseinfo>
+
+  <authorgroup>
+   <author>
+    <firstname>Julian</firstname>
+    <surname>Seward</surname>
+    <affiliation>
+     <orgname>&bz-url;</orgname>
+    </affiliation>
+   </author>
+  </authorgroup>
+
+  <legalnotice>
+
+  <para>This program, <computeroutput>bzip2</computeroutput>, the
+  associated library <computeroutput>libbzip2</computeroutput>, and
+  all documentation, are copyright &copy; &bz-lifespan; Julian Seward.
+  All rights reserved.</para>
+
+  <para>Redistribution and use in source and binary forms, with
+  or without modification, are permitted provided that the
+  following conditions are met:</para>
+
+  <itemizedlist mark='bullet'>
+
+   <listitem><para>Redistributions of source code must retain the
+   above copyright notice, this list of conditions and the
+   following disclaimer.</para></listitem>
+
+   <listitem><para>The origin of this software must not be
+   misrepresented; you must not claim that you wrote the original
+   software.  If you use this software in a product, an
+   acknowledgment in the product documentation would be
+   appreciated but is not required.</para></listitem>
+
+   <listitem><para>Altered source versions must be plainly marked
+   as such, and must not be misrepresented as being the original
+   software.</para></listitem>
+
+   <listitem><para>The name of the author may not be used to
+   endorse or promote products derived from this software without
+   specific prior written permission.</para></listitem>
+
+  </itemizedlist>
+
+  <para>THIS SOFTWARE IS PROVIDED BY THE AUTHOR "AS IS" AND ANY
+  EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,
+  THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
+  PARTICULAR PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE
+  AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
+  EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED
+  TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+  DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+  ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+  LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING
+  IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+  THE POSSIBILITY OF SUCH DAMAGE.</para>
+
+ <para>PATENTS: To the best of my knowledge,
+ <computeroutput>bzip2</computeroutput> and
+ <computeroutput>libbzip2</computeroutput> do not use any patented
+ algorithms.  However, I do not have the resources to carry
+ out a patent search.  Therefore I cannot give any guarantee of
+ the above statement.
+ </para>
+
+</legalnotice>
+
+</bookinfo>
+
+
+
+<chapter id="intro" xreflabel="Introduction">
+<title>Introduction</title>
+
+<para><computeroutput>bzip2</computeroutput> compresses files
+using the Burrows-Wheeler block-sorting text compression
+algorithm, and Huffman coding.  Compression is generally
+considerably better than that achieved by more conventional
+LZ77/LZ78-based compressors, and approaches the performance of
+the PPM family of statistical compressors.</para>
+
+<para><computeroutput>bzip2</computeroutput> is built on top of
+<computeroutput>libbzip2</computeroutput>, a flexible library for
+handling compressed data in the
+<computeroutput>bzip2</computeroutput> format.  This manual
+describes both how to use the program and how to work with the
+library interface.  Most of the manual is devoted to this
+library, not the program, which is good news if your interest is
+only in the program.</para>
+
+<itemizedlist mark='bullet'>
+
+ <listitem><para><xref linkend="using"/> describes how to use
+ <computeroutput>bzip2</computeroutput>; this is the only part
+ you need to read if you just want to know how to operate the
+ program.</para></listitem>
+
+ <listitem><para><xref linkend="libprog"/> describes the
+ programming interfaces in detail, and</para></listitem>
+
+ <listitem><para><xref linkend="misc"/> records some
+ miscellaneous notes which I thought ought to be recorded
+ somewhere.</para></listitem>
+
+</itemizedlist>
+
+</chapter>
+
+
+<chapter id="using" xreflabel="How to use bzip2">
+<title>How to use bzip2</title>
+
+<para>This chapter contains a copy of the
+<computeroutput>bzip2</computeroutput> man page, and nothing
+else.</para>
+
+<sect1 id="name" xreflabel="NAME">
+<title>NAME</title>
+
+<itemizedlist mark='bullet'>
+
+ <listitem><para><computeroutput>bzip2</computeroutput>,
+  <computeroutput>bunzip2</computeroutput> - a block-sorting file
+  compressor, v1.0.3</para></listitem>
+
+ <listitem><para><computeroutput>bzcat</computeroutput> -
+   decompresses files to stdout</para></listitem>
+
+ <listitem><para><computeroutput>bzip2recover</computeroutput> -
+   recovers data from damaged bzip2 files</para></listitem>
+
+</itemizedlist>
+
+</sect1>
+
+
+<sect1 id="synopsis" xreflabel="SYNOPSIS">
+<title>SYNOPSIS</title>
+
+<itemizedlist mark='bullet'>
+
+ <listitem><para><computeroutput>bzip2</computeroutput> [
+  -cdfkqstvzVL123456789 ] [ filenames ...  ]</para></listitem>
+
+ <listitem><para><computeroutput>bunzip2</computeroutput> [
+  -fkvsVL ] [ filenames ...  ]</para></listitem>
+
+ <listitem><para><computeroutput>bzcat</computeroutput> [ -s ] [
+  filenames ...  ]</para></listitem>
+
+ <listitem><para><computeroutput>bzip2recover</computeroutput>
+  filename</para></listitem>
+
+</itemizedlist>
+
+</sect1>
+
+
+<sect1 id="description" xreflabel="DESCRIPTION">
+<title>DESCRIPTION</title>
+
+<para><computeroutput>bzip2</computeroutput> compresses files
+using the Burrows-Wheeler block sorting text compression
+algorithm, and Huffman coding.  Compression is generally
+considerably better than that achieved by more conventional
+LZ77/LZ78-based compressors, and approaches the performance of
+the PPM family of statistical compressors.</para>
+
+<para>The command-line options are deliberately very similar to
+those of GNU <computeroutput>gzip</computeroutput>, but they are
+not identical.</para>
+
+<para><computeroutput>bzip2</computeroutput> expects a list of
+file names to accompany the command-line flags.  Each file is
+replaced by a compressed version of itself, with the name
+<computeroutput>original_name.bz2</computeroutput>.  Each
+compressed file has the same modification date, permissions, and,
+when possible, ownership as the corresponding original, so that
+these properties can be correctly restored at decompression time.
+File name handling is naive in the sense that there is no
+mechanism for preserving original file names, permissions,
+ownerships or dates in filesystems which lack these concepts, or
+have serious file name length restrictions, such as
+MS-DOS.</para>
+
+<para><computeroutput>bzip2</computeroutput> and
+<computeroutput>bunzip2</computeroutput> will by default not
+overwrite existing files.  If you want this to happen, specify
+the <computeroutput>-f</computeroutput> flag.</para>
+
+<para>If no file names are specified,
+<computeroutput>bzip2</computeroutput> compresses from standard
+input to standard output.  In this case,
+<computeroutput>bzip2</computeroutput> will decline to write
+compressed output to a terminal, as this would be entirely
+incomprehensible and therefore pointless.</para>
+
+<para><computeroutput>bunzip2</computeroutput> (or
+<computeroutput>bzip2 -d</computeroutput>) decompresses all
+specified files.  Files which were not created by
+<computeroutput>bzip2</computeroutput> will be detected and
+ignored, and a warning issued.
+<computeroutput>bzip2</computeroutput> attempts to guess the
+filename for the decompressed file from that of the compressed
+file as follows:</para>
+
+<itemizedlist mark='bullet'>
+
+ <listitem><para><computeroutput>filename.bz2 </computeroutput>
+  becomes
+  <computeroutput>filename</computeroutput></para></listitem>
+
+ <listitem><para><computeroutput>filename.bz </computeroutput>
+  becomes
+  <computeroutput>filename</computeroutput></para></listitem>
+
+ <listitem><para><computeroutput>filename.tbz2</computeroutput>
+  becomes
+  <computeroutput>filename.tar</computeroutput></para></listitem>
+
+ <listitem><para><computeroutput>filename.tbz </computeroutput>
+  becomes
+  <computeroutput>filename.tar</computeroutput></para></listitem>
+
+ <listitem><para><computeroutput>anyothername </computeroutput>
+  becomes
+  <computeroutput>anyothername.out</computeroutput></para></listitem>
+
+</itemizedlist>
+
+<para>If the file does not end in one of the recognised endings,
+<computeroutput>.bz2</computeroutput>,
+<computeroutput>.bz</computeroutput>,
+<computeroutput>.tbz2</computeroutput> or
+<computeroutput>.tbz</computeroutput>,
+<computeroutput>bzip2</computeroutput> complains that it cannot
+guess the name of the original file, and uses the original name
+with <computeroutput>.out</computeroutput> appended.</para>
+
+<para>As with compression, supplying no filenames causes
+decompression from standard input to standard output.</para>
+
+<para><computeroutput>bunzip2</computeroutput> will correctly
+decompress a file which is the concatenation of two or more
+compressed files.  The result is the concatenation of the
+corresponding uncompressed files.  Integrity testing
+(<computeroutput>-t</computeroutput>) of concatenated compressed
+files is also supported.</para>
+
+<para>You can also compress or decompress files to the standard
+output by giving the <computeroutput>-c</computeroutput> flag.
+Multiple files may be compressed and decompressed like this.  The
+resulting outputs are fed sequentially to stdout.  Compression of
+multiple files in this manner generates a stream containing
+multiple compressed file representations.  Such a stream can be
+decompressed correctly only by
+<computeroutput>bzip2</computeroutput> version 0.9.0 or later.
+Earlier versions of <computeroutput>bzip2</computeroutput> will
+stop after decompressing the first file in the stream.</para>
+
+<para><computeroutput>bzcat</computeroutput> (or
+<computeroutput>bzip2 -dc</computeroutput>) decompresses all
+specified files to the standard output.</para>
+
+<para><computeroutput>bzip2</computeroutput> will read arguments
+from the environment variables
+<computeroutput>BZIP2</computeroutput> and
+<computeroutput>BZIP</computeroutput>, in that order, and will
+process them before any arguments read from the command line.
+This gives a convenient way to supply default arguments.</para>
+
+<para>Compression is always performed, even if the compressed
+file is slightly larger than the original.  Files of less than
+about one hundred bytes tend to get larger, since the compression
+mechanism has a constant overhead in the region of 50 bytes.
+Random data (including the output of most file compressors) is
+coded at about 8.05 bits per byte, giving an expansion of around
+0.5%.</para>
+
+<para>As a self-check for your protection,
+<computeroutput>bzip2</computeroutput> uses 32-bit CRCs to make
+sure that the decompressed version of a file is identical to the
+original.  This guards against corruption of the compressed data,
+and against undetected bugs in
+<computeroutput>bzip2</computeroutput> (hopefully very unlikely).
+The chances of data corruption going undetected is microscopic,
+about one chance in four billion for each file processed.  Be
+aware, though, that the check occurs upon decompression, so it
+can only tell you that something is wrong.  It can't help you
+recover the original uncompressed data.  You can use
+<computeroutput>bzip2recover</computeroutput> to try to recover
+data from damaged files.</para>
+
+<para>Return values: 0 for a normal exit, 1 for environmental
+problems (file not found, invalid flags, I/O errors, etc.), 2
+to indicate a corrupt compressed file, 3 for an internal
+consistency error (eg, bug) which caused
+<computeroutput>bzip2</computeroutput> to panic.</para>
+
+</sect1>
+
+
+<sect1 id="options" xreflabel="OPTIONS">
+<title>OPTIONS</title>
+
+<variablelist>
+
+ <varlistentry>
+ <term><computeroutput>-c --stdout</computeroutput></term>
+ <listitem><para>Compress or decompress to standard
+  output.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term><computeroutput>-d --decompress</computeroutput></term>
+ <listitem><para>Force decompression.
+  <computeroutput>bzip2</computeroutput>,
+  <computeroutput>bunzip2</computeroutput> and
+  <computeroutput>bzcat</computeroutput> are really the same
+  program, and the decision about what actions to take is done on
+  the basis of which name is used.  This flag overrides that
+  mechanism, and forces bzip2 to decompress.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term><computeroutput>-z --compress</computeroutput></term>
+ <listitem><para>The complement to
+  <computeroutput>-d</computeroutput>: forces compression,
+  regardless of the invokation name.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term><computeroutput>-t --test</computeroutput></term>
+ <listitem><para>Check integrity of the specified file(s), but
+  don't decompress them.  This really performs a trial
+  decompression and throws away the result.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term><computeroutput>-f --force</computeroutput></term>
+ <listitem><para>Force overwrite of output files.  Normally,
+  <computeroutput>bzip2</computeroutput> will not overwrite
+  existing output files.  Also forces
+  <computeroutput>bzip2</computeroutput> to break hard links to
+  files, which it otherwise wouldn't do.</para>
+  <para><computeroutput>bzip2</computeroutput> normally declines
+  to decompress files which don't have the correct magic header
+  bytes. If forced (<computeroutput>-f</computeroutput>),
+  however, it will pass such files through unmodified. This is
+  how GNU <computeroutput>gzip</computeroutput> behaves.</para>
+ </listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term><computeroutput>-k --keep</computeroutput></term>
+ <listitem><para>Keep (don't delete) input files during
+  compression or decompression.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term><computeroutput>-s --small</computeroutput></term>
+ <listitem><para>Reduce memory usage, for compression,
+  decompression and testing.  Files are decompressed and tested
+  using a modified algorithm which only requires 2.5 bytes per
+  block byte.  This means any file can be decompressed in 2300k
+  of memory, albeit at about half the normal speed.</para>
+  <para>During compression, <computeroutput>-s</computeroutput>
+  selects a block size of 200k, which limits memory use to around
+  the same figure, at the expense of your compression ratio.  In
+  short, if your machine is low on memory (8 megabytes or less),
+  use <computeroutput>-s</computeroutput> for everything.  See
+  <xref linkend="memory-management"/> below.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term><computeroutput>-q --quiet</computeroutput></term>
+ <listitem><para>Suppress non-essential warning messages.
+  Messages pertaining to I/O errors and other critical events
+  will not be suppressed.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term><computeroutput>-v --verbose</computeroutput></term>
+ <listitem><para>Verbose mode -- show the compression ratio for
+  each file processed.  Further
+  <computeroutput>-v</computeroutput>'s increase the verbosity
+  level, spewing out lots of information which is primarily of
+  interest for diagnostic purposes.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term><computeroutput>-L --license -V --version</computeroutput></term>
+ <listitem><para>Display the software version, license terms and
+  conditions.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term><computeroutput>-1</computeroutput> (or
+ <computeroutput>--fast</computeroutput>) to
+ <computeroutput>-9</computeroutput> (or
+ <computeroutput>-best</computeroutput>)</term>
+ <listitem><para>Set the block size to 100 k, 200 k ...  900 k
+  when compressing.  Has no effect when decompressing.  See <xref
+  linkend="memory-management" /> below.  The
+  <computeroutput>--fast</computeroutput> and
+  <computeroutput>--best</computeroutput> aliases are primarily
+  for GNU <computeroutput>gzip</computeroutput> compatibility.
+  In particular, <computeroutput>--fast</computeroutput> doesn't
+  make things significantly faster.  And
+  <computeroutput>--best</computeroutput> merely selects the
+  default behaviour.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term><computeroutput>--</computeroutput></term>
+ <listitem><para>Treats all subsequent arguments as file names,
+  even if they start with a dash.  This is so you can handle
+  files with names beginning with a dash, for example:
+  <computeroutput>bzip2 --
+  -myfilename</computeroutput>.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+ <term><computeroutput>--repetitive-fast</computeroutput></term>
+ <term><computeroutput>--repetitive-best</computeroutput></term>
+ <listitem><para>These flags are redundant in versions 0.9.5 and
+  above.  They provided some coarse control over the behaviour of
+  the sorting algorithm in earlier versions, which was sometimes
+  useful.  0.9.5 and above have an improved algorithm which
+  renders these flags irrelevant.</para></listitem>
+ </varlistentry>
+
+</variablelist>
+
+</sect1>
+
+
+<sect1 id="memory-management" xreflabel="MEMORY MANAGEMENT">
+<title>MEMORY MANAGEMENT</title>
+
+<para><computeroutput>bzip2</computeroutput> compresses large
+files in blocks.  The block size affects both the compression
+ratio achieved, and the amount of memory needed for compression
+and decompression.  The flags <computeroutput>-1</computeroutput>
+through <computeroutput>-9</computeroutput> specify the block
+size to be 100,000 bytes through 900,000 bytes (the default)
+respectively.  At decompression time, the block size used for
+compression is read from the header of the compressed file, and
+<computeroutput>bunzip2</computeroutput> then allocates itself
+just enough memory to decompress the file.  Since block sizes are
+stored in compressed files, it follows that the flags
+<computeroutput>-1</computeroutput> to
+<computeroutput>-9</computeroutput> are irrelevant to and so
+ignored during decompression.</para>
+
+<para>Compression and decompression requirements, in bytes, can be
+estimated as:</para>
+<programlisting>
+Compression:   400k + ( 8 x block size )
+
+Decompression: 100k + ( 4 x block size ), or
+               100k + ( 2.5 x block size )
+</programlisting>
+
+<para>Larger block sizes give rapidly diminishing marginal
+returns.  Most of the compression comes from the first two or
+three hundred k of block size, a fact worth bearing in mind when
+using <computeroutput>bzip2</computeroutput> on small machines.
+It is also important to appreciate that the decompression memory
+requirement is set at compression time by the choice of block
+size.</para>
+
+<para>For files compressed with the default 900k block size,
+<computeroutput>bunzip2</computeroutput> will require about 3700
+kbytes to decompress.  To support decompression of any file on a
+4 megabyte machine, <computeroutput>bunzip2</computeroutput> has
+an option to decompress using approximately half this amount of
+memory, about 2300 kbytes.  Decompression speed is also halved,
+so you should use this option only where necessary.  The relevant
+flag is <computeroutput>-s</computeroutput>.</para>
+
+<para>In general, try and use the largest block size memory
+constraints allow, since that maximises the compression achieved.
+Compression and decompression speed are virtually unaffected by
+block size.</para>
+
+<para>Another significant point applies to files which fit in a
+single block -- that means most files you'd encounter using a
+large block size.  The amount of real memory touched is
+proportional to the size of the file, since the file is smaller
+than a block.  For example, compressing a file 20,000 bytes long
+with the flag <computeroutput>-9</computeroutput> will cause the
+compressor to allocate around 7600k of memory, but only touch
+400k + 20000 * 8 = 560 kbytes of it.  Similarly, the decompressor
+will allocate 3700k but only touch 100k + 20000 * 4 = 180
+kbytes.</para>
+
+<para>Here is a table which summarises the maximum memory usage
+for different block sizes.  Also recorded is the total compressed
+size for 14 files of the Calgary Text Compression Corpus
+totalling 3,141,622 bytes.  This column gives some feel for how
+compression varies with block size.  These figures tend to
+understate the advantage of larger block sizes for larger files,
+since the Corpus is dominated by smaller files.</para>
+
+<programlisting>
+        Compress   Decompress   Decompress   Corpus
+Flag     usage      usage       -s usage     Size
+
+ -1      1200k       500k         350k      914704
+ -2      2000k       900k         600k      877703
+ -3      2800k      1300k         850k      860338
+ -4      3600k      1700k        1100k      846899
+ -5      4400k      2100k        1350k      845160
+ -6      5200k      2500k        1600k      838626
+ -7      6100k      2900k        1850k      834096
+ -8      6800k      3300k        2100k      828642
+ -9      7600k      3700k        2350k      828642
+</programlisting>
+
+</sect1>
+
+
+<sect1 id="recovering" xreflabel="RECOVERING DATA FROM DAMAGED FILES">
+<title>RECOVERING DATA FROM DAMAGED FILES</title>
+
+<para><computeroutput>bzip2</computeroutput> compresses files in
+blocks, usually 900kbytes long.  Each block is handled
+independently.  If a media or transmission error causes a
+multi-block <computeroutput>.bz2</computeroutput> file to become
+damaged, it may be possible to recover data from the undamaged
+blocks in the file.</para>
+
+<para>The compressed representation of each block is delimited by
+a 48-bit pattern, which makes it possible to find the block
+boundaries with reasonable certainty.  Each block also carries
+its own 32-bit CRC, so damaged blocks can be distinguished from
+undamaged ones.</para>
+
+<para><computeroutput>bzip2recover</computeroutput> is a simple
+program whose purpose is to search for blocks in
+<computeroutput>.bz2</computeroutput> files, and write each block
+out into its own <computeroutput>.bz2</computeroutput> file.  You
+can then use <computeroutput>bzip2 -t</computeroutput> to test
+the integrity of the resulting files, and decompress those which
+are undamaged.</para>
+
+<para><computeroutput>bzip2recover</computeroutput> takes a
+single argument, the name of the damaged file, and writes a
+number of files <computeroutput>rec0001file.bz2</computeroutput>,
+<computeroutput>rec0002file.bz2</computeroutput>, etc, containing
+the extracted blocks.  The output filenames are designed so that
+the use of wildcards in subsequent processing -- for example,
+<computeroutput>bzip2 -dc rec*file.bz2 &#62;
+recovered_data</computeroutput> -- lists the files in the correct
+order.</para>
+
+<para><computeroutput>bzip2recover</computeroutput> should be of
+most use dealing with large <computeroutput>.bz2</computeroutput>
+files, as these will contain many blocks.  It is clearly futile
+to use it on damaged single-block files, since a damaged block
+cannot be recovered.  If you wish to minimise any potential data
+loss through media or transmission errors, you might consider
+compressing with a smaller block size.</para>
+
+</sect1>
+
+
+<sect1 id="performance" xreflabel="PERFORMANCE NOTES">
+<title>PERFORMANCE NOTES</title>
+
+<para>The sorting phase of compression gathers together similar
+strings in the file.  Because of this, files containing very long
+runs of repeated symbols, like "aabaabaabaab ..."  (repeated
+several hundred times) may compress more slowly than normal.
+Versions 0.9.5 and above fare much better than previous versions
+in this respect.  The ratio between worst-case and average-case
+compression time is in the region of 10:1.  For previous
+versions, this figure was more like 100:1.  You can use the
+<computeroutput>-vvvv</computeroutput> option to monitor progress
+in great detail, if you want.</para>
+
+<para>Decompression speed is unaffected by these
+phenomena.</para>
+
+<para><computeroutput>bzip2</computeroutput> usually allocates
+several megabytes of memory to operate in, and then charges all
+over it in a fairly random fashion.  This means that performance,
+both for compressing and decompressing, is largely determined by
+the speed at which your machine can service cache misses.
+Because of this, small changes to the code to reduce the miss
+rate have been observed to give disproportionately large
+performance improvements.  I imagine
+<computeroutput>bzip2</computeroutput> will perform best on
+machines with very large caches.</para>
+
+</sect1>
+
+
+
+<sect1 id="caveats" xreflabel="CAVEATS">
+<title>CAVEATS</title>
+
+<para>I/O error messages are not as helpful as they could be.
+<computeroutput>bzip2</computeroutput> tries hard to detect I/O
+errors and exit cleanly, but the details of what the problem is
+sometimes seem rather misleading.</para>
+
+<para>This manual page pertains to version &bz-version; of
+<computeroutput>bzip2</computeroutput>.  Compressed data created
+by this version is entirely forwards and backwards compatible
+with the previous public releases, versions 0.1pl2, 0.9.0 and
+0.9.5, 1.0.0, 1.0.1 and 1.0.2, but with the following exception: 0.9.0
+and above can correctly decompress multiple concatenated
+compressed files.  0.1pl2 cannot do this; it will stop after
+decompressing just the first file in the stream.</para>
+
+<para><computeroutput>bzip2recover</computeroutput> versions
+prior to 1.0.2 used 32-bit integers to represent bit positions in
+compressed files, so it could not handle compressed files more
+than 512 megabytes long.  Versions 1.0.2 and above use 64-bit ints
+on some platforms which support them (GNU supported targets, and
+Windows). To establish whether or not
+<computeroutput>bzip2recover</computeroutput> was built with such
+a limitation, run it without arguments. In any event you can
+build yourself an unlimited version if you can recompile it with
+<computeroutput>MaybeUInt64</computeroutput> set to be an
+unsigned 64-bit integer.</para>
+
+</sect1>
+
+
+
+<sect1 id="author" xreflabel="AUTHOR">
+<title>AUTHOR</title>
+
+<para>Julian Seward,
+<computeroutput>&bz-email;</computeroutput></para>
+
+<para>The ideas embodied in
+<computeroutput>bzip2</computeroutput> are due to (at least) the
+following people: Michael Burrows and David Wheeler (for the
+block sorting transformation), David Wheeler (again, for the
+Huffman coder), Peter Fenwick (for the structured coding model in
+the original <computeroutput>bzip</computeroutput>, and many
+refinements), and Alistair Moffat, Radford Neal and Ian Witten
+(for the arithmetic coder in the original
+<computeroutput>bzip</computeroutput>).  I am much indebted for
+their help, support and advice.  See the manual in the source
+distribution for pointers to sources of documentation.  Christian
+von Roques encouraged me to look for faster sorting algorithms,
+so as to speed up compression.  Bela Lubkin encouraged me to
+improve the worst-case compression performance.  
+Donna Robinson XMLised the documentation.
+Many people sent
+patches, helped with portability problems, lent machines, gave
+advice and were generally helpful.</para>
+
+</sect1>
+
+</chapter>
+
+
+
+<chapter id="libprog" xreflabel="Programming with libbzip2">
+<title>
+Programming with <computeroutput>libbzip2</computeroutput>
+</title>
+
+<para>This chapter describes the programming interface to
+<computeroutput>libbzip2</computeroutput>.</para>
+
+<para>For general background information, particularly about
+memory use and performance aspects, you'd be well advised to read
+<xref linkend="using"/> as well.</para>
+
+
+<sect1 id="top-level" xreflabel="Top-level structure">
+<title>Top-level structure</title>
+
+<para><computeroutput>libbzip2</computeroutput> is a flexible
+library for compressing and decompressing data in the
+<computeroutput>bzip2</computeroutput> data format.  Although
+packaged as a single entity, it helps to regard the library as
+three separate parts: the low level interface, and the high level
+interface, and some utility functions.</para>
+
+<para>The structure of
+<computeroutput>libbzip2</computeroutput>'s interfaces is similar
+to that of Jean-loup Gailly's and Mark Adler's excellent
+<computeroutput>zlib</computeroutput> library.</para>
+
+<para>All externally visible symbols have names beginning
+<computeroutput>BZ2_</computeroutput>.  This is new in version
+1.0.  The intention is to minimise pollution of the namespaces of
+library clients.</para>
+
+<para>To use any part of the library, you need to
+<computeroutput>#include &lt;bzlib.h&gt;</computeroutput>
+into your sources.</para>
+
+
+
+<sect2 id="ll-summary" xreflabel="Low-level summary">
+<title>Low-level summary</title>
+
+<para>This interface provides services for compressing and
+decompressing data in memory.  There's no provision for dealing
+with files, streams or any other I/O mechanisms, just straight
+memory-to-memory work.  In fact, this part of the library can be
+compiled without inclusion of
+<computeroutput>stdio.h</computeroutput>, which may be helpful
+for embedded applications.</para>
+
+<para>The low-level part of the library has no global variables
+and is therefore thread-safe.</para>
+
+<para>Six routines make up the low level interface:
+<computeroutput>BZ2_bzCompressInit</computeroutput>,
+<computeroutput>BZ2_bzCompress</computeroutput>, and
+<computeroutput>BZ2_bzCompressEnd</computeroutput> for
+compression, and a corresponding trio
+<computeroutput>BZ2_bzDecompressInit</computeroutput>,
+<computeroutput>BZ2_bzDecompress</computeroutput> and
+<computeroutput>BZ2_bzDecompressEnd</computeroutput> for
+decompression.  The <computeroutput>*Init</computeroutput>
+functions allocate memory for compression/decompression and do
+other initialisations, whilst the
+<computeroutput>*End</computeroutput> functions close down
+operations and release memory.</para>
+
+<para>The real work is done by
+<computeroutput>BZ2_bzCompress</computeroutput> and
+<computeroutput>BZ2_bzDecompress</computeroutput>.  These
+compress and decompress data from a user-supplied input buffer to
+a user-supplied output buffer.  These buffers can be any size;
+arbitrary quantities of data are handled by making repeated calls
+to these functions.  This is a flexible mechanism allowing a
+consumer-pull style of activity, or producer-push, or a mixture
+of both.</para>
+
+</sect2>
+
+
+<sect2 id="hl-summary" xreflabel="High-level summary">
+<title>High-level summary</title>
+
+<para>This interface provides some handy wrappers around the
+low-level interface to facilitate reading and writing
+<computeroutput>bzip2</computeroutput> format files
+(<computeroutput>.bz2</computeroutput> files).  The routines
+provide hooks to facilitate reading files in which the
+<computeroutput>bzip2</computeroutput> data stream is embedded
+within some larger-scale file structure, or where there are
+multiple <computeroutput>bzip2</computeroutput> data streams
+concatenated end-to-end.</para>
+
+<para>For reading files,
+<computeroutput>BZ2_bzReadOpen</computeroutput>,
+<computeroutput>BZ2_bzRead</computeroutput>,
+<computeroutput>BZ2_bzReadClose</computeroutput> and 
+<computeroutput>BZ2_bzReadGetUnused</computeroutput> are
+supplied.  For writing files,
+<computeroutput>BZ2_bzWriteOpen</computeroutput>,
+<computeroutput>BZ2_bzWrite</computeroutput> and
+<computeroutput>BZ2_bzWriteFinish</computeroutput> are
+available.</para>
+
+<para>As with the low-level library, no global variables are used
+so the library is per se thread-safe.  However, if I/O errors
+occur whilst reading or writing the underlying compressed files,
+you may have to consult <computeroutput>errno</computeroutput> to
+determine the cause of the error.  In that case, you'd need a C
+library which correctly supports
+<computeroutput>errno</computeroutput> in a multithreaded
+environment.</para>
+
+<para>To make the library a little simpler and more portable,
+<computeroutput>BZ2_bzReadOpen</computeroutput> and
+<computeroutput>BZ2_bzWriteOpen</computeroutput> require you to
+pass them file handles (<computeroutput>FILE*</computeroutput>s)
+which have previously been opened for reading or writing
+respectively.  That avoids portability problems associated with
+file operations and file attributes, whilst not being much of an
+imposition on the programmer.</para>
+
+</sect2>
+
+
+<sect2 id="util-fns-summary" xreflabel="Utility functions summary">
+<title>Utility functions summary</title>
+
+<para>For very simple needs,
+<computeroutput>BZ2_bzBuffToBuffCompress</computeroutput> and
+<computeroutput>BZ2_bzBuffToBuffDecompress</computeroutput> are
+provided.  These compress data in memory from one buffer to
+another buffer in a single function call.  You should assess
+whether these functions fulfill your memory-to-memory
+compression/decompression requirements before investing effort in
+understanding the more general but more complex low-level
+interface.</para>
+
+<para>Yoshioka Tsuneo
+(<computeroutput>QWF00133@niftyserve.or.jp</computeroutput> /
+<computeroutput>tsuneo-y@is.aist-nara.ac.jp</computeroutput>) has
+contributed some functions to give better
+<computeroutput>zlib</computeroutput> compatibility.  These
+functions are <computeroutput>BZ2_bzopen</computeroutput>,
+<computeroutput>BZ2_bzread</computeroutput>,
+<computeroutput>BZ2_bzwrite</computeroutput>,
+<computeroutput>BZ2_bzflush</computeroutput>,
+<computeroutput>BZ2_bzclose</computeroutput>,
+<computeroutput>BZ2_bzerror</computeroutput> and
+<computeroutput>BZ2_bzlibVersion</computeroutput>.  You may find
+these functions more convenient for simple file reading and
+writing, than those in the high-level interface.  These functions
+are not (yet) officially part of the library, and are minimally
+documented here.  If they break, you get to keep all the pieces.
+I hope to document them properly when time permits.</para>
+
+<para>Yoshioka also contributed modifications to allow the
+library to be built as a Windows DLL.</para>
+
+</sect2>
+
+</sect1>
+
+
+<sect1 id="err-handling" xreflabel="Error handling">
+<title>Error handling</title>
+
+<para>The library is designed to recover cleanly in all
+situations, including the worst-case situation of decompressing
+random data.  I'm not 100% sure that it can always do this, so
+you might want to add a signal handler to catch segmentation
+violations during decompression if you are feeling especially
+paranoid.  I would be interested in hearing more about the
+robustness of the library to corrupted compressed data.</para>
+
+<para>Version 1.0.3 more robust in this respect than any
+previous version.  Investigations with Valgrind (a tool for detecting
+problems with memory management) indicate
+that, at least for the few files I tested, all single-bit errors
+in the decompressed data are caught properly, with no
+segmentation faults, no uses of uninitialised data, no out of
+range reads or writes, and no infinite looping in the decompressor.
+So it's certainly pretty robust, although
+I wouldn't claim it to be totally bombproof.</para>
+
+<para>The file <computeroutput>bzlib.h</computeroutput> contains
+all definitions needed to use the library.  In particular, you
+should definitely not include
+<computeroutput>bzlib_private.h</computeroutput>.</para>
+
+<para>In <computeroutput>bzlib.h</computeroutput>, the various
+return values are defined.  The following list is not intended as
+an exhaustive description of the circumstances in which a given
+value may be returned -- those descriptions are given later.
+Rather, it is intended to convey the rough meaning of each return
+value.  The first five actions are normal and not intended to
+denote an error situation.</para>
+
+<variablelist>
+
+ <varlistentry>
+  <term><computeroutput>BZ_OK</computeroutput></term>
+  <listitem><para>The requested action was completed
+   successfully.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+  <term><computeroutput>BZ_RUN_OK, BZ_FLUSH_OK,
+    BZ_FINISH_OK</computeroutput></term>
+  <listitem><para>In 
+   <computeroutput>BZ2_bzCompress</computeroutput>, the requested
+   flush/finish/nothing-special action was completed
+   successfully.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+  <term><computeroutput>BZ_STREAM_END</computeroutput></term>
+  <listitem><para>Compression of data was completed, or the
+   logical stream end was detected during
+   decompression.</para></listitem>
+ </varlistentry>
+
+</variablelist>
+
+<para>The following return values indicate an error of some
+kind.</para>
+
+<variablelist>
+
+ <varlistentry>
+  <term><computeroutput>BZ_CONFIG_ERROR</computeroutput></term>
+  <listitem><para>Indicates that the library has been improperly
+   compiled on your platform -- a major configuration error.
+   Specifically, it means that
+   <computeroutput>sizeof(char)</computeroutput>,
+   <computeroutput>sizeof(short)</computeroutput> and
+   <computeroutput>sizeof(int)</computeroutput> are not 1, 2 and
+   4 respectively, as they should be.  Note that the library
+   should still work properly on 64-bit platforms which follow
+   the LP64 programming model -- that is, where
+   <computeroutput>sizeof(long)</computeroutput> and
+   <computeroutput>sizeof(void*)</computeroutput> are 8.  Under
+   LP64, <computeroutput>sizeof(int)</computeroutput> is still 4,
+   so <computeroutput>libbzip2</computeroutput>, which doesn't
+   use the <computeroutput>long</computeroutput> type, is
+   OK.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+  <term><computeroutput>BZ_SEQUENCE_ERROR</computeroutput></term>
+  <listitem><para>When using the library, it is important to call
+   the functions in the correct sequence and with data structures
+   (buffers etc) in the correct states.
+   <computeroutput>libbzip2</computeroutput> checks as much as it
+   can to ensure this is happening, and returns
+   <computeroutput>BZ_SEQUENCE_ERROR</computeroutput> if not.
+   Code which complies precisely with the function semantics, as
+   detailed below, should never receive this value; such an event
+   denotes buggy code which you should
+   investigate.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+  <term><computeroutput>BZ_PARAM_ERROR</computeroutput></term>
+  <listitem><para>Returned when a parameter to a function call is
+   out of range or otherwise manifestly incorrect.  As with
+   <computeroutput>BZ_SEQUENCE_ERROR</computeroutput>, this
+   denotes a bug in the client code.  The distinction between
+   <computeroutput>BZ_PARAM_ERROR</computeroutput> and
+   <computeroutput>BZ_SEQUENCE_ERROR</computeroutput> is a bit
+   hazy, but still worth making.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+  <term><computeroutput>BZ_MEM_ERROR</computeroutput></term>
+  <listitem><para>Returned when a request to allocate memory
+   failed.  Note that the quantity of memory needed to decompress
+   a stream cannot be determined until the stream's header has
+   been read.  So
+   <computeroutput>BZ2_bzDecompress</computeroutput> and
+   <computeroutput>BZ2_bzRead</computeroutput> may return
+   <computeroutput>BZ_MEM_ERROR</computeroutput> even though some
+   of the compressed data has been read.  The same is not true
+   for compression; once
+   <computeroutput>BZ2_bzCompressInit</computeroutput> or
+   <computeroutput>BZ2_bzWriteOpen</computeroutput> have
+   successfully completed,
+   <computeroutput>BZ_MEM_ERROR</computeroutput> cannot
+   occur.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+  <term><computeroutput>BZ_DATA_ERROR</computeroutput></term>
+  <listitem><para>Returned when a data integrity error is
+   detected during decompression.  Most importantly, this means
+   when stored and computed CRCs for the data do not match.  This
+   value is also returned upon detection of any other anomaly in
+   the compressed data.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+  <term><computeroutput>BZ_DATA_ERROR_MAGIC</computeroutput></term>
+  <listitem><para>As a special case of
+   <computeroutput>BZ_DATA_ERROR</computeroutput>, it is
+   sometimes useful to know when the compressed stream does not
+   start with the correct magic bytes (<computeroutput>'B' 'Z'
+   'h'</computeroutput>).</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+  <term><computeroutput>BZ_IO_ERROR</computeroutput></term>
+  <listitem><para>Returned by
+   <computeroutput>BZ2_bzRead</computeroutput> and
+   <computeroutput>BZ2_bzWrite</computeroutput> when there is an
+   error reading or writing in the compressed file, and by
+   <computeroutput>BZ2_bzReadOpen</computeroutput> and
+   <computeroutput>BZ2_bzWriteOpen</computeroutput> for attempts
+   to use a file for which the error indicator (viz,
+   <computeroutput>ferror(f)</computeroutput>) is set.  On
+   receipt of <computeroutput>BZ_IO_ERROR</computeroutput>, the
+   caller should consult <computeroutput>errno</computeroutput>
+   and/or <computeroutput>perror</computeroutput> to acquire
+   operating-system specific information about the
+   problem.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+  <term><computeroutput>BZ_UNEXPECTED_EOF</computeroutput></term>
+  <listitem><para>Returned by
+   <computeroutput>BZ2_bzRead</computeroutput> when the
+   compressed file finishes before the logical end of stream is
+   detected.</para></listitem>
+ </varlistentry>
+
+ <varlistentry>
+  <term><computeroutput>BZ_OUTBUFF_FULL</computeroutput></term>
+  <listitem><para>Returned by
+   <computeroutput>BZ2_bzBuffToBuffCompress</computeroutput> and
+   <computeroutput>BZ2_bzBuffToBuffDecompress</computeroutput> to
+   indicate that the output data will not fit into the output
+   buffer provided.</para></listitem>
+ </varlistentry>
+
+</variablelist>
+
+</sect1>
+
+
+
+<sect1 id="low-level" xreflabel=">Low-level interface">
+<title>Low-level interface</title>
+
+
+<sect2 id="bzcompress-init" xreflabel="BZ2_bzCompressInit">
+<title><computeroutput>BZ2_bzCompressInit</computeroutput></title>
+
+<programlisting>
+typedef struct {
+  char *next_in;
+  unsigned int avail_in;
+  unsigned int total_in_lo32;
+  unsigned int total_in_hi32;
+
+  char *next_out;
+  unsigned int avail_out;
+  unsigned int total_out_lo32;
+  unsigned int total_out_hi32;
+
+  void *state;
+
+  void *(*bzalloc)(void *,int,int);
+  void (*bzfree)(void *,void *);
+  void *opaque;
+} bz_stream;
+
+int BZ2_bzCompressInit ( bz_stream *strm, 
+                         int blockSize100k, 
+                         int verbosity,
+                         int workFactor );
+</programlisting>
+
+<para>Prepares for compression.  The
+<computeroutput>bz_stream</computeroutput> structure holds all
+data pertaining to the compression activity.  A
+<computeroutput>bz_stream</computeroutput> structure should be
+allocated and initialised prior to the call.  The fields of
+<computeroutput>bz_stream</computeroutput> comprise the entirety
+of the user-visible data.  <computeroutput>state</computeroutput>
+is a pointer to the private data structures required for
+compression.</para>
+
+<para>Custom memory allocators are supported, via fields
+<computeroutput>bzalloc</computeroutput>,
+<computeroutput>bzfree</computeroutput>, and
+<computeroutput>opaque</computeroutput>.  The value
+<computeroutput>opaque</computeroutput> is passed to as the first
+argument to all calls to <computeroutput>bzalloc</computeroutput>
+and <computeroutput>bzfree</computeroutput>, but is otherwise
+ignored by the library.  The call <computeroutput>bzalloc (
+opaque, n, m )</computeroutput> is expected to return a pointer
+<computeroutput>p</computeroutput> to <computeroutput>n *
+m</computeroutput> bytes of memory, and <computeroutput>bzfree (
+opaque, p )</computeroutput> should free that memory.</para>
+
+<para>If you don't want to use a custom memory allocator, set
+<computeroutput>bzalloc</computeroutput>,
+<computeroutput>bzfree</computeroutput> and
+<computeroutput>opaque</computeroutput> to
+<computeroutput>NULL</computeroutput>, and the library will then
+use the standard <computeroutput>malloc</computeroutput> /
+<computeroutput>free</computeroutput> routines.</para>
+
+<para>Before calling
+<computeroutput>BZ2_bzCompressInit</computeroutput>, fields
+<computeroutput>bzalloc</computeroutput>,
+<computeroutput>bzfree</computeroutput> and
+<computeroutput>opaque</computeroutput> should be filled
+appropriately, as just described.  Upon return, the internal
+state will have been allocated and initialised, and
+<computeroutput>total_in_lo32</computeroutput>,
+<computeroutput>total_in_hi32</computeroutput>,
+<computeroutput>total_out_lo32</computeroutput> and
+<computeroutput>total_out_hi32</computeroutput> will have been
+set to zero.  These four fields are used by the library to inform
+the caller of the total amount of data passed into and out of the
+library, respectively.  You should not try to change them.  As of
+version 1.0, 64-bit counts are maintained, even on 32-bit
+platforms, using the <computeroutput>_hi32</computeroutput>
+fields to store the upper 32 bits of the count.  So, for example,
+the total amount of data in is <computeroutput>(total_in_hi32
+&#60;&#60; 32) + total_in_lo32</computeroutput>.</para>
+
+<para>Parameter <computeroutput>blockSize100k</computeroutput>
+specifies the block size to be used for compression.  It should
+be a value between 1 and 9 inclusive, and the actual block size
+used is 100000 x this figure.  9 gives the best compression but
+takes most memory.</para>
+
+<para>Parameter <computeroutput>verbosity</computeroutput> should
+be set to a number between 0 and 4 inclusive.  0 is silent, and
+greater numbers give increasingly verbose monitoring/debugging
+output.  If the library has been compiled with
+<computeroutput>-DBZ_NO_STDIO</computeroutput>, no such output
+will appear for any verbosity setting.</para>
+
+<para>Parameter <computeroutput>workFactor</computeroutput>
+controls how the compression phase behaves when presented with
+worst case, highly repetitive, input data.  If compression runs
+into difficulties caused by repetitive data, the library switches
+from the standard sorting algorithm to a fallback algorithm.  The
+fallback is slower than the standard algorithm by perhaps a
+factor of three, but always behaves reasonably, no matter how bad
+the input.</para>
+
+<para>Lower values of <computeroutput>workFactor</computeroutput>
+reduce the amount of effort the standard algorithm will expend
+before resorting to the fallback.  You should set this parameter
+carefully; too low, and many inputs will be handled by the
+fallback algorithm and so compress rather slowly, too high, and
+your average-to-worst case compression times can become very
+large.  The default value of 30 gives reasonable behaviour over a
+wide range of circumstances.</para>
+
+<para>Allowable values range from 0 to 250 inclusive.  0 is a
+special case, equivalent to using the default value of 30.</para>
+
+<para>Note that the compressed output generated is the same
+regardless of whether or not the fallback algorithm is
+used.</para>
+
+<para>Be aware also that this parameter may disappear entirely in
+future versions of the library.  In principle it should be
+possible to devise a good way to automatically choose which
+algorithm to use.  Such a mechanism would render the parameter
+obsolete.</para>
+
+<para>Possible return values:</para>
+
+<programlisting>
+BZ_CONFIG_ERROR
+  if the library has been mis-compiled
+BZ_PARAM_ERROR
+  if strm is NULL 
+  or blockSize < 1 or blockSize > 9
+  or verbosity < 0 or verbosity > 4
+  or workFactor < 0 or workFactor > 250
+BZ_MEM_ERROR 
+  if not enough memory is available
+BZ_OK 
+  otherwise
+</programlisting>
+
+<para>Allowable next actions:</para>
+
+<programlisting>
+BZ2_bzCompress
+  if BZ_OK is returned
+  no specific action needed in case of error
+</programlisting>
+
+</sect2>
+
+
+<sect2 id="bzCompress" xreflabel="BZ2_bzCompress">
+<title><computeroutput>BZ2_bzCompress</computeroutput></title>
+
+<programlisting>
+int BZ2_bzCompress ( bz_stream *strm, int action );
+</programlisting>
+
+<para>Provides more input and/or output buffer space for the
+library.  The caller maintains input and output buffers, and
+calls <computeroutput>BZ2_bzCompress</computeroutput> to transfer
+data between them.</para>
+
+<para>Before each call to
+<computeroutput>BZ2_bzCompress</computeroutput>,
+<computeroutput>next_in</computeroutput> should point at the data
+to be compressed, and <computeroutput>avail_in</computeroutput>
+should indicate how many bytes the library may read.
+<computeroutput>BZ2_bzCompress</computeroutput> updates
+<computeroutput>next_in</computeroutput>,
+<computeroutput>avail_in</computeroutput> and
+<computeroutput>total_in</computeroutput> to reflect the number
+of bytes it has read.</para>
+
+<para>Similarly, <computeroutput>next_out</computeroutput> should
+point to a buffer in which the compressed data is to be placed,
+with <computeroutput>avail_out</computeroutput> indicating how
+much output space is available.
+<computeroutput>BZ2_bzCompress</computeroutput> updates
+<computeroutput>next_out</computeroutput>,
+<computeroutput>avail_out</computeroutput> and
+<computeroutput>total_out</computeroutput> to reflect the number
+of bytes output.</para>
+
+<para>You may provide and remove as little or as much data as you
+like on each call of
+<computeroutput>BZ2_bzCompress</computeroutput>.  In the limit,
+it is acceptable to supply and remove data one byte at a time,
+although this would be terribly inefficient.  You should always
+ensure that at least one byte of output space is available at
+each call.</para>
+
+<para>A second purpose of
+<computeroutput>BZ2_bzCompress</computeroutput> is to request a
+change of mode of the compressed stream.</para>
+
+<para>Conceptually, a compressed stream can be in one of four
+states: IDLE, RUNNING, FLUSHING and FINISHING.  Before
+initialisation
+(<computeroutput>BZ2_bzCompressInit</computeroutput>) and after
+termination (<computeroutput>BZ2_bzCompressEnd</computeroutput>),
+a stream is regarded as IDLE.</para>
+
+<para>Upon initialisation
+(<computeroutput>BZ2_bzCompressInit</computeroutput>), the stream
+is placed in the RUNNING state.  Subsequent calls to
+<computeroutput>BZ2_bzCompress</computeroutput> should pass
+<computeroutput>BZ_RUN</computeroutput> as the requested action;
+other actions are illegal and will result in
+<computeroutput>BZ_SEQUENCE_ERROR</computeroutput>.</para>
+
+<para>At some point, the calling program will have provided all
+the input data it wants to.  It will then want to finish up -- in
+effect, asking the library to process any data it might have
+buffered internally.  In this state,
+<computeroutput>BZ2_bzCompress</computeroutput> will no longer
+attempt to read data from
+<computeroutput>next_in</computeroutput>, but it will want to
+write data to <computeroutput>next_out</computeroutput>.  Because
+the output buffer supplied by the user can be arbitrarily small,
+the finishing-up operation cannot necessarily be done with a
+single call of
+<computeroutput>BZ2_bzCompress</computeroutput>.</para>
+
+<para>Instead, the calling program passes
+<computeroutput>BZ_FINISH</computeroutput> as an action to
+<computeroutput>BZ2_bzCompress</computeroutput>.  This changes
+the stream's state to FINISHING.  Any remaining input (ie,
+<computeroutput>next_in[0 .. avail_in-1]</computeroutput>) is
+compressed and transferred to the output buffer.  To do this,
+<computeroutput>BZ2_bzCompress</computeroutput> must be called
+repeatedly until all the output has been consumed.  At that
+point, <computeroutput>BZ2_bzCompress</computeroutput> returns
+<computeroutput>BZ_STREAM_END</computeroutput>, and the stream's
+state is set back to IDLE.
+<computeroutput>BZ2_bzCompressEnd</computeroutput> should then be
+called.</para>
+
+<para>Just to make sure the calling program does not cheat, the
+library makes a note of <computeroutput>avail_in</computeroutput>
+at the time of the first call to
+<computeroutput>BZ2_bzCompress</computeroutput> which has
+<computeroutput>BZ_FINISH</computeroutput> as an action (ie, at
+the time the program has announced its intention to not supply
+any more input).  By comparing this value with that of
+<computeroutput>avail_in</computeroutput> over subsequent calls
+to <computeroutput>BZ2_bzCompress</computeroutput>, the library
+can detect any attempts to slip in more data to compress.  Any
+calls for which this is detected will return
+<computeroutput>BZ_SEQUENCE_ERROR</computeroutput>.  This
+indicates a programming mistake which should be corrected.</para>
+
+<para>Instead of asking to finish, the calling program may ask
+<computeroutput>BZ2_bzCompress</computeroutput> to take all the
+remaining input, compress it and terminate the current
+(Burrows-Wheeler) compression block.  This could be useful for
+error control purposes.  The mechanism is analogous to that for
+finishing: call <computeroutput>BZ2_bzCompress</computeroutput>
+with an action of <computeroutput>BZ_FLUSH</computeroutput>,
+remove output data, and persist with the
+<computeroutput>BZ_FLUSH</computeroutput> action until the value
+<computeroutput>BZ_RUN</computeroutput> is returned.  As with
+finishing, <computeroutput>BZ2_bzCompress</computeroutput>
+detects any attempt to provide more input data once the flush has
+begun.</para>
+
+<para>Once the flush is complete, the stream returns to the
+normal RUNNING state.</para>
+
+<para>This all sounds pretty complex, but isn't really.  Here's a
+table which shows which actions are allowable in each state, what
+action will be taken, what the next state is, and what the
+non-error return values are.  Note that you can't explicitly ask
+what state the stream is in, but nor do you need to -- it can be
+inferred from the values returned by
+<computeroutput>BZ2_bzCompress</computeroutput>.</para>
+
+<programlisting>
+IDLE/any
+  Illegal.  IDLE state only exists after BZ2_bzCompressEnd or
+  before BZ2_bzCompressInit.
+  Return value = BZ_SEQUENCE_ERROR
+
+RUNNING/BZ_RUN
+  Compress from next_in to next_out as much as possible.
+  Next state = RUNNING
+  Return value = BZ_RUN_OK
+
+RUNNING/BZ_FLUSH
+  Remember current value of next_in. Compress from next_in
+  to next_out as much as possible, but do not accept any more input.
+  Next state = FLUSHING
+  Return value = BZ_FLUSH_OK
+
+RUNNING/BZ_FINISH
+  Remember current value of next_in. Compress from next_in
+  to next_out as much as possible, but do not accept any more input.
+  Next state = FINISHING
+  Return value = BZ_FINISH_OK
+
+FLUSHING/BZ_FLUSH
+  Compress from next_in to next_out as much as possible, 
+  but do not accept any more input.
+  If all the existing input has been used up and all compressed
+  output has been removed
+    Next state = RUNNING; Return value = BZ_RUN_OK
+  else
+    Next state = FLUSHING; Return value = BZ_FLUSH_OK
+
+FLUSHING/other     
+  Illegal.
+  Return value = BZ_SEQUENCE_ERROR
+
+FINISHING/BZ_FINISH
+  Compress from next_in to next_out as much as possible,
+  but to not accept any more input.  
+  If all the existing input has been used up and all compressed
+  output has been removed
+    Next state = IDLE; Return value = BZ_STREAM_END
+  else
+    Next state = FINISHING; Return value = BZ_FINISHING
+
+FINISHING/other
+  Illegal.
+  Return value = BZ_SEQUENCE_ERROR
+</programlisting>
+
+
+<para>That still looks complicated?  Well, fair enough.  The
+usual sequence of calls for compressing a load of data is:</para>
+
+<orderedlist>
+
+ <listitem><para>Get started with
+  <computeroutput>BZ2_bzCompressInit</computeroutput>.</para></listitem>
+
+ <listitem><para>Shovel data in and shlurp out its compressed form
+  using zero or more calls of
+  <computeroutput>BZ2_bzCompress</computeroutput> with action =
+  <computeroutput>BZ_RUN</computeroutput>.</para></listitem>
+
+ <listitem><para>Finish up. Repeatedly call
+  <computeroutput>BZ2_bzCompress</computeroutput> with action =
+  <computeroutput>BZ_FINISH</computeroutput>, copying out the
+  compressed output, until
+  <computeroutput>BZ_STREAM_END</computeroutput> is
+  returned.</para></listitem> <listitem><para>Close up and go home.  Call
+  <computeroutput>BZ2_bzCompressEnd</computeroutput>.</para></listitem>
+
+</orderedlist>
+
+<para>If the data you want to compress fits into your input
+buffer all at once, you can skip the calls of
+<computeroutput>BZ2_bzCompress ( ..., BZ_RUN )</computeroutput>
+and just do the <computeroutput>BZ2_bzCompress ( ..., BZ_FINISH
+)</computeroutput> calls.</para>
+
+<para>All required memory is allocated by
+<computeroutput>BZ2_bzCompressInit</computeroutput>.  The
+compression library can accept any data at all (obviously).  So
+you shouldn't get any error return values from the
+<computeroutput>BZ2_bzCompress</computeroutput> calls.  If you
+do, they will be
+<computeroutput>BZ_SEQUENCE_ERROR</computeroutput>, and indicate
+a bug in your programming.</para>
+
+<para>Trivial other possible return values:</para>
+
+<programlisting>
+BZ_PARAM_ERROR
+  if strm is NULL, or strm->s is NULL
+</programlisting>
+
+</sect2>
+
+
+<sect2 id="bzCompress-end" xreflabel="BZ2_bzCompressEnd">
+<title><computeroutput>BZ2_bzCompressEnd</computeroutput></title>
+
+<programlisting>
+int BZ2_bzCompressEnd ( bz_stream *strm );
+</programlisting>
+
+<para>Releases all memory associated with a compression
+stream.</para>
+
+<para>Possible return values:</para>
+
+<programlisting>
+BZ_PARAM_ERROR  if strm is NULL or strm->s is NULL
+BZ_OK           otherwise
+</programlisting>
+
+</sect2>
+
+
+<sect2 id="bzDecompress-init" xreflabel="BZ2_bzDecompressInit">
+<title><computeroutput>BZ2_bzDecompressInit</computeroutput></title>
+
+<programlisting>
+int BZ2_bzDecompressInit ( bz_stream *strm, int verbosity, int small );
+</programlisting>
+
+<para>Prepares for decompression.  As with
+<computeroutput>BZ2_bzCompressInit</computeroutput>, a
+<computeroutput>bz_stream</computeroutput> record should be
+allocated and initialised before the call.  Fields
+<computeroutput>bzalloc</computeroutput>,
+<computeroutput>bzfree</computeroutput> and
+<computeroutput>opaque</computeroutput> should be set if a custom
+memory allocator is required, or made
+<computeroutput>NULL</computeroutput> for the normal
+<computeroutput>malloc</computeroutput> /
+<computeroutput>free</computeroutput> routines.  Upon return, the
+internal state will have been initialised, and
+<computeroutput>total_in</computeroutput> and
+<computeroutput>total_out</computeroutput> will be zero.</para>
+
+<para>For the meaning of parameter
+<computeroutput>verbosity</computeroutput>, see
+<computeroutput>BZ2_bzCompressInit</computeroutput>.</para>
+
+<para>If <computeroutput>small</computeroutput> is nonzero, the
+library will use an alternative decompression algorithm which
+uses less memory but at the cost of decompressing more slowly
+(roughly speaking, half the speed, but the maximum memory
+requirement drops to around 2300k).  See <xref linkend="using"/>
+for more information on memory management.</para>
+
+<para>Note that the amount of memory needed to decompress a
+stream cannot be determined until the stream's header has been
+read, so even if
+<computeroutput>BZ2_bzDecompressInit</computeroutput> succeeds, a
+subsequent <computeroutput>BZ2_bzDecompress</computeroutput>
+could fail with
+<computeroutput>BZ_MEM_ERROR</computeroutput>.</para>
+
+<para>Possible return values:</para>
+
+<programlisting>
+BZ_CONFIG_ERROR
+  if the library has been mis-compiled
+BZ_PARAM_ERROR
+  if ( small != 0 && small != 1 )
+  or (verbosity <; 0 || verbosity > 4)
+BZ_MEM_ERROR
+  if insufficient memory is available
+</programlisting>
+
+<para>Allowable next actions:</para>
+
+<programlisting>
+BZ2_bzDecompress
+  if BZ_OK was returned
+  no specific action required in case of error
+</programlisting>
+
+</sect2>
+
+
+<sect2 id="bzDecompress" xreflabel="BZ2_bzDecompress">
+<title><computeroutput>BZ2_bzDecompress</computeroutput></title>
+
+<programlisting>
+int BZ2_bzDecompress ( bz_stream *strm );
+</programlisting>
+
+<para>Provides more input and/out output buffer space for the
+library.  The caller maintains input and output buffers, and uses
+<computeroutput>BZ2_bzDecompress</computeroutput> to transfer
+data between them.</para>
+
+<para>Before each call to
+<computeroutput>BZ2_bzDecompress</computeroutput>,
+<computeroutput>next_in</computeroutput> should point at the
+compressed data, and <computeroutput>avail_in</computeroutput>
+should indicate how many bytes the library may read.
+<computeroutput>BZ2_bzDecompress</computeroutput> updates
+<computeroutput>next_in</computeroutput>,
+<computeroutput>avail_in</computeroutput> and
+<computeroutput>total_in</computeroutput> to reflect the number
+of bytes it has read.</para>
+
+<para>Similarly, <computeroutput>next_out</computeroutput> should
+point to a buffer in which the uncompressed output is to be
+placed, with <computeroutput>avail_out</computeroutput>
+indicating how much output space is available.
+<computeroutput>BZ2_bzCompress</computeroutput> updates
+<computeroutput>next_out</computeroutput>,
+<computeroutput>avail_out</computeroutput> and
+<computeroutput>total_out</computeroutput> to reflect the number
+of bytes output.</para>
+
+<para>You may provide and remove as little or as much data as you
+like on each call of
+<computeroutput>BZ2_bzDecompress</computeroutput>.  In the limit,
+it is acceptable to supply and remove data one byte at a time,
+although this would be terribly inefficient.  You should always
+ensure that at least one byte of output space is available at
+each call.</para>
+
+<para>Use of <computeroutput>BZ2_bzDecompress</computeroutput> is
+simpler than
+<computeroutput>BZ2_bzCompress</computeroutput>.</para>
+
+<para>You should provide input and remove output as described
+above, and repeatedly call
+<computeroutput>BZ2_bzDecompress</computeroutput> until
+<computeroutput>BZ_STREAM_END</computeroutput> is returned.
+Appearance of <computeroutput>BZ_STREAM_END</computeroutput>
+denotes that <computeroutput>BZ2_bzDecompress</computeroutput>
+has detected the logical end of the compressed stream.
+<computeroutput>BZ2_bzDecompress</computeroutput> will not
+produce <computeroutput>BZ_STREAM_END</computeroutput> until all
+output data has been placed into the output buffer, so once
+<computeroutput>BZ_STREAM_END</computeroutput> appears, you are
+guaranteed to have available all the decompressed output, and
+<computeroutput>BZ2_bzDecompressEnd</computeroutput> can safely
+be called.</para>
+
+<para>If case of an error return value, you should call
+<computeroutput>BZ2_bzDecompressEnd</computeroutput> to clean up
+and release memory.</para>
+
+<para>Possible return values:</para>
+
+<programlisting>
+BZ_PARAM_ERROR
+  if strm is NULL or strm->s is NULL
+  or strm->avail_out < 1
+BZ_DATA_ERROR
+  if a data integrity error is detected in the compressed stream
+BZ_DATA_ERROR_MAGIC
+  if the compressed stream doesn't begin with the right magic bytes
+BZ_MEM_ERROR
+  if there wasn't enough memory available
+BZ_STREAM_END
+  if the logical end of the data stream was detected and all
+  output in has been consumed, eg s-->avail_out > 0
+BZ_OK
+  otherwise
+</programlisting>
+
+<para>Allowable next actions:</para>
+
+<programlisting>
+BZ2_bzDecompress
+  if BZ_OK was returned
+BZ2_bzDecompressEnd
+  otherwise
+</programlisting>
+
+</sect2>
+
+
+<sect2 id="bzDecompress-end" xreflabel="BZ2_bzDecompressEnd">
+<title><computeroutput>BZ2_bzDecompressEnd</computeroutput></title>
+
+<programlisting>
+int BZ2_bzDecompressEnd ( bz_stream *strm );
+</programlisting>
+
+<para>Releases all memory associated with a decompression
+stream.</para>
+
+<para>Possible return values:</para>
+
+<programlisting>
+BZ_PARAM_ERROR
+  if strm is NULL or strm->s is NULL
+BZ_OK
+  otherwise
+</programlisting>
+
+<para>Allowable next actions:</para>
+
+<programlisting>
+  None.
+</programlisting>
+
+</sect2>
+
+</sect1>
+
+
+<sect1 id="hl-interface" xreflabel="High-level interface">
+<title>High-level interface</title>
+
+<para>This interface provides functions for reading and writing
+<computeroutput>bzip2</computeroutput> format files.  First, some
+general points.</para>
+
+<itemizedlist mark='bullet'>
+
+ <listitem><para>All of the functions take an
+  <computeroutput>int*</computeroutput> first argument,
+  <computeroutput>bzerror</computeroutput>.  After each call,
+  <computeroutput>bzerror</computeroutput> should be consulted
+  first to determine the outcome of the call.  If
+  <computeroutput>bzerror</computeroutput> is
+  <computeroutput>BZ_OK</computeroutput>, the call completed
+  successfully, and only then should the return value of the
+  function (if any) be consulted.  If
+  <computeroutput>bzerror</computeroutput> is
+  <computeroutput>BZ_IO_ERROR</computeroutput>, there was an
+  error reading/writing the underlying compressed file, and you
+  should then consult <computeroutput>errno</computeroutput> /
+  <computeroutput>perror</computeroutput> to determine the cause
+  of the difficulty.  <computeroutput>bzerror</computeroutput>
+  may also be set to various other values; precise details are
+  given on a per-function basis below.</para></listitem>
+
+ <listitem><para>If <computeroutput>bzerror</computeroutput> indicates
+  an error (ie, anything except
+  <computeroutput>BZ_OK</computeroutput> and
+  <computeroutput>BZ_STREAM_END</computeroutput>), you should
+  immediately call
+  <computeroutput>BZ2_bzReadClose</computeroutput> (or
+  <computeroutput>BZ2_bzWriteClose</computeroutput>, depending on
+  whether you are attempting to read or to write) to free up all
+  resources associated with the stream.  Once an error has been
+  indicated, behaviour of all calls except
+  <computeroutput>BZ2_bzReadClose</computeroutput>
+  (<computeroutput>BZ2_bzWriteClose</computeroutput>) is
+  undefined.  The implication is that (1)
+  <computeroutput>bzerror</computeroutput> should be checked
+  after each call, and (2) if
+  <computeroutput>bzerror</computeroutput> indicates an error,
+  <computeroutput>BZ2_bzReadClose</computeroutput>
+  (<computeroutput>BZ2_bzWriteClose</computeroutput>) should then
+  be called to clean up.</para></listitem>
+
+ <listitem><para>The <computeroutput>FILE*</computeroutput> arguments
+  passed to <computeroutput>BZ2_bzReadOpen</computeroutput> /
+  <computeroutput>BZ2_bzWriteOpen</computeroutput> should be set
+  to binary mode.  Most Unix systems will do this by default, but
+  other platforms, including Windows and Mac, will not.  If you
+  omit this, you may encounter problems when moving code to new
+  platforms.</para></listitem>
+
+ <listitem><para>Memory allocation requests are handled by
+  <computeroutput>malloc</computeroutput> /
+  <computeroutput>free</computeroutput>.  At present there is no
+  facility for user-defined memory allocators in the file I/O
+  functions (could easily be added, though).</para></listitem>
+
+</itemizedlist>
+
+
+
+<sect2 id="bzreadopen" xreflabel="BZ2_bzReadOpen">
+<title><computeroutput>BZ2_bzReadOpen</computeroutput></title>
+
+<programlisting>
+typedef void BZFILE;
+
+BZFILE *BZ2_bzReadOpen( int *bzerror, FILE *f, 
+                        int verbosity, int small,
+                        void *unused, int nUnused );
+</programlisting>
+
+<para>Prepare to read compressed data from file handle
+<computeroutput>f</computeroutput>.
+<computeroutput>f</computeroutput> should refer to a file which
+has been opened for reading, and for which the error indicator
+(<computeroutput>ferror(f)</computeroutput>)is not set.  If
+<computeroutput>small</computeroutput> is 1, the library will try
+to decompress using less memory, at the expense of speed.</para>
+
+<para>For reasons explained below,
+<computeroutput>BZ2_bzRead</computeroutput> will decompress the
+<computeroutput>nUnused</computeroutput> bytes starting at
+<computeroutput>unused</computeroutput>, before starting to read
+from the file <computeroutput>f</computeroutput>.  At most
+<computeroutput>BZ_MAX_UNUSED</computeroutput> bytes may be
+supplied like this.  If this facility is not required, you should
+pass <computeroutput>NULL</computeroutput> and
+<computeroutput>0</computeroutput> for
+<computeroutput>unused</computeroutput> and
+n<computeroutput>Unused</computeroutput> respectively.</para>
+
+<para>For the meaning of parameters
+<computeroutput>small</computeroutput> and
+<computeroutput>verbosity</computeroutput>, see
+<computeroutput>BZ2_bzDecompressInit</computeroutput>.</para>
+
+<para>The amount of memory needed to decompress a file cannot be
+determined until the file's header has been read.  So it is
+possible that <computeroutput>BZ2_bzReadOpen</computeroutput>
+returns <computeroutput>BZ_OK</computeroutput> but a subsequent
+call of <computeroutput>BZ2_bzRead</computeroutput> will return
+<computeroutput>BZ_MEM_ERROR</computeroutput>.</para>
+
+<para>Possible assignments to
+<computeroutput>bzerror</computeroutput>:</para>
+
+<programlisting>
+BZ_CONFIG_ERROR
+  if the library has been mis-compiled
+BZ_PARAM_ERROR
+  if f is NULL
+  or small is neither 0 nor 1
+  or ( unused == NULL && nUnused != 0 )
+  or ( unused != NULL && !(0 <= nUnused <= BZ_MAX_UNUSED) )
+BZ_IO_ERROR
+  if ferror(f) is nonzero
+BZ_MEM_ERROR
+  if insufficient memory is available
+BZ_OK
+  otherwise.
+</programlisting>
+
+<para>Possible return values:</para>
+
+<programlisting>
+Pointer to an abstract BZFILE
+  if bzerror is BZ_OK
+NULL
+  otherwise
+</programlisting>
+
+<para>Allowable next actions:</para>
+
+<programlisting>
+BZ2_bzRead
+  if bzerror is BZ_OK
+BZ2_bzClose
+  otherwise
+</programlisting>
+
+</sect2>
+
+
+<sect2 id="bzread" xreflabel="BZ2_bzRead">
+<title><computeroutput>BZ2_bzRead</computeroutput></title>
+
+<programlisting>
+int BZ2_bzRead ( int *bzerror, BZFILE *b, void *buf, int len );
+</programlisting>
+
+<para>Reads up to <computeroutput>len</computeroutput>
+(uncompressed) bytes from the compressed file
+<computeroutput>b</computeroutput> into the buffer
+<computeroutput>buf</computeroutput>.  If the read was
+successful, <computeroutput>bzerror</computeroutput> is set to
+<computeroutput>BZ_OK</computeroutput> and the number of bytes
+read is returned.  If the logical end-of-stream was detected,
+<computeroutput>bzerror</computeroutput> will be set to
+<computeroutput>BZ_STREAM_END</computeroutput>, and the number of
+bytes read is returned.  All other
+<computeroutput>bzerror</computeroutput> values denote an
+error.</para>
+
+<para><computeroutput>BZ2_bzRead</computeroutput> will supply
+<computeroutput>len</computeroutput> bytes, unless the logical
+stream end is detected or an error occurs.  Because of this, it
+is possible to detect the stream end by observing when the number
+of bytes returned is less than the number requested.
+Nevertheless, this is regarded as inadvisable; you should instead
+check <computeroutput>bzerror</computeroutput> after every call
+and watch out for
+<computeroutput>BZ_STREAM_END</computeroutput>.</para>
+
+<para>Internally, <computeroutput>BZ2_bzRead</computeroutput>
+copies data from the compressed file in chunks of size
+<computeroutput>BZ_MAX_UNUSED</computeroutput> bytes before
+decompressing it.  If the file contains more bytes than strictly
+needed to reach the logical end-of-stream,
+<computeroutput>BZ2_bzRead</computeroutput> will almost certainly
+read some of the trailing data before signalling
+<computeroutput>BZ_SEQUENCE_END</computeroutput>.  To collect the
+read but unused data once
+<computeroutput>BZ_SEQUENCE_END</computeroutput> has appeared,
+call <computeroutput>BZ2_bzReadGetUnused</computeroutput>
+immediately before
+<computeroutput>BZ2_bzReadClose</computeroutput>.</para>
+
+<para>Possible assignments to
+<computeroutput>bzerror</computeroutput>:</para>
+
+<programlisting>
+BZ_PARAM_ERROR
+  if b is NULL or buf is NULL or len < 0
+BZ_SEQUENCE_ERROR
+  if b was opened with BZ2_bzWriteOpen
+BZ_IO_ERROR
+  if there is an error reading from the compressed file
+BZ_UNEXPECTED_EOF
+  if the compressed file ended before 
+  the logical end-of-stream was detected
+BZ_DATA_ERROR
+  if a data integrity error was detected in the compressed stream
+BZ_DATA_ERROR_MAGIC
+  if the stream does not begin with the requisite header bytes 
+  (ie, is not a bzip2 data file).  This is really 
+  a special case of BZ_DATA_ERROR.
+BZ_MEM_ERROR
+  if insufficient memory was available
+BZ_STREAM_END
+  if the logical end of stream was detected.
+BZ_OK
+  otherwise.
+</programlisting>
+
+<para>Possible return values:</para>
+
+<programlisting>
+number of bytes read
+  if bzerror is BZ_OK or BZ_STREAM_END
+undefined
+  otherwise
+</programlisting>
+
+<para>Allowable next actions:</para>
+
+<programlisting>
+collect data from buf, then BZ2_bzRead or BZ2_bzReadClose
+  if bzerror is BZ_OK
+collect data from buf, then BZ2_bzReadClose or BZ2_bzReadGetUnused
+  if bzerror is BZ_SEQUENCE_END
+BZ2_bzReadClose
+  otherwise
+</programlisting>
+
+</sect2>
+
+
+<sect2 id="bzreadgetunused" xreflabel="BZ2_bzReadGetUnused">
+<title><computeroutput>BZ2_bzReadGetUnused</computeroutput></title>
+
+<programlisting>
+void BZ2_bzReadGetUnused( int* bzerror, BZFILE *b, 
+                          void** unused, int* nUnused );
+</programlisting>
+
+<para>Returns data which was read from the compressed file but
+was not needed to get to the logical end-of-stream.
+<computeroutput>*unused</computeroutput> is set to the address of
+the data, and <computeroutput>*nUnused</computeroutput> to the
+number of bytes.  <computeroutput>*nUnused</computeroutput> will
+be set to a value between <computeroutput>0</computeroutput> and
+<computeroutput>BZ_MAX_UNUSED</computeroutput> inclusive.</para>
+
+<para>This function may only be called once
+<computeroutput>BZ2_bzRead</computeroutput> has signalled
+<computeroutput>BZ_STREAM_END</computeroutput> but before
+<computeroutput>BZ2_bzReadClose</computeroutput>.</para>
+
+<para>Possible assignments to
+<computeroutput>bzerror</computeroutput>:</para>
+
+<programlisting>
+BZ_PARAM_ERROR
+  if b is NULL
+  or unused is NULL or nUnused is NULL
+BZ_SEQUENCE_ERROR
+  if BZ_STREAM_END has not been signalled
+  or if b was opened with BZ2_bzWriteOpen
+BZ_OK
+  otherwise
+</programlisting>
+
+<para>Allowable next actions:</para>
+
+<programlisting>
+BZ2_bzReadClose
+</programlisting>
+
+</sect2>
+
+
+<sect2 id="bzreadclose" xreflabel="BZ2_bzReadClose">
+<title><computeroutput>BZ2_bzReadClose</computeroutput></title>
+
+<programlisting>
+void BZ2_bzReadClose ( int *bzerror, BZFILE *b );
+</programlisting>
+
+<para>Releases all memory pertaining to the compressed file
+<computeroutput>b</computeroutput>.
+<computeroutput>BZ2_bzReadClose</computeroutput> does not call
+<computeroutput>fclose</computeroutput> on the underlying file
+handle, so you should do that yourself if appropriate.
+<computeroutput>BZ2_bzReadClose</computeroutput> should be called
+to clean up after all error situations.</para>
+
+<para>Possible assignments to
+<computeroutput>bzerror</computeroutput>:</para>
+
+<programlisting>
+BZ_SEQUENCE_ERROR
+  if b was opened with BZ2_bzOpenWrite
+BZ_OK
+  otherwise
+</programlisting>
+
+<para>Allowable next actions:</para>
+
+<programlisting>
+none
+</programlisting>
+
+</sect2>
+
+
+<sect2 id="bzwriteopen" xreflabel="BZ2_bzWriteOpen">
+<title><computeroutput>BZ2_bzWriteOpen</computeroutput></title>
+
+<programlisting>
+BZFILE *BZ2_bzWriteOpen( int *bzerror, FILE *f, 
+                         int blockSize100k, int verbosity,
+                         int workFactor );
+</programlisting>
+
+<para>Prepare to write compressed data to file handle
+<computeroutput>f</computeroutput>.
+<computeroutput>f</computeroutput> should refer to a file which
+has been opened for writing, and for which the error indicator
+(<computeroutput>ferror(f)</computeroutput>)is not set.</para>
+
+<para>For the meaning of parameters
+<computeroutput>blockSize100k</computeroutput>,
+<computeroutput>verbosity</computeroutput> and
+<computeroutput>workFactor</computeroutput>, see
+<computeroutput>BZ2_bzCompressInit</computeroutput>.</para>
+
+<para>All required memory is allocated at this stage, so if the
+call completes successfully,
+<computeroutput>BZ_MEM_ERROR</computeroutput> cannot be signalled
+by a subsequent call to
+<computeroutput>BZ2_bzWrite</computeroutput>.</para>
+
+<para>Possible assignments to
+<computeroutput>bzerror</computeroutput>:</para>
+
+<programlisting>
+BZ_CONFIG_ERROR
+  if the library has been mis-compiled
+BZ_PARAM_ERROR
+  if f is NULL
+  or blockSize100k < 1 or blockSize100k > 9
+BZ_IO_ERROR
+  if ferror(f) is nonzero
+BZ_MEM_ERROR
+  if insufficient memory is available
+BZ_OK
+  otherwise
+</programlisting>
+
+<para>Possible return values:</para>
+
+<programlisting>
+Pointer to an abstract BZFILE
+  if bzerror is BZ_OK
+NULL
+  otherwise
+</programlisting>
+
+<para>Allowable next actions:</para>
+
+<programlisting>
+BZ2_bzWrite
+  if bzerror is BZ_OK
+  (you could go directly to BZ2_bzWriteClose, but this would be pretty pointless)
+BZ2_bzWriteClose
+  otherwise
+</programlisting>
+
+</sect2>
+
+
+<sect2 id="bzwrite" xreflabel="BZ2_bzWrite">
+<title><computeroutput>BZ2_bzWrite</computeroutput></title>
+
+<programlisting>
+void BZ2_bzWrite ( int *bzerror, BZFILE *b, void *buf, int len );
+</programlisting>
+
+<para>Absorbs <computeroutput>len</computeroutput> bytes from the
+buffer <computeroutput>buf</computeroutput>, eventually to be
+compressed and written to the file.</para>
+
+<para>Possible assignments to
+<computeroutput>bzerror</computeroutput>:</para>
+
+<programlisting>
+BZ_PARAM_ERROR
+  if b is NULL or buf is NULL or len < 0
+BZ_SEQUENCE_ERROR
+  if b was opened with BZ2_bzReadOpen
+BZ_IO_ERROR
+  if there is an error writing the compressed file.
+BZ_OK
+  otherwise
+</programlisting>
+
+</sect2>
+
+
+<sect2 id="bzwriteclose" xreflabel="BZ2_bzWriteClose">
+<title><computeroutput>BZ2_bzWriteClose</computeroutput></title>
+
+<programlisting>
+void BZ2_bzWriteClose( int *bzerror, BZFILE* f,
+                       int abandon,
+                       unsigned int* nbytes_in,
+                       unsigned int* nbytes_out );
+
+void BZ2_bzWriteClose64( int *bzerror, BZFILE* f,
+                         int abandon,
+                         unsigned int* nbytes_in_lo32,
+                         unsigned int* nbytes_in_hi32,
+                         unsigned int* nbytes_out_lo32,
+                         unsigned int* nbytes_out_hi32 );
+</programlisting>
+
+<para>Compresses and flushes to the compressed file all data so
+far supplied by <computeroutput>BZ2_bzWrite</computeroutput>.
+The logical end-of-stream markers are also written, so subsequent
+calls to <computeroutput>BZ2_bzWrite</computeroutput> are
+illegal.  All memory associated with the compressed file
+<computeroutput>b</computeroutput> is released.
+<computeroutput>fflush</computeroutput> is called on the
+compressed file, but it is not
+<computeroutput>fclose</computeroutput>'d.</para>
+
+<para>If <computeroutput>BZ2_bzWriteClose</computeroutput> is
+called to clean up after an error, the only action is to release
+the memory.  The library records the error codes issued by
+previous calls, so this situation will be detected automatically.
+There is no attempt to complete the compression operation, nor to
+<computeroutput>fflush</computeroutput> the compressed file.  You
+can force this behaviour to happen even in the case of no error,
+by passing a nonzero value to
+<computeroutput>abandon</computeroutput>.</para>
+
+<para>If <computeroutput>nbytes_in</computeroutput> is non-null,
+<computeroutput>*nbytes_in</computeroutput> will be set to be the
+total volume of uncompressed data handled.  Similarly,
+<computeroutput>nbytes_out</computeroutput> will be set to the
+total volume of compressed data written.  For compatibility with
+older versions of the library,
+<computeroutput>BZ2_bzWriteClose</computeroutput> only yields the
+lower 32 bits of these counts.  Use
+<computeroutput>BZ2_bzWriteClose64</computeroutput> if you want
+the full 64 bit counts.  These two functions are otherwise
+absolutely identical.</para>
+
+<para>Possible assignments to
+<computeroutput>bzerror</computeroutput>:</para>
+
+<programlisting>
+BZ_SEQUENCE_ERROR
+  if b was opened with BZ2_bzReadOpen
+BZ_IO_ERROR
+  if there is an error writing the compressed file
+BZ_OK
+  otherwise
+</programlisting>
+
+</sect2>
+
+
+<sect2 id="embed" xreflabel="Handling embedded compressed data streams">
+<title>Handling embedded compressed data streams</title>
+
+<para>The high-level library facilitates use of
+<computeroutput>bzip2</computeroutput> data streams which form
+some part of a surrounding, larger data stream.</para>
+
+<itemizedlist mark='bullet'>
+
+ <listitem><para>For writing, the library takes an open file handle,
+  writes compressed data to it,
+  <computeroutput>fflush</computeroutput>es it but does not
+  <computeroutput>fclose</computeroutput> it.  The calling
+  application can write its own data before and after the
+  compressed data stream, using that same file handle.</para></listitem>
+
+ <listitem><para>Reading is more complex, and the facilities are not as
+  general as they could be since generality is hard to reconcile
+  with efficiency.  <computeroutput>BZ2_bzRead</computeroutput>
+  reads from the compressed file in blocks of size
+  <computeroutput>BZ_MAX_UNUSED</computeroutput> bytes, and in
+  doing so probably will overshoot the logical end of compressed
+  stream.  To recover this data once decompression has ended,
+  call <computeroutput>BZ2_bzReadGetUnused</computeroutput> after
+  the last call of <computeroutput>BZ2_bzRead</computeroutput>
+  (the one returning
+  <computeroutput>BZ_STREAM_END</computeroutput>) but before
+  calling
+  <computeroutput>BZ2_bzReadClose</computeroutput>.</para></listitem>
+
+</itemizedlist>
+
+<para>This mechanism makes it easy to decompress multiple
+<computeroutput>bzip2</computeroutput> streams placed end-to-end.
+As the end of one stream, when
+<computeroutput>BZ2_bzRead</computeroutput> returns
+<computeroutput>BZ_STREAM_END</computeroutput>, call
+<computeroutput>BZ2_bzReadGetUnused</computeroutput> to collect
+the unused data (copy it into your own buffer somewhere).  That
+data forms the start of the next compressed stream.  To start
+uncompressing that next stream, call
+<computeroutput>BZ2_bzReadOpen</computeroutput> again, feeding in
+the unused data via the <computeroutput>unused</computeroutput> /
+<computeroutput>nUnused</computeroutput> parameters.  Keep doing
+this until <computeroutput>BZ_STREAM_END</computeroutput> return
+coincides with the physical end of file
+(<computeroutput>feof(f)</computeroutput>).  In this situation
+<computeroutput>BZ2_bzReadGetUnused</computeroutput> will of
+course return no data.</para>
+
+<para>This should give some feel for how the high-level interface
+can be used.  If you require extra flexibility, you'll have to
+bite the bullet and get to grips with the low-level
+interface.</para>
+
+</sect2>
+
+
+<sect2 id="std-rdwr" xreflabel="Standard file-reading/writing code">
+<title>Standard file-reading/writing code</title>
+
+<para>Here's how you'd write data to a compressed file:</para>
+
+<programlisting>
+FILE*   f;
+BZFILE* b;
+int     nBuf;
+char    buf[ /* whatever size you like */ ];
+int     bzerror;
+int     nWritten;
+
+f = fopen ( "myfile.bz2", "w" );
+if ( !f ) {
+ /* handle error */
+}
+b = BZ2_bzWriteOpen( &bzerror, f, 9 );
+if (bzerror != BZ_OK) {
+ BZ2_bzWriteClose ( b );
+ /* handle error */
+}
+
+while ( /* condition */ ) {
+ /* get data to write into buf, and set nBuf appropriately */
+ nWritten = BZ2_bzWrite ( &bzerror, b, buf, nBuf );
+ if (bzerror == BZ_IO_ERROR) { 
+   BZ2_bzWriteClose ( &bzerror, b );
+   /* handle error */
+ }
+}
+
+BZ2_bzWriteClose( &bzerror, b );
+if (bzerror == BZ_IO_ERROR) {
+ /* handle error */
+}
+</programlisting>
+
+<para>And to read from a compressed file:</para>
+
+<programlisting>
+FILE*   f;
+BZFILE* b;
+int     nBuf;
+char    buf[ /* whatever size you like */ ];
+int     bzerror;
+int     nWritten;
+
+f = fopen ( "myfile.bz2", "r" );
+if ( !f ) {
+  /* handle error */
+}
+b = BZ2_bzReadOpen ( &bzerror, f, 0, NULL, 0 );
+if ( bzerror != BZ_OK ) {
+  BZ2_bzReadClose ( &bzerror, b );
+  /* handle error */
+}
+
+bzerror = BZ_OK;
+while ( bzerror == BZ_OK && /* arbitrary other conditions */) {
+  nBuf = BZ2_bzRead ( &bzerror, b, buf, /* size of buf */ );
+  if ( bzerror == BZ_OK ) {
+    /* do something with buf[0 .. nBuf-1] */
+  }
+}
+if ( bzerror != BZ_STREAM_END ) {
+   BZ2_bzReadClose ( &bzerror, b );
+   /* handle error */
+} else {
+   BZ2_bzReadClose ( &bzerror );
+}
+</programlisting>
+
+</sect2>
+
+</sect1>
+
+
+<sect1 id="util-fns" xreflabel="Utility functions">
+<title>Utility functions</title>
+
+
+<sect2 id="bzbufftobuffcompress" xreflabel="BZ2_bzBuffToBuffCompress">
+<title><computeroutput>BZ2_bzBuffToBuffCompress</computeroutput></title>
+
+<programlisting>
+int BZ2_bzBuffToBuffCompress( char*         dest,
+                              unsigned int* destLen,
+                              char*         source,
+                              unsigned int  sourceLen,
+                              int           blockSize100k,
+                              int           verbosity,
+                              int           workFactor );
+</programlisting>
+
+<para>Attempts to compress the data in <computeroutput>source[0
+.. sourceLen-1]</computeroutput> into the destination buffer,
+<computeroutput>dest[0 .. *destLen-1]</computeroutput>.  If the
+destination buffer is big enough,
+<computeroutput>*destLen</computeroutput> is set to the size of
+the compressed data, and <computeroutput>BZ_OK</computeroutput>
+is returned.  If the compressed data won't fit,
+<computeroutput>*destLen</computeroutput> is unchanged, and
+<computeroutput>BZ_OUTBUFF_FULL</computeroutput> is
+returned.</para>
+
+<para>Compression in this manner is a one-shot event, done with a
+single call to this function.  The resulting compressed data is a
+complete <computeroutput>bzip2</computeroutput> format data
+stream.  There is no mechanism for making additional calls to
+provide extra input data.  If you want that kind of mechanism,
+use the low-level interface.</para>
+
+<para>For the meaning of parameters
+<computeroutput>blockSize100k</computeroutput>,
+<computeroutput>verbosity</computeroutput> and
+<computeroutput>workFactor</computeroutput>, see
+<computeroutput>BZ2_bzCompressInit</computeroutput>.</para>
+
+<para>To guarantee that the compressed data will fit in its
+buffer, allocate an output buffer of size 1% larger than the
+uncompressed data, plus six hundred extra bytes.</para>
+
+<para><computeroutput>BZ2_bzBuffToBuffDecompress</computeroutput>
+will not write data at or beyond
+<computeroutput>dest[*destLen]</computeroutput>, even in case of
+buffer overflow.</para>
+
+<para>Possible return values:</para>
+
+<programlisting>
+BZ_CONFIG_ERROR
+  if the library has been mis-compiled
+BZ_PARAM_ERROR
+  if dest is NULL or destLen is NULL
+  or blockSize100k < 1 or blockSize100k > 9
+  or verbosity < 0 or verbosity > 4
+  or workFactor < 0 or workFactor > 250
+BZ_MEM_ERROR
+  if insufficient memory is available 
+BZ_OUTBUFF_FULL
+  if the size of the compressed data exceeds *destLen
+BZ_OK
+  otherwise
+</programlisting>
+
+</sect2>
+
+
+<sect2 id="bzbufftobuffdecompress" xreflabel="BZ2_bzBuffToBuffDecompress">
+<title><computeroutput>BZ2_bzBuffToBuffDecompress</computeroutput></title>
+
+<programlisting>
+int BZ2_bzBuffToBuffDecompress( char*         dest,
+                                unsigned int* destLen,
+                                char*         source,
+                                unsigned int  sourceLen,
+                                int           small,
+                                int           verbosity );
+</programlisting>
+
+<para>Attempts to decompress the data in <computeroutput>source[0
+.. sourceLen-1]</computeroutput> into the destination buffer,
+<computeroutput>dest[0 .. *destLen-1]</computeroutput>.  If the
+destination buffer is big enough,
+<computeroutput>*destLen</computeroutput> is set to the size of
+the uncompressed data, and <computeroutput>BZ_OK</computeroutput>
+is returned.  If the compressed data won't fit,
+<computeroutput>*destLen</computeroutput> is unchanged, and
+<computeroutput>BZ_OUTBUFF_FULL</computeroutput> is
+returned.</para>
+
+<para><computeroutput>source</computeroutput> is assumed to hold
+a complete <computeroutput>bzip2</computeroutput> format data
+stream.
+<computeroutput>BZ2_bzBuffToBuffDecompress</computeroutput> tries
+to decompress the entirety of the stream into the output
+buffer.</para>
+
+<para>For the meaning of parameters
+<computeroutput>small</computeroutput> and
+<computeroutput>verbosity</computeroutput>, see
+<computeroutput>BZ2_bzDecompressInit</computeroutput>.</para>
+
+<para>Because the compression ratio of the compressed data cannot
+be known in advance, there is no easy way to guarantee that the
+output buffer will be big enough.  You may of course make
+arrangements in your code to record the size of the uncompressed
+data, but such a mechanism is beyond the scope of this
+library.</para>
+
+<para><computeroutput>BZ2_bzBuffToBuffDecompress</computeroutput>
+will not write data at or beyond
+<computeroutput>dest[*destLen]</computeroutput>, even in case of
+buffer overflow.</para>
+
+<para>Possible return values:</para>
+
+<programlisting>
+BZ_CONFIG_ERROR
+  if the library has been mis-compiled
+BZ_PARAM_ERROR
+  if dest is NULL or destLen is NULL
+  or small != 0 && small != 1
+  or verbosity < 0 or verbosity > 4
+BZ_MEM_ERROR
+  if insufficient memory is available 
+BZ_OUTBUFF_FULL
+  if the size of the compressed data exceeds *destLen
+BZ_DATA_ERROR
+  if a data integrity error was detected in the compressed data
+BZ_DATA_ERROR_MAGIC
+  if the compressed data doesn't begin with the right magic bytes
+BZ_UNEXPECTED_EOF
+  if the compressed data ends unexpectedly
+BZ_OK
+  otherwise
+</programlisting>
+
+</sect2>
+
+</sect1>
+
+
+<sect1 id="zlib-compat" xreflabel="zlib compatibility functions">
+<title><computeroutput>zlib</computeroutput> compatibility functions</title>
+
+<para>Yoshioka Tsuneo has contributed some functions to give
+better <computeroutput>zlib</computeroutput> compatibility.
+These functions are <computeroutput>BZ2_bzopen</computeroutput>,
+<computeroutput>BZ2_bzread</computeroutput>,
+<computeroutput>BZ2_bzwrite</computeroutput>,
+<computeroutput>BZ2_bzflush</computeroutput>,
+<computeroutput>BZ2_bzclose</computeroutput>,
+<computeroutput>BZ2_bzerror</computeroutput> and
+<computeroutput>BZ2_bzlibVersion</computeroutput>.  These
+functions are not (yet) officially part of the library.  If they
+break, you get to keep all the pieces.  Nevertheless, I think
+they work ok.</para>
+
+<programlisting>
+typedef void BZFILE;
+
+const char * BZ2_bzlibVersion ( void );
+</programlisting>
+
+<para>Returns a string indicating the library version.</para>
+
+<programlisting>
+BZFILE * BZ2_bzopen  ( const char *path, const char *mode );
+BZFILE * BZ2_bzdopen ( int        fd,    const char *mode );
+</programlisting>
+
+<para>Opens a <computeroutput>.bz2</computeroutput> file for
+reading or writing, using either its name or a pre-existing file
+descriptor.  Analogous to <computeroutput>fopen</computeroutput>
+and <computeroutput>fdopen</computeroutput>.</para>
+
+<programlisting>
+int BZ2_bzread  ( BZFILE* b, void* buf, int len );
+int BZ2_bzwrite ( BZFILE* b, void* buf, int len );
+</programlisting>
+
+<para>Reads/writes data from/to a previously opened
+<computeroutput>BZFILE</computeroutput>.  Analogous to
+<computeroutput>fread</computeroutput> and
+<computeroutput>fwrite</computeroutput>.</para>
+
+<programlisting>
+int  BZ2_bzflush ( BZFILE* b );
+void BZ2_bzclose ( BZFILE* b );
+</programlisting>
+
+<para>Flushes/closes a <computeroutput>BZFILE</computeroutput>.
+<computeroutput>BZ2_bzflush</computeroutput> doesn't actually do
+anything.  Analogous to <computeroutput>fflush</computeroutput>
+and <computeroutput>fclose</computeroutput>.</para>
+
+<programlisting>
+const char * BZ2_bzerror ( BZFILE *b, int *errnum )
+</programlisting>
+
+<para>Returns a string describing the more recent error status of
+<computeroutput>b</computeroutput>, and also sets
+<computeroutput>*errnum</computeroutput> to its numerical
+value.</para>
+
+</sect1>
+
+
+<sect1 id="stdio-free" 
+       xreflabel="Using the library in a stdio-free environment">
+<title>Using the library in a <computeroutput>stdio</computeroutput>-free environment</title>
+
+
+<sect2 id="stdio-bye" xreflabel="Getting rid of stdio">
+<title>Getting rid of <computeroutput>stdio</computeroutput></title>
+
+<para>In a deeply embedded application, you might want to use
+just the memory-to-memory functions.  You can do this
+conveniently by compiling the library with preprocessor symbol
+<computeroutput>BZ_NO_STDIO</computeroutput> defined.  Doing this
+gives you a library containing only the following eight
+functions:</para>
+
+<para><computeroutput>BZ2_bzCompressInit</computeroutput>,
+<computeroutput>BZ2_bzCompress</computeroutput>,
+<computeroutput>BZ2_bzCompressEnd</computeroutput>
+<computeroutput>BZ2_bzDecompressInit</computeroutput>,
+<computeroutput>BZ2_bzDecompress</computeroutput>,
+<computeroutput>BZ2_bzDecompressEnd</computeroutput>
+<computeroutput>BZ2_bzBuffToBuffCompress</computeroutput>,
+<computeroutput>BZ2_bzBuffToBuffDecompress</computeroutput></para>
+
+<para>When compiled like this, all functions will ignore
+<computeroutput>verbosity</computeroutput> settings.</para>
+
+</sect2>
+
+
+<sect2 id="critical-error" xreflabel="Critical error handling">
+<title>Critical error handling</title>
+
+<para><computeroutput>libbzip2</computeroutput> contains a number
+of internal assertion checks which should, needless to say, never
+be activated.  Nevertheless, if an assertion should fail,
+behaviour depends on whether or not the library was compiled with
+<computeroutput>BZ_NO_STDIO</computeroutput> set.</para>
+
+<para>For a normal compile, an assertion failure yields the
+message:</para>
+
+<blockquote>
+<para>bzip2/libbzip2: internal error number N.</para>
+<para>This is a bug in bzip2/libbzip2, &bz-version; of &bz-date;.
+Please report it to me at: &bz-email;.  If this happened
+when you were using some program which uses libbzip2 as a
+component, you should also report this bug to the author(s)
+of that program.  Please make an effort to report this bug;
+timely and accurate bug reports eventually lead to higher
+quality software.  Thanks.  Julian Seward, &bz-date;.
+</para></blockquote>
+
+<para>where <computeroutput>N</computeroutput> is some error code
+number.  If <computeroutput>N == 1007</computeroutput>, it also
+prints some extra text advising the reader that unreliable memory
+is often associated with internal error 1007. (This is a
+frequently-observed-phenomenon with versions 1.0.0/1.0.1).</para>
+
+<para><computeroutput>exit(3)</computeroutput> is then
+called.</para>
+
+<para>For a <computeroutput>stdio</computeroutput>-free library,
+assertion failures result in a call to a function declared
+as:</para>
+
+<programlisting>
+extern void bz_internal_error ( int errcode );
+</programlisting>
+
+<para>The relevant code is passed as a parameter.  You should
+supply such a function.</para>
+
+<para>In either case, once an assertion failure has occurred, any
+<computeroutput>bz_stream</computeroutput> records involved can
+be regarded as invalid.  You should not attempt to resume normal
+operation with them.</para>
+
+<para>You may, of course, change critical error handling to suit
+your needs.  As I said above, critical errors indicate bugs in
+the library and should not occur.  All "normal" error situations
+are indicated via error return codes from functions, and can be
+recovered from.</para>
+
+</sect2>
+
+</sect1>
+
+
+<sect1 id="win-dll" xreflabel="Making a Windows DLL">
+<title>Making a Windows DLL</title>
+
+<para>Everything related to Windows has been contributed by
+Yoshioka Tsuneo
+(<computeroutput>QWF00133@niftyserve.or.jp</computeroutput> /
+<computeroutput>tsuneo-y@is.aist-nara.ac.jp</computeroutput>), so
+you should send your queries to him (but perhaps Cc: me,
+<computeroutput>&bz-email;</computeroutput>).</para>
+
+<para>My vague understanding of what to do is: using Visual C++
+5.0, open the project file
+<computeroutput>libbz2.dsp</computeroutput>, and build.  That's
+all.</para>
+
+<para>If you can't open the project file for some reason, make a
+new one, naming these files:
+<computeroutput>blocksort.c</computeroutput>,
+<computeroutput>bzlib.c</computeroutput>,
+<computeroutput>compress.c</computeroutput>,
+<computeroutput>crctable.c</computeroutput>,
+<computeroutput>decompress.c</computeroutput>,
+<computeroutput>huffman.c</computeroutput>,
+<computeroutput>randtable.c</computeroutput> and
+<computeroutput>libbz2.def</computeroutput>.  You will also need
+to name the header files <computeroutput>bzlib.h</computeroutput>
+and <computeroutput>bzlib_private.h</computeroutput>.</para>
+
+<para>If you don't use VC++, you may need to define the
+proprocessor symbol
+<computeroutput>_WIN32</computeroutput>.</para>
+
+<para>Finally, <computeroutput>dlltest.c</computeroutput> is a
+sample program using the DLL.  It has a project file,
+<computeroutput>dlltest.dsp</computeroutput>.</para>
+
+<para>If you just want a makefile for Visual C, have a look at
+<computeroutput>makefile.msc</computeroutput>.</para>
+
+<para>Be aware that if you compile
+<computeroutput>bzip2</computeroutput> itself on Win32, you must
+set <computeroutput>BZ_UNIX</computeroutput> to 0 and
+<computeroutput>BZ_LCCWIN32</computeroutput> to 1, in the file
+<computeroutput>bzip2.c</computeroutput>, before compiling.
+Otherwise the resulting binary won't work correctly.</para>
+
+<para>I haven't tried any of this stuff myself, but it all looks
+plausible.</para>
+
+</sect1>
+
+</chapter>
+
+
+
+<chapter id="misc" xreflabel="Miscellanea">
+<title>Miscellanea</title>
+
+<para>These are just some random thoughts of mine.  Your mileage
+may vary.</para>
+
+
+<sect1 id="limits" xreflabel="Limitations of the compressed file format">
+<title>Limitations of the compressed file format</title>
+
+<para><computeroutput>bzip2-1.0.X</computeroutput>,
+<computeroutput>0.9.5</computeroutput> and
+<computeroutput>0.9.0</computeroutput> use exactly the same file
+format as the original version,
+<computeroutput>bzip2-0.1</computeroutput>.  This decision was
+made in the interests of stability.  Creating yet another
+incompatible compressed file format would create further
+confusion and disruption for users.</para>
+
+<para>Nevertheless, this is not a painless decision.  Development
+work since the release of
+<computeroutput>bzip2-0.1</computeroutput> in August 1997 has
+shown complexities in the file format which slow down
+decompression and, in retrospect, are unnecessary.  These
+are:</para>
+
+<itemizedlist mark='bullet'>
+
+ <listitem><para>The run-length encoder, which is the first of the
+   compression transformations, is entirely irrelevant.  The
+   original purpose was to protect the sorting algorithm from the
+   very worst case input: a string of repeated symbols.  But
+   algorithm steps Q6a and Q6b in the original Burrows-Wheeler
+   technical report (SRC-124) show how repeats can be handled
+   without difficulty in block sorting.</para></listitem>
+
+ <listitem><para>The randomisation mechanism doesn't really need to be
+   there.  Udi Manber and Gene Myers published a suffix array
+   construction algorithm a few years back, which can be employed
+   to sort any block, no matter how repetitive, in O(N log N)
+   time.  Subsequent work by Kunihiko Sadakane has produced a
+   derivative O(N (log N)^2) algorithm which usually outperforms
+   the Manber-Myers algorithm.</para>
+
+   <para>I could have changed to Sadakane's algorithm, but I find
+   it to be slower than <computeroutput>bzip2</computeroutput>'s
+   existing algorithm for most inputs, and the randomisation
+   mechanism protects adequately against bad cases.  I didn't
+   think it was a good tradeoff to make.  Partly this is due to
+   the fact that I was not flooded with email complaints about
+   <computeroutput>bzip2-0.1</computeroutput>'s performance on
+   repetitive data, so perhaps it isn't a problem for real
+   inputs.</para>
+
+   <para>Probably the best long-term solution, and the one I have
+   incorporated into 0.9.5 and above, is to use the existing
+   sorting algorithm initially, and fall back to a O(N (log N)^2)
+   algorithm if the standard algorithm gets into
+   difficulties.</para></listitem>
+
+  <listitem><para>The compressed file format was never designed to be
+   handled by a library, and I have had to jump though some hoops
+   to produce an efficient implementation of decompression.  It's
+   a bit hairy.  Try passing
+   <computeroutput>decompress.c</computeroutput> through the C
+   preprocessor and you'll see what I mean.  Much of this
+   complexity could have been avoided if the compressed size of
+   each block of data was recorded in the data stream.</para></listitem>
+
+ <listitem><para>An Adler-32 checksum, rather than a CRC32 checksum,
+   would be faster to compute.</para></listitem>
+
+</itemizedlist>
+
+<para>It would be fair to say that the
+<computeroutput>bzip2</computeroutput> format was frozen before I
+properly and fully understood the performance consequences of
+doing so.</para>
+
+<para>Improvements which I was able to incorporate into 0.9.0,
+despite using the same file format, are:</para>
+
+<itemizedlist mark='bullet'>
+
+ <listitem><para>Single array implementation of the inverse BWT.  This
+  significantly speeds up decompression, presumably because it
+  reduces the number of cache misses.</para></listitem>
+
+ <listitem><para>Faster inverse MTF transform for large MTF values.
+  The new implementation is based on the notion of sliding blocks
+  of values.</para></listitem>
+
+ <listitem><para><computeroutput>bzip2-0.9.0</computeroutput> now reads
+  and writes files with <computeroutput>fread</computeroutput>
+  and <computeroutput>fwrite</computeroutput>; version 0.1 used
+  <computeroutput>putc</computeroutput> and
+  <computeroutput>getc</computeroutput>.  Duh!  Well, you live
+  and learn.</para></listitem>
+
+</itemizedlist>
+
+<para>Further ahead, it would be nice to be able to do random
+access into files.  This will require some careful design of
+compressed file formats.</para>
+
+</sect1>
+
+
+<sect1 id="port-issues" xreflabel="Portability issues">
+<title>Portability issues</title>
+
+<para>After some consideration, I have decided not to use GNU
+<computeroutput>autoconf</computeroutput> to configure 0.9.5 or
+1.0.</para>
+
+<para><computeroutput>autoconf</computeroutput>, admirable and
+wonderful though it is, mainly assists with portability problems
+between Unix-like platforms.  But
+<computeroutput>bzip2</computeroutput> doesn't have much in the
+way of portability problems on Unix; most of the difficulties
+appear when porting to the Mac, or to Microsoft's operating
+systems.  <computeroutput>autoconf</computeroutput> doesn't help
+in those cases, and brings in a whole load of new
+complexity.</para>
+
+<para>Most people should be able to compile the library and
+program under Unix straight out-of-the-box, so to speak,
+especially if you have a version of GNU C available.</para>
+
+<para>There are a couple of
+<computeroutput>__inline__</computeroutput> directives in the
+code.  GNU C (<computeroutput>gcc</computeroutput>) should be
+able to handle them.  If you're not using GNU C, your C compiler
+shouldn't see them at all.  If your compiler does, for some
+reason, see them and doesn't like them, just
+<computeroutput>#define</computeroutput>
+<computeroutput>__inline__</computeroutput> to be
+<computeroutput>/* */</computeroutput>.  One easy way to do this
+is to compile with the flag
+<computeroutput>-D__inline__=</computeroutput>, which should be
+understood by most Unix compilers.</para>
+
+<para>If you still have difficulties, try compiling with the
+macro <computeroutput>BZ_STRICT_ANSI</computeroutput> defined.
+This should enable you to build the library in a strictly ANSI
+compliant environment.  Building the program itself like this is
+dangerous and not supported, since you remove
+<computeroutput>bzip2</computeroutput>'s checks against
+compressing directories, symbolic links, devices, and other
+not-really-a-file entities.  This could cause filesystem
+corruption!</para>
+
+<para>One other thing: if you create a
+<computeroutput>bzip2</computeroutput> binary for public distribution,
+please consider linking it statically (<computeroutput>gcc
+-static</computeroutput>).  This avoids all sorts of library-version
+issues that others may encounter later on.</para>
+
+<para>If you build <computeroutput>bzip2</computeroutput> on
+Win32, you must set <computeroutput>BZ_UNIX</computeroutput> to 0
+and <computeroutput>BZ_LCCWIN32</computeroutput> to 1, in the
+file <computeroutput>bzip2.c</computeroutput>, before compiling.
+Otherwise the resulting binary won't work correctly.</para>
+
+</sect1>
+
+
+<sect1 id="bugs" xreflabel="Reporting bugs">
+<title>Reporting bugs</title>
+
+<para>I tried pretty hard to make sure
+<computeroutput>bzip2</computeroutput> is bug free, both by
+design and by testing.  Hopefully you'll never need to read this
+section for real.</para>
+
+<para>Nevertheless, if <computeroutput>bzip2</computeroutput> dies
+with a segmentation fault, a bus error or an internal assertion
+failure, it will ask you to email me a bug report.  Experience from
+years of feedback of bzip2 users indicates that almost all these
+problems can be traced to either compiler bugs or hardware
+problems.</para>
+
+<itemizedlist mark='bullet'>
+
+ <listitem><para>Recompile the program with no optimisation, and
+  see if it works.  And/or try a different compiler.  I heard all
+  sorts of stories about various flavours of GNU C (and other
+  compilers) generating bad code for
+  <computeroutput>bzip2</computeroutput>, and I've run across two
+  such examples myself.</para>
+
+  <para>2.7.X versions of GNU C are known to generate bad code
+  from time to time, at high optimisation levels.  If you get
+  problems, try using the flags
+  <computeroutput>-O2</computeroutput>
+  <computeroutput>-fomit-frame-pointer</computeroutput>
+  <computeroutput>-fno-strength-reduce</computeroutput>.  You
+  should specifically <emphasis>not</emphasis> use
+  <computeroutput>-funroll-loops</computeroutput>.</para>
+
+  <para>You may notice that the Makefile runs six tests as part
+  of the build process.  If the program passes all of these, it's
+  a pretty good (but not 100%) indication that the compiler has
+  done its job correctly.</para></listitem>
+
+ <listitem><para>If <computeroutput>bzip2</computeroutput>
+  crashes randomly, and the crashes are not repeatable, you may
+  have a flaky memory subsystem.
+  <computeroutput>bzip2</computeroutput> really hammers your
+  memory hierarchy, and if it's a bit marginal, you may get these
+  problems.  Ditto if your disk or I/O subsystem is slowly
+  failing.  Yup, this really does happen.</para>
+
+  <para>Try using a different machine of the same type, and see
+  if you can repeat the problem.</para></listitem>
+
+  <listitem><para>This isn't really a bug, but ... If
+  <computeroutput>bzip2</computeroutput> tells you your file is
+  corrupted on decompression, and you obtained the file via FTP,
+  there is a possibility that you forgot to tell FTP to do a
+  binary mode transfer.  That absolutely will cause the file to
+  be non-decompressible.  You'll have to transfer it
+  again.</para></listitem>
+
+</itemizedlist>
+
+<para>If you've incorporated
+<computeroutput>libbzip2</computeroutput> into your own program
+and are getting problems, please, please, please, check that the
+parameters you are passing in calls to the library, are correct,
+and in accordance with what the documentation says is allowable.
+I have tried to make the library robust against such problems,
+but I'm sure I haven't succeeded.</para>
+
+<para>Finally, if the above comments don't help, you'll have to
+send me a bug report.  Now, it's just amazing how many people
+will send me a bug report saying something like:</para>
+
+<programlisting>
+bzip2 crashed with segmentation fault on my machine
+</programlisting>
+
+<para>and absolutely nothing else.  Needless to say, a such a
+report is <emphasis>totally, utterly, completely and
+comprehensively 100% useless; a waste of your time, my time, and
+net bandwidth</emphasis>.  With no details at all, there's no way
+I can possibly begin to figure out what the problem is.</para>
+
+<para>The rules of the game are: facts, facts, facts.  Don't omit
+them because "oh, they won't be relevant".  At the bare
+minimum:</para>
+
+<programlisting>
+Machine type.  Operating system version.  
+Exact version of bzip2 (do bzip2 -V).  
+Exact version of the compiler used.  
+Flags passed to the compiler.
+</programlisting>
+
+<para>However, the most important single thing that will help me
+is the file that you were trying to compress or decompress at the
+time the problem happened.  Without that, my ability to do
+anything more than speculate about the cause, is limited.</para>
+
+</sect1>
+
+
+<sect1 id="package" xreflabel="Did you get the right package?">
+<title>Did you get the right package?</title>
+
+<para><computeroutput>bzip2</computeroutput> is a resource hog.
+It soaks up large amounts of CPU cycles and memory.  Also, it
+gives very large latencies.  In the worst case, you can feed many
+megabytes of uncompressed data into the library before getting
+any compressed output, so this probably rules out applications
+requiring interactive behaviour.</para>
+
+<para>These aren't faults of my implementation, I hope, but more
+an intrinsic property of the Burrows-Wheeler transform
+(unfortunately).  Maybe this isn't what you want.</para>
+
+<para>If you want a compressor and/or library which is faster,
+uses less memory but gets pretty good compression, and has
+minimal latency, consider Jean-loup Gailly's and Mark Adler's
+work, <computeroutput>zlib-1.2.1</computeroutput> and
+<computeroutput>gzip-1.2.4</computeroutput>.  Look for them at 
+<ulink url="http://www.zlib.org">http://www.zlib.org</ulink> and 
+<ulink url="http://www.gzip.org">http://www.gzip.org</ulink>
+respectively.</para>
+
+<para>For something faster and lighter still, you might try Markus F
+X J Oberhumer's <computeroutput>LZO</computeroutput> real-time
+compression/decompression library, at 
+<ulink url="http://www.oberhumer.com/opensource">http://www.oberhumer.com/opensource</ulink>.</para>
+
+</sect1>
+
+
+
+<sect1 id="reading" xreflabel="Further Reading">
+<title>Further Reading</title>
+
+<para><computeroutput>bzip2</computeroutput> is not research
+work, in the sense that it doesn't present any new ideas.
+Rather, it's an engineering exercise based on existing
+ideas.</para>
+
+<para>Four documents describe essentially all the ideas behind
+<computeroutput>bzip2</computeroutput>:</para>
+
+<literallayout>Michael Burrows and D. J. Wheeler:
+  "A block-sorting lossless data compression algorithm"
+   10th May 1994. 
+   Digital SRC Research Report 124.
+   ftp://ftp.digital.com/pub/DEC/SRC/research-reports/SRC-124.ps.gz
+   If you have trouble finding it, try searching at the
+   New Zealand Digital Library, http://www.nzdl.org.
+
+Daniel S. Hirschberg and Debra A. LeLewer
+  "Efficient Decoding of Prefix Codes"
+   Communications of the ACM, April 1990, Vol 33, Number 4.
+   You might be able to get an electronic copy of this
+   from the ACM Digital Library.
+
+David J. Wheeler
+   Program bred3.c and accompanying document bred3.ps.
+   This contains the idea behind the multi-table Huffman coding scheme.
+   ftp://ftp.cl.cam.ac.uk/users/djw3/
+
+Jon L. Bentley and Robert Sedgewick
+  "Fast Algorithms for Sorting and Searching Strings"
+   Available from Sedgewick's web page,
+   www.cs.princeton.edu/~rs
+</literallayout>
+
+<para>The following paper gives valuable additional insights into
+the algorithm, but is not immediately the basis of any code used
+in bzip2.</para>
+
+<literallayout>Peter Fenwick:
+   Block Sorting Text Compression
+   Proceedings of the 19th Australasian Computer Science Conference,
+     Melbourne, Australia.  Jan 31 - Feb 2, 1996.
+   ftp://ftp.cs.auckland.ac.nz/pub/peter-f/ACSC96paper.ps</literallayout>
+
+<para>Kunihiko Sadakane's sorting algorithm, mentioned above, is
+available from:</para>
+
+<literallayout>http://naomi.is.s.u-tokyo.ac.jp/~sada/papers/Sada98b.ps.gz
+</literallayout>
+
+<para>The Manber-Myers suffix array construction algorithm is
+described in a paper available from:</para>
+
+<literallayout>http://www.cs.arizona.edu/people/gene/PAPERS/suffix.ps
+</literallayout>
+
+<para>Finally, the following papers document some
+investigations I made into the performance of sorting
+and decompression algorithms:</para>
+
+<literallayout>Julian Seward
+   On the Performance of BWT Sorting Algorithms
+   Proceedings of the IEEE Data Compression Conference 2000
+     Snowbird, Utah.  28-30 March 2000.
+
+Julian Seward
+   Space-time Tradeoffs in the Inverse B-W Transform
+   Proceedings of the IEEE Data Compression Conference 2001
+     Snowbird, Utah.  27-29 March 2001.
+</literallayout>
+
+</sect1>
+
+</chapter>
+
+</book>
diff --git a/randtable.c b/randtable.c
index 5c922e9..940462d 100644
--- a/randtable.c
+++ b/randtable.c
@@ -8,7 +8,7 @@
   This file is a part of bzip2 and/or libbzip2, a program and
   library for lossless, block-sorting data compression.
 
-  Copyright (C) 1996-2002 Julian R Seward.  All rights reserved.
+  Copyright (C) 1996-2005 Julian R Seward.  All rights reserved.
 
   Redistribution and use in source and binary forms, with or without
   modification, are permitted provided that the following conditions
@@ -42,7 +42,7 @@
   SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
   Julian Seward, Cambridge, UK.
-  jseward@acm.org
+  jseward@bzip.org
   bzip2/libbzip2 version 1.0 of 21 March 2000
 
   This program is based on (at least) the work of:
diff --git a/xmlproc.sh b/xmlproc.sh
new file mode 100755
index 0000000..6fe4d57
--- /dev/null
+++ b/xmlproc.sh
@@ -0,0 +1,99 @@
+#!/bin/bash
+# see the README in this directory for usage etc.
+
+usage() {
+  echo '';
+  echo 'Usage: xmlproc.sh -[option] <filename.xml>';
+  echo 'Specify a target from:';
+  echo '-v      verify xml file conforms to dtd';
+  echo '-html   output in html format (single file)';
+  echo '-ps     output in postscript format';
+  echo '-pdf    output in pdf format';
+  exit;
+}
+
+if test $# -ne 2; then
+  usage
+fi
+# assign the variable for the output type
+action=$1; shift
+# assign the output filename
+xmlfile=$1; shift
+# and check user input it correct
+if !(test -f $xmlfile); then
+  echo "No such file: $xmlfile";
+  exit;
+fi
+# some other stuff we will use
+OUT=output
+xsl_fo=bz-fo.xsl
+xsl_html=bz-html.xsl
+
+basename=$xmlfile
+basename=${basename//'.xml'/''}
+
+fofile="${basename}.fo"
+htmlfile="${basename}.html"
+pdffile="${basename}.pdf"
+psfile="${basename}.ps"
+xmlfmtfile="${basename}.fmt"
+
+# first process the xmlfile with CDATA tags
+./format.pl $xmlfile $xmlfmtfile
+# so the shell knows where the catalogs live
+export XML_CATALOG_FILES=/etc/xml/catalog
+
+# post-processing tidy up
+cleanup() {
+  echo "Cleaning up: # $@" 
+  while [ $# != 0 ]
+  do
+    arg=$1; shift;
+    echo "  deleting $arg";
+    rm $arg
+  done
+}
+
+case $action in
+  -v)
+   flags='--noout --xinclude --noblanks --postvalid'
+   dtd='--dtdvalid http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd'
+   xmllint $flags $dtd $xmlfmtfile 2> $OUT 
+   egrep 'error' $OUT 
+   rm $OUT
+  ;;
+
+  -html)
+   echo "Creating $htmlfile ..."
+   xsltproc --nonet --xinclude  -o $htmlfile $xsl_html $xmlfmtfile
+   cleanup $xmlfmtfile
+  ;;
+
+  -pdf)
+   echo "Creating $pdffile ..."
+   xsltproc --nonet --xinclude -o $fofile $xsl_fo $xmlfmtfile
+   pdfxmltex $fofile >$OUT </dev/null
+   pdfxmltex $fofile >$OUT </dev/null
+   pdfxmltex $fofile >$OUT </dev/null
+   cleanup $OUT $xmlfmtfile *.aux *.fo *.log *.out
+  ;;
+
+  -ps)
+   echo "Creating $psfile ..."
+   xsltproc --nonet --xinclude -o $fofile $xsl_fo $xmlfmtfile
+   pdfxmltex $fofile >$OUT </dev/null
+   pdfxmltex $fofile >$OUT </dev/null
+   pdfxmltex $fofile >$OUT </dev/null
+   pdftops $pdffile $psfile
+   cleanup $OUT $xmlfmtfile $pdffile *.aux *.fo *.log *.out
+#  passivetex is broken, so we can't go this route yet.
+#   xmltex $fofile >$OUT </dev/null
+#   xmltex $fofile >$OUT </dev/null
+#   xmltex $fofile >$OUT </dev/null
+#   dvips -R -q -o bzip-manual.ps *.dvi
+  ;;
+
+  *)
+  usage
+  ;;
+esac