5 0. Overview
6 1. Documentation
7 1.1. Overall documentation
8 1.2. Documentation for command-line tools
9 1.3. Documentation for liblzma
10 2. Version numbering
11 3. Reporting bugs
12 4. Translating the xz tool
13 5. Other implementations of the .xz format
14 6. Contact information
20 XZ Utils provide a general-purpose data-compression library plus
21 command-line tools. The native file format is the .xz format, but
22 also the legacy .lzma format is supported. The .xz format supports
23 multiple compression algorithms, which are called "filters" in the
24 context of XZ Utils. The primary filter is currently LZMA2. With
25 typical files, XZ Utils create about 30 % smaller files than gzip.
27 To ease adapting support for the .xz format into existing applications
28 and scripts, the API of liblzma is somewhat similar to the API of the
29 popular zlib library. For the same reason, the command-line tool xz
30 has a command-line syntax similar to that of gzip.
32 When aiming for the highest compression ratio, the LZMA2 encoder uses
33 a lot of CPU time and may use, depending on the settings, even
34 hundreds of megabytes of RAM. However, in fast modes, the LZMA2 encoder
35 competes with bzip2 in compression speed, RAM usage, and compression
38 LZMA2 is reasonably fast to decompress. It is a little slower than
39 gzip, but a lot faster than bzip2. Being fast to decompress means
40 that the .xz format is especially nice when the same file will be
41 decompressed very many times (usually on different computers), which
42 is the case e.g. when distributing software packages. In such
43 situations, it's not too bad if the compression takes some time,
44 since that needs to be done only once to benefit many people.
46 With some file types, combining (or "chaining") LZMA2 with an
47 additional filter can improve the compression ratio. A filter chain may
48 contain up to four filters, although usually only one or two are used.
49 For example, putting a BCJ (Branch/Call/Jump) filter before LZMA2
50 in the filter chain can improve compression ratio of executable files.
52 Since the .xz format allows adding new filter IDs, it is possible that
53 some day there will be a filter that is, for example, much faster to
54 compress than LZMA2 (but probably with worse compression ratio).
55 Similarly, it is possible that some day there is a filter that will
56 compress better than LZMA2.
58 XZ Utils doesn't support multithreaded compression or decompression
59 yet. It has been planned though and taken into account when designing
60 the .xz file format.
661.1. Overall documentation
68 README This file
70 INSTALL.generic Generic install instructions for those not familiar
71 with packages using GNU Autotools
72 INSTALL Installation instructions specific to XZ Utils
73 PACKAGERS Information to packagers of XZ Utils
75 COPYING XZ Utils copyright and license information
76 COPYING.GPLv2 GNU General Public License version 2
77 COPYING.GPLv3 GNU General Public License version 3
78 COPYING.LGPLv2.1 GNU Lesser General Public License version 2.1
80 AUTHORS The main authors of XZ Utils
81 THANKS Incomplete list of people who have helped making
82 this software
83 NEWS User-visible changes between XZ Utils releases
84 ChangeLog Detailed list of changes (commit log)
85 TODO Known bugs and some sort of to-do list
87 Note that only some of the above files are included in binary
911.2. Documentation for command-line tools
93 The command-line tools are documented as man pages. In source code
94 releases (and possibly also in some binary packages), the man pages
95 are also provided in plain text (ASCII only) and PDF formats in the
96 directory "doc/man" to make the man pages more accessible to those
97 whose operating system doesn't provide an easy way to view man pages.
1001.3. Documentation for liblzma
102 The liblzma API headers include short docs about each function
103 and data type as Doxygen tags. These docs should be quite OK as
104 a quick reference.
106 I have planned to write a bunch of very well documented example
107 programs, which (due to comments) should work as a tutorial to
108 various features of liblzma. No such example programs have been
109 written yet.
111 For now, if you have never used liblzma, libbzip2, or zlib, I
112 recommend learning the *basics* of the zlib API. Once you know that,
113 it should be easier to learn liblzma.
1192. Version numbering
122 The version number format of XZ Utils is X.Y.ZS:
124 - X is the major version. When this is incremented, the library
125 API and ABI break.
127 - Y is the minor version. It is incremented when new features
128 are added without breaking the existing API or ABI. An even Y
129 indicates a stable release and an odd Y indicates unstable
130 (alpha or beta version).
132 - Z is the revision. This has a different meaning for stable and
133 unstable releases:
135 * Stable: Z is incremented when bugs get fixed without adding
136 any new features. This is intended to be convenient for
137 downstream distributors that want bug fixes but don't want
138 any new features to minimize the risk of introducing new bugs.
140 * Unstable: Z is just a counter. API or ABI of features added
141 in earlier unstable releases having the same X.Y may break.
143 - S indicates stability of the release. It is missing from the
144 stable releases, where Y is an even number. When Y is odd, S
145 is either "alpha" or "beta" to make it very clear that such
146 versions are not stable releases. The same X.Y.Z combination is
147 not used for more than one stability level, i.e. after X.Y.Zalpha,
148 the next version can be X.Y.(Z+1)beta but not X.Y.Zbeta.
1513. Reporting bugs
154 Naturally it is easiest for me if you already know what causes the
155 unexpected behavior. Even better if you have a patch to propose.
156 However, quite often the reason for unexpected behavior is unknown,
157 so here are a few things to do before sending a bug report:
159 1. Try to create a small example how to reproduce the issue.
161 2. Compile XZ Utils with debugging code using configure switches
162 --enable-debug and, if possible, --disable-shared. If you are
163 using GCC, use CFLAGS='-O0 -ggdb3'. Don't strip the resulting
166 3. Turn on core dumps. The exact command depends on your shell;
167 for example in GNU bash it is done with "ulimit -c unlimited",
168 and in tcsh with "limit coredumpsize unlimited".
170 4. Try to reproduce the suspected bug. If you get "assertion failed"
171 message, be sure to include the complete message in your bug
172 report. If the application leaves a coredump, get a backtrace
173 using gdb:
174 $ gdb /path/to/app-binary # Load the app to the debugger.
175 (gdb) core core # Open the coredump.
176 (gdb) bt # Print the backtrace. Copy & paste to bug report.
177 (gdb) quit # Quit gdb.
179 Report your bug via email or IRC (see Contact information below).
180 Don't send core dump files or any executables. If you have a small
181 example file(s) (total size less than 256 KiB), please include
182 it/them as an attachment. If you have bigger test files, put them
183 online somewhere and include a URL to the file(s) in the bug report.
185 Always include the exact version number of XZ Utils in the bug report.
186 If you are using a snapshot from the git repository, use "git describe"
187 to get the exact snapshot version. If you are using XZ Utils shipped
188 in an operating system distribution, mention the distribution name,
189 distribution version, and exact xz package version; if you cannot
190 repeat the bug with the code compiled from unpatched source code,
191 you probably need to report a bug to your distribution's bug tracking
1954. Translating the xz tool
198 The messages from the xz tool have been translated into a few
199 languages. Before starting to translate into a new language, ask
200 the author whether someone else hasn't already started working on it.
202 Test your translation. Testing includes comparing the translated
203 output to the original English version by running the same commands
204 in both your target locale and with LC_ALL=C. Ask someone to
205 proof-read and test the translation.
207 Testing can be done e.g. by installing xz into a temporary directory:
209 ./configure --disable-shared --prefix=/tmp/xz-test
210 # <Edit the .po file in the po directory.>
211 make -C po update-po
212 make install
213 bash debug/translation.bash | less
214 bash debug/translation.bash | less -S # For --list outputs
216 Repeat the above as needed (no need to re-run configure though).
218 Note especially the following:
220 - The output of --help and --long-help must look nice on
221 an 80-column terminal. It's OK to add extra lines if needed.
223 - In contrast, don't add extra lines to error messages and such.
224 They are often preceded with e.g. a filename on the same line,
225 so you have no way to predict where to put a \n. Let the terminal
226 do the wrapping even if it looks ugly. Adding new lines will be
227 even uglier in the generic case even if it looks nice in a few
228 limited examples.
230 - Be careful with column alignment in tables and table-like output
231 (--list, --list --verbose --verbose, --info-memory, --help, and
234 * All descriptions of options in --help should start in the
235 same column (but it doesn't need to be the same column as
236 in the English messages; just be consistent if you change it).
237 Check that both --help and --long-help look OK, since they
238 share several strings.
240 * --list --verbose and --info-memory print lines that have
241 the format "Description: %s". If you need a longer
242 description, you can put extra space between the colon
243 and %s. Then you may need to add extra space to other
244 strings too so that the result as a whole looks good (all
245 values start at the same column).
247 * The columns of the actual tables in --list --verbose --verbose
248 should be aligned properly. Abbreviate if necessary. It might
249 be good to keep at least 2 or 3 spaces between column headings
250 and avoid spaces in the headings so that the columns stand out
251 better, but this is a matter of opinion. Do what you think
252 looks best.
254 - Be careful to put a period at the end of a sentence when the
255 original version has it, and don't put it when the original
256 doesn't have it. Similarly, be careful with \n characters
257 at the beginning and end of the strings.
259 - Read the TRANSLATORS comments that have been extracted from the
260 source code and included in xz.pot. If they suggest testing the
261 translation with some type of command, do it. If testing needs
262 input files, use e.g. tests/files/good-*.xz.
264 - When updating the translation, read the fuzzy (modified) strings
265 carefully, and don't mark them as updated before you actually
266 have updated them. Reading through the unchanged messages can be
267 good too; sometimes you may find a better wording for them.
269 - If you find language problems in the original English strings,
270 feel free to suggest improvements. Ask if something is unclear.
272 - The translated messages should be understandable (sometimes this
273 may be a problem with the original English messages too). Don't
274 make a direct word-by-word translation from English especially if
275 the result doesn't sound good in your language.
277 In short, take your time and pay attention to the details. Making
278 a good translation is not a quick and trivial thing to do. The
279 translated xz should look as polished as the English version.
2825. Other implementations of the .xz format
285 7-Zip and the p7zip port of 7-Zip support the .xz format starting
286 from the version 9.00alpha.
291 XZ Embedded is a limited implementation written for use in the Linux
292 kernel, but it is also suitable for other embedded use.
2976. Contact information
300 If you have questions, bug reports, patches etc. related to XZ Utils,
301 contact Lasse Collin <firstname.lastname@example.org> (in Finnish or English).
302 I'm sometimes slow at replying. If you haven't got a reply within two
303 weeks, assume that your email has got lost and resend it or use IRC.
305 You can find me also from #tukaani on Freenode; my nick is Larhzu.
306 The channel tends to be pretty quiet, so just ask your question and
307 someone may wake up.