Next: , Up: (dir)

Zutils Manual

This manual is for Zutils (version 1.13, 23 January 2024).
Copyright © 2009-2024 Antonio Diaz Diaz.

This manual is free documentation: you have unlimited permission to copy, distribute, and modify it.

Next: , Previous: Top, Up: Top

1 Introduction

Zutils is a collection of utilities able to process any combination of compressed and uncompressed files transparently. If any file given, including standard input, is compressed, its decompressed content is used. Compressed files are decompressed on the fly; no temporary files are created. Data format is detected by its identifier string (magic bytes), not by the file name extension. Empty files are considered uncompressed.

These utilities are not wrapper scripts but safer and more efficient C++ programs. In particular the option --recursive is very efficient in those utilities supporting it.

The utilities provided are zcat, zcmp, zdiff, zgrep, ztest, and zupdate.
The formats supported are bzip2, gzip, lzip, xz, and zstd.
Zutils uses external compressors. The compressor to be used for each format is configurable at runtime.

zcat, zcmp, zdiff, and zgrep are improved replacements for the shell scripts provided by GNU gzip. ztest is unique to zutils. zupdate is similar to gzip's znew.

When zcat, zcmp, zdiff, or zgrep need to try compressed file names, the search order is: lzip, gzip, bzip2, zstd, xz. (file.[lz|gz|bz2|zst|xz]).

NOTE: Bzip2 and lzip provide well-defined values of exit status, which makes them safe to use with zutils. Gzip and xz may return ambiguous warning values, making them less reliable back ends for zutils. Zstd currently does not even document its exit status in its man page. See compressor-requirements.

FORMAT NOTE 1: The option --format allows the processing of a subset of formats in recursive mode and when trying compressed file names. For example, use the following command to search for the string 'foo' in gzip and lzip files only: 'zgrep foo -r --format=gz,lz somedir somefile.tar'.

FORMAT NOTE 2: The standard POSIX compress format (.Z) is obsolete and is only supported through gzip. For this to work, the gzip program used (for example GNU gzip) must be able to decompress .Z files.

LANGUAGE NOTE: Uncompressed = not compressed = plain data; it may never have been compressed. Decompressed is used to refer to data which have undergone the process of decompression.

Next: , Previous: Introduction, Up: Top

2 Common options

The following options: are available in all the utilities. Rather than writing identical descriptions for each of the programs, they are described here. Remember to prepend ./ to any file name beginning with a hyphen, or use '--'.

-h
--help
Print an informative help message describing the options and exit. zgrep only supports the --help form of this option.


-V
--version
Print the version number on the standard output and exit. This version number should be included in all bug reports. In verbose mode, zdiff and zgrep print also the version of the diff or grep program used respectively. At verbosity level 1 (2 for zdiff and zgrep) or higher, print also the versions of the compressors used (perhaps limited by option --format). (The compressors used must support the option -V for this to work).
-M format_list
--format=format_list
Process only the formats listed in the comma-separated format_list. Valid formats are 'bz2', 'gz', 'lz', 'xz', 'zst', and 'un' for 'uncompressed', meaning "any file name without a known extension". This option excludes files based on extension, instead of format, because it is more efficient. The exclusion only applies to names generated automatically (for example when adding extensions to a file name or when operating recursively on directories). Files given in the command line are always processed.

Each format in format_list enables file names with the following extensions:

bz2 enables .bz2 .tbz .tbz2
gz enables .gz .tgz .Z
lz enables .lz .tlz
xz enables .xz .txz
zst enables .zst .tzst
un enables any other file name

-N
--no-rcfile
Don't read the runtime configuration file zutils.conf.
--bz2=command
--gz=command
--lz=command
--xz=command
--zst=command
Set program to be used as decompressor for the corresponding format. command may include arguments. For example --lz='plzip --threads=2'. zupdate uses --lz for compression, not for decompression (see lz-compressor). The name of the program can't begin with '-'. These options override the values set in zutils.conf. The compression program used must meet three requirements:

  1. When called with the option -d and without file names, it must read compressed data from the standard input and produce decompressed data on the standard output.
  2. If the option -q is passed to zutils, the compression program must also accept it.
  3. It must return 0 if no errors occurred, and a non-zero value otherwise.

Numbers given as arguments to options may be expressed in decimal, hexadecimal, or octal (using the same syntax as integer constants in C++), and may be followed by a multiplier and an optional 'B' for "byte".

Table of SI and binary prefixes (unit multipliers):

Prefix Value | Prefix Value
k kilobyte (10^3 = 1000) | Ki kibibyte (2^10 = 1024)
M megabyte (10^6) | Mi mebibyte (2^20)
G gigabyte (10^9) | Gi gibibyte (2^30)
T terabyte (10^12) | Ti tebibyte (2^40)
P petabyte (10^15) | Pi pebibyte (2^50)
E exabyte (10^18) | Ei exbibyte (2^60)
Z zettabyte (10^21) | Zi zebibyte (2^70)
Y yottabyte (10^24) | Yi yobibyte (2^80)
R ronnabyte (10^27) | Ri robibyte (2^90)
Q quettabyte (10^30) | Qi quebibyte (2^100)

Next: , Previous: Common options, Up: Top

3 The configuration file 'zutils.conf'

zutils.conf is the runtime configuration file for zutils. In it you may define the compressor name and options to be used for each format. zutils.conf is optional; you don't need to install it in order to run zutils.

The compressors specified in the command line override those specified in zutils.conf.

You may copy the system zutils.conf file ${sysconfdir}/zutils.conf to $XDG_CONFIG_HOME/zutils.conf and customize these options as you like. (XDG_CONFIG_HOME defaults to $HOME/.config). The file syntax is fairly obvious (and there are further instructions in it):

  1. Any line beginning with '#' is a comment line.
  2. Each non-comment line defines the command to be used for the corresponding format, with the syntax: 
              <format> = <compressor> [options]

    where <format> is one of 'bz2', 'gz', 'lz', 'xz', or 'zst'.

Next: , Previous: Configuration, Up: Top

4 Zcat

zcat copies each file argument to standard output in sequence. If any file given is compressed, its decompressed content is copied. If a file given does not exist, and its name does not end with one of the known extensions, zcat tries the compressed file names corresponding to the formats supported until one is found. See search-order. If a file fails to decompress, zcat continues copying the rest of the files.

If a file is specified as '-', data are read from standard input, decompressed if needed, and sent to standard output. Data read from standard input must be of the same type; all uncompressed or all in the same compressed format.

If no files are specified, recursive searches examine the current working directory, and nonrecursive searches read standard input.

The format for running zcat is: 

     zcat [options] [files]

Exit status is 0 if no errors occurred, 1 otherwise.

zcat supports the following options:

-A
--show-all
Equivalent to -vET.
-b
--number-nonblank
Number all nonblank output lines, starting with 1. The line count is unlimited.
-e
Equivalent to -vE.
-E
--show-ends
Print a '$' after the end of each line.
-n
--number
Number all output lines, starting with 1. The line count is unlimited.
-O format
--force-format=format
Force the compressed format given. Valid values for format are 'bz2', 'gz', 'lz', 'xz', 'zst', and 'un' for 'uncompressed'. If this option is used, the files are passed to the corresponding decompressor (or transmitted unmodified) without checking their format, and the exact file name must be given. Other names are not tried.
-q
--quiet
Quiet operation. Suppress all messages.
-r
--recursive
For each directory operand, read and process all files in that directory, recursively. Follow symbolic links given in the command line, but skip symbolic links that are encountered recursively.
-R
--dereference-recursive
For each directory operand, read and process all files in that directory, recursively, following all symbolic links.
-s
--squeeze-blank
Replace multiple adjacent blank lines with a single blank line.
-t
Equivalent to -vT.
-T
--show-tabs
Print TAB characters as '^I'.
-v
--show-nonprinting
Print control characters except for LF (newline) and TAB using '^' notation and precede characters larger than 127 with 'M-' (which stands for "meta").
--verbose
Verbose mode. Show error messages. Repeating it increases the verbosity level. See version.

Next: , Previous: Zcat, Up: Top

5 Zcmp

zcmp compares two files and, if they differ, writes to standard output the first byte and line number where they differ. Bytes and lines are numbered starting with 1. A hyphen '-' used as a file argument means standard input. If any file given is compressed, its decompressed content is used. Compressed files are decompressed on the fly; no temporary files are created.

The format for running zcmp is: 

     zcmp [options] file1 [file2]

This compares file1 to file2. The standard input is used only if file1 or file2 refers to standard input. If file2 is omitted zcmp tries to compare file1 with the corresponding uncompressed file (if file1 is compressed), and then with the corresponding compressed files of the remaining formats until one is found. See search-order.

An exit status of 0 means no differences were found, 1 means some differences were found, and 2 means trouble.

zcmp supports the following options:

-b
--print-bytes
Print the values of the differing bytes (in octal by default) followed by the bytes themselves in printable form. Print control bytes as a '^' followed by a letter, and precede bytes larger than 127 with 'M-' (which stands for "meta").
-H
--hexadecimal
Print the values of the differing bytes in hexadecimal instead of octal.
-i size
--ignore-initial=size
Ignore any differences in the first size bytes of the input files. Treat files with fewer than size bytes as if they were empty. If size is in the form 'size1:size2', ignore the first size1 bytes of the first input file and the first size2 bytes of the second input file.
-l
--list
Print the byte numbers (in decimal) and values (in octal by default) of all differing bytes. Bytes are numbered starting with 1.
-n count
--bytes=count
Compare at most count input bytes.
-O [format1][,format2]
--force-format=[format1][,format2]
Force the compressed formats given. If format1 or format2 is omitted, the corresponding format is automatically detected. Valid values for format are 'bz2', 'gz', 'lz', 'xz', 'zst', and 'un' for 'uncompressed'. If at least one format is specified with this option, the file is passed to the corresponding decompressor (or transmitted unmodified) without checking its format, and the exact file names of both file1 and file2 must be given. Other names are not tried.
-q
--quiet
--silent
Suppress diagnostics written to standard error, even the 'EOF on <name_of_shorter_file>' diagnostic. Byte differences are still written to standard output. (-q produces no output except byte differences).
-s
--script
Write nothing to standard output or standard error when files differ, not even the 'EOF on <name_of_shorter_file>' diagnostic; indicate differing files through exit status only. Diagnostic messages are still written to standard error when an error is encountered. (-s produces no output except error messages).
-v
--verbose
Verbose mode. Undoes the effect of --quiet. Further -v's increase the verbosity level. See version.

Next: , Previous: Zcmp, Up: Top

6 Zdiff

zdiff compares two files and, if they differ, writes to standard output the differences line by line. A hyphen '-' used as a file argument means standard input. If any file given is compressed, its decompressed content is used. zdiff is a front end to the program diff and has the limitation that messages from diff refer to temporary file names instead of those specified.

The format for running zdiff is: 

     zdiff [options] file1 [file2]

This compares file1 to file2. The standard input is used only if file1 or file2 refers to standard input. If file2 is omitted zdiff tries to compare file1 with the corresponding uncompressed file (if file1 is compressed), and then with the corresponding compressed files of the remaining formats until one is found. See search-order.

An exit status of 0 means no differences were found, 1 means some differences were found, and 2 means trouble.

zdiff supports the following options (some options only work if the diff program used supports them):

-a
--text
Treat all files as text.
-b
--ignore-space-change
Ignore changes in the amount of white space.
-B
--ignore-blank-lines
Ignore changes whose lines are all blank.
-c
Use the context output format.
-C n
--context=n
Same as -c but use n lines of context.
-d
--minimal
Try hard to find a smaller set of changes.
-E
--ignore-tab-expansion
Ignore changes due to tab expansion.
-i
--ignore-case
Ignore case differences. Consider uppercase and lowercase letters equivalent.
-O [format1][,format2]
--force-format=[format1][,format2]
Force the compressed formats given. If format1 or format2 is omitted, the corresponding format is automatically detected. Valid values for format are 'bz2', 'gz', 'lz', 'xz', 'zst', and 'un' for 'uncompressed'. If at least one format is specified with this option, the file is passed to the corresponding decompressor (or transmitted unmodified) without checking its format, and the exact file names of both file1 and file2 must be given. Other names are not tried.
-p
--show-c-function
Show which C function each change is in.
-q
--brief
Output only whether files differ.
-s
--report-identical-files
Report when two files are identical.
-t
--expand-tabs
Expand tabs to spaces in output.
-T
--initial-tab
Make tabs line up by prepending a tab.
-u
Use the unified output format.
-U n
--unified=n
Same as -u but use n lines of context.
-v
--verbose
When specified before --version, print the version of the diff program used. Further -v's increase the verbosity level. See version.
-w
--ignore-all-space
Ignore all white space.
-W columns
--width=columns
Output at most the specified number of print columns per line in side by side format.
-y
--side-by-side
Use the side by side output format.

Next: , Previous: Zdiff, Up: Top

7 Zgrep

zgrep is a front end to the program grep that allows transparent search on any combination of compressed and uncompressed files. If any file given is compressed, its decompressed content is used. If a file given does not exist, and its name does not end with one of the known extensions, zgrep tries the compressed file names corresponding to the formats supported until one is found. See search-order. If a file fails to decompress, zgrep continues searching the rest of the files.

If a file is specified as '-', data are read from standard input, decompressed if needed, and fed to grep. Data read from standard input must be of the same type; all uncompressed or all in the same compressed format.

If no files are specified, recursive searches examine the current working directory, and nonrecursive searches read standard input.

For efficiency reasons, zgrep does not always read all its input. For example, the shell command 'base64 -d foo | zgrep -q X' can cause zgrep to exit immediately after reading a line containing 'X', without bothering to read the rest of its input data. This in turn can cause base64 to exit with a nonzero status because base64 cannot write to its output pipe after zgrep exits.

The format for running zgrep is: 

     zgrep [options] pattern [files]

An exit status of 0 means at least one match was found, 1 means no matches were found, and 2 means trouble.

zgrep supports the following options (Some options only work if the grep program used supports them. Options -h, -H, -r, -R, and -Z are managed by zgrep and not passed to grep):

-a
--text
Treat all files as text.
-A n
--after-context=n
Print n lines of trailing context.
-b
--byte-offset
Print the byte offset of each line.
-B n
--before-context=n
Print n lines of leading context.
-c
--count
Only print a count of matching lines per file.
-C n
--context=n
Print n lines of output context.
--color[=when]
Show matched strings in color. when is 'never', 'always', or 'auto'.
-e pattern
--regexp=pattern
Use pattern as the pattern to match.
-E
--extended-regexp
Interpret pattern as an extended regular expression (ERE).
-f file
--file=file
Obtain patterns from file, one per line.
When searching in several files at once, command substitution can be used with -e to read file only once, for example if file is not a regular file: 'zgrep -e "$(cat file)" file1.lz file2.gz'
-F
--fixed-strings
Interpret pattern as a set of newline-separated strings.
-G
--basic-regexp
Interpret pattern as a basic regular expression (BRE). This is the default.
-h
--no-filename
Suppress the prefixing of file names on output when multiple files are searched.
-H
--with-filename
Print the file name for each match.
-i
--ignore-case
Ignore case distinctions.
-I
Ignore binary files.
-l
--files-with-matches
Only print names of files containing at least one match. Stop reading each file on the first match.
-L
--files-without-match
Only print names of files not containing any matches. Stop reading each file on the first match.
Note: option -L fails (prints wrong results, returns wrong status, and even hangs) when using GNU grep versions 3.2 to 3.4 inclusive because of a wrong change in the exit status of grep, which was reverted in GNU grep 3.5.
--label=label
Display input actually coming from standard input as input coming from file label.
--line-buffered
Use line buffering on output. This may cause a performance penalty.
-m n
--max-count=n
Stop after n matches.
-n
--line-number
Prefix each matched line with its line number in the input file.
-o
--only-matching
Show only the part of matching lines that actually matches pattern.
-O format
--force-format=format
Force the compressed format given. Valid values for format are 'bz2', 'gz', 'lz', 'xz', 'zst', and 'un' for 'uncompressed'. If this option is used, the files are passed to the corresponding decompressor (or transmitted unmodified) without checking their format, and the exact file name must be given. Other names are not tried.
-P
--perl-regexp
Interpret pattern as a Perl-compatible regular expression (PCRE).
-q
--quiet
--silent
Suppress all messages. Exit immediately with zero status if any match is found, even if an error was detected.
-r
--recursive
For each directory operand, read and process all files in that directory, recursively. Follow symbolic links given in the command line, but skip symbolic links that are encountered recursively.
-R
--dereference-recursive
For each directory operand, read and process all files in that directory, recursively, following all symbolic links.
-s
--no-messages
Suppress error messages about nonexistent or unreadable files.
-T
--initial-tab
Make sure that the first character of actual line content lies on a tab stop, so that the alignment of tabs looks normal.
-U
--binary
Use binary I/O on platforms affected by the bug known as "text mode I/O". (MS-DOS, MS-Windows, OS/2).
-v
--invert-match
Select non-matching lines.
--verbose
Verbose mode. Show error messages. When specified before --version, print the version of the grep program used. Repeating it increases the verbosity level. See version.
-w
--word-regexp
Match only whole words.
-x
--line-regexp
Match only whole lines.
-Z
--null
Output a zero byte (the ASCII NUL character) instead of the character that normally follows a file name. For example, 'zgrep -lZ' outputs a zero byte after each file name instead of the usual newline. This option makes the output unambiguous, even in the presence of file names containing unusual characters like newlines.

Next: , Previous: Zgrep, Up: Top

8 Ztest

ztest checks the integrity of the compressed files specified. It also warns if an uncompressed file has a compressed file name extension, or if a compressed file has a wrong compressed extension. Uncompressed files are otherwise ignored. If a file is specified as '-', the integrity of compressed data read from standard input is checked. Data read from standard input must be all in the same compressed format. If a file fails to decompress, does not exist, can't be opened, or is a terminal, ztest continues testing the rest of the files. A final diagnostic is shown at verbosity level 1 or higher if any file fails the test when testing multiple files.

If no files are specified, recursive searches examine the current working directory, and nonrecursive searches read standard input.

Bzip2, gzip, and lzip are the primary formats. Xz and zstd are optional. If the decompressor for the xz or zstd formats is not found, the corresponding files are ignored.

Note that error detection in the xz format is broken. First, some xz files lack integrity information. Second, not all xz decompressors can check the integrity of all xz files. Third, section 2.1.1.2 'Stream Flags' of the xz format specification allows xz decompressors to produce garbage output without issuing any warning. Therefore, xz files can't always be checked as reliably as files in the other formats can.

The format for running ztest is: 

     ztest [options] [files]

Exit status is 0 if all compressed files check OK, 1 if environmental problems (file not found, invalid command-line options, I/O errors, etc), 2 if any compressed file is corrupt or invalid, or if any file has an incorrect file name extension.

ztest supports the following options:

-O format
--force-format=format
Force the compressed format given. Valid values for format are 'bz2', 'gz', 'lz', 'xz', and 'zst'. If this option is used, the files are passed to the corresponding decompressor without checking their format, and any files in a format that the decompressor can't understand fail the test.
-q
--quiet
Quiet operation. Suppress all messages.
-r
--recursive
For each directory operand, read and process all files in that directory, recursively. Follow symbolic links given in the command line, but skip symbolic links that are encountered recursively.
-R
--dereference-recursive
For each directory operand, read and process all files in that directory, recursively, following all symbolic links.
-v
--verbose
Verbose mode. Show the check status for each file processed. Further -v's increase the verbosity level. See version.

Next: , Previous: Ztest, Up: Top

9 Zupdate

zupdate recompresses files from bzip2, gzip, xz, and zstd formats to lzip format. Each original is compared with the new file and then deleted. Only regular files with standard file name extensions are recompressed, other files are ignored. Compressed files are decompressed and then recompressed on the fly; no temporary files are created. If an error happens while recompressing a file, zupdate exits immediately without recompressing the rest of the files. The lzip format is chosen as destination because it is the most appropriate for long-term archiving.

If no files are specified, recursive searches examine the current working directory, and nonrecursive searches do nothing.

If the lzip-compressed version of a file already exists, the file is skipped unless the option --force is given. In this case, if the comparison with the existing lzip version fails, an error is returned and the original file is not deleted. The operation of zupdate is meant to be safe and not cause any data loss. Therefore, existing lzip-compressed files are never overwritten nor deleted.

Combining the options --force and --keep, as in 'zupdate -f -k *.gz', checks that there are no differences between each pair of files in a multiformat set of files.

The names of the original files must have one of the following extensions:
'.bz2', '.gz', '.xz', '.zst', or '.Z', which are recompressed to '.lz';
'.tbz', '.tbz2', '.tgz', '.txz', or '.tzst', which are recompressed to '.tlz'.
Keeping the combined extensions ('.tgz' --> '.tlz') may be useful when recompressing Slackware packages, for example.

Bzip2, gzip, and lzip are the primary formats. Xz and zstd are optional. If the decompressor for the xz or zstd formats is not found, the corresponding files are ignored.

Recompressing a file is much like copying or moving it. Therefore zupdate preserves the access and modification dates, permissions, and, if you have appropriate privileges, ownership of the file just as 'cp -p' does. (If the user ID or the group ID can't be duplicated, the file permission bits S_ISUID and S_ISGID are cleared).

The format for running zupdate is: 

     zupdate [options] [files]

Exit status is 0 if all the compressed files were successfully recompressed (if needed), compared, and deleted (if requested). 1 if a non-fatal error occurred (file not found or not regular, or has invalid format, or can't be deleted). 2 if a fatal error occurred (invalid command-line options, compressor can't be run, or comparison fails).

zupdate supports the following options:

-d dir
--destdir=dir
Write recompressed files to another directory, using dir as base directory, instead of writing them in the same directory as the original files. In recursive mode, this is done by replacing each directory specified in the command line with dir to produce the recompressed file names. For example, 'zupdate -r -d dir ../a' recompresses a file named ../a/b/c.gz to dir/b/c.lz. Regular files specified in the command line are recompressed directly into dir. For example, 'zupdate -d dir ../a/b/c.gz' writes the recompressed file to dir/c.lz.

This option allows recompressing files from a read-only file system to another place without the need to copy or link them to the destination directory first. (Remember to use option --keep when recompressing read-only files to avoid warnings about files that can't be deleted).

-e
--expand-extensions
Expand combined file name extensions; recompress '.tbz', '.tbz2', '.tgz', '.txz', and '.tzst' to 'tar.lz'.
-f
--force
Don't skip a file for which a lzip-compressed version already exists. --force compares the content of the input file with the content of the existing lzip file and deletes the input file if both contents are identical.
-i
--ignore-errors
Ignore non-fatal errors. (See exit status above).
-k
--keep
Keep (don't delete) the input file after comparing it with the lzip file. Use it when recompressing files from a read-only file system. (See option --destdir above).
-l
--lzip-verbose
Pass one option -v to the lzip compressor so that it shows the compression ratio for each file processed. Using lzip 1.15 or newer, a second -l shows the progress of compression. Use it together with -v to see the name of the file.
-q
--quiet
Quiet operation. Suppress all messages.
-r
--recursive
For each directory operand, read and process all files in that directory, recursively. Follow symbolic links given in the command line, but skip symbolic links that are encountered recursively.
-R
--dereference-recursive
For each directory operand, read and process all files in that directory, recursively, following all symbolic links.
-v
--verbose
Verbose mode. Show the files being processed. A second -v also shows the files being ignored and increases the verbosity level. See version.
-0 .. -9
Set the compression level of lzip. By default zupdate passes -9 to lzip. Custom compression options can be passed to lzip with the option --lz. For example --lz='lzip -9 -s64MiB'.


--lz=command
Set compression command. command may include arguments. For example --lz='plzip --threads=2'. The name of the program can't begin with '-'. This option overrides the value set in zutils.conf. The compression program used does not need to implement decompression (see compressor-requirements), but it must implement at least the compression level option -9 and the option -o file to write the compressed output to file. tarlz meets these requirements, and therefore can be used to recompress POSIX tar archives by using a command like 'zupdate --lz='tarlz -9 -z --no-solid' archive.tar.gz'.

Next: , Previous: Zupdate, Up: Top

10 Reporting bugs

There are probably bugs in zutils. There are certainly errors and omissions in this manual. If you report them, they will get fixed. If you don't, no one will ever know about them and they will remain unfixed for all eternity, if not longer.

If you find a bug in zutils, please send electronic mail to zutils-bug@nongnu.org. Include the version number, which you can find by running 'zupdate --version'.

Previous: Problems, Up: Top

Concept index