3 @setfilename coreutils.info
4 @settitle @sc{gnu} Coreutils
9 @include constants.texi
11 @c Define new indices.
15 @c Put everything in one index (arbitrarily chosen to be the concept index).
25 * Coreutils: (coreutils). Core GNU (file, text, shell) utilities.
26 * Common options: (coreutils)Common options. Common options.
27 * File permissions: (coreutils)File permissions. Access modes.
28 * Date input formats: (coreutils)Date input formats.
31 @c FIXME: the following need documentation
32 @c * [: (coreutils)[ invocation. File/string tests.
33 @c * pinky: (coreutils)pinky invocation. FIXME.
35 @dircategory Individual utilities
37 * arch: (coreutils)arch invocation. Print machine hardware name.
38 * base64: (coreutils)base64 invocation. Base64 encode/decode data.
39 * basename: (coreutils)basename invocation. Strip directory and suffix.
40 * cat: (coreutils)cat invocation. Concatenate and write files.
41 * chcon: (coreutils)chcon invocation. Change SELinux CTX of files.
42 * chgrp: (coreutils)chgrp invocation. Change file groups.
43 * chmod: (coreutils)chmod invocation. Change file permissions.
44 * chown: (coreutils)chown invocation. Change file owners/groups.
45 * chroot: (coreutils)chroot invocation. Specify the root directory.
46 * cksum: (coreutils)cksum invocation. Print POSIX CRC checksum.
47 * comm: (coreutils)comm invocation. Compare sorted files by line.
48 * cp: (coreutils)cp invocation. Copy files.
49 * csplit: (coreutils)csplit invocation. Split by context.
50 * cut: (coreutils)cut invocation. Print selected parts of lines.
51 * date: (coreutils)date invocation. Print/set system date and time.
52 * dd: (coreutils)dd invocation. Copy and convert a file.
53 * df: (coreutils)df invocation. Report file system disk usage.
54 * dir: (coreutils)dir invocation. List directories briefly.
55 * dircolors: (coreutils)dircolors invocation. Color setup for ls.
56 * dirname: (coreutils)dirname invocation. Strip last file name component.
57 * du: (coreutils)du invocation. Report on disk usage.
58 * echo: (coreutils)echo invocation. Print a line of text.
59 * env: (coreutils)env invocation. Modify the environment.
60 * expand: (coreutils)expand invocation. Convert tabs to spaces.
61 * expr: (coreutils)expr invocation. Evaluate expressions.
62 * factor: (coreutils)factor invocation. Print prime factors
63 * false: (coreutils)false invocation. Do nothing, unsuccessfully.
64 * fmt: (coreutils)fmt invocation. Reformat paragraph text.
65 * fold: (coreutils)fold invocation. Wrap long input lines.
66 * groups: (coreutils)groups invocation. Print group names a user is in.
67 * head: (coreutils)head invocation. Output the first part of files.
68 * hostid: (coreutils)hostid invocation. Print numeric host identifier.
69 * hostname: (coreutils)hostname invocation. Print or set system name.
70 * id: (coreutils)id invocation. Print user identity.
71 * install: (coreutils)install invocation. Copy and change attributes.
72 * join: (coreutils)join invocation. Join lines on a common field.
73 * kill: (coreutils)kill invocation. Send a signal to processes.
74 * link: (coreutils)link invocation. Make hard links between files.
75 * ln: (coreutils)ln invocation. Make links between files.
76 * logname: (coreutils)logname invocation. Print current login name.
77 * ls: (coreutils)ls invocation. List directory contents.
78 * md5sum: (coreutils)md5sum invocation. Print or check MD5 digests.
79 * mkdir: (coreutils)mkdir invocation. Create directories.
80 * mkfifo: (coreutils)mkfifo invocation. Create FIFOs (named pipes).
81 * mknod: (coreutils)mknod invocation. Create special files.
82 * mktemp: (coreutils)mktemp invocation. Create temporary files.
83 * mv: (coreutils)mv invocation. Rename files.
84 * nice: (coreutils)nice invocation. Modify niceness.
85 * nl: (coreutils)nl invocation. Number lines and write files.
86 * nohup: (coreutils)nohup invocation. Immunize to hangups.
87 * nproc: (coreutils)nproc invocation. Print the number of processors.
88 * od: (coreutils)od invocation. Dump files in octal, etc.
89 * paste: (coreutils)paste invocation. Merge lines of files.
90 * pathchk: (coreutils)pathchk invocation. Check file name portability.
91 * pr: (coreutils)pr invocation. Paginate or columnate files.
92 * printenv: (coreutils)printenv invocation. Print environment variables.
93 * printf: (coreutils)printf invocation. Format and print data.
94 * ptx: (coreutils)ptx invocation. Produce permuted indexes.
95 * pwd: (coreutils)pwd invocation. Print working directory.
96 * readlink: (coreutils)readlink invocation. Print referent of a symlink.
97 * realpath: (coreutils)readpath invocation. Print resolved file names.
98 * rm: (coreutils)rm invocation. Remove files.
99 * rmdir: (coreutils)rmdir invocation. Remove empty directories.
100 * runcon: (coreutils)runcon invocation. Run in specified SELinux CTX.
101 * seq: (coreutils)seq invocation. Print numeric sequences
102 * sha1sum: (coreutils)sha1sum invocation. Print or check SHA-1 digests.
103 * sha2: (coreutils)sha2 utilities. Print or check SHA-2 digests.
104 * shred: (coreutils)shred invocation. Remove files more securely.
105 * shuf: (coreutils)shuf invocation. Shuffling text files.
106 * sleep: (coreutils)sleep invocation. Delay for a specified time.
107 * sort: (coreutils)sort invocation. Sort text files.
108 * split: (coreutils)split invocation. Split into pieces.
109 * stat: (coreutils)stat invocation. Report file(system) status.
110 * stdbuf: (coreutils)stdbuf invocation. Modify stdio buffering.
111 * stty: (coreutils)stty invocation. Print/change terminal settings.
112 * su: (coreutils)su invocation. Modify user and group ID.
113 * sum: (coreutils)sum invocation. Print traditional checksum.
114 * sync: (coreutils)sync invocation. Synchronize memory and disk.
115 * tac: (coreutils)tac invocation. Reverse files.
116 * tail: (coreutils)tail invocation. Output the last part of files.
117 * tee: (coreutils)tee invocation. Redirect to multiple files.
118 * test: (coreutils)test invocation. File/string tests.
119 * timeout: (coreutils)timeout invocation. Run with time limit.
120 * touch: (coreutils)touch invocation. Change file timestamps.
121 * tr: (coreutils)tr invocation. Translate characters.
122 * true: (coreutils)true invocation. Do nothing, successfully.
123 * truncate: (coreutils)truncate invocation. Shrink/extend size of a file.
124 * tsort: (coreutils)tsort invocation. Topological sort.
125 * tty: (coreutils)tty invocation. Print terminal name.
126 * uname: (coreutils)uname invocation. Print system information.
127 * unexpand: (coreutils)unexpand invocation. Convert spaces to tabs.
128 * uniq: (coreutils)uniq invocation. Uniquify files.
129 * unlink: (coreutils)unlink invocation. Removal via unlink(2).
130 * uptime: (coreutils)uptime invocation. Print uptime and load.
131 * users: (coreutils)users invocation. Print current user names.
132 * vdir: (coreutils)vdir invocation. List directories verbosely.
133 * wc: (coreutils)wc invocation. Line, word, and byte counts.
134 * who: (coreutils)who invocation. Print who is logged in.
135 * whoami: (coreutils)whoami invocation. Print effective user ID.
136 * yes: (coreutils)yes invocation. Print a string indefinitely.
140 This manual documents version @value{VERSION} of the @sc{gnu} core
141 utilities, including the standard programs for text and file manipulation.
143 Copyright @copyright{} 1994-2012 Free Software Foundation, Inc.
146 Permission is granted to copy, distribute and/or modify this document
147 under the terms of the GNU Free Documentation License, Version 1.3 or
148 any later version published by the Free Software Foundation; with no
149 Invariant Sections, with no Front-Cover Texts, and with no Back-Cover
150 Texts. A copy of the license is included in the section entitled ``GNU
151 Free Documentation License''.
156 @title @sc{gnu} @code{Coreutils}
157 @subtitle Core GNU utilities
158 @subtitle for version @value{VERSION}, @value{UPDATED}
159 @author David MacKenzie et al.
162 @vskip 0pt plus 1filll
175 @cindex core utilities
176 @cindex text utilities
177 @cindex shell utilities
178 @cindex file utilities
181 * Introduction:: Caveats, overview, and authors
182 * Common options:: Common options
183 * Output of entire files:: cat tac nl od base64
184 * Formatting file contents:: fmt pr fold
185 * Output of parts of files:: head tail split csplit
186 * Summarizing files:: wc sum cksum md5sum sha1sum sha2
187 * Operating on sorted files:: sort shuf uniq comm ptx tsort
188 * Operating on fields:: cut paste join
189 * Operating on characters:: tr expand unexpand
190 * Directory listing:: ls dir vdir dircolors
191 * Basic operations:: cp dd install mv rm shred
192 * Special file types:: mkdir rmdir unlink mkfifo mknod ln link readlink
193 * Changing file attributes:: chgrp chmod chown touch
194 * Disk usage:: df du stat sync truncate
195 * Printing text:: echo printf yes
196 * Conditions:: false true test expr
198 * File name manipulation:: dirname basename pathchk mktemp realpath
199 * Working context:: pwd stty printenv tty
200 * User information:: id logname whoami groups users who
201 * System context:: date arch nproc uname hostname hostid uptime
202 * SELinux context:: chcon runcon
203 * Modified command invocation:: chroot env nice nohup stdbuf su timeout
204 * Process control:: kill
206 * Numeric operations:: factor seq
207 * File permissions:: Access modes
208 * Date input formats:: Specifying date strings
209 * Opening the software toolbox:: The software tools philosophy
210 * GNU Free Documentation License:: Copying and sharing this manual
211 * Concept index:: General index
214 --- The Detailed Node Listing ---
218 * Exit status:: Indicating program success or failure
219 * Backup options:: Backup options
220 * Block size:: Block size
221 * Floating point:: Floating point number representation
222 * Signal specifications:: Specifying signals
223 * Disambiguating names and IDs:: chgrp and chown owner and group syntax
224 * Random sources:: Sources of random data
225 * Target directory:: Target directory
226 * Trailing slashes:: Trailing slashes
227 * Traversing symlinks:: Traversing symlinks to directories
228 * Treating / specially:: Treating / specially
229 * Standards conformance:: Standards conformance
231 Output of entire files
233 * cat invocation:: Concatenate and write files
234 * tac invocation:: Concatenate and write files in reverse
235 * nl invocation:: Number lines and write files
236 * od invocation:: Write files in octal or other formats
237 * base64 invocation:: Transform data into printable data
239 Formatting file contents
241 * fmt invocation:: Reformat paragraph text
242 * pr invocation:: Paginate or columnate files for printing
243 * fold invocation:: Wrap input lines to fit in specified width
245 Output of parts of files
247 * head invocation:: Output the first part of files
248 * tail invocation:: Output the last part of files
249 * split invocation:: Split a file into fixed-size pieces
250 * csplit invocation:: Split a file into context-determined pieces
254 * wc invocation:: Print newline, word, and byte counts
255 * sum invocation:: Print checksum and block counts
256 * cksum invocation:: Print CRC checksum and byte counts
257 * md5sum invocation:: Print or check MD5 digests
258 * sha1sum invocation:: Print or check SHA-1 digests
259 * sha2 utilities:: Print or check SHA-2 digests
261 Operating on sorted files
263 * sort invocation:: Sort text files
264 * shuf invocation:: Shuffle text files
265 * uniq invocation:: Uniquify files
266 * comm invocation:: Compare two sorted files line by line
267 * ptx invocation:: Produce a permuted index of file contents
268 * tsort invocation:: Topological sort
270 @command{ptx}: Produce permuted indexes
272 * General options in ptx:: Options which affect general program behavior
273 * Charset selection in ptx:: Underlying character set considerations
274 * Input processing in ptx:: Input fields, contexts, and keyword selection
275 * Output formatting in ptx:: Types of output format, and sizing the fields
276 * Compatibility in ptx:: The @acronym{GNU} extensions to @command{ptx}
280 * cut invocation:: Print selected parts of lines
281 * paste invocation:: Merge lines of files
282 * join invocation:: Join lines on a common field
284 Operating on characters
286 * tr invocation:: Translate, squeeze, and/or delete characters
287 * expand invocation:: Convert tabs to spaces
288 * unexpand invocation:: Convert spaces to tabs
290 @command{tr}: Translate, squeeze, and/or delete characters
292 * Character sets:: Specifying sets of characters
293 * Translating:: Changing one set of characters to another
294 * Squeezing:: Squeezing repeats and deleting
298 * ls invocation:: List directory contents
299 * dir invocation:: Briefly list directory contents
300 * vdir invocation:: Verbosely list directory contents
301 * dircolors invocation:: Color setup for @command{ls}
303 @command{ls}: List directory contents
305 * Which files are listed:: Which files are listed
306 * What information is listed:: What information is listed
307 * Sorting the output:: Sorting the output
308 * Details about version sort:: More details about version sort
309 * General output formatting:: General output formatting
310 * Formatting the file names:: Formatting the file names
314 * cp invocation:: Copy files and directories
315 * dd invocation:: Convert and copy a file
316 * install invocation:: Copy files and set attributes
317 * mv invocation:: Move (rename) files
318 * rm invocation:: Remove files or directories
319 * shred invocation:: Remove files more securely
323 * link invocation:: Make a hard link via the link syscall
324 * ln invocation:: Make links between files
325 * mkdir invocation:: Make directories
326 * mkfifo invocation:: Make FIFOs (named pipes)
327 * mknod invocation:: Make block or character special files
328 * readlink invocation:: Print value of a symlink or canonical file name
329 * rmdir invocation:: Remove empty directories
330 * unlink invocation:: Remove files via unlink syscall
332 Changing file attributes
334 * chown invocation:: Change file owner and group
335 * chgrp invocation:: Change group ownership
336 * chmod invocation:: Change access permissions
337 * touch invocation:: Change file timestamps
341 * df invocation:: Report file system disk space usage
342 * du invocation:: Estimate file space usage
343 * stat invocation:: Report file or file system status
344 * sync invocation:: Synchronize data on disk with memory
345 * truncate invocation:: Shrink or extend the size of a file
349 * echo invocation:: Print a line of text
350 * printf invocation:: Format and print data
351 * yes invocation:: Print a string until interrupted
355 * false invocation:: Do nothing, unsuccessfully
356 * true invocation:: Do nothing, successfully
357 * test invocation:: Check file types and compare values
358 * expr invocation:: Evaluate expressions
360 @command{test}: Check file types and compare values
362 * File type tests:: File type tests
363 * Access permission tests:: Access permission tests
364 * File characteristic tests:: File characteristic tests
365 * String tests:: String tests
366 * Numeric tests:: Numeric tests
368 @command{expr}: Evaluate expression
370 * String expressions:: + : match substr index length
371 * Numeric expressions:: + - * / %
372 * Relations for expr:: | & < <= = == != >= >
373 * Examples of expr:: Examples of using @command{expr}
377 * tee invocation:: Redirect output to multiple files or processes
379 File name manipulation
381 * basename invocation:: Strip directory and suffix from a file name
382 * dirname invocation:: Strip last file name component
383 * pathchk invocation:: Check file name validity and portability
384 * mktemp invocation:: Create temporary file or directory
385 * realpath invocation:: Print resolved file names
389 * pwd invocation:: Print working directory
390 * stty invocation:: Print or change terminal characteristics
391 * printenv invocation:: Print all or some environment variables
392 * tty invocation:: Print file name of terminal on standard input
394 @command{stty}: Print or change terminal characteristics
396 * Control:: Control settings
397 * Input:: Input settings
398 * Output:: Output settings
399 * Local:: Local settings
400 * Combination:: Combination settings
401 * Characters:: Special characters
402 * Special:: Special settings
406 * id invocation:: Print user identity
407 * logname invocation:: Print current login name
408 * whoami invocation:: Print effective user ID
409 * groups invocation:: Print group names a user is in
410 * users invocation:: Print login names of users currently logged in
411 * who invocation:: Print who is currently logged in
415 * arch invocation:: Print machine hardware name
416 * date invocation:: Print or set system date and time
417 * nproc invocation:: Print the number of processors
418 * uname invocation:: Print system information
419 * hostname invocation:: Print or set system name
420 * hostid invocation:: Print numeric host identifier
421 * uptime invocation:: Print system uptime and load
423 @command{date}: Print or set system date and time
425 * Time conversion specifiers:: %[HIklMNpPrRsSTXzZ]
426 * Date conversion specifiers:: %[aAbBcCdDeFgGhjmuUVwWxyY]
427 * Literal conversion specifiers:: %[%nt]
428 * Padding and other flags:: Pad with zeros, spaces, etc.
429 * Setting the time:: Changing the system clock
430 * Options for date:: Instead of the current time
431 * Date input formats:: Specifying date strings
432 * Examples of date:: Examples
436 * chcon invocation:: Change SELinux context of file
437 * runcon invocation:: Run a command in specified SELinux context
439 Modified command invocation
441 * chroot invocation:: Run a command with a different root directory
442 * env invocation:: Run a command in a modified environment
443 * nice invocation:: Run a command with modified niceness
444 * nohup invocation:: Run a command immune to hangups
445 * stdbuf invocation:: Run a command with modified I/O buffering
446 * su invocation:: Run a command with substitute user and group ID
447 * timeout invocation:: Run a command with a time limit
451 * kill invocation:: Sending a signal to processes.
455 * sleep invocation:: Delay for a specified time
459 * factor invocation:: Print prime factors
460 * seq invocation:: Print numeric sequences
464 * Mode Structure:: Structure of file mode bits
465 * Symbolic Modes:: Mnemonic representation of file mode bits
466 * Numeric Modes:: File mode bits as octal numbers
467 * Directory Setuid and Setgid:: Set-user-ID and set-group-ID on directories
471 * General date syntax:: Common rules
472 * Calendar date items:: 19 Dec 1994
473 * Time of day items:: 9:20pm
474 * Time zone items:: @sc{est}, @sc{pdt}, @sc{gmt}
475 * Day of week items:: Monday and others
476 * Relative items in date strings:: next tuesday, 2 years ago
477 * Pure numbers in date strings:: 19931219, 1440
478 * Seconds since the Epoch:: @@1078100502
479 * Specifying time zone rules:: TZ="America/New_York", TZ="UTC0"
480 * Authors of parse_datetime:: Bellovin, Eggert, Salz, Berets, et al
482 Opening the software toolbox
484 * Toolbox introduction:: Toolbox introduction
485 * I/O redirection:: I/O redirection
486 * The who command:: The @command{who} command
487 * The cut command:: The @command{cut} command
488 * The sort command:: The @command{sort} command
489 * The uniq command:: The @command{uniq} command
490 * Putting the tools together:: Putting the tools together
494 * GNU Free Documentation License:: Copying and sharing this manual
501 @chapter Introduction
503 This manual is a work in progress: many sections make no attempt to explain
504 basic concepts in a way suitable for novices. Thus, if you are interested,
505 please get involved in improving this manual. The entire @sc{gnu} community
508 @cindex @acronym{POSIX}
509 The @sc{gnu} utilities documented here are mostly compatible with the
510 @acronym{POSIX} standard.
511 @cindex bugs, reporting
512 Please report bugs to @email{bug-coreutils@@gnu.org}. Remember
513 to include the version number, machine architecture, input files, and
514 any other information needed to reproduce the bug: your input, what you
515 expected, what you got, and why it is wrong. Diffs are welcome, but
516 please include a description of the problem as well, since this is
517 sometimes difficult to infer. @xref{Bugs, , , gcc, Using and Porting GNU CC}.
523 @cindex MacKenzie, D.
526 This manual was originally derived from the Unix man pages in the
527 distributions, which were written by David MacKenzie and updated by Jim
528 Meyering. What you are reading now is the authoritative documentation
529 for these utilities; the man pages are no longer being maintained. The
530 original @command{fmt} man page was written by Ross Paterson. Fran@,{c}ois
531 Pinard did the initial conversion to Texinfo format. Karl Berry did the
532 indexing, some reorganization, and editing of the results. Brian
533 Youmans of the Free Software Foundation office staff combined the
534 manuals for textutils, fileutils, and sh-utils to produce the present
535 omnibus manual. Richard Stallman contributed his usual invaluable
536 insights to the overall process.
539 @chapter Common options
543 @itemx @w{@kbd{--backup}[=@var{method}]}
546 @vindex VERSION_CONTROL
547 @cindex backups, making
548 @xref{Backup options}.
549 Make a backup of each file that would otherwise be overwritten or removed.
552 @macro optBackupSuffix
553 @item -S @var{suffix}
554 @itemx --suffix=@var{suffix}
557 Append @var{suffix} to each backup file made with @option{-b}.
558 @xref{Backup options}.
561 @macro optTargetDirectory
562 @item -t @var{directory}
563 @itemx @w{@kbd{--target-directory}=@var{directory}}
565 @opindex --target-directory
566 @cindex target directory
567 @cindex destination directory
568 Specify the destination @var{directory}.
569 @xref{Target directory}.
572 @macro optNoTargetDirectory
574 @itemx --no-target-directory
576 @opindex --no-target-directory
577 @cindex target directory
578 @cindex destination directory
579 Do not treat the last operand specially when it is a directory or a
580 symbolic link to a directory. @xref{Target directory}.
588 @cindex output @sc{nul}-byte-terminated lines
589 Output a zero byte (@acronym{ASCII} @sc{nul}) at the end of each line,
590 rather than a newline. This option enables other programs to parse the
591 output of @command{\cmd\} even when that output would contain data
592 with embedded newlines.
599 Append an SI-style abbreviation to each size, such as @samp{M} for
600 megabytes. Powers of 1000 are used, not 1024; @samp{M} stands for
601 1,000,000 bytes. This option is equivalent to
602 @option{--block-size=si}. Use the @option{-h} or
603 @option{--human-readable} option if
604 you prefer powers of 1024.
607 @macro optHumanReadable
609 @itemx --human-readable
611 @opindex --human-readable
612 @cindex human-readable output
613 Append a size letter to each size, such as @samp{M} for mebibytes.
614 Powers of 1024 are used, not 1000; @samp{M} stands for 1,048,576 bytes.
615 This option is equivalent to @option{--block-size=human-readable}.
616 Use the @option{--si} option if you prefer powers of 1000.
619 @macro optStripTrailingSlashes
620 @itemx @w{@kbd{--strip-trailing-slashes}}
621 @opindex --strip-trailing-slashes
622 @cindex stripping trailing slashes
623 Remove any trailing slashes from each @var{source} argument.
624 @xref{Trailing slashes}.
627 @macro mayConflictWithShellBuiltIn{cmd}
628 @cindex conflicts with shell built-ins
629 @cindex built-in shell commands, conflicts with
630 Due to shell aliases and built-in @command{\cmd\} functions, using an
631 unadorned @command{\cmd\} interactively or in a script may get you
632 different functionality than that described here. Invoke it via
633 @command{env} (i.e., @code{env \cmd\ @dots{}}) to avoid interference
638 @macro multiplierSuffixes{varName}
639 @var{\varName\} may be, or may be an integer optionally followed by,
640 one of the following multiplicative suffixes:
642 @samp{b} => 512 ("blocks")
643 @samp{KB} => 1000 (KiloBytes)
644 @samp{K} => 1024 (KibiBytes)
645 @samp{MB} => 1000*1000 (MegaBytes)
646 @samp{M} => 1024*1024 (MebiBytes)
647 @samp{GB} => 1000*1000*1000 (GigaBytes)
648 @samp{G} => 1024*1024*1024 (GibiBytes)
650 and so on for @samp{T}, @samp{P}, @samp{E}, @samp{Z}, and @samp{Y}.
653 @c FIXME: same as above, but no ``blocks'' line.
654 @macro multiplierSuffixesNoBlocks{varName}
655 @var{\varName\} may be, or may be an integer optionally followed by,
656 one of the following multiplicative suffixes:
658 @samp{KB} => 1000 (KiloBytes)
659 @samp{K} => 1024 (KibiBytes)
660 @samp{MB} => 1000*1000 (MegaBytes)
661 @samp{M} => 1024*1024 (MebiBytes)
662 @samp{GB} => 1000*1000*1000 (GigaBytes)
663 @samp{G} => 1024*1024*1024 (GibiBytes)
665 and so on for @samp{T}, @samp{P}, @samp{E}, @samp{Z}, and @samp{Y}.
668 @cindex common options
670 Certain options are available in all of these programs. Rather than
671 writing identical descriptions for each of the programs, they are
672 described here. (In fact, every @sc{gnu} program accepts (or should accept)
675 @vindex POSIXLY_CORRECT
676 Normally options and operands can appear in any order, and programs act
677 as if all the options appear before any operands. For example,
678 @samp{sort -r passwd -t :} acts like @samp{sort -r -t : passwd}, since
679 @samp{:} is an option-argument of @option{-t}. However, if the
680 @env{POSIXLY_CORRECT} environment variable is set, options must appear
681 before operands, unless otherwise specified for a particular command.
683 A few programs can usefully have trailing operands with leading
684 @samp{-}. With such a program, options must precede operands even if
685 @env{POSIXLY_CORRECT} is not set, and this fact is noted in the
686 program description. For example, the @command{env} command's options
687 must appear before its operands, since in some cases the operands
688 specify a command that itself contains options.
690 Most programs that accept long options recognize unambiguous
691 abbreviations of those options. For example, @samp{rmdir
692 --ignore-fail-on-non-empty} can be invoked as @samp{rmdir
693 --ignore-fail} or even @samp{rmdir --i}. Ambiguous options, such as
694 @samp{ls --h}, are identified as such.
696 Some of these programs recognize the @option{--help} and @option{--version}
697 options only when one of them is the sole command line argument. For
698 these programs, abbreviations of the long options are not always recognized.
705 Print a usage message listing all available options, then exit successfully.
709 @cindex version number, finding
710 Print the version number, then exit successfully.
714 @cindex option delimiter
715 Delimit the option list. Later arguments, if any, are treated as
716 operands even if they begin with @samp{-}. For example, @samp{sort --
717 -r} reads from the file named @file{-r}.
721 @cindex standard input
722 @cindex standard output
723 A single @samp{-} operand is not really an option, though it looks like one. It
724 stands for standard input, or for standard output if that is clear from
725 the context. For example, @samp{sort -} reads from standard input,
726 and is equivalent to plain @samp{sort}, and @samp{tee -} writes an
727 extra copy of its input to standard output. Unless otherwise
728 specified, @samp{-} can appear as any operand that requires a file
732 * Exit status:: Indicating program success or failure.
733 * Backup options:: -b -S, in some programs.
734 * Block size:: BLOCK_SIZE and --block-size, in some programs.
735 * Floating point:: Floating point number representation.
736 * Signal specifications:: Specifying signals using the --signal option.
737 * Disambiguating names and IDs:: chgrp and chown owner and group syntax
738 * Random sources:: --random-source, in some programs.
739 * Target directory:: Specifying a target directory, in some programs.
740 * Trailing slashes:: --strip-trailing-slashes, in some programs.
741 * Traversing symlinks:: -H, -L, or -P, in some programs.
742 * Treating / specially:: --preserve-root and --no-preserve-root.
743 * Special built-in utilities:: @command{break}, @command{:}, @dots{}
744 * Standards conformance:: Conformance to the @acronym{POSIX} standard.
752 An exit status of zero indicates success,
753 and a nonzero value indicates failure.
756 Nearly every command invocation yields an integral @dfn{exit status}
757 that can be used to change how other commands work.
758 For the vast majority of commands, an exit status of zero indicates
759 success. Failure is indicated by a nonzero value---typically
760 @samp{1}, though it may differ on unusual platforms as @acronym{POSIX}
761 requires only that it be nonzero.
763 However, some of the programs documented here do produce
764 other exit status values and a few associate different
765 meanings with the values @samp{0} and @samp{1}.
766 Here are some of the exceptions:
767 @command{chroot}, @command{env}, @command{expr}, @command{nice},
768 @command{nohup}, @command{printenv}, @command{sort}, @command{stdbuf},
769 @command{su}, @command{test}, @command{timeout}, @command{tty}.
773 @section Backup options
775 @cindex backup options
777 Some @sc{gnu} programs (at least @command{cp}, @command{install},
778 @command{ln}, and @command{mv}) optionally make backups of files
779 before writing new versions.
780 These options control the details of these backups. The options are also
781 briefly mentioned in the descriptions of the particular programs.
786 @itemx @w{@kbd{--backup}[=@var{method}]}
789 @vindex VERSION_CONTROL
790 @cindex backups, making
791 Make a backup of each file that would otherwise be overwritten or removed.
792 Without this option, the original versions are destroyed.
793 Use @var{method} to determine the type of backups to make.
794 When this option is used but @var{method} is not specified,
795 then the value of the @env{VERSION_CONTROL}
796 environment variable is used. And if @env{VERSION_CONTROL} is not set,
797 the default backup type is @samp{existing}.
799 Note that the short form of this option, @option{-b} does not accept any
800 argument. Using @option{-b} is equivalent to using @option{--backup=existing}.
802 @vindex version-control @r{Emacs variable}
803 This option corresponds to the Emacs variable @samp{version-control};
804 the values for @var{method} are the same as those used in Emacs.
805 This option also accepts more descriptive names.
806 The valid @var{method}s are (unique abbreviations are accepted):
811 @opindex none @r{backup method}
816 @opindex numbered @r{backup method}
817 Always make numbered backups.
821 @opindex existing @r{backup method}
822 Make numbered backups of files that already have them, simple backups
827 @opindex simple @r{backup method}
828 Always make simple backups. Please note @samp{never} is not to be
829 confused with @samp{none}.
833 @item -S @var{suffix}
834 @itemx --suffix=@var{suffix}
837 @cindex backup suffix
838 @vindex SIMPLE_BACKUP_SUFFIX
839 Append @var{suffix} to each backup file made with @option{-b}. If this
840 option is not specified, the value of the @env{SIMPLE_BACKUP_SUFFIX}
841 environment variable is used. And if @env{SIMPLE_BACKUP_SUFFIX} is not
842 set, the default is @samp{~}, just as in Emacs.
851 Some @sc{gnu} programs (at least @command{df}, @command{du}, and
852 @command{ls}) display sizes in ``blocks''. You can adjust the block size
853 and method of display to make sizes easier to read. The block size
854 used for display is independent of any file system block size.
855 Fractional block counts are rounded up to the nearest integer.
857 @opindex --block-size=@var{size}
860 @vindex DF_BLOCK_SIZE
861 @vindex DU_BLOCK_SIZE
862 @vindex LS_BLOCK_SIZE
863 @vindex POSIXLY_CORRECT@r{, and block size}
865 The default block size is chosen by examining the following environment
866 variables in turn; the first one that is set determines the block size.
871 This specifies the default block size for the @command{df} command.
872 Similarly, @env{DU_BLOCK_SIZE} specifies the default for @command{du} and
873 @env{LS_BLOCK_SIZE} for @command{ls}.
876 This specifies the default block size for all three commands, if the
877 above command-specific environment variables are not set.
880 This specifies the default block size for all values that are normally
881 printed as blocks, if neither @env{BLOCK_SIZE} nor the above
882 command-specific environment variables are set. Unlike the other
883 environment variables, @env{BLOCKSIZE} does not affect values that are
884 normally printed as byte counts, e.g., the file sizes contained in
887 @item POSIXLY_CORRECT
888 If neither @env{@var{command}_BLOCK_SIZE}, nor @env{BLOCK_SIZE}, nor
889 @env{BLOCKSIZE} is set, but this variable is set, the block size
894 If none of the above environment variables are set, the block size
895 currently defaults to 1024 bytes in most contexts, but this number may
896 change in the future. For @command{ls} file sizes, the block size
899 @cindex human-readable output
902 A block size specification can be a positive integer specifying the number
903 of bytes per block, or it can be @code{human-readable} or @code{si} to
904 select a human-readable format. Integers may be followed by suffixes
905 that are upward compatible with the
906 @uref{http://www.bipm.org/en/si/si_brochure/chapter3/prefixes.html, SI prefixes}
907 for decimal multiples and with the
908 @uref{http://physics.nist.gov/cuu/Units/binary.html, ISO/IEC 80000-13
909 (formerly IEC 60027-2) prefixes} for binary multiples.
911 With human-readable formats, output sizes are followed by a size letter
912 such as @samp{M} for megabytes. @code{BLOCK_SIZE=human-readable} uses
913 powers of 1024; @samp{M} stands for 1,048,576 bytes.
914 @code{BLOCK_SIZE=si} is similar, but uses powers of 1000 and appends
915 @samp{B}; @samp{MB} stands for 1,000,000 bytes.
918 A block size specification preceded by @samp{'} causes output sizes to
919 be displayed with thousands separators. The @env{LC_NUMERIC} locale
920 specifies the thousands separator and grouping. For example, in an
921 American English locale, @samp{--block-size="'1kB"} would cause a size
922 of 1234000 bytes to be displayed as @samp{1,234}. In the default C
923 locale, there is no thousands separator so a leading @samp{'} has no
926 An integer block size can be followed by a suffix to specify a
927 multiple of that size. A bare size letter,
928 or one followed by @samp{iB}, specifies
929 a multiple using powers of 1024. A size letter followed by @samp{B}
930 specifies powers of 1000 instead. For example, @samp{1M} and
931 @samp{1MiB} are equivalent to @samp{1048576}, whereas @samp{1MB} is
932 equivalent to @samp{1000000}.
934 A plain suffix without a preceding integer acts as if @samp{1} were
935 prepended, except that it causes a size indication to be appended to
936 the output. For example, @samp{--block-size="kB"} displays 3000 as
939 The following suffixes are defined. Large sizes like @code{1Y}
940 may be rejected by your computer due to limitations of its arithmetic.
944 @cindex kilobyte, definition of
945 kilobyte: @math{10^3 = 1000}.
949 @cindex kibibyte, definition of
950 kibibyte: @math{2^{10} = 1024}. @samp{K} is special: the SI prefix is
951 @samp{k} and the ISO/IEC 80000-13 prefix is @samp{Ki}, but tradition and
952 @acronym{POSIX} use @samp{k} to mean @samp{KiB}.
954 @cindex megabyte, definition of
955 megabyte: @math{10^6 = 1,000,000}.
958 @cindex mebibyte, definition of
959 mebibyte: @math{2^{20} = 1,048,576}.
961 @cindex gigabyte, definition of
962 gigabyte: @math{10^9 = 1,000,000,000}.
965 @cindex gibibyte, definition of
966 gibibyte: @math{2^{30} = 1,073,741,824}.
968 @cindex terabyte, definition of
969 terabyte: @math{10^{12} = 1,000,000,000,000}.
972 @cindex tebibyte, definition of
973 tebibyte: @math{2^{40} = 1,099,511,627,776}.
975 @cindex petabyte, definition of
976 petabyte: @math{10^{15} = 1,000,000,000,000,000}.
979 @cindex pebibyte, definition of
980 pebibyte: @math{2^{50} = 1,125,899,906,842,624}.
982 @cindex exabyte, definition of
983 exabyte: @math{10^{18} = 1,000,000,000,000,000,000}.
986 @cindex exbibyte, definition of
987 exbibyte: @math{2^{60} = 1,152,921,504,606,846,976}.
989 @cindex zettabyte, definition of
990 zettabyte: @math{10^{21} = 1,000,000,000,000,000,000,000}
993 @math{2^{70} = 1,180,591,620,717,411,303,424}.
995 @cindex yottabyte, definition of
996 yottabyte: @math{10^{24} = 1,000,000,000,000,000,000,000,000}.
999 @math{2^{80} = 1,208,925,819,614,629,174,706,176}.
1004 @opindex --block-size
1005 @opindex --human-readable
1008 Block size defaults can be overridden by an explicit
1009 @option{--block-size=@var{size}} option. The @option{-k}
1010 option is equivalent to @option{--block-size=1K}, which
1011 is the default unless the @env{POSIXLY_CORRECT} environment variable is
1012 set. The @option{-h} or @option{--human-readable} option is equivalent to
1013 @option{--block-size=human-readable}. The @option{--si} option is
1014 equivalent to @option{--block-size=si}.
1016 @node Floating point
1017 @section Floating point numbers
1018 @cindex floating point
1019 @cindex IEEE floating point
1021 Commands that accept or produce floating point numbers employ the
1022 floating point representation of the underlying system, and suffer
1023 from rounding error, overflow, and similar floating-point issues.
1024 Almost all modern systems use IEEE-754 floating point, and it is
1025 typically portable to assume IEEE-754 behavior these days. IEEE-754
1026 has positive and negative infinity, distinguishes positive from
1027 negative zero, and uses special values called NaNs to represent
1028 invalid computations such as dividing zero by itself. For more
1029 information, please see David Goldberg's paper
1030 @uref{http://@/www.validlab.com/@/goldberg/@/paper.pdf, What Every
1031 Computer Scientist Should Know About Floating-Point Arithmetic}.
1034 Commands that accept floating point numbers as options, operands or
1035 input use the standard C functions @code{strtod} and @code{strtold} to
1036 convert from text to floating point numbers. These floating point
1037 numbers therefore can use scientific notation like @code{1.0e-34} and
1038 @code{-10e100}. Modern C implementations also accept hexadecimal
1039 floating point numbers such as @code{-0x.ep-3}, which stands for
1040 @minus{}14/16 times @math{2^-3}, which equals @minus{}0.109375. The
1041 @env{LC_NUMERIC} locale determines the decimal-point character.
1042 @xref{Parsing of Floats,,, libc, The GNU C Library Reference Manual}.
1044 @node Signal specifications
1045 @section Signal specifications
1046 @cindex signals, specifying
1048 A @var{signal} may be a signal name like @samp{HUP}, or a signal
1049 number like @samp{1}, or an exit status of a process terminated by the
1050 signal. A signal name can be given in canonical form or prefixed by
1051 @samp{SIG}. The case of the letters is ignored. The following signal names
1052 and numbers are supported on all @acronym{POSIX} compliant systems:
1058 2. Terminal interrupt.
1064 9. Kill (cannot be caught or ignored).
1072 Other supported signal names have system-dependent corresponding
1073 numbers. All systems conforming to @acronym{POSIX} 1003.1-2001 also
1074 support the following signals:
1078 Access to an undefined portion of a memory object.
1080 Child process terminated, stopped, or continued.
1082 Continue executing, if stopped.
1084 Erroneous arithmetic operation.
1086 Illegal Instruction.
1088 Write on a pipe with no one to read it.
1090 Invalid memory reference.
1092 Stop executing (cannot be caught or ignored).
1096 Background process attempting read.
1098 Background process attempting write.
1100 High bandwidth data is available at a socket.
1102 User-defined signal 1.
1104 User-defined signal 2.
1108 @acronym{POSIX} 1003.1-2001 systems that support the @acronym{XSI} extension
1109 also support the following signals:
1115 Profiling timer expired.
1119 Trace/breakpoint trap.
1121 Virtual timer expired.
1123 CPU time limit exceeded.
1125 File size limit exceeded.
1129 @acronym{POSIX} 1003.1-2001 systems that support the @acronym{XRT} extension
1130 also support at least eight real-time signals called @samp{RTMIN},
1131 @samp{RTMIN+1}, @dots{}, @samp{RTMAX-1}, @samp{RTMAX}.
1133 @node Disambiguating names and IDs
1134 @section chown and chgrp: Disambiguating user names and IDs
1135 @cindex user names, disambiguating
1136 @cindex user IDs, disambiguating
1137 @cindex group names, disambiguating
1138 @cindex group IDs, disambiguating
1139 @cindex disambiguating group names and IDs
1141 Since the @var{owner} and @var{group} arguments to @command{chown} and
1142 @command{chgrp} may be specified as names or numeric IDs, there is an
1144 What if a user or group @emph{name} is a string of digits?
1145 @footnote{Using a number as a user name is common in some environments.}
1146 Should the command interpret it as a user name or as an ID?
1147 @acronym{POSIX} requires that @command{chown} and @command{chgrp}
1148 first attempt to resolve the specified string as a name, and
1149 only once that fails, then try to interpret it as an ID.
1150 This is troublesome when you want to specify a numeric ID, say 42,
1151 and it must work even in a pathological situation where
1152 @samp{42} is a user name that maps to some other user ID, say 1000.
1153 Simply invoking @code{chown 42 F}, will set @file{F}s owner ID to
1154 1000---not what you intended.
1156 GNU @command{chown} and @command{chgrp} provide a way to work around this,
1157 that at the same time may result in a significant performance improvement
1158 by eliminating a database look-up.
1159 Simply precede each numeric user ID and/or group ID with a @samp{+},
1160 in order to force its interpretation as an integer:
1164 chgrp +$numeric_group_id another-file
1168 GNU @command{chown} and @command{chgrp}
1169 skip the name look-up process for each @samp{+}-prefixed string,
1170 because a string containing @samp{+} is never a valid user or group name.
1171 This syntax is accepted on most common Unix systems, but not on Solaris 10.
1173 @node Random sources
1174 @section Sources of random data
1176 @cindex random sources
1178 The @command{shuf}, @command{shred}, and @command{sort} commands
1179 sometimes need random data to do their work. For example, @samp{sort
1180 -R} must choose a hash function at random, and it needs random data to
1181 make this selection.
1183 By default these commands use an internal pseudorandom generator
1184 initialized by a small amount of entropy, but can be directed to use
1185 an external source with the @option{--random-source=@var{file}} option.
1186 An error is reported if @var{file} does not contain enough bytes.
1188 For example, the device file @file{/dev/urandom} could be used as the
1189 source of random data. Typically, this device gathers environmental
1190 noise from device drivers and other sources into an entropy pool, and
1191 uses the pool to generate random bits. If the pool is short of data,
1192 the device reuses the internal pool to produce more bits, using a
1193 cryptographically secure pseudorandom number generator. But be aware
1194 that this device is not designed for bulk random data generation
1195 and is relatively slow.
1197 @file{/dev/urandom} suffices for most practical uses, but applications
1198 requiring high-value or long-term protection of private data may
1199 require an alternate data source like @file{/dev/random} or
1200 @file{/dev/arandom}. The set of available sources depends on your
1203 To reproduce the results of an earlier invocation of a command, you
1204 can save some random data into a file and then use that file as the
1205 random source in earlier and later invocations of the command.
1207 @node Target directory
1208 @section Target directory
1210 @cindex target directory
1212 The @command{cp}, @command{install}, @command{ln}, and @command{mv}
1213 commands normally treat the last operand specially when it is a
1214 directory or a symbolic link to a directory. For example, @samp{cp
1215 source dest} is equivalent to @samp{cp source dest/source} if
1216 @file{dest} is a directory. Sometimes this behavior is not exactly
1217 what is wanted, so these commands support the following options to
1218 allow more fine-grained control:
1223 @itemx --no-target-directory
1224 @opindex --no-target-directory
1225 @cindex target directory
1226 @cindex destination directory
1227 Do not treat the last operand specially when it is a directory or a
1228 symbolic link to a directory. This can help avoid race conditions in
1229 programs that operate in a shared area. For example, when the command
1230 @samp{mv /tmp/source /tmp/dest} succeeds, there is no guarantee that
1231 @file{/tmp/source} was renamed to @file{/tmp/dest}: it could have been
1232 renamed to @file{/tmp/dest/source} instead, if some other process
1233 created @file{/tmp/dest} as a directory. However, if @file{mv
1234 -T /tmp/source /tmp/dest} succeeds, there is no
1235 question that @file{/tmp/source} was renamed to @file{/tmp/dest}.
1237 In the opposite situation, where you want the last operand to be
1238 treated as a directory and want a diagnostic otherwise, you can use
1239 the @option{--target-directory} (@option{-t}) option.
1241 @item -t @var{directory}
1242 @itemx @w{@kbd{--target-directory}=@var{directory}}
1243 @opindex --target-directory
1244 @cindex target directory
1245 @cindex destination directory
1246 Use @var{directory} as the directory component of each destination
1249 The interface for most programs is that after processing options and a
1250 finite (possibly zero) number of fixed-position arguments, the remaining
1251 argument list is either expected to be empty, or is a list of items
1252 (usually files) that will all be handled identically. The @command{xargs}
1253 program is designed to work well with this convention.
1255 The commands in the @command{mv}-family are unusual in that they take
1256 a variable number of arguments with a special case at the @emph{end}
1257 (namely, the target directory). This makes it nontrivial to perform some
1258 operations, e.g., ``move all files from here to ../d/'', because
1259 @code{mv * ../d/} might exhaust the argument space, and @code{ls | xargs ...}
1260 doesn't have a clean way to specify an extra final argument for each
1261 invocation of the subject command. (It can be done by going through a
1262 shell command, but that requires more human labor and brain power than
1265 The @w{@kbd{--target-directory}} (@option{-t}) option allows the @command{cp},
1266 @command{install}, @command{ln}, and @command{mv} programs to be used
1267 conveniently with @command{xargs}. For example, you can move the files
1268 from the current directory to a sibling directory, @code{d} like this:
1271 ls | xargs mv -t ../d --
1274 However, this doesn't move files whose names begin with @samp{.}.
1275 If you use the @sc{gnu} @command{find} program, you can move those
1276 files too, with this command:
1279 find . -mindepth 1 -maxdepth 1 \
1283 But both of the above approaches fail if there are no files in the
1284 current directory, or if any file has a name containing a blank or
1285 some other special characters.
1286 The following example removes those limitations and requires both
1287 @sc{gnu} @command{find} and @sc{gnu} @command{xargs}:
1290 find . -mindepth 1 -maxdepth 1 -print0 \
1291 | xargs --null --no-run-if-empty \
1298 The @option{--target-directory} (@option{-t}) and
1299 @option{--no-target-directory} (@option{-T})
1300 options cannot be combined.
1302 @node Trailing slashes
1303 @section Trailing slashes
1305 @cindex trailing slashes
1307 Some @sc{gnu} programs (at least @command{cp} and @command{mv}) allow you to
1308 remove any trailing slashes from each @var{source} argument before
1309 operating on it. The @w{@kbd{--strip-trailing-slashes}} option enables
1312 This is useful when a @var{source} argument may have a trailing slash and
1313 @c FIXME: mv's behavior in this case is system-dependent
1314 specify a symbolic link to a directory. This scenario is in fact rather
1315 common because some shells can automatically append a trailing slash when
1316 performing file name completion on such symbolic links. Without this
1317 option, @command{mv}, for example, (via the system's rename function) must
1318 interpret a trailing slash as a request to dereference the symbolic link
1319 and so must rename the indirectly referenced @emph{directory} and not
1320 the symbolic link. Although it may seem surprising that such behavior
1321 be the default, it is required by @acronym{POSIX} and is consistent with
1322 other parts of that standard.
1324 @node Traversing symlinks
1325 @section Traversing symlinks
1327 @cindex symbolic link to directory, controlling traversal of
1329 The following options modify how @command{chown} and @command{chgrp}
1330 @c FIXME: note that 'du' has these options, too, but they have slightly
1331 @c different meaning.
1332 traverse a hierarchy when the @option{--recursive} (@option{-R})
1333 option is also specified.
1334 If more than one of the following options is specified, only the final
1336 These options specify whether processing a symbolic link to a directory
1337 entails operating on just the symbolic link or on all files in the
1338 hierarchy rooted at that directory.
1340 These options are independent of @option{--dereference} and
1341 @option{--no-dereference} (@option{-h}), which control whether to modify
1342 a symlink or its referent.
1349 @cindex symbolic link to directory, traverse if on the command line
1350 If @option{--recursive} (@option{-R}) is specified and
1351 a command line argument is a symbolic link to a directory, traverse it.
1358 @cindex symbolic link to directory, traverse each that is encountered
1359 In a recursive traversal, traverse every symbolic link to a directory
1360 that is encountered.
1367 @cindex symbolic link to directory, never traverse
1368 Do not traverse any symbolic links.
1369 This is the default if none of @option{-H}, @option{-L},
1370 or @option{-P} is specified.
1377 @node Treating / specially
1378 @section Treating @file{/} specially
1380 Certain commands can operate destructively on entire hierarchies.
1381 For example, if a user with appropriate privileges mistakenly runs
1382 @samp{rm -rf / tmp/junk}, that may remove
1383 all files on the entire system. Since there are so few
1384 legitimate uses for such a command,
1385 @sc{gnu} @command{rm} normally declines to operate on any directory
1386 that resolves to @file{/}. If you really want to try to remove all
1387 the files on your system, you can use the @option{--no-preserve-root}
1388 option, but the default behavior, specified by the
1389 @option{--preserve-option}, is safer for most purposes.
1391 The commands @command{chgrp}, @command{chmod} and @command{chown}
1392 can also operate destructively on entire hierarchies, so they too
1393 support these options. Although, unlike @command{rm}, they don't
1394 actually unlink files, these commands are arguably more dangerous
1395 when operating recursively on @file{/}, since they often work much
1396 more quickly, and hence damage more files before an alert user can
1397 interrupt them. Tradition and @acronym{POSIX} require these commands
1398 to operate recursively on @file{/}, so they default to
1399 @option{--no-preserve-root}, but using the @option{--preserve-root}
1400 option makes them safer for most purposes. For convenience you can
1401 specify @option{--preserve-root} in an alias or in a shell function.
1403 Note that the @option{--preserve-root} option also ensures
1404 that @command{chgrp} and @command{chown} do not modify @file{/}
1405 even when dereferencing a symlink pointing to @file{/}.
1407 @node Special built-in utilities
1408 @section Special built-in utilities
1410 Some programs like @command{nice} can invoke other programs; for
1411 example, the command @samp{nice cat file} invokes the program
1412 @command{cat} by executing the command @samp{cat file}. However,
1413 @dfn{special built-in utilities} like @command{exit} cannot be invoked
1414 this way. For example, the command @samp{nice exit} does not have a
1415 well-defined behavior: it may generate an error message instead of
1418 Here is a list of the special built-in utilities that are standardized
1419 by @acronym{POSIX} 1003.1-2004.
1422 @t{.@: : break continue eval exec exit export readonly
1423 return set shift times trap unset}
1426 For example, because @samp{.}, @samp{:}, and @samp{exec} are special,
1427 the commands @samp{nice . foo.sh}, @samp{nice :}, and @samp{nice exec
1428 pwd} do not work as you might expect.
1430 Many shells extend this list. For example, Bash has several extra
1431 special built-in utilities like @command{history}, and
1432 @command{suspend}, and with Bash the command @samp{nice suspend}
1433 generates an error message instead of suspending.
1435 @node Standards conformance
1436 @section Standards conformance
1438 @vindex POSIXLY_CORRECT
1439 In a few cases, the @sc{gnu} utilities' default behavior is
1440 incompatible with the @acronym{POSIX} standard. To suppress these
1441 incompatibilities, define the @env{POSIXLY_CORRECT} environment
1442 variable. Unless you are checking for @acronym{POSIX} conformance, you
1443 probably do not need to define @env{POSIXLY_CORRECT}.
1445 Newer versions of @acronym{POSIX} are occasionally incompatible with older
1446 versions. For example, older versions of @acronym{POSIX} required the
1447 command @samp{sort +1} to sort based on the second and succeeding
1448 fields in each input line, but starting with @acronym{POSIX} 1003.1-2001
1449 the same command is required to sort the file named @file{+1}, and you
1450 must instead use the command @samp{sort -k 2} to get the field-based
1453 @vindex _POSIX2_VERSION
1454 The @sc{gnu} utilities normally conform to the version of @acronym{POSIX}
1455 that is standard for your system. To cause them to conform to a
1456 different version of @acronym{POSIX}, define the @env{_POSIX2_VERSION}
1457 environment variable to a value of the form @var{yyyymm} specifying
1458 the year and month the standard was adopted. Three values are currently
1459 supported for @env{_POSIX2_VERSION}: @samp{199209} stands for
1460 @acronym{POSIX} 1003.2-1992, @samp{200112} stands for @acronym{POSIX}
1461 1003.1-2001, and @samp{200809} stands for @acronym{POSIX} 1003.1-2008.
1462 For example, if you have a newer system but are running software
1463 that assumes an older version of @acronym{POSIX} and uses @samp{sort +1}
1464 or @samp{tail +10}, you can work around any compatibility problems by setting
1465 @samp{_POSIX2_VERSION=199209} in your environment.
1467 @node Output of entire files
1468 @chapter Output of entire files
1470 @cindex output of entire files
1471 @cindex entire files, output of
1473 These commands read and write entire files, possibly transforming them
1477 * cat invocation:: Concatenate and write files.
1478 * tac invocation:: Concatenate and write files in reverse.
1479 * nl invocation:: Number lines and write files.
1480 * od invocation:: Write files in octal or other formats.
1481 * base64 invocation:: Transform data into printable data.
1484 @node cat invocation
1485 @section @command{cat}: Concatenate and write files
1488 @cindex concatenate and write files
1489 @cindex copying files
1491 @command{cat} copies each @var{file} (@samp{-} means standard input), or
1492 standard input if none are given, to standard output. Synopsis:
1495 cat [@var{option}] [@var{file}]@dots{}
1498 The program accepts the following options. Also see @ref{Common options}.
1506 Equivalent to @option{-vET}.
1509 @itemx --number-nonblank
1511 @opindex --number-nonblank
1512 Number all nonempty output lines, starting with 1.
1516 Equivalent to @option{-vE}.
1521 @opindex --show-ends
1522 Display a @samp{$} after the end of each line.
1528 Number all output lines, starting with 1. This option is ignored
1529 if @option{-b} is in effect.
1532 @itemx --squeeze-blank
1534 @opindex --squeeze-blank
1535 @cindex squeezing empty lines
1536 Suppress repeated adjacent empty lines; output just one empty line
1541 Equivalent to @option{-vT}.
1546 @opindex --show-tabs
1547 Display TAB characters as @samp{^I}.
1551 Ignored; for @acronym{POSIX} compatibility.
1554 @itemx --show-nonprinting
1556 @opindex --show-nonprinting
1557 Display control characters except for LFD and TAB using
1558 @samp{^} notation and precede characters that have the high bit set with
1563 On systems like MS-DOS that distinguish between text and binary files,
1564 @command{cat} normally reads and writes in binary mode. However,
1565 @command{cat} reads in text mode if one of the options
1566 @option{-bensAE} is used or if @command{cat} is reading from standard
1567 input and standard input is a terminal. Similarly, @command{cat}
1568 writes in text mode if one of the options @option{-bensAE} is used or
1569 if standard output is a terminal.
1576 # Output f's contents, then standard input, then g's contents.
1579 # Copy standard input to standard output.
1584 @node tac invocation
1585 @section @command{tac}: Concatenate and write files in reverse
1588 @cindex reversing files
1590 @command{tac} copies each @var{file} (@samp{-} means standard input), or
1591 standard input if none are given, to standard output, reversing the
1592 records (lines by default) in each separately. Synopsis:
1595 tac [@var{option}]@dots{} [@var{file}]@dots{}
1598 @dfn{Records} are separated by instances of a string (newline by
1599 default). By default, this separator string is attached to the end of
1600 the record that it follows in the file.
1602 The program accepts the following options. Also see @ref{Common options}.
1610 The separator is attached to the beginning of the record that it
1611 precedes in the file.
1617 Treat the separator string as a regular expression. Users of @command{tac}
1618 on MS-DOS/MS-Windows should note that, since @command{tac} reads files in
1619 binary mode, each line of a text file might end with a CR/LF pair
1620 instead of the Unix-style LF.
1622 @item -s @var{separator}
1623 @itemx --separator=@var{separator}
1625 @opindex --separator
1626 Use @var{separator} as the record separator, instead of newline.
1634 @section @command{nl}: Number lines and write files
1637 @cindex numbering lines
1638 @cindex line numbering
1640 @command{nl} writes each @var{file} (@samp{-} means standard input), or
1641 standard input if none are given, to standard output, with line numbers
1642 added to some or all of the lines. Synopsis:
1645 nl [@var{option}]@dots{} [@var{file}]@dots{}
1648 @cindex logical pages, numbering on
1649 @command{nl} decomposes its input into (logical) pages; by default, the
1650 line number is reset to 1 at the top of each logical page. @command{nl}
1651 treats all of the input files as a single document; it does not reset
1652 line numbers or logical pages between files.
1654 @cindex headers, numbering
1655 @cindex body, numbering
1656 @cindex footers, numbering
1657 A logical page consists of three sections: header, body, and footer.
1658 Any of the sections can be empty. Each can be numbered in a different
1659 style from the others.
1661 The beginnings of the sections of logical pages are indicated in the
1662 input file by a line containing exactly one of these delimiter strings:
1673 The two characters from which these strings are made can be changed from
1674 @samp{\} and @samp{:} via options (see below), but the pattern and
1675 length of each string cannot be changed.
1677 A section delimiter is replaced by an empty line on output. Any text
1678 that comes before the first section delimiter string in the input file
1679 is considered to be part of a body section, so @command{nl} treats a
1680 file that contains no section delimiters as a single body section.
1682 The program accepts the following options. Also see @ref{Common options}.
1686 @item -b @var{style}
1687 @itemx --body-numbering=@var{style}
1689 @opindex --body-numbering
1690 Select the numbering style for lines in the body section of each
1691 logical page. When a line is not numbered, the current line number
1692 is not incremented, but the line number separator character is still
1693 prepended to the line. The styles are:
1699 number only nonempty lines (default for body),
1701 do not number lines (default for header and footer),
1703 number only lines that contain a match for the basic regular
1704 expression @var{bre}.
1705 @xref{Regular Expressions, , Regular Expressions, grep, The GNU Grep Manual}.
1709 @itemx --section-delimiter=@var{cd}
1711 @opindex --section-delimiter
1712 @cindex section delimiters of pages
1713 Set the section delimiter characters to @var{cd}; default is
1714 @samp{\:}. If only @var{c} is given, the second remains @samp{:}.
1715 (Remember to protect @samp{\} or other metacharacters from shell
1716 expansion with quotes or extra backslashes.)
1718 @item -f @var{style}
1719 @itemx --footer-numbering=@var{style}
1721 @opindex --footer-numbering
1722 Analogous to @option{--body-numbering}.
1724 @item -h @var{style}
1725 @itemx --header-numbering=@var{style}
1727 @opindex --header-numbering
1728 Analogous to @option{--body-numbering}.
1730 @item -i @var{number}
1731 @itemx --line-increment=@var{number}
1733 @opindex --line-increment
1734 Increment line numbers by @var{number} (default 1).
1736 @item -l @var{number}
1737 @itemx --join-blank-lines=@var{number}
1739 @opindex --join-blank-lines
1740 @cindex empty lines, numbering
1741 @cindex blank lines, numbering
1742 Consider @var{number} (default 1) consecutive empty lines to be one
1743 logical line for numbering, and only number the last one. Where fewer
1744 than @var{number} consecutive empty lines occur, do not number them.
1745 An empty line is one that contains no characters, not even spaces
1748 @item -n @var{format}
1749 @itemx --number-format=@var{format}
1751 @opindex --number-format
1752 Select the line numbering format (default is @code{rn}):
1756 @opindex ln @r{format for @command{nl}}
1757 left justified, no leading zeros;
1759 @opindex rn @r{format for @command{nl}}
1760 right justified, no leading zeros;
1762 @opindex rz @r{format for @command{nl}}
1763 right justified, leading zeros.
1767 @itemx --no-renumber
1769 @opindex --no-renumber
1770 Do not reset the line number at the start of a logical page.
1772 @item -s @var{string}
1773 @itemx --number-separator=@var{string}
1775 @opindex --number-separator
1776 Separate the line number from the text line in the output with
1777 @var{string} (default is the TAB character).
1779 @item -v @var{number}
1780 @itemx --starting-line-number=@var{number}
1782 @opindex --starting-line-number
1783 Set the initial line number on each logical page to @var{number} (default 1).
1785 @item -w @var{number}
1786 @itemx --number-width=@var{number}
1788 @opindex --number-width
1789 Use @var{number} characters for line numbers (default 6).
1797 @section @command{od}: Write files in octal or other formats
1800 @cindex octal dump of files
1801 @cindex hex dump of files
1802 @cindex ASCII dump of files
1803 @cindex file contents, dumping unambiguously
1805 @command{od} writes an unambiguous representation of each @var{file}
1806 (@samp{-} means standard input), or standard input if none are given.
1810 od [@var{option}]@dots{} [@var{file}]@dots{}
1811 od [-abcdfilosx]@dots{} [@var{file}] [[+]@var{offset}[.][b]]
1812 od [@var{option}]@dots{} --traditional [@var{file}]@c
1813 [[+]@var{offset}[.][b] [[+]@var{label}[.][b]]]
1816 Each line of output consists of the offset in the input, followed by
1817 groups of data from the file. By default, @command{od} prints the offset in
1818 octal, and each group of file data is a C @code{short int}'s worth of input
1819 printed as a single octal number.
1821 If @var{offset} is given, it specifies how many input bytes to skip
1822 before formatting and writing. By default, it is interpreted as an
1823 octal number, but the optional trailing decimal point causes it to be
1824 interpreted as decimal. If no decimal is specified and the offset
1825 begins with @samp{0x} or @samp{0X} it is interpreted as a hexadecimal
1826 number. If there is a trailing @samp{b}, the number of bytes skipped
1827 will be @var{offset} multiplied by 512.
1829 If a command is of both the first and second forms, the second form is
1830 assumed if the last operand begins with @samp{+} or (if there are two
1831 operands) a digit. For example, in @samp{od foo 10} and @samp{od +10}
1832 the @samp{10} is an offset, whereas in @samp{od 10} the @samp{10} is a
1835 The program accepts the following options. Also see @ref{Common options}.
1839 @item -A @var{radix}
1840 @itemx --address-radix=@var{radix}
1842 @opindex --address-radix
1843 @cindex radix for file offsets
1844 @cindex file offset radix
1845 Select the base in which file offsets are printed. @var{radix} can
1846 be one of the following:
1856 none (do not print offsets).
1859 The default is octal.
1861 @item -j @var{bytes}
1862 @itemx --skip-bytes=@var{bytes}
1864 @opindex --skip-bytes
1865 Skip @var{bytes} input bytes before formatting and writing. If
1866 @var{bytes} begins with @samp{0x} or @samp{0X}, it is interpreted in
1867 hexadecimal; otherwise, if it begins with @samp{0}, in octal; otherwise,
1869 @multiplierSuffixes{bytes}
1871 @item -N @var{bytes}
1872 @itemx --read-bytes=@var{bytes}
1874 @opindex --read-bytes
1875 Output at most @var{bytes} bytes of the input. Prefixes and suffixes on
1876 @code{bytes} are interpreted as for the @option{-j} option.
1878 @item -S @var{bytes}
1879 @itemx --strings[=@var{bytes}]
1882 @cindex string constants, outputting
1883 Instead of the normal output, output only @dfn{string constants}: at
1884 least @var{bytes} consecutive @acronym{ASCII} graphic characters,
1885 followed by a zero byte (@acronym{ASCII} @sc{nul}).
1886 Prefixes and suffixes on @var{bytes} are interpreted as for the
1889 If @var{n} is omitted with @option{--strings}, the default is 3.
1892 @itemx --format=@var{type}
1895 Select the format in which to output the file data. @var{type} is a
1896 string of one or more of the below type indicator characters. If you
1897 include more than one type indicator character in a single @var{type}
1898 string, or use this option more than once, @command{od} writes one copy
1899 of each output line using each of the data types that you specified,
1900 in the order that you specified.
1902 Adding a trailing ``z'' to any type specification appends a display
1903 of the @acronym{ASCII} character representation of the printable characters
1904 to the output line generated by the type specification.
1908 named character, ignoring high-order bit
1910 @acronym{ASCII} character or backslash escape,
1914 floating point (@pxref{Floating point})
1923 The type @code{a} outputs things like @samp{sp} for space, @samp{nl} for
1924 newline, and @samp{nul} for a zero byte. Only the least significant
1925 seven bits of each byte is used; the high-order bit is ignored.
1926 Type @code{c} outputs
1927 @samp{ }, @samp{\n}, and @code{\0}, respectively.
1930 Except for types @samp{a} and @samp{c}, you can specify the number
1931 of bytes to use in interpreting each number in the given data type
1932 by following the type indicator character with a decimal integer.
1933 Alternately, you can specify the size of one of the C compiler's
1934 built-in data types by following the type indicator character with
1935 one of the following characters. For integers (@samp{d}, @samp{o},
1936 @samp{u}, @samp{x}):
1949 For floating point (@code{f}):
1961 @itemx --output-duplicates
1963 @opindex --output-duplicates
1964 Output consecutive lines that are identical. By default, when two or
1965 more consecutive output lines would be identical, @command{od} outputs only
1966 the first line, and puts just an asterisk on the following line to
1967 indicate the elision.
1970 @itemx --width[=@var{n}]
1973 Dump @code{n} input bytes per output line. This must be a multiple of
1974 the least common multiple of the sizes associated with the specified
1977 If this option is not given at all, the default is 16. If @var{n} is
1978 omitted, the default is 32.
1982 The next several options are shorthands for format specifications.
1983 @sc{gnu} @command{od} accepts any combination of shorthands and format
1984 specification options. These options accumulate.
1990 Output as named characters. Equivalent to @samp{-t a}.
1994 Output as octal bytes. Equivalent to @samp{-t o1}.
1998 Output as @acronym{ASCII} characters or backslash escapes. Equivalent to
2003 Output as unsigned decimal two-byte units. Equivalent to @samp{-t u2}.
2007 Output as floats. Equivalent to @samp{-t fF}.
2011 Output as decimal ints. Equivalent to @samp{-t dI}.
2015 Output as decimal long ints. Equivalent to @samp{-t dL}.
2019 Output as octal two-byte units. Equivalent to @option{-t o2}.
2023 Output as decimal two-byte units. Equivalent to @option{-t d2}.
2027 Output as hexadecimal two-byte units. Equivalent to @samp{-t x2}.
2030 @opindex --traditional
2031 Recognize the non-option label argument that traditional @command{od}
2032 accepted. The following syntax:
2035 od --traditional [@var{file}] [[+]@var{offset}[.][b] [[+]@var{label}[.][b]]]
2039 can be used to specify at most one file and optional arguments
2040 specifying an offset and a pseudo-start address, @var{label}.
2041 The @var{label} argument is interpreted
2042 just like @var{offset}, but it specifies an initial pseudo-address. The
2043 pseudo-addresses are displayed in parentheses following any normal
2050 @node base64 invocation
2051 @section @command{base64}: Transform data into printable data
2054 @cindex base64 encoding
2056 @command{base64} transforms data read from a file, or standard input,
2057 into (or from) base64 encoded form. The base64 encoded form uses
2058 printable @acronym{ASCII} characters to represent binary data.
2062 base64 [@var{option}]@dots{} [@var{file}]
2063 base64 --decode [@var{option}]@dots{} [@var{file}]
2066 The base64 encoding expands data to roughly 133% of the original.
2067 The format conforms to
2068 @uref{ftp://ftp.rfc-editor.org/in-notes/rfc4648.txt, RFC 4648}.
2070 The program accepts the following options. Also see @ref{Common options}.
2075 @itemx --wrap=@var{cols}
2079 @cindex column to wrap data after
2080 During encoding, wrap lines after @var{cols} characters. This must be
2083 The default is to wrap after 76 characters. Use the value 0 to
2084 disable line wrapping altogether.
2090 @cindex Decode base64 data
2091 @cindex Base64 decoding
2092 Change the mode of operation, from the default of encoding data, to
2093 decoding data. Input is expected to be base64 encoded data, and the
2094 output will be the original data.
2097 @itemx --ignore-garbage
2099 @opindex --ignore-garbage
2100 @cindex Ignore garbage in base64 stream
2101 When decoding, newlines are always accepted.
2102 During decoding, ignore unrecognized bytes,
2103 to permit distorted data to be decoded.
2110 @node Formatting file contents
2111 @chapter Formatting file contents
2113 @cindex formatting file contents
2115 These commands reformat the contents of files.
2118 * fmt invocation:: Reformat paragraph text.
2119 * pr invocation:: Paginate or columnate files for printing.
2120 * fold invocation:: Wrap input lines to fit in specified width.
2124 @node fmt invocation
2125 @section @command{fmt}: Reformat paragraph text
2128 @cindex reformatting paragraph text
2129 @cindex paragraphs, reformatting
2130 @cindex text, reformatting
2132 @command{fmt} fills and joins lines to produce output lines of (at most)
2133 a given number of characters (75 by default). Synopsis:
2136 fmt [@var{option}]@dots{} [@var{file}]@dots{}
2139 @command{fmt} reads from the specified @var{file} arguments (or standard
2140 input if none are given), and writes to standard output.
2142 By default, blank lines, spaces between words, and indentation are
2143 preserved in the output; successive input lines with different
2144 indentation are not joined; tabs are expanded on input and introduced on
2147 @cindex line-breaking
2148 @cindex sentences and line-breaking
2149 @cindex Knuth, Donald E.
2150 @cindex Plass, Michael F.
2151 @command{fmt} prefers breaking lines at the end of a sentence, and tries to
2152 avoid line breaks after the first word of a sentence or before the last
2153 word of a sentence. A @dfn{sentence break} is defined as either the end
2154 of a paragraph or a word ending in any of @samp{.?!}, followed by two
2155 spaces or end of line, ignoring any intervening parentheses or quotes.
2156 Like @TeX{}, @command{fmt} reads entire ``paragraphs'' before choosing line
2157 breaks; the algorithm is a variant of that given by Donald E. Knuth
2158 and Michael F. Plass in ``Breaking Paragraphs Into Lines'',
2159 @cite{Software---Practice & Experience} @b{11}, 11 (November 1981),
2162 The program accepts the following options. Also see @ref{Common options}.
2167 @itemx --crown-margin
2169 @opindex --crown-margin
2170 @cindex crown margin
2171 @dfn{Crown margin} mode: preserve the indentation of the first two
2172 lines within a paragraph, and align the left margin of each subsequent
2173 line with that of the second line.
2176 @itemx --tagged-paragraph
2178 @opindex --tagged-paragraph
2179 @cindex tagged paragraphs
2180 @dfn{Tagged paragraph} mode: like crown margin mode, except that if
2181 indentation of the first line of a paragraph is the same as the
2182 indentation of the second, the first line is treated as a one-line
2188 @opindex --split-only
2189 Split lines only. Do not join short lines to form longer ones. This
2190 prevents sample lines of code, and other such ``formatted'' text from
2191 being unduly combined.
2194 @itemx --uniform-spacing
2196 @opindex --uniform-spacing
2197 Uniform spacing. Reduce spacing between words to one space, and spacing
2198 between sentences to two spaces.
2201 @itemx -w @var{width}
2202 @itemx --width=@var{width}
2203 @opindex -@var{width}
2206 Fill output lines up to @var{width} characters (default 75). @command{fmt}
2207 initially tries to make lines about 7% shorter than this, to give it
2208 room to balance line lengths.
2210 @item -p @var{prefix}
2211 @itemx --prefix=@var{prefix}
2212 Only lines beginning with @var{prefix} (possibly preceded by whitespace)
2213 are subject to formatting. The prefix and any preceding whitespace are
2214 stripped for the formatting and then re-attached to each formatted output
2215 line. One use is to format certain kinds of program comments, while
2216 leaving the code unchanged.
2224 @section @command{pr}: Paginate or columnate files for printing
2227 @cindex printing, preparing files for
2228 @cindex multicolumn output, generating
2229 @cindex merging files in parallel
2231 @command{pr} writes each @var{file} (@samp{-} means standard input), or
2232 standard input if none are given, to standard output, paginating and
2233 optionally outputting in multicolumn format; optionally merges all
2234 @var{file}s, printing all in parallel, one per column. Synopsis:
2237 pr [@var{option}]@dots{} [@var{file}]@dots{}
2241 By default, a 5-line header is printed at each page: two blank lines;
2242 a line with the date, the file name, and the page count; and two more
2243 blank lines. A footer of five blank lines is also printed.
2244 The default @var{page_length} is 66
2245 lines. The default number of text lines is therefore 56.
2246 The text line of the header takes the form
2247 @samp{@var{date} @var{string} @var{page}}, with spaces inserted around
2248 @var{string} so that the line takes up the full @var{page_width}. Here,
2249 @var{date} is the date (see the @option{-D} or @option{--date-format}
2250 option for details), @var{string} is the centered header string, and
2251 @var{page} identifies the page number. The @env{LC_MESSAGES} locale
2252 category affects the spelling of @var{page}; in the default C locale, it
2253 is @samp{Page @var{number}} where @var{number} is the decimal page
2256 Form feeds in the input cause page breaks in the output. Multiple form
2257 feeds produce empty pages.
2259 Columns are of equal width, separated by an optional string (default
2260 is @samp{space}). For multicolumn output, lines will always be truncated to
2261 @var{page_width} (default 72), unless you use the @option{-J} option.
2263 column output no line truncation occurs by default. Use @option{-W} option to
2264 truncate lines in that case.
2266 The following changes were made in version 1.22i and apply to later
2267 versions of @command{pr}:
2268 @c FIXME: this whole section here sounds very awkward to me. I
2269 @c made a few small changes, but really it all needs to be redone. - Brian
2270 @c OK, I fixed another sentence or two, but some of it I just don't understand.
2275 Some small @var{letter options} (@option{-s}, @option{-w}) have been
2276 redefined for better @acronym{POSIX} compliance. The output of some further
2277 cases has been adapted to other Unix systems. These changes are not
2278 compatible with earlier versions of the program.
2281 Some @var{new capital letter} options (@option{-J}, @option{-S}, @option{-W})
2282 have been introduced to turn off unexpected interferences of small letter
2283 options. The @option{-N} option and the second argument @var{last_page}
2284 of @samp{+FIRST_PAGE} offer more flexibility. The detailed handling of
2285 form feeds set in the input files requires the @option{-T} option.
2288 Capital letter options override small letter ones.
2291 Some of the option-arguments (compare @option{-s}, @option{-e},
2292 @option{-i}, @option{-n}) cannot be specified as separate arguments from the
2293 preceding option letter (already stated in the @acronym{POSIX} specification).
2296 The program accepts the following options. Also see @ref{Common options}.
2300 @item +@var{first_page}[:@var{last_page}]
2301 @itemx --pages=@var{first_page}[:@var{last_page}]
2302 @c The two following @opindex lines evoke warnings because they contain ':'
2303 @c The 'info' spec does not permit that. If we use those lines, we end
2304 @c up with truncated index entries that don't work.
2305 @c @opindex +@var{first_page}[:@var{last_page}]
2306 @c @opindex --pages=@var{first_page}[:@var{last_page}]
2307 @opindex +@var{page_range}
2308 @opindex --pages=@var{page_range}
2309 Begin printing with page @var{first_page} and stop with @var{last_page}.
2310 Missing @samp{:@var{last_page}} implies end of file. While estimating
2311 the number of skipped pages each form feed in the input file results
2312 in a new page. Page counting with and without @samp{+@var{first_page}}
2313 is identical. By default, counting starts with the first page of input
2314 file (not first page printed). Line numbering may be altered by @option{-N}
2318 @itemx --columns=@var{column}
2319 @opindex -@var{column}
2321 @cindex down columns
2322 With each single @var{file}, produce @var{column} columns of output
2323 (default is 1) and print columns down, unless @option{-a} is used. The
2324 column width is automatically decreased as @var{column} increases; unless
2325 you use the @option{-W/-w} option to increase @var{page_width} as well.
2326 This option might well cause some lines to be truncated. The number of
2327 lines in the columns on each page are balanced. The options @option{-e}
2328 and @option{-i} are on for multiple text-column output. Together with
2329 @option{-J} option column alignment and line truncation is turned off.
2330 Lines of full length are joined in a free field format and @option{-S}
2331 option may set field separators. @option{-@var{column}} may not be used
2332 with @option{-m} option.
2338 @cindex across columns
2339 With each single @var{file}, print columns across rather than down. The
2340 @option{-@var{column}} option must be given with @var{column} greater than one.
2341 If a line is too long to fit in a column, it is truncated.
2344 @itemx --show-control-chars
2346 @opindex --show-control-chars
2347 Print control characters using hat notation (e.g., @samp{^G}); print
2348 other nonprinting characters in octal backslash notation. By default,
2349 nonprinting characters are not changed.
2352 @itemx --double-space
2354 @opindex --double-space
2355 @cindex double spacing
2356 Double space the output.
2358 @item -D @var{format}
2359 @itemx --date-format=@var{format}
2360 @cindex time formats
2361 @cindex formatting times
2362 Format header dates using @var{format}, using the same conventions as
2363 for the command @samp{date +@var{format}}; @xref{date invocation}.
2364 Except for directives, which start with
2365 @samp{%}, characters in @var{format} are printed unchanged. You can use
2366 this option to specify an arbitrary string in place of the header date,
2367 e.g., @option{--date-format="Monday morning"}.
2369 @vindex POSIXLY_CORRECT
2371 The default date format is @samp{%Y-%m-%d %H:%M} (for example,
2372 @samp{2001-12-04 23:59});
2373 but if the @env{POSIXLY_CORRECT} environment variable is set
2374 and the @env{LC_TIME} locale category specifies the @acronym{POSIX}
2375 locale, the default is @samp{%b %e %H:%M %Y} (for example,
2376 @samp{Dec@ @ 4 23:59 2001}.
2379 Time stamps are listed according to the time zone rules specified by
2380 the @env{TZ} environment variable, or by the system default rules if
2381 @env{TZ} is not set. @xref{TZ Variable,, Specifying the Time Zone
2382 with @env{TZ}, libc, The GNU C Library Reference Manual}.
2384 @item -e[@var{in-tabchar}[@var{in-tabwidth}]]
2385 @itemx --expand-tabs[=@var{in-tabchar}[@var{in-tabwidth}]]
2387 @opindex --expand-tabs
2389 Expand @var{tab}s to spaces on input. Optional argument @var{in-tabchar} is
2390 the input tab character (default is the TAB character). Second optional
2391 argument @var{in-tabwidth} is the input tab character's width (default
2399 @opindex --form-feed
2400 Use a form feed instead of newlines to separate output pages. This does
2401 not alter the default page length of 66 lines.
2403 @item -h @var{header}
2404 @itemx --header=@var{header}
2407 Replace the file name in the header with the centered string @var{header}.
2408 When using the shell, @var{header} should be quoted and should be
2409 separated from @option{-h} by a space.
2411 @item -i[@var{out-tabchar}[@var{out-tabwidth}]]
2412 @itemx --output-tabs[=@var{out-tabchar}[@var{out-tabwidth}]]
2414 @opindex --output-tabs
2416 Replace spaces with @var{tab}s on output. Optional argument @var{out-tabchar}
2417 is the output tab character (default is the TAB character). Second optional
2418 argument @var{out-tabwidth} is the output tab character's width (default
2424 @opindex --join-lines
2425 Merge lines of full length. Used together with the column options
2426 @option{-@var{column}}, @option{-a -@var{column}} or @option{-m}. Turns off
2427 @option{-W/-w} line truncation;
2428 no column alignment used; may be used with
2429 @option{--sep-string[=@var{string}]}. @option{-J} has been introduced
2430 (together with @option{-W} and @option{--sep-string})
2431 to disentangle the old (@acronym{POSIX}-compliant) options @option{-w} and
2432 @option{-s} along with the three column options.
2435 @item -l @var{page_length}
2436 @itemx --length=@var{page_length}
2439 Set the page length to @var{page_length} (default 66) lines, including
2440 the lines of the header [and the footer]. If @var{page_length} is less
2441 than or equal to 10, the header and footer are omitted, as if the
2442 @option{-t} option had been given.
2448 Merge and print all @var{file}s in parallel, one in each column. If a
2449 line is too long to fit in a column, it is truncated, unless the @option{-J}
2450 option is used. @option{--sep-string[=@var{string}]} may be used.
2452 some @var{file}s (form feeds set) produce empty columns, still marked
2453 by @var{string}. The result is a continuous line numbering and column
2454 marking throughout the whole merged file. Completely empty merged pages
2455 show no separators or line numbers. The default header becomes
2456 @samp{@var{date} @var{page}} with spaces inserted in the middle; this
2457 may be used with the @option{-h} or @option{--header} option to fill up
2458 the middle blank part.
2460 @item -n[@var{number-separator}[@var{digits}]]
2461 @itemx --number-lines[=@var{number-separator}[@var{digits}]]
2463 @opindex --number-lines
2464 Provide @var{digits} digit line numbering (default for @var{digits} is
2465 5). With multicolumn output the number occupies the first @var{digits}
2466 column positions of each text column or only each line of @option{-m}
2467 output. With single column output the number precedes each line just as
2468 @option{-m} does. Default counting of the line numbers starts with the
2469 first line of the input file (not the first line printed, compare the
2470 @option{--page} option and @option{-N} option).
2471 Optional argument @var{number-separator} is the character appended to
2472 the line number to separate it from the text followed. The default
2473 separator is the TAB character. In a strict sense a TAB is always
2474 printed with single column output only. The TAB width varies
2475 with the TAB position, e.g., with the left @var{margin} specified
2476 by @option{-o} option. With multicolumn output priority is given to
2477 @samp{equal width of output columns} (a @acronym{POSIX} specification).
2478 The TAB width is fixed to the value of the first column and does
2479 not change with different values of left @var{margin}. That means a
2480 fixed number of spaces is always printed in the place of the
2481 @var{number-separator} TAB. The tabification depends upon the output
2484 @item -N @var{line_number}
2485 @itemx --first-line-number=@var{line_number}
2487 @opindex --first-line-number
2488 Start line counting with the number @var{line_number} at first line of
2489 first page printed (in most cases not the first line of the input file).
2491 @item -o @var{margin}
2492 @itemx --indent=@var{margin}
2495 @cindex indenting lines
2497 Indent each line with a margin @var{margin} spaces wide (default is zero).
2498 The total page width is the size of the margin plus the @var{page_width}
2499 set with the @option{-W/-w} option. A limited overflow may occur with
2500 numbered single column output (compare @option{-n} option).
2503 @itemx --no-file-warnings
2505 @opindex --no-file-warnings
2506 Do not print a warning message when an argument @var{file} cannot be
2507 opened. (The exit status will still be nonzero, however.)
2509 @item -s[@var{char}]
2510 @itemx --separator[=@var{char}]
2512 @opindex --separator
2513 Separate columns by a single character @var{char}. The default for
2514 @var{char} is the TAB character without @option{-w} and @samp{no
2515 character} with @option{-w}. Without @option{-s} the default separator
2516 @samp{space} is set. @option{-s[char]} turns off line truncation of all
2517 three column options (@option{-COLUMN}|@option{-a -COLUMN}|@option{-m}) unless
2518 @option{-w} is set. This is a @acronym{POSIX}-compliant formulation.
2521 @item -S[@var{string}]
2522 @itemx --sep-string[=@var{string}]
2524 @opindex --sep-string
2525 Use @var{string} to separate output columns. The @option{-S} option doesn't
2526 affect the @option{-W/-w} option, unlike the @option{-s} option which does. It
2527 does not affect line truncation or column alignment.
2528 Without @option{-S}, and with @option{-J}, @command{pr} uses the default output
2530 Without @option{-S} or @option{-J}, @command{pr} uses a @samp{space}
2531 (same as @option{-S"@w{ }"}).
2532 If no @samp{@var{string}} argument is specified, @samp{""} is assumed.
2535 @itemx --omit-header
2537 @opindex --omit-header
2538 Do not print the usual header [and footer] on each page, and do not fill
2539 out the bottom of pages (with blank lines or a form feed). No page
2540 structure is produced, but form feeds set in the input files are retained.
2541 The predefined pagination is not changed. @option{-t} or @option{-T} may be
2542 useful together with other options; e.g.: @option{-t -e4}, expand TAB characters
2543 in the input file to 4 spaces but don't make any other changes. Use of
2544 @option{-t} overrides @option{-h}.
2547 @itemx --omit-pagination
2549 @opindex --omit-pagination
2550 Do not print header [and footer]. In addition eliminate all form feeds
2551 set in the input files.
2554 @itemx --show-nonprinting
2556 @opindex --show-nonprinting
2557 Print nonprinting characters in octal backslash notation.
2559 @item -w @var{page_width}
2560 @itemx --width=@var{page_width}
2563 Set page width to @var{page_width} characters for multiple text-column
2564 output only (default for @var{page_width} is 72). @option{-s[CHAR]} turns
2565 off the default page width and any line truncation and column alignment.
2566 Lines of full length are merged, regardless of the column options
2567 set. No @var{page_width} setting is possible with single column output.
2568 A @acronym{POSIX}-compliant formulation.
2570 @item -W @var{page_width}
2571 @itemx --page_width=@var{page_width}
2573 @opindex --page_width
2574 Set the page width to @var{page_width} characters. That's valid with and
2575 without a column option. Text lines are truncated, unless @option{-J}
2576 is used. Together with one of the three column options
2577 (@option{-@var{column}}, @option{-a -@var{column}} or @option{-m}) column
2578 alignment is always used. The separator options @option{-S} or @option{-s}
2579 don't affect the @option{-W} option. Default is 72 characters. Without
2580 @option{-W @var{page_width}} and without any of the column options NO line
2581 truncation is used (defined to keep downward compatibility and to meet
2582 most frequent tasks). That's equivalent to @option{-W 72 -J}. The header
2583 line is never truncated.
2590 @node fold invocation
2591 @section @command{fold}: Wrap input lines to fit in specified width
2594 @cindex wrapping long input lines
2595 @cindex folding long input lines
2597 @command{fold} writes each @var{file} (@option{-} means standard input), or
2598 standard input if none are given, to standard output, breaking long
2602 fold [@var{option}]@dots{} [@var{file}]@dots{}
2605 By default, @command{fold} breaks lines wider than 80 columns. The output
2606 is split into as many lines as necessary.
2608 @cindex screen columns
2609 @command{fold} counts screen columns by default; thus, a tab may count more
2610 than one column, backspace decreases the column count, and carriage
2611 return sets the column to zero.
2613 The program accepts the following options. Also see @ref{Common options}.
2621 Count bytes rather than columns, so that tabs, backspaces, and carriage
2622 returns are each counted as taking up one column, just like other
2629 Break at word boundaries: the line is broken after the last blank before
2630 the maximum line length. If the line contains no such blanks, the line
2631 is broken at the maximum line length as usual.
2633 @item -w @var{width}
2634 @itemx --width=@var{width}
2637 Use a maximum line length of @var{width} columns instead of 80.
2639 For compatibility @command{fold} supports an obsolete option syntax
2640 @option{-@var{width}}. New scripts should use @option{-w @var{width}}
2648 @node Output of parts of files
2649 @chapter Output of parts of files
2651 @cindex output of parts of files
2652 @cindex parts of files, output of
2654 These commands output pieces of the input.
2657 * head invocation:: Output the first part of files.
2658 * tail invocation:: Output the last part of files.
2659 * split invocation:: Split a file into pieces.
2660 * csplit invocation:: Split a file into context-determined pieces.
2663 @node head invocation
2664 @section @command{head}: Output the first part of files
2667 @cindex initial part of files, outputting
2668 @cindex first part of files, outputting
2670 @command{head} prints the first part (10 lines by default) of each
2671 @var{file}; it reads from standard input if no files are given or
2672 when given a @var{file} of @option{-}. Synopsis:
2675 head [@var{option}]@dots{} [@var{file}]@dots{}
2678 If more than one @var{file} is specified, @command{head} prints a
2679 one-line header consisting of:
2682 ==> @var{file name} <==
2686 before the output for each @var{file}.
2688 The program accepts the following options. Also see @ref{Common options}.
2693 @itemx --bytes=@var{k}
2696 Print the first @var{k} bytes, instead of initial lines.
2697 However, if @var{k} starts with a @samp{-},
2698 print all but the last @var{k} bytes of each file.
2699 @multiplierSuffixes{k}
2702 @itemx --lines=@var{k}
2705 Output the first @var{k} lines.
2706 However, if @var{k} starts with a @samp{-},
2707 print all but the last @var{k} lines of each file.
2708 Size multiplier suffixes are the same as with the @option{-c} option.
2716 Never print file name headers.
2722 Always print file name headers.
2726 For compatibility @command{head} also supports an obsolete option syntax
2727 @option{-@var{count}@var{options}}, which is recognized only if it is
2728 specified first. @var{count} is a decimal number optionally followed
2729 by a size letter (@samp{b}, @samp{k}, @samp{m}) as in @option{-c}, or
2730 @samp{l} to mean count by lines, or other option letters (@samp{cqv}).
2731 Scripts intended for standard hosts should use @option{-c @var{count}}
2732 or @option{-n @var{count}} instead. If your script must also run on
2733 hosts that support only the obsolete syntax, it is usually simpler to
2734 avoid @command{head}, e.g., by using @samp{sed 5q} instead of
2740 @node tail invocation
2741 @section @command{tail}: Output the last part of files
2744 @cindex last part of files, outputting
2746 @command{tail} prints the last part (10 lines by default) of each
2747 @var{file}; it reads from standard input if no files are given or
2748 when given a @var{file} of @samp{-}. Synopsis:
2751 tail [@var{option}]@dots{} [@var{file}]@dots{}
2754 If more than one @var{file} is specified, @command{tail} prints a
2755 one-line header consisting of:
2758 ==> @var{file name} <==
2762 before the output for each @var{file}.
2764 @cindex BSD @command{tail}
2765 @sc{gnu} @command{tail} can output any amount of data (some other versions of
2766 @command{tail} cannot). It also has no @option{-r} option (print in
2767 reverse), since reversing a file is really a different job from printing
2768 the end of a file; BSD @command{tail} (which is the one with @option{-r}) can
2769 only reverse files that are at most as large as its buffer, which is
2770 typically 32 KiB@. A more reliable and versatile way to reverse files is
2771 the @sc{gnu} @command{tac} command.
2773 The program accepts the following options. Also see @ref{Common options}.
2778 @itemx --bytes=@var{k}
2781 Output the last @var{k} bytes, instead of final lines.
2782 However, if @var{k} starts with a @samp{+}, start printing with the
2783 @var{k}th byte from the start of each file, instead of from the end.
2784 @multiplierSuffixes{k}
2787 @itemx --follow[=@var{how}]
2790 @cindex growing files
2791 @vindex name @r{follow option}
2792 @vindex descriptor @r{follow option}
2793 Loop forever trying to read more characters at the end of the file,
2794 presumably because the file is growing.
2795 If more than one file is given, @command{tail} prints a header whenever it
2796 gets output from a different file, to indicate which file that output is
2799 There are two ways to specify how you'd like to track files with this option,
2800 but that difference is noticeable only when a followed file is removed or
2802 If you'd like to continue to track the end of a growing file even after
2803 it has been unlinked, use @option{--follow=descriptor}. This is the default
2804 behavior, but it is not useful if you're tracking a log file that may be
2805 rotated (removed or renamed, then reopened). In that case, use
2806 @option{--follow=name} to track the named file, perhaps by reopening it
2807 periodically to see if it has been removed and recreated by some other program.
2808 Note that the inotify-based implementation handles this case without
2809 the need for any periodic reopening.
2811 No matter which method you use, if the tracked file is determined to have
2812 shrunk, @command{tail} prints a message saying the file has been truncated
2813 and resumes tracking the end of the file from the newly-determined endpoint.
2815 When a file is removed, @command{tail}'s behavior depends on whether it is
2816 following the name or the descriptor. When following by name, tail can
2817 detect that a file has been removed and gives a message to that effect,
2818 and if @option{--retry} has been specified it will continue checking
2819 periodically to see if the file reappears.
2820 When following a descriptor, tail does not detect that the file has
2821 been unlinked or renamed and issues no message; even though the file
2822 may no longer be accessible via its original name, it may still be
2825 The option values @samp{descriptor} and @samp{name} may be specified only
2826 with the long form of the option, not with @option{-f}.
2828 The @option{-f} option is ignored if
2829 no @var{file} operand is specified and standard input is a FIFO or a pipe.
2830 Likewise, the @option{-f} option has no effect for any
2831 operand specified as @samp{-}, when standard input is a FIFO or a pipe.
2833 With kernel inotify support, output is triggered by file changes
2834 and is generally very prompt.
2835 Otherwise, @command{tail} sleeps for one second between checks---
2836 use @option{--sleep-interval=@var{n}} to change that default---which can
2837 make the output appear slightly less responsive or bursty.
2838 When using tail without inotify support, you can make it more responsive
2839 by using a sub-second sleep interval, e.g., via an alias like this:
2842 alias tail='tail -s.1'
2847 This option is the same as @option{--follow=name --retry}. That is, tail
2848 will attempt to reopen a file when it is removed. Should this fail, tail
2849 will keep trying until it becomes accessible again.
2853 This option is useful mainly when following by name (i.e., with
2854 @option{--follow=name}).
2855 Without this option, when tail encounters a file that doesn't
2856 exist or is otherwise inaccessible, it reports that fact and
2857 never checks it again.
2859 @itemx --sleep-interval=@var{number}
2860 @opindex --sleep-interval
2861 Change the number of seconds to wait between iterations (the default is 1.0).
2862 During one iteration, every specified file is checked to see if it has
2864 Historical implementations of @command{tail} have required that
2865 @var{number} be an integer. However, GNU @command{tail} accepts
2866 an arbitrary floating point number. @xref{Floating point}.
2867 When @command{tail} uses inotify, this polling-related option
2868 is usually ignored. However, if you also specify @option{--pid=@var{p}},
2869 @command{tail} checks whether process @var{p} is alive at least
2870 every @var{number} seconds.
2872 @itemx --pid=@var{pid}
2874 When following by name or by descriptor, you may specify the process ID,
2875 @var{pid}, of the sole writer of all @var{file} arguments. Then, shortly
2876 after that process terminates, tail will also terminate. This will
2877 work properly only if the writer and the tailing process are running on
2878 the same machine. For example, to save the output of a build in a file
2879 and to watch the file grow, if you invoke @command{make} and @command{tail}
2880 like this then the tail process will stop when your build completes.
2881 Without this option, you would have had to kill the @code{tail -f}
2885 $ make >& makerr & tail --pid=$! -f makerr
2888 If you specify a @var{pid} that is not in use or that does not correspond
2889 to the process that is writing to the tailed files, then @command{tail}
2890 may terminate long before any @var{file}s stop growing or it may not
2891 terminate until long after the real writer has terminated.
2892 Note that @option{--pid} cannot be supported on some systems; @command{tail}
2893 will print a warning if this is the case.
2895 @itemx --max-unchanged-stats=@var{n}
2896 @opindex --max-unchanged-stats
2897 When tailing a file by name, if there have been @var{n} (default
2898 n=@value{DEFAULT_MAX_N_UNCHANGED_STATS_BETWEEN_OPENS}) consecutive
2899 iterations for which the file has not changed, then
2900 @code{open}/@code{fstat} the file to determine if that file name is
2901 still associated with the same device/inode-number pair as before.
2902 When following a log file that is rotated, this is approximately the
2903 number of seconds between when tail prints the last pre-rotation lines
2904 and when it prints the lines that have accumulated in the new log file.
2905 This option is meaningful only when polling (i.e., without inotify)
2906 and when following by name.
2909 @itemx --lines=@var{k}
2912 Output the last @var{k} lines.
2913 However, if @var{k} starts with a @samp{+}, start printing with the
2914 @var{k}th line from the start of each file, instead of from the end.
2915 Size multiplier suffixes are the same as with the @option{-c} option.
2923 Never print file name headers.
2929 Always print file name headers.
2933 For compatibility @command{tail} also supports an obsolete usage
2934 @samp{tail -[@var{count}][bcl][f] [@var{file}]}, which is recognized
2935 only if it does not conflict with the usage described
2936 above. This obsolete form uses exactly one option and at most one
2937 file. In the option, @var{count} is an optional decimal number optionally
2938 followed by a size letter (@samp{b}, @samp{c}, @samp{l}) to mean count
2939 by 512-byte blocks, bytes, or lines, optionally followed by @samp{f}
2940 which has the same meaning as @option{-f}.
2942 @vindex _POSIX2_VERSION
2943 On older systems, the leading @samp{-} can be replaced by @samp{+} in
2944 the obsolete option syntax with the same meaning as in counts, and
2945 obsolete usage overrides normal usage when the two conflict.
2946 This obsolete behavior can be enabled or disabled with the
2947 @env{_POSIX2_VERSION} environment variable (@pxref{Standards
2950 Scripts intended for use on standard hosts should avoid obsolete
2951 syntax and should use @option{-c @var{count}[b]}, @option{-n
2952 @var{count}}, and/or @option{-f} instead. If your script must also
2953 run on hosts that support only the obsolete syntax, you can often
2954 rewrite it to avoid problematic usages, e.g., by using @samp{sed -n
2955 '$p'} rather than @samp{tail -1}. If that's not possible, the script
2956 can use a test like @samp{if tail -c +1 </dev/null >/dev/null 2>&1;
2957 then @dots{}} to decide which syntax to use.
2959 Even if your script assumes the standard behavior, you should still
2960 beware usages whose behaviors differ depending on the @acronym{POSIX}
2961 version. For example, avoid @samp{tail - main.c}, since it might be
2962 interpreted as either @samp{tail main.c} or as @samp{tail -- -
2963 main.c}; avoid @samp{tail -c 4}, since it might mean either @samp{tail
2964 -c4} or @samp{tail -c 10 4}; and avoid @samp{tail +4}, since it might
2965 mean either @samp{tail ./+4} or @samp{tail -n +4}.
2970 @node split invocation
2971 @section @command{split}: Split a file into pieces.
2974 @cindex splitting a file into pieces
2975 @cindex pieces, splitting a file into
2977 @command{split} creates output files containing consecutive or interleaved
2978 sections of @var{input} (standard input if none is given or @var{input}
2979 is @samp{-}). Synopsis:
2982 split [@var{option}] [@var{input} [@var{prefix}]]
2985 By default, @command{split} puts 1000 lines of @var{input} (or whatever is
2986 left over for the last section), into each output file.
2988 @cindex output file name prefix
2989 The output files' names consist of @var{prefix} (@samp{x} by default)
2990 followed by a group of characters (@samp{aa}, @samp{ab}, @dots{} by
2991 default), such that concatenating the output files in traditional
2992 sorted order by file name produces the original input file (except
2993 @option{-r}). If the output file names are exhausted, @command{split}
2994 reports an error without deleting the output files that it did create.
2996 The program accepts the following options. Also see @ref{Common options}.
3000 @item -l @var{lines}
3001 @itemx --lines=@var{lines}
3004 Put @var{lines} lines of @var{input} into each output file.
3006 For compatibility @command{split} also supports an obsolete
3007 option syntax @option{-@var{lines}}. New scripts should use
3008 @option{-l @var{lines}} instead.
3011 @itemx --bytes=@var{size}
3014 Put @var{size} bytes of @var{input} into each output file.
3015 @multiplierSuffixes{size}
3018 @itemx --line-bytes=@var{size}
3020 @opindex --line-bytes
3021 Put into each output file as many complete lines of @var{input} as
3022 possible without exceeding @var{size} bytes. Individual lines longer than
3023 @var{size} bytes are broken into multiple files.
3024 @var{size} has the same format as for the @option{--bytes} option.
3026 @itemx --filter=@var{command}
3028 With this option, rather than simply writing to each output file,
3029 write through a pipe to the specified shell @var{command} for each output file.
3030 @var{command} should use the $FILE environment variable, which is set
3031 to a different output file name for each invocation of the command.
3032 For example, imagine that you have a 1TiB compressed file
3033 that, if uncompressed, would be too large to reside on disk,
3034 yet you must split it into individually-compressed pieces
3035 of a more manageable size.
3036 To do that, you might run this command:
3039 xz -dc BIG.xz | split -b200G --filter='xz > $FILE.xz' - big-
3042 Assuming a 10:1 compression ratio, that would create about fifty 20GiB files
3043 with names @file{big-xaa.xz}, @file{big-xab.xz}, @file{big-xac.xz}, etc.
3045 @item -n @var{chunks}
3046 @itemx --number=@var{chunks}
3050 Split @var{input} to @var{chunks} output files where @var{chunks} may be:
3053 @var{n} generate @var{n} files based on current size of @var{input}
3054 @var{k}/@var{n} only output @var{k}th of @var{n} to stdout
3055 l/@var{n} generate @var{n} files without splitting lines
3056 l/@var{k}/@var{n} likewise but only output @var{k}th of @var{n} to stdout
3057 r/@var{n} like @samp{l} but use round robin distribution
3058 r/@var{k}/@var{n} likewise but only output @var{k}th of @var{n} to stdout
3061 Any excess bytes remaining after dividing the @var{input}
3062 into @var{n} chunks, are assigned to the last chunk.
3063 Any excess bytes appearing after the initial calculation are discarded
3064 (except when using @samp{r} mode).
3066 All @var{n} files are created even if there are fewer than @var{n} lines,
3067 or the @var{input} is truncated.
3069 For @samp{l} mode, chunks are approximately @var{input} size / @var{n}.
3070 The @var{input} is partitioned into @var{n} equal sized portions, with
3071 the last assigned any excess. If a line @emph{starts} within a partition
3072 it is written completely to the corresponding file. Since lines
3073 are not split even if they overlap a partition, the files written
3074 can be larger or smaller than the partition size, and even empty
3075 if a line is so long as to completely overlap the partition.
3077 For @samp{r} mode, the size of @var{input} is irrelevant,
3078 and so can be a pipe for example.
3080 @item -a @var{length}
3081 @itemx --suffix-length=@var{length}
3083 @opindex --suffix-length
3084 Use suffixes of length @var{length}. The default @var{length} is 2.
3087 @itemx --numeric-suffixes
3089 @opindex --numeric-suffixes
3090 Use digits in suffixes rather than lower-case letters.
3093 @itemx --elide-empty-files
3095 @opindex --elide-empty-files
3096 Suppress the generation of zero-length output files. This can happen
3097 with the @option{--number} option if a file is (truncated to be) shorter
3098 than the number requested, or if a line is so long as to completely
3099 span a chunk. The output file sequence numbers, always run consecutively
3100 even when this option is specified.
3105 @opindex --unbuffered
3106 Immediately copy input to output in @option{--number r/...} mode,
3107 which is a much slower mode of operation.
3111 Write a diagnostic just before each output file is opened.
3117 Here are a few examples to illustrate how the
3118 @option{--number} (@option{-n}) option works:
3120 Notice how, by default, one line may be split onto two or more:
3123 $ seq -w 6 10 > k; split -n3 k; head xa?
3136 Use the "l/" modifier to suppress that:
3139 $ seq -w 6 10 > k; split -nl/3 k; head xa?
3152 Use the "r/" modifier to distribute lines in a round-robin fashion:
3155 $ seq -w 6 10 > k; split -nr/3 k; head xa?
3168 You can also extract just the Kth chunk.
3169 This extracts and prints just the 7th "chunk" of 33:
3172 $ seq 100 > k; split -nl/7/33 k
3179 @node csplit invocation
3180 @section @command{csplit}: Split a file into context-determined pieces
3183 @cindex context splitting
3184 @cindex splitting a file into pieces by context
3186 @command{csplit} creates zero or more output files containing sections of
3187 @var{input} (standard input if @var{input} is @samp{-}). Synopsis:
3190 csplit [@var{option}]@dots{} @var{input} @var{pattern}@dots{}
3193 The contents of the output files are determined by the @var{pattern}
3194 arguments, as detailed below. An error occurs if a @var{pattern}
3195 argument refers to a nonexistent line of the input file (e.g., if no
3196 remaining line matches a given regular expression). After every
3197 @var{pattern} has been matched, any remaining input is copied into one
3200 By default, @command{csplit} prints the number of bytes written to each
3201 output file after it has been created.
3203 The types of pattern arguments are:
3208 Create an output file containing the input up to but not including line
3209 @var{n} (a positive integer). If followed by a repeat count, also
3210 create an output file containing the next @var{n} lines of the input
3211 file once for each repeat.
3213 @item /@var{regexp}/[@var{offset}]
3214 Create an output file containing the current line up to (but not
3215 including) the next line of the input file that contains a match for
3216 @var{regexp}. The optional @var{offset} is an integer.
3217 If it is given, the input up to (but not including) the
3218 matching line plus or minus @var{offset} is put into the output file,
3219 and the line after that begins the next section of input.
3221 @item %@var{regexp}%[@var{offset}]
3222 Like the previous type, except that it does not create an output
3223 file, so that section of the input file is effectively ignored.
3225 @item @{@var{repeat-count}@}
3226 Repeat the previous pattern @var{repeat-count} additional
3227 times. The @var{repeat-count} can either be a positive integer or an
3228 asterisk, meaning repeat as many times as necessary until the input is
3233 The output files' names consist of a prefix (@samp{xx} by default)
3234 followed by a suffix. By default, the suffix is an ascending sequence
3235 of two-digit decimal numbers from @samp{00} to @samp{99}. In any case,
3236 concatenating the output files in sorted order by file name produces the
3237 original input file.
3239 By default, if @command{csplit} encounters an error or receives a hangup,
3240 interrupt, quit, or terminate signal, it removes any output files
3241 that it has created so far before it exits.
3243 The program accepts the following options. Also see @ref{Common options}.
3247 @item -f @var{prefix}
3248 @itemx --prefix=@var{prefix}
3251 @cindex output file name prefix
3252 Use @var{prefix} as the output file name prefix.
3254 @item -b @var{suffix}
3255 @itemx --suffix=@var{suffix}
3258 @cindex output file name suffix
3259 Use @var{suffix} as the output file name suffix. When this option is
3260 specified, the suffix string must include exactly one
3261 @code{printf(3)}-style conversion specification, possibly including
3262 format specification flags, a field width, a precision specifications,
3263 or all of these kinds of modifiers. The format letter must convert a
3264 binary unsigned integer argument to readable form. The format letters
3265 @samp{d} and @samp{i} are aliases for @samp{u}, and the
3266 @samp{u}, @samp{o}, @samp{x}, and @samp{X} conversions are allowed. The
3267 entire @var{suffix} is given (with the current output file number) to
3268 @code{sprintf(3)} to form the file name suffixes for each of the
3269 individual output files in turn. If this option is used, the
3270 @option{--digits} option is ignored.
3272 @item -n @var{digits}
3273 @itemx --digits=@var{digits}
3276 Use output file names containing numbers that are @var{digits} digits
3277 long instead of the default 2.
3282 @opindex --keep-files
3283 Do not remove output files when errors are encountered.
3286 @itemx --elide-empty-files
3288 @opindex --elide-empty-files
3289 Suppress the generation of zero-length output files. (In cases where
3290 the section delimiters of the input file are supposed to mark the first
3291 lines of each of the sections, the first output file will generally be a
3292 zero-length file unless you use this option.) The output file sequence
3293 numbers always run consecutively starting from 0, even when this option
3304 Do not print counts of output file sizes.
3310 Here is an example of its usage.
3311 First, create an empty directory for the exercise,
3318 Now, split the sequence of 1..14 on lines that end with 0 or 5:
3321 $ seq 14 | csplit - '/[05]$/' '@{*@}'
3327 Each number printed above is the size of an output
3328 file that csplit has just created.
3329 List the names of those output files:
3336 Use @command{head} to show their contents:
3361 @node Summarizing files
3362 @chapter Summarizing files
3364 @cindex summarizing files
3366 These commands generate just a few numbers representing entire
3370 * wc invocation:: Print newline, word, and byte counts.
3371 * sum invocation:: Print checksum and block counts.
3372 * cksum invocation:: Print CRC checksum and byte counts.
3373 * md5sum invocation:: Print or check MD5 digests.
3374 * sha1sum invocation:: Print or check SHA-1 digests.
3375 * sha2 utilities:: Print or check SHA-2 digests.
3380 @section @command{wc}: Print newline, word, and byte counts
3384 @cindex character count
3388 @command{wc} counts the number of bytes, characters, whitespace-separated
3389 words, and newlines in each given @var{file}, or standard input if none
3390 are given or for a @var{file} of @samp{-}. Synopsis:
3393 wc [@var{option}]@dots{} [@var{file}]@dots{}
3396 @cindex total counts
3397 @command{wc} prints one line of counts for each file, and if the file was
3398 given as an argument, it prints the file name following the counts. If
3399 more than one @var{file} is given, @command{wc} prints a final line
3400 containing the cumulative counts, with the file name @file{total}. The
3401 counts are printed in this order: newlines, words, characters, bytes,
3402 maximum line length.
3403 Each count is printed right-justified in a field with at least one
3404 space between fields so that the numbers and file names normally line
3405 up nicely in columns. The width of the count fields varies depending
3406 on the inputs, so you should not depend on a particular field width.
3407 However, as a @acronym{GNU} extension, if only one count is printed,
3408 it is guaranteed to be printed without leading spaces.
3410 By default, @command{wc} prints three counts: the newline, words, and byte
3411 counts. Options can specify that only certain counts be printed.
3412 Options do not undo others previously given, so
3419 prints both the byte counts and the word counts.
3421 With the @option{--max-line-length} option, @command{wc} prints the length
3422 of the longest line per file, and if there is more than one file it
3423 prints the maximum (not the sum) of those lengths. The line lengths here
3424 are measured in screen columns, according to the current locale and
3425 assuming tab positions in every 8th column.
3427 The program accepts the following options. Also see @ref{Common options}.
3435 Print only the byte counts.
3441 Print only the character counts.
3447 Print only the word counts.
3453 Print only the newline counts.
3456 @itemx --max-line-length
3458 @opindex --max-line-length
3459 Print only the maximum line lengths.
3461 @macro filesZeroFromOption{cmd,withTotalOption,subListOutput}
3462 @itemx --files0-from=@var{file}
3463 @opindex --files0-from=@var{file}
3464 @c This is commented out to avoid a texi2dvi failure.
3465 @c texi2dvi (GNU Texinfo 4.11) 1.104
3466 @c @cindex including files from @command{\cmd\}
3467 Disallow processing files named on the command line, and instead process
3468 those named in file @var{file}; each name being terminated by a zero byte
3469 (@acronym{ASCII} @sc{nul}).
3470 This is useful \withTotalOption\
3471 when the list of file names is so long that it may exceed a command line
3473 In such cases, running @command{\cmd\} via @command{xargs} is undesirable
3474 because it splits the list into pieces and makes @command{\cmd\} print
3475 \subListOutput\ for each sublist rather than for the entire list.
3476 One way to produce a list of @acronym{ASCII} @sc{nul} terminated file
3477 names is with @sc{gnu}
3478 @command{find}, using its @option{-print0} predicate.
3479 If @var{file} is @samp{-} then the @acronym{ASCII} @sc{nul} terminated
3480 file names are read from standard input.
3482 @filesZeroFromOption{wc,,a total}
3484 For example, to find the length of the longest line in any @file{.c} or
3485 @file{.h} file in the current hierarchy, do this:
3488 find . -name '*.[ch]' -print0 |
3489 wc -L --files0-from=- | tail -n1
3497 @node sum invocation
3498 @section @command{sum}: Print checksum and block counts
3501 @cindex 16-bit checksum
3502 @cindex checksum, 16-bit
3504 @command{sum} computes a 16-bit checksum for each given @var{file}, or
3505 standard input if none are given or for a @var{file} of @samp{-}. Synopsis:
3508 sum [@var{option}]@dots{} [@var{file}]@dots{}
3511 @command{sum} prints the checksum for each @var{file} followed by the
3512 number of blocks in the file (rounded up). If more than one @var{file}
3513 is given, file names are also printed (by default). (With the
3514 @option{--sysv} option, corresponding file names are printed when there is
3515 at least one file argument.)
3517 By default, @sc{gnu} @command{sum} computes checksums using an algorithm
3518 compatible with BSD @command{sum} and prints file sizes in units of
3521 The program accepts the following options. Also see @ref{Common options}.
3527 @cindex BSD @command{sum}
3528 Use the default (BSD compatible) algorithm. This option is included for
3529 compatibility with the System V @command{sum}. Unless @option{-s} was also
3530 given, it has no effect.
3536 @cindex System V @command{sum}
3537 Compute checksums using an algorithm compatible with System V
3538 @command{sum}'s default, and print file sizes in units of 512-byte blocks.
3542 @command{sum} is provided for compatibility; the @command{cksum} program (see
3543 next section) is preferable in new applications.
3548 @node cksum invocation
3549 @section @command{cksum}: Print CRC checksum and byte counts
3552 @cindex cyclic redundancy check
3553 @cindex CRC checksum
3555 @command{cksum} computes a cyclic redundancy check (CRC) checksum for each
3556 given @var{file}, or standard input if none are given or for a
3557 @var{file} of @samp{-}. Synopsis:
3560 cksum [@var{option}]@dots{} [@var{file}]@dots{}
3563 @command{cksum} prints the CRC checksum for each file along with the number
3564 of bytes in the file, and the file name unless no arguments were given.
3566 @command{cksum} is typically used to ensure that files
3567 transferred by unreliable means (e.g., netnews) have not been corrupted,
3568 by comparing the @command{cksum} output for the received files with the
3569 @command{cksum} output for the original files (typically given in the
3572 The CRC algorithm is specified by the @acronym{POSIX} standard. It is not
3573 compatible with the BSD or System V @command{sum} algorithms (see the
3574 previous section); it is more robust.
3576 The only options are @option{--help} and @option{--version}. @xref{Common
3582 @node md5sum invocation
3583 @section @command{md5sum}: Print or check MD5 digests
3587 @cindex 128-bit checksum
3588 @cindex checksum, 128-bit
3589 @cindex fingerprint, 128-bit
3590 @cindex message-digest, 128-bit
3592 @command{md5sum} computes a 128-bit checksum (or @dfn{fingerprint} or
3593 @dfn{message-digest}) for each specified @var{file}.
3595 Note: The MD5 digest is more reliable than a simple CRC (provided by
3596 the @command{cksum} command) for detecting accidental file corruption,
3597 as the chances of accidentally having two files with identical MD5
3598 are vanishingly small. However, it should not be considered secure
3599 against malicious tampering: although finding a file with a given MD5
3600 fingerprint is considered infeasible at the moment, it is known how
3601 to modify certain files, including digital certificates, so that they
3602 appear valid when signed with an MD5 digest.
3603 For more secure hashes, consider using SHA-2. @xref{sha2 utilities}.
3605 If a @var{file} is specified as @samp{-} or if no files are given
3606 @command{md5sum} computes the checksum for the standard input.
3607 @command{md5sum} can also determine whether a file and checksum are
3608 consistent. Synopsis:
3611 md5sum [@var{option}]@dots{} [@var{file}]@dots{}
3614 For each @var{file}, @samp{md5sum} outputs the MD5 checksum, a flag
3615 indicating binary or text input mode, and the file name.
3616 If @var{file} contains a backslash or newline, the
3617 line is started with a backslash, and each problematic character in
3618 the file name is escaped with a backslash, making the output
3619 unambiguous even in the presence of arbitrary file names.
3620 If @var{file} is omitted or specified as @samp{-}, standard input is read.
3622 The program accepts the following options. Also see @ref{Common options}.
3630 @cindex binary input files
3631 Treat each input file as binary, by reading it in binary mode and
3632 outputting a @samp{*} flag. This is the inverse of @option{--text}.
3633 On systems like @acronym{GNU} that do not distinguish between binary
3634 and text files, this option merely flags each input mode as binary:
3635 the MD5 checksum is unaffected. This option is the default on systems
3636 like MS-DOS that distinguish between binary and text files, except
3637 for reading standard input when standard input is a terminal.
3641 Read file names and checksum information (not data) from each
3642 @var{file} (or from stdin if no @var{file} was specified) and report
3643 whether the checksums match the contents of the named files.
3644 The input to this mode of @command{md5sum} is usually the output of
3645 a prior, checksum-generating run of @samp{md5sum}.
3646 Each valid line of input consists of an MD5 checksum, a binary/text
3647 flag, and then a file name.
3648 Binary mode is indicated with @samp{*}, text with @samp{ } (space).
3649 For each such line, @command{md5sum} reads the named file and computes its
3650 MD5 checksum. Then, if the computed message digest does not match the
3651 one on the line with the file name, the file is noted as having
3652 failed the test. Otherwise, the file passes the test.
3653 By default, for each valid line, one line is written to standard
3654 output indicating whether the named file passed the test.
3655 After all checks have been performed, if there were any failures,
3656 a warning is issued to standard error.
3657 Use the @option{--status} option to inhibit that output.
3658 If any listed file cannot be opened or read, if any valid line has
3659 an MD5 checksum inconsistent with the associated file, or if no valid
3660 line is found, @command{md5sum} exits with nonzero status. Otherwise,
3661 it exits successfully.
3665 @cindex verifying MD5 checksums
3666 This option is useful only when verifying checksums.
3667 When verifying checksums, don't generate an 'OK' message per successfully
3668 checked file. Files that fail the verification are reported in the
3669 default one-line-per-file format. If there is any checksum mismatch,
3670 print a warning summarizing the failures to standard error.
3674 @cindex verifying MD5 checksums
3675 This option is useful only when verifying checksums.
3676 When verifying checksums, don't generate the default one-line-per-file
3677 diagnostic and don't output the warning summarizing any failures.
3678 Failures to open or read a file still evoke individual diagnostics to
3680 If all listed files are readable and are consistent with the associated
3681 MD5 checksums, exit successfully. Otherwise exit with a status code
3682 indicating there was a failure.
3688 @cindex text input files
3689 Treat each input file as text, by reading it in text mode and
3690 outputting a @samp{ } flag. This is the inverse of @option{--binary}.
3691 This option is the default on systems like @acronym{GNU} that do not
3692 distinguish between binary and text files. On other systems, it is
3693 the default for reading standard input when standard input is a
3700 @cindex verifying MD5 checksums
3701 When verifying checksums, warn about improperly formatted MD5 checksum lines.
3702 This option is useful only if all but a few lines in the checked input
3707 @cindex verifying MD5 checksums
3708 When verifying checksums,
3709 if one or more input line is invalid,
3710 exit nonzero after all warnings have been issued.
3717 @node sha1sum invocation
3718 @section @command{sha1sum}: Print or check SHA-1 digests
3722 @cindex 160-bit checksum
3723 @cindex checksum, 160-bit
3724 @cindex fingerprint, 160-bit
3725 @cindex message-digest, 160-bit
3727 @command{sha1sum} computes a 160-bit checksum for each specified
3728 @var{file}. The usage and options of this command are precisely the
3729 same as for @command{md5sum}. @xref{md5sum invocation}.
3731 Note: The SHA-1 digest is more secure than MD5, and no collisions of
3732 it are known (different files having the same fingerprint). However,
3733 it is known that they can be produced with considerable, but not
3734 unreasonable, resources. For this reason, it is generally considered
3735 that SHA-1 should be gradually phased out in favor of the more secure
3736 SHA-2 hash algorithms. @xref{sha2 utilities}.
3739 @node sha2 utilities
3740 @section sha2 utilities: Print or check SHA-2 digests
3747 @cindex 224-bit checksum
3748 @cindex 256-bit checksum
3749 @cindex 384-bit checksum
3750 @cindex 512-bit checksum
3751 @cindex checksum, 224-bit
3752 @cindex checksum, 256-bit
3753 @cindex checksum, 384-bit
3754 @cindex checksum, 512-bit
3755 @cindex fingerprint, 224-bit
3756 @cindex fingerprint, 256-bit
3757 @cindex fingerprint, 384-bit
3758 @cindex fingerprint, 512-bit
3759 @cindex message-digest, 224-bit
3760 @cindex message-digest, 256-bit
3761 @cindex message-digest, 384-bit
3762 @cindex message-digest, 512-bit
3764 The commands @command{sha224sum}, @command{sha256sum},
3765 @command{sha384sum} and @command{sha512sum} compute checksums of
3766 various lengths (respectively 224, 256, 384 and 512 bits),
3767 collectively known as the SHA-2 hashes. The usage and options of
3768 these commands are precisely the same as for @command{md5sum}.
3769 @xref{md5sum invocation}.
3771 Note: The SHA384 and SHA512 digests are considerably slower to
3772 compute, especially on 32-bit computers, than SHA224 or SHA256.
3775 @node Operating on sorted files
3776 @chapter Operating on sorted files
3778 @cindex operating on sorted files
3779 @cindex sorted files, operations on
3781 These commands work with (or produce) sorted files.
3784 * sort invocation:: Sort text files.
3785 * shuf invocation:: Shuffle text files.
3786 * uniq invocation:: Uniquify files.
3787 * comm invocation:: Compare two sorted files line by line.
3788 * ptx invocation:: Produce a permuted index of file contents.
3789 * tsort invocation:: Topological sort.
3793 @node sort invocation
3794 @section @command{sort}: Sort text files
3797 @cindex sorting files
3799 @command{sort} sorts, merges, or compares all the lines from the given
3800 files, or standard input if none are given or for a @var{file} of
3801 @samp{-}. By default, @command{sort} writes the results to standard
3805 sort [@var{option}]@dots{} [@var{file}]@dots{}
3808 @command{sort} has three modes of operation: sort (the default), merge,
3809 and check for sortedness. The following options change the operation
3816 @itemx --check=diagnose-first
3819 @cindex checking for sortedness
3820 Check whether the given file is already sorted: if it is not all
3821 sorted, print a diagnostic containing the first out-of-order line and
3822 exit with a status of 1.
3823 Otherwise, exit successfully.
3824 At most one input file can be given.
3827 @itemx --check=quiet
3828 @itemx --check=silent
3831 @cindex checking for sortedness
3832 Exit successfully if the given file is already sorted, and
3833 exit with status 1 otherwise.
3834 At most one input file can be given.
3835 This is like @option{-c}, except it does not print a diagnostic.
3841 @cindex merging sorted files
3842 Merge the given files by sorting them as a group. Each input file must
3843 always be individually sorted. It always works to sort instead of
3844 merge; merging is provided because it is faster, in the case where it
3849 @cindex sort stability
3850 @cindex sort's last-resort comparison
3851 A pair of lines is compared as follows:
3852 @command{sort} compares each pair of fields, in the
3853 order specified on the command line, according to the associated
3854 ordering options, until a difference is found or no fields are left.
3855 If no key fields are specified, @command{sort} uses a default key of
3856 the entire line. Finally, as a last resort when all keys compare
3857 equal, @command{sort} compares entire lines as if no ordering options
3858 other than @option{--reverse} (@option{-r}) were specified. The
3859 @option{--stable} (@option{-s}) option disables this @dfn{last-resort
3860 comparison} so that lines in which all fields compare equal are left
3861 in their original relative order. The @option{--unique}
3862 (@option{-u}) option also disables the last-resort comparison.
3866 Unless otherwise specified, all comparisons use the character collating
3867 sequence specified by the @env{LC_COLLATE} locale.@footnote{If you
3868 use a non-@acronym{POSIX} locale (e.g., by setting @env{LC_ALL}
3869 to @samp{en_US}), then @command{sort} may produce output that is sorted
3870 differently than you're accustomed to. In that case, set the @env{LC_ALL}
3871 environment variable to @samp{C}. Note that setting only @env{LC_COLLATE}
3872 has two problems. First, it is ineffective if @env{LC_ALL} is also set.
3873 Second, it has undefined behavior if @env{LC_CTYPE} (or @env{LANG}, if
3874 @env{LC_CTYPE} is unset) is set to an incompatible value. For example,
3875 you get undefined behavior if @env{LC_CTYPE} is @code{ja_JP.PCK} but
3876 @env{LC_COLLATE} is @code{en_US.UTF-8}.}
3878 @sc{gnu} @command{sort} (as specified for all @sc{gnu} utilities) has no
3879 limit on input line length or restrictions on bytes allowed within lines.
3880 In addition, if the final byte of an input file is not a newline, @sc{gnu}
3881 @command{sort} silently supplies one. A line's trailing newline is not
3882 part of the line for comparison purposes.
3884 @cindex exit status of @command{sort}
3888 0 if no error occurred
3889 1 if invoked with @option{-c} or @option{-C} and the input is not sorted
3890 2 if an error occurred
3894 If the environment variable @env{TMPDIR} is set, @command{sort} uses its
3895 value as the directory for temporary files instead of @file{/tmp}. The
3896 @option{--temporary-directory} (@option{-T}) option in turn overrides
3897 the environment variable.
3899 The following options affect the ordering of output lines. They may be
3900 specified globally or as part of a specific key field. If no key
3901 fields are specified, global options apply to comparison of entire
3902 lines; otherwise the global options are inherited by key fields that do
3903 not specify any special options of their own. In pre-@acronym{POSIX}
3904 versions of @command{sort}, global options affect only later key fields,
3905 so portable shell scripts should specify global options first.
3910 @itemx --ignore-leading-blanks
3912 @opindex --ignore-leading-blanks
3913 @cindex blanks, ignoring leading
3915 Ignore leading blanks when finding sort keys in each line.
3916 By default a blank is a space or a tab, but the @env{LC_CTYPE} locale
3917 can change this. Note blanks may be ignored by your locale's collating
3918 rules, but without this option they will be significant for character
3919 positions specified in keys with the @option{-k} option.
3922 @itemx --dictionary-order
3924 @opindex --dictionary-order
3925 @cindex dictionary order
3926 @cindex phone directory order
3927 @cindex telephone directory order
3929 Sort in @dfn{phone directory} order: ignore all characters except
3930 letters, digits and blanks when sorting.
3931 By default letters and digits are those of @acronym{ASCII} and a blank
3932 is a space or a tab, but the @env{LC_CTYPE} locale can change this.
3935 @itemx --ignore-case
3937 @opindex --ignore-case
3938 @cindex ignoring case
3939 @cindex case folding
3941 Fold lowercase characters into the equivalent uppercase characters when
3942 comparing so that, for example, @samp{b} and @samp{B} sort as equal.
3943 The @env{LC_CTYPE} locale determines character types.
3944 When used with @option{--unique} those lower case equivalent lines are
3945 thrown away. (There is currently no way to throw away the upper case
3946 equivalent instead. (Any @option{--reverse} given would only affect
3947 the final result, after the throwing away.))
3950 @itemx --general-numeric-sort
3951 @itemx --sort=general-numeric
3953 @opindex --general-numeric-sort
3955 @cindex general numeric sort
3957 Sort numerically, converting a prefix of each line to a long
3958 double-precision floating point number. @xref{Floating point}.
3959 Do not report overflow, underflow, or conversion errors.
3960 Use the following collating sequence:
3964 Lines that do not start with numbers (all considered to be equal).
3966 NaNs (``Not a Number'' values, in IEEE floating point arithmetic)
3967 in a consistent but machine-dependent order.
3971 Finite numbers in ascending numeric order (with @math{-0} and @math{+0} equal).
3976 Use this option only if there is no alternative; it is much slower than
3977 @option{--numeric-sort} (@option{-n}) and it can lose information when
3978 converting to floating point.
3981 @itemx --human-numeric-sort
3982 @itemx --sort=human-numeric
3984 @opindex --human-numeric-sort
3986 @cindex human numeric sort
3988 Sort numerically, first by numeric sign (negative, zero, or positive);
3989 then by @acronym{SI} suffix (either empty, or @samp{k} or @samp{K}, or
3990 one of @samp{MGTPEZY}, in that order; @pxref{Block size}); and finally
3991 by numeric value. For example, @samp{1023M} sorts before @samp{1G}
3992 because @samp{M} (mega) precedes @samp{G} (giga) as an @acronym{SI}
3993 suffix. This option sorts values that are consistently scaled to the
3994 nearest suffix, regardless of whether suffixes denote powers of 1000
3995 or 1024, and it therefore sorts the output of any single invocation of
3996 the @command{df}, @command{du}, or @command{ls} commands that are
3997 invoked with their @option{--human-readable} or @option{--si} options.
3998 The syntax for numbers is the same as for the @option{--numeric-sort}
3999 option; the @acronym{SI} suffix must immediately follow the number.
4002 @itemx --ignore-nonprinting
4004 @opindex --ignore-nonprinting
4005 @cindex nonprinting characters, ignoring
4006 @cindex unprintable characters, ignoring
4008 Ignore nonprinting characters.
4009 The @env{LC_CTYPE} locale determines character types.
4010 This option has no effect if the stronger @option{--dictionary-order}
4011 (@option{-d}) option is also given.
4017 @opindex --month-sort
4019 @cindex months, sorting by
4021 An initial string, consisting of any amount of blanks, followed
4022 by a month name abbreviation, is folded to UPPER case and
4023 compared in the order @samp{JAN} < @samp{FEB} < @dots{} < @samp{DEC}.
4024 Invalid names compare low to valid names. The @env{LC_TIME} locale
4025 category determines the month spellings.
4026 By default a blank is a space or a tab, but the @env{LC_CTYPE} locale
4030 @itemx --numeric-sort
4031 @itemx --sort=numeric
4033 @opindex --numeric-sort
4035 @cindex numeric sort
4037 Sort numerically. The number begins each line and consists
4038 of optional blanks, an optional @samp{-} sign, and zero or more
4039 digits possibly separated by thousands separators, optionally followed
4040 by a decimal-point character and zero or more digits. An empty
4041 number is treated as @samp{0}. The @env{LC_NUMERIC}
4042 locale specifies the decimal-point character and thousands separator.
4043 By default a blank is a space or a tab, but the @env{LC_CTYPE} locale
4046 Comparison is exact; there is no rounding error.
4048 Neither a leading @samp{+} nor exponential notation is recognized.
4049 To compare such strings numerically, use the
4050 @option{--general-numeric-sort} (@option{-g}) option.
4053 @itemx --version-sort
4055 @opindex --version-sort
4056 @cindex version number sort
4057 Sort by version name and number. It behaves like a standard sort,
4058 except that each sequence of decimal digits is treated numerically
4059 as an index/version number. (@xref{Details about version sort}.)
4065 @cindex reverse sorting
4066 Reverse the result of comparison, so that lines with greater key values
4067 appear earlier in the output instead of later.
4070 @itemx --random-sort
4071 @itemx --sort=random
4073 @opindex --random-sort
4076 Sort by hashing the input keys and then sorting the hash values.
4077 Choose the hash function at random, ensuring that it is free of
4078 collisions so that differing keys have differing hash values. This is
4079 like a random permutation of the inputs (@pxref{shuf invocation}),
4080 except that keys with the same value sort together.
4082 If multiple random sort fields are specified, the same random hash
4083 function is used for all fields. To use different random hash
4084 functions for different fields, you can invoke @command{sort} more
4087 The choice of hash function is affected by the
4088 @option{--random-source} option.
4096 @item --compress-program=@var{prog}
4097 Compress any temporary files with the program @var{prog}.
4099 With no arguments, @var{prog} must compress standard input to standard
4100 output, and when given the @option{-d} option it must decompress
4101 standard input to standard output.
4103 Terminate with an error if @var{prog} exits with nonzero status.
4105 White space and the backslash character should not appear in
4106 @var{prog}; they are reserved for future use.
4108 @filesZeroFromOption{sort,,sorted output}
4110 @item -k @var{pos1}[,@var{pos2}]
4111 @itemx --key=@var{pos1}[,@var{pos2}]
4115 Specify a sort field that consists of the part of the line between
4116 @var{pos1} and @var{pos2} (or the end of the line, if @var{pos2} is
4117 omitted), @emph{inclusive}.
4119 Each @var{pos} has the form @samp{@var{f}[.@var{c}][@var{opts}]},
4120 where @var{f} is the number of the field to use, and @var{c} is the number
4121 of the first character from the beginning of the field. Fields and character
4122 positions are numbered starting with 1; a character position of zero in
4123 @var{pos2} indicates the field's last character. If @samp{.@var{c}} is
4124 omitted from @var{pos1}, it defaults to 1 (the beginning of the field);
4125 if omitted from @var{pos2}, it defaults to 0 (the end of the field).
4126 @var{opts} are ordering options, allowing individual keys to be sorted
4127 according to different rules; see below for details. Keys can span
4130 Example: To sort on the second field, use @option{--key=2,2}
4131 (@option{-k 2,2}). See below for more notes on keys and more examples.
4132 See also the @option{--debug} option to help determine the part
4133 of the line being used in the sort.
4136 Highlight the portion of each line used for sorting.
4137 Also issue warnings about questionable usage to stderr.
4139 @item --batch-size=@var{nmerge}
4140 @opindex --batch-size
4141 @cindex number of inputs to merge, nmerge
4142 Merge at most @var{nmerge} inputs at once.
4144 When @command{sort} has to merge more than @var{nmerge} inputs,
4145 it merges them in groups of @var{nmerge}, saving the result in
4146 a temporary file, which is then used as an input in a subsequent merge.
4148 A large value of @var{nmerge} may improve merge performance and decrease
4149 temporary storage utilization at the expense of increased memory usage
4150 and I/O. Conversely a small value of @var{nmerge} may reduce memory
4151 requirements and I/O at the expense of temporary storage consumption and
4154 The value of @var{nmerge} must be at least 2. The default value is
4155 currently 16, but this is implementation-dependent and may change in
4158 The value of @var{nmerge} may be bounded by a resource limit for open
4159 file descriptors. The commands @samp{ulimit -n} or @samp{getconf
4160 OPEN_MAX} may display limits for your systems; these limits may be
4161 modified further if your program already has some files open, or if
4162 the operating system has other limits on the number of open files. If
4163 the value of @var{nmerge} exceeds the resource limit, @command{sort}
4164 silently uses a smaller value.
4166 @item -o @var{output-file}
4167 @itemx --output=@var{output-file}
4170 @cindex overwriting of input, allowed
4171 Write output to @var{output-file} instead of standard output.
4172 Normally, @command{sort} reads all input before opening
4173 @var{output-file}, so you can safely sort a file in place by using
4174 commands like @code{sort -o F F} and @code{cat F | sort -o F}.
4175 However, @command{sort} with @option{--merge} (@option{-m}) can open
4176 the output file before reading all input, so a command like @code{cat
4177 F | sort -m -o F - G} is not safe as @command{sort} might start
4178 writing @file{F} before @command{cat} is done reading it.
4180 @vindex POSIXLY_CORRECT
4181 On newer systems, @option{-o} cannot appear after an input file if
4182 @env{POSIXLY_CORRECT} is set, e.g., @samp{sort F -o F}. Portable
4183 scripts should specify @option{-o @var{output-file}} before any input
4186 @item --random-source=@var{file}
4187 @opindex --random-source
4188 @cindex random source for sorting
4189 Use @var{file} as a source of random data used to determine which
4190 random hash function to use with the @option{-R} option. @xref{Random
4197 @cindex sort stability
4198 @cindex sort's last-resort comparison
4200 Make @command{sort} stable by disabling its last-resort comparison.
4201 This option has no effect if no fields or global ordering options
4202 other than @option{--reverse} (@option{-r}) are specified.
4205 @itemx --buffer-size=@var{size}
4207 @opindex --buffer-size
4208 @cindex size for main memory sorting
4209 Use a main-memory sort buffer of the given @var{size}. By default,
4210 @var{size} is in units of 1024 bytes. Appending @samp{%} causes
4211 @var{size} to be interpreted as a percentage of physical memory.
4212 Appending @samp{K} multiplies @var{size} by 1024 (the default),
4213 @samp{M} by 1,048,576, @samp{G} by 1,073,741,824, and so on for
4214 @samp{T}, @samp{P}, @samp{E}, @samp{Z}, and @samp{Y}. Appending
4215 @samp{b} causes @var{size} to be interpreted as a byte count, with no
4218 This option can improve the performance of @command{sort} by causing it
4219 to start with a larger or smaller sort buffer than the default.
4220 However, this option affects only the initial buffer size. The buffer
4221 grows beyond @var{size} if @command{sort} encounters input lines larger
4224 @item -t @var{separator}
4225 @itemx --field-separator=@var{separator}
4227 @opindex --field-separator
4228 @cindex field separator character
4229 Use character @var{separator} as the field separator when finding the
4230 sort keys in each line. By default, fields are separated by the empty
4231 string between a non-blank character and a blank character.
4232 By default a blank is a space or a tab, but the @env{LC_CTYPE} locale
4235 That is, given the input line @w{@samp{ foo bar}}, @command{sort} breaks it
4236 into fields @w{@samp{ foo}} and @w{@samp{ bar}}. The field separator is
4237 not considered to be part of either the field preceding or the field
4238 following, so with @samp{sort @w{-t " "}} the same input line has
4239 three fields: an empty field, @samp{foo}, and @samp{bar}.
4240 However, fields that extend to the end of the line,
4241 as @option{-k 2}, or fields consisting of a range, as @option{-k 2,3},
4242 retain the field separators present between the endpoints of the range.
4244 To specify @acronym{ASCII} @sc{nul} as the field separator,
4245 use the two-character string @samp{\0}, e.g., @samp{sort -t '\0'}.
4247 @item -T @var{tempdir}
4248 @itemx --temporary-directory=@var{tempdir}
4250 @opindex --temporary-directory
4251 @cindex temporary directory
4253 Use directory @var{tempdir} to store temporary files, overriding the
4254 @env{TMPDIR} environment variable. If this option is given more than
4255 once, temporary files are stored in all the directories given. If you
4256 have a large sort or merge that is I/O-bound, you can often improve
4257 performance by using this option to specify directories on different
4258 disks and controllers.
4260 @item --parallel=@var{n}
4262 @cindex multithreaded sort
4263 Set the number of sorts run in parallel to @var{n}. By default,
4264 @var{n} is set to the number of available processors, but limited
4265 to 8, as there are diminishing performance gains after that.
4266 Note also that using @var{n} threads increases the memory usage by
4267 a factor of log @var{n}. Also see @ref{nproc invocation}.
4273 @cindex uniquifying output
4275 Normally, output only the first of a sequence of lines that compare
4276 equal. For the @option{--check} (@option{-c} or @option{-C}) option,
4277 check that no pair of consecutive lines compares equal.
4279 This option also disables the default last-resort comparison.
4281 The commands @code{sort -u} and @code{sort | uniq} are equivalent, but
4282 this equivalence does not extend to arbitrary @command{sort} options.
4283 For example, @code{sort -n -u} inspects only the value of the initial
4284 numeric string when checking for uniqueness, whereas @code{sort -n |
4285 uniq} inspects the entire line. @xref{uniq invocation}.
4287 @macro zeroTerminatedOption
4289 @itemx --zero-terminated
4291 @opindex --zero-terminated
4292 @cindex process zero-terminated items
4293 Delimit items with a zero byte rather than a newline (@acronym{ASCII} @sc{lf}).
4294 I.E. treat input as items separated by @acronym{ASCII} @sc{nul}
4295 and terminate output items with @acronym{ASCII} @sc{nul}.
4296 This option can be useful in conjunction with @samp{perl -0} or
4297 @samp{find -print0} and @samp{xargs -0} which do the same in order to
4298 reliably handle arbitrary file names (even those containing blanks
4299 or other special characters).
4301 @zeroTerminatedOption
4305 Historical (BSD and System V) implementations of @command{sort} have
4306 differed in their interpretation of some options, particularly
4307 @option{-b}, @option{-f}, and @option{-n}.
4308 @sc{gnu} sort follows the @acronym{POSIX}
4309 behavior, which is usually (but not always!) like the System V behavior.
4310 According to @acronym{POSIX}, @option{-n} no longer implies @option{-b}. For
4311 consistency, @option{-M} has been changed in the same way. This may
4312 affect the meaning of character positions in field specifications in
4313 obscure cases. The only fix is to add an explicit @option{-b}.
4315 A position in a sort field specified with @option{-k} may have any
4316 of the option letters @samp{MbdfghinRrV} appended to it, in which case no
4317 global ordering options are inherited by that particular field. The
4318 @option{-b} option may be independently attached to either or both of
4319 the start and end positions of a field specification, and if it is
4320 inherited from the global options it will be attached to both.
4321 If input lines can contain leading or adjacent blanks and @option{-t}
4322 is not used, then @option{-k} is typically combined with @option{-b} or
4323 an option that implicitly ignores leading blanks (@samp{Mghn}) as otherwise
4324 the varying numbers of leading blanks in fields can cause confusing results.
4326 If the start position in a sort field specifier falls after the end of
4327 the line or after the end field, the field is empty. If the @option{-b}
4328 option was specified, the @samp{.@var{c}} part of a field specification
4329 is counted from the first nonblank character of the field.
4331 @vindex _POSIX2_VERSION
4332 @vindex POSIXLY_CORRECT
4333 On older systems, @command{sort} supports an obsolete origin-zero
4334 syntax @samp{+@var{pos1} [-@var{pos2}]} for specifying sort keys.
4335 The obsolete sequence @samp{sort +@var{a}.@var{x} -@var{b}.@var{y}}
4336 is equivalent to @samp{sort -k @var{a+1}.@var{x+1},@var{b}} if @var{y}
4337 is @samp{0} or absent, otherwise it is equivalent to @samp{sort -k
4338 @var{a+1}.@var{x+1},@var{b+1}.@var{y}}.
4340 This obsolete behavior can be enabled or disabled with the
4341 @env{_POSIX2_VERSION} environment variable (@pxref{Standards
4342 conformance}); it can also be enabled when @env{POSIXLY_CORRECT} is
4343 not set by using the obsolete syntax with @samp{-@var{pos2}} present.
4345 Scripts intended for use on standard hosts should avoid obsolete
4346 syntax and should use @option{-k} instead. For example, avoid
4347 @samp{sort +2}, since it might be interpreted as either @samp{sort
4348 ./+2} or @samp{sort -k 3}. If your script must also run on hosts that
4349 support only the obsolete syntax, it can use a test like @samp{if sort
4350 -k 1 </dev/null >/dev/null 2>&1; then @dots{}} to decide which syntax
4353 Here are some examples to illustrate various combinations of options.
4358 Sort in descending (reverse) numeric order.
4365 Run no more than 4 sorts concurrently, using a buffer size of 10M.
4368 sort --parallel=4 -S 10M
4372 Sort alphabetically, omitting the first and second fields
4373 and the blanks at the start of the third field.
4374 This uses a single key composed of the characters beginning
4375 at the start of the first nonblank character in field three
4376 and extending to the end of each line.
4383 Sort numerically on the second field and resolve ties by sorting
4384 alphabetically on the third and fourth characters of field five.
4385 Use @samp{:} as the field delimiter.
4388 sort -t : -k 2,2n -k 5.3,5.4
4391 Note that if you had written @option{-k 2n} instead of @option{-k 2,2n}
4392 @command{sort} would have used all characters beginning in the second field
4393 and extending to the end of the line as the primary @emph{numeric}
4394 key. For the large majority of applications, treating keys spanning
4395 more than one field as numeric will not do what you expect.
4397 Also note that the @samp{n} modifier was applied to the field-end
4398 specifier for the first key. It would have been equivalent to
4399 specify @option{-k 2n,2} or @option{-k 2n,2n}. All modifiers except
4400 @samp{b} apply to the associated @emph{field}, regardless of whether
4401 the modifier character is attached to the field-start and/or the
4402 field-end part of the key specifier.
4405 Sort the password file on the fifth field and ignore any
4406 leading blanks. Sort lines with equal values in field five
4407 on the numeric user ID in field three. Fields are separated
4411 sort -t : -k 5b,5 -k 3,3n /etc/passwd
4412 sort -t : -n -k 5b,5 -k 3,3 /etc/passwd
4413 sort -t : -b -k 5,5 -k 3,3n /etc/passwd
4416 These three commands have equivalent effect. The first specifies that
4417 the first key's start position ignores leading blanks and the second
4418 key is sorted numerically. The other two commands rely on global
4419 options being inherited by sort keys that lack modifiers. The inheritance
4420 works in this case because @option{-k 5b,5b} and @option{-k 5b,5} are
4421 equivalent, as the location of a field-end lacking a @samp{.@var{c}}
4422 character position is not affected by whether initial blanks are
4426 Sort a set of log files, primarily by IPv4 address and secondarily by
4427 time stamp. If two lines' primary and secondary keys are identical,
4428 output the lines in the same order that they were input. The log
4429 files contain lines that look like this:
4432 4.150.156.3 - - [01/Apr/2004:06:31:51 +0000] message 1
4433 211.24.3.231 - - [24/Apr/2004:20:17:39 +0000] message 2
4436 Fields are separated by exactly one space. Sort IPv4 addresses
4437 lexicographically, e.g., 212.61.52.2 sorts before 212.129.233.201
4438 because 61 is less than 129.
4441 sort -s -t ' ' -k 4.9n -k 4.5M -k 4.2n -k 4.14,4.21 file*.log |
4442 sort -s -t '.' -k 1,1n -k 2,2n -k 3,3n -k 4,4n
4445 This example cannot be done with a single @command{sort} invocation,
4446 since IPv4 address components are separated by @samp{.} while dates
4447 come just after a space. So it is broken down into two invocations of
4448 @command{sort}: the first sorts by time stamp and the second by IPv4
4449 address. The time stamp is sorted by year, then month, then day, and
4450 finally by hour-minute-second field, using @option{-k} to isolate each
4451 field. Except for hour-minute-second there's no need to specify the
4452 end of each key field, since the @samp{n} and @samp{M} modifiers sort
4453 based on leading prefixes that cannot cross field boundaries. The
4454 IPv4 addresses are sorted lexicographically. The second sort uses
4455 @samp{-s} so that ties in the primary key are broken by the secondary
4456 key; the first sort uses @samp{-s} so that the combination of the two
4460 Generate a tags file in case-insensitive sorted order.
4463 find src -type f -print0 | sort -z -f | xargs -0 etags --append
4466 The use of @option{-print0}, @option{-z}, and @option{-0} in this case means
4467 that file names that contain blanks or other special characters are
4469 by the sort operation.
4471 @c This example is a bit contrived and needs more explanation.
4473 @c Sort records separated by an arbitrary string by using a pipe to convert
4474 @c each record delimiter string to @samp{\0}, then using sort's -z option,
4475 @c and converting each @samp{\0} back to the original record delimiter.
4478 @c printf 'c\n\nb\n\na\n' |
4479 @c perl -0pe 's/\n\n/\n\0/g' |
4481 @c perl -0pe 's/\0/\n/g'
4485 Use the common @acronym{DSU, Decorate Sort Undecorate} idiom to
4486 sort lines according to their length.
4489 awk '@{print length, $0@}' /etc/passwd | sort -n | cut -f2- -d' '
4492 In general this technique can be used to sort data that the @command{sort}
4493 command does not support, or is inefficient at, sorting directly.
4496 Shuffle a list of directories, but preserve the order of files within
4497 each directory. For instance, one could use this to generate a music
4498 playlist in which albums are shuffled but the songs of each album are
4502 ls */* | sort -t / -k 1,1R -k 2,2
4508 @node shuf invocation
4509 @section @command{shuf}: Shuffling text
4512 @cindex shuffling files
4514 @command{shuf} shuffles its input by outputting a random permutation
4515 of its input lines. Each output permutation is equally likely.
4519 shuf [@var{option}]@dots{} [@var{file}]
4520 shuf -e [@var{option}]@dots{} [@var{arg}]@dots{}
4521 shuf -i @var{lo}-@var{hi} [@var{option}]@dots{}
4524 @command{shuf} has three modes of operation that affect where it
4525 obtains its input lines. By default, it reads lines from standard
4526 input. The following options change the operation mode:
4534 @cindex command-line operands to shuffle
4535 Treat each command-line operand as an input line.
4537 @item -i @var{lo}-@var{hi}
4538 @itemx --input-range=@var{lo}-@var{hi}
4540 @opindex --input-range
4541 @cindex input range to shuffle
4542 Act as if input came from a file containing the range of unsigned
4543 decimal integers @var{lo}@dots{}@var{hi}, one per line.
4547 @command{shuf}'s other options can affect its behavior in all
4552 @item -n @var{lines}
4553 @itemx --head-count=@var{count}
4555 @opindex --head-count
4556 @cindex head of output
4557 Output at most @var{count} lines. By default, all input lines are
4560 @item -o @var{output-file}
4561 @itemx --output=@var{output-file}
4564 @cindex overwriting of input, allowed
4565 Write output to @var{output-file} instead of standard output.
4566 @command{shuf} reads all input before opening
4567 @var{output-file}, so you can safely shuffle a file in place by using
4568 commands like @code{shuf -o F <F} and @code{cat F | shuf -o F}.
4570 @item --random-source=@var{file}
4571 @opindex --random-source
4572 @cindex random source for shuffling
4573 Use @var{file} as a source of random data used to determine which
4574 permutation to generate. @xref{Random sources}.
4576 @zeroTerminatedOption
4592 might produce the output
4602 Similarly, the command:
4605 shuf -e clubs hearts diamonds spades
4619 and the command @samp{shuf -i 1-4} might output:
4629 These examples all have four input lines, so @command{shuf} might
4630 produce any of the twenty-four possible permutations of the input. In
4631 general, if there are @var{n} input lines, there are @var{n}! (i.e.,
4632 @var{n} factorial, or @var{n} * (@var{n} - 1) * @dots{} * 1) possible
4633 output permutations.
4638 @node uniq invocation
4639 @section @command{uniq}: Uniquify files
4642 @cindex uniquify files
4644 @command{uniq} writes the unique lines in the given @file{input}, or
4645 standard input if nothing is given or for an @var{input} name of
4649 uniq [@var{option}]@dots{} [@var{input} [@var{output}]]
4652 By default, @command{uniq} prints its input lines, except that
4653 it discards all but the first of adjacent repeated lines, so that
4654 no output lines are repeated. Optionally, it can instead discard
4655 lines that are not repeated, or all repeated lines.
4657 The input need not be sorted, but repeated input lines are detected
4658 only if they are adjacent. If you want to discard non-adjacent
4659 duplicate lines, perhaps you want to use @code{sort -u}.
4660 @xref{sort invocation}.
4663 Comparisons honor the rules specified by the @env{LC_COLLATE}
4666 If no @var{output} file is specified, @command{uniq} writes to standard
4669 The program accepts the following options. Also see @ref{Common options}.
4674 @itemx --skip-fields=@var{n}
4676 @opindex --skip-fields
4677 Skip @var{n} fields on each line before checking for uniqueness. Use
4678 a null string for comparison if a line has fewer than @var{n} fields. Fields
4679 are sequences of non-space non-tab characters that are separated from
4680 each other by at least one space or tab.
4682 For compatibility @command{uniq} supports an obsolete option syntax
4683 @option{-@var{n}}. New scripts should use @option{-f @var{n}} instead.
4686 @itemx --skip-chars=@var{n}
4688 @opindex --skip-chars
4689 Skip @var{n} characters before checking for uniqueness. Use a null string
4690 for comparison if a line has fewer than @var{n} characters. If you use both
4691 the field and character skipping options, fields are skipped over first.
4693 @vindex _POSIX2_VERSION
4694 On older systems, @command{uniq} supports an obsolete option syntax
4696 This obsolete behavior can be enabled or disabled with the
4697 @env{_POSIX2_VERSION} environment variable (@pxref{Standards
4698 conformance}), but portable scripts should avoid commands whose
4699 behavior depends on this variable.
4700 For example, use @samp{uniq ./+10} or @samp{uniq -s 10} rather than
4701 the ambiguous @samp{uniq +10}.
4707 Print the number of times each line occurred along with the line.
4710 @itemx --ignore-case
4712 @opindex --ignore-case
4713 Ignore differences in case when comparing lines.
4719 @cindex repeated lines, outputting
4720 Discard lines that are not repeated. When used by itself, this option
4721 causes @command{uniq} to print the first copy of each repeated line,
4725 @itemx --all-repeated[=@var{delimit-method}]
4727 @opindex --all-repeated
4728 @cindex all repeated lines, outputting
4729 Do not discard the second and subsequent repeated input lines,
4730 but discard lines that are not repeated.
4731 This option is useful mainly in conjunction with other options e.g.,
4732 to ignore case or to compare only selected fields.
4733 The optional @var{delimit-method} tells how to delimit
4734 groups of repeated lines, and must be one of the following:
4739 Do not delimit groups of repeated lines.
4740 This is equivalent to @option{--all-repeated} (@option{-D}).
4743 Output a newline before each group of repeated lines.
4744 With @option{--zero-terminated} (@option{-z}), use a zero
4745 byte (@acronym{ASCII} @sc{nul}) instead of a newline.
4748 Separate groups of repeated lines with a single newline.
4749 With @option{--zero-terminated} (@option{-z}), use a zero
4750 byte (@acronym{ASCII} @sc{nul}) instead of a newline.
4751 This is the same as using @samp{prepend}, except that
4752 no delimiter is inserted before the first group, and hence
4753 may be better suited for output direct to users.
4756 Note that when groups are delimited and the input stream contains
4757 two or more consecutive blank lines, then the output is ambiguous.
4758 To avoid that, filter the input through @samp{tr -s '\n'} to replace
4759 each sequence of consecutive newlines with a single newline.
4761 This is a @sc{gnu} extension.
4762 @c FIXME: give an example showing *how* it's useful
4768 @cindex unique lines, outputting
4769 Discard the first repeated line. When used by itself, this option
4770 causes @command{uniq} to print unique lines, and nothing else.
4773 @itemx --check-chars=@var{n}
4775 @opindex --check-chars
4776 Compare at most @var{n} characters on each line (after skipping any specified
4777 fields and characters). By default the entire rest of the lines are
4780 @zeroTerminatedOption
4787 @node comm invocation
4788 @section @command{comm}: Compare two sorted files line by line
4791 @cindex line-by-line comparison
4792 @cindex comparing sorted files
4794 @command{comm} writes to standard output lines that are common, and lines
4795 that are unique, to two input files; a file name of @samp{-} means
4796 standard input. Synopsis:
4799 comm [@var{option}]@dots{} @var{file1} @var{file2}
4803 Before @command{comm} can be used, the input files must be sorted using the
4804 collating sequence specified by the @env{LC_COLLATE} locale.
4805 If an input file ends in a non-newline
4806 character, a newline is silently appended. The @command{sort} command with
4807 no options always outputs a file that is suitable input to @command{comm}.
4809 @cindex differing lines
4810 @cindex common lines
4811 With no options, @command{comm} produces three-column output. Column one
4812 contains lines unique to @var{file1}, column two contains lines unique
4813 to @var{file2}, and column three contains lines common to both files.
4814 Columns are separated by a single TAB character.
4815 @c FIXME: when there's an option to supply an alternative separator
4816 @c string, append "by default" to the above sentence.
4821 The options @option{-1}, @option{-2}, and @option{-3} suppress printing of
4822 the corresponding columns (and separators). Also see @ref{Common options}.
4824 Unlike some other comparison utilities, @command{comm} has an exit
4825 status that does not depend on the result of the comparison.
4826 Upon normal completion @command{comm} produces an exit code of zero.
4827 If there is an error it exits with nonzero status.
4829 @macro checkOrderOption{cmd}
4830 If the @option{--check-order} option is given, unsorted inputs will
4831 cause a fatal error message. If the option @option{--nocheck-order}
4832 is given, unsorted inputs will never cause an error message. If neither
4833 of these options is given, wrongly sorted inputs are diagnosed
4834 only if an input file is found to contain unpairable
4836 lines, and when both input files are non empty.
4838 @ifclear JOIN_COMMAND
4841 If an input file is diagnosed as being unsorted, the @command{\cmd\}
4842 command will exit with a nonzero status (and the output should not be used).
4844 Forcing @command{\cmd\} to process wrongly sorted input files
4845 containing unpairable lines by specifying @option{--nocheck-order} is
4846 not guaranteed to produce any particular output. The output will
4847 probably not correspond with whatever you hoped it would be.
4849 @checkOrderOption{comm}
4854 Fail with an error message if either input file is wrongly ordered.
4856 @item --nocheck-order
4857 Do not check that both input files are in sorted order.
4861 @item --output-delimiter=@var{str}
4862 Print @var{str} between adjacent output columns,
4863 rather than the default of a single TAB character.
4865 The delimiter @var{str} may not be empty.
4869 @node ptx invocation
4870 @section @command{ptx}: Produce permuted indexes
4874 @command{ptx} reads a text file and essentially produces a permuted index, with
4875 each keyword in its context. The calling sketch is either one of:
4878 ptx [@var{option} @dots{}] [@var{file} @dots{}]
4879 ptx -G [@var{option} @dots{}] [@var{input} [@var{output}]]
4882 The @option{-G} (or its equivalent: @option{--traditional}) option disables
4883 all @sc{gnu} extensions and reverts to traditional mode, thus introducing some
4884 limitations and changing several of the program's default option values.
4885 When @option{-G} is not specified, @sc{gnu} extensions are always enabled.
4886 @sc{gnu} extensions to @command{ptx} are documented wherever appropriate in this
4887 document. For the full list, see @xref{Compatibility in ptx}.
4889 Individual options are explained in the following sections.
4891 When @sc{gnu} extensions are enabled, there may be zero, one or several
4892 @var{file}s after the options. If there is no @var{file}, the program
4893 reads the standard input. If there is one or several @var{file}s, they
4894 give the name of input files which are all read in turn, as if all the
4895 input files were concatenated. However, there is a full contextual
4896 break between each file and, when automatic referencing is requested,
4897 file names and line numbers refer to individual text input files. In
4898 all cases, the program outputs the permuted index to the standard
4901 When @sc{gnu} extensions are @emph{not} enabled, that is, when the program
4902 operates in traditional mode, there may be zero, one or two parameters
4903 besides the options. If there are no parameters, the program reads the
4904 standard input and outputs the permuted index to the standard output.
4905 If there is only one parameter, it names the text @var{input} to be read
4906 instead of the standard input. If two parameters are given, they give
4907 respectively the name of the @var{input} file to read and the name of
4908 the @var{output} file to produce. @emph{Be very careful} to note that,
4909 in this case, the contents of file given by the second parameter is
4910 destroyed. This behavior is dictated by System V @command{ptx}
4911 compatibility; @sc{gnu} Standards normally discourage output parameters not
4912 introduced by an option.
4914 Note that for @emph{any} file named as the value of an option or as an
4915 input text file, a single dash @kbd{-} may be used, in which case
4916 standard input is assumed. However, it would not make sense to use this
4917 convention more than once per program invocation.
4920 * General options in ptx:: Options which affect general program behavior.
4921 * Charset selection in ptx:: Underlying character set considerations.
4922 * Input processing in ptx:: Input fields, contexts, and keyword selection.
4923 * Output formatting in ptx:: Types of output format, and sizing the fields.
4924 * Compatibility in ptx::
4928 @node General options in ptx
4929 @subsection General options
4934 @itemx --traditional
4935 As already explained, this option disables all @sc{gnu} extensions to
4936 @command{ptx} and switches to traditional mode.
4939 Print a short help on standard output, then exit without further
4943 Print the program version on standard output, then exit without further
4951 @node Charset selection in ptx
4952 @subsection Charset selection
4954 @c FIXME: People don't necessarily know what an IBM-PC was these days.
4955 As it is set up now, the program assumes that the input file is coded
4956 using 8-bit @acronym{ISO} 8859-1 code, also known as Latin-1 character set,
4957 @emph{unless} it is compiled for MS-DOS, in which case it uses the
4958 character set of the IBM-PC@. (@sc{gnu} @command{ptx} is not known to work on
4959 smaller MS-DOS machines anymore.) Compared to 7-bit @acronym{ASCII}, the set
4960 of characters which are letters is different; this alters the behavior
4961 of regular expression matching. Thus, the default regular expression
4962 for a keyword allows foreign or diacriticized letters. Keyword sorting,
4963 however, is still crude; it obeys the underlying character set ordering
4969 @itemx --ignore-case
4970 Fold lower case letters to upper case for sorting.
4975 @node Input processing in ptx
4976 @subsection Word selection and input processing
4981 @itemx --break-file=@var{file}
4983 This option provides an alternative (to @option{-W}) method of describing
4984 which characters make up words. It introduces the name of a
4985 file which contains a list of characters which can@emph{not} be part of
4986 one word; this file is called the @dfn{Break file}. Any character which
4987 is not part of the Break file is a word constituent. If both options
4988 @option{-b} and @option{-W} are specified, then @option{-W} has precedence and
4989 @option{-b} is ignored.
4991 When @sc{gnu} extensions are enabled, the only way to avoid newline as a
4992 break character is to write all the break characters in the file with no
4993 newline at all, not even at the end of the file. When @sc{gnu} extensions
4994 are disabled, spaces, tabs and newlines are always considered as break
4995 characters even if not included in the Break file.
4998 @itemx --ignore-file=@var{file}
5000 The file associated with this option contains a list of words which will
5001 never be taken as keywords in concordance output. It is called the
5002 @dfn{Ignore file}. The file contains exactly one word in each line; the
5003 end of line separation of words is not subject to the value of the
5007 @itemx --only-file=@var{file}
5009 The file associated with this option contains a list of words which will
5010 be retained in concordance output; any word not mentioned in this file
5011 is ignored. The file is called the @dfn{Only file}. The file contains
5012 exactly one word in each line; the end of line separation of words is
5013 not subject to the value of the @option{-S} option.
5015 There is no default for the Only file. When both an Only file and an
5016 Ignore file are specified, a word is considered a keyword only
5017 if it is listed in the Only file and not in the Ignore file.
5022 On each input line, the leading sequence of non-white space characters will be
5023 taken to be a reference that has the purpose of identifying this input
5024 line in the resulting permuted index. For more information about reference
5025 production, see @xref{Output formatting in ptx}.
5026 Using this option changes the default value for option @option{-S}.
5028 Using this option, the program does not try very hard to remove
5029 references from contexts in output, but it succeeds in doing so
5030 @emph{when} the context ends exactly at the newline. If option
5031 @option{-r} is used with @option{-S} default value, or when @sc{gnu} extensions
5032 are disabled, this condition is always met and references are completely
5033 excluded from the output contexts.
5035 @item -S @var{regexp}
5036 @itemx --sentence-regexp=@var{regexp}
5038 This option selects which regular expression will describe the end of a
5039 line or the end of a sentence. In fact, this regular expression is not
5040 the only distinction between end of lines or end of sentences, and input
5041 line boundaries have no special significance outside this option. By
5042 default, when @sc{gnu} extensions are enabled and if @option{-r} option is not
5043 used, end of sentences are used. In this case, this @var{regex} is
5044 imported from @sc{gnu} Emacs:
5047 [.?!][]\"')@}]*\\($\\|\t\\| \\)[ \t\n]*
5050 Whenever @sc{gnu} extensions are disabled or if @option{-r} option is used, end
5051 of lines are used; in this case, the default @var{regexp} is just:
5057 Using an empty @var{regexp} is equivalent to completely disabling end of
5058 line or end of sentence recognition. In this case, the whole file is
5059 considered to be a single big line or sentence. The user might want to
5060 disallow all truncation flag generation as well, through option @option{-F
5061 ""}. @xref{Regexps, , Syntax of Regular Expressions, emacs, The GNU Emacs
5064 When the keywords happen to be near the beginning of the input line or
5065 sentence, this often creates an unused area at the beginning of the
5066 output context line; when the keywords happen to be near the end of the
5067 input line or sentence, this often creates an unused area at the end of
5068 the output context line. The program tries to fill those unused areas
5069 by wrapping around context in them; the tail of the input line or
5070 sentence is used to fill the unused area on the left of the output line;
5071 the head of the input line or sentence is used to fill the unused area
5072 on the right of the output line.
5074 As a matter of convenience to the user, many usual backslashed escape
5075 sequences from the C language are recognized and converted to the
5076 corresponding characters by @command{ptx} itself.
5078 @item -W @var{regexp}
5079 @itemx --word-regexp=@var{regexp}
5081 This option selects which regular expression will describe each keyword.
5082 By default, if @sc{gnu} extensions are enabled, a word is a sequence of
5083 letters; the @var{regexp} used is @samp{\w+}. When @sc{gnu} extensions are
5084 disabled, a word is by default anything which ends with a space, a tab
5085 or a newline; the @var{regexp} used is @samp{[^ \t\n]+}.
5087 An empty @var{regexp} is equivalent to not using this option.
5088 @xref{Regexps, , Syntax of Regular Expressions, emacs, The GNU Emacs
5091 As a matter of convenience to the user, many usual backslashed escape
5092 sequences, as found in the C language, are recognized and converted to
5093 the corresponding characters by @command{ptx} itself.
5098 @node Output formatting in ptx
5099 @subsection Output formatting
5101 Output format is mainly controlled by the @option{-O} and @option{-T} options
5102 described in the table below. When neither @option{-O} nor @option{-T} are
5103 selected, and if @sc{gnu} extensions are enabled, the program chooses an
5104 output format suitable for a dumb terminal. Each keyword occurrence is
5105 output to the center of one line, surrounded by its left and right
5106 contexts. Each field is properly justified, so the concordance output
5107 can be readily observed. As a special feature, if automatic
5108 references are selected by option @option{-A} and are output before the
5109 left context, that is, if option @option{-R} is @emph{not} selected, then
5110 a colon is added after the reference; this nicely interfaces with @sc{gnu}
5111 Emacs @code{next-error} processing. In this default output format, each
5112 white space character, like newline and tab, is merely changed to
5113 exactly one space, with no special attempt to compress consecutive
5114 spaces. This might change in the future. Except for those white space
5115 characters, every other character of the underlying set of 256
5116 characters is transmitted verbatim.
5118 Output format is further controlled by the following options.
5122 @item -g @var{number}
5123 @itemx --gap-size=@var{number}
5125 Select the size of the minimum white space gap between the fields on the
5128 @item -w @var{number}
5129 @itemx --width=@var{number}
5131 Select the maximum output width of each final line. If references are
5132 used, they are included or excluded from the maximum output width
5133 depending on the value of option @option{-R}. If this option is not
5134 selected, that is, when references are output before the left context,
5135 the maximum output width takes into account the maximum length of all
5136 references. If this option is selected, that is, when references are
5137 output after the right context, the maximum output width does not take
5138 into account the space taken by references, nor the gap that precedes
5142 @itemx --auto-reference
5144 Select automatic references. Each input line will have an automatic
5145 reference made up of the file name and the line ordinal, with a single
5146 colon between them. However, the file name will be empty when standard
5147 input is being read. If both @option{-A} and @option{-r} are selected, then
5148 the input reference is still read and skipped, but the automatic
5149 reference is used at output time, overriding the input reference.
5152 @itemx --right-side-refs
5154 In the default output format, when option @option{-R} is not used, any
5155 references produced by the effect of options @option{-r} or @option{-A} are
5156 placed to the far right of output lines, after the right context. With
5157 default output format, when the @option{-R} option is specified, references
5158 are rather placed at the beginning of each output line, before the left
5159 context. For any other output format, option @option{-R} is
5160 ignored, with one exception: with @option{-R} the width of references
5161 is @emph{not} taken into account in total output width given by @option{-w}.
5163 This option is automatically selected whenever @sc{gnu} extensions are
5166 @item -F @var{string}
5167 @itemx --flac-truncation=@var{string}
5169 This option will request that any truncation in the output be reported
5170 using the string @var{string}. Most output fields theoretically extend
5171 towards the beginning or the end of the current line, or current
5172 sentence, as selected with option @option{-S}. But there is a maximum
5173 allowed output line width, changeable through option @option{-w}, which is
5174 further divided into space for various output fields. When a field has
5175 to be truncated because it cannot extend beyond the beginning or the end of
5176 the current line to fit in, then a truncation occurs. By default,
5177 the string used is a single slash, as in @option{-F /}.
5179 @var{string} may have more than one character, as in @option{-F ...}.
5180 Also, in the particular case when @var{string} is empty (@option{-F ""}),
5181 truncation flagging is disabled, and no truncation marks are appended in
5184 As a matter of convenience to the user, many usual backslashed escape
5185 sequences, as found in the C language, are recognized and converted to
5186 the corresponding characters by @command{ptx} itself.
5188 @item -M @var{string}
5189 @itemx --macro-name=@var{string}
5191 Select another @var{string} to be used instead of @samp{xx}, while
5192 generating output suitable for @command{nroff}, @command{troff} or @TeX{}.
5195 @itemx --format=roff
5197 Choose an output format suitable for @command{nroff} or @command{troff}
5198 processing. Each output line will look like:
5201 .xx "@var{tail}" "@var{before}" "@var{keyword_and_after}"@c
5202 "@var{head}" "@var{ref}"
5205 so it will be possible to write a @samp{.xx} roff macro to take care of
5206 the output typesetting. This is the default output format when @sc{gnu}
5207 extensions are disabled. Option @option{-M} can be used to change
5208 @samp{xx} to another macro name.
5210 In this output format, each non-graphical character, like newline and
5211 tab, is merely changed to exactly one space, with no special attempt to
5212 compress consecutive spaces. Each quote character: @kbd{"} is doubled
5213 so it will be correctly processed by @command{nroff} or @command{troff}.
5218 Choose an output format suitable for @TeX{} processing. Each output
5219 line will look like:
5222 \xx @{@var{tail}@}@{@var{before}@}@{@var{keyword}@}@c
5223 @{@var{after}@}@{@var{head}@}@{@var{ref}@}
5227 so it will be possible to write a @code{\xx} definition to take care of
5228 the output typesetting. Note that when references are not being
5229 produced, that is, neither option @option{-A} nor option @option{-r} is
5230 selected, the last parameter of each @code{\xx} call is inhibited.
5231 Option @option{-M} can be used to change @samp{xx} to another macro
5234 In this output format, some special characters, like @kbd{$}, @kbd{%},
5235 @kbd{&}, @kbd{#} and @kbd{_} are automatically protected with a
5236 backslash. Curly brackets @kbd{@{}, @kbd{@}} are protected with a
5237 backslash and a pair of dollar signs (to force mathematical mode). The
5238 backslash itself produces the sequence @code{\backslash@{@}}.
5239 Circumflex and tilde diacritical marks produce the sequence @code{^\@{ @}} and
5240 @code{~\@{ @}} respectively. Other diacriticized characters of the
5241 underlying character set produce an appropriate @TeX{} sequence as far
5242 as possible. The other non-graphical characters, like newline and tab,
5243 and all other characters which are not part of @acronym{ASCII}, are merely
5244 changed to exactly one space, with no special attempt to compress
5245 consecutive spaces. Let me know how to improve this special character
5246 processing for @TeX{}.
5251 @node Compatibility in ptx
5252 @subsection The @sc{gnu} extensions to @command{ptx}
5254 This version of @command{ptx} contains a few features which do not exist in
5255 System V @command{ptx}. These extra features are suppressed by using the
5256 @option{-G} command line option, unless overridden by other command line
5257 options. Some @sc{gnu} extensions cannot be recovered by overriding, so the
5258 simple rule is to avoid @option{-G} if you care about @sc{gnu} extensions.
5259 Here are the differences between this program and System V @command{ptx}.
5264 This program can read many input files at once, it always writes the
5265 resulting concordance on standard output. On the other hand, System V
5266 @command{ptx} reads only one file and sends the result to standard output
5267 or, if a second @var{file} parameter is given on the command, to that
5270 Having output parameters not introduced by options is a dangerous
5271 practice which @sc{gnu} avoids as far as possible. So, for using @command{ptx}
5272 portably between @sc{gnu} and System V, you should always use it with a
5273 single input file, and always expect the result on standard output. You
5274 might also want to automatically configure in a @option{-G} option to
5275 @command{ptx} calls in products using @command{ptx}, if the configurator finds
5276 that the installed @command{ptx} accepts @option{-G}.
5279 The only options available in System V @command{ptx} are options @option{-b},
5280 @option{-f}, @option{-g}, @option{-i}, @option{-o}, @option{-r}, @option{-t} and
5281 @option{-w}. All other options are @sc{gnu} extensions and are not repeated in
5282 this enumeration. Moreover, some options have a slightly different
5283 meaning when @sc{gnu} extensions are enabled, as explained below.
5286 By default, concordance output is not formatted for @command{troff} or
5287 @command{nroff}. It is rather formatted for a dumb terminal. @command{troff}
5288 or @command{nroff} output may still be selected through option @option{-O}.
5291 Unless @option{-R} option is used, the maximum reference width is
5292 subtracted from the total output line width. With @sc{gnu} extensions
5293 disabled, width of references is not taken into account in the output
5294 line width computations.
5297 All 256 bytes, even @acronym{ASCII} @sc{nul} bytes, are always read and
5298 processed from input file with no adverse effect, even if @sc{gnu} extensions
5299 are disabled. However, System V @command{ptx} does not accept 8-bit characters,
5300 a few control characters are rejected, and the tilde @kbd{~} is also rejected.
5303 Input line length is only limited by available memory, even if @sc{gnu}
5304 extensions are disabled. However, System V @command{ptx} processes only
5305 the first 200 characters in each line.
5308 The break (non-word) characters default to be every character except all
5309 letters of the underlying character set, diacriticized or not. When @sc{gnu}
5310 extensions are disabled, the break characters default to space, tab and
5314 The program makes better use of output line width. If @sc{gnu} extensions
5315 are disabled, the program rather tries to imitate System V @command{ptx},
5316 but still, there are some slight disposition glitches this program does
5317 not completely reproduce.
5320 The user can specify both an Ignore file and an Only file. This is not
5321 allowed with System V @command{ptx}.
5326 @node tsort invocation
5327 @section @command{tsort}: Topological sort
5330 @cindex topological sort
5332 @command{tsort} performs a topological sort on the given @var{file}, or
5333 standard input if no input file is given or for a @var{file} of
5334 @samp{-}. For more details and some history, see @ref{tsort background}.
5338 tsort [@var{option}] [@var{file}]
5341 @command{tsort} reads its input as pairs of strings, separated by blanks,
5342 indicating a partial ordering. The output is a total ordering that
5343 corresponds to the given partial ordering.
5357 will produce the output
5368 Consider a more realistic example.
5369 You have a large set of functions all in one file, and they may all be
5370 declared static except one. Currently that one (say @code{main}) is the
5371 first function defined in the file, and the ones it calls directly follow
5372 it, followed by those they call, etc. Let's say that you are determined
5373 to take advantage of prototypes, so you have to choose between declaring
5374 all of those functions (which means duplicating a lot of information from
5375 the definitions) and rearranging the functions so that as many as possible
5376 are defined before they are used. One way to automate the latter process
5377 is to get a list for each function of the functions it calls directly.
5378 Many programs can generate such lists. They describe a call graph.
5379 Consider the following list, in which a given line indicates that the
5380 function on the left calls the one on the right directly.
5386 tail_file pretty_name
5387 tail_file write_header
5389 tail_forever recheck
5390 tail_forever pretty_name
5391 tail_forever write_header
5392 tail_forever dump_remainder
5395 tail_lines start_lines
5396 tail_lines dump_remainder
5397 tail_lines file_lines
5398 tail_lines pipe_lines
5400 tail_bytes start_bytes
5401 tail_bytes dump_remainder
5402 tail_bytes pipe_bytes
5403 file_lines dump_remainder
5407 then you can use @command{tsort} to produce an ordering of those
5408 functions that satisfies your requirement.
5411 example$ tsort call-graph | tac
5431 @command{tsort} detects any cycles in the input and writes the first cycle
5432 encountered to standard error.
5434 Note that for a given partial ordering, generally there is no unique
5435 total ordering. In the context of the call graph above, the function
5436 @code{parse_options} may be placed anywhere in the list as long as it
5437 precedes @code{main}.
5439 The only options are @option{--help} and @option{--version}. @xref{Common
5445 * tsort background:: Where tsort came from.
5448 @node tsort background
5449 @subsection @command{tsort}: Background
5451 @command{tsort} exists because very early versions of the Unix linker processed
5452 an archive file exactly once, and in order. As @command{ld} read each object
5453 in the archive, it decided whether it was needed in the program based on
5454 whether it defined any symbols which were undefined at that point in
5457 This meant that dependencies within the archive had to be handled
5458 specially. For example, @code{scanf} probably calls @code{read}. That means
5459 that in a single pass through an archive, it was important for @code{scanf.o}
5460 to appear before read.o, because otherwise a program which calls
5461 @code{scanf} but not @code{read} might end up with an unexpected unresolved
5462 reference to @code{read}.
5464 The way to address this problem was to first generate a set of
5465 dependencies of one object file on another. This was done by a shell
5466 script called @command{lorder}. The GNU tools don't provide a version of
5467 lorder, as far as I know, but you can still find it in BSD
5470 Then you ran @command{tsort} over the @command{lorder} output, and you used the
5471 resulting sort to define the order in which you added objects to the archive.
5473 This whole procedure has been obsolete since about 1980, because
5474 Unix archives now contain a symbol table (traditionally built by
5475 @command{ranlib}, now generally built by @command{ar} itself), and the Unix
5476 linker uses the symbol table to effectively make multiple passes over
5479 Anyhow, that's where tsort came from. To solve an old problem with
5480 the way the linker handled archive files, which has since been solved
5484 @node Operating on fields
5485 @chapter Operating on fields
5488 * cut invocation:: Print selected parts of lines.
5489 * paste invocation:: Merge lines of files.
5490 * join invocation:: Join lines on a common field.
5494 @node cut invocation
5495 @section @command{cut}: Print selected parts of lines
5498 @command{cut} writes to standard output selected parts of each line of each
5499 input file, or standard input if no files are given or for a file name of
5503 cut @var{option}@dots{} [@var{file}]@dots{}
5506 In the table which follows, the @var{byte-list}, @var{character-list},
5507 and @var{field-list} are one or more numbers or ranges (two numbers
5508 separated by a dash) separated by commas. Bytes, characters, and
5509 fields are numbered starting at 1. Incomplete ranges may be
5510 given: @option{-@var{m}} means @samp{1-@var{m}}; @samp{@var{n}-} means
5511 @samp{@var{n}} through end of line or last field. The list elements
5512 can be repeated, can overlap, and can be specified in any order; but
5513 the selected input is written in the same order that it is read, and
5514 is written exactly once.
5516 The program accepts the following options. Also see @ref{Common
5521 @item -b @var{byte-list}
5522 @itemx --bytes=@var{byte-list}
5525 Select for printing only the bytes in positions listed in
5526 @var{byte-list}. Tabs and backspaces are treated like any other
5527 character; they take up 1 byte. If an output delimiter is specified,
5528 (see the description of @option{--output-delimiter}), then output that
5529 string between ranges of selected bytes.
5531 @item -c @var{character-list}
5532 @itemx --characters=@var{character-list}
5534 @opindex --characters
5535 Select for printing only the characters in positions listed in
5536 @var{character-list}. The same as @option{-b} for now, but
5537 internationalization will change that. Tabs and backspaces are
5538 treated like any other character; they take up 1 character. If an
5539 output delimiter is specified, (see the description of
5540 @option{--output-delimiter}), then output that string between ranges
5543 @item -f @var{field-list}
5544 @itemx --fields=@var{field-list}
5547 Select for printing only the fields listed in @var{field-list}.
5548 Fields are separated by a TAB character by default. Also print any
5549 line that contains no delimiter character, unless the
5550 @option{--only-delimited} (@option{-s}) option is specified.
5552 Note @command{awk} supports more sophisticated field processing,
5553 and by default will use (and discard) runs of blank characters to
5554 separate fields, and ignore leading and trailing blanks.
5557 awk '{print $2}' # print the second field
5558 awk '{print $NF-1}' # print the penultimate field
5559 awk '{print $2,$1}' # reorder the first two fields
5563 In the unlikely event that @command{awk} is unavailable,
5564 one can use the @command{join} command, to process blank
5565 characters as @command{awk} does above.
5568 join -a1 -o 1.2 - /dev/null # print the second field
5569 join -a1 -o 1.2,1.1 - /dev/null # reorder the first two fields
5573 @item -d @var{input_delim_byte}
5574 @itemx --delimiter=@var{input_delim_byte}
5576 @opindex --delimiter
5577 With @option{-f}, use the first byte of @var{input_delim_byte} as
5578 the input fields separator (default is TAB).
5582 Do not split multi-byte characters (no-op for now).
5585 @itemx --only-delimited
5587 @opindex --only-delimited
5588 For @option{-f}, do not print lines that do not contain the field separator
5589 character. Normally, any line without a field separator is printed verbatim.
5591 @item --output-delimiter=@var{output_delim_string}
5592 @opindex --output-delimiter
5593 With @option{-f}, output fields are separated by @var{output_delim_string}.
5594 The default with @option{-f} is to use the input delimiter.
5595 When using @option{-b} or @option{-c} to select ranges of byte or
5596 character offsets (as opposed to ranges of fields),
5597 output @var{output_delim_string} between non-overlapping
5598 ranges of selected bytes.
5601 @opindex --complement
5602 This option is a @acronym{GNU} extension.
5603 Select for printing the complement of the bytes, characters or fields
5604 selected with the @option{-b}, @option{-c} or @option{-f} options.
5605 In other words, do @emph{not} print the bytes, characters or fields
5606 specified via those options. This option is useful when you have
5607 many fields and want to print all but a few of them.
5614 @node paste invocation
5615 @section @command{paste}: Merge lines of files
5618 @cindex merging files
5620 @command{paste} writes to standard output lines consisting of sequentially
5621 corresponding lines of each given file, separated by a TAB character.
5622 Standard input is used for a file name of @samp{-} or if no input files
5644 paste [@var{option}]@dots{} [@var{file}]@dots{}
5647 The program accepts the following options. Also see @ref{Common options}.
5655 Paste the lines of one file at a time rather than one line from each
5656 file. Using the above example data:
5659 $ paste -s num2 let3
5664 @item -d @var{delim-list}
5665 @itemx --delimiters=@var{delim-list}
5667 @opindex --delimiters
5668 Consecutively use the characters in @var{delim-list} instead of
5669 TAB to separate merged lines. When @var{delim-list} is
5670 exhausted, start again at its beginning. Using the above example data:
5673 $ paste -d '%_' num2 let3 num2
5684 @node join invocation
5685 @section @command{join}: Join lines on a common field
5688 @cindex common field, joining on
5690 @command{join} writes to standard output a line for each pair of input
5691 lines that have identical join fields. Synopsis:
5694 join [@var{option}]@dots{} @var{file1} @var{file2}
5697 Either @var{file1} or @var{file2} (but not both) can be @samp{-},
5698 meaning standard input. @var{file1} and @var{file2} should be
5699 sorted on the join fields.
5702 Normally, the sort order is that of the
5703 collating sequence specified by the @env{LC_COLLATE} locale. Unless
5704 the @option{-t} option is given, the sort comparison ignores blanks at
5705 the start of the join field, as in @code{sort -b}. If the
5706 @option{--ignore-case} option is given, the sort comparison ignores
5707 the case of characters in the join field, as in @code{sort -f}.
5709 The @command{sort} and @command{join} commands should use consistent
5710 locales and options if the output of @command{sort} is fed to
5711 @command{join}. You can use a command like @samp{sort -k 1b,1} to
5712 sort a file on its default join field, but if you select a non-default
5713 locale, join field, separator, or comparison options, then you should
5714 do so consistently between @command{join} and @command{sort}.
5715 If @samp{join -t ''} is specified then the whole line is considered which
5716 matches the default operation of sort.
5718 If the input has no unpairable lines, a @acronym{GNU} extension is
5719 available; the sort order can be any order that considers two fields
5720 to be equal if and only if the sort comparison described above
5721 considers them to be equal. For example:
5739 @checkOrderOption{join}
5744 @item the join field is the first field in each line;
5745 @item fields in the input are separated by one or more blanks, with leading
5746 blanks on the line ignored;
5747 @item fields in the output are separated by a space;
5748 @item each output line consists of the join field, the remaining
5749 fields from @var{file1}, then the remaining fields from @var{file2}.
5752 The program accepts the following options. Also see @ref{Common options}.
5756 @item -a @var{file-number}
5758 Print a line for each unpairable line in file @var{file-number} (either
5759 @samp{1} or @samp{2}), in addition to the normal output.
5762 Fail with an error message if either input file is wrongly ordered.
5764 @item --nocheck-order
5765 Do not check that both input files are in sorted order. This is the default.
5767 @item -e @var{string}
5769 Replace those output fields that are missing in the input with @var{string}.
5770 I.E. missing fields specified with the @option{-12jo} options.
5774 Treat the first line of each input file as a header line. The header lines will
5775 be joined and printed as the first output line. If @option{-o} is used to
5776 specify output format, the header line will be printed according to the
5777 specified format. The header lines will not be checked for ordering even if
5778 @option{--check-order} is specified. Also if the header lines from each file
5779 do not match, the heading fields from the first file will be used.
5782 @itemx --ignore-case
5784 @opindex --ignore-case
5785 Ignore differences in case when comparing keys.
5786 With this option, the lines of the input files must be ordered in the same way.
5787 Use @samp{sort -f} to produce this ordering.
5789 @item -1 @var{field}
5791 Join on field @var{field} (a positive integer) of file 1.
5793 @item -2 @var{field}
5795 Join on field @var{field} (a positive integer) of file 2.
5797 @item -j @var{field}
5798 Equivalent to @option{-1 @var{field} -2 @var{field}}.
5800 @item -o @var{field-list}
5802 If the keyword @samp{auto} is specified, infer the output format from
5803 the first line in each file. This is the same as the default output format
5804 but also ensures the same number of fields are output for each line.
5805 Missing fields are replaced with the @option{-e} option and extra fields
5808 Otherwise, construct each output line according to the format in
5809 @var{field-list}. Each element in @var{field-list} is either the single
5810 character @samp{0} or has the form @var{m.n} where the file number, @var{m},
5811 is @samp{1} or @samp{2} and @var{n} is a positive field number.
5813 A field specification of @samp{0} denotes the join field.
5814 In most cases, the functionality of the @samp{0} field spec
5815 may be reproduced using the explicit @var{m.n} that corresponds
5816 to the join field. However, when printing unpairable lines
5817 (using either of the @option{-a} or @option{-v} options), there is no way
5818 to specify the join field using @var{m.n} in @var{field-list}
5819 if there are unpairable lines in both files.
5820 To give @command{join} that functionality, @acronym{POSIX} invented the @samp{0}
5821 field specification notation.
5823 The elements in @var{field-list}
5824 are separated by commas or blanks.
5825 Blank separators typically need to be quoted for the shell. For
5826 example, the commands @samp{join -o 1.2,2.2} and @samp{join -o '1.2
5827 2.2'} are equivalent.
5829 All output lines---including those printed because of any -a or -v
5830 option---are subject to the specified @var{field-list}.
5833 Use character @var{char} as the input and output field separator.
5834 Treat as significant each occurrence of @var{char} in the input file.
5835 Use @samp{sort -t @var{char}}, without the @option{-b} option of
5836 @samp{sort}, to produce this ordering. If @samp{join -t ''} is specified,
5837 the whole line is considered, matching the default operation of sort.
5838 If @samp{-t '\0'} is specified then the @acronym{ASCII} @sc{nul}
5839 character is used to delimit the fields.
5841 @item -v @var{file-number}
5842 Print a line for each unpairable line in file @var{file-number}
5843 (either @samp{1} or @samp{2}), instead of the normal output.
5850 @node Operating on characters
5851 @chapter Operating on characters
5853 @cindex operating on characters
5855 This commands operate on individual characters.
5858 * tr invocation:: Translate, squeeze, and/or delete characters.
5859 * expand invocation:: Convert tabs to spaces.
5860 * unexpand invocation:: Convert spaces to tabs.
5865 @section @command{tr}: Translate, squeeze, and/or delete characters
5872 tr [@var{option}]@dots{} @var{set1} [@var{set2}]
5875 @command{tr} copies standard input to standard output, performing
5876 one of the following operations:
5880 translate, and optionally squeeze repeated characters in the result,
5882 squeeze repeated characters,
5886 delete characters, then squeeze repeated characters from the result.
5889 The @var{set1} and (if given) @var{set2} arguments define ordered
5890 sets of characters, referred to below as @var{set1} and @var{set2}. These
5891 sets are the characters of the input that @command{tr} operates on.
5892 The @option{--complement} (@option{-c}, @option{-C}) option replaces
5894 complement (all of the characters that are not in @var{set1}).
5896 Currently @command{tr} fully supports only single-byte characters.
5897 Eventually it will support multibyte characters; when it does, the
5898 @option{-C} option will cause it to complement the set of characters,
5899 whereas @option{-c} will cause it to complement the set of values.
5900 This distinction will matter only when some values are not characters,
5901 and this is possible only in locales using multibyte encodings when
5902 the input contains encoding errors.
5904 The program accepts the @option{--help} and @option{--version}
5905 options. @xref{Common options}. Options must precede operands.
5910 * Character sets:: Specifying sets of characters.
5911 * Translating:: Changing one set of characters to another.
5912 * Squeezing:: Squeezing repeats and deleting.
5916 @node Character sets
5917 @subsection Specifying sets of characters
5919 @cindex specifying sets of characters
5921 The format of the @var{set1} and @var{set2} arguments resembles
5922 the format of regular expressions; however, they are not regular
5923 expressions, only lists of characters. Most characters simply
5924 represent themselves in these strings, but the strings can contain
5925 the shorthands listed below, for convenience. Some of them can be
5926 used only in @var{set1} or @var{set2}, as noted below.
5930 @item Backslash escapes
5931 @cindex backslash escapes
5933 The following backslash escape sequences are recognized:
5951 The 8-bit character with the value given by @var{ooo}, which is 1 to 3
5952 octal digits. Note that @samp{\400} is interpreted as the two-byte
5953 sequence, @samp{\040} @samp{0}.
5958 While a backslash followed by a character not listed above is
5959 interpreted as that character, the backslash also effectively
5960 removes any special significance, so it is useful to escape
5961 @samp{[}, @samp{]}, @samp{*}, and @samp{-}.
5966 The notation @samp{@var{m}-@var{n}} expands to all of the characters
5967 from @var{m} through @var{n}, in ascending order. @var{m} should
5968 collate before @var{n}; if it doesn't, an error results. As an example,
5969 @samp{0-9} is the same as @samp{0123456789}.
5971 @sc{gnu} @command{tr} does not support the System V syntax that uses square
5972 brackets to enclose ranges. Translations specified in that format
5973 sometimes work as expected, since the brackets are often transliterated
5974 to themselves. However, they should be avoided because they sometimes
5975 behave unexpectedly. For example, @samp{tr -d '[0-9]'} deletes brackets
5978 Many historically common and even accepted uses of ranges are not
5979 portable. For example, on @acronym{EBCDIC} hosts using the @samp{A-Z}
5980 range will not do what most would expect because @samp{A} through @samp{Z}
5981 are not contiguous as they are in @acronym{ASCII}.
5982 If you can rely on a @acronym{POSIX} compliant version of @command{tr}, then
5983 the best way to work around this is to use character classes (see below).
5984 Otherwise, it is most portable (and most ugly) to enumerate the members
5987 @item Repeated characters
5988 @cindex repeated characters
5990 The notation @samp{[@var{c}*@var{n}]} in @var{set2} expands to @var{n}
5991 copies of character @var{c}. Thus, @samp{[y*6]} is the same as
5992 @samp{yyyyyy}. The notation @samp{[@var{c}*]} in @var{string2} expands
5993 to as many copies of @var{c} as are needed to make @var{set2} as long as
5994 @var{set1}. If @var{n} begins with @samp{0}, it is interpreted in
5995 octal, otherwise in decimal.
5997 @item Character classes
5998 @cindex character classes
6000 The notation @samp{[:@var{class}:]} expands to all of the characters in
6001 the (predefined) class @var{class}. The characters expand in no
6002 particular order, except for the @code{upper} and @code{lower} classes,
6003 which expand in ascending order. When the @option{--delete} (@option{-d})
6004 and @option{--squeeze-repeats} (@option{-s}) options are both given, any
6005 character class can be used in @var{set2}. Otherwise, only the
6006 character classes @code{lower} and @code{upper} are accepted in
6007 @var{set2}, and then only if the corresponding character class
6008 (@code{upper} and @code{lower}, respectively) is specified in the same
6009 relative position in @var{set1}. Doing this specifies case conversion.
6010 The class names are given below; an error results when an invalid class
6022 Horizontal whitespace.
6031 Printable characters, not including space.
6037 Printable characters, including space.
6040 Punctuation characters.
6043 Horizontal or vertical whitespace.
6052 @item Equivalence classes
6053 @cindex equivalence classes
6055 The syntax @samp{[=@var{c}=]} expands to all of the characters that are
6056 equivalent to @var{c}, in no particular order. Equivalence classes are
6057 a relatively recent invention intended to support non-English alphabets.
6058 But there seems to be no standard way to define them or determine their
6059 contents. Therefore, they are not fully implemented in @sc{gnu} @command{tr};
6060 each character's equivalence class consists only of that character,
6061 which is of no particular use.
6067 @subsection Translating
6069 @cindex translating characters
6071 @command{tr} performs translation when @var{set1} and @var{set2} are
6072 both given and the @option{--delete} (@option{-d}) option is not given.
6073 @command{tr} translates each character of its input that is in @var{set1}
6074 to the corresponding character in @var{set2}. Characters not in
6075 @var{set1} are passed through unchanged. When a character appears more
6076 than once in @var{set1} and the corresponding characters in @var{set2}
6077 are not all the same, only the final one is used. For example, these
6078 two commands are equivalent:
6085 A common use of @command{tr} is to convert lowercase characters to
6086 uppercase. This can be done in many ways. Here are three of them:
6089 tr abcdefghijklmnopqrstuvwxyz ABCDEFGHIJKLMNOPQRSTUVWXYZ
6091 tr '[:lower:]' '[:upper:]'
6095 But note that using ranges like @code{a-z} above is not portable.
6097 When @command{tr} is performing translation, @var{set1} and @var{set2}
6098 typically have the same length. If @var{set1} is shorter than
6099 @var{set2}, the extra characters at the end of @var{set2} are ignored.
6101 On the other hand, making @var{set1} longer than @var{set2} is not
6102 portable; @acronym{POSIX} says that the result is undefined. In this situation,
6103 BSD @command{tr} pads @var{set2} to the length of @var{set1} by repeating
6104 the last character of @var{set2} as many times as necessary. System V
6105 @command{tr} truncates @var{set1} to the length of @var{set2}.
6107 By default, @sc{gnu} @command{tr} handles this case like BSD @command{tr}.
6108 When the @option{--truncate-set1} (@option{-t}) option is given,
6109 @sc{gnu} @command{tr} handles this case like the System V @command{tr}
6110 instead. This option is ignored for operations other than translation.
6112 Acting like System V @command{tr} in this case breaks the relatively common
6116 tr -cs A-Za-z0-9 '\012'
6120 because it converts only zero bytes (the first element in the
6121 complement of @var{set1}), rather than all non-alphanumerics, to
6125 By the way, the above idiom is not portable because it uses ranges, and
6126 it assumes that the octal code for newline is 012.
6127 Assuming a @acronym{POSIX} compliant @command{tr}, here is a better
6131 tr -cs '[:alnum:]' '[\n*]'
6136 @subsection Squeezing repeats and deleting
6138 @cindex squeezing repeat characters
6139 @cindex deleting characters
6141 When given just the @option{--delete} (@option{-d}) option, @command{tr}
6142 removes any input characters that are in @var{set1}.
6144 When given just the @option{--squeeze-repeats} (@option{-s}) option,
6145 @command{tr} replaces each input sequence of a repeated character that
6146 is in @var{set1} with a single occurrence of that character.
6148 When given both @option{--delete} and @option{--squeeze-repeats}, @command{tr}
6149 first performs any deletions using @var{set1}, then squeezes repeats
6150 from any remaining characters using @var{set2}.
6152 The @option{--squeeze-repeats} option may also be used when translating,
6153 in which case @command{tr} first performs translation, then squeezes
6154 repeats from any remaining characters using @var{set2}.
6156 Here are some examples to illustrate various combinations of options:
6161 Remove all zero bytes:
6168 Put all words on lines by themselves. This converts all
6169 non-alphanumeric characters to newlines, then squeezes each string
6170 of repeated newlines into a single newline:
6173 tr -cs '[:alnum:]' '[\n*]'
6177 Convert each sequence of repeated newlines to a single newline:
6184 Find doubled occurrences of words in a document.
6185 @c Separate the following two "the"s, so typo checkers don't complain.
6186 For example, people often write ``the @w{}the'' with the repeated words
6187 separated by a newline. The Bourne shell script below works first
6188 by converting each sequence of punctuation and blank characters to a
6189 single newline. That puts each ``word'' on a line by itself.
6190 Next it maps all uppercase characters to lower case, and finally it
6191 runs @command{uniq} with the @option{-d} option to print out only the words
6197 | tr -s '[:punct:][:blank:]' '[\n*]' \
6198 | tr '[:upper:]' '[:lower:]' \
6203 Deleting a small set of characters is usually straightforward. For example,
6204 to remove all @samp{a}s, @samp{x}s, and @samp{M}s you would do this:
6210 However, when @samp{-} is one of those characters, it can be tricky because
6211 @samp{-} has special meanings. Performing the same task as above but also
6212 removing all @samp{-} characters, we might try @code{tr -d -axM}, but
6213 that would fail because @command{tr} would try to interpret @option{-a} as
6214 a command-line option. Alternatively, we could try putting the hyphen
6215 inside the string, @code{tr -d a-xM}, but that wouldn't work either because
6216 it would make @command{tr} interpret @code{a-x} as the range of characters
6217 @samp{a}@dots{}@samp{x} rather than the three.
6218 One way to solve the problem is to put the hyphen at the end of the list
6225 Or you can use @samp{--} to terminate option processing:
6231 More generally, use the character class notation @code{[=c=]}
6232 with @samp{-} (or any other character) in place of the @samp{c}:
6238 Note how single quotes are used in the above example to protect the
6239 square brackets from interpretation by a shell.
6244 @node expand invocation
6245 @section @command{expand}: Convert tabs to spaces
6248 @cindex tabs to spaces, converting
6249 @cindex converting tabs to spaces
6251 @command{expand} writes the contents of each given @var{file}, or standard
6252 input if none are given or for a @var{file} of @samp{-}, to standard
6253 output, with tab characters converted to the appropriate number of
6257 expand [@var{option}]@dots{} [@var{file}]@dots{}
6260 By default, @command{expand} converts all tabs to spaces. It preserves
6261 backspace characters in the output; they decrement the column count for
6262 tab calculations. The default action is equivalent to @option{-t 8} (set
6263 tabs every 8 columns).
6265 The program accepts the following options. Also see @ref{Common options}.
6269 @item -t @var{tab1}[,@var{tab2}]@dots{}
6270 @itemx --tabs=@var{tab1}[,@var{tab2}]@dots{}
6273 @cindex tab stops, setting
6274 If only one tab stop is given, set the tabs @var{tab1} spaces apart
6275 (default is 8). Otherwise, set the tabs at columns @var{tab1},
6276 @var{tab2}, @dots{} (numbered from 0), and replace any tabs beyond the
6277 last tab stop given with single spaces. Tab stops can be separated by
6278 blanks as well as by commas.
6280 For compatibility, GNU @command{expand} also accepts the obsolete
6281 option syntax, @option{-@var{t1}[,@var{t2}]@dots{}}. New scripts
6282 should use @option{-t @var{t1}[,@var{t2}]@dots{}} instead.
6288 @cindex initial tabs, converting
6289 Only convert initial tabs (those that precede all non-space or non-tab
6290 characters) on each line to spaces.
6297 @node unexpand invocation
6298 @section @command{unexpand}: Convert spaces to tabs
6302 @command{unexpand} writes the contents of each given @var{file}, or
6303 standard input if none are given or for a @var{file} of @samp{-}, to
6304 standard output, converting blanks at the beginning of each line into
6305 as many tab characters as needed. In the default @acronym{POSIX}
6306 locale, a @dfn{blank} is a space or a tab; other locales may specify
6307 additional blank characters. Synopsis:
6310 unexpand [@var{option}]@dots{} [@var{file}]@dots{}
6313 By default, @command{unexpand} converts only initial blanks (those
6314 that precede all non-blank characters) on each line. It
6315 preserves backspace characters in the output; they decrement the column
6316 count for tab calculations. By default, tabs are set at every 8th
6319 The program accepts the following options. Also see @ref{Common options}.
6323 @item -t @var{tab1}[,@var{tab2}]@dots{}
6324 @itemx --tabs=@var{tab1}[,@var{tab2}]@dots{}
6327 If only one tab stop is given, set the tabs @var{tab1} columns apart
6328 instead of the default 8. Otherwise, set the tabs at columns
6329 @var{tab1}, @var{tab2}, @dots{} (numbered from 0), and leave blanks
6330 beyond the tab stops given unchanged. Tab stops can be separated by
6331 blanks as well as by commas. This option implies the @option{-a} option.
6333 For compatibility, GNU @command{unexpand} supports the obsolete option syntax,
6334 @option{-@var{tab1}[,@var{tab2}]@dots{}}, where tab stops must be
6335 separated by commas. (Unlike @option{-t}, this obsolete option does
6336 not imply @option{-a}.) New scripts should use @option{--first-only -t
6337 @var{tab1}[,@var{tab2}]@dots{}} instead.
6343 Also convert all sequences of two or more blanks just before a tab stop,
6344 even if they occur after non-blank characters in a line.
6351 @node Directory listing
6352 @chapter Directory listing
6354 This chapter describes the @command{ls} command and its variants @command{dir}
6355 and @command{vdir}, which list information about files.
6358 * ls invocation:: List directory contents.
6359 * dir invocation:: Briefly ls.
6360 * vdir invocation:: Verbosely ls.
6361 * dircolors invocation:: Color setup for ls, etc.
6366 @section @command{ls}: List directory contents
6369 @cindex directory listing
6371 The @command{ls} program lists information about files (of any type,
6372 including directories). Options and file arguments can be intermixed
6373 arbitrarily, as usual.
6375 For non-option command-line arguments that are directories, by default
6376 @command{ls} lists the contents of directories, not recursively, and
6377 omitting files with names beginning with @samp{.}. For other non-option
6378 arguments, by default @command{ls} lists just the file name. If no
6379 non-option argument is specified, @command{ls} operates on the current
6380 directory, acting as if it had been invoked with a single argument of @samp{.}.
6383 By default, the output is sorted alphabetically, according to the locale
6384 settings in effect.@footnote{If you use a non-@acronym{POSIX}
6385 locale (e.g., by setting @env{LC_ALL} to @samp{en_US}), then @command{ls} may
6386 produce output that is sorted differently than you're accustomed to.
6387 In that case, set the @env{LC_ALL} environment variable to @samp{C}.}
6388 If standard output is
6389 a terminal, the output is in columns (sorted vertically) and control
6390 characters are output as question marks; otherwise, the output is listed
6391 one per line and control characters are output as-is.
6393 Because @command{ls} is such a fundamental program, it has accumulated many
6394 options over the years. They are described in the subsections below;
6395 within each section, options are listed alphabetically (ignoring case).
6396 The division of options into the subsections is not absolute, since some
6397 options affect more than one aspect of @command{ls}'s operation.
6399 @cindex exit status of @command{ls}
6404 1 minor problems (e.g., failure to access a file or directory not
6405 specified as a command line argument. This happens when listing a
6406 directory in which entries are actively being removed or renamed.)
6407 2 serious trouble (e.g., memory exhausted, invalid option, failure
6408 to access a file or directory specified as a command line argument
6409 or a directory loop)
6412 Also see @ref{Common options}.
6415 * Which files are listed::
6416 * What information is listed::
6417 * Sorting the output::
6418 * Details about version sort::
6419 * General output formatting::
6420 * Formatting file timestamps::
6421 * Formatting the file names::
6425 @node Which files are listed
6426 @subsection Which files are listed
6428 These options determine which files @command{ls} lists information for.
6429 By default, @command{ls} lists files and the contents of any
6430 directories on the command line, except that in directories it ignores
6431 files whose names start with @samp{.}.
6439 In directories, do not ignore file names that start with @samp{.}.
6444 @opindex --almost-all
6445 In directories, do not ignore all file names that start with @samp{.};
6446 ignore only @file{.} and @file{..}. The @option{--all} (@option{-a})
6447 option overrides this option.
6450 @itemx --ignore-backups
6452 @opindex --ignore-backups
6453 @cindex backup files, ignoring
6454 In directories, ignore files that end with @samp{~}. This option is
6455 equivalent to @samp{--ignore='*~' --ignore='.*~'}.
6460 @opindex --directory
6461 List just the names of directories, as with other types of files, rather
6462 than listing their contents.
6463 @c The following sentence is the same as the one for -F.
6464 Do not follow symbolic links listed on the
6465 command line unless the @option{--dereference-command-line} (@option{-H}),
6466 @option{--dereference} (@option{-L}), or
6467 @option{--dereference-command-line-symlink-to-dir} options are specified.
6470 @itemx --dereference-command-line
6472 @opindex --dereference-command-line
6473 @cindex symbolic links, dereferencing
6474 If a command line argument specifies a symbolic link, show information
6475 for the file the link references rather than for the link itself.
6477 @itemx --dereference-command-line-symlink-to-dir
6478 @opindex --dereference-command-line-symlink-to-dir
6479 @cindex symbolic links, dereferencing
6480 Do not dereference symbolic links, with one exception:
6481 if a command line argument specifies a symbolic link that refers to
6482 a directory, show information for that directory rather than for the
6484 This is the default behavior when no other dereferencing-related
6485 option has been specified (@option{--classify} (@option{-F}),
6486 @option{--directory} (@option{-d}),
6488 @option{--dereference} (@option{-L}), or
6489 @option{--dereference-command-line} (@option{-H})).
6491 @item --group-directories-first
6492 @opindex --group-directories-first
6493 Group all the directories before the files and then sort the
6494 directories and the files separately using the selected sort key
6495 (see --sort option).
6496 That is, this option specifies a primary sort key,
6497 and the --sort option specifies a secondary key.
6498 However, any use of @option{--sort=none}
6499 (@option{-U}) disables this option altogether.
6501 @item --hide=PATTERN
6502 @opindex --hide=@var{pattern}
6503 In directories, ignore files whose names match the shell pattern
6504 @var{pattern}, unless the @option{--all} (@option{-a}) or
6505 @option{--almost-all} (@option{-A}) is also given. This
6506 option acts like @option{--ignore=@var{pattern}} except that it has no
6507 effect if @option{--all} (@option{-a}) or @option{--almost-all}
6508 (@option{-A}) is also given.
6510 This option can be useful in shell aliases. For example, if
6511 @command{lx} is an alias for @samp{ls --hide='*~'} and @command{ly} is
6512 an alias for @samp{ls --ignore='*~'}, then the command @samp{lx -A}
6513 lists the file @file{README~} even though @samp{ly -A} would not.
6515 @item -I @var{pattern}
6516 @itemx --ignore=@var{pattern}
6518 @opindex --ignore=@var{pattern}
6519 In directories, ignore files whose names match the shell pattern
6520 (not regular expression) @var{pattern}. As
6521 in the shell, an initial @samp{.} in a file name does not match a
6522 wildcard at the start of @var{pattern}. Sometimes it is useful
6523 to give this option several times. For example,
6526 $ ls --ignore='.??*' --ignore='.[^.]' --ignore='#*'
6529 The first option ignores names of length 3 or more that start with @samp{.},
6530 the second ignores all two-character names that start with @samp{.}
6531 except @samp{..}, and the third ignores names that start with @samp{#}.
6534 @itemx --dereference
6536 @opindex --dereference
6537 @cindex symbolic links, dereferencing
6538 When showing file information for a symbolic link, show information
6539 for the file the link references rather than the link itself.
6540 However, even with this option, @command{ls} still prints the name
6541 of the link itself, not the name of the file that the link points to.
6546 @opindex --recursive
6547 @cindex recursive directory listing
6548 @cindex directory listing, recursive
6549 List the contents of all directories recursively.
6554 @node What information is listed
6555 @subsection What information is listed
6557 These options affect the information that @command{ls} displays. By
6558 default, only file names are shown.
6564 @cindex hurd, author, printing
6565 List each file's author when producing long format directory listings.
6566 In GNU/Hurd, file authors can differ from their owners, but in other
6567 operating systems the two are the same.
6573 @cindex dired Emacs mode support
6574 With the long listing (@option{-l}) format, print an additional line after
6578 //DIRED// @var{beg1} @var{end1} @var{beg2} @var{end2} @dots{}
6582 The @var{begn} and @var{endn} are unsigned integers that record the
6583 byte position of the beginning and end of each file name in the output.
6584 This makes it easy for Emacs to find the names, even when they contain
6585 unusual characters such as space or newline, without fancy searching.
6587 If directories are being listed recursively (@option{-R}), output a similar
6588 line with offsets for each subdirectory name:
6591 //SUBDIRED// @var{beg1} @var{end1} @dots{}
6594 Finally, output a line of the form:
6597 //DIRED-OPTIONS// --quoting-style=@var{word}
6601 where @var{word} is the quoting style (@pxref{Formatting the file names}).
6603 Here is an actual example:
6606 $ mkdir -p a/sub/deeper a/sub2
6608 $ touch a/sub/deeper/file
6609 $ ls -gloRF --dired a
6612 -rw-r--r-- 1 0 Jun 10 12:27 f1
6613 -rw-r--r-- 1 0 Jun 10 12:27 f2
6614 drwxr-xr-x 3 4096 Jun 10 12:27 sub/
6615 drwxr-xr-x 2 4096 Jun 10 12:27 sub2/
6619 drwxr-xr-x 2 4096 Jun 10 12:27 deeper/
6623 -rw-r--r-- 1 0 Jun 10 12:27 file
6627 //DIRED// 48 50 84 86 120 123 158 162 217 223 282 286
6628 //SUBDIRED// 2 3 167 172 228 240 290 296
6629 //DIRED-OPTIONS// --quoting-style=literal
6632 Note that the pairs of offsets on the @samp{//DIRED//} line above delimit
6633 these names: @file{f1}, @file{f2}, @file{sub}, @file{sub2}, @file{deeper},
6635 The offsets on the @samp{//SUBDIRED//} line delimit the following
6636 directory names: @file{a}, @file{a/sub}, @file{a/sub/deeper}, @file{a/sub2}.
6638 Here is an example of how to extract the fifth entry name, @samp{deeper},
6639 corresponding to the pair of offsets, 222 and 228:
6642 $ ls -gloRF --dired a > out
6643 $ dd bs=1 skip=222 count=6 < out 2>/dev/null; echo
6647 Note that although the listing above includes a trailing slash
6648 for the @samp{deeper} entry, the offsets select the name without
6649 the trailing slash. However, if you invoke @command{ls} with @option{--dired}
6650 along with an option like @option{--escape} (aka @option{-b}) and operate
6651 on a file whose name contains special characters, notice that the backslash
6656 $ ls -blog --dired 'a b'
6657 -rw-r--r-- 1 0 Jun 10 12:28 a\ b
6659 //DIRED-OPTIONS// --quoting-style=escape
6662 If you use a quoting style that adds quote marks
6663 (e.g., @option{--quoting-style=c}), then the offsets include the quote marks.
6664 So beware that the user may select the quoting style via the environment
6665 variable @env{QUOTING_STYLE}. Hence, applications using @option{--dired}
6666 should either specify an explicit @option{--quoting-style=literal} option
6667 (aka @option{-N} or @option{--literal}) on the command line, or else be
6668 prepared to parse the escaped names.
6671 @opindex --full-time
6672 Produce long format directory listings, and list times in full. It is
6673 equivalent to using @option{--format=long} with
6674 @option{--time-style=full-iso} (@pxref{Formatting file timestamps}).
6678 Produce long format directory listings, but don't display owner information.
6684 Inhibit display of group information in a long format directory listing.
6685 (This is the default in some non-@sc{gnu} versions of @command{ls}, so we
6686 provide this option for compatibility.)
6694 @cindex inode number, printing
6695 Print the inode number (also called the file serial number and index
6696 number) of each file to the left of the file name. (This number
6697 uniquely identifies each file within a particular file system.)
6700 @itemx --format=long
6701 @itemx --format=verbose
6704 @opindex long ls @r{format}
6705 @opindex verbose ls @r{format}
6706 In addition to the name of each file, print the file type, file mode bits,
6707 number of hard links, owner name, group name, size, and
6708 timestamp (@pxref{Formatting file timestamps}), normally
6709 the modification time. Print question marks for information that
6710 cannot be determined.
6712 Normally the size is printed as a byte count without punctuation, but
6713 this can be overridden (@pxref{Block size}). For example, @option{-h}
6714 prints an abbreviated, human-readable count, and
6715 @samp{--block-size="'1"} prints a byte count with the thousands
6716 separator of the current locale.
6718 For each directory that is listed, preface the files with a line
6719 @samp{total @var{blocks}}, where @var{blocks} is the total disk allocation
6720 for all files in that directory. The block size currently defaults to 1024
6721 bytes, but this can be overridden (@pxref{Block size}).
6722 The @var{blocks} computed counts each hard link separately;
6723 this is arguably a deficiency.
6725 The file type is one of the following characters:
6727 @c The commented-out entries are ones we're not sure about.
6735 character special file
6737 high performance (``contiguous data'') file
6741 door (Solaris 2.5 and up)
6743 @c semaphore, if this is a distinct file type
6747 @c multiplexed file (7th edition Unix; obsolete)
6749 off-line (``migrated'') file (Cray DMF)
6751 network special file (HP-UX)
6755 port (Solaris 10 and up)
6757 @c message queue, if this is a distinct file type
6761 @c shared memory object, if this is a distinct file type
6763 @c typed memory object, if this is a distinct file type
6765 @c whiteout (4.4BSD; not implemented)
6767 some other file type
6770 @cindex permissions, output by @command{ls}
6771 The file mode bits listed are similar to symbolic mode specifications
6772 (@pxref{Symbolic Modes}). But @command{ls} combines multiple bits into the
6773 third character of each set of permissions as follows:
6777 If the set-user-ID or set-group-ID bit and the corresponding executable bit
6781 If the set-user-ID or set-group-ID bit is set but the corresponding
6782 executable bit is not set.
6785 If the restricted deletion flag or sticky bit, and the
6786 other-executable bit, are both set. The restricted deletion flag is
6787 another name for the sticky bit. @xref{Mode Structure}.
6790 If the restricted deletion flag or sticky bit is set but the
6791 other-executable bit is not set.
6794 If the executable bit is set and none of the above apply.
6800 Following the file mode bits is a single character that specifies
6801 whether an alternate access method such as an access control list
6802 applies to the file. When the character following the file mode bits is a
6803 space, there is no alternate access method. When it is a printing
6804 character, then there is such a method.
6806 GNU @command{ls} uses a @samp{.} character to indicate a file
6807 with an SELinux security context, but no other alternate access method.
6809 A file with any other combination of alternate access methods
6810 is marked with a @samp{+} character.
6813 @itemx --numeric-uid-gid
6815 @opindex --numeric-uid-gid
6816 @cindex numeric uid and gid
6817 @cindex numeric user and group IDs
6818 Produce long format directory listings, but
6819 display numeric user and group IDs instead of the owner and group names.
6823 Produce long format directory listings, but don't display group information.
6824 It is equivalent to using @option{--format=long} with @option{--no-group} .
6830 @cindex disk allocation
6831 @cindex size of files, reporting
6832 Print the disk allocation of each file to the left of the file name.
6833 This is the amount of disk space used by the file, which is usually a
6834 bit more than the file's size, but it can be less if the file has holes.
6836 Normally the disk allocation is printed in units of
6837 1024 bytes, but this can be overridden (@pxref{Block size}).
6839 @cindex NFS mounts from BSD to HP-UX
6840 For files that are NFS-mounted from an HP-UX system to a BSD system,
6841 this option reports sizes that are half the correct values. On HP-UX
6842 systems, it reports sizes that are twice the correct values for files
6843 that are NFS-mounted from BSD systems. This is due to a flaw in HP-UX;
6844 it also affects the HP-UX @command{ls} program.
6853 @cindex security context
6854 Display the SELinux security context or @samp{?} if none is found.
6855 When used with the @option{-l} option, print the security context
6856 to the left of the size column.
6861 @node Sorting the output
6862 @subsection Sorting the output
6864 @cindex sorting @command{ls} output
6865 These options change the order in which @command{ls} sorts the information
6866 it outputs. By default, sorting is done by character code
6867 (e.g., @acronym{ASCII} order).
6873 @itemx --time=status
6876 @opindex ctime@r{, printing or sorting by}
6877 @opindex status time@r{, printing or sorting by}
6878 @opindex use time@r{, printing or sorting files by}
6879 If the long listing format (e.g., @option{-l}, @option{-o}) is being used,
6880 print the status change time (the @samp{ctime} in the inode) instead of
6881 the modification time.
6882 When explicitly sorting by time (@option{--sort=time} or @option{-t})
6883 or when not using a long listing format,
6884 sort according to the status change time.
6888 @cindex unsorted directory listing
6889 @cindex directory order, listing by
6890 Primarily, like @option{-U}---do not sort; list the files in whatever
6891 order they are stored in the directory. But also enable @option{-a} (list
6892 all files) and disable @option{-l}, @option{--color}, and @option{-s} (if they
6893 were specified before the @option{-f}).
6899 @cindex reverse sorting
6900 Reverse whatever the sorting method is---e.g., list files in reverse
6901 alphabetical order, youngest first, smallest first, or whatever.
6907 @opindex size of files@r{, sorting files by}
6908 Sort by file size, largest first.
6914 @opindex modification time@r{, sorting files by}
6915 Sort by modification time (the @samp{mtime} in the inode), newest first.
6919 @itemx --time=access
6923 @opindex use time@r{, printing or sorting files by}
6924 @opindex atime@r{, printing or sorting files by}
6925 @opindex access time@r{, printing or sorting files by}
6926 If the long listing format (e.g., @option{--format=long}) is being used,
6927 print the last access time (the @samp{atime} in the inode).
6928 When explicitly sorting by time (@option{--sort=time} or @option{-t})
6929 or when not using a long listing format, sort according to the access time.
6935 @opindex none@r{, sorting option for @command{ls}}
6936 Do not sort; list the files in whatever order they are
6937 stored in the directory. (Do not do any of the other unrelated things
6938 that @option{-f} does.) This is especially useful when listing very large
6939 directories, since not doing any sorting can be noticeably faster.
6942 @itemx --sort=version
6945 @opindex version@r{, sorting option for @command{ls}}
6946 Sort by version name and number, lowest first. It behaves like a default
6947 sort, except that each sequence of decimal digits is treated numerically
6948 as an index/version number. (@xref{Details about version sort}.)
6951 @itemx --sort=extension
6954 @opindex extension@r{, sorting files by}
6955 Sort directory contents alphabetically by file extension (characters
6956 after the last @samp{.}); files with no extension are sorted first.
6961 @node Details about version sort
6962 @subsection Details about version sort
6964 Version sorting handles the fact that file names frequently include indices or
6965 version numbers. Standard sorting usually does not produce the order that one
6966 expects because comparisons are made on a character-by-character basis.
6967 Version sorting is especially useful when browsing directories that contain
6968 many files with indices/version numbers in their names:
6972 abc.zml-1.gz abc.zml-1.gz
6973 abc.zml-12.gz abc.zml-2.gz
6974 abc.zml-2.gz abc.zml-12.gz
6977 Version-sorted strings are compared such that if @var{ver1} and @var{ver2}
6978 are version numbers and @var{prefix} and @var{suffix} (@var{suffix} matching
6979 the regular expression @samp{(\.[A-Za-z~][A-Za-z0-9~]*)*}) are strings then
6980 @var{ver1} < @var{ver2} implies that the name composed of
6981 ``@var{prefix} @var{ver1} @var{suffix}'' sorts before
6982 ``@var{prefix} @var{ver2} @var{suffix}''.
6984 Note also that leading zeros of numeric parts are ignored:
6988 abc-1.007.tgz abc-1.01a.tgz
6989 abc-1.012b.tgz abc-1.007.tgz
6990 abc-1.01a.tgz abc-1.012b.tgz
6993 This functionality is implemented using gnulib's @code{filevercmp} function,
6994 which has some caveats worth noting.
6997 @item @env{LC_COLLATE} is ignored, which means @samp{ls -v} and @samp{sort -V}
6998 will sort non-numeric prefixes as if the @env{LC_COLLATE} locale category
6999 was set to @samp{C}.
7000 @item Some suffixes will not be matched by the regular
7001 expression mentioned above. Consequently these examples may
7002 not sort as you expect:
7010 abc-1.2.3.4.x86_64.rpm
7011 abc-1.2.3.x86_64.rpm
7015 @node General output formatting
7016 @subsection General output formatting
7018 These options affect the appearance of the overall output.
7023 @itemx --format=single-column
7026 @opindex single-column @r{output of files}
7027 List one file per line. This is the default for @command{ls} when standard
7028 output is not a terminal.
7031 @itemx --format=vertical
7034 @opindex vertical @r{sorted files in columns}
7035 List files in columns, sorted vertically. This is the default for
7036 @command{ls} if standard output is a terminal. It is always the default
7037 for the @command{dir} program.
7038 @sc{gnu} @command{ls} uses variable width columns to display as many files as
7039 possible in the fewest lines.
7041 @item --color [=@var{when}]
7043 @cindex color, distinguishing file types with
7044 Specify whether to use color for distinguishing file types. @var{when}
7045 may be omitted, or one of:
7048 @vindex none @r{color option}
7049 - Do not use color at all. This is the default.
7051 @vindex auto @r{color option}
7052 @cindex terminal, using color iff
7053 - Only use color if standard output is a terminal.
7055 @vindex always @r{color option}
7058 Specifying @option{--color} and no @var{when} is equivalent to
7059 @option{--color=always}.
7060 Piping a colorized listing through a pager like @command{more} or
7061 @command{less} usually produces unreadable results. However, using
7062 @code{more -f} does seem to work.
7065 @vindex SHELL @r{environment variable, and color}
7066 Note that using the @option{--color} option may incur a noticeable
7067 performance penalty when run in a directory with very many entries,
7068 because the default settings require that @command{ls} @code{stat} every
7069 single file it lists.
7070 However, if you would like most of the file-type coloring
7071 but can live without the other coloring options (e.g.,
7072 executable, orphan, sticky, other-writable, capability), use
7073 @command{dircolors} to set the @env{LS_COLORS} environment variable like this,
7075 eval $(dircolors -p | perl -pe \
7076 's/^((CAP|S[ET]|O[TR]|M|E)\w+).*/$1 00/' | dircolors -)
7078 and on a @code{dirent.d_type}-capable file system, @command{ls}
7079 will perform only one @code{stat} call per command line argument.
7083 @itemx --indicator-style=classify
7086 @opindex --indicator-style
7087 @cindex file type and executables, marking
7088 @cindex executables and file type, marking
7089 Append a character to each file name indicating the file type. Also,
7090 for regular files that are executable, append @samp{*}. The file type
7091 indicators are @samp{/} for directories, @samp{@@} for symbolic links,
7092 @samp{|} for FIFOs, @samp{=} for sockets, @samp{>} for doors,
7093 and nothing for regular files.
7094 @c The following sentence is the same as the one for -d.
7095 Do not follow symbolic links listed on the
7096 command line unless the @option{--dereference-command-line} (@option{-H}),
7097 @option{--dereference} (@option{-L}), or
7098 @option{--dereference-command-line-symlink-to-dir} options are specified.
7101 @itemx --indicator-style=file-type
7102 @opindex --file-type
7103 @opindex --indicator-style
7104 @cindex file type, marking
7105 Append a character to each file name indicating the file type. This is
7106 like @option{-F}, except that executables are not marked.
7108 @item --indicator-style=@var{word}
7109 @opindex --indicator-style
7110 Append a character indicator with style @var{word} to entry names,
7115 Do not append any character indicator; this is the default.
7117 Append @samp{/} for directories. This is the same as the @option{-p}
7120 Append @samp{/} for directories, @samp{@@} for symbolic links, @samp{|}
7121 for FIFOs, @samp{=} for sockets, and nothing for regular files. This is
7122 the same as the @option{--file-type} option.
7124 Append @samp{*} for executable regular files, otherwise behave as for
7125 @samp{file-type}. This is the same as the @option{-F} or
7126 @option{--classify} option.
7132 @opindex --kibibytes
7133 Set the default block size to its normal value of 1024 bytes,
7134 overriding any contrary specification in environment variables
7135 (@pxref{Block size}). This option is in turn overridden by the
7136 @option{--block-size}, @option{-h} or @option{--human-readable}, and
7137 @option{--si} options.
7139 The @option{-k} or @option{--kibibytes} option affects the
7140 per-directory block count written by the @option{-l} and similar
7141 options, and the size written by the @option{-s} or @option{--size}
7142 option. It does not affect the file size written by @option{-l}.
7145 @itemx --format=commas
7148 @opindex commas@r{, outputting between files}
7149 List files horizontally, with as many as will fit on each line,
7150 separated by @samp{, } (a comma and a space).
7153 @itemx --indicator-style=slash
7155 @opindex --indicator-style
7156 @cindex file type, marking
7157 Append a @samp{/} to directory names.
7160 @itemx --format=across
7161 @itemx --format=horizontal
7164 @opindex across@r{, listing files}
7165 @opindex horizontal@r{, listing files}
7166 List the files in columns, sorted horizontally.
7169 @itemx --tabsize=@var{cols}
7172 Assume that each tab stop is @var{cols} columns wide. The default is 8.
7173 @command{ls} uses tabs where possible in the output, for efficiency. If
7174 @var{cols} is zero, do not use tabs at all.
7176 @c FIXME: remove in 2009, if Apple Terminal has been fixed for long enough.
7177 Some terminal emulators (at least Apple Terminal 1.5 (133) from Mac OS X 10.4.8)
7178 do not properly align columns to the right of a TAB following a
7179 non-@acronym{ASCII} byte. If you use such a terminal emulator, use the
7180 @option{-T0} option or put @code{TABSIZE=0} in your environment to tell
7181 @command{ls} to align using spaces, not tabs.
7184 @itemx --width=@var{cols}
7188 Assume the screen is @var{cols} columns wide. The default is taken
7189 from the terminal settings if possible; otherwise the environment
7190 variable @env{COLUMNS} is used if it is set; otherwise the default
7196 @node Formatting file timestamps
7197 @subsection Formatting file timestamps
7199 By default, file timestamps are listed in abbreviated form, using
7200 a date like @samp{Mar 30@ @ 2002} for non-recent timestamps, and a
7201 date-without-year and time like @samp{Mar 30 23:45} for recent timestamps.
7202 This format can change depending on the current locale as detailed below.
7205 A timestamp is considered to be @dfn{recent} if it is less than six
7206 months old, and is not dated in the future. If a timestamp dated
7207 today is not listed in recent form, the timestamp is in the future,
7208 which means you probably have clock skew problems which may break
7209 programs like @command{make} that rely on file timestamps.
7212 Time stamps are listed according to the time zone rules specified by
7213 the @env{TZ} environment variable, or by the system default rules if
7214 @env{TZ} is not set. @xref{TZ Variable,, Specifying the Time Zone
7215 with @env{TZ}, libc, The GNU C Library Reference Manual}.
7217 The following option changes how file timestamps are printed.
7220 @item --time-style=@var{style}
7221 @opindex --time-style
7223 List timestamps in style @var{style}. The @var{style} should
7224 be one of the following:
7229 List timestamps using @var{format}, where @var{format} is interpreted
7230 like the format argument of @command{date} (@pxref{date invocation}).
7231 For example, @option{--time-style="+%Y-%m-%d %H:%M:%S"} causes
7232 @command{ls} to list timestamps like @samp{2002-03-30 23:45:56}. As
7233 with @command{date}, @var{format}'s interpretation is affected by the
7234 @env{LC_TIME} locale category.
7236 If @var{format} contains two format strings separated by a newline,
7237 the former is used for non-recent files and the latter for recent
7238 files; if you want output columns to line up, you may need to insert
7239 spaces in one of the two formats.
7242 List timestamps in full using @acronym{ISO} 8601 date, time, and time zone
7243 format with nanosecond precision, e.g., @samp{2002-03-30
7244 23:45:56.477817180 -0700}. This style is equivalent to
7245 @samp{+%Y-%m-%d %H:%M:%S.%N %z}.
7247 This is useful because the time output includes all the information that
7248 is available from the operating system. For example, this can help
7249 explain @command{make}'s behavior, since @acronym{GNU} @command{make}
7250 uses the full timestamp to determine whether a file is out of date.
7253 List @acronym{ISO} 8601 date and time in minutes, e.g.,
7254 @samp{2002-03-30 23:45}. These timestamps are shorter than
7255 @samp{full-iso} timestamps, and are usually good enough for everyday
7256 work. This style is equivalent to @samp{+%Y-%m-%d %H:%M}.
7259 List @acronym{ISO} 8601 dates for non-recent timestamps (e.g.,
7260 @samp{2002-03-30@ }), and @acronym{ISO} 8601 month, day, hour, and
7261 minute for recent timestamps (e.g., @samp{03-30 23:45}). These
7262 timestamps are uglier than @samp{long-iso} timestamps, but they carry
7263 nearly the same information in a smaller space and their brevity helps
7264 @command{ls} output fit within traditional 80-column output lines.
7265 The following two @command{ls} invocations are equivalent:
7270 ls -l --time-style="+%Y-%m-%d $newline%m-%d %H:%M"
7271 ls -l --time-style="iso"
7276 List timestamps in a locale-dependent form. For example, a Finnish
7277 locale might list non-recent timestamps like @samp{maalis 30@ @ 2002}
7278 and recent timestamps like @samp{maalis 30 23:45}. Locale-dependent
7279 timestamps typically consume more space than @samp{iso} timestamps and
7280 are harder for programs to parse because locale conventions vary so
7281 widely, but they are easier for many people to read.
7283 The @env{LC_TIME} locale category specifies the timestamp format. The
7284 default @acronym{POSIX} locale uses timestamps like @samp{Mar 30@
7285 @ 2002} and @samp{Mar 30 23:45}; in this locale, the following two
7286 @command{ls} invocations are equivalent:
7291 ls -l --time-style="+%b %e %Y$newline%b %e %H:%M"
7292 ls -l --time-style="locale"
7295 Other locales behave differently. For example, in a German locale,
7296 @option{--time-style="locale"} might be equivalent to
7297 @option{--time-style="+%e. %b %Y $newline%e. %b %H:%M"}
7298 and might generate timestamps like @samp{30. M@"ar 2002@ } and
7299 @samp{30. M@"ar 23:45}.
7301 @item posix-@var{style}
7303 List @acronym{POSIX}-locale timestamps if the @env{LC_TIME} locale
7304 category is @acronym{POSIX}, @var{style} timestamps otherwise. For
7305 example, the @samp{posix-long-iso} style lists
7306 timestamps like @samp{Mar 30@ @ 2002} and @samp{Mar 30 23:45} when in
7307 the @acronym{POSIX} locale, and like @samp{2002-03-30 23:45} otherwise.
7312 You can specify the default value of the @option{--time-style} option
7313 with the environment variable @env{TIME_STYLE}; if @env{TIME_STYLE} is not set
7314 the default style is @samp{locale}. @acronym{GNU} Emacs 21.3 and
7315 later use the @option{--dired} option and therefore can parse any date
7316 format, but if you are using Emacs 21.1 or 21.2 and specify a
7317 non-@acronym{POSIX} locale you may need to set
7318 @samp{TIME_STYLE="posix-long-iso"}.
7320 To avoid certain denial-of-service attacks, timestamps that would be
7321 longer than 1000 bytes may be treated as errors.
7324 @node Formatting the file names
7325 @subsection Formatting the file names
7327 These options change how file names themselves are printed.
7333 @itemx --quoting-style=escape
7336 @opindex --quoting-style
7337 @cindex backslash sequences for file names
7338 Quote nongraphic characters in file names using alphabetic and octal
7339 backslash sequences like those used in C.
7343 @itemx --quoting-style=literal
7346 @opindex --quoting-style
7347 Do not quote file names. However, with @command{ls} nongraphic
7348 characters are still printed as question marks if the output is a
7349 terminal and you do not specify the @option{--show-control-chars}
7353 @itemx --hide-control-chars
7355 @opindex --hide-control-chars
7356 Print question marks instead of nongraphic characters in file names.
7357 This is the default if the output is a terminal and the program is
7362 @itemx --quoting-style=c
7364 @opindex --quote-name
7365 @opindex --quoting-style
7366 Enclose file names in double quotes and quote nongraphic characters as
7369 @item --quoting-style=@var{word}
7370 @opindex --quoting-style
7371 @cindex quoting style
7372 Use style @var{word} to quote file names and other strings that may
7373 contain arbitrary characters. The @var{word} should
7374 be one of the following:
7378 Output strings as-is; this is the same as the @option{-N} or
7379 @option{--literal} option.
7381 Quote strings for the shell if they contain shell metacharacters or would
7382 cause ambiguous output.
7383 The quoting is suitable for @acronym{POSIX}-compatible shells like
7384 @command{bash}, but it does not always work for incompatible shells
7387 Quote strings for the shell, even if they would normally not require quoting.
7389 Quote strings as for C character string literals, including the
7390 surrounding double-quote characters; this is the same as the
7391 @option{-Q} or @option{--quote-name} option.
7393 Quote strings as for C character string literals, except omit the
7394 surrounding double-quote
7395 characters; this is the same as the @option{-b} or @option{--escape} option.
7397 Quote strings as for C character string literals, except use
7398 surrounding quotation marks appropriate for the
7401 @c Use @t instead of @samp to avoid duplicate quoting in some output styles.
7402 Quote strings as for C character string literals, except use
7403 surrounding quotation marks appropriate for the locale, and quote
7404 @t{'like this'} instead of @t{"like
7405 this"} in the default C locale. This looks nicer on many displays.
7408 You can specify the default value of the @option{--quoting-style} option
7409 with the environment variable @env{QUOTING_STYLE}. If that environment
7410 variable is not set, the default value is @samp{literal}, but this
7411 default may change to @samp{shell} in a future version of this package.
7413 @item --show-control-chars
7414 @opindex --show-control-chars
7415 Print nongraphic characters as-is in file names.
7416 This is the default unless the output is a terminal and the program is
7422 @node dir invocation
7423 @section @command{dir}: Briefly list directory contents
7426 @cindex directory listing, brief
7428 @command{dir} is equivalent to @code{ls -C
7429 -b}; that is, by default files are listed in columns, sorted vertically,
7430 and special characters are represented by backslash escape sequences.
7432 @xref{ls invocation, @command{ls}}.
7435 @node vdir invocation
7436 @section @command{vdir}: Verbosely list directory contents
7439 @cindex directory listing, verbose
7441 @command{vdir} is equivalent to @code{ls -l
7442 -b}; that is, by default files are listed in long format and special
7443 characters are represented by backslash escape sequences.
7445 @node dircolors invocation
7446 @section @command{dircolors}: Color setup for @command{ls}
7450 @cindex setup for color
7452 @command{dircolors} outputs a sequence of shell commands to set up the
7453 terminal for color output from @command{ls} (and @command{dir}, etc.).
7457 eval "`dircolors [@var{option}]@dots{} [@var{file}]`"
7460 If @var{file} is specified, @command{dircolors} reads it to determine which
7461 colors to use for which file types and extensions. Otherwise, a
7462 precompiled database is used. For details on the format of these files,
7463 run @samp{dircolors --print-database}.
7465 To make @command{dircolors} read a @file{~/.dircolors} file if it
7466 exists, you can put the following lines in your @file{~/.bashrc} (or
7467 adapt them to your favorite shell):
7471 test -r $d && eval "$(dircolors $d)"
7475 @vindex SHELL @r{environment variable, and color}
7476 The output is a shell command to set the @env{LS_COLORS} environment
7477 variable. You can specify the shell syntax to use on the command line,
7478 or @command{dircolors} will guess it from the value of the @env{SHELL}
7479 environment variable.
7481 The program accepts the following options. Also see @ref{Common options}.
7486 @itemx --bourne-shell
7489 @opindex --bourne-shell
7490 @cindex Bourne shell syntax for color setup
7491 @cindex @command{sh} syntax for color setup
7492 Output Bourne shell commands. This is the default if the @env{SHELL}
7493 environment variable is set and does not end with @samp{csh} or
7502 @cindex C shell syntax for color setup
7503 @cindex @command{csh} syntax for color setup
7504 Output C shell commands. This is the default if @code{SHELL} ends with
7505 @command{csh} or @command{tcsh}.
7508 @itemx --print-database
7510 @opindex --print-database
7511 @cindex color database, printing
7512 @cindex database for color setup, printing
7513 @cindex printing color database
7514 Print the (compiled-in) default color configuration database. This
7515 output is itself a valid configuration file, and is fairly descriptive
7516 of the possibilities.
7523 @node Basic operations
7524 @chapter Basic operations
7526 @cindex manipulating files
7528 This chapter describes the commands for basic file manipulation:
7529 copying, moving (renaming), and deleting (removing).
7532 * cp invocation:: Copy files.
7533 * dd invocation:: Convert and copy a file.
7534 * install invocation:: Copy files and set attributes.
7535 * mv invocation:: Move (rename) files.
7536 * rm invocation:: Remove files or directories.
7537 * shred invocation:: Remove files more securely.
7542 @section @command{cp}: Copy files and directories
7545 @cindex copying files and directories
7546 @cindex files, copying
7547 @cindex directories, copying
7549 @command{cp} copies files (or, optionally, directories). The copy is
7550 completely independent of the original. You can either copy one file to
7551 another, or copy arbitrarily many files to a destination directory.
7555 cp [@var{option}]@dots{} [-T] @var{source} @var{dest}
7556 cp [@var{option}]@dots{} @var{source}@dots{} @var{directory}
7557 cp [@var{option}]@dots{} -t @var{directory} @var{source}@dots{}
7562 If two file names are given, @command{cp} copies the first file to the
7566 If the @option{--target-directory} (@option{-t}) option is given, or
7567 failing that if the last file is a directory and the
7568 @option{--no-target-directory} (@option{-T}) option is not given,
7569 @command{cp} copies each @var{source} file to the specified directory,
7570 using the @var{source}s' names.
7573 Generally, files are written just as they are read. For exceptions,
7574 see the @option{--sparse} option below.
7576 By default, @command{cp} does not copy directories. However, the
7577 @option{-R}, @option{-a}, and @option{-r} options cause @command{cp} to
7578 copy recursively by descending into source directories and copying files
7579 to corresponding destination directories.
7581 When copying from a symbolic link, @command{cp} normally follows the
7582 link only when not copying
7583 recursively. This default can be overridden with the
7584 @option{--archive} (@option{-a}), @option{-d}, @option{--dereference}
7585 (@option{-L}), @option{--no-dereference} (@option{-P}), and
7586 @option{-H} options. If more than one of these options is specified,
7587 the last one silently overrides the others.
7589 When copying to a symbolic link, @command{cp} follows the
7590 link only when it refers to an existing regular file.
7591 However, when copying to a dangling symbolic link, @command{cp}
7592 refuses by default, and fails with a diagnostic, since the operation
7593 is inherently dangerous. This behavior is contrary to historical
7594 practice and to @acronym{POSIX}.
7595 Set @env{POSIXLY_CORRECT} to make @command{cp} attempt to create
7596 the target of a dangling destination symlink, in spite of the possible risk.
7597 Also, when an option like
7598 @option{--backup} or @option{--link} acts to rename or remove the
7599 destination before copying, @command{cp} renames or removes the
7600 symbolic link rather than the file it points to.
7602 By default, @command{cp} copies the contents of special files only
7603 when not copying recursively. This default can be overridden with the
7604 @option{--copy-contents} option.
7606 @cindex self-backups
7607 @cindex backups, making only
7608 @command{cp} generally refuses to copy a file onto itself, with the
7609 following exception: if @option{--force --backup} is specified with
7610 @var{source} and @var{dest} identical, and referring to a regular file,
7611 @command{cp} will make a backup file, either regular or numbered, as
7612 specified in the usual ways (@pxref{Backup options}). This is useful when
7613 you simply want to make a backup of an existing file before changing it.
7615 The program accepts the following options. Also see @ref{Common options}.
7622 Preserve as much as possible of the structure and attributes of the
7623 original files in the copy (but do not attempt to preserve internal
7624 directory structure; i.e., @samp{ls -U} may list the entries in a copied
7625 directory in a different order).
7626 Try to preserve SELinux security context and extended attributes (xattr),
7627 but ignore any failure to do that and print no corresponding diagnostic.
7628 Equivalent to @option{-dR --preserve=all} with the reduced diagnostics.
7630 @itemx --attributes-only
7631 @opindex --attributes-only
7632 Preserve the specified attributes of the original files in the copy,
7633 but do not copy any data. See the @option{--preserve} option for
7634 controlling which attributes to copy.
7637 @itemx @w{@kbd{--backup}[=@var{method}]}
7640 @vindex VERSION_CONTROL
7641 @cindex backups, making
7642 @xref{Backup options}.
7643 Make a backup of each file that would otherwise be overwritten or removed.
7644 As a special case, @command{cp} makes a backup of @var{source} when the force
7645 and backup options are given and @var{source} and @var{dest} are the same
7646 name for an existing, regular file. One useful application of this
7647 combination of options is this tiny Bourne shell script:
7651 # Usage: backup FILE...
7652 # Create a @sc{gnu}-style backup of each listed FILE.
7654 cp --backup --force -- "$i" "$i"
7658 @item --copy-contents
7659 @cindex directories, copying recursively
7660 @cindex copying directories recursively
7661 @cindex recursively copying directories
7662 @cindex non-directories, copying as special files
7663 If copying recursively, copy the contents of any special files (e.g.,
7664 FIFOs and device files) as if they were regular files. This means
7665 trying to read the data in each source file and writing it to the
7666 destination. It is usually a mistake to use this option, as it
7667 normally has undesirable effects on special files like FIFOs and the
7668 ones typically found in the @file{/dev} directory. In most cases,
7669 @code{cp -R --copy-contents} will hang indefinitely trying to read
7670 from FIFOs and special files like @file{/dev/console}, and it will
7671 fill up your destination disk if you use it to copy @file{/dev/zero}.
7672 This option has no effect unless copying recursively, and it does not
7673 affect the copying of symbolic links.
7677 @cindex symbolic links, copying
7678 @cindex hard links, preserving
7679 Copy symbolic links as symbolic links rather than copying the files that
7680 they point to, and preserve hard links between source files in the copies.
7681 Equivalent to @option{--no-dereference --preserve=links}.
7687 When copying without this option and an existing destination file cannot
7688 be opened for writing, the copy fails. However, with @option{--force}),
7689 when a destination file cannot be opened, @command{cp} then removes it and
7690 tries to open it again. Contrast this behavior with that enabled by
7691 @option{--link} and @option{--symbolic-link}, whereby the destination file
7692 is never opened but rather is removed unconditionally. Also see the
7693 description of @option{--remove-destination}.
7695 This option is independent of the @option{--interactive} or
7696 @option{-i} option: neither cancels the effect of the other.
7698 This option is redundant if the @option{--no-clobber} or @option{-n} option is
7703 If a command line argument specifies a symbolic link, then copy the
7704 file it points to rather than the symbolic link itself. However,
7705 copy (preserving its nature) any symbolic link that is encountered
7706 via recursive traversal.
7709 @itemx --interactive
7711 @opindex --interactive
7712 When copying a file other than a directory, prompt whether to
7713 overwrite an existing destination file. The @option{-i} option overrides
7714 a previous @option{-n} option.
7720 Make hard links instead of copies of non-directories.
7723 @itemx --dereference
7725 @opindex --dereference
7726 Follow symbolic links when copying from them.
7727 With this option, @command{cp} cannot create a symbolic link.
7728 For example, a symlink (to regular file) in the source tree will be copied to
7729 a regular file in the destination tree.
7734 @opindex --no-clobber
7735 Do not overwrite an existing file. The @option{-n} option overrides a previous
7736 @option{-i} option. This option is mutually exclusive with @option{-b} or
7737 @option{--backup} option.
7740 @itemx --no-dereference
7742 @opindex --no-dereference
7743 @cindex symbolic links, copying
7744 Copy symbolic links as symbolic links rather than copying the files that
7745 they point to. This option affects only symbolic links in the source;
7746 symbolic links in the destination are always followed if possible.
7749 @itemx @w{@kbd{--preserve}[=@var{attribute_list}]}
7752 @cindex file information, preserving, extended attributes, xattr
7753 Preserve the specified attributes of the original files.
7754 If specified, the @var{attribute_list} must be a comma-separated list
7755 of one or more of the following strings:
7759 Preserve the file mode bits and access control lists.
7761 Preserve the owner and group. On most modern systems,
7762 only users with appropriate privileges may change the owner of a file,
7764 may preserve the group ownership of a file only if they happen to be
7765 a member of the desired group.
7767 Preserve the times of last access and last modification, when possible.
7768 On older systems, it is not possible to preserve these attributes
7769 when the affected file is a symbolic link.
7770 However, many systems now provide the @code{utimensat} function,
7771 which makes it possible even for symbolic links.
7773 Preserve in the destination files
7774 any links between corresponding source files.
7775 Note that with @option{-L} or @option{-H}, this option can convert
7776 symbolic links to hard links. For example,
7778 $ mkdir c; : > a; ln -s a b; cp -aH a b c; ls -i1 c
7783 Note the inputs: @file{b} is a symlink to regular file @file{a},
7784 yet the files in destination directory, @file{c/}, are hard-linked.
7785 Since @option{-a} implies @option{--preserve=links}, and since @option{-H}
7786 tells @command{cp} to dereference command line arguments, it sees two files
7787 with the same inode number, and preserves the perceived hard link.
7789 Here is a similar example that exercises @command{cp}'s @option{-L} option:
7791 $ mkdir b c; (cd b; : > a; ln -s a b); cp -aL b c; ls -i1 c/b
7797 Preserve SELinux security context of the file, or fail with full diagnostics.
7799 Preserve extended attributes of the file, or fail with full diagnostics.
7800 If @command{cp} is built without xattr support, ignore this option.
7801 If SELinux context, ACLs or Capabilities are implemented using xattrs,
7802 they are preserved by this option as well.
7804 Preserve all file attributes.
7805 Equivalent to specifying all of the above, but with the difference
7806 that failure to preserve SELinux security context or extended attributes
7807 does not change @command{cp}'s exit status. In contrast to @option{-a},
7808 all but @samp{Operation not supported} warnings are output.
7811 Using @option{--preserve} with no @var{attribute_list} is equivalent
7812 to @option{--preserve=mode,ownership,timestamps}.
7814 In the absence of this option, each destination file is created with the
7815 mode bits of the corresponding source file, minus the bits set in the
7816 umask and minus the set-user-ID and set-group-ID bits.
7817 @xref{File permissions}.
7819 @itemx @w{@kbd{--no-preserve}=@var{attribute_list}}
7820 @cindex file information, preserving
7821 Do not preserve the specified attributes. The @var{attribute_list}
7822 has the same form as for @option{--preserve}.
7826 @cindex parent directories and @command{cp}
7827 Form the name of each destination file by appending to the target
7828 directory a slash and the specified name of the source file. The last
7829 argument given to @command{cp} must be the name of an existing directory.
7830 For example, the command:
7833 cp --parents a/b/c existing_dir
7837 copies the file @file{a/b/c} to @file{existing_dir/a/b/c}, creating
7838 any missing intermediate directories.
7845 @opindex --recursive
7846 @cindex directories, copying recursively
7847 @cindex copying directories recursively
7848 @cindex recursively copying directories
7849 @cindex non-directories, copying as special files
7850 Copy directories recursively. By default, do not follow symbolic
7851 links in the source; see the @option{--archive} (@option{-a}), @option{-d},
7852 @option{--dereference} (@option{-L}), @option{--no-dereference}
7853 (@option{-P}), and @option{-H} options. Special files are copied by
7854 creating a destination file of the same type as the source; see the
7855 @option{--copy-contents} option. It is not portable to use
7856 @option{-r} to copy symbolic links or special files. On some
7857 non-@sc{gnu} systems, @option{-r} implies the equivalent of
7858 @option{-L} and @option{--copy-contents} for historical reasons.
7859 Also, it is not portable to use @option{-R} to copy symbolic links
7860 unless you also specify @option{-P}, as @acronym{POSIX} allows
7861 implementations that dereference symbolic links by default.
7863 @item --reflink[=@var{when}]
7864 @opindex --reflink[=@var{when}]
7867 @cindex copy on write
7868 Perform a lightweight, copy-on-write (COW) copy, if supported by the
7869 file system. Once it has succeeded, beware that the source and destination
7870 files share the same disk data blocks as long as they remain unmodified.
7871 Thus, if a disk I/O error affects data blocks of one of the files,
7872 the other suffers the same fate.
7874 The @var{when} value can be one of the following:
7878 The default behavior: if the copy-on-write operation is not supported
7879 then report the failure for each file and exit with a failure status.
7882 If the copy-on-write operation is not supported then fall back
7883 to the standard copy behaviour.
7886 This option is overridden by the @option{--link}, @option{--symbolic-link}
7887 and @option{--attributes-only} options, thus allowing it to be used
7888 to configure the default data copying behavior for @command{cp}.
7889 For example, with the following alias, @command{cp} will use the
7890 minimum amount of space supported by the file system.
7893 alias cp='cp --reflink=auto --sparse=always'
7896 @item --remove-destination
7897 @opindex --remove-destination
7898 Remove each existing destination file before attempting to open it
7899 (contrast with @option{-f} above).
7901 @item --sparse=@var{when}
7902 @opindex --sparse=@var{when}
7903 @cindex sparse files, copying
7904 @cindex holes, copying files with
7905 @findex read @r{system call, and holes}
7906 A @dfn{sparse file} contains @dfn{holes}---a sequence of zero bytes that
7907 does not occupy any physical disk blocks; the @samp{read} system call
7908 reads these as zeros. This can both save considerable disk space and
7909 increase speed, since many binary files contain lots of consecutive zero
7910 bytes. By default, @command{cp} detects holes in input source files via a crude
7911 heuristic and makes the corresponding output file sparse as well.
7912 Only regular files may be sparse.
7914 The @var{when} value can be one of the following:
7918 The default behavior: if the input file is sparse, attempt to make
7919 the output file sparse, too. However, if an output file exists but
7920 refers to a non-regular file, then do not attempt to make it sparse.
7923 For each sufficiently long sequence of zero bytes in the input file,
7924 attempt to create a corresponding hole in the output file, even if the
7925 input file does not appear to be sparse.
7926 This is useful when the input file resides on a file system
7927 that does not support sparse files
7928 (for example, @samp{efs} file systems in SGI IRIX 5.3 and earlier),
7929 but the output file is on a type of file system that does support them.
7930 Holes may be created only in regular files, so if the destination file
7931 is of some other type, @command{cp} does not even try to make it sparse.
7934 Never make the output file sparse.
7935 This is useful in creating a file for use with the @command{mkswap} command,
7936 since such a file must not have any holes.
7939 @optStripTrailingSlashes
7942 @itemx --symbolic-link
7944 @opindex --symbolic-link
7945 @cindex symbolic links, copying with
7946 Make symbolic links instead of copies of non-directories. All source
7947 file names must be absolute (starting with @samp{/}) unless the
7948 destination files are in the current directory. This option merely
7949 results in an error message on systems that do not support symbolic links.
7955 @optNoTargetDirectory
7961 @cindex newer files, copying only
7962 Do not copy a non-directory that has an existing destination with the
7963 same or newer modification time. If time stamps are being preserved,
7964 the comparison is to the source time stamp truncated to the
7965 resolutions of the destination file system and of the system calls
7966 used to update time stamps; this avoids duplicate work if several
7967 @samp{cp -pu} commands are executed with the same source and destination.
7968 If @option{--preserve=links} is also specified (like with @samp{cp -au}
7969 for example), that will take precedence. Consequently, depending on the
7970 order that files are processed from the source, newer files in the destination
7971 may be replaced, to mirror hard links in the source.
7977 Print the name of each file before copying it.
7980 @itemx --one-file-system
7982 @opindex --one-file-system
7983 @cindex file systems, omitting copying to different
7984 Skip subdirectories that are on different file systems from the one that
7985 the copy started on.
7986 However, mount point directories @emph{are} copied.
7994 @section @command{dd}: Convert and copy a file
7997 @cindex converting while copying a file
7999 @command{dd} copies a file (from standard input to standard output, by
8000 default) with a changeable I/O block size, while optionally performing
8001 conversions on it. Synopses:
8004 dd [@var{operand}]@dots{}
8008 The only options are @option{--help} and @option{--version}.
8009 @xref{Common options}. @command{dd} accepts the following operands.
8015 Read from @var{file} instead of standard input.
8019 Write to @var{file} instead of standard output. Unless
8020 @samp{conv=notrunc} is given, @command{dd} truncates @var{file} to zero
8021 bytes (or the size specified with @samp{seek=}).
8023 @item ibs=@var{bytes}
8025 @cindex block size of input
8026 @cindex input block size
8027 Set the input block size to @var{bytes}.
8028 This makes @command{dd} read @var{bytes} per block.
8029 The default is 512 bytes.
8031 @item obs=@var{bytes}
8033 @cindex block size of output
8034 @cindex output block size
8035 Set the output block size to @var{bytes}.
8036 This makes @command{dd} write @var{bytes} per block.
8037 The default is 512 bytes.
8039 @item bs=@var{bytes}
8042 Set both input and output block sizes to @var{bytes}.
8043 This makes @command{dd} read and write @var{bytes} per block,
8044 overriding any @samp{ibs} and @samp{obs} settings.
8045 In addition, if no data-transforming @option{conv} option is specified,
8046 input is copied to the output as soon as it's read,
8047 even if it is smaller than the block size.
8049 @item cbs=@var{bytes}
8051 @cindex block size of conversion
8052 @cindex conversion block size
8053 @cindex fixed-length records, converting to variable-length
8054 @cindex variable-length records, converting to fixed-length
8055 Set the conversion block size to @var{bytes}.
8056 When converting variable-length records to fixed-length ones
8057 (@option{conv=block}) or the reverse (@option{conv=unblock}),
8058 use @var{bytes} as the fixed record length.
8062 Skip @var{n} @samp{ibs}-byte blocks in the input file before copying.
8063 If @samp{iflag=skip_bytes} is specified, @var{n} is interpreted
8064 as a byte count rather than a block count.
8068 Skip @var{n} @samp{obs}-byte blocks in the output file before copying.
8069 if @samp{oflag=seek_bytes} is specified, @var{n} is interpreted
8070 as a byte count rather than a block count.
8074 Copy @var{n} @samp{ibs}-byte blocks from the input file, instead
8075 of everything until the end of the file.
8076 if @samp{iflag=count_bytes} is specified, @var{n} is interpreted
8077 as a byte count rather than a block count.
8081 Do not print the overall transfer rate and volume statistics
8082 that normally make up the third status line when @command{dd} exits.
8084 @item conv=@var{conversion}[,@var{conversion}]@dots{}
8086 Convert the file as specified by the @var{conversion} argument(s).
8087 (No spaces around any comma(s).)
8094 @opindex ascii@r{, converting to}
8095 Convert @acronym{EBCDIC} to @acronym{ASCII},
8096 using the conversion table specified by @acronym{POSIX}.
8097 This provides a 1:1 translation for all 256 bytes.
8100 @opindex ebcdic@r{, converting to}
8101 Convert @acronym{ASCII} to @acronym{EBCDIC}.
8102 This is the inverse of the @samp{ascii} conversion.
8105 @opindex alternate ebcdic@r{, converting to}
8106 Convert @acronym{ASCII} to alternate @acronym{EBCDIC},
8107 using the alternate conversion table specified by @acronym{POSIX}.
8108 This is not a 1:1 translation, but reflects common historical practice
8109 for @samp{~}, @samp{[}, and @samp{]}.
8111 The @samp{ascii}, @samp{ebcdic}, and @samp{ibm} conversions are
8115 @opindex block @r{(space-padding)}
8116 For each line in the input, output @samp{cbs} bytes, replacing the
8117 input newline with a space and padding with spaces as necessary.
8121 Remove any trailing spaces in each @samp{cbs}-sized input block,
8122 and append a newline.
8124 The @samp{block} and @samp{unblock} conversions are mutually exclusive.
8127 @opindex lcase@r{, converting to}
8128 Change uppercase letters to lowercase.
8131 @opindex ucase@r{, converting to}
8132 Change lowercase letters to uppercase.
8134 The @samp{lcase} and @samp{ucase} conversions are mutually exclusive.
8137 @opindex swab @r{(byte-swapping)}
8138 @cindex byte-swapping
8139 Swap every pair of input bytes. @sc{gnu} @command{dd}, unlike others, works
8140 when an odd number of bytes are read---the last byte is simply copied
8141 (since there is nothing to swap it with).
8144 @opindex sync @r{(padding with @acronym{ASCII} @sc{nul}s)}
8145 Pad every input block to size of @samp{ibs} with trailing zero bytes.
8146 When used with @samp{block} or @samp{unblock}, pad with spaces instead of
8151 The following ``conversions'' are really file flags
8152 and don't affect internal processing:
8157 @cindex creating output file, requiring
8158 Fail if the output file already exists; @command{dd} must create the
8163 @cindex creating output file, avoiding
8164 Do not create the output file; the output file must already exist.
8166 The @samp{excl} and @samp{nocreat} conversions are mutually exclusive.
8170 @cindex truncating output file, avoiding
8171 Do not truncate the output file.
8175 @cindex read errors, ignoring
8176 Continue after read errors.
8180 @cindex synchronized data writes, before finishing
8181 Synchronize output data just before finishing. This forces a physical
8182 write of output data.
8186 @cindex synchronized data and metadata writes, before finishing
8187 Synchronize output data and metadata just before finishing. This
8188 forces a physical write of output data and metadata.
8192 @item iflag=@var{flag}[,@var{flag}]@dots{}
8194 Access the input file using the flags specified by the @var{flag}
8195 argument(s). (No spaces around any comma(s).)
8197 @item oflag=@var{flag}[,@var{flag}]@dots{}
8199 Access the output file using the flags specified by the @var{flag}
8200 argument(s). (No spaces around any comma(s).)
8202 Here are the flags. Not every flag is supported on every operating
8209 @cindex appending to the output file
8210 Write in append mode, so that even if some other process is writing to
8211 this file, every @command{dd} write will append to the current
8212 contents of the file. This flag makes sense only for output.
8213 If you combine this flag with the @samp{of=@var{file}} operand,
8214 you should also specify @samp{conv=notrunc} unless you want the
8215 output file to be truncated before being appended to.
8219 @cindex concurrent I/O
8220 Use concurrent I/O mode for data. This mode performs direct I/O
8221 and drops the @acronym{POSIX} requirement to serialize all I/O to the same file.
8222 A file cannot be opened in CIO mode and with a standard open at the
8228 Use direct I/O for data, avoiding the buffer cache.
8229 Note that the kernel may impose restrictions on read or write buffer sizes.
8230 For example, with an ext4 destination file system and a linux-based kernel,
8231 using @samp{oflag=direct} will cause writes to fail with @code{EINVAL} if the
8232 output buffer size is not a multiple of 512.
8236 @cindex directory I/O
8238 Fail unless the file is a directory. Most operating systems do not
8239 allow I/O to a directory, so this flag has limited utility.
8243 @cindex synchronized data reads
8244 Use synchronized I/O for data. For the output file, this forces a
8245 physical write of output data on each write. For the input file,
8246 this flag can matter when reading from a remote file that has been
8247 written to synchronously by some other process. Metadata (e.g.,
8248 last-access and last-modified time) is not necessarily synchronized.
8252 @cindex synchronized data and metadata I/O
8253 Use synchronized I/O for both data and metadata.
8257 @cindex discarding file cache
8258 Discard the data cache for a file.
8259 When count=0 all cache is discarded,
8260 otherwise the cache is dropped for the processed
8261 portion of the file. Also when count=0
8262 failure to discard the cache is diagnosed
8263 and reflected in the exit status.
8264 Here as some usage examples:
8267 # Advise to drop cache for whole file
8268 dd if=ifile iflag=nocache count=0
8270 # Ensure drop cache for the whole file
8271 dd of=ofile oflag=nocache conv=notrunc,fdatasync count=0
8273 # Drop cache for part of file
8274 dd if=ifile iflag=nocache skip=10 count=10 of=/dev/null
8276 # Stream data using just the read-ahead cache
8277 dd if=ifile of=ofile iflag=nocache oflag=nocache
8282 @cindex nonblocking I/O
8283 Use non-blocking I/O.
8288 Do not update the file's access time.
8289 Some older file systems silently ignore this flag, so it is a good
8290 idea to test it on your files before relying on it.
8294 @cindex controlling terminal
8295 Do not assign the file to be a controlling terminal for @command{dd}.
8296 This has no effect when the file is not a terminal.
8297 On many hosts (e.g., @acronym{GNU}/Linux hosts), this option has no effect
8302 @cindex symbolic links, following
8303 Do not follow symbolic links.
8308 Fail if the file has multiple hard links.
8313 Use binary I/O. This option has an effect only on nonstandard
8314 platforms that distinguish binary from text I/O.
8319 Use text I/O. Like @samp{binary}, this option has no effect on
8324 Accumulate full blocks from input. The @code{read} system call
8325 may return early if a full block is not available.
8326 When that happens, continue calling @code{read} to fill the remainder
8328 This flag can be used only with @code{iflag}.
8331 @opindex count_bytes
8332 Interpret the @samp{count=} operand as a byte count,
8333 rather than a block count, which allows specifying
8334 a length that is not a multiple of the I/O block size.
8335 This flag can be used only with @code{iflag}.
8339 Interpret the @samp{skip=} operand as a byte count,
8340 rather than a block count, which allows specifying
8341 an offset that is not a multiple of the I/O block size.
8342 This flag can be used only with @code{iflag}.
8346 Interpret the @samp{seek=} operand as a byte count,
8347 rather than a block count, which allows specifying
8348 an offset that is not a multiple of the I/O block size.
8349 This flag can be used only with @code{oflag}.
8353 These flags are not supported on all systems, and @samp{dd} rejects
8354 attempts to use them when they are not supported. When reading from
8355 standard input or writing to standard output, the @samp{nofollow} and
8356 @samp{noctty} flags should not be specified, and the other flags
8357 (e.g., @samp{nonblock}) can affect how other processes behave with the
8358 affected file descriptors, even after @command{dd} exits.
8362 @cindex multipliers after numbers
8363 The numeric-valued strings above (@var{bytes} and @var{blocks}) can be
8364 followed by a multiplier: @samp{b}=512, @samp{c}=1,
8365 @samp{w}=2, @samp{x@var{m}}=@var{m}, or any of the
8366 standard block size suffixes like @samp{k}=1024 (@pxref{Block size}).
8368 Any block size you specify via @samp{bs=}, @samp{ibs=}, @samp{obs=}, @samp{cbs=}
8369 should not be too large---values larger than a few megabytes
8370 are generally wasteful or (as in the gigabyte..exabyte case) downright
8371 counterproductive or error-inducing.
8373 To process data that is at an offset or size that is not a
8374 multiple of the I/O@ block size, you can use the @samp{skip_bytes},
8375 @samp{seek_bytes} and @samp{count_bytes} flags. Alternatively
8376 the traditional method of separate @command{dd} invocations can be used.
8377 For example, the following shell commands copy data
8378 in 512 KiB blocks between a disk and a tape, but do not save
8379 or restore a 4 KiB label at the start of the disk:
8382 disk=/dev/rdsk/c0t1d0s2
8385 # Copy all but the label from disk to tape.
8386 (dd bs=4k skip=1 count=0 && dd bs=512k) <$disk >$tape
8388 # Copy from tape back to disk, but leave the disk label alone.
8389 (dd bs=4k seek=1 count=0 && dd bs=512k) <$tape >$disk
8392 Sending an @samp{INFO} signal to a running @command{dd}
8393 process makes it print I/O statistics to standard error
8394 and then resume copying. In the example below,
8395 @command{dd} is run in the background to copy 10 million blocks.
8396 The @command{kill} command makes it output intermediate I/O statistics,
8397 and when @command{dd} completes normally or is killed by the
8398 @code{SIGINT} signal, it outputs the final statistics.
8401 $ dd if=/dev/zero of=/dev/null count=10MB & pid=$!
8402 $ kill -s INFO $pid; wait $pid
8403 3385223+0 records in
8404 3385223+0 records out
8405 1733234176 bytes (1.7 GB) copied, 6.42173 seconds, 270 MB/s
8406 10000000+0 records in
8407 10000000+0 records out
8408 5120000000 bytes (5.1 GB) copied, 18.913 seconds, 271 MB/s
8411 @vindex POSIXLY_CORRECT
8412 On systems lacking the @samp{INFO} signal @command{dd} responds to the
8413 @samp{USR1} signal instead, unless the @env{POSIXLY_CORRECT}
8414 environment variable is set.
8419 @node install invocation
8420 @section @command{install}: Copy files and set attributes
8423 @cindex copying files and setting attributes
8425 @command{install} copies files while setting their file mode bits and, if
8426 possible, their owner and group. Synopses:
8429 install [@var{option}]@dots{} [-T] @var{source} @var{dest}
8430 install [@var{option}]@dots{} @var{source}@dots{} @var{directory}
8431 install [@var{option}]@dots{} -t @var{directory} @var{source}@dots{}
8432 install [@var{option}]@dots{} -d @var{directory}@dots{}
8437 If two file names are given, @command{install} copies the first file to the
8441 If the @option{--target-directory} (@option{-t}) option is given, or
8442 failing that if the last file is a directory and the
8443 @option{--no-target-directory} (@option{-T}) option is not given,
8444 @command{install} copies each @var{source} file to the specified
8445 directory, using the @var{source}s' names.
8448 If the @option{--directory} (@option{-d}) option is given,
8449 @command{install} creates each @var{directory} and any missing parent
8450 directories. Parent directories are created with mode
8451 @samp{u=rwx,go=rx} (755), regardless of the @option{-m} option or the
8452 current umask. @xref{Directory Setuid and Setgid}, for how the
8453 set-user-ID and set-group-ID bits of parent directories are inherited.
8456 @cindex Makefiles, installing programs in
8457 @command{install} is similar to @command{cp}, but allows you to control the
8458 attributes of destination files. It is typically used in Makefiles to
8459 copy programs into their destination directories. It refuses to copy
8460 files onto themselves.
8462 @cindex extended attributes, xattr
8463 @command{install} never preserves extended attributes (xattr).
8465 The program accepts the following options. Also see @ref{Common options}.
8475 Compare each pair of source and destination files, and if the destination has
8476 identical content and any specified owner, group, permissions, and possibly
8477 SELinux context, then do not modify the destination at all.
8481 Ignored; for compatibility with old Unix versions of @command{install}.
8485 Create any missing parent directories of @var{dest},
8486 then copy @var{source} to @var{dest}.
8487 This option is ignored if a destination directory is specified
8488 via @option{--target-directory=DIR}.
8493 @opindex --directory
8494 @cindex directories, creating with given attributes
8495 @cindex parent directories, creating missing
8496 @cindex leading directories, creating missing
8497 Create any missing parent directories, giving them the default
8498 attributes. Then create each given directory, setting their owner,
8499 group and mode as given on the command line or to the defaults.
8501 @item -g @var{group}
8502 @itemx --group=@var{group}
8505 @cindex group ownership of installed files, setting
8506 Set the group ownership of installed files or directories to
8507 @var{group}. The default is the process's current group. @var{group}
8508 may be either a group name or a numeric group ID.
8511 @itemx --mode=@var{mode}
8514 @cindex permissions of installed files, setting
8515 Set the file mode bits for the installed file or directory to @var{mode},
8516 which can be either an octal number, or a symbolic mode as in
8517 @command{chmod}, with @samp{a=} (no access allowed to anyone) as the
8518 point of departure (@pxref{File permissions}).
8519 The default mode is @samp{u=rwx,go=rx,a-s}---read, write, and
8520 execute for the owner, read and execute for group and other, and with
8521 set-user-ID and set-group-ID disabled.
8522 This default is not quite the same as @samp{755}, since it disables
8523 instead of preserving set-user-ID and set-group-ID on directories.
8524 @xref{Directory Setuid and Setgid}.
8526 @item -o @var{owner}
8527 @itemx --owner=@var{owner}
8530 @cindex ownership of installed files, setting
8531 @cindex appropriate privileges
8532 @vindex root @r{as default owner}
8533 If @command{install} has appropriate privileges (is run as root), set the
8534 ownership of installed files or directories to @var{owner}. The default
8535 is @code{root}. @var{owner} may be either a user name or a numeric user
8538 @item --preserve-context
8539 @opindex --preserve-context
8541 @cindex security context
8542 Preserve the SELinux security context of files and directories.
8543 Failure to preserve the context in all of the files or directories
8544 will result in an exit status of 1. If SELinux is disabled then
8545 print a warning and ignore the option.
8548 @itemx --preserve-timestamps
8550 @opindex --preserve-timestamps
8551 @cindex timestamps of installed files, preserving
8552 Set the time of last access and the time of last modification of each
8553 installed file to match those of each corresponding original file.
8554 When a file is installed without this option, its last access and
8555 last modification times are both set to the time of installation.
8556 This option is useful if you want to use the last modification times
8557 of installed files to keep track of when they were last built as opposed
8558 to when they were last installed.
8564 @cindex symbol table information, stripping
8565 @cindex stripping symbol table information
8566 Strip the symbol tables from installed binary executables.
8568 @itemx --strip-program=@var{program}
8569 @opindex --strip-program
8570 @cindex symbol table information, stripping, program
8571 Program used to strip binaries.
8577 @optNoTargetDirectory
8583 Print the name of each file before copying it.
8585 @item -Z @var{context}
8586 @itemx --context=@var{context}
8590 @cindex security context
8591 Set the default SELinux security context to be used for any
8592 created files and directories. If SELinux is disabled then
8593 print a warning and ignore the option.
8601 @section @command{mv}: Move (rename) files
8605 @command{mv} moves or renames files (or directories). Synopses:
8608 mv [@var{option}]@dots{} [-T] @var{source} @var{dest}
8609 mv [@var{option}]@dots{} @var{source}@dots{} @var{directory}
8610 mv [@var{option}]@dots{} -t @var{directory} @var{source}@dots{}
8615 If two file names are given, @command{mv} moves the first file to the
8619 If the @option{--target-directory} (@option{-t}) option is given, or
8620 failing that if the last file is a directory and the
8621 @option{--no-target-directory} (@option{-T}) option is not given,
8622 @command{mv} moves each @var{source} file to the specified
8623 directory, using the @var{source}s' names.
8626 @command{mv} can move any type of file from one file system to another.
8627 Prior to version @code{4.0} of the fileutils,
8628 @command{mv} could move only regular files between file systems.
8629 For example, now @command{mv} can move an entire directory hierarchy
8630 including special device files from one partition to another. It first
8631 uses some of the same code that's used by @code{cp -a} to copy the
8632 requested directories and files, then (assuming the copy succeeded)
8633 it removes the originals. If the copy fails, then the part that was
8634 copied to the destination partition is removed. If you were to copy
8635 three directories from one partition to another and the copy of the first
8636 directory succeeded, but the second didn't, the first would be left on
8637 the destination partition and the second and third would be left on the
8640 @cindex extended attributes, xattr
8641 @command{mv} always tries to copy extended attributes (xattr), which may
8642 include SELinux context, ACLs or Capabilities.
8643 Upon failure all but @samp{Operation not supported} warnings are output.
8645 @cindex prompting, and @command{mv}
8646 If a destination file exists but is normally unwritable, standard input
8647 is a terminal, and the @option{-f} or @option{--force} option is not given,
8648 @command{mv} prompts the user for whether to replace the file. (You might
8649 own the file, or have write permission on its directory.) If the
8650 response is not affirmative, the file is skipped.
8652 @emph{Warning}: Avoid specifying a source name with a trailing slash,
8653 when it might be a symlink to a directory.
8654 Otherwise, @command{mv} may do something very surprising, since
8655 its behavior depends on the underlying rename system call.
8656 On a system with a modern Linux-based kernel, it fails with
8657 @code{errno=ENOTDIR}.
8658 However, on other systems (at least FreeBSD 6.1 and Solaris 10) it silently
8659 renames not the symlink but rather the directory referenced by the symlink.
8660 @xref{Trailing slashes}.
8662 The program accepts the following options. Also see @ref{Common options}.
8672 @cindex prompts, omitting
8673 Do not prompt the user before removing a destination file.
8675 If you specify more than one of the @option{-i}, @option{-f}, @option{-n}
8676 options, only the final one takes effect.
8681 @itemx --interactive
8683 @opindex --interactive
8684 @cindex prompts, forcing
8685 Prompt whether to overwrite each existing destination file, regardless
8687 If the response is not affirmative, the file is skipped.
8693 @opindex --no-clobber
8694 @cindex prompts, omitting
8695 Do not overwrite an existing file.
8697 This option is mutually exclusive with @option{-b} or @option{--backup} option.
8703 @cindex newer files, moving only
8704 Do not move a non-directory that has an existing destination with the
8705 same or newer modification time.
8706 If the move is across file system boundaries, the comparison is to the
8707 source time stamp truncated to the resolutions of the destination file
8708 system and of the system calls used to update time stamps; this avoids
8709 duplicate work if several @samp{mv -u} commands are executed with the
8710 same source and destination.
8716 Print the name of each file before moving it.
8718 @optStripTrailingSlashes
8724 @optNoTargetDirectory
8732 @section @command{rm}: Remove files or directories
8735 @cindex removing files or directories
8737 @command{rm} removes each given @var{file}. By default, it does not remove
8738 directories. Synopsis:
8741 rm [@var{option}]@dots{} [@var{file}]@dots{}
8744 @cindex prompting, and @command{rm}
8745 If the @option{-I} or @option{--interactive=once} option is given,
8746 and there are more than three files or the @option{-r}, @option{-R},
8747 or @option{--recursive} are given, then @command{rm} prompts the user
8748 for whether to proceed with the entire operation. If the response is
8749 not affirmative, the entire command is aborted.
8751 Otherwise, if a file is unwritable, standard input is a terminal, and
8752 the @option{-f} or @option{--force} option is not given, or the
8753 @option{-i} or @option{--interactive=always} option @emph{is} given,
8754 @command{rm} prompts the user for whether to remove the file.
8755 If the response is not affirmative, the file is skipped.
8757 Any attempt to remove a file whose last file name component is
8758 @file{.} or @file{..} is rejected without any prompting.
8760 @emph{Warning}: If you use @command{rm} to remove a file, it is usually
8761 possible to recover the contents of that file. If you want more assurance
8762 that the contents are truly unrecoverable, consider using @command{shred}.
8764 The program accepts the following options. Also see @ref{Common options}.
8772 Ignore nonexistent files and missing operands, and never prompt the user.
8773 Ignore any previous @option{--interactive} (@option{-i}) option.
8777 Prompt whether to remove each file.
8778 If the response is not affirmative, the file is skipped.
8779 Ignore any previous @option{--force} (@option{-f}) option.
8780 Equivalent to @option{--interactive=always}.
8784 Prompt once whether to proceed with the command, if more than three
8785 files are named or if a recursive removal is requested. Ignore any
8786 previous @option{--force} (@option{-f}) option. Equivalent to
8787 @option{--interactive=once}.
8789 @itemx --interactive [=@var{when}]
8790 @opindex --interactive
8791 Specify when to issue an interactive prompt. @var{when} may be
8795 @vindex never @r{interactive option}
8796 - Do not prompt at all.
8798 @vindex once @r{interactive option}
8799 - Prompt once if more than three files are named or if a recursive
8800 removal is requested. Equivalent to @option{-I}.
8802 @vindex always @r{interactive option}
8803 - Prompt for every file being removed. Equivalent to @option{-i}.
8805 @option{--interactive} with no @var{when} is equivalent to
8806 @option{--interactive=always}.
8808 @itemx --one-file-system
8809 @opindex --one-file-system
8810 @cindex one file system, restricting @command{rm} to
8811 When removing a hierarchy recursively, skip any directory that is on a
8812 file system different from that of the corresponding command line argument.
8815 This option is useful when removing a build ``chroot'' hierarchy,
8816 which normally contains no valuable data. However, it is not uncommon
8817 to bind-mount @file{/home} into such a hierarchy, to make it easier to
8818 use one's start-up file. The catch is that it's easy to forget to
8819 unmount @file{/home}. Then, when you use @command{rm -rf} to remove
8820 your normally throw-away chroot, that command will remove everything
8821 under @file{/home}, too.
8822 Use the @option{--one-file-system} option, and it will
8823 warn about and skip directories on other file systems.
8824 Of course, this will not save your @file{/home} if it and your
8825 chroot happen to be on the same file system.
8827 @itemx --preserve-root
8828 @opindex --preserve-root
8829 @cindex root directory, disallow recursive destruction
8830 Fail upon any attempt to remove the root directory, @file{/},
8831 when used with the @option{--recursive} option.
8832 This is the default behavior.
8833 @xref{Treating / specially}.
8835 @itemx --no-preserve-root
8836 @opindex --no-preserve-root
8837 @cindex root directory, allow recursive destruction
8838 Do not treat @file{/} specially when removing recursively.
8839 This option is not recommended unless you really want to
8840 remove all the files on your computer.
8841 @xref{Treating / specially}.
8848 @opindex --recursive
8849 @cindex directories, removing (recursively)
8850 Remove the listed directories and their contents recursively.
8856 Print the name of each file before removing it.
8860 @cindex files beginning with @samp{-}, removing
8861 @cindex @samp{-}, removing files beginning with
8862 One common question is how to remove files whose names begin with a
8863 @samp{-}. @sc{gnu} @command{rm}, like every program that uses the @code{getopt}
8864 function to parse its arguments, lets you use the @samp{--} option to
8865 indicate that all following arguments are non-options. To remove a file
8866 called @file{-f} in the current directory, you could type either:
8879 @opindex - @r{and Unix @command{rm}}
8880 The Unix @command{rm} program's use of a single @samp{-} for this purpose
8881 predates the development of the getopt standard syntax.
8886 @node shred invocation
8887 @section @command{shred}: Remove files more securely
8890 @cindex data, erasing
8891 @cindex erasing data
8893 @command{shred} overwrites devices or files, to help prevent even
8894 very expensive hardware from recovering the data.
8896 Ordinarily when you remove a file (@pxref{rm invocation}), the data is
8897 not actually destroyed. Only the index listing where the file is
8898 stored is destroyed, and the storage is made available for reuse.
8899 There are undelete utilities that will attempt to reconstruct the index
8900 and can bring the file back if the parts were not reused.
8902 On a busy system with a nearly-full drive, space can get reused in a few
8903 seconds. But there is no way to know for sure. If you have sensitive
8904 data, you may want to be sure that recovery is not possible by actually
8905 overwriting the file with non-sensitive data.
8907 However, even after doing that, it is possible to take the disk back
8908 to a laboratory and use a lot of sensitive (and expensive) equipment
8909 to look for the faint ``echoes'' of the original data underneath the
8910 overwritten data. If the data has only been overwritten once, it's not
8913 The best way to remove something irretrievably is to destroy the media
8914 it's on with acid, melt it down, or the like. For cheap removable media
8915 like floppy disks, this is the preferred method. However, hard drives
8916 are expensive and hard to melt, so the @command{shred} utility tries
8917 to achieve a similar effect non-destructively.
8919 This uses many overwrite passes, with the data patterns chosen to
8920 maximize the damage they do to the old data. While this will work on
8921 floppies, the patterns are designed for best effect on hard drives.
8922 For more details, see the source code and Peter Gutmann's paper
8923 @uref{http://www.cs.auckland.ac.nz/~pgut001/pubs/secure_del.html,
8924 @cite{Secure Deletion of Data from Magnetic and Solid-State Memory}},
8925 from the proceedings of the Sixth @acronym{USENIX} Security Symposium (San Jose,
8926 California, July 22--25, 1996).
8928 @strong{Please note} that @command{shred} relies on a very important assumption:
8929 that the file system overwrites data in place. This is the traditional
8930 way to do things, but many modern file system designs do not satisfy this
8931 assumption. Exceptions include:
8936 Log-structured or journaled file systems, such as those supplied with
8937 AIX and Solaris, and JFS, ReiserFS, XFS, Ext3 (in @code{data=journal} mode),
8938 BFS, NTFS, etc.@: when they are configured to journal @emph{data}.
8941 File systems that write redundant data and carry on even if some writes
8942 fail, such as RAID-based file systems.
8945 File systems that make snapshots, such as Network Appliance's NFS server.
8948 File systems that cache in temporary locations, such as NFS version 3
8952 Compressed file systems.
8955 In the particular case of ext3 file systems, the above disclaimer applies (and
8956 @command{shred} is thus of limited effectiveness) only in @code{data=journal}
8957 mode, which journals file data in addition to just metadata. In both
8958 the @code{data=ordered} (default) and @code{data=writeback} modes,
8959 @command{shred} works as usual. Ext3 journaling modes can be changed
8960 by adding the @code{data=something} option to the mount options for a
8961 particular file system in the @file{/etc/fstab} file, as documented in
8962 the mount man page (man mount).
8964 If you are not sure how your file system operates, then you should assume
8965 that it does not overwrite data in place, which means that shred cannot
8966 reliably operate on regular files in your file system.
8968 Generally speaking, it is more reliable to shred a device than a file,
8969 since this bypasses the problem of file system design mentioned above.
8970 However, even shredding devices is not always completely reliable. For
8971 example, most disks map out bad sectors invisibly to the application; if
8972 the bad sectors contain sensitive data, @command{shred} won't be able to
8975 @command{shred} makes no attempt to detect or report this problem, just as
8976 it makes no attempt to do anything about backups. However, since it is
8977 more reliable to shred devices than files, @command{shred} by default does
8978 not truncate or remove the output file. This default is more suitable
8979 for devices, which typically cannot be truncated and should not be
8982 Finally, consider the risk of backups and mirrors.
8983 File system backups and remote mirrors may contain copies of the
8984 file that cannot be removed, and that will allow a shredded file
8985 to be recovered later. So if you keep any data you may later want
8986 to destroy using @command{shred}, be sure that it is not backed up or mirrored.
8989 shred [@var{option}]@dots{} @var{file}[@dots{}]
8992 The program accepts the following options. Also see @ref{Common options}.
9000 @cindex force deletion
9001 Override file permissions if necessary to allow overwriting.
9004 @itemx -n @var{number}
9005 @itemx --iterations=@var{number}
9006 @opindex -n @var{number}
9007 @opindex --iterations=@var{number}
9008 @cindex iterations, selecting the number of
9009 By default, @command{shred} uses @value{SHRED_DEFAULT_PASSES} passes of
9010 overwrite. You can reduce this to save time, or increase it if you think it's
9011 appropriate. After 25 passes all of the internal overwrite patterns will have
9012 been used at least once.
9014 @item --random-source=@var{file}
9015 @opindex --random-source
9016 @cindex random source for shredding
9017 Use @var{file} as a source of random data used to overwrite and to
9018 choose pass ordering. @xref{Random sources}.
9020 @item -s @var{bytes}
9021 @itemx --size=@var{bytes}
9022 @opindex -s @var{bytes}
9023 @opindex --size=@var{bytes}
9024 @cindex size of file to shred
9025 Shred the first @var{bytes} bytes of the file. The default is to shred
9026 the whole file. @var{bytes} can be followed by a size specification like
9027 @samp{K}, @samp{M}, or @samp{G} to specify a multiple. @xref{Block size}.
9033 @cindex removing files after shredding
9034 After shredding a file, truncate it (if possible) and then remove it.
9035 If a file has multiple links, only the named links will be removed.
9041 Display to standard error all status updates as sterilization proceeds.
9047 By default, @command{shred} rounds the size of a regular file up to the next
9048 multiple of the file system block size to fully erase the last block
9050 Use @option{--exact} to suppress that behavior.
9051 Thus, by default if you shred a 10-byte regular file on a system with 512-byte
9052 blocks, the resulting file will be 512 bytes long. With this option,
9053 shred does not increase the apparent size of the file.
9059 Normally, the last pass that @command{shred} writes is made up of
9060 random data. If this would be conspicuous on your hard drive (for
9061 example, because it looks like encrypted data), or you just think
9062 it's tidier, the @option{--zero} option adds an additional overwrite pass with
9063 all zero bits. This is in addition to the number of passes specified
9064 by the @option{--iterations} option.
9068 You might use the following command to erase all trace of the
9069 file system you'd created on the floppy disk in your first drive.
9070 That command takes about 20 minutes to erase a ``1.44MB'' (actually
9074 shred --verbose /dev/fd0
9077 Similarly, to erase all data on a selected partition of
9078 your hard disk, you could give a command like this:
9081 shred --verbose /dev/sda5
9084 On modern disks, a single pass should be adequate,
9085 and it will take one third the time of the default three-pass approach.
9088 # 1 pass, write pseudo-random data; 3x faster than the default
9089 shred --verbose -n1 /dev/sda5
9092 To be on the safe side, use at least one pass that overwrites using
9093 pseudo-random data. I.e., don't be tempted to use @samp{-n0 --zero},
9094 in case some disk controller optimizes the process of writing blocks
9095 of all zeros, and thereby does not clear all bytes in a block.
9096 Some SSDs may do just that.
9098 A @var{file} of @samp{-} denotes standard output.
9099 The intended use of this is to shred a removed temporary file.
9106 echo "Hello, world" >&3
9111 However, the command @samp{shred - >file} does not shred the contents
9112 of @var{file}, since the shell truncates @var{file} before invoking
9113 @command{shred}. Use the command @samp{shred file} or (if using a
9114 Bourne-compatible shell) the command @samp{shred - 1<>file} instead.
9119 @node Special file types
9120 @chapter Special file types
9122 @cindex special file types
9123 @cindex file types, special
9125 This chapter describes commands which create special types of files (and
9126 @command{rmdir}, which removes directories, one special file type).
9128 @cindex special file types
9130 Although Unix-like operating systems have markedly fewer special file
9131 types than others, not @emph{everything} can be treated only as the
9132 undifferentiated byte stream of @dfn{normal files}. For example, when a
9133 file is created or removed, the system must record this information,
9134 which it does in a @dfn{directory}---a special type of file. Although
9135 you can read directories as normal files, if you're curious, in order
9136 for the system to do its job it must impose a structure, a certain
9137 order, on the bytes of the file. Thus it is a ``special'' type of file.
9139 Besides directories, other special file types include named pipes
9140 (FIFOs), symbolic links, sockets, and so-called @dfn{special files}.
9143 * link invocation:: Make a hard link via the link syscall
9144 * ln invocation:: Make links between files.
9145 * mkdir invocation:: Make directories.
9146 * mkfifo invocation:: Make FIFOs (named pipes).
9147 * mknod invocation:: Make block or character special files.
9148 * readlink invocation:: Print value of a symlink or canonical file name.
9149 * rmdir invocation:: Remove empty directories.
9150 * unlink invocation:: Remove files via the unlink syscall
9154 @node link invocation
9155 @section @command{link}: Make a hard link via the link syscall
9158 @cindex links, creating
9159 @cindex hard links, creating
9160 @cindex creating links (hard only)
9162 @command{link} creates a single hard link at a time.
9163 It is a minimalist interface to the system-provided
9164 @code{link} function. @xref{Hard Links, , , libc,
9165 The GNU C Library Reference Manual}.
9166 It avoids the bells and whistles of the more commonly-used
9167 @command{ln} command (@pxref{ln invocation}).
9171 link @var{filename} @var{linkname}
9174 @var{filename} must specify an existing file, and @var{linkname}
9175 must specify a nonexistent entry in an existing directory.
9176 @command{link} simply calls @code{link (@var{filename}, @var{linkname})}
9179 On a @acronym{GNU} system, this command acts like @samp{ln --directory
9180 --no-target-directory @var{filename} @var{linkname}}. However, the
9181 @option{--directory} and @option{--no-target-directory} options are
9182 not specified by @acronym{POSIX}, and the @command{link} command is
9183 more portable in practice.
9185 If @var{filename} is a symbolic link, it is unspecified whether
9186 @var{linkname} will be a hard link to the symbolic link or to the
9187 target of the symbolic link. Use @command{ln -P} or @command{ln -L}
9188 to specify which behavior is desired.
9194 @section @command{ln}: Make links between files
9197 @cindex links, creating
9198 @cindex hard links, creating
9199 @cindex symbolic (soft) links, creating
9200 @cindex creating links (hard or soft)
9202 @cindex file systems and hard links
9203 @command{ln} makes links between files. By default, it makes hard links;
9204 with the @option{-s} option, it makes symbolic (or @dfn{soft}) links.
9208 ln [@var{option}]@dots{} [-T] @var{target} @var{linkname}
9209 ln [@var{option}]@dots{} @var{target}
9210 ln [@var{option}]@dots{} @var{target}@dots{} @var{directory}
9211 ln [@var{option}]@dots{} -t @var{directory} @var{target}@dots{}
9217 If two file names are given, @command{ln} creates a link to the first
9218 file from the second.
9221 If one @var{target} is given, @command{ln} creates a link to that file
9222 in the current directory.
9225 If the @option{--target-directory} (@option{-t}) option is given, or
9226 failing that if the last file is a directory and the
9227 @option{--no-target-directory} (@option{-T}) option is not given,
9228 @command{ln} creates a link to each @var{target} file in the specified
9229 directory, using the @var{target}s' names.
9233 Normally @command{ln} does not remove existing files. Use the
9234 @option{--force} (@option{-f}) option to remove them unconditionally,
9235 the @option{--interactive} (@option{-i}) option to remove them
9236 conditionally, and the @option{--backup} (@option{-b}) option to
9239 @cindex hard link, defined
9240 @cindex inode, and hard links
9241 A @dfn{hard link} is another name for an existing file; the link and the
9242 original are indistinguishable. Technically speaking, they share the
9243 same inode, and the inode contains all the information about a
9244 file---indeed, it is not incorrect to say that the inode @emph{is} the
9245 file. Most systems prohibit making a hard link to
9246 a directory; on those where it is allowed, only the super-user can do
9247 so (and with caution, since creating a cycle will cause problems to many
9248 other utilities). Hard links cannot cross file system boundaries. (These
9249 restrictions are not mandated by @acronym{POSIX}, however.)
9251 @cindex dereferencing symbolic links
9252 @cindex symbolic link, defined
9253 @dfn{Symbolic links} (@dfn{symlinks} for short), on the other hand, are
9254 a special file type (which not all kernels support: System V release 3
9255 (and older) systems lack symlinks) in which the link file actually
9256 refers to a different file, by name. When most operations (opening,
9257 reading, writing, and so on) are passed the symbolic link file, the
9258 kernel automatically @dfn{dereferences} the link and operates on the
9259 target of the link. But some operations (e.g., removing) work on the
9260 link file itself, rather than on its target. The owner and group of a
9261 symlink are not significant to file access performed through
9262 the link, but do have implications on deleting a symbolic link from a
9263 directory with the restricted deletion bit set. On the GNU system,
9264 the mode of a symlink has no significance and cannot be changed, but
9265 on some BSD systems, the mode can be changed and will affect whether
9266 the symlink will be traversed in file name resolution. @xref{Symbolic Links,,,
9267 libc, The GNU C Library Reference Manual}.
9269 Symbolic links can contain arbitrary strings; a @dfn{dangling symlink}
9270 occurs when the string in the symlink does not resolve to a file.
9271 There are no restrictions against creating dangling symbolic links.
9272 There are trade-offs to using absolute or relative symlinks. An
9273 absolute symlink always points to the same file, even if the directory
9274 containing the link is moved. However, if the symlink is visible from
9275 more than one machine (such as on a networked file system), the file
9276 pointed to might not always be the same. A relative symbolic link is
9277 resolved in relation to the directory that contains the link, and is
9278 often useful in referring to files on the same device without regards
9279 to what name that device is mounted on when accessed via networked
9282 When creating a relative symlink in a different location than the
9283 current directory, the resolution of the symlink will be different
9284 than the resolution of the same string from the current directory.
9285 Therefore, many users prefer to first change directories to the
9286 location where the relative symlink will be created, so that
9287 tab-completion or other file resolution will find the same target as
9288 what will be placed in the symlink.
9290 The program accepts the following options. Also see @ref{Common options}.
9301 @opindex --directory
9302 @cindex hard links to directories
9303 Allow users with appropriate privileges to attempt to make hard links
9305 However, note that this will probably fail due to
9306 system restrictions, even for the super-user.
9312 Remove existing destination files.
9315 @itemx --interactive
9317 @opindex --interactive
9318 @cindex prompting, and @command{ln}
9319 Prompt whether to remove existing destination files.
9325 If @option{-s} is not in effect, and the source file is a symbolic
9326 link, create the hard link to the file referred to by the symbolic
9327 link, rather than the symbolic link itself.
9330 @itemx --no-dereference
9332 @opindex --no-dereference
9333 Do not treat the last operand specially when it is a symbolic link to
9334 a directory. Instead, treat it as if it were a normal file.
9336 When the destination is an actual directory (not a symlink to one),
9337 there is no ambiguity. The link is created in that directory.
9338 But when the specified destination is a symlink to a directory,
9339 there are two ways to treat the user's request. @command{ln} can
9340 treat the destination just as it would a normal directory and create
9341 the link in it. On the other hand, the destination can be viewed as a
9342 non-directory---as the symlink itself. In that case, @command{ln}
9343 must delete or backup that symlink before creating the new link.
9344 The default is to treat a destination that is a symlink to a directory
9345 just like a directory.
9347 This option is weaker than the @option{--no-target-directory}
9348 (@option{-T}) option, so it has no effect if both options are given.
9354 If @option{-s} is not in effect, and the source file is a symbolic
9355 link, create the hard link to the symbolic link itself. On platforms
9356 where this is not supported by the kernel, this option creates a
9357 symbolic link with identical contents; since symbolic link contents
9358 cannot be edited, any file name resolution performed through either
9359 link will be the same as if a hard link had been created.
9365 Make symbolic links instead of hard links. This option merely produces
9366 an error message on systems that do not support symbolic links.
9372 @optNoTargetDirectory
9378 Print the name of each file after linking it successfully.
9382 @cindex hard links to symbolic links
9383 @cindex symbolic links and @command{ln}
9384 If @option{-L} and @option{-P} are both given, the last one takes
9385 precedence. If @option{-s} is also given, @option{-L} and @option{-P}
9386 are silently ignored. If neither option is given, then this
9387 implementation defaults to @option{-P} if the system @code{link} supports
9388 hard links to symbolic links (such as the GNU system), and @option{-L}
9389 if @code{link} follows symbolic links (such as on BSD).
9398 # Create link ../a pointing to a in that directory.
9399 # Not really useful because it points to itself.
9404 # Change to the target before creating symlinks to avoid being confused.
9410 # Hard coded file names don't move well.
9411 ln -s $(pwd)/a /some/dir/
9415 # Relative file names survive directory moves and also
9416 # work across networked file systems.
9417 ln -s afile anotherfile
9418 ln -s ../adir/afile yetanotherfile
9422 @node mkdir invocation
9423 @section @command{mkdir}: Make directories
9426 @cindex directories, creating
9427 @cindex creating directories
9429 @command{mkdir} creates directories with the specified names. Synopsis:
9432 mkdir [@var{option}]@dots{} @var{name}@dots{}
9435 @command{mkdir} creates each directory @var{name} in the order given.
9436 It reports an error if @var{name} already exists, unless the
9437 @option{-p} option is given and @var{name} is a directory.
9439 The program accepts the following options. Also see @ref{Common options}.
9444 @itemx --mode=@var{mode}
9447 @cindex modes of created directories, setting
9448 Set the file permission bits of created directories to @var{mode},
9449 which uses the same syntax as
9450 in @command{chmod} and uses @samp{a=rwx} (read, write and execute allowed for
9451 everyone) for the point of the departure. @xref{File permissions}.
9453 Normally the directory has the desired file mode bits at the moment it
9454 is created. As a @acronym{GNU} extension, @var{mode} may also mention
9455 special mode bits, but in this case there may be a temporary window
9456 during which the directory exists but its special mode bits are
9457 incorrect. @xref{Directory Setuid and Setgid}, for how the
9458 set-user-ID and set-group-ID bits of directories are inherited unless
9459 overridden in this way.
9465 @cindex parent directories, creating
9466 Make any missing parent directories for each argument, setting their
9467 file permission bits to the umask modified by @samp{u+wx}. Ignore
9468 existing parent directories, and do not change their file permission
9471 To set the file permission bits of any newly-created parent
9472 directories to a value that includes @samp{u+wx}, you can set the
9473 umask before invoking @command{mkdir}. For example, if the shell
9474 command @samp{(umask u=rwx,go=rx; mkdir -p P/Q)} creates the parent
9475 @file{P} it sets the parent's permission bits to @samp{u=rwx,go=rx}.
9476 To set a parent's special mode bits as well, you can invoke
9477 @command{chmod} after @command{mkdir}. @xref{Directory Setuid and
9478 Setgid}, for how the set-user-ID and set-group-ID bits of
9479 newly-created parent directories are inherited.
9485 Print a message for each created directory. This is most useful with
9488 @item -Z @var{context}
9489 @itemx --context=@var{context}
9493 @cindex security context
9494 Set the default SELinux security context to be used for created directories.
9501 @node mkfifo invocation
9502 @section @command{mkfifo}: Make FIFOs (named pipes)
9505 @cindex FIFOs, creating
9506 @cindex named pipes, creating
9507 @cindex creating FIFOs (named pipes)
9509 @command{mkfifo} creates FIFOs (also called @dfn{named pipes}) with the
9510 specified names. Synopsis:
9513 mkfifo [@var{option}] @var{name}@dots{}
9516 A @dfn{FIFO} is a special file type that permits independent processes
9517 to communicate. One process opens the FIFO file for writing, and
9518 another for reading, after which data can flow as with the usual
9519 anonymous pipe in shells or elsewhere.
9521 The program accepts the following option. Also see @ref{Common options}.
9526 @itemx --mode=@var{mode}
9529 @cindex modes of created FIFOs, setting
9530 Set the mode of created FIFOs to @var{mode}, which is symbolic as in
9531 @command{chmod} and uses @samp{a=rw} (read and write allowed for everyone)
9532 for the point of departure. @var{mode} should specify only file
9533 permission bits. @xref{File permissions}.
9535 @item -Z @var{context}
9536 @itemx --context=@var{context}
9540 @cindex security context
9541 Set the default SELinux security context to be used for created FIFOs.
9548 @node mknod invocation
9549 @section @command{mknod}: Make block or character special files
9552 @cindex block special files, creating
9553 @cindex character special files, creating
9555 @command{mknod} creates a FIFO, character special file, or block special
9556 file with the specified name. Synopsis:
9559 mknod [@var{option}]@dots{} @var{name} @var{type} [@var{major} @var{minor}]
9562 @cindex special files
9563 @cindex block special files
9564 @cindex character special files
9565 Unlike the phrase ``special file type'' above, the term @dfn{special
9566 file} has a technical meaning on Unix: something that can generate or
9567 receive data. Usually this corresponds to a physical piece of hardware,
9568 e.g., a printer or a disk. (These files are typically created at
9569 system-configuration time.) The @command{mknod} command is what creates
9570 files of this type. Such devices can be read either a character at a
9571 time or a ``block'' (many characters) at a time, hence we say there are
9572 @dfn{block special} files and @dfn{character special} files.
9574 @c mknod is a shell built-in at least with OpenBSD's /bin/sh
9575 @mayConflictWithShellBuiltIn{mknod}
9577 The arguments after @var{name} specify the type of file to make:
9582 @opindex p @r{for FIFO file}
9586 @opindex b @r{for block special file}
9587 for a block special file
9590 @c Don't document the 'u' option -- it's just a synonym for 'c'.
9591 @c Do *any* versions of mknod still use it?
9593 @opindex c @r{for character special file}
9594 @c @opindex u @r{for character special file}
9595 for a character special file
9599 When making a block or character special file, the major and minor
9600 device numbers must be given after the file type.
9601 If a major or minor device number begins with @samp{0x} or @samp{0X},
9602 it is interpreted as hexadecimal; otherwise, if it begins with @samp{0},
9603 as octal; otherwise, as decimal.
9605 The program accepts the following option. Also see @ref{Common options}.
9610 @itemx --mode=@var{mode}
9613 Set the mode of created files to @var{mode}, which is symbolic as in
9614 @command{chmod} and uses @samp{a=rw} as the point of departure.
9615 @var{mode} should specify only file permission bits.
9616 @xref{File permissions}.
9618 @item -Z @var{context}
9619 @itemx --context=@var{context}
9623 @cindex security context
9624 Set the default SELinux security context to be used for created files.
9631 @node readlink invocation
9632 @section @command{readlink}: Print value of a symlink or canonical file name
9635 @cindex displaying value of a symbolic link
9636 @cindex canonical file name
9637 @cindex canonicalize a file name
9640 @command{readlink} may work in one of two supported modes:
9646 @command{readlink} outputs the value of the given symbolic link.
9647 If @command{readlink} is invoked with an argument other than the name
9648 of a symbolic link, it produces no output and exits with a nonzero exit code.
9650 @item Canonicalize mode
9652 @command{readlink} outputs the absolute name of the given file which contains
9653 no @file{.}, @file{..} components nor any repeated separators
9654 (@file{/}) or symbolic links.
9659 readlink [@var{option}] @var{file}
9662 By default, @command{readlink} operates in readlink mode.
9664 The program accepts the following options. Also see @ref{Common options}.
9669 @itemx --canonicalize
9671 @opindex --canonicalize
9672 Activate canonicalize mode.
9673 If any component of the file name except the last one is missing or unavailable,
9674 @command{readlink} produces no output and exits with a nonzero exit
9675 code. A trailing slash is ignored.
9678 @itemx --canonicalize-existing
9680 @opindex --canonicalize-existing
9681 Activate canonicalize mode.
9682 If any component is missing or unavailable, @command{readlink} produces
9683 no output and exits with a nonzero exit code. A trailing slash
9684 requires that the name resolve to a directory.
9687 @itemx --canonicalize-missing
9689 @opindex --canonicalize-missing
9690 Activate canonicalize mode.
9691 If any component is missing or unavailable, @command{readlink} treats it
9697 @opindex --no-newline
9698 Do not output the trailing newline.
9708 Suppress most error messages.
9714 Report error messages.
9718 The @command{readlink} utility first appeared in OpenBSD 2.1.
9720 The @command{realpath} command without options, operates like
9721 @command{readlink} in canonicalize mode.
9726 @node rmdir invocation
9727 @section @command{rmdir}: Remove empty directories
9730 @cindex removing empty directories
9731 @cindex directories, removing empty
9733 @command{rmdir} removes empty directories. Synopsis:
9736 rmdir [@var{option}]@dots{} @var{directory}@dots{}
9739 If any @var{directory} argument does not refer to an existing empty
9740 directory, it is an error.
9742 The program accepts the following options. Also see @ref{Common options}.
9746 @item --ignore-fail-on-non-empty
9747 @opindex --ignore-fail-on-non-empty
9748 @cindex directory deletion, ignoring failures
9749 Ignore each failure to remove a directory that is solely because
9750 the directory is non-empty.
9756 @cindex parent directories, removing
9757 Remove @var{directory}, then try to remove each component of @var{directory}.
9758 So, for example, @samp{rmdir -p a/b/c} is similar to @samp{rmdir a/b/c a/b a}.
9759 As such, it fails if any of those directories turns out not to be empty.
9760 Use the @option{--ignore-fail-on-non-empty} option to make it so such
9761 a failure does not evoke a diagnostic and does not cause @command{rmdir} to
9762 exit unsuccessfully.
9768 @cindex directory deletion, reporting
9769 Give a diagnostic for each successful removal.
9770 @var{directory} is removed.
9774 @xref{rm invocation}, for how to remove non-empty directories (recursively).
9779 @node unlink invocation
9780 @section @command{unlink}: Remove files via the unlink syscall
9783 @cindex removing files or directories (via the unlink syscall)
9785 @command{unlink} deletes a single specified file name.
9786 It is a minimalist interface to the system-provided
9787 @code{unlink} function. @xref{Deleting Files, , , libc,
9788 The GNU C Library Reference Manual}. Synopsis:
9789 It avoids the bells and whistles of the more commonly-used
9790 @command{rm} command (@pxref{rm invocation}).
9793 unlink @var{filename}
9796 On some systems @code{unlink} can be used to delete the name of a
9797 directory. On others, it can be used that way only by a privileged user.
9798 In the GNU system @code{unlink} can never delete the name of a directory.
9800 The @command{unlink} command honors the @option{--help} and
9801 @option{--version} options. To remove a file whose name begins with
9802 @samp{-}, prefix the name with @samp{./}, e.g., @samp{unlink ./--help}.
9807 @node Changing file attributes
9808 @chapter Changing file attributes
9810 @cindex changing file attributes
9811 @cindex file attributes, changing
9812 @cindex attributes, file
9814 A file is not merely its contents, a name, and a file type
9815 (@pxref{Special file types}). A file also has an owner (a user ID), a
9816 group (a group ID), permissions (what the owner can do with the file,
9817 what people in the group can do, and what everyone else can do), various
9818 timestamps, and other information. Collectively, we call these a file's
9821 These commands change file attributes.
9824 * chgrp invocation:: Change file groups.
9825 * chmod invocation:: Change access permissions.
9826 * chown invocation:: Change file owners and groups.
9827 * touch invocation:: Change file timestamps.
9831 @node chown invocation
9832 @section @command{chown}: Change file owner and group
9835 @cindex file ownership, changing
9836 @cindex group ownership, changing
9837 @cindex changing file ownership
9838 @cindex changing group ownership
9840 @command{chown} changes the user and/or group ownership of each given @var{file}
9841 to @var{new-owner} or to the user and group of an existing reference file.
9845 chown [@var{option}]@dots{} @{@var{new-owner} | --reference=@var{ref_file}@}@c
9849 If used, @var{new-owner} specifies the new owner and/or group as follows
9850 (with no embedded white space):
9853 [@var{owner}] [ : [@var{group}] ]
9860 If only an @var{owner} (a user name or numeric user ID) is given, that
9861 user is made the owner of each given file, and the files' group is not
9864 @item owner@samp{:}group
9865 If the @var{owner} is followed by a colon and a @var{group} (a
9866 group name or numeric group ID), with no spaces between them, the group
9867 ownership of the files is changed as well (to @var{group}).
9870 If a colon but no group name follows @var{owner}, that user is
9871 made the owner of the files and the group of the files is changed to
9872 @var{owner}'s login group.
9875 If the colon and following @var{group} are given, but the owner
9876 is omitted, only the group of the files is changed; in this case,
9877 @command{chown} performs the same function as @command{chgrp}.
9880 If only a colon is given, or if @var{new-owner} is empty, neither the
9881 owner nor the group is changed.
9885 If @var{owner} or @var{group} is intended to represent a numeric user
9886 or group ID, then you may specify it with a leading @samp{+}.
9887 @xref{Disambiguating names and IDs}.
9889 Some older scripts may still use @samp{.} in place of the @samp{:} separator.
9890 @acronym{POSIX} 1003.1-2001 (@pxref{Standards conformance}) does not
9891 require support for that, but for backward compatibility @acronym{GNU}
9892 @command{chown} supports @samp{.} so long as no ambiguity results.
9893 New scripts should avoid the use of @samp{.} because it is not
9894 portable, and because it has undesirable results if the entire
9895 @var{owner@samp{.}group} happens to identify a user whose name
9898 The @command{chown} command sometimes clears the set-user-ID or
9899 set-group-ID permission bits. This behavior depends on the policy and
9900 functionality of the underlying @code{chown} system call, which may
9901 make system-dependent file mode modifications outside the control of
9902 the @command{chown} command. For example, the @command{chown} command
9903 might not affect those bits when invoked by a user with appropriate
9904 privileges, or when the
9905 bits signify some function other than executable permission (e.g.,
9907 When in doubt, check the underlying system behavior.
9909 The program accepts the following options. Also see @ref{Common options}.
9917 @cindex changed owners, verbosely describing
9918 Verbosely describe the action for each @var{file} whose ownership
9927 @cindex error messages, omitting
9928 Do not print error messages about files whose ownership cannot be
9931 @itemx @w{@kbd{--from}=@var{old-owner}}
9933 @cindex symbolic links, changing owner
9934 Change a @var{file}'s ownership only if it has current attributes specified
9935 by @var{old-owner}. @var{old-owner} has the same form as @var{new-owner}
9937 This option is useful primarily from a security standpoint in that
9938 it narrows considerably the window of potential abuse.
9939 For example, to reflect a user ID numbering change for one user's files
9940 without an option like this, @code{root} might run
9943 find / -owner OLDUSER -print0 | xargs -0 chown -h NEWUSER
9946 But that is dangerous because the interval between when the @command{find}
9947 tests the existing file's owner and when the @command{chown} is actually run
9949 One way to narrow the gap would be to invoke chown for each file
9953 find / -owner OLDUSER -exec chown -h NEWUSER @{@} \;
9956 But that is very slow if there are many affected files.
9957 With this option, it is safer (the gap is narrower still)
9958 though still not perfect:
9961 chown -h -R --from=OLDUSER NEWUSER /
9965 @opindex --dereference
9966 @cindex symbolic links, changing owner
9968 Do not act on symbolic links themselves but rather on what they point to.
9969 This is the default.
9972 @itemx --no-dereference
9974 @opindex --no-dereference
9975 @cindex symbolic links, changing owner
9977 Act on symbolic links themselves instead of what they point to.
9978 This mode relies on the @code{lchown} system call.
9979 On systems that do not provide the @code{lchown} system call,
9980 @command{chown} fails when a file specified on the command line
9982 By default, no diagnostic is issued for symbolic links encountered
9983 during a recursive traversal, but see @option{--verbose}.
9985 @itemx --preserve-root
9986 @opindex --preserve-root
9987 @cindex root directory, disallow recursive modification
9988 Fail upon any attempt to recursively change the root directory, @file{/}.
9989 Without @option{--recursive}, this option has no effect.
9990 @xref{Treating / specially}.
9992 @itemx --no-preserve-root
9993 @opindex --no-preserve-root
9994 @cindex root directory, allow recursive modification
9995 Cancel the effect of any preceding @option{--preserve-root} option.
9996 @xref{Treating / specially}.
9998 @item --reference=@var{ref_file}
9999 @opindex --reference
10000 Change the user and group of each @var{file} to be the same as those of
10001 @var{ref_file}. If @var{ref_file} is a symbolic link, do not use the
10002 user and group of the symbolic link, but rather those of the file it
10009 Output a diagnostic for every file processed.
10010 If a symbolic link is encountered during a recursive traversal
10011 on a system without the @code{lchown} system call, and @option{--no-dereference}
10012 is in effect, then issue a diagnostic saying neither the symbolic link nor
10013 its referent is being changed.
10018 @opindex --recursive
10019 @cindex recursively changing file ownership
10020 Recursively change ownership of directories and their contents.
10023 @xref{Traversing symlinks}.
10026 @xref{Traversing symlinks}.
10029 @xref{Traversing symlinks}.
10038 # Change the owner of /u to "root".
10041 # Likewise, but also change its group to "staff".
10042 chown root:staff /u
10044 # Change the owner of /u and subfiles to "root".
10049 @node chgrp invocation
10050 @section @command{chgrp}: Change group ownership
10053 @cindex group ownership, changing
10054 @cindex changing group ownership
10056 @command{chgrp} changes the group ownership of each given @var{file}
10057 to @var{group} (which can be either a group name or a numeric group ID)
10058 or to the group of an existing reference file. Synopsis:
10061 chgrp [@var{option}]@dots{} @{@var{group} | --reference=@var{ref_file}@}@c
10065 If @var{group} is intended to represent a
10066 numeric group ID, then you may specify it with a leading @samp{+}.
10067 @xref{Disambiguating names and IDs}.
10069 The program accepts the following options. Also see @ref{Common options}.
10077 @cindex changed files, verbosely describing
10078 Verbosely describe the action for each @var{file} whose group actually
10087 @cindex error messages, omitting
10088 Do not print error messages about files whose group cannot be
10091 @item --dereference
10092 @opindex --dereference
10093 @cindex symbolic links, changing owner
10095 Do not act on symbolic links themselves but rather on what they point to.
10096 This is the default.
10099 @itemx --no-dereference
10101 @opindex --no-dereference
10102 @cindex symbolic links, changing group
10104 Act on symbolic links themselves instead of what they point to.
10105 This mode relies on the @code{lchown} system call.
10106 On systems that do not provide the @code{lchown} system call,
10107 @command{chgrp} fails when a file specified on the command line
10108 is a symbolic link.
10109 By default, no diagnostic is issued for symbolic links encountered
10110 during a recursive traversal, but see @option{--verbose}.
10112 @itemx --preserve-root
10113 @opindex --preserve-root
10114 @cindex root directory, disallow recursive modification
10115 Fail upon any attempt to recursively change the root directory, @file{/}.
10116 Without @option{--recursive}, this option has no effect.
10117 @xref{Treating / specially}.
10119 @itemx --no-preserve-root
10120 @opindex --no-preserve-root
10121 @cindex root directory, allow recursive modification
10122 Cancel the effect of any preceding @option{--preserve-root} option.
10123 @xref{Treating / specially}.
10125 @item --reference=@var{ref_file}
10126 @opindex --reference
10127 Change the group of each @var{file} to be the same as that of
10128 @var{ref_file}. If @var{ref_file} is a symbolic link, do not use the
10129 group of the symbolic link, but rather that of the file it refers to.
10135 Output a diagnostic for every file processed.
10136 If a symbolic link is encountered during a recursive traversal
10137 on a system without the @code{lchown} system call, and @option{--no-dereference}
10138 is in effect, then issue a diagnostic saying neither the symbolic link nor
10139 its referent is being changed.
10144 @opindex --recursive
10145 @cindex recursively changing group ownership
10146 Recursively change the group ownership of directories and their contents.
10149 @xref{Traversing symlinks}.
10152 @xref{Traversing symlinks}.
10155 @xref{Traversing symlinks}.
10164 # Change the group of /u to "staff".
10167 # Change the group of /u and subfiles to "staff".
10172 @node chmod invocation
10173 @section @command{chmod}: Change access permissions
10176 @cindex changing access permissions
10177 @cindex access permissions, changing
10178 @cindex permissions, changing access
10180 @command{chmod} changes the access permissions of the named files. Synopsis:
10183 chmod [@var{option}]@dots{} @{@var{mode} | --reference=@var{ref_file}@}@c
10187 @cindex symbolic links, permissions of
10188 @command{chmod} never changes the permissions of symbolic links, since
10189 the @command{chmod} system call cannot change their permissions.
10190 This is not a problem since the permissions of symbolic links are
10191 never used. However, for each symbolic link listed on the command
10192 line, @command{chmod} changes the permissions of the pointed-to file.
10193 In contrast, @command{chmod} ignores symbolic links encountered during
10194 recursive directory traversals.
10196 A successful use of @command{chmod} clears the set-group-ID bit of a
10197 regular file if the file's group ID does not match the user's
10198 effective group ID or one of the user's supplementary group IDs,
10199 unless the user has appropriate privileges. Additional restrictions
10200 may cause the set-user-ID and set-group-ID bits of @var{mode} or
10201 @var{ref_file} to be ignored. This behavior depends on the policy and
10202 functionality of the underlying @code{chmod} system call. When in
10203 doubt, check the underlying system behavior.
10205 If used, @var{mode} specifies the new file mode bits.
10206 For details, see the section on @ref{File permissions}.
10207 If you really want @var{mode} to have a leading @samp{-}, you should
10208 use @option{--} first, e.g., @samp{chmod -- -w file}. Typically,
10209 though, @samp{chmod a-w file} is preferable, and @command{chmod -w
10210 file} (without the @option{--}) complains if it behaves differently
10211 from what @samp{chmod a-w file} would do.
10213 The program accepts the following options. Also see @ref{Common options}.
10221 Verbosely describe the action for each @var{file} whose permissions
10230 @cindex error messages, omitting
10231 Do not print error messages about files whose permissions cannot be
10234 @itemx --preserve-root
10235 @opindex --preserve-root
10236 @cindex root directory, disallow recursive modification
10237 Fail upon any attempt to recursively change the root directory, @file{/}.
10238 Without @option{--recursive}, this option has no effect.
10239 @xref{Treating / specially}.
10241 @itemx --no-preserve-root
10242 @opindex --no-preserve-root
10243 @cindex root directory, allow recursive modification
10244 Cancel the effect of any preceding @option{--preserve-root} option.
10245 @xref{Treating / specially}.
10251 Verbosely describe the action or non-action taken for every @var{file}.
10253 @item --reference=@var{ref_file}
10254 @opindex --reference
10255 Change the mode of each @var{file} to be the same as that of @var{ref_file}.
10256 @xref{File permissions}.
10257 If @var{ref_file} is a symbolic link, do not use the mode
10258 of the symbolic link, but rather that of the file it refers to.
10263 @opindex --recursive
10264 @cindex recursively changing access permissions
10265 Recursively change permissions of directories and their contents.
10272 @node touch invocation
10273 @section @command{touch}: Change file timestamps
10276 @cindex changing file timestamps
10277 @cindex file timestamps, changing
10278 @cindex timestamps, changing file
10280 @command{touch} changes the access and/or modification times of the
10281 specified files. Synopsis:
10284 touch [@var{option}]@dots{} @var{file}@dots{}
10287 @cindex empty files, creating
10288 Any @var{file} argument that does not exist is created empty, unless
10289 option @option{--no-create} (@option{-c}) or @option{--no-dereference}
10290 (@option{-h}) was in effect.
10292 A @var{file} argument string of @samp{-} is handled specially and
10293 causes @command{touch} to change the times of the file associated with
10297 By default, @command{touch} sets file timestamps to the current time.
10298 Because @command{touch} acts on its operands left to right, the
10299 resulting timestamps of earlier and later operands may disagree.
10300 Also, the determination of what time is ``current'' depends on the
10301 platform. Platforms with network file systems often use different
10302 clocks for the operating system and for file systems; because
10303 @command{touch} typically uses file systems' clocks by default, clock
10304 skew can cause the resulting file timestamps to appear to be in a
10305 program's ``future'' or ``past''.
10307 @cindex file timestamp resolution
10308 The @command{touch} command sets the file's timestamp to the greatest
10309 representable value that is not greater than the requested time. This
10310 can differ from the requested time for several reasons. First, the
10311 requested time may have a higher resolution than supported. Second, a
10312 file system may use different resolutions for different types of
10313 times. Third, file timestamps may use a different resolution than
10314 operating system timestamps. Fourth, the operating system primitives
10315 used to update timestamps may employ yet a different resolution. For
10316 example, in theory a file system might use 10-microsecond resolution
10317 for access time and 100-nanosecond resolution for modification time,
10318 and the operating system might use nanosecond resolution for the
10319 current time and microsecond resolution for the primitive that
10320 @command{touch} uses to set a file's timestamp to an arbitrary value.
10322 @cindex permissions, for changing file timestamps
10323 When setting file timestamps to the current time, @command{touch} can
10324 change the timestamps for files that the user does not own but has
10325 write permission for. Otherwise, the user must own the files. Some
10326 older systems have a further restriction: the user must own the files
10327 unless both the access and modification times are being set to the
10330 Although @command{touch} provides options for changing two of the times---the
10331 times of last access and modification---of a file, there is actually
10332 a standard third one as well: the inode change time. This is often
10333 referred to as a file's @code{ctime}.
10334 The inode change time represents the time when the file's meta-information
10335 last changed. One common example of this is when the permissions of a
10336 file change. Changing the permissions doesn't access the file, so
10337 the atime doesn't change, nor does it modify the file, so the mtime
10338 doesn't change. Yet, something about the file itself has changed,
10339 and this must be noted somewhere. This is the job of the ctime field.
10340 This is necessary, so that, for example, a backup program can make a
10341 fresh copy of the file, including the new permissions value.
10342 Another operation that modifies a file's ctime without affecting
10343 the others is renaming. In any case, it is not possible, in normal
10344 operations, for a user to change the ctime field to a user-specified value.
10345 Some operating systems and file systems support a fourth time: the
10346 birth time, when the file was first created; by definition, this
10347 timestamp never changes.
10350 Time stamps assume the time zone rules specified by the @env{TZ}
10351 environment variable, or by the system default rules if @env{TZ} is
10352 not set. @xref{TZ Variable,, Specifying the Time Zone with @env{TZ},
10353 libc, The GNU C Library Reference Manual}.
10354 You can avoid ambiguities during
10355 daylight saving transitions by using @sc{utc} time stamps.
10357 The program accepts the following options. Also see @ref{Common options}.
10362 @itemx --time=atime
10363 @itemx --time=access
10367 @opindex atime@r{, changing}
10368 @opindex access @r{time, changing}
10369 @opindex use @r{time, changing}
10370 Change the access time only.
10375 @opindex --no-create
10376 Do not warn about or create files that do not exist.
10379 @itemx --date=@var{time}
10383 Use @var{time} instead of the current time. It can contain month names,
10384 time zones, @samp{am} and @samp{pm}, @samp{yesterday}, etc. For
10385 example, @option{--date="2004-02-27 14:19:13.489392193 +0530"}
10386 specifies the instant of time that is 489,392,193 nanoseconds after
10387 February 27, 2004 at 2:19:13 PM in a time zone that is 5 hours and 30
10388 minutes east of @acronym{UTC}. @xref{Date input formats}.
10389 File systems that do not support high-resolution time stamps
10390 silently ignore any excess precision here.
10394 @cindex BSD @command{touch} compatibility
10395 Ignored; for compatibility with BSD versions of @command{touch}.
10398 @itemx --no-dereference
10400 @opindex --no-dereference
10401 @cindex symbolic links, changing time
10403 Attempt to change the timestamps of a symbolic link, rather than what
10404 the link refers to. When using this option, empty files are not
10405 created, but option @option{-c} must also be used to avoid warning
10406 about files that do not exist. Not all systems support changing the
10407 timestamps of symlinks, since underlying system support for this
10408 action was not required until @acronym{POSIX} 2008. Also, on some
10409 systems, the mere act of examining a symbolic link changes the access
10410 time, such that only changes to the modification time will persist
10411 long enough to be observable. When coupled with option @option{-r}, a
10412 reference timestamp is taken from a symbolic link rather than the file
10416 @itemx --time=mtime
10417 @itemx --time=modify
10420 @opindex mtime@r{, changing}
10421 @opindex modify @r{time, changing}
10422 Change the modification time only.
10424 @item -r @var{file}
10425 @itemx --reference=@var{file}
10427 @opindex --reference
10428 Use the times of the reference @var{file} instead of the current time.
10429 If this option is combined with the @option{--date=@var{time}}
10430 (@option{-d @var{time}}) option, the reference @var{file}'s time is
10431 the origin for any relative @var{time}s given, but is otherwise ignored.
10432 For example, @samp{-r foo -d '-5 seconds'} specifies a time stamp
10433 equal to five seconds before the corresponding time stamp for @file{foo}.
10434 If @var{file} is a symbolic link, the reference timestamp is taken
10435 from the target of the symlink, unless @option{-h} was also in effect.
10437 @item -t [[@var{cc}]@var{yy}]@var{mmddhhmm}[.@var{ss}]
10438 Use the argument (optional four-digit or two-digit years, months,
10439 days, hours, minutes, optional seconds) instead of the current time.
10440 If the year is specified with only two digits, then @var{cc}
10441 is 20 for years in the range 0 @dots{} 68, and 19 for years in
10442 69 @dots{} 99. If no digits of the year are specified,
10443 the argument is interpreted as a date in the current year.
10444 Note that @var{ss} may be @samp{60}, to accommodate leap seconds.
10448 @vindex _POSIX2_VERSION
10449 On older systems, @command{touch} supports an obsolete syntax, as follows.
10450 If no timestamp is given with any of the @option{-d}, @option{-r}, or
10451 @option{-t} options, and if there are two or more @var{file}s and the
10452 first @var{file} is of the form @samp{@var{mmddhhmm}[@var{yy}]} and this
10453 would be a valid argument to the @option{-t} option (if the @var{yy}, if
10454 any, were moved to the front), and if the represented year
10455 is in the range 1969--1999, that argument is interpreted as the time
10456 for the other files instead of as a file name.
10457 This obsolete behavior can be enabled or disabled with the
10458 @env{_POSIX2_VERSION} environment variable (@pxref{Standards
10459 conformance}), but portable scripts should avoid commands whose
10460 behavior depends on this variable.
10461 For example, use @samp{touch ./12312359 main.c} or @samp{touch -t
10462 12312359 main.c} rather than the ambiguous @samp{touch 12312359 main.c}.
10468 @chapter Disk usage
10472 No disk can hold an infinite amount of data. These commands report
10473 how much disk storage is in use or available, report other file and
10474 file status information, and write buffers to disk.
10477 * df invocation:: Report file system disk space usage.
10478 * du invocation:: Estimate file space usage.
10479 * stat invocation:: Report file or file system status.
10480 * sync invocation:: Synchronize memory and disk.
10481 * truncate invocation:: Shrink or extend the size of a file.
10485 @node df invocation
10486 @section @command{df}: Report file system disk space usage
10489 @cindex file system disk usage
10490 @cindex disk usage by file system
10492 @command{df} reports the amount of disk space used and available on
10493 file systems. Synopsis:
10496 df [@var{option}]@dots{} [@var{file}]@dots{}
10499 With no arguments, @command{df} reports the space used and available on all
10500 currently mounted file systems (of all types). Otherwise, @command{df}
10501 reports on the file system containing each argument @var{file}.
10503 Normally the disk space is printed in units of
10504 1024 bytes, but this can be overridden (@pxref{Block size}).
10505 Non-integer quantities are rounded up to the next higher unit.
10507 @cindex disk device file
10508 @cindex device file, disk
10509 If an argument @var{file} is a disk device file containing a mounted
10510 file system, @command{df} shows the space available on that file system
10511 rather than on the file system containing the device node (i.e., the root
10512 file system). @sc{gnu} @command{df} does not attempt to determine the
10514 on unmounted file systems, because on most kinds of systems doing so
10515 requires extremely nonportable intimate knowledge of file system
10518 The program accepts the following options. Also see @ref{Common options}.
10526 @cindex automounter file systems
10527 @cindex ignore file systems
10528 Include in the listing dummy file systems, which
10529 are omitted by default. Such file systems are typically special-purpose
10530 pseudo-file-systems, such as automounter entries.
10532 @item -B @var{size}
10533 @itemx --block-size=@var{size}
10535 @opindex --block-size
10536 @cindex file system sizes
10537 Scale sizes by @var{size} before printing them (@pxref{Block size}).
10538 For example, @option{-BG} prints sizes in units of 1,073,741,824 bytes.
10542 @cindex grand total of disk size, usage and available space
10543 Print a grand total of all arguments after all arguments have
10544 been processed. This can be used to find out the total disk size, usage
10545 and available space of all listed devices.
10551 Equivalent to @option{--si}.
10557 @cindex inode usage
10558 List inode usage information instead of block usage. An inode (short
10559 for index node) contains information about a file such as its owner,
10560 permissions, timestamps, and location on the disk.
10564 @cindex kibibytes for file system sizes
10565 Print sizes in 1024-byte blocks, overriding the default block size
10566 (@pxref{Block size}).
10567 This option is equivalent to @option{--block-size=1K}.
10573 @cindex file system types, limiting output to certain
10574 Limit the listing to local file systems. By default, remote file systems
10579 @cindex file system space, retrieving old data more quickly
10580 Do not invoke the @code{sync} system call before getting any usage data.
10581 This may make @command{df} run significantly faster on systems with many
10582 disks, but on some systems (notably SunOS) the results may be slightly
10583 out of date. This is the default.
10586 @itemx --portability
10588 @opindex --portability
10589 @cindex one-line output format
10590 @cindex @acronym{POSIX} output format
10591 @cindex portable output format
10592 @cindex output format, portable
10593 Use the @acronym{POSIX} output format. This is like the default format except
10598 The information about each file system is always printed on exactly
10599 one line; a mount device is never put on a line by itself. This means
10600 that if the mount device name is more than 20 characters long (e.g., for
10601 some network mounts), the columns are misaligned.
10604 The labels in the header output line are changed to conform to @acronym{POSIX}.
10607 The default block size and output format are unaffected by the
10608 @env{DF_BLOCK_SIZE}, @env{BLOCK_SIZE} and @env{BLOCKSIZE} environment
10609 variables. However, the default block size is still affected by
10610 @env{POSIXLY_CORRECT}: it is 512 if @env{POSIXLY_CORRECT} is set, 1024
10611 otherwise. @xref{Block size}.
10618 @cindex file system space, retrieving current data more slowly
10619 Invoke the @code{sync} system call before getting any usage data. On
10620 some systems (notably SunOS), doing this yields more up to date results,
10621 but in general this option makes @command{df} much slower, especially when
10622 there are many or very busy file systems.
10624 @item -t @var{fstype}
10625 @itemx --type=@var{fstype}
10628 @cindex file system types, limiting output to certain
10629 Limit the listing to file systems of type @var{fstype}. Multiple
10630 file system types can be specified by giving multiple @option{-t} options.
10631 By default, nothing is omitted.
10634 @itemx --print-type
10636 @opindex --print-type
10637 @cindex file system types, printing
10638 Print each file system's type. The types printed here are the same ones
10639 you can include or exclude with @option{-t} and @option{-x}. The particular
10640 types printed are whatever is supported by the system. Here are some of
10641 the common names (this list is certainly not exhaustive):
10646 @cindex @acronym{NFS} file system type
10647 An @acronym{NFS} file system, i.e., one mounted over a network from another
10648 machine. This is the one type name which seems to be used uniformly by
10651 @item 4.2@r{, }ufs@r{, }efs@dots{}
10652 @cindex Linux file system types
10653 @cindex local file system types
10654 @opindex 4.2 @r{file system type}
10655 @opindex ufs @r{file system type}
10656 @opindex efs @r{file system type}
10657 A file system on a locally-mounted hard disk. (The system might even
10658 support more than one type here; Linux does.)
10660 @item hsfs@r{, }cdfs
10661 @cindex CD-ROM file system type
10662 @cindex High Sierra file system
10663 @opindex hsfs @r{file system type}
10664 @opindex cdfs @r{file system type}
10665 A file system on a CD-ROM drive. HP-UX uses @samp{cdfs}, most other
10666 systems use @samp{hsfs} (@samp{hs} for ``High Sierra'').
10669 @cindex PC file system
10670 @cindex DOS file system
10671 @cindex MS-DOS file system
10672 @cindex diskette file system
10674 An MS-DOS file system, usually on a diskette.
10678 @item -x @var{fstype}
10679 @itemx --exclude-type=@var{fstype}
10681 @opindex --exclude-type
10682 Limit the listing to file systems not of type @var{fstype}.
10683 Multiple file system types can be eliminated by giving multiple
10684 @option{-x} options. By default, no file system types are omitted.
10687 Ignored; for compatibility with System V versions of @command{df}.
10692 Failure includes the case where no output is generated, so you can
10693 inspect the exit status of a command like @samp{df -t ext3 -t reiserfs
10694 @var{dir}} to test whether @var{dir} is on a file system of type
10695 @samp{ext3} or @samp{reiserfs}.
10698 @node du invocation
10699 @section @command{du}: Estimate file space usage
10702 @cindex file space usage
10703 @cindex disk usage for files
10705 @command{du} reports the amount of disk space used by the specified files
10706 and for each subdirectory (of directory arguments). Synopsis:
10709 du [@var{option}]@dots{} [@var{file}]@dots{}
10712 With no arguments, @command{du} reports the disk space for the current
10713 directory. Normally the disk space is printed in units of
10714 1024 bytes, but this can be overridden (@pxref{Block size}).
10715 Non-integer quantities are rounded up to the next higher unit.
10717 If two or more hard links point to the same file, only one of the hard
10718 links is counted. The @var{file} argument order affects which links
10719 are counted, and changing the argument order may change the numbers
10720 that @command{du} outputs.
10722 The program accepts the following options. Also see @ref{Common options}.
10730 Show counts for all files, not just directories.
10732 @itemx --apparent-size
10733 @opindex --apparent-size
10734 Print apparent sizes, rather than disk usage. The apparent size of a
10735 file is the number of bytes reported by @code{wc -c} on regular files,
10736 or more generally, @code{ls -l --block-size=1} or @code{stat --format=%s}.
10737 For example, a file containing the word @samp{zoo} with no newline would,
10738 of course, have an apparent size of 3. Such a small file may require
10739 anywhere from 0 to 16 KiB or more of disk space, depending on
10740 the type and configuration of the file system on which the file resides.
10741 However, a sparse file created with this command:
10744 dd bs=1 seek=2GiB if=/dev/null of=big
10748 has an apparent size of 2 GiB, yet on most modern
10749 systems, it actually uses almost no disk space.
10755 Equivalent to @code{--apparent-size --block-size=1}.
10757 @item -B @var{size}
10758 @itemx --block-size=@var{size}
10760 @opindex --block-size
10762 Scale sizes by @var{size} before printing them (@pxref{Block size}).
10763 For example, @option{-BG} prints sizes in units of 1,073,741,824 bytes.
10769 @cindex grand total of disk space
10770 Print a grand total of all arguments after all arguments have
10771 been processed. This can be used to find out the total disk usage of
10772 a given set of files or directories.
10775 @itemx --dereference-args
10777 @opindex --dereference-args
10778 Dereference symbolic links that are command line arguments.
10779 Does not affect other symbolic links. This is helpful for finding
10780 out the disk usage of directories, such as @file{/usr/tmp}, which
10781 are often symbolic links.
10783 @c --files0-from=FILE
10784 @filesZeroFromOption{du,, with the @option{--total} (@option{-c}) option}
10790 Equivalent to @option{--dereference-args} (@option{-D}).
10794 @cindex kibibytes for file sizes
10795 Print sizes in 1024-byte blocks, overriding the default block size
10796 (@pxref{Block size}).
10797 This option is equivalent to @option{--block-size=1K}.
10800 @itemx --count-links
10802 @opindex --count-links
10803 @cindex hard links, counting in @command{du}
10804 Count the size of all files, even if they have appeared already (as a
10808 @itemx --dereference
10810 @opindex --dereference
10811 @cindex symbolic links, dereferencing in @command{du}
10812 Dereference symbolic links (show the disk space used by the file
10813 or directory that the link points to instead of the space used by
10818 @cindex mebibytes for file sizes
10819 Print sizes in 1,048,576-byte blocks, overriding the default block size
10820 (@pxref{Block size}).
10821 This option is equivalent to @option{--block-size=1M}.
10824 @itemx --no-dereference
10826 @opindex --no-dereference
10827 @cindex symbolic links, dereferencing in @command{du}
10828 For each symbolic links encountered by @command{du},
10829 consider the disk space used by the symbolic link.
10831 @item -d @var{depth}
10832 @item --max-depth=@var{depth}
10833 @opindex -d @var{depth}
10834 @opindex --max-depth=@var{depth}
10835 @cindex limiting output of @command{du}
10836 Show the total for each directory (and file if --all) that is at
10837 most MAX_DEPTH levels down from the root of the hierarchy. The root
10838 is at level 0, so @code{du --max-depth=0} is equivalent to @code{du -s}.
10847 @opindex --summarize
10848 Display only a total for each argument.
10851 @itemx --separate-dirs
10853 @opindex --separate-dirs
10854 Normally, in the output of @command{du} (when not using @option{--summarize}),
10855 the size listed next to a directory name, @var{d}, represents the sum
10856 of sizes of all entries beneath @var{d} as well as the size of @var{d} itself.
10857 With @option{--separate-dirs}, the size reported for a directory name,
10858 @var{d}, is merely the @code{stat.st_size}-derived size of the directory
10863 @cindex last modified dates, displaying in @command{du}
10864 Show time of the most recent modification of any file in the directory,
10865 or any of its subdirectories.
10867 @itemx --time=ctime
10868 @itemx --time=status
10871 @opindex ctime@r{, show the most recent}
10872 @opindex status time@r{, show the most recent}
10873 @opindex use time@r{, show the most recent}
10874 Show the most recent status change time (the @samp{ctime} in the inode) of
10875 any file in the directory, instead of the modification time.
10877 @itemx --time=atime
10878 @itemx --time=access
10880 @opindex atime@r{, show the most recent}
10881 @opindex access time@r{, show the most recent}
10882 Show the most recent access time (the @samp{atime} in the inode) of
10883 any file in the directory, instead of the modification time.
10885 @item --time-style=@var{style}
10886 @opindex --time-style
10888 List timestamps in style @var{style}. This option has an effect only if
10889 the @option{--time} option is also specified. The @var{style} should
10890 be one of the following:
10893 @item +@var{format}
10895 List timestamps using @var{format}, where @var{format} is interpreted
10896 like the format argument of @command{date} (@pxref{date invocation}).
10897 For example, @option{--time-style="+%Y-%m-%d %H:%M:%S"} causes
10898 @command{du} to list timestamps like @samp{2002-03-30 23:45:56}. As
10899 with @command{date}, @var{format}'s interpretation is affected by the
10900 @env{LC_TIME} locale category.
10903 List timestamps in full using @acronym{ISO} 8601 date, time, and time zone
10904 format with nanosecond precision, e.g., @samp{2002-03-30
10905 23:45:56.477817180 -0700}. This style is equivalent to
10906 @samp{+%Y-%m-%d %H:%M:%S.%N %z}.
10909 List @acronym{ISO} 8601 date and time in minutes, e.g.,
10910 @samp{2002-03-30 23:45}. These timestamps are shorter than
10911 @samp{full-iso} timestamps, and are usually good enough for everyday
10912 work. This style is equivalent to @samp{+%Y-%m-%d %H:%M}.
10915 List @acronym{ISO} 8601 dates for timestamps, e.g., @samp{2002-03-30}.
10916 This style is equivalent to @samp{+%Y-%m-%d}.
10920 You can specify the default value of the @option{--time-style} option
10921 with the environment variable @env{TIME_STYLE}; if @env{TIME_STYLE} is not set
10922 the default style is @samp{long-iso}. For compatibility with @command{ls},
10923 if @env{TIME_STYLE} begins with @samp{+} and contains a newline,
10924 the newline and any later characters are ignored; if @env{TIME_STYLE}
10925 begins with @samp{posix-} the @samp{posix-} is ignored; and if
10926 @env{TIME_STYLE} is @samp{locale} it is ignored.
10929 @itemx --one-file-system
10931 @opindex --one-file-system
10932 @cindex one file system, restricting @command{du} to
10933 Skip directories that are on different file systems from the one that
10934 the argument being processed is on.
10936 @item --exclude=@var{pattern}
10937 @opindex --exclude=@var{pattern}
10938 @cindex excluding files from @command{du}
10939 When recursing, skip subdirectories or files matching @var{pattern}.
10940 For example, @code{du --exclude='*.o'} excludes files whose names
10943 @item -X @var{file}
10944 @itemx --exclude-from=@var{file}
10945 @opindex -X @var{file}
10946 @opindex --exclude-from=@var{file}
10947 @cindex excluding files from @command{du}
10948 Like @option{--exclude}, except take the patterns to exclude from @var{file},
10949 one per line. If @var{file} is @samp{-}, take the patterns from standard
10954 @cindex NFS mounts from BSD to HP-UX
10955 On BSD systems, @command{du} reports sizes that are half the correct
10956 values for files that are NFS-mounted from HP-UX systems. On HP-UX
10957 systems, it reports sizes that are twice the correct values for
10958 files that are NFS-mounted from BSD systems. This is due to a flaw
10959 in HP-UX; it also affects the HP-UX @command{du} program.
10964 @node stat invocation
10965 @section @command{stat}: Report file or file system status
10968 @cindex file status
10969 @cindex file system status
10971 @command{stat} displays information about the specified file(s). Synopsis:
10974 stat [@var{option}]@dots{} [@var{file}]@dots{}
10977 With no option, @command{stat} reports all information about the given files.
10978 But it also can be used to report the information of the file systems the
10979 given files are located on. If the files are links, @command{stat} can
10980 also give information about the files the links point to.
10982 @mayConflictWithShellBuiltIn{stat}
10987 @itemx --dereference
10989 @opindex --dereference
10990 @cindex symbolic links, dereferencing in @command{stat}
10991 Change how @command{stat} treats symbolic links.
10992 With this option, @command{stat} acts on the file referenced
10993 by each symbolic link argument.
10994 Without it, @command{stat} acts on any symbolic link argument directly.
10997 @itemx --file-system
10999 @opindex --file-system
11000 @cindex file systems
11001 Report information about the file systems where the given files are located
11002 instead of information about the files themselves.
11003 This option implies the @option{-L} option.
11006 @itemx --format=@var{format}
11008 @opindex --format=@var{format}
11009 @cindex output format
11010 Use @var{format} rather than the default format.
11011 @var{format} is automatically newline-terminated, so
11012 running a command like the following with two or more @var{file}
11013 operands produces a line of output for each operand:
11015 $ stat --format=%d:%i / /usr
11020 @itemx --printf=@var{format}
11021 @opindex --printf=@var{format}
11022 @cindex output format
11023 Use @var{format} rather than the default format.
11024 Like @option{--format}, but interpret backslash escapes,
11025 and do not output a mandatory trailing newline.
11026 If you want a newline, include @samp{\n} in the @var{format}.
11027 Here's how you would use @option{--printf} to print the device
11028 and inode numbers of @file{/} and @file{/usr}:
11030 $ stat --printf='%d:%i\n' / /usr
11039 @cindex terse output
11040 Print the information in terse form, suitable for parsing by other programs.
11044 The valid @var{format} directives for files with @option{--format} and
11045 @option{--printf} are:
11048 @item %a - Access rights in octal
11049 @item %A - Access rights in human readable form
11050 @item %b - Number of blocks allocated (see @samp{%B})
11051 @item %B - The size in bytes of each block reported by @samp{%b}
11052 @item %C - The SELinux security context of a file, if available
11053 @item %d - Device number in decimal
11054 @item %D - Device number in hex
11055 @item %f - Raw mode in hex
11056 @item %F - File type
11057 @item %g - Group ID of owner
11058 @item %G - Group name of owner
11059 @item %h - Number of hard links
11060 @item %i - Inode number
11061 @item %m - Mount point (See note below)
11062 @item %n - File name
11063 @item %N - Quoted file name with dereference if symbolic link
11064 @item %o - Optimal I/O transfer size hint
11065 @item %s - Total size, in bytes
11066 @item %t - Major device type in hex
11067 @item %T - Minor device type in hex
11068 @item %u - User ID of owner
11069 @item %U - User name of owner
11070 @item %w - Time of file birth, or @samp{-} if unknown
11071 @item %W - Time of file birth as seconds since Epoch, or @samp{0}
11072 @item %x - Time of last access
11073 @item %X - Time of last access as seconds since Epoch
11074 @item %y - Time of last modification
11075 @item %Y - Time of last modification as seconds since Epoch
11076 @item %z - Time of last change
11077 @item %Z - Time of last change as seconds since Epoch
11080 The @samp{%W}, @samp{%X}, @samp{%Y}, and @samp{%Z} formats accept a
11081 precision preceded by a period to specify the number of digits to
11082 print after the decimal point. For example, @samp{%.3X} outputs the
11083 last access time to millisecond precision. If a period is given but no
11084 precision, @command{stat} uses 9 digits, so @samp{%.X} is equivalent to
11085 @samp{%.9X}. When discarding excess precision, time stamps are truncated
11086 toward minus infinity.
11090 $ stat -c '[%015Y]' /usr
11093 $ stat -c '[%15Y]' /usr
11095 $ stat -c '[%-15Y]' /usr
11098 $ stat -c '[%.3Y]' /usr
11100 $ stat -c '[%.Y]' /usr
11101 [1288929712.114951834]
11104 The mount point printed by @samp{%m} is similar to that output
11105 by @command{df}, except that:
11108 stat does not dereference symlinks by default
11109 (unless @option{-L} is specified)
11111 stat does not search for specified device nodes in the
11112 file system list, instead operating on them directly
11115 stat outputs the alias for a bind mounted file, rather than
11116 the initial mount point of its backing device.
11117 One can recursively call stat until there is no change in output,
11118 to get the current base mount point
11121 When listing file system information (@option{--file-system} (@option{-f})),
11122 you must use a different set of @var{format} directives:
11125 @item %a - Free blocks available to non-super-user
11126 @item %b - Total data blocks in file system
11127 @item %c - Total file nodes in file system
11128 @item %d - Free file nodes in file system
11129 @item %f - Free blocks in file system
11130 @item %i - File System ID in hex
11131 @item %l - Maximum length of file names
11132 @item %n - File name
11133 @item %s - Block size (for faster transfers)
11134 @item %S - Fundamental block size (for block counts)
11135 @item %t - Type in hex
11136 @item %T - Type in human readable form
11140 Time stamps are listed according to the time zone rules specified by
11141 the @env{TZ} environment variable, or by the system default rules if
11142 @env{TZ} is not set. @xref{TZ Variable,, Specifying the Time Zone
11143 with @env{TZ}, libc, The GNU C Library Reference Manual}.
11148 @node sync invocation
11149 @section @command{sync}: Synchronize data on disk with memory
11152 @cindex synchronize disk and memory
11154 @cindex superblock, writing
11155 @cindex inodes, written buffered
11156 @command{sync} writes any data buffered in memory out to disk. This can
11157 include (but is not limited to) modified superblocks, modified inodes,
11158 and delayed reads and writes. This must be implemented by the kernel;
11159 The @command{sync} program does nothing but exercise the @code{sync} system
11162 @cindex crashes and corruption
11163 The kernel keeps data in memory to avoid doing (relatively slow) disk
11164 reads and writes. This improves performance, but if the computer
11165 crashes, data may be lost or the file system corrupted as a
11166 result. The @command{sync} command ensures everything in memory
11167 is written to disk.
11169 Any arguments are ignored, except for a lone @option{--help} or
11170 @option{--version} (@pxref{Common options}).
11175 @node truncate invocation
11176 @section @command{truncate}: Shrink or extend the size of a file
11179 @cindex truncating, file sizes
11181 @command{truncate} shrinks or extends the size of each @var{file} to the
11182 specified size. Synopsis:
11185 truncate @var{option}@dots{} @var{file}@dots{}
11188 @cindex files, creating
11189 Any @var{file} that does not exist is created.
11191 @cindex sparse files, creating
11192 @cindex holes, creating files with
11193 If a @var{file} is larger than the specified size, the extra data is lost.
11194 If a @var{file} is shorter, it is extended and the extended part (or hole)
11195 reads as zero bytes.
11197 The program accepts the following options. Also see @ref{Common options}.
11204 @opindex --no-create
11205 Do not create files that do not exist.
11210 @opindex --io-blocks
11211 Treat @var{size} as number of I/O blocks of the @var{file} rather than bytes.
11213 @item -r @var{rfile}
11214 @itemx --reference=@var{rfile}
11216 @opindex --reference
11217 Base the size of each @var{file} on the size of @var{rfile}.
11219 @item -s @var{size}
11220 @itemx --size=@var{size}
11223 Set or adjust the size of each @var{file} according to @var{size}.
11224 @multiplierSuffixesNoBlocks{size}
11226 @var{size} may also be prefixed by one of the following to adjust
11227 the size of each @var{file} based on their current size:
11229 @samp{+} => extend by
11230 @samp{-} => reduce by
11231 @samp{<} => at most
11232 @samp{>} => at least
11233 @samp{/} => round down to multiple of
11234 @samp{%} => round up to multiple of
11242 @node Printing text
11243 @chapter Printing text
11245 @cindex printing text, commands for
11246 @cindex commands for printing text
11248 This section describes commands that display text strings.
11251 * echo invocation:: Print a line of text.
11252 * printf invocation:: Format and print data.
11253 * yes invocation:: Print a string until interrupted.
11257 @node echo invocation
11258 @section @command{echo}: Print a line of text
11261 @cindex displaying text
11262 @cindex printing text
11263 @cindex text, displaying
11264 @cindex arbitrary text, displaying
11266 @command{echo} writes each given @var{string} to standard output, with a
11267 space between each and a newline after the last one. Synopsis:
11270 echo [@var{option}]@dots{} [@var{string}]@dots{}
11273 @mayConflictWithShellBuiltIn{echo}
11275 The program accepts the following options. Also see @ref{Common options}.
11276 Options must precede operands, and the normally-special argument
11277 @samp{--} has no special meaning and is treated like any other
11283 Do not output the trailing newline.
11287 @cindex backslash escapes
11288 Enable interpretation of the following backslash-escaped characters in
11297 produce no further output
11313 the eight-bit value that is the octal number @var{nnn}
11314 (zero to three octal digits), if @var{nnn} is
11315 a nine-bit value, the ninth bit is ignored
11317 the eight-bit value that is the octal number @var{nnn}
11318 (one to three octal digits), if @var{nnn} is
11319 a nine-bit value, the ninth bit is ignored
11321 the eight-bit value that is the hexadecimal number @var{hh}
11322 (one or two hexadecimal digits)
11327 @cindex backslash escapes
11328 Disable interpretation of backslash escapes in each @var{string}.
11329 This is the default. If @option{-e} and @option{-E} are both
11330 specified, the last one given takes effect.
11334 @vindex POSIXLY_CORRECT
11335 If the @env{POSIXLY_CORRECT} environment variable is set, then when
11336 @command{echo}'s first argument is not @option{-n} it outputs
11337 option-like arguments instead of treating them as options. For
11338 example, @code{echo -ne hello} outputs @samp{-ne hello} instead of
11339 plain @samp{hello}.
11341 @acronym{POSIX} does not require support for any options, and says
11342 that the behavior of @command{echo} is implementation-defined if any
11343 @var{string} contains a backslash or if the first argument is
11344 @option{-n}. Portable programs can use the @command{printf} command
11345 if they need to omit trailing newlines or output control characters or
11346 backslashes. @xref{printf invocation}.
11351 @node printf invocation
11352 @section @command{printf}: Format and print data
11355 @command{printf} does formatted printing of text. Synopsis:
11358 printf @var{format} [@var{argument}]@dots{}
11361 @command{printf} prints the @var{format} string, interpreting @samp{%}
11362 directives and @samp{\} escapes to format numeric and string arguments
11363 in a way that is mostly similar to the C @samp{printf} function.
11364 @xref{Output Conversion Syntax,, @command{printf} format directives,
11365 libc, The GNU C Library Reference Manual}, for details.
11366 The differences are listed below.
11368 @mayConflictWithShellBuiltIn{printf}
11373 The @var{format} argument is reused as necessary to convert all the
11374 given @var{argument}s. For example, the command @samp{printf %s a b}
11378 Missing @var{argument}s are treated as null strings or as zeros,
11379 depending on whether the context expects a string or a number. For
11380 example, the command @samp{printf %sx%d} prints @samp{x0}.
11384 An additional escape, @samp{\c}, causes @command{printf} to produce no
11385 further output. For example, the command @samp{printf 'A%sC\cD%sF' B
11386 E} prints @samp{ABC}.
11389 The hexadecimal escape sequence @samp{\x@var{hh}} has at most two
11390 digits, as opposed to C where it can have an unlimited number of
11391 digits. For example, the command @samp{printf '\x07e'} prints two
11392 bytes, whereas the C statement @samp{printf ("\x07e")} prints just
11397 @command{printf} has an additional directive, @samp{%b}, which prints its
11398 argument string with @samp{\} escapes interpreted in the same way as in
11399 the @var{format} string, except that octal escapes are of the form
11400 @samp{\0@var{ooo}} where @var{ooo} is 0 to 3 octal digits. If
11401 @samp{\@var{ooo}} is nine-bit value, ignore the ninth bit.
11402 If a precision is also given, it limits the number of bytes printed
11403 from the converted string.
11406 Numeric arguments must be single C constants, possibly with leading
11407 @samp{+} or @samp{-}. For example, @samp{printf %.4d -3} outputs
11411 @vindex POSIXLY_CORRECT
11412 If the leading character of a numeric argument is @samp{"} or @samp{'}
11413 then its value is the numeric value of the immediately following
11414 character. Any remaining characters are silently ignored if the
11415 @env{POSIXLY_CORRECT} environment variable is set; otherwise, a
11416 warning is printed. For example, @samp{printf "%d" "'a"} outputs
11417 @samp{97} on hosts that use the @acronym{ASCII} character set, since
11418 @samp{a} has the numeric value 97 in @acronym{ASCII}.
11423 A floating-point argument must use a period before any fractional
11424 digits, but is printed according to the @env{LC_NUMERIC} category of the
11425 current locale. For example, in a locale whose radix character is a
11426 comma, the command @samp{printf %g 3.14} outputs @samp{3,14} whereas
11427 the command @samp{printf %g 3,14} is an error.
11428 @xref{Floating point}.
11432 @command{printf} interprets @samp{\@var{ooo}} in @var{format} as an octal number
11433 (if @var{ooo} is 1 to 3 octal digits) specifying a byte to print,
11434 and @samp{\x@var{hh}} as a hexadecimal number (if @var{hh} is 1 to 2 hex
11435 digits) specifying a character to print.
11436 Note however that when @samp{\@var{ooo}} specifies a number larger than 255,
11437 @command{printf} ignores the ninth bit.
11438 For example, @samp{printf '\400'} is equivalent to @samp{printf '\0'}.
11443 @cindex ISO/IEC 10646
11445 @command{printf} interprets two character syntaxes introduced in
11446 @acronym{ISO} C 99:
11447 @samp{\u} for 16-bit Unicode (@acronym{ISO}/@acronym{IEC} 10646)
11448 characters, specified as
11449 four hexadecimal digits @var{hhhh}, and @samp{\U} for 32-bit Unicode
11450 characters, specified as eight hexadecimal digits @var{hhhhhhhh}.
11451 @command{printf} outputs the Unicode characters
11452 according to the @env{LC_CTYPE} locale. Unicode characters in the ranges
11453 U+0000...U+009F, U+D800...U+DFFF cannot be specified by this syntax, except
11454 for U+0024 ($), U+0040 (@@), and U+0060 (@`).
11456 The processing of @samp{\u} and @samp{\U} requires a full-featured
11457 @code{iconv} facility. It is activated on systems with glibc 2.2 (or newer),
11458 or when @code{libiconv} is installed prior to this package. Otherwise
11459 @samp{\u} and @samp{\U} will print as-is.
11461 The only options are a lone @option{--help} or
11462 @option{--version}. @xref{Common options}.
11463 Options must precede operands.
11465 The Unicode character syntaxes are useful for writing strings in a locale
11466 independent way. For example, a string containing the Euro currency symbol
11469 $ env printf '\u20AC 14.95'
11473 will be output correctly in all locales supporting the Euro symbol
11474 (@acronym{ISO}-8859-15, UTF-8, and others). Similarly, a Chinese string
11477 $ env printf '\u4e2d\u6587'
11481 will be output correctly in all Chinese locales (GB2312, BIG5, UTF-8, etc).
11483 Note that in these examples, the @command{printf} command has been
11484 invoked via @command{env} to ensure that we run the program found via
11485 your shell's search path, and not a shell alias or a built-in function.
11487 For larger strings, you don't need to look up the hexadecimal code
11488 values of each character one by one. @acronym{ASCII} characters mixed with \u
11489 escape sequences is also known as the JAVA source file encoding. You can
11490 use GNU recode 3.5c (or newer) to convert strings to this encoding. Here
11491 is how to convert a piece of text into a shell script which will output
11492 this text in a locale-independent way:
11495 $ LC_CTYPE=zh_CN.big5 /usr/local/bin/printf \
11496 '\u4e2d\u6587\n' > sample.txt
11497 $ recode BIG5..JAVA < sample.txt \
11498 | sed -e "s|^|/usr/local/bin/printf '|" -e "s|$|\\\\n'|" \
11505 @node yes invocation
11506 @section @command{yes}: Print a string until interrupted
11509 @cindex repeated output of a string
11511 @command{yes} prints the command line arguments, separated by spaces and
11512 followed by a newline, forever until it is killed. If no arguments are
11513 given, it prints @samp{y} followed by a newline forever until killed.
11515 Upon a write error, @command{yes} exits with status @samp{1}.
11517 The only options are a lone @option{--help} or @option{--version}.
11518 To output an argument that begins with
11519 @samp{-}, precede it with @option{--}, e.g., @samp{yes -- --help}.
11520 @xref{Common options}.
11524 @chapter Conditions
11527 @cindex commands for exit status
11528 @cindex exit status commands
11530 This section describes commands that are primarily useful for their exit
11531 status, rather than their output. Thus, they are often used as the
11532 condition of shell @code{if} statements, or as the last command in a
11536 * false invocation:: Do nothing, unsuccessfully.
11537 * true invocation:: Do nothing, successfully.
11538 * test invocation:: Check file types and compare values.
11539 * expr invocation:: Evaluate expressions.
11543 @node false invocation
11544 @section @command{false}: Do nothing, unsuccessfully
11547 @cindex do nothing, unsuccessfully
11548 @cindex failure exit status
11549 @cindex exit status of @command{false}
11551 @command{false} does nothing except return an exit status of 1, meaning
11552 @dfn{failure}. It can be used as a place holder in shell scripts
11553 where an unsuccessful command is needed.
11554 In most modern shells, @command{false} is a built-in command, so when
11555 you use @samp{false} in a script, you're probably using the built-in
11556 command, not the one documented here.
11558 @command{false} honors the @option{--help} and @option{--version} options.
11560 This version of @command{false} is implemented as a C program, and is thus
11561 more secure and faster than a shell script implementation, and may safely
11562 be used as a dummy shell for the purpose of disabling accounts.
11564 Note that @command{false} (unlike all other programs documented herein)
11565 exits unsuccessfully, even when invoked with
11566 @option{--help} or @option{--version}.
11568 Portable programs should not assume that the exit status of
11569 @command{false} is 1, as it is greater than 1 on some
11570 non-@acronym{GNU} hosts.
11573 @node true invocation
11574 @section @command{true}: Do nothing, successfully
11577 @cindex do nothing, successfully
11579 @cindex successful exit
11580 @cindex exit status of @command{true}
11582 @command{true} does nothing except return an exit status of 0, meaning
11583 @dfn{success}. It can be used as a place holder in shell scripts
11584 where a successful command is needed, although the shell built-in
11585 command @code{:} (colon) may do the same thing faster.
11586 In most modern shells, @command{true} is a built-in command, so when
11587 you use @samp{true} in a script, you're probably using the built-in
11588 command, not the one documented here.
11590 @command{true} honors the @option{--help} and @option{--version} options.
11592 Note, however, that it is possible to cause @command{true}
11593 to exit with nonzero status: with the @option{--help} or @option{--version}
11594 option, and with standard
11595 output already closed or redirected to a file that evokes an I/O error.
11596 For example, using a Bourne-compatible shell:
11599 $ ./true --version >&-
11600 ./true: write error: Bad file number
11601 $ ./true --version > /dev/full
11602 ./true: write error: No space left on device
11605 This version of @command{true} is implemented as a C program, and is thus
11606 more secure and faster than a shell script implementation, and may safely
11607 be used as a dummy shell for the purpose of disabling accounts.
11609 @node test invocation
11610 @section @command{test}: Check file types and compare values
11613 @cindex check file types
11614 @cindex compare values
11615 @cindex expression evaluation
11617 @command{test} returns a status of 0 (true) or 1 (false) depending on the
11618 evaluation of the conditional expression @var{expr}. Each part of the
11619 expression must be a separate argument.
11621 @command{test} has file status checks, string operators, and numeric
11622 comparison operators.
11624 @command{test} has an alternate form that uses opening and closing
11625 square brackets instead a leading @samp{test}. For example, instead
11626 of @samp{test -d /}, you can write @samp{[ -d / ]}. The square
11627 brackets must be separate arguments; for example, @samp{[-d /]} does
11628 not have the desired effect. Since @samp{test @var{expr}} and @samp{[
11629 @var{expr} ]} have the same meaning, only the former form is discussed
11635 test @var{expression}
11637 [ @var{expression} ]
11642 @mayConflictWithShellBuiltIn{test}
11644 If @var{expression} is omitted, @command{test} returns false.
11645 If @var{expression} is a single argument,
11646 @command{test} returns false if the argument is null and true
11647 otherwise. The argument
11648 can be any string, including strings like @samp{-d}, @samp{-1},
11649 @samp{--}, @samp{--help}, and @samp{--version} that most other
11650 programs would treat as options. To get help and version information,
11651 invoke the commands @samp{[ --help} and @samp{[ --version}, without
11652 the usual closing brackets. @xref{Common options}.
11654 @cindex exit status of @command{test}
11658 0 if the expression is true,
11659 1 if the expression is false,
11660 2 if an error occurred.
11664 * File type tests:: -[bcdfhLpSt]
11665 * Access permission tests:: -[gkruwxOG]
11666 * File characteristic tests:: -e -s -nt -ot -ef
11667 * String tests:: -z -n = == !=
11668 * Numeric tests:: -eq -ne -lt -le -gt -ge
11669 * Connectives for test:: ! -a -o
11673 @node File type tests
11674 @subsection File type tests
11676 @cindex file type tests
11678 These options test for particular types of files. (Everything's a file,
11679 but not all files are the same!)
11683 @item -b @var{file}
11685 @cindex block special check
11686 True if @var{file} exists and is a block special device.
11688 @item -c @var{file}
11690 @cindex character special check
11691 True if @var{file} exists and is a character special device.
11693 @item -d @var{file}
11695 @cindex directory check
11696 True if @var{file} exists and is a directory.
11698 @item -f @var{file}
11700 @cindex regular file check
11701 True if @var{file} exists and is a regular file.
11703 @item -h @var{file}
11704 @itemx -L @var{file}
11707 @cindex symbolic link check
11708 True if @var{file} exists and is a symbolic link.
11709 Unlike all other file-related tests, this test does not dereference
11710 @var{file} if it is a symbolic link.
11712 @item -p @var{file}
11714 @cindex named pipe check
11715 True if @var{file} exists and is a named pipe.
11717 @item -S @var{file}
11719 @cindex socket check
11720 True if @var{file} exists and is a socket.
11724 @cindex terminal check
11725 True if @var{fd} is a file descriptor that is associated with a
11731 @node Access permission tests
11732 @subsection Access permission tests
11734 @cindex access permission tests
11735 @cindex permission tests
11737 These options test for particular access permissions.
11741 @item -g @var{file}
11743 @cindex set-group-ID check
11744 True if @var{file} exists and has its set-group-ID bit set.
11746 @item -k @var{file}
11748 @cindex sticky bit check
11749 True if @var{file} exists and has its @dfn{sticky} bit set.
11751 @item -r @var{file}
11753 @cindex readable file check
11754 True if @var{file} exists and read permission is granted.
11756 @item -u @var{file}
11758 @cindex set-user-ID check
11759 True if @var{file} exists and has its set-user-ID bit set.
11761 @item -w @var{file}
11763 @cindex writable file check
11764 True if @var{file} exists and write permission is granted.
11766 @item -x @var{file}
11768 @cindex executable file check
11769 True if @var{file} exists and execute permission is granted
11770 (or search permission, if it is a directory).
11772 @item -O @var{file}
11774 @cindex owned by effective user ID check
11775 True if @var{file} exists and is owned by the current effective user ID.
11777 @item -G @var{file}
11779 @cindex owned by effective group ID check
11780 True if @var{file} exists and is owned by the current effective group ID.
11784 @node File characteristic tests
11785 @subsection File characteristic tests
11787 @cindex file characteristic tests
11789 These options test other file characteristics.
11793 @item -e @var{file}
11795 @cindex existence-of-file check
11796 True if @var{file} exists.
11798 @item -s @var{file}
11800 @cindex nonempty file check
11801 True if @var{file} exists and has a size greater than zero.
11803 @item @var{file1} -nt @var{file2}
11805 @cindex newer-than file check
11806 True if @var{file1} is newer (according to modification date) than
11807 @var{file2}, or if @var{file1} exists and @var{file2} does not.
11809 @item @var{file1} -ot @var{file2}
11811 @cindex older-than file check
11812 True if @var{file1} is older (according to modification date) than
11813 @var{file2}, or if @var{file2} exists and @var{file1} does not.
11815 @item @var{file1} -ef @var{file2}
11817 @cindex same file check
11818 @cindex hard link check
11819 True if @var{file1} and @var{file2} have the same device and inode
11820 numbers, i.e., if they are hard links to each other.
11826 @subsection String tests
11828 @cindex string tests
11830 These options test string characteristics. You may need to quote
11831 @var{string} arguments for the shell. For example:
11837 The quotes here prevent the wrong arguments from being passed to
11838 @command{test} if @samp{$V} is empty or contains special characters.
11842 @item -z @var{string}
11844 @cindex zero-length string check
11845 True if the length of @var{string} is zero.
11847 @item -n @var{string}
11848 @itemx @var{string}
11850 @cindex nonzero-length string check
11851 True if the length of @var{string} is nonzero.
11853 @item @var{string1} = @var{string2}
11855 @cindex equal string check
11856 True if the strings are equal.
11858 @item @var{string1} == @var{string2}
11860 @cindex equal string check
11861 True if the strings are equal (synonym for =).
11863 @item @var{string1} != @var{string2}
11865 @cindex not-equal string check
11866 True if the strings are not equal.
11871 @node Numeric tests
11872 @subsection Numeric tests
11874 @cindex numeric tests
11875 @cindex arithmetic tests
11877 Numeric relational operators. The arguments must be entirely numeric
11878 (possibly negative), or the special expression @w{@code{-l @var{string}}},
11879 which evaluates to the length of @var{string}.
11883 @item @var{arg1} -eq @var{arg2}
11884 @itemx @var{arg1} -ne @var{arg2}
11885 @itemx @var{arg1} -lt @var{arg2}
11886 @itemx @var{arg1} -le @var{arg2}
11887 @itemx @var{arg1} -gt @var{arg2}
11888 @itemx @var{arg1} -ge @var{arg2}
11895 These arithmetic binary operators return true if @var{arg1} is equal,
11896 not-equal, less-than, less-than-or-equal, greater-than, or
11897 greater-than-or-equal than @var{arg2}, respectively.
11904 test -1 -gt -2 && echo yes
11906 test -l abc -gt 1 && echo yes
11909 @error{} test: integer expression expected before -eq
11913 @node Connectives for test
11914 @subsection Connectives for @command{test}
11916 @cindex logical connectives
11917 @cindex connectives, logical
11919 The usual logical connectives.
11925 True if @var{expr} is false.
11927 @item @var{expr1} -a @var{expr2}
11929 @cindex logical and operator
11930 @cindex and operator
11931 True if both @var{expr1} and @var{expr2} are true.
11933 @item @var{expr1} -o @var{expr2}
11935 @cindex logical or operator
11936 @cindex or operator
11937 True if either @var{expr1} or @var{expr2} is true.
11942 @node expr invocation
11943 @section @command{expr}: Evaluate expressions
11946 @cindex expression evaluation
11947 @cindex evaluation of expressions
11949 @command{expr} evaluates an expression and writes the result on standard
11950 output. Each token of the expression must be a separate argument.
11952 Operands are either integers or strings. Integers consist of one or
11953 more decimal digits, with an optional leading @samp{-}.
11954 @command{expr} converts
11955 anything appearing in an operand position to an integer or a string
11956 depending on the operation being applied to it.
11958 Strings are not quoted for @command{expr} itself, though you may need to
11959 quote them to protect characters with special meaning to the shell,
11960 e.g., spaces. However, regardless of whether it is quoted, a string
11961 operand should not be a parenthesis or any of @command{expr}'s
11962 operators like @code{+}, so you cannot safely pass an arbitrary string
11963 @code{$str} to expr merely by quoting it to the shell. One way to
11964 work around this is to use the @sc{gnu} extension @code{+},
11965 (e.g., @code{+ "$str" = foo}); a more portable way is to use
11966 @code{@w{" $str"}} and to adjust the rest of the expression to take
11967 the leading space into account (e.g., @code{@w{" $str" = " foo"}}).
11969 You should not pass a negative integer or a string with leading
11970 @samp{-} as @command{expr}'s first argument, as it might be
11971 misinterpreted as an option; this can be avoided by parenthesization.
11972 Also, portable scripts should not use a string operand that happens to
11973 take the form of an integer; this can be worked around by inserting
11974 leading spaces as mentioned above.
11976 @cindex parentheses for grouping
11977 Operators may be given as infix symbols or prefix keywords. Parentheses
11978 may be used for grouping in the usual manner. You must quote
11979 parentheses and many operators to avoid the shell evaluating them,
11982 When built with support for the GNU MP library, @command{expr} uses
11983 arbitrary-precision arithmetic; otherwise, it uses native arithmetic
11984 types and may fail due to arithmetic overflow.
11986 The only options are @option{--help} and @option{--version}. @xref{Common
11987 options}. Options must precede operands.
11989 @cindex exit status of @command{expr}
11993 0 if the expression is neither null nor 0,
11994 1 if the expression is null or 0,
11995 2 if the expression is invalid,
11996 3 if an internal error occurred (e.g., arithmetic overflow).
12000 * String expressions:: + : match substr index length
12001 * Numeric expressions:: + - * / %
12002 * Relations for expr:: | & < <= = == != >= >
12003 * Examples of expr:: Examples.
12007 @node String expressions
12008 @subsection String expressions
12010 @cindex string expressions
12011 @cindex expressions, string
12013 @command{expr} supports pattern matching and other string operators. These
12014 have higher precedence than both the numeric and relational operators (in
12015 the next sections).
12019 @item @var{string} : @var{regex}
12020 @cindex pattern matching
12021 @cindex regular expression matching
12022 @cindex matching patterns
12023 Perform pattern matching. The arguments are converted to strings and the
12024 second is considered to be a (basic, a la GNU @code{grep}) regular
12025 expression, with a @code{^} implicitly prepended. The first argument is
12026 then matched against this regular expression.
12028 If the match succeeds and @var{regex} uses @samp{\(} and @samp{\)}, the
12029 @code{:} expression returns the part of @var{string} that matched the
12030 subexpression; otherwise, it returns the number of characters matched.
12032 If the match fails, the @code{:} operator returns the null string if
12033 @samp{\(} and @samp{\)} are used in @var{regex}, otherwise 0.
12035 @kindex \( @r{regexp operator}
12036 Only the first @samp{\( @dots{} \)} pair is relevant to the return
12037 value; additional pairs are meaningful only for grouping the regular
12038 expression operators.
12040 @kindex \+ @r{regexp operator}
12041 @kindex \? @r{regexp operator}
12042 @kindex \| @r{regexp operator}
12043 In the regular expression, @code{\+}, @code{\?}, and @code{\|} are
12044 operators which respectively match one or more, zero or one, or separate
12045 alternatives. SunOS and other @command{expr}'s treat these as regular
12046 characters. (@acronym{POSIX} allows either behavior.)
12047 @xref{Top, , Regular Expression Library, regex, Regex}, for details of
12048 regular expression syntax. Some examples are in @ref{Examples of expr}.
12050 @item match @var{string} @var{regex}
12052 An alternative way to do pattern matching. This is the same as
12053 @w{@samp{@var{string} : @var{regex}}}.
12055 @item substr @var{string} @var{position} @var{length}
12057 Returns the substring of @var{string} beginning at @var{position}
12058 with length at most @var{length}. If either @var{position} or
12059 @var{length} is negative, zero, or non-numeric, returns the null string.
12061 @item index @var{string} @var{charset}
12063 Returns the first position in @var{string} where the first character in
12064 @var{charset} was found. If no character in @var{charset} is found in
12065 @var{string}, return 0.
12067 @item length @var{string}
12069 Returns the length of @var{string}.
12071 @item + @var{token}
12073 Interpret @var{token} as a string, even if it is a keyword like @var{match}
12074 or an operator like @code{/}.
12075 This makes it possible to test @code{expr length + "$x"} or
12076 @code{expr + "$x" : '.*/\(.\)'} and have it do the right thing even if
12077 the value of @var{$x} happens to be (for example) @code{/} or @code{index}.
12078 This operator is a @acronym{GNU} extension. Portable shell scripts should use
12079 @code{@w{" $token"} : @w{' \(.*\)'}} instead of @code{+ "$token"}.
12083 To make @command{expr} interpret keywords as strings, you must use the
12084 @code{quote} operator.
12087 @node Numeric expressions
12088 @subsection Numeric expressions
12090 @cindex numeric expressions
12091 @cindex expressions, numeric
12093 @command{expr} supports the usual numeric operators, in order of increasing
12094 precedence. These numeric operators have lower precedence than the
12095 string operators described in the previous section, and higher precedence
12096 than the connectives (next section).
12104 @cindex subtraction
12105 Addition and subtraction. Both arguments are converted to integers;
12106 an error occurs if this cannot be done.
12112 @cindex multiplication
12115 Multiplication, division, remainder. Both arguments are converted to
12116 integers; an error occurs if this cannot be done.
12121 @node Relations for expr
12122 @subsection Relations for @command{expr}
12124 @cindex connectives, logical
12125 @cindex logical connectives
12126 @cindex relations, numeric or string
12128 @command{expr} supports the usual logical connectives and relations. These
12129 have lower precedence than the string and numeric operators
12130 (previous sections). Here is the list, lowest-precedence operator first.
12136 @cindex logical or operator
12137 @cindex or operator
12138 Returns its first argument if that is neither null nor zero, otherwise
12139 its second argument if it is neither null nor zero, otherwise 0. It
12140 does not evaluate its second argument if its first argument is neither
12145 @cindex logical and operator
12146 @cindex and operator
12147 Return its first argument if neither argument is null or zero, otherwise
12148 0. It does not evaluate its second argument if its first argument is
12151 @item < <= = == != >= >
12158 @cindex comparison operators
12160 Compare the arguments and return 1 if the relation is true, 0 otherwise.
12161 @code{==} is a synonym for @code{=}. @command{expr} first tries to convert
12162 both arguments to integers and do a numeric comparison; if either
12163 conversion fails, it does a lexicographic comparison using the character
12164 collating sequence specified by the @env{LC_COLLATE} locale.
12169 @node Examples of expr
12170 @subsection Examples of using @command{expr}
12172 @cindex examples of @command{expr}
12173 Here are a few examples, including quoting for shell metacharacters.
12175 To add 1 to the shell variable @code{foo}, in Bourne-compatible shells:
12178 foo=`expr $foo + 1`
12181 To print the non-directory part of the file name stored in
12182 @code{$fname}, which need not contain a @code{/}:
12185 expr $fname : '.*/\(.*\)' '|' $fname
12188 An example showing that @code{\+} is an operator:
12196 expr abc : 'a\(.\)c'
12198 expr index abcdef cz
12201 @error{} expr: syntax error
12202 expr index + index a
12208 @chapter Redirection
12210 @cindex redirection
12211 @cindex commands for redirection
12213 Unix shells commonly provide several forms of @dfn{redirection}---ways
12214 to change the input source or output destination of a command. But one
12215 useful redirection is performed by a separate command, not by the shell;
12216 it's described here.
12219 * tee invocation:: Redirect output to multiple files or processes.
12223 @node tee invocation
12224 @section @command{tee}: Redirect output to multiple files or processes
12227 @cindex pipe fitting
12228 @cindex destinations, multiple output
12229 @cindex read from stdin and write to stdout and files
12231 The @command{tee} command copies standard input to standard output and also
12232 to any files given as arguments. This is useful when you want not only
12233 to send some data down a pipe, but also to save a copy. Synopsis:
12236 tee [@var{option}]@dots{} [@var{file}]@dots{}
12239 If a file being written to does not already exist, it is created. If a
12240 file being written to already exists, the data it previously contained
12241 is overwritten unless the @option{-a} option is used.
12243 A @var{file} of @samp{-} causes @command{tee} to send another copy of
12244 input to standard output, but this is typically not that useful as the
12245 copies are interleaved.
12247 The program accepts the following options. Also see @ref{Common options}.
12254 Append standard input to the given files rather than overwriting
12258 @itemx --ignore-interrupts
12260 @opindex --ignore-interrupts
12261 Ignore interrupt signals.
12265 The @command{tee} command is useful when you happen to be transferring a large
12266 amount of data and also want to summarize that data without reading
12267 it a second time. For example, when you are downloading a DVD image,
12268 you often want to verify its signature or checksum right away.
12269 The inefficient way to do it is simply:
12272 wget http://example.com/some.iso && sha1sum some.iso
12275 One problem with the above is that it makes you wait for the
12276 download to complete before starting the time-consuming SHA1 computation.
12277 Perhaps even more importantly, the above requires reading
12278 the DVD image a second time (the first was from the network).
12280 The efficient way to do it is to interleave the download
12281 and SHA1 computation. Then, you'll get the checksum for
12282 free, because the entire process parallelizes so well:
12285 # slightly contrived, to demonstrate process substitution
12286 wget -O - http://example.com/dvd.iso \
12287 | tee >(sha1sum > dvd.sha1) > dvd.iso
12290 That makes @command{tee} write not just to the expected output file,
12291 but also to a pipe running @command{sha1sum} and saving the final
12292 checksum in a file named @file{dvd.sha1}.
12294 Note, however, that this example relies on a feature of modern shells
12295 called @dfn{process substitution}
12296 (the @samp{>(command)} syntax, above;
12297 @xref{Process Substitution,,Process Substitution, bashref,
12298 The Bash Reference Manual}.),
12299 so it works with @command{zsh}, @command{bash}, and @command{ksh},
12300 but not with @command{/bin/sh}. So if you write code like this
12301 in a shell script, be sure to start the script with @samp{#!/bin/bash}.
12303 Since the above example writes to one file and one process,
12304 a more conventional and portable use of @command{tee} is even better:
12307 wget -O - http://example.com/dvd.iso \
12308 | tee dvd.iso | sha1sum > dvd.sha1
12311 You can extend this example to make @command{tee} write to two processes,
12312 computing MD5 and SHA1 checksums in parallel. In this case,
12313 process substitution is required:
12316 wget -O - http://example.com/dvd.iso \
12317 | tee >(sha1sum > dvd.sha1) \
12318 >(md5sum > dvd.md5) \
12322 This technique is also useful when you want to make a @emph{compressed}
12323 copy of the contents of a pipe.
12324 Consider a tool to graphically summarize disk usage data from @samp{du -ak}.
12325 For a large hierarchy, @samp{du -ak} can run for a long time,
12326 and can easily produce terabytes of data, so you won't want to
12327 rerun the command unnecessarily. Nor will you want to save
12328 the uncompressed output.
12330 Doing it the inefficient way, you can't even start the GUI
12331 until after you've compressed all of the @command{du} output:
12334 du -ak | gzip -9 > /tmp/du.gz
12335 gzip -d /tmp/du.gz | xdiskusage -a
12338 With @command{tee} and process substitution, you start the GUI
12339 right away and eliminate the decompression completely:
12342 du -ak | tee >(gzip -9 > /tmp/du.gz) | xdiskusage -a
12345 Finally, if you regularly create more than one type of
12346 compressed tarball at once, for example when @code{make dist} creates
12347 both @command{gzip}-compressed and @command{bzip2}-compressed tarballs,
12348 there may be a better way.
12349 Typical @command{automake}-generated @file{Makefile} rules create
12350 the two compressed tar archives with commands in sequence, like this
12351 (slightly simplified):
12354 tardir=your-pkg-M.N
12355 tar chof - "$tardir" | gzip -9 -c > your-pkg-M.N.tar.gz
12356 tar chof - "$tardir" | bzip2 -9 -c > your-pkg-M.N.tar.bz2
12359 However, if the hierarchy you are archiving and compressing is larger
12360 than a couple megabytes, and especially if you are using a multi-processor
12361 system with plenty of memory, then you can do much better by reading the
12362 directory contents only once and running the compression programs in parallel:
12365 tardir=your-pkg-M.N
12366 tar chof - "$tardir" \
12367 | tee >(gzip -9 -c > your-pkg-M.N.tar.gz) \
12368 | bzip2 -9 -c > your-pkg-M.N.tar.bz2
12374 @node File name manipulation
12375 @chapter File name manipulation
12377 @cindex file name manipulation
12378 @cindex manipulation of file names
12379 @cindex commands for file name manipulation
12381 This section describes commands that manipulate file names.
12384 * basename invocation:: Strip directory and suffix from a file name.
12385 * dirname invocation:: Strip last file name component.
12386 * pathchk invocation:: Check file name validity and portability.
12387 * mktemp invocation:: Create temporary file or directory.
12388 * realpath invocation:: Print resolved file names.
12392 @node basename invocation
12393 @section @command{basename}: Strip directory and suffix from a file name
12396 @cindex strip directory and suffix from file names
12397 @cindex directory, stripping from file names
12398 @cindex suffix, stripping from file names
12399 @cindex file names, stripping directory and suffix
12400 @cindex leading directory components, stripping
12402 @command{basename} removes any leading directory components from
12403 @var{name}. Synopsis:
12406 basename @var{name} [@var{suffix}]
12409 If @var{suffix} is specified and is identical to the end of @var{name},
12410 it is removed from @var{name} as well. Note that since trailing slashes
12411 are removed prior to suffix matching, @var{suffix} will do nothing if it
12412 contains slashes. @command{basename} prints the result on standard
12415 @c This test is used both here and in the section on dirname.
12416 @macro basenameAndDirname
12417 Together, @command{basename} and @command{dirname} are designed such
12418 that if @samp{ls "$name"} succeeds, then the command sequence @samp{cd
12419 "$(dirname "$name")"; ls "$(basename "$name")"} will, too. This works
12420 for everything except file names containing a trailing newline.
12422 @basenameAndDirname
12424 @acronym{POSIX} allows the implementation to define the results if
12425 @var{name} is empty or @samp{//}. In the former case, @acronym{GNU}
12426 @command{basename} returns the empty string. In the latter case, the
12427 result is @samp{//} on platforms where @var{//} is distinct from
12428 @var{/}, and @samp{/} on platforms where there is no difference.
12430 The only options are @option{--help} and @option{--version}. @xref{Common
12431 options}. Options must precede operands.
12439 basename /usr/bin/sort
12442 basename include/stdio.h .h
12446 @node dirname invocation
12447 @section @command{dirname}: Strip last file name component
12450 @cindex directory components, printing
12451 @cindex stripping non-directory suffix
12452 @cindex non-directory suffix, stripping
12454 @command{dirname} prints all but the final slash-delimited component of
12455 @var{name}. Slashes on either side of the final component are also
12456 removed. If the string contains no slash, @command{dirname} prints
12457 @samp{.} (meaning the current directory). Synopsis:
12463 @var{name} need not be a file name, but if it is, this operation
12464 effectively lists the directory that contains the final component,
12465 including the case when the final component is itself a directory.
12467 @basenameAndDirname
12469 @acronym{POSIX} allows the implementation to define the results if
12470 @var{name} is @samp{//}. With @acronym{GNU} @command{dirname}, the
12471 result is @samp{//} on platforms where @var{//} is distinct from
12472 @var{/}, and @samp{/} on platforms where there is no difference.
12474 The only options are @option{--help} and @option{--version}. @xref{Common
12482 # Output "/usr/bin".
12483 dirname /usr/bin/sort
12484 dirname /usr/bin//.//
12491 @node pathchk invocation
12492 @section @command{pathchk}: Check file name validity and portability
12495 @cindex file names, checking validity and portability
12496 @cindex valid file names, checking for
12497 @cindex portable file names, checking for
12499 @command{pathchk} checks validity and portability of file names. Synopsis:
12502 pathchk [@var{option}]@dots{} @var{name}@dots{}
12505 For each @var{name}, @command{pathchk} prints an error message if any of
12506 these conditions is true:
12510 One of the existing directories in @var{name} does not have search
12511 (execute) permission,
12513 The length of @var{name} is larger than the maximum supported by the
12516 The length of one component of @var{name} is longer than
12517 its file system's maximum.
12520 A nonexistent @var{name} is not an error, so long a file with that
12521 name could be created under the above conditions.
12523 The program accepts the following options. Also see @ref{Common options}.
12524 Options must precede operands.
12530 Instead of performing checks based on the underlying file system,
12531 print an error message if any of these conditions is true:
12535 A file name is empty.
12538 A file name contains a character outside the @acronym{POSIX} portable file
12539 name character set, namely, the ASCII letters and digits, @samp{.},
12540 @samp{_}, @samp{-}, and @samp{/}.
12543 The length of a file name or one of its components exceeds the
12544 @acronym{POSIX} minimum limits for portability.
12549 Print an error message if a file name is empty, or if it contains a component
12550 that begins with @samp{-}.
12552 @item --portability
12553 @opindex --portability
12554 Print an error message if a file name is not portable to all @acronym{POSIX}
12555 hosts. This option is equivalent to @samp{-p -P}.
12559 @cindex exit status of @command{pathchk}
12563 0 if all specified file names passed all checks,
12567 @node mktemp invocation
12568 @section @command{mktemp}: Create temporary file or directory
12571 @cindex file names, creating temporary
12572 @cindex directory, creating temporary
12573 @cindex temporary files and directories
12575 @command{mktemp} manages the creation of temporary files and
12576 directories. Synopsis:
12579 mktemp [@var{option}]@dots{} [@var{template}]
12582 Safely create a temporary file or directory based on @var{template},
12583 and print its name. If given, @var{template} must include at least
12584 three consecutive @samp{X}s in the last component. If omitted, the template
12585 @samp{tmp.XXXXXXXXXX} is used, and option @option{--tmpdir} is
12586 implied. The final run of @samp{X}s in the @var{template} will be replaced
12587 by alpha-numeric characters; thus, on a case-sensitive file system,
12588 and with a @var{template} including a run of @var{n} instances of @samp{X},
12589 there are @samp{62**@var{n}} potential file names.
12591 Older scripts used to create temporary files by simply joining the
12592 name of the program with the process id (@samp{$$}) as a suffix.
12593 However, that naming scheme is easily predictable, and suffers from a
12594 race condition where the attacker can create an appropriately named
12595 symbolic link, such that when the script then opens a handle to what
12596 it thought was an unused file, it is instead modifying an existing
12597 file. Using the same scheme to create a directory is slightly safer,
12598 since the @command{mkdir} will fail if the target already exists, but
12599 it is still inferior because it allows for denial of service attacks.
12600 Therefore, modern scripts should use the @command{mktemp} command to
12601 guarantee that the generated name will be unpredictable, and that
12602 knowledge of the temporary file name implies that the file was created
12603 by the current script and cannot be modified by other users.
12605 When creating a file, the resulting file has read and write
12606 permissions for the current user, but no permissions for the group or
12607 others; these permissions are reduced if the current umask is more
12610 Here are some examples (although note that if you repeat them, you
12611 will most likely get different file names):
12616 Create a temporary file in the current directory.
12623 Create a temporary file with a known suffix.
12625 $ mktemp --suffix=.txt file-XXXX
12627 $ mktemp file-XXXX-XXXX.txt
12632 Create a secure fifo relative to the user's choice of @env{TMPDIR},
12633 but falling back to the current directory rather than @file{/tmp}.
12634 Note that @command{mktemp} does not create fifos, but can create a
12635 secure directory in which the fifo can live. Exit the shell if the
12636 directory or fifo could not be created.
12638 $ dir=$(mktemp -p "$@{TMPDIR:-.@}" -d dir-XXXX) || exit 1
12640 $ mkfifo "$fifo" || @{ rmdir "$dir"; exit 1; @}
12644 Create and use a temporary file if possible, but ignore failure. The
12645 file will reside in the directory named by @env{TMPDIR}, if specified,
12646 or else in @file{/tmp}.
12648 $ file=$(mktemp -q) && @{
12649 > # Safe to use $file only within this block. Use quotes,
12650 > # since $TMPDIR, and thus $file, may contain whitespace.
12651 > echo ... > "$file"
12657 Act as a semi-random character generator (it is not fully random,
12658 since it is impacted by the contents of the current directory). To
12659 avoid security holes, do not use the resulting names to create a file.
12669 The program accepts the following options. Also see @ref{Common options}.
12676 @opindex --directory
12677 Create a directory rather than a file. The directory will have read,
12678 write, and search permissions for the current user, but no permissions
12679 for the group or others; these permissions are reduced if the current
12680 umask is more restrictive.
12686 Suppress diagnostics about failure to create a file or directory. The
12687 exit status will still reflect whether a file was created.
12693 Generate a temporary name that does not name an existing file, without
12694 changing the file system contents. Using the output of this command
12695 to create a new file is inherently unsafe, as there is a window of
12696 time between generating the name and using it where another process
12697 can create an object by the same name.
12700 @itemx --tmpdir[=@var{dir}]
12703 Treat @var{template} relative to the directory @var{dir}. If
12704 @var{dir} is not specified (only possible with the long option
12705 @option{--tmpdir}) or is the empty string, use the value of
12706 @env{TMPDIR} if available, otherwise use @samp{/tmp}. If this is
12707 specified, @var{template} must not be absolute. However,
12708 @var{template} can still contain slashes, although intermediate
12709 directories must already exist.
12711 @item --suffix=@var{suffix}
12713 Append @var{suffix} to the @var{template}. @var{suffix} must not
12714 contain slash. If @option{--suffix} is specified, @var{template} must
12715 end in @samp{X}; if it is not specified, then an appropriate
12716 @option{--suffix} is inferred by finding the last @samp{X} in
12717 @var{template}. This option exists for use with the default
12718 @var{template} and for the creation of a @var{suffix} that starts with
12723 Treat @var{template} as a single file relative to the value of
12724 @env{TMPDIR} if available, or to the directory specified by
12725 @option{-p}, otherwise to @samp{/tmp}. @var{template} must not
12726 contain slashes. This option is deprecated; the use of @option{-p}
12727 without @option{-t} offers better defaults (by favoring the command
12728 line over @env{TMPDIR}) and more flexibility (by allowing intermediate
12733 @cindex exit status of @command{mktemp}
12737 0 if the file was created,
12742 @node realpath invocation
12743 @section @command{realpath}: Print the resolved file name.
12746 @cindex file names, canonicalization
12747 @cindex symlinks, resolution
12748 @cindex canonical file name
12749 @cindex canonicalize a file name
12753 @command{realpath} expands all symbolic links and resolves references to
12754 @samp{/./}, @samp{/../} and extra @samp{/} characters. By default,
12755 all but the last component of the specified files must exist. Synopsis:
12758 realpath [@var{option}]@dots{} @var{file}@dots{}
12761 The program accepts the following options. Also see @ref{Common options}.
12766 @itemx --canonicalize-existing
12768 @opindex --canonicalize-existing
12769 Ensure that all components of the specified file names exist.
12770 If any component is missing or unavailable, @command{realpath} will output
12771 a diagnostic unless the @option{-q} option is specified, and exit with a
12772 nonzero exit code. A trailing slash requires that the name resolve to a
12776 @itemx --canonicalize-missing
12778 @opindex --canonicalize-missing
12779 If any component of a specified file name is missing or unavailable,
12780 treat it as a directory.
12786 Symbolic links are resolved in the specified file names,
12787 but they are resolved after any subsequent @samp{..} components are processed.
12792 @opindex --physical
12793 Symbolic links are resolved in the specified file names,
12794 and they are resolved before any subsequent @samp{..} components are processed.
12795 This is the default mode of operation.
12801 Suppress diagnostic messages for specified file names.
12805 @itemx --no-symlinks
12808 @opindex --no-symlinks
12809 Do not resolve symbolic links. Only resolve references to
12810 @samp{/./}, @samp{/../} and remove extra @samp{/} characters.
12811 When combined with the @option{-m} option, realpath operates
12812 only on the file name, and does not touch any actual file.
12818 Separate output items with @sc{nul} characters.
12820 @itemx --relative-to=@var{file}
12821 @opindex --relative-to
12823 Print the resolved file names relative to the specified file.
12824 Note this option honors the @option{-m} and @option{-e} options
12825 pertaining to file existence.
12827 @itemx --relative-base=@var{base}
12828 @opindex --relative-base
12829 This option is valid when used with @option{--relative-to}, and will restrict
12830 the output of @option{--relative-to} so that relative names are output,
12831 only when @var{file}s are descendants of @var{base}. Otherwise output the
12832 absolute file name. Note: this option honors the @option{-m} and @option{-e}
12833 options pertaining to file existence. For example:
12836 realpath --relative-to=/usr /tmp /usr/bin
12839 realpath --relative-base=/usr --relative-to=/usr /tmp /usr/bin
12846 @cindex exit status of @command{realpath}
12850 0 if all file names were printed without issue.
12855 @node Working context
12856 @chapter Working context
12858 @cindex working context
12859 @cindex commands for printing the working context
12861 This section describes commands that display or alter the context in
12862 which you are working: the current directory, the terminal settings, and
12863 so forth. See also the user-related commands in the next section.
12866 * pwd invocation:: Print working directory.
12867 * stty invocation:: Print or change terminal characteristics.
12868 * printenv invocation:: Print environment variables.
12869 * tty invocation:: Print file name of terminal on standard input.
12873 @node pwd invocation
12874 @section @command{pwd}: Print working directory
12877 @cindex print name of current directory
12878 @cindex current working directory, printing
12879 @cindex working directory, printing
12882 @command{pwd} prints the name of the current directory. Synopsis:
12885 pwd [@var{option}]@dots{}
12888 The program accepts the following options. Also see @ref{Common options}.
12895 If the contents of the environment variable @env{PWD} provide an
12896 absolute name of the current directory with no @samp{.} or @samp{..}
12897 components, but possibly with symbolic links, then output those
12898 contents. Otherwise, fall back to default @option{-P} handling.
12903 @opindex --physical
12904 Print a fully resolved name for the current directory. That is, all
12905 components of the printed name will be actual directory names---none
12906 will be symbolic links.
12909 @cindex symbolic links and @command{pwd}
12910 If @option{-L} and @option{-P} are both given, the last one takes
12911 precedence. If neither option is given, then this implementation uses
12912 @option{-P} as the default unless the @env{POSIXLY_CORRECT}
12913 environment variable is set.
12915 @mayConflictWithShellBuiltIn{pwd}
12920 @node stty invocation
12921 @section @command{stty}: Print or change terminal characteristics
12924 @cindex change or print terminal settings
12925 @cindex terminal settings
12926 @cindex line settings of terminal
12928 @command{stty} prints or changes terminal characteristics, such as baud rate.
12932 stty [@var{option}] [@var{setting}]@dots{}
12933 stty [@var{option}]
12936 If given no line settings, @command{stty} prints the baud rate, line
12937 discipline number (on systems that support it), and line settings
12938 that have been changed from the values set by @samp{stty sane}.
12939 By default, mode reading and setting are performed on the tty line
12940 connected to standard input, although this can be modified by the
12941 @option{--file} option.
12943 @command{stty} accepts many non-option arguments that change aspects of
12944 the terminal line operation, as described below.
12946 The program accepts the following options. Also see @ref{Common options}.
12953 Print all current settings in human-readable form. This option may not
12954 be used in combination with any line settings.
12956 @item -F @var{device}
12957 @itemx --file=@var{device}
12960 Set the line opened by the file name specified in @var{device} instead of
12961 the tty line connected to standard input. This option is necessary
12962 because opening a @acronym{POSIX} tty requires use of the
12963 @code{O_NONDELAY} flag to prevent a @acronym{POSIX} tty from blocking
12964 until the carrier detect line is high if
12965 the @code{clocal} flag is not set. Hence, it is not always possible
12966 to allow the shell to open the device in the traditional manner.
12972 @cindex machine-readable @command{stty} output
12973 Print all current settings in a form that can be used as an argument to
12974 another @command{stty} command to restore the current settings. This option
12975 may not be used in combination with any line settings.
12979 Many settings can be turned off by preceding them with a @samp{-}.
12980 Such arguments are marked below with ``May be negated'' in their
12981 description. The descriptions themselves refer to the positive
12982 case, that is, when @emph{not} negated (unless stated otherwise,
12985 Some settings are not available on all @acronym{POSIX} systems, since they use
12986 extensions. Such arguments are marked below with
12987 ``Non-@acronym{POSIX}'' in their description. On non-@acronym{POSIX}
12988 systems, those or other settings also may not
12989 be available, but it's not feasible to document all the variations: just
12995 * Control:: Control settings
12996 * Input:: Input settings
12997 * Output:: Output settings
12998 * Local:: Local settings
12999 * Combination:: Combination settings
13000 * Characters:: Special characters
13001 * Special:: Special settings
13006 @subsection Control settings
13008 @cindex control settings
13014 @cindex two-way parity
13015 Generate parity bit in output and expect parity bit in input.
13021 @cindex even parity
13022 Set odd parity (even if negated). May be negated.
13029 @cindex character size
13030 @cindex eight-bit characters
13031 Set character size to 5, 6, 7, or 8 bits.
13036 Send a hangup signal when the last process closes the tty. May be
13042 Use two stop bits per character (one if negated). May be negated.
13046 Allow input to be received. May be negated.
13050 @cindex modem control
13051 Disable modem control signals. May be negated.
13055 @cindex hardware flow control
13056 @cindex flow control, hardware
13057 @cindex RTS/CTS flow control
13058 Enable RTS/CTS flow control. Non-@acronym{POSIX}. May be negated.
13063 @subsection Input settings
13065 @cindex input settings
13066 These settings control operations on data received from the terminal.
13071 @cindex breaks, ignoring
13072 Ignore break characters. May be negated.
13076 @cindex breaks, cause interrupts
13077 Make breaks cause an interrupt signal. May be negated.
13081 @cindex parity, ignoring
13082 Ignore characters with parity errors. May be negated.
13086 @cindex parity errors, marking
13087 Mark parity errors (with a 255-0-character sequence). May be negated.
13091 Enable input parity checking. May be negated.
13095 @cindex eight-bit input
13096 Clear high (8th) bit of input characters. May be negated.
13100 @cindex newline, translating to return
13101 Translate newline to carriage return. May be negated.
13105 @cindex return, ignoring
13106 Ignore carriage return. May be negated.
13110 @cindex return, translating to newline
13111 Translate carriage return to newline. May be negated.
13115 @cindex input encoding, UTF-8
13116 Assume input characters are UTF-8 encoded. May be negated.
13120 @kindex C-s/C-q flow control
13121 @cindex XON/XOFF flow control
13122 Enable XON/XOFF flow control (that is, @kbd{CTRL-S}/@kbd{CTRL-Q}). May
13129 @cindex software flow control
13130 @cindex flow control, software
13131 Enable sending of @code{stop} character when the system input buffer
13132 is almost full, and @code{start} character when it becomes almost
13133 empty again. May be negated.
13137 @cindex uppercase, translating to lowercase
13138 Translate uppercase characters to lowercase. Non-@acronym{POSIX}. May be
13139 negated. Note ilcuc is not implemented, as one would not be able to issue
13140 almost any (lowercase) Unix command, after invoking it.
13144 Allow any character to restart output (only the start character
13145 if negated). Non-@acronym{POSIX}. May be negated.
13149 @cindex beeping at input buffer full
13150 Enable beeping and not flushing input buffer if a character arrives
13151 when the input buffer is full. Non-@acronym{POSIX}. May be negated.
13156 @subsection Output settings
13158 @cindex output settings
13159 These settings control operations on data sent to the terminal.
13164 Postprocess output. May be negated.
13168 @cindex lowercase, translating to output
13169 Translate lowercase characters to uppercase. Non-@acronym{POSIX}. May be
13170 negated. (Note ouclc is not currently implemented.)
13174 @cindex return, translating to newline
13175 Translate carriage return to newline. Non-@acronym{POSIX}. May be negated.
13179 @cindex newline, translating to crlf
13180 Translate newline to carriage return-newline. Non-@acronym{POSIX}. May be
13185 Do not print carriage returns in the first column. Non-@acronym{POSIX}.
13190 Newline performs a carriage return. Non-@acronym{POSIX}. May be negated.
13194 @cindex pad instead of timing for delaying
13195 Use fill (padding) characters instead of timing for delays.
13196 Non-@acronym{POSIX}.
13201 @cindex pad character
13202 Use @acronym{ASCII} @sc{del} characters for fill instead of
13203 @acronym{ASCII} @sc{nul} characters. Non-@acronym{POSIX}.
13209 Newline delay style. Non-@acronym{POSIX}.
13216 Carriage return delay style. Non-@acronym{POSIX}.
13222 @opindex tab@var{n}
13223 Horizontal tab delay style. Non-@acronym{POSIX}.
13228 Backspace delay style. Non-@acronym{POSIX}.
13233 Vertical tab delay style. Non-@acronym{POSIX}.
13238 Form feed delay style. Non-@acronym{POSIX}.
13243 @subsection Local settings
13245 @cindex local settings
13250 Enable @code{interrupt}, @code{quit}, and @code{suspend} special
13251 characters. May be negated.
13255 Enable @code{erase}, @code{kill}, @code{werase}, and @code{rprnt}
13256 special characters. May be negated.
13260 Enable non-@acronym{POSIX} special characters. May be negated.
13264 Echo input characters. May be negated.
13270 Echo @code{erase} characters as backspace-space-backspace. May be
13275 @cindex newline echoing after @code{kill}
13276 Echo a newline after a @code{kill} character. May be negated.
13280 @cindex newline, echoing
13281 Echo newline even if not echoing other characters. May be negated.
13285 @cindex flushing, disabling
13286 Disable flushing after @code{interrupt} and @code{quit} special
13287 characters. May be negated.
13291 @cindex case translation
13292 Enable input and output of uppercase characters by preceding their
13293 lowercase equivalents with @samp{\}, when @code{icanon} is set.
13294 Non-@acronym{POSIX}. May be negated.
13298 @cindex background jobs, stopping at terminal write
13299 Stop background jobs that try to write to the terminal. Non-@acronym{POSIX}.
13306 Echo erased characters backward, between @samp{\} and @samp{/}.
13307 Non-@acronym{POSIX}. May be negated.
13313 @cindex control characters, using @samp{^@var{c}}
13314 @cindex hat notation for control characters
13315 Echo control characters in hat notation (@samp{^@var{c}}) instead
13316 of literally. Non-@acronym{POSIX}. May be negated.
13322 Echo the @code{kill} special character by erasing each character on
13323 the line as indicated by the @code{echoprt} and @code{echoe} settings,
13324 instead of by the @code{echoctl} and @code{echok} settings.
13325 Non-@acronym{POSIX}.
13331 @subsection Combination settings
13333 @cindex combination settings
13334 Combination settings:
13341 Same as @code{parenb -parodd cs7}. May be negated. If negated, same
13342 as @code{-parenb cs8}.
13346 Same as @code{parenb parodd cs7}. May be negated. If negated, same
13347 as @code{-parenb cs8}.
13351 Same as @code{-icrnl -onlcr}. May be negated. If negated, same as
13352 @code{icrnl -inlcr -igncr onlcr -ocrnl -onlret}.
13356 Reset the @code{erase} and @code{kill} special characters to their default
13363 @c This is too long to write inline.
13365 cread -ignbrk brkint -inlcr -igncr icrnl -ixoff
13366 -iuclc -ixany imaxbel opost -olcuc -ocrnl onlcr
13367 -onocr -onlret -ofill -ofdel nl0 cr0 tab0 bs0 vt0
13368 ff0 isig icanon iexten echo echoe echok -echonl
13369 -noflsh -xcase -tostop -echoprt echoctl echoke
13373 and also sets all special characters to their default values.
13377 Same as @code{brkint ignpar istrip icrnl ixon opost isig icanon}, plus
13378 sets the @code{eof} and @code{eol} characters to their default values
13379 if they are the same as the @code{min} and @code{time} characters.
13380 May be negated. If negated, same as @code{raw}.
13387 -ignbrk -brkint -ignpar -parmrk -inpck -istrip
13388 -inlcr -igncr -icrnl -ixon -ixoff -iuclc -ixany
13389 -imaxbel -opost -isig -icanon -xcase min 1 time 0
13393 May be negated. If negated, same as @code{cooked}.
13397 Same as @option{-icanon}. May be negated. If negated, same as
13402 @cindex eight-bit characters
13403 Same as @code{-parenb -istrip cs8}. May be negated. If negated,
13404 same as @code{parenb istrip cs7}.
13408 Same as @option{-parenb -istrip -opost cs8}. May be negated.
13409 If negated, same as @code{parenb istrip opost cs7}.
13413 Same as @option{-ixany}. Non-@acronym{POSIX}. May be negated.
13417 Same as @code{tab0}. Non-@acronym{POSIX}. May be negated. If negated, same
13424 Same as @code{xcase iuclc olcuc}. Non-@acronym{POSIX}. May be negated.
13425 (Used for terminals with uppercase characters only.)
13429 Same as @code{echoe echoctl echoke}.
13433 Same as @code{echoe echoctl echoke -ixany intr ^C erase ^? kill C-u}.
13438 @subsection Special characters
13440 @cindex special characters
13441 @cindex characters, special
13443 The special characters' default values vary from system to system.
13444 They are set with the syntax @samp{name value}, where the names are
13445 listed below and the value can be given either literally, in hat
13446 notation (@samp{^@var{c}}), or as an integer which may start with
13447 @samp{0x} to indicate hexadecimal, @samp{0} to indicate octal, or
13448 any other digit to indicate decimal.
13450 @cindex disabling special characters
13451 @kindex u@r{, and disabling special characters}
13452 For GNU stty, giving a value of @code{^-} or @code{undef} disables that
13453 special character. (This is incompatible with Ultrix @command{stty},
13454 which uses a value of @samp{u} to disable a special character. GNU
13455 @command{stty} treats a value @samp{u} like any other, namely to set that
13456 special character to @key{U}.)
13462 Send an interrupt signal.
13466 Send a quit signal.
13470 Erase the last character typed.
13474 Erase the current line.
13478 Send an end of file (terminate the input).
13486 Alternate character to end the line. Non-@acronym{POSIX}.
13490 Switch to a different shell layer. Non-@acronym{POSIX}.
13494 Restart the output after stopping it.
13502 Send a terminal stop signal.
13506 Send a terminal stop signal after flushing the input. Non-@acronym{POSIX}.
13510 Redraw the current line. Non-@acronym{POSIX}.
13514 Erase the last word typed. Non-@acronym{POSIX}.
13518 Enter the next character typed literally, even if it is a special
13519 character. Non-@acronym{POSIX}.
13524 @subsection Special settings
13526 @cindex special settings
13531 Set the minimum number of characters that will satisfy a read until
13532 the time value has expired, when @option{-icanon} is set.
13536 Set the number of tenths of a second before reads time out if the minimum
13537 number of characters have not been read, when @option{-icanon} is set.
13539 @item ispeed @var{n}
13541 Set the input speed to @var{n}.
13543 @item ospeed @var{n}
13545 Set the output speed to @var{n}.
13549 Tell the tty kernel driver that the terminal has @var{n} rows.
13550 Non-@acronym{POSIX}.
13553 @itemx columns @var{n}
13556 Tell the kernel that the terminal has @var{n} columns. Non-@acronym{POSIX}.
13562 Print the number of rows and columns that the kernel thinks the
13563 terminal has. (Systems that don't support rows and columns in the kernel
13564 typically use the environment variables @env{LINES} and @env{COLUMNS}
13565 instead; however, GNU @command{stty} does not know anything about them.)
13566 Non-@acronym{POSIX}.
13570 Use line discipline @var{n}. Non-@acronym{POSIX}.
13574 Print the terminal speed.
13577 @cindex baud rate, setting
13578 Set the input and output speeds to @var{n}. @var{n} can be one of: 0
13579 50 75 110 134 134.5 150 200 300 600 1200 1800 2400 4800 9600 19200
13580 38400 @code{exta} @code{extb}. @code{exta} is the same as 19200;
13581 @code{extb} is the same as 38400. Many systems, including GNU/Linux,
13582 support higher speeds. The @command{stty} command includes support
13599 4000000 where the system supports these.
13600 0 hangs up the line if @option{-clocal} is set.
13604 @node printenv invocation
13605 @section @command{printenv}: Print all or some environment variables
13608 @cindex printing all or some environment variables
13609 @cindex environment variables, printing
13611 @command{printenv} prints environment variable values. Synopsis:
13614 printenv [@var{option}] [@var{variable}]@dots{}
13617 If no @var{variable}s are specified, @command{printenv} prints the value of
13618 every environment variable. Otherwise, it prints the value of each
13619 @var{variable} that is set, and nothing for those that are not set.
13621 The program accepts the following option. Also see @ref{Common options}.
13629 @cindex exit status of @command{printenv}
13633 0 if all variables specified were found
13634 1 if at least one specified variable was not found
13635 2 if a write error occurred
13639 @node tty invocation
13640 @section @command{tty}: Print file name of terminal on standard input
13643 @cindex print terminal file name
13644 @cindex terminal file name, printing
13646 @command{tty} prints the file name of the terminal connected to its standard
13647 input. It prints @samp{not a tty} if standard input is not a terminal.
13651 tty [@var{option}]@dots{}
13654 The program accepts the following option. Also see @ref{Common options}.
13664 Print nothing; only return an exit status.
13668 @cindex exit status of @command{tty}
13672 0 if standard input is a terminal
13673 1 if standard input is not a terminal
13674 2 if given incorrect arguments
13675 3 if a write error occurs
13679 @node User information
13680 @chapter User information
13682 @cindex user information, commands for
13683 @cindex commands for printing user information
13685 This section describes commands that print user-related information:
13686 logins, groups, and so forth.
13689 * id invocation:: Print user identity.
13690 * logname invocation:: Print current login name.
13691 * whoami invocation:: Print effective user ID.
13692 * groups invocation:: Print group names a user is in.
13693 * users invocation:: Print login names of users currently logged in.
13694 * who invocation:: Print who is currently logged in.
13698 @node id invocation
13699 @section @command{id}: Print user identity
13702 @cindex real user and group IDs, printing
13703 @cindex effective user and group IDs, printing
13704 @cindex printing real and effective user and group IDs
13706 @command{id} prints information about the given user, or the process
13707 running it if no user is specified. Synopsis:
13710 id [@var{option}]@dots{} [@var{username}]
13713 @vindex POSIXLY_CORRECT
13714 By default, it prints the real user ID, real group ID, effective user ID
13715 if different from the real user ID, effective group ID if different from
13716 the real group ID, and supplemental group IDs.
13717 In addition, if SELinux
13718 is enabled and the @env{POSIXLY_CORRECT} environment variable is not set,
13719 then print @samp{context=@var{c}}, where @var{c} is the security context.
13721 Each of these numeric values is preceded by an identifying string and
13722 followed by the corresponding user or group name in parentheses.
13724 The options cause @command{id} to print only part of the above information.
13725 Also see @ref{Common options}.
13732 Print only the group ID.
13738 Print only the group ID and the supplementary groups.
13744 Print the user or group name instead of the ID number. Requires
13745 @option{-u}, @option{-g}, or @option{-G}.
13751 Print the real, instead of effective, user or group ID. Requires
13752 @option{-u}, @option{-g}, or @option{-G}.
13758 Print only the user ID.
13765 @cindex security context
13766 Print only the security context of the current user.
13767 If SELinux is disabled then print a warning and
13768 set the exit status to 1.
13774 @macro primaryAndSupplementaryGroups{cmd,arg}
13775 Primary and supplementary groups for a process are normally inherited
13776 from its parent and are usually unchanged since login. This means
13777 that if you change the group database after logging in, @command{\cmd\}
13778 will not reflect your changes within your existing login session.
13779 Running @command{\cmd\} with a \arg\ causes the user and group
13780 database to be consulted afresh, and so will give a different result.
13782 @primaryAndSupplementaryGroups{id,user argument}
13784 @node logname invocation
13785 @section @command{logname}: Print current login name
13788 @cindex printing user's login name
13789 @cindex login name, printing
13790 @cindex user name, printing
13793 @command{logname} prints the calling user's name, as found in a
13794 system-maintained file (often @file{/var/run/utmp} or
13795 @file{/etc/utmp}), and exits with a status of 0. If there is no entry
13796 for the calling process, @command{logname} prints
13797 an error message and exits with a status of 1.
13799 The only options are @option{--help} and @option{--version}. @xref{Common
13805 @node whoami invocation
13806 @section @command{whoami}: Print effective user ID
13809 @cindex effective user ID, printing
13810 @cindex printing the effective user ID
13812 @command{whoami} prints the user name associated with the current
13813 effective user ID. It is equivalent to the command @samp{id -un}.
13815 The only options are @option{--help} and @option{--version}. @xref{Common
13821 @node groups invocation
13822 @section @command{groups}: Print group names a user is in
13825 @cindex printing groups a user is in
13826 @cindex supplementary groups, printing
13828 @command{groups} prints the names of the primary and any supplementary
13829 groups for each given @var{username}, or the current process if no names
13830 are given. If more than one name is given, the name of each user is
13832 the list of that user's groups and the user name is separated from the
13833 group list by a colon. Synopsis:
13836 groups [@var{username}]@dots{}
13839 The group lists are equivalent to the output of the command @samp{id -Gn}.
13841 @primaryAndSupplementaryGroups{groups,list of users}
13843 The only options are @option{--help} and @option{--version}. @xref{Common
13849 @node users invocation
13850 @section @command{users}: Print login names of users currently logged in
13853 @cindex printing current usernames
13854 @cindex usernames, printing current
13856 @cindex login sessions, printing users with
13857 @command{users} prints on a single line a blank-separated list of user
13858 names of users currently logged in to the current host. Each user name
13859 corresponds to a login session, so if a user has more than one login
13860 session, that user's name will appear the same number of times in the
13869 With no @var{file} argument, @command{users} extracts its information from
13870 a system-maintained file (often @file{/var/run/utmp} or
13871 @file{/etc/utmp}). If a file argument is given, @command{users} uses
13872 that file instead. A common choice is @file{/var/log/wtmp}.
13874 The only options are @option{--help} and @option{--version}. @xref{Common
13880 @node who invocation
13881 @section @command{who}: Print who is currently logged in
13884 @cindex printing current user information
13885 @cindex information, about current users
13887 @command{who} prints information about users who are currently logged on.
13891 @command{who} [@var{option}] [@var{file}] [am i]
13894 @cindex terminal lines, currently used
13896 @cindex remote hostname
13897 If given no non-option arguments, @command{who} prints the following
13898 information for each user currently logged on: login name, terminal
13899 line, login time, and remote hostname or X display.
13903 If given one non-option argument, @command{who} uses that instead of
13904 a default system-maintained file (often @file{/var/run/utmp} or
13905 @file{/etc/utmp}) as the name of the file containing the record of
13906 users logged on. @file{/var/log/wtmp} is commonly given as an argument
13907 to @command{who} to look at who has previously logged on.
13911 If given two non-option arguments, @command{who} prints only the entry
13912 for the user running it (determined from its standard input), preceded
13913 by the hostname. Traditionally, the two arguments given are @samp{am
13914 i}, as in @samp{who am i}.
13917 Time stamps are listed according to the time zone rules specified by
13918 the @env{TZ} environment variable, or by the system default rules if
13919 @env{TZ} is not set. @xref{TZ Variable,, Specifying the Time Zone
13920 with @env{TZ}, libc, The GNU C Library Reference Manual}.
13922 The program accepts the following options. Also see @ref{Common options}.
13930 Same as @samp{-b -d --login -p -r -t -T -u}.
13936 Print the date and time of last system boot.
13942 Print information corresponding to dead processes.
13948 Print a line of column headings.
13954 List only the entries that correspond to processes via which the
13955 system is waiting for a user to login. The user name is always @samp{LOGIN}.
13959 Attempt to canonicalize hostnames found in utmp through a DNS lookup. This
13960 is not the default because it can cause significant delays on systems with
13961 automatic dial-up internet access.
13965 Same as @samp{who am i}.
13971 List active processes spawned by init.
13977 Print only the login names and the number of users logged on.
13978 Overrides all other options.
13983 @opindex --runlevel
13984 Print the current (and maybe previous) run-level of the init process.
13988 Ignored; for compatibility with other versions of @command{who}.
13994 Print last system clock change.
13999 After the login time, print the number of hours and minutes that the
14000 user has been idle. @samp{.} means the user was active in the last minute.
14001 @samp{old} means the user has been idle for more than 24 hours.
14012 @opindex --writable
14013 @cindex message status
14014 @pindex write@r{, allowed}
14015 After each login name print a character indicating the user's message status:
14018 @samp{+} allowing @code{write} messages
14019 @samp{-} disallowing @code{write} messages
14020 @samp{?} cannot find terminal device
14028 @node System context
14029 @chapter System context
14031 @cindex system context
14032 @cindex context, system
14033 @cindex commands for system context
14035 This section describes commands that print or change system-wide
14039 * date invocation:: Print or set system date and time.
14040 * arch invocation:: Print machine hardware name.
14041 * nproc invocation:: Print the number of processors.
14042 * uname invocation:: Print system information.
14043 * hostname invocation:: Print or set system name.
14044 * hostid invocation:: Print numeric host identifier.
14045 * uptime invocation:: Print system uptime and load.
14048 @node date invocation
14049 @section @command{date}: Print or set system date and time
14052 @cindex time, printing or setting
14053 @cindex printing the current time
14058 date [@var{option}]@dots{} [+@var{format}]
14059 date [-u|--utc|--universal] @c this avoids a newline in the output
14060 [ MMDDhhmm[[CC]YY][.ss] ]
14064 Invoking @command{date} with no @var{format} argument is equivalent to invoking
14065 it with a default format that depends on the @env{LC_TIME} locale category.
14066 In the default C locale, this format is @samp{'+%a %b %e %H:%M:%S %Z %Y'},
14067 so the output looks like @samp{Thu Mar @ 3 13:47:51 PST 2005}.
14070 Normally, @command{date} uses the time zone rules indicated by the
14071 @env{TZ} environment variable, or the system default rules if @env{TZ}
14072 is not set. @xref{TZ Variable,, Specifying the Time Zone with
14073 @env{TZ}, libc, The GNU C Library Reference Manual}.
14075 @findex strftime @r{and @command{date}}
14076 @cindex time formats
14077 @cindex formatting times
14078 If given an argument that starts with a @samp{+}, @command{date} prints the
14079 current date and time (or the date and time specified by the
14080 @option{--date} option, see below) in the format defined by that argument,
14081 which is similar to that of the @code{strftime} function. Except for
14082 conversion specifiers, which start with @samp{%}, characters in the
14083 format string are printed unchanged. The conversion specifiers are
14089 * Time conversion specifiers:: %[HIklMNpPrRsSTXzZ]
14090 * Date conversion specifiers:: %[aAbBcCdDeFgGhjmuUVwWxyY]
14091 * Literal conversion specifiers:: %[%nt]
14092 * Padding and other flags:: Pad with zeros, spaces, etc.
14093 * Setting the time:: Changing the system clock.
14094 * Options for date:: Instead of the current time.
14096 * Date input formats:: Specifying date strings.
14098 * Examples of date:: Examples.
14101 @node Time conversion specifiers
14102 @subsection Time conversion specifiers
14104 @cindex time conversion specifiers
14105 @cindex conversion specifiers, time
14107 @command{date} conversion specifiers related to times.
14111 hour (@samp{00}@dots{}@samp{23})
14113 hour (@samp{01}@dots{}@samp{12})
14115 hour, space padded (@samp{ 0}@dots{}@samp{23}); equivalent to @samp{%_H}.
14116 This is a @acronym{GNU} extension.
14118 hour, space padded (@samp{ 1}@dots{}@samp{12}); equivalent to @samp{%_I}.
14119 This is a @acronym{GNU} extension.
14121 minute (@samp{00}@dots{}@samp{59})
14123 nanoseconds (@samp{000000000}@dots{}@samp{999999999}).
14124 This is a @acronym{GNU} extension.
14126 locale's equivalent of either @samp{AM} or @samp{PM};
14127 blank in many locales.
14128 Noon is treated as @samp{PM} and midnight as @samp{AM}.
14130 like @samp{%p}, except lower case.
14131 This is a @acronym{GNU} extension.
14133 locale's 12-hour clock time (e.g., @samp{11:11:04 PM})
14135 24-hour hour and minute. Same as @samp{%H:%M}.
14137 @cindex epoch, seconds since
14138 @cindex seconds since the epoch
14139 @cindex beginning of time
14140 seconds since the epoch, i.e., since 1970-01-01 00:00:00 UTC.
14141 Leap seconds are not counted unless leap second support is available.
14142 @xref{%s-examples}, for examples.
14143 This is a @acronym{GNU} extension.
14145 second (@samp{00}@dots{}@samp{60}).
14146 This may be @samp{60} if leap seconds are supported.
14148 24-hour hour, minute, and second. Same as @samp{%H:%M:%S}.
14150 locale's time representation (e.g., @samp{23:13:48})
14152 @w{@acronym{RFC} 2822/@acronym{ISO} 8601} style numeric time zone
14153 (e.g., @samp{-0600} or @samp{+0530}), or nothing if no
14154 time zone is determinable. This value reflects the numeric time zone
14155 appropriate for the current time, using the time zone rules specified
14156 by the @env{TZ} environment variable.
14157 The time (and optionally, the time zone rules) can be overridden
14158 by the @option{--date} option.
14160 @w{@acronym{RFC} 3339/@acronym{ISO} 8601} style numeric time zone with
14161 @samp{:} (e.g., @samp{-06:00} or @samp{+05:30}), or nothing if no time
14162 zone is determinable.
14163 This is a @acronym{GNU} extension.
14165 Numeric time zone to the nearest second with @samp{:} (e.g.,
14166 @samp{-06:00:00} or @samp{+05:30:00}), or nothing if no time zone is
14168 This is a @acronym{GNU} extension.
14170 Numeric time zone with @samp{:} using the minimum necessary precision
14171 (e.g., @samp{-06}, @samp{+05:30}, or @samp{-04:56:02}), or nothing if
14172 no time zone is determinable.
14173 This is a @acronym{GNU} extension.
14175 alphabetic time zone abbreviation (e.g., @samp{EDT}), or nothing if no
14176 time zone is determinable. See @samp{%z} for how it is determined.
14180 @node Date conversion specifiers
14181 @subsection Date conversion specifiers
14183 @cindex date conversion specifiers
14184 @cindex conversion specifiers, date
14186 @command{date} conversion specifiers related to dates.
14190 locale's abbreviated weekday name (e.g., @samp{Sun})
14192 locale's full weekday name, variable length (e.g., @samp{Sunday})
14194 locale's abbreviated month name (e.g., @samp{Jan})
14196 locale's full month name, variable length (e.g., @samp{January})
14198 locale's date and time (e.g., @samp{Thu Mar @ 3 23:05:25 2005})
14200 century. This is like @samp{%Y}, except the last two digits are omitted.
14201 For example, it is @samp{20} if @samp{%Y} is @samp{2000},
14202 and is @samp{-0} if @samp{%Y} is @samp{-001}.
14203 It is normally at least two characters, but it may be more.
14205 day of month (e.g., @samp{01})
14207 date; same as @samp{%m/%d/%y}
14209 day of month, space padded; same as @samp{%_d}
14211 full date in @acronym{ISO} 8601 format; same as @samp{%Y-%m-%d}.
14212 This is a good choice for a date format, as it is standard and
14213 is easy to sort in the usual case where years are in the range
14216 year corresponding to the @acronym{ISO} week number, but without the century
14217 (range @samp{00} through @samp{99}). This has the same format and value
14218 as @samp{%y}, except that if the @acronym{ISO} week number (see
14220 to the previous or next year, that year is used instead.
14222 year corresponding to the @acronym{ISO} week number. This has the
14223 same format and value as @samp{%Y}, except that if the @acronym{ISO}
14225 @samp{%V}) belongs to the previous or next year, that year is used
14227 It is normally useful only if @samp{%V} is also used;
14228 for example, the format @samp{%G-%m-%d} is probably a mistake,
14229 since it combines the ISO week number year with the conventional month and day.
14233 day of year (@samp{001}@dots{}@samp{366})
14235 month (@samp{01}@dots{}@samp{12})
14237 day of week (@samp{1}@dots{}@samp{7}) with @samp{1} corresponding to Monday
14239 week number of year, with Sunday as the first day of the week
14240 (@samp{00}@dots{}@samp{53}).
14241 Days in a new year preceding the first Sunday are in week zero.
14243 @acronym{ISO} week number, that is, the
14244 week number of year, with Monday as the first day of the week
14245 (@samp{01}@dots{}@samp{53}).
14246 If the week containing January 1 has four or more days in
14247 the new year, then it is considered week 1; otherwise, it is week 53 of
14248 the previous year, and the next week is week 1. (See the @acronym{ISO} 8601
14251 day of week (@samp{0}@dots{}@samp{6}) with 0 corresponding to Sunday
14253 week number of year, with Monday as first day of week
14254 (@samp{00}@dots{}@samp{53}).
14255 Days in a new year preceding the first Monday are in week zero.
14257 locale's date representation (e.g., @samp{12/31/99})
14259 last two digits of year (@samp{00}@dots{}@samp{99})
14261 year. This is normally at least four characters, but it may be more.
14262 Year @samp{0000} precedes year @samp{0001}, and year @samp{-001}
14263 precedes year @samp{0000}.
14267 @node Literal conversion specifiers
14268 @subsection Literal conversion specifiers
14270 @cindex literal conversion specifiers
14271 @cindex conversion specifiers, literal
14273 @command{date} conversion specifiers that produce literal strings.
14285 @node Padding and other flags
14286 @subsection Padding and other flags
14288 @cindex numeric field padding
14289 @cindex padding of numeric fields
14290 @cindex fields, padding numeric
14292 Unless otherwise specified, @command{date} normally pads numeric fields
14293 with zeros, so that, for
14294 example, numeric months are always output as two digits.
14295 Seconds since the epoch are not padded, though,
14296 since there is no natural width for them.
14298 As a @acronym{GNU} extension, @command{date} recognizes any of the
14299 following optional flags after the @samp{%}:
14303 (hyphen) Do not pad the field; useful if the output is intended for
14306 (underscore) Pad with spaces; useful if you need a fixed
14307 number of characters in the output, but zeros are too distracting.
14309 (zero) Pad with zeros even if the conversion specifier
14310 would normally pad with spaces.
14312 Use upper case characters if possible.
14314 Use opposite case characters if possible.
14315 A field that is normally upper case becomes lower case, and vice versa.
14319 Here are some examples of padding:
14322 date +%d/%m -d "Feb 1"
14324 date +%-d/%-m -d "Feb 1"
14326 date +%_d/%_m -d "Feb 1"
14330 As a @acronym{GNU} extension, you can specify the field width
14331 (after any flag, if present) as a decimal number. If the natural size of the
14332 output of the field has less than the specified number of characters,
14333 the result is written right adjusted and padded to the given
14334 size. For example, @samp{%9B} prints the right adjusted month name in
14335 a field of width 9.
14337 An optional modifier can follow the optional flag and width
14338 specification. The modifiers are:
14342 Use the locale's alternate representation for date and time. This
14343 modifier applies to the @samp{%c}, @samp{%C}, @samp{%x}, @samp{%X},
14344 @samp{%y} and @samp{%Y} conversion specifiers. In a Japanese locale, for
14345 example, @samp{%Ex} might yield a date format based on the Japanese
14349 Use the locale's alternate numeric symbols for numbers. This modifier
14350 applies only to numeric conversion specifiers.
14353 If the format supports the modifier but no alternate representation
14354 is available, it is ignored.
14357 @node Setting the time
14358 @subsection Setting the time
14360 @cindex setting the time
14361 @cindex time setting
14362 @cindex appropriate privileges
14364 If given an argument that does not start with @samp{+}, @command{date} sets
14365 the system clock to the date and time specified by that argument (as
14366 described below). You must have appropriate privileges to set the
14367 system clock. Note for changes to persist across a reboot, the
14368 hardware clock may need to be updated from the system clock, which
14369 might not happen automatically on your system.
14371 The argument must consist entirely of digits, which have the following
14384 first two digits of year (optional)
14386 last two digits of year (optional)
14391 Note, the @option{--date} and @option{--set} options may not be used with an
14392 argument in the above format. The @option{--universal} option may be used
14393 with such an argument to indicate that the specified date and time are
14394 relative to Coordinated Universal Time rather than to the local time zone.
14397 @node Options for date
14398 @subsection Options for @command{date}
14400 @cindex @command{date} options
14401 @cindex options for @command{date}
14403 The program accepts the following options. Also see @ref{Common options}.
14407 @item -d @var{datestr}
14408 @itemx --date=@var{datestr}
14411 @cindex parsing date strings
14412 @cindex date strings, parsing
14413 @cindex arbitrary date strings, parsing
14416 @opindex next @var{day}
14417 @opindex last @var{day}
14418 Display the date and time specified in @var{datestr} instead of the
14419 current date and time. @var{datestr} can be in almost any common
14420 format. It can contain month names, time zones, @samp{am} and @samp{pm},
14421 @samp{yesterday}, etc. For example, @option{--date="2004-02-27
14422 14:19:13.489392193 +0530"} specifies the instant of time that is
14423 489,392,193 nanoseconds after February 27, 2004 at 2:19:13 PM in a
14424 time zone that is 5 hours and 30 minutes east of @acronym{UTC}.@*
14425 Note: input currently must be in locale independent format. E.g., the
14426 LC_TIME=C below is needed to print back the correct date in many locales:
14428 date -d "$(LC_TIME=C date)"
14430 @xref{Date input formats}.
14432 @item -f @var{datefile}
14433 @itemx --file=@var{datefile}
14436 Parse each line in @var{datefile} as with @option{-d} and display the
14437 resulting date and time. If @var{datefile} is @samp{-}, use standard
14438 input. This is useful when you have many dates to process, because the
14439 system overhead of starting up the @command{date} executable many times can
14442 @item -I[@var{timespec}]
14443 @itemx --iso-8601[=@var{timespec}]
14444 @opindex -I[@var{timespec}]
14445 @opindex --iso-8601[=@var{timespec}]
14446 Display the date using the @acronym{ISO} 8601 format, @samp{%Y-%m-%d}.
14448 The argument @var{timespec} specifies the number of additional
14449 terms of the time to include. It can be one of the following:
14452 Print just the date. This is the default if @var{timespec} is omitted.
14455 Append the hour of the day to the date.
14458 Append the hours and minutes.
14461 Append the hours, minutes and seconds.
14464 Append the hours, minutes, seconds and nanoseconds.
14467 If showing any time terms, then include the time zone using the format
14470 @item -r @var{file}
14471 @itemx --reference=@var{file}
14473 @opindex --reference
14474 Display the date and time of the last modification of @var{file},
14475 instead of the current date and time.
14482 @opindex --rfc-2822
14483 Display the date and time using the format @samp{%a, %d %b %Y %H:%M:%S
14484 %z}, evaluated in the C locale so abbreviations are always in English.
14488 Fri, 09 Sep 2005 13:51:39 -0700
14491 This format conforms to
14492 @uref{ftp://ftp.rfc-editor.org/in-notes/rfc2822.txt, Internet
14493 @acronym{RFCs} 2822} and
14494 @uref{ftp://ftp.rfc-editor.org/in-notes/rfc822.txt, 822}, the
14495 current and previous standards for Internet email.
14497 @item --rfc-3339=@var{timespec}
14498 @opindex --rfc-3339=@var{timespec}
14499 Display the date using a format specified by
14500 @uref{ftp://ftp.rfc-editor.org/in-notes/rfc3339.txt, Internet
14501 @acronym{RFC} 3339}. This is a subset of the @acronym{ISO} 8601
14502 format, except that it also permits applications to use a space rather
14503 than a @samp{T} to separate dates from times. Unlike the other
14504 standard formats, @acronym{RFC} 3339 format is always suitable as
14505 input for the @option{--date} (@option{-d}) and @option{--file}
14506 (@option{-f}) options, regardless of the current locale.
14508 The argument @var{timespec} specifies how much of the time to include.
14509 It can be one of the following:
14513 Print just the full-date, e.g., @samp{2005-09-14}.
14514 This is equivalent to the format @samp{%Y-%m-%d}.
14517 Print the full-date and full-time separated by a space, e.g.,
14518 @samp{2005-09-14 00:56:06+05:30}. The output ends with a numeric
14519 time-offset; here the @samp{+05:30} means that local time is five
14520 hours and thirty minutes east of @acronym{UTC}. This is equivalent to
14521 the format @samp{%Y-%m-%d %H:%M:%S%:z}.
14524 Like @samp{seconds}, but also print nanoseconds, e.g.,
14525 @samp{2005-09-14 00:56:06.998458565+05:30}.
14526 This is equivalent to the format @samp{%Y-%m-%d %H:%M:%S.%N%:z}.
14530 @item -s @var{datestr}
14531 @itemx --set=@var{datestr}
14534 Set the date and time to @var{datestr}. See @option{-d} above.
14535 See also @ref{Setting the time}.
14542 @opindex --universal
14543 @cindex Coordinated Universal Time
14545 @cindex Greenwich Mean Time
14548 Use Coordinated Universal Time (@acronym{UTC}) by operating as if the
14549 @env{TZ} environment variable were set to the string @samp{UTC0}.
14551 Universal Time is often called ``Greenwich Mean Time'' (@sc{gmt}) for
14552 historical reasons.
14556 @node Examples of date
14557 @subsection Examples of @command{date}
14559 @cindex examples of @command{date}
14561 Here are a few examples. Also see the documentation for the @option{-d}
14562 option in the previous section.
14567 To print the date of the day before yesterday:
14570 date --date='2 days ago'
14574 To print the date of the day three months and one day hence:
14577 date --date='3 months 1 day'
14581 To print the day of year of Christmas in the current year:
14584 date --date='25 Dec' +%j
14588 To print the current full month name and the day of the month:
14594 But this may not be what you want because for the first nine days of
14595 the month, the @samp{%d} expands to a zero-padded two-digit field,
14596 for example @samp{date -d 1may '+%B %d'} will print @samp{May 01}.
14599 To print a date without the leading zero for one-digit days
14600 of the month, you can use the (@acronym{GNU} extension)
14601 @samp{-} flag to suppress
14602 the padding altogether:
14605 date -d 1may '+%B %-d
14609 To print the current date and time in the format required by many
14610 non-@acronym{GNU} versions of @command{date} when setting the system clock:
14613 date +%m%d%H%M%Y.%S
14617 To set the system clock forward by two minutes:
14620 date --set='+2 minutes'
14624 To print the date in @acronym{RFC} 2822 format,
14625 use @samp{date --rfc-2822}. Here is some example output:
14628 Fri, 09 Sep 2005 13:51:39 -0700
14631 @anchor{%s-examples}
14633 To convert a date string to the number of seconds since the epoch
14634 (which is 1970-01-01 00:00:00 UTC), use the @option{--date} option with
14635 the @samp{%s} format. That can be useful in sorting and/or graphing
14636 and/or comparing data by date. The following command outputs the
14637 number of the seconds since the epoch for the time two minutes after the
14641 date --date='1970-01-01 00:02:00 +0000' +%s
14645 If you do not specify time zone information in the date string,
14646 @command{date} uses your computer's idea of the time zone when
14647 interpreting the string. For example, if your computer's time zone is
14648 that of Cambridge, Massachusetts, which was then 5 hours (i.e., 18,000
14649 seconds) behind UTC:
14652 # local time zone used
14653 date --date='1970-01-01 00:02:00' +%s
14658 If you're sorting or graphing dated data, your raw date values may be
14659 represented as seconds since the epoch. But few people can look at
14660 the date @samp{946684800} and casually note ``Oh, that's the first second
14661 of the year 2000 in Greenwich, England.''
14664 date --date='2000-01-01 UTC' +%s
14668 An alternative is to use the @option{--utc} (@option{-u}) option.
14669 Then you may omit @samp{UTC} from the date string. Although this
14670 produces the same result for @samp{%s} and many other format sequences,
14671 with a time zone offset different from zero, it would give a different
14672 result for zone-dependent formats like @samp{%z}.
14675 date -u --date=2000-01-01 +%s
14679 To convert such an unwieldy number of seconds back to
14680 a more readable form, use a command like this:
14683 # local time zone used
14684 date -d '1970-01-01 UTC 946684800 seconds' +"%Y-%m-%d %T %z"
14685 1999-12-31 19:00:00 -0500
14688 Or if you do not mind depending on the @samp{@@} feature present since
14689 coreutils 5.3.0, you could shorten this to:
14692 date -d @@946684800 +"%F %T %z"
14693 1999-12-31 19:00:00 -0500
14696 Often it is better to output UTC-relative date and time:
14699 date -u -d '1970-01-01 946684800 seconds' +"%Y-%m-%d %T %z"
14700 2000-01-01 00:00:00 +0000
14706 @node arch invocation
14707 @section @command{arch}: Print machine hardware name
14710 @cindex print machine hardware name
14711 @cindex system information, printing
14713 @command{arch} prints the machine hardware name,
14714 and is equivalent to @samp{uname -m}.
14718 arch [@var{option}]
14721 The program accepts the @ref{Common options} only.
14726 @node nproc invocation
14727 @section @command{nproc}: Print the number of available processors
14730 @cindex Print the number of processors
14731 @cindex system information, printing
14733 Print the number of processing units available to the current process,
14734 which may be less than the number of online processors.
14735 If this information is not accessible, then print the number of
14736 processors installed. If the @env{OMP_NUM_THREADS} environment variable is
14737 set, then it will determine the returned value. The result is guaranteed to be
14738 greater than zero. Synopsis:
14741 nproc [@var{option}]
14744 The program accepts the following options. Also see @ref{Common options}.
14750 Print the number of installed processors on the system, which may
14751 be greater than the number online or available to the current process.
14752 The @env{OMP_NUM_THREADS} environment variable is not honored in this case.
14754 @item --ignore=@var{number}
14756 If possible, exclude this @var{number} of processing units.
14763 @node uname invocation
14764 @section @command{uname}: Print system information
14767 @cindex print system information
14768 @cindex system information, printing
14770 @command{uname} prints information about the machine and operating system
14771 it is run on. If no options are given, @command{uname} acts as if the
14772 @option{-s} option were given. Synopsis:
14775 uname [@var{option}]@dots{}
14778 If multiple options or @option{-a} are given, the selected information is
14779 printed in this order:
14782 @var{kernel-name} @var{nodename} @var{kernel-release} @var{kernel-version}
14783 @var{machine} @var{processor} @var{hardware-platform} @var{operating-system}
14786 The information may contain internal spaces, so such output cannot be
14787 parsed reliably. In the following example, @var{release} is
14788 @samp{2.2.18ss.e820-bda652a #4 SMP Tue Jun 5 11:24:08 PDT 2001}:
14792 @result{} Linux dum 2.2.18 #4 SMP Tue Jun 5 11:24:08 PDT 2001 i686@c
14793 unknown unknown GNU/Linux
14797 The program accepts the following options. Also see @ref{Common options}.
14805 Print all of the below information, except omit the processor type
14806 and the hardware platform name if they are unknown.
14809 @itemx --hardware-platform
14811 @opindex --hardware-platform
14812 @cindex implementation, hardware
14813 @cindex hardware platform
14814 @cindex platform, hardware
14815 Print the hardware platform name
14816 (sometimes called the hardware implementation).
14817 Print @samp{unknown} if the kernel does not make this information
14818 easily available, as is the case with Linux kernels.
14824 @cindex machine type
14825 @cindex hardware class
14826 @cindex hardware type
14827 Print the machine hardware name (sometimes called the hardware class
14833 @opindex --nodename
14836 @cindex network node name
14837 Print the network node hostname.
14842 @opindex --processor
14843 @cindex host processor type
14844 Print the processor type (sometimes called the instruction set
14845 architecture or ISA).
14846 Print @samp{unknown} if the kernel does not make this information
14847 easily available, as is the case with Linux kernels.
14850 @itemx --operating-system
14852 @opindex --operating-system
14853 @cindex operating system name
14854 Print the name of the operating system.
14857 @itemx --kernel-release
14859 @opindex --kernel-release
14860 @cindex kernel release
14861 @cindex release of kernel
14862 Print the kernel release.
14865 @itemx --kernel-name
14867 @opindex --kernel-name
14868 @cindex kernel name
14869 @cindex name of kernel
14870 Print the kernel name.
14871 @acronym{POSIX} 1003.1-2001 (@pxref{Standards conformance}) calls this
14872 ``the implementation of the operating system'', because the
14873 @acronym{POSIX} specification itself has no notion of ``kernel''.
14874 The kernel name might be the same as the operating system name printed
14875 by the @option{-o} or @option{--operating-system} option, but it might
14876 differ. Some operating systems (e.g., FreeBSD, HP-UX) have the same
14877 name as their underlying kernels; others (e.g., GNU/Linux, Solaris)
14881 @itemx --kernel-version
14883 @opindex --kernel-version
14884 @cindex kernel version
14885 @cindex version of kernel
14886 Print the kernel version.
14893 @node hostname invocation
14894 @section @command{hostname}: Print or set system name
14897 @cindex setting the hostname
14898 @cindex printing the hostname
14899 @cindex system name, printing
14900 @cindex appropriate privileges
14902 With no arguments, @command{hostname} prints the name of the current host
14903 system. With one argument, it sets the current host name to the
14904 specified string. You must have appropriate privileges to set the host
14908 hostname [@var{name}]
14911 The only options are @option{--help} and @option{--version}. @xref{Common
14917 @node hostid invocation
14918 @section @command{hostid}: Print numeric host identifier
14921 @cindex printing the host identifier
14923 @command{hostid} prints the numeric identifier of the current host
14924 in hexadecimal. This command accepts no arguments.
14925 The only options are @option{--help} and @option{--version}.
14926 @xref{Common options}.
14928 For example, here's what it prints on one system I use:
14935 On that system, the 32-bit quantity happens to be closely
14936 related to the system's Internet address, but that isn't always
14941 @node uptime invocation
14942 @section @command{uptime}: Print system uptime and load
14945 @cindex printing the system uptime and load
14947 @command{uptime} prints the current time, the system's uptime, the
14948 number of logged-in users and the current load average.
14950 If an argument is specified, it is used as the file to be read
14951 to discover how many users are logged in. If no argument is
14952 specified, a system default is used (@command{uptime --help} indicates
14953 the default setting).
14955 The only options are @option{--help} and @option{--version}.
14956 @xref{Common options}.
14958 For example, here's what it prints right now on one system I use:
14962 14:07 up 3:35, 3 users, load average: 1.39, 1.15, 1.04
14965 The precise method of calculation of load average varies somewhat
14966 between systems. Some systems calculate it as the average number of
14967 runnable processes over the last 1, 5 and 15 minutes, but some systems
14968 also include processes in the uninterruptible sleep state (that is,
14969 those processes which are waiting for disk I/O). The Linux kernel
14970 includes uninterruptible processes.
14972 @node SELinux context
14973 @chapter SELinux context
14975 @cindex SELinux context
14976 @cindex SELinux, context
14977 @cindex commands for SELinux context
14979 This section describes commands for operations with SELinux
14983 * chcon invocation:: Change SELinux context of file
14984 * runcon invocation:: Run a command in specified SELinux context
14987 @node chcon invocation
14988 @section @command{chcon}: Change SELinux context of file
14991 @cindex changing security context
14992 @cindex change SELinux context
14994 @command{chcon} changes the SELinux security context of the selected files.
14998 chcon [@var{option}]@dots{} @var{context} @var{file}@dots{}
14999 chcon [@var{option}]@dots{} [-u @var{user}] [-r @var{role}] [-l @var{range}]@c
15000 [-t @var{type}] @var{file}@dots{}
15001 chcon [@var{option}]@dots{} --reference=@var{rfile} @var{file}@dots{}
15004 Change the SELinux security context of each @var{file} to @var{context}.
15005 With @option{--reference}, change the security context of each @var{file}
15006 to that of @var{rfile}.
15008 The program accepts the following options. Also see @ref{Common options}.
15013 @itemx --no-dereference
15015 @opindex --no-dereference
15016 @cindex no dereference
15017 Affect symbolic links instead of any referenced file.
15019 @item --reference=@var{rfile}
15020 @opindex --reference
15021 @cindex reference file
15022 Use @var{rfile}'s security context rather than specifying a @var{context} value.
15027 @opindex --recursive
15028 Operate on files and directories recursively.
15031 @xref{Traversing symlinks}.
15034 @xref{Traversing symlinks}.
15037 @xref{Traversing symlinks}.
15044 Output a diagnostic for every file processed.
15046 @item -u @var{user}
15047 @itemx --user=@var{user}
15050 Set user @var{user} in the target security context.
15052 @item -r @var{role}
15053 @itemx --role=@var{role}
15056 Set role @var{role} in the target security context.
15058 @item -t @var{type}
15059 @itemx --type=@var{type}
15062 Set type @var{type} in the target security context.
15064 @item -l @var{range}
15065 @itemx --range=@var{range}
15068 Set range @var{range} in the target security context.
15074 @node runcon invocation
15075 @section @command{runcon}: Run a command in specified SELinux context
15078 @cindex run with security context
15081 @command{runcon} runs file in specified SELinux security context.
15085 runcon @var{context} @var{command} [@var{args}]
15086 runcon [ -c ] [-u @var{user}] [-r @var{role}] [-t @var{type}]@c
15087 [-l @var{range}] @var{command} [@var{args}]
15090 Run @var{command} with completely-specified @var{context}, or with
15091 current or transitioned security context modified by one or more of @var{level},
15092 @var{role}, @var{type} and @var{user}.
15094 If none of @option{-c}, @option{-t}, @option{-u}, @option{-r}, or @option{-l}
15095 is specified, the first argument is used as the complete context.
15096 Any additional arguments after @var{command}
15097 are interpreted as arguments to the command.
15099 With neither @var{context} nor @var{command}, print the current
15102 The program accepts the following options. Also see @ref{Common options}.
15110 Compute process transition context before modifying.
15112 @item -u @var{user}
15113 @itemx --user=@var{user}
15116 Set user @var{user} in the target security context.
15118 @item -r @var{role}
15119 @itemx --role=@var{role}
15122 Set role @var{role} in the target security context.
15124 @item -t @var{type}
15125 @itemx --type=@var{type}
15128 Set type @var{type} in the target security context.
15130 @item -l @var{range}
15131 @itemx --range=@var{range}
15134 Set range @var{range} in the target security context.
15138 @cindex exit status of @command{runcon}
15142 126 if @var{command} is found but cannot be invoked
15143 127 if @command{runcon} itself fails or if @var{command} cannot be found
15144 the exit status of @var{command} otherwise
15147 @node Modified command invocation
15148 @chapter Modified command invocation
15150 @cindex modified command invocation
15151 @cindex invocation of commands, modified
15152 @cindex commands for invoking other commands
15154 This section describes commands that run other commands in some context
15155 different than the current one: a modified environment, as a different
15159 * chroot invocation:: Modify the root directory.
15160 * env invocation:: Modify environment variables.
15161 * nice invocation:: Modify niceness.
15162 * nohup invocation:: Immunize to hangups.
15163 * stdbuf invocation:: Modify buffering of standard streams.
15164 * su invocation:: Modify user and group ID.
15165 * timeout invocation:: Run with time limit.
15169 @node chroot invocation
15170 @section @command{chroot}: Run a command with a different root directory
15173 @cindex running a program in a specified root directory
15174 @cindex root directory, running a program in a specified
15176 @command{chroot} runs a command with a specified root directory.
15177 On many systems, only the super-user can do this.@footnote{However,
15178 some systems (e.g., FreeBSD) can be configured to allow certain regular
15179 users to use the @code{chroot} system call, and hence to run this program.
15180 Also, on Cygwin, anyone can run the @command{chroot} command, because the
15181 underlying function is non-privileged due to lack of support in MS-Windows.}
15185 chroot @var{option} @var{newroot} [@var{command} [@var{args}]@dots{}]
15186 chroot @var{option}
15189 Ordinarily, file names are looked up starting at the root of the
15190 directory structure, i.e., @file{/}. @command{chroot} changes the root to
15191 the directory @var{newroot} (which must exist) and then runs
15192 @var{command} with optional @var{args}. If @var{command} is not
15193 specified, the default is the value of the @env{SHELL} environment
15194 variable or @command{/bin/sh} if not set, invoked with the @option{-i} option.
15195 @var{command} must not be a special built-in utility
15196 (@pxref{Special built-in utilities}).
15198 The program accepts the following options. Also see @ref{Common options}.
15199 Options must precede operands.
15203 @itemx --userspec=@var{user}[:@var{group}]
15204 @opindex --userspec
15205 By default, @var{command} is run with the same credentials
15206 as the invoking process.
15207 Use this option to run it as a different @var{user} and/or with a
15208 different primary @var{group}.
15210 @itemx --groups=@var{groups}
15212 Use this option to specify the supplementary @var{groups} to be
15213 used by the new process.
15214 The items in the list (names or numeric IDs) must be separated by commas.
15218 Here are a few tips to help avoid common problems in using chroot.
15219 To start with a simple example, make @var{command} refer to a statically
15220 linked binary. If you were to use a dynamically linked executable, then
15221 you'd have to arrange to have the shared libraries in the right place under
15222 your new root directory.
15224 For example, if you create a statically linked @command{ls} executable,
15225 and put it in @file{/tmp/empty}, you can run this command as root:
15228 $ chroot /tmp/empty /ls -Rl /
15231 Then you'll see output like this:
15236 -rwxr-xr-x 1 0 0 1041745 Aug 16 11:17 ls
15239 If you want to use a dynamically linked executable, say @command{bash},
15240 then first run @samp{ldd bash} to see what shared objects it needs.
15241 Then, in addition to copying the actual binary, also copy the listed
15242 files to the required positions under your intended new root directory.
15243 Finally, if the executable requires any other files (e.g., data, state,
15244 device files), copy them into place, too.
15246 @cindex exit status of @command{chroot}
15250 125 if @command{chroot} itself fails
15251 126 if @var{command} is found but cannot be invoked
15252 127 if @var{command} cannot be found
15253 the exit status of @var{command} otherwise
15257 @node env invocation
15258 @section @command{env}: Run a command in a modified environment
15261 @cindex environment, running a program in a modified
15262 @cindex modified environment, running a program in a
15263 @cindex running a program in a modified environment
15265 @command{env} runs a command with a modified environment. Synopses:
15268 env [@var{option}]@dots{} [@var{name}=@var{value}]@dots{} @c
15269 [@var{command} [@var{args}]@dots{}]
15273 Operands of the form @samp{@var{variable}=@var{value}} set
15274 the environment variable @var{variable} to value @var{value}.
15275 @var{value} may be empty (@samp{@var{variable}=}). Setting a variable
15276 to an empty value is different from unsetting it.
15277 These operands are evaluated left-to-right, so if two operands
15278 mention the same variable the earlier is ignored.
15280 Environment variable names can be empty, and can contain any
15281 characters other than @samp{=} and @acronym{ASCII} @sc{nul}.
15282 However, it is wise to limit yourself to names that
15283 consist solely of underscores, digits, and @acronym{ASCII} letters,
15284 and that begin with a non-digit, as applications like the shell do not
15285 work well with other names.
15288 The first operand that does not contain the character @samp{=}
15289 specifies the program to invoke; it is
15290 searched for according to the @env{PATH} environment variable. Any
15291 remaining arguments are passed as arguments to that program.
15292 The program should not be a special built-in utility
15293 (@pxref{Special built-in utilities}).
15295 Modifications to @env{PATH} take effect prior to searching for
15296 @var{command}. Use caution when reducing @env{PATH}; behavior is
15297 not portable when @env{PATH} is undefined or omits key directories
15298 such as @file{/bin}.
15300 In the rare case that a utility contains a @samp{=} in the name, the
15301 only way to disambiguate it from a variable assignment is to use an
15302 intermediate command for @var{command}, and pass the problematic
15303 program name via @var{args}. For example, if @file{./prog=} is an
15304 executable in the current @env{PATH}:
15307 env prog= true # runs 'true', with prog= in environment
15308 env ./prog= true # runs 'true', with ./prog= in environment
15309 env -- prog= true # runs 'true', with prog= in environment
15310 env sh -c '\prog= true' # runs 'prog=' with argument 'true'
15311 env sh -c 'exec "$@@"' sh prog= true # also runs 'prog='
15314 @cindex environment, printing
15316 If no command name is specified following the environment
15317 specifications, the resulting environment is printed. This is like
15318 specifying the @command{printenv} program.
15320 For some examples, suppose the environment passed to @command{env}
15321 contains @samp{LOGNAME=rms}, @samp{EDITOR=emacs}, and
15322 @samp{PATH=.:/gnubin:/hacks}:
15327 Output the current environment.
15329 $ env | LC_ALL=C sort
15332 PATH=.:/gnubin:/hacks
15336 Run @command{foo} with a reduced environment, preserving only the
15337 original @env{PATH} to avoid problems in locating @command{foo}.
15339 env - PATH="$PATH" foo
15343 Run @command{foo} with the environment containing @samp{LOGNAME=rms},
15344 @samp{EDITOR=emacs}, and @samp{PATH=.:/gnubin:/hacks}, and guarantees
15345 that @command{foo} was found in the file system rather than as a shell
15352 Run @command{nemacs} with the environment containing @samp{LOGNAME=foo},
15353 @samp{EDITOR=emacs}, @samp{PATH=.:/gnubin:/hacks}, and
15354 @samp{DISPLAY=gnu:0}.
15356 env DISPLAY=gnu:0 LOGNAME=foo nemacs
15360 Attempt to run the program @command{/energy/--} (as that is the only
15361 possible path search result); if the command exists, the environment
15362 will contain @samp{LOGNAME=rms} and @samp{PATH=/energy}, and the
15363 arguments will be @samp{e=mc2}, @samp{bar}, and @samp{baz}.
15365 env -u EDITOR PATH=/energy -- e=mc2 bar baz
15371 The program accepts the following options. Also see @ref{Common options}.
15372 Options must precede operands.
15378 @item -u @var{name}
15379 @itemx --unset=@var{name}
15382 Remove variable @var{name} from the environment, if it was in the
15387 @itemx --ignore-environment
15390 @opindex --ignore-environment
15391 Start with an empty environment, ignoring the inherited environment.
15395 @cindex exit status of @command{env}
15399 0 if no @var{command} is specified and the environment is output
15400 125 if @command{env} itself fails
15401 126 if @var{command} is found but cannot be invoked
15402 127 if @var{command} cannot be found
15403 the exit status of @var{command} otherwise
15407 @node nice invocation
15408 @section @command{nice}: Run a command with modified niceness
15412 @cindex scheduling, affecting
15413 @cindex appropriate privileges
15415 @command{nice} prints or modifies a process's @dfn{niceness},
15416 a parameter that affects whether the process is scheduled favorably.
15420 nice [@var{option}]@dots{} [@var{command} [@var{arg}]@dots{}]
15423 If no arguments are given, @command{nice} prints the current niceness.
15424 Otherwise, @command{nice} runs the given @var{command} with its
15425 niceness adjusted. By default, its niceness is incremented by 10.
15427 Niceness values range at least from @minus{}20 (process has high priority
15428 and gets more resources, thus slowing down other processes) through 19
15429 (process has lower priority and runs slowly itself, but has less impact
15430 on the speed of other running processes). Some systems
15431 may have a wider range of nicenesses; conversely, other systems may
15432 enforce more restrictive limits. An attempt to set the niceness
15433 outside the supported range is treated as an attempt to use the
15434 minimum or maximum supported value.
15436 A niceness should not be confused with a scheduling priority, which
15437 lets applications determine the order in which threads are scheduled
15438 to run. Unlike a priority, a niceness is merely advice to the
15439 scheduler, which the scheduler is free to ignore. Also, as a point of
15440 terminology, @acronym{POSIX} defines the behavior of @command{nice} in
15441 terms of a @dfn{nice value}, which is the nonnegative difference
15442 between a niceness and the minimum niceness. Though @command{nice}
15443 conforms to @acronym{POSIX}, its documentation and diagnostics use the
15444 term ``niceness'' for compatibility with historical practice.
15446 @var{command} must not be a special built-in utility (@pxref{Special
15447 built-in utilities}).
15449 @mayConflictWithShellBuiltIn{nice}
15451 The program accepts the following option. Also see @ref{Common options}.
15452 Options must precede operands.
15455 @item -n @var{adjustment}
15456 @itemx --adjustment=@var{adjustment}
15458 @opindex --adjustment
15459 Add @var{adjustment} instead of 10 to the command's niceness. If
15460 @var{adjustment} is negative and you lack appropriate privileges,
15461 @command{nice} issues a warning but otherwise acts as if you specified
15464 For compatibility @command{nice} also supports an obsolete
15465 option syntax @option{-@var{adjustment}}. New scripts should use
15466 @option{-n @var{adjustment}} instead.
15470 @cindex exit status of @command{nice}
15474 0 if no @var{command} is specified and the niceness is output
15475 125 if @command{nice} itself fails
15476 126 if @var{command} is found but cannot be invoked
15477 127 if @var{command} cannot be found
15478 the exit status of @var{command} otherwise
15481 It is sometimes useful to run a non-interactive program with reduced niceness.
15484 $ nice factor 4611686018427387903
15487 Since @command{nice} prints the current niceness,
15488 you can invoke it through itself to demonstrate how it works.
15490 The default behavior is to increase the niceness by @samp{10}:
15501 The @var{adjustment} is relative to the current niceness. In the
15502 next example, the first @command{nice} invocation runs the second one
15503 with niceness 10, and it in turn runs the final one with a niceness
15507 $ nice nice -n 3 nice
15511 Specifying a niceness larger than the supported range
15512 is the same as specifying the maximum supported value:
15515 $ nice -n 10000000000 nice
15519 Only a privileged user may run a process with lower niceness:
15523 nice: cannot set niceness: Permission denied
15525 $ sudo nice -n -1 nice
15530 @node nohup invocation
15531 @section @command{nohup}: Run a command immune to hangups
15534 @cindex hangups, immunity to
15535 @cindex immunity to hangups
15536 @cindex logging out and continuing to run
15539 @command{nohup} runs the given @var{command} with hangup signals ignored,
15540 so that the command can continue running in the background after you log
15544 nohup @var{command} [@var{arg}]@dots{}
15547 If standard input is a terminal, it is redirected from
15548 @file{/dev/null} so that terminal sessions do not mistakenly consider
15549 the terminal to be used by the command. This is a @acronym{GNU}
15550 extension; programs intended to be portable to non-@acronym{GNU} hosts
15551 should use @samp{nohup @var{command} [@var{arg}]@dots{} </dev/null}
15555 If standard output is a terminal, the command's standard output is appended
15556 to the file @file{nohup.out}; if that cannot be written to, it is appended
15557 to the file @file{$HOME/nohup.out}; and if that cannot be written to, the
15558 command is not run.
15559 Any @file{nohup.out} or @file{$HOME/nohup.out} file created by
15560 @command{nohup} is made readable and writable only to the user,
15561 regardless of the current umask settings.
15563 If standard error is a terminal, it is normally redirected to the same file
15564 descriptor as the (possibly-redirected) standard output.
15565 However, if standard output is closed, standard error terminal output
15566 is instead appended to the file @file{nohup.out} or
15567 @file{$HOME/nohup.out} as above.
15569 To capture the command's output to a file other than @file{nohup.out}
15570 you can redirect it. For example, to capture the output of
15574 nohup make > make.log
15577 @command{nohup} does not automatically put the command it runs in the
15578 background; you must do that explicitly, by ending the command line
15579 with an @samp{&}. Also, @command{nohup} does not alter the
15580 niceness of @var{command}; use @command{nice} for that,
15581 e.g., @samp{nohup nice @var{command}}.
15583 @var{command} must not be a special built-in utility (@pxref{Special
15584 built-in utilities}).
15586 The only options are @option{--help} and @option{--version}. @xref{Common
15587 options}. Options must precede operands.
15589 @cindex exit status of @command{nohup}
15593 125 if @command{nohup} itself fails, and @env{POSIXLY_CORRECT} is not set
15594 126 if @var{command} is found but cannot be invoked
15595 127 if @var{command} cannot be found
15596 the exit status of @var{command} otherwise
15599 If @env{POSIXLY_CORRECT} is set, internal failures give status 127
15603 @node stdbuf invocation
15604 @section @command{stdbuf}: Run a command with modified I/O stream buffering
15607 @cindex standard streams, buffering
15608 @cindex line buffered
15610 @command{stdbuf} allows one to modify the buffering operations of the
15611 three standard I/O streams associated with a program. Synopsis:
15614 stdbuf @var{option}@dots{} @var{command}
15617 @var{command} must start with the name of a program that
15620 uses the ISO C @code{FILE} streams for input/output (note the
15621 programs @command{dd} and @command{cat} don't do that),
15624 does not adjust the buffering of its standard streams (note the
15625 program @command{tee} is not in this category).
15628 Any additional @var{arg}s are passed as additional arguments to the
15631 The program accepts the following options. Also see @ref{Common options}.
15635 @item -i @var{mode}
15636 @itemx --input=@var{mode}
15639 Adjust the standard input stream buffering.
15641 @item -o @var{mode}
15642 @itemx --output=@var{mode}
15645 Adjust the standard output stream buffering.
15647 @item -e @var{mode}
15648 @itemx --error=@var{mode}
15651 Adjust the standard error stream buffering.
15655 The @var{mode} can be specified as follows:
15660 Set the stream to line buffered mode.
15661 In this mode data is coalesced until a newline is output or
15662 input is read from any stream attached to a terminal device.
15663 This option is invalid with standard input.
15666 Disable buffering of the selected stream.
15667 In this mode, data is output immediately and only the
15668 amount of data requested is read from input.
15669 Note the difference in function for input and output.
15670 Disabling buffering for input will not influence the responsiveness
15671 or blocking behavior of the stream input functions.
15672 For example @code{fread} will still block until @code{EOF} or error,
15673 even if the underlying @code{read} returns less data than requested.
15676 Specify the size of the buffer to use in fully buffered mode.
15677 @multiplierSuffixesNoBlocks{size}
15681 @cindex exit status of @command{stdbuf}
15685 125 if @command{stdbuf} itself fails
15686 126 if @var{command} is found but cannot be invoked
15687 127 if @var{command} cannot be found
15688 the exit status of @var{command} otherwise
15692 @node su invocation
15693 @section @command{su}: Run a command with substitute user and group ID
15696 @cindex substitute user and group IDs
15697 @cindex user ID, switching
15698 @cindex super-user, becoming
15699 @cindex root, becoming
15701 @command{su} allows one user to temporarily become another user. It runs a
15702 command (often an interactive shell) with the real and effective user
15703 ID, group ID, and supplemental groups of a given @var{user}. Synopsis:
15706 su [@var{option}]@dots{} [@var{user} [@var{arg}]@dots{}]
15709 @cindex passwd entry, and @command{su} shell
15711 @flindex /etc/passwd
15712 If no @var{user} is given, the default is @code{root}, the super-user.
15713 The shell to use is taken from @var{user}'s @code{passwd} entry, or
15714 @file{/bin/sh} if none is specified there. If @var{user} has a
15715 password, @command{su} prompts for the password unless run by a user with
15716 effective user ID of zero (the super-user).
15722 @cindex login shell
15723 By default, @command{su} does not change the current directory.
15724 It sets the environment variables @env{HOME} and @env{SHELL}
15725 from the password entry for @var{user}, and if @var{user} is not
15726 the super-user, sets @env{USER} and @env{LOGNAME} to @var{user}.
15727 By default, the shell is not a login shell.
15729 Any additional @var{arg}s are passed as additional arguments to the
15732 @cindex @option{-su}
15733 GNU @command{su} does not treat @file{/bin/sh} or any other shells specially
15734 (e.g., by setting @code{argv[0]} to @option{-su}, passing @option{-c} only
15735 to certain shells, etc.).
15738 @command{su} can optionally be compiled to use @code{syslog} to report
15739 failed, and optionally successful, @command{su} attempts. (If the system
15740 supports @code{syslog}.) However, GNU @command{su} does not check if the
15741 user is a member of the @code{wheel} group; see below.
15743 The program accepts the following options. Also see @ref{Common options}.
15746 @item -c @var{command}
15747 @itemx --command=@var{command}
15750 Pass @var{command}, a single command line to run, to the shell with
15751 a @option{-c} option instead of starting an interactive shell.
15758 @cindex file name pattern expansion, disabled
15759 @cindex globbing, disabled
15760 Pass the @option{-f} option to the shell. This probably only makes sense
15761 if the shell run is @command{csh} or @command{tcsh}, for which the @option{-f}
15762 option prevents reading the startup file (@file{.cshrc}). With
15763 Bourne-like shells, the @option{-f} option disables file name pattern
15764 expansion (globbing), which is not likely to be useful.
15772 @c other variables already indexed above
15775 @cindex login shell, creating
15776 Make the shell a login shell. This means the following. Unset all
15777 environment variables except @env{TERM}, @env{HOME}, and @env{SHELL}
15778 (which are set as described above), and @env{USER} and @env{LOGNAME}
15779 (which are set, even for the super-user, as described above), and set
15780 @env{PATH} to a compiled-in default value. Change to @var{user}'s home
15781 directory. Prepend @samp{-} to the shell's name, intended to make it
15782 read its login startup file(s).
15786 @itemx --preserve-environment
15789 @opindex --preserve-environment
15790 @cindex environment, preserving
15791 @flindex /etc/shells
15792 @cindex restricted shell
15793 Do not change the environment variables @env{HOME}, @env{USER},
15794 @env{LOGNAME}, or @env{SHELL}. Run the shell given in the environment
15795 variable @env{SHELL} instead of the shell from @var{user}'s passwd
15796 entry, unless the user running @command{su} is not the super-user and
15797 @var{user}'s shell is restricted. A @dfn{restricted shell} is one that
15798 is not listed in the file @file{/etc/shells}, or in a compiled-in list
15799 if that file does not exist. Parts of what this option does can be
15800 overridden by @option{--login} and @option{--shell}.
15802 @item -s @var{shell}
15803 @itemx --shell=@var{shell}
15806 Run @var{shell} instead of the shell from @var{user}'s passwd entry,
15807 unless the user running @command{su} is not the super-user and @var{user}'s
15808 shell is restricted (see @option{-m} just above).
15812 @cindex exit status of @command{su}
15816 125 if @command{su} itself fails
15817 126 if subshell is found but cannot be invoked
15818 127 if subshell cannot be found
15819 the exit status of the subshell otherwise
15822 @cindex wheel group, not supported
15823 @cindex group wheel, not supported
15825 @subsection Why GNU @command{su} does not support the @samp{wheel} group
15827 (This section is by Richard Stallman.)
15831 Sometimes a few of the users try to hold total power over all the
15832 rest. For example, in 1984, a few users at the MIT AI lab decided to
15833 seize power by changing the operator password on the Twenex system and
15834 keeping it secret from everyone else. (I was able to thwart this coup
15835 and give power back to the users by patching the kernel, but I
15836 wouldn't know how to do that in Unix.)
15838 However, occasionally the rulers do tell someone. Under the usual
15839 @command{su} mechanism, once someone learns the root password who
15840 sympathizes with the ordinary users, he or she can tell the rest. The
15841 ``wheel group'' feature would make this impossible, and thus cement the
15842 power of the rulers.
15844 I'm on the side of the masses, not that of the rulers. If you are
15845 used to supporting the bosses and sysadmins in whatever they do, you
15846 might find this idea strange at first.
15849 @node timeout invocation
15850 @section @command{timeout}: Run a command with a time limit
15854 @cindex run commands with bounded time
15856 @command{timeout} runs the given @var{command} and kills it if it is
15857 still running after the specified time interval. Synopsis:
15860 timeout [@var{option}] @var{duration} @var{command} [@var{arg}]@dots{}
15863 @var{command} must not be a special built-in utility (@pxref{Special
15864 built-in utilities}).
15866 The program accepts the following options. Also see @ref{Common options}.
15867 Options must precede operands.
15870 @itemx --foreground
15871 @opindex --foreground
15872 Don't create a separate background program group, so that
15873 the managed @var{command} can use the foreground TTY normally.
15874 This is needed to support timing out commands not started
15875 directly from an interactive shell, in two situations.
15878 @var{command} is interactive and needs to read from the terminal for example
15880 the user wants to support sending signals directly to @var{command}
15881 from the terminal (like Ctrl-C for example)
15884 Note in this mode of operation, any children of @var{command}
15885 will not be timed out.
15887 @item -k @var{duration}
15888 @itemx --kill-after=@var{duration}
15890 @opindex --kill-after
15891 Ensure the monitored @var{command} is killed by also sending a @samp{KILL}
15892 signal, after the specified @var{duration}. Without this option, if the
15893 selected signal proves not to be fatal, @command{timeout} does not kill
15896 @item -s @var{signal}
15897 @itemx --signal=@var{signal}
15900 Send this @var{signal} to @var{command} on timeout, rather than the
15901 default @samp{TERM} signal. @var{signal} may be a name like @samp{HUP}
15902 or a number. Also see @xref{Signal specifications}.
15906 @var{duration} is a floating point number followed by an optional unit:
15908 @samp{s} for seconds (the default)
15909 @samp{m} for minutes
15913 A duration of 0 disables the associated timeout.
15914 Note that the actual timeout duration is dependent on system conditions,
15915 which should be especially considered when specifying sub-second timeouts.
15917 @cindex exit status of @command{timeout}
15921 124 if @var{command} times out
15922 125 if @command{timeout} itself fails
15923 126 if @var{command} is found but cannot be invoked
15924 127 if @var{command} cannot be found
15925 the exit status of @var{command} otherwise
15929 @node Process control
15930 @chapter Process control
15932 @cindex processes, commands for controlling
15933 @cindex commands for controlling processes
15936 * kill invocation:: Sending a signal to processes.
15940 @node kill invocation
15941 @section @command{kill}: Send a signal to processes
15944 @cindex send a signal to processes
15946 The @command{kill} command sends a signal to processes, causing them
15947 to terminate or otherwise act upon receiving the signal in some way.
15948 Alternatively, it lists information about signals. Synopses:
15951 kill [-s @var{signal} | --signal @var{signal} | -@var{signal}] @var{pid}@dots{}
15952 kill [-l | --list | -t | --table] [@var{signal}]@dots{}
15955 @mayConflictWithShellBuiltIn{kill}
15957 The first form of the @command{kill} command sends a signal to all
15958 @var{pid} arguments. The default signal to send if none is specified
15959 is @samp{TERM}. The special signal number @samp{0} does not denote a
15960 valid signal, but can be used to test whether the @var{pid} arguments
15961 specify processes to which a signal could be sent.
15963 If @var{pid} is positive, the signal is sent to the process with the
15964 process ID @var{pid}. If @var{pid} is zero, the signal is sent to all
15965 processes in the process group of the current process. If @var{pid}
15966 is @minus{}1, the signal is sent to all processes for which the user has
15967 permission to send a signal. If @var{pid} is less than @minus{}1, the signal
15968 is sent to all processes in the process group that equals the absolute
15969 value of @var{pid}.
15971 If @var{pid} is not positive, a system-dependent set of system
15972 processes is excluded from the list of processes to which the signal
15975 If a negative @var{pid} argument is desired as the first one, it
15976 should be preceded by @option{--}. However, as a common extension to
15977 @acronym{POSIX}, @option{--} is not required with @samp{kill
15978 -@var{signal} -@var{pid}}. The following commands are equivalent:
15987 The first form of the @command{kill} command succeeds if every @var{pid}
15988 argument specifies at least one process that the signal was sent to.
15990 The second form of the @command{kill} command lists signal information.
15991 Either the @option{-l} or @option{--list} option, or the @option{-t}
15992 or @option{--table} option must be specified. Without any
15993 @var{signal} argument, all supported signals are listed. The output
15994 of @option{-l} or @option{--list} is a list of the signal names, one
15995 per line; if @var{signal} is already a name, the signal number is
15996 printed instead. The output of @option{-t} or @option{--table} is a
15997 table of signal numbers, names, and descriptions. This form of the
15998 @command{kill} command succeeds if all @var{signal} arguments are valid
15999 and if there is no output error.
16001 The @command{kill} command also supports the @option{--help} and
16002 @option{--version} options. @xref{Common options}.
16004 A @var{signal} may be a signal name like @samp{HUP}, or a signal
16005 number like @samp{1}, or an exit status of a process terminated by the
16006 signal. A signal name can be given in canonical form or prefixed by
16007 @samp{SIG}. The case of the letters is ignored, except for the
16008 @option{-@var{signal}} option which must use upper case to avoid
16009 ambiguity with lower case option letters. For a list of supported
16010 signal names and numbers see @xref{Signal specifications}.
16015 @cindex delaying commands
16016 @cindex commands for delaying
16018 @c Perhaps @command{wait} or other commands should be described here also?
16021 * sleep invocation:: Delay for a specified time.
16025 @node sleep invocation
16026 @section @command{sleep}: Delay for a specified time
16029 @cindex delay for a specified time
16031 @command{sleep} pauses for an amount of time specified by the sum of
16032 the values of the command line arguments.
16036 sleep @var{number}[smhd]@dots{}
16040 Each argument is a number followed by an optional unit; the default
16041 is seconds. The units are:
16054 Historical implementations of @command{sleep} have required that
16055 @var{number} be an integer, and only accepted a single argument
16056 without a suffix. However, GNU @command{sleep} accepts
16057 arbitrary floating point numbers. @xref{Floating point}.
16059 The only options are @option{--help} and @option{--version}. @xref{Common
16062 @c sleep is a shell built-in at least with Solaris 11's /bin/sh
16063 @mayConflictWithShellBuiltIn{sleep}
16068 @node Numeric operations
16069 @chapter Numeric operations
16071 @cindex numeric operations
16072 These programs do numerically-related operations.
16075 * factor invocation:: Show factors of numbers.
16076 * seq invocation:: Print sequences of numbers.
16080 @node factor invocation
16081 @section @command{factor}: Print prime factors
16084 @cindex prime factors
16086 @command{factor} prints prime factors. Synopses:
16089 factor [@var{number}]@dots{}
16090 factor @var{option}
16093 If no @var{number} is specified on the command line, @command{factor} reads
16094 numbers from standard input, delimited by newlines, tabs, or spaces.
16096 The @command{factor} command supports only a small number of options:
16100 Print a short help on standard output, then exit without further
16104 Print the program version on standard output, then exit without further
16108 Factoring the product of the eighth and ninth Mersenne primes
16109 takes about 30 milliseconds of CPU time on a 2.2 GHz Athlon.
16112 M8=$(echo 2^31-1|bc)
16113 M9=$(echo 2^61-1|bc)
16114 n=$(echo "$M8 * $M9" | bc)
16115 /usr/bin/time -f %U factor $n
16116 4951760154835678088235319297: 2147483647 2305843009213693951
16120 Similarly, factoring the eighth Fermat number @math{2^{256}+1} takes
16121 about 20 seconds on the same machine.
16123 Factoring large numbers is, in general, hard. The Pollard Rho
16124 algorithm used by @command{factor} is particularly effective for
16125 numbers with relatively small factors. If you wish to factor large
16126 numbers which do not have small factors (for example, numbers which
16127 are the product of two large primes), other methods are far better.
16129 If @command{factor} is built without using GNU MP, only
16130 single-precision arithmetic is available, and so large numbers
16131 (typically @math{2^{64}} and above) will not be supported. The single-precision
16132 code uses an algorithm which is designed for factoring smaller
16138 @node seq invocation
16139 @section @command{seq}: Print numeric sequences
16142 @cindex numeric sequences
16143 @cindex sequence of numbers
16145 @command{seq} prints a sequence of numbers to standard output. Synopses:
16148 seq [@var{option}]@dots{} @var{last}
16149 seq [@var{option}]@dots{} @var{first} @var{last}
16150 seq [@var{option}]@dots{} @var{first} @var{increment} @var{last}
16153 @command{seq} prints the numbers from @var{first} to @var{last} by
16154 @var{increment}. By default, each number is printed on a separate line.
16155 When @var{increment} is not specified, it defaults to @samp{1},
16156 even when @var{first} is larger than @var{last}.
16157 @var{first} also defaults to @samp{1}. So @code{seq 1} prints
16158 @samp{1}, but @code{seq 0} and @code{seq 10 5} produce no output.
16159 Floating-point numbers may be specified. @xref{Floating point}.
16161 The program accepts the following options. Also see @ref{Common options}.
16162 Options must precede operands.
16165 @item -f @var{format}
16166 @itemx --format=@var{format}
16167 @opindex -f @var{format}
16168 @opindex --format=@var{format}
16169 @cindex formatting of numbers in @command{seq}
16170 Print all numbers using @var{format}.
16171 @var{format} must contain exactly one of the @samp{printf}-style
16172 floating point conversion specifications @samp{%a}, @samp{%e},
16173 @samp{%f}, @samp{%g}, @samp{%A}, @samp{%E}, @samp{%F}, @samp{%G}.
16174 The @samp{%} may be followed by zero or more flags taken from the set
16175 @samp{-+#0 '}, then an optional width containing one or more digits,
16176 then an optional precision consisting of a @samp{.} followed by zero
16177 or more digits. @var{format} may also contain any number of @samp{%%}
16178 conversion specifications. All conversion specifications have the
16179 same meaning as with @samp{printf}.
16181 The default format is derived from @var{first}, @var{step}, and
16182 @var{last}. If these all use a fixed point decimal representation,
16183 the default format is @samp{%.@var{p}f}, where @var{p} is the minimum
16184 precision that can represent the output numbers exactly. Otherwise,
16185 the default format is @samp{%g}.
16187 @item -s @var{string}
16188 @itemx --separator=@var{string}
16189 @cindex separator for numbers in @command{seq}
16190 Separate numbers with @var{string}; default is a newline.
16191 The output always terminates with a newline.
16194 @itemx --equal-width
16195 Print all numbers with the same width, by padding with leading zeros.
16196 @var{first}, @var{step}, and @var{last} should all use a fixed point
16197 decimal representation.
16198 (To have other kinds of padding, use @option{--format}).
16202 You can get finer-grained control over output with @option{-f}:
16205 $ seq -f '(%9.2E)' -9e5 1.1e6 1.3e6
16211 If you want hexadecimal integer output, you can use @command{printf}
16212 to perform the conversion:
16215 $ printf '%x\n' `seq 1048575 1024 1050623`
16221 For very long lists of numbers, use xargs to avoid
16222 system limitations on the length of an argument list:
16225 $ seq 1000000 | xargs printf '%x\n' | tail -n 3
16231 To generate octal output, use the printf @code{%o} format instead
16234 On most systems, seq can produce whole-number output for values up to
16235 at least @math{2^{53}}. Larger integers are approximated. The details
16236 differ depending on your floating-point implementation.
16237 @xref{Floating point}. A common
16238 case is that @command{seq} works with integers through @math{2^{64}},
16239 and larger integers may not be numerically correct:
16242 $ seq 18446744073709551616 1 18446744073709551618
16243 18446744073709551616
16244 18446744073709551616
16245 18446744073709551618
16248 Be careful when using @command{seq} with outlandish values: otherwise
16249 you may see surprising results, as @command{seq} uses floating point
16250 internally. For example, on the x86 platform, where the internal
16251 representation uses a 64-bit fraction, the command:
16254 seq 1 0.0000000000000000001 1.0000000000000000009
16257 outputs 1.0000000000000000007 twice and skips 1.0000000000000000008.
16262 @node File permissions
16263 @chapter File permissions
16266 @include parse-datetime.texi
16270 @node Opening the software toolbox
16271 @chapter Opening the Software Toolbox
16273 An earlier version of this chapter appeared in
16274 @uref{http://www.linuxjournal.com/article.php?sid=2762, the
16275 @cite{What's GNU?} column of the June 1994 @cite{Linux Journal}}.
16276 It was written by Arnold Robbins.
16279 * Toolbox introduction:: Toolbox introduction
16280 * I/O redirection:: I/O redirection
16281 * The who command:: The @command{who} command
16282 * The cut command:: The @command{cut} command
16283 * The sort command:: The @command{sort} command
16284 * The uniq command:: The @command{uniq} command
16285 * Putting the tools together:: Putting the tools together
16289 @node Toolbox introduction
16290 @unnumberedsec Toolbox Introduction
16292 This month's column is only peripherally related to the GNU Project, in
16293 that it describes a number of the GNU tools on your GNU/Linux system
16295 might be used. What it's really about is the ``Software Tools'' philosophy
16296 of program development and usage.
16298 The software tools philosophy was an important and integral concept
16299 in the initial design and development of Unix (of which Linux and GNU are
16300 essentially clones). Unfortunately, in the modern day press of
16301 Internetworking and flashy GUIs, it seems to have fallen by the
16302 wayside. This is a shame, since it provides a powerful mental model
16303 for solving many kinds of problems.
16305 Many people carry a Swiss Army knife around in their pants pockets (or
16306 purse). A Swiss Army knife is a handy tool to have: it has several knife
16307 blades, a screwdriver, tweezers, toothpick, nail file, corkscrew, and perhaps
16308 a number of other things on it. For the everyday, small miscellaneous jobs
16309 where you need a simple, general purpose tool, it's just the thing.
16311 On the other hand, an experienced carpenter doesn't build a house using
16312 a Swiss Army knife. Instead, he has a toolbox chock full of specialized
16313 tools---a saw, a hammer, a screwdriver, a plane, and so on. And he knows
16314 exactly when and where to use each tool; you won't catch him hammering nails
16315 with the handle of his screwdriver.
16317 The Unix developers at Bell Labs were all professional programmers and trained
16318 computer scientists. They had found that while a one-size-fits-all program
16319 might appeal to a user because there's only one program to use, in practice
16324 difficult to write,
16327 difficult to maintain and
16331 difficult to extend to meet new situations.
16334 Instead, they felt that programs should be specialized tools. In short, each
16335 program ``should do one thing well.'' No more and no less. Such programs are
16336 simpler to design, write, and get right---they only do one thing.
16338 Furthermore, they found that with the right machinery for hooking programs
16339 together, that the whole was greater than the sum of the parts. By combining
16340 several special purpose programs, you could accomplish a specific task
16341 that none of the programs was designed for, and accomplish it much more
16342 quickly and easily than if you had to write a special purpose program.
16343 We will see some (classic) examples of this further on in the column.
16344 (An important additional point was that, if necessary, take a detour
16345 and build any software tools you may need first, if you don't already
16346 have something appropriate in the toolbox.)
16348 @node I/O redirection
16349 @unnumberedsec I/O Redirection
16351 Hopefully, you are familiar with the basics of I/O redirection in the
16352 shell, in particular the concepts of ``standard input,'' ``standard output,''
16353 and ``standard error''. Briefly, ``standard input'' is a data source, where
16354 data comes from. A program should not need to either know or care if the
16355 data source is a disk file, a keyboard, a magnetic tape, or even a punched
16356 card reader. Similarly, ``standard output'' is a data sink, where data goes
16357 to. The program should neither know nor care where this might be.
16358 Programs that only read their standard input, do something to the data,
16359 and then send it on, are called @dfn{filters}, by analogy to filters in a
16362 With the Unix shell, it's very easy to set up data pipelines:
16365 program_to_create_data | filter1 | ... | filterN > final.pretty.data
16368 We start out by creating the raw data; each filter applies some successive
16369 transformation to the data, until by the time it comes out of the pipeline,
16370 it is in the desired form.
16372 This is fine and good for standard input and standard output. Where does the
16373 standard error come in to play? Well, think about @command{filter1} in
16374 the pipeline above. What happens if it encounters an error in the data it
16375 sees? If it writes an error message to standard output, it will just
16376 disappear down the pipeline into @command{filter2}'s input, and the
16377 user will probably never see it. So programs need a place where they can send
16378 error messages so that the user will notice them. This is standard error,
16379 and it is usually connected to your console or window, even if you have
16380 redirected standard output of your program away from your screen.
16382 For filter programs to work together, the format of the data has to be
16383 agreed upon. The most straightforward and easiest format to use is simply
16384 lines of text. Unix data files are generally just streams of bytes, with
16385 lines delimited by the @acronym{ASCII} @sc{lf} (Line Feed) character,
16386 conventionally called a ``newline'' in the Unix literature. (This is
16387 @code{'\n'} if you're a C programmer.) This is the format used by all
16388 the traditional filtering programs. (Many earlier operating systems
16389 had elaborate facilities and special purpose programs for managing
16390 binary data. Unix has always shied away from such things, under the
16391 philosophy that it's easiest to simply be able to view and edit your
16392 data with a text editor.)
16394 OK, enough introduction. Let's take a look at some of the tools, and then
16395 we'll see how to hook them together in interesting ways. In the following
16396 discussion, we will only present those command line options that interest
16397 us. As you should always do, double check your system documentation
16398 for the full story.
16400 @node The who command
16401 @unnumberedsec The @command{who} Command
16403 The first program is the @command{who} command. By itself, it generates a
16404 list of the users who are currently logged in. Although I'm writing
16405 this on a single-user system, we'll pretend that several people are
16410 @print{} arnold console Jan 22 19:57
16411 @print{} miriam ttyp0 Jan 23 14:19(:0.0)
16412 @print{} bill ttyp1 Jan 21 09:32(:0.0)
16413 @print{} arnold ttyp2 Jan 23 20:48(:0.0)
16416 Here, the @samp{$} is the usual shell prompt, at which I typed @samp{who}.
16417 There are three people logged in, and I am logged in twice. On traditional
16418 Unix systems, user names are never more than eight characters long. This
16419 little bit of trivia will be useful later. The output of @command{who} is nice,
16420 but the data is not all that exciting.
16422 @node The cut command
16423 @unnumberedsec The @command{cut} Command
16425 The next program we'll look at is the @command{cut} command. This program
16426 cuts out columns or fields of input data. For example, we can tell it
16427 to print just the login name and full name from the @file{/etc/passwd}
16428 file. The @file{/etc/passwd} file has seven fields, separated by
16432 arnold:xyzzy:2076:10:Arnold D. Robbins:/home/arnold:/bin/bash
16435 To get the first and fifth fields, we would use @command{cut} like this:
16438 $ cut -d: -f1,5 /etc/passwd
16439 @print{} root:Operator
16441 @print{} arnold:Arnold D. Robbins
16442 @print{} miriam:Miriam A. Robbins
16446 With the @option{-c} option, @command{cut} will cut out specific characters
16447 (i.e., columns) in the input lines. This is useful for input data
16448 that has fixed width fields, and does not have a field separator. For
16449 example, list the Monday dates for the current month:
16451 @c Is using cal ok? Looked at gcal, but I don't like it.
16462 @node The sort command
16463 @unnumberedsec The @command{sort} Command
16465 Next we'll look at the @command{sort} command. This is one of the most
16466 powerful commands on a Unix-style system; one that you will often find
16467 yourself using when setting up fancy data plumbing.
16470 command reads and sorts each file named on the command line. It then
16471 merges the sorted data and writes it to standard output. It will read
16472 standard input if no files are given on the command line (thus
16473 making it into a filter). The sort is based on the character collating
16474 sequence or based on user-supplied ordering criteria.
16477 @node The uniq command
16478 @unnumberedsec The @command{uniq} Command
16480 Finally (at least for now), we'll look at the @command{uniq} program. When
16481 sorting data, you will often end up with duplicate lines, lines that
16482 are identical. Usually, all you need is one instance of each line.
16483 This is where @command{uniq} comes in. The @command{uniq} program reads its
16484 standard input. It prints only one
16485 copy of each repeated line. It does have several options. Later on,
16486 we'll use the @option{-c} option, which prints each unique line, preceded
16487 by a count of the number of times that line occurred in the input.
16490 @node Putting the tools together
16491 @unnumberedsec Putting the Tools Together
16493 Now, let's suppose this is a large ISP server system with dozens of users
16494 logged in. The management wants the system administrator to write a
16496 generate a sorted list of logged in users. Furthermore, even if a user
16497 is logged in multiple times, his or her name should only show up in the
16500 The administrator could sit down with the system documentation and write a C
16501 program that did this. It would take perhaps a couple of hundred lines
16502 of code and about two hours to write it, test it, and debug it.
16503 However, knowing the software toolbox, the administrator can instead start out
16504 by generating just a list of logged on users:
16514 Next, sort the list:
16517 $ who | cut -c1-8 | sort
16524 Finally, run the sorted list through @command{uniq}, to weed out duplicates:
16527 $ who | cut -c1-8 | sort | uniq
16533 The @command{sort} command actually has a @option{-u} option that does what
16534 @command{uniq} does. However, @command{uniq} has other uses for which one
16535 cannot substitute @samp{sort -u}.
16537 The administrator puts this pipeline into a shell script, and makes it
16539 all the users on the system (@samp{#} is the system administrator,
16540 or @code{root}, prompt):
16543 # cat > /usr/local/bin/listusers
16544 who | cut -c1-8 | sort | uniq
16546 # chmod +x /usr/local/bin/listusers
16549 There are four major points to note here. First, with just four
16550 programs, on one command line, the administrator was able to save about two
16551 hours worth of work. Furthermore, the shell pipeline is just about as
16552 efficient as the C program would be, and it is much more efficient in
16553 terms of programmer time. People time is much more expensive than
16554 computer time, and in our modern ``there's never enough time to do
16555 everything'' society, saving two hours of programmer time is no mean
16558 Second, it is also important to emphasize that with the
16559 @emph{combination} of the tools, it is possible to do a special
16560 purpose job never imagined by the authors of the individual programs.
16562 Third, it is also valuable to build up your pipeline in stages, as we did here.
16563 This allows you to view the data at each stage in the pipeline, which helps
16564 you acquire the confidence that you are indeed using these tools correctly.
16566 Finally, by bundling the pipeline in a shell script, other users can use
16567 your command, without having to remember the fancy plumbing you set up for
16568 them. In terms of how you run them, shell scripts and compiled programs are
16571 After the previous warm-up exercise, we'll look at two additional, more
16572 complicated pipelines. For them, we need to introduce two more tools.
16574 The first is the @command{tr} command, which stands for ``transliterate.''
16575 The @command{tr} command works on a character-by-character basis, changing
16576 characters. Normally it is used for things like mapping upper case to
16580 $ echo ThIs ExAmPlE HaS MIXED case! | tr '[:upper:]' '[:lower:]'
16581 @print{} this example has mixed case!
16584 There are several options of interest:
16588 work on the complement of the listed characters, i.e.,
16589 operations apply to characters not in the given set
16592 delete characters in the first set from the output
16595 squeeze repeated characters in the output into just one character.
16598 We will be using all three options in a moment.
16600 The other command we'll look at is @command{comm}. The @command{comm}
16601 command takes two sorted input files as input data, and prints out the
16602 files' lines in three columns. The output columns are the data lines
16603 unique to the first file, the data lines unique to the second file, and
16604 the data lines that are common to both. The @option{-1}, @option{-2}, and
16605 @option{-3} command line options @emph{omit} the respective columns. (This is
16606 non-intuitive and takes a little getting used to.) For example:
16628 The file name @file{-} tells @command{comm} to read standard input
16629 instead of a regular file.
16631 Now we're ready to build a fancy pipeline. The first application is a word
16632 frequency counter. This helps an author determine if he or she is over-using
16635 The first step is to change the case of all the letters in our input file
16636 to one case. ``The'' and ``the'' are the same word when doing counting.
16639 $ tr '[:upper:]' '[:lower:]' < whats.gnu | ...
16642 The next step is to get rid of punctuation. Quoted words and unquoted words
16643 should be treated identically; it's easiest to just get the punctuation out of
16647 $ tr '[:upper:]' '[:lower:]' < whats.gnu | tr -cd '[:alnum:]_ \n' | ...
16650 The second @command{tr} command operates on the complement of the listed
16651 characters, which are all the letters, the digits, the underscore, and
16652 the blank. The @samp{\n} represents the newline character; it has to
16653 be left alone. (The @acronym{ASCII} tab character should also be included for
16654 good measure in a production script.)
16656 At this point, we have data consisting of words separated by blank space.
16657 The words only contain alphanumeric characters (and the underscore). The
16658 next step is break the data apart so that we have one word per line. This
16659 makes the counting operation much easier, as we will see shortly.
16662 $ tr '[:upper:]' '[:lower:]' < whats.gnu | tr -cd '[:alnum:]_ \n' |
16663 > tr -s ' ' '\n' | ...
16666 This command turns blanks into newlines. The @option{-s} option squeezes
16667 multiple newline characters in the output into just one. This helps us
16668 avoid blank lines. (The @samp{>} is the shell's ``secondary prompt.''
16669 This is what the shell prints when it notices you haven't finished
16670 typing in all of a command.)
16672 We now have data consisting of one word per line, no punctuation, all one
16673 case. We're ready to count each word:
16676 $ tr '[:upper:]' '[:lower:]' < whats.gnu | tr -cd '[:alnum:]_ \n' |
16677 > tr -s ' ' '\n' | sort | uniq -c | ...
16680 At this point, the data might look something like this:
16693 The output is sorted by word, not by count! What we want is the most
16694 frequently used words first. Fortunately, this is easy to accomplish,
16695 with the help of two more @command{sort} options:
16699 do a numeric sort, not a textual one
16702 reverse the order of the sort
16705 The final pipeline looks like this:
16708 $ tr '[:upper:]' '[:lower:]' < whats.gnu | tr -cd '[:alnum:]_ \n' |
16709 > tr -s ' ' '\n' | sort | uniq -c | sort -n -r
16718 Whew! That's a lot to digest. Yet, the same principles apply. With six
16719 commands, on two lines (really one long one split for convenience), we've
16720 created a program that does something interesting and useful, in much
16721 less time than we could have written a C program to do the same thing.
16723 A minor modification to the above pipeline can give us a simple spelling
16724 checker! To determine if you've spelled a word correctly, all you have to
16725 do is look it up in a dictionary. If it is not there, then chances are
16726 that your spelling is incorrect. So, we need a dictionary.
16727 The conventional location for a dictionary is @file{/usr/dict/words}.
16728 On my GNU/Linux system,@footnote{Redhat Linux 6.1, for the November 2000
16729 revision of this article.}
16730 this is a sorted, 45,402 word dictionary.
16732 Now, how to compare our file with the dictionary? As before, we generate
16733 a sorted list of words, one per line:
16736 $ tr '[:upper:]' '[:lower:]' < whats.gnu | tr -cd '[:alnum:]_ \n' |
16737 > tr -s ' ' '\n' | sort -u | ...
16740 Now, all we need is a list of words that are @emph{not} in the
16741 dictionary. Here is where the @command{comm} command comes in.
16744 $ tr '[:upper:]' '[:lower:]' < whats.gnu | tr -cd '[:alnum:]_ \n' |
16745 > tr -s ' ' '\n' | sort -u |
16746 > comm -23 - /usr/dict/words
16749 The @option{-2} and @option{-3} options eliminate lines that are only in the
16750 dictionary (the second file), and lines that are in both files. Lines
16751 only in the first file (standard input, our stream of words), are
16752 words that are not in the dictionary. These are likely candidates for
16753 spelling errors. This pipeline was the first cut at a production
16754 spelling checker on Unix.
16756 There are some other tools that deserve brief mention.
16760 search files for text that matches a regular expression
16763 count lines, words, characters
16766 a T-fitting for data pipes, copies data to files and to standard output
16769 the stream editor, an advanced tool
16772 a data manipulation language, another advanced tool
16775 The software tools philosophy also espoused the following bit of
16776 advice: ``Let someone else do the hard part.'' This means, take
16777 something that gives you most of what you need, and then massage it the
16778 rest of the way until it's in the form that you want.
16784 Each program should do one thing well. No more, no less.
16787 Combining programs with appropriate plumbing leads to results where
16788 the whole is greater than the sum of the parts. It also leads to novel
16789 uses of programs that the authors might never have imagined.
16792 Programs should never print extraneous header or trailer data, since these
16793 could get sent on down a pipeline. (A point we didn't mention earlier.)
16796 Let someone else do the hard part.
16799 Know your toolbox! Use each program appropriately. If you don't have an
16800 appropriate tool, build one.
16803 As of this writing, all the programs we've discussed are available via
16804 anonymous @command{ftp} from: @*
16805 @uref{ftp://gnudist.gnu.org/textutils/textutils-1.22.tar.gz}. (There may
16806 be more recent versions available now.)
16808 None of what I have presented in this column is new. The Software Tools
16809 philosophy was first introduced in the book @cite{Software Tools}, by
16810 Brian Kernighan and P.J. Plauger (Addison-Wesley, ISBN 0-201-03669-X).
16811 This book showed how to write and use software tools. It was written in
16812 1976, using a preprocessor for FORTRAN named @command{ratfor} (RATional
16813 FORtran). At the time, C was not as ubiquitous as it is now; FORTRAN
16814 was. The last chapter presented a @command{ratfor} to FORTRAN
16815 processor, written in @command{ratfor}. @command{ratfor} looks an awful
16816 lot like C; if you know C, you won't have any problem following the
16819 In 1981, the book was updated and made available as @cite{Software Tools
16820 in Pascal} (Addison-Wesley, ISBN 0-201-10342-7). Both books are
16821 still in print and are well worth
16822 reading if you're a programmer. They certainly made a major change in
16823 how I view programming.
16825 The programs in both books are available from
16826 @uref{http://cm.bell-labs.com/who/bwk, Brian Kernighan's home page}.
16827 For a number of years, there was an active
16828 Software Tools Users Group, whose members had ported the original
16829 @command{ratfor} programs to essentially every computer system with a
16830 FORTRAN compiler. The popularity of the group waned in the middle 1980s
16831 as Unix began to spread beyond universities.
16833 With the current proliferation of GNU code and other clones of Unix programs,
16834 these programs now receive little attention; modern C versions are
16835 much more efficient and do more than these programs do. Nevertheless, as
16836 exposition of good programming style, and evangelism for a still-valuable
16837 philosophy, these books are unparalleled, and I recommend them highly.
16839 Acknowledgment: I would like to express my gratitude to Brian Kernighan
16840 of Bell Labs, the original Software Toolsmith, for reviewing this column.
16842 @node GNU Free Documentation License
16843 @appendix GNU Free Documentation License
16847 @node Concept index
16854 @c Local variables:
16855 @c texinfo-column-for-description: 32