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.
34 @c * mktemp: (coreutils)mktemp invocation. FIXME.
36 @dircategory Individual utilities
38 * arch: (coreutils)arch invocation. Print machine hardware name.
39 * base64: (coreutils)base64 invocation. Base64 encode/decode data.
40 * basename: (coreutils)basename invocation. Strip directory and suffix.
41 * cat: (coreutils)cat invocation. Concatenate and write files.
42 * chcon: (coreutils)chcon invocation. Change SELinux CTX of files.
43 * chgrp: (coreutils)chgrp invocation. Change file groups.
44 * chmod: (coreutils)chmod invocation. Change file permissions.
45 * chown: (coreutils)chown invocation. Change file owners/groups.
46 * chroot: (coreutils)chroot invocation. Specify the root directory.
47 * cksum: (coreutils)cksum invocation. Print POSIX CRC checksum.
48 * comm: (coreutils)comm invocation. Compare sorted files by line.
49 * cp: (coreutils)cp invocation. Copy files.
50 * csplit: (coreutils)csplit invocation. Split by context.
51 * cut: (coreutils)cut invocation. Print selected parts of lines.
52 * date: (coreutils)date invocation. Print/set system date and time.
53 * dd: (coreutils)dd invocation. Copy and convert a file.
54 * df: (coreutils)df invocation. Report file system disk usage.
55 * dir: (coreutils)dir invocation. List directories briefly.
56 * dircolors: (coreutils)dircolors invocation. Color setup for ls.
57 * dirname: (coreutils)dirname invocation. Strip non-directory suffix.
58 * du: (coreutils)du invocation. Report on disk usage.
59 * echo: (coreutils)echo invocation. Print a line of text.
60 * env: (coreutils)env invocation. Modify the environment.
61 * expand: (coreutils)expand invocation. Convert tabs to spaces.
62 * expr: (coreutils)expr invocation. Evaluate expressions.
63 * factor: (coreutils)factor invocation. Print prime factors
64 * false: (coreutils)false invocation. Do nothing, unsuccessfully.
65 * fmt: (coreutils)fmt invocation. Reformat paragraph text.
66 * fold: (coreutils)fold invocation. Wrap long input lines.
67 * groups: (coreutils)groups invocation. Print group names a user is in.
68 * head: (coreutils)head invocation. Output the first part of files.
69 * hostid: (coreutils)hostid invocation. Print numeric host identifier.
70 * hostname: (coreutils)hostname invocation. Print or set system name.
71 * id: (coreutils)id invocation. Print user identity.
72 * install: (coreutils)install invocation. Copy and change attributes.
73 * join: (coreutils)join invocation. Join lines on a common field.
74 * kill: (coreutils)kill invocation. Send a signal to processes.
75 * link: (coreutils)link invocation. Make hard links between files.
76 * ln: (coreutils)ln invocation. Make links between files.
77 * logname: (coreutils)logname invocation. Print current login name.
78 * ls: (coreutils)ls invocation. List directory contents.
79 * md5sum: (coreutils)md5sum invocation. Print or check MD5 digests.
80 * mkdir: (coreutils)mkdir invocation. Create directories.
81 * mkfifo: (coreutils)mkfifo invocation. Create FIFOs (named pipes).
82 * mknod: (coreutils)mknod invocation. Create special 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 * od: (coreutils)od invocation. Dump files in octal, etc.
88 * paste: (coreutils)paste invocation. Merge lines of files.
89 * pathchk: (coreutils)pathchk invocation. Check file name portability.
90 * pr: (coreutils)pr invocation. Paginate or columnate files.
91 * printenv: (coreutils)printenv invocation. Print environment variables.
92 * printf: (coreutils)printf invocation. Format and print data.
93 * ptx: (coreutils)ptx invocation. Produce permuted indexes.
94 * pwd: (coreutils)pwd invocation. Print working directory.
95 * readlink: (coreutils)readlink invocation. Print referent of a symlink.
96 * rm: (coreutils)rm invocation. Remove files.
97 * rmdir: (coreutils)rmdir invocation. Remove empty directories.
98 * runcon: (coreutils)runcon invocation. Run in specified SELinux CTX.
99 * seq: (coreutils)seq invocation. Print numeric sequences
100 * sha1sum: (coreutils)sha1sum invocation. Print or check SHA-1 digests.
101 * sha2: (coreutils)sha2 utilities. Print or check SHA-2 digests.
102 * shred: (coreutils)shred invocation. Remove files more securely.
103 * shuf: (coreutils)shuf invocation. Shuffling text files.
104 * sleep: (coreutils)sleep invocation. Delay for a specified time.
105 * sort: (coreutils)sort invocation. Sort text files.
106 * split: (coreutils)split invocation. Split into fixed-size pieces.
107 * stat: (coreutils)stat invocation. Report file(system) status.
108 * stdbuf: (coreutils)stdbuf invocation. Modify stdio buffering.
109 * stty: (coreutils)stty invocation. Print/change terminal settings.
110 * su: (coreutils)su invocation. Modify user and group ID.
111 * sum: (coreutils)sum invocation. Print traditional checksum.
112 * sync: (coreutils)sync invocation. Synchronize memory and disk.
113 * tac: (coreutils)tac invocation. Reverse files.
114 * tail: (coreutils)tail invocation. Output the last part of files.
115 * tee: (coreutils)tee invocation. Redirect to multiple files.
116 * test: (coreutils)test invocation. File/string tests.
117 * timeout: (coreutils)timeout invocation. Run with time limit.
118 * touch: (coreutils)touch invocation. Change file timestamps.
119 * tr: (coreutils)tr invocation. Translate characters.
120 * true: (coreutils)true invocation. Do nothing, successfully.
121 * truncate: (coreutils)truncate invocation. Shrink/extend size of a file.
122 * tsort: (coreutils)tsort invocation. Topological sort.
123 * tty: (coreutils)tty invocation. Print terminal name.
124 * uname: (coreutils)uname invocation. Print system information.
125 * unexpand: (coreutils)unexpand invocation. Convert spaces to tabs.
126 * uniq: (coreutils)uniq invocation. Uniquify files.
127 * unlink: (coreutils)unlink invocation. Removal via unlink(2).
128 * uptime: (coreutils)uptime invocation. Print uptime and load.
129 * users: (coreutils)users invocation. Print current user names.
130 * vdir: (coreutils)vdir invocation. List directories verbosely.
131 * wc: (coreutils)wc invocation. Line, word, and byte counts.
132 * who: (coreutils)who invocation. Print who is logged in.
133 * whoami: (coreutils)whoami invocation. Print effective user ID.
134 * yes: (coreutils)yes invocation. Print a string indefinitely.
138 This manual documents version @value{VERSION} of the @sc{gnu} core
139 utilities, including the standard programs for text and file manipulation.
141 Copyright @copyright{} 1994-1996, 2000-2009 Free Software Foundation, Inc.
144 Permission is granted to copy, distribute and/or modify this document
145 under the terms of the GNU Free Documentation License, Version 1.3 or
146 any later version published by the Free Software Foundation; with no
147 Invariant Sections, with no Front-Cover Texts, and with no Back-Cover
148 Texts. A copy of the license is included in the section entitled ``GNU
149 Free Documentation License''.
154 @title @sc{gnu} @code{Coreutils}
155 @subtitle Core GNU utilities
156 @subtitle for version @value{VERSION}, @value{UPDATED}
157 @author David MacKenzie et al.
160 @vskip 0pt plus 1filll
173 @cindex core utilities
174 @cindex text utilities
175 @cindex shell utilities
176 @cindex file utilities
179 * Introduction:: Caveats, overview, and authors.
180 * Common options:: Common options.
181 * Output of entire files:: cat tac nl od
182 * Formatting file contents:: fmt pr fold
183 * Output of parts of files:: head tail split csplit
184 * Summarizing files:: wc sum cksum md5sum sha1sum sha2
185 * Operating on sorted files:: sort shuf uniq comm ptx tsort
186 * Operating on fields within a line:: cut paste join
187 * Operating on characters:: tr expand unexpand
188 * Directory listing:: ls dir vdir dircolors
189 * Basic operations:: cp dd install mv rm shred
190 * Special file types:: ln mkdir rmdir mkfifo mknod
191 * Changing file attributes:: chgrp chmod chown touch
192 * Disk usage:: df du stat sync truncate
193 * Printing text:: echo printf yes
194 * Conditions:: false true test expr
196 * File name manipulation:: dirname basename pathchk
197 * Working context:: pwd stty printenv tty
198 * User information:: id logname whoami groups users who
199 * System context:: date uname hostname hostid uptime
200 * SELinux context:: chcon runcon
201 * Modified command invocation:: chroot env nice nohup stdbuf su timeout
202 * Process control:: kill
204 * Numeric operations:: factor seq
205 * File permissions:: Access modes.
206 * Date input formats:: Specifying date strings.
207 * Opening the software toolbox:: The software tools philosophy.
208 * GNU Free Documentation License:: Copying and sharing this manual.
209 * Concept index:: General index.
212 --- The Detailed Node Listing ---
216 * Exit status:: Indicating program success or failure.
217 * Backup options:: Backup options
218 * Block size:: Block size
219 * Signal specifications:: Specifying signals
220 * Disambiguating names and IDs:: chgrp and chown owner and group syntax
221 * Random sources:: Sources of random data
222 * Target directory:: Target directory
223 * Trailing slashes:: Trailing slashes
224 * Traversing symlinks:: Traversing symlinks to directories
225 * Treating / specially:: Treating / specially
226 * Standards conformance:: Standards conformance
228 Output of entire files
230 * cat invocation:: Concatenate and write files.
231 * tac invocation:: Concatenate and write files in reverse.
232 * nl invocation:: Number lines and write files.
233 * od invocation:: Write files in octal or other formats.
234 * base64 invocation:: Transform data into printable data.
236 Formatting file contents
238 * fmt invocation:: Reformat paragraph text.
239 * pr invocation:: Paginate or columnate files for printing.
240 * fold invocation:: Wrap input lines to fit in specified width.
242 Output of parts of files
244 * head invocation:: Output the first part of files.
245 * tail invocation:: Output the last part of files.
246 * split invocation:: Split a file into fixed-size pieces.
247 * csplit invocation:: Split a file into context-determined pieces.
251 * wc invocation:: Print newline, word, and byte counts.
252 * sum invocation:: Print checksum and block counts.
253 * cksum invocation:: Print CRC checksum and byte counts.
254 * md5sum invocation:: Print or check MD5 digests.
255 * sha1sum invocation:: Print or check SHA-1 digests.
256 * sha2 utilities:: Print or check SHA-2 digests.
258 Operating on sorted files
260 * sort invocation:: Sort text files.
261 * shuf invocation:: Shuffle text files.
262 * uniq invocation:: Uniquify files.
263 * comm invocation:: Compare two sorted files line by line.
264 * ptx invocation:: Produce a permuted index of file contents.
265 * tsort invocation:: Topological sort.
267 @command{ptx}: Produce permuted indexes
269 * General options in ptx:: Options which affect general program behavior.
270 * Charset selection in ptx:: Underlying character set considerations.
271 * Input processing in ptx:: Input fields, contexts, and keyword selection.
272 * Output formatting in ptx:: Types of output format, and sizing the fields.
273 * Compatibility in ptx:: The @acronym{GNU} extensions to @command{ptx}
275 Operating on fields within a line
277 * cut invocation:: Print selected parts of lines.
278 * paste invocation:: Merge lines of files.
279 * join invocation:: Join lines on a common field.
281 Operating on characters
283 * tr invocation:: Translate, squeeze, and/or delete characters.
284 * expand invocation:: Convert tabs to spaces.
285 * unexpand invocation:: Convert spaces to tabs.
287 @command{tr}: Translate, squeeze, and/or delete characters
289 * Character sets:: Specifying sets of characters.
290 * Translating:: Changing one set of characters to another.
291 * Squeezing:: Squeezing repeats and deleting.
295 * ls invocation:: List directory contents
296 * dir invocation:: Briefly list directory contents
297 * vdir invocation:: Verbosely list directory contents
298 * dircolors invocation:: Color setup for @command{ls}
300 @command{ls}: List directory contents
302 * Which files are listed:: Which files are listed
303 * What information is listed:: What information is listed
304 * Sorting the output:: Sorting the output
305 * More details about version sort:: More details about version sort
306 * General output formatting:: General output formatting
307 * Formatting the file names:: Formatting the file names
311 * cp invocation:: Copy files and directories
312 * dd invocation:: Convert and copy a file
313 * install invocation:: Copy files and set attributes
314 * mv invocation:: Move (rename) files
315 * rm invocation:: Remove files or directories
316 * shred invocation:: Remove files more securely
320 * link invocation:: Make a hard link via the link syscall
321 * ln invocation:: Make links between files
322 * mkdir invocation:: Make directories
323 * mkfifo invocation:: Make FIFOs (named pipes)
324 * mknod invocation:: Make block or character special files
325 * readlink invocation:: Print value of a symlink or canonical file name
326 * rmdir invocation:: Remove empty directories
327 * unlink invocation:: Remove files via unlink syscall
329 Changing file attributes
331 * chown invocation:: Change file owner and group
332 * chgrp invocation:: Change group ownership
333 * chmod invocation:: Change access permissions
334 * touch invocation:: Change file timestamps
338 * df invocation:: Report file system disk space usage
339 * du invocation:: Estimate file space usage
340 * stat invocation:: Report file or file system status
341 * sync invocation:: Synchronize data on disk with memory
342 * truncate invocation:: Shrink or extend the size of a file
346 * echo invocation:: Print a line of text
347 * printf invocation:: Format and print data
348 * yes invocation:: Print a string until interrupted
352 * false invocation:: Do nothing, unsuccessfully
353 * true invocation:: Do nothing, successfully
354 * test invocation:: Check file types and compare values
355 * expr invocation:: Evaluate expressions
357 @command{test}: Check file types and compare values
359 * File type tests:: File type tests
360 * Access permission tests:: Access permission tests
361 * File characteristic tests:: File characteristic tests
362 * String tests:: String tests
363 * Numeric tests:: Numeric tests
365 @command{expr}: Evaluate expression
367 * String expressions:: + : match substr index length
368 * Numeric expressions:: + - * / %
369 * Relations for expr:: | & < <= = == != >= >
370 * Examples of expr:: Examples of using @command{expr}
374 * tee invocation:: Redirect output to multiple files or processes
376 File name manipulation
378 * basename invocation:: Strip directory and suffix from a file name
379 * dirname invocation:: Strip non-directory suffix from a file name
380 * pathchk invocation:: Check file name validity and portability
384 * pwd invocation:: Print working directory
385 * stty invocation:: Print or change terminal characteristics
386 * printenv invocation:: Print all or some environment variables
387 * tty invocation:: Print file name of terminal on standard input
389 @command{stty}: Print or change terminal characteristics
391 * Control:: Control settings
392 * Input:: Input settings
393 * Output:: Output settings
394 * Local:: Local settings
395 * Combination:: Combination settings
396 * Characters:: Special characters
397 * Special:: Special settings
401 * id invocation:: Print user identity
402 * logname invocation:: Print current login name
403 * whoami invocation:: Print effective user ID
404 * groups invocation:: Print group names a user is in
405 * users invocation:: Print login names of users currently logged in
406 * who invocation:: Print who is currently logged in
410 * arch invocation:: Print machine hardware name
411 * date invocation:: Print or set system date and time
412 * uname invocation:: Print system information
413 * hostname invocation:: Print or set system name
414 * hostid invocation:: Print numeric host identifier
415 * uptime invocation:: Print system uptime and load
417 @command{date}: Print or set system date and time
419 * Time conversion specifiers:: %[HIklMNpPrRsSTXzZ]
420 * Date conversion specifiers:: %[aAbBcCdDeFgGhjmuUVwWxyY]
421 * Literal conversion specifiers:: %[%nt]
422 * Padding and other flags:: Pad with zeros, spaces, etc.
423 * Setting the time:: Changing the system clock.
424 * Options for date:: Instead of the current time.
425 * Date input formats:: Specifying date strings.
426 * Examples of date:: Examples.
430 * chcon invocation:: Change SELinux context of file
431 * runcon invocation:: Run a command in specified SELinux context
433 Modified command invocation
435 * chroot invocation:: Run a command with a different root directory
436 * env invocation:: Run a command in a modified environment
437 * nice invocation:: Run a command with modified niceness
438 * nohup invocation:: Run a command immune to hangups
439 * stdbuf invocation:: Run a command with modified I/O buffering
440 * su invocation:: Run a command with substitute user and group ID
441 * timeout invocation:: Run a command with a time limit
445 * kill invocation:: Sending a signal to processes.
449 * sleep invocation:: Delay for a specified time
453 * factor invocation:: Print prime factors
454 * seq invocation:: Print numeric sequences
458 * Mode Structure:: Structure of file mode bits.
459 * Symbolic Modes:: Mnemonic representation of file mode bits.
460 * Numeric Modes:: File mode bits as octal numbers.
461 * Directory Setuid and Setgid:: Set-user-ID and set-group-ID on directories.
465 * General date syntax:: Common rules.
466 * Calendar date items:: 19 Dec 1994.
467 * Time of day items:: 9:20pm.
468 * Time zone items:: @sc{est}, @sc{pdt}, @sc{gmt}.
469 * Day of week items:: Monday and others.
470 * Relative items in date strings:: next tuesday, 2 years ago.
471 * Pure numbers in date strings:: 19931219, 1440.
472 * Seconds since the Epoch:: @@1078100502.
473 * Specifying time zone rules:: TZ="America/New_York", TZ="UTC0".
474 * Authors of get_date:: Bellovin, Eggert, Salz, Berets, et al.
476 Opening the software toolbox
478 * Toolbox introduction:: Toolbox introduction
479 * I/O redirection:: I/O redirection
480 * The who command:: The @command{who} command
481 * The cut command:: The @command{cut} command
482 * The sort command:: The @command{sort} command
483 * The uniq command:: The @command{uniq} command
484 * Putting the tools together:: Putting the tools together
488 * GNU Free Documentation License:: Copying and sharing this manual.
495 @chapter Introduction
497 This manual is a work in progress: many sections make no attempt to explain
498 basic concepts in a way suitable for novices. Thus, if you are interested,
499 please get involved in improving this manual. The entire @sc{gnu} community
502 @cindex @acronym{POSIX}
503 The @sc{gnu} utilities documented here are mostly compatible with the
504 @acronym{POSIX} standard.
505 @cindex bugs, reporting
506 Please report bugs to @email{bug-coreutils@@gnu.org}. Remember
507 to include the version number, machine architecture, input files, and
508 any other information needed to reproduce the bug: your input, what you
509 expected, what you got, and why it is wrong. Diffs are welcome, but
510 please include a description of the problem as well, since this is
511 sometimes difficult to infer. @xref{Bugs, , , gcc, Using and Porting GNU CC}.
517 @cindex MacKenzie, D.
520 This manual was originally derived from the Unix man pages in the
521 distributions, which were written by David MacKenzie and updated by Jim
522 Meyering. What you are reading now is the authoritative documentation
523 for these utilities; the man pages are no longer being maintained. The
524 original @command{fmt} man page was written by Ross Paterson. Fran@,{c}ois
525 Pinard did the initial conversion to Texinfo format. Karl Berry did the
526 indexing, some reorganization, and editing of the results. Brian
527 Youmans of the Free Software Foundation office staff combined the
528 manuals for textutils, fileutils, and sh-utils to produce the present
529 omnibus manual. Richard Stallman contributed his usual invaluable
530 insights to the overall process.
533 @chapter Common options
537 @itemx @w{@kbd{--backup}[=@var{method}]}
540 @vindex VERSION_CONTROL
541 @cindex backups, making
542 @xref{Backup options}.
543 Make a backup of each file that would otherwise be overwritten or removed.
546 @macro optBackupSuffix
547 @item -S @var{suffix}
548 @itemx --suffix=@var{suffix}
551 Append @var{suffix} to each backup file made with @option{-b}.
552 @xref{Backup options}.
555 @macro optTargetDirectory
556 @item -t @var{directory}
557 @itemx @w{@kbd{--target-directory}=@var{directory}}
559 @opindex --target-directory
560 @cindex target directory
561 @cindex destination directory
562 Specify the destination @var{directory}.
563 @xref{Target directory}.
566 @macro optNoTargetDirectory
568 @itemx --no-target-directory
570 @opindex --no-target-directory
571 @cindex target directory
572 @cindex destination directory
573 Do not treat the last operand specially when it is a directory or a
574 symbolic link to a directory. @xref{Target directory}.
581 Append an SI-style abbreviation to each size, such as @samp{M} for
582 megabytes. Powers of 1000 are used, not 1024; @samp{M} stands for
583 1,000,000 bytes. This option is equivalent to
584 @option{--block-size=si}. Use the @option{-h} or
585 @option{--human-readable} option if
586 you prefer powers of 1024.
589 @macro optHumanReadable
591 @itemx --human-readable
593 @opindex --human-readable
594 @cindex human-readable output
595 Append a size letter to each size, such as @samp{M} for mebibytes.
596 Powers of 1024 are used, not 1000; @samp{M} stands for 1,048,576 bytes.
597 This option is equivalent to @option{--block-size=human-readable}.
598 Use the @option{--si} option if you prefer powers of 1000.
601 @macro optStripTrailingSlashes
602 @itemx @w{@kbd{--strip-trailing-slashes}}
603 @opindex --strip-trailing-slashes
604 @cindex stripping trailing slashes
605 Remove any trailing slashes from each @var{source} argument.
606 @xref{Trailing slashes}.
609 @macro mayConflictWithShellBuiltIn{cmd}
610 @cindex conflicts with shell built-ins
611 @cindex built-in shell commands, conflicts with
612 Due to shell aliases and built-in @command{\cmd\} command, using an
613 unadorned @command{\cmd\} interactively or in a script may get you
614 different functionality than that described here. Invoke it via
615 @command{env} (i.e., @code{env \cmd\ @dots{}}) to avoid interference
620 @macro multiplierSuffixes{varName}
621 @var{\varName\} may be, or may be an integer optionally followed by,
622 one of the following multiplicative suffixes:
624 @samp{b} => 512 ("blocks")
625 @samp{KB} => 1000 (KiloBytes)
626 @samp{K} => 1024 (KibiBytes)
627 @samp{MB} => 1000*1000 (MegaBytes)
628 @samp{M} => 1024*1024 (MebiBytes)
629 @samp{GB} => 1000*1000*1000 (GigaBytes)
630 @samp{G} => 1024*1024*1024 (GibiBytes)
632 and so on for @samp{T}, @samp{P}, @samp{E}, @samp{Z}, and @samp{Y}.
635 @c FIXME: same as above, but no ``blocks'' line.
636 @macro multiplierSuffixesNoBlocks{varName}
637 @var{\varName\} may be, or may be an integer optionally followed by,
638 one of the following multiplicative suffixes:
640 @samp{KB} => 1000 (KiloBytes)
641 @samp{K} => 1024 (KibiBytes)
642 @samp{MB} => 1000*1000 (MegaBytes)
643 @samp{M} => 1024*1024 (MebiBytes)
644 @samp{GB} => 1000*1000*1000 (GigaBytes)
645 @samp{G} => 1024*1024*1024 (GibiBytes)
647 and so on for @samp{T}, @samp{P}, @samp{E}, @samp{Z}, and @samp{Y}.
650 @cindex common options
652 Certain options are available in all of these programs. Rather than
653 writing identical descriptions for each of the programs, they are
654 described here. (In fact, every @sc{gnu} program accepts (or should accept)
657 @vindex POSIXLY_CORRECT
658 Normally options and operands can appear in any order, and programs act
659 as if all the options appear before any operands. For example,
660 @samp{sort -r passwd -t :} acts like @samp{sort -r -t : passwd}, since
661 @samp{:} is an option-argument of @option{-t}. However, if the
662 @env{POSIXLY_CORRECT} environment variable is set, options must appear
663 before operands, unless otherwise specified for a particular command.
665 A few programs can usefully have trailing operands with leading
666 @samp{-}. With such a program, options must precede operands even if
667 @env{POSIXLY_CORRECT} is not set, and this fact is noted in the
668 program description. For example, the @command{env} command's options
669 must appear before its operands, since in some cases the operands
670 specify a command that itself contains options.
672 Most programs that accept long options recognize unambiguous
673 abbreviations of those options. For example, @samp{rmdir
674 --ignore-fail-on-non-empty} can be invoked as @samp{rmdir
675 --ignore-fail} or even @samp{rmdir --i}. Ambiguous options, such as
676 @samp{ls --h}, are identified as such.
678 Some of these programs recognize the @option{--help} and @option{--version}
679 options only when one of them is the sole command line argument. For
680 these programs, abbreviations of the long options are not always recognized.
687 Print a usage message listing all available options, then exit successfully.
691 @cindex version number, finding
692 Print the version number, then exit successfully.
696 @cindex option delimiter
697 Delimit the option list. Later arguments, if any, are treated as
698 operands even if they begin with @samp{-}. For example, @samp{sort --
699 -r} reads from the file named @file{-r}.
703 @cindex standard input
704 @cindex standard output
705 A single @samp{-} operand is not really an option, though it looks like one. It
706 stands for standard input, or for standard output if that is clear from
707 the context. For example, @samp{sort -} reads from standard input,
708 and is equivalent to plain @samp{sort}, and @samp{tee -} writes an
709 extra copy of its input to standard output. Unless otherwise
710 specified, @samp{-} can appear as any operand that requires a file
714 * Exit status:: Indicating program success or failure.
715 * Backup options:: -b -S, in some programs.
716 * Block size:: BLOCK_SIZE and --block-size, in some programs.
717 * Signal specifications:: Specifying signals using the --signal option.
718 * Disambiguating names and IDs:: chgrp and chown owner and group syntax
719 * Random sources:: --random-source, in some programs.
720 * Target directory:: Specifying a target directory, in some programs.
721 * Trailing slashes:: --strip-trailing-slashes, in some programs.
722 * Traversing symlinks:: -H, -L, or -P, in some programs.
723 * Treating / specially:: --preserve-root and --no-preserve-root.
724 * Special built-in utilities:: @command{break}, @command{:}, @command{eval}, @dots{}
725 * Standards conformance:: Conformance to the @acronym{POSIX} standard.
733 An exit status of zero indicates success,
734 and a nonzero value indicates failure.
737 Nearly every command invocation yields an integral @dfn{exit status}
738 that can be used to change how other commands work.
739 For the vast majority of commands, an exit status of zero indicates
740 success. Failure is indicated by a nonzero value---typically
741 @samp{1}, though it may differ on unusual platforms as @acronym{POSIX}
742 requires only that it be nonzero.
744 However, some of the programs documented here do produce
745 other exit status values and a few associate different
746 meanings with the values @samp{0} and @samp{1}.
747 Here are some of the exceptions:
748 @command{chroot}, @command{env}, @command{expr},
749 @command{nice}, @command{nohup}, @command{printenv}, @command{sort},
750 @command{su}, @command{test}, @command{timeout}, @command{tty}.
754 @section Backup options
756 @cindex backup options
758 Some @sc{gnu} programs (at least @command{cp}, @command{install},
759 @command{ln}, and @command{mv}) optionally make backups of files
760 before writing new versions.
761 These options control the details of these backups. The options are also
762 briefly mentioned in the descriptions of the particular programs.
767 @itemx @w{@kbd{--backup}[=@var{method}]}
770 @vindex VERSION_CONTROL
771 @cindex backups, making
772 Make a backup of each file that would otherwise be overwritten or removed.
773 Without this option, the original versions are destroyed.
774 Use @var{method} to determine the type of backups to make.
775 When this option is used but @var{method} is not specified,
776 then the value of the @env{VERSION_CONTROL}
777 environment variable is used. And if @env{VERSION_CONTROL} is not set,
778 the default backup type is @samp{existing}.
780 Note that the short form of this option, @option{-b} does not accept any
781 argument. Using @option{-b} is equivalent to using @option{--backup=existing}.
783 @vindex version-control @r{Emacs variable}
784 This option corresponds to the Emacs variable @samp{version-control};
785 the values for @var{method} are the same as those used in Emacs.
786 This option also accepts more descriptive names.
787 The valid @var{method}s are (unique abbreviations are accepted):
792 @opindex none @r{backup method}
797 @opindex numbered @r{backup method}
798 Always make numbered backups.
802 @opindex existing @r{backup method}
803 Make numbered backups of files that already have them, simple backups
808 @opindex simple @r{backup method}
809 Always make simple backups. Please note @samp{never} is not to be
810 confused with @samp{none}.
814 @item -S @var{suffix}
815 @itemx --suffix=@var{suffix}
818 @cindex backup suffix
819 @vindex SIMPLE_BACKUP_SUFFIX
820 Append @var{suffix} to each backup file made with @option{-b}. If this
821 option is not specified, the value of the @env{SIMPLE_BACKUP_SUFFIX}
822 environment variable is used. And if @env{SIMPLE_BACKUP_SUFFIX} is not
823 set, the default is @samp{~}, just as in Emacs.
832 Some @sc{gnu} programs (at least @command{df}, @command{du}, and
833 @command{ls}) display sizes in ``blocks''. You can adjust the block size
834 and method of display to make sizes easier to read. The block size
835 used for display is independent of any file system block size.
836 Fractional block counts are rounded up to the nearest integer.
838 @opindex --block-size=@var{size}
841 @vindex DF_BLOCK_SIZE
842 @vindex DU_BLOCK_SIZE
843 @vindex LS_BLOCK_SIZE
844 @vindex POSIXLY_CORRECT@r{, and block size}
846 The default block size is chosen by examining the following environment
847 variables in turn; the first one that is set determines the block size.
852 This specifies the default block size for the @command{df} command.
853 Similarly, @env{DU_BLOCK_SIZE} specifies the default for @command{du} and
854 @env{LS_BLOCK_SIZE} for @command{ls}.
857 This specifies the default block size for all three commands, if the
858 above command-specific environment variables are not set.
861 This specifies the default block size for all values that are normally
862 printed as blocks, if neither @env{BLOCK_SIZE} nor the above
863 command-specific environment variables are set. Unlike the other
864 environment variables, @env{BLOCKSIZE} does not affect values that are
865 normally printed as byte counts, e.g., the file sizes contained in
868 @item POSIXLY_CORRECT
869 If neither @env{@var{command}_BLOCK_SIZE}, nor @env{BLOCK_SIZE}, nor
870 @env{BLOCKSIZE} is set, but this variable is set, the block size
875 If none of the above environment variables are set, the block size
876 currently defaults to 1024 bytes in most contexts, but this number may
877 change in the future. For @command{ls} file sizes, the block size
880 @cindex human-readable output
883 A block size specification can be a positive integer specifying the number
884 of bytes per block, or it can be @code{human-readable} or @code{si} to
885 select a human-readable format. Integers may be followed by suffixes
886 that are upward compatible with the
887 @uref{http://www.bipm.fr/enus/3_SI/si-prefixes.html, SI prefixes}
888 for decimal multiples and with the
889 @uref{http://physics.nist.gov/cuu/Units/binary.html, IEC 60027-2
890 prefixes for binary multiples}.
892 With human-readable formats, output sizes are followed by a size letter
893 such as @samp{M} for megabytes. @code{BLOCK_SIZE=human-readable} uses
894 powers of 1024; @samp{M} stands for 1,048,576 bytes.
895 @code{BLOCK_SIZE=si} is similar, but uses powers of 1000 and appends
896 @samp{B}; @samp{MB} stands for 1,000,000 bytes.
899 A block size specification preceded by @samp{'} causes output sizes to
900 be displayed with thousands separators. The @env{LC_NUMERIC} locale
901 specifies the thousands separator and grouping. For example, in an
902 American English locale, @samp{--block-size="'1kB"} would cause a size
903 of 1234000 bytes to be displayed as @samp{1,234}. In the default C
904 locale, there is no thousands separator so a leading @samp{'} has no
907 An integer block size can be followed by a suffix to specify a
908 multiple of that size. A bare size letter,
909 or one followed by @samp{iB}, specifies
910 a multiple using powers of 1024. A size letter followed by @samp{B}
911 specifies powers of 1000 instead. For example, @samp{1M} and
912 @samp{1MiB} are equivalent to @samp{1048576}, whereas @samp{1MB} is
913 equivalent to @samp{1000000}.
915 A plain suffix without a preceding integer acts as if @samp{1} were
916 prepended, except that it causes a size indication to be appended to
917 the output. For example, @samp{--block-size="kB"} displays 3000 as
920 The following suffixes are defined. Large sizes like @code{1Y}
921 may be rejected by your computer due to limitations of its arithmetic.
925 @cindex kilobyte, definition of
926 kilobyte: @math{10^3 = 1000}.
930 @cindex kibibyte, definition of
931 kibibyte: @math{2^{10} = 1024}. @samp{K} is special: the SI prefix is
932 @samp{k} and the IEC 60027-2 prefix is @samp{Ki}, but tradition and
933 @acronym{POSIX} use @samp{k} to mean @samp{KiB}.
935 @cindex megabyte, definition of
936 megabyte: @math{10^6 = 1,000,000}.
939 @cindex mebibyte, definition of
940 mebibyte: @math{2^{20} = 1,048,576}.
942 @cindex gigabyte, definition of
943 gigabyte: @math{10^9 = 1,000,000,000}.
946 @cindex gibibyte, definition of
947 gibibyte: @math{2^{30} = 1,073,741,824}.
949 @cindex terabyte, definition of
950 terabyte: @math{10^{12} = 1,000,000,000,000}.
953 @cindex tebibyte, definition of
954 tebibyte: @math{2^{40} = 1,099,511,627,776}.
956 @cindex petabyte, definition of
957 petabyte: @math{10^{15} = 1,000,000,000,000,000}.
960 @cindex pebibyte, definition of
961 pebibyte: @math{2^{50} = 1,125,899,906,842,624}.
963 @cindex exabyte, definition of
964 exabyte: @math{10^{18} = 1,000,000,000,000,000,000}.
967 @cindex exbibyte, definition of
968 exbibyte: @math{2^{60} = 1,152,921,504,606,846,976}.
970 @cindex zettabyte, definition of
971 zettabyte: @math{10^{21} = 1,000,000,000,000,000,000,000}
974 @math{2^{70} = 1,180,591,620,717,411,303,424}.
975 (@samp{Zi} is a @acronym{GNU} extension to IEC 60027-2.)
977 @cindex yottabyte, definition of
978 yottabyte: @math{10^{24} = 1,000,000,000,000,000,000,000,000}.
981 @math{2^{80} = 1,208,925,819,614,629,174,706,176}.
982 (@samp{Yi} is a @acronym{GNU} extension to IEC 60027-2.)
987 @opindex --block-size
988 @opindex --human-readable
991 Block size defaults can be overridden by an explicit
992 @option{--block-size=@var{size}} option. The @option{-k}
993 option is equivalent to @option{--block-size=1K}, which
994 is the default unless the @env{POSIXLY_CORRECT} environment variable is
995 set. The @option{-h} or @option{--human-readable} option is equivalent to
996 @option{--block-size=human-readable}. The @option{--si} option is
997 equivalent to @option{--block-size=si}.
999 @node Signal specifications
1000 @section Signal specifications
1001 @cindex signals, specifying
1003 A @var{signal} may be a signal name like @samp{HUP}, or a signal
1004 number like @samp{1}, or an exit status of a process terminated by the
1005 signal. A signal name can be given in canonical form or prefixed by
1006 @samp{SIG}. The case of the letters is ignored. The following signal names
1007 and numbers are supported on all @acronym{POSIX} compliant systems:
1013 2. Terminal interrupt.
1019 9. Kill (cannot be caught or ignored).
1027 Other supported signal names have system-dependent corresponding
1028 numbers. All systems conforming to @acronym{POSIX} 1003.1-2001 also
1029 support the following signals:
1033 Access to an undefined portion of a memory object.
1035 Child process terminated, stopped, or continued.
1037 Continue executing, if stopped.
1039 Erroneous arithmetic operation.
1041 Illegal Instruction.
1043 Write on a pipe with no one to read it.
1045 Invalid memory reference.
1047 Stop executing (cannot be caught or ignored).
1051 Background process attempting read.
1053 Background process attempting write.
1055 High bandwidth data is available at a socket.
1057 User-defined signal 1.
1059 User-defined signal 2.
1063 @acronym{POSIX} 1003.1-2001 systems that support the @acronym{XSI} extension
1064 also support the following signals:
1070 Profiling timer expired.
1074 Trace/breakpoint trap.
1076 Virtual timer expired.
1078 CPU time limit exceeded.
1080 File size limit exceeded.
1084 @acronym{POSIX} 1003.1-2001 systems that support the @acronym{XRT} extension
1085 also support at least eight real-time signals called @samp{RTMIN},
1086 @samp{RTMIN+1}, @dots{}, @samp{RTMAX-1}, @samp{RTMAX}.
1088 @node Disambiguating names and IDs
1089 @section chown and chgrp: Disambiguating user names and IDs
1090 @cindex user names, disambiguating
1091 @cindex user IDs, disambiguating
1092 @cindex group names, disambiguating
1093 @cindex group IDs, disambiguating
1094 @cindex disambiguating group names and IDs
1096 Since the @var{owner} and @var{group} arguments to @command{chown} and
1097 @command{chgrp} may be specified as names or numeric IDs, there is an
1099 What if a user or group @emph{name} is a string of digits?
1100 @footnote{Using a number as a user name is common in some environments.}
1101 Should the command interpret it as a user name or as an ID?
1102 @acronym{POSIX} requires that @command{chown} and @command{chgrp}
1103 first attempt to resolve the specified string as a name, and
1104 only once that fails, then try to interpret it as an ID.
1105 This is troublesome when you want to specify a numeric ID, say 42,
1106 and it must work even in a pathological situation where
1107 @samp{42} is a user name that maps to some other user ID, say 1000.
1108 Simply invoking @code{chown 42 F}, will set @file{F}s owner ID to
1109 1000---not what you intended.
1111 GNU @command{chown} and @command{chgrp} provide a way to work around this,
1112 that at the same time may result in a significant performance improvement
1113 by eliminating a database look-up.
1114 Simply precede each numeric user ID and/or group ID with a @samp{+},
1115 in order to force its interpretation as an integer:
1119 chgrp +$numeric_group_id another-file
1123 GNU @command{chown} and @command{chgrp}
1124 skip the name look-up process for each @samp{+}-prefixed string,
1125 because a string containing @samp{+} is never a valid user or group name.
1126 This syntax is accepted on most common Unix systems, but not on Solaris 10.
1128 @node Random sources
1129 @section Sources of random data
1131 @cindex random sources
1133 The @command{shuf}, @command{shred}, and @command{sort} commands
1134 sometimes need random data to do their work. For example, @samp{sort
1135 -R} must choose a hash function at random, and it needs random data to
1136 make this selection.
1138 By default these commands use an internal pseudorandom generator
1139 initialized by a small amount of entropy, but can be directed to use
1140 an external source with the @option{--random-source=@var{file}} option.
1141 An error is reported if @var{file} does not contain enough bytes.
1143 For example, the device file @file{/dev/urandom} could be used as the
1144 source of random data. Typically, this device gathers environmental
1145 noise from device drivers and other sources into an entropy pool, and
1146 uses the pool to generate random bits. If the pool is short of data,
1147 the device reuses the internal pool to produce more bits, using a
1148 cryptographically secure pseudorandom number generator. But be aware
1149 that this device is not designed for bulk random data generation
1150 and is relatively slow.
1152 @file{/dev/urandom} suffices for most practical uses, but applications
1153 requiring high-value or long-term protection of private data may
1154 require an alternate data source like @file{/dev/random} or
1155 @file{/dev/arandom}. The set of available sources depends on your
1158 To reproduce the results of an earlier invocation of a command, you
1159 can save some random data into a file and then use that file as the
1160 random source in earlier and later invocations of the command.
1162 @node Target directory
1163 @section Target directory
1165 @cindex target directory
1167 The @command{cp}, @command{install}, @command{ln}, and @command{mv}
1168 commands normally treat the last operand specially when it is a
1169 directory or a symbolic link to a directory. For example, @samp{cp
1170 source dest} is equivalent to @samp{cp source dest/source} if
1171 @file{dest} is a directory. Sometimes this behavior is not exactly
1172 what is wanted, so these commands support the following options to
1173 allow more fine-grained control:
1178 @itemx --no-target-directory
1179 @opindex --no-target-directory
1180 @cindex target directory
1181 @cindex destination directory
1182 Do not treat the last operand specially when it is a directory or a
1183 symbolic link to a directory. This can help avoid race conditions in
1184 programs that operate in a shared area. For example, when the command
1185 @samp{mv /tmp/source /tmp/dest} succeeds, there is no guarantee that
1186 @file{/tmp/source} was renamed to @file{/tmp/dest}: it could have been
1187 renamed to @file{/tmp/dest/source} instead, if some other process
1188 created @file{/tmp/dest} as a directory. However, if @file{mv
1189 -T /tmp/source /tmp/dest} succeeds, there is no
1190 question that @file{/tmp/source} was renamed to @file{/tmp/dest}.
1192 In the opposite situation, where you want the last operand to be
1193 treated as a directory and want a diagnostic otherwise, you can use
1194 the @option{--target-directory} (@option{-t}) option.
1196 @item -t @var{directory}
1197 @itemx @w{@kbd{--target-directory}=@var{directory}}
1198 @opindex --target-directory
1199 @cindex target directory
1200 @cindex destination directory
1201 Use @var{directory} as the directory component of each destination
1204 The interface for most programs is that after processing options and a
1205 finite (possibly zero) number of fixed-position arguments, the remaining
1206 argument list is either expected to be empty, or is a list of items
1207 (usually files) that will all be handled identically. The @command{xargs}
1208 program is designed to work well with this convention.
1210 The commands in the @command{mv}-family are unusual in that they take
1211 a variable number of arguments with a special case at the @emph{end}
1212 (namely, the target directory). This makes it nontrivial to perform some
1213 operations, e.g., ``move all files from here to ../d/'', because
1214 @code{mv * ../d/} might exhaust the argument space, and @code{ls | xargs ...}
1215 doesn't have a clean way to specify an extra final argument for each
1216 invocation of the subject command. (It can be done by going through a
1217 shell command, but that requires more human labor and brain power than
1220 The @w{@kbd{--target-directory}} (@option{-t}) option allows the @command{cp},
1221 @command{install}, @command{ln}, and @command{mv} programs to be used
1222 conveniently with @command{xargs}. For example, you can move the files
1223 from the current directory to a sibling directory, @code{d} like this:
1226 ls | xargs mv -t ../d --
1229 However, this doesn't move files whose names begin with @samp{.}.
1230 If you use the @sc{gnu} @command{find} program, you can move those
1231 files too, with this command:
1234 find . -mindepth 1 -maxdepth 1 \
1238 But both of the above approaches fail if there are no files in the
1239 current directory, or if any file has a name containing a blank or
1240 some other special characters.
1241 The following example removes those limitations and requires both
1242 @sc{gnu} @command{find} and @sc{gnu} @command{xargs}:
1245 find . -mindepth 1 -maxdepth 1 -print0 \
1246 | xargs --null --no-run-if-empty \
1253 The @option{--target-directory} (@option{-t}) and
1254 @option{--no-target-directory} (@option{-T})
1255 options cannot be combined.
1257 @node Trailing slashes
1258 @section Trailing slashes
1260 @cindex trailing slashes
1262 Some @sc{gnu} programs (at least @command{cp} and @command{mv}) allow you to
1263 remove any trailing slashes from each @var{source} argument before
1264 operating on it. The @w{@kbd{--strip-trailing-slashes}} option enables
1267 This is useful when a @var{source} argument may have a trailing slash and
1268 @c FIXME: mv's behavior in this case is system-dependent
1269 specify a symbolic link to a directory. This scenario is in fact rather
1270 common because some shells can automatically append a trailing slash when
1271 performing file name completion on such symbolic links. Without this
1272 option, @command{mv}, for example, (via the system's rename function) must
1273 interpret a trailing slash as a request to dereference the symbolic link
1274 and so must rename the indirectly referenced @emph{directory} and not
1275 the symbolic link. Although it may seem surprising that such behavior
1276 be the default, it is required by @acronym{POSIX} and is consistent with
1277 other parts of that standard.
1279 @node Traversing symlinks
1280 @section Traversing symlinks
1282 @cindex symbolic link to directory, controlling traversal of
1284 The following options modify how @command{chown} and @command{chgrp}
1285 @c FIXME: note that `du' has these options, too, but they have slightly
1286 @c different meaning.
1287 traverse a hierarchy when the @option{--recursive} (@option{-R})
1288 option is also specified.
1289 If more than one of the following options is specified, only the final
1291 These options specify whether processing a symbolic link to a directory
1292 entails operating on just the symbolic link or on all files in the
1293 hierarchy rooted at that directory.
1295 These options are independent of @option{--dereference} and
1296 @option{--no-dereference} (@option{-h}), which control whether to modify
1297 a symlink or its referent.
1304 @cindex symbolic link to directory, traverse each that is specified on the command line
1305 If @option{--recursive} (@option{-R}) is specified and
1306 a command line argument is a symbolic link to a directory, traverse it.
1313 @cindex symbolic link to directory, traverse each that is encountered
1314 In a recursive traversal, traverse every symbolic link to a directory
1315 that is encountered.
1322 @cindex symbolic link to directory, never traverse
1323 Do not traverse any symbolic links.
1324 This is the default if none of @option{-H}, @option{-L},
1325 or @option{-P} is specified.
1332 @node Treating / specially
1333 @section Treating @file{/} specially
1335 Certain commands can operate destructively on entire hierarchies.
1336 For example, if a user with appropriate privileges mistakenly runs
1337 @samp{rm -rf / tmp/junk}, that may remove
1338 all files on the entire system. Since there are so few
1339 legitimate uses for such a command,
1340 @sc{gnu} @command{rm} normally declines to operate on any directory
1341 that resolves to @file{/}. If you really want to try to remove all
1342 the files on your system, you can use the @option{--no-preserve-root}
1343 option, but the default behavior, specified by the
1344 @option{--preserve-option}, is safer for most purposes.
1346 The commands @command{chgrp}, @command{chmod} and @command{chown}
1347 can also operate destructively on entire hierarchies, so they too
1348 support these options. Although, unlike @command{rm}, they don't
1349 actually unlink files, these commands are arguably more dangerous
1350 when operating recursively on @file{/}, since they often work much
1351 more quickly, and hence damage more files before an alert user can
1352 interrupt them. Tradition and @acronym{POSIX} require these commands
1353 to operate recursively on @file{/}, so they default to
1354 @option{--no-preserve-root}, but using the @option{--preserve-root}
1355 option makes them safer for most purposes. For convenience you can
1356 specify @option{--preserve-root} in an alias or in a shell function.
1358 Note that the @option{--preserve-root} option also ensures
1359 that @command{chgrp} and @command{chown} do not modify @file{/}
1360 even when dereferencing a symlink pointing to @file{/}.
1362 @node Special built-in utilities
1363 @section Special built-in utilities
1365 Some programs like @command{nice} can invoke other programs; for
1366 example, the command @samp{nice cat file} invokes the program
1367 @command{cat} by executing the command @samp{cat file}. However,
1368 @dfn{special built-in utilities} like @command{exit} cannot be invoked
1369 this way. For example, the command @samp{nice exit} does not have a
1370 well-defined behavior: it may generate an error message instead of
1373 Here is a list of the special built-in utilities that are standardized
1374 by @acronym{POSIX} 1003.1-2004.
1377 @t{.@: : break continue eval exec exit export readonly
1378 return set shift times trap unset}
1381 For example, because @samp{.}, @samp{:}, and @samp{exec} are special,
1382 the commands @samp{nice . foo.sh}, @samp{nice :}, and @samp{nice exec
1383 pwd} do not work as you might expect.
1385 Many shells extend this list. For example, Bash has several extra
1386 special built-in utilities like @command{history}, and
1387 @command{suspend}, and with Bash the command @samp{nice suspend}
1388 generates an error message instead of suspending.
1390 @node Standards conformance
1391 @section Standards conformance
1393 @vindex POSIXLY_CORRECT
1394 In a few cases, the @sc{gnu} utilities' default behavior is
1395 incompatible with the @acronym{POSIX} standard. To suppress these
1396 incompatibilities, define the @env{POSIXLY_CORRECT} environment
1397 variable. Unless you are checking for @acronym{POSIX} conformance, you
1398 probably do not need to define @env{POSIXLY_CORRECT}.
1400 Newer versions of @acronym{POSIX} are occasionally incompatible with older
1401 versions. For example, older versions of @acronym{POSIX} required the
1402 command @samp{sort +1} to sort based on the second and succeeding
1403 fields in each input line, but starting with @acronym{POSIX} 1003.1-2001
1404 the same command is required to sort the file named @file{+1}, and you
1405 must instead use the command @samp{sort -k 2} to get the field-based
1408 @vindex _POSIX2_VERSION
1409 The @sc{gnu} utilities normally conform to the version of @acronym{POSIX}
1410 that is standard for your system. To cause them to conform to a
1411 different version of @acronym{POSIX}, define the @env{_POSIX2_VERSION}
1412 environment variable to a value of the form @var{yyyymm} specifying
1413 the year and month the standard was adopted. Two values are currently
1414 supported for @env{_POSIX2_VERSION}: @samp{199209} stands for
1415 @acronym{POSIX} 1003.2-1992, and @samp{200112} stands for @acronym{POSIX}
1416 1003.1-2001. For example, if you have a newer system but are running software
1417 that assumes an older version of @acronym{POSIX} and uses @samp{sort +1}
1418 or @samp{tail +10}, you can work around any compatibility problems by setting
1419 @samp{_POSIX2_VERSION=199209} in your environment.
1421 @node Output of entire files
1422 @chapter Output of entire files
1424 @cindex output of entire files
1425 @cindex entire files, output of
1427 These commands read and write entire files, possibly transforming them
1431 * cat invocation:: Concatenate and write files.
1432 * tac invocation:: Concatenate and write files in reverse.
1433 * nl invocation:: Number lines and write files.
1434 * od invocation:: Write files in octal or other formats.
1435 * base64 invocation:: Transform data into printable data.
1438 @node cat invocation
1439 @section @command{cat}: Concatenate and write files
1442 @cindex concatenate and write files
1443 @cindex copying files
1445 @command{cat} copies each @var{file} (@samp{-} means standard input), or
1446 standard input if none are given, to standard output. Synopsis:
1449 cat [@var{option}] [@var{file}]@dots{}
1452 The program accepts the following options. Also see @ref{Common options}.
1460 Equivalent to @option{-vET}.
1463 @itemx --number-nonblank
1465 @opindex --number-nonblank
1466 Number all nonempty output lines, starting with 1.
1470 Equivalent to @option{-vE}.
1475 @opindex --show-ends
1476 Display a @samp{$} after the end of each line.
1482 Number all output lines, starting with 1.
1485 @itemx --squeeze-blank
1487 @opindex --squeeze-blank
1488 @cindex squeezing empty lines
1489 Suppress repeated adjacent empty lines; output just one empty line
1494 Equivalent to @option{-vT}.
1499 @opindex --show-tabs
1500 Display TAB characters as @samp{^I}.
1504 Ignored; for @acronym{POSIX} compatibility.
1507 @itemx --show-nonprinting
1509 @opindex --show-nonprinting
1510 Display control characters except for LFD and TAB using
1511 @samp{^} notation and precede characters that have the high bit set with
1516 On systems like MS-DOS that distinguish between text and binary files,
1517 @command{cat} normally reads and writes in binary mode. However,
1518 @command{cat} reads in text mode if one of the options
1519 @option{-bensAE} is used or if @command{cat} is reading from standard
1520 input and standard input is a terminal. Similarly, @command{cat}
1521 writes in text mode if one of the options @option{-bensAE} is used or
1522 if standard output is a terminal.
1529 # Output f's contents, then standard input, then g's contents.
1532 # Copy standard input to standard output.
1537 @node tac invocation
1538 @section @command{tac}: Concatenate and write files in reverse
1541 @cindex reversing files
1543 @command{tac} copies each @var{file} (@samp{-} means standard input), or
1544 standard input if none are given, to standard output, reversing the
1545 records (lines by default) in each separately. Synopsis:
1548 tac [@var{option}]@dots{} [@var{file}]@dots{}
1551 @dfn{Records} are separated by instances of a string (newline by
1552 default). By default, this separator string is attached to the end of
1553 the record that it follows in the file.
1555 The program accepts the following options. Also see @ref{Common options}.
1563 The separator is attached to the beginning of the record that it
1564 precedes in the file.
1570 Treat the separator string as a regular expression. Users of @command{tac}
1571 on MS-DOS/MS-Windows should note that, since @command{tac} reads files in
1572 binary mode, each line of a text file might end with a CR/LF pair
1573 instead of the Unix-style LF.
1575 @item -s @var{separator}
1576 @itemx --separator=@var{separator}
1578 @opindex --separator
1579 Use @var{separator} as the record separator, instead of newline.
1587 @section @command{nl}: Number lines and write files
1590 @cindex numbering lines
1591 @cindex line numbering
1593 @command{nl} writes each @var{file} (@samp{-} means standard input), or
1594 standard input if none are given, to standard output, with line numbers
1595 added to some or all of the lines. Synopsis:
1598 nl [@var{option}]@dots{} [@var{file}]@dots{}
1601 @cindex logical pages, numbering on
1602 @command{nl} decomposes its input into (logical) pages; by default, the
1603 line number is reset to 1 at the top of each logical page. @command{nl}
1604 treats all of the input files as a single document; it does not reset
1605 line numbers or logical pages between files.
1607 @cindex headers, numbering
1608 @cindex body, numbering
1609 @cindex footers, numbering
1610 A logical page consists of three sections: header, body, and footer.
1611 Any of the sections can be empty. Each can be numbered in a different
1612 style from the others.
1614 The beginnings of the sections of logical pages are indicated in the
1615 input file by a line containing exactly one of these delimiter strings:
1626 The two characters from which these strings are made can be changed from
1627 @samp{\} and @samp{:} via options (see below), but the pattern and
1628 length of each string cannot be changed.
1630 A section delimiter is replaced by an empty line on output. Any text
1631 that comes before the first section delimiter string in the input file
1632 is considered to be part of a body section, so @command{nl} treats a
1633 file that contains no section delimiters as a single body section.
1635 The program accepts the following options. Also see @ref{Common options}.
1639 @item -b @var{style}
1640 @itemx --body-numbering=@var{style}
1642 @opindex --body-numbering
1643 Select the numbering style for lines in the body section of each
1644 logical page. When a line is not numbered, the current line number
1645 is not incremented, but the line number separator character is still
1646 prepended to the line. The styles are:
1652 number only nonempty lines (default for body),
1654 do not number lines (default for header and footer),
1656 number only lines that contain a match for the basic regular
1657 expression @var{bre}.
1658 @xref{Regular Expressions, , Regular Expressions, grep, The GNU Grep Manual}.
1662 @itemx --section-delimiter=@var{cd}
1664 @opindex --section-delimiter
1665 @cindex section delimiters of pages
1666 Set the section delimiter characters to @var{cd}; default is
1667 @samp{\:}. If only @var{c} is given, the second remains @samp{:}.
1668 (Remember to protect @samp{\} or other metacharacters from shell
1669 expansion with quotes or extra backslashes.)
1671 @item -f @var{style}
1672 @itemx --footer-numbering=@var{style}
1674 @opindex --footer-numbering
1675 Analogous to @option{--body-numbering}.
1677 @item -h @var{style}
1678 @itemx --header-numbering=@var{style}
1680 @opindex --header-numbering
1681 Analogous to @option{--body-numbering}.
1683 @item -i @var{number}
1684 @itemx --line-increment=@var{number}
1686 @opindex --line-increment
1687 Increment line numbers by @var{number} (default 1).
1689 @item -l @var{number}
1690 @itemx --join-blank-lines=@var{number}
1692 @opindex --join-blank-lines
1693 @cindex empty lines, numbering
1694 @cindex blank lines, numbering
1695 Consider @var{number} (default 1) consecutive empty lines to be one
1696 logical line for numbering, and only number the last one. Where fewer
1697 than @var{number} consecutive empty lines occur, do not number them.
1698 An empty line is one that contains no characters, not even spaces
1701 @item -n @var{format}
1702 @itemx --number-format=@var{format}
1704 @opindex --number-format
1705 Select the line numbering format (default is @code{rn}):
1709 @opindex ln @r{format for @command{nl}}
1710 left justified, no leading zeros;
1712 @opindex rn @r{format for @command{nl}}
1713 right justified, no leading zeros;
1715 @opindex rz @r{format for @command{nl}}
1716 right justified, leading zeros.
1720 @itemx --no-renumber
1722 @opindex --no-renumber
1723 Do not reset the line number at the start of a logical page.
1725 @item -s @var{string}
1726 @itemx --number-separator=@var{string}
1728 @opindex --number-separator
1729 Separate the line number from the text line in the output with
1730 @var{string} (default is the TAB character).
1732 @item -v @var{number}
1733 @itemx --starting-line-number=@var{number}
1735 @opindex --starting-line-number
1736 Set the initial line number on each logical page to @var{number} (default 1).
1738 @item -w @var{number}
1739 @itemx --number-width=@var{number}
1741 @opindex --number-width
1742 Use @var{number} characters for line numbers (default 6).
1750 @section @command{od}: Write files in octal or other formats
1753 @cindex octal dump of files
1754 @cindex hex dump of files
1755 @cindex ASCII dump of files
1756 @cindex file contents, dumping unambiguously
1758 @command{od} writes an unambiguous representation of each @var{file}
1759 (@samp{-} means standard input), or standard input if none are given.
1763 od [@var{option}]@dots{} [@var{file}]@dots{}
1764 od [-abcdfilosx]@dots{} [@var{file}] [[+]@var{offset}[.][b]]
1765 od [@var{option}]@dots{} --traditional [@var{file}] [[+]@var{offset}[.][b] [[+]@var{label}[.][b]]]
1768 Each line of output consists of the offset in the input, followed by
1769 groups of data from the file. By default, @command{od} prints the offset in
1770 octal, and each group of file data is a C @code{short int}'s worth of input
1771 printed as a single octal number.
1773 If @var{offset} is given, it specifies how many input bytes to skip
1774 before formatting and writing. By default, it is interpreted as an
1775 octal number, but the optional trailing decimal point causes it to be
1776 interpreted as decimal. If no decimal is specified and the offset
1777 begins with @samp{0x} or @samp{0X} it is interpreted as a hexadecimal
1778 number. If there is a trailing @samp{b}, the number of bytes skipped
1779 will be @var{offset} multiplied by 512.
1781 If a command is of both the first and second forms, the second form is
1782 assumed if the last operand begins with @samp{+} or (if there are two
1783 operands) a digit. For example, in @samp{od foo 10} and @samp{od +10}
1784 the @samp{10} is an offset, whereas in @samp{od 10} the @samp{10} is a
1787 The program accepts the following options. Also see @ref{Common options}.
1791 @item -A @var{radix}
1792 @itemx --address-radix=@var{radix}
1794 @opindex --address-radix
1795 @cindex radix for file offsets
1796 @cindex file offset radix
1797 Select the base in which file offsets are printed. @var{radix} can
1798 be one of the following:
1808 none (do not print offsets).
1811 The default is octal.
1813 @item -j @var{bytes}
1814 @itemx --skip-bytes=@var{bytes}
1816 @opindex --skip-bytes
1817 Skip @var{bytes} input bytes before formatting and writing. If
1818 @var{bytes} begins with @samp{0x} or @samp{0X}, it is interpreted in
1819 hexadecimal; otherwise, if it begins with @samp{0}, in octal; otherwise,
1821 @multiplierSuffixes{bytes}
1823 @item -N @var{bytes}
1824 @itemx --read-bytes=@var{bytes}
1826 @opindex --read-bytes
1827 Output at most @var{bytes} bytes of the input. Prefixes and suffixes on
1828 @code{bytes} are interpreted as for the @option{-j} option.
1830 @item -S @var{bytes}
1831 @itemx --strings[=@var{bytes}]
1834 @cindex string constants, outputting
1835 Instead of the normal output, output only @dfn{string constants}: at
1836 least @var{bytes} consecutive @acronym{ASCII} graphic characters,
1837 followed by a zero byte (@acronym{ASCII} @sc{nul}).
1838 Prefixes and suffixes on @code{bytes} are interpreted as for the
1841 If @var{n} is omitted with @option{--strings}, the default is 3.
1844 @itemx --format=@var{type}
1847 Select the format in which to output the file data. @var{type} is a
1848 string of one or more of the below type indicator characters. If you
1849 include more than one type indicator character in a single @var{type}
1850 string, or use this option more than once, @command{od} writes one copy
1851 of each output line using each of the data types that you specified,
1852 in the order that you specified.
1854 Adding a trailing ``z'' to any type specification appends a display
1855 of the @acronym{ASCII} character representation of the printable characters
1856 to the output line generated by the type specification.
1860 named character, ignoring high-order bit
1862 @acronym{ASCII} character or backslash escape,
1875 The type @code{a} outputs things like @samp{sp} for space, @samp{nl} for
1876 newline, and @samp{nul} for a zero byte. Only the least significant
1877 seven bits of each byte is used; the high-order bit is ignored.
1878 Type @code{c} outputs
1879 @samp{ }, @samp{\n}, and @code{\0}, respectively.
1882 Except for types @samp{a} and @samp{c}, you can specify the number
1883 of bytes to use in interpreting each number in the given data type
1884 by following the type indicator character with a decimal integer.
1885 Alternately, you can specify the size of one of the C compiler's
1886 built-in data types by following the type indicator character with
1887 one of the following characters. For integers (@samp{d}, @samp{o},
1888 @samp{u}, @samp{x}):
1901 For floating point (@code{f}):
1913 @itemx --output-duplicates
1915 @opindex --output-duplicates
1916 Output consecutive lines that are identical. By default, when two or
1917 more consecutive output lines would be identical, @command{od} outputs only
1918 the first line, and puts just an asterisk on the following line to
1919 indicate the elision.
1922 @itemx --width[=@var{n}]
1925 Dump @code{n} input bytes per output line. This must be a multiple of
1926 the least common multiple of the sizes associated with the specified
1929 If this option is not given at all, the default is 16. If @var{n} is
1930 omitted, the default is 32.
1934 The next several options are shorthands for format specifications.
1935 @sc{gnu} @command{od} accepts any combination of shorthands and format
1936 specification options. These options accumulate.
1942 Output as named characters. Equivalent to @samp{-t a}.
1946 Output as octal bytes. Equivalent to @samp{-t o1}.
1950 Output as @acronym{ASCII} characters or backslash escapes. Equivalent to
1955 Output as unsigned decimal two-byte units. Equivalent to @samp{-t u2}.
1959 Output as floats. Equivalent to @samp{-t fF}.
1963 Output as decimal ints. Equivalent to @samp{-t dI}.
1967 Output as decimal long ints. Equivalent to @samp{-t dL}.
1971 Output as octal two-byte units. Equivalent to @option{-t o2}.
1975 Output as decimal two-byte units. Equivalent to @option{-t d2}.
1979 Output as hexadecimal two-byte units. Equivalent to @samp{-t x2}.
1982 @opindex --traditional
1983 Recognize the non-option label argument that traditional @command{od}
1984 accepted. The following syntax:
1987 od --traditional [@var{file}] [[+]@var{offset}[.][b] [[+]@var{label}[.][b]]]
1991 can be used to specify at most one file and optional arguments
1992 specifying an offset and a pseudo-start address, @var{label}.
1993 The @var{label} argument is interpreted
1994 just like @var{offset}, but it specifies an initial pseudo-address. The
1995 pseudo-addresses are displayed in parentheses following any normal
2002 @node base64 invocation
2003 @section @command{base64}: Transform data into printable data
2006 @cindex base64 encoding
2008 @command{base64} transforms data read from a file, or standard input,
2009 into (or from) base64 encoded form. The base64 encoded form uses
2010 printable @acronym{ASCII} characters to represent binary data.
2014 base64 [@var{option}]@dots{} [@var{file}]
2015 base64 --decode [@var{option}]@dots{} [@var{file}]
2018 The base64 encoding expands data to roughly 133% of the original.
2019 The format conforms to
2020 @uref{ftp://ftp.rfc-editor.org/in-notes/rfc4648.txt, RFC 4648}.
2022 The program accepts the following options. Also see @ref{Common options}.
2027 @itemx --wrap=@var{cols}
2031 @cindex column to wrap data after
2032 During encoding, wrap lines after @var{cols} characters. This must be
2035 The default is to wrap after 76 characters. Use the value 0 to
2036 disable line wrapping altogether.
2042 @cindex Decode base64 data
2043 @cindex Base64 decoding
2044 Change the mode of operation, from the default of encoding data, to
2045 decoding data. Input is expected to be base64 encoded data, and the
2046 output will be the original data.
2049 @itemx --ignore-garbage
2051 @opindex --ignore-garbage
2052 @cindex Ignore garbage in base64 stream
2053 When decoding, newlines are always accepted.
2054 During decoding, ignore unrecognized bytes,
2055 to permit distorted data to be decoded.
2062 @node Formatting file contents
2063 @chapter Formatting file contents
2065 @cindex formatting file contents
2067 These commands reformat the contents of files.
2070 * fmt invocation:: Reformat paragraph text.
2071 * pr invocation:: Paginate or columnate files for printing.
2072 * fold invocation:: Wrap input lines to fit in specified width.
2076 @node fmt invocation
2077 @section @command{fmt}: Reformat paragraph text
2080 @cindex reformatting paragraph text
2081 @cindex paragraphs, reformatting
2082 @cindex text, reformatting
2084 @command{fmt} fills and joins lines to produce output lines of (at most)
2085 a given number of characters (75 by default). Synopsis:
2088 fmt [@var{option}]@dots{} [@var{file}]@dots{}
2091 @command{fmt} reads from the specified @var{file} arguments (or standard
2092 input if none are given), and writes to standard output.
2094 By default, blank lines, spaces between words, and indentation are
2095 preserved in the output; successive input lines with different
2096 indentation are not joined; tabs are expanded on input and introduced on
2099 @cindex line-breaking
2100 @cindex sentences and line-breaking
2101 @cindex Knuth, Donald E.
2102 @cindex Plass, Michael F.
2103 @command{fmt} prefers breaking lines at the end of a sentence, and tries to
2104 avoid line breaks after the first word of a sentence or before the last
2105 word of a sentence. A @dfn{sentence break} is defined as either the end
2106 of a paragraph or a word ending in any of @samp{.?!}, followed by two
2107 spaces or end of line, ignoring any intervening parentheses or quotes.
2108 Like @TeX{}, @command{fmt} reads entire ``paragraphs'' before choosing line
2109 breaks; the algorithm is a variant of that given by Donald E. Knuth
2110 and Michael F. Plass in ``Breaking Paragraphs Into Lines'',
2111 @cite{Software---Practice & Experience} @b{11}, 11 (November 1981),
2114 The program accepts the following options. Also see @ref{Common options}.
2119 @itemx --crown-margin
2121 @opindex --crown-margin
2122 @cindex crown margin
2123 @dfn{Crown margin} mode: preserve the indentation of the first two
2124 lines within a paragraph, and align the left margin of each subsequent
2125 line with that of the second line.
2128 @itemx --tagged-paragraph
2130 @opindex --tagged-paragraph
2131 @cindex tagged paragraphs
2132 @dfn{Tagged paragraph} mode: like crown margin mode, except that if
2133 indentation of the first line of a paragraph is the same as the
2134 indentation of the second, the first line is treated as a one-line
2140 @opindex --split-only
2141 Split lines only. Do not join short lines to form longer ones. This
2142 prevents sample lines of code, and other such ``formatted'' text from
2143 being unduly combined.
2146 @itemx --uniform-spacing
2148 @opindex --uniform-spacing
2149 Uniform spacing. Reduce spacing between words to one space, and spacing
2150 between sentences to two spaces.
2153 @itemx -w @var{width}
2154 @itemx --width=@var{width}
2155 @opindex -@var{width}
2158 Fill output lines up to @var{width} characters (default 75). @command{fmt}
2159 initially tries to make lines about 7% shorter than this, to give it
2160 room to balance line lengths.
2162 @item -p @var{prefix}
2163 @itemx --prefix=@var{prefix}
2164 Only lines beginning with @var{prefix} (possibly preceded by whitespace)
2165 are subject to formatting. The prefix and any preceding whitespace are
2166 stripped for the formatting and then re-attached to each formatted output
2167 line. One use is to format certain kinds of program comments, while
2168 leaving the code unchanged.
2176 @section @command{pr}: Paginate or columnate files for printing
2179 @cindex printing, preparing files for
2180 @cindex multicolumn output, generating
2181 @cindex merging files in parallel
2183 @command{pr} writes each @var{file} (@samp{-} means standard input), or
2184 standard input if none are given, to standard output, paginating and
2185 optionally outputting in multicolumn format; optionally merges all
2186 @var{file}s, printing all in parallel, one per column. Synopsis:
2189 pr [@var{option}]@dots{} [@var{file}]@dots{}
2193 By default, a 5-line header is printed at each page: two blank lines;
2194 a line with the date, the file name, and the page count; and two more
2195 blank lines. A footer of five blank lines is also printed.
2196 The default @var{page_length} is 66
2197 lines. The default number of text lines is therefore 56.
2198 The text line of the header takes the form
2199 @samp{@var{date} @var{string} @var{page}}, with spaces inserted around
2200 @var{string} so that the line takes up the full @var{page_width}. Here,
2201 @var{date} is the date (see the @option{-D} or @option{--date-format}
2202 option for details), @var{string} is the centered header string, and
2203 @var{page} identifies the page number. The @env{LC_MESSAGES} locale
2204 category affects the spelling of @var{page}; in the default C locale, it
2205 is @samp{Page @var{number}} where @var{number} is the decimal page
2208 Form feeds in the input cause page breaks in the output. Multiple form
2209 feeds produce empty pages.
2211 Columns are of equal width, separated by an optional string (default
2212 is @samp{space}). For multicolumn output, lines will always be truncated to
2213 @var{page_width} (default 72), unless you use the @option{-J} option.
2215 column output no line truncation occurs by default. Use @option{-W} option to
2216 truncate lines in that case.
2218 The following changes were made in version 1.22i and apply to later
2219 versions of @command{pr}:
2220 @c FIXME: this whole section here sounds very awkward to me. I
2221 @c made a few small changes, but really it all needs to be redone. - Brian
2222 @c OK, I fixed another sentence or two, but some of it I just don't understand.
2227 Some small @var{letter options} (@option{-s}, @option{-w}) have been
2228 redefined for better @acronym{POSIX} compliance. The output of some further
2229 cases has been adapted to other Unix systems. These changes are not
2230 compatible with earlier versions of the program.
2233 Some @var{new capital letter} options (@option{-J}, @option{-S}, @option{-W})
2234 have been introduced to turn off unexpected interferences of small letter
2235 options. The @option{-N} option and the second argument @var{last_page}
2236 of @samp{+FIRST_PAGE} offer more flexibility. The detailed handling of
2237 form feeds set in the input files requires the @option{-T} option.
2240 Capital letter options override small letter ones.
2243 Some of the option-arguments (compare @option{-s}, @option{-e},
2244 @option{-i}, @option{-n}) cannot be specified as separate arguments from the
2245 preceding option letter (already stated in the @acronym{POSIX} specification).
2248 The program accepts the following options. Also see @ref{Common options}.
2252 @item +@var{first_page}[:@var{last_page}]
2253 @itemx --pages=@var{first_page}[:@var{last_page}]
2254 @c The two following @opindex lines evoke warnings because they contain `:'
2255 @c The `info' spec does not permit that. If we use those lines, we end
2256 @c up with truncated index entries that don't work.
2257 @c @opindex +@var{first_page}[:@var{last_page}]
2258 @c @opindex --pages=@var{first_page}[:@var{last_page}]
2259 @opindex +@var{page_range}
2260 @opindex --pages=@var{page_range}
2261 Begin printing with page @var{first_page} and stop with @var{last_page}.
2262 Missing @samp{:@var{last_page}} implies end of file. While estimating
2263 the number of skipped pages each form feed in the input file results
2264 in a new page. Page counting with and without @samp{+@var{first_page}}
2265 is identical. By default, counting starts with the first page of input
2266 file (not first page printed). Line numbering may be altered by @option{-N}
2270 @itemx --columns=@var{column}
2271 @opindex -@var{column}
2273 @cindex down columns
2274 With each single @var{file}, produce @var{column} columns of output
2275 (default is 1) and print columns down, unless @option{-a} is used. The
2276 column width is automatically decreased as @var{column} increases; unless
2277 you use the @option{-W/-w} option to increase @var{page_width} as well.
2278 This option might well cause some lines to be truncated. The number of
2279 lines in the columns on each page are balanced. The options @option{-e}
2280 and @option{-i} are on for multiple text-column output. Together with
2281 @option{-J} option column alignment and line truncation is turned off.
2282 Lines of full length are joined in a free field format and @option{-S}
2283 option may set field separators. @option{-@var{column}} may not be used
2284 with @option{-m} option.
2290 @cindex across columns
2291 With each single @var{file}, print columns across rather than down. The
2292 @option{-@var{column}} option must be given with @var{column} greater than one.
2293 If a line is too long to fit in a column, it is truncated.
2296 @itemx --show-control-chars
2298 @opindex --show-control-chars
2299 Print control characters using hat notation (e.g., @samp{^G}); print
2300 other nonprinting characters in octal backslash notation. By default,
2301 nonprinting characters are not changed.
2304 @itemx --double-space
2306 @opindex --double-space
2307 @cindex double spacing
2308 Double space the output.
2310 @item -D @var{format}
2311 @itemx --date-format=@var{format}
2312 @cindex time formats
2313 @cindex formatting times
2314 Format header dates using @var{format}, using the same conventions as
2315 for the command @samp{date +@var{format}}; @xref{date invocation}.
2316 Except for directives, which start with
2317 @samp{%}, characters in @var{format} are printed unchanged. You can use
2318 this option to specify an arbitrary string in place of the header date,
2319 e.g., @option{--date-format="Monday morning"}.
2321 @vindex POSIXLY_CORRECT
2323 The default date format is @samp{%Y-%m-%d %H:%M} (for example,
2324 @samp{2001-12-04 23:59});
2325 but if the @env{POSIXLY_CORRECT} environment variable is set
2326 and the @env{LC_TIME} locale category specifies the @acronym{POSIX}
2327 locale, the default is @samp{%b %e %H:%M %Y} (for example,
2328 @samp{Dec@ @ 4 23:59 2001}.
2331 Time stamps are listed according to the time zone rules specified by
2332 the @env{TZ} environment variable, or by the system default rules if
2333 @env{TZ} is not set. @xref{TZ Variable,, Specifying the Time Zone
2334 with @env{TZ}, libc, The GNU C Library Reference Manual}.
2336 @item -e[@var{in-tabchar}[@var{in-tabwidth}]]
2337 @itemx --expand-tabs[=@var{in-tabchar}[@var{in-tabwidth}]]
2339 @opindex --expand-tabs
2341 Expand @var{tab}s to spaces on input. Optional argument @var{in-tabchar} is
2342 the input tab character (default is the TAB character). Second optional
2343 argument @var{in-tabwidth} is the input tab character's width (default
2351 @opindex --form-feed
2352 Use a form feed instead of newlines to separate output pages. This does
2353 not alter the default page length of 66 lines.
2355 @item -h @var{header}
2356 @itemx --header=@var{header}
2359 Replace the file name in the header with the centered string @var{header}.
2360 When using the shell, @var{header} should be quoted and should be
2361 separated from @option{-h} by a space.
2363 @item -i[@var{out-tabchar}[@var{out-tabwidth}]]
2364 @itemx --output-tabs[=@var{out-tabchar}[@var{out-tabwidth}]]
2366 @opindex --output-tabs
2368 Replace spaces with @var{tab}s on output. Optional argument @var{out-tabchar}
2369 is the output tab character (default is the TAB character). Second optional
2370 argument @var{out-tabwidth} is the output tab character's width (default
2376 @opindex --join-lines
2377 Merge lines of full length. Used together with the column options
2378 @option{-@var{column}}, @option{-a -@var{column}} or @option{-m}. Turns off
2379 @option{-W/-w} line truncation;
2380 no column alignment used; may be used with
2381 @option{--sep-string[=@var{string}]}. @option{-J} has been introduced
2382 (together with @option{-W} and @option{--sep-string})
2383 to disentangle the old (@acronym{POSIX}-compliant) options @option{-w} and
2384 @option{-s} along with the three column options.
2387 @item -l @var{page_length}
2388 @itemx --length=@var{page_length}
2391 Set the page length to @var{page_length} (default 66) lines, including
2392 the lines of the header [and the footer]. If @var{page_length} is less
2393 than or equal to 10, the header and footer are omitted, as if the
2394 @option{-t} option had been given.
2400 Merge and print all @var{file}s in parallel, one in each column. If a
2401 line is too long to fit in a column, it is truncated, unless the @option{-J}
2402 option is used. @option{--sep-string[=@var{string}]} may be used.
2404 some @var{file}s (form feeds set) produce empty columns, still marked
2405 by @var{string}. The result is a continuous line numbering and column
2406 marking throughout the whole merged file. Completely empty merged pages
2407 show no separators or line numbers. The default header becomes
2408 @samp{@var{date} @var{page}} with spaces inserted in the middle; this
2409 may be used with the @option{-h} or @option{--header} option to fill up
2410 the middle blank part.
2412 @item -n[@var{number-separator}[@var{digits}]]
2413 @itemx --number-lines[=@var{number-separator}[@var{digits}]]
2415 @opindex --number-lines
2416 Provide @var{digits} digit line numbering (default for @var{digits} is
2417 5). With multicolumn output the number occupies the first @var{digits}
2418 column positions of each text column or only each line of @option{-m}
2419 output. With single column output the number precedes each line just as
2420 @option{-m} does. Default counting of the line numbers starts with the
2421 first line of the input file (not the first line printed, compare the
2422 @option{--page} option and @option{-N} option).
2423 Optional argument @var{number-separator} is the character appended to
2424 the line number to separate it from the text followed. The default
2425 separator is the TAB character. In a strict sense a TAB is always
2426 printed with single column output only. The TAB width varies
2427 with the TAB position, e.g., with the left @var{margin} specified
2428 by @option{-o} option. With multicolumn output priority is given to
2429 @samp{equal width of output columns} (a @acronym{POSIX} specification).
2430 The TAB width is fixed to the value of the first column and does
2431 not change with different values of left @var{margin}. That means a
2432 fixed number of spaces is always printed in the place of the
2433 @var{number-separator} TAB. The tabification depends upon the output
2436 @item -N @var{line_number}
2437 @itemx --first-line-number=@var{line_number}
2439 @opindex --first-line-number
2440 Start line counting with the number @var{line_number} at first line of
2441 first page printed (in most cases not the first line of the input file).
2443 @item -o @var{margin}
2444 @itemx --indent=@var{margin}
2447 @cindex indenting lines
2449 Indent each line with a margin @var{margin} spaces wide (default is zero).
2450 The total page width is the size of the margin plus the @var{page_width}
2451 set with the @option{-W/-w} option. A limited overflow may occur with
2452 numbered single column output (compare @option{-n} option).
2455 @itemx --no-file-warnings
2457 @opindex --no-file-warnings
2458 Do not print a warning message when an argument @var{file} cannot be
2459 opened. (The exit status will still be nonzero, however.)
2461 @item -s[@var{char}]
2462 @itemx --separator[=@var{char}]
2464 @opindex --separator
2465 Separate columns by a single character @var{char}. The default for
2466 @var{char} is the TAB character without @option{-w} and @samp{no
2467 character} with @option{-w}. Without @option{-s} the default separator
2468 @samp{space} is set. @option{-s[char]} turns off line truncation of all
2469 three column options (@option{-COLUMN}|@option{-a -COLUMN}|@option{-m}) unless
2470 @option{-w} is set. This is a @acronym{POSIX}-compliant formulation.
2473 @item -S@var{string}
2474 @itemx --sep-string[=@var{string}]
2476 @opindex --sep-string
2477 Use @var{string} to separate output columns. The @option{-S} option doesn't
2478 affect the @option{-W/-w} option, unlike the @option{-s} option which does. It
2479 does not affect line truncation or column alignment.
2480 Without @option{-S}, and with @option{-J}, @command{pr} uses the default output
2482 Without @option{-S} or @option{-J}, @command{pr} uses a @samp{space}
2483 (same as @option{-S"@w{ }"}). @option{--sep-string} with no
2484 @samp{=@var{string}} is equivalent to @option{--sep-string=""}.
2487 @itemx --omit-header
2489 @opindex --omit-header
2490 Do not print the usual header [and footer] on each page, and do not fill
2491 out the bottom of pages (with blank lines or a form feed). No page
2492 structure is produced, but form feeds set in the input files are retained.
2493 The predefined pagination is not changed. @option{-t} or @option{-T} may be
2494 useful together with other options; e.g.: @option{-t -e4}, expand TAB characters
2495 in the input file to 4 spaces but don't make any other changes. Use of
2496 @option{-t} overrides @option{-h}.
2499 @itemx --omit-pagination
2501 @opindex --omit-pagination
2502 Do not print header [and footer]. In addition eliminate all form feeds
2503 set in the input files.
2506 @itemx --show-nonprinting
2508 @opindex --show-nonprinting
2509 Print nonprinting characters in octal backslash notation.
2511 @item -w @var{page_width}
2512 @itemx --width=@var{page_width}
2515 Set page width to @var{page_width} characters for multiple text-column
2516 output only (default for @var{page_width} is 72). @option{-s[CHAR]} turns
2517 off the default page width and any line truncation and column alignment.
2518 Lines of full length are merged, regardless of the column options
2519 set. No @var{page_width} setting is possible with single column output.
2520 A @acronym{POSIX}-compliant formulation.
2522 @item -W @var{page_width}
2523 @itemx --page_width=@var{page_width}
2525 @opindex --page_width
2526 Set the page width to @var{page_width} characters. That's valid with and
2527 without a column option. Text lines are truncated, unless @option{-J}
2528 is used. Together with one of the three column options
2529 (@option{-@var{column}}, @option{-a -@var{column}} or @option{-m}) column
2530 alignment is always used. The separator options @option{-S} or @option{-s}
2531 don't affect the @option{-W} option. Default is 72 characters. Without
2532 @option{-W @var{page_width}} and without any of the column options NO line
2533 truncation is used (defined to keep downward compatibility and to meet
2534 most frequent tasks). That's equivalent to @option{-W 72 -J}. The header
2535 line is never truncated.
2542 @node fold invocation
2543 @section @command{fold}: Wrap input lines to fit in specified width
2546 @cindex wrapping long input lines
2547 @cindex folding long input lines
2549 @command{fold} writes each @var{file} (@option{-} means standard input), or
2550 standard input if none are given, to standard output, breaking long
2554 fold [@var{option}]@dots{} [@var{file}]@dots{}
2557 By default, @command{fold} breaks lines wider than 80 columns. The output
2558 is split into as many lines as necessary.
2560 @cindex screen columns
2561 @command{fold} counts screen columns by default; thus, a tab may count more
2562 than one column, backspace decreases the column count, and carriage
2563 return sets the column to zero.
2565 The program accepts the following options. Also see @ref{Common options}.
2573 Count bytes rather than columns, so that tabs, backspaces, and carriage
2574 returns are each counted as taking up one column, just like other
2581 Break at word boundaries: the line is broken after the last blank before
2582 the maximum line length. If the line contains no such blanks, the line
2583 is broken at the maximum line length as usual.
2585 @item -w @var{width}
2586 @itemx --width=@var{width}
2589 Use a maximum line length of @var{width} columns instead of 80.
2591 For compatibility @command{fold} supports an obsolete option syntax
2592 @option{-@var{width}}. New scripts should use @option{-w @var{width}}
2600 @node Output of parts of files
2601 @chapter Output of parts of files
2603 @cindex output of parts of files
2604 @cindex parts of files, output of
2606 These commands output pieces of the input.
2609 * head invocation:: Output the first part of files.
2610 * tail invocation:: Output the last part of files.
2611 * split invocation:: Split a file into fixed-size pieces.
2612 * csplit invocation:: Split a file into context-determined pieces.
2615 @node head invocation
2616 @section @command{head}: Output the first part of files
2619 @cindex initial part of files, outputting
2620 @cindex first part of files, outputting
2622 @command{head} prints the first part (10 lines by default) of each
2623 @var{file}; it reads from standard input if no files are given or
2624 when given a @var{file} of @option{-}. Synopsis:
2627 head [@var{option}]@dots{} [@var{file}]@dots{}
2630 If more than one @var{file} is specified, @command{head} prints a
2631 one-line header consisting of:
2634 ==> @var{file name} <==
2638 before the output for each @var{file}.
2640 The program accepts the following options. Also see @ref{Common options}.
2645 @itemx --bytes=@var{k}
2648 Print the first @var{k} bytes, instead of initial lines.
2649 However, if @var{k} starts with a @samp{-},
2650 print all but the last @var{k} bytes of each file.
2651 @multiplierSuffixes{k}
2654 @itemx --lines=@var{k}
2657 Output the first @var{k} lines.
2658 However, if @var{k} starts with a @samp{-},
2659 print all but the last @var{k} lines of each file.
2660 Size multiplier suffixes are the same as with the @option{-c} option.
2668 Never print file name headers.
2674 Always print file name headers.
2678 For compatibility @command{head} also supports an obsolete option syntax
2679 @option{-@var{count}@var{options}}, which is recognized only if it is
2680 specified first. @var{count} is a decimal number optionally followed
2681 by a size letter (@samp{b}, @samp{k}, @samp{m}) as in @option{-c}, or
2682 @samp{l} to mean count by lines, or other option letters (@samp{cqv}).
2683 Scripts intended for standard hosts should use @option{-c @var{count}}
2684 or @option{-n @var{count}} instead. If your script must also run on
2685 hosts that support only the obsolete syntax, it is usually simpler to
2686 avoid @command{head}, e.g., by using @samp{sed 5q} instead of
2692 @node tail invocation
2693 @section @command{tail}: Output the last part of files
2696 @cindex last part of files, outputting
2698 @command{tail} prints the last part (10 lines by default) of each
2699 @var{file}; it reads from standard input if no files are given or
2700 when given a @var{file} of @samp{-}. Synopsis:
2703 tail [@var{option}]@dots{} [@var{file}]@dots{}
2706 If more than one @var{file} is specified, @command{tail} prints a
2707 one-line header consisting of:
2710 ==> @var{file name} <==
2714 before the output for each @var{file}.
2716 @cindex BSD @command{tail}
2717 @sc{gnu} @command{tail} can output any amount of data (some other versions of
2718 @command{tail} cannot). It also has no @option{-r} option (print in
2719 reverse), since reversing a file is really a different job from printing
2720 the end of a file; BSD @command{tail} (which is the one with @option{-r}) can
2721 only reverse files that are at most as large as its buffer, which is
2722 typically 32 KiB@. A more reliable and versatile way to reverse files is
2723 the @sc{gnu} @command{tac} command.
2725 The program accepts the following options. Also see @ref{Common options}.
2730 @itemx --bytes=@var{k}
2733 Output the last @var{k} bytes, instead of final lines.
2734 However, if @var{k} starts with a @samp{+}, start printing with the
2735 @var{k}th byte from the start of each file, instead of from the end.
2736 @multiplierSuffixes{k}
2739 @itemx --follow[=@var{how}]
2742 @cindex growing files
2743 @vindex name @r{follow option}
2744 @vindex descriptor @r{follow option}
2745 Loop forever trying to read more characters at the end of the file,
2746 presumably because the file is growing.
2747 If more than one file is given, @command{tail} prints a header whenever it
2748 gets output from a different file, to indicate which file that output is
2751 There are two ways to specify how you'd like to track files with this option,
2752 but that difference is noticeable only when a followed file is removed or
2754 If you'd like to continue to track the end of a growing file even after
2755 it has been unlinked, use @option{--follow=descriptor}. This is the default
2756 behavior, but it is not useful if you're tracking a log file that may be
2757 rotated (removed or renamed, then reopened). In that case, use
2758 @option{--follow=name} to track the named file by reopening it periodically
2759 to see if it has been removed and recreated by some other program.
2761 No matter which method you use, if the tracked file is determined to have
2762 shrunk, @command{tail} prints a message saying the file has been truncated
2763 and resumes tracking the end of the file from the newly-determined endpoint.
2765 When a file is removed, @command{tail}'s behavior depends on whether it is
2766 following the name or the descriptor. When following by name, tail can
2767 detect that a file has been removed and gives a message to that effect,
2768 and if @option{--retry} has been specified it will continue checking
2769 periodically to see if the file reappears.
2770 When following a descriptor, tail does not detect that the file has
2771 been unlinked or renamed and issues no message; even though the file
2772 may no longer be accessible via its original name, it may still be
2775 The option values @samp{descriptor} and @samp{name} may be specified only
2776 with the long form of the option, not with @option{-f}.
2778 The @option{-f} option is ignored if
2779 no @var{file} operand is specified and standard input is a FIFO or a pipe.
2780 Likewise, the @option{-f} option has no effect for any
2781 operand specified as @samp{-}, when standard input is a FIFO or a pipe.
2785 This option is the same as @option{--follow=name --retry}. That is, tail
2786 will attempt to reopen a file when it is removed. Should this fail, tail
2787 will keep trying until it becomes accessible again.
2791 This option is useful mainly when following by name (i.e., with
2792 @option{--follow=name}).
2793 Without this option, when tail encounters a file that doesn't
2794 exist or is otherwise inaccessible, it reports that fact and
2795 never checks it again.
2797 @itemx --sleep-interval=@var{number}
2798 @opindex --sleep-interval
2799 Change the number of seconds to wait between iterations (the default is 1.0).
2800 During one iteration, every specified file is checked to see if it has
2802 Historical implementations of @command{tail} have required that
2803 @var{number} be an integer. However, GNU @command{tail} accepts
2804 an arbitrary floating point number (using a period before any
2807 @itemx --pid=@var{pid}
2809 When following by name or by descriptor, you may specify the process ID,
2810 @var{pid}, of the sole writer of all @var{file} arguments. Then, shortly
2811 after that process terminates, tail will also terminate. This will
2812 work properly only if the writer and the tailing process are running on
2813 the same machine. For example, to save the output of a build in a file
2814 and to watch the file grow, if you invoke @command{make} and @command{tail}
2815 like this then the tail process will stop when your build completes.
2816 Without this option, you would have had to kill the @code{tail -f}
2820 $ make >& makerr & tail --pid=$! -f makerr
2823 If you specify a @var{pid} that is not in use or that does not correspond
2824 to the process that is writing to the tailed files, then @command{tail}
2825 may terminate long before any @var{file}s stop growing or it may not
2826 terminate until long after the real writer has terminated.
2827 Note that @option{--pid} cannot be supported on some systems; @command{tail}
2828 will print a warning if this is the case.
2830 @itemx --max-unchanged-stats=@var{n}
2831 @opindex --max-unchanged-stats
2832 When tailing a file by name, if there have been @var{n} (default
2833 n=@value{DEFAULT_MAX_N_UNCHANGED_STATS_BETWEEN_OPENS}) consecutive
2834 iterations for which the file has not changed, then
2835 @code{open}/@code{fstat} the file to determine if that file name is
2836 still associated with the same device/inode-number pair as before.
2837 When following a log file that is rotated, this is approximately the
2838 number of seconds between when tail prints the last pre-rotation lines
2839 and when it prints the lines that have accumulated in the new log file.
2840 This option is meaningful only when following by name.
2843 @itemx --lines=@var{k}
2846 Output the last @var{k} lines.
2847 However, if @var{k} starts with a @samp{+}, start printing with the
2848 @var{k}th line from the start of each file, instead of from the end.
2849 Size multiplier suffixes are the same as with the @option{-c} option.
2857 Never print file name headers.
2863 Always print file name headers.
2867 For compatibility @command{tail} also supports an obsolete usage
2868 @samp{tail -[@var{count}][bcl][f] [@var{file}]}, which is recognized
2869 only if it does not conflict with the usage described
2870 above. This obsolete form uses exactly one option and at most one
2871 file. In the option, @var{count} is an optional decimal number optionally
2872 followed by a size letter (@samp{b}, @samp{c}, @samp{l}) to mean count
2873 by 512-byte blocks, bytes, or lines, optionally followed by @samp{f}
2874 which has the same meaning as @option{-f}.
2876 @vindex _POSIX2_VERSION
2877 On older systems, the leading @samp{-} can be replaced by @samp{+} in
2878 the obsolete option syntax with the same meaning as in counts, and
2879 obsolete usage overrides normal usage when the two conflict.
2880 This obsolete behavior can be enabled or disabled with the
2881 @env{_POSIX2_VERSION} environment variable (@pxref{Standards
2884 Scripts intended for use on standard hosts should avoid obsolete
2885 syntax and should use @option{-c @var{count}[b]}, @option{-n
2886 @var{count}}, and/or @option{-f} instead. If your script must also
2887 run on hosts that support only the obsolete syntax, you can often
2888 rewrite it to avoid problematic usages, e.g., by using @samp{sed -n
2889 '$p'} rather than @samp{tail -1}. If that's not possible, the script
2890 can use a test like @samp{if tail -c +1 </dev/null >/dev/null 2>&1;
2891 then @dots{}} to decide which syntax to use.
2893 Even if your script assumes the standard behavior, you should still
2894 beware usages whose behaviors differ depending on the @acronym{POSIX}
2895 version. For example, avoid @samp{tail - main.c}, since it might be
2896 interpreted as either @samp{tail main.c} or as @samp{tail -- -
2897 main.c}; avoid @samp{tail -c 4}, since it might mean either @samp{tail
2898 -c4} or @samp{tail -c 10 4}; and avoid @samp{tail +4}, since it might
2899 mean either @samp{tail ./+4} or @samp{tail -n +4}.
2904 @node split invocation
2905 @section @command{split}: Split a file into fixed-size pieces
2908 @cindex splitting a file into pieces
2909 @cindex pieces, splitting a file into
2911 @command{split} creates output files containing consecutive sections of
2912 @var{input} (standard input if none is given or @var{input} is
2913 @samp{-}). Synopsis:
2916 split [@var{option}] [@var{input} [@var{prefix}]]
2919 By default, @command{split} puts 1000 lines of @var{input} (or whatever is
2920 left over for the last section), into each output file.
2922 @cindex output file name prefix
2923 The output files' names consist of @var{prefix} (@samp{x} by default)
2924 followed by a group of characters (@samp{aa}, @samp{ab}, @dots{} by
2925 default), such that concatenating the output files in traditional
2926 sorted order by file name produces
2927 the original input file. If the output file names are exhausted,
2928 @command{split} reports an error without deleting the output files
2931 The program accepts the following options. Also see @ref{Common options}.
2935 @item -l @var{lines}
2936 @itemx --lines=@var{lines}
2939 Put @var{lines} lines of @var{input} into each output file.
2941 For compatibility @command{split} also supports an obsolete
2942 option syntax @option{-@var{lines}}. New scripts should use @option{-l
2943 @var{lines}} instead.
2946 @itemx --bytes=@var{size}
2949 Put @var{size} bytes of @var{input} into each output file.
2950 @multiplierSuffixes{size}
2953 @itemx --line-bytes=@var{size}
2955 @opindex --line-bytes
2956 Put into each output file as many complete lines of @var{input} as
2957 possible without exceeding @var{size} bytes. Individual lines longer than
2958 @var{size} bytes are broken into multiple files.
2959 @var{size} has the same format as for the @option{--bytes} option.
2961 @item -a @var{length}
2962 @itemx --suffix-length=@var{length}
2964 @opindex --suffix-length
2965 Use suffixes of length @var{length}. The default @var{length} is 2.
2968 @itemx --numeric-suffixes
2970 @opindex --numeric-suffixes
2971 Use digits in suffixes rather than lower-case letters.
2975 Write a diagnostic just before each output file is opened.
2982 @node csplit invocation
2983 @section @command{csplit}: Split a file into context-determined pieces
2986 @cindex context splitting
2987 @cindex splitting a file into pieces by context
2989 @command{csplit} creates zero or more output files containing sections of
2990 @var{input} (standard input if @var{input} is @samp{-}). Synopsis:
2993 csplit [@var{option}]@dots{} @var{input} @var{pattern}@dots{}
2996 The contents of the output files are determined by the @var{pattern}
2997 arguments, as detailed below. An error occurs if a @var{pattern}
2998 argument refers to a nonexistent line of the input file (e.g., if no
2999 remaining line matches a given regular expression). After every
3000 @var{pattern} has been matched, any remaining input is copied into one
3003 By default, @command{csplit} prints the number of bytes written to each
3004 output file after it has been created.
3006 The types of pattern arguments are:
3011 Create an output file containing the input up to but not including line
3012 @var{n} (a positive integer). If followed by a repeat count, also
3013 create an output file containing the next @var{n} lines of the input
3014 file once for each repeat.
3016 @item /@var{regexp}/[@var{offset}]
3017 Create an output file containing the current line up to (but not
3018 including) the next line of the input file that contains a match for
3019 @var{regexp}. The optional @var{offset} is an integer.
3020 If it is given, the input up to (but not including) the
3021 matching line plus or minus @var{offset} is put into the output file,
3022 and the line after that begins the next section of input.
3024 @item %@var{regexp}%[@var{offset}]
3025 Like the previous type, except that it does not create an output
3026 file, so that section of the input file is effectively ignored.
3028 @item @{@var{repeat-count}@}
3029 Repeat the previous pattern @var{repeat-count} additional
3030 times. The @var{repeat-count} can either be a positive integer or an
3031 asterisk, meaning repeat as many times as necessary until the input is
3036 The output files' names consist of a prefix (@samp{xx} by default)
3037 followed by a suffix. By default, the suffix is an ascending sequence
3038 of two-digit decimal numbers from @samp{00} to @samp{99}. In any case,
3039 concatenating the output files in sorted order by file name produces the
3040 original input file.
3042 By default, if @command{csplit} encounters an error or receives a hangup,
3043 interrupt, quit, or terminate signal, it removes any output files
3044 that it has created so far before it exits.
3046 The program accepts the following options. Also see @ref{Common options}.
3050 @item -f @var{prefix}
3051 @itemx --prefix=@var{prefix}
3054 @cindex output file name prefix
3055 Use @var{prefix} as the output file name prefix.
3057 @item -b @var{suffix}
3058 @itemx --suffix=@var{suffix}
3061 @cindex output file name suffix
3062 Use @var{suffix} as the output file name suffix. When this option is
3063 specified, the suffix string must include exactly one
3064 @code{printf(3)}-style conversion specification, possibly including
3065 format specification flags, a field width, a precision specifications,
3066 or all of these kinds of modifiers. The format letter must convert a
3067 binary integer argument to readable form; thus, only @samp{d}, @samp{i},
3068 @samp{u}, @samp{o}, @samp{x}, and @samp{X} conversions are allowed. The
3069 entire @var{suffix} is given (with the current output file number) to
3070 @code{sprintf(3)} to form the file name suffixes for each of the
3071 individual output files in turn. If this option is used, the
3072 @option{--digits} option is ignored.
3074 @item -n @var{digits}
3075 @itemx --digits=@var{digits}
3078 Use output file names containing numbers that are @var{digits} digits
3079 long instead of the default 2.
3084 @opindex --keep-files
3085 Do not remove output files when errors are encountered.
3088 @itemx --elide-empty-files
3090 @opindex --elide-empty-files
3091 Suppress the generation of zero-length output files. (In cases where
3092 the section delimiters of the input file are supposed to mark the first
3093 lines of each of the sections, the first output file will generally be a
3094 zero-length file unless you use this option.) The output file sequence
3095 numbers always run consecutively starting from 0, even when this option
3106 Do not print counts of output file sizes.
3112 Here is an example of its usage.
3113 First, create an empty directory for the exercise,
3120 Now, split the sequence of 1..14 on lines that end with 0 or 5:
3123 $ seq 14 | csplit - '/[05]$/' '@{*@}'
3129 Each number printed above is the size of an output
3130 file that csplit has just created.
3131 List the names of those output files:
3138 Use @command{head} to show their contents:
3163 @node Summarizing files
3164 @chapter Summarizing files
3166 @cindex summarizing files
3168 These commands generate just a few numbers representing entire
3172 * wc invocation:: Print newline, word, and byte counts.
3173 * sum invocation:: Print checksum and block counts.
3174 * cksum invocation:: Print CRC checksum and byte counts.
3175 * md5sum invocation:: Print or check MD5 digests.
3176 * sha1sum invocation:: Print or check SHA-1 digests.
3177 * sha2 utilities:: Print or check SHA-2 digests.
3182 @section @command{wc}: Print newline, word, and byte counts
3186 @cindex character count
3190 @command{wc} counts the number of bytes, characters, whitespace-separated
3191 words, and newlines in each given @var{file}, or standard input if none
3192 are given or for a @var{file} of @samp{-}. Synopsis:
3195 wc [@var{option}]@dots{} [@var{file}]@dots{}
3198 @cindex total counts
3199 @command{wc} prints one line of counts for each file, and if the file was
3200 given as an argument, it prints the file name following the counts. If
3201 more than one @var{file} is given, @command{wc} prints a final line
3202 containing the cumulative counts, with the file name @file{total}. The
3203 counts are printed in this order: newlines, words, characters, bytes,
3204 maximum line length.
3205 Each count is printed right-justified in a field with at least one
3206 space between fields so that the numbers and file names normally line
3207 up nicely in columns. The width of the count fields varies depending
3208 on the inputs, so you should not depend on a particular field width.
3209 However, as a @acronym{GNU} extension, if only one count is printed,
3210 it is guaranteed to be printed without leading spaces.
3212 By default, @command{wc} prints three counts: the newline, words, and byte
3213 counts. Options can specify that only certain counts be printed.
3214 Options do not undo others previously given, so
3221 prints both the byte counts and the word counts.
3223 With the @option{--max-line-length} option, @command{wc} prints the length
3224 of the longest line per file, and if there is more than one file it
3225 prints the maximum (not the sum) of those lengths. The line lengths here
3226 are measured in screen columns, according to the current locale and
3227 assuming tab positions in every 8th column.
3229 The program accepts the following options. Also see @ref{Common options}.
3237 Print only the byte counts.
3243 Print only the character counts.
3249 Print only the word counts.
3255 Print only the newline counts.
3258 @itemx --max-line-length
3260 @opindex --max-line-length
3261 Print only the maximum line lengths.
3263 @macro filesZeroFromOption{cmd,withTotalOption,subListOutput}
3264 @itemx --files0-from=@var{file}
3265 @opindex --files0-from=@var{file}
3266 @c This is commented out to avoid a texi2dvi failure.
3267 @c texi2dvi (GNU Texinfo 4.11) 1.104
3268 @c @cindex including files from @command{\cmd\}
3269 Disallow processing files named on the command line, and instead process
3270 those named in file @var{file}; each name being terminated by a zero byte
3271 (@acronym{ASCII} @sc{nul}).
3272 This is useful \withTotalOption\
3273 when the list of file names is so long that it may exceed a command line
3275 In such cases, running @command{\cmd\} via @command{xargs} is undesirable
3276 because it splits the list into pieces and makes @command{\cmd\} print
3277 \subListOutput\ for each sublist rather than for the entire list.
3278 One way to produce a list of @acronym{ASCII} @sc{nul} terminated file names is with @sc{gnu}
3279 @command{find}, using its @option{-print0} predicate.
3280 If @var{file} is @samp{-} then the @acronym{ASCII} @sc{nul} terminated file names
3281 are read from standard input.
3283 @filesZeroFromOption{wc,,a total}
3285 For example, to find the length of the longest line in any @file{.c} or
3286 @file{.h} file in the current hierarchy, do this:
3289 find . -name '*.[ch]' -print0 |
3290 wc -L --files0-from=- | tail -n1
3298 @node sum invocation
3299 @section @command{sum}: Print checksum and block counts
3302 @cindex 16-bit checksum
3303 @cindex checksum, 16-bit
3305 @command{sum} computes a 16-bit checksum for each given @var{file}, or
3306 standard input if none are given or for a @var{file} of @samp{-}. Synopsis:
3309 sum [@var{option}]@dots{} [@var{file}]@dots{}
3312 @command{sum} prints the checksum for each @var{file} followed by the
3313 number of blocks in the file (rounded up). If more than one @var{file}
3314 is given, file names are also printed (by default). (With the
3315 @option{--sysv} option, corresponding file names are printed when there is
3316 at least one file argument.)
3318 By default, @sc{gnu} @command{sum} computes checksums using an algorithm
3319 compatible with BSD @command{sum} and prints file sizes in units of
3322 The program accepts the following options. Also see @ref{Common options}.
3328 @cindex BSD @command{sum}
3329 Use the default (BSD compatible) algorithm. This option is included for
3330 compatibility with the System V @command{sum}. Unless @option{-s} was also
3331 given, it has no effect.
3337 @cindex System V @command{sum}
3338 Compute checksums using an algorithm compatible with System V
3339 @command{sum}'s default, and print file sizes in units of 512-byte blocks.
3343 @command{sum} is provided for compatibility; the @command{cksum} program (see
3344 next section) is preferable in new applications.
3349 @node cksum invocation
3350 @section @command{cksum}: Print CRC checksum and byte counts
3353 @cindex cyclic redundancy check
3354 @cindex CRC checksum
3356 @command{cksum} computes a cyclic redundancy check (CRC) checksum for each
3357 given @var{file}, or standard input if none are given or for a
3358 @var{file} of @samp{-}. Synopsis:
3361 cksum [@var{option}]@dots{} [@var{file}]@dots{}
3364 @command{cksum} prints the CRC checksum for each file along with the number
3365 of bytes in the file, and the file name unless no arguments were given.
3367 @command{cksum} is typically used to ensure that files
3368 transferred by unreliable means (e.g., netnews) have not been corrupted,
3369 by comparing the @command{cksum} output for the received files with the
3370 @command{cksum} output for the original files (typically given in the
3373 The CRC algorithm is specified by the @acronym{POSIX} standard. It is not
3374 compatible with the BSD or System V @command{sum} algorithms (see the
3375 previous section); it is more robust.
3377 The only options are @option{--help} and @option{--version}. @xref{Common
3383 @node md5sum invocation
3384 @section @command{md5sum}: Print or check MD5 digests
3388 @cindex 128-bit checksum
3389 @cindex checksum, 128-bit
3390 @cindex fingerprint, 128-bit
3391 @cindex message-digest, 128-bit
3393 @command{md5sum} computes a 128-bit checksum (or @dfn{fingerprint} or
3394 @dfn{message-digest}) for each specified @var{file}.
3396 Note: The MD5 digest is more reliable than a simple CRC (provided by
3397 the @command{cksum} command) for detecting accidental file corruption,
3398 as the chances of accidentally having two files with identical MD5
3399 are vanishingly small. However, it should not be considered truly
3400 secure against malicious tampering: although finding a file with a
3401 given MD5 fingerprint, or modifying a file so as to retain its MD5 are
3402 considered infeasible at the moment, it is known how to produce
3403 different files with identical MD5 (a ``collision''), something which
3404 can be a security issue in certain contexts. For more secure hashes,
3405 consider using SHA-1 or SHA-2. @xref{sha1sum invocation}, and
3406 @ref{sha2 utilities}.
3408 If a @var{file} is specified as @samp{-} or if no files are given
3409 @command{md5sum} computes the checksum for the standard input.
3410 @command{md5sum} can also determine whether a file and checksum are
3411 consistent. Synopsis:
3414 md5sum [@var{option}]@dots{} [@var{file}]@dots{}
3417 For each @var{file}, @samp{md5sum} outputs the MD5 checksum, a flag
3418 indicating a binary or text input file, and the file name.
3419 If @var{file} contains a backslash or newline, the
3420 line is started with a backslash, and each problematic character in
3421 the file name is escaped with a backslash, making the output
3422 unambiguous even in the presence of arbitrary file names.
3423 If @var{file} is omitted or specified as @samp{-}, standard input is read.
3425 The program accepts the following options. Also see @ref{Common options}.
3433 @cindex binary input files
3434 Treat each input file as binary, by reading it in binary mode and
3435 outputting a @samp{*} flag. This is the inverse of @option{--text}.
3436 On systems like @acronym{GNU} that do not distinguish between binary
3437 and text files, this option merely flags each input file as binary:
3438 the MD5 checksum is unaffected. This option is the default on systems
3439 like MS-DOS that distinguish between binary and text files, except
3440 for reading standard input when standard input is a terminal.
3444 Read file names and checksum information (not data) from each
3445 @var{file} (or from stdin if no @var{file} was specified) and report
3446 whether the checksums match the contents of the named files.
3447 The input to this mode of @command{md5sum} is usually the output of
3448 a prior, checksum-generating run of @samp{md5sum}.
3449 Each valid line of input consists of an MD5 checksum, a binary/text
3450 flag, and then a file name.
3451 Binary files are marked with @samp{*}, text with @samp{ }.
3452 For each such line, @command{md5sum} reads the named file and computes its
3453 MD5 checksum. Then, if the computed message digest does not match the
3454 one on the line with the file name, the file is noted as having
3455 failed the test. Otherwise, the file passes the test.
3456 By default, for each valid line, one line is written to standard
3457 output indicating whether the named file passed the test.
3458 After all checks have been performed, if there were any failures,
3459 a warning is issued to standard error.
3460 Use the @option{--status} option to inhibit that output.
3461 If any listed file cannot be opened or read, if any valid line has
3462 an MD5 checksum inconsistent with the associated file, or if no valid
3463 line is found, @command{md5sum} exits with nonzero status. Otherwise,
3464 it exits successfully.
3468 @cindex verifying MD5 checksums
3469 This option is useful only when verifying checksums.
3470 When verifying checksums, don't generate an 'OK' message per successfully
3471 checked file. Files that fail the verification are reported in the
3472 default one-line-per-file format. If there is any checksum mismatch,
3473 print a warning summarizing the failures to standard error.
3477 @cindex verifying MD5 checksums
3478 This option is useful only when verifying checksums.
3479 When verifying checksums, don't generate the default one-line-per-file
3480 diagnostic and don't output the warning summarizing any failures.
3481 Failures to open or read a file still evoke individual diagnostics to
3483 If all listed files are readable and are consistent with the associated
3484 MD5 checksums, exit successfully. Otherwise exit with a status code
3485 indicating there was a failure.
3491 @cindex text input files
3492 Treat each input file as text, by reading it in text mode and
3493 outputting a @samp{ } flag. This is the inverse of @option{--binary}.
3494 This option is the default on systems like @acronym{GNU} that do not
3495 distinguish between binary and text files. On other systems, it is
3496 the default for reading standard input when standard input is a
3503 @cindex verifying MD5 checksums
3504 When verifying checksums, warn about improperly formatted MD5 checksum lines.
3505 This option is useful only if all but a few lines in the checked input
3513 @node sha1sum invocation
3514 @section @command{sha1sum}: Print or check SHA-1 digests
3518 @cindex 160-bit checksum
3519 @cindex checksum, 160-bit
3520 @cindex fingerprint, 160-bit
3521 @cindex message-digest, 160-bit
3523 @command{sha1sum} computes a 160-bit checksum for each specified
3524 @var{file}. The usage and options of this command are precisely the
3525 same as for @command{md5sum}. @xref{md5sum invocation}.
3527 Note: The SHA-1 digest is more secure than MD5, and no collisions of
3528 it are known (different files having the same fingerprint). However,
3529 it is known that they can be produced with considerable, but not
3530 unreasonable, resources. For this reason, it is generally considered
3531 that SHA-1 should be gradually phased out in favor of the more secure
3532 SHA-2 hash algorithms. @xref{sha2 utilities}.
3535 @node sha2 utilities
3536 @section sha2 utilities: Print or check SHA-2 digests
3543 @cindex 224-bit checksum
3544 @cindex 256-bit checksum
3545 @cindex 384-bit checksum
3546 @cindex 512-bit checksum
3547 @cindex checksum, 224-bit
3548 @cindex checksum, 256-bit
3549 @cindex checksum, 384-bit
3550 @cindex checksum, 512-bit
3551 @cindex fingerprint, 224-bit
3552 @cindex fingerprint, 256-bit
3553 @cindex fingerprint, 384-bit
3554 @cindex fingerprint, 512-bit
3555 @cindex message-digest, 224-bit
3556 @cindex message-digest, 256-bit
3557 @cindex message-digest, 384-bit
3558 @cindex message-digest, 512-bit
3560 The commands @command{sha224sum}, @command{sha256sum},
3561 @command{sha384sum} and @command{sha512sum} compute checksums of
3562 various lengths (respectively 224, 256, 384 and 512 bits),
3563 collectively known as the SHA-2 hashes. The usage and options of
3564 these commands are precisely the same as for @command{md5sum}.
3565 @xref{md5sum invocation}.
3567 Note: The SHA384 and SHA512 digests are considerably slower to
3568 compute, especially on 32-bit computers, than SHA224 or SHA256.
3571 @node Operating on sorted files
3572 @chapter Operating on sorted files
3574 @cindex operating on sorted files
3575 @cindex sorted files, operations on
3577 These commands work with (or produce) sorted files.
3580 * sort invocation:: Sort text files.
3581 * shuf invocation:: Shuffle text files.
3582 * uniq invocation:: Uniquify files.
3583 * comm invocation:: Compare two sorted files line by line.
3584 * ptx invocation:: Produce a permuted index of file contents.
3585 * tsort invocation:: Topological sort.
3589 @node sort invocation
3590 @section @command{sort}: Sort text files
3593 @cindex sorting files
3595 @command{sort} sorts, merges, or compares all the lines from the given
3596 files, or standard input if none are given or for a @var{file} of
3597 @samp{-}. By default, @command{sort} writes the results to standard
3601 sort [@var{option}]@dots{} [@var{file}]@dots{}
3604 @command{sort} has three modes of operation: sort (the default), merge,
3605 and check for sortedness. The following options change the operation
3612 @itemx --check=diagnose-first
3615 @cindex checking for sortedness
3616 Check whether the given file is already sorted: if it is not all
3617 sorted, print a diagnostic containing the first out-of-order line and
3618 exit with a status of 1.
3619 Otherwise, exit successfully.
3620 At most one input file can be given.
3623 @itemx --check=quiet
3624 @itemx --check=silent
3627 @cindex checking for sortedness
3628 Exit successfully if the given file is already sorted, and
3629 exit with status 1 otherwise.
3630 At most one input file can be given.
3631 This is like @option{-c}, except it does not print a diagnostic.
3637 @cindex merging sorted files
3638 Merge the given files by sorting them as a group. Each input file must
3639 always be individually sorted. It always works to sort instead of
3640 merge; merging is provided because it is faster, in the case where it
3645 @cindex sort stability
3646 @cindex sort's last-resort comparison
3647 A pair of lines is compared as follows:
3648 @command{sort} compares each pair of fields, in the
3649 order specified on the command line, according to the associated
3650 ordering options, until a difference is found or no fields are left.
3651 If no key fields are specified, @command{sort} uses a default key of
3652 the entire line. Finally, as a last resort when all keys compare
3653 equal, @command{sort} compares entire lines as if no ordering options
3654 other than @option{--reverse} (@option{-r}) were specified. The
3655 @option{--stable} (@option{-s}) option disables this @dfn{last-resort
3656 comparison} so that lines in which all fields compare equal are left
3657 in their original relative order. The @option{--unique}
3658 (@option{-u}) option also disables the last-resort comparison.
3662 Unless otherwise specified, all comparisons use the character collating
3663 sequence specified by the @env{LC_COLLATE} locale.@footnote{If you
3664 use a non-@acronym{POSIX} locale (e.g., by setting @env{LC_ALL}
3665 to @samp{en_US}), then @command{sort} may produce output that is sorted
3666 differently than you're accustomed to. In that case, set the @env{LC_ALL}
3667 environment variable to @samp{C}. Note that setting only @env{LC_COLLATE}
3668 has two problems. First, it is ineffective if @env{LC_ALL} is also set.
3669 Second, it has undefined behavior if @env{LC_CTYPE} (or @env{LANG}, if
3670 @env{LC_CTYPE} is unset) is set to an incompatible value. For example,
3671 you get undefined behavior if @env{LC_CTYPE} is @code{ja_JP.PCK} but
3672 @env{LC_COLLATE} is @code{en_US.UTF-8}.}
3674 @sc{gnu} @command{sort} (as specified for all @sc{gnu} utilities) has no
3675 limit on input line length or restrictions on bytes allowed within lines.
3676 In addition, if the final byte of an input file is not a newline, @sc{gnu}
3677 @command{sort} silently supplies one. A line's trailing newline is not
3678 part of the line for comparison purposes.
3680 @cindex exit status of @command{sort}
3684 0 if no error occurred
3685 1 if invoked with @option{-c} or @option{-C} and the input is not sorted
3686 2 if an error occurred
3690 If the environment variable @env{TMPDIR} is set, @command{sort} uses its
3691 value as the directory for temporary files instead of @file{/tmp}. The
3692 @option{--temporary-directory} (@option{-T}) option in turn overrides
3693 the environment variable.
3695 The following options affect the ordering of output lines. They may be
3696 specified globally or as part of a specific key field. If no key
3697 fields are specified, global options apply to comparison of entire
3698 lines; otherwise the global options are inherited by key fields that do
3699 not specify any special options of their own. In pre-@acronym{POSIX}
3700 versions of @command{sort}, global options affect only later key fields,
3701 so portable shell scripts should specify global options first.
3706 @itemx --ignore-leading-blanks
3708 @opindex --ignore-leading-blanks
3709 @cindex blanks, ignoring leading
3711 Ignore leading blanks when finding sort keys in each line.
3712 By default a blank is a space or a tab, but the @env{LC_CTYPE} locale
3713 can change this. Note blanks may be ignored by your locale's collating
3714 rules, but without this option they will be significant for character
3715 positions specified in keys with the @option{-k} option.
3718 @itemx --dictionary-order
3720 @opindex --dictionary-order
3721 @cindex dictionary order
3722 @cindex phone directory order
3723 @cindex telephone directory order
3725 Sort in @dfn{phone directory} order: ignore all characters except
3726 letters, digits and blanks when sorting.
3727 By default letters and digits are those of @acronym{ASCII} and a blank
3728 is a space or a tab, but the @env{LC_CTYPE} locale can change this.
3731 @itemx --ignore-case
3733 @opindex --ignore-case
3734 @cindex ignoring case
3735 @cindex case folding
3737 Fold lowercase characters into the equivalent uppercase characters when
3738 comparing so that, for example, @samp{b} and @samp{B} sort as equal.
3739 The @env{LC_CTYPE} locale determines character types.
3740 When used with @option{--unique} those lower case equivalent lines are
3741 thrown away. (There is currently no way to throw away the upper case
3742 equivalent instead. (Any @option{--reverse} given would only affect
3743 the final result, after the throwing away.))
3746 @itemx --general-numeric-sort
3747 @itemx --sort=general-numeric
3749 @opindex --general-numeric-sort
3751 @cindex general numeric sort
3753 Sort numerically, using the standard C function @code{strtod} to convert
3754 a prefix of each line to a double-precision floating point number.
3755 This allows floating point numbers to be specified in scientific notation,
3756 like @code{1.0e-34} and @code{10e100}.
3757 The @env{LC_NUMERIC} locale determines the decimal-point character.
3758 Do not report overflow, underflow, or conversion errors.
3759 Use the following collating sequence:
3763 Lines that do not start with numbers (all considered to be equal).
3765 NaNs (``Not a Number'' values, in IEEE floating point arithmetic)
3766 in a consistent but machine-dependent order.
3770 Finite numbers in ascending numeric order (with @math{-0} and @math{+0} equal).
3775 Use this option only if there is no alternative; it is much slower than
3776 @option{--numeric-sort} (@option{-n}) and it can lose information when
3777 converting to floating point.
3780 @itemx --human-numeric-sort
3781 @itemx --sort=human-numeric
3783 @opindex --human-numeric-sort
3785 @cindex human numeric sort
3787 Sort numerically, as per the @option{--numeric-sort} option below, and in
3788 addition handle IEC or SI suffixes like MiB, MB etc (@ref{Block size}).
3789 Note a mixture of IEC and SI suffixes is not supported and will
3790 be flagged as an error. Also the numbers must be abbreviated uniformly.
3791 I.E. values with different precisions like 6000K and 5M will be sorted
3795 @itemx --ignore-nonprinting
3797 @opindex --ignore-nonprinting
3798 @cindex nonprinting characters, ignoring
3799 @cindex unprintable characters, ignoring
3801 Ignore nonprinting characters.
3802 The @env{LC_CTYPE} locale determines character types.
3803 This option has no effect if the stronger @option{--dictionary-order}
3804 (@option{-d}) option is also given.
3810 @opindex --month-sort
3812 @cindex months, sorting by
3814 An initial string, consisting of any amount of blanks, followed
3815 by a month name abbreviation, is folded to UPPER case and
3816 compared in the order @samp{JAN} < @samp{FEB} < @dots{} < @samp{DEC}.
3817 Invalid names compare low to valid names. The @env{LC_TIME} locale
3818 category determines the month spellings.
3819 By default a blank is a space or a tab, but the @env{LC_CTYPE} locale
3823 @itemx --numeric-sort
3824 @itemx --sort=numeric
3826 @opindex --numeric-sort
3828 @cindex numeric sort
3830 Sort numerically. The number begins each line and consists
3831 of optional blanks, an optional @samp{-} sign, and zero or more
3832 digits possibly separated by thousands separators, optionally followed
3833 by a decimal-point character and zero or more digits. An empty
3834 number is treated as @samp{0}. The @env{LC_NUMERIC}
3835 locale specifies the decimal-point character and thousands separator.
3836 By default a blank is a space or a tab, but the @env{LC_CTYPE} locale
3839 Comparison is exact; there is no rounding error.
3841 Neither a leading @samp{+} nor exponential notation is recognized.
3842 To compare such strings numerically, use the
3843 @option{--general-numeric-sort} (@option{-g}) option.
3846 @itemx --version-sort
3848 @opindex --version-sort
3849 @cindex version number sort
3851 Sort per @code{strverscmp(3)}. This is a normal string comparison, except
3852 that embedded decimal numbers are sorted by numeric value
3853 (see @option{--numeric-sort} above).
3859 @cindex reverse sorting
3860 Reverse the result of comparison, so that lines with greater key values
3861 appear earlier in the output instead of later.
3864 @itemx --random-sort
3865 @itemx --sort=random
3867 @opindex --random-sort
3870 Sort by hashing the input keys and then sorting the hash values.
3871 Choose the hash function at random, ensuring that it is free of
3872 collisions so that differing keys have differing hash values. This is
3873 like a random permutation of the inputs (@pxref{shuf invocation}),
3874 except that keys with the same value sort together.
3876 If multiple random sort fields are specified, the same random hash
3877 function is used for all fields. To use different random hash
3878 functions for different fields, you can invoke @command{sort} more
3881 The choice of hash function is affected by the
3882 @option{--random-source} option.
3890 @item --compress-program=@var{prog}
3891 Compress any temporary files with the program @var{prog}.
3893 With no arguments, @var{prog} must compress standard input to standard
3894 output, and when given the @option{-d} option it must decompress
3895 standard input to standard output.
3897 Terminate with an error if @var{prog} exits with nonzero status.
3899 White space and the backslash character should not appear in
3900 @var{prog}; they are reserved for future use.
3902 @filesZeroFromOption{sort,,sorted output}
3904 @item -k @var{pos1}[,@var{pos2}]
3905 @itemx --key=@var{pos1}[,@var{pos2}]
3909 Specify a sort field that consists of the part of the line between
3910 @var{pos1} and @var{pos2} (or the end of the line, if @var{pos2} is
3911 omitted), @emph{inclusive}.
3913 Each @var{pos} has the form @samp{@var{f}[.@var{c}][@var{opts}]},
3914 where @var{f} is the number of the field to use, and @var{c} is the number
3915 of the first character from the beginning of the field. Fields and character
3916 positions are numbered starting with 1; a character position of zero in
3917 @var{pos2} indicates the field's last character. If @samp{.@var{c}} is
3918 omitted from @var{pos1}, it defaults to 1 (the beginning of the field);
3919 if omitted from @var{pos2}, it defaults to 0 (the end of the field).
3920 @var{opts} are ordering options, allowing individual keys to be sorted
3921 according to different rules; see below for details. Keys can span
3924 Example: To sort on the second field, use @option{--key=2,2}
3925 (@option{-k 2,2}). See below for more notes on keys and more examples.
3927 @item --batch-size=@var{nmerge}
3928 @opindex --batch-size
3929 @cindex number of inputs to merge, nmerge
3930 Merge at most @var{nmerge} inputs at once.
3932 When @command{sort} has to merge more than @var{nmerge} inputs,
3933 it merges them in groups of @var{nmerge}, saving the result in
3934 a temporary file, which is then used as an input in a subsequent merge.
3936 A large value of @var{nmerge} may improve merge performance and decrease
3937 temporary storage utilization at the expense of increased memory usage
3938 and I/0. Conversely a small value of @var{nmerge} may reduce memory
3939 requirements and I/0 at the expense of temporary storage consumption and
3942 The value of @var{nmerge} must be at least 2. The default value is
3943 currently 16, but this is implementation-dependent and may change in
3946 The value of @var{nmerge} may be bounded by a resource limit for open
3947 file descriptors. The commands @samp{ulimit -n} or @samp{getconf
3948 OPEN_MAX} may display limits for your systems; these limits may be
3949 modified further if your program already has some files open, or if
3950 the operating system has other limits on the number of open files. If
3951 the value of @var{nmerge} exceeds the resource limit, @command{sort}
3952 silently uses a smaller value.
3954 @item -o @var{output-file}
3955 @itemx --output=@var{output-file}
3958 @cindex overwriting of input, allowed
3959 Write output to @var{output-file} instead of standard output.
3960 Normally, @command{sort} reads all input before opening
3961 @var{output-file}, so you can safely sort a file in place by using
3962 commands like @code{sort -o F F} and @code{cat F | sort -o F}.
3963 However, @command{sort} with @option{--merge} (@option{-m}) can open
3964 the output file before reading all input, so a command like @code{cat
3965 F | sort -m -o F - G} is not safe as @command{sort} might start
3966 writing @file{F} before @command{cat} is done reading it.
3968 @vindex POSIXLY_CORRECT
3969 On newer systems, @option{-o} cannot appear after an input file if
3970 @env{POSIXLY_CORRECT} is set, e.g., @samp{sort F -o F}. Portable
3971 scripts should specify @option{-o @var{output-file}} before any input
3974 @item --random-source=@var{file}
3975 @opindex --random-source
3976 @cindex random source for sorting
3977 Use @var{file} as a source of random data used to determine which
3978 random hash function to use with the @option{-R} option. @xref{Random
3985 @cindex sort stability
3986 @cindex sort's last-resort comparison
3988 Make @command{sort} stable by disabling its last-resort comparison.
3989 This option has no effect if no fields or global ordering options
3990 other than @option{--reverse} (@option{-r}) are specified.
3993 @itemx --buffer-size=@var{size}
3995 @opindex --buffer-size
3996 @cindex size for main memory sorting
3997 Use a main-memory sort buffer of the given @var{size}. By default,
3998 @var{size} is in units of 1024 bytes. Appending @samp{%} causes
3999 @var{size} to be interpreted as a percentage of physical memory.
4000 Appending @samp{K} multiplies @var{size} by 1024 (the default),
4001 @samp{M} by 1,048,576, @samp{G} by 1,073,741,824, and so on for
4002 @samp{T}, @samp{P}, @samp{E}, @samp{Z}, and @samp{Y}. Appending
4003 @samp{b} causes @var{size} to be interpreted as a byte count, with no
4006 This option can improve the performance of @command{sort} by causing it
4007 to start with a larger or smaller sort buffer than the default.
4008 However, this option affects only the initial buffer size. The buffer
4009 grows beyond @var{size} if @command{sort} encounters input lines larger
4012 @item -t @var{separator}
4013 @itemx --field-separator=@var{separator}
4015 @opindex --field-separator
4016 @cindex field separator character
4017 Use character @var{separator} as the field separator when finding the
4018 sort keys in each line. By default, fields are separated by the empty
4019 string between a non-blank character and a blank character.
4020 By default a blank is a space or a tab, but the @env{LC_CTYPE} locale
4023 That is, given the input line @w{@samp{ foo bar}}, @command{sort} breaks it
4024 into fields @w{@samp{ foo}} and @w{@samp{ bar}}. The field separator is
4025 not considered to be part of either the field preceding or the field
4026 following, so with @samp{sort @w{-t " "}} the same input line has
4027 three fields: an empty field, @samp{foo}, and @samp{bar}.
4028 However, fields that extend to the end of the line,
4029 as @option{-k 2}, or fields consisting of a range, as @option{-k 2,3},
4030 retain the field separators present between the endpoints of the range.
4032 To specify @acronym{ASCII} @sc{nul} as the field separator,
4033 use the two-character string @samp{\0}, e.g., @samp{sort -t '\0'}.
4035 @item -T @var{tempdir}
4036 @itemx --temporary-directory=@var{tempdir}
4038 @opindex --temporary-directory
4039 @cindex temporary directory
4041 Use directory @var{tempdir} to store temporary files, overriding the
4042 @env{TMPDIR} environment variable. If this option is given more than
4043 once, temporary files are stored in all the directories given. If you
4044 have a large sort or merge that is I/O-bound, you can often improve
4045 performance by using this option to specify directories on different
4046 disks and controllers.
4052 @cindex uniquifying output
4054 Normally, output only the first of a sequence of lines that compare
4055 equal. For the @option{--check} (@option{-c} or @option{-C}) option,
4056 check that no pair of consecutive lines compares equal.
4058 This option also disables the default last-resort comparison.
4060 The commands @code{sort -u} and @code{sort | uniq} are equivalent, but
4061 this equivalence does not extend to arbitrary @command{sort} options.
4062 For example, @code{sort -n -u} inspects only the value of the initial
4063 numeric string when checking for uniqueness, whereas @code{sort -n |
4064 uniq} inspects the entire line. @xref{uniq invocation}.
4066 @macro zeroTerminatedOption
4068 @itemx --zero-terminated
4070 @opindex --zero-terminated
4071 @cindex process zero-terminated items
4072 Delimit items with a zero byte rather than a newline (@acronym{ASCII} @sc{lf}).
4073 I.E. treat input as items separated by @acronym{ASCII} @sc{nul}
4074 and terminate output items with @acronym{ASCII} @sc{nul}.
4075 This option can be useful in conjunction with @samp{perl -0} or
4076 @samp{find -print0} and @samp{xargs -0} which do the same in order to
4077 reliably handle arbitrary file names (even those containing blanks
4078 or other special characters).
4080 @zeroTerminatedOption
4084 Historical (BSD and System V) implementations of @command{sort} have
4085 differed in their interpretation of some options, particularly
4086 @option{-b}, @option{-f}, and @option{-n}. @sc{gnu} sort follows the @acronym{POSIX}
4087 behavior, which is usually (but not always!) like the System V behavior.
4088 According to @acronym{POSIX}, @option{-n} no longer implies @option{-b}. For
4089 consistency, @option{-M} has been changed in the same way. This may
4090 affect the meaning of character positions in field specifications in
4091 obscure cases. The only fix is to add an explicit @option{-b}.
4093 A position in a sort field specified with @option{-k} may have any
4094 of the option letters @samp{MbdfghinRrV} appended to it, in which case no
4095 global ordering options are inherited by that particular field. The
4096 @option{-b} option may be independently attached to either or both of
4097 the start and end positions of a field specification, and if it is
4098 inherited from the global options it will be attached to both.
4099 If input lines can contain leading or adjacent blanks and @option{-t}
4100 is not used, then @option{-k} is typically combined with @option{-b} or
4101 an option that implicitly ignores leading blanks (@samp{MghnV}) as otherwise
4102 the varying numbers of leading blanks in fields can cause confusing results.
4104 If the start position in a sort field specifier falls after the end of
4105 the line or after the end field, the field is empty. If the @option{-b}
4106 option was specified, the @samp{.@var{c}} part of a field specification
4107 is counted from the first nonblank character of the field.
4109 @vindex _POSIX2_VERSION
4110 @vindex POSIXLY_CORRECT
4111 On older systems, @command{sort} supports an obsolete origin-zero
4112 syntax @samp{+@var{pos1} [-@var{pos2}]} for specifying sort keys.
4113 This obsolete behavior can be enabled or disabled with the
4114 @env{_POSIX2_VERSION} environment variable (@pxref{Standards
4115 conformance}); it can also be enabled when @env{POSIXLY_CORRECT} is
4116 not set by using the obsolete syntax with @samp{-@var{pos2}} present.
4118 Scripts intended for use on standard hosts should avoid obsolete
4119 syntax and should use @option{-k} instead. For example, avoid
4120 @samp{sort +2}, since it might be interpreted as either @samp{sort
4121 ./+2} or @samp{sort -k 3}. If your script must also run on hosts that
4122 support only the obsolete syntax, it can use a test like @samp{if sort
4123 -k 1 </dev/null >/dev/null 2>&1; then @dots{}} to decide which syntax
4126 Here are some examples to illustrate various combinations of options.
4131 Sort in descending (reverse) numeric order.
4138 Sort alphabetically, omitting the first and second fields
4139 and the blanks at the start of the third field.
4140 This uses a single key composed of the characters beginning
4141 at the start of the first nonblank character in field three
4142 and extending to the end of each line.
4149 Sort numerically on the second field and resolve ties by sorting
4150 alphabetically on the third and fourth characters of field five.
4151 Use @samp{:} as the field delimiter.
4154 sort -t : -k 2,2n -k 5.3,5.4
4157 Note that if you had written @option{-k 2n} instead of @option{-k 2,2n}
4158 @command{sort} would have used all characters beginning in the second field
4159 and extending to the end of the line as the primary @emph{numeric}
4160 key. For the large majority of applications, treating keys spanning
4161 more than one field as numeric will not do what you expect.
4163 Also note that the @samp{n} modifier was applied to the field-end
4164 specifier for the first key. It would have been equivalent to
4165 specify @option{-k 2n,2} or @option{-k 2n,2n}. All modifiers except
4166 @samp{b} apply to the associated @emph{field}, regardless of whether
4167 the modifier character is attached to the field-start and/or the
4168 field-end part of the key specifier.
4171 Sort the password file on the fifth field and ignore any
4172 leading blanks. Sort lines with equal values in field five
4173 on the numeric user ID in field three. Fields are separated
4177 sort -t : -k 5b,5 -k 3,3n /etc/passwd
4178 sort -t : -n -k 5b,5 -k 3,3 /etc/passwd
4179 sort -t : -b -k 5,5 -k 3,3n /etc/passwd
4182 These three commands have equivalent effect. The first specifies that
4183 the first key's start position ignores leading blanks and the second
4184 key is sorted numerically. The other two commands rely on global
4185 options being inherited by sort keys that lack modifiers. The inheritance
4186 works in this case because @option{-k 5b,5b} and @option{-k 5b,5} are
4187 equivalent, as the location of a field-end lacking a @samp{.@var{c}}
4188 character position is not affected by whether initial blanks are
4192 Sort a set of log files, primarily by IPv4 address and secondarily by
4193 time stamp. If two lines' primary and secondary keys are identical,
4194 output the lines in the same order that they were input. The log
4195 files contain lines that look like this:
4198 4.150.156.3 - - [01/Apr/2004:06:31:51 +0000] message 1
4199 211.24.3.231 - - [24/Apr/2004:20:17:39 +0000] message 2
4202 Fields are separated by exactly one space. Sort IPv4 addresses
4203 lexicographically, e.g., 212.61.52.2 sorts before 212.129.233.201
4204 because 61 is less than 129.
4207 sort -s -t ' ' -k 4.9n -k 4.5M -k 4.2n -k 4.14,4.21 file*.log |
4208 sort -s -t '.' -k 1,1n -k 2,2n -k 3,3n -k 4,4n
4211 This example cannot be done with a single @command{sort} invocation,
4212 since IPv4 address components are separated by @samp{.} while dates
4213 come just after a space. So it is broken down into two invocations of
4214 @command{sort}: the first sorts by time stamp and the second by IPv4
4215 address. The time stamp is sorted by year, then month, then day, and
4216 finally by hour-minute-second field, using @option{-k} to isolate each
4217 field. Except for hour-minute-second there's no need to specify the
4218 end of each key field, since the @samp{n} and @samp{M} modifiers sort
4219 based on leading prefixes that cannot cross field boundaries. The
4220 IPv4 addresses are sorted lexicographically. The second sort uses
4221 @samp{-s} so that ties in the primary key are broken by the secondary
4222 key; the first sort uses @samp{-s} so that the combination of the two
4226 Generate a tags file in case-insensitive sorted order.
4229 find src -type f -print0 | sort -z -f | xargs -0 etags --append
4232 The use of @option{-print0}, @option{-z}, and @option{-0} in this case means
4233 that file names that contain blanks or other special characters are
4235 by the sort operation.
4237 @c This example is a bit contrived and needs more explanation.
4239 @c Sort records separated by an arbitrary string by using a pipe to convert
4240 @c each record delimiter string to @samp{\0}, then using sort's -z option,
4241 @c and converting each @samp{\0} back to the original record delimiter.
4244 @c printf 'c\n\nb\n\na\n'|perl -0pe 's/\n\n/\n\0/g'|sort -z|perl -0pe 's/\0/\n/g'
4248 Use the common @acronym{DSU, Decorate Sort Undecorate} idiom to
4249 sort lines according to their length.
4252 awk '@{print length, $0@}' /etc/passwd | sort -n | cut -f2- -d' '
4255 In general this technique can be used to sort data that the @command{sort}
4256 command does not support, or is inefficient at, sorting directly.
4259 Shuffle a list of directories, but preserve the order of files within
4260 each directory. For instance, one could use this to generate a music
4261 playlist in which albums are shuffled but the songs of each album are
4265 ls */* | sort -t / -k 1,1R -k 2,2
4271 @node shuf invocation
4272 @section @command{shuf}: Shuffling text
4275 @cindex shuffling files
4277 @command{shuf} shuffles its input by outputting a random permutation
4278 of its input lines. Each output permutation is equally likely.
4282 shuf [@var{option}]@dots{} [@var{file}]
4283 shuf -e [@var{option}]@dots{} [@var{arg}]@dots{}
4284 shuf -i @var{lo}-@var{hi} [@var{option}]@dots{}
4287 @command{shuf} has three modes of operation that affect where it
4288 obtains its input lines. By default, it reads lines from standard
4289 input. The following options change the operation mode:
4297 @cindex command-line operands to shuffle
4298 Treat each command-line operand as an input line.
4300 @item -i @var{lo}-@var{hi}
4301 @itemx --input-range=@var{lo}-@var{hi}
4303 @opindex --input-range
4304 @cindex input range to shuffle
4305 Act as if input came from a file containing the range of unsigned
4306 decimal integers @var{lo}@dots{}@var{hi}, one per line.
4310 @command{shuf}'s other options can affect its behavior in all
4315 @item -n @var{lines}
4316 @itemx --head-count=@var{count}
4318 @opindex --head-count
4319 @cindex head of output
4320 Output at most @var{count} lines. By default, all input lines are
4323 @item -o @var{output-file}
4324 @itemx --output=@var{output-file}
4327 @cindex overwriting of input, allowed
4328 Write output to @var{output-file} instead of standard output.
4329 @command{shuf} reads all input before opening
4330 @var{output-file}, so you can safely shuffle a file in place by using
4331 commands like @code{shuf -o F <F} and @code{cat F | shuf -o F}.
4333 @item --random-source=@var{file}
4334 @opindex --random-source
4335 @cindex random source for shuffling
4336 Use @var{file} as a source of random data used to determine which
4337 permutation to generate. @xref{Random sources}.
4339 @zeroTerminatedOption
4355 might produce the output
4365 Similarly, the command:
4368 shuf -e clubs hearts diamonds spades
4382 and the command @samp{shuf -i 1-4} might output:
4392 These examples all have four input lines, so @command{shuf} might
4393 produce any of the twenty-four possible permutations of the input. In
4394 general, if there are @var{n} input lines, there are @var{n}! (i.e.,
4395 @var{n} factorial, or @var{n} * (@var{n} - 1) * @dots{} * 1) possible
4396 output permutations.
4401 @node uniq invocation
4402 @section @command{uniq}: Uniquify files
4405 @cindex uniquify files
4407 @command{uniq} writes the unique lines in the given @file{input}, or
4408 standard input if nothing is given or for an @var{input} name of
4412 uniq [@var{option}]@dots{} [@var{input} [@var{output}]]
4415 By default, @command{uniq} prints its input lines, except that
4416 it discards all but the first of adjacent repeated lines, so that
4417 no output lines are repeated. Optionally, it can instead discard
4418 lines that are not repeated, or all repeated lines.
4420 The input need not be sorted, but repeated input lines are detected
4421 only if they are adjacent. If you want to discard non-adjacent
4422 duplicate lines, perhaps you want to use @code{sort -u}.
4423 @xref{sort invocation}.
4426 Comparisons honor the rules specified by the @env{LC_COLLATE}
4429 If no @var{output} file is specified, @command{uniq} writes to standard
4432 The program accepts the following options. Also see @ref{Common options}.
4437 @itemx --skip-fields=@var{n}
4439 @opindex --skip-fields
4440 Skip @var{n} fields on each line before checking for uniqueness. Use
4441 a null string for comparison if a line has fewer than @var{n} fields. Fields
4442 are sequences of non-space non-tab characters that are separated from
4443 each other by at least one space or tab.
4445 For compatibility @command{uniq} supports an obsolete option syntax
4446 @option{-@var{n}}. New scripts should use @option{-f @var{n}} instead.
4449 @itemx --skip-chars=@var{n}
4451 @opindex --skip-chars
4452 Skip @var{n} characters before checking for uniqueness. Use a null string
4453 for comparison if a line has fewer than @var{n} characters. If you use both
4454 the field and character skipping options, fields are skipped over first.
4456 @vindex _POSIX2_VERSION
4457 On older systems, @command{uniq} supports an obsolete option syntax
4459 This obsolete behavior can be enabled or disabled with the
4460 @env{_POSIX2_VERSION} environment variable (@pxref{Standards
4461 conformance}), but portable scripts should avoid commands whose
4462 behavior depends on this variable.
4463 For example, use @samp{uniq ./+10} or @samp{uniq -s 10} rather than
4464 the ambiguous @samp{uniq +10}.
4470 Print the number of times each line occurred along with the line.
4473 @itemx --ignore-case
4475 @opindex --ignore-case
4476 Ignore differences in case when comparing lines.
4482 @cindex repeated lines, outputting
4483 Discard lines that are not repeated. When used by itself, this option
4484 causes @command{uniq} to print the first copy of each repeated line,
4488 @itemx --all-repeated[=@var{delimit-method}]
4490 @opindex --all-repeated
4491 @cindex all repeated lines, outputting
4492 Do not discard the second and subsequent repeated input lines,
4493 but discard lines that are not repeated.
4494 This option is useful mainly in conjunction with other options e.g.,
4495 to ignore case or to compare only selected fields.
4496 The optional @var{delimit-method} tells how to delimit
4497 groups of repeated lines, and must be one of the following:
4502 Do not delimit groups of repeated lines.
4503 This is equivalent to @option{--all-repeated} (@option{-D}).
4506 Output a newline before each group of repeated lines.
4507 With @option{--zero-terminated} (@option{-z}), use a zero
4508 byte (@acronym{ASCII} @sc{nul}) instead of a newline.
4511 Separate groups of repeated lines with a single newline.
4512 With @option{--zero-terminated} (@option{-z}), use a zero
4513 byte (@acronym{ASCII} @sc{nul}) instead of a newline.
4514 This is the same as using @samp{prepend}, except that
4515 no delimiter is inserted before the first group, and hence
4516 may be better suited for output direct to users.
4519 Note that when groups are delimited and the input stream contains
4520 two or more consecutive blank lines, then the output is ambiguous.
4521 To avoid that, filter the input through @samp{tr -s '\n'} to replace
4522 each sequence of consecutive newlines with a single newline.
4524 This is a @sc{gnu} extension.
4525 @c FIXME: give an example showing *how* it's useful
4531 @cindex unique lines, outputting
4532 Discard the first repeated line. When used by itself, this option
4533 causes @command{uniq} to print unique lines, and nothing else.
4536 @itemx --check-chars=@var{n}
4538 @opindex --check-chars
4539 Compare at most @var{n} characters on each line (after skipping any specified
4540 fields and characters). By default the entire rest of the lines are
4543 @zeroTerminatedOption
4550 @node comm invocation
4551 @section @command{comm}: Compare two sorted files line by line
4554 @cindex line-by-line comparison
4555 @cindex comparing sorted files
4557 @command{comm} writes to standard output lines that are common, and lines
4558 that are unique, to two input files; a file name of @samp{-} means
4559 standard input. Synopsis:
4562 comm [@var{option}]@dots{} @var{file1} @var{file2}
4566 Before @command{comm} can be used, the input files must be sorted using the
4567 collating sequence specified by the @env{LC_COLLATE} locale.
4568 If an input file ends in a non-newline
4569 character, a newline is silently appended. The @command{sort} command with
4570 no options always outputs a file that is suitable input to @command{comm}.
4572 @cindex differing lines
4573 @cindex common lines
4574 With no options, @command{comm} produces three-column output. Column one
4575 contains lines unique to @var{file1}, column two contains lines unique
4576 to @var{file2}, and column three contains lines common to both files.
4577 Columns are separated by a single TAB character.
4578 @c FIXME: when there's an option to supply an alternative separator
4579 @c string, append `by default' to the above sentence.
4584 The options @option{-1}, @option{-2}, and @option{-3} suppress printing of
4585 the corresponding columns (and separators). Also see @ref{Common options}.
4587 Unlike some other comparison utilities, @command{comm} has an exit
4588 status that does not depend on the result of the comparison.
4589 Upon normal completion @command{comm} produces an exit code of zero.
4590 If there is an error it exits with nonzero status.
4592 @macro checkOrderOption{cmd}
4593 If the @option{--check-order} option is given, unsorted inputs will
4594 cause a fatal error message. If the option @option{--nocheck-order}
4595 is given, unsorted inputs will never cause an error message. If
4596 neither of these options is given, wrongly sorted inputs are diagnosed
4597 only if an input file is found to contain unpairable lines. If an
4598 input file is diagnosed as being unsorted, the @command{\cmd\} command
4599 will exit with a nonzero status (and the output should not be used).
4601 Forcing @command{\cmd\} to process wrongly sorted input files
4602 containing unpairable lines by specifying @option{--nocheck-order} is
4603 not guaranteed to produce any particular output. The output will
4604 probably not correspond with whatever you hoped it would be.
4606 @checkOrderOption{comm}
4611 Fail with an error message if either input file is wrongly ordered.
4613 @item --nocheck-order
4614 Do not check that both input files are in sorted order.
4618 @item --output-delimiter=@var{str}
4619 Print @var{str} between adjacent output columns,
4620 rather than the default of a single TAB character.
4622 The delimiter @var{str} may not be empty.
4626 @node ptx invocation
4627 @section @command{ptx}: Produce permuted indexes
4631 @command{ptx} reads a text file and essentially produces a permuted index, with
4632 each keyword in its context. The calling sketch is either one of:
4635 ptx [@var{option} @dots{}] [@var{file} @dots{}]
4636 ptx -G [@var{option} @dots{}] [@var{input} [@var{output}]]
4639 The @option{-G} (or its equivalent: @option{--traditional}) option disables
4640 all @sc{gnu} extensions and reverts to traditional mode, thus introducing some
4641 limitations and changing several of the program's default option values.
4642 When @option{-G} is not specified, @sc{gnu} extensions are always enabled.
4643 @sc{gnu} extensions to @command{ptx} are documented wherever appropriate in this
4644 document. For the full list, see @xref{Compatibility in ptx}.
4646 Individual options are explained in the following sections.
4648 When @sc{gnu} extensions are enabled, there may be zero, one or several
4649 @var{file}s after the options. If there is no @var{file}, the program
4650 reads the standard input. If there is one or several @var{file}s, they
4651 give the name of input files which are all read in turn, as if all the
4652 input files were concatenated. However, there is a full contextual
4653 break between each file and, when automatic referencing is requested,
4654 file names and line numbers refer to individual text input files. In
4655 all cases, the program outputs the permuted index to the standard
4658 When @sc{gnu} extensions are @emph{not} enabled, that is, when the program
4659 operates in traditional mode, there may be zero, one or two parameters
4660 besides the options. If there are no parameters, the program reads the
4661 standard input and outputs the permuted index to the standard output.
4662 If there is only one parameter, it names the text @var{input} to be read
4663 instead of the standard input. If two parameters are given, they give
4664 respectively the name of the @var{input} file to read and the name of
4665 the @var{output} file to produce. @emph{Be very careful} to note that,
4666 in this case, the contents of file given by the second parameter is
4667 destroyed. This behavior is dictated by System V @command{ptx}
4668 compatibility; @sc{gnu} Standards normally discourage output parameters not
4669 introduced by an option.
4671 Note that for @emph{any} file named as the value of an option or as an
4672 input text file, a single dash @kbd{-} may be used, in which case
4673 standard input is assumed. However, it would not make sense to use this
4674 convention more than once per program invocation.
4677 * General options in ptx:: Options which affect general program behavior.
4678 * Charset selection in ptx:: Underlying character set considerations.
4679 * Input processing in ptx:: Input fields, contexts, and keyword selection.
4680 * Output formatting in ptx:: Types of output format, and sizing the fields.
4681 * Compatibility in ptx::
4685 @node General options in ptx
4686 @subsection General options
4691 @itemx --traditional
4692 As already explained, this option disables all @sc{gnu} extensions to
4693 @command{ptx} and switches to traditional mode.
4696 Print a short help on standard output, then exit without further
4700 Print the program version on standard output, then exit without further
4708 @node Charset selection in ptx
4709 @subsection Charset selection
4711 @c FIXME: People don't necessarily know what an IBM-PC was these days.
4712 As it is set up now, the program assumes that the input file is coded
4713 using 8-bit @acronym{ISO} 8859-1 code, also known as Latin-1 character set,
4714 @emph{unless} it is compiled for MS-DOS, in which case it uses the
4715 character set of the IBM-PC@. (@sc{gnu} @command{ptx} is not known to work on
4716 smaller MS-DOS machines anymore.) Compared to 7-bit @acronym{ASCII}, the set
4717 of characters which are letters is different; this alters the behavior
4718 of regular expression matching. Thus, the default regular expression
4719 for a keyword allows foreign or diacriticized letters. Keyword sorting,
4720 however, is still crude; it obeys the underlying character set ordering
4726 @itemx --ignore-case
4727 Fold lower case letters to upper case for sorting.
4732 @node Input processing in ptx
4733 @subsection Word selection and input processing
4738 @itemx --break-file=@var{file}
4740 This option provides an alternative (to @option{-W}) method of describing
4741 which characters make up words. It introduces the name of a
4742 file which contains a list of characters which can@emph{not} be part of
4743 one word; this file is called the @dfn{Break file}. Any character which
4744 is not part of the Break file is a word constituent. If both options
4745 @option{-b} and @option{-W} are specified, then @option{-W} has precedence and
4746 @option{-b} is ignored.
4748 When @sc{gnu} extensions are enabled, the only way to avoid newline as a
4749 break character is to write all the break characters in the file with no
4750 newline at all, not even at the end of the file. When @sc{gnu} extensions
4751 are disabled, spaces, tabs and newlines are always considered as break
4752 characters even if not included in the Break file.
4755 @itemx --ignore-file=@var{file}
4757 The file associated with this option contains a list of words which will
4758 never be taken as keywords in concordance output. It is called the
4759 @dfn{Ignore file}. The file contains exactly one word in each line; the
4760 end of line separation of words is not subject to the value of the
4764 @itemx --only-file=@var{file}
4766 The file associated with this option contains a list of words which will
4767 be retained in concordance output; any word not mentioned in this file
4768 is ignored. The file is called the @dfn{Only file}. The file contains
4769 exactly one word in each line; the end of line separation of words is
4770 not subject to the value of the @option{-S} option.
4772 There is no default for the Only file. When both an Only file and an
4773 Ignore file are specified, a word is considered a keyword only
4774 if it is listed in the Only file and not in the Ignore file.
4779 On each input line, the leading sequence of non-white space characters will be
4780 taken to be a reference that has the purpose of identifying this input
4781 line in the resulting permuted index. For more information about reference
4782 production, see @xref{Output formatting in ptx}.
4783 Using this option changes the default value for option @option{-S}.
4785 Using this option, the program does not try very hard to remove
4786 references from contexts in output, but it succeeds in doing so
4787 @emph{when} the context ends exactly at the newline. If option
4788 @option{-r} is used with @option{-S} default value, or when @sc{gnu} extensions
4789 are disabled, this condition is always met and references are completely
4790 excluded from the output contexts.
4792 @item -S @var{regexp}
4793 @itemx --sentence-regexp=@var{regexp}
4795 This option selects which regular expression will describe the end of a
4796 line or the end of a sentence. In fact, this regular expression is not
4797 the only distinction between end of lines or end of sentences, and input
4798 line boundaries have no special significance outside this option. By
4799 default, when @sc{gnu} extensions are enabled and if @option{-r} option is not
4800 used, end of sentences are used. In this case, this @var{regex} is
4801 imported from @sc{gnu} Emacs:
4804 [.?!][]\"')@}]*\\($\\|\t\\| \\)[ \t\n]*
4807 Whenever @sc{gnu} extensions are disabled or if @option{-r} option is used, end
4808 of lines are used; in this case, the default @var{regexp} is just:
4814 Using an empty @var{regexp} is equivalent to completely disabling end of
4815 line or end of sentence recognition. In this case, the whole file is
4816 considered to be a single big line or sentence. The user might want to
4817 disallow all truncation flag generation as well, through option @option{-F
4818 ""}. @xref{Regexps, , Syntax of Regular Expressions, emacs, The GNU Emacs
4821 When the keywords happen to be near the beginning of the input line or
4822 sentence, this often creates an unused area at the beginning of the
4823 output context line; when the keywords happen to be near the end of the
4824 input line or sentence, this often creates an unused area at the end of
4825 the output context line. The program tries to fill those unused areas
4826 by wrapping around context in them; the tail of the input line or
4827 sentence is used to fill the unused area on the left of the output line;
4828 the head of the input line or sentence is used to fill the unused area
4829 on the right of the output line.
4831 As a matter of convenience to the user, many usual backslashed escape
4832 sequences from the C language are recognized and converted to the
4833 corresponding characters by @command{ptx} itself.
4835 @item -W @var{regexp}
4836 @itemx --word-regexp=@var{regexp}
4838 This option selects which regular expression will describe each keyword.
4839 By default, if @sc{gnu} extensions are enabled, a word is a sequence of
4840 letters; the @var{regexp} used is @samp{\w+}. When @sc{gnu} extensions are
4841 disabled, a word is by default anything which ends with a space, a tab
4842 or a newline; the @var{regexp} used is @samp{[^ \t\n]+}.
4844 An empty @var{regexp} is equivalent to not using this option.
4845 @xref{Regexps, , Syntax of Regular Expressions, emacs, The GNU Emacs
4848 As a matter of convenience to the user, many usual backslashed escape
4849 sequences, as found in the C language, are recognized and converted to
4850 the corresponding characters by @command{ptx} itself.
4855 @node Output formatting in ptx
4856 @subsection Output formatting
4858 Output format is mainly controlled by the @option{-O} and @option{-T} options
4859 described in the table below. When neither @option{-O} nor @option{-T} are
4860 selected, and if @sc{gnu} extensions are enabled, the program chooses an
4861 output format suitable for a dumb terminal. Each keyword occurrence is
4862 output to the center of one line, surrounded by its left and right
4863 contexts. Each field is properly justified, so the concordance output
4864 can be readily observed. As a special feature, if automatic
4865 references are selected by option @option{-A} and are output before the
4866 left context, that is, if option @option{-R} is @emph{not} selected, then
4867 a colon is added after the reference; this nicely interfaces with @sc{gnu}
4868 Emacs @code{next-error} processing. In this default output format, each
4869 white space character, like newline and tab, is merely changed to
4870 exactly one space, with no special attempt to compress consecutive
4871 spaces. This might change in the future. Except for those white space
4872 characters, every other character of the underlying set of 256
4873 characters is transmitted verbatim.
4875 Output format is further controlled by the following options.
4879 @item -g @var{number}
4880 @itemx --gap-size=@var{number}
4882 Select the size of the minimum white space gap between the fields on the
4885 @item -w @var{number}
4886 @itemx --width=@var{number}
4888 Select the maximum output width of each final line. If references are
4889 used, they are included or excluded from the maximum output width
4890 depending on the value of option @option{-R}. If this option is not
4891 selected, that is, when references are output before the left context,
4892 the maximum output width takes into account the maximum length of all
4893 references. If this option is selected, that is, when references are
4894 output after the right context, the maximum output width does not take
4895 into account the space taken by references, nor the gap that precedes
4899 @itemx --auto-reference
4901 Select automatic references. Each input line will have an automatic
4902 reference made up of the file name and the line ordinal, with a single
4903 colon between them. However, the file name will be empty when standard
4904 input is being read. If both @option{-A} and @option{-r} are selected, then
4905 the input reference is still read and skipped, but the automatic
4906 reference is used at output time, overriding the input reference.
4909 @itemx --right-side-refs
4911 In the default output format, when option @option{-R} is not used, any
4912 references produced by the effect of options @option{-r} or @option{-A} are
4913 placed to the far right of output lines, after the right context. With
4914 default output format, when the @option{-R} option is specified, references
4915 are rather placed at the beginning of each output line, before the left
4916 context. For any other output format, option @option{-R} is
4917 ignored, with one exception: with @option{-R} the width of references
4918 is @emph{not} taken into account in total output width given by @option{-w}.
4920 This option is automatically selected whenever @sc{gnu} extensions are
4923 @item -F @var{string}
4924 @itemx --flac-truncation=@var{string}
4926 This option will request that any truncation in the output be reported
4927 using the string @var{string}. Most output fields theoretically extend
4928 towards the beginning or the end of the current line, or current
4929 sentence, as selected with option @option{-S}. But there is a maximum
4930 allowed output line width, changeable through option @option{-w}, which is
4931 further divided into space for various output fields. When a field has
4932 to be truncated because it cannot extend beyond the beginning or the end of
4933 the current line to fit in, then a truncation occurs. By default,
4934 the string used is a single slash, as in @option{-F /}.
4936 @var{string} may have more than one character, as in @option{-F ...}.
4937 Also, in the particular case when @var{string} is empty (@option{-F ""}),
4938 truncation flagging is disabled, and no truncation marks are appended in
4941 As a matter of convenience to the user, many usual backslashed escape
4942 sequences, as found in the C language, are recognized and converted to
4943 the corresponding characters by @command{ptx} itself.
4945 @item -M @var{string}
4946 @itemx --macro-name=@var{string}
4948 Select another @var{string} to be used instead of @samp{xx}, while
4949 generating output suitable for @command{nroff}, @command{troff} or @TeX{}.
4952 @itemx --format=roff
4954 Choose an output format suitable for @command{nroff} or @command{troff}
4955 processing. Each output line will look like:
4958 .xx "@var{tail}" "@var{before}" "@var{keyword_and_after}" "@var{head}" "@var{ref}"
4961 so it will be possible to write a @samp{.xx} roff macro to take care of
4962 the output typesetting. This is the default output format when @sc{gnu}
4963 extensions are disabled. Option @option{-M} can be used to change
4964 @samp{xx} to another macro name.
4966 In this output format, each non-graphical character, like newline and
4967 tab, is merely changed to exactly one space, with no special attempt to
4968 compress consecutive spaces. Each quote character: @kbd{"} is doubled
4969 so it will be correctly processed by @command{nroff} or @command{troff}.
4974 Choose an output format suitable for @TeX{} processing. Each output
4975 line will look like:
4978 \xx @{@var{tail}@}@{@var{before}@}@{@var{keyword}@}@{@var{after}@}@{@var{head}@}@{@var{ref}@}
4982 so it will be possible to write a @code{\xx} definition to take care of
4983 the output typesetting. Note that when references are not being
4984 produced, that is, neither option @option{-A} nor option @option{-r} is
4985 selected, the last parameter of each @code{\xx} call is inhibited.
4986 Option @option{-M} can be used to change @samp{xx} to another macro
4989 In this output format, some special characters, like @kbd{$}, @kbd{%},
4990 @kbd{&}, @kbd{#} and @kbd{_} are automatically protected with a
4991 backslash. Curly brackets @kbd{@{}, @kbd{@}} are protected with a
4992 backslash and a pair of dollar signs (to force mathematical mode). The
4993 backslash itself produces the sequence @code{\backslash@{@}}.
4994 Circumflex and tilde diacritical marks produce the sequence @code{^\@{ @}} and
4995 @code{~\@{ @}} respectively. Other diacriticized characters of the
4996 underlying character set produce an appropriate @TeX{} sequence as far
4997 as possible. The other non-graphical characters, like newline and tab,
4998 and all other characters which are not part of @acronym{ASCII}, are merely
4999 changed to exactly one space, with no special attempt to compress
5000 consecutive spaces. Let me know how to improve this special character
5001 processing for @TeX{}.
5006 @node Compatibility in ptx
5007 @subsection The @sc{gnu} extensions to @command{ptx}
5009 This version of @command{ptx} contains a few features which do not exist in
5010 System V @command{ptx}. These extra features are suppressed by using the
5011 @option{-G} command line option, unless overridden by other command line
5012 options. Some @sc{gnu} extensions cannot be recovered by overriding, so the
5013 simple rule is to avoid @option{-G} if you care about @sc{gnu} extensions.
5014 Here are the differences between this program and System V @command{ptx}.
5019 This program can read many input files at once, it always writes the
5020 resulting concordance on standard output. On the other hand, System V
5021 @command{ptx} reads only one file and sends the result to standard output
5022 or, if a second @var{file} parameter is given on the command, to that
5025 Having output parameters not introduced by options is a dangerous
5026 practice which @sc{gnu} avoids as far as possible. So, for using @command{ptx}
5027 portably between @sc{gnu} and System V, you should always use it with a
5028 single input file, and always expect the result on standard output. You
5029 might also want to automatically configure in a @option{-G} option to
5030 @command{ptx} calls in products using @command{ptx}, if the configurator finds
5031 that the installed @command{ptx} accepts @option{-G}.
5034 The only options available in System V @command{ptx} are options @option{-b},
5035 @option{-f}, @option{-g}, @option{-i}, @option{-o}, @option{-r}, @option{-t} and
5036 @option{-w}. All other options are @sc{gnu} extensions and are not repeated in
5037 this enumeration. Moreover, some options have a slightly different
5038 meaning when @sc{gnu} extensions are enabled, as explained below.
5041 By default, concordance output is not formatted for @command{troff} or
5042 @command{nroff}. It is rather formatted for a dumb terminal. @command{troff}
5043 or @command{nroff} output may still be selected through option @option{-O}.
5046 Unless @option{-R} option is used, the maximum reference width is
5047 subtracted from the total output line width. With @sc{gnu} extensions
5048 disabled, width of references is not taken into account in the output
5049 line width computations.
5052 All 256 bytes, even @acronym{ASCII} @sc{nul} bytes, are always read and
5053 processed from input file with no adverse effect, even if @sc{gnu} extensions
5054 are disabled. However, System V @command{ptx} does not accept 8-bit characters,
5055 a few control characters are rejected, and the tilde @kbd{~} is also rejected.
5058 Input line length is only limited by available memory, even if @sc{gnu}
5059 extensions are disabled. However, System V @command{ptx} processes only
5060 the first 200 characters in each line.
5063 The break (non-word) characters default to be every character except all
5064 letters of the underlying character set, diacriticized or not. When @sc{gnu}
5065 extensions are disabled, the break characters default to space, tab and
5069 The program makes better use of output line width. If @sc{gnu} extensions
5070 are disabled, the program rather tries to imitate System V @command{ptx},
5071 but still, there are some slight disposition glitches this program does
5072 not completely reproduce.
5075 The user can specify both an Ignore file and an Only file. This is not
5076 allowed with System V @command{ptx}.
5081 @node tsort invocation
5082 @section @command{tsort}: Topological sort
5085 @cindex topological sort
5087 @command{tsort} performs a topological sort on the given @var{file}, or
5088 standard input if no input file is given or for a @var{file} of
5089 @samp{-}. For more details and some history, see @ref{tsort background}.
5093 tsort [@var{option}] [@var{file}]
5096 @command{tsort} reads its input as pairs of strings, separated by blanks,
5097 indicating a partial ordering. The output is a total ordering that
5098 corresponds to the given partial ordering.
5112 will produce the output
5123 Consider a more realistic example.
5124 You have a large set of functions all in one file, and they may all be
5125 declared static except one. Currently that one (say @code{main}) is the
5126 first function defined in the file, and the ones it calls directly follow
5127 it, followed by those they call, etc. Let's say that you are determined
5128 to take advantage of prototypes, so you have to choose between declaring
5129 all of those functions (which means duplicating a lot of information from
5130 the definitions) and rearranging the functions so that as many as possible
5131 are defined before they are used. One way to automate the latter process
5132 is to get a list for each function of the functions it calls directly.
5133 Many programs can generate such lists. They describe a call graph.
5134 Consider the following list, in which a given line indicates that the
5135 function on the left calls the one on the right directly.
5141 tail_file pretty_name
5142 tail_file write_header
5144 tail_forever recheck
5145 tail_forever pretty_name
5146 tail_forever write_header
5147 tail_forever dump_remainder
5150 tail_lines start_lines
5151 tail_lines dump_remainder
5152 tail_lines file_lines
5153 tail_lines pipe_lines
5155 tail_bytes start_bytes
5156 tail_bytes dump_remainder
5157 tail_bytes pipe_bytes
5158 file_lines dump_remainder
5162 then you can use @command{tsort} to produce an ordering of those
5163 functions that satisfies your requirement.
5166 example$ tsort call-graph | tac
5186 @command{tsort} detects any cycles in the input and writes the first cycle
5187 encountered to standard error.
5189 Note that for a given partial ordering, generally there is no unique
5190 total ordering. In the context of the call graph above, the function
5191 @code{parse_options} may be placed anywhere in the list as long as it
5192 precedes @code{main}.
5194 The only options are @option{--help} and @option{--version}. @xref{Common
5200 * tsort background:: Where tsort came from.
5203 @node tsort background
5204 @subsection @command{tsort}: Background
5206 @command{tsort} exists because very early versions of the Unix linker processed
5207 an archive file exactly once, and in order. As @command{ld} read each object
5208 in the archive, it decided whether it was needed in the program based on
5209 whether it defined any symbols which were undefined at that point in
5212 This meant that dependencies within the archive had to be handled
5213 specially. For example, @code{scanf} probably calls @code{read}. That means
5214 that in a single pass through an archive, it was important for @code{scanf.o}
5215 to appear before read.o, because otherwise a program which calls
5216 @code{scanf} but not @code{read} might end up with an unexpected unresolved
5217 reference to @code{read}.
5219 The way to address this problem was to first generate a set of
5220 dependencies of one object file on another. This was done by a shell
5221 script called @command{lorder}. The GNU tools don't provide a version of
5222 lorder, as far as I know, but you can still find it in BSD
5225 Then you ran @command{tsort} over the @command{lorder} output, and you used the
5226 resulting sort to define the order in which you added objects to the archive.
5228 This whole procedure has been obsolete since about 1980, because
5229 Unix archives now contain a symbol table (traditionally built by
5230 @command{ranlib}, now generally built by @command{ar} itself), and the Unix
5231 linker uses the symbol table to effectively make multiple passes over
5234 Anyhow, that's where tsort came from. To solve an old problem with
5235 the way the linker handled archive files, which has since been solved
5239 @node Operating on fields within a line
5240 @chapter Operating on fields within a line
5243 * cut invocation:: Print selected parts of lines.
5244 * paste invocation:: Merge lines of files.
5245 * join invocation:: Join lines on a common field.
5249 @node cut invocation
5250 @section @command{cut}: Print selected parts of lines
5253 @command{cut} writes to standard output selected parts of each line of each
5254 input file, or standard input if no files are given or for a file name of
5258 cut @var{option}@dots{} [@var{file}]@dots{}
5261 In the table which follows, the @var{byte-list}, @var{character-list},
5262 and @var{field-list} are one or more numbers or ranges (two numbers
5263 separated by a dash) separated by commas. Bytes, characters, and
5264 fields are numbered starting at 1. Incomplete ranges may be
5265 given: @option{-@var{m}} means @samp{1-@var{m}}; @samp{@var{n}-} means
5266 @samp{@var{n}} through end of line or last field. The list elements
5267 can be repeated, can overlap, and can be specified in any order; but
5268 the selected input is written in the same order that it is read, and
5269 is written exactly once.
5271 The program accepts the following options. Also see @ref{Common
5276 @item -b @var{byte-list}
5277 @itemx --bytes=@var{byte-list}
5280 Select for printing only the bytes in positions listed in
5281 @var{byte-list}. Tabs and backspaces are treated like any other
5282 character; they take up 1 byte. If an output delimiter is specified,
5283 (see the description of @option{--output-delimiter}), then output that
5284 string between ranges of selected bytes.
5286 @item -c @var{character-list}
5287 @itemx --characters=@var{character-list}
5289 @opindex --characters
5290 Select for printing only the characters in positions listed in
5291 @var{character-list}. The same as @option{-b} for now, but
5292 internationalization will change that. Tabs and backspaces are
5293 treated like any other character; they take up 1 character. If an
5294 output delimiter is specified, (see the description of
5295 @option{--output-delimiter}), then output that string between ranges
5298 @item -f @var{field-list}
5299 @itemx --fields=@var{field-list}
5302 Select for printing only the fields listed in @var{field-list}.
5303 Fields are separated by a TAB character by default. Also print any
5304 line that contains no delimiter character, unless the
5305 @option{--only-delimited} (@option{-s}) option is specified
5307 @item -d @var{input_delim_byte}
5308 @itemx --delimiter=@var{input_delim_byte}
5310 @opindex --delimiter
5311 With @option{-f}, use the first byte of @var{input_delim_byte} as
5312 the input fields separator (default is TAB).
5316 Do not split multi-byte characters (no-op for now).
5319 @itemx --only-delimited
5321 @opindex --only-delimited
5322 For @option{-f}, do not print lines that do not contain the field separator
5323 character. Normally, any line without a field separator is printed verbatim.
5325 @item --output-delimiter=@var{output_delim_string}
5326 @opindex --output-delimiter
5327 With @option{-f}, output fields are separated by @var{output_delim_string}.
5328 The default with @option{-f} is to use the input delimiter.
5329 When using @option{-b} or @option{-c} to select ranges of byte or
5330 character offsets (as opposed to ranges of fields),
5331 output @var{output_delim_string} between non-overlapping
5332 ranges of selected bytes.
5335 @opindex --complement
5336 This option is a @acronym{GNU} extension.
5337 Select for printing the complement of the bytes, characters or fields
5338 selected with the @option{-b}, @option{-c} or @option{-f} options.
5339 In other words, do @emph{not} print the bytes, characters or fields
5340 specified via those options. This option is useful when you have
5341 many fields and want to print all but a few of them.
5348 @node paste invocation
5349 @section @command{paste}: Merge lines of files
5352 @cindex merging files
5354 @command{paste} writes to standard output lines consisting of sequentially
5355 corresponding lines of each given file, separated by a TAB character.
5356 Standard input is used for a file name of @samp{-} or if no input files
5378 paste [@var{option}]@dots{} [@var{file}]@dots{}
5381 The program accepts the following options. Also see @ref{Common options}.
5389 Paste the lines of one file at a time rather than one line from each
5390 file. Using the above example data:
5393 $ paste -s num2 let3
5398 @item -d @var{delim-list}
5399 @itemx --delimiters=@var{delim-list}
5401 @opindex --delimiters
5402 Consecutively use the characters in @var{delim-list} instead of
5403 TAB to separate merged lines. When @var{delim-list} is
5404 exhausted, start again at its beginning. Using the above example data:
5407 $ paste -d '%_' num2 let3 num2
5418 @node join invocation
5419 @section @command{join}: Join lines on a common field
5422 @cindex common field, joining on
5424 @command{join} writes to standard output a line for each pair of input
5425 lines that have identical join fields. Synopsis:
5428 join [@var{option}]@dots{} @var{file1} @var{file2}
5431 Either @var{file1} or @var{file2} (but not both) can be @samp{-},
5432 meaning standard input. @var{file1} and @var{file2} should be
5433 sorted on the join fields.
5436 Normally, the sort order is that of the
5437 collating sequence specified by the @env{LC_COLLATE} locale. Unless
5438 the @option{-t} option is given, the sort comparison ignores blanks at
5439 the start of the join field, as in @code{sort -b}. If the
5440 @option{--ignore-case} option is given, the sort comparison ignores
5441 the case of characters in the join field, as in @code{sort -f}.
5443 The @command{sort} and @command{join} commands should use consistent
5444 locales and options if the output of @command{sort} is fed to
5445 @command{join}. You can use a command like @samp{sort -k 1b,1} to
5446 sort a file on its default join field, but if you select a non-default
5447 locale, join field, separator, or comparison options, then you should
5448 do so consistently between @command{join} and @command{sort}.
5450 If the input has no unpairable lines, a @acronym{GNU} extension is
5451 available; the sort order can be any order that considers two fields
5452 to be equal if and only if the sort comparison described above
5453 considers them to be equal. For example:
5470 @checkOrderOption{join}
5474 @item the join field is the first field in each line;
5475 @item fields in the input are separated by one or more blanks, with leading
5476 blanks on the line ignored;
5477 @item fields in the output are separated by a space;
5478 @item each output line consists of the join field, the remaining
5479 fields from @var{file1}, then the remaining fields from @var{file2}.
5482 The program accepts the following options. Also see @ref{Common options}.
5486 @item -a @var{file-number}
5488 Print a line for each unpairable line in file @var{file-number} (either
5489 @samp{1} or @samp{2}), in addition to the normal output.
5492 Fail with an error message if either input file is wrongly ordered.
5494 @item --nocheck-order
5495 Do not check that both input files are in sorted order. This is the default.
5497 @item -e @var{string}
5499 Replace those output fields that are missing in the input with
5503 @itemx --ignore-case
5505 @opindex --ignore-case
5506 Ignore differences in case when comparing keys.
5507 With this option, the lines of the input files must be ordered in the same way.
5508 Use @samp{sort -f} to produce this ordering.
5510 @item -1 @var{field}
5512 Join on field @var{field} (a positive integer) of file 1.
5514 @item -2 @var{field}
5516 Join on field @var{field} (a positive integer) of file 2.
5518 @item -j @var{field}
5519 Equivalent to @option{-1 @var{field} -2 @var{field}}.
5521 @item -o @var{field-list}
5522 Construct each output line according to the format in @var{field-list}.
5523 Each element in @var{field-list} is either the single character @samp{0} or
5524 has the form @var{m.n} where the file number, @var{m}, is @samp{1} or
5525 @samp{2} and @var{n} is a positive field number.
5527 A field specification of @samp{0} denotes the join field.
5528 In most cases, the functionality of the @samp{0} field spec
5529 may be reproduced using the explicit @var{m.n} that corresponds
5530 to the join field. However, when printing unpairable lines
5531 (using either of the @option{-a} or @option{-v} options), there is no way
5532 to specify the join field using @var{m.n} in @var{field-list}
5533 if there are unpairable lines in both files.
5534 To give @command{join} that functionality, @acronym{POSIX} invented the @samp{0}
5535 field specification notation.
5537 The elements in @var{field-list}
5538 are separated by commas or blanks.
5539 Blank separators typically need to be quoted for the shell. For
5540 example, the commands @samp{join -o 1.2,2.2} and @samp{join -o '1.2
5541 2.2'} are equivalent.
5543 All output lines---including those printed because of any -a or -v
5544 option---are subject to the specified @var{field-list}.
5547 Use character @var{char} as the input and output field separator.
5548 Treat as significant each occurrence of @var{char} in the input file.
5549 Use @samp{sort -t @var{char}}, without the @option{-b} option of
5550 @samp{sort}, to produce this ordering.
5552 @item -v @var{file-number}
5553 Print a line for each unpairable line in file @var{file-number}
5554 (either @samp{1} or @samp{2}), instead of the normal output.
5561 @node Operating on characters
5562 @chapter Operating on characters
5564 @cindex operating on characters
5566 This commands operate on individual characters.
5569 * tr invocation:: Translate, squeeze, and/or delete characters.
5570 * expand invocation:: Convert tabs to spaces.
5571 * unexpand invocation:: Convert spaces to tabs.
5576 @section @command{tr}: Translate, squeeze, and/or delete characters
5583 tr [@var{option}]@dots{} @var{set1} [@var{set2}]
5586 @command{tr} copies standard input to standard output, performing
5587 one of the following operations:
5591 translate, and optionally squeeze repeated characters in the result,
5593 squeeze repeated characters,
5597 delete characters, then squeeze repeated characters from the result.
5600 The @var{set1} and (if given) @var{set2} arguments define ordered
5601 sets of characters, referred to below as @var{set1} and @var{set2}. These
5602 sets are the characters of the input that @command{tr} operates on.
5603 The @option{--complement} (@option{-c}, @option{-C}) option replaces
5605 complement (all of the characters that are not in @var{set1}).
5607 Currently @command{tr} fully supports only single-byte characters.
5608 Eventually it will support multibyte characters; when it does, the
5609 @option{-C} option will cause it to complement the set of characters,
5610 whereas @option{-c} will cause it to complement the set of values.
5611 This distinction will matter only when some values are not characters,
5612 and this is possible only in locales using multibyte encodings when
5613 the input contains encoding errors.
5615 The program accepts the @option{--help} and @option{--version}
5616 options. @xref{Common options}. Options must precede operands.
5621 * Character sets:: Specifying sets of characters.
5622 * Translating:: Changing one set of characters to another.
5623 * Squeezing:: Squeezing repeats and deleting.
5627 @node Character sets
5628 @subsection Specifying sets of characters
5630 @cindex specifying sets of characters
5632 The format of the @var{set1} and @var{set2} arguments resembles
5633 the format of regular expressions; however, they are not regular
5634 expressions, only lists of characters. Most characters simply
5635 represent themselves in these strings, but the strings can contain
5636 the shorthands listed below, for convenience. Some of them can be
5637 used only in @var{set1} or @var{set2}, as noted below.
5641 @item Backslash escapes
5642 @cindex backslash escapes
5644 The following backslash escape sequences are recognized:
5662 The character with the value given by @var{ooo}, which is 1 to 3
5668 While a backslash followed by a character not listed above is
5669 interpreted as that character, the backslash also effectively
5670 removes any special significance, so it is useful to escape
5671 @samp{[}, @samp{]}, @samp{*}, and @samp{-}.
5676 The notation @samp{@var{m}-@var{n}} expands to all of the characters
5677 from @var{m} through @var{n}, in ascending order. @var{m} should
5678 collate before @var{n}; if it doesn't, an error results. As an example,
5679 @samp{0-9} is the same as @samp{0123456789}.
5681 @sc{gnu} @command{tr} does not support the System V syntax that uses square
5682 brackets to enclose ranges. Translations specified in that format
5683 sometimes work as expected, since the brackets are often transliterated
5684 to themselves. However, they should be avoided because they sometimes
5685 behave unexpectedly. For example, @samp{tr -d '[0-9]'} deletes brackets
5688 Many historically common and even accepted uses of ranges are not
5689 portable. For example, on @acronym{EBCDIC} hosts using the @samp{A-Z}
5690 range will not do what most would expect because @samp{A} through @samp{Z}
5691 are not contiguous as they are in @acronym{ASCII}.
5692 If you can rely on a @acronym{POSIX} compliant version of @command{tr}, then
5693 the best way to work around this is to use character classes (see below).
5694 Otherwise, it is most portable (and most ugly) to enumerate the members
5697 @item Repeated characters
5698 @cindex repeated characters
5700 The notation @samp{[@var{c}*@var{n}]} in @var{set2} expands to @var{n}
5701 copies of character @var{c}. Thus, @samp{[y*6]} is the same as
5702 @samp{yyyyyy}. The notation @samp{[@var{c}*]} in @var{string2} expands
5703 to as many copies of @var{c} as are needed to make @var{set2} as long as
5704 @var{set1}. If @var{n} begins with @samp{0}, it is interpreted in
5705 octal, otherwise in decimal.
5707 @item Character classes
5708 @cindex character classes
5710 The notation @samp{[:@var{class}:]} expands to all of the characters in
5711 the (predefined) class @var{class}. The characters expand in no
5712 particular order, except for the @code{upper} and @code{lower} classes,
5713 which expand in ascending order. When the @option{--delete} (@option{-d})
5714 and @option{--squeeze-repeats} (@option{-s}) options are both given, any
5715 character class can be used in @var{set2}. Otherwise, only the
5716 character classes @code{lower} and @code{upper} are accepted in
5717 @var{set2}, and then only if the corresponding character class
5718 (@code{upper} and @code{lower}, respectively) is specified in the same
5719 relative position in @var{set1}. Doing this specifies case conversion.
5720 The class names are given below; an error results when an invalid class
5732 Horizontal whitespace.
5741 Printable characters, not including space.
5747 Printable characters, including space.
5750 Punctuation characters.
5753 Horizontal or vertical whitespace.
5762 @item Equivalence classes
5763 @cindex equivalence classes
5765 The syntax @samp{[=@var{c}=]} expands to all of the characters that are
5766 equivalent to @var{c}, in no particular order. Equivalence classes are
5767 a relatively recent invention intended to support non-English alphabets.
5768 But there seems to be no standard way to define them or determine their
5769 contents. Therefore, they are not fully implemented in @sc{gnu} @command{tr};
5770 each character's equivalence class consists only of that character,
5771 which is of no particular use.
5777 @subsection Translating
5779 @cindex translating characters
5781 @command{tr} performs translation when @var{set1} and @var{set2} are
5782 both given and the @option{--delete} (@option{-d}) option is not given.
5783 @command{tr} translates each character of its input that is in @var{set1}
5784 to the corresponding character in @var{set2}. Characters not in
5785 @var{set1} are passed through unchanged. When a character appears more
5786 than once in @var{set1} and the corresponding characters in @var{set2}
5787 are not all the same, only the final one is used. For example, these
5788 two commands are equivalent:
5795 A common use of @command{tr} is to convert lowercase characters to
5796 uppercase. This can be done in many ways. Here are three of them:
5799 tr abcdefghijklmnopqrstuvwxyz ABCDEFGHIJKLMNOPQRSTUVWXYZ
5801 tr '[:lower:]' '[:upper:]'
5805 But note that using ranges like @code{a-z} above is not portable.
5807 When @command{tr} is performing translation, @var{set1} and @var{set2}
5808 typically have the same length. If @var{set1} is shorter than
5809 @var{set2}, the extra characters at the end of @var{set2} are ignored.
5811 On the other hand, making @var{set1} longer than @var{set2} is not
5812 portable; @acronym{POSIX} says that the result is undefined. In this situation,
5813 BSD @command{tr} pads @var{set2} to the length of @var{set1} by repeating
5814 the last character of @var{set2} as many times as necessary. System V
5815 @command{tr} truncates @var{set1} to the length of @var{set2}.
5817 By default, @sc{gnu} @command{tr} handles this case like BSD @command{tr}.
5818 When the @option{--truncate-set1} (@option{-t}) option is given,
5819 @sc{gnu} @command{tr} handles this case like the System V @command{tr}
5820 instead. This option is ignored for operations other than translation.
5822 Acting like System V @command{tr} in this case breaks the relatively common
5826 tr -cs A-Za-z0-9 '\012'
5830 because it converts only zero bytes (the first element in the
5831 complement of @var{set1}), rather than all non-alphanumerics, to
5835 By the way, the above idiom is not portable because it uses ranges, and
5836 it assumes that the octal code for newline is 012.
5837 Assuming a @acronym{POSIX} compliant @command{tr}, here is a better way to write it:
5840 tr -cs '[:alnum:]' '[\n*]'
5845 @subsection Squeezing repeats and deleting
5847 @cindex squeezing repeat characters
5848 @cindex deleting characters
5850 When given just the @option{--delete} (@option{-d}) option, @command{tr}
5851 removes any input characters that are in @var{set1}.
5853 When given just the @option{--squeeze-repeats} (@option{-s}) option,
5854 @command{tr} replaces each input sequence of a repeated character that
5855 is in @var{set1} with a single occurrence of that character.
5857 When given both @option{--delete} and @option{--squeeze-repeats}, @command{tr}
5858 first performs any deletions using @var{set1}, then squeezes repeats
5859 from any remaining characters using @var{set2}.
5861 The @option{--squeeze-repeats} option may also be used when translating,
5862 in which case @command{tr} first performs translation, then squeezes
5863 repeats from any remaining characters using @var{set2}.
5865 Here are some examples to illustrate various combinations of options:
5870 Remove all zero bytes:
5877 Put all words on lines by themselves. This converts all
5878 non-alphanumeric characters to newlines, then squeezes each string
5879 of repeated newlines into a single newline:
5882 tr -cs '[:alnum:]' '[\n*]'
5886 Convert each sequence of repeated newlines to a single newline:
5893 Find doubled occurrences of words in a document.
5894 @c Separate the following two "the"s, so typo checkers don't complain.
5895 For example, people often write ``the @w{}the'' with the repeated words
5896 separated by a newline. The Bourne shell script below works first
5897 by converting each sequence of punctuation and blank characters to a
5898 single newline. That puts each ``word'' on a line by itself.
5899 Next it maps all uppercase characters to lower case, and finally it
5900 runs @command{uniq} with the @option{-d} option to print out only the words
5906 | tr -s '[:punct:][:blank:]' '[\n*]' \
5907 | tr '[:upper:]' '[:lower:]' \
5912 Deleting a small set of characters is usually straightforward. For example,
5913 to remove all @samp{a}s, @samp{x}s, and @samp{M}s you would do this:
5919 However, when @samp{-} is one of those characters, it can be tricky because
5920 @samp{-} has special meanings. Performing the same task as above but also
5921 removing all @samp{-} characters, we might try @code{tr -d -axM}, but
5922 that would fail because @command{tr} would try to interpret @option{-a} as
5923 a command-line option. Alternatively, we could try putting the hyphen
5924 inside the string, @code{tr -d a-xM}, but that wouldn't work either because
5925 it would make @command{tr} interpret @code{a-x} as the range of characters
5926 @samp{a}@dots{}@samp{x} rather than the three.
5927 One way to solve the problem is to put the hyphen at the end of the list
5934 Or you can use @samp{--} to terminate option processing:
5940 More generally, use the character class notation @code{[=c=]}
5941 with @samp{-} (or any other character) in place of the @samp{c}:
5947 Note how single quotes are used in the above example to protect the
5948 square brackets from interpretation by a shell.
5953 @node expand invocation
5954 @section @command{expand}: Convert tabs to spaces
5957 @cindex tabs to spaces, converting
5958 @cindex converting tabs to spaces
5960 @command{expand} writes the contents of each given @var{file}, or standard
5961 input if none are given or for a @var{file} of @samp{-}, to standard
5962 output, with tab characters converted to the appropriate number of
5966 expand [@var{option}]@dots{} [@var{file}]@dots{}
5969 By default, @command{expand} converts all tabs to spaces. It preserves
5970 backspace characters in the output; they decrement the column count for
5971 tab calculations. The default action is equivalent to @option{-t 8} (set
5972 tabs every 8 columns).
5974 The program accepts the following options. Also see @ref{Common options}.
5978 @item -t @var{tab1}[,@var{tab2}]@dots{}
5979 @itemx --tabs=@var{tab1}[,@var{tab2}]@dots{}
5982 @cindex tab stops, setting
5983 If only one tab stop is given, set the tabs @var{tab1} spaces apart
5984 (default is 8). Otherwise, set the tabs at columns @var{tab1},
5985 @var{tab2}, @dots{} (numbered from 0), and replace any tabs beyond the
5986 last tab stop given with single spaces. Tab stops can be separated by
5987 blanks as well as by commas.
5989 For compatibility, GNU @command{expand} also accepts the obsolete
5990 option syntax, @option{-@var{t1}[,@var{t2}]@dots{}}. New scripts
5991 should use @option{-t @var{t1}[,@var{t2}]@dots{}} instead.
5997 @cindex initial tabs, converting
5998 Only convert initial tabs (those that precede all non-space or non-tab
5999 characters) on each line to spaces.
6006 @node unexpand invocation
6007 @section @command{unexpand}: Convert spaces to tabs
6011 @command{unexpand} writes the contents of each given @var{file}, or
6012 standard input if none are given or for a @var{file} of @samp{-}, to
6013 standard output, converting blanks at the beginning of each line into
6014 as many tab characters as needed. In the default @acronym{POSIX}
6015 locale, a @dfn{blank} is a space or a tab; other locales may specify
6016 additional blank characters. Synopsis:
6019 unexpand [@var{option}]@dots{} [@var{file}]@dots{}
6022 By default, @command{unexpand} converts only initial blanks (those
6023 that precede all non-blank characters) on each line. It
6024 preserves backspace characters in the output; they decrement the column
6025 count for tab calculations. By default, tabs are set at every 8th
6028 The program accepts the following options. Also see @ref{Common options}.
6032 @item -t @var{tab1}[,@var{tab2}]@dots{}
6033 @itemx --tabs=@var{tab1}[,@var{tab2}]@dots{}
6036 If only one tab stop is given, set the tabs @var{tab1} columns apart
6037 instead of the default 8. Otherwise, set the tabs at columns
6038 @var{tab1}, @var{tab2}, @dots{} (numbered from 0), and leave blanks
6039 beyond the tab stops given unchanged. Tab stops can be separated by
6040 blanks as well as by commas. This option implies the @option{-a} option.
6042 For compatibility, GNU @command{unexpand} supports the obsolete option syntax,
6043 @option{-@var{tab1}[,@var{tab2}]@dots{}}, where tab stops must be
6044 separated by commas. (Unlike @option{-t}, this obsolete option does
6045 not imply @option{-a}.) New scripts should use @option{--first-only -t
6046 @var{tab1}[,@var{tab2}]@dots{}} instead.
6052 Also convert all sequences of two or more blanks just before a tab stop,
6053 even if they occur after non-blank characters in a line.
6060 @node Directory listing
6061 @chapter Directory listing
6063 This chapter describes the @command{ls} command and its variants @command{dir}
6064 and @command{vdir}, which list information about files.
6067 * ls invocation:: List directory contents.
6068 * dir invocation:: Briefly ls.
6069 * vdir invocation:: Verbosely ls.
6070 * dircolors invocation:: Color setup for ls, etc.
6075 @section @command{ls}: List directory contents
6078 @cindex directory listing
6080 The @command{ls} program lists information about files (of any type,
6081 including directories). Options and file arguments can be intermixed
6082 arbitrarily, as usual.
6084 For non-option command-line arguments that are directories, by default
6085 @command{ls} lists the contents of directories, not recursively, and
6086 omitting files with names beginning with @samp{.}. For other non-option
6087 arguments, by default @command{ls} lists just the file name. If no
6088 non-option argument is specified, @command{ls} operates on the current
6089 directory, acting as if it had been invoked with a single argument of @samp{.}.
6092 By default, the output is sorted alphabetically, according to the locale
6093 settings in effect.@footnote{If you use a non-@acronym{POSIX}
6094 locale (e.g., by setting @env{LC_ALL} to @samp{en_US}), then @command{ls} may
6095 produce output that is sorted differently than you're accustomed to.
6096 In that case, set the @env{LC_ALL} environment variable to @samp{C}.}
6097 If standard output is
6098 a terminal, the output is in columns (sorted vertically) and control
6099 characters are output as question marks; otherwise, the output is listed
6100 one per line and control characters are output as-is.
6102 Because @command{ls} is such a fundamental program, it has accumulated many
6103 options over the years. They are described in the subsections below;
6104 within each section, options are listed alphabetically (ignoring case).
6105 The division of options into the subsections is not absolute, since some
6106 options affect more than one aspect of @command{ls}'s operation.
6108 @cindex exit status of @command{ls}
6113 1 minor problems (e.g., failure to access a file or directory not
6114 specified as a command line argument. This happens when listing a
6115 directory in which entries are actively being removed or renamed.)
6116 2 serious trouble (e.g., memory exhausted, invalid option or failure
6117 to access file or directory specified as a command line argument)
6120 Also see @ref{Common options}.
6123 * Which files are listed::
6124 * What information is listed::
6125 * Sorting the output::
6126 * More details about version sort::
6127 * General output formatting::
6128 * Formatting file timestamps::
6129 * Formatting the file names::
6133 @node Which files are listed
6134 @subsection Which files are listed
6136 These options determine which files @command{ls} lists information for.
6137 By default, @command{ls} lists files and the contents of any
6138 directories on the command line, except that in directories it ignores
6139 files whose names start with @samp{.}.
6147 In directories, do not ignore file names that start with @samp{.}.
6152 @opindex --almost-all
6153 In directories, do not ignore all file names that start with @samp{.};
6154 ignore only @file{.} and @file{..}. The @option{--all} (@option{-a})
6155 option overrides this option.
6158 @itemx --ignore-backups
6160 @opindex --ignore-backups
6161 @cindex backup files, ignoring
6162 In directories, ignore files that end with @samp{~}. This option is
6163 equivalent to @samp{--ignore='*~' --ignore='.*~'}.
6168 @opindex --directory
6169 List just the names of directories, as with other types of files, rather
6170 than listing their contents.
6171 @c The following sentence is the same as the one for -F.
6172 Do not follow symbolic links listed on the
6173 command line unless the @option{--dereference-command-line} (@option{-H}),
6174 @option{--dereference} (@option{-L}), or
6175 @option{--dereference-command-line-symlink-to-dir} options are specified.
6178 @itemx --dereference-command-line
6180 @opindex --dereference-command-line
6181 @cindex symbolic links, dereferencing
6182 If a command line argument specifies a symbolic link, show information
6183 for the file the link references rather than for the link itself.
6185 @itemx --dereference-command-line-symlink-to-dir
6186 @opindex --dereference-command-line-symlink-to-dir
6187 @cindex symbolic links, dereferencing
6188 Do not dereference symbolic links, with one exception:
6189 if a command line argument specifies a symbolic link that refers to
6190 a directory, show information for that directory rather than for the
6192 This is the default behavior when no other dereferencing-related
6193 option has been specified (@option{--classify} (@option{-F}),
6194 @option{--directory} (@option{-d}),
6196 @option{--dereference} (@option{-L}), or
6197 @option{--dereference-command-line} (@option{-H})).
6199 @item --group-directories-first
6200 @opindex --group-directories-first
6201 Group all the directories before the files and then sort the
6202 directories and the files separately using the selected sort key
6203 (see --sort option).
6204 That is, this option specifies a primary sort key,
6205 and the --sort option specifies a secondary key.
6206 However, any use of @option{--sort=none}
6207 (@option{-U}) disables this option altogether.
6209 @item --hide=PATTERN
6210 @opindex --hide=@var{pattern}
6211 In directories, ignore files whose names match the shell pattern
6212 @var{pattern}, unless the @option{--all} (@option{-a}) or
6213 @option{--almost-all} (@option{-A}) is also given. This
6214 option acts like @option{--ignore=@var{pattern}} except that it has no
6215 effect if @option{--all} (@option{-a}) or @option{--almost-all}
6216 (@option{-A}) is also given.
6218 This option can be useful in shell aliases. For example, if
6219 @command{lx} is an alias for @samp{ls --hide='*~'} and @command{ly} is
6220 an alias for @samp{ls --ignore='*~'}, then the command @samp{lx -A}
6221 lists the file @file{README~} even though @samp{ly -A} would not.
6223 @item -I @var{pattern}
6224 @itemx --ignore=@var{pattern}
6226 @opindex --ignore=@var{pattern}
6227 In directories, ignore files whose names match the shell pattern
6228 (not regular expression) @var{pattern}. As
6229 in the shell, an initial @samp{.} in a file name does not match a
6230 wildcard at the start of @var{pattern}. Sometimes it is useful
6231 to give this option several times. For example,
6234 $ ls --ignore='.??*' --ignore='.[^.]' --ignore='#*'
6237 The first option ignores names of length 3 or more that start with @samp{.},
6238 the second ignores all two-character names that start with @samp{.}
6239 except @samp{..}, and the third ignores names that start with @samp{#}.
6242 @itemx --dereference
6244 @opindex --dereference
6245 @cindex symbolic links, dereferencing
6246 When showing file information for a symbolic link, show information
6247 for the file the link references rather than the link itself.
6248 However, even with this option, @command{ls} still prints the name
6249 of the link itself, not the name of the file that the link points to.
6254 @opindex --recursive
6255 @cindex recursive directory listing
6256 @cindex directory listing, recursive
6257 List the contents of all directories recursively.
6262 @node What information is listed
6263 @subsection What information is listed
6265 These options affect the information that @command{ls} displays. By
6266 default, only file names are shown.
6272 @cindex hurd, author, printing
6273 List each file's author when producing long format directory listings.
6274 In GNU/Hurd, file authors can differ from their owners, but in other
6275 operating systems the two are the same.
6281 @cindex dired Emacs mode support
6282 With the long listing (@option{-l}) format, print an additional line after
6286 //DIRED// @var{beg1} @var{end1} @var{beg2} @var{end2} @dots{}
6290 The @var{begn} and @var{endn} are unsigned integers that record the
6291 byte position of the beginning and end of each file name in the output.
6292 This makes it easy for Emacs to find the names, even when they contain
6293 unusual characters such as space or newline, without fancy searching.
6295 If directories are being listed recursively (@option{-R}), output a similar
6296 line with offsets for each subdirectory name:
6299 //SUBDIRED// @var{beg1} @var{end1} @dots{}
6302 Finally, output a line of the form:
6305 //DIRED-OPTIONS// --quoting-style=@var{word}
6309 where @var{word} is the quoting style (@pxref{Formatting the file names}).
6311 Here is an actual example:
6314 $ mkdir -p a/sub/deeper a/sub2
6316 $ touch a/sub/deeper/file
6317 $ ls -gloRF --dired a
6320 -rw-r--r-- 1 0 Jun 10 12:27 f1
6321 -rw-r--r-- 1 0 Jun 10 12:27 f2
6322 drwxr-xr-x 3 4096 Jun 10 12:27 sub/
6323 drwxr-xr-x 2 4096 Jun 10 12:27 sub2/
6327 drwxr-xr-x 2 4096 Jun 10 12:27 deeper/
6331 -rw-r--r-- 1 0 Jun 10 12:27 file
6335 //DIRED// 48 50 84 86 120 123 158 162 217 223 282 286
6336 //SUBDIRED// 2 3 167 172 228 240 290 296
6337 //DIRED-OPTIONS// --quoting-style=literal
6340 Note that the pairs of offsets on the @samp{//DIRED//} line above delimit
6341 these names: @file{f1}, @file{f2}, @file{sub}, @file{sub2}, @file{deeper},
6343 The offsets on the @samp{//SUBDIRED//} line delimit the following
6344 directory names: @file{a}, @file{a/sub}, @file{a/sub/deeper}, @file{a/sub2}.
6346 Here is an example of how to extract the fifth entry name, @samp{deeper},
6347 corresponding to the pair of offsets, 222 and 228:
6350 $ ls -gloRF --dired a > out
6351 $ dd bs=1 skip=222 count=6 < out 2>/dev/null; echo
6355 Note that although the listing above includes a trailing slash
6356 for the @samp{deeper} entry, the offsets select the name without
6357 the trailing slash. However, if you invoke @command{ls} with @option{--dired}
6358 along with an option like @option{--escape} (aka @option{-b}) and operate
6359 on a file whose name contains special characters, notice that the backslash
6364 $ ls -blog --dired 'a b'
6365 -rw-r--r-- 1 0 Jun 10 12:28 a\ b
6367 //DIRED-OPTIONS// --quoting-style=escape
6370 If you use a quoting style that adds quote marks
6371 (e.g., @option{--quoting-style=c}), then the offsets include the quote marks.
6372 So beware that the user may select the quoting style via the environment
6373 variable @env{QUOTING_STYLE}. Hence, applications using @option{--dired}
6374 should either specify an explicit @option{--quoting-style=literal} option
6375 (aka @option{-N} or @option{--literal}) on the command line, or else be
6376 prepared to parse the escaped names.
6379 @opindex --full-time
6380 Produce long format directory listings, and list times in full. It is
6381 equivalent to using @option{--format=long} with
6382 @option{--time-style=full-iso} (@pxref{Formatting file timestamps}).
6386 Produce long format directory listings, but don't display owner information.
6392 Inhibit display of group information in a long format directory listing.
6393 (This is the default in some non-@sc{gnu} versions of @command{ls}, so we
6394 provide this option for compatibility.)
6402 @cindex inode number, printing
6403 Print the inode number (also called the file serial number and index
6404 number) of each file to the left of the file name. (This number
6405 uniquely identifies each file within a particular file system.)
6408 @itemx --format=long
6409 @itemx --format=verbose
6412 @opindex long ls @r{format}
6413 @opindex verbose ls @r{format}
6414 In addition to the name of each file, print the file type, file mode bits,
6415 number of hard links, owner name, group name, size, and
6416 timestamp (@pxref{Formatting file timestamps}), normally
6417 the modification time. Print question marks for information that
6418 cannot be determined.
6420 Normally the size is printed as a byte count without punctuation, but
6421 this can be overridden (@pxref{Block size}). For example, @option{-h}
6422 prints an abbreviated, human-readable count, and
6423 @samp{--block-size="'1"} prints a byte count with the thousands
6424 separator of the current locale.
6426 For each directory that is listed, preface the files with a line
6427 @samp{total @var{blocks}}, where @var{blocks} is the total disk allocation
6428 for all files in that directory. The block size currently defaults to 1024
6429 bytes, but this can be overridden (@pxref{Block size}).
6430 The @var{blocks} computed counts each hard link separately;
6431 this is arguably a deficiency.
6433 The file type is one of the following characters:
6435 @c The commented-out entries are ones we're not sure about.
6443 character special file
6445 high performance (``contiguous data'') file
6449 door (Solaris 2.5 and up)
6451 @c semaphore, if this is a distinct file type
6455 @c multiplexed file (7th edition Unix; obsolete)
6457 off-line (``migrated'') file (Cray DMF)
6459 network special file (HP-UX)
6463 port (Solaris 10 and up)
6465 @c message queue, if this is a distinct file type
6469 @c shared memory object, if this is a distinct file type
6471 @c typed memory object, if this is a distinct file type
6473 @c whiteout (4.4BSD; not implemented)
6475 some other file type
6478 @cindex permissions, output by @command{ls}
6479 The file mode bits listed are similar to symbolic mode specifications
6480 (@pxref{Symbolic Modes}). But @command{ls} combines multiple bits into the
6481 third character of each set of permissions as follows:
6485 If the set-user-ID or set-group-ID bit and the corresponding executable bit
6489 If the set-user-ID or set-group-ID bit is set but the corresponding
6490 executable bit is not set.
6493 If the restricted deletion flag or sticky bit, and the
6494 other-executable bit, are both set. The restricted deletion flag is
6495 another name for the sticky bit. @xref{Mode Structure}.
6498 If the restricted deletion flag or sticky bit is set but the
6499 other-executable bit is not set.
6502 If the executable bit is set and none of the above apply.
6508 Following the file mode bits is a single character that specifies
6509 whether an alternate access method such as an access control list
6510 applies to the file. When the character following the file mode bits is a
6511 space, there is no alternate access method. When it is a printing
6512 character, then there is such a method.
6514 GNU @command{ls} uses a @samp{.} character to indicate a file
6515 with an SELinux security context, but no other alternate access method.
6517 A file with any other combination of alternate access methods
6518 is marked with a @samp{+} character.
6521 @itemx --numeric-uid-gid
6523 @opindex --numeric-uid-gid
6524 @cindex numeric uid and gid
6525 @cindex numeric user and group IDs
6526 Produce long format directory listings, but
6527 display numeric user and group IDs instead of the owner and group names.
6531 Produce long format directory listings, but don't display group information.
6532 It is equivalent to using @option{--format=long} with @option{--no-group} .
6538 @cindex disk allocation
6539 @cindex size of files, reporting
6540 Print the disk allocation of each file to the left of the file name.
6541 This is the amount of disk space used by the file, which is usually a
6542 bit more than the file's size, but it can be less if the file has holes.
6544 Normally the disk allocation is printed in units of
6545 1024 bytes, but this can be overridden (@pxref{Block size}).
6547 @cindex NFS mounts from BSD to HP-UX
6548 For files that are NFS-mounted from an HP-UX system to a BSD system,
6549 this option reports sizes that are half the correct values. On HP-UX
6550 systems, it reports sizes that are twice the correct values for files
6551 that are NFS-mounted from BSD systems. This is due to a flaw in HP-UX;
6552 it also affects the HP-UX @command{ls} program.
6561 @cindex security context
6562 Display the SELinux security context or @samp{?} if none is found.
6563 When used with the @option{-l} option, print the security context
6564 to the left of the size column.
6569 @node Sorting the output
6570 @subsection Sorting the output
6572 @cindex sorting @command{ls} output
6573 These options change the order in which @command{ls} sorts the information
6574 it outputs. By default, sorting is done by character code
6575 (e.g., @acronym{ASCII} order).
6581 @itemx --time=status
6584 @opindex ctime@r{, printing or sorting by}
6585 @opindex status time@r{, printing or sorting by}
6586 @opindex use time@r{, printing or sorting files by}
6587 If the long listing format (e.g., @option{-l}, @option{-o}) is being used,
6588 print the status change time (the @samp{ctime} in the inode) instead of
6589 the modification time.
6590 When explicitly sorting by time (@option{--sort=time} or @option{-t})
6591 or when not using a long listing format,
6592 sort according to the status change time.
6596 @cindex unsorted directory listing
6597 @cindex directory order, listing by
6598 Primarily, like @option{-U}---do not sort; list the files in whatever
6599 order they are stored in the directory. But also enable @option{-a} (list
6600 all files) and disable @option{-l}, @option{--color}, and @option{-s} (if they
6601 were specified before the @option{-f}).
6607 @cindex reverse sorting
6608 Reverse whatever the sorting method is---e.g., list files in reverse
6609 alphabetical order, youngest first, smallest first, or whatever.
6615 @opindex size of files@r{, sorting files by}
6616 Sort by file size, largest first.
6622 @opindex modification time@r{, sorting files by}
6623 Sort by modification time (the @samp{mtime} in the inode), newest first.
6627 @itemx --time=access
6631 @opindex use time@r{, printing or sorting files by}
6632 @opindex atime@r{, printing or sorting files by}
6633 @opindex access time@r{, printing or sorting files by}
6634 If the long listing format (e.g., @option{--format=long}) is being used,
6635 print the last access time (the @samp{atime} in the inode).
6636 When explicitly sorting by time (@option{--sort=time} or @option{-t})
6637 or when not using a long listing format, sort according to the access time.
6643 @opindex none@r{, sorting option for @command{ls}}
6644 Do not sort; list the files in whatever order they are
6645 stored in the directory. (Do not do any of the other unrelated things
6646 that @option{-f} does.) This is especially useful when listing very large
6647 directories, since not doing any sorting can be noticeably faster.
6650 @itemx --sort=version
6653 @opindex version@r{, sorting option for @command{ls}}
6654 Sort by version name and number, lowest first. It behaves like a default
6655 sort, except that each sequence of decimal digits is treated numerically
6656 as an index/version number. (@xref{More details about version sort}.)
6659 @itemx --sort=extension
6662 @opindex extension@r{, sorting files by}
6663 Sort directory contents alphabetically by file extension (characters
6664 after the last @samp{.}); files with no extension are sorted first.
6669 @node More details about version sort
6670 @subsection More details about version sort
6672 The version sort takes into account the fact that file names frequently include
6673 indices or version numbers. Standard sorting functions usually do not produce
6674 the ordering that people expect because comparisons are made on a
6675 character-by-character basis. The version
6676 sort addresses this problem, and is especially useful when browsing
6677 directories that contain many files with indices/version numbers in their
6682 foo.zml-1.gz foo.zml-1.gz
6683 foo.zml-100.gz foo.zml-2.gz
6684 foo.zml-12.gz foo.zml-6.gz
6685 foo.zml-13.gz foo.zml-12.gz
6686 foo.zml-2.gz foo.zml-13.gz
6687 foo.zml-25.gz foo.zml-25.gz
6688 foo.zml-6.gz foo.zml-100.gz
6691 Version-sorted strings are compared such that if @var{ver1} and @var{ver2}
6692 are version numbers and @var{prefix} and @var{suffix} (@var{suffix} matching
6693 the regular expression @samp{(\.[A-Za-z~][A-Za-z0-9~]*)*}) are strings then
6694 @var{ver1} < @var{ver2} implies that the name composed of
6695 ``@var{prefix} @var{ver1} @var{suffix}'' sorts before
6696 ``@var{prefix} @var{ver2} @var{suffix}''.
6698 Note also that leading zeros of numeric parts are ignored:
6702 abc-1.007.tgz abc-1.01a.tgz
6703 abc-1.012b.tgz abc-1.007.tgz
6704 abc-1.01a.tgz abc-1.012b.tgz
6707 This functionality is implemented using gnulib's @code{filevercmp} function.
6708 One result of that implementation decision is that @samp{ls -v}
6709 and @samp{sort -V} do not use the locale category, @env{LC_COLLATE},
6710 which means non-numeric prefixes are sorted as if @env{LC_COLLATE} were set
6713 @node General output formatting
6714 @subsection General output formatting
6716 These options affect the appearance of the overall output.
6721 @itemx --format=single-column
6724 @opindex single-column @r{output of files}
6725 List one file per line. This is the default for @command{ls} when standard
6726 output is not a terminal.
6729 @itemx --format=vertical
6732 @opindex vertical @r{sorted files in columns}
6733 List files in columns, sorted vertically. This is the default for
6734 @command{ls} if standard output is a terminal. It is always the default
6735 for the @command{dir} program.
6736 @sc{gnu} @command{ls} uses variable width columns to display as many files as
6737 possible in the fewest lines.
6739 @item --color [=@var{when}]
6741 @cindex color, distinguishing file types with
6742 Specify whether to use color for distinguishing file types. @var{when}
6743 may be omitted, or one of:
6746 @vindex none @r{color option}
6747 - Do not use color at all. This is the default.
6749 @vindex auto @r{color option}
6750 @cindex terminal, using color iff
6751 - Only use color if standard output is a terminal.
6753 @vindex always @r{color option}
6756 Specifying @option{--color} and no @var{when} is equivalent to
6757 @option{--color=always}.
6758 Piping a colorized listing through a pager like @command{more} or
6759 @command{less} usually produces unreadable results. However, using
6760 @code{more -f} does seem to work.
6764 @itemx --indicator-style=classify
6767 @opindex --indicator-style
6768 @cindex file type and executables, marking
6769 @cindex executables and file type, marking
6770 Append a character to each file name indicating the file type. Also,
6771 for regular files that are executable, append @samp{*}. The file type
6772 indicators are @samp{/} for directories, @samp{@@} for symbolic links,
6773 @samp{|} for FIFOs, @samp{=} for sockets, @samp{>} for doors,
6774 and nothing for regular files.
6775 @c The following sentence is the same as the one for -d.
6776 Do not follow symbolic links listed on the
6777 command line unless the @option{--dereference-command-line} (@option{-H}),
6778 @option{--dereference} (@option{-L}), or
6779 @option{--dereference-command-line-symlink-to-dir} options are specified.
6782 @itemx --indicator-style=file-type
6783 @opindex --file-type
6784 @opindex --indicator-style
6785 @cindex file type, marking
6786 Append a character to each file name indicating the file type. This is
6787 like @option{-F}, except that executables are not marked.
6789 @item --indicator-style=@var{word}
6790 @opindex --indicator-style
6791 Append a character indicator with style @var{word} to entry names,
6796 Do not append any character indicator; this is the default.
6798 Append @samp{/} for directories. This is the same as the @option{-p}
6801 Append @samp{/} for directories, @samp{@@} for symbolic links, @samp{|}
6802 for FIFOs, @samp{=} for sockets, and nothing for regular files. This is
6803 the same as the @option{--file-type} option.
6805 Append @samp{*} for executable regular files, otherwise behave as for
6806 @samp{file-type}. This is the same as the @option{-F} or
6807 @option{--classify} option.
6812 Print file sizes in 1024-byte blocks, overriding the default block
6813 size (@pxref{Block size}).
6814 This option is equivalent to @option{--block-size=1K}.
6817 @itemx --format=commas
6820 @opindex commas@r{, outputting between files}
6821 List files horizontally, with as many as will fit on each line,
6822 separated by @samp{, } (a comma and a space).
6825 @itemx --indicator-style=slash
6827 @opindex --indicator-style
6828 @cindex file type, marking
6829 Append a @samp{/} to directory names.
6832 @itemx --format=across
6833 @itemx --format=horizontal
6836 @opindex across@r{, listing files}
6837 @opindex horizontal@r{, listing files}
6838 List the files in columns, sorted horizontally.
6841 @itemx --tabsize=@var{cols}
6844 Assume that each tab stop is @var{cols} columns wide. The default is 8.
6845 @command{ls} uses tabs where possible in the output, for efficiency. If
6846 @var{cols} is zero, do not use tabs at all.
6848 @c FIXME: remove in 2009, if Apple Terminal has been fixed for long enough.
6849 Some terminal emulators (at least Apple Terminal 1.5 (133) from Mac OS X 10.4.8)
6850 do not properly align columns to the right of a TAB following a
6851 non-@acronym{ASCII} byte. If you use such a terminal emulator, use the
6852 @option{-T0} option or put @code{TABSIZE=0} in your environment to tell
6853 @command{ls} to align using spaces, not tabs.
6856 @itemx --width=@var{cols}
6860 Assume the screen is @var{cols} columns wide. The default is taken
6861 from the terminal settings if possible; otherwise the environment
6862 variable @env{COLUMNS} is used if it is set; otherwise the default
6868 @node Formatting file timestamps
6869 @subsection Formatting file timestamps
6871 By default, file timestamps are listed in abbreviated form. Most
6872 locales use a timestamp like @samp{2002-03-30 23:45}. However, the
6873 default @acronym{POSIX} locale uses a date like @samp{Mar 30@ @ 2002}
6874 for non-recent timestamps, and a date-without-year and time like
6875 @samp{Mar 30 23:45} for recent timestamps.
6877 A timestamp is considered to be @dfn{recent} if it is less than six
6878 months old, and is not dated in the future. If a timestamp dated
6879 today is not listed in recent form, the timestamp is in the future,
6880 which means you probably have clock skew problems which may break
6881 programs like @command{make} that rely on file timestamps.
6884 Time stamps are listed according to the time zone rules specified by
6885 the @env{TZ} environment variable, or by the system default rules if
6886 @env{TZ} is not set. @xref{TZ Variable,, Specifying the Time Zone
6887 with @env{TZ}, libc, The GNU C Library Reference Manual}.
6889 The following option changes how file timestamps are printed.
6892 @item --time-style=@var{style}
6893 @opindex --time-style
6895 List timestamps in style @var{style}. The @var{style} should
6896 be one of the following:
6901 List timestamps using @var{format}, where @var{format} is interpreted
6902 like the format argument of @command{date} (@pxref{date invocation}).
6903 For example, @option{--time-style="+%Y-%m-%d %H:%M:%S"} causes
6904 @command{ls} to list timestamps like @samp{2002-03-30 23:45:56}. As
6905 with @command{date}, @var{format}'s interpretation is affected by the
6906 @env{LC_TIME} locale category.
6908 If @var{format} contains two format strings separated by a newline,
6909 the former is used for non-recent files and the latter for recent
6910 files; if you want output columns to line up, you may need to insert
6911 spaces in one of the two formats.
6914 List timestamps in full using @acronym{ISO} 8601 date, time, and time zone
6915 format with nanosecond precision, e.g., @samp{2002-03-30
6916 23:45:56.477817180 -0700}. This style is equivalent to
6917 @samp{+%Y-%m-%d %H:%M:%S.%N %z}.
6919 This is useful because the time output includes all the information that
6920 is available from the operating system. For example, this can help
6921 explain @command{make}'s behavior, since @acronym{GNU} @command{make}
6922 uses the full timestamp to determine whether a file is out of date.
6925 List @acronym{ISO} 8601 date and time in minutes, e.g.,
6926 @samp{2002-03-30 23:45}. These timestamps are shorter than
6927 @samp{full-iso} timestamps, and are usually good enough for everyday
6928 work. This style is equivalent to @samp{+%Y-%m-%d %H:%M}.
6931 List @acronym{ISO} 8601 dates for non-recent timestamps (e.g.,
6932 @samp{2002-03-30@ }), and @acronym{ISO} 8601 month, day, hour, and
6933 minute for recent timestamps (e.g., @samp{03-30 23:45}). These
6934 timestamps are uglier than @samp{long-iso} timestamps, but they carry
6935 nearly the same information in a smaller space and their brevity helps
6936 @command{ls} output fit within traditional 80-column output lines.
6937 The following two @command{ls} invocations are equivalent:
6942 ls -l --time-style="+%Y-%m-%d $newline%m-%d %H:%M"
6943 ls -l --time-style="iso"
6948 List timestamps in a locale-dependent form. For example, a Finnish
6949 locale might list non-recent timestamps like @samp{maalis 30@ @ 2002}
6950 and recent timestamps like @samp{maalis 30 23:45}. Locale-dependent
6951 timestamps typically consume more space than @samp{iso} timestamps and
6952 are harder for programs to parse because locale conventions vary so
6953 widely, but they are easier for many people to read.
6955 The @env{LC_TIME} locale category specifies the timestamp format. The
6956 default @acronym{POSIX} locale uses timestamps like @samp{Mar 30@
6957 @ 2002} and @samp{Mar 30 23:45}; in this locale, the following two
6958 @command{ls} invocations are equivalent:
6963 ls -l --time-style="+%b %e %Y$newline%b %e %H:%M"
6964 ls -l --time-style="locale"
6967 Other locales behave differently. For example, in a German locale,
6968 @option{--time-style="locale"} might be equivalent to
6969 @option{--time-style="+%e. %b %Y $newline%e. %b %H:%M"}
6970 and might generate timestamps like @samp{30. M@"ar 2002@ } and
6971 @samp{30. M@"ar 23:45}.
6973 @item posix-@var{style}
6975 List @acronym{POSIX}-locale timestamps if the @env{LC_TIME} locale
6976 category is @acronym{POSIX}, @var{style} timestamps otherwise. For
6977 example, the @samp{posix-long-iso} style lists
6978 timestamps like @samp{Mar 30@ @ 2002} and @samp{Mar 30 23:45} when in
6979 the @acronym{POSIX} locale, and like @samp{2002-03-30 23:45} otherwise.
6984 You can specify the default value of the @option{--time-style} option
6985 with the environment variable @env{TIME_STYLE}; if @env{TIME_STYLE} is not set
6986 the default style is @samp{locale}. @acronym{GNU} Emacs 21.3 and
6987 later use the @option{--dired} option and therefore can parse any date
6988 format, but if you are using Emacs 21.1 or 21.2 and specify a
6989 non-@acronym{POSIX} locale you may need to set
6990 @samp{TIME_STYLE="posix-long-iso"}.
6992 To avoid certain denial-of-service attacks, timestamps that would be
6993 longer than 1000 bytes may be treated as errors.
6996 @node Formatting the file names
6997 @subsection Formatting the file names
6999 These options change how file names themselves are printed.
7005 @itemx --quoting-style=escape
7008 @opindex --quoting-style
7009 @cindex backslash sequences for file names
7010 Quote nongraphic characters in file names using alphabetic and octal
7011 backslash sequences like those used in C.
7015 @itemx --quoting-style=literal
7018 @opindex --quoting-style
7019 Do not quote file names. However, with @command{ls} nongraphic
7020 characters are still printed as question marks if the output is a
7021 terminal and you do not specify the @option{--show-control-chars}
7025 @itemx --hide-control-chars
7027 @opindex --hide-control-chars
7028 Print question marks instead of nongraphic characters in file names.
7029 This is the default if the output is a terminal and the program is
7034 @itemx --quoting-style=c
7036 @opindex --quote-name
7037 @opindex --quoting-style
7038 Enclose file names in double quotes and quote nongraphic characters as
7041 @item --quoting-style=@var{word}
7042 @opindex --quoting-style
7043 @cindex quoting style
7044 Use style @var{word} to quote file names and other strings that may
7045 contain arbitrary characters. The @var{word} should
7046 be one of the following:
7050 Output strings as-is; this is the same as the @option{-N} or
7051 @option{--literal} option.
7053 Quote strings for the shell if they contain shell metacharacters or would
7054 cause ambiguous output.
7055 The quoting is suitable for @acronym{POSIX}-compatible shells like
7056 @command{bash}, but it does not always work for incompatible shells
7059 Quote strings for the shell, even if they would normally not require quoting.
7061 Quote strings as for C character string literals, including the
7062 surrounding double-quote characters; this is the same as the
7063 @option{-Q} or @option{--quote-name} option.
7065 Quote strings as for C character string literals, except omit the
7066 surrounding double-quote
7067 characters; this is the same as the @option{-b} or @option{--escape} option.
7069 Quote strings as for C character string literals, except use
7070 surrounding quotation marks appropriate for the
7073 @c Use @t instead of @samp to avoid duplicate quoting in some output styles.
7074 Quote strings as for C character string literals, except use
7075 surrounding quotation marks appropriate for the locale, and quote
7076 @t{`like this'} instead of @t{"like
7077 this"} in the default C locale. This looks nicer on many displays.
7080 You can specify the default value of the @option{--quoting-style} option
7081 with the environment variable @env{QUOTING_STYLE}. If that environment
7082 variable is not set, the default value is @samp{literal}, but this
7083 default may change to @samp{shell} in a future version of this package.
7085 @item --show-control-chars
7086 @opindex --show-control-chars
7087 Print nongraphic characters as-is in file names.
7088 This is the default unless the output is a terminal and the program is
7094 @node dir invocation
7095 @section @command{dir}: Briefly list directory contents
7098 @cindex directory listing, brief
7100 @command{dir} is equivalent to @code{ls -C
7101 -b}; that is, by default files are listed in columns, sorted vertically,
7102 and special characters are represented by backslash escape sequences.
7104 @xref{ls invocation, @command{ls}}.
7107 @node vdir invocation
7108 @section @command{vdir}: Verbosely list directory contents
7111 @cindex directory listing, verbose
7113 @command{vdir} is equivalent to @code{ls -l
7114 -b}; that is, by default files are listed in long format and special
7115 characters are represented by backslash escape sequences.
7117 @node dircolors invocation
7118 @section @command{dircolors}: Color setup for @command{ls}
7122 @cindex setup for color
7124 @command{dircolors} outputs a sequence of shell commands to set up the
7125 terminal for color output from @command{ls} (and @command{dir}, etc.).
7129 eval "`dircolors [@var{option}]@dots{} [@var{file}]`"
7132 If @var{file} is specified, @command{dircolors} reads it to determine which
7133 colors to use for which file types and extensions. Otherwise, a
7134 precompiled database is used. For details on the format of these files,
7135 run @samp{dircolors --print-database}.
7137 To make @command{dircolors} read a @file{~/.dircolors} file if it
7138 exists, you can put the following lines in your @file{~/.bashrc} (or
7139 adapt them to your favorite shell):
7143 test -r $d && eval "$(dircolors $d)"
7147 @vindex SHELL @r{environment variable, and color}
7148 The output is a shell command to set the @env{LS_COLORS} environment
7149 variable. You can specify the shell syntax to use on the command line,
7150 or @command{dircolors} will guess it from the value of the @env{SHELL}
7151 environment variable.
7153 The program accepts the following options. Also see @ref{Common options}.
7158 @itemx --bourne-shell
7161 @opindex --bourne-shell
7162 @cindex Bourne shell syntax for color setup
7163 @cindex @command{sh} syntax for color setup
7164 Output Bourne shell commands. This is the default if the @env{SHELL}
7165 environment variable is set and does not end with @samp{csh} or
7174 @cindex C shell syntax for color setup
7175 @cindex @command{csh} syntax for color setup
7176 Output C shell commands. This is the default if @code{SHELL} ends with
7177 @command{csh} or @command{tcsh}.
7180 @itemx --print-database
7182 @opindex --print-database
7183 @cindex color database, printing
7184 @cindex database for color setup, printing
7185 @cindex printing color database
7186 Print the (compiled-in) default color configuration database. This
7187 output is itself a valid configuration file, and is fairly descriptive
7188 of the possibilities.
7195 @node Basic operations
7196 @chapter Basic operations
7198 @cindex manipulating files
7200 This chapter describes the commands for basic file manipulation:
7201 copying, moving (renaming), and deleting (removing).
7204 * cp invocation:: Copy files.
7205 * dd invocation:: Convert and copy a file.
7206 * install invocation:: Copy files and set attributes.
7207 * mv invocation:: Move (rename) files.
7208 * rm invocation:: Remove files or directories.
7209 * shred invocation:: Remove files more securely.
7214 @section @command{cp}: Copy files and directories
7217 @cindex copying files and directories
7218 @cindex files, copying
7219 @cindex directories, copying
7221 @command{cp} copies files (or, optionally, directories). The copy is
7222 completely independent of the original. You can either copy one file to
7223 another, or copy arbitrarily many files to a destination directory.
7227 cp [@var{option}]@dots{} [-T] @var{source} @var{dest}
7228 cp [@var{option}]@dots{} @var{source}@dots{} @var{directory}
7229 cp [@var{option}]@dots{} -t @var{directory} @var{source}@dots{}
7234 If two file names are given, @command{cp} copies the first file to the
7238 If the @option{--target-directory} (@option{-t}) option is given, or
7239 failing that if the last file is a directory and the
7240 @option{--no-target-directory} (@option{-T}) option is not given,
7241 @command{cp} copies each @var{source} file to the specified directory,
7242 using the @var{source}s' names.
7245 Generally, files are written just as they are read. For exceptions,
7246 see the @option{--sparse} option below.
7248 By default, @command{cp} does not copy directories. However, the
7249 @option{-R}, @option{-a}, and @option{-r} options cause @command{cp} to
7250 copy recursively by descending into source directories and copying files
7251 to corresponding destination directories.
7253 When copying from a symbolic link, @command{cp} normally follows the
7254 link only when not copying
7255 recursively. This default can be overridden with the
7256 @option{--archive} (@option{-a}), @option{-d}, @option{--dereference}
7257 (@option{-L}), @option{--no-dereference} (@option{-P}), and
7258 @option{-H} options. If more than one of these options is specified,
7259 the last one silently overrides the others.
7261 When copying to a symbolic link, @command{cp} follows the
7262 link only when it refers to an existing regular file.
7263 However, when copying to a dangling symbolic link, @command{cp}
7264 refuses by default, and fails with a diagnostic, since the operation
7265 is inherently dangerous. This behavior is contrary to historical
7266 practice and to @acronym{POSIX}.
7267 Set @env{POSIXLY_CORRECT} to make @command{cp} attempt to create
7268 the target of a dangling destination symlink, in spite of the possible risk.
7269 Also, when an option like
7270 @option{--backup} or @option{--link} acts to rename or remove the
7271 destination before copying, @command{cp} renames or removes the
7272 symbolic link rather than the file it points to.
7274 By default, @command{cp} copies the contents of special files only
7275 when not copying recursively. This default can be overridden with the
7276 @option{--copy-contents} option.
7278 @cindex self-backups
7279 @cindex backups, making only
7280 @command{cp} generally refuses to copy a file onto itself, with the
7281 following exception: if @option{--force --backup} is specified with
7282 @var{source} and @var{dest} identical, and referring to a regular file,
7283 @command{cp} will make a backup file, either regular or numbered, as
7284 specified in the usual ways (@pxref{Backup options}). This is useful when
7285 you simply want to make a backup of an existing file before changing it.
7287 The program accepts the following options. Also see @ref{Common options}.
7294 Preserve as much as possible of the structure and attributes of the
7295 original files in the copy (but do not attempt to preserve internal
7296 directory structure; i.e., @samp{ls -U} may list the entries in a copied
7297 directory in a different order).
7298 Try to preserve SELinux security context and extended attributes (xattr),
7299 but ignore any failure to do that and print no corresponding diagnostic.
7300 Equivalent to @option{-dR --preserve=all} with the reduced diagnostics.
7303 @itemx @w{@kbd{--backup}[=@var{method}]}
7306 @vindex VERSION_CONTROL
7307 @cindex backups, making
7308 @xref{Backup options}.
7309 Make a backup of each file that would otherwise be overwritten or removed.
7310 As a special case, @command{cp} makes a backup of @var{source} when the force
7311 and backup options are given and @var{source} and @var{dest} are the same
7312 name for an existing, regular file. One useful application of this
7313 combination of options is this tiny Bourne shell script:
7317 # Usage: backup FILE...
7318 # Create a @sc{gnu}-style backup of each listed FILE.
7320 cp --backup --force -- "$i" "$i"
7324 @item --copy-contents
7325 @cindex directories, copying recursively
7326 @cindex copying directories recursively
7327 @cindex recursively copying directories
7328 @cindex non-directories, copying as special files
7329 If copying recursively, copy the contents of any special files (e.g.,
7330 FIFOs and device files) as if they were regular files. This means
7331 trying to read the data in each source file and writing it to the
7332 destination. It is usually a mistake to use this option, as it
7333 normally has undesirable effects on special files like FIFOs and the
7334 ones typically found in the @file{/dev} directory. In most cases,
7335 @code{cp -R --copy-contents} will hang indefinitely trying to read
7336 from FIFOs and special files like @file{/dev/console}, and it will
7337 fill up your destination disk if you use it to copy @file{/dev/zero}.
7338 This option has no effect unless copying recursively, and it does not
7339 affect the copying of symbolic links.
7343 @cindex symbolic links, copying
7344 @cindex hard links, preserving
7345 Copy symbolic links as symbolic links rather than copying the files that
7346 they point to, and preserve hard links between source files in the copies.
7347 Equivalent to @option{--no-dereference --preserve=links}.
7353 When copying without this option and an existing destination file cannot
7354 be opened for writing, the copy fails. However, with @option{--force}),
7355 when a destination file cannot be opened, @command{cp} then removes it and
7356 tries to open it again. Contrast this behavior with that enabled by
7357 @option{--link} and @option{--symbolic-link}, whereby the destination file
7358 is never opened but rather is removed unconditionally. Also see the
7359 description of @option{--remove-destination}.
7361 This option is independent of the @option{--interactive} or
7362 @option{-i} option: neither cancels the effect of the other.
7364 This option is redundant if the @option{--no-clobber} or @option{-n} option is
7369 If a command line argument specifies a symbolic link, then copy the
7370 file it points to rather than the symbolic link itself. However,
7371 copy (preserving its nature) any symbolic link that is encountered
7372 via recursive traversal.
7375 @itemx --interactive
7377 @opindex --interactive
7378 When copying a file other than a directory, prompt whether to
7379 overwrite an existing destination file. The @option{-i} option overrides
7380 a previous @option{-n} option.
7386 Make hard links instead of copies of non-directories.
7389 @itemx --dereference
7391 @opindex --dereference
7392 Follow symbolic links when copying from them.
7393 With this option, @command{cp} cannot create a symbolic link.
7394 For example, a symlink (to regular file) in the source tree will be copied to
7395 a regular file in the destination tree.
7400 @opindex --no-clobber
7401 Do not overwrite an existing file. The @option{-n} option overrides a previous
7402 @option{-i} option. This option is mutually exclusive with @option{-b} or
7403 @option{--backup} option.
7406 @itemx --no-dereference
7408 @opindex --no-dereference
7409 @cindex symbolic links, copying
7410 Copy symbolic links as symbolic links rather than copying the files that
7411 they point to. This option affects only symbolic links in the source;
7412 symbolic links in the destination are always followed if possible.
7415 @itemx @w{@kbd{--preserve}[=@var{attribute_list}]}
7418 @cindex file information, preserving, extended attributes, xattr
7419 Preserve the specified attributes of the original files.
7420 If specified, the @var{attribute_list} must be a comma-separated list
7421 of one or more of the following strings:
7425 Preserve the file mode bits and access control lists.
7427 Preserve the owner and group. On most modern systems,
7428 only users with appropriate privileges may change the owner of a file,
7430 may preserve the group ownership of a file only if they happen to be
7431 a member of the desired group.
7433 Preserve the times of last access and last modification, when possible.
7434 On older systems, it is not possible to preserve these attributes
7435 when the affected file is a symbolic link.
7436 However, many systems now provide the @code{utimensat} function,
7437 which makes it possible even for symbolic links.
7439 Preserve in the destination files
7440 any links between corresponding source files.
7441 Note that with @option{-L} or @option{-H}, this option can convert
7442 symbolic links to hard links. For example,
7444 $ mkdir c; : > a; ln -s a b; cp -aH a b c; ls -i1 c
7449 Note the inputs: @file{b} is a symlink to regular file @file{a},
7450 yet the files in destination directory, @file{c/}, are hard-linked.
7451 Since @option{-a} implies @option{--preserve=links}, and since @option{-H}
7452 tells @command{cp} to dereference command line arguments, it sees two files
7453 with the same inode number, and preserves the perceived hard link.
7455 Here is a similar example that exercises @command{cp}'s @option{-L} option:
7457 $ mkdir b c; (cd b; : > a; ln -s a b); cp -aL b c; ls -i1 c/b
7463 Preserve SELinux security context of the file. @command{cp} will fail
7464 if the preserving of SELinux security context is not succesful.
7466 Preserve extended attributes if @command{cp} is built with xattr support,
7467 and xattrs are supported and enabled on your file system.
7468 If SELinux context and/or ACLs are implemented using xattrs,
7469 they are preserved by this option as well.
7471 Preserve all file attributes.
7472 Equivalent to specifying all of the above, but with the difference
7473 that failure to preserve SELinux security context or extended attributes
7474 does not change @command{cp}'s exit status.
7475 @command{cp} does diagnose such failures.
7478 Using @option{--preserve} with no @var{attribute_list} is equivalent
7479 to @option{--preserve=mode,ownership,timestamps}.
7481 In the absence of this option, each destination file is created with the
7482 mode bits of the corresponding source file, minus the bits set in the
7483 umask and minus the set-user-ID and set-group-ID bits.
7484 @xref{File permissions}.
7486 @itemx @w{@kbd{--no-preserve}=@var{attribute_list}}
7487 @cindex file information, preserving
7488 Do not preserve the specified attributes. The @var{attribute_list}
7489 has the same form as for @option{--preserve}.
7493 @cindex parent directories and @command{cp}
7494 Form the name of each destination file by appending to the target
7495 directory a slash and the specified name of the source file. The last
7496 argument given to @command{cp} must be the name of an existing directory.
7497 For example, the command:
7500 cp --parents a/b/c existing_dir
7504 copies the file @file{a/b/c} to @file{existing_dir/a/b/c}, creating
7505 any missing intermediate directories.
7512 @opindex --recursive
7513 @cindex directories, copying recursively
7514 @cindex copying directories recursively
7515 @cindex recursively copying directories
7516 @cindex non-directories, copying as special files
7517 Copy directories recursively. By default, do not follow symbolic
7518 links in the source; see the @option{--archive} (@option{-a}), @option{-d},
7519 @option{--dereference} (@option{-L}), @option{--no-dereference}
7520 (@option{-P}), and @option{-H} options. Special files are copied by
7521 creating a destination file of the same type as the source; see the
7522 @option{--copy-contents} option. It is not portable to use
7523 @option{-r} to copy symbolic links or special files. On some
7524 non-@sc{gnu} systems, @option{-r} implies the equivalent of
7525 @option{-L} and @option{--copy-contents} for historical reasons.
7526 Also, it is not portable to use @option{-R} to copy symbolic links
7527 unless you also specify @option{-P}, as @acronym{POSIX} allows
7528 implementations that dereference symbolic links by default.
7530 @item --reflink[=@var{when}]
7531 @opindex --reflink[=@var{when}]
7534 @cindex copy on write
7535 Perform a lightweight, copy-on-write (COW) copy.
7536 Copying with this option can succeed only on some file systems.
7537 Once it has succeeded, beware that the source and destination files
7538 share the same disk data blocks as long as they remain unmodified.
7539 Thus, if a disk I/O error affects data blocks of one of the files,
7540 the other suffers the exact same fate.
7542 The @var{when} value can be one of the following:
7546 The default behavior: if the copy-on-write operation is not supported
7547 then report the failure for each file and exit with a failure status.
7550 If the copy-on-write operation is not supported then fall back
7551 to the standard copy behaviour.
7555 @item --remove-destination
7556 @opindex --remove-destination
7557 Remove each existing destination file before attempting to open it
7558 (contrast with @option{-f} above).
7560 @item --sparse=@var{when}
7561 @opindex --sparse=@var{when}
7562 @cindex sparse files, copying
7563 @cindex holes, copying files with
7564 @findex read @r{system call, and holes}
7565 A @dfn{sparse file} contains @dfn{holes}---a sequence of zero bytes that
7566 does not occupy any physical disk blocks; the @samp{read} system call
7567 reads these as zeros. This can both save considerable disk space and
7568 increase speed, since many binary files contain lots of consecutive zero
7569 bytes. By default, @command{cp} detects holes in input source files via a crude
7570 heuristic and makes the corresponding output file sparse as well.
7571 Only regular files may be sparse.
7573 The @var{when} value can be one of the following:
7577 The default behavior: if the input file is sparse, attempt to make
7578 the output file sparse, too. However, if an output file exists but
7579 refers to a non-regular file, then do not attempt to make it sparse.
7582 For each sufficiently long sequence of zero bytes in the input file,
7583 attempt to create a corresponding hole in the output file, even if the
7584 input file does not appear to be sparse.
7585 This is useful when the input file resides on a file system
7586 that does not support sparse files
7587 (for example, @samp{efs} file systems in SGI IRIX 5.3 and earlier),
7588 but the output file is on a type of file system that does support them.
7589 Holes may be created only in regular files, so if the destination file
7590 is of some other type, @command{cp} does not even try to make it sparse.
7593 Never make the output file sparse.
7594 This is useful in creating a file for use with the @command{mkswap} command,
7595 since such a file must not have any holes.
7598 @optStripTrailingSlashes
7601 @itemx --symbolic-link
7603 @opindex --symbolic-link
7604 @cindex symbolic links, copying with
7605 Make symbolic links instead of copies of non-directories. All source
7606 file names must be absolute (starting with @samp{/}) unless the
7607 destination files are in the current directory. This option merely
7608 results in an error message on systems that do not support symbolic links.
7614 @optNoTargetDirectory
7620 @cindex newer files, copying only
7621 Do not copy a non-directory that has an existing destination with the
7622 same or newer modification time. If time stamps are being preserved,
7623 the comparison is to the source time stamp truncated to the
7624 resolutions of the destination file system and of the system calls
7625 used to update time stamps; this avoids duplicate work if several
7626 @samp{cp -pu} commands are executed with the same source and
7633 Print the name of each file before copying it.
7636 @itemx --one-file-system
7638 @opindex --one-file-system
7639 @cindex file systems, omitting copying to different
7640 Skip subdirectories that are on different file systems from the one that
7641 the copy started on.
7642 However, mount point directories @emph{are} copied.
7650 @section @command{dd}: Convert and copy a file
7653 @cindex converting while copying a file
7655 @command{dd} copies a file (from standard input to standard output, by
7656 default) with a changeable I/O block size, while optionally performing
7657 conversions on it. Synopses:
7660 dd [@var{operand}]@dots{}
7664 The only options are @option{--help} and @option{--version}.
7665 @xref{Common options}. @command{dd} accepts the following operands.
7671 Read from @var{file} instead of standard input.
7675 Write to @var{file} instead of standard output. Unless
7676 @samp{conv=notrunc} is given, @command{dd} truncates @var{file} to zero
7677 bytes (or the size specified with @samp{seek=}).
7679 @item ibs=@var{bytes}
7681 @cindex block size of input
7682 @cindex input block size
7683 Set the input block size to @var{bytes}.
7684 This makes @command{dd} read @var{bytes} per block.
7685 The default is 512 bytes.
7687 @item obs=@var{bytes}
7689 @cindex block size of output
7690 @cindex output block size
7691 Set the output block size to @var{bytes}.
7692 This makes @command{dd} write @var{bytes} per block.
7693 The default is 512 bytes.
7695 @item bs=@var{bytes}
7698 Set both input and output block sizes to @var{bytes}.
7699 This makes @command{dd} read and write @var{bytes} per block,
7700 overriding any @samp{ibs} and @samp{obs} settings.
7701 In addition, if no data-transforming @option{conv} option is specified,
7702 each input block is copied to the output as a single block,
7703 without aggregating short reads.
7705 @item cbs=@var{bytes}
7707 @cindex block size of conversion
7708 @cindex conversion block size
7709 @cindex fixed-length records, converting to variable-length
7710 @cindex variable-length records, converting to fixed-length
7711 Set the conversion block size to @var{bytes}.
7712 When converting variable-length records to fixed-length ones
7713 (@option{conv=block}) or the reverse (@option{conv=unblock}),
7714 use @var{bytes} as the fixed record length.
7716 @item skip=@var{blocks}
7718 Skip @var{blocks} @samp{ibs}-byte blocks in the input file before copying.
7720 @item seek=@var{blocks}
7722 Skip @var{blocks} @samp{obs}-byte blocks in the output file before copying.
7724 @item count=@var{blocks}
7726 Copy @var{blocks} @samp{ibs}-byte blocks from the input file, instead
7727 of everything until the end of the file.
7731 Do not print the overall transfer rate and volume statistics
7732 that normally make up the third status line when @command{dd} exits.
7734 @item conv=@var{conversion}[,@var{conversion}]@dots{}
7736 Convert the file as specified by the @var{conversion} argument(s).
7737 (No spaces around any comma(s).)
7744 @opindex ascii@r{, converting to}
7745 Convert @acronym{EBCDIC} to @acronym{ASCII},
7746 using the conversion table specified by @acronym{POSIX}.
7747 This provides a 1:1 translation for all 256 bytes.
7750 @opindex ebcdic@r{, converting to}
7751 Convert @acronym{ASCII} to @acronym{EBCDIC}.
7752 This is the inverse of the @samp{ascii} conversion.
7755 @opindex alternate ebcdic@r{, converting to}
7756 Convert @acronym{ASCII} to alternate @acronym{EBCDIC},
7757 using the alternate conversion table specified by @acronym{POSIX}.
7758 This is not a 1:1 translation, but reflects common historical practice
7759 for @samp{~}, @samp{[}, and @samp{]}.
7761 The @samp{ascii}, @samp{ebcdic}, and @samp{ibm} conversions are
7765 @opindex block @r{(space-padding)}
7766 For each line in the input, output @samp{cbs} bytes, replacing the
7767 input newline with a space and padding with spaces as necessary.
7771 Remove any trailing spaces in each @samp{cbs}-sized input block,
7772 and append a newline.
7774 The @samp{block} and @samp{unblock} conversions are mutually exclusive.
7777 @opindex lcase@r{, converting to}
7778 Change uppercase letters to lowercase.
7781 @opindex ucase@r{, converting to}
7782 Change lowercase letters to uppercase.
7784 The @samp{lcase} and @samp{ucase} conversions are mutually exclusive.
7787 @opindex swab @r{(byte-swapping)}
7788 @cindex byte-swapping
7789 Swap every pair of input bytes. @sc{gnu} @command{dd}, unlike others, works
7790 when an odd number of bytes are read---the last byte is simply copied
7791 (since there is nothing to swap it with).
7795 @cindex read errors, ignoring
7796 Continue after read errors.
7800 @cindex creating output file, avoiding
7801 Do not create the output file; the output file must already exist.
7805 @cindex creating output file, requiring
7806 Fail if the output file already exists; @command{dd} must create the
7809 The @samp{excl} and @samp{nocreat} conversions are mutually exclusive.
7813 @cindex truncating output file, avoiding
7814 Do not truncate the output file.
7817 @opindex sync @r{(padding with @acronym{ASCII} @sc{nul}s)}
7818 Pad every input block to size of @samp{ibs} with trailing zero bytes.
7819 When used with @samp{block} or @samp{unblock}, pad with spaces instead of
7824 @cindex synchronized data writes, before finishing
7825 Synchronize output data just before finishing. This forces a physical
7826 write of output data.
7830 @cindex synchronized data and metadata writes, before finishing
7831 Synchronize output data and metadata just before finishing. This
7832 forces a physical write of output data and metadata.
7836 @item iflag=@var{flag}[,@var{flag}]@dots{}
7838 Access the input file using the flags specified by the @var{flag}
7839 argument(s). (No spaces around any comma(s).)
7841 @item oflag=@var{flag}[,@var{flag}]@dots{}
7843 Access the output file using the flags specified by the @var{flag}
7844 argument(s). (No spaces around any comma(s).)
7846 Here are the flags. Not every flag is supported on every operating
7853 @cindex appending to the output file
7854 Write in append mode, so that even if some other process is writing to
7855 this file, every @command{dd} write will append to the current
7856 contents of the file. This flag makes sense only for output.
7857 If you combine this flag with the @samp{of=@var{file}} operand,
7858 you should also specify @samp{conv=notrunc} unless you want the
7859 output file to be truncated before being appended to.
7863 @cindex concurrent I/O
7864 Use concurrent I/O mode for data. This mode performs direct I/O
7865 and drops the @acronym{POSIX} requirement to serialize all I/O to the same file.
7866 A file cannot be opened in CIO mode and with a standard open at the
7872 Use direct I/O for data, avoiding the buffer cache.
7873 Note that the kernel may impose restrictions on read or write buffer sizes.
7874 For example, with an ext4 destination file system and a linux-based kernel,
7875 using @samp{oflag=direct} will cause writes to fail with @code{EINVAL} if the
7876 output buffer size is not a multiple of 512.
7880 @cindex directory I/O
7882 Fail unless the file is a directory. Most operating systems do not
7883 allow I/O to a directory, so this flag has limited utility.
7887 @cindex synchronized data reads
7888 Use synchronized I/O for data. For the output file, this forces a
7889 physical write of output data on each write. For the input file,
7890 this flag can matter when reading from a remote file that has been
7891 written to synchronously by some other process. Metadata (e.g.,
7892 last-access and last-modified time) is not necessarily synchronized.
7896 @cindex synchronized data and metadata I/O
7897 Use synchronized I/O for both data and metadata.
7901 @cindex nonblocking I/O
7902 Use non-blocking I/O.
7907 Do not update the file's access time.
7908 Some older file systems silently ignore this flag, so it is a good
7909 idea to test it on your files before relying on it.
7913 @cindex controlling terminal
7914 Do not assign the file to be a controlling terminal for @command{dd}.
7915 This has no effect when the file is not a terminal.
7916 On many hosts (e.g., @acronym{GNU}/Linux hosts), this option has no effect
7921 @cindex symbolic links, following
7922 Do not follow symbolic links.
7927 Fail if the file has multiple hard links.
7932 Use binary I/O. This option has an effect only on nonstandard
7933 platforms that distinguish binary from text I/O.
7938 Use text I/O. Like @samp{binary}, this option has no effect on
7943 Accumulate full blocks from input. The @code{read} system call
7944 may return early if a full block is not available.
7945 When that happens, continue calling @code{read} to fill the remainder
7947 This flag can be used only with @code{iflag}.
7951 These flags are not supported on all systems, and @samp{dd} rejects
7952 attempts to use them when they are not supported. When reading from
7953 standard input or writing to standard output, the @samp{nofollow} and
7954 @samp{noctty} flags should not be specified, and the other flags
7955 (e.g., @samp{nonblock}) can affect how other processes behave with the
7956 affected file descriptors, even after @command{dd} exits.
7960 @cindex multipliers after numbers
7961 The numeric-valued strings above (@var{bytes} and @var{blocks}) can be
7962 followed by a multiplier: @samp{b}=512, @samp{c}=1,
7963 @samp{w}=2, @samp{x@var{m}}=@var{m}, or any of the
7964 standard block size suffixes like @samp{k}=1024 (@pxref{Block size}).
7966 Use different @command{dd} invocations to use different block sizes for
7967 skipping and I/O@. For example, the following shell commands copy data
7968 in 512 KiB blocks between a disk and a tape, but do not save or restore a
7969 4 KiB label at the start of the disk:
7972 disk=/dev/rdsk/c0t1d0s2
7975 # Copy all but the label from disk to tape.
7976 (dd bs=4k skip=1 count=0 && dd bs=512k) <$disk >$tape
7978 # Copy from tape back to disk, but leave the disk label alone.
7979 (dd bs=4k seek=1 count=0 && dd bs=512k) <$tape >$disk
7982 Sending an @samp{INFO} signal to a running @command{dd}
7983 process makes it print I/O statistics to standard error
7984 and then resume copying. In the example below,
7985 @command{dd} is run in the background to copy 10 million blocks.
7986 The @command{kill} command makes it output intermediate I/O statistics,
7987 and when @command{dd} completes normally or is killed by the
7988 @code{SIGINT} signal, it outputs the final statistics.
7991 $ dd if=/dev/zero of=/dev/null count=10MB & pid=$!
7992 $ kill -s INFO $pid; wait $pid
7993 3385223+0 records in
7994 3385223+0 records out
7995 1733234176 bytes (1.7 GB) copied, 6.42173 seconds, 270 MB/s
7996 10000000+0 records in
7997 10000000+0 records out
7998 5120000000 bytes (5.1 GB) copied, 18.913 seconds, 271 MB/s
8001 @vindex POSIXLY_CORRECT
8002 On systems lacking the @samp{INFO} signal @command{dd} responds to the
8003 @samp{USR1} signal instead, unless the @env{POSIXLY_CORRECT}
8004 environment variable is set.
8009 @node install invocation
8010 @section @command{install}: Copy files and set attributes
8013 @cindex copying files and setting attributes
8015 @command{install} copies files while setting their file mode bits and, if
8016 possible, their owner and group. Synopses:
8019 install [@var{option}]@dots{} [-T] @var{source} @var{dest}
8020 install [@var{option}]@dots{} @var{source}@dots{} @var{directory}
8021 install [@var{option}]@dots{} -t @var{directory} @var{source}@dots{}
8022 install [@var{option}]@dots{} -d @var{directory}@dots{}
8027 If two file names are given, @command{install} copies the first file to the
8031 If the @option{--target-directory} (@option{-t}) option is given, or
8032 failing that if the last file is a directory and the
8033 @option{--no-target-directory} (@option{-T}) option is not given,
8034 @command{install} copies each @var{source} file to the specified
8035 directory, using the @var{source}s' names.
8038 If the @option{--directory} (@option{-d}) option is given,
8039 @command{install} creates each @var{directory} and any missing parent
8040 directories. Parent directories are created with mode
8041 @samp{u=rwx,go=rx} (755), regardless of the @option{-m} option or the
8042 current umask. @xref{Directory Setuid and Setgid}, for how the
8043 set-user-ID and set-group-ID bits of parent directories are inherited.
8046 @cindex Makefiles, installing programs in
8047 @command{install} is similar to @command{cp}, but allows you to control the
8048 attributes of destination files. It is typically used in Makefiles to
8049 copy programs into their destination directories. It refuses to copy
8050 files onto themselves.
8052 @cindex extended attributes, xattr
8053 @command{install} never preserves extended attributes (xattr).
8055 The program accepts the following options. Also see @ref{Common options}.
8065 Compare each pair of source and destination files, and if the destination has
8066 identical content and any specified owner, group, permissions, and possibly
8067 SELinux context, then do not modify the destination at all.
8071 Ignored; for compatibility with old Unix versions of @command{install}.
8075 Create any missing parent directories of @var{dest},
8076 then copy @var{source} to @var{dest}.
8077 This option is ignored if a destination directory is specified
8078 via @option{--target-directory=DIR}.
8083 @opindex --directory
8084 @cindex directories, creating with given attributes
8085 @cindex parent directories, creating missing
8086 @cindex leading directories, creating missing
8087 Create any missing parent directories, giving them the default
8088 attributes. Then create each given directory, setting their owner,
8089 group and mode as given on the command line or to the defaults.
8091 @item -g @var{group}
8092 @itemx --group=@var{group}
8095 @cindex group ownership of installed files, setting
8096 Set the group ownership of installed files or directories to
8097 @var{group}. The default is the process's current group. @var{group}
8098 may be either a group name or a numeric group ID.
8101 @itemx --mode=@var{mode}
8104 @cindex permissions of installed files, setting
8105 Set the file mode bits for the installed file or directory to @var{mode},
8106 which can be either an octal number, or a symbolic mode as in
8107 @command{chmod}, with @samp{a=} (no access allowed to anyone) as the
8108 point of departure (@pxref{File permissions}).
8109 The default mode is @samp{u=rwx,go=rx,a-s}---read, write, and
8110 execute for the owner, read and execute for group and other, and with
8111 set-user-ID and set-group-ID disabled.
8112 This default is not quite the same as @samp{755}, since it disables
8113 instead of preserving set-user-ID and set-group-ID on directories.
8114 @xref{Directory Setuid and Setgid}.
8116 @item -o @var{owner}
8117 @itemx --owner=@var{owner}
8120 @cindex ownership of installed files, setting
8121 @cindex appropriate privileges
8122 @vindex root @r{as default owner}
8123 If @command{install} has appropriate privileges (is run as root), set the
8124 ownership of installed files or directories to @var{owner}. The default
8125 is @code{root}. @var{owner} may be either a user name or a numeric user
8128 @item --preserve-context
8129 @opindex --preserve-context
8131 @cindex security context
8132 Preserve the SELinux security context of files and directories.
8133 Failure to preserve the context in all of the files or directories
8134 will result in an exit status of 1. If SELinux is disabled then
8135 print a warning and ignore the option.
8138 @itemx --preserve-timestamps
8140 @opindex --preserve-timestamps
8141 @cindex timestamps of installed files, preserving
8142 Set the time of last access and the time of last modification of each
8143 installed file to match those of each corresponding original file.
8144 When a file is installed without this option, its last access and
8145 last modification times are both set to the time of installation.
8146 This option is useful if you want to use the last modification times
8147 of installed files to keep track of when they were last built as opposed
8148 to when they were last installed.
8154 @cindex symbol table information, stripping
8155 @cindex stripping symbol table information
8156 Strip the symbol tables from installed binary executables.
8158 @itemx --strip-program=@var{program}
8159 @opindex --strip-program
8160 @cindex symbol table information, stripping, program
8161 Program used to strip binaries.
8167 @optNoTargetDirectory
8173 Print the name of each file before copying it.
8175 @item -Z @var{context}
8176 @itemx --context=@var{context}
8180 @cindex security context
8181 Set the default SELinux security context to be used for any
8182 created files and directories. If SELinux is disabled then
8183 print a warning and ignore the option.
8191 @section @command{mv}: Move (rename) files
8195 @command{mv} moves or renames files (or directories). Synopses:
8198 mv [@var{option}]@dots{} [-T] @var{source} @var{dest}
8199 mv [@var{option}]@dots{} @var{source}@dots{} @var{directory}
8200 mv [@var{option}]@dots{} -t @var{directory} @var{source}@dots{}
8205 If two file names are given, @command{mv} moves the first file to the
8209 If the @option{--target-directory} (@option{-t}) option is given, or
8210 failing that if the last file is a directory and the
8211 @option{--no-target-directory} (@option{-T}) option is not given,
8212 @command{mv} moves each @var{source} file to the specified
8213 directory, using the @var{source}s' names.
8216 @command{mv} can move any type of file from one file system to another.
8217 Prior to version @code{4.0} of the fileutils,
8218 @command{mv} could move only regular files between file systems.
8219 For example, now @command{mv} can move an entire directory hierarchy
8220 including special device files from one partition to another. It first
8221 uses some of the same code that's used by @code{cp -a} to copy the
8222 requested directories and files, then (assuming the copy succeeded)
8223 it removes the originals. If the copy fails, then the part that was
8224 copied to the destination partition is removed. If you were to copy
8225 three directories from one partition to another and the copy of the first
8226 directory succeeded, but the second didn't, the first would be left on
8227 the destination partition and the second and third would be left on the
8230 @cindex extended attributes, xattr
8231 @command{mv} always tries to copy extended attributes (xattr).
8233 @cindex prompting, and @command{mv}
8234 If a destination file exists but is normally unwritable, standard input
8235 is a terminal, and the @option{-f} or @option{--force} option is not given,
8236 @command{mv} prompts the user for whether to replace the file. (You might
8237 own the file, or have write permission on its directory.) If the
8238 response is not affirmative, the file is skipped.
8240 @emph{Warning}: Avoid specifying a source name with a trailing slash,
8241 when it might be a symlink to a directory.
8242 Otherwise, @command{mv} may do something very surprising, since
8243 its behavior depends on the underlying rename system call.
8244 On a system with a modern Linux-based kernel, it fails with @code{errno=ENOTDIR}.
8245 However, on other systems (at least FreeBSD 6.1 and Solaris 10) it silently
8246 renames not the symlink but rather the directory referenced by the symlink.
8247 @xref{Trailing slashes}.
8249 The program accepts the following options. Also see @ref{Common options}.
8259 @cindex prompts, omitting
8260 Do not prompt the user before removing a destination file.
8262 If you specify more than one of the @option{-i}, @option{-f}, @option{-n}
8263 options, only the final one takes effect.
8268 @itemx --interactive
8270 @opindex --interactive
8271 @cindex prompts, forcing
8272 Prompt whether to overwrite each existing destination file, regardless
8274 If the response is not affirmative, the file is skipped.
8280 @opindex --no-clobber
8281 @cindex prompts, omitting
8282 Do not overwrite an existing file.
8284 This option is mutually exclusive with @option{-b} or @option{--backup} option.
8290 @cindex newer files, moving only
8291 Do not move a non-directory that has an existing destination with the
8292 same or newer modification time.
8293 If the move is across file system boundaries, the comparison is to the
8294 source time stamp truncated to the resolutions of the destination file
8295 system and of the system calls used to update time stamps; this avoids
8296 duplicate work if several @samp{mv -u} commands are executed with the
8297 same source and destination.
8303 Print the name of each file before moving it.
8305 @optStripTrailingSlashes
8311 @optNoTargetDirectory
8319 @section @command{rm}: Remove files or directories
8322 @cindex removing files or directories
8324 @command{rm} removes each given @var{file}. By default, it does not remove
8325 directories. Synopsis:
8328 rm [@var{option}]@dots{} [@var{file}]@dots{}
8331 @cindex prompting, and @command{rm}
8332 If the @option{-I} or @option{--interactive=once} option is given,
8333 and there are more than three files or the @option{-r}, @option{-R},
8334 or @option{--recursive} are given, then @command{rm} prompts the user
8335 for whether to proceed with the entire operation. If the response is
8336 not affirmative, the entire command is aborted.
8338 Otherwise, if a file is unwritable, standard input is a terminal, and
8339 the @option{-f} or @option{--force} option is not given, or the
8340 @option{-i} or @option{--interactive=always} option @emph{is} given,
8341 @command{rm} prompts the user for whether to remove the file.
8342 If the response is not affirmative, the file is skipped.
8344 Any attempt to remove a file whose last file name component is
8345 @file{.} or @file{..} is rejected without any prompting.
8347 @emph{Warning}: If you use @command{rm} to remove a file, it is usually
8348 possible to recover the contents of that file. If you want more assurance
8349 that the contents are truly unrecoverable, consider using @command{shred}.
8351 The program accepts the following options. Also see @ref{Common options}.
8359 Ignore nonexistent files and never prompt the user.
8360 Ignore any previous @option{--interactive} (@option{-i}) option.
8364 Prompt whether to remove each file.
8365 If the response is not affirmative, the file is skipped.
8366 Ignore any previous @option{--force} (@option{-f}) option.
8367 Equivalent to @option{--interactive=always}.
8371 Prompt once whether to proceed with the command, if more than three
8372 files are named or if a recursive removal is requested. Ignore any
8373 previous @option{--force} (@option{-f}) option. Equivalent to
8374 @option{--interactive=once}.
8376 @itemx --interactive [=@var{when}]
8377 @opindex --interactive
8378 Specify when to issue an interactive prompt. @var{when} may be
8382 @vindex never @r{interactive option}
8383 - Do not prompt at all.
8385 @vindex once @r{interactive option}
8386 - Prompt once if more than three files are named or if a recursive
8387 removal is requested. Equivalent to @option{-I}.
8389 @vindex always @r{interactive option}
8390 - Prompt for every file being removed. Equivalent to @option{-i}.
8392 @option{--interactive} with no @var{when} is equivalent to
8393 @option{--interactive=always}.
8395 @itemx --one-file-system
8396 @opindex --one-file-system
8397 @cindex one file system, restricting @command{rm} to
8398 When removing a hierarchy recursively, skip any directory that is on a
8399 file system different from that of the corresponding command line argument.
8401 This option is useful when removing a build ``chroot'' hierarchy,
8402 which normally contains no valuable data. However, it is not uncommon
8403 to bind-mount @file{/home} into such a hierarchy, to make it easier to
8404 use one's start-up file. The catch is that it's easy to forget to
8405 unmount @file{/home}. Then, when you use @command{rm -rf} to remove
8406 your normally throw-away chroot, that command will remove everything
8407 under @file{/home}, too.
8408 Use the @option{--one-file-system} option, and it will
8409 warn about and skip directories on other file systems.
8410 Of course, this will not save your @file{/home} if it and your
8411 chroot happen to be on the same file system.
8413 @itemx --preserve-root
8414 @opindex --preserve-root
8415 @cindex root directory, disallow recursive destruction
8416 Fail upon any attempt to remove the root directory, @file{/},
8417 when used with the @option{--recursive} option.
8418 This is the default behavior.
8419 @xref{Treating / specially}.
8421 @itemx --no-preserve-root
8422 @opindex --no-preserve-root
8423 @cindex root directory, allow recursive destruction
8424 Do not treat @file{/} specially when removing recursively.
8425 This option is not recommended unless you really want to
8426 remove all the files on your computer.
8427 @xref{Treating / specially}.
8434 @opindex --recursive
8435 @cindex directories, removing (recursively)
8436 Remove the listed directories and their contents recursively.
8442 Print the name of each file before removing it.
8446 @cindex files beginning with @samp{-}, removing
8447 @cindex @samp{-}, removing files beginning with
8448 One common question is how to remove files whose names begin with a
8449 @samp{-}. @sc{gnu} @command{rm}, like every program that uses the @code{getopt}
8450 function to parse its arguments, lets you use the @samp{--} option to
8451 indicate that all following arguments are non-options. To remove a file
8452 called @file{-f} in the current directory, you could type either:
8465 @opindex - @r{and Unix @command{rm}}
8466 The Unix @command{rm} program's use of a single @samp{-} for this purpose
8467 predates the development of the getopt standard syntax.
8472 @node shred invocation
8473 @section @command{shred}: Remove files more securely
8476 @cindex data, erasing
8477 @cindex erasing data
8479 @command{shred} overwrites devices or files, to help prevent even
8480 very expensive hardware from recovering the data.
8482 Ordinarily when you remove a file (@pxref{rm invocation}), the data is
8483 not actually destroyed. Only the index listing where the file is
8484 stored is destroyed, and the storage is made available for reuse.
8485 There are undelete utilities that will attempt to reconstruct the index
8486 and can bring the file back if the parts were not reused.
8488 On a busy system with a nearly-full drive, space can get reused in a few
8489 seconds. But there is no way to know for sure. If you have sensitive
8490 data, you may want to be sure that recovery is not possible by actually
8491 overwriting the file with non-sensitive data.
8493 However, even after doing that, it is possible to take the disk back
8494 to a laboratory and use a lot of sensitive (and expensive) equipment
8495 to look for the faint ``echoes'' of the original data underneath the
8496 overwritten data. If the data has only been overwritten once, it's not
8499 The best way to remove something irretrievably is to destroy the media
8500 it's on with acid, melt it down, or the like. For cheap removable media
8501 like floppy disks, this is the preferred method. However, hard drives
8502 are expensive and hard to melt, so the @command{shred} utility tries
8503 to achieve a similar effect non-destructively.
8505 This uses many overwrite passes, with the data patterns chosen to
8506 maximize the damage they do to the old data. While this will work on
8507 floppies, the patterns are designed for best effect on hard drives.
8508 For more details, see the source code and Peter Gutmann's paper
8509 @uref{http://www.cs.auckland.ac.nz/~pgut001/pubs/secure_del.html,
8510 @cite{Secure Deletion of Data from Magnetic and Solid-State Memory}},
8511 from the proceedings of the Sixth @acronym{USENIX} Security Symposium (San Jose,
8512 California, July 22--25, 1996).
8514 @strong{Please note} that @command{shred} relies on a very important assumption:
8515 that the file system overwrites data in place. This is the traditional
8516 way to do things, but many modern file system designs do not satisfy this
8517 assumption. Exceptions include:
8522 Log-structured or journaled file systems, such as those supplied with
8523 AIX and Solaris, and JFS, ReiserFS, XFS, Ext3 (in @code{data=journal} mode),
8524 BFS, NTFS, etc.@: when they are configured to journal @emph{data}.
8527 File systems that write redundant data and carry on even if some writes
8528 fail, such as RAID-based file systems.
8531 File systems that make snapshots, such as Network Appliance's NFS server.
8534 File systems that cache in temporary locations, such as NFS version 3
8538 Compressed file systems.
8541 In the particular case of ext3 file systems, the above disclaimer applies (and
8542 @command{shred} is thus of limited effectiveness) only in @code{data=journal}
8543 mode, which journals file data in addition to just metadata. In both
8544 the @code{data=ordered} (default) and @code{data=writeback} modes,
8545 @command{shred} works as usual. Ext3 journaling modes can be changed
8546 by adding the @code{data=something} option to the mount options for a
8547 particular file system in the @file{/etc/fstab} file, as documented in
8548 the mount man page (man mount).
8550 If you are not sure how your file system operates, then you should assume
8551 that it does not overwrite data in place, which means that shred cannot
8552 reliably operate on regular files in your file system.
8554 Generally speaking, it is more reliable to shred a device than a file,
8555 since this bypasses the problem of file system design mentioned above.
8556 However, even shredding devices is not always completely reliable. For
8557 example, most disks map out bad sectors invisibly to the application; if
8558 the bad sectors contain sensitive data, @command{shred} won't be able to
8561 @command{shred} makes no attempt to detect or report this problem, just as
8562 it makes no attempt to do anything about backups. However, since it is
8563 more reliable to shred devices than files, @command{shred} by default does
8564 not truncate or remove the output file. This default is more suitable
8565 for devices, which typically cannot be truncated and should not be
8568 Finally, consider the risk of backups and mirrors.
8569 File system backups and remote mirrors may contain copies of the
8570 file that cannot be removed, and that will allow a shredded file
8571 to be recovered later. So if you keep any data you may later want
8572 to destroy using @command{shred}, be sure that it is not backed up or mirrored.
8575 shred [@var{option}]@dots{} @var{file}[@dots{}]
8578 The program accepts the following options. Also see @ref{Common options}.
8586 @cindex force deletion
8587 Override file permissions if necessary to allow overwriting.
8590 @itemx -n @var{number}
8591 @itemx --iterations=@var{number}
8592 @opindex -n @var{number}
8593 @opindex --iterations=@var{number}
8594 @cindex iterations, selecting the number of
8595 By default, @command{shred} uses @value{SHRED_DEFAULT_PASSES} passes of
8596 overwrite. You can reduce this to save time, or increase it if you think it's
8597 appropriate. After 25 passes all of the internal overwrite patterns will have
8598 been used at least once.
8600 @item --random-source=@var{file}
8601 @opindex --random-source
8602 @cindex random source for shredding
8603 Use @var{file} as a source of random data used to overwrite and to
8604 choose pass ordering. @xref{Random sources}.
8606 @item -s @var{bytes}
8607 @itemx --size=@var{bytes}
8608 @opindex -s @var{bytes}
8609 @opindex --size=@var{bytes}
8610 @cindex size of file to shred
8611 Shred the first @var{bytes} bytes of the file. The default is to shred
8612 the whole file. @var{bytes} can be followed by a size specification like
8613 @samp{K}, @samp{M}, or @samp{G} to specify a multiple. @xref{Block size}.
8619 @cindex removing files after shredding
8620 After shredding a file, truncate it (if possible) and then remove it.
8621 If a file has multiple links, only the named links will be removed.
8627 Display to standard error all status updates as sterilization proceeds.
8633 By default, @command{shred} rounds the size of a regular file up to the next
8634 multiple of the file system block size to fully erase the last block of the file.
8635 Use @option{--exact} to suppress that behavior.
8636 Thus, by default if you shred a 10-byte regular file on a system with 512-byte
8637 blocks, the resulting file will be 512 bytes long. With this option,
8638 shred does not increase the apparent size of the file.
8644 Normally, the last pass that @command{shred} writes is made up of
8645 random data. If this would be conspicuous on your hard drive (for
8646 example, because it looks like encrypted data), or you just think
8647 it's tidier, the @option{--zero} option adds an additional overwrite pass with
8648 all zero bits. This is in addition to the number of passes specified
8649 by the @option{--iterations} option.
8653 You might use the following command to erase all trace of the
8654 file system you'd created on the floppy disk in your first drive.
8655 That command takes about 20 minutes to erase a ``1.44MB'' (actually
8659 shred --verbose /dev/fd0
8662 Similarly, to erase all data on a selected partition of
8663 your hard disk, you could give a command like this:
8666 shred --verbose /dev/sda5
8669 A @var{file} of @samp{-} denotes standard output.
8670 The intended use of this is to shred a removed temporary file.
8674 i=`tempfile -m 0600`
8677 echo "Hello, world" >&3
8682 However, the command @samp{shred - >file} does not shred the contents
8683 of @var{file}, since the shell truncates @var{file} before invoking
8684 @command{shred}. Use the command @samp{shred file} or (if using a
8685 Bourne-compatible shell) the command @samp{shred - 1<>file} instead.
8690 @node Special file types
8691 @chapter Special file types
8693 @cindex special file types
8694 @cindex file types, special
8696 This chapter describes commands which create special types of files (and
8697 @command{rmdir}, which removes directories, one special file type).
8699 @cindex special file types
8701 Although Unix-like operating systems have markedly fewer special file
8702 types than others, not @emph{everything} can be treated only as the
8703 undifferentiated byte stream of @dfn{normal files}. For example, when a
8704 file is created or removed, the system must record this information,
8705 which it does in a @dfn{directory}---a special type of file. Although
8706 you can read directories as normal files, if you're curious, in order
8707 for the system to do its job it must impose a structure, a certain
8708 order, on the bytes of the file. Thus it is a ``special'' type of file.
8710 Besides directories, other special file types include named pipes
8711 (FIFOs), symbolic links, sockets, and so-called @dfn{special files}.
8714 * link invocation:: Make a hard link via the link syscall
8715 * ln invocation:: Make links between files.
8716 * mkdir invocation:: Make directories.
8717 * mkfifo invocation:: Make FIFOs (named pipes).
8718 * mknod invocation:: Make block or character special files.
8719 * readlink invocation:: Print value of a symlink or canonical file name.
8720 * rmdir invocation:: Remove empty directories.
8721 * unlink invocation:: Remove files via the unlink syscall
8725 @node link invocation
8726 @section @command{link}: Make a hard link via the link syscall
8729 @cindex links, creating
8730 @cindex hard links, creating
8731 @cindex creating links (hard only)
8733 @command{link} creates a single hard link at a time.
8734 It is a minimalist interface to the system-provided
8735 @code{link} function. @xref{Hard Links, , , libc,
8736 The GNU C Library Reference Manual}.
8737 It avoids the bells and whistles of the more commonly-used
8738 @command{ln} command (@pxref{ln invocation}).
8742 link @var{filename} @var{linkname}
8745 @var{filename} must specify an existing file, and @var{linkname}
8746 must specify a nonexistent entry in an existing directory.
8747 @command{link} simply calls @code{link (@var{filename}, @var{linkname})}
8750 On a @acronym{GNU} system, this command acts like @samp{ln --directory
8751 --no-target-directory @var{filename} @var{linkname}}. However, the
8752 @option{--directory} and @option{--no-target-directory} options are
8753 not specified by @acronym{POSIX}, and the @command{link} command is
8754 more portable in practice.
8760 @section @command{ln}: Make links between files
8763 @cindex links, creating
8764 @cindex hard links, creating
8765 @cindex symbolic (soft) links, creating
8766 @cindex creating links (hard or soft)
8768 @cindex file systems and hard links
8769 @command{ln} makes links between files. By default, it makes hard links;
8770 with the @option{-s} option, it makes symbolic (or @dfn{soft}) links.
8774 ln [@var{option}]@dots{} [-T] @var{target} @var{linkname}
8775 ln [@var{option}]@dots{} @var{target}
8776 ln [@var{option}]@dots{} @var{target}@dots{} @var{directory}
8777 ln [@var{option}]@dots{} -t @var{directory} @var{target}@dots{}
8783 If two file names are given, @command{ln} creates a link to the first
8784 file from the second.
8787 If one @var{target} is given, @command{ln} creates a link to that file
8788 in the current directory.
8791 If the @option{--target-directory} (@option{-t}) option is given, or
8792 failing that if the last file is a directory and the
8793 @option{--no-target-directory} (@option{-T}) option is not given,
8794 @command{ln} creates a link to each @var{target} file in the specified
8795 directory, using the @var{target}s' names.
8799 Normally @command{ln} does not remove existing files. Use the
8800 @option{--force} (@option{-f}) option to remove them unconditionally,
8801 the @option{--interactive} (@option{-i}) option to remove them
8802 conditionally, and the @option{--backup} (@option{-b}) option to
8805 @cindex hard link, defined
8806 @cindex inode, and hard links
8807 A @dfn{hard link} is another name for an existing file; the link and the
8808 original are indistinguishable. Technically speaking, they share the
8809 same inode, and the inode contains all the information about a
8810 file---indeed, it is not incorrect to say that the inode @emph{is} the
8811 file. On all existing implementations, you cannot make a hard link to
8812 a directory, and hard links cannot cross file system boundaries. (These
8813 restrictions are not mandated by @acronym{POSIX}, however.)
8815 @cindex dereferencing symbolic links
8816 @cindex symbolic link, defined
8817 @dfn{Symbolic links} (@dfn{symlinks} for short), on the other hand, are
8818 a special file type (which not all kernels support: System V release 3
8819 (and older) systems lack symlinks) in which the link file actually
8820 refers to a different file, by name. When most operations (opening,
8821 reading, writing, and so on) are passed the symbolic link file, the
8822 kernel automatically @dfn{dereferences} the link and operates on the
8823 target of the link. But some operations (e.g., removing) work on the
8824 link file itself, rather than on its target. The owner, group, and
8825 mode of a symlink are not significant to file access performed through
8826 the link. @xref{Symbolic Links,,,
8827 libc, The GNU C Library Reference Manual}.
8829 Symbolic links can contain arbitrary strings; a @dfn{dangling symlink}
8830 occurs when the string in the symlink does not resolve to a file.
8831 There are no restrictions against creating dangling symbolic links.
8832 There are trade-offs to using absolute or relative symlinks. An
8833 absolute symlink always points to the same file, even if the directory
8834 containing the link is moved. However, if the symlink is visible from
8835 more than one machine (such as on a networked file system), the file
8836 pointed to might not always be the same. A relative symbolic link is
8837 resolved in relation to the directory that contains the link, and is
8838 often useful in referring to files on the same device without regards
8839 to what name that device is mounted on when accessed via networked
8842 When creating a relative symlink in a different location than the
8843 current directory, the resolution of the symlink will be different
8844 than the resolution of the same string from the current directory.
8845 Therefore, many users prefer to first change directories to the
8846 location where the relative symlink will be created, so that
8847 tab-completion or other file resolution will find the same target as
8848 what will be placed in the symlink.
8850 The program accepts the following options. Also see @ref{Common options}.
8861 @opindex --directory
8862 @cindex hard links to directories
8863 Allow users with appropriate privileges to attempt to make hard links
8865 However, note that this will probably fail due to
8866 system restrictions, even for the super-user.
8872 Remove existing destination files.
8875 @itemx --interactive
8877 @opindex --interactive
8878 @cindex prompting, and @command{ln}
8879 Prompt whether to remove existing destination files.
8882 @itemx --no-dereference
8884 @opindex --no-dereference
8885 Do not treat the last operand specially when it is a symbolic link to
8886 a directory. Instead, treat it as if it were a normal file.
8888 When the destination is an actual directory (not a symlink to one),
8889 there is no ambiguity. The link is created in that directory.
8890 But when the specified destination is a symlink to a directory,
8891 there are two ways to treat the user's request. @command{ln} can
8892 treat the destination just as it would a normal directory and create
8893 the link in it. On the other hand, the destination can be viewed as a
8894 non-directory---as the symlink itself. In that case, @command{ln}
8895 must delete or backup that symlink before creating the new link.
8896 The default is to treat a destination that is a symlink to a directory
8897 just like a directory.
8899 This option is weaker than the @option{--no-target-directory}
8900 (@option{-T}) option, so it has no effect if both options are given.
8906 Make symbolic links instead of hard links. This option merely produces
8907 an error message on systems that do not support symbolic links.
8913 @optNoTargetDirectory
8919 Print the name of each file after linking it successfully.
8930 # Create link ../a pointing to a in that directory.
8931 # Not really useful because it points to itself.
8936 # Change to the target before creating symlinks to avoid being confused.
8942 # Hard coded file names don't move well.
8943 ln -s $(pwd)/a /some/dir/
8947 # Relative file names survive directory moves and also
8948 # work across networked file systems.
8949 ln -s afile anotherfile
8950 ln -s ../adir/afile yetanotherfile
8954 @node mkdir invocation
8955 @section @command{mkdir}: Make directories
8958 @cindex directories, creating
8959 @cindex creating directories
8961 @command{mkdir} creates directories with the specified names. Synopsis:
8964 mkdir [@var{option}]@dots{} @var{name}@dots{}
8967 @command{mkdir} creates each directory @var{name} in the order given.
8968 It reports an error if @var{name} already exists, unless the
8969 @option{-p} option is given and @var{name} is a directory.
8971 The program accepts the following options. Also see @ref{Common options}.
8976 @itemx --mode=@var{mode}
8979 @cindex modes of created directories, setting
8980 Set the file permission bits of created directories to @var{mode},
8981 which uses the same syntax as
8982 in @command{chmod} and uses @samp{a=rwx} (read, write and execute allowed for
8983 everyone) for the point of the departure. @xref{File permissions}.
8985 Normally the directory has the desired file mode bits at the moment it
8986 is created. As a @acronym{GNU} extension, @var{mode} may also mention
8987 special mode bits, but in this case there may be a temporary window
8988 during which the directory exists but its special mode bits are
8989 incorrect. @xref{Directory Setuid and Setgid}, for how the
8990 set-user-ID and set-group-ID bits of directories are inherited unless
8991 overridden in this way.
8997 @cindex parent directories, creating
8998 Make any missing parent directories for each argument, setting their
8999 file permission bits to the umask modified by @samp{u+wx}. Ignore
9000 existing parent directories, and do not change their file permission
9003 To set the file permission bits of any newly-created parent
9004 directories to a value that includes @samp{u+wx}, you can set the
9005 umask before invoking @command{mkdir}. For example, if the shell
9006 command @samp{(umask u=rwx,go=rx; mkdir -p P/Q)} creates the parent
9007 @file{P} it sets the parent's permission bits to @samp{u=rwx,go=rx}.
9008 To set a parent's special mode bits as well, you can invoke
9009 @command{chmod} after @command{mkdir}. @xref{Directory Setuid and
9010 Setgid}, for how the set-user-ID and set-group-ID bits of
9011 newly-created parent directories are inherited.
9017 Print a message for each created directory. This is most useful with
9020 @item -Z @var{context}
9021 @itemx --context=@var{context}
9025 @cindex security context
9026 Set the default SELinux security context to be used for created directories.
9033 @node mkfifo invocation
9034 @section @command{mkfifo}: Make FIFOs (named pipes)
9037 @cindex FIFOs, creating
9038 @cindex named pipes, creating
9039 @cindex creating FIFOs (named pipes)
9041 @command{mkfifo} creates FIFOs (also called @dfn{named pipes}) with the
9042 specified names. Synopsis:
9045 mkfifo [@var{option}] @var{name}@dots{}
9048 A @dfn{FIFO} is a special file type that permits independent processes
9049 to communicate. One process opens the FIFO file for writing, and
9050 another for reading, after which data can flow as with the usual
9051 anonymous pipe in shells or elsewhere.
9053 The program accepts the following option. Also see @ref{Common options}.
9058 @itemx --mode=@var{mode}
9061 @cindex modes of created FIFOs, setting
9062 Set the mode of created FIFOs to @var{mode}, which is symbolic as in
9063 @command{chmod} and uses @samp{a=rw} (read and write allowed for everyone)
9064 for the point of departure. @var{mode} should specify only file
9065 permission bits. @xref{File permissions}.
9067 @item -Z @var{context}
9068 @itemx --context=@var{context}
9072 @cindex security context
9073 Set the default SELinux security context to be used for created FIFOs.
9080 @node mknod invocation
9081 @section @command{mknod}: Make block or character special files
9084 @cindex block special files, creating
9085 @cindex character special files, creating
9087 @command{mknod} creates a FIFO, character special file, or block special
9088 file with the specified name. Synopsis:
9091 mknod [@var{option}]@dots{} @var{name} @var{type} [@var{major} @var{minor}]
9094 @cindex special files
9095 @cindex block special files
9096 @cindex character special files
9097 Unlike the phrase ``special file type'' above, the term @dfn{special
9098 file} has a technical meaning on Unix: something that can generate or
9099 receive data. Usually this corresponds to a physical piece of hardware,
9100 e.g., a printer or a disk. (These files are typically created at
9101 system-configuration time.) The @command{mknod} command is what creates
9102 files of this type. Such devices can be read either a character at a
9103 time or a ``block'' (many characters) at a time, hence we say there are
9104 @dfn{block special} files and @dfn{character special} files.
9106 @c mknod is a shell built-in at least with OpenBSD's /bin/sh
9107 @mayConflictWithShellBuiltIn{mknod}
9109 The arguments after @var{name} specify the type of file to make:
9114 @opindex p @r{for FIFO file}
9118 @opindex b @r{for block special file}
9119 for a block special file
9122 @c Don't document the `u' option -- it's just a synonym for `c'.
9123 @c Do *any* versions of mknod still use it?
9125 @opindex c @r{for character special file}
9126 @c @opindex u @r{for character special file}
9127 for a character special file
9131 When making a block or character special file, the major and minor
9132 device numbers must be given after the file type.
9133 If a major or minor device number begins with @samp{0x} or @samp{0X},
9134 it is interpreted as hexadecimal; otherwise, if it begins with @samp{0},
9135 as octal; otherwise, as decimal.
9137 The program accepts the following option. Also see @ref{Common options}.
9142 @itemx --mode=@var{mode}
9145 Set the mode of created files to @var{mode}, which is symbolic as in
9146 @command{chmod} and uses @samp{a=rw} as the point of departure.
9147 @var{mode} should specify only file permission bits.
9148 @xref{File permissions}.
9150 @item -Z @var{context}
9151 @itemx --context=@var{context}
9155 @cindex security context
9156 Set the default SELinux security context to be used for created files.
9163 @node readlink invocation
9164 @section @command{readlink}: Print value of a symlink or canonical file name
9167 @cindex displaying value of a symbolic link
9168 @cindex canonical file name
9169 @cindex canonicalize a file name
9173 @command{readlink} may work in one of two supported modes:
9179 @command{readlink} outputs the value of the given symbolic link.
9180 If @command{readlink} is invoked with an argument other than the name
9181 of a symbolic link, it produces no output and exits with a nonzero exit code.
9183 @item Canonicalize mode
9185 @command{readlink} outputs the absolute name of the given file which contains
9186 no @file{.}, @file{..} components nor any repeated separators
9187 (@file{/}) or symbolic links.
9192 readlink [@var{option}] @var{file}
9195 By default, @command{readlink} operates in readlink mode.
9197 The program accepts the following options. Also see @ref{Common options}.
9202 @itemx --canonicalize
9204 @opindex --canonicalize
9205 Activate canonicalize mode.
9206 If any component of the file name except the last one is missing or unavailable,
9207 @command{readlink} produces no output and exits with a nonzero exit code.
9210 @itemx --canonicalize-existing
9212 @opindex --canonicalize-existing
9213 Activate canonicalize mode.
9214 If any component is missing or unavailable, @command{readlink} produces
9215 no output and exits with a nonzero exit code.
9218 @itemx --canonicalize-missing
9220 @opindex --canonicalize-missing
9221 Activate canonicalize mode.
9222 If any component is missing or unavailable, @command{readlink} treats it
9228 @opindex --no-newline
9229 Do not output the trailing newline.
9239 Suppress most error messages.
9245 Report error messages.
9249 The @command{readlink} utility first appeared in OpenBSD 2.1.
9251 There is a @command{realpath} command on some systems
9252 which operates like @command{readlink} in canonicalize mode.
9257 @node rmdir invocation
9258 @section @command{rmdir}: Remove empty directories
9261 @cindex removing empty directories
9262 @cindex directories, removing empty
9264 @command{rmdir} removes empty directories. Synopsis:
9267 rmdir [@var{option}]@dots{} @var{directory}@dots{}
9270 If any @var{directory} argument does not refer to an existing empty
9271 directory, it is an error.
9273 The program accepts the following options. Also see @ref{Common options}.
9277 @item --ignore-fail-on-non-empty
9278 @opindex --ignore-fail-on-non-empty
9279 @cindex directory deletion, ignoring failures
9280 Ignore each failure to remove a directory that is solely because
9281 the directory is non-empty.
9287 @cindex parent directories, removing
9288 Remove @var{directory}, then try to remove each component of @var{directory}.
9289 So, for example, @samp{rmdir -p a/b/c} is similar to @samp{rmdir a/b/c a/b a}.
9290 As such, it fails if any of those directories turns out not to be empty.
9291 Use the @option{--ignore-fail-on-non-empty} option to make it so such
9292 a failure does not evoke a diagnostic and does not cause @command{rmdir} to
9293 exit unsuccessfully.
9299 @cindex directory deletion, reporting
9300 Give a diagnostic for each successful removal.
9301 @var{directory} is removed.
9305 @xref{rm invocation}, for how to remove non-empty directories (recursively).
9310 @node unlink invocation
9311 @section @command{unlink}: Remove files via the unlink syscall
9314 @cindex removing files or directories (via the unlink syscall)
9316 @command{unlink} deletes a single specified file name.
9317 It is a minimalist interface to the system-provided
9318 @code{unlink} function. @xref{Deleting Files, , , libc,
9319 The GNU C Library Reference Manual}. Synopsis:
9320 It avoids the bells and whistles of the more commonly-used
9321 @command{rm} command (@pxref{rm invocation}).
9324 unlink @var{filename}
9327 On some systems @code{unlink} can be used to delete the name of a
9328 directory. On others, it can be used that way only by a privileged user.
9329 In the GNU system @code{unlink} can never delete the name of a directory.
9331 The @command{unlink} command honors the @option{--help} and
9332 @option{--version} options. To remove a file whose name begins with
9333 @samp{-}, prefix the name with @samp{./}, e.g., @samp{unlink ./--help}.
9338 @node Changing file attributes
9339 @chapter Changing file attributes
9341 @cindex changing file attributes
9342 @cindex file attributes, changing
9343 @cindex attributes, file
9345 A file is not merely its contents, a name, and a file type
9346 (@pxref{Special file types}). A file also has an owner (a user ID), a
9347 group (a group ID), permissions (what the owner can do with the file,
9348 what people in the group can do, and what everyone else can do), various
9349 timestamps, and other information. Collectively, we call these a file's
9352 These commands change file attributes.
9355 * chgrp invocation:: Change file groups.
9356 * chmod invocation:: Change access permissions.
9357 * chown invocation:: Change file owners and groups.
9358 * touch invocation:: Change file timestamps.
9362 @node chown invocation
9363 @section @command{chown}: Change file owner and group
9366 @cindex file ownership, changing
9367 @cindex group ownership, changing
9368 @cindex changing file ownership
9369 @cindex changing group ownership
9371 @command{chown} changes the user and/or group ownership of each given @var{file}
9372 to @var{new-owner} or to the user and group of an existing reference file.
9376 chown [@var{option}]@dots{} @{@var{new-owner} | --reference=@var{ref_file}@} @var{file}@dots{}
9379 If used, @var{new-owner} specifies the new owner and/or group as follows
9380 (with no embedded white space):
9383 [@var{owner}] [ : [@var{group}] ]
9390 If only an @var{owner} (a user name or numeric user ID) is given, that
9391 user is made the owner of each given file, and the files' group is not
9394 @item owner@samp{:}group
9395 If the @var{owner} is followed by a colon and a @var{group} (a
9396 group name or numeric group ID), with no spaces between them, the group
9397 ownership of the files is changed as well (to @var{group}).
9400 If a colon but no group name follows @var{owner}, that user is
9401 made the owner of the files and the group of the files is changed to
9402 @var{owner}'s login group.
9405 If the colon and following @var{group} are given, but the owner
9406 is omitted, only the group of the files is changed; in this case,
9407 @command{chown} performs the same function as @command{chgrp}.
9410 If only a colon is given, or if @var{new-owner} is empty, neither the
9411 owner nor the group is changed.
9415 If @var{owner} or @var{group} is intended to represent a numeric user
9416 or group ID, then you may specify it with a leading @samp{+}.
9417 @xref{Disambiguating names and IDs}.
9419 Some older scripts may still use @samp{.} in place of the @samp{:} separator.
9420 @acronym{POSIX} 1003.1-2001 (@pxref{Standards conformance}) does not
9421 require support for that, but for backward compatibility @acronym{GNU}
9422 @command{chown} supports @samp{.} so long as no ambiguity results.
9423 New scripts should avoid the use of @samp{.} because it is not
9424 portable, and because it has undesirable results if the entire
9425 @var{owner@samp{.}group} happens to identify a user whose name
9428 The @command{chown} command sometimes clears the set-user-ID or
9429 set-group-ID permission bits. This behavior depends on the policy and
9430 functionality of the underlying @code{chown} system call, which may
9431 make system-dependent file mode modifications outside the control of
9432 the @command{chown} command. For example, the @command{chown} command
9433 might not affect those bits when invoked by a user with appropriate
9434 privileges, or when the
9435 bits signify some function other than executable permission (e.g.,
9437 When in doubt, check the underlying system behavior.
9439 The program accepts the following options. Also see @ref{Common options}.
9447 @cindex changed owners, verbosely describing
9448 Verbosely describe the action for each @var{file} whose ownership
9457 @cindex error messages, omitting
9458 Do not print error messages about files whose ownership cannot be
9461 @itemx @w{@kbd{--from}=@var{old-owner}}
9463 @cindex symbolic links, changing owner
9464 Change a @var{file}'s ownership only if it has current attributes specified
9465 by @var{old-owner}. @var{old-owner} has the same form as @var{new-owner}
9467 This option is useful primarily from a security standpoint in that
9468 it narrows considerably the window of potential abuse.
9469 For example, to reflect a user ID numbering change for one user's files
9470 without an option like this, @code{root} might run
9473 find / -owner OLDUSER -print0 | xargs -0 chown -h NEWUSER
9476 But that is dangerous because the interval between when the @command{find}
9477 tests the existing file's owner and when the @command{chown} is actually run
9479 One way to narrow the gap would be to invoke chown for each file
9483 find / -owner OLDUSER -exec chown -h NEWUSER @{@} \;
9486 But that is very slow if there are many affected files.
9487 With this option, it is safer (the gap is narrower still)
9488 though still not perfect:
9491 chown -h -R --from=OLDUSER NEWUSER /
9495 @opindex --dereference
9496 @cindex symbolic links, changing owner
9498 Do not act on symbolic links themselves but rather on what they point to.
9499 This is the default.
9502 @itemx --no-dereference
9504 @opindex --no-dereference
9505 @cindex symbolic links, changing owner
9507 Act on symbolic links themselves instead of what they point to.
9508 This mode relies on the @code{lchown} system call.
9509 On systems that do not provide the @code{lchown} system call,
9510 @command{chown} fails when a file specified on the command line
9512 By default, no diagnostic is issued for symbolic links encountered
9513 during a recursive traversal, but see @option{--verbose}.
9515 @itemx --preserve-root
9516 @opindex --preserve-root
9517 @cindex root directory, disallow recursive modification
9518 Fail upon any attempt to recursively change the root directory, @file{/}.
9519 Without @option{--recursive}, this option has no effect.
9520 @xref{Treating / specially}.
9522 @itemx --no-preserve-root
9523 @opindex --no-preserve-root
9524 @cindex root directory, allow recursive modification
9525 Cancel the effect of any preceding @option{--preserve-root} option.
9526 @xref{Treating / specially}.
9528 @item --reference=@var{ref_file}
9529 @opindex --reference
9530 Change the user and group of each @var{file} to be the same as those of
9531 @var{ref_file}. If @var{ref_file} is a symbolic link, do not use the
9532 user and group of the symbolic link, but rather those of the file it
9539 Output a diagnostic for every file processed.
9540 If a symbolic link is encountered during a recursive traversal
9541 on a system without the @code{lchown} system call, and @option{--no-dereference}
9542 is in effect, then issue a diagnostic saying neither the symbolic link nor
9543 its referent is being changed.
9548 @opindex --recursive
9549 @cindex recursively changing file ownership
9550 Recursively change ownership of directories and their contents.
9553 @xref{Traversing symlinks}.
9556 @xref{Traversing symlinks}.
9559 @xref{Traversing symlinks}.
9568 # Change the owner of /u to "root".
9571 # Likewise, but also change its group to "staff".
9574 # Change the owner of /u and subfiles to "root".
9579 @node chgrp invocation
9580 @section @command{chgrp}: Change group ownership
9583 @cindex group ownership, changing
9584 @cindex changing group ownership
9586 @command{chgrp} changes the group ownership of each given @var{file}
9587 to @var{group} (which can be either a group name or a numeric group ID)
9588 or to the group of an existing reference file. Synopsis:
9591 chgrp [@var{option}]@dots{} @{@var{group} | --reference=@var{ref_file}@} @var{file}@dots{}
9594 If @var{group} is intended to represent a
9595 numeric group ID, then you may specify it with a leading @samp{+}.
9596 @xref{Disambiguating names and IDs}.
9598 The program accepts the following options. Also see @ref{Common options}.
9606 @cindex changed files, verbosely describing
9607 Verbosely describe the action for each @var{file} whose group actually
9616 @cindex error messages, omitting
9617 Do not print error messages about files whose group cannot be
9621 @opindex --dereference
9622 @cindex symbolic links, changing owner
9624 Do not act on symbolic links themselves but rather on what they point to.
9625 This is the default.
9628 @itemx --no-dereference
9630 @opindex --no-dereference
9631 @cindex symbolic links, changing group
9633 Act on symbolic links themselves instead of what they point to.
9634 This mode relies on the @code{lchown} system call.
9635 On systems that do not provide the @code{lchown} system call,
9636 @command{chgrp} fails when a file specified on the command line
9638 By default, no diagnostic is issued for symbolic links encountered
9639 during a recursive traversal, but see @option{--verbose}.
9641 @itemx --preserve-root
9642 @opindex --preserve-root
9643 @cindex root directory, disallow recursive modification
9644 Fail upon any attempt to recursively change the root directory, @file{/}.
9645 Without @option{--recursive}, this option has no effect.
9646 @xref{Treating / specially}.
9648 @itemx --no-preserve-root
9649 @opindex --no-preserve-root
9650 @cindex root directory, allow recursive modification
9651 Cancel the effect of any preceding @option{--preserve-root} option.
9652 @xref{Treating / specially}.
9654 @item --reference=@var{ref_file}
9655 @opindex --reference
9656 Change the group of each @var{file} to be the same as that of
9657 @var{ref_file}. If @var{ref_file} is a symbolic link, do not use the
9658 group of the symbolic link, but rather that of the file it refers to.
9664 Output a diagnostic for every file processed.
9665 If a symbolic link is encountered during a recursive traversal
9666 on a system without the @code{lchown} system call, and @option{--no-dereference}
9667 is in effect, then issue a diagnostic saying neither the symbolic link nor
9668 its referent is being changed.
9673 @opindex --recursive
9674 @cindex recursively changing group ownership
9675 Recursively change the group ownership of directories and their contents.
9678 @xref{Traversing symlinks}.
9681 @xref{Traversing symlinks}.
9684 @xref{Traversing symlinks}.
9693 # Change the group of /u to "staff".
9696 # Change the group of /u and subfiles to "staff".
9701 @node chmod invocation
9702 @section @command{chmod}: Change access permissions
9705 @cindex changing access permissions
9706 @cindex access permissions, changing
9707 @cindex permissions, changing access
9709 @command{chmod} changes the access permissions of the named files. Synopsis:
9712 chmod [@var{option}]@dots{} @{@var{mode} | --reference=@var{ref_file}@} @var{file}@dots{}
9715 @cindex symbolic links, permissions of
9716 @command{chmod} never changes the permissions of symbolic links, since
9717 the @command{chmod} system call cannot change their permissions.
9718 This is not a problem since the permissions of symbolic links are
9719 never used. However, for each symbolic link listed on the command
9720 line, @command{chmod} changes the permissions of the pointed-to file.
9721 In contrast, @command{chmod} ignores symbolic links encountered during
9722 recursive directory traversals.
9724 A successful use of @command{chmod} clears the set-group-ID bit of a
9725 regular file if the file's group ID does not match the user's
9726 effective group ID or one of the user's supplementary group IDs,
9727 unless the user has appropriate privileges. Additional restrictions
9728 may cause the set-user-ID and set-group-ID bits of @var{mode} or
9729 @var{ref_file} to be ignored. This behavior depends on the policy and
9730 functionality of the underlying @code{chmod} system call. When in
9731 doubt, check the underlying system behavior.
9733 If used, @var{mode} specifies the new file mode bits.
9734 For details, see the section on @ref{File permissions}.
9735 If you really want @var{mode} to have a leading @samp{-}, you should
9736 use @option{--} first, e.g., @samp{chmod -- -w file}. Typically,
9737 though, @samp{chmod a-w file} is preferable, and @command{chmod -w
9738 file} (without the @option{--}) complains if it behaves differently
9739 from what @samp{chmod a-w file} would do.
9741 The program accepts the following options. Also see @ref{Common options}.
9749 Verbosely describe the action for each @var{file} whose permissions
9758 @cindex error messages, omitting
9759 Do not print error messages about files whose permissions cannot be
9762 @itemx --preserve-root
9763 @opindex --preserve-root
9764 @cindex root directory, disallow recursive modification
9765 Fail upon any attempt to recursively change the root directory, @file{/}.
9766 Without @option{--recursive}, this option has no effect.
9767 @xref{Treating / specially}.
9769 @itemx --no-preserve-root
9770 @opindex --no-preserve-root
9771 @cindex root directory, allow recursive modification
9772 Cancel the effect of any preceding @option{--preserve-root} option.
9773 @xref{Treating / specially}.
9779 Verbosely describe the action or non-action taken for every @var{file}.
9781 @item --reference=@var{ref_file}
9782 @opindex --reference
9783 Change the mode of each @var{file} to be the same as that of @var{ref_file}.
9784 @xref{File permissions}.
9785 If @var{ref_file} is a symbolic link, do not use the mode
9786 of the symbolic link, but rather that of the file it refers to.
9791 @opindex --recursive
9792 @cindex recursively changing access permissions
9793 Recursively change permissions of directories and their contents.
9800 @node touch invocation
9801 @section @command{touch}: Change file timestamps
9804 @cindex changing file timestamps
9805 @cindex file timestamps, changing
9806 @cindex timestamps, changing file
9808 @command{touch} changes the access and/or modification times of the
9809 specified files. Synopsis:
9812 touch [@var{option}]@dots{} @var{file}@dots{}
9815 @cindex empty files, creating
9816 Any @var{file} argument that does not exist is created empty.
9818 A @var{file} argument string of @samp{-} is handled specially and
9819 causes @command{touch} to change the times of the file associated with
9822 @cindex permissions, for changing file timestamps
9823 If changing both the access and modification times to the current
9824 time, @command{touch} can change the timestamps for files that the user
9825 running it does not own but has write permission for. Otherwise, the
9826 user must own the files.
9828 Although @command{touch} provides options for changing two of the times---the
9829 times of last access and modification---of a file, there is actually
9830 a third one as well: the inode change time. This is often referred to
9831 as a file's @code{ctime}.
9832 The inode change time represents the time when the file's meta-information
9833 last changed. One common example of this is when the permissions of a
9834 file change. Changing the permissions doesn't access the file, so
9835 the atime doesn't change, nor does it modify the file, so the mtime
9836 doesn't change. Yet, something about the file itself has changed,
9837 and this must be noted somewhere. This is the job of the ctime field.
9838 This is necessary, so that, for example, a backup program can make a
9839 fresh copy of the file, including the new permissions value.
9840 Another operation that modifies a file's ctime without affecting
9841 the others is renaming. In any case, it is not possible, in normal
9842 operations, for a user to change the ctime field to a user-specified value.
9845 Time stamps assume the time zone rules specified by the @env{TZ}
9846 environment variable, or by the system default rules if @env{TZ} is
9847 not set. @xref{TZ Variable,, Specifying the Time Zone with @env{TZ},
9848 libc, The GNU C Library Reference Manual}.
9849 You can avoid ambiguities during
9850 daylight saving transitions by using @sc{utc} time stamps.
9852 The program accepts the following options. Also see @ref{Common options}.
9858 @itemx --time=access
9862 @opindex atime@r{, changing}
9863 @opindex access @r{time, changing}
9864 @opindex use @r{time, changing}
9865 Change the access time only.
9870 @opindex --no-create
9871 Do not create files that do not exist.
9874 @itemx --date=@var{time}
9878 Use @var{time} instead of the current time. It can contain month names,
9879 time zones, @samp{am} and @samp{pm}, @samp{yesterday}, etc. For
9880 example, @option{--date="2004-02-27 14:19:13.489392193 +0530"}
9881 specifies the instant of time that is 489,392,193 nanoseconds after
9882 February 27, 2004 at 2:19:13 PM in a time zone that is 5 hours and 30
9883 minutes east of @acronym{UTC}. @xref{Date input formats}.
9884 File systems that do not support high-resolution time stamps
9885 silently ignore any excess precision here.
9889 @cindex BSD @command{touch} compatibility
9890 Ignored; for compatibility with BSD versions of @command{touch}.
9894 @itemx --time=modify
9897 @opindex mtime@r{, changing}
9898 @opindex modify @r{time, changing}
9899 Change the modification time only.
9902 @itemx --reference=@var{file}
9904 @opindex --reference
9905 Use the times of the reference @var{file} instead of the current time.
9906 If this option is combined with the @option{--date=@var{time}}
9907 (@option{-d @var{time}}) option, the reference @var{file}'s time is
9908 the origin for any relative @var{time}s given, but is otherwise ignored.
9909 For example, @samp{-r foo -d '-5 seconds'} specifies a time stamp
9910 equal to five seconds before the corresponding time stamp for @file{foo}.
9912 @item -t [[@var{cc}]@var{yy}]@var{mmddhhmm}[.@var{ss}]
9913 Use the argument (optional four-digit or two-digit years, months,
9914 days, hours, minutes, optional seconds) instead of the current time.
9915 If the year is specified with only two digits, then @var{cc}
9916 is 20 for years in the range 0 @dots{} 68, and 19 for years in
9917 69 @dots{} 99. If no digits of the year are specified,
9918 the argument is interpreted as a date in the current year.
9922 @vindex _POSIX2_VERSION
9923 On older systems, @command{touch} supports an obsolete syntax, as follows.
9924 If no timestamp is given with any of the @option{-d}, @option{-r}, or
9925 @option{-t} options, and if there are two or more @var{file}s and the
9926 first @var{file} is of the form @samp{@var{mmddhhmm}[@var{yy}]} and this
9927 would be a valid argument to the @option{-t} option (if the @var{yy}, if
9928 any, were moved to the front), and if the represented year
9929 is in the range 1969--1999, that argument is interpreted as the time
9930 for the other files instead of as a file name.
9931 This obsolete behavior can be enabled or disabled with the
9932 @env{_POSIX2_VERSION} environment variable (@pxref{Standards
9933 conformance}), but portable scripts should avoid commands whose
9934 behavior depends on this variable.
9935 For example, use @samp{touch ./12312359 main.c} or @samp{touch -t
9936 12312359 main.c} rather than the ambiguous @samp{touch 12312359 main.c}.
9946 No disk can hold an infinite amount of data. These commands report
9947 how much disk storage is in use or available, report other file and
9948 file status information, and write buffers to disk.
9951 * df invocation:: Report file system disk space usage.
9952 * du invocation:: Estimate file space usage.
9953 * stat invocation:: Report file or file system status.
9954 * sync invocation:: Synchronize memory and disk.
9955 * truncate invocation:: Shrink or extend the size of a file.
9960 @section @command{df}: Report file system disk space usage
9963 @cindex file system disk usage
9964 @cindex disk usage by file system
9966 @command{df} reports the amount of disk space used and available on
9967 file systems. Synopsis:
9970 df [@var{option}]@dots{} [@var{file}]@dots{}
9973 With no arguments, @command{df} reports the space used and available on all
9974 currently mounted file systems (of all types). Otherwise, @command{df}
9975 reports on the file system containing each argument @var{file}.
9977 Normally the disk space is printed in units of
9978 1024 bytes, but this can be overridden (@pxref{Block size}).
9979 Non-integer quantities are rounded up to the next higher unit.
9981 @cindex disk device file
9982 @cindex device file, disk
9983 If an argument @var{file} is a disk device file containing a mounted
9984 file system, @command{df} shows the space available on that file system
9985 rather than on the file system containing the device node (i.e., the root
9986 file system). @sc{gnu} @command{df} does not attempt to determine the disk usage
9987 on unmounted file systems, because on most kinds of systems doing so
9988 requires extremely nonportable intimate knowledge of file system
9991 The program accepts the following options. Also see @ref{Common options}.
9999 @cindex automounter file systems
10000 @cindex ignore file systems
10001 Include in the listing dummy file systems, which
10002 are omitted by default. Such file systems are typically special-purpose
10003 pseudo-file-systems, such as automounter entries.
10005 @item -B @var{size}
10006 @itemx --block-size=@var{size}
10008 @opindex --block-size
10009 @cindex file system sizes
10010 Scale sizes by @var{size} before printing them (@pxref{Block size}).
10011 For example, @option{-BG} prints sizes in units of 1,073,741,824 bytes.
10015 @cindex grand total of disk size, usage and available space
10016 Print a grand total of all arguments after all arguments have
10017 been processed. This can be used to find out the total disk size, usage
10018 and available space of all listed devices.
10024 Equivalent to @option{--si}.
10030 @cindex inode usage
10031 List inode usage information instead of block usage. An inode (short
10032 for index node) contains information about a file such as its owner,
10033 permissions, timestamps, and location on the disk.
10037 @cindex kibibytes for file system sizes
10038 Print sizes in 1024-byte blocks, overriding the default block size
10039 (@pxref{Block size}).
10040 This option is equivalent to @option{--block-size=1K}.
10046 @cindex file system types, limiting output to certain
10047 Limit the listing to local file systems. By default, remote file systems
10052 @cindex file system space, retrieving old data more quickly
10053 Do not invoke the @code{sync} system call before getting any usage data.
10054 This may make @command{df} run significantly faster on systems with many
10055 disks, but on some systems (notably SunOS) the results may be slightly
10056 out of date. This is the default.
10059 @itemx --portability
10061 @opindex --portability
10062 @cindex one-line output format
10063 @cindex @acronym{POSIX} output format
10064 @cindex portable output format
10065 @cindex output format, portable
10066 Use the @acronym{POSIX} output format. This is like the default format except
10071 The information about each file system is always printed on exactly
10072 one line; a mount device is never put on a line by itself. This means
10073 that if the mount device name is more than 20 characters long (e.g., for
10074 some network mounts), the columns are misaligned.
10077 The labels in the header output line are changed to conform to @acronym{POSIX}.
10080 The default block size and output format are unaffected by the
10081 @env{DF_BLOCK_SIZE}, @env{BLOCK_SIZE} and @env{BLOCKSIZE} environment
10082 variables. However, the default block size is still affected by
10083 @env{POSIXLY_CORRECT}: it is 512 if @env{POSIXLY_CORRECT} is set, 1024
10084 otherwise. @xref{Block size}.
10091 @cindex file system space, retrieving current data more slowly
10092 Invoke the @code{sync} system call before getting any usage data. On
10093 some systems (notably SunOS), doing this yields more up to date results,
10094 but in general this option makes @command{df} much slower, especially when
10095 there are many or very busy file systems.
10097 @item -t @var{fstype}
10098 @itemx --type=@var{fstype}
10101 @cindex file system types, limiting output to certain
10102 Limit the listing to file systems of type @var{fstype}. Multiple
10103 file system types can be specified by giving multiple @option{-t} options.
10104 By default, nothing is omitted.
10107 @itemx --print-type
10109 @opindex --print-type
10110 @cindex file system types, printing
10111 Print each file system's type. The types printed here are the same ones
10112 you can include or exclude with @option{-t} and @option{-x}. The particular
10113 types printed are whatever is supported by the system. Here are some of
10114 the common names (this list is certainly not exhaustive):
10119 @cindex @acronym{NFS} file system type
10120 An @acronym{NFS} file system, i.e., one mounted over a network from another
10121 machine. This is the one type name which seems to be used uniformly by
10124 @item 4.2@r{, }ufs@r{, }efs@dots{}
10125 @cindex Linux file system types
10126 @cindex local file system types
10127 @opindex 4.2 @r{file system type}
10128 @opindex ufs @r{file system type}
10129 @opindex efs @r{file system type}
10130 A file system on a locally-mounted hard disk. (The system might even
10131 support more than one type here; Linux does.)
10133 @item hsfs@r{, }cdfs
10134 @cindex CD-ROM file system type
10135 @cindex High Sierra file system
10136 @opindex hsfs @r{file system type}
10137 @opindex cdfs @r{file system type}
10138 A file system on a CD-ROM drive. HP-UX uses @samp{cdfs}, most other
10139 systems use @samp{hsfs} (@samp{hs} for ``High Sierra'').
10142 @cindex PC file system
10143 @cindex DOS file system
10144 @cindex MS-DOS file system
10145 @cindex diskette file system
10147 An MS-DOS file system, usually on a diskette.
10151 @item -x @var{fstype}
10152 @itemx --exclude-type=@var{fstype}
10154 @opindex --exclude-type
10155 Limit the listing to file systems not of type @var{fstype}.
10156 Multiple file system types can be eliminated by giving multiple
10157 @option{-x} options. By default, no file system types are omitted.
10160 Ignored; for compatibility with System V versions of @command{df}.
10165 Failure includes the case where no output is generated, so you can
10166 inspect the exit status of a command like @samp{df -t ext3 -t reiserfs
10167 @var{dir}} to test whether @var{dir} is on a file system of type
10168 @samp{ext3} or @samp{reiserfs}.
10171 @node du invocation
10172 @section @command{du}: Estimate file space usage
10175 @cindex file space usage
10176 @cindex disk usage for files
10178 @command{du} reports the amount of disk space used by the specified files
10179 and for each subdirectory (of directory arguments). Synopsis:
10182 du [@var{option}]@dots{} [@var{file}]@dots{}
10185 With no arguments, @command{du} reports the disk space for the current
10186 directory. Normally the disk space is printed in units of
10187 1024 bytes, but this can be overridden (@pxref{Block size}).
10188 Non-integer quantities are rounded up to the next higher unit.
10190 If two or more hard links point to the same file, only one of the hard
10191 links is counted. The @var{file} argument order affects which links
10192 are counted, and changing the argument order may change the numbers
10193 that @command{du} outputs.
10195 The program accepts the following options. Also see @ref{Common options}.
10203 Show counts for all files, not just directories.
10205 @itemx --apparent-size
10206 @opindex --apparent-size
10207 Print apparent sizes, rather than disk usage. The apparent size of a
10208 file is the number of bytes reported by @code{wc -c} on regular files,
10209 or more generally, @code{ls -l --block-size=1} or @code{stat --format=%s}.
10210 For example, a file containing the word @samp{zoo} with no newline would,
10211 of course, have an apparent size of 3. Such a small file may require
10212 anywhere from 0 to 16 KiB or more of disk space, depending on
10213 the type and configuration of the file system on which the file resides.
10214 However, a sparse file created with this command:
10217 dd bs=1 seek=2GiB if=/dev/null of=big
10221 has an apparent size of 2 GiB, yet on most modern
10222 systems, it actually uses almost no disk space.
10228 Equivalent to @code{--apparent-size --block-size=1}.
10230 @item -B @var{size}
10231 @itemx --block-size=@var{size}
10233 @opindex --block-size
10235 Scale sizes by @var{size} before printing them (@pxref{Block size}).
10236 For example, @option{-BG} prints sizes in units of 1,073,741,824 bytes.
10242 @cindex grand total of disk space
10243 Print a grand total of all arguments after all arguments have
10244 been processed. This can be used to find out the total disk usage of
10245 a given set of files or directories.
10248 @itemx --dereference-args
10250 @opindex --dereference-args
10251 Dereference symbolic links that are command line arguments.
10252 Does not affect other symbolic links. This is helpful for finding
10253 out the disk usage of directories, such as @file{/usr/tmp}, which
10254 are often symbolic links.
10256 @c --files0-from=FILE
10257 @filesZeroFromOption{du,, with the @option{--total} (@option{-c}) option}
10263 Equivalent to @option{--dereference-args} (@option{-D}).
10267 @cindex kibibytes for file sizes
10268 Print sizes in 1024-byte blocks, overriding the default block size
10269 (@pxref{Block size}).
10270 This option is equivalent to @option{--block-size=1K}.
10273 @itemx --count-links
10275 @opindex --count-links
10276 @cindex hard links, counting in @command{du}
10277 Count the size of all files, even if they have appeared already (as a
10281 @itemx --dereference
10283 @opindex --dereference
10284 @cindex symbolic links, dereferencing in @command{du}
10285 Dereference symbolic links (show the disk space used by the file
10286 or directory that the link points to instead of the space used by
10291 @cindex mebibytes for file sizes
10292 Print sizes in 1,048,576-byte blocks, overriding the default block size
10293 (@pxref{Block size}).
10294 This option is equivalent to @option{--block-size=1M}.
10297 @itemx --no-dereference
10299 @opindex --no-dereference
10300 @cindex symbolic links, dereferencing in @command{du}
10301 For each symbolic links encountered by @command{du},
10302 consider the disk space used by the symbolic link.
10304 @item --max-depth=@var{depth}
10305 @opindex --max-depth=@var{depth}
10306 @cindex limiting output of @command{du}
10307 Show the total for each directory (and file if --all) that is at
10308 most MAX_DEPTH levels down from the root of the hierarchy. The root
10309 is at level 0, so @code{du --max-depth=0} is equivalent to @code{du -s}.
10315 @cindex output null-byte-terminated lines
10316 Output a zero byte (@acronym{ASCII} @sc{nul}) at the end of each line,
10317 rather than a newline. This option enables other programs to parse the
10318 output of @command{du} even when that output would contain file names
10319 with embedded newlines.
10326 @opindex --summarize
10327 Display only a total for each argument.
10330 @itemx --separate-dirs
10332 @opindex --separate-dirs
10333 Normally, in the output of @command{du} (when not using @option{--summarize}),
10334 the size listed next to a directory name, @var{d}, represents the sum
10335 of sizes of all entries beneath @var{d} as well as the size of @var{d} itself.
10336 With @option{--separate-dirs}, the size reported for a directory name,
10337 @var{d}, is merely the @code{stat.st_size}-derived size of the directory
10342 @cindex last modified dates, displaying in @command{du}
10343 Show time of the most recent modification of any file in the directory,
10344 or any of its subdirectories.
10346 @itemx --time=ctime
10347 @itemx --time=status
10350 @opindex ctime@r{, show the most recent}
10351 @opindex status time@r{, show the most recent}
10352 @opindex use time@r{, show the most recent}
10353 Show the most recent status change time (the @samp{ctime} in the inode) of
10354 any file in the directory, instead of the modification time.
10356 @itemx --time=atime
10357 @itemx --time=access
10359 @opindex atime@r{, show the most recent}
10360 @opindex access time@r{, show the most recent}
10361 Show the most recent access time (the @samp{atime} in the inode) of
10362 any file in the directory, instead of the modification time.
10364 @item --time-style=@var{style}
10365 @opindex --time-style
10367 List timestamps in style @var{style}. This option has an effect only if
10368 the @option{--time} option is also specified. The @var{style} should
10369 be one of the following:
10372 @item +@var{format}
10374 List timestamps using @var{format}, where @var{format} is interpreted
10375 like the format argument of @command{date} (@pxref{date invocation}).
10376 For example, @option{--time-style="+%Y-%m-%d %H:%M:%S"} causes
10377 @command{du} to list timestamps like @samp{2002-03-30 23:45:56}. As
10378 with @command{date}, @var{format}'s interpretation is affected by the
10379 @env{LC_TIME} locale category.
10382 List timestamps in full using @acronym{ISO} 8601 date, time, and time zone
10383 format with nanosecond precision, e.g., @samp{2002-03-30
10384 23:45:56.477817180 -0700}. This style is equivalent to
10385 @samp{+%Y-%m-%d %H:%M:%S.%N %z}.
10388 List @acronym{ISO} 8601 date and time in minutes, e.g.,
10389 @samp{2002-03-30 23:45}. These timestamps are shorter than
10390 @samp{full-iso} timestamps, and are usually good enough for everyday
10391 work. This style is equivalent to @samp{+%Y-%m-%d %H:%M}.
10394 List @acronym{ISO} 8601 dates for timestamps, e.g., @samp{2002-03-30}.
10395 This style is equivalent to @samp{+%Y-%m-%d}.
10399 You can specify the default value of the @option{--time-style} option
10400 with the environment variable @env{TIME_STYLE}; if @env{TIME_STYLE} is not set
10401 the default style is @samp{long-iso}. For compatibility with @command{ls},
10402 if @env{TIME_STYLE} begins with @samp{+} and contains a newline,
10403 the newline and any later characters are ignored; if @env{TIME_STYLE}
10404 begins with @samp{posix-} the @samp{posix-} is ignored; and if
10405 @env{TIME_STYLE} is @samp{locale} it is ignored.
10408 @itemx --one-file-system
10410 @opindex --one-file-system
10411 @cindex one file system, restricting @command{du} to
10412 Skip directories that are on different file systems from the one that
10413 the argument being processed is on.
10415 @item --exclude=@var{pattern}
10416 @opindex --exclude=@var{pattern}
10417 @cindex excluding files from @command{du}
10418 When recursing, skip subdirectories or files matching @var{pattern}.
10419 For example, @code{du --exclude='*.o'} excludes files whose names
10422 @item -X @var{file}
10423 @itemx --exclude-from=@var{file}
10424 @opindex -X @var{file}
10425 @opindex --exclude-from=@var{file}
10426 @cindex excluding files from @command{du}
10427 Like @option{--exclude}, except take the patterns to exclude from @var{file},
10428 one per line. If @var{file} is @samp{-}, take the patterns from standard
10433 @cindex NFS mounts from BSD to HP-UX
10434 On BSD systems, @command{du} reports sizes that are half the correct
10435 values for files that are NFS-mounted from HP-UX systems. On HP-UX
10436 systems, it reports sizes that are twice the correct values for
10437 files that are NFS-mounted from BSD systems. This is due to a flaw
10438 in HP-UX; it also affects the HP-UX @command{du} program.
10443 @node stat invocation
10444 @section @command{stat}: Report file or file system status
10447 @cindex file status
10448 @cindex file system status
10450 @command{stat} displays information about the specified file(s). Synopsis:
10453 stat [@var{option}]@dots{} [@var{file}]@dots{}
10456 With no option, @command{stat} reports all information about the given files.
10457 But it also can be used to report the information of the file systems the
10458 given files are located on. If the files are links, @command{stat} can
10459 also give information about the files the links point to.
10461 @mayConflictWithShellBuiltIn{stat}
10466 @itemx --dereference
10468 @opindex --dereference
10469 @cindex symbolic links, dereferencing in @command{stat}
10470 Change how @command{stat} treats symbolic links.
10471 With this option, @command{stat} acts on the file referenced
10472 by each symbolic link argument.
10473 Without it, @command{stat} acts on any symbolic link argument directly.
10476 @itemx --file-system
10478 @opindex --file-system
10479 @cindex file systems
10480 Report information about the file systems where the given files are located
10481 instead of information about the files themselves.
10484 @itemx --format=@var{format}
10486 @opindex --format=@var{format}
10487 @cindex output format
10488 Use @var{format} rather than the default format.
10489 @var{format} is automatically newline-terminated, so
10490 running a command like the following with two or more @var{file}
10491 operands produces a line of output for each operand:
10493 $ stat --format=%d:%i / /usr
10498 @itemx --printf=@var{format}
10499 @opindex --printf=@var{format}
10500 @cindex output format
10501 Use @var{format} rather than the default format.
10502 Like @option{--format}, but interpret backslash escapes,
10503 and do not output a mandatory trailing newline.
10504 If you want a newline, include @samp{\n} in the @var{format}.
10505 Here's how you would use @option{--printf} to print the device
10506 and inode numbers of @file{/} and @file{/usr}:
10508 $ stat --printf='%d:%i\n' / /usr
10517 @cindex terse output
10518 Print the information in terse form, suitable for parsing by other programs.
10522 The valid @var{format} directives for files with @option{--format} and
10523 @option{--printf} are:
10526 @item %a - Access rights in octal
10527 @item %A - Access rights in human readable form
10528 @item %b - Number of blocks allocated (see @samp{%B})
10529 @item %B - The size in bytes of each block reported by @samp{%b}
10530 @item %d - Device number in decimal
10531 @item %D - Device number in hex
10532 @item %f - Raw mode in hex
10533 @item %F - File type
10534 @item %g - Group ID of owner
10535 @item %G - Group name of owner
10536 @item %h - Number of hard links
10537 @item %i - Inode number
10538 @item %n - File name
10539 @item %N - Quoted file name with dereference if symbolic link
10540 @item %o - I/O block size
10541 @item %s - Total size, in bytes
10542 @item %t - Major device type in hex
10543 @item %T - Minor device type in hex
10544 @item %u - User ID of owner
10545 @item %U - User name of owner
10546 @item %x - Time of last access
10547 @item %X - Time of last access as seconds since Epoch
10548 @item %y - Time of last modification
10549 @item %Y - Time of last modification as seconds since Epoch
10550 @item %z - Time of last change
10551 @item %Z - Time of last change as seconds since Epoch
10554 When listing file system information (@option{--file-system} (@option{-f})),
10555 you must use a different set of @var{format} directives:
10558 @item %a - Free blocks available to non-super-user
10559 @item %b - Total data blocks in file system
10560 @item %c - Total file nodes in file system
10561 @item %d - Free file nodes in file system
10562 @item %f - Free blocks in file system
10563 @item %i - File System ID in hex
10564 @item %l - Maximum length of file names
10565 @item %n - File name
10566 @item %s - Block size (for faster transfers)
10567 @item %S - Fundamental block size (for block counts)
10568 @item %t - Type in hex
10569 @item %T - Type in human readable form
10573 Time stamps are listed according to the time zone rules specified by
10574 the @env{TZ} environment variable, or by the system default rules if
10575 @env{TZ} is not set. @xref{TZ Variable,, Specifying the Time Zone
10576 with @env{TZ}, libc, The GNU C Library Reference Manual}.
10581 @node sync invocation
10582 @section @command{sync}: Synchronize data on disk with memory
10585 @cindex synchronize disk and memory
10587 @cindex superblock, writing
10588 @cindex inodes, written buffered
10589 @command{sync} writes any data buffered in memory out to disk. This can
10590 include (but is not limited to) modified superblocks, modified inodes,
10591 and delayed reads and writes. This must be implemented by the kernel;
10592 The @command{sync} program does nothing but exercise the @code{sync} system
10595 @cindex crashes and corruption
10596 The kernel keeps data in memory to avoid doing (relatively slow) disk
10597 reads and writes. This improves performance, but if the computer
10598 crashes, data may be lost or the file system corrupted as a
10599 result. The @command{sync} command ensures everything in memory
10600 is written to disk.
10602 Any arguments are ignored, except for a lone @option{--help} or
10603 @option{--version} (@pxref{Common options}).
10608 @node truncate invocation
10609 @section @command{truncate}: Shrink or extend the size of a file
10612 @cindex truncating, file sizes
10614 @command{truncate} shrinks or extends the size of each @var{file} to the
10615 specified size. Synopsis:
10618 truncate @var{option}@dots{} @var{file}@dots{}
10621 @cindex files, creating
10622 Any @var{file} that does not exist is created.
10624 @cindex sparse files, creating
10625 @cindex holes, creating files with
10626 If a @var{file} is larger than the specified size, the extra data is lost.
10627 If a @var{file} is shorter, it is extended and the extended part (or hole)
10628 reads as zero bytes.
10630 The program accepts the following options. Also see @ref{Common options}.
10637 @opindex --no-create
10638 Do not create files that do not exist.
10643 @opindex --io-blocks
10644 Treat @var{size} as number of I/O blocks of the @var{file} rather than bytes.
10646 @item -r @var{rfile}
10647 @itemx --reference=@var{rfile}
10649 @opindex --reference
10650 Set the size of each @var{file} to the same size as @var{rfile}.
10652 @item -s @var{size}
10653 @itemx --size=@var{size}
10656 Set the size of each @var{file} to this @var{size}.
10657 @multiplierSuffixesNoBlocks{size}
10659 @var{size} may also be prefixed by one of the following to adjust
10660 the size of each @var{file} based on their current size:
10662 @samp{+} => extend by
10663 @samp{-} => reduce by
10664 @samp{<} => at most
10665 @samp{>} => at least
10666 @samp{/} => round down to multiple of
10667 @samp{%} => round up to multiple of
10675 @node Printing text
10676 @chapter Printing text
10678 @cindex printing text, commands for
10679 @cindex commands for printing text
10681 This section describes commands that display text strings.
10684 * echo invocation:: Print a line of text.
10685 * printf invocation:: Format and print data.
10686 * yes invocation:: Print a string until interrupted.
10690 @node echo invocation
10691 @section @command{echo}: Print a line of text
10694 @cindex displaying text
10695 @cindex printing text
10696 @cindex text, displaying
10697 @cindex arbitrary text, displaying
10699 @command{echo} writes each given @var{string} to standard output, with a
10700 space between each and a newline after the last one. Synopsis:
10703 echo [@var{option}]@dots{} [@var{string}]@dots{}
10706 @mayConflictWithShellBuiltIn{echo}
10708 The program accepts the following options. Also see @ref{Common options}.
10709 Options must precede operands, and the normally-special argument
10710 @samp{--} has no special meaning and is treated like any other
10716 Do not output the trailing newline.
10720 @cindex backslash escapes
10721 Enable interpretation of the following backslash-escaped characters in
10730 produce no further output
10744 the eight-bit value that is the octal number @var{nnn}
10745 (zero to three octal digits)
10747 the eight-bit value that is the octal number @var{nnn}
10748 (one to three octal digits)
10750 the eight-bit value that is the hexadecimal number @var{hh}
10751 (one or two hexadecimal digits)
10756 @cindex backslash escapes
10757 Disable interpretation of backslash escapes in each @var{string}.
10758 This is the default. If @option{-e} and @option{-E} are both
10759 specified, the last one given takes effect.
10763 @vindex POSIXLY_CORRECT
10764 If the @env{POSIXLY_CORRECT} environment variable is set, then when
10765 @command{echo}'s first argument is not @option{-n} it outputs
10766 option-like arguments instead of treating them as options. For
10767 example, @code{echo -ne hello} outputs @samp{-ne hello} instead of
10768 plain @samp{hello}.
10770 @acronym{POSIX} does not require support for any options, and says
10771 that the behavior of @command{echo} is implementation-defined if any
10772 @var{string} contains a backslash or if the first argument is
10773 @option{-n}. Portable programs can use the @command{printf} command
10774 if they need to omit trailing newlines or output control characters or
10775 backslashes. @xref{printf invocation}.
10780 @node printf invocation
10781 @section @command{printf}: Format and print data
10784 @command{printf} does formatted printing of text. Synopsis:
10787 printf @var{format} [@var{argument}]@dots{}
10790 @command{printf} prints the @var{format} string, interpreting @samp{%}
10791 directives and @samp{\} escapes to format numeric and string arguments
10792 in a way that is mostly similar to the C @samp{printf} function.
10793 @xref{Output Conversion Syntax,, @command{printf} format directives,
10794 libc, The GNU C Library Reference Manual}, for details.
10795 The differences are listed below.
10797 @mayConflictWithShellBuiltIn{printf}
10802 The @var{format} argument is reused as necessary to convert all the
10803 given @var{argument}s. For example, the command @samp{printf %s a b}
10807 Missing @var{argument}s are treated as null strings or as zeros,
10808 depending on whether the context expects a string or a number. For
10809 example, the command @samp{printf %sx%d} prints @samp{x0}.
10813 An additional escape, @samp{\c}, causes @command{printf} to produce no
10814 further output. For example, the command @samp{printf 'A%sC\cD%sF' B
10815 E} prints @samp{ABC}.
10818 The hexadecimal escape sequence @samp{\x@var{hh}} has at most two
10819 digits, as opposed to C where it can have an unlimited number of
10820 digits. For example, the command @samp{printf '\x07e'} prints two
10821 bytes, whereas the C statement @samp{printf ("\x07e")} prints just
10826 @command{printf} has an additional directive, @samp{%b}, which prints its
10827 argument string with @samp{\} escapes interpreted in the same way as in
10828 the @var{format} string, except that octal escapes are of the form
10829 @samp{\0@var{ooo}} where @var{ooo} is 0 to 3 octal digits.
10830 If a precision is also given, it limits the number of bytes printed
10831 from the converted string.
10834 Numeric arguments must be single C constants, possibly with leading
10835 @samp{+} or @samp{-}. For example, @samp{printf %.4d -3} outputs
10839 @vindex POSIXLY_CORRECT
10840 If the leading character of a numeric argument is @samp{"} or @samp{'}
10841 then its value is the numeric value of the immediately following
10842 character. Any remaining characters are silently ignored if the
10843 @env{POSIXLY_CORRECT} environment variable is set; otherwise, a
10844 warning is printed. For example, @samp{printf "%d" "'a"} outputs
10845 @samp{97} on hosts that use the @acronym{ASCII} character set, since
10846 @samp{a} has the numeric value 97 in @acronym{ASCII}.
10851 A floating-point argument must use a period before any fractional
10852 digits, but is printed according to the @env{LC_NUMERIC} category of the
10853 current locale. For example, in a locale whose radix character is a
10854 comma, the command @samp{printf %g 3.14} outputs @samp{3,14} whereas
10855 the command @samp{printf %g 3,14} is an error.
10859 @command{printf} interprets @samp{\@var{ooo}} in @var{format} as an octal number
10860 (if @var{ooo} is 1 to 3 octal digits) specifying a character to print,
10861 and @samp{\x@var{hh}} as a hexadecimal number (if @var{hh} is 1 to 2 hex
10862 digits) specifying a character to print.
10867 @cindex ISO/IEC 10646
10869 @command{printf} interprets two character syntaxes introduced in
10870 @acronym{ISO} C 99:
10871 @samp{\u} for 16-bit Unicode (@acronym{ISO}/@acronym{IEC} 10646)
10872 characters, specified as
10873 four hexadecimal digits @var{hhhh}, and @samp{\U} for 32-bit Unicode
10874 characters, specified as eight hexadecimal digits @var{hhhhhhhh}.
10875 @command{printf} outputs the Unicode characters
10876 according to the @env{LC_CTYPE} locale. Unicode characters in the ranges
10877 U+0000...U+009F, U+D800...U+DFFF cannot be specified by this syntax, except
10878 for U+0024 ($), U+0040 (@@), and U+0060 (@`).
10880 The processing of @samp{\u} and @samp{\U} requires a full-featured
10881 @code{iconv} facility. It is activated on systems with glibc 2.2 (or newer),
10882 or when @code{libiconv} is installed prior to this package. Otherwise
10883 @samp{\u} and @samp{\U} will print as-is.
10885 The only options are a lone @option{--help} or
10886 @option{--version}. @xref{Common options}.
10887 Options must precede operands.
10889 The Unicode character syntaxes are useful for writing strings in a locale
10890 independent way. For example, a string containing the Euro currency symbol
10893 $ env printf '\u20AC 14.95'
10897 will be output correctly in all locales supporting the Euro symbol
10898 (@acronym{ISO}-8859-15, UTF-8, and others). Similarly, a Chinese string
10901 $ env printf '\u4e2d\u6587'
10905 will be output correctly in all Chinese locales (GB2312, BIG5, UTF-8, etc).
10907 Note that in these examples, the @command{printf} command has been
10908 invoked via @command{env} to ensure that we run the program found via
10909 your shell's search path, and not a shell alias or a built-in function.
10911 For larger strings, you don't need to look up the hexadecimal code
10912 values of each character one by one. @acronym{ASCII} characters mixed with \u
10913 escape sequences is also known as the JAVA source file encoding. You can
10914 use GNU recode 3.5c (or newer) to convert strings to this encoding. Here
10915 is how to convert a piece of text into a shell script which will output
10916 this text in a locale-independent way:
10919 $ LC_CTYPE=zh_CN.big5 /usr/local/bin/printf \
10920 '\u4e2d\u6587\n' > sample.txt
10921 $ recode BIG5..JAVA < sample.txt \
10922 | sed -e "s|^|/usr/local/bin/printf '|" -e "s|$|\\\\n'|" \
10929 @node yes invocation
10930 @section @command{yes}: Print a string until interrupted
10933 @cindex repeated output of a string
10935 @command{yes} prints the command line arguments, separated by spaces and
10936 followed by a newline, forever until it is killed. If no arguments are
10937 given, it prints @samp{y} followed by a newline forever until killed.
10939 Upon a write error, @command{yes} exits with status @samp{1}.
10941 The only options are a lone @option{--help} or @option{--version}.
10942 To output an argument that begins with
10943 @samp{-}, precede it with @option{--}, e.g., @samp{yes -- --help}.
10944 @xref{Common options}.
10948 @chapter Conditions
10951 @cindex commands for exit status
10952 @cindex exit status commands
10954 This section describes commands that are primarily useful for their exit
10955 status, rather than their output. Thus, they are often used as the
10956 condition of shell @code{if} statements, or as the last command in a
10960 * false invocation:: Do nothing, unsuccessfully.
10961 * true invocation:: Do nothing, successfully.
10962 * test invocation:: Check file types and compare values.
10963 * expr invocation:: Evaluate expressions.
10967 @node false invocation
10968 @section @command{false}: Do nothing, unsuccessfully
10971 @cindex do nothing, unsuccessfully
10972 @cindex failure exit status
10973 @cindex exit status of @command{false}
10975 @command{false} does nothing except return an exit status of 1, meaning
10976 @dfn{failure}. It can be used as a place holder in shell scripts
10977 where an unsuccessful command is needed.
10978 In most modern shells, @command{false} is a built-in command, so when
10979 you use @samp{false} in a script, you're probably using the built-in
10980 command, not the one documented here.
10982 @command{false} honors the @option{--help} and @option{--version} options.
10984 This version of @command{false} is implemented as a C program, and is thus
10985 more secure and faster than a shell script implementation, and may safely
10986 be used as a dummy shell for the purpose of disabling accounts.
10988 Note that @command{false} (unlike all other programs documented herein)
10989 exits unsuccessfully, even when invoked with
10990 @option{--help} or @option{--version}.
10992 Portable programs should not assume that the exit status of
10993 @command{false} is 1, as it is greater than 1 on some
10994 non-@acronym{GNU} hosts.
10997 @node true invocation
10998 @section @command{true}: Do nothing, successfully
11001 @cindex do nothing, successfully
11003 @cindex successful exit
11004 @cindex exit status of @command{true}
11006 @command{true} does nothing except return an exit status of 0, meaning
11007 @dfn{success}. It can be used as a place holder in shell scripts
11008 where a successful command is needed, although the shell built-in
11009 command @code{:} (colon) may do the same thing faster.
11010 In most modern shells, @command{true} is a built-in command, so when
11011 you use @samp{true} in a script, you're probably using the built-in
11012 command, not the one documented here.
11014 @command{true} honors the @option{--help} and @option{--version} options.
11016 Note, however, that it is possible to cause @command{true}
11017 to exit with nonzero status: with the @option{--help} or @option{--version}
11018 option, and with standard
11019 output already closed or redirected to a file that evokes an I/O error.
11020 For example, using a Bourne-compatible shell:
11023 $ ./true --version >&-
11024 ./true: write error: Bad file number
11025 $ ./true --version > /dev/full
11026 ./true: write error: No space left on device
11029 This version of @command{true} is implemented as a C program, and is thus
11030 more secure and faster than a shell script implementation, and may safely
11031 be used as a dummy shell for the purpose of disabling accounts.
11033 @node test invocation
11034 @section @command{test}: Check file types and compare values
11037 @cindex check file types
11038 @cindex compare values
11039 @cindex expression evaluation
11041 @command{test} returns a status of 0 (true) or 1 (false) depending on the
11042 evaluation of the conditional expression @var{expr}. Each part of the
11043 expression must be a separate argument.
11045 @command{test} has file status checks, string operators, and numeric
11046 comparison operators.
11048 @command{test} has an alternate form that uses opening and closing
11049 square brackets instead a leading @samp{test}. For example, instead
11050 of @samp{test -d /}, you can write @samp{[ -d / ]}. The square
11051 brackets must be separate arguments; for example, @samp{[-d /]} does
11052 not have the desired effect. Since @samp{test @var{expr}} and @samp{[
11053 @var{expr} ]} have the same meaning, only the former form is discussed
11059 test @var{expression}
11061 [ @var{expression} ]
11066 @mayConflictWithShellBuiltIn{test}
11068 If @var{expression} is omitted, @command{test} returns false.
11069 If @var{expression} is a single argument,
11070 @command{test} returns false if the argument is null and true otherwise. The argument
11071 can be any string, including strings like @samp{-d}, @samp{-1},
11072 @samp{--}, @samp{--help}, and @samp{--version} that most other
11073 programs would treat as options. To get help and version information,
11074 invoke the commands @samp{[ --help} and @samp{[ --version}, without
11075 the usual closing brackets. @xref{Common options}.
11077 @cindex exit status of @command{test}
11081 0 if the expression is true,
11082 1 if the expression is false,
11083 2 if an error occurred.
11087 * File type tests:: -[bcdfhLpSt]
11088 * Access permission tests:: -[gkruwxOG]
11089 * File characteristic tests:: -e -s -nt -ot -ef
11090 * String tests:: -z -n = !=
11091 * Numeric tests:: -eq -ne -lt -le -gt -ge
11092 * Connectives for test:: ! -a -o
11096 @node File type tests
11097 @subsection File type tests
11099 @cindex file type tests
11101 These options test for particular types of files. (Everything's a file,
11102 but not all files are the same!)
11106 @item -b @var{file}
11108 @cindex block special check
11109 True if @var{file} exists and is a block special device.
11111 @item -c @var{file}
11113 @cindex character special check
11114 True if @var{file} exists and is a character special device.
11116 @item -d @var{file}
11118 @cindex directory check
11119 True if @var{file} exists and is a directory.
11121 @item -f @var{file}
11123 @cindex regular file check
11124 True if @var{file} exists and is a regular file.
11126 @item -h @var{file}
11127 @itemx -L @var{file}
11130 @cindex symbolic link check
11131 True if @var{file} exists and is a symbolic link.
11132 Unlike all other file-related tests, this test does not dereference
11133 @var{file} if it is a symbolic link.
11135 @item -p @var{file}
11137 @cindex named pipe check
11138 True if @var{file} exists and is a named pipe.
11140 @item -S @var{file}
11142 @cindex socket check
11143 True if @var{file} exists and is a socket.
11147 @cindex terminal check
11148 True if @var{fd} is a file descriptor that is associated with a
11154 @node Access permission tests
11155 @subsection Access permission tests
11157 @cindex access permission tests
11158 @cindex permission tests
11160 These options test for particular access permissions.
11164 @item -g @var{file}
11166 @cindex set-group-ID check
11167 True if @var{file} exists and has its set-group-ID bit set.
11169 @item -k @var{file}
11171 @cindex sticky bit check
11172 True if @var{file} exists and has its @dfn{sticky} bit set.
11174 @item -r @var{file}
11176 @cindex readable file check
11177 True if @var{file} exists and read permission is granted.
11179 @item -u @var{file}
11181 @cindex set-user-ID check
11182 True if @var{file} exists and has its set-user-ID bit set.
11184 @item -w @var{file}
11186 @cindex writable file check
11187 True if @var{file} exists and write permission is granted.
11189 @item -x @var{file}
11191 @cindex executable file check
11192 True if @var{file} exists and execute permission is granted
11193 (or search permission, if it is a directory).
11195 @item -O @var{file}
11197 @cindex owned by effective user ID check
11198 True if @var{file} exists and is owned by the current effective user ID.
11200 @item -G @var{file}
11202 @cindex owned by effective group ID check
11203 True if @var{file} exists and is owned by the current effective group ID.
11207 @node File characteristic tests
11208 @subsection File characteristic tests
11210 @cindex file characteristic tests
11212 These options test other file characteristics.
11216 @item -e @var{file}
11218 @cindex existence-of-file check
11219 True if @var{file} exists.
11221 @item -s @var{file}
11223 @cindex nonempty file check
11224 True if @var{file} exists and has a size greater than zero.
11226 @item @var{file1} -nt @var{file2}
11228 @cindex newer-than file check
11229 True if @var{file1} is newer (according to modification date) than
11230 @var{file2}, or if @var{file1} exists and @var{file2} does not.
11232 @item @var{file1} -ot @var{file2}
11234 @cindex older-than file check
11235 True if @var{file1} is older (according to modification date) than
11236 @var{file2}, or if @var{file2} exists and @var{file1} does not.
11238 @item @var{file1} -ef @var{file2}
11240 @cindex same file check
11241 @cindex hard link check
11242 True if @var{file1} and @var{file2} have the same device and inode
11243 numbers, i.e., if they are hard links to each other.
11249 @subsection String tests
11251 @cindex string tests
11253 These options test string characteristics. You may need to quote
11254 @var{string} arguments for the shell. For example:
11260 The quotes here prevent the wrong arguments from being passed to
11261 @command{test} if @samp{$V} is empty or contains special characters.
11265 @item -z @var{string}
11267 @cindex zero-length string check
11268 True if the length of @var{string} is zero.
11270 @item -n @var{string}
11271 @itemx @var{string}
11273 @cindex nonzero-length string check
11274 True if the length of @var{string} is nonzero.
11276 @item @var{string1} = @var{string2}
11278 @cindex equal string check
11279 True if the strings are equal.
11281 @item @var{string1} != @var{string2}
11283 @cindex not-equal string check
11284 True if the strings are not equal.
11289 @node Numeric tests
11290 @subsection Numeric tests
11292 @cindex numeric tests
11293 @cindex arithmetic tests
11295 Numeric relational operators. The arguments must be entirely numeric
11296 (possibly negative), or the special expression @w{@code{-l @var{string}}},
11297 which evaluates to the length of @var{string}.
11301 @item @var{arg1} -eq @var{arg2}
11302 @itemx @var{arg1} -ne @var{arg2}
11303 @itemx @var{arg1} -lt @var{arg2}
11304 @itemx @var{arg1} -le @var{arg2}
11305 @itemx @var{arg1} -gt @var{arg2}
11306 @itemx @var{arg1} -ge @var{arg2}
11313 These arithmetic binary operators return true if @var{arg1} is equal,
11314 not-equal, less-than, less-than-or-equal, greater-than, or
11315 greater-than-or-equal than @var{arg2}, respectively.
11322 test -1 -gt -2 && echo yes
11324 test -l abc -gt 1 && echo yes
11327 @error{} test: integer expression expected before -eq
11331 @node Connectives for test
11332 @subsection Connectives for @command{test}
11334 @cindex logical connectives
11335 @cindex connectives, logical
11337 The usual logical connectives.
11343 True if @var{expr} is false.
11345 @item @var{expr1} -a @var{expr2}
11347 @cindex logical and operator
11348 @cindex and operator
11349 True if both @var{expr1} and @var{expr2} are true.
11351 @item @var{expr1} -o @var{expr2}
11353 @cindex logical or operator
11354 @cindex or operator
11355 True if either @var{expr1} or @var{expr2} is true.
11360 @node expr invocation
11361 @section @command{expr}: Evaluate expressions
11364 @cindex expression evaluation
11365 @cindex evaluation of expressions
11367 @command{expr} evaluates an expression and writes the result on standard
11368 output. Each token of the expression must be a separate argument.
11370 Operands are either integers or strings. Integers consist of one or
11371 more decimal digits, with an optional leading @samp{-}.
11372 @command{expr} converts
11373 anything appearing in an operand position to an integer or a string
11374 depending on the operation being applied to it.
11376 Strings are not quoted for @command{expr} itself, though you may need to
11377 quote them to protect characters with special meaning to the shell,
11378 e.g., spaces. However, regardless of whether it is quoted, a string
11379 operand should not be a parenthesis or any of @command{expr}'s
11380 operators like @code{+}, so you cannot safely pass an arbitrary string
11381 @code{$str} to expr merely by quoting it to the shell. One way to
11382 work around this is to use the @sc{gnu} extension @code{+},
11383 (e.g., @code{+ "$str" = foo}); a more portable way is to use
11384 @code{@w{" $str"}} and to adjust the rest of the expression to take
11385 the leading space into account (e.g., @code{@w{" $str" = " foo"}}).
11387 You should not pass a negative integer or a string with leading
11388 @samp{-} as @command{expr}'s first argument, as it might be
11389 misinterpreted as an option; this can be avoided by parenthesization.
11390 Also, portable scripts should not use a string operand that happens to
11391 take the form of an integer; this can be worked around by inserting
11392 leading spaces as mentioned above.
11394 @cindex parentheses for grouping
11395 Operators may be given as infix symbols or prefix keywords. Parentheses
11396 may be used for grouping in the usual manner. You must quote
11397 parentheses and many operators to avoid the shell evaluating them,
11400 When built with support for the GNU MP library, @command{expr} uses
11401 arbitrary-precision arithmetic; otherwise, it uses native arithmetic
11402 types and may fail due to arithmetic overflow.
11404 The only options are @option{--help} and @option{--version}. @xref{Common
11405 options}. Options must precede operands.
11407 @cindex exit status of @command{expr}
11411 0 if the expression is neither null nor 0,
11412 1 if the expression is null or 0,
11413 2 if the expression is invalid,
11414 3 if an internal error occurred (e.g., arithmetic overflow).
11418 * String expressions:: + : match substr index length
11419 * Numeric expressions:: + - * / %
11420 * Relations for expr:: | & < <= = == != >= >
11421 * Examples of expr:: Examples.
11425 @node String expressions
11426 @subsection String expressions
11428 @cindex string expressions
11429 @cindex expressions, string
11431 @command{expr} supports pattern matching and other string operators. These
11432 have higher precedence than both the numeric and relational operators (in
11433 the next sections).
11437 @item @var{string} : @var{regex}
11438 @cindex pattern matching
11439 @cindex regular expression matching
11440 @cindex matching patterns
11441 Perform pattern matching. The arguments are converted to strings and the
11442 second is considered to be a (basic, a la GNU @code{grep}) regular
11443 expression, with a @code{^} implicitly prepended. The first argument is
11444 then matched against this regular expression.
11446 If the match succeeds and @var{regex} uses @samp{\(} and @samp{\)}, the
11447 @code{:} expression returns the part of @var{string} that matched the
11448 subexpression; otherwise, it returns the number of characters matched.
11450 If the match fails, the @code{:} operator returns the null string if
11451 @samp{\(} and @samp{\)} are used in @var{regex}, otherwise 0.
11453 @kindex \( @r{regexp operator}
11454 Only the first @samp{\( @dots{} \)} pair is relevant to the return
11455 value; additional pairs are meaningful only for grouping the regular
11456 expression operators.
11458 @kindex \+ @r{regexp operator}
11459 @kindex \? @r{regexp operator}
11460 @kindex \| @r{regexp operator}
11461 In the regular expression, @code{\+}, @code{\?}, and @code{\|} are
11462 operators which respectively match one or more, zero or one, or separate
11463 alternatives. SunOS and other @command{expr}'s treat these as regular
11464 characters. (@acronym{POSIX} allows either behavior.)
11465 @xref{Top, , Regular Expression Library, regex, Regex}, for details of
11466 regular expression syntax. Some examples are in @ref{Examples of expr}.
11468 @item match @var{string} @var{regex}
11470 An alternative way to do pattern matching. This is the same as
11471 @w{@samp{@var{string} : @var{regex}}}.
11473 @item substr @var{string} @var{position} @var{length}
11475 Returns the substring of @var{string} beginning at @var{position}
11476 with length at most @var{length}. If either @var{position} or
11477 @var{length} is negative, zero, or non-numeric, returns the null string.
11479 @item index @var{string} @var{charset}
11481 Returns the first position in @var{string} where the first character in
11482 @var{charset} was found. If no character in @var{charset} is found in
11483 @var{string}, return 0.
11485 @item length @var{string}
11487 Returns the length of @var{string}.
11489 @item + @var{token}
11491 Interpret @var{token} as a string, even if it is a keyword like @var{match}
11492 or an operator like @code{/}.
11493 This makes it possible to test @code{expr length + "$x"} or
11494 @code{expr + "$x" : '.*/\(.\)'} and have it do the right thing even if
11495 the value of @var{$x} happens to be (for example) @code{/} or @code{index}.
11496 This operator is a @acronym{GNU} extension. Portable shell scripts should use
11497 @code{@w{" $token"} : @w{' \(.*\)'}} instead of @code{+ "$token"}.
11501 To make @command{expr} interpret keywords as strings, you must use the
11502 @code{quote} operator.
11505 @node Numeric expressions
11506 @subsection Numeric expressions
11508 @cindex numeric expressions
11509 @cindex expressions, numeric
11511 @command{expr} supports the usual numeric operators, in order of increasing
11512 precedence. These numeric operators have lower precedence than the
11513 string operators described in the previous section, and higher precedence
11514 than the connectives (next section).
11522 @cindex subtraction
11523 Addition and subtraction. Both arguments are converted to integers;
11524 an error occurs if this cannot be done.
11530 @cindex multiplication
11533 Multiplication, division, remainder. Both arguments are converted to
11534 integers; an error occurs if this cannot be done.
11539 @node Relations for expr
11540 @subsection Relations for @command{expr}
11542 @cindex connectives, logical
11543 @cindex logical connectives
11544 @cindex relations, numeric or string
11546 @command{expr} supports the usual logical connectives and relations. These
11547 have lower precedence than the string and numeric operators
11548 (previous sections). Here is the list, lowest-precedence operator first.
11554 @cindex logical or operator
11555 @cindex or operator
11556 Returns its first argument if that is neither null nor zero, otherwise
11557 its second argument if it is neither null nor zero, otherwise 0. It
11558 does not evaluate its second argument if its first argument is neither
11563 @cindex logical and operator
11564 @cindex and operator
11565 Return its first argument if neither argument is null or zero, otherwise
11566 0. It does not evaluate its second argument if its first argument is
11569 @item < <= = == != >= >
11576 @cindex comparison operators
11578 Compare the arguments and return 1 if the relation is true, 0 otherwise.
11579 @code{==} is a synonym for @code{=}. @command{expr} first tries to convert
11580 both arguments to integers and do a numeric comparison; if either
11581 conversion fails, it does a lexicographic comparison using the character
11582 collating sequence specified by the @env{LC_COLLATE} locale.
11587 @node Examples of expr
11588 @subsection Examples of using @command{expr}
11590 @cindex examples of @command{expr}
11591 Here are a few examples, including quoting for shell metacharacters.
11593 To add 1 to the shell variable @code{foo}, in Bourne-compatible shells:
11596 foo=`expr $foo + 1`
11599 To print the non-directory part of the file name stored in
11600 @code{$fname}, which need not contain a @code{/}:
11603 expr $fname : '.*/\(.*\)' '|' $fname
11606 An example showing that @code{\+} is an operator:
11614 expr abc : 'a\(.\)c'
11616 expr index abcdef cz
11619 @error{} expr: syntax error
11620 expr index + index a
11626 @chapter Redirection
11628 @cindex redirection
11629 @cindex commands for redirection
11631 Unix shells commonly provide several forms of @dfn{redirection}---ways
11632 to change the input source or output destination of a command. But one
11633 useful redirection is performed by a separate command, not by the shell;
11634 it's described here.
11637 * tee invocation:: Redirect output to multiple files or processes.
11641 @node tee invocation
11642 @section @command{tee}: Redirect output to multiple files or processes
11645 @cindex pipe fitting
11646 @cindex destinations, multiple output
11647 @cindex read from stdin and write to stdout and files
11649 The @command{tee} command copies standard input to standard output and also
11650 to any files given as arguments. This is useful when you want not only
11651 to send some data down a pipe, but also to save a copy. Synopsis:
11654 tee [@var{option}]@dots{} [@var{file}]@dots{}
11657 If a file being written to does not already exist, it is created. If a
11658 file being written to already exists, the data it previously contained
11659 is overwritten unless the @option{-a} option is used.
11661 A @var{file} of @samp{-} causes @command{tee} to send another copy of
11662 input to standard output, but this is typically not that useful as the
11663 copies are interleaved.
11665 The program accepts the following options. Also see @ref{Common options}.
11672 Append standard input to the given files rather than overwriting
11676 @itemx --ignore-interrupts
11678 @opindex --ignore-interrupts
11679 Ignore interrupt signals.
11683 The @command{tee} command is useful when you happen to be transferring a large
11684 amount of data and also want to summarize that data without reading
11685 it a second time. For example, when you are downloading a DVD image,
11686 you often want to verify its signature or checksum right away.
11687 The inefficient way to do it is simply:
11690 wget http://example.com/some.iso && sha1sum some.iso
11693 One problem with the above is that it makes you wait for the
11694 download to complete before starting the time-consuming SHA1 computation.
11695 Perhaps even more importantly, the above requires reading
11696 the DVD image a second time (the first was from the network).
11698 The efficient way to do it is to interleave the download
11699 and SHA1 computation. Then, you'll get the checksum for
11700 free, because the entire process parallelizes so well:
11703 # slightly contrived, to demonstrate process substitution
11704 wget -O - http://example.com/dvd.iso \
11705 | tee >(sha1sum > dvd.sha1) > dvd.iso
11708 That makes @command{tee} write not just to the expected output file,
11709 but also to a pipe running @command{sha1sum} and saving the final
11710 checksum in a file named @file{dvd.sha1}.
11712 Note, however, that this example relies on a feature of modern shells
11713 called @dfn{process substitution}
11714 (the @samp{>(command)} syntax, above;
11715 @xref{Process Substitution,,Process Substitution, bashref,
11716 The Bash Reference Manual}.),
11717 so it works with @command{zsh}, @command{bash}, and @command{ksh},
11718 but not with @command{/bin/sh}. So if you write code like this
11719 in a shell script, be sure to start the script with @samp{#!/bin/bash}.
11721 Since the above example writes to one file and one process,
11722 a more conventional and portable use of @command{tee} is even better:
11725 wget -O - http://example.com/dvd.iso \
11726 | tee dvd.iso | sha1sum > dvd.sha1
11729 You can extend this example to make @command{tee} write to two processes,
11730 computing MD5 and SHA1 checksums in parallel. In this case,
11731 process substitution is required:
11734 wget -O - http://example.com/dvd.iso \
11735 | tee >(sha1sum > dvd.sha1) \
11736 >(md5sum > dvd.md5) \
11740 This technique is also useful when you want to make a @emph{compressed}
11741 copy of the contents of a pipe.
11742 Consider a tool to graphically summarize disk usage data from @samp{du -ak}.
11743 For a large hierarchy, @samp{du -ak} can run for a long time,
11744 and can easily produce terabytes of data, so you won't want to
11745 rerun the command unnecessarily. Nor will you want to save
11746 the uncompressed output.
11748 Doing it the inefficient way, you can't even start the GUI
11749 until after you've compressed all of the @command{du} output:
11752 du -ak | gzip -9 > /tmp/du.gz
11753 gzip -d /tmp/du.gz | xdiskusage -a
11756 With @command{tee} and process substitution, you start the GUI
11757 right away and eliminate the decompression completely:
11760 du -ak | tee >(gzip -9 > /tmp/du.gz) | xdiskusage -a
11763 Finally, if you regularly create more than one type of
11764 compressed tarball at once, for example when @code{make dist} creates
11765 both @command{gzip}-compressed and @command{bzip2}-compressed tarballs,
11766 there may be a better way.
11767 Typical @command{automake}-generated @file{Makefile} rules create
11768 the two compressed tar archives with commands in sequence, like this
11769 (slightly simplified):
11772 tardir=your-pkg-M.N
11773 tar chof - "$tardir" | gzip -9 -c > your-pkg-M.N.tar.gz
11774 tar chof - "$tardir" | bzip2 -9 -c > your-pkg-M.N.tar.bz2
11777 However, if the hierarchy you are archiving and compressing is larger
11778 than a couple megabytes, and especially if you are using a multi-processor
11779 system with plenty of memory, then you can do much better by reading the
11780 directory contents only once and running the compression programs in parallel:
11783 tardir=your-pkg-M.N
11784 tar chof - "$tardir" \
11785 | tee >(gzip -9 -c > your-pkg-M.N.tar.gz) \
11786 | bzip2 -9 -c > your-pkg-M.N.tar.bz2
11792 @node File name manipulation
11793 @chapter File name manipulation
11795 @cindex file name manipulation
11796 @cindex manipulation of file names
11797 @cindex commands for file name manipulation
11799 This section describes commands that manipulate file names.
11802 * basename invocation:: Strip directory and suffix from a file name.
11803 * dirname invocation:: Strip non-directory suffix from a file name.
11804 * pathchk invocation:: Check file name validity and portability.
11808 @node basename invocation
11809 @section @command{basename}: Strip directory and suffix from a file name
11812 @cindex strip directory and suffix from file names
11813 @cindex directory, stripping from file names
11814 @cindex suffix, stripping from file names
11815 @cindex file names, stripping directory and suffix
11816 @cindex leading directory components, stripping
11818 @command{basename} removes any leading directory components from
11819 @var{name}. Synopsis:
11822 basename @var{name} [@var{suffix}]
11825 If @var{suffix} is specified and is identical to the end of @var{name},
11826 it is removed from @var{name} as well. Note that since trailing slashes
11827 are removed prior to suffix matching, @var{suffix} will do nothing if it
11828 contains slashes. @command{basename} prints the result on standard
11831 @c This test is used both here and in the section on dirname.
11832 @macro basenameAndDirname
11833 Together, @command{basename} and @command{dirname} are designed such
11834 that if @samp{ls "$name"} succeeds, then the command sequence @samp{cd
11835 "$(dirname "$name")"; ls "$(basename "$name")"} will, too. This works
11836 for everything except file names containing a trailing newline.
11838 @basenameAndDirname
11840 @acronym{POSIX} allows the implementation to define the results if
11841 @var{name} is empty or @samp{//}. In the former case, @acronym{GNU}
11842 @command{basename} returns the empty string. In the latter case, the
11843 result is @samp{//} on platforms where @var{//} is distinct from
11844 @var{/}, and @samp{/} on platforms where there is no difference.
11846 The only options are @option{--help} and @option{--version}. @xref{Common
11847 options}. Options must precede operands.
11855 basename /usr/bin/sort
11858 basename include/stdio.h .h
11862 @node dirname invocation
11863 @section @command{dirname}: Strip non-directory suffix from a file name
11866 @cindex directory components, printing
11867 @cindex stripping non-directory suffix
11868 @cindex non-directory suffix, stripping
11870 @command{dirname} prints all but the final slash-delimited component of
11871 a string (presumably a file name). Synopsis:
11877 If @var{name} is a single component, @command{dirname} prints @samp{.}
11878 (meaning the current directory).
11880 @basenameAndDirname
11882 @acronym{POSIX} allows the implementation to define the results if
11883 @var{name} is @samp{//}. With @acronym{GNU} @command{dirname}, the
11884 result is @samp{//} on platforms where @var{//} is distinct from
11885 @var{/}, and @samp{/} on platforms where there is no difference.
11887 The only options are @option{--help} and @option{--version}. @xref{Common
11895 # Output "/usr/bin".
11896 dirname /usr/bin/sort
11903 @node pathchk invocation
11904 @section @command{pathchk}: Check file name validity and portability
11907 @cindex file names, checking validity and portability
11908 @cindex valid file names, checking for
11909 @cindex portable file names, checking for
11911 @command{pathchk} checks validity and portability of file names. Synopsis:
11914 pathchk [@var{option}]@dots{} @var{name}@dots{}
11917 For each @var{name}, @command{pathchk} prints an error message if any of
11918 these conditions is true:
11922 One of the existing directories in @var{name} does not have search
11923 (execute) permission,
11925 The length of @var{name} is larger than the maximum supported by the
11928 The length of one component of @var{name} is longer than
11929 its file system's maximum.
11932 A nonexistent @var{name} is not an error, so long a file with that
11933 name could be created under the above conditions.
11935 The program accepts the following options. Also see @ref{Common options}.
11936 Options must precede operands.
11942 Instead of performing checks based on the underlying file system,
11943 print an error message if any of these conditions is true:
11947 A file name is empty.
11950 A file name contains a character outside the @acronym{POSIX} portable file
11951 name character set, namely, the ASCII letters and digits, @samp{.},
11952 @samp{_}, @samp{-}, and @samp{/}.
11955 The length of a file name or one of its components exceeds the
11956 @acronym{POSIX} minimum limits for portability.
11961 Print an error message if a file name is empty, or if it contains a component
11962 that begins with @samp{-}.
11964 @item --portability
11965 @opindex --portability
11966 Print an error message if a file name is not portable to all @acronym{POSIX}
11967 hosts. This option is equivalent to @samp{-p -P}.
11971 @cindex exit status of @command{pathchk}
11975 0 if all specified file names passed all checks,
11980 @node Working context
11981 @chapter Working context
11983 @cindex working context
11984 @cindex commands for printing the working context
11986 This section describes commands that display or alter the context in
11987 which you are working: the current directory, the terminal settings, and
11988 so forth. See also the user-related commands in the next section.
11991 * pwd invocation:: Print working directory.
11992 * stty invocation:: Print or change terminal characteristics.
11993 * printenv invocation:: Print environment variables.
11994 * tty invocation:: Print file name of terminal on standard input.
11998 @node pwd invocation
11999 @section @command{pwd}: Print working directory
12002 @cindex print name of current directory
12003 @cindex current working directory, printing
12004 @cindex working directory, printing
12007 @command{pwd} prints the name of the current directory. Synopsis:
12010 pwd [@var{option}]@dots{}
12013 The program accepts the following options. Also see @ref{Common options}.
12020 If the contents of the environment variable @env{PWD} provide an
12021 absolute name of the current directory with no @samp{.} or @samp{..}
12022 components, but possibly with symbolic links, then output those
12023 contents. Otherwise, fall back to default @option{-P} handling.
12028 @opindex --physical
12029 Print a fully resolved name for the current directory. That is, all
12030 components of the printed name will be actual directory names---none
12031 will be symbolic links.
12034 @cindex symbolic links and @command{pwd}
12035 If @option{-L} and @option{-P} are both given, the last one takes
12036 precedence. If neither option is given, then this implementation uses
12037 @option{-P} as the default unless the @env{POSIXLY_CORRECT}
12038 environment variable is set.
12040 @mayConflictWithShellBuiltIn{pwd}
12045 @node stty invocation
12046 @section @command{stty}: Print or change terminal characteristics
12049 @cindex change or print terminal settings
12050 @cindex terminal settings
12051 @cindex line settings of terminal
12053 @command{stty} prints or changes terminal characteristics, such as baud rate.
12057 stty [@var{option}] [@var{setting}]@dots{}
12058 stty [@var{option}]
12061 If given no line settings, @command{stty} prints the baud rate, line
12062 discipline number (on systems that support it), and line settings
12063 that have been changed from the values set by @samp{stty sane}.
12064 By default, mode reading and setting are performed on the tty line
12065 connected to standard input, although this can be modified by the
12066 @option{--file} option.
12068 @command{stty} accepts many non-option arguments that change aspects of
12069 the terminal line operation, as described below.
12071 The program accepts the following options. Also see @ref{Common options}.
12078 Print all current settings in human-readable form. This option may not
12079 be used in combination with any line settings.
12081 @item -F @var{device}
12082 @itemx --file=@var{device}
12085 Set the line opened by the file name specified in @var{device} instead of
12086 the tty line connected to standard input. This option is necessary
12087 because opening a @acronym{POSIX} tty requires use of the @code{O_NONDELAY} flag to
12088 prevent a @acronym{POSIX} tty from blocking until the carrier detect line is high if
12089 the @code{clocal} flag is not set. Hence, it is not always possible
12090 to allow the shell to open the device in the traditional manner.
12096 @cindex machine-readable @command{stty} output
12097 Print all current settings in a form that can be used as an argument to
12098 another @command{stty} command to restore the current settings. This option
12099 may not be used in combination with any line settings.
12103 Many settings can be turned off by preceding them with a @samp{-}.
12104 Such arguments are marked below with ``May be negated'' in their
12105 description. The descriptions themselves refer to the positive
12106 case, that is, when @emph{not} negated (unless stated otherwise,
12109 Some settings are not available on all @acronym{POSIX} systems, since they use
12110 extensions. Such arguments are marked below with ``Non-@acronym{POSIX}'' in their
12111 description. On non-@acronym{POSIX} systems, those or other settings also may not
12112 be available, but it's not feasible to document all the variations: just
12118 * Control:: Control settings
12119 * Input:: Input settings
12120 * Output:: Output settings
12121 * Local:: Local settings
12122 * Combination:: Combination settings
12123 * Characters:: Special characters
12124 * Special:: Special settings
12129 @subsection Control settings
12131 @cindex control settings
12137 @cindex two-way parity
12138 Generate parity bit in output and expect parity bit in input.
12144 @cindex even parity
12145 Set odd parity (even if negated). May be negated.
12152 @cindex character size
12153 @cindex eight-bit characters
12154 Set character size to 5, 6, 7, or 8 bits.
12159 Send a hangup signal when the last process closes the tty. May be
12165 Use two stop bits per character (one if negated). May be negated.
12169 Allow input to be received. May be negated.
12173 @cindex modem control
12174 Disable modem control signals. May be negated.
12178 @cindex hardware flow control
12179 @cindex flow control, hardware
12180 @cindex RTS/CTS flow control
12181 Enable RTS/CTS flow control. Non-@acronym{POSIX}. May be negated.
12186 @subsection Input settings
12188 @cindex input settings
12189 These settings control operations on data received from the terminal.
12194 @cindex breaks, ignoring
12195 Ignore break characters. May be negated.
12199 @cindex breaks, cause interrupts
12200 Make breaks cause an interrupt signal. May be negated.
12204 @cindex parity, ignoring
12205 Ignore characters with parity errors. May be negated.
12209 @cindex parity errors, marking
12210 Mark parity errors (with a 255-0-character sequence). May be negated.
12214 Enable input parity checking. May be negated.
12218 @cindex eight-bit input
12219 Clear high (8th) bit of input characters. May be negated.
12223 @cindex newline, translating to return
12224 Translate newline to carriage return. May be negated.
12228 @cindex return, ignoring
12229 Ignore carriage return. May be negated.
12233 @cindex return, translating to newline
12234 Translate carriage return to newline. May be negated.
12238 @cindex input encoding, UTF-8
12239 Assume input characters are UTF-8 encoded. May be negated.
12243 @kindex C-s/C-q flow control
12244 @cindex XON/XOFF flow control
12245 Enable XON/XOFF flow control (that is, @kbd{CTRL-S}/@kbd{CTRL-Q}). May
12252 @cindex software flow control
12253 @cindex flow control, software
12254 Enable sending of @code{stop} character when the system input buffer
12255 is almost full, and @code{start} character when it becomes almost
12256 empty again. May be negated.
12260 @cindex uppercase, translating to lowercase
12261 Translate uppercase characters to lowercase. Non-@acronym{POSIX}. May be
12262 negated. Note ilcuc is not implemented, as one would not be able to issue
12263 almost any (lowercase) Unix command, after invoking it.
12267 Allow any character to restart output (only the start character
12268 if negated). Non-@acronym{POSIX}. May be negated.
12272 @cindex beeping at input buffer full
12273 Enable beeping and not flushing input buffer if a character arrives
12274 when the input buffer is full. Non-@acronym{POSIX}. May be negated.
12279 @subsection Output settings
12281 @cindex output settings
12282 These settings control operations on data sent to the terminal.
12287 Postprocess output. May be negated.
12291 @cindex lowercase, translating to output
12292 Translate lowercase characters to uppercase. Non-@acronym{POSIX}. May be
12293 negated. (Note ouclc is not currently implemented.)
12297 @cindex return, translating to newline
12298 Translate carriage return to newline. Non-@acronym{POSIX}. May be negated.
12302 @cindex newline, translating to crlf
12303 Translate newline to carriage return-newline. Non-@acronym{POSIX}. May be
12308 Do not print carriage returns in the first column. Non-@acronym{POSIX}.
12313 Newline performs a carriage return. Non-@acronym{POSIX}. May be negated.
12317 @cindex pad instead of timing for delaying
12318 Use fill (padding) characters instead of timing for delays. Non-@acronym{POSIX}.
12323 @cindex pad character
12324 Use @acronym{ASCII} @sc{del} characters for fill instead of
12325 @acronym{ASCII} @sc{nul} characters. Non-@acronym{POSIX}.
12331 Newline delay style. Non-@acronym{POSIX}.
12338 Carriage return delay style. Non-@acronym{POSIX}.
12344 @opindex tab@var{n}
12345 Horizontal tab delay style. Non-@acronym{POSIX}.
12350 Backspace delay style. Non-@acronym{POSIX}.
12355 Vertical tab delay style. Non-@acronym{POSIX}.
12360 Form feed delay style. Non-@acronym{POSIX}.
12365 @subsection Local settings
12367 @cindex local settings
12372 Enable @code{interrupt}, @code{quit}, and @code{suspend} special
12373 characters. May be negated.
12377 Enable @code{erase}, @code{kill}, @code{werase}, and @code{rprnt}
12378 special characters. May be negated.
12382 Enable non-@acronym{POSIX} special characters. May be negated.
12386 Echo input characters. May be negated.
12392 Echo @code{erase} characters as backspace-space-backspace. May be
12397 @cindex newline echoing after @code{kill}
12398 Echo a newline after a @code{kill} character. May be negated.
12402 @cindex newline, echoing
12403 Echo newline even if not echoing other characters. May be negated.
12407 @cindex flushing, disabling
12408 Disable flushing after @code{interrupt} and @code{quit} special
12409 characters. May be negated.
12413 @cindex case translation
12414 Enable input and output of uppercase characters by preceding their
12415 lowercase equivalents with @samp{\}, when @code{icanon} is set.
12416 Non-@acronym{POSIX}. May be negated.
12420 @cindex background jobs, stopping at terminal write
12421 Stop background jobs that try to write to the terminal. Non-@acronym{POSIX}.
12428 Echo erased characters backward, between @samp{\} and @samp{/}.
12429 Non-@acronym{POSIX}. May be negated.
12435 @cindex control characters, using @samp{^@var{c}}
12436 @cindex hat notation for control characters
12437 Echo control characters in hat notation (@samp{^@var{c}}) instead
12438 of literally. Non-@acronym{POSIX}. May be negated.
12444 Echo the @code{kill} special character by erasing each character on
12445 the line as indicated by the @code{echoprt} and @code{echoe} settings,
12446 instead of by the @code{echoctl} and @code{echok} settings. Non-@acronym{POSIX}.
12452 @subsection Combination settings
12454 @cindex combination settings
12455 Combination settings:
12462 Same as @code{parenb -parodd cs7}. May be negated. If negated, same
12463 as @code{-parenb cs8}.
12467 Same as @code{parenb parodd cs7}. May be negated. If negated, same
12468 as @code{-parenb cs8}.
12472 Same as @code{-icrnl -onlcr}. May be negated. If negated, same as
12473 @code{icrnl -inlcr -igncr onlcr -ocrnl -onlret}.
12477 Reset the @code{erase} and @code{kill} special characters to their default
12484 @c This is too long to write inline.
12486 cread -ignbrk brkint -inlcr -igncr icrnl -ixoff
12487 -iuclc -ixany imaxbel opost -olcuc -ocrnl onlcr
12488 -onocr -onlret -ofill -ofdel nl0 cr0 tab0 bs0 vt0
12489 ff0 isig icanon iexten echo echoe echok -echonl
12490 -noflsh -xcase -tostop -echoprt echoctl echoke
12494 and also sets all special characters to their default values.
12498 Same as @code{brkint ignpar istrip icrnl ixon opost isig icanon}, plus
12499 sets the @code{eof} and @code{eol} characters to their default values
12500 if they are the same as the @code{min} and @code{time} characters.
12501 May be negated. If negated, same as @code{raw}.
12508 -ignbrk -brkint -ignpar -parmrk -inpck -istrip
12509 -inlcr -igncr -icrnl -ixon -ixoff -iuclc -ixany
12510 -imaxbel -opost -isig -icanon -xcase min 1 time 0
12514 May be negated. If negated, same as @code{cooked}.
12518 Same as @option{-icanon}. May be negated. If negated, same as
12523 @cindex eight-bit characters
12524 Same as @code{-parenb -istrip cs8}. May be negated. If negated,
12525 same as @code{parenb istrip cs7}.
12529 Same as @option{-parenb -istrip -opost cs8}. May be negated.
12530 If negated, same as @code{parenb istrip opost cs7}.
12534 Same as @option{-ixany}. Non-@acronym{POSIX}. May be negated.
12538 Same as @code{tab0}. Non-@acronym{POSIX}. May be negated. If negated, same
12545 Same as @code{xcase iuclc olcuc}. Non-@acronym{POSIX}. May be negated.
12546 (Used for terminals with uppercase characters only.)
12550 Same as @code{echoe echoctl echoke}.
12554 Same as @code{echoe echoctl echoke -ixany intr ^C erase ^? kill C-u}.
12559 @subsection Special characters
12561 @cindex special characters
12562 @cindex characters, special
12564 The special characters' default values vary from system to system.
12565 They are set with the syntax @samp{name value}, where the names are
12566 listed below and the value can be given either literally, in hat
12567 notation (@samp{^@var{c}}), or as an integer which may start with
12568 @samp{0x} to indicate hexadecimal, @samp{0} to indicate octal, or
12569 any other digit to indicate decimal.
12571 @cindex disabling special characters
12572 @kindex u@r{, and disabling special characters}
12573 For GNU stty, giving a value of @code{^-} or @code{undef} disables that
12574 special character. (This is incompatible with Ultrix @command{stty},
12575 which uses a value of @samp{u} to disable a special character. GNU
12576 @command{stty} treats a value @samp{u} like any other, namely to set that
12577 special character to @key{U}.)
12583 Send an interrupt signal.
12587 Send a quit signal.
12591 Erase the last character typed.
12595 Erase the current line.
12599 Send an end of file (terminate the input).
12607 Alternate character to end the line. Non-@acronym{POSIX}.
12611 Switch to a different shell layer. Non-@acronym{POSIX}.
12615 Restart the output after stopping it.
12623 Send a terminal stop signal.
12627 Send a terminal stop signal after flushing the input. Non-@acronym{POSIX}.
12631 Redraw the current line. Non-@acronym{POSIX}.
12635 Erase the last word typed. Non-@acronym{POSIX}.
12639 Enter the next character typed literally, even if it is a special
12640 character. Non-@acronym{POSIX}.
12645 @subsection Special settings
12647 @cindex special settings
12652 Set the minimum number of characters that will satisfy a read until
12653 the time value has expired, when @option{-icanon} is set.
12657 Set the number of tenths of a second before reads time out if the minimum
12658 number of characters have not been read, when @option{-icanon} is set.
12660 @item ispeed @var{n}
12662 Set the input speed to @var{n}.
12664 @item ospeed @var{n}
12666 Set the output speed to @var{n}.
12670 Tell the tty kernel driver that the terminal has @var{n} rows. Non-@acronym{POSIX}.
12673 @itemx columns @var{n}
12676 Tell the kernel that the terminal has @var{n} columns. Non-@acronym{POSIX}.
12682 Print the number of rows and columns that the kernel thinks the
12683 terminal has. (Systems that don't support rows and columns in the kernel
12684 typically use the environment variables @env{LINES} and @env{COLUMNS}
12685 instead; however, GNU @command{stty} does not know anything about them.)
12686 Non-@acronym{POSIX}.
12690 Use line discipline @var{n}. Non-@acronym{POSIX}.
12694 Print the terminal speed.
12697 @cindex baud rate, setting
12698 Set the input and output speeds to @var{n}. @var{n} can be one of: 0
12699 50 75 110 134 134.5 150 200 300 600 1200 1800 2400 4800 9600 19200
12700 38400 @code{exta} @code{extb}. @code{exta} is the same as 19200;
12701 @code{extb} is the same as 38400. Many systems, including GNU/Linux,
12702 support higher speeds. The @command{stty} command includes support
12719 4000000 where the system supports these.
12720 0 hangs up the line if @option{-clocal} is set.
12724 @node printenv invocation
12725 @section @command{printenv}: Print all or some environment variables
12728 @cindex printing all or some environment variables
12729 @cindex environment variables, printing
12731 @command{printenv} prints environment variable values. Synopsis:
12734 printenv [@var{option}] [@var{variable}]@dots{}
12737 If no @var{variable}s are specified, @command{printenv} prints the value of
12738 every environment variable. Otherwise, it prints the value of each
12739 @var{variable} that is set, and nothing for those that are not set.
12741 The only options are a lone @option{--help} or @option{--version}.
12742 @xref{Common options}.
12744 @cindex exit status of @command{printenv}
12748 0 if all variables specified were found
12749 1 if at least one specified variable was not found
12750 2 if a write error occurred
12754 @node tty invocation
12755 @section @command{tty}: Print file name of terminal on standard input
12758 @cindex print terminal file name
12759 @cindex terminal file name, printing
12761 @command{tty} prints the file name of the terminal connected to its standard
12762 input. It prints @samp{not a tty} if standard input is not a terminal.
12766 tty [@var{option}]@dots{}
12769 The program accepts the following option. Also see @ref{Common options}.
12779 Print nothing; only return an exit status.
12783 @cindex exit status of @command{tty}
12787 0 if standard input is a terminal
12788 1 if standard input is not a terminal
12789 2 if given incorrect arguments
12790 3 if a write error occurs
12794 @node User information
12795 @chapter User information
12797 @cindex user information, commands for
12798 @cindex commands for printing user information
12800 This section describes commands that print user-related information:
12801 logins, groups, and so forth.
12804 * id invocation:: Print user identity.
12805 * logname invocation:: Print current login name.
12806 * whoami invocation:: Print effective user ID.
12807 * groups invocation:: Print group names a user is in.
12808 * users invocation:: Print login names of users currently logged in.
12809 * who invocation:: Print who is currently logged in.
12813 @node id invocation
12814 @section @command{id}: Print user identity
12817 @cindex real user and group IDs, printing
12818 @cindex effective user and group IDs, printing
12819 @cindex printing real and effective user and group IDs
12821 @command{id} prints information about the given user, or the process
12822 running it if no user is specified. Synopsis:
12825 id [@var{option}]@dots{} [@var{username}]
12828 By default, it prints the real user ID, real group ID, effective user ID
12829 if different from the real user ID, effective group ID if different from
12830 the real group ID, and supplemental group IDs.
12832 Each of these numeric values is preceded by an identifying string and
12833 followed by the corresponding user or group name in parentheses.
12835 The options cause @command{id} to print only part of the above information.
12836 Also see @ref{Common options}.
12843 Print only the group ID.
12849 Print only the group ID and the supplementary groups.
12855 Print the user or group name instead of the ID number. Requires
12856 @option{-u}, @option{-g}, or @option{-G}.
12862 Print the real, instead of effective, user or group ID. Requires
12863 @option{-u}, @option{-g}, or @option{-G}.
12869 Print only the user ID.
12876 @cindex security context
12877 Print only the security context of the current user.
12878 If SELinux is disabled then print a warning and
12879 set the exit status to 1.
12885 @macro primaryAndSupplementaryGroups{cmd,arg}
12886 Primary and supplementary groups for a process are normally inherited
12887 from its parent and are usually unchanged since login. This means
12888 that if you change the group database after logging in, @command{\cmd\}
12889 will not reflect your changes within your existing login session.
12890 Running @command{\cmd\} with a \arg\ causes the user and group
12891 database to be consulted afresh, and so will give a different result.
12893 @primaryAndSupplementaryGroups{id,user argument}
12895 @node logname invocation
12896 @section @command{logname}: Print current login name
12899 @cindex printing user's login name
12900 @cindex login name, printing
12901 @cindex user name, printing
12904 @command{logname} prints the calling user's name, as found in a
12905 system-maintained file (often @file{/var/run/utmp} or
12906 @file{/etc/utmp}), and exits with a status of 0. If there is no entry
12907 for the calling process, @command{logname} prints
12908 an error message and exits with a status of 1.
12910 The only options are @option{--help} and @option{--version}. @xref{Common
12916 @node whoami invocation
12917 @section @command{whoami}: Print effective user ID
12920 @cindex effective user ID, printing
12921 @cindex printing the effective user ID
12923 @command{whoami} prints the user name associated with the current
12924 effective user ID. It is equivalent to the command @samp{id -un}.
12926 The only options are @option{--help} and @option{--version}. @xref{Common
12932 @node groups invocation
12933 @section @command{groups}: Print group names a user is in
12936 @cindex printing groups a user is in
12937 @cindex supplementary groups, printing
12939 @command{groups} prints the names of the primary and any supplementary
12940 groups for each given @var{username}, or the current process if no names
12941 are given. If more than one name is given, the name of each user is
12943 the list of that user's groups and the user name is separated from the
12944 group list by a colon. Synopsis:
12947 groups [@var{username}]@dots{}
12950 The group lists are equivalent to the output of the command @samp{id -Gn}.
12952 @primaryAndSupplementaryGroups{groups,list of users}
12954 The only options are @option{--help} and @option{--version}. @xref{Common
12960 @node users invocation
12961 @section @command{users}: Print login names of users currently logged in
12964 @cindex printing current usernames
12965 @cindex usernames, printing current
12967 @cindex login sessions, printing users with
12968 @command{users} prints on a single line a blank-separated list of user
12969 names of users currently logged in to the current host. Each user name
12970 corresponds to a login session, so if a user has more than one login
12971 session, that user's name will appear the same number of times in the
12980 With no @var{file} argument, @command{users} extracts its information from
12981 a system-maintained file (often @file{/var/run/utmp} or
12982 @file{/etc/utmp}). If a file argument is given, @command{users} uses
12983 that file instead. A common choice is @file{/var/log/wtmp}.
12985 The only options are @option{--help} and @option{--version}. @xref{Common
12991 @node who invocation
12992 @section @command{who}: Print who is currently logged in
12995 @cindex printing current user information
12996 @cindex information, about current users
12998 @command{who} prints information about users who are currently logged on.
13002 @command{who} [@var{option}] [@var{file}] [am i]
13005 @cindex terminal lines, currently used
13007 @cindex remote hostname
13008 If given no non-option arguments, @command{who} prints the following
13009 information for each user currently logged on: login name, terminal
13010 line, login time, and remote hostname or X display.
13014 If given one non-option argument, @command{who} uses that instead of
13015 a default system-maintained file (often @file{/var/run/utmp} or
13016 @file{/etc/utmp}) as the name of the file containing the record of
13017 users logged on. @file{/var/log/wtmp} is commonly given as an argument
13018 to @command{who} to look at who has previously logged on.
13022 If given two non-option arguments, @command{who} prints only the entry
13023 for the user running it (determined from its standard input), preceded
13024 by the hostname. Traditionally, the two arguments given are @samp{am
13025 i}, as in @samp{who am i}.
13028 Time stamps are listed according to the time zone rules specified by
13029 the @env{TZ} environment variable, or by the system default rules if
13030 @env{TZ} is not set. @xref{TZ Variable,, Specifying the Time Zone
13031 with @env{TZ}, libc, The GNU C Library Reference Manual}.
13033 The program accepts the following options. Also see @ref{Common options}.
13041 Same as @samp{-b -d --login -p -r -t -T -u}.
13047 Print the date and time of last system boot.
13053 Print information corresponding to dead processes.
13059 Print a line of column headings.
13065 List only the entries that correspond to processes via which the
13066 system is waiting for a user to login. The user name is always @samp{LOGIN}.
13070 Attempt to canonicalize hostnames found in utmp through a DNS lookup. This
13071 is not the default because it can cause significant delays on systems with
13072 automatic dial-up internet access.
13076 Same as @samp{who am i}.
13082 List active processes spawned by init.
13088 Print only the login names and the number of users logged on.
13089 Overrides all other options.
13094 @opindex --runlevel
13095 Print the current (and maybe previous) run-level of the init process.
13099 Ignored; for compatibility with other versions of @command{who}.
13105 Print last system clock change.
13110 After the login time, print the number of hours and minutes that the
13111 user has been idle. @samp{.} means the user was active in the last minute.
13112 @samp{old} means the user has been idle for more than 24 hours.
13123 @opindex --writable
13124 @cindex message status
13125 @pindex write@r{, allowed}
13126 After each login name print a character indicating the user's message status:
13129 @samp{+} allowing @code{write} messages
13130 @samp{-} disallowing @code{write} messages
13131 @samp{?} cannot find terminal device
13139 @node System context
13140 @chapter System context
13142 @cindex system context
13143 @cindex context, system
13144 @cindex commands for system context
13146 This section describes commands that print or change system-wide
13150 * date invocation:: Print or set system date and time.
13151 * arch invocation:: Print machine hardware name.
13152 * uname invocation:: Print system information.
13153 * hostname invocation:: Print or set system name.
13154 * hostid invocation:: Print numeric host identifier.
13155 * uptime invocation:: Print system uptime and load.
13158 @node date invocation
13159 @section @command{date}: Print or set system date and time
13162 @cindex time, printing or setting
13163 @cindex printing the current time
13168 date [@var{option}]@dots{} [+@var{format}]
13169 date [-u|--utc|--universal] @c this avoids a newline in the output
13170 [ MMDDhhmm[[CC]YY][.ss] ]
13174 Invoking @command{date} with no @var{format} argument is equivalent to invoking
13175 it with a default format that depends on the @env{LC_TIME} locale category.
13176 In the default C locale, this format is @samp{'+%a %b %e %H:%M:%S %Z %Y'},
13177 so the output looks like @samp{Thu Mar @ 3 13:47:51 PST 2005}.
13180 Normally, @command{date} uses the time zone rules indicated by the
13181 @env{TZ} environment variable, or the system default rules if @env{TZ}
13182 is not set. @xref{TZ Variable,, Specifying the Time Zone with
13183 @env{TZ}, libc, The GNU C Library Reference Manual}.
13185 @findex strftime @r{and @command{date}}
13186 @cindex time formats
13187 @cindex formatting times
13188 If given an argument that starts with a @samp{+}, @command{date} prints the
13189 current date and time (or the date and time specified by the
13190 @option{--date} option, see below) in the format defined by that argument,
13191 which is similar to that of the @code{strftime} function. Except for
13192 conversion specifiers, which start with @samp{%}, characters in the
13193 format string are printed unchanged. The conversion specifiers are
13199 * Time conversion specifiers:: %[HIklMNpPrRsSTXzZ]
13200 * Date conversion specifiers:: %[aAbBcCdDeFgGhjmuUVwWxyY]
13201 * Literal conversion specifiers:: %[%nt]
13202 * Padding and other flags:: Pad with zeros, spaces, etc.
13203 * Setting the time:: Changing the system clock.
13204 * Options for date:: Instead of the current time.
13206 * Date input formats:: Specifying date strings.
13208 * Examples of date:: Examples.
13211 @node Time conversion specifiers
13212 @subsection Time conversion specifiers
13214 @cindex time conversion specifiers
13215 @cindex conversion specifiers, time
13217 @command{date} conversion specifiers related to times.
13221 hour (@samp{00}@dots{}@samp{23})
13223 hour (@samp{01}@dots{}@samp{12})
13225 hour (@samp{ 0}@dots{}@samp{23}).
13226 This is a @acronym{GNU} extension.
13228 hour (@samp{ 1}@dots{}@samp{12}).
13229 This is a @acronym{GNU} extension.
13231 minute (@samp{00}@dots{}@samp{59})
13233 nanoseconds (@samp{000000000}@dots{}@samp{999999999}).
13234 This is a @acronym{GNU} extension.
13236 locale's equivalent of either @samp{AM} or @samp{PM};
13237 blank in many locales.
13238 Noon is treated as @samp{PM} and midnight as @samp{AM}.
13240 like @samp{%p}, except lower case.
13241 This is a @acronym{GNU} extension.
13243 locale's 12-hour clock time (e.g., @samp{11:11:04 PM})
13245 24-hour hour and minute. Same as @samp{%H:%M}.
13246 This is a @acronym{GNU} extension.
13248 @cindex epoch, seconds since
13249 @cindex seconds since the epoch
13250 @cindex beginning of time
13251 seconds since the epoch, i.e., since 1970-01-01 00:00:00 UTC.
13252 Leap seconds are not counted unless leap second support is available.
13253 @xref{%s-examples}, for examples.
13254 This is a @acronym{GNU} extension.
13256 second (@samp{00}@dots{}@samp{60}).
13257 This may be @samp{60} if leap seconds are supported.
13259 24-hour hour, minute, and second. Same as @samp{%H:%M:%S}.
13261 locale's time representation (e.g., @samp{23:13:48})
13263 @w{@acronym{RFC} 2822/@acronym{ISO} 8601} style numeric time zone
13264 (e.g., @samp{-0600} or @samp{+0530}), or nothing if no
13265 time zone is determinable. This value reflects the numeric time zone
13266 appropriate for the current time, using the time zone rules specified
13267 by the @env{TZ} environment variable.
13268 The time (and optionally, the time zone rules) can be overridden
13269 by the @option{--date} option.
13270 This is a @acronym{GNU} extension.
13272 @w{@acronym{RFC} 3339/@acronym{ISO} 8601} style numeric time zone with
13273 @samp{:} (e.g., @samp{-06:00} or @samp{+05:30}), or nothing if no time
13274 zone is determinable.
13275 This is a @acronym{GNU} extension.
13277 Numeric time zone to the nearest second with @samp{:} (e.g.,
13278 @samp{-06:00:00} or @samp{+05:30:00}), or nothing if no time zone is
13280 This is a @acronym{GNU} extension.
13282 Numeric time zone with @samp{:} using the minimum necessary precision
13283 (e.g., @samp{-06}, @samp{+05:30}, or @samp{-04:56:02}), or nothing if
13284 no time zone is determinable.
13285 This is a @acronym{GNU} extension.
13287 alphabetic time zone abbreviation (e.g., @samp{EDT}), or nothing if no
13288 time zone is determinable. See @samp{%z} for how it is determined.
13292 @node Date conversion specifiers
13293 @subsection Date conversion specifiers
13295 @cindex date conversion specifiers
13296 @cindex conversion specifiers, date
13298 @command{date} conversion specifiers related to dates.
13302 locale's abbreviated weekday name (e.g., @samp{Sun})
13304 locale's full weekday name, variable length (e.g., @samp{Sunday})
13306 locale's abbreviated month name (e.g., @samp{Jan})
13308 locale's full month name, variable length (e.g., @samp{January})
13310 locale's date and time (e.g., @samp{Thu Mar @ 3 23:05:25 2005})
13312 century. This is like @samp{%Y}, except the last two digits are omitted.
13313 For example, it is @samp{20} if @samp{%Y} is @samp{2000},
13314 and is @samp{-0} if @samp{%Y} is @samp{-001}.
13315 It is normally at least two characters, but it may be more.
13317 day of month (e.g., @samp{01})
13319 date; same as @samp{%m/%d/%y}
13321 day of month, space padded; same as @samp{%_d}
13323 full date in @acronym{ISO} 8601 format; same as @samp{%Y-%m-%d}.
13324 This is a good choice for a date format, as it is standard and
13325 is easy to sort in the usual case where years are in the range
13327 This is a @acronym{GNU} extension.
13329 year corresponding to the @acronym{ISO} week number, but without the century
13330 (range @samp{00} through @samp{99}). This has the same format and value
13331 as @samp{%y}, except that if the @acronym{ISO} week number (see
13333 to the previous or next year, that year is used instead.
13334 This is a @acronym{GNU} extension.
13336 year corresponding to the @acronym{ISO} week number. This has the
13337 same format and value as @samp{%Y}, except that if the @acronym{ISO}
13339 @samp{%V}) belongs to the previous or next year, that year is used
13341 It is normally useful only if @samp{%V} is also used;
13342 for example, the format @samp{%G-%m-%d} is probably a mistake,
13343 since it combines the ISO week number year with the conventional month and day.
13344 This is a @acronym{GNU} extension.
13348 day of year (@samp{001}@dots{}@samp{366})
13350 month (@samp{01}@dots{}@samp{12})
13352 day of week (@samp{1}@dots{}@samp{7}) with @samp{1} corresponding to Monday
13354 week number of year, with Sunday as the first day of the week
13355 (@samp{00}@dots{}@samp{53}).
13356 Days in a new year preceding the first Sunday are in week zero.
13358 @acronym{ISO} week number, that is, the
13359 week number of year, with Monday as the first day of the week
13360 (@samp{01}@dots{}@samp{53}).
13361 If the week containing January 1 has four or more days in
13362 the new year, then it is considered week 1; otherwise, it is week 53 of
13363 the previous year, and the next week is week 1. (See the @acronym{ISO} 8601
13366 day of week (@samp{0}@dots{}@samp{6}) with 0 corresponding to Sunday
13368 week number of year, with Monday as first day of week
13369 (@samp{00}@dots{}@samp{53}).
13370 Days in a new year preceding the first Monday are in week zero.
13372 locale's date representation (e.g., @samp{12/31/99})
13374 last two digits of year (@samp{00}@dots{}@samp{99})
13376 year. This is normally at least four characters, but it may be more.
13377 Year @samp{0000} precedes year @samp{0001}, and year @samp{-001}
13378 precedes year @samp{0000}.
13382 @node Literal conversion specifiers
13383 @subsection Literal conversion specifiers
13385 @cindex literal conversion specifiers
13386 @cindex conversion specifiers, literal
13388 @command{date} conversion specifiers that produce literal strings.
13400 @node Padding and other flags
13401 @subsection Padding and other flags
13403 @cindex numeric field padding
13404 @cindex padding of numeric fields
13405 @cindex fields, padding numeric
13407 Unless otherwise specified, @command{date} normally pads numeric fields
13408 with zeros, so that, for
13409 example, numeric months are always output as two digits.
13410 Seconds since the epoch are not padded, though,
13411 since there is no natural width for them.
13413 As a @acronym{GNU} extension, @command{date} recognizes any of the
13414 following optional flags after the @samp{%}:
13418 (hyphen) Do not pad the field; useful if the output is intended for
13421 (underscore) Pad with spaces; useful if you need a fixed
13422 number of characters in the output, but zeros are too distracting.
13424 (zero) Pad with zeros even if the conversion specifier
13425 would normally pad with spaces.
13427 Use upper case characters if possible.
13429 Use opposite case characters if possible.
13430 A field that is normally upper case becomes lower case, and vice versa.
13434 Here are some examples of padding:
13437 date +%d/%m -d "Feb 1"
13439 date +%-d/%-m -d "Feb 1"
13441 date +%_d/%_m -d "Feb 1"
13445 As a @acronym{GNU} extension, you can specify the field width
13446 (after any flag, if present) as a decimal number. If the natural size of the
13447 output of the field has less than the specified number of characters,
13448 the result is written right adjusted and padded to the given
13449 size. For example, @samp{%9B} prints the right adjusted month name in
13450 a field of width 9.
13452 An optional modifier can follow the optional flag and width
13453 specification. The modifiers are:
13457 Use the locale's alternate representation for date and time. This
13458 modifier applies to the @samp{%c}, @samp{%C}, @samp{%x}, @samp{%X},
13459 @samp{%y} and @samp{%Y} conversion specifiers. In a Japanese locale, for
13460 example, @samp{%Ex} might yield a date format based on the Japanese
13464 Use the locale's alternate numeric symbols for numbers. This modifier
13465 applies only to numeric conversion specifiers.
13468 If the format supports the modifier but no alternate representation
13469 is available, it is ignored.
13472 @node Setting the time
13473 @subsection Setting the time
13475 @cindex setting the time
13476 @cindex time setting
13477 @cindex appropriate privileges
13479 If given an argument that does not start with @samp{+}, @command{date} sets
13480 the system clock to the date and time specified by that argument (as
13481 described below). You must have appropriate privileges to set the
13482 system clock. The @option{--date} and @option{--set} options may not be
13483 used with such an argument. The @option{--universal} option may be used
13484 with such an argument to indicate that the specified date and time are
13485 relative to Coordinated Universal Time rather than to the local time
13488 The argument must consist entirely of digits, which have the following
13501 first two digits of year (optional)
13503 last two digits of year (optional)
13508 The @option{--set} option also sets the system clock; see the next section.
13511 @node Options for date
13512 @subsection Options for @command{date}
13514 @cindex @command{date} options
13515 @cindex options for @command{date}
13517 The program accepts the following options. Also see @ref{Common options}.
13521 @item -d @var{datestr}
13522 @itemx --date=@var{datestr}
13525 @cindex parsing date strings
13526 @cindex date strings, parsing
13527 @cindex arbitrary date strings, parsing
13530 @opindex next @var{day}
13531 @opindex last @var{day}
13532 Display the date and time specified in @var{datestr} instead of the
13533 current date and time. @var{datestr} can be in almost any common
13534 format. It can contain month names, time zones, @samp{am} and @samp{pm},
13535 @samp{yesterday}, etc. For example, @option{--date="2004-02-27
13536 14:19:13.489392193 +0530"} specifies the instant of time that is
13537 489,392,193 nanoseconds after February 27, 2004 at 2:19:13 PM in a
13538 time zone that is 5 hours and 30 minutes east of @acronym{UTC}.@*
13539 Note: input currently must be in locale independent format. E.g., the
13540 LC_TIME=C below is needed to print back the correct date in many locales:
13542 date -d "$(LC_TIME=C date)"
13544 @xref{Date input formats}.
13546 @item -f @var{datefile}
13547 @itemx --file=@var{datefile}
13550 Parse each line in @var{datefile} as with @option{-d} and display the
13551 resulting date and time. If @var{datefile} is @samp{-}, use standard
13552 input. This is useful when you have many dates to process, because the
13553 system overhead of starting up the @command{date} executable many times can
13556 @item -r @var{file}
13557 @itemx --reference=@var{file}
13559 @opindex --reference
13560 Display the date and time of the last modification of @var{file},
13561 instead of the current date and time.
13568 @opindex --rfc-2822
13569 Display the date and time using the format @samp{%a, %d %b %Y %H:%M:%S
13570 %z}, evaluated in the C locale so abbreviations are always in English.
13574 Fri, 09 Sep 2005 13:51:39 -0700
13577 This format conforms to
13578 @uref{ftp://ftp.rfc-editor.org/in-notes/rfc2822.txt, Internet
13579 @acronym{RFCs} 2822} and
13580 @uref{ftp://ftp.rfc-editor.org/in-notes/rfc822.txt, 822}, the
13581 current and previous standards for Internet email.
13583 @item --rfc-3339=@var{timespec}
13584 @opindex --rfc-3339=@var{timespec}
13585 Display the date using a format specified by
13586 @uref{ftp://ftp.rfc-editor.org/in-notes/rfc3339.txt, Internet
13587 @acronym{RFC} 3339}. This is a subset of the @acronym{ISO} 8601
13588 format, except that it also permits applications to use a space rather
13589 than a @samp{T} to separate dates from times. Unlike the other
13590 standard formats, @acronym{RFC} 3339 format is always suitable as
13591 input for the @option{--date} (@option{-d}) and @option{--file}
13592 (@option{-f}) options, regardless of the current locale.
13594 The argument @var{timespec} specifies how much of the time to include.
13595 It can be one of the following:
13599 Print just the full-date, e.g., @samp{2005-09-14}.
13600 This is equivalent to the format @samp{%Y-%m-%d}.
13603 Print the full-date and full-time separated by a space, e.g.,
13604 @samp{2005-09-14 00:56:06+05:30}. The output ends with a numeric
13605 time-offset; here the @samp{+05:30} means that local time is five
13606 hours and thirty minutes east of @acronym{UTC}. This is equivalent to
13607 the format @samp{%Y-%m-%d %H:%M:%S%:z}.
13610 Like @samp{seconds}, but also print nanoseconds, e.g.,
13611 @samp{2005-09-14 00:56:06.998458565+05:30}.
13612 This is equivalent to the format @samp{%Y-%m-%d %H:%M:%S.%N%:z}.
13616 @item -s @var{datestr}
13617 @itemx --set=@var{datestr}
13620 Set the date and time to @var{datestr}. See @option{-d} above.
13627 @opindex --universal
13628 @cindex Coordinated Universal Time
13630 @cindex Greenwich Mean Time
13633 Use Coordinated Universal Time (@acronym{UTC}) by operating as if the
13634 @env{TZ} environment variable were set to the string @samp{UTC0}.
13636 Universal Time is often called ``Greenwich Mean Time'' (@sc{gmt}) for
13637 historical reasons.
13641 @node Examples of date
13642 @subsection Examples of @command{date}
13644 @cindex examples of @command{date}
13646 Here are a few examples. Also see the documentation for the @option{-d}
13647 option in the previous section.
13652 To print the date of the day before yesterday:
13655 date --date='2 days ago'
13659 To print the date of the day three months and one day hence:
13662 date --date='3 months 1 day'
13666 To print the day of year of Christmas in the current year:
13669 date --date='25 Dec' +%j
13673 To print the current full month name and the day of the month:
13679 But this may not be what you want because for the first nine days of
13680 the month, the @samp{%d} expands to a zero-padded two-digit field,
13681 for example @samp{date -d 1may '+%B %d'} will print @samp{May 01}.
13684 To print a date without the leading zero for one-digit days
13685 of the month, you can use the (@acronym{GNU} extension)
13686 @samp{-} flag to suppress
13687 the padding altogether:
13690 date -d 1may '+%B %-d
13694 To print the current date and time in the format required by many
13695 non-@acronym{GNU} versions of @command{date} when setting the system clock:
13698 date +%m%d%H%M%Y.%S
13702 To set the system clock forward by two minutes:
13705 date --set='+2 minutes'
13709 To print the date in @acronym{RFC} 2822 format,
13710 use @samp{date --rfc-2822}. Here is some example output:
13713 Fri, 09 Sep 2005 13:51:39 -0700
13716 @anchor{%s-examples}
13718 To convert a date string to the number of seconds since the epoch
13719 (which is 1970-01-01 00:00:00 UTC), use the @option{--date} option with
13720 the @samp{%s} format. That can be useful in sorting and/or graphing
13721 and/or comparing data by date. The following command outputs the
13722 number of the seconds since the epoch for the time two minutes after the
13726 date --date='1970-01-01 00:02:00 +0000' +%s
13730 If you do not specify time zone information in the date string,
13731 @command{date} uses your computer's idea of the time zone when
13732 interpreting the string. For example, if your computer's time zone is
13733 that of Cambridge, Massachusetts, which was then 5 hours (i.e., 18,000
13734 seconds) behind UTC:
13737 # local time zone used
13738 date --date='1970-01-01 00:02:00' +%s
13743 If you're sorting or graphing dated data, your raw date values may be
13744 represented as seconds since the epoch. But few people can look at
13745 the date @samp{946684800} and casually note ``Oh, that's the first second
13746 of the year 2000 in Greenwich, England.''
13749 date --date='2000-01-01 UTC' +%s
13753 An alternative is to use the @option{--utc} (@option{-u}) option.
13754 Then you may omit @samp{UTC} from the date string. Although this
13755 produces the same result for @samp{%s} and many other format sequences,
13756 with a time zone offset different from zero, it would give a different
13757 result for zone-dependent formats like @samp{%z}.
13760 date -u --date=2000-01-01 +%s
13764 To convert such an unwieldy number of seconds back to
13765 a more readable form, use a command like this:
13768 # local time zone used
13769 date -d '1970-01-01 UTC 946684800 seconds' +"%Y-%m-%d %T %z"
13770 1999-12-31 19:00:00 -0500
13773 Or if you do not mind depending on the @samp{@@} feature present since
13774 coreutils 5.3.0, you could shorten this to:
13777 date -d @@946684800 +"%F %T %z"
13778 1999-12-31 19:00:00 -0500
13781 Often it is better to output UTC-relative date and time:
13784 date -u -d '1970-01-01 946684800 seconds' +"%Y-%m-%d %T %z"
13785 2000-01-01 00:00:00 +0000
13791 @node arch invocation
13792 @section @command{arch}: Print machine hardware name
13795 @cindex print machine hardware name
13796 @cindex system information, printing
13798 @command{arch} prints the machine hardware name,
13799 and is equivalent to @samp{uname -m}.
13803 arch [@var{option}]
13806 The program accepts the @ref{Common options} only.
13811 @node uname invocation
13812 @section @command{uname}: Print system information
13815 @cindex print system information
13816 @cindex system information, printing
13818 @command{uname} prints information about the machine and operating system
13819 it is run on. If no options are given, @command{uname} acts as if the
13820 @option{-s} option were given. Synopsis:
13823 uname [@var{option}]@dots{}
13826 If multiple options or @option{-a} are given, the selected information is
13827 printed in this order:
13830 @var{kernel-name} @var{nodename} @var{kernel-release} @var{kernel-version}
13831 @var{machine} @var{processor} @var{hardware-platform} @var{operating-system}
13834 The information may contain internal spaces, so such output cannot be
13835 parsed reliably. In the following example, @var{release} is
13836 @samp{2.2.18ss.e820-bda652a #4 SMP Tue Jun 5 11:24:08 PDT 2001}:
13840 @result{} Linux dum 2.2.18 #4 SMP Tue Jun 5 11:24:08 PDT 2001 i686 unknown unknown GNU/Linux
13844 The program accepts the following options. Also see @ref{Common options}.
13852 Print all of the below information, except omit the processor type
13853 and the hardware platform name if they are unknown.
13856 @itemx --hardware-platform
13858 @opindex --hardware-platform
13859 @cindex implementation, hardware
13860 @cindex hardware platform
13861 @cindex platform, hardware
13862 Print the hardware platform name
13863 (sometimes called the hardware implementation).
13864 Print @samp{unknown} if the kernel does not make this information
13865 easily available, as is the case with Linux kernels.
13871 @cindex machine type
13872 @cindex hardware class
13873 @cindex hardware type
13874 Print the machine hardware name (sometimes called the hardware class
13880 @opindex --nodename
13883 @cindex network node name
13884 Print the network node hostname.
13889 @opindex --processor
13890 @cindex host processor type
13891 Print the processor type (sometimes called the instruction set
13892 architecture or ISA).
13893 Print @samp{unknown} if the kernel does not make this information
13894 easily available, as is the case with Linux kernels.
13897 @itemx --operating-system
13899 @opindex --operating-system
13900 @cindex operating system name
13901 Print the name of the operating system.
13904 @itemx --kernel-release
13906 @opindex --kernel-release
13907 @cindex kernel release
13908 @cindex release of kernel
13909 Print the kernel release.
13912 @itemx --kernel-name
13914 @opindex --kernel-name
13915 @cindex kernel name
13916 @cindex name of kernel
13917 Print the kernel name.
13918 @acronym{POSIX} 1003.1-2001 (@pxref{Standards conformance}) calls this
13919 ``the implementation of the operating system'', because the
13920 @acronym{POSIX} specification itself has no notion of ``kernel''.
13921 The kernel name might be the same as the operating system name printed
13922 by the @option{-o} or @option{--operating-system} option, but it might
13923 differ. Some operating systems (e.g., FreeBSD, HP-UX) have the same
13924 name as their underlying kernels; others (e.g., GNU/Linux, Solaris)
13928 @itemx --kernel-version
13930 @opindex --kernel-version
13931 @cindex kernel version
13932 @cindex version of kernel
13933 Print the kernel version.
13940 @node hostname invocation
13941 @section @command{hostname}: Print or set system name
13944 @cindex setting the hostname
13945 @cindex printing the hostname
13946 @cindex system name, printing
13947 @cindex appropriate privileges
13949 With no arguments, @command{hostname} prints the name of the current host
13950 system. With one argument, it sets the current host name to the
13951 specified string. You must have appropriate privileges to set the host
13955 hostname [@var{name}]
13958 The only options are @option{--help} and @option{--version}. @xref{Common
13964 @node hostid invocation
13965 @section @command{hostid}: Print numeric host identifier
13968 @cindex printing the host identifier
13970 @command{hostid} prints the numeric identifier of the current host
13971 in hexadecimal. This command accepts no arguments.
13972 The only options are @option{--help} and @option{--version}.
13973 @xref{Common options}.
13975 For example, here's what it prints on one system I use:
13982 On that system, the 32-bit quantity happens to be closely
13983 related to the system's Internet address, but that isn't always
13988 @node uptime invocation
13989 @section @command{uptime}: Print system uptime and load
13992 @cindex printing the system uptime and load
13994 @command{uptime} prints the current time, the system's uptime, the
13995 number of logged-in users and the current load average.
13997 If an argument is specified, it is used as the file to be read
13998 to discover how many users are logged in. If no argument is
13999 specified, a system default is used (@command{uptime --help} indicates
14000 the default setting).
14002 The only options are @option{--help} and @option{--version}.
14003 @xref{Common options}.
14005 For example, here's what it prints right now on one system I use:
14009 14:07 up 3:35, 3 users, load average: 1.39, 1.15, 1.04
14012 The precise method of calculation of load average varies somewhat
14013 between systems. Some systems calculate it as the average number of
14014 runnable processes over the last 1, 5 and 15 minutes, but some systems
14015 also include processes in the uninterruptible sleep state (that is,
14016 those processes which are waiting for disk I/O). The Linux kernel
14017 includes uninterruptible processes.
14019 @node SELinux context
14020 @chapter SELinux context
14022 @cindex SELinux context
14023 @cindex SELinux, context
14024 @cindex commands for SELinux context
14026 This section describes commands for operations with SELinux
14030 * chcon invocation:: Change SELinux context of file
14031 * runcon invocation:: Run a command in specified SELinux context
14034 @node chcon invocation
14035 @section @command{chcon}: Change SELinux context of file
14038 @cindex changing security context
14039 @cindex change SELinux context
14041 @command{chcon} changes the SELinux security context of the selected files.
14045 chcon [@var{option}]@dots{} @var{context} @var{file}@dots{}
14046 chcon [@var{option}]@dots{} [-u @var{user}] [-r @var{role}] [-l @var{range}] [-t @var{type}] @var{file}@dots{}
14047 chcon [@var{option}]@dots{} --reference=@var{rfile} @var{file}@dots{}
14050 Change the SELinux security context of each @var{file} to @var{context}.
14051 With @option{--reference}, change the security context of each @var{file}
14052 to that of @var{rfile}.
14054 The program accepts the following options. Also see @ref{Common options}.
14059 @itemx --no-dereference
14061 @opindex --no-dereference
14062 @cindex no dereference
14063 Affect symbolic links instead of any referenced file.
14065 @item --reference=@var{rfile}
14066 @opindex --reference
14067 @cindex reference file
14068 Use @var{rfile}'s security context rather than specifying a @var{context} value.
14073 @opindex --recursive
14074 Operate on files and directories recursively.
14077 @xref{Traversing symlinks}.
14080 @xref{Traversing symlinks}.
14083 @xref{Traversing symlinks}.
14090 Output a diagnostic for every file processed.
14092 @item -u @var{user}
14093 @itemx --user=@var{user}
14096 Set user @var{user} in the target security context.
14098 @item -r @var{role}
14099 @itemx --role=@var{role}
14102 Set role @var{role} in the target security context.
14104 @item -t @var{type}
14105 @itemx --type=@var{type}
14108 Set type @var{type} in the target security context.
14110 @item -l @var{range}
14111 @itemx --range=@var{range}
14114 Set range @var{range} in the target security context.
14120 @node runcon invocation
14121 @section @command{runcon}: Run a command in specified SELinux context
14124 @cindex run with security context
14127 @command{runcon} runs file in specified SELinux security context.
14131 runcon @var{context} @var{command} [@var{args}]
14132 runcon [ -c ] [-u @var{user}] [-r @var{role}] [-t @var{type}] [-l @var{range}] @var{command} [@var{args}]
14135 Run @var{command} with completely-specified @var{context}, or with
14136 current or transitioned security context modified by one or more of @var{level},
14137 @var{role}, @var{type} and @var{user}.
14139 If none of @option{-c}, @option{-t}, @option{-u}, @option{-r}, or @option{-l}
14140 is specified, the first argument is used as the complete context.
14141 Any additional arguments after @var{command}
14142 are interpreted as arguments to the command.
14144 With neither @var{context} nor @var{command}, print the current security context.
14146 The program accepts the following options. Also see @ref{Common options}.
14154 Compute process transition context before modifying.
14156 @item -u @var{user}
14157 @itemx --user=@var{user}
14160 Set user @var{user} in the target security context.
14162 @item -r @var{role}
14163 @itemx --role=@var{role}
14166 Set role @var{role} in the target security context.
14168 @item -t @var{type}
14169 @itemx --type=@var{type}
14172 Set type @var{type} in the target security context.
14174 @item -l @var{range}
14175 @itemx --range=@var{range}
14178 Set range @var{range} in the target security context.
14182 @cindex exit status of @command{runcon}
14186 126 if @var{command} is found but cannot be invoked
14187 127 if @command{runcon} itself fails or if @var{command} cannot be found
14188 the exit status of @var{command} otherwise
14191 @node Modified command invocation
14192 @chapter Modified command invocation
14194 @cindex modified command invocation
14195 @cindex invocation of commands, modified
14196 @cindex commands for invoking other commands
14198 This section describes commands that run other commands in some context
14199 different than the current one: a modified environment, as a different
14203 * chroot invocation:: Modify the root directory.
14204 * env invocation:: Modify environment variables.
14205 * nice invocation:: Modify niceness.
14206 * nohup invocation:: Immunize to hangups.
14207 * stdbuf invocation:: Modify buffering of standard streams.
14208 * su invocation:: Modify user and group ID.
14209 * timeout invocation:: Run with time limit.
14213 @node chroot invocation
14214 @section @command{chroot}: Run a command with a different root directory
14217 @cindex running a program in a specified root directory
14218 @cindex root directory, running a program in a specified
14220 @command{chroot} runs a command with a specified root directory.
14221 On many systems, only the super-user can do this.@footnote{However,
14222 some systems (e.g., FreeBSD) can be configured to allow certain regular
14223 users to use the @code{chroot} system call, and hence to run this program.
14224 Also, on Cygwin, anyone can run the @command{chroot} command, because the
14225 underlying function is non-privileged due to lack of support in MS-Windows.}
14229 chroot @var{option} @var{newroot} [@var{command} [@var{args}]@dots{}]
14230 chroot @var{option}
14233 Ordinarily, file names are looked up starting at the root of the
14234 directory structure, i.e., @file{/}. @command{chroot} changes the root to
14235 the directory @var{newroot} (which must exist) and then runs
14236 @var{command} with optional @var{args}. If @var{command} is not
14237 specified, the default is the value of the @env{SHELL} environment
14238 variable or @command{/bin/sh} if not set, invoked with the @option{-i} option.
14239 @var{command} must not be a special built-in utility
14240 (@pxref{Special built-in utilities}).
14242 The program accepts the following options. Also see @ref{Common options}.
14243 Options must precede operands.
14247 @itemx --userspec=@var{user}[:@var{group}]
14248 @opindex --userspec
14249 By default, @var{command} is run with the same credentials
14250 as the invoking process.
14251 Use this option to run it as a different @var{user} and/or with a
14252 different primary @var{group}.
14254 @itemx --groups=@var{groups}
14256 Use this option to specify the supplementary @var{groups} to be
14257 used by the new process.
14258 The items in the list (names or numeric IDs) must be separated by commas.
14262 Here are a few tips to help avoid common problems in using chroot.
14263 To start with a simple example, make @var{command} refer to a statically
14264 linked binary. If you were to use a dynamically linked executable, then
14265 you'd have to arrange to have the shared libraries in the right place under
14266 your new root directory.
14268 For example, if you create a statically linked @command{ls} executable,
14269 and put it in @file{/tmp/empty}, you can run this command as root:
14272 $ chroot /tmp/empty /ls -Rl /
14275 Then you'll see output like this:
14280 -rwxr-xr-x 1 0 0 1041745 Aug 16 11:17 ls
14283 If you want to use a dynamically linked executable, say @command{bash},
14284 then first run @samp{ldd bash} to see what shared objects it needs.
14285 Then, in addition to copying the actual binary, also copy the listed
14286 files to the required positions under your intended new root directory.
14287 Finally, if the executable requires any other files (e.g., data, state,
14288 device files), copy them into place, too.
14290 @cindex exit status of @command{chroot}
14294 1 if @command{chroot} itself fails
14295 126 if @var{command} is found but cannot be invoked
14296 127 if @var{command} cannot be found
14297 the exit status of @var{command} otherwise
14301 @node env invocation
14302 @section @command{env}: Run a command in a modified environment
14305 @cindex environment, running a program in a modified
14306 @cindex modified environment, running a program in a
14307 @cindex running a program in a modified environment
14309 @command{env} runs a command with a modified environment. Synopses:
14312 env [@var{option}]@dots{} [@var{name}=@var{value}]@dots{} @c
14313 [@var{command} [@var{args}]@dots{}]
14317 Operands of the form @samp{@var{variable}=@var{value}} set
14318 the environment variable @var{variable} to value @var{value}.
14319 @var{value} may be empty (@samp{@var{variable}=}). Setting a variable
14320 to an empty value is different from unsetting it.
14321 These operands are evaluated left-to-right, so if two operands
14322 mention the same variable the earlier is ignored.
14324 Environment variable names can be empty, and can contain any
14325 characters other than @samp{=} and @acronym{ASCII} @sc{nul}.
14326 However, it is wise to limit yourself to names that
14327 consist solely of underscores, digits, and @acronym{ASCII} letters,
14328 and that begin with a non-digit, as applications like the shell do not
14329 work well with other names.
14332 The first operand that does not contain the character @samp{=}
14333 specifies the program to invoke; it is
14334 searched for according to the @env{PATH} environment variable. Any
14335 remaining arguments are passed as arguments to that program.
14336 The program should not be a special built-in utility
14337 (@pxref{Special built-in utilities}).
14339 @cindex environment, printing
14341 If no command name is specified following the environment
14342 specifications, the resulting environment is printed. This is like
14343 specifying the @command{printenv} program.
14345 The program accepts the following options. Also see @ref{Common options}.
14346 Options must precede operands.
14350 @item -u @var{name}
14351 @itemx --unset=@var{name}
14354 Remove variable @var{name} from the environment, if it was in the
14359 @itemx --ignore-environment
14362 @opindex --ignore-environment
14363 Start with an empty environment, ignoring the inherited environment.
14367 @cindex exit status of @command{env}
14371 0 if no @var{command} is specified and the environment is output
14372 1 if @command{env} itself fails
14373 126 if @var{command} is found but cannot be invoked
14374 127 if @var{command} cannot be found
14375 the exit status of @var{command} otherwise
14379 @node nice invocation
14380 @section @command{nice}: Run a command with modified niceness
14384 @cindex scheduling, affecting
14385 @cindex appropriate privileges
14387 @command{nice} prints or modifies a process's @dfn{niceness},
14388 a parameter that affects whether the process is scheduled favorably.
14392 nice [@var{option}]@dots{} [@var{command} [@var{arg}]@dots{}]
14395 If no arguments are given, @command{nice} prints the current niceness.
14396 Otherwise, @command{nice} runs the given @var{command} with its
14397 niceness adjusted. By default, its niceness is incremented by 10.
14399 Niceness values range at least from @minus{}20 (process has high priority
14400 and gets more resources, thus slowing down other processes) through 19
14401 (process has lower priority and runs slowly itself, but has less impact
14402 on the speed of other running processes). Some systems
14403 may have a wider range of nicenesses; conversely, other systems may
14404 enforce more restrictive limits. An attempt to set the niceness
14405 outside the supported range is treated as an attempt to use the
14406 minimum or maximum supported value.
14408 A niceness should not be confused with a scheduling priority, which
14409 lets applications determine the order in which threads are scheduled
14410 to run. Unlike a priority, a niceness is merely advice to the
14411 scheduler, which the scheduler is free to ignore. Also, as a point of
14412 terminology, @acronym{POSIX} defines the behavior of @command{nice} in
14413 terms of a @dfn{nice value}, which is the nonnegative difference
14414 between a niceness and the minimum niceness. Though @command{nice}
14415 conforms to @acronym{POSIX}, its documentation and diagnostics use the
14416 term ``niceness'' for compatibility with historical practice.
14418 @var{command} must not be a special built-in utility (@pxref{Special
14419 built-in utilities}).
14421 @mayConflictWithShellBuiltIn{nice}
14423 The program accepts the following option. Also see @ref{Common options}.
14424 Options must precede operands.
14427 @item -n @var{adjustment}
14428 @itemx --adjustment=@var{adjustment}
14430 @opindex --adjustment
14431 Add @var{adjustment} instead of 10 to the command's niceness. If
14432 @var{adjustment} is negative and you lack appropriate privileges,
14433 @command{nice} issues a warning but otherwise acts as if you specified
14436 For compatibility @command{nice} also supports an obsolete
14437 option syntax @option{-@var{adjustment}}. New scripts should use
14438 @option{-n @var{adjustment}} instead.
14442 @cindex exit status of @command{nice}
14446 0 if no @var{command} is specified and the niceness is output
14447 1 if @command{nice} itself fails
14448 126 if @var{command} is found but cannot be invoked
14449 127 if @var{command} cannot be found
14450 the exit status of @var{command} otherwise
14453 It is sometimes useful to run a non-interactive program with reduced niceness.
14456 $ nice factor 4611686018427387903
14459 Since @command{nice} prints the current niceness,
14460 you can invoke it through itself to demonstrate how it works.
14462 The default behavior is to increase the niceness by @samp{10}:
14473 The @var{adjustment} is relative to the current niceness. In the
14474 next example, the first @command{nice} invocation runs the second one
14475 with niceness 10, and it in turn runs the final one with a niceness
14479 $ nice nice -n 3 nice
14483 Specifying a niceness larger than the supported range
14484 is the same as specifying the maximum supported value:
14487 $ nice -n 10000000000 nice
14491 Only a privileged user may run a process with lower niceness:
14495 nice: cannot set niceness: Permission denied
14497 $ sudo nice -n -1 nice
14502 @node nohup invocation
14503 @section @command{nohup}: Run a command immune to hangups
14506 @cindex hangups, immunity to
14507 @cindex immunity to hangups
14508 @cindex logging out and continuing to run
14511 @command{nohup} runs the given @var{command} with hangup signals ignored,
14512 so that the command can continue running in the background after you log
14516 nohup @var{command} [@var{arg}]@dots{}
14519 If standard input is a terminal, it is redirected from
14520 @file{/dev/null} so that terminal sessions do not mistakenly consider
14521 the terminal to be used by the command. This is a @acronym{GNU}
14522 extension; programs intended to be portable to non-@acronym{GNU} hosts
14523 should use @samp{nohup @var{command} [@var{arg}]@dots{} </dev/null}
14527 If standard output is a terminal, the command's standard output is appended
14528 to the file @file{nohup.out}; if that cannot be written to, it is appended
14529 to the file @file{$HOME/nohup.out}; and if that cannot be written to, the
14530 command is not run.
14531 Any @file{nohup.out} or @file{$HOME/nohup.out} file created by
14532 @command{nohup} is made readable and writable only to the user,
14533 regardless of the current umask settings.
14535 If standard error is a terminal, it is normally redirected to the same file
14536 descriptor as the (possibly-redirected) standard output.
14537 However, if standard output is closed, standard error terminal output
14538 is instead appended to the file @file{nohup.out} or
14539 @file{$HOME/nohup.out} as above.
14541 To capture the command's output to a file other than @file{nohup.out}
14542 you can redirect it. For example, to capture the output of
14546 nohup make > make.log
14549 @command{nohup} does not automatically put the command it runs in the
14550 background; you must do that explicitly, by ending the command line
14551 with an @samp{&}. Also, @command{nohup} does not alter the
14552 niceness of @var{command}; use @command{nice} for that,
14553 e.g., @samp{nohup nice @var{command}}.
14555 @var{command} must not be a special built-in utility (@pxref{Special
14556 built-in utilities}).
14558 The only options are @option{--help} and @option{--version}. @xref{Common
14559 options}. Options must precede operands.
14561 @cindex exit status of @command{nohup}
14565 126 if @var{command} is found but cannot be invoked
14566 127 if @command{nohup} itself fails or if @var{command} cannot be found
14567 the exit status of @var{command} otherwise
14571 @node stdbuf invocation
14572 @section @command{stdbuf}: Run a command with modified I/O stream buffering
14575 @cindex standard streams, buffering
14576 @cindex line buffered
14578 @command{stdbuf} allows one to modify the buffering operations of the
14579 three standard I/O streams associated with a program. Synopsis:
14582 stdbuf @var{option}@dots{} @var{command}
14585 Any additional @var{arg}s are passed as additional arguments to the
14588 The program accepts the following options. Also see @ref{Common options}.
14592 @item -i @var{mode}
14593 @itemx --input=@var{mode}
14596 Adjust the standard input stream buffering.
14598 @item -o @var{mode}
14599 @itemx --output=@var{mode}
14602 Adjust the standard output stream buffering.
14604 @item -e @var{mode}
14605 @itemx --error=@var{mode}
14608 Adjust the standard error stream buffering.
14612 The @var{mode} can be specified as follows:
14617 Set the stream to line buffered mode.
14618 In this mode data is coalesced until a newline is output or
14619 input is read from any stream attached to a terminal device.
14620 This option is invalid with standard input.
14623 Disable buffering of the selected stream.
14624 In this mode data is output immediately and only the
14625 amount of data requested is read from input.
14628 Specify the size of the buffer to use in fully buffered mode.
14629 @multiplierSuffixesNoBlocks{size}
14633 NOTE: If @var{command} adjusts the buffering of its standard streams
14634 (@command{tee} does for e.g.) then that will override corresponding settings
14635 changed by @command{stdbuf}. Also some filters (like @command{dd} and
14636 @command{cat} etc.) don't use streams for I/O, and are thus unaffected
14637 by @command{stdbuf} settings.
14639 @cindex exit status of @command{stdbuf}
14643 125 if @command{stdbuf} itself fails
14644 126 if @var{command} is found but cannot be invoked
14645 127 if @var{command} cannot be found
14646 the exit status of @var{command} otherwise
14650 @node su invocation
14651 @section @command{su}: Run a command with substitute user and group ID
14654 @cindex substitute user and group IDs
14655 @cindex user ID, switching
14656 @cindex super-user, becoming
14657 @cindex root, becoming
14659 @command{su} allows one user to temporarily become another user. It runs a
14660 command (often an interactive shell) with the real and effective user
14661 ID, group ID, and supplemental groups of a given @var{user}. Synopsis:
14664 su [@var{option}]@dots{} [@var{user} [@var{arg}]@dots{}]
14667 @cindex passwd entry, and @command{su} shell
14669 @flindex /etc/passwd
14670 If no @var{user} is given, the default is @code{root}, the super-user.
14671 The shell to use is taken from @var{user}'s @code{passwd} entry, or
14672 @file{/bin/sh} if none is specified there. If @var{user} has a
14673 password, @command{su} prompts for the password unless run by a user with
14674 effective user ID of zero (the super-user).
14680 @cindex login shell
14681 By default, @command{su} does not change the current directory.
14682 It sets the environment variables @env{HOME} and @env{SHELL}
14683 from the password entry for @var{user}, and if @var{user} is not
14684 the super-user, sets @env{USER} and @env{LOGNAME} to @var{user}.
14685 By default, the shell is not a login shell.
14687 Any additional @var{arg}s are passed as additional arguments to the
14690 @cindex @option{-su}
14691 GNU @command{su} does not treat @file{/bin/sh} or any other shells specially
14692 (e.g., by setting @code{argv[0]} to @option{-su}, passing @option{-c} only
14693 to certain shells, etc.).
14696 @command{su} can optionally be compiled to use @code{syslog} to report
14697 failed, and optionally successful, @command{su} attempts. (If the system
14698 supports @code{syslog}.) However, GNU @command{su} does not check if the
14699 user is a member of the @code{wheel} group; see below.
14701 The program accepts the following options. Also see @ref{Common options}.
14704 @item -c @var{command}
14705 @itemx --command=@var{command}
14708 Pass @var{command}, a single command line to run, to the shell with
14709 a @option{-c} option instead of starting an interactive shell.
14716 @cindex file name pattern expansion, disabled
14717 @cindex globbing, disabled
14718 Pass the @option{-f} option to the shell. This probably only makes sense
14719 if the shell run is @command{csh} or @command{tcsh}, for which the @option{-f}
14720 option prevents reading the startup file (@file{.cshrc}). With
14721 Bourne-like shells, the @option{-f} option disables file name pattern
14722 expansion (globbing), which is not likely to be useful.
14730 @c other variables already indexed above
14733 @cindex login shell, creating
14734 Make the shell a login shell. This means the following. Unset all
14735 environment variables except @env{TERM}, @env{HOME}, and @env{SHELL}
14736 (which are set as described above), and @env{USER} and @env{LOGNAME}
14737 (which are set, even for the super-user, as described above), and set
14738 @env{PATH} to a compiled-in default value. Change to @var{user}'s home
14739 directory. Prepend @samp{-} to the shell's name, intended to make it
14740 read its login startup file(s).
14744 @itemx --preserve-environment
14747 @opindex --preserve-environment
14748 @cindex environment, preserving
14749 @flindex /etc/shells
14750 @cindex restricted shell
14751 Do not change the environment variables @env{HOME}, @env{USER},
14752 @env{LOGNAME}, or @env{SHELL}. Run the shell given in the environment
14753 variable @env{SHELL} instead of the shell from @var{user}'s passwd
14754 entry, unless the user running @command{su} is not the super-user and
14755 @var{user}'s shell is restricted. A @dfn{restricted shell} is one that
14756 is not listed in the file @file{/etc/shells}, or in a compiled-in list
14757 if that file does not exist. Parts of what this option does can be
14758 overridden by @option{--login} and @option{--shell}.
14760 @item -s @var{shell}
14761 @itemx --shell=@var{shell}
14764 Run @var{shell} instead of the shell from @var{user}'s passwd entry,
14765 unless the user running @command{su} is not the super-user and @var{user}'s
14766 shell is restricted (see @option{-m} just above).
14770 @cindex exit status of @command{su}
14774 1 if @command{su} itself fails
14775 126 if subshell is found but cannot be invoked
14776 127 if subshell cannot be found
14777 the exit status of the subshell otherwise
14780 @cindex wheel group, not supported
14781 @cindex group wheel, not supported
14783 @subsection Why GNU @command{su} does not support the @samp{wheel} group
14785 (This section is by Richard Stallman.)
14789 Sometimes a few of the users try to hold total power over all the
14790 rest. For example, in 1984, a few users at the MIT AI lab decided to
14791 seize power by changing the operator password on the Twenex system and
14792 keeping it secret from everyone else. (I was able to thwart this coup
14793 and give power back to the users by patching the kernel, but I
14794 wouldn't know how to do that in Unix.)
14796 However, occasionally the rulers do tell someone. Under the usual
14797 @command{su} mechanism, once someone learns the root password who
14798 sympathizes with the ordinary users, he or she can tell the rest. The
14799 ``wheel group'' feature would make this impossible, and thus cement the
14800 power of the rulers.
14802 I'm on the side of the masses, not that of the rulers. If you are
14803 used to supporting the bosses and sysadmins in whatever they do, you
14804 might find this idea strange at first.
14807 @node timeout invocation
14808 @section @command{timeout}: Run a command with a time limit
14812 @cindex run commands with bounded time
14814 @command{timeout} runs the given @var{command} and kills it if it is
14815 still running after the specified time interval. Synopsis:
14818 timeout [@var{option}] @var{number}[smhd] @var{command} [@var{arg}]@dots{}
14822 @var{number} is an integer followed by an optional unit; the default
14823 is seconds. The units are:
14836 @var{command} must not be a special built-in utility (@pxref{Special
14837 built-in utilities}).
14839 The program accepts the following option. Also see @ref{Common options}.
14840 Options must precede operands.
14843 @item -s @var{signal}
14844 @itemx --signal=@var{signal}
14847 Send this @var{signal} to @var{command} on timeout, rather than the
14848 default @samp{TERM} signal. @var{signal} may be a name like @samp{HUP}
14849 or a number. Also see @xref{Signal specifications}.
14853 @cindex exit status of @command{timeout}
14857 124 if @var{command} times out
14858 125 if @command{timeout} itself fails
14859 126 if @var{command} is found but cannot be invoked
14860 127 if @var{command} cannot be found
14861 the exit status of @var{command} otherwise
14865 @node Process control
14866 @chapter Process control
14868 @cindex processes, commands for controlling
14869 @cindex commands for controlling processes
14872 * kill invocation:: Sending a signal to processes.
14876 @node kill invocation
14877 @section @command{kill}: Send a signal to processes
14880 @cindex send a signal to processes
14882 The @command{kill} command sends a signal to processes, causing them
14883 to terminate or otherwise act upon receiving the signal in some way.
14884 Alternatively, it lists information about signals. Synopses:
14887 kill [-s @var{signal} | --signal @var{signal} | -@var{signal}] @var{pid}@dots{}
14888 kill [-l | --list | -t | --table] [@var{signal}]@dots{}
14891 @mayConflictWithShellBuiltIn{kill}
14893 The first form of the @command{kill} command sends a signal to all
14894 @var{pid} arguments. The default signal to send if none is specified
14895 is @samp{TERM}. The special signal number @samp{0} does not denote a
14896 valid signal, but can be used to test whether the @var{pid} arguments
14897 specify processes to which a signal could be sent.
14899 If @var{pid} is positive, the signal is sent to the process with the
14900 process ID @var{pid}. If @var{pid} is zero, the signal is sent to all
14901 processes in the process group of the current process. If @var{pid}
14902 is @minus{}1, the signal is sent to all processes for which the user has
14903 permission to send a signal. If @var{pid} is less than @minus{}1, the signal
14904 is sent to all processes in the process group that equals the absolute
14905 value of @var{pid}.
14907 If @var{pid} is not positive, a system-dependent set of system
14908 processes is excluded from the list of processes to which the signal
14911 If a negative @var{pid} argument is desired as the first one, it
14912 should be preceded by @option{--}. However, as a common extension to
14913 @acronym{POSIX}, @option{--} is not required with @samp{kill
14914 -@var{signal} -@var{pid}}. The following commands are equivalent:
14923 The first form of the @command{kill} command succeeds if every @var{pid}
14924 argument specifies at least one process that the signal was sent to.
14926 The second form of the @command{kill} command lists signal information.
14927 Either the @option{-l} or @option{--list} option, or the @option{-t}
14928 or @option{--table} option must be specified. Without any
14929 @var{signal} argument, all supported signals are listed. The output
14930 of @option{-l} or @option{--list} is a list of the signal names, one
14931 per line; if @var{signal} is already a name, the signal number is
14932 printed instead. The output of @option{-t} or @option{--table} is a
14933 table of signal numbers, names, and descriptions. This form of the
14934 @command{kill} command succeeds if all @var{signal} arguments are valid
14935 and if there is no output error.
14937 The @command{kill} command also supports the @option{--help} and
14938 @option{--version} options. @xref{Common options}.
14940 A @var{signal} may be a signal name like @samp{HUP}, or a signal
14941 number like @samp{1}, or an exit status of a process terminated by the
14942 signal. A signal name can be given in canonical form or prefixed by
14943 @samp{SIG}. The case of the letters is ignored, except for the
14944 @option{-@var{signal}} option which must use upper case to avoid
14945 ambiguity with lower case option letters. For a list of supported
14946 signal names and numbers see @xref{Signal specifications}.
14951 @cindex delaying commands
14952 @cindex commands for delaying
14954 @c Perhaps @command{wait} or other commands should be described here also?
14957 * sleep invocation:: Delay for a specified time.
14961 @node sleep invocation
14962 @section @command{sleep}: Delay for a specified time
14965 @cindex delay for a specified time
14967 @command{sleep} pauses for an amount of time specified by the sum of
14968 the values of the command line arguments.
14972 sleep @var{number}[smhd]@dots{}
14976 Each argument is a number followed by an optional unit; the default
14977 is seconds. The units are:
14990 Historical implementations of @command{sleep} have required that
14991 @var{number} be an integer, and only accepted a single argument
14992 without a suffix. However, GNU @command{sleep} accepts
14993 arbitrary floating point numbers (using a period before any fractional
14996 The only options are @option{--help} and @option{--version}. @xref{Common
14999 @c sleep is a shell built-in at least with Solaris 11's /bin/sh
15000 @mayConflictWithShellBuiltIn{sleep}
15005 @node Numeric operations
15006 @chapter Numeric operations
15008 @cindex numeric operations
15009 These programs do numerically-related operations.
15012 * factor invocation:: Show factors of numbers.
15013 * seq invocation:: Print sequences of numbers.
15017 @node factor invocation
15018 @section @command{factor}: Print prime factors
15021 @cindex prime factors
15023 @command{factor} prints prime factors. Synopses:
15026 factor [@var{number}]@dots{}
15027 factor @var{option}
15030 If no @var{number} is specified on the command line, @command{factor} reads
15031 numbers from standard input, delimited by newlines, tabs, or spaces.
15033 The @command{factor} command supports only a small number of options:
15037 Print a short help on standard output, then exit without further
15041 Print the program version on standard output, then exit without further
15045 Factoring the product of the eighth and ninth Mersenne primes
15046 takes about 30 milliseconds of CPU time on a 2.2 GHz Athlon.
15049 M8=`echo 2^31-1|bc` ; M9=`echo 2^61-1|bc`
15050 /usr/bin/time -f '%U' factor $(echo "$M8 * $M9" | bc)
15051 4951760154835678088235319297: 2147483647 2305843009213693951
15055 Similarly, factoring the eighth Fermat number @math{2^{256}+1} takes
15056 about 20 seconds on the same machine.
15058 Factoring large prime numbers is, in general, hard. The Pollard Rho
15059 algorithm used by @command{factor} is particularly effective for
15060 numbers with relatively small factors. If you wish to factor large
15061 numbers which do not have small factors (for example, numbers which
15062 are the product of two large primes), other methods are far better.
15064 If @command{factor} is built without using GNU MP, only
15065 single-precision arithmetic is available, and so large numbers
15066 (typically @math{2^{64}} and above) will not be supported. The single-precision
15067 code uses an algorithm which is designed for factoring smaller
15073 @node seq invocation
15074 @section @command{seq}: Print numeric sequences
15077 @cindex numeric sequences
15078 @cindex sequence of numbers
15080 @command{seq} prints a sequence of numbers to standard output. Synopses:
15083 seq [@var{option}]@dots{} @var{last}
15084 seq [@var{option}]@dots{} @var{first} @var{last}
15085 seq [@var{option}]@dots{} @var{first} @var{increment} @var{last}
15088 @command{seq} prints the numbers from @var{first} to @var{last} by
15089 @var{increment}. By default, each number is printed on a separate line.
15090 When @var{increment} is not specified, it defaults to @samp{1},
15091 even when @var{first} is larger than @var{last}.
15092 @var{first} also defaults to @samp{1}. So @code{seq 1} prints
15093 @samp{1}, but @code{seq 0} and @code{seq 10 5} produce no output.
15094 Floating-point numbers
15095 may be specified (using a period before any fractional digits).
15097 The program accepts the following options. Also see @ref{Common options}.
15098 Options must precede operands.
15101 @item -f @var{format}
15102 @itemx --format=@var{format}
15103 @opindex -f @var{format}
15104 @opindex --format=@var{format}
15105 @cindex formatting of numbers in @command{seq}
15106 Print all numbers using @var{format}.
15107 @var{format} must contain exactly one of the @samp{printf}-style
15108 floating point conversion specifications @samp{%a}, @samp{%e},
15109 @samp{%f}, @samp{%g}, @samp{%A}, @samp{%E}, @samp{%F}, @samp{%G}.
15110 The @samp{%} may be followed by zero or more flags taken from the set
15111 @samp{-+#0 '}, then an optional width containing one or more digits,
15112 then an optional precision consisting of a @samp{.} followed by zero
15113 or more digits. @var{format} may also contain any number of @samp{%%}
15114 conversion specifications. All conversion specifications have the
15115 same meaning as with @samp{printf}.
15117 The default format is derived from @var{first}, @var{step}, and
15118 @var{last}. If these all use a fixed point decimal representation,
15119 the default format is @samp{%.@var{p}f}, where @var{p} is the minimum
15120 precision that can represent the output numbers exactly. Otherwise,
15121 the default format is @samp{%g}.
15123 @item -s @var{string}
15124 @itemx --separator=@var{string}
15125 @cindex separator for numbers in @command{seq}
15126 Separate numbers with @var{string}; default is a newline.
15127 The output always terminates with a newline.
15130 @itemx --equal-width
15131 Print all numbers with the same width, by padding with leading zeros.
15132 @var{first}, @var{step}, and @var{last} should all use a fixed point
15133 decimal representation.
15134 (To have other kinds of padding, use @option{--format}).
15138 You can get finer-grained control over output with @option{-f}:
15141 $ seq -f '(%9.2E)' -9e5 1.1e6 1.3e6
15147 If you want hexadecimal integer output, you can use @command{printf}
15148 to perform the conversion:
15151 $ printf '%x\n' `seq 1048575 1024 1050623`
15157 For very long lists of numbers, use xargs to avoid
15158 system limitations on the length of an argument list:
15161 $ seq 1000000 | xargs printf '%x\n' | tail -n 3
15167 To generate octal output, use the printf @code{%o} format instead
15170 On most systems, seq can produce whole-number output for values up to
15171 at least @math{2^{53}}. Larger integers are approximated. The details
15172 differ depending on your floating-point implementation, but a common
15173 case is that @command{seq} works with integers through @math{2^{64}},
15174 and larger integers may not be numerically correct:
15177 $ seq 18446744073709551616 1 18446744073709551618
15178 18446744073709551616
15179 18446744073709551616
15180 18446744073709551618
15183 Be careful when using @command{seq} with outlandish values: otherwise
15184 you may see surprising results, as @command{seq} uses floating point
15185 internally. For example, on the x86 platform, where the internal
15186 representation uses a 64-bit fraction, the command:
15189 seq 1 0.0000000000000000001 1.0000000000000000009
15192 outputs 1.0000000000000000007 twice and skips 1.0000000000000000008.
15197 @node File permissions
15198 @chapter File permissions
15201 @include getdate.texi
15205 @node Opening the software toolbox
15206 @chapter Opening the Software Toolbox
15208 An earlier version of this chapter appeared in
15209 @uref{http://www.linuxjournal.com/article.php?sid=2762, the
15210 @cite{What's GNU?} column of @cite{Linux Journal}, 2 (June, 1994)}.
15211 It was written by Arnold Robbins.
15214 * Toolbox introduction:: Toolbox introduction
15215 * I/O redirection:: I/O redirection
15216 * The who command:: The @command{who} command
15217 * The cut command:: The @command{cut} command
15218 * The sort command:: The @command{sort} command
15219 * The uniq command:: The @command{uniq} command
15220 * Putting the tools together:: Putting the tools together
15224 @node Toolbox introduction
15225 @unnumberedsec Toolbox Introduction
15227 This month's column is only peripherally related to the GNU Project, in
15228 that it describes a number of the GNU tools on your GNU/Linux system and how they
15229 might be used. What it's really about is the ``Software Tools'' philosophy
15230 of program development and usage.
15232 The software tools philosophy was an important and integral concept
15233 in the initial design and development of Unix (of which Linux and GNU are
15234 essentially clones). Unfortunately, in the modern day press of
15235 Internetworking and flashy GUIs, it seems to have fallen by the
15236 wayside. This is a shame, since it provides a powerful mental model
15237 for solving many kinds of problems.
15239 Many people carry a Swiss Army knife around in their pants pockets (or
15240 purse). A Swiss Army knife is a handy tool to have: it has several knife
15241 blades, a screwdriver, tweezers, toothpick, nail file, corkscrew, and perhaps
15242 a number of other things on it. For the everyday, small miscellaneous jobs
15243 where you need a simple, general purpose tool, it's just the thing.
15245 On the other hand, an experienced carpenter doesn't build a house using
15246 a Swiss Army knife. Instead, he has a toolbox chock full of specialized
15247 tools---a saw, a hammer, a screwdriver, a plane, and so on. And he knows
15248 exactly when and where to use each tool; you won't catch him hammering nails
15249 with the handle of his screwdriver.
15251 The Unix developers at Bell Labs were all professional programmers and trained
15252 computer scientists. They had found that while a one-size-fits-all program
15253 might appeal to a user because there's only one program to use, in practice
15258 difficult to write,
15261 difficult to maintain and
15265 difficult to extend to meet new situations.
15268 Instead, they felt that programs should be specialized tools. In short, each
15269 program ``should do one thing well.'' No more and no less. Such programs are
15270 simpler to design, write, and get right---they only do one thing.
15272 Furthermore, they found that with the right machinery for hooking programs
15273 together, that the whole was greater than the sum of the parts. By combining
15274 several special purpose programs, you could accomplish a specific task
15275 that none of the programs was designed for, and accomplish it much more
15276 quickly and easily than if you had to write a special purpose program.
15277 We will see some (classic) examples of this further on in the column.
15278 (An important additional point was that, if necessary, take a detour
15279 and build any software tools you may need first, if you don't already
15280 have something appropriate in the toolbox.)
15282 @node I/O redirection
15283 @unnumberedsec I/O Redirection
15285 Hopefully, you are familiar with the basics of I/O redirection in the
15286 shell, in particular the concepts of ``standard input,'' ``standard output,''
15287 and ``standard error''. Briefly, ``standard input'' is a data source, where
15288 data comes from. A program should not need to either know or care if the
15289 data source is a disk file, a keyboard, a magnetic tape, or even a punched
15290 card reader. Similarly, ``standard output'' is a data sink, where data goes
15291 to. The program should neither know nor care where this might be.
15292 Programs that only read their standard input, do something to the data,
15293 and then send it on, are called @dfn{filters}, by analogy to filters in a
15296 With the Unix shell, it's very easy to set up data pipelines:
15299 program_to_create_data | filter1 | ... | filterN > final.pretty.data
15302 We start out by creating the raw data; each filter applies some successive
15303 transformation to the data, until by the time it comes out of the pipeline,
15304 it is in the desired form.
15306 This is fine and good for standard input and standard output. Where does the
15307 standard error come in to play? Well, think about @command{filter1} in
15308 the pipeline above. What happens if it encounters an error in the data it
15309 sees? If it writes an error message to standard output, it will just
15310 disappear down the pipeline into @command{filter2}'s input, and the
15311 user will probably never see it. So programs need a place where they can send
15312 error messages so that the user will notice them. This is standard error,
15313 and it is usually connected to your console or window, even if you have
15314 redirected standard output of your program away from your screen.
15316 For filter programs to work together, the format of the data has to be
15317 agreed upon. The most straightforward and easiest format to use is simply
15318 lines of text. Unix data files are generally just streams of bytes, with
15319 lines delimited by the @acronym{ASCII} @sc{lf} (Line Feed) character,
15320 conventionally called a ``newline'' in the Unix literature. (This is
15321 @code{'\n'} if you're a C programmer.) This is the format used by all
15322 the traditional filtering programs. (Many earlier operating systems
15323 had elaborate facilities and special purpose programs for managing
15324 binary data. Unix has always shied away from such things, under the
15325 philosophy that it's easiest to simply be able to view and edit your
15326 data with a text editor.)
15328 OK, enough introduction. Let's take a look at some of the tools, and then
15329 we'll see how to hook them together in interesting ways. In the following
15330 discussion, we will only present those command line options that interest
15331 us. As you should always do, double check your system documentation
15332 for the full story.
15334 @node The who command
15335 @unnumberedsec The @command{who} Command
15337 The first program is the @command{who} command. By itself, it generates a
15338 list of the users who are currently logged in. Although I'm writing
15339 this on a single-user system, we'll pretend that several people are
15344 @print{} arnold console Jan 22 19:57
15345 @print{} miriam ttyp0 Jan 23 14:19(:0.0)
15346 @print{} bill ttyp1 Jan 21 09:32(:0.0)
15347 @print{} arnold ttyp2 Jan 23 20:48(:0.0)
15350 Here, the @samp{$} is the usual shell prompt, at which I typed @samp{who}.
15351 There are three people logged in, and I am logged in twice. On traditional
15352 Unix systems, user names are never more than eight characters long. This
15353 little bit of trivia will be useful later. The output of @command{who} is nice,
15354 but the data is not all that exciting.
15356 @node The cut command
15357 @unnumberedsec The @command{cut} Command
15359 The next program we'll look at is the @command{cut} command. This program
15360 cuts out columns or fields of input data. For example, we can tell it
15361 to print just the login name and full name from the @file{/etc/passwd}
15362 file. The @file{/etc/passwd} file has seven fields, separated by
15366 arnold:xyzzy:2076:10:Arnold D. Robbins:/home/arnold:/bin/bash
15369 To get the first and fifth fields, we would use @command{cut} like this:
15372 $ cut -d: -f1,5 /etc/passwd
15373 @print{} root:Operator
15375 @print{} arnold:Arnold D. Robbins
15376 @print{} miriam:Miriam A. Robbins
15380 With the @option{-c} option, @command{cut} will cut out specific characters
15381 (i.e., columns) in the input lines. This is useful for input data
15382 that has fixed width fields, and does not have a field separator. For
15383 example, list the Monday dates for the current month:
15385 @c Is using cal ok? Looked at gcal, but I don't like it.
15396 @node The sort command
15397 @unnumberedsec The @command{sort} Command
15399 Next we'll look at the @command{sort} command. This is one of the most
15400 powerful commands on a Unix-style system; one that you will often find
15401 yourself using when setting up fancy data plumbing.
15404 command reads and sorts each file named on the command line. It then
15405 merges the sorted data and writes it to standard output. It will read
15406 standard input if no files are given on the command line (thus
15407 making it into a filter). The sort is based on the character collating
15408 sequence or based on user-supplied ordering criteria.
15411 @node The uniq command
15412 @unnumberedsec The @command{uniq} Command
15414 Finally (at least for now), we'll look at the @command{uniq} program. When
15415 sorting data, you will often end up with duplicate lines, lines that
15416 are identical. Usually, all you need is one instance of each line.
15417 This is where @command{uniq} comes in. The @command{uniq} program reads its
15418 standard input. It prints only one
15419 copy of each repeated line. It does have several options. Later on,
15420 we'll use the @option{-c} option, which prints each unique line, preceded
15421 by a count of the number of times that line occurred in the input.
15424 @node Putting the tools together
15425 @unnumberedsec Putting the Tools Together
15427 Now, let's suppose this is a large ISP server system with dozens of users
15428 logged in. The management wants the system administrator to write a program that will
15429 generate a sorted list of logged in users. Furthermore, even if a user
15430 is logged in multiple times, his or her name should only show up in the
15433 The administrator could sit down with the system documentation and write a C
15434 program that did this. It would take perhaps a couple of hundred lines
15435 of code and about two hours to write it, test it, and debug it.
15436 However, knowing the software toolbox, the administrator can instead start out
15437 by generating just a list of logged on users:
15447 Next, sort the list:
15450 $ who | cut -c1-8 | sort
15457 Finally, run the sorted list through @command{uniq}, to weed out duplicates:
15460 $ who | cut -c1-8 | sort | uniq
15466 The @command{sort} command actually has a @option{-u} option that does what
15467 @command{uniq} does. However, @command{uniq} has other uses for which one
15468 cannot substitute @samp{sort -u}.
15470 The administrator puts this pipeline into a shell script, and makes it available for
15471 all the users on the system (@samp{#} is the system administrator,
15472 or @code{root}, prompt):
15475 # cat > /usr/local/bin/listusers
15476 who | cut -c1-8 | sort | uniq
15478 # chmod +x /usr/local/bin/listusers
15481 There are four major points to note here. First, with just four
15482 programs, on one command line, the administrator was able to save about two
15483 hours worth of work. Furthermore, the shell pipeline is just about as
15484 efficient as the C program would be, and it is much more efficient in
15485 terms of programmer time. People time is much more expensive than
15486 computer time, and in our modern ``there's never enough time to do
15487 everything'' society, saving two hours of programmer time is no mean
15490 Second, it is also important to emphasize that with the
15491 @emph{combination} of the tools, it is possible to do a special
15492 purpose job never imagined by the authors of the individual programs.
15494 Third, it is also valuable to build up your pipeline in stages, as we did here.
15495 This allows you to view the data at each stage in the pipeline, which helps
15496 you acquire the confidence that you are indeed using these tools correctly.
15498 Finally, by bundling the pipeline in a shell script, other users can use
15499 your command, without having to remember the fancy plumbing you set up for
15500 them. In terms of how you run them, shell scripts and compiled programs are
15503 After the previous warm-up exercise, we'll look at two additional, more
15504 complicated pipelines. For them, we need to introduce two more tools.
15506 The first is the @command{tr} command, which stands for ``transliterate.''
15507 The @command{tr} command works on a character-by-character basis, changing
15508 characters. Normally it is used for things like mapping upper case to
15512 $ echo ThIs ExAmPlE HaS MIXED case! | tr '[:upper:]' '[:lower:]'
15513 @print{} this example has mixed case!
15516 There are several options of interest:
15520 work on the complement of the listed characters, i.e.,
15521 operations apply to characters not in the given set
15524 delete characters in the first set from the output
15527 squeeze repeated characters in the output into just one character.
15530 We will be using all three options in a moment.
15532 The other command we'll look at is @command{comm}. The @command{comm}
15533 command takes two sorted input files as input data, and prints out the
15534 files' lines in three columns. The output columns are the data lines
15535 unique to the first file, the data lines unique to the second file, and
15536 the data lines that are common to both. The @option{-1}, @option{-2}, and
15537 @option{-3} command line options @emph{omit} the respective columns. (This is
15538 non-intuitive and takes a little getting used to.) For example:
15560 The file name @file{-} tells @command{comm} to read standard input
15561 instead of a regular file.
15563 Now we're ready to build a fancy pipeline. The first application is a word
15564 frequency counter. This helps an author determine if he or she is over-using
15567 The first step is to change the case of all the letters in our input file
15568 to one case. ``The'' and ``the'' are the same word when doing counting.
15571 $ tr '[:upper:]' '[:lower:]' < whats.gnu | ...
15574 The next step is to get rid of punctuation. Quoted words and unquoted words
15575 should be treated identically; it's easiest to just get the punctuation out of
15579 $ tr '[:upper:]' '[:lower:]' < whats.gnu | tr -cd '[:alnum:]_ \n' | ...
15582 The second @command{tr} command operates on the complement of the listed
15583 characters, which are all the letters, the digits, the underscore, and
15584 the blank. The @samp{\n} represents the newline character; it has to
15585 be left alone. (The @acronym{ASCII} tab character should also be included for
15586 good measure in a production script.)
15588 At this point, we have data consisting of words separated by blank space.
15589 The words only contain alphanumeric characters (and the underscore). The
15590 next step is break the data apart so that we have one word per line. This
15591 makes the counting operation much easier, as we will see shortly.
15594 $ tr '[:upper:]' '[:lower:]' < whats.gnu | tr -cd '[:alnum:]_ \n' |
15595 > tr -s ' ' '\n' | ...
15598 This command turns blanks into newlines. The @option{-s} option squeezes
15599 multiple newline characters in the output into just one. This helps us
15600 avoid blank lines. (The @samp{>} is the shell's ``secondary prompt.''
15601 This is what the shell prints when it notices you haven't finished
15602 typing in all of a command.)
15604 We now have data consisting of one word per line, no punctuation, all one
15605 case. We're ready to count each word:
15608 $ tr '[:upper:]' '[:lower:]' < whats.gnu | tr -cd '[:alnum:]_ \n' |
15609 > tr -s ' ' '\n' | sort | uniq -c | ...
15612 At this point, the data might look something like this:
15625 The output is sorted by word, not by count! What we want is the most
15626 frequently used words first. Fortunately, this is easy to accomplish,
15627 with the help of two more @command{sort} options:
15631 do a numeric sort, not a textual one
15634 reverse the order of the sort
15637 The final pipeline looks like this:
15640 $ tr '[:upper:]' '[:lower:]' < whats.gnu | tr -cd '[:alnum:]_ \n' |
15641 > tr -s ' ' '\n' | sort | uniq -c | sort -n -r
15650 Whew! That's a lot to digest. Yet, the same principles apply. With six
15651 commands, on two lines (really one long one split for convenience), we've
15652 created a program that does something interesting and useful, in much
15653 less time than we could have written a C program to do the same thing.
15655 A minor modification to the above pipeline can give us a simple spelling
15656 checker! To determine if you've spelled a word correctly, all you have to
15657 do is look it up in a dictionary. If it is not there, then chances are
15658 that your spelling is incorrect. So, we need a dictionary.
15659 The conventional location for a dictionary is @file{/usr/dict/words}.
15660 On my GNU/Linux system,@footnote{Redhat Linux 6.1, for the November 2000
15661 revision of this article.}
15662 this is a sorted, 45,402 word dictionary.
15664 Now, how to compare our file with the dictionary? As before, we generate
15665 a sorted list of words, one per line:
15668 $ tr '[:upper:]' '[:lower:]' < whats.gnu | tr -cd '[:alnum:]_ \n' |
15669 > tr -s ' ' '\n' | sort -u | ...
15672 Now, all we need is a list of words that are @emph{not} in the
15673 dictionary. Here is where the @command{comm} command comes in.
15676 $ tr '[:upper:]' '[:lower:]' < whats.gnu | tr -cd '[:alnum:]_ \n' |
15677 > tr -s ' ' '\n' | sort -u |
15678 > comm -23 - /usr/dict/words
15681 The @option{-2} and @option{-3} options eliminate lines that are only in the
15682 dictionary (the second file), and lines that are in both files. Lines
15683 only in the first file (standard input, our stream of words), are
15684 words that are not in the dictionary. These are likely candidates for
15685 spelling errors. This pipeline was the first cut at a production
15686 spelling checker on Unix.
15688 There are some other tools that deserve brief mention.
15692 search files for text that matches a regular expression
15695 count lines, words, characters
15698 a T-fitting for data pipes, copies data to files and to standard output
15701 the stream editor, an advanced tool
15704 a data manipulation language, another advanced tool
15707 The software tools philosophy also espoused the following bit of
15708 advice: ``Let someone else do the hard part.'' This means, take
15709 something that gives you most of what you need, and then massage it the
15710 rest of the way until it's in the form that you want.
15716 Each program should do one thing well. No more, no less.
15719 Combining programs with appropriate plumbing leads to results where
15720 the whole is greater than the sum of the parts. It also leads to novel
15721 uses of programs that the authors might never have imagined.
15724 Programs should never print extraneous header or trailer data, since these
15725 could get sent on down a pipeline. (A point we didn't mention earlier.)
15728 Let someone else do the hard part.
15731 Know your toolbox! Use each program appropriately. If you don't have an
15732 appropriate tool, build one.
15735 As of this writing, all the programs we've discussed are available via
15736 anonymous @command{ftp} from: @*
15737 @uref{ftp://gnudist.gnu.org/textutils/textutils-1.22.tar.gz}. (There may
15738 be more recent versions available now.)
15740 None of what I have presented in this column is new. The Software Tools
15741 philosophy was first introduced in the book @cite{Software Tools}, by
15742 Brian Kernighan and P.J. Plauger (Addison-Wesley, ISBN 0-201-03669-X).
15743 This book showed how to write and use software tools. It was written in
15744 1976, using a preprocessor for FORTRAN named @command{ratfor} (RATional
15745 FORtran). At the time, C was not as ubiquitous as it is now; FORTRAN
15746 was. The last chapter presented a @command{ratfor} to FORTRAN
15747 processor, written in @command{ratfor}. @command{ratfor} looks an awful
15748 lot like C; if you know C, you won't have any problem following the
15751 In 1981, the book was updated and made available as @cite{Software Tools
15752 in Pascal} (Addison-Wesley, ISBN 0-201-10342-7). Both books are
15753 still in print and are well worth
15754 reading if you're a programmer. They certainly made a major change in
15755 how I view programming.
15757 The programs in both books are available from
15758 @uref{http://cm.bell-labs.com/who/bwk, Brian Kernighan's home page}.
15759 For a number of years, there was an active
15760 Software Tools Users Group, whose members had ported the original
15761 @command{ratfor} programs to essentially every computer system with a
15762 FORTRAN compiler. The popularity of the group waned in the middle 1980s
15763 as Unix began to spread beyond universities.
15765 With the current proliferation of GNU code and other clones of Unix programs,
15766 these programs now receive little attention; modern C versions are
15767 much more efficient and do more than these programs do. Nevertheless, as
15768 exposition of good programming style, and evangelism for a still-valuable
15769 philosophy, these books are unparalleled, and I recommend them highly.
15771 Acknowledgment: I would like to express my gratitude to Brian Kernighan
15772 of Bell Labs, the original Software Toolsmith, for reviewing this column.
15774 @node GNU Free Documentation License
15775 @appendix GNU Free Documentation License
15779 @node Concept index
15786 @c Local variables:
15787 @c texinfo-column-for-description: 32