Bash (Unix shell)

From Infogalactic: the planetary knowledge core
Jump to: navigation, search
Bash screenshot.png
Screenshot of a Bash session
Original author(s) Brian Fox
Initial release June 8, 1989; 29 years ago (1989-06-08)
Stable release 4.3.42 / August 13, 2015; 3 years ago (2015-08-13)[1]
Written in C
Operating system Cross-platform
Platform GNU
Available in English, multilingual (gettext)
Type Unix shell, command language
License GNU GPL v3+[2]

Bash is a Unix shell and command language written by Brian Fox for the GNU Project as a free software replacement for the Bourne shell.[3][4] Released in 1989,[5] it has been distributed widely as the shell for the GNU operating system and as a default shell on Linux and OS X. It has been ported to Microsoft Windows and distributed with Cygwin and MinGW, to DOS by the DJGPP project, to Novell NetWare and to Android via various terminal emulation applications. In the late 1990s, Bash was a minor player among multiple commonly used shells; at present Bash has overwhelming favor.

Bash is a command processor that typically runs in a text window, where the user types commands that cause actions. Bash can also read commands from a file, called a script. Like all Unix shells, it supports filename globbing (wildcard matching), piping, here documents, command substitution, variables and control structures for condition-testing and iteration. The keywords, syntax and other basic features of the language were all copied from sh. Other features, e.g., history, were copied from csh and ksh. Bash is a POSIX shell, but with a number of extensions.

The name itself is an acronym, a pun, and a description. As an acronym, it stands for Bourne-again shell, referring to its objective as a free replacement for the Bourne shell.[6] As a pun, it expressed that objective in a phrase that sounds similar to born again, a term for spiritual rebirth.[7][8] The name is also descriptive of what it did, bashing together the features of sh, csh, and ksh.[9]

A security hole in Bash dating from version 1.03 (August 1989),[10] dubbed Shellshock, was discovered in early September 2014.[11][12]


Brian Fox began coding Bash on January 10, 1988[13] after Richard Stallman became dissatisfied with the lack of progress being made by a prior developer.[3] Stallman and the Free Software Foundation (FSF) considered a free shell that could run existing shell scripts so strategic to a completely free system built from BSD and GNU code that this was one of the few projects they funded themselves, with Fox undertaking the work as an employee of FSF.[3][14] Fox released Bash as a beta, version .99, on June 8, 1989[5] and remained the primary maintainer until sometime between mid-1992[15] and mid-1994,[16] when he was laid off from FSF[17] and his responsibility was transitioned to another early contributor, Chet Ramey.[18][19][20]

In September 2014, Stéphane Chazelas, a Unix/Linux, network and telecom specialist working in the UK,[21] discovered a security bug in the program. The bug, first disclosed on September 24, was named Shellshock and assigned the numbers CVE-2014-6271, CVE-2014-6277[22] and CVE-2014-7169. The bug was regarded as severe, since CGI scripts using Bash could be vulnerable, enabling arbitrary code execution. The bug is related to how Bash passes function definitions to subshells through environment variables.[23]


The Bash command syntax is a superset of the Bourne shell command syntax. Bash can execute the vast majority of Bourne shell scripts without modification, with the exception of Bourne shell scripts stumbling into fringe syntax behavior interpreted differently in Bash or attempting to run a system command matching a newer Bash builtin, etc. Bash command syntax includes ideas drawn from the Korn shell (ksh) and the C shell (csh) such as command line editing, command history, the directory stack, the $RANDOM and $PPID variables, and POSIX command substitution syntax $(…).

When a user presses the tab key within an interactive command-shell, Bash automatically uses command line completion to match partly typed program names, filenames and variable names. The Bash command-line completion system is very flexible and customizable, and is often packaged with functions that complete arguments and filenames for specific programs and tasks.

Bash's syntax has many extensions lacking in the Bourne shell. Bash can perform integer calculations ("arithmetic evaluation") without spawning external processes. It uses the ((…)) command and the $((…)) variable syntax for this purpose. Its syntax simplifies I/O redirection. For example, it can redirect standard output (stdout) and standard error (stderr) at the same time using the &> operator. This is simpler to type than the Bourne shell equivalent 'command > file 2>&1'. Bash supports process substitution using the <(command) and >(command)syntax, which substitutes the output of (or input to) a command where a filename is normally used. (This is implemented through /proc/fd/ unnamed pipes on systems which support that, or via temporary named pipes where necessary).

When using the 'function' keyword, Bash function declarations are not compatible with Bourne/Korn/POSIX scripts (the Korn shell has the same problem when using 'function'), but Bash accepts the same function declaration syntax as the Bourne and Korn shells, and is POSIX-conformant. Because of these and other differences, Bash shell scripts are rarely runnable under the Bourne or Korn shell interpreters unless deliberately written with that compatibility in mind, which is becoming less common as Linux becomes more widespread. But in POSIX mode,[24] Bash conformance with POSIX is nearly perfect.[citation needed]

Bash supports here documents. Since version 2.05b Bash can redirect standard input (stdin) from a "here string" using the <<< operator.

Bash 3.0 supports in-process regular expression matching using a syntax reminiscent of Perl.[25]

Bash 4.0 introduced support for associative arrays.[24][26] Associative arrays allow a fake support for multi-dimensional (indexed) arrays, in a similar way to AWK:

$ declare -A a         # declare an associative array 'a' faking a bi-dimensional indexed array
$ i=1; j=2             # initialize some indices
$ a[$i,$j]=5           # associate value "5" to key "$i,$j" (i.e. "1,2")
$ echo ${a[$i,$j]}     # print the stored value at key "$i,$j"

Brace expansion

Brace expansion, also called alternation, is a feature copied from the C shell. It generates a set of alternative combinations. Generated results need not exist as files. The results of each expanded string are not sorted and left to right order is preserved:

$ echo a{p,c,d,b}e 
ape ace ade abe
$ echo {a,b,c}{d,e,f} 
ad ae af bd be bf cd ce cf

Users should not use brace expansions in portable shell scripts, because the Bourne shell does not produce the same output.


# A traditional shell does not produce the same output
echo a{p,c,d,b}e # a{p,c,d,b}e

When brace expansion is combined with wildcards, the braces are expanded first, and then the resulting wildcards are substituted normally. Hence, a listing of JPEG and PNG images in the current directory could be obtained using:

ls *.{jpg,jpeg,png}    # expands to *.jpg *.jpeg *.png - after which,
                       # the wildcards are processed
echo *.{png,jp{e,}g}   # echo just show the expansions -
                       # and braces in braces are possible.

In addition to alternation, brace expansion can be used for sequential ranges between two integers or characters separated by double dots. Newer versions of Bash allow a third integer to specify the increment.

$ echo {1..10}
1 2 3 4 5 6 7 8 9 10
$ echo file{1..4}.txt
file1.txt file2.txt file3.txt file4.txt
$ echo {a..e}
a b c d e
$ echo {1..10..3}
1 4 7 10
$ echo {a..j..3}
a d g j

When brace expansion is combined with variable expansion the variable expansion is performed after the brace expansion, which in some cases may necessitate the use of the built-in eval function thus:

$ start=1; end=10
$ echo {$start..$end}  # fails to expand due to the evaluation order
$ eval echo {$start..$end} # variable expansion occurs then resulting string is evaluated
1 2 3 4 5 6 7 8 9 10

Startup scripts

When Bash starts, it executes the commands in a variety of dot files. Though similar to Bash shell script commands, which have execute permission enabled and an interpreter directive like #!/bin/bash, the initialization files used by Bash require neither.

Execution order of startup files

When started as an interactive login shell

Bash reads and executes /etc/profile (if it exists). (Often this file calls /etc/bash.bashrc.)

After reading that file, it looks for ~/.bash_profile, ~/.bash_login, and ~/.profile in that order, and reads and executes the first one that exists and is readable.

When a login shell exits

Bash reads and executes ~/.bash_logout (if it exists).

When started as an interactive shell (but not a login shell)

Bash reads and executes ~/.bashrc (if it exists). This may be inhibited by using the --norc option. The --rcfile file option forces Bash to read and execute commands from file instead of ~/.bashrc.

Comparison with the Bourne shell and csh startup sequences

Elements of Bash were derived from the Bourne shell and csh, and allow limited startup file sharing with the Bourne shell and provide some startup features familiar to users of the csh.

Setting inheritable environment variables

The Bourne shell uses the ~/.profile at login to set environment variables that subprocesses then inherit. Bash can use the ~/.profile in a compatible way, by executing it explicitly from the Bash-specific ~/.bash_profile or ~/.bash_login with the line below. Bash-specific syntax can be kept out of the ~/.profile to keep the latter compatible with the Bourne shell.

. ~/.profile
Aliases and Functions

These two facilities, aliases from csh and the more general functions that largely supersede them from Bourne shell, were not typically inheritable from the login shell, and had to be redefined in each subshell spawned from the login shell. Although there is an ENV environment variable that could be applied to the problem, both csh and Bash support per-subshell startup files that address it directly. In Bash, the ~/.bashrc is called for interactive subshells. If user-defined functions from the ~/.bashrc are desired in the login shell as well, the ~/.bash_login can include the line below after any setting up of environment variables:

. ~/.bashrc
Commands performed only at login and logout

The csh supports a ~/.login file for purposes of tasks performed only during initial login, such as displaying system load, disk status, whether email has come in, logging the login time, etc. The Bourne shell can emulate this in the ~/.profile, but doesn't predefine a file name. To achieve similar semantics to the csh model, the ~/.bash_profile can contain the line below, after the environment setup and function setup:

. ~/.bash_login

Likewise, the csh has a ~/.logout file run only when the login shell exits. The Bash equivalent is ~/.bash_logout, and requires no special setup. In the Bourne shell, the trap built-in can be used to achieve a similar effect.

Legacy-compatible Bash startup example

The skeleton ~/.bash_profile below is compatible with the Bourne shell and gives semantics similar to csh for the ~/.bashrc and ~/.bash_login. The [ -r filename ] are tests to see if the filename exists and is readable, simply skipping the part after the && if it's not.

[ -r ~/.profile ] && . ~/.profile             # set up environment, once, Bourne-sh syntax only.
if [ -n "$PS1" ] ; then                       # are we interactive?
   [ -r ~/.bashrc     ] && . ~/.bashrc        # tty/prompt/function setup for interactive shells.
   [ -r ~/.bash_login ] && . ~/.bash_login    # any at-login tasks for login shell only.

Operating system issues in Bash startup

Some versions of Unix and Linux contain Bash system startup scripts, generally under the /etc directories. Bash calls these as part of its standard initialization, but other startup files can read them in a different order than the documented Bash startup sequence. The default content of the root user's files may also have issues, as well as the skeleton files the system provides to new user accounts upon setup. The startup scripts that launch the X window system may also do surprising things with the user's Bash startup scripts in an attempt to set up user-environment variables before launching the window manager. These issues can often be addressed using a ~/.xsession or ~/.xprofile file to read the ~/.profile — which provides the environment variables which Bash shell windows spawned from the window manager need, such as xterm or Gnome Terminal.


Invoking Bash with the --posix option or stating set -o posix in a script causes Bash to conform very closely to the POSIX 1003.2 standard.[27] Bash shell scripts intended for portability should at least take into account the Bourne shell it intends to replace. Bash has certain features that the traditional Bourne shell lacks. They include:[27]

  • Certain extended invocation options
  • Command substitution using $( ) notation (this feature is part of the POSIX 1003.2 standard though)
  • Brace expansion
  • Certain array operations, and associative arrays
  • The double brackets extended test construct
  • The double-parentheses arithmetic-evaluation construct
  • Certain string-manipulation operations
  • Process substitution
  • A Regular Expression matching operator
  • Bash-specific builtins
  • Coprocesses

Keyboard shortcuts

Bash uses readline to provide keyboard shortcuts for command line editing using the default (Emacs) key bindings. Vi-bindings can be enabled by running set -o vi.[28]

Process management

The Bash shell has two modes of execution for commands: batch (Unix), and concurrent mode.

To execute commands in batch (i.e., in sequence) they must be separated by the character ";":

command1; command2

in this example, when command1 is finished, command2 is executed.

To have a concurrent execution of command1 and command2, they must be executed in the Bash shell in the following way:

command1 & command2

In this case command1 is executed in background (symbol &), returning immediately the control to the shell that executes command2.


  • Normally a command is executed in foreground (fg). The control of the shell returns to the user after the command finishes its execution and the special variable $! is set to the process ID of the newly spawned background job.
  • With the symbol & after the command, it can be executed in background (bg). The shell is ready to execute other commands, concurrently to the first command.
  • A program in the running state and in foreground (fg) can be suspended pressing Ctrl+z
  • A suspended program can be resumed in foreground using the command fg or background using the command bg.
  • A background process may become blocked on attempts to write to the terminal device (as distinct from simple writes to its standard output).
  • The shell can synchronize with the completion of specific background jobs using the wait command by supplying a process ID or job number, or await completion of all background jobs by calling wait with no arguments.

Conditional Execution

Bash supplies "conditional execution" command separators which allow for the execution of a command to be contingent on the exit code set by a precedent command. For example:

cd "$SOMEWHERE" && ./do_something || echo "An error occurred" >&2

Where ./do_something is only executed if the cd (change directory) command was "successful" (returned an exit status of zero) and the echo command would only be executed if either the cd' or the ./do_something command return an "error" (non-zero exit status).

For all commands the exit status is stored in the special variable $?. Bash also support if ...;then ...;else ...;fi and case $VARIABLE in $pattern)...;;$other_pattern)...;; esac forms of conditional command evaluation.

Bug reporting

An external command called bashbug reports Bash shell bugs.[29][30] When the command is invoked, it brings up the user's default editor with a form to fill in. The form is mailed to the Bash maintainers (or optionally to other email addresses).

See also


  1. "Bash-4.3 patch list". May 19, 2015. Retrieved May 19, 2015.<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
  2. GNU Project. "README file". Bash is free software, distributed under the terms of the [GNU] General Public License as published by the Free Software Foundation, version 3 of the License (or any later version).<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
  3. 3.0 3.1 3.2 Lua error in Module:Citation/CS1/Identifiers at line 47: attempt to index field 'wikibase' (a nil value).
  4. Hamilton, Naomi (May 30, 2008), "The A-Z of Programming Languages: BASH/Bourne-Again Shell", Computerworld: 2, retrieved March 21, 2011, When Richard Stallman decided to create a full replacement for the then-encumbered Unix systems, he knew that he would eventually have to have replacements for all of the common utilities, especially the standard shell, and those replacements would have to have acceptable licensing.<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
  5. 5.0 5.1 Brian Fox (forwarded by Leonard H. Tower Jr.) (June 8, 1989). "Bash is in beta release!". Newsgroupgnu.announce. Retrieved October 28, 2010.<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
  6. C Programming by Al Stevens, Dr. Dobb's Journal, July 1, 2001
  7. Richard Stallman (November 12, 2010). "About the GNU Project". Free Software Foundation. Archived from the original on April 24, 2011. Retrieved March 13, 2011. "Bourne Again Shell" is a play on the name Bourne Shell, which was the usual shell on Unix.<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
  8. Gattol, Markus (March 13, 2011), Bourne-again Shell, retrieved March 13, 2011, The name is a pun on the name of the Bourne shell (sh), an early and important Unix shell written by Stephen Bourne and distributed with Version 7 Unix circa 1978, and the concept of being "born again".<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
  9. Ian Darwin (June 13, 1989). "at&t-free ksh (was: job control is a bug, not a feature)". Newsgroupcomp.os.minix. Retrieved March 21, 2011. Yup, the gnu project's Born Again Shell ("bash") is an attempt at bashing all the features of sh together with many of those from both csh and ksh.<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
  10. Chazelas, Stephane (4 October 2014). "oss-sec mailing list archives". Retrieved 4 October 2014.<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
  11. Leyden, John (September 24, 2014). "Patch Bash NOW: 'Shell Shock' bug blasts OS X, Linux systems wide open". The Register. Retrieved September 25, 2014.<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
  12. Perlroth, Nicole (September 25, 2014). "Security Experts Expect 'Shellshock' Software Bug in Bash to Be Significant". New York Times. Retrieved September 25, 2014.<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
  13. Brian Fox (August 29, 1996), shell.c, Free Software Foundation, Birthdate: Sunday, January 10th, 1988. Initial author: Brian Fox<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
  14. Richard Stallman (October 3, 2010). "About the GNU Project". Free Software Foundation. Archived from the original on April 24, 2011. Retrieved March 21, 2011. Free Software Foundation employees have written and maintained a number of GNU software packages. Two notable ones are the C library and the shell. ... We funded development of these programs because the GNU Project was not just about tools or a development environment. Our goal was a complete operating system, and these programs were needed for that goal.<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
  15. Lua error in Module:Citation/CS1/Identifiers at line 47: attempt to index field 'wikibase' (a nil value).
  16. Ramey, Chet (August 1, 1994). "Bash - the GNU shell (Reflections and Lessons Learned)". Linux Journal. Archived from the original on December 5, 2008. Retrieved November 13, 2008.<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
  17. Chet Ramey (October 31, 2010), Dates in your Computerworld interview, retrieved October 31, 2010<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
  18. Chet Ramey (June 12, 1989). "Bash 0.99 fixes & improvements". Newsgroupgnu.bash.bug. Retrieved November 1, 2010.<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
  19. Chet Ramey (July 24, 1989). "Some bash-1.02 fixes". Newsgroupgnu.bash.bug. Retrieved October 30, 2010.<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
  20. Brian Fox (March 2, 1990). "Availability of bash 1.05". Newsgroupgnu.bash.bug. Retrieved October 30, 2010.<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
  23. Huzaifa Sidhpurwala (2014-09-24). "Bash specially-crafted environment variables code injection attack". Redhat.<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
  24. 24.0 24.1 "6.11 Bash POSIX Mode", The GNU Bash Reference Manual, for Bash, Version 4.1, December 23, 2009, archived from the original on December 3, 2010, retrieved October 26, 2010<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
  25. The syntax matches that shown on the regex(7) man page.
  26. "The shell provides associative array variables, with the appropriate support to create, delete, assign values to, and expand them."
  27. 27.0 27.1 Mendel Cooper. "Portability Issues". The Linux Documentation Project.<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
  28. "BASH Help - A Bash Tutorial". October 5, 2012. Retrieved July 21, 2013.<templatestyles src="Module:Citation/CS1/styles.css"></templatestyles>
  29. bashbug(1),
  30. "Linux / Unix Command: bashbug",

External links