Visit Apps Ved →

parenthesis

Shell parentheses for subshells, arrays, and arithmetic

TLDR

Run commands in a subshell
$ (cd [/tmp] && [command])
Create an array
$ array=([one] [two] [three])
Command grouping without subshell
$ { [command1]; [command2]; }
Function definition
$ function_name() { [commands]; }
Arithmetic evaluation
$ (( count++ ))

SYNOPSIS

( commands )((arithmetic))

DESCRIPTION

Parentheses ( ) in shell have several distinct uses depending on context:Subshell execution: Commands in (...) run in a child shell. Environment changes (cd, variable assignments) don't affect the parent shell.Array literals: In bash/zsh, array=(a b c) creates an array.Function definition: name() { ... } defines a function (parentheses are part of syntax, not grouping).Command substitution: $(...) captures command output.Arithmetic: ((...)) performs arithmetic evaluation, $((...)) expands to the result.

SUBSHELL BEHAVIOR

$ # Changes don't affect parent
(cd /tmp; pwd)  # prints /tmp
pwd             # still original directory

# Variables don't leak
(x=5)
echo $x         # empty or original value

ARITHMETIC (( ))

$ (( x = 5 + 3 ))    # Assignment
(( x++ ))          # Increment
(( x > 5 )) && echo "big"  # Condition
result=$(( a * b ))         # Capture result

CAVEATS

Subshells have overhead from process creation. For simple grouping without isolation, use braces { ...; } instead.{ } grouping requires a space after { and semicolon before }; parentheses don't.Subshell exit status is visible to parent, but variable changes are not.Nested parentheses may need careful quoting to avoid syntax issues.

SEE ALSO

bash(1), sh(1), test(1)